Copyright (c) 2016-present, RxJava Contributors.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in
compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is
distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See
the License for the specific language governing permissions and limitations under the License.
/**
 * Copyright (c) 2016-present, RxJava Contributors.
 *
 * Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in
 * compliance with the License. You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software distributed under the License is
 * distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See
 * the License for the specific language governing permissions and limitations under the License.
 */
package io.reactivex;

import java.util.*;
import java.util.concurrent.*;

import org.reactivestreams.*;

import io.reactivex.annotations.*;
import io.reactivex.disposables.Disposable;
import io.reactivex.exceptions.Exceptions;
import io.reactivex.flowables.*;
import io.reactivex.functions.*;
import io.reactivex.internal.functions.*;
import io.reactivex.internal.fuseable.*;
import io.reactivex.internal.operators.flowable.*;
import io.reactivex.internal.operators.mixed.*;
import io.reactivex.internal.operators.observable.*;
import io.reactivex.internal.schedulers.ImmediateThinScheduler;
import io.reactivex.internal.subscribers.*;
import io.reactivex.internal.util.*;
import io.reactivex.parallel.ParallelFlowable;
import io.reactivex.plugins.RxJavaPlugins;
import io.reactivex.schedulers.*;
import io.reactivex.subscribers.*;

The Flowable class that implements the Reactive Streams
Pattern and offers factory methods, intermediate operators and the ability to consume reactive dataflows.
 Reactive Streams operates with Publishers which Flowable extends. Many operators therefore accept general Publishers directly and allow direct interoperation with other Reactive Streams implementations. 
 The Flowable hosts the default buffer size of 128 elements for operators, accessible via bufferSize(), that can be overridden globally via the system parameter rx2.buffer-size. Most operators, however, have overloads that allow setting their internal buffer size explicitly. 

The documentation for this class makes use of marble diagrams. The following legend explains these diagrams:


 The Flowable follows the protocol 

     onSubscribe onNext* (onError | onComplete)?
 where the stream can be disposed through the Subscription instance provided to consumers through Subscriber.onSubscribe(Subscription). Unlike the Observable.subscribe() of version 1.x, subscribe(Subscriber) does not allow external cancellation of a subscription and the Subscriber instance is expected to expose such capability if needed. 
 Flowables support backpressure and require Subscribers to signal demand via Subscription.request(long). 

Example:

Disposable d = Flowable.just("Hello world!")
    .delay(1, TimeUnit.SECONDS)
    .subscribeWith(new DisposableSubscriber<String>() {
        @Override public void onStart() {
            System.out.println("Start!");
            request(1);
        }
        @Override public void onNext(String t) {
            System.out.println(t);
            request(1);
        }
        @Override public void onError(Throwable t) {
            t.printStackTrace();
        }
        @Override public void onComplete() {
            System.out.println("Done!");
        }
    });
Thread.sleep(500);
// the sequence can now be cancelled via dispose()
d.dispose();

 The Reactive Streams specification is relatively strict when defining interactions between Publishers and Subscribers, so much so that there is a significant performance penalty due certain timing requirements and the need to prepare for invalid request amounts via Subscription.request(long). Therefore, RxJava has introduced the FlowableSubscriber interface that indicates the consumer can be driven with relaxed rules. All RxJava operators are implemented with these relaxed rules in mind. If the subscribing Subscriber does not implement this interface, for example, due to it being from another Reactive Streams compliant library, the Flowable will automatically apply a compliance wrapper around it. 
 Flowable is an abstract class, but it is not advised to implement sources and custom operators by extending the class directly due to the large amounts of Reactive Streams
rules to be followed to the letter. See the wiki for
some guidance if such custom implementations are necessary.
 The recommended way of creating custom Flowables is by using the create(FlowableOnSubscribe<Object>, BackpressureStrategy) factory method: 

Flowable<String> source = Flowable.create(new FlowableOnSubscribe<String>() {
    @Override
    public void subscribe(FlowableEmitter<String> emitter) throws Exception {
        // signal an item
        emitter.onNext("Hello");
        // could be some blocking operation
        Thread.sleep(1000);
        
        // the consumer might have cancelled the flow
        if (emitter.isCancelled() {
            return;
        }
        emitter.onNext("World");
        Thread.sleep(1000);
        // the end-of-sequence has to be signaled, otherwise the
        // consumers may never finish
        emitter.onComplete();
    }
}, BackpressureStrategy.BUFFER);
System.out.println("Subscribe!");
source.subscribe(System.out::println);
System.out.println("Done!");

 RxJava reactive sources, such as Flowable, are generally synchronous and sequential in nature. In the ReactiveX design, the location (thread) where operators run is orthogonal to when the operators can work with data. This means that asynchrony and parallelism has to be explicitly expressed via operators such as subscribeOn(Scheduler), observeOn(Scheduler) and parallel(). In general, operators featuring a Scheduler parameter are introducing this type of asynchrony into the flow. 

For more information see the ReactiveX
documentation.
Type parameters:
<T> – 
           the type of the items emitted by the Flowable
See Also:
Observable
ParallelFlowable
DisposableSubscriber/**
 * The Flowable class that implements the <a href="https://github.com/reactive-streams/reactive-streams-jvm">Reactive Streams</a>
 * Pattern and offers factory methods, intermediate operators and the ability to consume reactive dataflows.
 * <p>
 * Reactive Streams operates with {@link Publisher}s which {@code Flowable} extends. Many operators
 * therefore accept general {@code Publisher}s directly and allow direct interoperation with other
 * Reactive Streams implementations.
 * <p>
 * The Flowable hosts the default buffer size of 128 elements for operators, accessible via {@link #bufferSize()},
 * that can be overridden globally via the system parameter {@code rx2.buffer-size}. Most operators, however, have
 * overloads that allow setting their internal buffer size explicitly.
 * <p>
 * The documentation for this class makes use of marble diagrams. The following legend explains these diagrams:
 * <p>
 * <img width="640" height="317" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/legend.png" alt="">
 * <p>
 * The {@code Flowable} follows the protocol
 * <pre><code>
 *      onSubscribe onNext* (onError | onComplete)?
 * </code></pre>
 * where the stream can be disposed through the {@link Subscription} instance provided to consumers through
 * {@link Subscriber#onSubscribe(Subscription)}.
 * Unlike the {@code Observable.subscribe()} of version 1.x, {@link #subscribe(Subscriber)} does not allow external cancellation
 * of a subscription and the {@link Subscriber} instance is expected to expose such capability if needed.
 * <p>
 * Flowables support backpressure and require {@link Subscriber}s to signal demand via {@link Subscription#request(long)}.
 * <p>
 * Example:
 * <pre><code>
 * Disposable d = Flowable.just("Hello world!")
 *     .delay(1, TimeUnit.SECONDS)
 *     .subscribeWith(new DisposableSubscriber&lt;String&gt;() {
 *         &#64;Override public void onStart() {
 *             System.out.println("Start!");
 *             request(1);
 *         }
 *         &#64;Override public void onNext(String t) {
 *             System.out.println(t);
 *             request(1);
 *         }
 *         &#64;Override public void onError(Throwable t) {
 *             t.printStackTrace();
 *         }
 *         &#64;Override public void onComplete() {
 *             System.out.println("Done!");
 *         }
 *     });
 *
 * Thread.sleep(500);
 * // the sequence can now be cancelled via dispose()
 * d.dispose();
 * </code></pre>
 * <p>
 * The Reactive Streams specification is relatively strict when defining interactions between {@code Publisher}s and {@code Subscriber}s, so much so
 * that there is a significant performance penalty due certain timing requirements and the need to prepare for invalid 
 * request amounts via {@link Subscription#request(long)}.
 * Therefore, RxJava has introduced the {@link FlowableSubscriber} interface that indicates the consumer can be driven with relaxed rules.
 * All RxJava operators are implemented with these relaxed rules in mind.
 * If the subscribing {@code Subscriber} does not implement this interface, for example, due to it being from another Reactive Streams compliant
 * library, the Flowable will automatically apply a compliance wrapper around it.
 * <p>
 * {@code Flowable} is an abstract class, but it is not advised to implement sources and custom operators by extending the class directly due
 * to the large amounts of <a href="https://github.com/reactive-streams/reactive-streams-jvm#specification">Reactive Streams</a>
 * rules to be followed to the letter. See <a href="https://github.com/ReactiveX/RxJava/wiki/Writing-operators-for-2.0">the wiki</a> for
 * some guidance if such custom implementations are necessary.
 * <p>
 * The recommended way of creating custom {@code Flowable}s is by using the {@link #create(FlowableOnSubscribe, BackpressureStrategy)} factory method:
 * <pre><code>
 * Flowable&lt;String&gt; source = Flowable.create(new FlowableOnSubscribe&lt;String&gt;() {
 *     &#64;Override
 *     public void subscribe(FlowableEmitter&lt;String&gt; emitter) throws Exception {
 *
 *         // signal an item
 *         emitter.onNext("Hello");
 *
 *         // could be some blocking operation
 *         Thread.sleep(1000);
 *         
 *         // the consumer might have cancelled the flow
 *         if (emitter.isCancelled() {
 *             return;
 *         }
 *
 *         emitter.onNext("World");
 *
 *         Thread.sleep(1000);
 *
 *         // the end-of-sequence has to be signaled, otherwise the
 *         // consumers may never finish
 *         emitter.onComplete();
 *     }
 * }, BackpressureStrategy.BUFFER);
 *
 * System.out.println("Subscribe!");
 * 
 * source.subscribe(System.out::println);
 * 
 * System.out.println("Done!");
 * </code></pre>
 * <p>
 * RxJava reactive sources, such as {@code Flowable}, are generally synchronous and sequential in nature. In the ReactiveX design, the location (thread)
 * where operators run is <i>orthogonal</i> to when the operators can work with data. This means that asynchrony and parallelism
 * has to be explicitly expressed via operators such as {@link #subscribeOn(Scheduler)}, {@link #observeOn(Scheduler)} and {@link #parallel()}. In general,
 * operators featuring a {@link Scheduler} parameter are introducing this type of asynchrony into the flow. 
 * <p>
 * For more information see the <a href="http://reactivex.io/documentation/Publisher.html">ReactiveX
 * documentation</a>.
 *
 * @param <T>
 *            the type of the items emitted by the Flowable
 * @see Observable
 * @see ParallelFlowable
 * @see io.reactivex.subscribers.DisposableSubscriber
 */
public abstract class Flowable<T> implements Publisher<T> {
    
The default buffer size. /** The default buffer size. */
    static final int BUFFER_SIZE;
    static {
        BUFFER_SIZE = Math.max(1, Integer.getInteger("rx2.buffer-size", 128));
    }

    
Mirrors the one Publisher in an Iterable of several Publishers that first either emits an item or sends
a termination notification.



 
Backpressure:
 
The operator itself doesn't interfere with backpressure which is determined by the winning Publisher's backpressure behavior.
 
Scheduler:
 
amb does not operate by default on a particular Scheduler.

Params:
sources – 
           an Iterable of Publishers sources competing to react first. A subscription to each Publisher will
           occur in the same order as in this Iterable.
Type parameters:
<T> – the common element type
See Also:
ReactiveX operators documentation: Amb
Returns:
a Flowable that emits the same sequence as whichever of the source Publishers first
        emitted an item or sent a termination notification/**
     * Mirrors the one Publisher in an Iterable of several Publishers that first either emits an item or sends
     * a termination notification.
     * <p>
     * <img width="640" height="385" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/amb.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator itself doesn't interfere with backpressure which is determined by the winning
     *  {@code Publisher}'s backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code amb} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element type
     * @param sources
     *            an Iterable of Publishers sources competing to react first. A subscription to each Publisher will
     *            occur in the same order as in this Iterable.
     * @return a Flowable that emits the same sequence as whichever of the source Publishers first
     *         emitted an item or sent a termination notification
     * @see <a href="http://reactivex.io/documentation/operators/amb.html">ReactiveX operators documentation: Amb</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> amb(Iterable<? extends Publisher<? extends T>> sources) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        return RxJavaPlugins.onAssembly(new FlowableAmb<T>(null, sources));
    }

    
Mirrors the one Publisher in an array of several Publishers that first either emits an item or sends
a termination notification.



 
Backpressure:
 
The operator itself doesn't interfere with backpressure which is determined by the winning Publisher's backpressure behavior.
 
Scheduler:
 
ambArray does not operate by default on a particular Scheduler.

Params:
sources – 
           an array of Publisher sources competing to react first. A subscription to each Publisher will
           occur in the same order as in this Iterable.
Type parameters:
<T> – the common element type
See Also:
ReactiveX operators documentation: Amb
Returns:
a Flowable that emits the same sequence as whichever of the source Publishers first
        emitted an item or sent a termination notification/**
     * Mirrors the one Publisher in an array of several Publishers that first either emits an item or sends
     * a termination notification.
     * <p>
     * <img width="640" height="385" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/amb.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator itself doesn't interfere with backpressure which is determined by the winning
     *  {@code Publisher}'s backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code ambArray} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element type
     * @param sources
     *            an array of Publisher sources competing to react first. A subscription to each Publisher will
     *            occur in the same order as in this Iterable.
     * @return a Flowable that emits the same sequence as whichever of the source Publishers first
     *         emitted an item or sent a termination notification
     * @see <a href="http://reactivex.io/documentation/operators/amb.html">ReactiveX operators documentation: Amb</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> ambArray(Publisher<? extends T>... sources) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        int len = sources.length;
        if (len == 0) {
            return empty();
        } else
        if (len == 1) {
            return fromPublisher(sources[0]);
        }
        return RxJavaPlugins.onAssembly(new FlowableAmb<T>(sources, null));
    }

    
Returns the default internal buffer size used by most async operators.
The value can be overridden via system parameter rx2.buffer-size before the Flowable class is loaded.
Returns:
the default internal buffer size./**
     * Returns the default internal buffer size used by most async operators.
     * <p>The value can be overridden via system parameter {@code rx2.buffer-size}
     * <em>before</em> the Flowable class is loaded.
     * @return the default internal buffer size.
     */
    public static int bufferSize() {
        return BUFFER_SIZE;
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatest(Publisher<? extends T>[] sources, Function<? super Object[], ? extends R> combiner) {
        return combineLatest(sources, combiner, bufferSize());
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If there are no source Publishers provided, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If there are no source Publishers provided, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatest(Function<? super Object[], ? extends R> combiner, Publisher<? extends T>... sources) {
        return combineLatest(sources, combiner, bufferSize());
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
bufferSize – 
           the internal buffer size and prefetch amount applied to every source Flowable
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @param bufferSize
     *            the internal buffer size and prefetch amount applied to every source Flowable
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatest(Publisher<? extends T>[] sources, Function<? super Object[], ? extends R> combiner, int bufferSize) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        if (sources.length == 0) {
            return empty();
        }
        ObjectHelper.requireNonNull(combiner, "combiner is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableCombineLatest<T, R>(sources, combiner, bufferSize, false));
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatest(Iterable<? extends Publisher<? extends T>> sources,
            Function<? super Object[], ? extends R> combiner) {
        return combineLatest(sources, combiner, bufferSize());
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting any items and
without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
bufferSize – 
           the internal buffer size and prefetch amount applied to every source Flowable
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting any items and
     * without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @param bufferSize
     *            the internal buffer size and prefetch amount applied to every source Flowable
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatest(Iterable<? extends Publisher<? extends T>> sources,
            Function<? super Object[], ? extends R> combiner, int bufferSize) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        ObjectHelper.requireNonNull(combiner, "combiner is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableCombineLatest<T, R>(sources, combiner, bufferSize, false));
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatestDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatestDelayError(Publisher<? extends T>[] sources,
            Function<? super Object[], ? extends R> combiner) {
        return combineLatestDelayError(sources, combiner, bufferSize());
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function and delays any error from the sources until
all source Publishers terminate.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If there are no source Publishers provided, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatestDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function and delays any error from the sources until
     * all source Publishers terminate.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If there are no source Publishers provided, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatestDelayError(Function<? super Object[], ? extends R> combiner,
            Publisher<? extends T>... sources) {
        return combineLatestDelayError(sources, combiner, bufferSize());
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publisher, where this
aggregation is defined by a specified function and delays any error from the sources until
all source Publishers terminate.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If there are no source Publishers provided, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatestDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
bufferSize – 
           the internal buffer size and prefetch amount applied to every source Publisher
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publisher, where this
     * aggregation is defined by a specified function and delays any error from the sources until
     * all source Publishers terminate.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If there are no source Publishers provided, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @param bufferSize
     *            the internal buffer size and prefetch amount applied to every source Publisher
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatestDelayError(Function<? super Object[], ? extends R> combiner,
            int bufferSize, Publisher<? extends T>... sources) {
        return combineLatestDelayError(sources, combiner, bufferSize);
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function and delays any error from the sources until
all source Publishers terminate.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatestDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
bufferSize – 
           the internal buffer size and prefetch amount applied to every source Flowable
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function and delays any error from the sources until
     * all source Publishers terminate.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided array of source Publishers is empty, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @param bufferSize
     *            the internal buffer size and prefetch amount applied to every source Flowable
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatestDelayError(Publisher<? extends T>[] sources,
            Function<? super Object[], ? extends R> combiner, int bufferSize) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        ObjectHelper.requireNonNull(combiner, "combiner is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        if (sources.length == 0) {
            return empty();
        }
        return RxJavaPlugins.onAssembly(new FlowableCombineLatest<T, R>(sources, combiner, bufferSize, true));
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function and delays any error from the sources until
all source Publishers terminate.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatestDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function and delays any error from the sources until
     * all source Publishers terminate.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatestDelayError(Iterable<? extends Publisher<? extends T>> sources,
            Function<? super Object[], ? extends R> combiner) {
        return combineLatestDelayError(sources, combiner, bufferSize());
    }

    
Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
the source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function and delays any error from the sources until
all source Publishers terminate.
 Note on method signature: since Java doesn't allow creating a generic array with new T[], the implementation of this operator has to create an Object[] instead. Unfortunately, a Function<Integer[], R> passed to the method would trigger a ClassCastException. 

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.

If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting
any items and without any calls to the combiner function.

 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatestDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the collection of source Publishers
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
bufferSize – 
           the internal buffer size and prefetch amount applied to every source Flowable
Type parameters:
<T> – 
           the common base type of source values
<R> – 
           the result type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines a collection of source Publishers by emitting an item that aggregates the latest values of each of
     * the source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function and delays any error from the sources until
     * all source Publishers terminate.
     * <p>
     * Note on method signature: since Java doesn't allow creating a generic array with {@code new T[]}, the
     * implementation of this operator has to create an {@code Object[]} instead. Unfortunately, a
     * {@code Function<Integer[], R>} passed to the method would trigger a {@code ClassCastException}.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * If the provided iterable of source Publishers is empty, the resulting sequence completes immediately without emitting
     * any items and without any calls to the combiner function.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatestDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the common base type of source values
     * @param <R>
     *            the result type
     * @param sources
     *            the collection of source Publishers
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @param bufferSize
     *            the internal buffer size and prefetch amount applied to every source Flowable
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T, R> Flowable<R> combineLatestDelayError(Iterable<? extends Publisher<? extends T>> sources,
            Function<? super Object[], ? extends R> combiner, int bufferSize) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        ObjectHelper.requireNonNull(combiner, "combiner is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableCombineLatest<T, R>(sources, combiner, bufferSize, true));
    }

    
Combines two source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from either of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines two source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from either of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            BiFunction<? super T1, ? super T2, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        Function<Object[], R> f = Functions.toFunction(combiner);
        return combineLatest(f, source1, source2);
    }

    
Combines three source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
source3 – 
           the third source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<T3> – the element type of the third source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines three source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <T3> the element type of the third source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param source3
     *            the third source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            Publisher<? extends T3> source3,
            Function3<? super T1, ? super T2, ? super T3, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        return combineLatest(Functions.toFunction(combiner), source1, source2, source3);
    }

    
Combines four source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
source3 – 
           the third source Publisher
source4 – 
           the fourth source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<T3> – the element type of the third source
<T4> – the element type of the fourth source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines four source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <T3> the element type of the third source
     * @param <T4> the element type of the fourth source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param source3
     *            the third source Publisher
     * @param source4
     *            the fourth source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            Publisher<? extends T3> source3, Publisher<? extends T4> source4,
            Function4<? super T1, ? super T2, ? super T3, ? super T4, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4);
    }

    
Combines five source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
source3 – 
           the third source Publisher
source4 – 
           the fourth source Publisher
source5 – 
           the fifth source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<T3> – the element type of the third source
<T4> – the element type of the fourth source
<T5> – the element type of the fifth source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines five source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <T3> the element type of the third source
     * @param <T4> the element type of the fourth source
     * @param <T5> the element type of the fifth source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param source3
     *            the third source Publisher
     * @param source4
     *            the fourth source Publisher
     * @param source5
     *            the fifth source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            Publisher<? extends T3> source3, Publisher<? extends T4> source4,
            Publisher<? extends T5> source5,
            Function5<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5);
    }

    
Combines six source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
source3 – 
           the third source Publisher
source4 – 
           the fourth source Publisher
source5 – 
           the fifth source Publisher
source6 – 
           the sixth source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<T3> – the element type of the third source
<T4> – the element type of the fourth source
<T5> – the element type of the fifth source
<T6> – the element type of the sixth source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines six source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <T3> the element type of the third source
     * @param <T4> the element type of the fourth source
     * @param <T5> the element type of the fifth source
     * @param <T6> the element type of the sixth source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param source3
     *            the third source Publisher
     * @param source4
     *            the fourth source Publisher
     * @param source5
     *            the fifth source Publisher
     * @param source6
     *            the sixth source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            Publisher<? extends T3> source3, Publisher<? extends T4> source4,
            Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Function6<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6);
    }

    
Combines seven source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
source3 – 
           the third source Publisher
source4 – 
           the fourth source Publisher
source5 – 
           the fifth source Publisher
source6 – 
           the sixth source Publisher
source7 – 
           the seventh source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<T3> – the element type of the third source
<T4> – the element type of the fourth source
<T5> – the element type of the fifth source
<T6> – the element type of the sixth source
<T7> – the element type of the seventh source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines seven source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <T3> the element type of the third source
     * @param <T4> the element type of the fourth source
     * @param <T5> the element type of the fifth source
     * @param <T6> the element type of the sixth source
     * @param <T7> the element type of the seventh source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param source3
     *            the third source Publisher
     * @param source4
     *            the fourth source Publisher
     * @param source5
     *            the fifth source Publisher
     * @param source6
     *            the sixth source Publisher
     * @param source7
     *            the seventh source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, T7, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            Publisher<? extends T3> source3, Publisher<? extends T4> source4,
            Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Publisher<? extends T7> source7,
            Function7<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? super T7, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        ObjectHelper.requireNonNull(source7, "source7 is null");
        return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6, source7);
    }

    
Combines eight source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
source3 – 
           the third source Publisher
source4 – 
           the fourth source Publisher
source5 – 
           the fifth source Publisher
source6 – 
           the sixth source Publisher
source7 – 
           the seventh source Publisher
source8 – 
           the eighth source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<T3> – the element type of the third source
<T4> – the element type of the fourth source
<T5> – the element type of the fifth source
<T6> – the element type of the sixth source
<T7> – the element type of the seventh source
<T8> – the element type of the eighth source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines eight source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <T3> the element type of the third source
     * @param <T4> the element type of the fourth source
     * @param <T5> the element type of the fifth source
     * @param <T6> the element type of the sixth source
     * @param <T7> the element type of the seventh source
     * @param <T8> the element type of the eighth source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param source3
     *            the third source Publisher
     * @param source4
     *            the fourth source Publisher
     * @param source5
     *            the fifth source Publisher
     * @param source6
     *            the sixth source Publisher
     * @param source7
     *            the seventh source Publisher
     * @param source8
     *            the eighth source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, T7, T8, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            Publisher<? extends T3> source3, Publisher<? extends T4> source4,
            Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Publisher<? extends T7> source7, Publisher<? extends T8> source8,
            Function8<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? super T7, ? super T8, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        ObjectHelper.requireNonNull(source7, "source7 is null");
        ObjectHelper.requireNonNull(source8, "source8 is null");
        return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6, source7, source8);
    }

    
Combines nine source Publishers by emitting an item that aggregates the latest values of each of the
source Publishers each time an item is received from any of the source Publishers, where this
aggregation is defined by a specified function.

If any of the sources never produces an item but only terminates (normally or with an error), the
resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
If that input source is also synchronous, other sources after it will not be subscribed to.



 
Backpressure:
 
The returned Publisher honors backpressure from downstream. The source Publishers are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal MissingBackpressureException) and may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
combineLatest does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           the second source Publisher
source3 – 
           the third source Publisher
source4 – 
           the fourth source Publisher
source5 – 
           the fifth source Publisher
source6 – 
           the sixth source Publisher
source7 – 
           the seventh source Publisher
source8 – 
           the eighth source Publisher
source9 – 
           the ninth source Publisher
combiner – 
           the aggregation function used to combine the items emitted by the source Publishers
Type parameters:
<T1> – the element type of the first source
<T2> – the element type of the second source
<T3> – the element type of the third source
<T4> – the element type of the fourth source
<T5> – the element type of the fifth source
<T6> – the element type of the sixth source
<T7> – the element type of the seventh source
<T8> – the element type of the eighth source
<T9> – the element type of the ninth source
<R> – the combined output type
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that emits items that are the result of combining the items emitted by the source
        Publishers by means of the given aggregation function/**
     * Combines nine source Publishers by emitting an item that aggregates the latest values of each of the
     * source Publishers each time an item is received from any of the source Publishers, where this
     * aggregation is defined by a specified function.
     * <p>
     * If any of the sources never produces an item but only terminates (normally or with an error), the
     * resulting sequence terminates immediately (normally or with all the errors accumulated until that point).
     * If that input source is also synchronous, other sources after it will not be subscribed to.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/combineLatest.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code Publisher} honors backpressure from downstream. The source {@code Publisher}s
     *   are requested in a bounded manner, however, their backpressure is not enforced (the operator won't signal
     *   {@code MissingBackpressureException}) and may lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code combineLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the element type of the first source
     * @param <T2> the element type of the second source
     * @param <T3> the element type of the third source
     * @param <T4> the element type of the fourth source
     * @param <T5> the element type of the fifth source
     * @param <T6> the element type of the sixth source
     * @param <T7> the element type of the seventh source
     * @param <T8> the element type of the eighth source
     * @param <T9> the element type of the ninth source
     * @param <R> the combined output type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            the second source Publisher
     * @param source3
     *            the third source Publisher
     * @param source4
     *            the fourth source Publisher
     * @param source5
     *            the fifth source Publisher
     * @param source6
     *            the sixth source Publisher
     * @param source7
     *            the seventh source Publisher
     * @param source8
     *            the eighth source Publisher
     * @param source9
     *            the ninth source Publisher
     * @param combiner
     *            the aggregation function used to combine the items emitted by the source Publishers
     * @return a Flowable that emits items that are the result of combining the items emitted by the source
     *         Publishers by means of the given aggregation function
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, T7, T8, T9, R> Flowable<R> combineLatest(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            Publisher<? extends T3> source3, Publisher<? extends T4> source4,
            Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Publisher<? extends T7> source7, Publisher<? extends T8> source8,
            Publisher<? extends T9> source9,
            Function9<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? super T7, ? super T8, ? super T9, ? extends R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        ObjectHelper.requireNonNull(source7, "source7 is null");
        ObjectHelper.requireNonNull(source8, "source8 is null");
        ObjectHelper.requireNonNull(source9, "source9 is null");
        return combineLatest(Functions.toFunction(combiner), source1, source2, source3, source4, source5, source6, source7, source8, source9);
    }

    
Concatenates elements of each Publisher provided via an Iterable sequence into a single sequence
of elements without interleaving them.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
concat does not operate by default on a particular Scheduler.

Params:
sources – the Iterable sequence of Publishers
Type parameters:
<T> – the common value type of the sources
Returns:
the new Flowable instance/**
     * Concatenates elements of each Publisher provided via an Iterable sequence into a single sequence
     * of elements without interleaving them.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the common value type of the sources
     * @param sources the Iterable sequence of Publishers
     * @return the new Flowable instance
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concat(Iterable<? extends Publisher<? extends T>> sources) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        // unlike general sources, fromIterable can only throw on a boundary because it is consumed only there
        return fromIterable(sources).concatMapDelayError((Function)Functions.identity(), 2, false);
    }

    
Returns a Flowable that emits the items emitted by each of the Publishers emitted by the source
Publisher, one after the other, without interleaving them.



 
Backpressure:
 
The operator honors backpressure from downstream. Both the outer and inner Publisher sources are expected to honor backpressure as well. If the outer violates this, a MissingBackpressureException is signaled. If any of the inner Publishers violates this, it may throw an IllegalStateException when an inner Publisher completes.
 
Scheduler:
 
concat does not operate by default on a particular Scheduler.

Params:
sources – 
           a Publisher that emits Publishers
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Concat
Returns:
a Flowable that emits items all of the items emitted by the Publishers emitted by Publishers, one after the other, without interleaving them/**
     * Returns a Flowable that emits the items emitted by each of the Publishers emitted by the source
     * Publisher, one after the other, without interleaving them.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}
     *  sources are expected to honor backpressure as well. If the outer violates this, a
     *  {@code MissingBackpressureException} is signaled. If any of the inner {@code Publisher}s violates
     *  this, it <em>may</em> throw an {@code IllegalStateException} when an inner {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            a Publisher that emits Publishers
     * @return a Flowable that emits items all of the items emitted by the Publishers emitted by
     *         {@code Publishers}, one after the other, without interleaving them
     * @see <a href="http://reactivex.io/documentation/operators/concat.html">ReactiveX operators documentation: Concat</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concat(Publisher<? extends Publisher<? extends T>> sources) {
        return concat(sources, bufferSize());
    }

    
Returns a Flowable that emits the items emitted by each of the Publishers emitted by the source
Publisher, one after the other, without interleaving them.



 
Backpressure:
 
The operator honors backpressure from downstream. Both the outer and inner Publisher sources are expected to honor backpressure as well. If the outer violates this, a MissingBackpressureException is signaled. If any of the inner Publishers violates this, it may throw an IllegalStateException when an inner Publisher completes.
 
Scheduler:
 
concat does not operate by default on a particular Scheduler.

Params:
sources – 
           a Publisher that emits Publishers
prefetch – 
           the number of Publishers to prefetch from the sources sequence.
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Concat
Returns:
a Flowable that emits items all of the items emitted by the Publishers emitted by Publishers, one after the other, without interleaving them/**
     * Returns a Flowable that emits the items emitted by each of the Publishers emitted by the source
     * Publisher, one after the other, without interleaving them.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}
     *  sources are expected to honor backpressure as well. If the outer violates this, a
     *  {@code MissingBackpressureException} is signaled. If any of the inner {@code Publisher}s violates
     *  this, it <em>may</em> throw an {@code IllegalStateException} when an inner {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            a Publisher that emits Publishers
     * @param prefetch
     *            the number of Publishers to prefetch from the sources sequence.
     * @return a Flowable that emits items all of the items emitted by the Publishers emitted by
     *         {@code Publishers}, one after the other, without interleaving them
     * @see <a href="http://reactivex.io/documentation/operators/concat.html">ReactiveX operators documentation: Concat</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concat(Publisher<? extends Publisher<? extends T>> sources, int prefetch) {
        return fromPublisher(sources).concatMap((Function)Functions.identity(), prefetch);
    }

    
Returns a Flowable that emits the items emitted by two Publishers, one after the other, without
interleaving them.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
concat does not operate by default on a particular Scheduler.

Params:
source1 – 
           a Publisher to be concatenated
source2 – 
           a Publisher to be concatenated
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Concat
Returns:
a Flowable that emits items emitted by the two source Publishers, one after the other,
        without interleaving them/**
     * Returns a Flowable that emits the items emitted by two Publishers, one after the other, without
     * interleaving them.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be concatenated
     * @param source2
     *            a Publisher to be concatenated
     * @return a Flowable that emits items emitted by the two source Publishers, one after the other,
     *         without interleaving them
     * @see <a href="http://reactivex.io/documentation/operators/concat.html">ReactiveX operators documentation: Concat</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concat(Publisher<? extends T> source1, Publisher<? extends T> source2) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        return concatArray(source1, source2);
    }

    
Returns a Flowable that emits the items emitted by three Publishers, one after the other, without
interleaving them.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
concat does not operate by default on a particular Scheduler.

Params:
source1 – 
           a Publisher to be concatenated
source2 – 
           a Publisher to be concatenated
source3 – 
           a Publisher to be concatenated
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Concat
Returns:
a Flowable that emits items emitted by the three source Publishers, one after the other,
        without interleaving them/**
     * Returns a Flowable that emits the items emitted by three Publishers, one after the other, without
     * interleaving them.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be concatenated
     * @param source2
     *            a Publisher to be concatenated
     * @param source3
     *            a Publisher to be concatenated
     * @return a Flowable that emits items emitted by the three source Publishers, one after the other,
     *         without interleaving them
     * @see <a href="http://reactivex.io/documentation/operators/concat.html">ReactiveX operators documentation: Concat</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concat(
            Publisher<? extends T> source1, Publisher<? extends T> source2,
            Publisher<? extends T> source3) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        return concatArray(source1, source2, source3);
    }

    
Returns a Flowable that emits the items emitted by four Publishers, one after the other, without
interleaving them.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
concat does not operate by default on a particular Scheduler.

Params:
source1 – 
           a Publisher to be concatenated
source2 – 
           a Publisher to be concatenated
source3 – 
           a Publisher to be concatenated
source4 – 
           a Publisher to be concatenated
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Concat
Returns:
a Flowable that emits items emitted by the four source Publishers, one after the other,
        without interleaving them/**
     * Returns a Flowable that emits the items emitted by four Publishers, one after the other, without
     * interleaving them.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be concatenated
     * @param source2
     *            a Publisher to be concatenated
     * @param source3
     *            a Publisher to be concatenated
     * @param source4
     *            a Publisher to be concatenated
     * @return a Flowable that emits items emitted by the four source Publishers, one after the other,
     *         without interleaving them
     * @see <a href="http://reactivex.io/documentation/operators/concat.html">ReactiveX operators documentation: Concat</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concat(
            Publisher<? extends T> source1, Publisher<? extends T> source2,
            Publisher<? extends T> source3, Publisher<? extends T> source4) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        return concatArray(source1, source2, source3, source4);
    }

    
Concatenates a variable number of Publisher sources.

Note: named this way because of overload conflict with concat(Publisher<Publisher>).



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
concatArray does not operate by default on a particular Scheduler.

Params:
sources – the array of sources
Type parameters:
<T> – the common base value type
Throws:
NullPointerException – if sources is null
Returns:
the new Publisher instance/**
     * Concatenates a variable number of Publisher sources.
     * <p>
     * Note: named this way because of overload conflict with concat(Publisher&lt;Publisher&gt;).
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatArray} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param sources the array of sources
     * @param <T> the common base value type
     * @return the new Publisher instance
     * @throws NullPointerException if sources is null
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatArray(Publisher<? extends T>... sources) {
        if (sources.length == 0) {
            return empty();
        } else
        if (sources.length == 1) {
            return fromPublisher(sources[0]);
        }
        return RxJavaPlugins.onAssembly(new FlowableConcatArray<T>(sources, false));
    }

    
Concatenates a variable number of Publisher sources and delays errors from any of them
till all terminate.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
concatArrayDelayError does not operate by default on a particular Scheduler.

Params:
sources – the array of sources
Type parameters:
<T> – the common base value type
Throws:
NullPointerException – if sources is null
Returns:
the new Flowable instance/**
     * Concatenates a variable number of Publisher sources and delays errors from any of them
     * till all terminate.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatArrayDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param sources the array of sources
     * @param <T> the common base value type
     * @return the new Flowable instance
     * @throws NullPointerException if sources is null
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatArrayDelayError(Publisher<? extends T>... sources) {
        if (sources.length == 0) {
            return empty();
        } else
        if (sources.length == 1) {
            return fromPublisher(sources[0]);
        }
        return RxJavaPlugins.onAssembly(new FlowableConcatArray<T>(sources, true));
    }

    
Concatenates an array of Publishers eagerly into a single stream of values.



Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them
in order, each one after the previous one completes.

 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – an array of Publishers that need to be eagerly concatenated
Type parameters:
<T> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Concatenates an array of Publishers eagerly into a single stream of values.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/Flowable.concatArrayEager.png" alt="">
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them
     * in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, the operator will signal a
     *  {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources an array of Publishers that need to be eagerly concatenated
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatArrayEager(Publisher<? extends T>... sources) {
        return concatArrayEager(bufferSize(), bufferSize(), sources);
    }

    
Concatenates an array of Publishers eagerly into a single stream of values.



Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them
in order, each one after the previous one completes.

 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – an array of Publishers that need to be eagerly concatenated
maxConcurrency – the maximum number of concurrent subscriptions at a time, Integer.MAX_VALUE
                      is interpreted as an indication to subscribe to all sources at once
prefetch – the number of elements to prefetch from each Publisher source
Type parameters:
<T> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Concatenates an array of Publishers eagerly into a single stream of values.
     * <p>
     * <img width="640" height="406" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/Flowable.concatArrayEager.nn.png" alt="">
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them
     * in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, the operator will signal a
     *  {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources an array of Publishers that need to be eagerly concatenated
     * @param maxConcurrency the maximum number of concurrent subscriptions at a time, Integer.MAX_VALUE
     *                       is interpreted as an indication to subscribe to all sources at once
     * @param prefetch the number of elements to prefetch from each Publisher source
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @SuppressWarnings({ "rawtypes", "unchecked" })
    public static <T> Flowable<T> concatArrayEager(int maxConcurrency, int prefetch, Publisher<? extends T>... sources) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapEager(new FlowableFromArray(sources), Functions.identity(), maxConcurrency, prefetch, ErrorMode.IMMEDIATE));
    }

    
Concatenates an array of Publishers eagerly into a single stream of values and delaying any errors until all sources terminate. 


 Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the source Publishers. The operator buffers the values emitted by these Publishers and then drains them in order, each one after the previous one completes. 

 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – an array of Publishers that need to be eagerly concatenated
Type parameters:
<T> – the value type
Returns:
the new Flowable instance with the specified concatenation behavior
Since:
2.2.1 - experimental/**
     * Concatenates an array of {@link Publisher}s eagerly into a single stream of values
     * and delaying any errors until all sources terminate.
     * <p>
     * <img width="640" height="358" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/Flowable.concatArrayEagerDelayError.png" alt="">
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source {@code Publisher}s. The operator buffers the values emitted by these {@code Publisher}s
     * and then drains them in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, the operator will signal a
     *  {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources an array of {@code Publisher}s that need to be eagerly concatenated
     * @return the new Flowable instance with the specified concatenation behavior
     * @since 2.2.1 - experimental
     */
    @CheckReturnValue
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T> Flowable<T> concatArrayEagerDelayError(Publisher<? extends T>... sources) {
        return concatArrayEagerDelayError(bufferSize(), bufferSize(), sources);
    }

    
Concatenates an array of Publishers eagerly into a single stream of values and delaying any errors until all sources terminate. 


 Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the source Publishers. The operator buffers the values emitted by these Publishers and then drains them in order, each one after the previous one completes. 

 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – an array of Publishers that need to be eagerly concatenated
maxConcurrency – the maximum number of concurrent subscriptions at a time, Integer.MAX_VALUE
                      is interpreted as indication to subscribe to all sources at once
prefetch – the number of elements to prefetch from each Publisher source
Type parameters:
<T> – the value type
Returns:
the new Flowable instance with the specified concatenation behavior
Since:
2.2.1 - experimental/**
     * Concatenates an array of {@link Publisher}s eagerly into a single stream of values
     * and delaying any errors until all sources terminate.
     * <p>
     * <img width="640" height="359" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/Flowable.concatArrayEagerDelayError.nn.png" alt="">
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source {@code Publisher}s. The operator buffers the values emitted by these {@code Publisher}s
     * and then drains them in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, the operator will signal a
     *  {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources an array of {@code Publisher}s that need to be eagerly concatenated
     * @param maxConcurrency the maximum number of concurrent subscriptions at a time, Integer.MAX_VALUE
     *                       is interpreted as indication to subscribe to all sources at once
     * @param prefetch the number of elements to prefetch from each {@code Publisher} source
     * @return the new Flowable instance with the specified concatenation behavior
     * @since 2.2.1 - experimental
     */
    @SuppressWarnings({ "rawtypes", "unchecked" })
    @CheckReturnValue
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.FULL)
    public static <T> Flowable<T> concatArrayEagerDelayError(int maxConcurrency, int prefetch, Publisher<? extends T>... sources) {
        return fromArray(sources).concatMapEagerDelayError((Function)Functions.identity(), maxConcurrency, prefetch, true);
    }

    
Concatenates the Iterable sequence of Publishers into a single sequence by subscribing to each Publisher,
one after the other, one at a time and delays any errors till the all inner Publishers terminate.

 
Backpressure:
 
The operator honors backpressure from downstream. Both the outer and inner Publisher sources are expected to honor backpressure as well. If the outer violates this, a MissingBackpressureException is signaled. If any of the inner Publishers violates this, it may throw an IllegalStateException when an inner Publisher completes.
 
Scheduler:
 
concatDelayError does not operate by default on a particular Scheduler.

Params:
sources – the Iterable sequence of Publishers
Type parameters:
<T> – the common element base type
Returns:
the new Publisher with the concatenating behavior/**
     * Concatenates the Iterable sequence of Publishers into a single sequence by subscribing to each Publisher,
     * one after the other, one at a time and delays any errors till the all inner Publishers terminate.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}
     *  sources are expected to honor backpressure as well. If the outer violates this, a
     *  {@code MissingBackpressureException} is signaled. If any of the inner {@code Publisher}s violates
     *  this, it <em>may</em> throw an {@code IllegalStateException} when an inner {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources the Iterable sequence of Publishers
     * @return the new Publisher with the concatenating behavior
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatDelayError(Iterable<? extends Publisher<? extends T>> sources) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        return fromIterable(sources).concatMapDelayError((Function)Functions.identity());
    }

    
Concatenates the Publisher sequence of Publishers into a single sequence by subscribing to each inner Publisher,
one after the other, one at a time and delays any errors till the all inner and the outer Publishers terminate.

 
Backpressure:
 
concatDelayError fully supports backpressure.
 
Scheduler:
 
concatDelayError does not operate by default on a particular Scheduler.

Params:
sources – the Publisher sequence of Publishers
Type parameters:
<T> – the common element base type
Returns:
the new Publisher with the concatenating behavior/**
     * Concatenates the Publisher sequence of Publishers into a single sequence by subscribing to each inner Publisher,
     * one after the other, one at a time and delays any errors till the all inner and the outer Publishers terminate.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>{@code concatDelayError} fully supports backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources the Publisher sequence of Publishers
     * @return the new Publisher with the concatenating behavior
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatDelayError(Publisher<? extends Publisher<? extends T>> sources) {
        return concatDelayError(sources, bufferSize(), true);
    }

    
Concatenates the Publisher sequence of Publishers into a single sequence by subscribing to each inner Publisher,
one after the other, one at a time and delays any errors till the all inner and the outer Publishers terminate.

 
Backpressure:
 
concatDelayError fully supports backpressure.
 
Scheduler:
 
concatDelayError does not operate by default on a particular Scheduler.

Params:
sources – the Publisher sequence of Publishers
prefetch – the number of elements to prefetch from the outer Publisher
tillTheEnd – if true exceptions from the outer and all inner Publishers are delayed to the end
                  if false, exception from the outer Publisher is delayed till the current Publisher terminates
Type parameters:
<T> – the common element base type
Returns:
the new Publisher with the concatenating behavior/**
     * Concatenates the Publisher sequence of Publishers into a single sequence by subscribing to each inner Publisher,
     * one after the other, one at a time and delays any errors till the all inner and the outer Publishers terminate.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>{@code concatDelayError} fully supports backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources the Publisher sequence of Publishers
     * @param prefetch the number of elements to prefetch from the outer Publisher
     * @param tillTheEnd if true exceptions from the outer and all inner Publishers are delayed to the end
     *                   if false, exception from the outer Publisher is delayed till the current Publisher terminates
     * @return the new Publisher with the concatenating behavior
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatDelayError(Publisher<? extends Publisher<? extends T>> sources, int prefetch, boolean tillTheEnd) {
        return fromPublisher(sources).concatMapDelayError((Function)Functions.identity(), prefetch, tillTheEnd);
    }

    
Concatenates a Publisher sequence of Publishers eagerly into a single stream of values.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
emitted source Publishers as they are observed. The operator buffers the values emitted by these
Publishers and then drains them in order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream and both the outer and inner Publishers are expected to support backpressure. Violating this assumption, the operator will signal MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – a sequence of Publishers that need to be eagerly concatenated
Type parameters:
<T> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Concatenates a Publisher sequence of Publishers eagerly into a single stream of values.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * emitted source Publishers as they are observed. The operator buffers the values emitted by these
     * Publishers and then drains them in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream and both the outer and inner Publishers are
     *  expected to support backpressure. Violating this assumption, the operator will
     *  signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources a sequence of Publishers that need to be eagerly concatenated
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatEager(Publisher<? extends Publisher<? extends T>> sources) {
        return concatEager(sources, bufferSize(), bufferSize());
    }

    
Concatenates a Publisher sequence of Publishers eagerly into a single stream of values.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
emitted source Publishers as they are observed. The operator buffers the values emitted by these
Publishers and then drains them in order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream and both the outer and inner Publishers are expected to support backpressure. Violating this assumption, the operator will signal MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – a sequence of Publishers that need to be eagerly concatenated
maxConcurrency – the maximum number of concurrently running inner Publishers; Integer.MAX_VALUE
                      is interpreted as all inner Publishers can be active at the same time
prefetch – the number of elements to prefetch from each inner Publisher source
Type parameters:
<T> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Concatenates a Publisher sequence of Publishers eagerly into a single stream of values.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * emitted source Publishers as they are observed. The operator buffers the values emitted by these
     * Publishers and then drains them in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream and both the outer and inner Publishers are
     *  expected to support backpressure. Violating this assumption, the operator will
     *  signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources a sequence of Publishers that need to be eagerly concatenated
     * @param maxConcurrency the maximum number of concurrently running inner Publishers; Integer.MAX_VALUE
     *                       is interpreted as all inner Publishers can be active at the same time
     * @param prefetch the number of elements to prefetch from each inner Publisher source
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @SuppressWarnings({ "rawtypes", "unchecked" })
    public static <T> Flowable<T> concatEager(Publisher<? extends Publisher<? extends T>> sources, int maxConcurrency, int prefetch) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapEagerPublisher(sources, Functions.identity(), maxConcurrency, prefetch, ErrorMode.IMMEDIATE));
    }

    
Concatenates a sequence of Publishers eagerly into a single stream of values.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them
in order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream and the inner Publishers are expected to support backpressure. Violating this assumption, the operator will signal MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – a sequence of Publishers that need to be eagerly concatenated
Type parameters:
<T> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Concatenates a sequence of Publishers eagerly into a single stream of values.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them
     * in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream and the inner Publishers are
     *  expected to support backpressure. Violating this assumption, the operator will
     *  signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources a sequence of Publishers that need to be eagerly concatenated
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> concatEager(Iterable<? extends Publisher<? extends T>> sources) {
        return concatEager(sources, bufferSize(), bufferSize());
    }

    
Concatenates a sequence of Publishers eagerly into a single stream of values.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them
in order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream and both the outer and inner Publishers are expected to support backpressure. Violating this assumption, the operator will signal MissingBackpressureException.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
sources – a sequence of Publishers that need to be eagerly concatenated
maxConcurrency – the maximum number of concurrently running inner Publishers; Integer.MAX_VALUE
                      is interpreted as all inner Publishers can be active at the same time
prefetch – the number of elements to prefetch from each inner Publisher source
Type parameters:
<T> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Concatenates a sequence of Publishers eagerly into a single stream of values.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them
     * in order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream and both the outer and inner Publishers are
     *  expected to support backpressure. Violating this assumption, the operator will
     *  signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type
     * @param sources a sequence of Publishers that need to be eagerly concatenated
     * @param maxConcurrency the maximum number of concurrently running inner Publishers; Integer.MAX_VALUE
     *                       is interpreted as all inner Publishers can be active at the same time
     * @param prefetch the number of elements to prefetch from each inner Publisher source
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @SuppressWarnings({ "rawtypes", "unchecked" })
    public static <T> Flowable<T> concatEager(Iterable<? extends Publisher<? extends T>> sources, int maxConcurrency, int prefetch) {
        ObjectHelper.requireNonNull(sources, "sources is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapEager(new FlowableFromIterable(sources), Functions.identity(), maxConcurrency, prefetch, ErrorMode.IMMEDIATE));
    }

    
Provides an API (via a cold Flowable) that bridges the reactive world with the callback-style,
generally non-backpressured world.

Example:

Flowable.<Event>create(emitter -> {
    Callback listener = new Callback() {
        @Override
        public void onEvent(Event e) {
            emitter.onNext(e);
            if (e.isLast()) {
                emitter.onComplete();
            }
        }
        @Override
        public void onFailure(Exception e) {
            emitter.onError(e);
        }
    };
    AutoCloseable c = api.someMethod(listener);
    emitter.setCancellable(c::close);
}, BackpressureStrategy.BUFFER);


You should call the FlowableEmitter onNext, onError and onComplete methods in a serialized fashion. The
rest of its methods are thread-safe.

 
Backpressure:
 
The backpressure behavior is determined by the mode parameter.
 
Scheduler:
 
create does not operate by default on a particular Scheduler.

Params:
source – the emitter that is called when a Subscriber subscribes to the returned Flowable
mode – the backpressure mode to apply if the downstream Subscriber doesn't request (fast) enough
Type parameters:
<T> – the element type
See Also:
FlowableOnSubscribe
BackpressureStrategy
Cancellable
Returns:
the new Flowable instance/**
     * Provides an API (via a cold Flowable) that bridges the reactive world with the callback-style,
     * generally non-backpressured world.
     * <p>
     * Example:
     * <pre><code>
     * Flowable.&lt;Event&gt;create(emitter -&gt; {
     *     Callback listener = new Callback() {
     *         &#64;Override
     *         public void onEvent(Event e) {
     *             emitter.onNext(e);
     *             if (e.isLast()) {
     *                 emitter.onComplete();
     *             }
     *         }
     *
     *         &#64;Override
     *         public void onFailure(Exception e) {
     *             emitter.onError(e);
     *         }
     *     };
     *
     *     AutoCloseable c = api.someMethod(listener);
     *
     *     emitter.setCancellable(c::close);
     *
     * }, BackpressureStrategy.BUFFER);
     * </code></pre>
     * <p>
     * You should call the FlowableEmitter onNext, onError and onComplete methods in a serialized fashion. The
     * rest of its methods are thread-safe.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The backpressure behavior is determined by the {@code mode} parameter.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code create} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the element type
     * @param source the emitter that is called when a Subscriber subscribes to the returned {@code Flowable}
     * @param mode the backpressure mode to apply if the downstream Subscriber doesn't request (fast) enough
     * @return the new Flowable instance
     * @see FlowableOnSubscribe
     * @see BackpressureStrategy
     * @see Cancellable
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> create(FlowableOnSubscribe<T> source, BackpressureStrategy mode) {
        ObjectHelper.requireNonNull(source, "source is null");
        ObjectHelper.requireNonNull(mode, "mode is null");
        return RxJavaPlugins.onAssembly(new FlowableCreate<T>(source, mode));
    }

    
Returns a Flowable that calls a Publisher factory to create a Publisher for each new Subscriber
that subscribes. That is, for each subscriber, the actual Publisher that subscriber observes is
determined by the factory function.


 The defer Subscriber allows you to defer or delay emitting items from a Publisher until such time as a Subscriber subscribes to the Publisher. This allows a Subscriber to easily obtain updates or a refreshed version of the sequence. 

 
Backpressure:
 
The operator itself doesn't interfere with backpressure which is determined by the Publisher returned by the supplier.
 
Scheduler:
 
defer does not operate by default on a particular Scheduler.

Params:
supplier –  the Publisher factory function to invoke for each Subscriber that subscribes to the resulting Publisher
Type parameters:
<T> – 
           the type of the items emitted by the Publisher
See Also:
ReactiveX operators documentation: Defer
Returns:
a Flowable whose Subscribers' subscriptions trigger an invocation of the given Publisher factory function/**
     * Returns a Flowable that calls a Publisher factory to create a Publisher for each new Subscriber
     * that subscribes. That is, for each subscriber, the actual Publisher that subscriber observes is
     * determined by the factory function.
     * <p>
     * <img width="640" height="340" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/defer.png" alt="">
     * <p>
     * The defer Subscriber allows you to defer or delay emitting items from a Publisher until such time as a
     * Subscriber subscribes to the Publisher. This allows a {@link Subscriber} to easily obtain updates or a
     * refreshed version of the sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator itself doesn't interfere with backpressure which is determined by the {@code Publisher}
     *  returned by the {@code supplier}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code defer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param supplier
     *            the Publisher factory function to invoke for each {@link Subscriber} that subscribes to the
     *            resulting Publisher
     * @param <T>
     *            the type of the items emitted by the Publisher
     * @return a Flowable whose {@link Subscriber}s' subscriptions trigger an invocation of the given
     *         Publisher factory function
     * @see <a href="http://reactivex.io/documentation/operators/defer.html">ReactiveX operators documentation: Defer</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> defer(Callable<? extends Publisher<? extends T>> supplier) {
        ObjectHelper.requireNonNull(supplier, "supplier is null");
        return RxJavaPlugins.onAssembly(new FlowableDefer<T>(supplier));
    }

    
Returns a Flowable that emits no items to the Subscriber and immediately invokes its onComplete method. 



 
Backpressure:
 
This source doesn't produce any elements and effectively ignores downstream backpressure.
 
Scheduler:
 
empty does not operate by default on a particular Scheduler.

Type parameters:
<T> – 
           the type of the items (ostensibly) emitted by the Publisher
See Also:
ReactiveX operators documentation: Empty
Returns:
a Flowable that emits no items to the Subscriber but immediately invokes the Subscriber's onComplete method/**
     * Returns a Flowable that emits no items to the {@link Subscriber} and immediately invokes its
     * {@link Subscriber#onComplete onComplete} method.
     * <p>
     * <img width="640" height="190" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/empty.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This source doesn't produce any elements and effectively ignores downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code empty} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the type of the items (ostensibly) emitted by the Publisher
     * @return a Flowable that emits no items to the {@link Subscriber} but immediately invokes the
     *         {@link Subscriber}'s {@link Subscriber#onComplete() onComplete} method
     * @see <a href="http://reactivex.io/documentation/operators/empty-never-throw.html">ReactiveX operators documentation: Empty</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    @SuppressWarnings("unchecked")
    public static <T> Flowable<T> empty() {
        return RxJavaPlugins.onAssembly((Flowable<T>) FlowableEmpty.INSTANCE);
    }

    
Returns a Flowable that invokes a Subscriber's onError method when the Subscriber subscribes to it. 



 
Backpressure:
 
This source doesn't produce any elements and effectively ignores downstream backpressure.
 
Scheduler:
 
error does not operate by default on a particular Scheduler.

Params:
supplier – 
           a Callable factory to return a Throwable for each individual Subscriber
Type parameters:
<T> – 
           the type of the items (ostensibly) emitted by the Publisher
See Also:
ReactiveX operators documentation: Throw
Returns:
a Flowable that invokes the Subscriber's onError method when the Subscriber subscribes to it/**
     * Returns a Flowable that invokes a {@link Subscriber}'s {@link Subscriber#onError onError} method when the
     * Subscriber subscribes to it.
     * <p>
     * <img width="640" height="190" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/error.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This source doesn't produce any elements and effectively ignores downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code error} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param supplier
     *            a Callable factory to return a Throwable for each individual Subscriber
     * @param <T>
     *            the type of the items (ostensibly) emitted by the Publisher
     * @return a Flowable that invokes the {@link Subscriber}'s {@link Subscriber#onError onError} method when
     *         the Subscriber subscribes to it
     * @see <a href="http://reactivex.io/documentation/operators/empty-never-throw.html">ReactiveX operators documentation: Throw</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> error(Callable<? extends Throwable> supplier) {
        ObjectHelper.requireNonNull(supplier, "supplier is null");
        return RxJavaPlugins.onAssembly(new FlowableError<T>(supplier));
    }

    
Returns a Flowable that invokes a Subscriber's onError method when the Subscriber subscribes to it. 



 
Backpressure:
 
This source doesn't produce any elements and effectively ignores downstream backpressure.
 
Scheduler:
 
error does not operate by default on a particular Scheduler.

Params:
throwable –  the particular Throwable to pass to onError
Type parameters:
<T> – 
           the type of the items (ostensibly) emitted by the Publisher
See Also:
ReactiveX operators documentation: Throw
Returns:
a Flowable that invokes the Subscriber's onError method when the Subscriber subscribes to it/**
     * Returns a Flowable that invokes a {@link Subscriber}'s {@link Subscriber#onError onError} method when the
     * Subscriber subscribes to it.
     * <p>
     * <img width="640" height="190" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/error.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This source doesn't produce any elements and effectively ignores downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code error} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param throwable
     *            the particular Throwable to pass to {@link Subscriber#onError onError}
     * @param <T>
     *            the type of the items (ostensibly) emitted by the Publisher
     * @return a Flowable that invokes the {@link Subscriber}'s {@link Subscriber#onError onError} method when
     *         the Subscriber subscribes to it
     * @see <a href="http://reactivex.io/documentation/operators/empty-never-throw.html">ReactiveX operators documentation: Throw</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> error(final Throwable throwable) {
        ObjectHelper.requireNonNull(throwable, "throwable is null");
        return error(Functions.justCallable(throwable));
    }

    
Converts an Array into a Publisher that emits the items in the Array.



 
Backpressure:
 
The operator honors backpressure from downstream and iterates the given array on demand (i.e., when requested).
 
Scheduler:
 
fromArray does not operate by default on a particular Scheduler.

Params:
items – 
           the array of elements
Type parameters:
<T> – 
           the type of items in the Array and the type of items to be emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: From
Returns:
a Flowable that emits each item in the source Array/**
     * Converts an Array into a Publisher that emits the items in the Array.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/from.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and iterates the given {@code array}
     *  on demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code fromArray} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param items
     *            the array of elements
     * @param <T>
     *            the type of items in the Array and the type of items to be emitted by the resulting Publisher
     * @return a Flowable that emits each item in the source Array
     * @see <a href="http://reactivex.io/documentation/operators/from.html">ReactiveX operators documentation: From</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> fromArray(T... items) {
        ObjectHelper.requireNonNull(items, "items is null");
        if (items.length == 0) {
            return empty();
        }
        if (items.length == 1) {
            return just(items[0]);
        }
        return RxJavaPlugins.onAssembly(new FlowableFromArray<T>(items));
    }

    
Returns a Flowable that, when a Subscriber subscribes to it, invokes a function you specify and then
emits the value returned from that function.



This allows you to defer the execution of the function you specify until a Subscriber subscribes to the
Publisher. That is to say, it makes the function "lazy."

  
Backpressure:
  
The operator honors backpressure from downstream.
  
Scheduler:
  
fromCallable does not operate by default on a particular Scheduler.
  
Error handling:
  
 If the Callable throws an exception, the respective Throwable is delivered to the downstream via Subscriber.onError(Throwable), except when the downstream has canceled this Flowable source. In this latter case, the Throwable is delivered to the global error handler via RxJavaPlugins.onError(Throwable) as an UndeliverableException. 

Params:
supplier –  a function, the execution of which should be deferred; fromCallable will invoke this function only when a Subscriber subscribes to the Publisher that fromCallable returns
Type parameters:
<T> – 
        the type of the item emitted by the Publisher
See Also:
defer(Callable<? extends Publisher<? extends Object>>)
Returns:
a Flowable whose Subscribers' subscriptions trigger an invocation of the given function
Since:
2.0/**
     * Returns a Flowable that, when a Subscriber subscribes to it, invokes a function you specify and then
     * emits the value returned from that function.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/fromCallable.png" alt="">
     * <p>
     * This allows you to defer the execution of the function you specify until a Subscriber subscribes to the
     * Publisher. That is to say, it makes the function "lazy."
     * <dl>
     *   <dt><b>Backpressure:</b></dt>
     *   <dd>The operator honors backpressure from downstream.</dd>
     *   <dt><b>Scheduler:</b></dt>
     *   <dd>{@code fromCallable} does not operate by default on a particular {@link Scheduler}.</dd>
     *   <dt><b>Error handling:</b></dt>
     *   <dd> If the {@link Callable} throws an exception, the respective {@link Throwable} is
     *   delivered to the downstream via {@link Subscriber#onError(Throwable)},
     *   except when the downstream has canceled this {@code Flowable} source.
     *   In this latter case, the {@code Throwable} is delivered to the global error handler via
     *   {@link RxJavaPlugins#onError(Throwable)} as an {@link io.reactivex.exceptions.UndeliverableException UndeliverableException}.
     *   </dd>
     * </dl>
     *
     * @param supplier
     *         a function, the execution of which should be deferred; {@code fromCallable} will invoke this
     *         function only when a Subscriber subscribes to the Publisher that {@code fromCallable} returns
     * @param <T>
     *         the type of the item emitted by the Publisher
     * @return a Flowable whose {@link Subscriber}s' subscriptions trigger an invocation of the given function
     * @see #defer(Callable)
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> fromCallable(Callable<? extends T> supplier) {
        ObjectHelper.requireNonNull(supplier, "supplier is null");
        return RxJavaPlugins.onAssembly(new FlowableFromCallable<T>(supplier));
    }

    
Converts a Future into a Publisher. 


 You can convert any object that supports the Future interface into a Publisher that emits the return value of the Future.get method of that object by passing the object into the from method. 

Important note: This Publisher is blocking on the thread it gets subscribed on; you cannot cancel it.
 Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the cancellation effect: futurePublisher.doOnCancel(() -> future.cancel(true));. 

 
Backpressure:
 
The operator honors backpressure from downstream.
 
Scheduler:
 
fromFuture does not operate by default on a particular Scheduler.

Params:
future –  the source Future
Type parameters:
<T> –  the type of object that the Future returns, and also the type of item to be emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: From
Returns:
a Flowable that emits the item from the source Future/**
     * Converts a {@link Future} into a Publisher.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/from.Future.png" alt="">
     * <p>
     * You can convert any object that supports the {@link Future} interface into a Publisher that emits the
     * return value of the {@link Future#get} method of that object by passing the object into the {@code from}
     * method.
     * <p>
     * <em>Important note:</em> This Publisher is blocking on the thread it gets subscribed on; you cannot cancel it.
     * <p>
     * Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the
     * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code fromFuture} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param future
     *            the source {@link Future}
     * @param <T>
     *            the type of object that the {@link Future} returns, and also the type of item to be emitted by
     *            the resulting Publisher
     * @return a Flowable that emits the item from the source {@link Future}
     * @see <a href="http://reactivex.io/documentation/operators/from.html">ReactiveX operators documentation: From</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> fromFuture(Future<? extends T> future) {
        ObjectHelper.requireNonNull(future, "future is null");
        return RxJavaPlugins.onAssembly(new FlowableFromFuture<T>(future, 0L, null));
    }

    
Converts a Future into a Publisher, with a timeout on the Future. 


 You can convert any object that supports the Future interface into a Publisher that emits the return value of the Future.get method of that object by passing the object into the fromFuture method. 
 Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the cancellation effect: futurePublisher.doOnCancel(() -> future.cancel(true));. 

Important note: This Publisher is blocking on the thread it gets subscribed on; you cannot cancel it.

 
Backpressure:
 
The operator honors backpressure from downstream.
 
Scheduler:
 
fromFuture does not operate by default on a particular Scheduler.

Params:
future –  the source Future
timeout –  the maximum time to wait before calling get
unit –  the TimeUnit of the timeout argument
Type parameters:
<T> –  the type of object that the Future returns, and also the type of item to be emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: From
Returns:
a Flowable that emits the item from the source Future/**
     * Converts a {@link Future} into a Publisher, with a timeout on the Future.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/from.Future.png" alt="">
     * <p>
     * You can convert any object that supports the {@link Future} interface into a Publisher that emits the
     * return value of the {@link Future#get} method of that object by passing the object into the {@code fromFuture}
     * method.
     * <p>
     * Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the
     * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}.
     * <p>
     * <em>Important note:</em> This Publisher is blocking on the thread it gets subscribed on; you cannot cancel it.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code fromFuture} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param future
     *            the source {@link Future}
     * @param timeout
     *            the maximum time to wait before calling {@code get}
     * @param unit
     *            the {@link TimeUnit} of the {@code timeout} argument
     * @param <T>
     *            the type of object that the {@link Future} returns, and also the type of item to be emitted by
     *            the resulting Publisher
     * @return a Flowable that emits the item from the source {@link Future}
     * @see <a href="http://reactivex.io/documentation/operators/from.html">ReactiveX operators documentation: From</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> fromFuture(Future<? extends T> future, long timeout, TimeUnit unit) {
        ObjectHelper.requireNonNull(future, "future is null");
        ObjectHelper.requireNonNull(unit, "unit is null");
        return RxJavaPlugins.onAssembly(new FlowableFromFuture<T>(future, timeout, unit));
    }

    
Converts a Future into a Publisher, with a timeout on the Future. 


 You can convert any object that supports the Future interface into a Publisher that emits the return value of the Future.get method of that object by passing the object into the from method. 
 Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the cancellation effect: futurePublisher.doOnCancel(() -> future.cancel(true));. 

Important note: This Publisher is blocking; you cannot cancel it.

 
Backpressure:
 
The operator honors backpressure from downstream.
 
Scheduler:
 
fromFuture does not operate by default on a particular Scheduler.

Params:
future –  the source Future
timeout –  the maximum time to wait before calling get
unit –  the TimeUnit of the timeout argument
scheduler –  the Scheduler to wait for the Future on. Use a Scheduler such as Schedulers.io() that can block and wait on the Future
Type parameters:
<T> –  the type of object that the Future returns, and also the type of item to be emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: From
Returns:
a Flowable that emits the item from the source Future/**
     * Converts a {@link Future} into a Publisher, with a timeout on the Future.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/from.Future.png" alt="">
     * <p>
     * You can convert any object that supports the {@link Future} interface into a Publisher that emits the
     * return value of the {@link Future#get} method of that object by passing the object into the {@code from}
     * method.
     * <p>
     * Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the
     * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}.
     * <p>
     * <em>Important note:</em> This Publisher is blocking; you cannot cancel it.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code fromFuture} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param future
     *            the source {@link Future}
     * @param timeout
     *            the maximum time to wait before calling {@code get}
     * @param unit
     *            the {@link TimeUnit} of the {@code timeout} argument
     * @param scheduler
     *            the {@link Scheduler} to wait for the Future on. Use a Scheduler such as
     *            {@link Schedulers#io()} that can block and wait on the Future
     * @param <T>
     *            the type of object that the {@link Future} returns, and also the type of item to be emitted by
     *            the resulting Publisher
     * @return a Flowable that emits the item from the source {@link Future}
     * @see <a href="http://reactivex.io/documentation/operators/from.html">ReactiveX operators documentation: From</a>
     */
    @SuppressWarnings({ "unchecked", "cast" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public static <T> Flowable<T> fromFuture(Future<? extends T> future, long timeout, TimeUnit unit, Scheduler scheduler) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return fromFuture((Future<T>)future, timeout, unit).subscribeOn(scheduler);
    }

    
Converts a Future, operating on a specified Scheduler, into a Publisher. 


 You can convert any object that supports the Future interface into a Publisher that emits the return value of the Future.get method of that object by passing the object into the from method. 
 Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the cancellation effect: futurePublisher.doOnCancel(() -> future.cancel(true));. 

 
Backpressure:
 
The operator honors backpressure from downstream.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
future –  the source Future
scheduler –  the Scheduler to wait for the Future on. Use a Scheduler such as Schedulers.io() that can block and wait on the Future
Type parameters:
<T> –  the type of object that the Future returns, and also the type of item to be emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: From
Returns:
a Flowable that emits the item from the source Future/**
     * Converts a {@link Future}, operating on a specified {@link Scheduler}, into a Publisher.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/from.Future.s.png" alt="">
     * <p>
     * You can convert any object that supports the {@link Future} interface into a Publisher that emits the
     * return value of the {@link Future#get} method of that object by passing the object into the {@code from}
     * method.
     * <p>
     * Unlike 1.x, canceling the Flowable won't cancel the future. If necessary, one can use composition to achieve the
     * cancellation effect: {@code futurePublisher.doOnCancel(() -> future.cancel(true));}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param future
     *            the source {@link Future}
     * @param scheduler
     *            the {@link Scheduler} to wait for the Future on. Use a Scheduler such as
     *            {@link Schedulers#io()} that can block and wait on the Future
     * @param <T>
     *            the type of object that the {@link Future} returns, and also the type of item to be emitted by
     *            the resulting Publisher
     * @return a Flowable that emits the item from the source {@link Future}
     * @see <a href="http://reactivex.io/documentation/operators/from.html">ReactiveX operators documentation: From</a>
     */
    @SuppressWarnings({ "cast", "unchecked" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public static <T> Flowable<T> fromFuture(Future<? extends T> future, Scheduler scheduler) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return fromFuture((Future<T>)future).subscribeOn(scheduler);
    }

    
Converts an Iterable sequence into a Publisher that emits the items in the sequence. 



 
Backpressure:
 
The operator honors backpressure from downstream and iterates the given iterable on demand (i.e., when requested).
 
Scheduler:
 
fromIterable does not operate by default on a particular Scheduler.

Params:
source –  the source Iterable sequence
Type parameters:
<T> –  the type of items in the Iterable sequence and the type of items to be emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: From
Returns:
a Flowable that emits each item in the source Iterable sequence/**
     * Converts an {@link Iterable} sequence into a Publisher that emits the items in the sequence.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/from.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and iterates the given {@code iterable}
     *  on demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code fromIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param source
     *            the source {@link Iterable} sequence
     * @param <T>
     *            the type of items in the {@link Iterable} sequence and the type of items to be emitted by the
     *            resulting Publisher
     * @return a Flowable that emits each item in the source {@link Iterable} sequence
     * @see <a href="http://reactivex.io/documentation/operators/from.html">ReactiveX operators documentation: From</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> fromIterable(Iterable<? extends T> source) {
        ObjectHelper.requireNonNull(source, "source is null");
        return RxJavaPlugins.onAssembly(new FlowableFromIterable<T>(source));
    }

    
Converts an arbitrary Reactive Streams Publisher into a Flowable if not already a
Flowable.
 The Publisher must follow the Reactive Streams specification.
Violating the specification may result in undefined behavior.
 If possible, use create(FlowableOnSubscribe<Object>, BackpressureStrategy) to create a source-like Flowable instead. 
 Note that even though Publisher appears to be a functional interface, it is not recommended to implement it through a lambda as the specification requires state management that is not achievable with a stateless lambda. 

 
Backpressure:
 
The operator is a pass-through for backpressure and its behavior is determined by the
 backpressure behavior of the wrapped publisher.
 
Scheduler:
 
fromPublisher does not operate by default on a particular Scheduler.

Params:
source – the Publisher to convert
Type parameters:
<T> – the value type of the flow
Throws:
NullPointerException – if the source Publisher is null
See Also:
create(FlowableOnSubscribe<Object>, BackpressureStrategy)
Returns:
the new Flowable instance/**
     * Converts an arbitrary Reactive Streams Publisher into a Flowable if not already a
     * Flowable.
     * <p>
     * The {@link Publisher} must follow the
     * <a href="https://github.com/reactive-streams/reactive-streams-jvm#reactive-streams">Reactive Streams specification</a>.
     * Violating the specification may result in undefined behavior.
     * <p>
     * If possible, use {@link #create(FlowableOnSubscribe, BackpressureStrategy)} to create a
     * source-like {@code Flowable} instead.
     * <p>
     * Note that even though {@link Publisher} appears to be a functional interface, it
     * is not recommended to implement it through a lambda as the specification requires
     * state management that is not achievable with a stateless lambda.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator is a pass-through for backpressure and its behavior is determined by the
     *  backpressure behavior of the wrapped publisher.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code fromPublisher} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type of the flow
     * @param source the Publisher to convert
     * @return the new Flowable instance
     * @throws NullPointerException if the {@code source} {@code Publisher} is null
     * @see #create(FlowableOnSubscribe, BackpressureStrategy)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    @SuppressWarnings("unchecked")
    public static <T> Flowable<T> fromPublisher(final Publisher<? extends T> source) {
        if (source instanceof Flowable) {
            return RxJavaPlugins.onAssembly((Flowable<T>)source);
        }
        ObjectHelper.requireNonNull(source, "source is null");

        return RxJavaPlugins.onAssembly(new FlowableFromPublisher<T>(source));
    }

    
Returns a cold, synchronous, stateless and backpressure-aware generator of values.
 Note that the Emitter.onNext, Emitter.onError and Emitter.onComplete methods provided to the function via the Emitter instance should be called synchronously, never concurrently and only while the function body is executing. Calling them from multiple threads or outside the function call is not supported and leads to an undefined behavior. 

 
Backpressure:
 
The operator honors downstream backpressure.
 
Scheduler:
 
generate does not operate by default on a particular Scheduler.

Params:
generator – the Consumer called whenever a particular downstream Subscriber has requested a value. The callback then should call onNext, onError or onComplete to signal a value or a terminal event. Signaling multiple onNext in a call will make the operator signal IllegalStateException.
Type parameters:
<T> – the generated value type
Returns:
the new Flowable instance/**
     * Returns a cold, synchronous, stateless and backpressure-aware generator of values.
     * <p>
     * Note that the {@link Emitter#onNext}, {@link Emitter#onError} and
     * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously,
     * never concurrently and only while the function body is executing. Calling them from multiple threads
     * or outside the function call is not supported and leads to an undefined behavior.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code generate} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the generated value type
     * @param generator the Consumer called whenever a particular downstream Subscriber has
     * requested a value. The callback then should call {@code onNext}, {@code onError} or
     * {@code onComplete} to signal a value or a terminal event. Signaling multiple {@code onNext}
     * in a call will make the operator signal {@code IllegalStateException}.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> generate(final Consumer<Emitter<T>> generator) {
        ObjectHelper.requireNonNull(generator, "generator is null");
        return generate(Functions.nullSupplier(),
                FlowableInternalHelper.<T, Object>simpleGenerator(generator),
                Functions.emptyConsumer());
    }

    
Returns a cold, synchronous, stateful and backpressure-aware generator of values.
 Note that the Emitter.onNext, Emitter.onError and Emitter.onComplete methods provided to the function via the Emitter instance should be called synchronously, never concurrently and only while the function body is executing. Calling them from multiple threads or outside the function call is not supported and leads to an undefined behavior. 

 
Backpressure:
 
The operator honors downstream backpressure.
 
Scheduler:
 
generate does not operate by default on a particular Scheduler.

Params:
initialState – the Callable to generate the initial state for each Subscriber
generator – the Consumer called with the current state whenever a particular downstream Subscriber has requested a value. The callback then should call onNext, onError or onComplete to signal a value or a terminal event. Signaling multiple onNext in a call will make the operator signal IllegalStateException.
Type parameters:
<S> – the type of the per-Subscriber state
<T> – the generated value type
Returns:
the new Flowable instance/**
     * Returns a cold, synchronous, stateful and backpressure-aware generator of values.
     * <p>
     * Note that the {@link Emitter#onNext}, {@link Emitter#onError} and
     * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously,
     * never concurrently and only while the function body is executing. Calling them from multiple threads
     * or outside the function call is not supported and leads to an undefined behavior.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code generate} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <S> the type of the per-Subscriber state
     * @param <T> the generated value type
     * @param initialState the Callable to generate the initial state for each Subscriber
     * @param generator the Consumer called with the current state whenever a particular downstream Subscriber has
     * requested a value. The callback then should call {@code onNext}, {@code onError} or
     * {@code onComplete} to signal a value or a terminal event. Signaling multiple {@code onNext}
     * in a call will make the operator signal {@code IllegalStateException}.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, S> Flowable<T> generate(Callable<S> initialState, final BiConsumer<S, Emitter<T>> generator) {
        ObjectHelper.requireNonNull(generator, "generator is null");
        return generate(initialState, FlowableInternalHelper.<T, S>simpleBiGenerator(generator),
                Functions.emptyConsumer());
    }

    
Returns a cold, synchronous, stateful and backpressure-aware generator of values.
 Note that the Emitter.onNext, Emitter.onError and Emitter.onComplete methods provided to the function via the Emitter instance should be called synchronously, never concurrently and only while the function body is executing. Calling them from multiple threads or outside the function call is not supported and leads to an undefined behavior. 

 
Backpressure:
 
The operator honors downstream backpressure.
 
Scheduler:
 
generate does not operate by default on a particular Scheduler.

Params:
initialState – the Callable to generate the initial state for each Subscriber
generator – the Consumer called with the current state whenever a particular downstream Subscriber has requested a value. The callback then should call onNext, onError or onComplete to signal a value or a terminal event. Signaling multiple onNext in a call will make the operator signal IllegalStateException.
disposeState – the Consumer that is called with the current state when the generator
terminates the sequence or it gets canceled
Type parameters:
<S> – the type of the per-Subscriber state
<T> – the generated value type
Returns:
the new Flowable instance/**
     * Returns a cold, synchronous, stateful and backpressure-aware generator of values.
     * <p>
     * Note that the {@link Emitter#onNext}, {@link Emitter#onError} and
     * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously,
     * never concurrently and only while the function body is executing. Calling them from multiple threads
     * or outside the function call is not supported and leads to an undefined behavior.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code generate} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <S> the type of the per-Subscriber state
     * @param <T> the generated value type
     * @param initialState the Callable to generate the initial state for each Subscriber
     * @param generator the Consumer called with the current state whenever a particular downstream Subscriber has
     * requested a value. The callback then should call {@code onNext}, {@code onError} or
     * {@code onComplete} to signal a value or a terminal event. Signaling multiple {@code onNext}
     * in a call will make the operator signal {@code IllegalStateException}.
     * @param disposeState the Consumer that is called with the current state when the generator
     * terminates the sequence or it gets canceled
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, S> Flowable<T> generate(Callable<S> initialState, final BiConsumer<S, Emitter<T>> generator,
            Consumer<? super S> disposeState) {
        ObjectHelper.requireNonNull(generator, "generator is null");
        return generate(initialState, FlowableInternalHelper.<T, S>simpleBiGenerator(generator), disposeState);
    }

    
Returns a cold, synchronous, stateful and backpressure-aware generator of values.
 Note that the Emitter.onNext, Emitter.onError and Emitter.onComplete methods provided to the function via the Emitter instance should be called synchronously, never concurrently and only while the function body is executing. Calling them from multiple threads or outside the function call is not supported and leads to an undefined behavior. 

 
Backpressure:
 
The operator honors downstream backpressure.
 
Scheduler:
 
generate does not operate by default on a particular Scheduler.

Params:
initialState – the Callable to generate the initial state for each Subscriber
generator – the Function called with the current state whenever a particular downstream Subscriber has requested a value. The callback then should call onNext, onError or onComplete to signal a value or a terminal event and should return a (new) state for the next invocation. Signaling multiple onNext in a call will make the operator signal IllegalStateException.
Type parameters:
<S> – the type of the per-Subscriber state
<T> – the generated value type
Returns:
the new Flowable instance/**
     * Returns a cold, synchronous, stateful and backpressure-aware generator of values.
     * <p>
     * Note that the {@link Emitter#onNext}, {@link Emitter#onError} and
     * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously,
     * never concurrently and only while the function body is executing. Calling them from multiple threads
     * or outside the function call is not supported and leads to an undefined behavior.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code generate} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <S> the type of the per-Subscriber state
     * @param <T> the generated value type
     * @param initialState the Callable to generate the initial state for each Subscriber
     * @param generator the Function called with the current state whenever a particular downstream Subscriber has
     * requested a value. The callback then should call {@code onNext}, {@code onError} or
     * {@code onComplete} to signal a value or a terminal event and should return a (new) state for
     * the next invocation. Signaling multiple {@code onNext}
     * in a call will make the operator signal {@code IllegalStateException}.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, S> Flowable<T> generate(Callable<S> initialState, BiFunction<S, Emitter<T>, S> generator) {
        return generate(initialState, generator, Functions.emptyConsumer());
    }

    
Returns a cold, synchronous, stateful and backpressure-aware generator of values.
 Note that the Emitter.onNext, Emitter.onError and Emitter.onComplete methods provided to the function via the Emitter instance should be called synchronously, never concurrently and only while the function body is executing. Calling them from multiple threads or outside the function call is not supported and leads to an undefined behavior. 

 
Backpressure:
 
The operator honors downstream backpressure.
 
Scheduler:
 
generate does not operate by default on a particular Scheduler.

Params:
initialState – the Callable to generate the initial state for each Subscriber
generator – the Function called with the current state whenever a particular downstream Subscriber has requested a value. The callback then should call onNext, onError or onComplete to signal a value or a terminal event and should return a (new) state for the next invocation. Signaling multiple onNext in a call will make the operator signal IllegalStateException.
disposeState – the Consumer that is called with the current state when the generator
terminates the sequence or it gets canceled
Type parameters:
<S> – the type of the per-Subscriber state
<T> – the generated value type
Returns:
the new Flowable instance/**
     * Returns a cold, synchronous, stateful and backpressure-aware generator of values.
     * <p>
     * Note that the {@link Emitter#onNext}, {@link Emitter#onError} and
     * {@link Emitter#onComplete} methods provided to the function via the {@link Emitter} instance should be called synchronously,
     * never concurrently and only while the function body is executing. Calling them from multiple threads
     * or outside the function call is not supported and leads to an undefined behavior.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code generate} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <S> the type of the per-Subscriber state
     * @param <T> the generated value type
     * @param initialState the Callable to generate the initial state for each Subscriber
     * @param generator the Function called with the current state whenever a particular downstream Subscriber has
     * requested a value. The callback then should call {@code onNext}, {@code onError} or
     * {@code onComplete} to signal a value or a terminal event and should return a (new) state for
     * the next invocation. Signaling multiple {@code onNext}
     * in a call will make the operator signal {@code IllegalStateException}.
     * @param disposeState the Consumer that is called with the current state when the generator
     * terminates the sequence or it gets canceled
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, S> Flowable<T> generate(Callable<S> initialState, BiFunction<S, Emitter<T>, S> generator, Consumer<? super S> disposeState) {
        ObjectHelper.requireNonNull(initialState, "initialState is null");
        ObjectHelper.requireNonNull(generator, "generator is null");
        ObjectHelper.requireNonNull(disposeState, "disposeState is null");
        return RxJavaPlugins.onAssembly(new FlowableGenerate<T, S>(initialState, generator, disposeState));
    }

    
Returns a Flowable that emits a 0L after the initialDelay and ever-increasing numbers after each period of time thereafter. 



 
Backpressure:
 
The operator generates values based on time and ignores downstream backpressure which may lead to MissingBackpressureException at some point in the chain. Consumers should consider applying one of the onBackpressureXXX operators as well.
 
Scheduler:
 
interval operates by default on the computation Scheduler.

Params:
initialDelay – 
           the initial delay time to wait before emitting the first value of 0L
period – 
           the period of time between emissions of the subsequent numbers
unit –  the time unit for both initialDelay and period
See Also:
ReactiveX operators documentation: Interval
Returns:
a Flowable that emits a 0L after the initialDelay and ever-increasing numbers after each period of time thereafter
Since:
1.0.12/**
     * Returns a Flowable that emits a {@code 0L} after the {@code initialDelay} and ever-increasing numbers
     * after each {@code period} of time thereafter.
     * <p>
     * <img width="640" height="200" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timer.p.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator generates values based on time and ignores downstream backpressure which
     *  may lead to {@code MissingBackpressureException} at some point in the chain.
     *  Consumers should consider applying one of the {@code onBackpressureXXX} operators as well.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code interval} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param initialDelay
     *            the initial delay time to wait before emitting the first value of 0L
     * @param period
     *            the period of time between emissions of the subsequent numbers
     * @param unit
     *            the time unit for both {@code initialDelay} and {@code period}
     * @return a Flowable that emits a 0L after the {@code initialDelay} and ever-increasing numbers after
     *         each {@code period} of time thereafter
     * @see <a href="http://reactivex.io/documentation/operators/interval.html">ReactiveX operators documentation: Interval</a>
     * @since 1.0.12
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public static Flowable<Long> interval(long initialDelay, long period, TimeUnit unit) {
        return interval(initialDelay, period, unit, Schedulers.computation());
    }

    
Returns a Flowable that emits a 0L after the initialDelay and ever-increasing numbers after each period of time thereafter, on a specified Scheduler. 



 
Backpressure:
 
The operator generates values based on time and ignores downstream backpressure which may lead to MissingBackpressureException at some point in the chain. Consumers should consider applying one of the onBackpressureXXX operators as well.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
initialDelay – 
           the initial delay time to wait before emitting the first value of 0L
period – 
           the period of time between emissions of the subsequent numbers
unit –  the time unit for both initialDelay and period
scheduler – 
           the Scheduler on which the waiting happens and items are emitted
See Also:
ReactiveX operators documentation: Interval
Returns:
a Flowable that emits a 0L after the initialDelay and ever-increasing numbers after each period of time thereafter, while running on the given Scheduler
Since:
1.0.12/**
     * Returns a Flowable that emits a {@code 0L} after the {@code initialDelay} and ever-increasing numbers
     * after each {@code period} of time thereafter, on a specified {@link Scheduler}.
     * <p>
     * <img width="640" height="200" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timer.ps.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator generates values based on time and ignores downstream backpressure which
     *  may lead to {@code MissingBackpressureException} at some point in the chain.
     *  Consumers should consider applying one of the {@code onBackpressureXXX} operators as well.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param initialDelay
     *            the initial delay time to wait before emitting the first value of 0L
     * @param period
     *            the period of time between emissions of the subsequent numbers
     * @param unit
     *            the time unit for both {@code initialDelay} and {@code period}
     * @param scheduler
     *            the Scheduler on which the waiting happens and items are emitted
     * @return a Flowable that emits a 0L after the {@code initialDelay} and ever-increasing numbers after
     *         each {@code period} of time thereafter, while running on the given Scheduler
     * @see <a href="http://reactivex.io/documentation/operators/interval.html">ReactiveX operators documentation: Interval</a>
     * @since 1.0.12
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public static Flowable<Long> interval(long initialDelay, long period, TimeUnit unit, Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableInterval(Math.max(0L, initialDelay), Math.max(0L, period), unit, scheduler));
    }

    
Returns a Flowable that emits a sequential number every specified interval of time.



 
Backpressure:
 
The operator signals a MissingBackpressureException if the downstream is not ready to receive the next value.
 
Scheduler:
 
interval operates by default on the computation Scheduler.

Params:
period – 
           the period size in time units (see below)
unit – 
           time units to use for the interval size
See Also:
ReactiveX operators documentation: Interval
Returns:
a Flowable that emits a sequential number each time interval/**
     * Returns a Flowable that emits a sequential number every specified interval of time.
     * <p>
     * <img width="640" height="195" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/interval.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator signals a {@code MissingBackpressureException} if the downstream
     *  is not ready to receive the next value.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code interval} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param period
     *            the period size in time units (see below)
     * @param unit
     *            time units to use for the interval size
     * @return a Flowable that emits a sequential number each time interval
     * @see <a href="http://reactivex.io/documentation/operators/interval.html">ReactiveX operators documentation: Interval</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public static Flowable<Long> interval(long period, TimeUnit unit) {
        return interval(period, period, unit, Schedulers.computation());
    }

    
Returns a Flowable that emits a sequential number every specified interval of time, on a
specified Scheduler.



 
Backpressure:
 
The operator generates values based on time and ignores downstream backpressure which may lead to MissingBackpressureException at some point in the chain. Consumers should consider applying one of the onBackpressureXXX operators as well.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
period – 
           the period size in time units (see below)
unit – 
           time units to use for the interval size
scheduler – 
           the Scheduler to use for scheduling the items
See Also:
ReactiveX operators documentation: Interval
Returns:
a Flowable that emits a sequential number each time interval/**
     * Returns a Flowable that emits a sequential number every specified interval of time, on a
     * specified Scheduler.
     * <p>
     * <img width="640" height="200" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/interval.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator generates values based on time and ignores downstream backpressure which
     *  may lead to {@code MissingBackpressureException} at some point in the chain.
     *  Consumers should consider applying one of the {@code onBackpressureXXX} operators as well.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param period
     *            the period size in time units (see below)
     * @param unit
     *            time units to use for the interval size
     * @param scheduler
     *            the Scheduler to use for scheduling the items
     * @return a Flowable that emits a sequential number each time interval
     * @see <a href="http://reactivex.io/documentation/operators/interval.html">ReactiveX operators documentation: Interval</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public static Flowable<Long> interval(long period, TimeUnit unit, Scheduler scheduler) {
        return interval(period, period, unit, scheduler);
    }

    
Signals a range of long values, the first after some initial delay and the rest periodically after.

The sequence completes immediately after the last value (start + count - 1) has been reached.

 
Backpressure:
 
The operator signals a MissingBackpressureException if the downstream can't keep up.
 
Scheduler:
 
intervalRange by default operates on the computation Scheduler.

Params:
start – that start value of the range
count – the number of values to emit in total, if zero, the operator emits an onComplete after the initial delay.
initialDelay – the initial delay before signaling the first value (the start)
period – the period between subsequent values
unit – the unit of measure of the initialDelay and period amounts
Returns:
the new Flowable instance/**
     * Signals a range of long values, the first after some initial delay and the rest periodically after.
     * <p>
     * The sequence completes immediately after the last value (start + count - 1) has been reached.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator signals a {@code MissingBackpressureException} if the downstream can't keep up.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code intervalRange} by default operates on the {@link Schedulers#computation() computation} {@link Scheduler}.</dd>
     * </dl>
     * @param start that start value of the range
     * @param count the number of values to emit in total, if zero, the operator emits an onComplete after the initial delay.
     * @param initialDelay the initial delay before signaling the first value (the start)
     * @param period the period between subsequent values
     * @param unit the unit of measure of the initialDelay and period amounts
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public static Flowable<Long> intervalRange(long start, long count, long initialDelay, long period, TimeUnit unit) {
        return intervalRange(start, count, initialDelay, period, unit, Schedulers.computation());
    }

    
Signals a range of long values, the first after some initial delay and the rest periodically after.

The sequence completes immediately after the last value (start + count - 1) has been reached.

 
Backpressure:
 
The operator signals a MissingBackpressureException if the downstream can't keep up.
 
Scheduler:
 
you provide the Scheduler.

Params:
start – that start value of the range
count – the number of values to emit in total, if zero, the operator emits an onComplete after the initial delay.
initialDelay – the initial delay before signaling the first value (the start)
period – the period between subsequent values
unit – the unit of measure of the initialDelay and period amounts
scheduler – the target scheduler where the values and terminal signals will be emitted
Returns:
the new Flowable instance/**
     * Signals a range of long values, the first after some initial delay and the rest periodically after.
     * <p>
     * The sequence completes immediately after the last value (start + count - 1) has been reached.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator signals a {@code MissingBackpressureException} if the downstream can't keep up.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>you provide the {@link Scheduler}.</dd>
     * </dl>
     * @param start that start value of the range
     * @param count the number of values to emit in total, if zero, the operator emits an onComplete after the initial delay.
     * @param initialDelay the initial delay before signaling the first value (the start)
     * @param period the period between subsequent values
     * @param unit the unit of measure of the initialDelay and period amounts
     * @param scheduler the target scheduler where the values and terminal signals will be emitted
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public static Flowable<Long> intervalRange(long start, long count, long initialDelay, long period, TimeUnit unit, Scheduler scheduler) {
        if (count < 0L) {
            throw new IllegalArgumentException("count >= 0 required but it was " + count);
        }
        if (count == 0L) {
            return Flowable.<Long>empty().delay(initialDelay, unit, scheduler);
        }

        long end = start + (count - 1);
        if (start > 0 && end < 0) {
            throw new IllegalArgumentException("Overflow! start + count is bigger than Long.MAX_VALUE");
        }
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");

        return RxJavaPlugins.onAssembly(new FlowableIntervalRange(start, end, Math.max(0L, initialDelay), Math.max(0L, period), unit, scheduler));
    }

    
Returns a Flowable that signals the given (constant reference) item and then completes.


 Note that the item is taken and re-emitted as is and not computed by any means by just. Use fromCallable(Callable<? extends Object>) to generate a single item on demand (when Subscribers subscribe to it). 
 See the multi-parameter overloads of just to emit more than one (constant reference) items one after the other. Use fromArray(Object...) to emit an arbitrary number of items that are known upfront. 
 To emit the items of an Iterable sequence (such as a List), use fromIterable(Iterable<? extends Object>). 

 
Backpressure:
 
The operator honors backpressure from downstream.
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item – 
           the item to emit
Type parameters:
<T> – 
           the type of that item
See Also:
ReactiveX operators documentation: Just
just(Object, Object)
fromCallable(Callable<? extends Object>)
fromArray(Object...)
fromIterable(Iterable<? extends Object>)
Returns:
a Flowable that emits value as a single item and then completes/**
     * Returns a Flowable that signals the given (constant reference) item and then completes.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.png" alt="">
     * <p>
     * Note that the item is taken and re-emitted as is and not computed by any means by {@code just}. Use {@link #fromCallable(Callable)}
     * to generate a single item on demand (when {@code Subscriber}s subscribe to it).
     * <p>
     * See the multi-parameter overloads of {@code just} to emit more than one (constant reference) items one after the other.
     * Use {@link #fromArray(Object...)} to emit an arbitrary number of items that are known upfront.
     * <p>
     * To emit the items of an {@link Iterable} sequence (such as a {@link java.util.List}), use {@link #fromIterable(Iterable)}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item
     *            the item to emit
     * @param <T>
     *            the type of that item
     * @return a Flowable that emits {@code value} as a single item and then completes
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     * @see #just(Object, Object)
     * @see #fromCallable(Callable)
     * @see #fromArray(Object...)
     * @see #fromIterable(Iterable)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item) {
        ObjectHelper.requireNonNull(item, "item is null");
        return RxJavaPlugins.onAssembly(new FlowableJust<T>(item));
    }

    
Converts two items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts two items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");

        return fromArray(item1, item2);
    }

    
Converts three items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts three items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");

        return fromArray(item1, item2, item3);
    }

    
Converts four items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
item4 – 
           fourth item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts four items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param item4
     *            fourth item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3, T item4) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");
        ObjectHelper.requireNonNull(item4, "item4 is null");

        return fromArray(item1, item2, item3, item4);
    }

    
Converts five items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
item4 – 
           fourth item
item5 – 
           fifth item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts five items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param item4
     *            fourth item
     * @param item5
     *            fifth item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3, T item4, T item5) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");
        ObjectHelper.requireNonNull(item4, "item4 is null");
        ObjectHelper.requireNonNull(item5, "item5 is null");

        return fromArray(item1, item2, item3, item4, item5);
    }

    
Converts six items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
item4 – 
           fourth item
item5 – 
           fifth item
item6 – 
           sixth item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts six items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param item4
     *            fourth item
     * @param item5
     *            fifth item
     * @param item6
     *            sixth item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3, T item4, T item5, T item6) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");
        ObjectHelper.requireNonNull(item4, "item4 is null");
        ObjectHelper.requireNonNull(item5, "item5 is null");
        ObjectHelper.requireNonNull(item6, "item6 is null");

        return fromArray(item1, item2, item3, item4, item5, item6);
    }

    
Converts seven items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
item4 – 
           fourth item
item5 – 
           fifth item
item6 – 
           sixth item
item7 – 
           seventh item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts seven items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param item4
     *            fourth item
     * @param item5
     *            fifth item
     * @param item6
     *            sixth item
     * @param item7
     *            seventh item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3, T item4, T item5, T item6, T item7) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");
        ObjectHelper.requireNonNull(item4, "item4 is null");
        ObjectHelper.requireNonNull(item5, "item5 is null");
        ObjectHelper.requireNonNull(item6, "item6 is null");
        ObjectHelper.requireNonNull(item7, "item7 is null");

        return fromArray(item1, item2, item3, item4, item5, item6, item7);
    }

    
Converts eight items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
item4 – 
           fourth item
item5 – 
           fifth item
item6 – 
           sixth item
item7 – 
           seventh item
item8 – 
           eighth item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts eight items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param item4
     *            fourth item
     * @param item5
     *            fifth item
     * @param item6
     *            sixth item
     * @param item7
     *            seventh item
     * @param item8
     *            eighth item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3, T item4, T item5, T item6, T item7, T item8) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");
        ObjectHelper.requireNonNull(item4, "item4 is null");
        ObjectHelper.requireNonNull(item5, "item5 is null");
        ObjectHelper.requireNonNull(item6, "item6 is null");
        ObjectHelper.requireNonNull(item7, "item7 is null");
        ObjectHelper.requireNonNull(item8, "item8 is null");

        return fromArray(item1, item2, item3, item4, item5, item6, item7, item8);
    }

    
Converts nine items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
item4 – 
           fourth item
item5 – 
           fifth item
item6 – 
           sixth item
item7 – 
           seventh item
item8 – 
           eighth item
item9 – 
           ninth item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts nine items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param item4
     *            fourth item
     * @param item5
     *            fifth item
     * @param item6
     *            sixth item
     * @param item7
     *            seventh item
     * @param item8
     *            eighth item
     * @param item9
     *            ninth item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3, T item4, T item5, T item6, T item7, T item8, T item9) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");
        ObjectHelper.requireNonNull(item4, "item4 is null");
        ObjectHelper.requireNonNull(item5, "item5 is null");
        ObjectHelper.requireNonNull(item6, "item6 is null");
        ObjectHelper.requireNonNull(item7, "item7 is null");
        ObjectHelper.requireNonNull(item8, "item8 is null");
        ObjectHelper.requireNonNull(item9, "item9 is null");

        return fromArray(item1, item2, item3, item4, item5, item6, item7, item8, item9);
    }

    
Converts ten items into a Publisher that emits those items.



 
Backpressure:
 
The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).
 
Scheduler:
 
just does not operate by default on a particular Scheduler.

Params:
item1 – 
           first item
item2 – 
           second item
item3 – 
           third item
item4 – 
           fourth item
item5 – 
           fifth item
item6 – 
           sixth item
item7 – 
           seventh item
item8 – 
           eighth item
item9 – 
           ninth item
item10 – 
           tenth item
Type parameters:
<T> – 
           the type of these items
See Also:
ReactiveX operators documentation: Just
Returns:
a Flowable that emits each item/**
     * Converts ten items into a Publisher that emits those items.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/just.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals each value on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code just} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item1
     *            first item
     * @param item2
     *            second item
     * @param item3
     *            third item
     * @param item4
     *            fourth item
     * @param item5
     *            fifth item
     * @param item6
     *            sixth item
     * @param item7
     *            seventh item
     * @param item8
     *            eighth item
     * @param item9
     *            ninth item
     * @param item10
     *            tenth item
     * @param <T>
     *            the type of these items
     * @return a Flowable that emits each item
     * @see <a href="http://reactivex.io/documentation/operators/just.html">ReactiveX operators documentation: Just</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> just(T item1, T item2, T item3, T item4, T item5, T item6, T item7, T item8, T item9, T item10) {
        ObjectHelper.requireNonNull(item1, "item1 is null");
        ObjectHelper.requireNonNull(item2, "item2 is null");
        ObjectHelper.requireNonNull(item3, "item3 is null");
        ObjectHelper.requireNonNull(item4, "item4 is null");
        ObjectHelper.requireNonNull(item5, "item5 is null");
        ObjectHelper.requireNonNull(item6, "item6 is null");
        ObjectHelper.requireNonNull(item7, "item7 is null");
        ObjectHelper.requireNonNull(item8, "item8 is null");
        ObjectHelper.requireNonNull(item9, "item9 is null");
        ObjectHelper.requireNonNull(item10, "item10 is null");

        return fromArray(item1, item2, item3, item4, item5, item6, item7, item8, item9, item10);
    }

    
Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the
number of concurrent subscriptions to these Publishers.


 You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Iterable<? extends Publisher<? extends Object>>, int, int) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
sources – 
           the Iterable of Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
bufferSize – 
           the number of items to prefetch from each inner Publisher
Type parameters:
<T> – the common element base type
Throws:
IllegalArgumentException –  if maxConcurrency is less than or equal to 0
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Iterable<? extends Publisher<? extends Object>>, int, int)
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the
     * number of concurrent subscriptions to these Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Iterable, int, int)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the Iterable of Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @param bufferSize
     *            the number of items to prefetch from each inner Publisher
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @throws IllegalArgumentException
     *             if {@code maxConcurrency} is less than or equal to 0
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Iterable, int, int)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(Iterable<? extends Publisher<? extends T>> sources, int maxConcurrency, int bufferSize) {
        return fromIterable(sources).flatMap((Function)Functions.identity(), false, maxConcurrency, bufferSize);
    }

    
Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the
number of concurrent subscriptions to these Publishers.


 You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeArray does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeArrayDelayError(int, int, Publisher<? extends Object>[]) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
sources – 
           the array of Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
bufferSize – 
           the number of items to prefetch from each inner Publisher
Type parameters:
<T> – the common element base type
Throws:
IllegalArgumentException –  if maxConcurrency is less than or equal to 0
See Also:
ReactiveX operators documentation: Merge
mergeArrayDelayError(int, int, Publisher<? extends Object>...)
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the
     * number of concurrent subscriptions to these Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeArray} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeArrayDelayError(int, int, Publisher[])} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the array of Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @param bufferSize
     *            the number of items to prefetch from each inner Publisher
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @throws IllegalArgumentException
     *             if {@code maxConcurrency} is less than or equal to 0
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeArrayDelayError(int, int, Publisher...)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeArray(int maxConcurrency, int bufferSize, Publisher<? extends T>... sources) {
        return fromArray(sources).flatMap((Function)Functions.identity(), false, maxConcurrency, bufferSize);
    }

    
Flattens an Iterable of Publishers into one Publisher, without any transformation.


 You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Iterable<? extends Publisher<? extends Object>>) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
sources – 
           the Iterable of Publishers
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Iterable<? extends Publisher<? extends Object>>)
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an Iterable of Publishers into one Publisher, without any transformation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Iterable)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the Iterable of Publishers
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Iterable)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(Iterable<? extends Publisher<? extends T>> sources) {
        return fromIterable(sources).flatMap((Function)Functions.identity());
    }

    
Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the
number of concurrent subscriptions to these Publishers.


 You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Iterable<? extends Publisher<? extends Object>>, int) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
sources – 
           the Iterable of Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
Type parameters:
<T> – the common element base type
Throws:
IllegalArgumentException –  if maxConcurrency is less than or equal to 0
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Iterable<? extends Publisher<? extends Object>>, int)
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an Iterable of Publishers into one Publisher, without any transformation, while limiting the
     * number of concurrent subscriptions to these Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Iterable, int)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the Iterable of Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @throws IllegalArgumentException
     *             if {@code maxConcurrency} is less than or equal to 0
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Iterable, int)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(Iterable<? extends Publisher<? extends T>> sources, int maxConcurrency) {
        return fromIterable(sources).flatMap((Function)Functions.identity(), maxConcurrency);
    }

    
Flattens a Publisher that emits Publishers into a single Publisher that emits the items emitted by
those Publishers, without any transformation.


 You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in unbounded mode (i.e., no backpressure is applied to it). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Publisher<? extends Publisher<? extends Object>>) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
sources – 
           a Publisher that emits Publishers
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Publisher<? extends Publisher<? extends Object>>)
Returns:
a Flowable that emits items that are the result of flattening the Publishers emitted by the source Publisher/**
     * Flattens a Publisher that emits Publishers into a single Publisher that emits the items emitted by
     * those Publishers, without any transformation.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.oo.png" alt="">
     * <p>
     * You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed
     *  in unbounded mode (i.e., no backpressure is applied to it). The inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Publisher)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            a Publisher that emits Publishers
     * @return a Flowable that emits items that are the result of flattening the Publishers emitted by the
     *         {@code source} Publisher
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Publisher)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(Publisher<? extends Publisher<? extends T>> sources) {
        return merge(sources, bufferSize());
    }

    
Flattens a Publisher that emits Publishers into a single Publisher that emits the items emitted by
those Publishers, without any transformation, while limiting the maximum number of concurrent
subscriptions to these Publishers.


 You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. Both the outer and inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Publisher<? extends Publisher<? extends Object>>, int) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
sources – 
           a Publisher that emits Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
Type parameters:
<T> – the common element base type
Throws:
IllegalArgumentException –  if maxConcurrency is less than or equal to 0
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Publisher<? extends Publisher<? extends Object>>, int)
Returns:
a Flowable that emits items that are the result of flattening the Publishers emitted by the source Publisher
Since:
1.1.0/**
     * Flattens a Publisher that emits Publishers into a single Publisher that emits the items emitted by
     * those Publishers, without any transformation, while limiting the maximum number of concurrent
     * subscriptions to these Publishers.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.oo.png" alt="">
     * <p>
     * You can combine the items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Publisher, int)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            a Publisher that emits Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @return a Flowable that emits items that are the result of flattening the Publishers emitted by the
     *         {@code source} Publisher
     * @throws IllegalArgumentException
     *             if {@code maxConcurrency} is less than or equal to 0
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Publisher, int)
     * @since 1.1.0
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(Publisher<? extends Publisher<? extends T>> sources, int maxConcurrency) {
        return fromPublisher(sources).flatMap((Function)Functions.identity(), maxConcurrency);
    }

    
Flattens an Array of Publishers into one Publisher, without any transformation.


 You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeArray does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeArrayDelayError(Publisher<? extends Object>...) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
sources – 
           the array of Publishers
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
mergeArrayDelayError(Publisher<? extends Object>...)
Returns:
a Flowable that emits all of the items emitted by the Publishers in the Array/**
     * Flattens an Array of Publishers into one Publisher, without any transformation.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.io.png" alt="">
     * <p>
     * You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeArray} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeArrayDelayError(Publisher...)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the array of Publishers
     * @return a Flowable that emits all of the items emitted by the Publishers in the Array
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeArrayDelayError(Publisher...)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeArray(Publisher<? extends T>... sources) {
        return fromArray(sources).flatMap((Function)Functions.identity(), sources.length);
    }

    
Flattens two Publishers into a single Publisher, without any transformation.


 You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Publisher<? extends Object>, Publisher<? extends Object>) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
source1 – 
           a Publisher to be merged
source2 – 
           a Publisher to be merged
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Publisher<? extends Object>, Publisher<? extends Object>)
Returns:
a Flowable that emits all of the items emitted by the source Publishers/**
     * Flattens two Publishers into a single Publisher, without any transformation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Publisher, Publisher)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be merged
     * @param source2
     *            a Publisher to be merged
     * @return a Flowable that emits all of the items emitted by the source Publishers
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Publisher, Publisher)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(Publisher<? extends T> source1, Publisher<? extends T> source2) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        return fromArray(source1, source2).flatMap((Function)Functions.identity(), false, 2);
    }

    
Flattens three Publishers into a single Publisher, without any transformation.


 You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
source1 – 
           a Publisher to be merged
source2 – 
           a Publisher to be merged
source3 – 
           a Publisher to be merged
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>)
Returns:
a Flowable that emits all of the items emitted by the source Publishers/**
     * Flattens three Publishers into a single Publisher, without any transformation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Publisher, Publisher, Publisher)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be merged
     * @param source2
     *            a Publisher to be merged
     * @param source3
     *            a Publisher to be merged
     * @return a Flowable that emits all of the items emitted by the source Publishers
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Publisher, Publisher, Publisher)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(Publisher<? extends T> source1, Publisher<? extends T> source2, Publisher<? extends T> source3) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        return fromArray(source1, source2, source3).flatMap((Function)Functions.identity(), false, 3);
    }

    
Flattens four Publishers into a single Publisher, without any transformation.


 You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by using the merge method. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
merge does not operate by default on a particular Scheduler.
 
Error handling:
 
If any of the source Publishers signal a Throwable via onError, the resulting Flowable terminates with that Throwable and all other source Publishers are canceled. If more than one Publisher signals an error, the resulting Flowable may terminate with the first one's error or, depending on the concurrency of the sources, may terminate with a CompositeException containing two or more of the various error signals. Throwables that didn't make into the composite will be sent (individually) to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. Similarly, Throwables signaled by source(s) after the returned Flowable has been canceled or terminated with a (composite) error will be sent to the same global error handler. Use mergeDelayError(Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>) to merge sources and terminate only when all source Publishers have completed or failed with an error. 

Params:
source1 – 
           a Publisher to be merged
source2 – 
           a Publisher to be merged
source3 – 
           a Publisher to be merged
source4 – 
           a Publisher to be merged
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
mergeDelayError(Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>)
Returns:
a Flowable that emits all of the items emitted by the source Publishers/**
     * Flattens four Publishers into a single Publisher, without any transformation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code merge} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code merge} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If any of the source {@code Publisher}s signal a {@code Throwable} via {@code onError}, the resulting
     *  {@code Flowable} terminates with that {@code Throwable} and all other source {@code Publisher}s are canceled.
     *  If more than one {@code Publisher} signals an error, the resulting {@code Flowable} may terminate with the
     *  first one's error or, depending on the concurrency of the sources, may terminate with a
     *  {@code CompositeException} containing two or more of the various error signals.
     *  {@code Throwable}s that didn't make into the composite will be sent (individually) to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors. Similarly, {@code Throwable}s
     *  signaled by source(s) after the returned {@code Flowable} has been canceled or terminated with a
     *  (composite) error will be sent to the same global error handler.
     *  Use {@link #mergeDelayError(Publisher, Publisher, Publisher, Publisher)} to merge sources and terminate only when all source {@code Publisher}s
     *  have completed or failed with an error.
     *  </dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be merged
     * @param source2
     *            a Publisher to be merged
     * @param source3
     *            a Publisher to be merged
     * @param source4
     *            a Publisher to be merged
     * @return a Flowable that emits all of the items emitted by the source Publishers
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @see #mergeDelayError(Publisher, Publisher, Publisher, Publisher)
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> merge(
            Publisher<? extends T> source1, Publisher<? extends T> source2,
            Publisher<? extends T> source3, Publisher<? extends T> source4) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        return fromArray(source1, source2, source3, source4).flatMap((Function)Functions.identity(), false, 4);
    }

    
Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all
successfully emitted items from each of the source Publishers without being interrupted by an error
notification from one of them.
 This behaves like merge(Publisher<? extends Publisher<? extends Object>>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. All inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the Iterable of Publishers
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all
     * successfully emitted items from each of the source Publishers without being interrupted by an error
     * notification from one of them.
     * <p>
     * This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an
     * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that
     * error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. All inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the Iterable of Publishers
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(Iterable<? extends Publisher<? extends T>> sources) {
        return fromIterable(sources).flatMap((Function)Functions.identity(), true);
    }

    
Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all
successfully emitted items from each of the source Publishers without being interrupted by an error
notification from one of them, while limiting the number of concurrent subscriptions to these Publishers.
 This behaves like merge(Publisher<? extends Publisher<? extends Object>>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. All inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the Iterable of Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
bufferSize – 
           the number of items to prefetch from each inner Publisher
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all
     * successfully emitted items from each of the source Publishers without being interrupted by an error
     * notification from one of them, while limiting the number of concurrent subscriptions to these Publishers.
     * <p>
     * This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an
     * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that
     * error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. All inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the Iterable of Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @param bufferSize
     *            the number of items to prefetch from each inner Publisher
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(Iterable<? extends Publisher<? extends T>> sources, int maxConcurrency, int bufferSize) {
        return fromIterable(sources).flatMap((Function)Functions.identity(), true, maxConcurrency, bufferSize);
    }

    
Flattens an array of Publishers into one Publisher, in a way that allows a Subscriber to receive all
successfully emitted items from each of the source Publishers without being interrupted by an error
notification from one of them, while limiting the number of concurrent subscriptions to these Publishers.
 This behaves like merge(Publisher<? extends Publisher<? extends Object>>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. All source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeArrayDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the array of Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
bufferSize – 
           the number of items to prefetch from each inner Publisher
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an array of Publishers into one Publisher, in a way that allows a Subscriber to receive all
     * successfully emitted items from each of the source Publishers without being interrupted by an error
     * notification from one of them, while limiting the number of concurrent subscriptions to these Publishers.
     * <p>
     * This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an
     * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that
     * error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. All source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeArrayDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the array of Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @param bufferSize
     *            the number of items to prefetch from each inner Publisher
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeArrayDelayError(int maxConcurrency, int bufferSize, Publisher<? extends T>... sources) {
        return fromArray(sources).flatMap((Function)Functions.identity(), true, maxConcurrency, bufferSize);
    }

    
Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all
successfully emitted items from each of the source Publishers without being interrupted by an error
notification from one of them, while limiting the number of concurrent subscriptions to these Publishers.
 This behaves like merge(Publisher<? extends Publisher<? extends Object>>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. All inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the Iterable of Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an Iterable of Publishers into one Publisher, in a way that allows a Subscriber to receive all
     * successfully emitted items from each of the source Publishers without being interrupted by an error
     * notification from one of them, while limiting the number of concurrent subscriptions to these Publishers.
     * <p>
     * This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an
     * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that
     * error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. All inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the Iterable of Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(Iterable<? extends Publisher<? extends T>> sources, int maxConcurrency) {
        return fromIterable(sources).flatMap((Function)Functions.identity(), true, maxConcurrency);
    }

    
Flattens a Publisher that emits Publishers into one Publisher, in a way that allows a Subscriber to
receive all successfully emitted items from all of the source Publishers without being interrupted by
an error notification from one of them.
 This behaves like merge(Publisher<? extends Publisher<? extends Object>>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in unbounded mode (i.e., no backpressure is applied to it). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           a Publisher that emits Publishers
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits all of the items emitted by the Publishers emitted by the source Publisher/**
     * Flattens a Publisher that emits Publishers into one Publisher, in a way that allows a Subscriber to
     * receive all successfully emitted items from all of the source Publishers without being interrupted by
     * an error notification from one of them.
     * <p>
     * This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an
     * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that
     * error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed
     *  in unbounded mode (i.e., no backpressure is applied to it). The inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            a Publisher that emits Publishers
     * @return a Flowable that emits all of the items emitted by the Publishers emitted by the
     *         {@code source} Publisher
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(Publisher<? extends Publisher<? extends T>> sources) {
        return mergeDelayError(sources, bufferSize());
    }

    
Flattens a Publisher that emits Publishers into one Publisher, in a way that allows a Subscriber to
receive all successfully emitted items from all of the source Publishers without being interrupted by
an error notification from one of them, while limiting the
number of concurrent subscriptions to these Publishers.
 This behaves like merge(Publisher<? extends Publisher<? extends Object>>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. Both the outer and inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           a Publisher that emits Publishers
maxConcurrency – 
           the maximum number of Publishers that may be subscribed to concurrently
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits all of the items emitted by the Publishers emitted by the source Publisher
Since:
2.0/**
     * Flattens a Publisher that emits Publishers into one Publisher, in a way that allows a Subscriber to
     * receive all successfully emitted items from all of the source Publishers without being interrupted by
     * an error notification from one of them, while limiting the
     * number of concurrent subscriptions to these Publishers.
     * <p>
     * This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an
     * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that
     * error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            a Publisher that emits Publishers
     * @param maxConcurrency
     *            the maximum number of Publishers that may be subscribed to concurrently
     * @return a Flowable that emits all of the items emitted by the Publishers emitted by the
     *         {@code source} Publisher
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     * @since 2.0
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(Publisher<? extends Publisher<? extends T>> sources, int maxConcurrency) {
        return fromPublisher(sources).flatMap((Function)Functions.identity(), true, maxConcurrency);
    }

    
Flattens an array of Publishers into one Flowable, in a way that allows a Subscriber to receive all
successfully emitted items from each of the source Publishers without being interrupted by an error
notification from one of them.
 This behaves like merge(Publisher<? extends Publisher<? extends Object>>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. Both the outer and inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeArrayDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the Iterable of Publishers
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits items that are the result of flattening the items emitted by the
        Publishers in the Iterable/**
     * Flattens an array of Publishers into one Flowable, in a way that allows a Subscriber to receive all
     * successfully emitted items from each of the source Publishers without being interrupted by an error
     * notification from one of them.
     * <p>
     * This behaves like {@link #merge(Publisher)} except that if any of the merged Publishers notify of an
     * error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from propagating that
     * error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both the outer and inner {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeArrayDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param sources
     *            the Iterable of Publishers
     * @return a Flowable that emits items that are the result of flattening the items emitted by the
     *         Publishers in the Iterable
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeArrayDelayError(Publisher<? extends T>... sources) {
        return fromArray(sources).flatMap((Function)Functions.identity(), true, sources.length);
    }

    
Flattens two Publishers into one Publisher, in a way that allows a Subscriber to receive all
successfully emitted items from each of the source Publishers without being interrupted by an error
notification from one of them.
 This behaves like merge(Publisher<? extends Object>, Publisher<? extends Object>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if both merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
source1 – 
           a Publisher to be merged
source2 – 
           a Publisher to be merged
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits all of the items that are emitted by the two source Publishers/**
     * Flattens two Publishers into one Publisher, in a way that allows a Subscriber to receive all
     * successfully emitted items from each of the source Publishers without being interrupted by an error
     * notification from one of them.
     * <p>
     * This behaves like {@link #merge(Publisher, Publisher)} except that if any of the merged Publishers
     * notify of an error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain from
     * propagating that error notification until all of the merged Publishers have finished emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if both merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be merged
     * @param source2
     *            a Publisher to be merged
     * @return a Flowable that emits all of the items that are emitted by the two source Publishers
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(Publisher<? extends T> source1, Publisher<? extends T> source2) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        return fromArray(source1, source2).flatMap((Function)Functions.identity(), true, 2);
    }

    
Flattens three Publishers into one Publisher, in a way that allows a Subscriber to receive all
successfully emitted items from all of the source Publishers without being interrupted by an error
notification from one of them.
 This behaves like merge(Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
source1 – 
           a Publisher to be merged
source2 – 
           a Publisher to be merged
source3 – 
           a Publisher to be merged
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits all of the items that are emitted by the source Publishers/**
     * Flattens three Publishers into one Publisher, in a way that allows a Subscriber to receive all
     * successfully emitted items from all of the source Publishers without being interrupted by an error
     * notification from one of them.
     * <p>
     * This behaves like {@link #merge(Publisher, Publisher, Publisher)} except that if any of the merged
     * Publishers notify of an error via {@link Subscriber#onError onError}, {@code mergeDelayError} will refrain
     * from propagating that error notification until all of the merged Publishers have finished emitting
     * items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be merged
     * @param source2
     *            a Publisher to be merged
     * @param source3
     *            a Publisher to be merged
     * @return a Flowable that emits all of the items that are emitted by the source Publishers
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(Publisher<? extends T> source1, Publisher<? extends T> source2, Publisher<? extends T> source3) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        return fromArray(source1, source2, source3).flatMap((Function)Functions.identity(), true, 3);
    }

    
Flattens four Publishers into one Publisher, in a way that allows a Subscriber to receive all
successfully emitted items from all of the source Publishers without being interrupted by an error
notification from one of them.
 This behaves like merge(Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>, Publisher<? extends Object>) except that if any of the merged Publishers notify of an error via onError, mergeDelayError will refrain from propagating that error notification until all of the merged Publishers have finished emitting items. 


 Even if multiple merged Publishers send onError notifications, mergeDelayError will only invoke the onError method of its Subscribers once. 

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeDelayError does not operate by default on a particular Scheduler.

Params:
source1 – 
           a Publisher to be merged
source2 – 
           a Publisher to be merged
source3 – 
           a Publisher to be merged
source4 – 
           a Publisher to be merged
Type parameters:
<T> – the common element base type
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits all of the items that are emitted by the source Publishers/**
     * Flattens four Publishers into one Publisher, in a way that allows a Subscriber to receive all
     * successfully emitted items from all of the source Publishers without being interrupted by an error
     * notification from one of them.
     * <p>
     * This behaves like {@link #merge(Publisher, Publisher, Publisher, Publisher)} except that if any of
     * the merged Publishers notify of an error via {@link Subscriber#onError onError}, {@code mergeDelayError}
     * will refrain from propagating that error notification until all of the merged Publishers have finished
     * emitting items.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeDelayError.png" alt="">
     * <p>
     * Even if multiple merged Publishers send {@code onError} notifications, {@code mergeDelayError} will only
     * invoke the {@code onError} method of its Subscribers once.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element base type
     * @param source1
     *            a Publisher to be merged
     * @param source2
     *            a Publisher to be merged
     * @param source3
     *            a Publisher to be merged
     * @param source4
     *            a Publisher to be merged
     * @return a Flowable that emits all of the items that are emitted by the source Publishers
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> mergeDelayError(
            Publisher<? extends T> source1, Publisher<? extends T> source2,
            Publisher<? extends T> source3, Publisher<? extends T> source4) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        return fromArray(source1, source2, source3, source4).flatMap((Function)Functions.identity(), true, 4);
    }

    
Returns a Flowable that never sends any items or notifications to a Subscriber. 



This Publisher is useful primarily for testing purposes.

 
Backpressure:
 
This source doesn't produce any elements and effectively ignores downstream backpressure.
 
Scheduler:
 
never does not operate by default on a particular Scheduler.

Type parameters:
<T> – 
           the type of items (not) emitted by the Publisher
See Also:
ReactiveX operators documentation: Never
Returns:
a Flowable that never emits any items or sends any notifications to a Subscriber/**
     * Returns a Flowable that never sends any items or notifications to a {@link Subscriber}.
     * <p>
     * <img width="640" height="185" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/never.png" alt="">
     * <p>
     * This Publisher is useful primarily for testing purposes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This source doesn't produce any elements and effectively ignores downstream backpressure.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code never} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T>
     *            the type of items (not) emitted by the Publisher
     * @return a Flowable that never emits any items or sends any notifications to a {@link Subscriber}
     * @see <a href="http://reactivex.io/documentation/operators/empty-never-throw.html">ReactiveX operators documentation: Never</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    @SuppressWarnings("unchecked")
    public static <T> Flowable<T> never() {
        return RxJavaPlugins.onAssembly((Flowable<T>) FlowableNever.INSTANCE);
    }

    
Returns a Flowable that emits a sequence of Integers within a specified range.



 
Backpressure:
 
The operator honors backpressure from downstream and signals values on-demand (i.e., when requested).
 
Scheduler:
 
range does not operate by default on a particular Scheduler.

Params:
start – 
           the value of the first Integer in the sequence
count – 
           the number of sequential Integers to generate
Throws:
IllegalArgumentException –  if count is less than zero, or if start + count − 1 exceeds Integer.MAX_VALUE
See Also:
ReactiveX operators documentation: Range
Returns:
a Flowable that emits a range of sequential Integers/**
     * Returns a Flowable that emits a sequence of Integers within a specified range.
     * <p>
     * <img width="640" height="195" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/range.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals values on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code range} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param start
     *            the value of the first Integer in the sequence
     * @param count
     *            the number of sequential Integers to generate
     * @return a Flowable that emits a range of sequential Integers
     * @throws IllegalArgumentException
     *             if {@code count} is less than zero, or if {@code start} + {@code count} &minus; 1 exceeds
     *             {@code Integer.MAX_VALUE}
     * @see <a href="http://reactivex.io/documentation/operators/range.html">ReactiveX operators documentation: Range</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static Flowable<Integer> range(int start, int count) {
        if (count < 0) {
            throw new IllegalArgumentException("count >= 0 required but it was " + count);
        } else
        if (count == 0) {
            return empty();
        } else
        if (count == 1) {
            return just(start);
        } else
        if ((long)start + (count - 1) > Integer.MAX_VALUE) {
            throw new IllegalArgumentException("Integer overflow");
        }
        return RxJavaPlugins.onAssembly(new FlowableRange(start, count));
    }

    
Returns a Flowable that emits a sequence of Longs within a specified range.



 
Backpressure:
 
The operator honors backpressure from downstream and signals values on-demand (i.e., when requested).
 
Scheduler:
 
rangeLong does not operate by default on a particular Scheduler.

Params:
start – 
           the value of the first Long in the sequence
count – 
           the number of sequential Longs to generate
Throws:
IllegalArgumentException –  if count is less than zero, or if start + count − 1 exceeds Long.MAX_VALUE
See Also:
ReactiveX operators documentation: Range
Returns:
a Flowable that emits a range of sequential Longs/**
     * Returns a Flowable that emits a sequence of Longs within a specified range.
     * <p>
     * <img width="640" height="195" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/range.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and signals values on-demand (i.e., when requested).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code rangeLong} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param start
     *            the value of the first Long in the sequence
     * @param count
     *            the number of sequential Longs to generate
     * @return a Flowable that emits a range of sequential Longs
     * @throws IllegalArgumentException
     *             if {@code count} is less than zero, or if {@code start} + {@code count} &minus; 1 exceeds
     *             {@code Long.MAX_VALUE}
     * @see <a href="http://reactivex.io/documentation/operators/range.html">ReactiveX operators documentation: Range</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static Flowable<Long> rangeLong(long start, long count) {
        if (count < 0) {
            throw new IllegalArgumentException("count >= 0 required but it was " + count);
        }

        if (count == 0) {
            return empty();
        }

        if (count == 1) {
            return just(start);
        }

        long end = start + (count - 1);
        if (start > 0 && end < 0) {
            throw new IllegalArgumentException("Overflow! start + count is bigger than Long.MAX_VALUE");
        }

        return RxJavaPlugins.onAssembly(new FlowableRangeLong(start, count));
    }

    
Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
same by comparing the items emitted by each Publisher pairwise.



 
Backpressure:
 
This operator honors downstream backpressure and expects both of its sources
 to honor backpressure as well. If violated, the operator will emit a MissingBackpressureException.
 
Scheduler:
 
sequenceEqual does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first Publisher to compare
source2 – 
           the second Publisher to compare
Type parameters:
<T> – 
           the type of items emitted by each Publisher
See Also:
ReactiveX operators documentation: SequenceEqual
Returns:
a Flowable that emits a Boolean value that indicates whether the two sequences are the same/**
     * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
     * same by comparing the items emitted by each Publisher pairwise.
     * <p>
     * <img width="640" height="385" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sequenceEqual.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator honors downstream backpressure and expects both of its sources
     *  to honor backpressure as well. If violated, the operator will emit a MissingBackpressureException.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param source1
     *            the first Publisher to compare
     * @param source2
     *            the second Publisher to compare
     * @param <T>
     *            the type of items emitted by each Publisher
     * @return a Flowable that emits a Boolean value that indicates whether the two sequences are the same
     * @see <a href="http://reactivex.io/documentation/operators/sequenceequal.html">ReactiveX operators documentation: SequenceEqual</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Single<Boolean> sequenceEqual(Publisher<? extends T> source1, Publisher<? extends T> source2) {
        return sequenceEqual(source1, source2, ObjectHelper.equalsPredicate(), bufferSize());
    }

    
Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
same by comparing the items emitted by each Publisher pairwise based on the results of a specified
equality function.



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator signals a MissingBackpressureException.
 
Scheduler:
 
sequenceEqual does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first Publisher to compare
source2 – 
           the second Publisher to compare
isEqual – 
           a function used to compare items emitted by each Publisher
Type parameters:
<T> – 
           the type of items emitted by each Publisher
See Also:
ReactiveX operators documentation: SequenceEqual
Returns:
a Single that emits a Boolean value that indicates whether the two Publisher sequences
        are the same according to the specified function/**
     * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
     * same by comparing the items emitted by each Publisher pairwise based on the results of a specified
     * equality function.
     * <p>
     * <img width="640" height="385" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sequenceEqual.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator signals a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param source1
     *            the first Publisher to compare
     * @param source2
     *            the second Publisher to compare
     * @param isEqual
     *            a function used to compare items emitted by each Publisher
     * @param <T>
     *            the type of items emitted by each Publisher
     * @return a Single that emits a Boolean value that indicates whether the two Publisher sequences
     *         are the same according to the specified function
     * @see <a href="http://reactivex.io/documentation/operators/sequenceequal.html">ReactiveX operators documentation: SequenceEqual</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Single<Boolean> sequenceEqual(Publisher<? extends T> source1, Publisher<? extends T> source2,
            BiPredicate<? super T, ? super T> isEqual) {
        return sequenceEqual(source1, source2, isEqual, bufferSize());
    }

    
Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
same by comparing the items emitted by each Publisher pairwise based on the results of a specified
equality function.



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers are expected to honor backpressure; if violated, the operator signals a MissingBackpressureException.
 
Scheduler:
 
sequenceEqual does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first Publisher to compare
source2 – 
           the second Publisher to compare
isEqual – 
           a function used to compare items emitted by each Publisher
bufferSize – 
           the number of items to prefetch from the first and second source Publisher
Type parameters:
<T> – 
           the type of items emitted by each Publisher
See Also:
ReactiveX operators documentation: SequenceEqual
Returns:
a Single that emits a Boolean value that indicates whether the two Publisher sequences
        are the same according to the specified function/**
     * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
     * same by comparing the items emitted by each Publisher pairwise based on the results of a specified
     * equality function.
     * <p>
     * <img width="640" height="385" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sequenceEqual.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator signals a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param source1
     *            the first Publisher to compare
     * @param source2
     *            the second Publisher to compare
     * @param isEqual
     *            a function used to compare items emitted by each Publisher
     * @param bufferSize
     *            the number of items to prefetch from the first and second source Publisher
     * @param <T>
     *            the type of items emitted by each Publisher
     * @return a Single that emits a Boolean value that indicates whether the two Publisher sequences
     *         are the same according to the specified function
     * @see <a href="http://reactivex.io/documentation/operators/sequenceequal.html">ReactiveX operators documentation: SequenceEqual</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Single<Boolean> sequenceEqual(Publisher<? extends T> source1, Publisher<? extends T> source2,
            BiPredicate<? super T, ? super T> isEqual, int bufferSize) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(isEqual, "isEqual is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableSequenceEqualSingle<T>(source1, source2, isEqual, bufferSize));
    }

    
Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
same by comparing the items emitted by each Publisher pairwise.



 
Backpressure:
 
This operator honors downstream backpressure and expects both of its sources
 to honor backpressure as well. If violated, the operator will emit a MissingBackpressureException.
 
Scheduler:
 
sequenceEqual does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first Publisher to compare
source2 – 
           the second Publisher to compare
bufferSize – 
           the number of items to prefetch from the first and second source Publisher
Type parameters:
<T> – 
           the type of items emitted by each Publisher
See Also:
ReactiveX operators documentation: SequenceEqual
Returns:
a Single that emits a Boolean value that indicates whether the two sequences are the same/**
     * Returns a Single that emits a Boolean value that indicates whether two Publisher sequences are the
     * same by comparing the items emitted by each Publisher pairwise.
     * <p>
     * <img width="640" height="385" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sequenceEqual.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator honors downstream backpressure and expects both of its sources
     *  to honor backpressure as well. If violated, the operator will emit a MissingBackpressureException.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sequenceEqual} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param source1
     *            the first Publisher to compare
     * @param source2
     *            the second Publisher to compare
     * @param bufferSize
     *            the number of items to prefetch from the first and second source Publisher
     * @param <T>
     *            the type of items emitted by each Publisher
     * @return a Single that emits a Boolean value that indicates whether the two sequences are the same
     * @see <a href="http://reactivex.io/documentation/operators/sequenceequal.html">ReactiveX operators documentation: SequenceEqual</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Single<Boolean> sequenceEqual(Publisher<? extends T> source1, Publisher<? extends T> source2, int bufferSize) {
        return sequenceEqual(source1, source2, ObjectHelper.equalsPredicate(), bufferSize);
    }

    
Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
most recently emitted of those Publishers.


 switchOnNext subscribes to a Publisher that emits Publishers. Each time it observes one of these emitted Publishers, the Publisher returned by switchOnNext begins emitting the items emitted by that Publisher. When a new Publisher is emitted, switchOnNext stops emitting items from the earlier-emitted Publisher and begins emitting items from the new one. 

The resulting Publisher completes if both the outer Publisher and the last inner Publisher, if any, complete.
If the outer Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.

 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchOnNext does not operate by default on a particular Scheduler.

Params:
sources – 
           the source Publisher that emits Publishers
bufferSize – 
           the number of items to prefetch from the inner Publishers
Type parameters:
<T> – the item type
See Also:
ReactiveX operators documentation: Switch
Returns:
a Flowable that emits the items emitted by the Publisher most recently emitted by the source
        Publisher/**
     * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
     * most recently emitted of those Publishers.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchDo.png" alt="">
     * <p>
     * {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of
     * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items
     * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items
     * from the earlier-emitted Publisher and begins emitting items from the new one.
     * <p>
     * The resulting Publisher completes if both the outer Publisher and the last inner Publisher, if any, complete.
     * If the outer Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchOnNext} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the item type
     * @param sources
     *            the source Publisher that emits Publishers
     * @param bufferSize
     *            the number of items to prefetch from the inner Publishers
     * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/switch.html">ReactiveX operators documentation: Switch</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> switchOnNext(Publisher<? extends Publisher<? extends T>> sources, int bufferSize) {
        return fromPublisher(sources).switchMap((Function)Functions.identity(), bufferSize);
    }

    
Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
most recently emitted of those Publishers.


 switchOnNext subscribes to a Publisher that emits Publishers. Each time it observes one of these emitted Publishers, the Publisher returned by switchOnNext begins emitting the items emitted by that Publisher. When a new Publisher is emitted, switchOnNext stops emitting items from the earlier-emitted Publisher and begins emitting items from the new one. 

The resulting Publisher completes if both the outer Publisher and the last inner Publisher, if any, complete.
If the outer Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.

 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchOnNext does not operate by default on a particular Scheduler.

Params:
sources – 
           the source Publisher that emits Publishers
Type parameters:
<T> – the item type
See Also:
ReactiveX operators documentation: Switch
Returns:
a Flowable that emits the items emitted by the Publisher most recently emitted by the source
        Publisher/**
     * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
     * most recently emitted of those Publishers.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchDo.png" alt="">
     * <p>
     * {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of
     * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items
     * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items
     * from the earlier-emitted Publisher and begins emitting items from the new one.
     * <p>
     * The resulting Publisher completes if both the outer Publisher and the last inner Publisher, if any, complete.
     * If the outer Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchOnNext} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the item type
     * @param sources
     *            the source Publisher that emits Publishers
     * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/switch.html">ReactiveX operators documentation: Switch</a>
     */
    @SuppressWarnings({ "unchecked", "rawtypes" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> switchOnNext(Publisher<? extends Publisher<? extends T>> sources) {
        return fromPublisher(sources).switchMap((Function)Functions.identity());
    }

    
Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
most recently emitted of those Publishers and delays any exception until all Publishers terminate.


 switchOnNext subscribes to a Publisher that emits Publishers. Each time it observes one of these emitted Publishers, the Publisher returned by switchOnNext begins emitting the items emitted by that Publisher. When a new Publisher is emitted, switchOnNext stops emitting items from the earlier-emitted Publisher and begins emitting items from the new one. 

The resulting Publisher completes if both the main Publisher and the last inner Publisher, if any, complete.
If the main Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.

 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchOnNextDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the source Publisher that emits Publishers
Type parameters:
<T> – the item type
See Also:
ReactiveX operators documentation: Switch
Returns:
a Flowable that emits the items emitted by the Publisher most recently emitted by the source
        Publisher
Since:
2.0/**
     * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
     * most recently emitted of those Publishers and delays any exception until all Publishers terminate.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchDo.png" alt="">
     * <p>
     * {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of
     * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items
     * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items
     * from the earlier-emitted Publisher and begins emitting items from the new one.
     * <p>
     * The resulting Publisher completes if both the main Publisher and the last inner Publisher, if any, complete.
     * If the main Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
     * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchOnNextDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the item type
     * @param sources
     *            the source Publisher that emits Publishers
     * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/switch.html">ReactiveX operators documentation: Switch</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> switchOnNextDelayError(Publisher<? extends Publisher<? extends T>> sources) {
        return switchOnNextDelayError(sources, bufferSize());
    }

    
Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
most recently emitted of those Publishers and delays any exception until all Publishers terminate.


 switchOnNext subscribes to a Publisher that emits Publishers. Each time it observes one of these emitted Publishers, the Publisher returned by switchOnNext begins emitting the items emitted by that Publisher. When a new Publisher is emitted, switchOnNext stops emitting items from the earlier-emitted Publisher and begins emitting items from the new one. 

The resulting Publisher completes if both the main Publisher and the last inner Publisher, if any, complete.
If the main Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.

 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchOnNextDelayError does not operate by default on a particular Scheduler.

Params:
sources – 
           the source Publisher that emits Publishers
prefetch – 
           the number of items to prefetch from the inner Publishers
Type parameters:
<T> – the item type
See Also:
ReactiveX operators documentation: Switch
Returns:
a Flowable that emits the items emitted by the Publisher most recently emitted by the source
        Publisher
Since:
2.0/**
     * Converts a Publisher that emits Publishers into a Publisher that emits the items emitted by the
     * most recently emitted of those Publishers and delays any exception until all Publishers terminate.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchDo.png" alt="">
     * <p>
     * {@code switchOnNext} subscribes to a Publisher that emits Publishers. Each time it observes one of
     * these emitted Publishers, the Publisher returned by {@code switchOnNext} begins emitting the items
     * emitted by that Publisher. When a new Publisher is emitted, {@code switchOnNext} stops emitting items
     * from the earlier-emitted Publisher and begins emitting items from the new one.
     * <p>
     * The resulting Publisher completes if both the main Publisher and the last inner Publisher, if any, complete.
     * If the main Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
     * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchOnNextDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the item type
     * @param sources
     *            the source Publisher that emits Publishers
     * @param prefetch
     *            the number of items to prefetch from the inner Publishers
     * @return a Flowable that emits the items emitted by the Publisher most recently emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/switch.html">ReactiveX operators documentation: Switch</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> switchOnNextDelayError(Publisher<? extends Publisher<? extends T>> sources, int prefetch) {
        return fromPublisher(sources).switchMapDelayError(Functions.<Publisher<? extends T>>identity(), prefetch);
    }

    
Returns a Flowable that emits 0L after a specified delay, and then completes. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. If the downstream needs a slower rate it should slow the timer or use something like onBackpressureDrop.
 
Scheduler:
 
timer operates by default on the computation Scheduler.

Params:
delay –  the initial delay before emitting a single 0L
unit –  time units to use for delay
See Also:
ReactiveX operators documentation: Timer
Returns:
a Flowable that emits 0L after a specified delay, and then completes/**
     * Returns a Flowable that emits {@code 0L} after a specified delay, and then completes.
     * <p>
     * <img width="640" height="200" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. If the downstream needs a slower rate
     *      it should slow the timer or use something like {@link #onBackpressureDrop}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code timer} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param delay
     *            the initial delay before emitting a single {@code 0L}
     * @param unit
     *            time units to use for {@code delay}
     * @return a Flowable that emits {@code 0L} after a specified delay, and then completes
     * @see <a href="http://reactivex.io/documentation/operators/timer.html">ReactiveX operators documentation: Timer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public static Flowable<Long> timer(long delay, TimeUnit unit) {
        return timer(delay, unit, Schedulers.computation());
    }

    
Returns a Flowable that emits 0L after a specified delay, on a specified Scheduler, and then completes. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. If the downstream needs a slower rate it should slow the timer or use something like onBackpressureDrop.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
delay – 
           the initial delay before emitting a single 0L
unit –  time units to use for delay
scheduler –  the Scheduler to use for scheduling the item
See Also:
ReactiveX operators documentation: Timer
Returns:
a Flowable that emits 0L after a specified delay, on a specified Scheduler, and then completes/**
     * Returns a Flowable that emits {@code 0L} after a specified delay, on a specified Scheduler, and then
     * completes.
     * <p>
     * <img width="640" height="200" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timer.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. If the downstream needs a slower rate
     *      it should slow the timer or use something like {@link #onBackpressureDrop}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param delay
     *            the initial delay before emitting a single 0L
     * @param unit
     *            time units to use for {@code delay}
     * @param scheduler
     *            the {@link Scheduler} to use for scheduling the item
     * @return a Flowable that emits {@code 0L} after a specified delay, on a specified Scheduler, and then
     *         completes
     * @see <a href="http://reactivex.io/documentation/operators/timer.html">ReactiveX operators documentation: Timer</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public static Flowable<Long> timer(long delay, TimeUnit unit, Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");

        return RxJavaPlugins.onAssembly(new FlowableTimer(Math.max(0L, delay), unit, scheduler));
    }

    
Create a Flowable by wrapping a Publisher which has to be implemented according
to the Reactive Streams specification by handling backpressure and
cancellation correctly; no safeguards are provided by the Flowable itself.

 
Backpressure:
 
This operator is a pass-through for backpressure and the behavior is determined by the
 provided Publisher implementation.
 
Scheduler:
 
unsafeCreate by default doesn't operate on any particular Scheduler.

Params:
onSubscribe – the Publisher instance to wrap
Type parameters:
<T> – the value type emitted
Throws:
IllegalArgumentException – if onSubscribe is a subclass of Flowable; such instances don't need conversion and is possibly a port remnant from 1.x or one should use hide() instead.
Returns:
the new Flowable instance/**
     * Create a Flowable by wrapping a Publisher <em>which has to be implemented according
     * to the Reactive Streams specification by handling backpressure and
     * cancellation correctly; no safeguards are provided by the Flowable itself</em>.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator is a pass-through for backpressure and the behavior is determined by the
     *  provided Publisher implementation.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code unsafeCreate} by default doesn't operate on any particular {@link Scheduler}.</dd>
     * </dl>
     * @param <T> the value type emitted
     * @param onSubscribe the Publisher instance to wrap
     * @return the new Flowable instance
     * @throws IllegalArgumentException if {@code onSubscribe} is a subclass of {@code Flowable}; such
     * instances don't need conversion and is possibly a port remnant from 1.x or one should use {@link #hide()}
     * instead.
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.NONE)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T> Flowable<T> unsafeCreate(Publisher<T> onSubscribe) {
        ObjectHelper.requireNonNull(onSubscribe, "onSubscribe is null");
        if (onSubscribe instanceof Flowable) {
            throw new IllegalArgumentException("unsafeCreate(Flowable) should be upgraded");
        }
        return RxJavaPlugins.onAssembly(new FlowableFromPublisher<T>(onSubscribe));
    }

    
Constructs a Publisher that creates a dependent resource object which is disposed of on cancellation.



 
Backpressure:
 
The operator is a pass-through for backpressure and otherwise depends on the backpressure support of the Publisher returned by the resourceFactory.
 
Scheduler:
 
using does not operate by default on a particular Scheduler.

Params:
resourceSupplier – 
           the factory function to create a resource object that depends on the Publisher
sourceSupplier – 
           the factory function to create a Publisher
resourceDisposer – 
           the function that will dispose of the resource
Type parameters:
<T> – the element type of the generated Publisher
<D> – the type of the resource associated with the output sequence
See Also:
ReactiveX operators documentation: Using
Returns:
the Publisher whose lifetime controls the lifetime of the dependent resource object/**
     * Constructs a Publisher that creates a dependent resource object which is disposed of on cancellation.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/using.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator is a pass-through for backpressure and otherwise depends on the
     *  backpressure support of the Publisher returned by the {@code resourceFactory}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code using} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the element type of the generated Publisher
     * @param <D> the type of the resource associated with the output sequence
     * @param resourceSupplier
     *            the factory function to create a resource object that depends on the Publisher
     * @param sourceSupplier
     *            the factory function to create a Publisher
     * @param resourceDisposer
     *            the function that will dispose of the resource
     * @return the Publisher whose lifetime controls the lifetime of the dependent resource object
     * @see <a href="http://reactivex.io/documentation/operators/using.html">ReactiveX operators documentation: Using</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, D> Flowable<T> using(Callable<? extends D> resourceSupplier,
            Function<? super D, ? extends Publisher<? extends T>> sourceSupplier, Consumer<? super D> resourceDisposer) {
        return using(resourceSupplier, sourceSupplier, resourceDisposer, true);
    }

    
Constructs a Publisher that creates a dependent resource object which is disposed of just before termination if you have set disposeEagerly to true and cancellation does not occur before termination. Otherwise, resource disposal will occur on cancellation. Eager disposal is particularly appropriate for a synchronous Publisher that reuses resources. disposeAction will only be called once per subscription. 



 
Backpressure:
 
The operator is a pass-through for backpressure and otherwise depends on the backpressure support of the Publisher returned by the resourceFactory.
 
Scheduler:
 
using does not operate by default on a particular Scheduler.

Params:
resourceSupplier – 
           the factory function to create a resource object that depends on the Publisher
sourceSupplier – 
           the factory function to create a Publisher
resourceDisposer – 
           the function that will dispose of the resource
eager –  if true then disposal will happen either on cancellation or just before emission of a terminal event (onComplete or onError).
Type parameters:
<T> – the element type of the generated Publisher
<D> – the type of the resource associated with the output sequence
See Also:
ReactiveX operators documentation: Using
Returns:
the Publisher whose lifetime controls the lifetime of the dependent resource object
Since:
2.0/**
     * Constructs a Publisher that creates a dependent resource object which is disposed of just before
     * termination if you have set {@code disposeEagerly} to {@code true} and cancellation does not occur
     * before termination. Otherwise, resource disposal will occur on cancellation.  Eager disposal is
     * particularly appropriate for a synchronous Publisher that reuses resources. {@code disposeAction} will
     * only be called once per subscription.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/using.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator is a pass-through for backpressure and otherwise depends on the
     *  backpressure support of the Publisher returned by the {@code resourceFactory}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code using} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the element type of the generated Publisher
     * @param <D> the type of the resource associated with the output sequence
     * @param resourceSupplier
     *            the factory function to create a resource object that depends on the Publisher
     * @param sourceSupplier
     *            the factory function to create a Publisher
     * @param resourceDisposer
     *            the function that will dispose of the resource
     * @param eager
     *            if {@code true} then disposal will happen either on cancellation or just before emission of
     *            a terminal event ({@code onComplete} or {@code onError}).
     * @return the Publisher whose lifetime controls the lifetime of the dependent resource object
     * @see <a href="http://reactivex.io/documentation/operators/using.html">ReactiveX operators documentation: Using</a>
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, D> Flowable<T> using(Callable<? extends D> resourceSupplier,
            Function<? super D, ? extends Publisher<? extends T>> sourceSupplier,
                    Consumer<? super D> resourceDisposer, boolean eager) {
        ObjectHelper.requireNonNull(resourceSupplier, "resourceSupplier is null");
        ObjectHelper.requireNonNull(sourceSupplier, "sourceSupplier is null");
        ObjectHelper.requireNonNull(resourceDisposer, "resourceDisposer is null");
        return RxJavaPlugins.onAssembly(new FlowableUsing<T, D>(resourceSupplier, sourceSupplier, resourceDisposer, eager));
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
items emitted, in sequence, by an Iterable of other Publishers.
 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each of the source Publishers; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(Arrays.asList(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -> a)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 



 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
sources – 
           an Iterable of source Publishers
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T> – the common value type
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * items emitted, in sequence, by an Iterable of other Publishers.
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each of the source Publishers;
     * the second item emitted by the new Publisher will be the result of the function applied to the second
     * item emitted by each of those Publishers; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@code onNext} as many times as
     * the number of {@code onNext} invocations of the source Publisher that emits the fewest items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(Arrays.asList(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -&gt; a)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common value type
     * @param <R> the zipped result type
     * @param sources
     *            an Iterable of source Publishers
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, R> Flowable<R> zip(Iterable<? extends Publisher<? extends T>> sources, Function<? super Object[], ? extends R> zipper) {
        ObjectHelper.requireNonNull(zipper, "zipper is null");
        ObjectHelper.requireNonNull(sources, "sources is null");
        return RxJavaPlugins.onAssembly(new FlowableZip<T, R>(null, sources, zipper, bufferSize(), false));
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
n items emitted, in sequence, by the n Publishers emitted by a specified Publisher.
 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each of the Publishers emitted by the source Publisher; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while cancel the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(just(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -> a)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 



 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
sources – 
           a Publisher of source Publishers
zipper –  a function that, when applied to an item emitted by each of the Publishers emitted by ws, results in an item that will be emitted by the resulting Publisher
Type parameters:
<T> – the value type of the inner Publishers
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * <i>n</i> items emitted, in sequence, by the <i>n</i> Publishers emitted by a specified Publisher.
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each of the Publishers emitted
     * by the source Publisher; the second item emitted by the new Publisher will be the result of the
     * function applied to the second item emitted by each of those Publishers; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@code onNext} as many times as
     * the number of {@code onNext} invocations of the source Publisher that emits the fewest items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while cancel the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(just(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -&gt; a)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.o.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the value type of the inner Publishers
     * @param <R> the zipped result type
     * @param sources
     *            a Publisher of source Publishers
     * @param zipper
     *            a function that, when applied to an item emitted by each of the Publishers emitted by
     *            {@code ws}, results in an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings({ "rawtypes", "unchecked", "cast" })
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, R> Flowable<R> zip(Publisher<? extends Publisher<? extends T>> sources,
            final Function<? super Object[], ? extends R> zipper) {
        ObjectHelper.requireNonNull(zipper, "zipper is null");
        return fromPublisher(sources).toList().flatMapPublisher((Function)FlowableInternalHelper.<T, R>zipIterable(zipper));
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
two items emitted, in sequence, by two other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by o1 and the first item emitted by o2; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by o1 and the second item emitted by o2; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results
           in an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * two items emitted, in sequence, by two other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by {@code o1} and the first item
     * emitted by {@code o2}; the second item emitted by the new Publisher will be the result of the function
     * applied to the second item emitted by {@code o1} and the second item emitted by {@code o2}; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results
     *            in an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            BiFunction<? super T1, ? super T2, ? extends R> zipper) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
two items emitted, in sequence, by two other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by o1 and the first item emitted by o2; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by o1 and the second item emitted by o2; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results
           in an item that will be emitted by the resulting Publisher
delayError – delay errors from any of the source Publishers till the other terminates
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * two items emitted, in sequence, by two other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by {@code o1} and the first item
     * emitted by {@code o2}; the second item emitted by the new Publisher will be the result of the function
     * applied to the second item emitted by {@code o1} and the second item emitted by {@code o2}; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results
     *            in an item that will be emitted by the resulting Publisher
     * @param delayError delay errors from any of the source Publishers till the other terminates
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            BiFunction<? super T1, ? super T2, ? extends R> zipper, boolean delayError) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        return zipArray(Functions.toFunction(zipper), delayError, bufferSize(), source1, source2);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
two items emitted, in sequence, by two other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by o1 and the first item emitted by o2; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by o1 and the second item emitted by o2; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results
           in an item that will be emitted by the resulting Publisher
delayError – delay errors from any of the source Publishers till the other terminates
bufferSize – the number of elements to prefetch from each source Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * two items emitted, in sequence, by two other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by {@code o1} and the first item
     * emitted by {@code o2}; the second item emitted by the new Publisher will be the result of the function
     * applied to the second item emitted by {@code o1} and the second item emitted by {@code o2}; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), (a, b) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results
     *            in an item that will be emitted by the resulting Publisher
     * @param delayError delay errors from any of the source Publishers till the other terminates
     * @param bufferSize the number of elements to prefetch from each source Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2,
            BiFunction<? super T1, ? super T2, ? extends R> zipper, boolean delayError, int bufferSize) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        return zipArray(Functions.toFunction(zipper), delayError, bufferSize, source1, source2);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
three items emitted, in sequence, by three other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by o1, the first item emitted by o2, and the first item emitted by o3; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by o1, the second item emitted by o2, and the second item emitted by o3; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
source3 – 
           a third source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<T3> – the value type of the third source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * three items emitted, in sequence, by three other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by {@code o1}, the first item
     * emitted by {@code o2}, and the first item emitted by {@code o3}; the second item emitted by the new
     * Publisher will be the result of the function applied to the second item emitted by {@code o1}, the
     * second item emitted by {@code o2}, and the second item emitted by {@code o3}; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <T3> the value type of the third source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param source3
     *            a third source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2, Publisher<? extends T3> source3,
            Function3<? super T1, ? super T2, ? super T3, ? extends R> zipper) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
four items emitted, in sequence, by four other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by o1, the first item emitted by o2, the first item emitted by o3, and the first item emitted by 04; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
source3 – 
           a third source Publisher
source4 – 
           a fourth source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<T3> – the value type of the third source
<T4> – the value type of the fourth source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * four items emitted, in sequence, by four other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by {@code o1}, the first item
     * emitted by {@code o2}, the first item emitted by {@code o3}, and the first item emitted by {@code 04};
     * the second item emitted by the new Publisher will be the result of the function applied to the second
     * item emitted by each of those Publishers; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <T3> the value type of the third source
     * @param <T4> the value type of the fourth source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param source3
     *            a third source Publisher
     * @param source4
     *            a fourth source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2, Publisher<? extends T3> source3,
            Publisher<? extends T4> source4,
            Function4<? super T1, ? super T2, ? super T3, ? super T4, ? extends R> zipper) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
five items emitted, in sequence, by five other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by o1, the first item emitted by o2, the first item emitted by o3, the first item emitted by o4, and the first item emitted by o5; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
source3 – 
           a third source Publisher
source4 – 
           a fourth source Publisher
source5 – 
           a fifth source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<T3> – the value type of the third source
<T4> – the value type of the fourth source
<T5> – the value type of the fifth source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * five items emitted, in sequence, by five other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by {@code o1}, the first item
     * emitted by {@code o2}, the first item emitted by {@code o3}, the first item emitted by {@code o4}, and
     * the first item emitted by {@code o5}; the second item emitted by the new Publisher will be the result of
     * the function applied to the second item emitted by each of those Publishers; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <T3> the value type of the third source
     * @param <T4> the value type of the fourth source
     * @param <T5> the value type of the fifth source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param source3
     *            a third source Publisher
     * @param source4
     *            a fourth source Publisher
     * @param source5
     *            a fifth source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2, Publisher<? extends T3> source3,
            Publisher<? extends T4> source4, Publisher<? extends T5> source5,
            Function5<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? extends R> zipper) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
six items emitted, in sequence, by six other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each source Publisher, the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers, and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
source3 – 
           a third source Publisher
source4 – 
           a fourth source Publisher
source5 – 
           a fifth source Publisher
source6 – 
           a sixth source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<T3> – the value type of the third source
<T4> – the value type of the fourth source
<T5> – the value type of the fifth source
<T6> – the value type of the sixth source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * six items emitted, in sequence, by six other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each source Publisher, the
     * second item emitted by the new Publisher will be the result of the function applied to the second item
     * emitted by each of those Publishers, and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <T3> the value type of the third source
     * @param <T4> the value type of the fourth source
     * @param <T5> the value type of the fifth source
     * @param <T6> the value type of the sixth source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param source3
     *            a third source Publisher
     * @param source4
     *            a fourth source Publisher
     * @param source5
     *            a fifth source Publisher
     * @param source6
     *            a sixth source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2, Publisher<? extends T3> source3,
            Publisher<? extends T4> source4, Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Function6<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? extends R> zipper) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
seven items emitted, in sequence, by seven other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each source Publisher, the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers, and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
source3 – 
           a third source Publisher
source4 – 
           a fourth source Publisher
source5 – 
           a fifth source Publisher
source6 – 
           a sixth source Publisher
source7 – 
           a seventh source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<T3> – the value type of the third source
<T4> – the value type of the fourth source
<T5> – the value type of the fifth source
<T6> – the value type of the sixth source
<T7> – the value type of the seventh source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * seven items emitted, in sequence, by seven other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each source Publisher, the
     * second item emitted by the new Publisher will be the result of the function applied to the second item
     * emitted by each of those Publishers, and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <T3> the value type of the third source
     * @param <T4> the value type of the fourth source
     * @param <T5> the value type of the fifth source
     * @param <T6> the value type of the sixth source
     * @param <T7> the value type of the seventh source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param source3
     *            a third source Publisher
     * @param source4
     *            a fourth source Publisher
     * @param source5
     *            a fifth source Publisher
     * @param source6
     *            a sixth source Publisher
     * @param source7
     *            a seventh source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, T7, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2, Publisher<? extends T3> source3,
            Publisher<? extends T4> source4, Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Publisher<? extends T7> source7,
            Function7<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? super T7, ? extends R> zipper) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        ObjectHelper.requireNonNull(source7, "source7 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6, source7);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
eight items emitted, in sequence, by eight other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each source Publisher, the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers, and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g, h) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
source3 – 
           a third source Publisher
source4 – 
           a fourth source Publisher
source5 – 
           a fifth source Publisher
source6 – 
           a sixth source Publisher
source7 – 
           a seventh source Publisher
source8 – 
           an eighth source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<T3> – the value type of the third source
<T4> – the value type of the fourth source
<T5> – the value type of the fifth source
<T6> – the value type of the sixth source
<T7> – the value type of the seventh source
<T8> – the value type of the eighth source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * eight items emitted, in sequence, by eight other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each source Publisher, the
     * second item emitted by the new Publisher will be the result of the function applied to the second item
     * emitted by each of those Publishers, and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g, h) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <T3> the value type of the third source
     * @param <T4> the value type of the fourth source
     * @param <T5> the value type of the fifth source
     * @param <T6> the value type of the sixth source
     * @param <T7> the value type of the seventh source
     * @param <T8> the value type of the eighth source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param source3
     *            a third source Publisher
     * @param source4
     *            a fourth source Publisher
     * @param source5
     *            a fifth source Publisher
     * @param source6
     *            a sixth source Publisher
     * @param source7
     *            a seventh source Publisher
     * @param source8
     *            an eighth source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, T7, T8, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2, Publisher<? extends T3> source3,
            Publisher<? extends T4> source4, Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Publisher<? extends T7> source7, Publisher<? extends T8> source8,
            Function8<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? super T7, ? super T8, ? extends R> zipper) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        ObjectHelper.requireNonNull(source7, "source7 is null");
        ObjectHelper.requireNonNull(source8, "source8 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6, source7, source8);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
nine items emitted, in sequence, by nine other Publishers.


 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each source Publisher, the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers, and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g, h, i) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zip does not operate by default on a particular Scheduler.

Params:
source1 – 
           the first source Publisher
source2 – 
           a second source Publisher
source3 – 
           a third source Publisher
source4 – 
           a fourth source Publisher
source5 – 
           a fifth source Publisher
source6 – 
           a sixth source Publisher
source7 – 
           a seventh source Publisher
source8 – 
           an eighth source Publisher
source9 – 
           a ninth source Publisher
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
Type parameters:
<T1> – the value type of the first source
<T2> – the value type of the second source
<T3> – the value type of the third source
<T4> – the value type of the fourth source
<T5> – the value type of the fifth source
<T6> – the value type of the sixth source
<T7> – the value type of the seventh source
<T8> – the value type of the eighth source
<T9> – the value type of the ninth source
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * nine items emitted, in sequence, by nine other Publishers.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each source Publisher, the
     * second item emitted by the new Publisher will be the result of the function applied to the second item
     * emitted by each of those Publishers, and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@link Subscriber#onNext onNext}
     * as many times as the number of {@code onNext} invocations of the source Publisher that emits the fewest
     * items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2), ..., (a, b, c, d, e, f, g, h, i) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the value type of the first source
     * @param <T2> the value type of the second source
     * @param <T3> the value type of the third source
     * @param <T4> the value type of the fourth source
     * @param <T5> the value type of the fifth source
     * @param <T6> the value type of the sixth source
     * @param <T7> the value type of the seventh source
     * @param <T8> the value type of the eighth source
     * @param <T9> the value type of the ninth source
     * @param <R> the zipped result type
     * @param source1
     *            the first source Publisher
     * @param source2
     *            a second source Publisher
     * @param source3
     *            a third source Publisher
     * @param source4
     *            a fourth source Publisher
     * @param source5
     *            a fifth source Publisher
     * @param source6
     *            a sixth source Publisher
     * @param source7
     *            a seventh source Publisher
     * @param source8
     *            an eighth source Publisher
     * @param source9
     *            a ninth source Publisher
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T1, T2, T3, T4, T5, T6, T7, T8, T9, R> Flowable<R> zip(
            Publisher<? extends T1> source1, Publisher<? extends T2> source2, Publisher<? extends T3> source3,
            Publisher<? extends T4> source4, Publisher<? extends T5> source5, Publisher<? extends T6> source6,
            Publisher<? extends T7> source7, Publisher<? extends T8> source8, Publisher<? extends T9> source9,
            Function9<? super T1, ? super T2, ? super T3, ? super T4, ? super T5, ? super T6, ? super T7, ? super T8, ? super T9, ? extends R> zipper) {

        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        ObjectHelper.requireNonNull(source5, "source5 is null");
        ObjectHelper.requireNonNull(source6, "source6 is null");
        ObjectHelper.requireNonNull(source7, "source7 is null");
        ObjectHelper.requireNonNull(source8, "source8 is null");
        ObjectHelper.requireNonNull(source9, "source9 is null");
        return zipArray(Functions.toFunction(zipper), false, bufferSize(), source1, source2, source3, source4, source5, source6, source7, source8, source9);
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
items emitted, in sequence, by an array of other Publishers.
 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each of the source Publishers; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(new Publisher[]{range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)}, (a) ->
a)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 



 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zipArray does not operate by default on a particular Scheduler.

Params:
sources – 
           an array of source Publishers
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
delayError – 
           delay errors signaled by any of the source Publisher until all Publishers terminate
bufferSize – 
           the number of elements to prefetch from each source Publisher
Type parameters:
<T> – the common element type
<R> – the result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * items emitted, in sequence, by an array of other Publishers.
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each of the source Publishers;
     * the second item emitted by the new Publisher will be the result of the function applied to the second
     * item emitted by each of those Publishers; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@code onNext} as many times as
     * the number of {@code onNext} invocations of the source Publisher that emits the fewest items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(new Publisher[]{range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)}, (a) -&gt;
     * a)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zipArray} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T> the common element type
     * @param <R> the result type
     * @param sources
     *            an array of source Publishers
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @param delayError
     *            delay errors signaled by any of the source Publisher until all Publishers terminate
     * @param bufferSize
     *            the number of elements to prefetch from each source Publisher
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, R> Flowable<R> zipArray(Function<? super Object[], ? extends R> zipper,
            boolean delayError, int bufferSize, Publisher<? extends T>... sources) {
        if (sources.length == 0) {
            return empty();
        }
        ObjectHelper.requireNonNull(zipper, "zipper is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableZip<T, R>(sources, null, zipper, bufferSize, delayError));
    }

    
Returns a Flowable that emits the results of a specified combiner function applied to combinations of
items emitted, in sequence, by an Iterable of other Publishers.
 zip applies this function in strict sequence, so the first item emitted by the new Publisher will be the result of the function applied to the first item emitted by each of the source Publishers; the second item emitted by the new Publisher will be the result of the function applied to the second item emitted by each of those Publishers; and so forth. 
 The resulting Publisher<R> returned from zip will invoke onNext as many times as the number of onNext invocations of the source Publisher that emits the fewest items. 
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
zip(Arrays.asList(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -> a)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 



 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zipIterable does not operate by default on a particular Scheduler.

Params:
sources – 
           an Iterable of source Publishers
zipper – 
           a function that, when applied to an item emitted by each of the source Publishers, results in
           an item that will be emitted by the resulting Publisher
delayError – 
           delay errors signaled by any of the source Publisher until all Publishers terminate
bufferSize – 
           the number of elements to prefetch from each source Publisher
Type parameters:
<T> – the common source value type
<R> – the zipped result type
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that emits the zipped results/**
     * Returns a Flowable that emits the results of a specified combiner function applied to combinations of
     * items emitted, in sequence, by an Iterable of other Publishers.
     * <p>
     * {@code zip} applies this function in strict sequence, so the first item emitted by the new Publisher
     * will be the result of the function applied to the first item emitted by each of the source Publishers;
     * the second item emitted by the new Publisher will be the result of the function applied to the second
     * item emitted by each of those Publishers; and so forth.
     * <p>
     * The resulting {@code Publisher<R>} returned from {@code zip} will invoke {@code onNext} as many times as
     * the number of {@code onNext} invocations of the source Publisher that emits the fewest items.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>zip(Arrays.asList(range(1, 5).doOnComplete(action1), range(6, 5).doOnComplete(action2)), (a) -&gt; a)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zipIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     *
     * @param sources
     *            an Iterable of source Publishers
     * @param zipper
     *            a function that, when applied to an item emitted by each of the source Publishers, results in
     *            an item that will be emitted by the resulting Publisher
     * @param delayError
     *            delay errors signaled by any of the source Publisher until all Publishers terminate
     * @param bufferSize
     *            the number of elements to prefetch from each source Publisher
     * @param <T> the common source value type
     * @param <R> the zipped result type
     * @return a Flowable that emits the zipped results
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public static <T, R> Flowable<R> zipIterable(Iterable<? extends Publisher<? extends T>> sources,
            Function<? super Object[], ? extends R> zipper, boolean delayError,
            int bufferSize) {
        ObjectHelper.requireNonNull(zipper, "zipper is null");
        ObjectHelper.requireNonNull(sources, "sources is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableZip<T, R>(null, sources, zipper, bufferSize, delayError));
    }

    // ***************************************************************************************************
    // Instance operators
    // ***************************************************************************************************

    
Returns a Single that emits a Boolean that indicates whether all of the items emitted by the source
Publisher satisfy a condition.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
all does not operate by default on a particular Scheduler.

Params:
predicate – 
           a function that evaluates an item and returns a Boolean
See Also:
ReactiveX operators documentation: All
Returns:
a Single that emits true if all items emitted by the source Publisher satisfy the predicate; otherwise, false/**
     * Returns a Single that emits a Boolean that indicates whether all of the items emitted by the source
     * Publisher satisfy a condition.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/all.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code all} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param predicate
     *            a function that evaluates an item and returns a Boolean
     * @return a Single that emits {@code true} if all items emitted by the source Publisher satisfy the
     *         predicate; otherwise, {@code false}
     * @see <a href="http://reactivex.io/documentation/operators/all.html">ReactiveX operators documentation: All</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<Boolean> all(Predicate<? super T> predicate) {
        ObjectHelper.requireNonNull(predicate, "predicate is null");
        return RxJavaPlugins.onAssembly(new FlowableAllSingle<T>(this, predicate));
    }

    
Mirrors the Publisher (current or provided) that first either emits an item or sends a termination
notification.



 
Backpressure:
 
The operator itself doesn't interfere with backpressure which is determined by the winning Publisher's backpressure behavior.
 
Scheduler:
 
ambWith does not operate by default on a particular Scheduler.

Params:
other – 
           a Publisher competing to react first. A subscription to this provided Publisher will occur after subscribing
           to the current Publisher.
See Also:
ReactiveX operators documentation: Amb
Returns:
a Flowable that emits the same sequence as whichever of the source Publishers first
        emitted an item or sent a termination notification/**
     * Mirrors the Publisher (current or provided) that first either emits an item or sends a termination
     * notification.
     * <p>
     * <img width="640" height="385" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/amb.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator itself doesn't interfere with backpressure which is determined by the winning
     *  {@code Publisher}'s backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code ambWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param other
     *            a Publisher competing to react first. A subscription to this provided Publisher will occur after subscribing
     *            to the current Publisher.
     * @return a Flowable that emits the same sequence as whichever of the source Publishers first
     *         emitted an item or sent a termination notification
     * @see <a href="http://reactivex.io/documentation/operators/amb.html">ReactiveX operators documentation: Amb</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> ambWith(Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return ambArray(this, other);
    }

    
Returns a Single that emits true if any item emitted by the source Publisher satisfies a specified condition, otherwise false. Note: this always emits false if the source Publisher is empty. 


 In Rx.Net this is the any operator but we renamed it in RxJava to better match Java naming idioms. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
any does not operate by default on a particular Scheduler.

Params:
predicate – 
           the condition to test items emitted by the source Publisher
See Also:
ReactiveX operators documentation: Contains
Returns:
a Single that emits a Boolean that indicates whether any item emitted by the source Publisher satisfies the predicate/**
     * Returns a Single that emits {@code true} if any item emitted by the source Publisher satisfies a
     * specified condition, otherwise {@code false}. <em>Note:</em> this always emits {@code false} if the
     * source Publisher is empty.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/exists.png" alt="">
     * <p>
     * In Rx.Net this is the {@code any} operator but we renamed it in RxJava to better match Java naming
     * idioms.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code any} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param predicate
     *            the condition to test items emitted by the source Publisher
     * @return a Single that emits a Boolean that indicates whether any item emitted by the source
     *         Publisher satisfies the {@code predicate}
     * @see <a href="http://reactivex.io/documentation/operators/contains.html">ReactiveX operators documentation: Contains</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<Boolean> any(Predicate<? super T> predicate) {
        ObjectHelper.requireNonNull(predicate, "predicate is null");
        return RxJavaPlugins.onAssembly(new FlowableAnySingle<T>(this, predicate));
    }

    
Calls the specified converter function during assembly time and returns its resulting value.

This allows fluent conversion to any other type.

 
Backpressure:
 
The backpressure behavior depends on what happens in the converter function.
 
Scheduler:
 
as does not operate by default on a particular Scheduler.

History: 2.1.7 - experimental
Params:
converter – the function that receives the current Flowable instance and returns a value
Type parameters:
<R> – the resulting object type
Throws:
NullPointerException – if converter is null
Returns:
the converted value
Since:
2.2/**
     * Calls the specified converter function during assembly time and returns its resulting value.
     * <p>
     * This allows fluent conversion to any other type.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The backpressure behavior depends on what happens in the {@code converter} function.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code as} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.7 - experimental
     * @param <R> the resulting object type
     * @param converter the function that receives the current Flowable instance and returns a value
     * @return the converted value
     * @throws NullPointerException if converter is null
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> R as(@NonNull FlowableConverter<T, ? extends R> converter) {
        return ObjectHelper.requireNonNull(converter, "converter is null").apply(this);
    }

    
Returns the first item emitted by this Flowable, or throws NoSuchElementException if it emits no items. 

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingFirst does not operate by default on a particular Scheduler.
 
Error handling:
 
If the source signals an error, the operator wraps a checked Exception into RuntimeException and throws that. Otherwise, RuntimeExceptions and Errors are rethrown as they are.

Throws:
NoSuchElementException –  if this Flowable emits no items
See Also:
ReactiveX documentation: First
Returns:
the first item emitted by this Flowable/**
     * Returns the first item emitted by this {@code Flowable}, or throws
     * {@code NoSuchElementException} if it emits no items.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingFirst} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If the source signals an error, the operator wraps a checked {@link Exception}
     *  into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and
     *  {@link Error}s are rethrown as they are.</dd>
     * </dl>
     *
     * @return the first item emitted by this {@code Flowable}
     * @throws NoSuchElementException
     *             if this {@code Flowable} emits no items
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final T blockingFirst() {
        BlockingFirstSubscriber<T> s = new BlockingFirstSubscriber<T>();
        subscribe(s);
        T v = s.blockingGet();
        if (v != null) {
            return v;
        }
        throw new NoSuchElementException();
    }

    
Returns the first item emitted by this Flowable, or a default value if it emits no items. 

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingFirst does not operate by default on a particular Scheduler.
 
Error handling:
 
If the source signals an error, the operator wraps a checked Exception into RuntimeException and throws that. Otherwise, RuntimeExceptions and Errors are rethrown as they are.

Params:
defaultItem –  a default value to return if this Flowable emits no items
See Also:
ReactiveX documentation: First
Returns:
the first item emitted by this Flowable, or the default value if it emits no items/**
     * Returns the first item emitted by this {@code Flowable}, or a default value if it emits no
     * items.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingFirst} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If the source signals an error, the operator wraps a checked {@link Exception}
     *  into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and
     *  {@link Error}s are rethrown as they are.</dd>
     * </dl>
     *
     * @param defaultItem
     *            a default value to return if this {@code Flowable} emits no items
     * @return the first item emitted by this {@code Flowable}, or the default value if it emits no
     *         items
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final T blockingFirst(T defaultItem) {
        BlockingFirstSubscriber<T> s = new BlockingFirstSubscriber<T>();
        subscribe(s);
        T v = s.blockingGet();
        return v != null ? v : defaultItem;
    }

    
Consumes the upstream Flowable in a blocking fashion and invokes the given Consumer with each upstream item on the current thread until the
upstream terminates.



Note: the method will only return if the upstream terminates or the current
thread is interrupted.
 This method executes the Consumer on the current thread while subscribe(Consumer) executes the consumer on the original caller thread of the sequence. 

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingForEach does not operate by default on a particular Scheduler.
 
Error handling:
 
If the source signals an error, the operator wraps a checked Exception into RuntimeException and throws that. Otherwise, RuntimeExceptions and Errors are rethrown as they are.

Params:
onNext –  the Consumer to invoke for each item emitted by the Flowable
Throws:
RuntimeException – 
            if an error occurs
See Also:
ReactiveX documentation: Subscribe
subscribe(Consumer)/**
     * Consumes the upstream {@code Flowable} in a blocking fashion and invokes the given
     * {@code Consumer} with each upstream item on the <em>current thread</em> until the
     * upstream terminates.
     * <p>
     * <img width="640" height="330" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.forEach.png" alt="">
     * <p>
     * <em>Note:</em> the method will only return if the upstream terminates or the current
     * thread is interrupted.
     * <p>
     * This method executes the {@code Consumer} on the current thread while
     * {@link #subscribe(Consumer)} executes the consumer on the original caller thread of the
     * sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingForEach} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If the source signals an error, the operator wraps a checked {@link Exception}
     *  into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and
     *  {@link Error}s are rethrown as they are.</dd>
     * </dl>
     *
     * @param onNext
     *            the {@link Consumer} to invoke for each item emitted by the {@code Flowable}
     * @throws RuntimeException
     *             if an error occurs
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX documentation: Subscribe</a>
     * @see #subscribe(Consumer)
     */
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingForEach(Consumer<? super T> onNext) {
        Iterator<T> it = blockingIterable().iterator();
        while (it.hasNext()) {
            try {
                onNext.accept(it.next());
            } catch (Throwable e) {
                Exceptions.throwIfFatal(e);
                ((Disposable)it).dispose();
                throw ExceptionHelper.wrapOrThrow(e);
            }
        }
    }

    
Converts this Flowable into an Iterable. 



 
Backpressure:
 
The operator expects the upstream to honor backpressure otherwise the returned Iterable's iterator will throw a MissingBackpressureException.
 
Scheduler:
 
blockingIterable does not operate by default on a particular Scheduler.

See Also:
ReactiveX documentation: To
Returns:
an Iterable version of this Flowable/**
     * Converts this {@code Flowable} into an {@link Iterable}.
     * <p>
     * <img width="640" height="315" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.toIterable.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to honor backpressure otherwise the returned
     *  Iterable's iterator will throw a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return an {@link Iterable} version of this {@code Flowable}
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Iterable<T> blockingIterable() {
        return blockingIterable(bufferSize());
    }

    
Converts this Flowable into an Iterable. 



 
Backpressure:
 
The operator expects the upstream to honor backpressure otherwise the returned Iterable's iterator will throw a MissingBackpressureException. 
 
Scheduler:
 
blockingIterable does not operate by default on a particular Scheduler.

Params:
bufferSize – the number of items to prefetch from the current Flowable
See Also:
ReactiveX documentation: To
Returns:
an Iterable version of this Flowable/**
     * Converts this {@code Flowable} into an {@link Iterable}.
     * <p>
     * <img width="640" height="315" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.toIterable.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to honor backpressure otherwise the returned
     *  Iterable's iterator will throw a {@code MissingBackpressureException}.
     *  </dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param bufferSize the number of items to prefetch from the current Flowable
     * @return an {@link Iterable} version of this {@code Flowable}
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Iterable<T> blockingIterable(int bufferSize) {
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return new BlockingFlowableIterable<T>(this, bufferSize);
    }

    
Returns the last item emitted by this Flowable, or throws NoSuchElementException if this Flowable emits no items. 



 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingLast does not operate by default on a particular Scheduler.
 
Error handling:
 
If the source signals an error, the operator wraps a checked Exception into RuntimeException and throws that. Otherwise, RuntimeExceptions and Errors are rethrown as they are.

Throws:
NoSuchElementException –  if this Flowable emits no items
See Also:
ReactiveX documentation: Last
Returns:
the last item emitted by this Flowable/**
     * Returns the last item emitted by this {@code Flowable}, or throws
     * {@code NoSuchElementException} if this {@code Flowable} emits no items.
     * <p>
     * <img width="640" height="315" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.last.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingLast} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If the source signals an error, the operator wraps a checked {@link Exception}
     *  into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and
     *  {@link Error}s are rethrown as they are.</dd>
     * </dl>
     *
     * @return the last item emitted by this {@code Flowable}
     * @throws NoSuchElementException
     *             if this {@code Flowable} emits no items
     * @see <a href="http://reactivex.io/documentation/operators/last.html">ReactiveX documentation: Last</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final T blockingLast() {
        BlockingLastSubscriber<T> s = new BlockingLastSubscriber<T>();
        subscribe(s);
        T v = s.blockingGet();
        if (v != null) {
            return v;
        }
        throw new NoSuchElementException();
    }

    
Returns the last item emitted by this Flowable, or a default value if it emits no items. 



 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingLast does not operate by default on a particular Scheduler.
 
Error handling:
 
If the source signals an error, the operator wraps a checked Exception into RuntimeException and throws that. Otherwise, RuntimeExceptions and Errors are rethrown as they are.

Params:
defaultItem –  a default value to return if this Flowable emits no items
See Also:
ReactiveX documentation: Last
Returns:
the last item emitted by the Flowable, or the default value if it emits no items/**
     * Returns the last item emitted by this {@code Flowable}, or a default value if it emits no
     * items.
     * <p>
     * <img width="640" height="310" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.lastOrDefault.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingLast} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If the source signals an error, the operator wraps a checked {@link Exception}
     *  into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and
     *  {@link Error}s are rethrown as they are.</dd>
     * </dl>
     *
     * @param defaultItem
     *            a default value to return if this {@code Flowable} emits no items
     * @return the last item emitted by the {@code Flowable}, or the default value if it emits no
     *         items
     * @see <a href="http://reactivex.io/documentation/operators/last.html">ReactiveX documentation: Last</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final T blockingLast(T defaultItem) {
        BlockingLastSubscriber<T> s = new BlockingLastSubscriber<T>();
        subscribe(s);
        T v = s.blockingGet();
        return v != null ? v : defaultItem;
    }

    
Returns an Iterable that returns the latest item emitted by this Flowable, waiting if necessary for one to become available. 
 If this Flowable produces items faster than Iterator.next takes them, onNext events might be skipped, but onError or onComplete events are not. 
 Note also that an onNext directly followed by onComplete might hide the onNext event. 

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingLatest does not operate by default on a particular Scheduler.

See Also:
ReactiveX documentation: First
Returns:
an Iterable that always returns the latest item emitted by this Flowable/**
     * Returns an {@link Iterable} that returns the latest item emitted by this {@code Flowable},
     * waiting if necessary for one to become available.
     * <p>
     * If this {@code Flowable} produces items faster than {@code Iterator.next} takes them,
     * {@code onNext} events might be skipped, but {@code onError} or {@code onComplete} events are not.
     * <p>
     * Note also that an {@code onNext} directly followed by {@code onComplete} might hide the {@code onNext}
     * event.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return an Iterable that always returns the latest item emitted by this {@code Flowable}
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Iterable<T> blockingLatest() {
        return new BlockingFlowableLatest<T>(this);
    }

    
Returns an Iterable that always returns the item most recently emitted by this Flowable. 



 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingMostRecent does not operate by default on a particular Scheduler.

Params:
initialItem –  the initial item that the Iterable sequence will yield if this Flowable has not yet emitted an item
See Also:
ReactiveX documentation: First
Returns:
an Iterable that on each iteration returns the item that this Flowable has most recently emitted/**
     * Returns an {@link Iterable} that always returns the item most recently emitted by this
     * {@code Flowable}.
     * <p>
     * <img width="640" height="490" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.mostRecent.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingMostRecent} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param initialItem
     *            the initial item that the {@link Iterable} sequence will yield if this
     *            {@code Flowable} has not yet emitted an item
     * @return an {@link Iterable} that on each iteration returns the item that this {@code Flowable}
     *         has most recently emitted
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Iterable<T> blockingMostRecent(T initialItem) {
        return new BlockingFlowableMostRecent<T>(this, initialItem);
    }

    
Returns an Iterable that blocks until this Flowable emits another item, then returns that item. 



 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingNext does not operate by default on a particular Scheduler.

See Also:
ReactiveX documentation: TakeLast
Returns:
an Iterable that blocks upon each iteration until this Flowable emits a new item, whereupon the Iterable returns that item/**
     * Returns an {@link Iterable} that blocks until this {@code Flowable} emits another item, then
     * returns that item.
     * <p>
     * <img width="640" height="490" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.next.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingNext} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return an {@link Iterable} that blocks upon each iteration until this {@code Flowable} emits
     *         a new item, whereupon the Iterable returns that item
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Iterable<T> blockingNext() {
        return new BlockingFlowableNext<T>(this);
    }

    
If this Flowable completes after emitting a single item, return that item, otherwise throw a NoSuchElementException. 



 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingSingle does not operate by default on a particular Scheduler.
 
Error handling:
 
If the source signals an error, the operator wraps a checked Exception into RuntimeException and throws that. Otherwise, RuntimeExceptions and Errors are rethrown as they are.

See Also:
ReactiveX documentation: First
Returns:
the single item emitted by this Flowable/**
     * If this {@code Flowable} completes after emitting a single item, return that item, otherwise
     * throw a {@code NoSuchElementException}.
     * <p>
     * <img width="640" height="315" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.single.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSingle} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If the source signals an error, the operator wraps a checked {@link Exception}
     *  into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and
     *  {@link Error}s are rethrown as they are.</dd>
     * </dl>
     *
     * @return the single item emitted by this {@code Flowable}
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final T blockingSingle() {
        return singleOrError().blockingGet();
    }

    
If this Flowable completes after emitting a single item, return that item; if it emits more than one item, throw an IllegalArgumentException; if it emits no items, return a default value. 



 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingSingle does not operate by default on a particular Scheduler.
 
Error handling:
 
If the source signals an error, the operator wraps a checked Exception into RuntimeException and throws that. Otherwise, RuntimeExceptions and Errors are rethrown as they are.

Params:
defaultItem –  a default value to return if this Flowable emits no items
See Also:
ReactiveX documentation: First
Returns:
the single item emitted by this Flowable, or the default value if it emits no items/**
     * If this {@code Flowable} completes after emitting a single item, return that item; if it emits
     * more than one item, throw an {@code IllegalArgumentException}; if it emits no items, return a default
     * value.
     * <p>
     * <img width="640" height="315" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/B.singleOrDefault.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSingle} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If the source signals an error, the operator wraps a checked {@link Exception}
     *  into {@link RuntimeException} and throws that. Otherwise, {@code RuntimeException}s and
     *  {@link Error}s are rethrown as they are.</dd>
     * </dl>
     *
     * @param defaultItem
     *            a default value to return if this {@code Flowable} emits no items
     * @return the single item emitted by this {@code Flowable}, or the default value if it emits no
     *         items
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final T blockingSingle(T defaultItem) {
        return single(defaultItem).blockingGet();
    }

    
Returns a Future representing the only value emitted by this Flowable. 


 If the Flowable emits more than one item, Future will receive an IndexOutOfBoundsException. If the Flowable is empty, Future will receive a NoSuchElementException. The Flowable source has to terminate in order for the returned Future to terminate as well. 
 If the Flowable may emit more than one item, use Flowable.toList().toFuture(). 

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
toFuture does not operate by default on a particular Scheduler.

See Also:
ReactiveX documentation: To
Returns:
a Future that expects a single item to be emitted by this Flowable/**
     * Returns a {@link Future} representing the only value emitted by this {@code Flowable}.
     * <p>
     * <img width="640" height="324" src="https://github.com/ReactiveX/RxJava/wiki/images/rx-operators/Flowable.toFuture.png" alt="">
     * <p>
     * If the {@link Flowable} emits more than one item, {@link java.util.concurrent.Future} will receive an
     * {@link java.lang.IndexOutOfBoundsException}. If the {@link Flowable} is empty, {@link java.util.concurrent.Future}
     * will receive a {@link java.util.NoSuchElementException}. The {@code Flowable} source has to terminate in order
     * for the returned {@code Future} to terminate as well.
     * <p>
     * If the {@code Flowable} may emit more than one item, use {@code Flowable.toList().toFuture()}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toFuture} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a {@link Future} that expects a single item to be emitted by this {@code Flowable}
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Future<T> toFuture() {
        return subscribeWith(new FutureSubscriber<T>());
    }

    
Runs the source Flowable to a terminal event, ignoring any values and rethrowing any exception.

Note that calling this method will block the caller thread until the upstream terminates
normally or with an error. Therefore, calling this method from special threads such as the
Android Main Thread or the Swing Event Dispatch Thread is not recommended.

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

See Also:
blockingSubscribe(Consumer)
blockingSubscribe(Consumer, Consumer)
blockingSubscribe(Consumer, Consumer, Action)
Since:
2.0/**
     * Runs the source Flowable to a terminal event, ignoring any values and rethrowing any exception.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally or with an error. Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @since 2.0
     * @see #blockingSubscribe(Consumer)
     * @see #blockingSubscribe(Consumer, Consumer)
     * @see #blockingSubscribe(Consumer, Consumer, Action)
     */
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe() {
        FlowableBlockingSubscribe.subscribe(this);
    }

    
Subscribes to the source and calls the given callbacks on the current thread.
 If the Flowable emits an error, it is wrapped into an OnErrorNotImplementedException and routed to the RxJavaPlugins.onError handler. Using the overloads blockingSubscribe(Consumer, Consumer) or blockingSubscribe(Consumer, Consumer, Action) instead is recommended. 

Note that calling this method will block the caller thread until the upstream terminates
normally or with an error. Therefore, calling this method from special threads such as the
Android Main Thread or the Swing Event Dispatch Thread is not recommended.

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

Params:
onNext – the callback action for each source value
See Also:
blockingSubscribe(Consumer, Consumer)
blockingSubscribe(Consumer, Consumer, Action)
Since:
2.0/**
     * Subscribes to the source and calls the given callbacks <strong>on the current thread</strong>.
     * <p>
     * If the Flowable emits an error, it is wrapped into an
     * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException}
     * and routed to the RxJavaPlugins.onError handler.
     * Using the overloads {@link #blockingSubscribe(Consumer, Consumer)}
     * or {@link #blockingSubscribe(Consumer, Consumer, Action)} instead is recommended.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally or with an error. Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param onNext the callback action for each source value
     * @since 2.0
     * @see #blockingSubscribe(Consumer, Consumer)
     * @see #blockingSubscribe(Consumer, Consumer, Action)
     */
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe(Consumer<? super T> onNext) {
        FlowableBlockingSubscribe.subscribe(this, onNext, Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION);
    }

    
Subscribes to the source and calls the given callbacks on the current thread.
 If the Flowable emits an error, it is wrapped into an OnErrorNotImplementedException and routed to the RxJavaPlugins.onError handler. Using the overloads blockingSubscribe(Consumer, Consumer) or blockingSubscribe(Consumer, Consumer, Action) instead is recommended. 

Note that calling this method will block the caller thread until the upstream terminates
normally or with an error. Therefore, calling this method from special threads such as the
Android Main Thread or the Swing Event Dispatch Thread is not recommended.

 
Backpressure:
 
The operator consumes the source Flowable in an bounded manner (up to bufferSize outstanding request amount for items).
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

History: 2.1.15 - experimental
Params:
onNext – the callback action for each source value
bufferSize – the size of the buffer
See Also:
blockingSubscribe(Consumer, Consumer)
blockingSubscribe(Consumer, Consumer, Action)
Since:
2.2/**
     * Subscribes to the source and calls the given callbacks <strong>on the current thread</strong>.
     * <p>
     * If the Flowable emits an error, it is wrapped into an
     * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException}
     * and routed to the RxJavaPlugins.onError handler.
     * Using the overloads {@link #blockingSubscribe(Consumer, Consumer)}
     * or {@link #blockingSubscribe(Consumer, Consumer, Action)} instead is recommended.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally or with an error. Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an bounded manner (up to bufferSize
     *  outstanding request amount for items).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.15 - experimental
     * @param onNext the callback action for each source value
     * @param bufferSize the size of the buffer
     * @see #blockingSubscribe(Consumer, Consumer)
     * @see #blockingSubscribe(Consumer, Consumer, Action)
     * @since 2.2
     */
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe(Consumer<? super T> onNext, int bufferSize) {
        FlowableBlockingSubscribe.subscribe(this, onNext, Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION, bufferSize);
    }

    
Subscribes to the source and calls the given callbacks on the current thread.

Note that calling this method will block the caller thread until the upstream terminates
normally or with an error. Therefore, calling this method from special threads such as the
Android Main Thread or the Swing Event Dispatch Thread is not recommended.

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

Params:
onNext – the callback action for each source value
onError – the callback action for an error event
See Also:
blockingSubscribe(Consumer, Consumer, Action)
Since:
2.0/**
     * Subscribes to the source and calls the given callbacks <strong>on the current thread</strong>.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally or with an error. Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param onNext the callback action for each source value
     * @param onError the callback action for an error event
     * @since 2.0
     * @see #blockingSubscribe(Consumer, Consumer, Action)
     */
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe(Consumer<? super T> onNext, Consumer<? super Throwable> onError) {
        FlowableBlockingSubscribe.subscribe(this, onNext, onError, Functions.EMPTY_ACTION);
    }

    
Subscribes to the source and calls the given callbacks on the current thread.

Note that calling this method will block the caller thread until the upstream terminates
normally or with an error. Therefore, calling this method from special threads such as the
Android Main Thread or the Swing Event Dispatch Thread is not recommended.

 
Backpressure:
 
The operator consumes the source Flowable in an bounded manner (up to bufferSize outstanding request amount for items).
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

History: 2.1.15 - experimental
Params:
onNext – the callback action for each source value
onError – the callback action for an error event
bufferSize – the size of the buffer
See Also:
blockingSubscribe(Consumer, Consumer, Action)
Since:
2.2/**
     * Subscribes to the source and calls the given callbacks <strong>on the current thread</strong>.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally or with an error. Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an bounded manner (up to bufferSize
     *  outstanding request amount for items).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.15 - experimental
     * @param onNext the callback action for each source value
     * @param onError the callback action for an error event
     * @param bufferSize the size of the buffer
     * @since 2.2
     * @see #blockingSubscribe(Consumer, Consumer, Action)
     */
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe(Consumer<? super T> onNext, Consumer<? super Throwable> onError,
        int bufferSize) {
        FlowableBlockingSubscribe.subscribe(this, onNext, onError, Functions.EMPTY_ACTION, bufferSize);
    }

    
Subscribes to the source and calls the given callbacks on the current thread.

Note that calling this method will block the caller thread until the upstream terminates
normally or with an error. Therefore, calling this method from special threads such as the
Android Main Thread or the Swing Event Dispatch Thread is not recommended.

 
Backpressure:
 
The operator consumes the source Flowable in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

Params:
onNext – the callback action for each source value
onError – the callback action for an error event
onComplete – the callback action for the completion event.
Since:
2.0/**
     * Subscribes to the source and calls the given callbacks <strong>on the current thread</strong>.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally or with an error. Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param onNext the callback action for each source value
     * @param onError the callback action for an error event
     * @param onComplete the callback action for the completion event.
     * @since 2.0
     */
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe(Consumer<? super T> onNext, Consumer<? super Throwable> onError, Action onComplete) {
        FlowableBlockingSubscribe.subscribe(this, onNext, onError, onComplete);
    }

    
Subscribes to the source and calls the given callbacks on the current thread.

Note that calling this method will block the caller thread until the upstream terminates
normally or with an error. Therefore, calling this method from special threads such as the
Android Main Thread or the Swing Event Dispatch Thread is not recommended.

 
Backpressure:
 
The operator consumes the source Flowable in an bounded manner (up to bufferSize outstanding request amount for items).
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

History: 2.1.15 - experimental
Params:
onNext – the callback action for each source value
onError – the callback action for an error event
onComplete – the callback action for the completion event.
bufferSize – the size of the buffer
Since:
2.2/**
     * Subscribes to the source and calls the given callbacks <strong>on the current thread</strong>.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally or with an error. Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Flowable} in an bounded manner (up to bufferSize
     *  outstanding request amount for items).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.15 - experimental
     * @param onNext the callback action for each source value
     * @param onError the callback action for an error event
     * @param onComplete the callback action for the completion event.
     * @param bufferSize the size of the buffer
     * @since 2.2
     */
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe(Consumer<? super T> onNext, Consumer<? super Throwable> onError, Action onComplete,
        int bufferSize) {
        FlowableBlockingSubscribe.subscribe(this, onNext, onError, onComplete, bufferSize);
    }

    
Subscribes to the source and calls the Subscriber methods on the current thread.
 Note that calling this method will block the caller thread until the upstream terminates normally, with an error or the Subscriber cancels the Subscription it receives via Subscriber.onSubscribe(Subscription). Therefore, calling this method from special threads such as the Android Main Thread or the Swing Event Dispatch Thread is not recommended. 

 
Backpressure:
 
The supplied Subscriber determines how backpressure is applied.
 
Scheduler:
 
blockingSubscribe does not operate by default on a particular Scheduler.

The cancellation and backpressure is composed through.
Params:
subscriber – the subscriber to forward events and calls to in the current thread
Since:
2.0/**
     * Subscribes to the source and calls the {@link Subscriber} methods <strong>on the current thread</strong>.
     * <p>
     * Note that calling this method will block the caller thread until the upstream terminates
     * normally, with an error or the {@code Subscriber} cancels the {@link Subscription} it receives via
     * {@link Subscriber#onSubscribe(Subscription)}.
     * Therefore, calling this method from special threads such as the
     * Android Main Thread or the Swing Event Dispatch Thread is not recommended.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The supplied {@code Subscriber} determines how backpressure is applied.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code blockingSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * The cancellation and backpressure is composed through.
     * @param subscriber the subscriber to forward events and calls to in the current thread
     * @since 2.0
     */
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void blockingSubscribe(Subscriber<? super T> subscriber) {
        FlowableBlockingSubscribe.subscribe(this, subscriber);
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers, each containing count items. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
The operator honors backpressure from downstream and expects the source Publisher to honor it as well, although not enforced; violation may lead to MissingBackpressureException somewhere downstream.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum number of items in each buffer before it should be emitted
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits connected, non-overlapping buffers, each containing at most count items from the source Publisher/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers, each containing {@code count} items. When the source
     * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the
     * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on
     * immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer3.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as
     *  well, although not enforced; violation <em>may</em> lead to {@code MissingBackpressureException} somewhere
     *  downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum number of items in each buffer before it should be emitted
     * @return a Flowable that emits connected, non-overlapping buffers, each containing at most
     *         {@code count} items from the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<List<T>> buffer(int count) {
        return buffer(count, count);
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits buffers every skip items, each containing count items. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
The operator honors backpressure from downstream and expects the source Publisher to honor it as well, although not enforced; violation may lead to MissingBackpressureException somewhere downstream.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum size of each buffer before it should be emitted
skip –  how many items emitted by the source Publisher should be skipped before starting a new buffer. Note that when skip and count are equal, this is the same operation as buffer(int).
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits buffers for every skip item from the source Publisher and containing at most count items/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits buffers every {@code skip} items, each containing {@code count} items. When the source
     * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the
     * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on
     * immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer4.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as
     *  well, although not enforced; violation <em>may</em> lead to {@code MissingBackpressureException} somewhere
     *  downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum size of each buffer before it should be emitted
     * @param skip
     *            how many items emitted by the source Publisher should be skipped before starting a new
     *            buffer. Note that when {@code skip} and {@code count} are equal, this is the same operation as
     *            {@link #buffer(int)}.
     * @return a Flowable that emits buffers for every {@code skip} item from the source Publisher and
     *         containing at most {@code count} items
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<List<T>> buffer(int count, int skip) {
        return buffer(count, skip, ArrayListSupplier.<T>asCallable());
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits buffers every skip items, each containing count items. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
The operator honors backpressure from downstream and expects the source Publisher to honor it as well, although not enforced; violation may lead to MissingBackpressureException somewhere downstream.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum size of each buffer before it should be emitted
skip –  how many items emitted by the source Publisher should be skipped before starting a new buffer. Note that when skip and count are equal, this is the same operation as buffer(int).
bufferSupplier – 
           a factory function that returns an instance of the collection subclass to be used and returned
           as the buffer
Type parameters:
<U> – the collection subclass type to buffer into
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits buffers for every skip item from the source Publisher and containing at most count items/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits buffers every {@code skip} items, each containing {@code count} items. When the source
     * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the
     * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on
     * immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer4.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as
     *  well, although not enforced; violation <em>may</em> lead to {@code MissingBackpressureException} somewhere
     *  downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the collection subclass type to buffer into
     * @param count
     *            the maximum size of each buffer before it should be emitted
     * @param skip
     *            how many items emitted by the source Publisher should be skipped before starting a new
     *            buffer. Note that when {@code skip} and {@code count} are equal, this is the same operation as
     *            {@link #buffer(int)}.
     * @param bufferSupplier
     *            a factory function that returns an instance of the collection subclass to be used and returned
     *            as the buffer
     * @return a Flowable that emits buffers for every {@code skip} item from the source Publisher and
     *         containing at most {@code count} items
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U extends Collection<? super T>> Flowable<U> buffer(int count, int skip, Callable<U> bufferSupplier) {
        ObjectHelper.verifyPositive(count, "count");
        ObjectHelper.verifyPositive(skip, "skip");
        ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableBuffer<T, U>(this, count, skip, bufferSupplier));
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers, each containing count items. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
The operator honors backpressure from downstream and expects the source Publisher to honor it as well, although not enforced; violation may lead to MissingBackpressureException somewhere downstream.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum number of items in each buffer before it should be emitted
bufferSupplier – 
           a factory function that returns an instance of the collection subclass to be used and returned
           as the buffer
Type parameters:
<U> – the collection subclass type to buffer into
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits connected, non-overlapping buffers, each containing at most count items from the source Publisher/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers, each containing {@code count} items. When the source
     * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the
     * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on
     * immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer3.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and expects the source {@code Publisher} to honor it as
     *  well, although not enforced; violation <em>may</em> lead to {@code MissingBackpressureException} somewhere
     *  downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the collection subclass type to buffer into
     * @param count
     *            the maximum number of items in each buffer before it should be emitted
     * @param bufferSupplier
     *            a factory function that returns an instance of the collection subclass to be used and returned
     *            as the buffer
     * @return a Flowable that emits connected, non-overlapping buffers, each containing at most
     *         {@code count} items from the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U extends Collection<? super T>> Flowable<U> buffer(int count, Callable<U> bufferSupplier) {
        return buffer(count, count, bufferSupplier);
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher starts a new buffer periodically, as determined by the timeskip argument. It emits each buffer after a fixed timespan, specified by the timespan argument. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer operates by default on the computation Scheduler.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted
timeskip – 
           the period of time after which a new buffer will be created
unit –  the unit of time that applies to the timespan and timeskip arguments
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits new buffers of items emitted by the source Publisher periodically after
        a fixed timespan has elapsed/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher starts a new buffer periodically, as determined by the {@code timeskip} argument. It emits
     * each buffer after a fixed timespan, specified by the {@code timespan} argument. When the source
     * Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the
     * source Publisher. Note that if the source Publisher issues an onError notification the event is passed on
     * immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer7.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each buffer collects items before it is emitted
     * @param timeskip
     *            the period of time after which a new buffer will be created
     * @param unit
     *            the unit of time that applies to the {@code timespan} and {@code timeskip} arguments
     * @return a Flowable that emits new buffers of items emitted by the source Publisher periodically after
     *         a fixed timespan has elapsed
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<List<T>> buffer(long timespan, long timeskip, TimeUnit unit) {
        return buffer(timespan, timeskip, unit, Schedulers.computation(), ArrayListSupplier.<T>asCallable());
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher starts a new buffer periodically, as determined by the timeskip argument, and on the specified scheduler. It emits each buffer after a fixed timespan, specified by the timespan argument. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted
timeskip – 
           the period of time after which a new buffer will be created
unit –  the unit of time that applies to the timespan and timeskip arguments
scheduler –  the Scheduler to use when determining the end and start of a buffer
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits new buffers of items emitted by the source Publisher periodically after
        a fixed timespan has elapsed/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher starts a new buffer periodically, as determined by the {@code timeskip} argument, and on the
     * specified {@code scheduler}. It emits each buffer after a fixed timespan, specified by the
     * {@code timespan} argument. When the source Publisher completes, the resulting Publisher emits the current buffer
     * and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError
     * notification the event is passed on immediately without first emitting the buffer it is in the process of
     * assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer7.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each buffer collects items before it is emitted
     * @param timeskip
     *            the period of time after which a new buffer will be created
     * @param unit
     *            the unit of time that applies to the {@code timespan} and {@code timeskip} arguments
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a buffer
     * @return a Flowable that emits new buffers of items emitted by the source Publisher periodically after
     *         a fixed timespan has elapsed
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<List<T>> buffer(long timespan, long timeskip, TimeUnit unit, Scheduler scheduler) {
        return buffer(timespan, timeskip, unit, scheduler, ArrayListSupplier.<T>asCallable());
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher starts a new buffer periodically, as determined by the timeskip argument, and on the specified scheduler. It emits each buffer after a fixed timespan, specified by the timespan argument. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted
timeskip – 
           the period of time after which a new buffer will be created
unit –  the unit of time that applies to the timespan and timeskip arguments
scheduler –  the Scheduler to use when determining the end and start of a buffer
bufferSupplier – 
           a factory function that returns an instance of the collection subclass to be used and returned
           as the buffer
Type parameters:
<U> – the collection subclass type to buffer into
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits new buffers of items emitted by the source Publisher periodically after
        a fixed timespan has elapsed/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher starts a new buffer periodically, as determined by the {@code timeskip} argument, and on the
     * specified {@code scheduler}. It emits each buffer after a fixed timespan, specified by the
     * {@code timespan} argument. When the source Publisher completes, the resulting Publisher emits the current buffer
     * and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError
     * notification the event is passed on immediately without first emitting the buffer it is in the process of
     * assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer7.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param <U> the collection subclass type to buffer into
     * @param timespan
     *            the period of time each buffer collects items before it is emitted
     * @param timeskip
     *            the period of time after which a new buffer will be created
     * @param unit
     *            the unit of time that applies to the {@code timespan} and {@code timeskip} arguments
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a buffer
     * @param bufferSupplier
     *            a factory function that returns an instance of the collection subclass to be used and returned
     *            as the buffer
     * @return a Flowable that emits new buffers of items emitted by the source Publisher periodically after
     *         a fixed timespan has elapsed
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final <U extends Collection<? super T>> Flowable<U> buffer(long timespan, long timeskip, TimeUnit unit,
            Scheduler scheduler, Callable<U> bufferSupplier) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableBufferTimed<T, U>(this, timespan, timeskip, unit, scheduler, bufferSupplier, Integer.MAX_VALUE, false));
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the timespan argument. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer operates by default on the computation Scheduler.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted and replaced with a new
           buffer
unit –  the unit of time that applies to the timespan argument
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits connected, non-overlapping buffers of items emitted by the source
        Publisher within a fixed duration/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the
     * {@code timespan} argument. When the source Publisher completes, the resulting Publisher emits the current buffer
     * and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError
     * notification the event is passed on immediately without first emitting the buffer it is in the process of
     * assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer5.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each buffer collects items before it is emitted and replaced with a new
     *            buffer
     * @param unit
     *            the unit of time that applies to the {@code timespan} argument
     * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source
     *         Publisher within a fixed duration
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<List<T>> buffer(long timespan, TimeUnit unit) {
        return buffer(timespan, unit, Schedulers.computation(), Integer.MAX_VALUE);
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the timespan argument or a maximum size specified by the count argument (whichever is reached first). When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer operates by default on the computation Scheduler.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted and replaced with a new
           buffer
unit –  the unit of time which applies to the timespan argument
count – 
           the maximum size of each buffer before it is emitted
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits connected, non-overlapping buffers of items emitted by the source
        Publisher, after a fixed duration or when the buffer reaches maximum capacity (whichever occurs
        first)/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the
     * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached
     * first). When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the
     * notification from the source Publisher. Note that if the source Publisher issues an onError notification the event
     * is passed on immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer6.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each buffer collects items before it is emitted and replaced with a new
     *            buffer
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param count
     *            the maximum size of each buffer before it is emitted
     * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source
     *         Publisher, after a fixed duration or when the buffer reaches maximum capacity (whichever occurs
     *         first)
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<List<T>> buffer(long timespan, TimeUnit unit, int count) {
        return buffer(timespan, unit, Schedulers.computation(), count);
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the timespan argument as measured on the specified scheduler, or a maximum size specified by the count argument (whichever is reached first). When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted and replaced with a new
           buffer
unit –  the unit of time which applies to the timespan argument
scheduler –  the Scheduler to use when determining the end and start of a buffer
count – 
           the maximum size of each buffer before it is emitted
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits connected, non-overlapping buffers of items emitted by the source
        Publisher after a fixed duration or when the buffer reaches maximum capacity (whichever occurs
        first)/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the
     * {@code timespan} argument as measured on the specified {@code scheduler}, or a maximum size specified by
     * the {@code count} argument (whichever is reached first). When the source Publisher completes, the resulting
     * Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the
     * source Publisher issues an onError notification the event is passed on immediately without first emitting the
     * buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer6.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each buffer collects items before it is emitted and replaced with a new
     *            buffer
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a buffer
     * @param count
     *            the maximum size of each buffer before it is emitted
     * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source
     *         Publisher after a fixed duration or when the buffer reaches maximum capacity (whichever occurs
     *         first)
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<List<T>> buffer(long timespan, TimeUnit unit, Scheduler scheduler, int count) {
        return buffer(timespan, unit, scheduler, count, ArrayListSupplier.<T>asCallable(), false);
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the timespan argument as measured on the specified scheduler, or a maximum size specified by the count argument (whichever is reached first). When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted and replaced with a new
           buffer
unit –  the unit of time which applies to the timespan argument
scheduler –  the Scheduler to use when determining the end and start of a buffer
count – 
           the maximum size of each buffer before it is emitted
bufferSupplier – 
           a factory function that returns an instance of the collection subclass to be used and returned
           as the buffer
restartTimerOnMaxSize – if true the time window is restarted when the max capacity of the current buffer
           is reached
Type parameters:
<U> – the collection subclass type to buffer into
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits connected, non-overlapping buffers of items emitted by the source
        Publisher after a fixed duration or when the buffer reaches maximum capacity (whichever occurs
        first)/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the
     * {@code timespan} argument as measured on the specified {@code scheduler}, or a maximum size specified by
     * the {@code count} argument (whichever is reached first). When the source Publisher completes, the resulting
     * Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the
     * source Publisher issues an onError notification the event is passed on immediately without first emitting the
     * buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer6.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param <U> the collection subclass type to buffer into
     * @param timespan
     *            the period of time each buffer collects items before it is emitted and replaced with a new
     *            buffer
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a buffer
     * @param count
     *            the maximum size of each buffer before it is emitted
     * @param bufferSupplier
     *            a factory function that returns an instance of the collection subclass to be used and returned
     *            as the buffer
     * @param restartTimerOnMaxSize if true the time window is restarted when the max capacity of the current buffer
     *            is reached
     * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source
     *         Publisher after a fixed duration or when the buffer reaches maximum capacity (whichever occurs
     *         first)
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final <U extends Collection<? super T>> Flowable<U> buffer(
            long timespan, TimeUnit unit,
            Scheduler scheduler, int count,
            Callable<U> bufferSupplier,
            boolean restartTimerOnMaxSize) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null");
        ObjectHelper.verifyPositive(count, "count");
        return RxJavaPlugins.onAssembly(new FlowableBufferTimed<T, U>(this, timespan, timespan, unit, scheduler, bufferSupplier, count, restartTimerOnMaxSize));
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the timespan argument and on the specified scheduler. When the source Publisher completes, the resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that if the source Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it uses time. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each buffer collects items before it is emitted and replaced with a new
           buffer
unit –  the unit of time which applies to the timespan argument
scheduler –  the Scheduler to use when determining the end and start of a buffer
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits connected, non-overlapping buffers of items emitted by the source
        Publisher within a fixed duration/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers, each of a fixed duration specified by the
     * {@code timespan} argument and on the specified {@code scheduler}. When the source Publisher completes, the
     * resulting Publisher emits the current buffer and propagates the notification from the source Publisher. Note that
     * if the source Publisher issues an onError notification the event is passed on immediately without first emitting
     * the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer5.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time. It requests {@code Long.MAX_VALUE}
     *      upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each buffer collects items before it is emitted and replaced with a new
     *            buffer
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a buffer
     * @return a Flowable that emits connected, non-overlapping buffers of items emitted by the source
     *         Publisher within a fixed duration
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<List<T>> buffer(long timespan, TimeUnit unit, Scheduler scheduler) {
        return buffer(timespan, unit, scheduler, Integer.MAX_VALUE, ArrayListSupplier.<T>asCallable(), false);
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits buffers that it creates when the specified openingIndicator Publisher emits an item, and closes when the Publisher returned from closingIndicator emits an item. If any of the source Publisher, openingIndicator or closingIndicator issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it is instead controlled by the given Publishers and buffers data. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
openingIndicator – 
           the Publisher that, when it emits an item, causes a new buffer to be created
closingIndicator –  the Function that is used to produce a Publisher for every buffer created. When this Publisher emits an item, the associated buffer is emitted.
Type parameters:
<TOpening> – the element type of the buffer-opening Publisher
<TClosing> – the element type of the individual buffer-closing Publishers
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits buffers, containing items from the source Publisher, that are created
        and closed when the specified Publishers emit items/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits buffers that it creates when the specified {@code openingIndicator} Publisher emits an
     * item, and closes when the Publisher returned from {@code closingIndicator} emits an item. If any of the source
     * Publisher, {@code openingIndicator} or {@code closingIndicator} issues an onError notification the event is passed
     * on immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="470" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer2.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it is instead controlled by the given Publishers and
     *      buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <TOpening> the element type of the buffer-opening Publisher
     * @param <TClosing> the element type of the individual buffer-closing Publishers
     * @param openingIndicator
     *            the Publisher that, when it emits an item, causes a new buffer to be created
     * @param closingIndicator
     *            the {@link Function} that is used to produce a Publisher for every buffer created. When this
     *            Publisher emits an item, the associated buffer is emitted.
     * @return a Flowable that emits buffers, containing items from the source Publisher, that are created
     *         and closed when the specified Publishers emit items
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <TOpening, TClosing> Flowable<List<T>> buffer(
            Flowable<? extends TOpening> openingIndicator,
            Function<? super TOpening, ? extends Publisher<? extends TClosing>> closingIndicator) {
        return buffer(openingIndicator, closingIndicator, ArrayListSupplier.<T>asCallable());
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits buffers that it creates when the specified openingIndicator Publisher emits an item, and closes when the Publisher returned from closingIndicator emits an item. If any of the source Publisher, openingIndicator or closingIndicator issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 



 
Backpressure:
 
This operator does not support backpressure as it is instead controlled by the given Publishers and buffers data. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
openingIndicator – 
           the Publisher that, when it emits an item, causes a new buffer to be created
closingIndicator –  the Function that is used to produce a Publisher for every buffer created. When this Publisher emits an item, the associated buffer is emitted.
bufferSupplier – 
           a factory function that returns an instance of the collection subclass to be used and returned
           as the buffer
Type parameters:
<U> – the collection subclass type to buffer into
<TOpening> – the element type of the buffer-opening Publisher
<TClosing> – the element type of the individual buffer-closing Publishers
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits buffers, containing items from the source Publisher, that are created
        and closed when the specified Publishers emit items/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits buffers that it creates when the specified {@code openingIndicator} Publisher emits an
     * item, and closes when the Publisher returned from {@code closingIndicator} emits an item. If any of the source
     * Publisher, {@code openingIndicator} or {@code closingIndicator} issues an onError notification the event is passed
     * on immediately without first emitting the buffer it is in the process of assembling.
     * <p>
     * <img width="640" height="470" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer2.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it is instead controlled by the given Publishers and
     *      buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the collection subclass type to buffer into
     * @param <TOpening> the element type of the buffer-opening Publisher
     * @param <TClosing> the element type of the individual buffer-closing Publishers
     * @param openingIndicator
     *            the Publisher that, when it emits an item, causes a new buffer to be created
     * @param closingIndicator
     *            the {@link Function} that is used to produce a Publisher for every buffer created. When this
     *            Publisher emits an item, the associated buffer is emitted.
     * @param bufferSupplier
     *            a factory function that returns an instance of the collection subclass to be used and returned
     *            as the buffer
     * @return a Flowable that emits buffers, containing items from the source Publisher, that are created
     *         and closed when the specified Publishers emit items
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <TOpening, TClosing, U extends Collection<? super T>> Flowable<U> buffer(
            Flowable<? extends TOpening> openingIndicator,
            Function<? super TOpening, ? extends Publisher<? extends TClosing>> closingIndicator,
            Callable<U> bufferSupplier) {
        ObjectHelper.requireNonNull(openingIndicator, "openingIndicator is null");
        ObjectHelper.requireNonNull(closingIndicator, "closingIndicator is null");
        ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableBufferBoundary<T, U, TOpening, TClosing>(this, openingIndicator, closingIndicator, bufferSupplier));
    }

    
Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the
specified boundary Publisher emits an item.



Completion of either the source or the boundary Publisher causes the returned Publisher to emit the
latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification
the event is passed on immediately without first emitting the buffer it is in the process of assembling.

 
Backpressure:
 
This operator does not support backpressure as it is instead controlled by the Publisher boundary and buffers data. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
boundaryIndicator – 
           the boundary Publisher
Type parameters:
<B> – 
           the boundary value type (ignored)
See Also:
buffer(Publisher, int)
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits buffered items from the source Publisher when the boundary Publisher
        emits an item/**
     * Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the
     * specified boundary Publisher emits an item.
     * <p>
     * <img width="640" height="395" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer8.png" alt="">
     * <p>
     * Completion of either the source or the boundary Publisher causes the returned Publisher to emit the
     * latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification
     * the event is passed on immediately without first emitting the buffer it is in the process of assembling.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it is instead controlled by the {@code Publisher}
     *      {@code boundary} and buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey
     *      downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <B>
     *            the boundary value type (ignored)
     * @param boundaryIndicator
     *            the boundary Publisher
     * @return a Flowable that emits buffered items from the source Publisher when the boundary Publisher
     *         emits an item
     * @see #buffer(Publisher, int)
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B> Flowable<List<T>> buffer(Publisher<B> boundaryIndicator) {
        return buffer(boundaryIndicator, ArrayListSupplier.<T>asCallable());
    }

    
Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the
specified boundary Publisher emits an item.



Completion of either the source or the boundary Publisher causes the returned Publisher to emit the
latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification
the event is passed on immediately without first emitting the buffer it is in the process of assembling.

 
Backpressure:
 
This operator does not support backpressure as it is instead controlled by the Publisher boundary and buffers data. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
boundaryIndicator – 
           the boundary Publisher
initialCapacity – 
           the initial capacity of each buffer chunk
Type parameters:
<B> – 
           the boundary value type (ignored)
See Also:
ReactiveX operators documentation: Buffer
buffer(Publisher)
Returns:
a Flowable that emits buffered items from the source Publisher when the boundary Publisher
        emits an item/**
     * Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the
     * specified boundary Publisher emits an item.
     * <p>
     * <img width="640" height="395" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer8.png" alt="">
     * <p>
     * Completion of either the source or the boundary Publisher causes the returned Publisher to emit the
     * latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification
     * the event is passed on immediately without first emitting the buffer it is in the process of assembling.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it is instead controlled by the {@code Publisher}
     *      {@code boundary} and buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey
     *      downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <B>
     *            the boundary value type (ignored)
     * @param boundaryIndicator
     *            the boundary Publisher
     * @param initialCapacity
     *            the initial capacity of each buffer chunk
     * @return a Flowable that emits buffered items from the source Publisher when the boundary Publisher
     *         emits an item
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     * @see #buffer(Publisher)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B> Flowable<List<T>> buffer(Publisher<B> boundaryIndicator, final int initialCapacity) {
        ObjectHelper.verifyPositive(initialCapacity, "initialCapacity");
        return buffer(boundaryIndicator, Functions.<T>createArrayList(initialCapacity));
    }

    
Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the
specified boundary Publisher emits an item.



Completion of either the source or the boundary Publisher causes the returned Publisher to emit the
latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification
the event is passed on immediately without first emitting the buffer it is in the process of assembling.

 
Backpressure:
 
This operator does not support backpressure as it is instead controlled by the Publisher boundary and buffers data. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
boundaryIndicator – 
           the boundary Publisher
bufferSupplier – 
           a factory function that returns an instance of the collection subclass to be used and returned
           as the buffer
Type parameters:
<U> – the collection subclass type to buffer into
<B> – 
           the boundary value type (ignored)
See Also:
buffer(Publisher, int)
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits buffered items from the source Publisher when the boundary Publisher
        emits an item/**
     * Returns a Flowable that emits non-overlapping buffered items from the source Publisher each time the
     * specified boundary Publisher emits an item.
     * <p>
     * <img width="640" height="395" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer8.png" alt="">
     * <p>
     * Completion of either the source or the boundary Publisher causes the returned Publisher to emit the
     * latest buffer and complete. If either the source Publisher or the boundary Publisher issues an onError notification
     * the event is passed on immediately without first emitting the buffer it is in the process of assembling.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it is instead controlled by the {@code Publisher}
     *      {@code boundary} and buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey
     *      downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the collection subclass type to buffer into
     * @param <B>
     *            the boundary value type (ignored)
     * @param boundaryIndicator
     *            the boundary Publisher
     * @param bufferSupplier
     *            a factory function that returns an instance of the collection subclass to be used and returned
     *            as the buffer
     * @return a Flowable that emits buffered items from the source Publisher when the boundary Publisher
     *         emits an item
     * @see #buffer(Publisher, int)
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B, U extends Collection<? super T>> Flowable<U> buffer(Publisher<B> boundaryIndicator, Callable<U> bufferSupplier) {
        ObjectHelper.requireNonNull(boundaryIndicator, "boundaryIndicator is null");
        ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableBufferExactBoundary<T, U, B>(this, boundaryIndicator, bufferSupplier));
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers. It emits the current buffer and replaces it with a new buffer whenever the Publisher produced by the specified boundaryIndicatorSupplier emits an item. 


 If either the source Publisher or the boundary Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 

 
Backpressure:
 
This operator does not support backpressure as it is instead controlled by the given Publishers and buffers data. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
boundaryIndicatorSupplier –  a Callable that produces a Publisher that governs the boundary between buffers. Whenever the supplied Publisher emits an item, buffer emits the current buffer and begins to fill a new one
Type parameters:
<B> – the value type of the boundary-providing Publisher
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits a connected, non-overlapping buffer of items from the source Publisher each time the Publisher created with the closingIndicator argument emits an item/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers. It emits the current buffer and replaces it with a
     * new buffer whenever the Publisher produced by the specified {@code boundaryIndicatorSupplier} emits an item.
     * <p>
     * <img width="640" height="395" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer1.png" alt="">
     * <p>
     * If either the source {@code Publisher} or the boundary {@code Publisher} issues an {@code onError} notification the event is passed on
     * immediately without first emitting the buffer it is in the process of assembling.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it is instead controlled by the given Publishers and
     *      buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <B> the value type of the boundary-providing Publisher
     * @param boundaryIndicatorSupplier
     *            a {@link Callable} that produces a Publisher that governs the boundary between buffers.
     *            Whenever the supplied {@code Publisher} emits an item, {@code buffer} emits the current buffer and
     *            begins to fill a new one
     * @return a Flowable that emits a connected, non-overlapping buffer of items from the source Publisher
     *         each time the Publisher created with the {@code closingIndicator} argument emits an item
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B> Flowable<List<T>> buffer(Callable<? extends Publisher<B>> boundaryIndicatorSupplier) {
        return buffer(boundaryIndicatorSupplier, ArrayListSupplier.<T>asCallable());
    }

    
Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping buffers. It emits the current buffer and replaces it with a new buffer whenever the Publisher produced by the specified boundaryIndicatorSupplier emits an item. 


 If either the source Publisher or the boundary Publisher issues an onError notification the event is passed on immediately without first emitting the buffer it is in the process of assembling. 

 
Backpressure:
 
This operator does not support backpressure as it is instead controlled by the given Publishers and buffers data. It requests Long.MAX_VALUE upstream and does not obey downstream requests.
 
Scheduler:
 
This version of buffer does not operate by default on a particular Scheduler.

Params:
boundaryIndicatorSupplier –  a Callable that produces a Publisher that governs the boundary between buffers. Whenever the supplied Publisher emits an item, buffer emits the current buffer and begins to fill a new one
bufferSupplier – 
           a factory function that returns an instance of the collection subclass to be used and returned
           as the buffer
Type parameters:
<U> – the collection subclass type to buffer into
<B> – the value type of the boundary-providing Publisher
See Also:
ReactiveX operators documentation: Buffer
Returns:
a Flowable that emits a connected, non-overlapping buffer of items from the source Publisher each time the Publisher created with the closingIndicator argument emits an item/**
     * Returns a Flowable that emits buffers of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping buffers. It emits the current buffer and replaces it with a
     * new buffer whenever the Publisher produced by the specified {@code boundaryIndicatorSupplier} emits an item.
     * <p>
     * <img width="640" height="395" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/buffer1.png" alt="">
     * <p>
     * If either the source {@code Publisher} or the boundary {@code Publisher} issues an {@code onError} notification the event is passed on
     * immediately without first emitting the buffer it is in the process of assembling.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it is instead controlled by the given Publishers and
     *      buffers data. It requests {@code Long.MAX_VALUE} upstream and does not obey downstream requests.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code buffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the collection subclass type to buffer into
     * @param <B> the value type of the boundary-providing Publisher
     * @param boundaryIndicatorSupplier
     *            a {@link Callable} that produces a Publisher that governs the boundary between buffers.
     *            Whenever the supplied {@code Publisher} emits an item, {@code buffer} emits the current buffer and
     *            begins to fill a new one
     * @param bufferSupplier
     *            a factory function that returns an instance of the collection subclass to be used and returned
     *            as the buffer
     * @return a Flowable that emits a connected, non-overlapping buffer of items from the source Publisher
     *         each time the Publisher created with the {@code closingIndicator} argument emits an item
     * @see <a href="http://reactivex.io/documentation/operators/buffer.html">ReactiveX operators documentation: Buffer</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B, U extends Collection<? super T>> Flowable<U> buffer(Callable<? extends Publisher<B>> boundaryIndicatorSupplier,
            Callable<U> bufferSupplier) {
        ObjectHelper.requireNonNull(boundaryIndicatorSupplier, "boundaryIndicatorSupplier is null");
        ObjectHelper.requireNonNull(bufferSupplier, "bufferSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableBufferBoundarySupplier<T, U, B>(this, boundaryIndicatorSupplier, bufferSupplier));
    }

    
Returns a Flowable that subscribes to this Publisher lazily, caches all of its events
and replays them, in the same order as received, to all the downstream subscribers.


 This is useful when you want a Publisher to cache responses and you can't control the subscribe/cancel behavior of all the Subscribers. 
 The operator subscribes only when the first downstream subscriber subscribes and maintains a single subscription towards this Publisher. In contrast, the operator family of replay() that return a ConnectableFlowable require an explicit call to ConnectableFlowable.connect(). 

Note: You sacrifice the ability to cancel the origin when you use the cache Subscriber so be careful not to use this Subscriber on Publishers that emit an infinite or very large number of items that will use up memory. A possible workaround is to apply `takeUntil` with a predicate or another source before (and perhaps after) the application of cache(). 

AtomicBoolean shouldStop = new AtomicBoolean();
source.takeUntil(v -> shouldStop.get())
      .cache()
      .takeUntil(v -> shouldStop.get())
      .subscribe(...);
 Since the operator doesn't allow clearing the cached values either, the possible workaround is to forget all references to it via onTerminateDetach() applied along with the previous workaround: 

AtomicBoolean shouldStop = new AtomicBoolean();
source.takeUntil(v -> shouldStop.get())
      .onTerminateDetach()
      .cache()
      .takeUntil(v -> shouldStop.get())
      .onTerminateDetach()
      .subscribe(...);


 
Backpressure:
 
The operator consumes this Publisher in an unbounded fashion but respects the backpressure
 of each downstream Subscriber individually.
 
Scheduler:
 
cache does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that, when first subscribed to, caches all of its items and notifications for the
        benefit of subsequent subscribers/**
     * Returns a Flowable that subscribes to this Publisher lazily, caches all of its events
     * and replays them, in the same order as received, to all the downstream subscribers.
     * <p>
     * <img width="640" height="410" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/cache.png" alt="">
     * <p>
     * This is useful when you want a Publisher to cache responses and you can't control the
     * subscribe/cancel behavior of all the {@link Subscriber}s.
     * <p>
     * The operator subscribes only when the first downstream subscriber subscribes and maintains
     * a single subscription towards this Publisher. In contrast, the operator family of {@link #replay()}
     * that return a {@link ConnectableFlowable} require an explicit call to {@link ConnectableFlowable#connect()}.
     * <p>
     * <em>Note:</em> You sacrifice the ability to cancel the origin when you use the {@code cache}
     * Subscriber so be careful not to use this Subscriber on Publishers that emit an infinite or very large number
     * of items that will use up memory.
     * A possible workaround is to apply `takeUntil` with a predicate or
     * another source before (and perhaps after) the application of cache().
     * <pre><code>
     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -&gt; shouldStop.get())
     *       .cache()
     *       .takeUntil(v -&gt; shouldStop.get())
     *       .subscribe(...);
     * </code></pre>
     * Since the operator doesn't allow clearing the cached values either, the possible workaround is
     * to forget all references to it via {@link #onTerminateDetach()} applied along with the previous
     * workaround:
     * <pre><code>
     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -&gt; shouldStop.get())
     *       .onTerminateDetach()
     *       .cache()
     *       .takeUntil(v -&gt; shouldStop.get())
     *       .onTerminateDetach()
     *       .subscribe(...);
     * </code></pre>
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes this Publisher in an unbounded fashion but respects the backpressure
     *  of each downstream Subscriber individually.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code cache} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that, when first subscribed to, caches all of its items and notifications for the
     *         benefit of subsequent subscribers
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> cache() {
        return cacheWithInitialCapacity(16);
    }

    
Returns a Flowable that subscribes to this Publisher lazily, caches all of its events
and replays them, in the same order as received, to all the downstream subscribers.


 This is useful when you want a Publisher to cache responses and you can't control the subscribe/cancel behavior of all the Subscribers. 
 The operator subscribes only when the first downstream subscriber subscribes and maintains a single subscription towards this Publisher. In contrast, the operator family of replay() that return a ConnectableFlowable require an explicit call to ConnectableFlowable.connect(). 

Note: You sacrifice the ability to cancel the origin when you use the cache Subscriber so be careful not to use this Subscriber on Publishers that emit an infinite or very large number of items that will use up memory. A possible workaround is to apply `takeUntil` with a predicate or another source before (and perhaps after) the application of cache(). 

AtomicBoolean shouldStop = new AtomicBoolean();
source.takeUntil(v -> shouldStop.get())
      .cache()
      .takeUntil(v -> shouldStop.get())
      .subscribe(...);
 Since the operator doesn't allow clearing the cached values either, the possible workaround is to forget all references to it via onTerminateDetach() applied along with the previous workaround: 

AtomicBoolean shouldStop = new AtomicBoolean();
source.takeUntil(v -> shouldStop.get())
      .onTerminateDetach()
      .cache()
      .takeUntil(v -> shouldStop.get())
      .onTerminateDetach()
      .subscribe(...);


 
Backpressure:
 
The operator consumes this Publisher in an unbounded fashion but respects the backpressure
 of each downstream Subscriber individually.
 
Scheduler:
 
cacheWithInitialCapacity does not operate by default on a particular Scheduler.


Note: The capacity hint is not an upper bound on cache size. For that, consider replay(int) in combination with ConnectableFlowable.autoConnect() or similar. 
Params:
initialCapacity – hint for number of items to cache (for optimizing underlying data structure)
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that, when first subscribed to, caches all of its items and notifications for the
        benefit of subsequent subscribers/**
     * Returns a Flowable that subscribes to this Publisher lazily, caches all of its events
     * and replays them, in the same order as received, to all the downstream subscribers.
     * <p>
     * <img width="640" height="410" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/cache.png" alt="">
     * <p>
     * This is useful when you want a Publisher to cache responses and you can't control the
     * subscribe/cancel behavior of all the {@link Subscriber}s.
     * <p>
     * The operator subscribes only when the first downstream subscriber subscribes and maintains
     * a single subscription towards this Publisher. In contrast, the operator family of {@link #replay()}
     * that return a {@link ConnectableFlowable} require an explicit call to {@link ConnectableFlowable#connect()}.
     * <p>
     * <em>Note:</em> You sacrifice the ability to cancel the origin when you use the {@code cache}
     * Subscriber so be careful not to use this Subscriber on Publishers that emit an infinite or very large number
     * of items that will use up memory.
     * A possible workaround is to apply `takeUntil` with a predicate or
     * another source before (and perhaps after) the application of cache().
     * <pre><code>
     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -&gt; shouldStop.get())
     *       .cache()
     *       .takeUntil(v -&gt; shouldStop.get())
     *       .subscribe(...);
     * </code></pre>
     * Since the operator doesn't allow clearing the cached values either, the possible workaround is
     * to forget all references to it via {@link #onTerminateDetach()} applied along with the previous
     * workaround:
     * <pre><code>
     * AtomicBoolean shouldStop = new AtomicBoolean();
     *
     * source.takeUntil(v -&gt; shouldStop.get())
     *       .onTerminateDetach()
     *       .cache()
     *       .takeUntil(v -&gt; shouldStop.get())
     *       .onTerminateDetach()
     *       .subscribe(...);
     * </code></pre>
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes this Publisher in an unbounded fashion but respects the backpressure
     *  of each downstream Subscriber individually.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code cacheWithInitialCapacity} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>
     * <em>Note:</em> The capacity hint is not an upper bound on cache size. For that, consider
     * {@link #replay(int)} in combination with {@link ConnectableFlowable#autoConnect()} or similar.
     *
     * @param initialCapacity hint for number of items to cache (for optimizing underlying data structure)
     * @return a Flowable that, when first subscribed to, caches all of its items and notifications for the
     *         benefit of subsequent subscribers
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> cacheWithInitialCapacity(int initialCapacity) {
        ObjectHelper.verifyPositive(initialCapacity, "initialCapacity");
        return RxJavaPlugins.onAssembly(new FlowableCache<T>(this, initialCapacity));
    }

    
Returns a Flowable that emits the items emitted by the source Publisher, converted to the specified
type.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
cast does not operate by default on a particular Scheduler.

Params:
clazz –  the target class type that cast will cast the items emitted by the source Publisher into before emitting them from the resulting Publisher
Type parameters:
<U> – the output value type cast to
See Also:
ReactiveX operators documentation: Map
Returns:
a Flowable that emits each item from the source Publisher after converting it to the
        specified type/**
     * Returns a Flowable that emits the items emitted by the source Publisher, converted to the specified
     * type.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/cast.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code cast} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the output value type cast to
     * @param clazz
     *            the target class type that {@code cast} will cast the items emitted by the source Publisher
     *            into before emitting them from the resulting Publisher
     * @return a Flowable that emits each item from the source Publisher after converting it to the
     *         specified type
     * @see <a href="http://reactivex.io/documentation/operators/map.html">ReactiveX operators documentation: Map</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<U> cast(final Class<U> clazz) {
        ObjectHelper.requireNonNull(clazz, "clazz is null");
        return map(Functions.castFunction(clazz));
    }

    
Collects items emitted by the finite source Publisher into a single mutable data structure and returns
a Single that emits this structure.


 This is a simplified version of reduce that does not need to return the state on each pass. 
 Note that this operator requires the upstream to signal onComplete for the accumulator object to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
This operator does not support backpressure because by intent it will receive all values and reduce them to a single onNext.
 
Scheduler:
 
collect does not operate by default on a particular Scheduler.

Params:
initialItemSupplier – 
          the mutable data structure that will collect the items
collector –  a function that accepts the state and an emitted item, and modifies state accordingly
Type parameters:
<U> – the accumulator and output type
See Also:
ReactiveX operators documentation: Reduce
Returns:
a Single that emits the result of collecting the values emitted by the source Publisher
        into a single mutable data structure/**
     * Collects items emitted by the finite source Publisher into a single mutable data structure and returns
     * a Single that emits this structure.
     * <p>
     * <img width="640" height="330" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/collect.png" alt="">
     * <p>
     * This is a simplified version of {@code reduce} that does not need to return the state on each pass.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure because by intent it will receive all values and reduce
     *      them to a single {@code onNext}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code collect} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the accumulator and output type
     * @param initialItemSupplier
     *           the mutable data structure that will collect the items
     * @param collector
     *           a function that accepts the {@code state} and an emitted item, and modifies {@code state}
     *           accordingly
     * @return a Single that emits the result of collecting the values emitted by the source Publisher
     *         into a single mutable data structure
     * @see <a href="http://reactivex.io/documentation/operators/reduce.html">ReactiveX operators documentation: Reduce</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Single<U> collect(Callable<? extends U> initialItemSupplier, BiConsumer<? super U, ? super T> collector) {
        ObjectHelper.requireNonNull(initialItemSupplier, "initialItemSupplier is null");
        ObjectHelper.requireNonNull(collector, "collector is null");
        return RxJavaPlugins.onAssembly(new FlowableCollectSingle<T, U>(this, initialItemSupplier, collector));
    }

    
Collects items emitted by the finite source Publisher into a single mutable data structure and returns
a Single that emits this structure.


 This is a simplified version of reduce that does not need to return the state on each pass. 
 Note that this operator requires the upstream to signal onComplete for the accumulator object to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
This operator does not support backpressure because by intent it will receive all values and reduce them to a single onNext.
 
Scheduler:
 
collectInto does not operate by default on a particular Scheduler.

Params:
initialItem – 
          the mutable data structure that will collect the items
collector –  a function that accepts the state and an emitted item, and modifies state accordingly
Type parameters:
<U> – the accumulator and output type
See Also:
ReactiveX operators documentation: Reduce
Returns:
a Single that emits the result of collecting the values emitted by the source Publisher
        into a single mutable data structure/**
     * Collects items emitted by the finite source Publisher into a single mutable data structure and returns
     * a Single that emits this structure.
     * <p>
     * <img width="640" height="330" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/collect.png" alt="">
     * <p>
     * This is a simplified version of {@code reduce} that does not need to return the state on each pass.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure because by intent it will receive all values and reduce
     *      them to a single {@code onNext}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code collectInto} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the accumulator and output type
     * @param initialItem
     *           the mutable data structure that will collect the items
     * @param collector
     *           a function that accepts the {@code state} and an emitted item, and modifies {@code state}
     *           accordingly
     * @return a Single that emits the result of collecting the values emitted by the source Publisher
     *         into a single mutable data structure
     * @see <a href="http://reactivex.io/documentation/operators/reduce.html">ReactiveX operators documentation: Reduce</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Single<U> collectInto(final U initialItem, BiConsumer<? super U, ? super T> collector) {
        ObjectHelper.requireNonNull(initialItem, "initialItem is null");
        return collect(Functions.justCallable(initialItem), collector);
    }

    
Transform a Publisher by applying a particular Transformer function to it.
 This method operates on the Publisher itself whereas lift operates on the Publisher's Subscribers or Subscribers. 
 If the operator you are creating is designed to act on the individual items emitted by a source Publisher, use lift. If your operator is designed to transform the source Publisher as a whole (for instance, by applying a particular set of existing RxJava operators to it) use compose. 

 
Backpressure:
 
The operator itself doesn't interfere with the backpressure behavior which only depends on what kind of Publisher the transformer returns.
 
Scheduler:
 
compose does not operate by default on a particular Scheduler.

Params:
composer – implements the function that transforms the source Publisher
Type parameters:
<R> – the value type of the output Publisher
See Also:
RxJava wiki: Implementing Your Own Operators
Returns:
the source Publisher, transformed by the transformer function/**
     * Transform a Publisher by applying a particular Transformer function to it.
     * <p>
     * This method operates on the Publisher itself whereas {@link #lift} operates on the Publisher's
     * Subscribers or Subscribers.
     * <p>
     * If the operator you are creating is designed to act on the individual items emitted by a source
     * Publisher, use {@link #lift}. If your operator is designed to transform the source Publisher as a whole
     * (for instance, by applying a particular set of existing RxJava operators to it) use {@code compose}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator itself doesn't interfere with the backpressure behavior which only depends
     *  on what kind of {@code Publisher} the transformer returns.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code compose} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the value type of the output Publisher
     * @param composer implements the function that transforms the source Publisher
     * @return the source Publisher, transformed by the transformer function
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Implementing-Your-Own-Operators">RxJava wiki: Implementing Your Own Operators</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> compose(FlowableTransformer<? super T, ? extends R> composer) {
        return fromPublisher(((FlowableTransformer<T, R>) ObjectHelper.requireNonNull(composer, "composer is null")).apply(this));
    }

    
Returns a new Flowable that emits items resulting from applying a function that you supply to each item
emitted by the source Publisher, where that function returns a Publisher, and then emitting the items
that result from concatenating those resulting Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. Both this and the inner Publishers are expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException. If any of the inner Publishers doesn't honor backpressure, that may throw an IllegalStateException when that Publisher completes.
 
Scheduler:
 
concatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
Type parameters:
<R> – the type of the inner Publisher sources and thus the output type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the result of applying the transformation function to each item emitted
        by the source Publisher and concatenating the Publishers obtained from this transformation/**
     * Returns a new Flowable that emits items resulting from applying a function that you supply to each item
     * emitted by the source Publisher, where that function returns a Publisher, and then emitting the items
     * that result from concatenating those resulting Publishers.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor
     *  backpressure, that <em>may</em> throw an {@code IllegalStateException} when that
     *  {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the type of the inner Publisher sources and thus the output type
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @return a Flowable that emits the result of applying the transformation function to each item emitted
     *         by the source Publisher and concatenating the Publishers obtained from this transformation
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMap(Function<? super T, ? extends Publisher<? extends R>> mapper) {
        return concatMap(mapper, 2);
    }

    
Returns a new Flowable that emits items resulting from applying a function that you supply to each item
emitted by the source Publisher, where that function returns a Publisher, and then emitting the items
that result from concatenating those resulting Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. Both this and the inner Publishers are expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException. If any of the inner Publishers doesn't honor backpressure, that may throw an IllegalStateException when that Publisher completes.
 
Scheduler:
 
concatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
prefetch – 
           the number of elements to prefetch from the current Flowable
Type parameters:
<R> – the type of the inner Publisher sources and thus the output type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the result of applying the transformation function to each item emitted
        by the source Publisher and concatenating the Publishers obtained from this transformation/**
     * Returns a new Flowable that emits items resulting from applying a function that you supply to each item
     * emitted by the source Publisher, where that function returns a Publisher, and then emitting the items
     * that result from concatenating those resulting Publishers.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor
     *  backpressure, that <em>may</em> throw an {@code IllegalStateException} when that
     *  {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the type of the inner Publisher sources and thus the output type
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @param prefetch
     *            the number of elements to prefetch from the current Flowable
     * @return a Flowable that emits the result of applying the transformation function to each item emitted
     *         by the source Publisher and concatenating the Publishers obtained from this transformation
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMap(Function<? super T, ? extends Publisher<? extends R>> mapper, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        if (this instanceof ScalarCallable) {
            @SuppressWarnings("unchecked")
            T v = ((ScalarCallable<T>)this).call();
            if (v == null) {
                return empty();
            }
            return FlowableScalarXMap.scalarXMap(v, mapper);
        }
        return RxJavaPlugins.onAssembly(new FlowableConcatMap<T, R>(this, mapper, prefetch, ErrorMode.IMMEDIATE));
    }

    
Maps the upstream items into CompletableSources and subscribes to them one after the other completes. 



 
Backpressure:
 
The operator expects the upstream to support backpressure. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapCompletable does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a CompletableSource to become the next source to be subscribed to
See Also:
concatMapCompletableDelayError(Function)
Returns:
a new Completable instance
Since:
2.2/**
     * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the
     * other completes.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapCompletable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param mapper the function called with the upstream item and should return
     *               a {@code CompletableSource} to become the next source to
     *               be subscribed to
     * @return a new Completable instance
     * @see #concatMapCompletableDelayError(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.FULL)
    public final Completable concatMapCompletable(Function<? super T, ? extends CompletableSource> mapper) {
        return concatMapCompletable(mapper, 2);
    }

    
Maps the upstream items into CompletableSources and subscribes to them one after the other completes. 



 
Backpressure:
 
The operator expects the upstream to support backpressure. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapCompletable does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a CompletableSource to become the next source to be subscribed to
prefetch – The number of upstream items to prefetch so that fresh items are ready to be mapped when a previous CompletableSource terminates. The operator replenishes after half of the prefetch amount has been consumed and turned into CompletableSources.
See Also:
concatMapCompletableDelayError(Function, boolean, int)
Returns:
a new Completable instance
Since:
2.2/**
     * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the
     * other completes.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapCompletable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param mapper the function called with the upstream item and should return
     *               a {@code CompletableSource} to become the next source to
     *               be subscribed to
     * @param prefetch The number of upstream items to prefetch so that fresh items are
     *                 ready to be mapped when a previous {@code CompletableSource} terminates.
     *                 The operator replenishes after half of the prefetch amount has been consumed
     *                 and turned into {@code CompletableSource}s.
     * @return a new Completable instance
     * @see #concatMapCompletableDelayError(Function, boolean, int)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.FULL)
    public final Completable concatMapCompletable(Function<? super T, ? extends CompletableSource> mapper, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapCompletable<T>(this, mapper, ErrorMode.IMMEDIATE, prefetch));
    }

    
Maps the upstream items into CompletableSources and subscribes to them one after the other terminates, delaying all errors till both this Flowable and all inner CompletableSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapCompletableDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a CompletableSource to become the next source to be subscribed to
See Also:
concatMapCompletable(Function, int)
Returns:
a new Completable instance
Since:
2.2/**
     * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the
     * other terminates, delaying all errors till both this {@code Flowable} and all
     * inner {@code CompletableSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param mapper the function called with the upstream item and should return
     *               a {@code CompletableSource} to become the next source to
     *               be subscribed to
     * @return a new Completable instance
     * @see #concatMapCompletable(Function, int)
     * @since 2.2
     */
    @CheckReturnValue
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.FULL)
    public final Completable concatMapCompletableDelayError(Function<? super T, ? extends CompletableSource> mapper) {
        return concatMapCompletableDelayError(mapper, true, 2);
    }

    
Maps the upstream items into CompletableSources and subscribes to them one after the other terminates, optionally delaying all errors till both this Flowable and all inner CompletableSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapCompletableDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a CompletableSource to become the next source to be subscribed to
tillTheEnd – If true, errors from this Flowable or any of the inner CompletableSources are delayed until all of them terminate. If false, an error from this Flowable is delayed until the current inner CompletableSource terminates and only then is it emitted to the downstream.
See Also:
concatMapCompletable(Function)
Returns:
a new Completable instance
Since:
2.2/**
     * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the
     * other terminates, optionally delaying all errors till both this {@code Flowable} and all
     * inner {@code CompletableSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param mapper the function called with the upstream item and should return
     *               a {@code CompletableSource} to become the next source to
     *               be subscribed to
     * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the
     *                   inner {@code CompletableSource}s are delayed until all
     *                   of them terminate. If {@code false}, an error from this
     *                   {@code Flowable} is delayed until the current inner
     *                   {@code CompletableSource} terminates and only then is
     *                   it emitted to the downstream.
     * @return a new Completable instance
     * @see #concatMapCompletable(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.FULL)
    public final Completable concatMapCompletableDelayError(Function<? super T, ? extends CompletableSource> mapper, boolean tillTheEnd) {
        return concatMapCompletableDelayError(mapper, tillTheEnd, 2);
    }

    
Maps the upstream items into CompletableSources and subscribes to them one after the other terminates, optionally delaying all errors till both this Flowable and all inner CompletableSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapCompletableDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a CompletableSource to become the next source to be subscribed to
tillTheEnd – If true, errors from this Flowable or any of the inner CompletableSources are delayed until all of them terminate. If false, an error from this Flowable is delayed until the current inner CompletableSource terminates and only then is it emitted to the downstream.
prefetch – The number of upstream items to prefetch so that fresh items are ready to be mapped when a previous CompletableSource terminates. The operator replenishes after half of the prefetch amount has been consumed and turned into CompletableSources.
See Also:
concatMapCompletable(Function, int)
Returns:
a new Completable instance
Since:
2.2/**
     * Maps the upstream items into {@link CompletableSource}s and subscribes to them one after the
     * other terminates, optionally delaying all errors till both this {@code Flowable} and all
     * inner {@code CompletableSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param mapper the function called with the upstream item and should return
     *               a {@code CompletableSource} to become the next source to
     *               be subscribed to
     * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the
     *                   inner {@code CompletableSource}s are delayed until all
     *                   of them terminate. If {@code false}, an error from this
     *                   {@code Flowable} is delayed until the current inner
     *                   {@code CompletableSource} terminates and only then is
     *                   it emitted to the downstream.
     * @param prefetch The number of upstream items to prefetch so that fresh items are
     *                 ready to be mapped when a previous {@code CompletableSource} terminates.
     *                 The operator replenishes after half of the prefetch amount has been consumed
     *                 and turned into {@code CompletableSource}s.
     * @return a new Completable instance
     * @see #concatMapCompletable(Function, int)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.FULL)
    public final Completable concatMapCompletableDelayError(Function<? super T, ? extends CompletableSource> mapper, boolean tillTheEnd, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapCompletable<T>(this, mapper, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY, prefetch));
    }

    
Maps each of the items into a Publisher, subscribes to them one after the other,
one at a time and emits their values in order
while delaying any error from either this or any of the inner Publishers
till all of them terminate.

 
Backpressure:
 
The operator honors backpressure from downstream. Both this and the inner Publishers are expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException. If any of the inner Publishers doesn't honor backpressure, that may throw an IllegalStateException when that Publisher completes.
 
Scheduler:
 
concatMapDelayError does not operate by default on a particular Scheduler.

Params:
mapper – the function that maps the items of this Publisher into the inner Publishers.
Type parameters:
<R> – the result value type
Returns:
the new Publisher instance with the concatenation behavior/**
     * Maps each of the items into a Publisher, subscribes to them one after the other,
     * one at a time and emits their values in order
     * while delaying any error from either this or any of the inner Publishers
     * till all of them terminate.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor
     *  backpressure, that <em>may</em> throw an {@code IllegalStateException} when that
     *  {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the result value type
     * @param mapper the function that maps the items of this Publisher into the inner Publishers.
     * @return the new Publisher instance with the concatenation behavior
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapDelayError(Function<? super T, ? extends Publisher<? extends R>> mapper) {
        return concatMapDelayError(mapper, 2, true);
    }

    
Maps each of the items into a Publisher, subscribes to them one after the other,
one at a time and emits their values in order
while delaying any error from either this or any of the inner Publishers
till all of them terminate.

 
Backpressure:
 
The operator honors backpressure from downstream. Both this and the inner Publishers are expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException. If any of the inner Publishers doesn't honor backpressure, that may throw an IllegalStateException when that Publisher completes.
 
Scheduler:
 
concatMapDelayError does not operate by default on a particular Scheduler.

Params:
mapper – the function that maps the items of this Publisher into the inner Publishers.
prefetch – 
           the number of elements to prefetch from the current Flowable
tillTheEnd – 
           if true, all errors from the outer and inner Publisher sources are delayed until the end,
           if false, an error from the main source is signaled when the current Publisher source terminates
Type parameters:
<R> – the result value type
Returns:
the new Publisher instance with the concatenation behavior/**
     * Maps each of the items into a Publisher, subscribes to them one after the other,
     * one at a time and emits their values in order
     * while delaying any error from either this or any of the inner Publishers
     * till all of them terminate.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both this and the inner {@code Publisher}s are
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}. If any of the inner {@code Publisher}s doesn't honor
     *  backpressure, that <em>may</em> throw an {@code IllegalStateException} when that
     *  {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the result value type
     * @param mapper the function that maps the items of this Publisher into the inner Publishers.
     * @param prefetch
     *            the number of elements to prefetch from the current Flowable
     * @param tillTheEnd
     *            if true, all errors from the outer and inner Publisher sources are delayed until the end,
     *            if false, an error from the main source is signaled when the current Publisher source terminates
     * @return the new Publisher instance with the concatenation behavior
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapDelayError(Function<? super T, ? extends Publisher<? extends R>> mapper,
            int prefetch, boolean tillTheEnd) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        if (this instanceof ScalarCallable) {
            @SuppressWarnings("unchecked")
            T v = ((ScalarCallable<T>)this).call();
            if (v == null) {
                return empty();
            }
            return FlowableScalarXMap.scalarXMap(v, mapper);
        }
        return RxJavaPlugins.onAssembly(new FlowableConcatMap<T, R>(this, mapper, prefetch, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY));
    }

    
Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
Publisher.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
mapper – the function that maps a sequence of values into a sequence of Publishers that will be
              eagerly concatenated
Type parameters:
<R> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
     * Publisher.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
     * order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     *      are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the value type
     * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be
     *               eagerly concatenated
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapEager(Function<? super T, ? extends Publisher<? extends R>> mapper) {
        return concatMapEager(mapper, bufferSize(), bufferSize());
    }

    
Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
Publisher.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
mapper – the function that maps a sequence of values into a sequence of Publishers that will be
              eagerly concatenated
maxConcurrency – the maximum number of concurrent subscribed Publishers
prefetch – hints about the number of expected values from each inner Publisher, must be positive
Type parameters:
<R> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
     * Publisher.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
     * order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     *      are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the value type
     * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be
     *               eagerly concatenated
     * @param maxConcurrency the maximum number of concurrent subscribed Publishers
     * @param prefetch hints about the number of expected values from each inner Publisher, must be positive
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapEager(Function<? super T, ? extends Publisher<? extends R>> mapper,
            int maxConcurrency, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapEager<T, R>(this, mapper, maxConcurrency, prefetch, ErrorMode.IMMEDIATE));
    }

    
Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
Publisher.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
mapper – the function that maps a sequence of values into a sequence of Publishers that will be
              eagerly concatenated
tillTheEnd – 
           if true, all errors from the outer and inner Publisher sources are delayed until the end,
           if false, an error from the main source is signaled when the current Publisher source terminates
Type parameters:
<R> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
     * Publisher.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
     * order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     *      are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the value type
     * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be
     *               eagerly concatenated
     * @param tillTheEnd
     *            if true, all errors from the outer and inner Publisher sources are delayed until the end,
     *            if false, an error from the main source is signaled when the current Publisher source terminates
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapEagerDelayError(Function<? super T, ? extends Publisher<? extends R>> mapper,
            boolean tillTheEnd) {
        return concatMapEagerDelayError(mapper, bufferSize(), bufferSize(), tillTheEnd);
    }

    
Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
Publisher.

Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
order, each one after the previous one completes.

 
Backpressure:
 
Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
mapper – the function that maps a sequence of values into a sequence of Publishers that will be
              eagerly concatenated
maxConcurrency – the maximum number of concurrent subscribed Publishers
prefetch – 
              the number of elements to prefetch from each source Publisher
tillTheEnd – 
              if true, exceptions from the current Flowable and all the inner Publishers are delayed until
              all of them terminate, if false, exception from the current Flowable is delayed until the
              currently running Publisher terminates
Type parameters:
<R> – the value type
Returns:
the new Publisher instance with the specified concatenation behavior
Since:
2.0/**
     * Maps a sequence of values into Publishers and concatenates these Publishers eagerly into a single
     * Publisher.
     * <p>
     * Eager concatenation means that once a subscriber subscribes, this operator subscribes to all of the
     * source Publishers. The operator buffers the values emitted by these Publishers and then drains them in
     * order, each one after the previous one completes.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Backpressure is honored towards the downstream, however, due to the eagerness requirement, sources
     *      are subscribed to in unbounded mode and their values are queued up in an unbounded buffer.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the value type
     * @param mapper the function that maps a sequence of values into a sequence of Publishers that will be
     *               eagerly concatenated
     * @param maxConcurrency the maximum number of concurrent subscribed Publishers
     * @param prefetch
     *               the number of elements to prefetch from each source Publisher
     * @param tillTheEnd
     *               if true, exceptions from the current Flowable and all the inner Publishers are delayed until
     *               all of them terminate, if false, exception from the current Flowable is delayed until the
     *               currently running Publisher terminates
     * @return the new Publisher instance with the specified concatenation behavior
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapEagerDelayError(Function<? super T, ? extends Publisher<? extends R>> mapper,
            int maxConcurrency, int prefetch, boolean tillTheEnd) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapEager<T, R>(this, mapper, maxConcurrency, prefetch, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY));
    }

    
Returns a Flowable that concatenate each item emitted by the source Publisher with the values in an
Iterable corresponding to that item that is generated by a selector.

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers is expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapIterable does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns an Iterable sequence of values for when given an item emitted by the
           source Publisher
Type parameters:
<U> – 
           the type of item emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of concatenating the items emitted by the source Publisher with the values in the Iterables corresponding to those items, as generated by collectionSelector/**
     * Returns a Flowable that concatenate each item emitted by the source Publisher with the values in an
     * Iterable corresponding to that item that is generated by a selector.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s is
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of item emitted by the resulting Publisher
     * @param mapper
     *            a function that returns an Iterable sequence of values for when given an item emitted by the
     *            source Publisher
     * @return a Flowable that emits the results of concatenating the items emitted by the source Publisher with
     *         the values in the Iterables corresponding to those items, as generated by {@code collectionSelector}
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<U> concatMapIterable(Function<? super T, ? extends Iterable<? extends U>> mapper) {
        return concatMapIterable(mapper, 2);
    }

    
Returns a Flowable that concatenate each item emitted by the source Publisher with the values in an
Iterable corresponding to that item that is generated by a selector.

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers is expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapIterable does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns an Iterable sequence of values for when given an item emitted by the
           source Publisher
prefetch – 
           the number of elements to prefetch from the current Flowable
Type parameters:
<U> – 
           the type of item emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of concatenating the items emitted by the source Publisher with the values in the Iterables corresponding to those items, as generated by collectionSelector/**
     * Returns a Flowable that concatenate each item emitted by the source Publisher with the values in an
     * Iterable corresponding to that item that is generated by a selector.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s is
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of item emitted by the resulting Publisher
     * @param mapper
     *            a function that returns an Iterable sequence of values for when given an item emitted by the
     *            source Publisher
     * @param prefetch
     *            the number of elements to prefetch from the current Flowable
     * @return a Flowable that emits the results of concatenating the items emitted by the source Publisher with
     *         the values in the Iterables corresponding to those items, as generated by {@code collectionSelector}
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<U> concatMapIterable(final Function<? super T, ? extends Iterable<? extends U>> mapper, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableFlattenIterable<T, U>(this, mapper, prefetch));
    }

    
Maps the upstream items into MaybeSources and subscribes to them one after the other succeeds or completes, emits their success value if available or terminates immediately if either this Flowable or the current inner MaybeSource fail. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapMaybe does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a MaybeSource to become the next source to be subscribed to
Type parameters:
<R> – the result type of the inner MaybeSources
See Also:
concatMapMaybeDelayError(Function)
concatMapMaybe(Function, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the
     * other succeeds or completes, emits their success value if available or terminates immediately if
     * either this {@code Flowable} or the current inner {@code MaybeSource} fail.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapMaybe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code MaybeSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code MaybeSource} to become the next source to
     *               be subscribed to
     * @return a new Flowable instance
     * @see #concatMapMaybeDelayError(Function)
     * @see #concatMapMaybe(Function, int)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapMaybe(Function<? super T, ? extends MaybeSource<? extends R>> mapper) {
        return concatMapMaybe(mapper, 2);
    }

    
Maps the upstream items into MaybeSources and subscribes to them one after the other succeeds or completes, emits their success value if available or terminates immediately if either this Flowable or the current inner MaybeSource fail. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapMaybe does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a MaybeSource to become the next source to be subscribed to
prefetch – The number of upstream items to prefetch so that fresh items are ready to be mapped when a previous MaybeSource terminates. The operator replenishes after half of the prefetch amount has been consumed and turned into MaybeSources.
Type parameters:
<R> – the result type of the inner MaybeSources
See Also:
concatMapMaybe(Function)
concatMapMaybeDelayError(Function, boolean, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the
     * other succeeds or completes, emits their success value if available or terminates immediately if
     * either this {@code Flowable} or the current inner {@code MaybeSource} fail.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapMaybe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code MaybeSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code MaybeSource} to become the next source to
     *               be subscribed to
     * @param prefetch The number of upstream items to prefetch so that fresh items are
     *                 ready to be mapped when a previous {@code MaybeSource} terminates.
     *                 The operator replenishes after half of the prefetch amount has been consumed
     *                 and turned into {@code MaybeSource}s.
     * @return a new Flowable instance
     * @see #concatMapMaybe(Function)
     * @see #concatMapMaybeDelayError(Function, boolean, int)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapMaybe(Function<? super T, ? extends MaybeSource<? extends R>> mapper, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapMaybe<T, R>(this, mapper, ErrorMode.IMMEDIATE, prefetch));
    }

    
Maps the upstream items into MaybeSources and subscribes to them one after the other terminates, emits their success value if available and delaying all errors till both this Flowable and all inner MaybeSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapMaybeDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a MaybeSource to become the next source to be subscribed to
Type parameters:
<R> – the result type of the inner MaybeSources
See Also:
concatMapMaybe(Function)
concatMapMaybeDelayError(Function, boolean)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the
     * other terminates, emits their success value if available and delaying all errors
     * till both this {@code Flowable} and all inner {@code MaybeSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code MaybeSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code MaybeSource} to become the next source to
     *               be subscribed to
     * @return a new Flowable instance
     * @see #concatMapMaybe(Function)
     * @see #concatMapMaybeDelayError(Function, boolean)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapMaybeDelayError(Function<? super T, ? extends MaybeSource<? extends R>> mapper) {
        return concatMapMaybeDelayError(mapper, true, 2);
    }

    
Maps the upstream items into MaybeSources and subscribes to them one after the other terminates, emits their success value if available and optionally delaying all errors till both this Flowable and all inner MaybeSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapMaybeDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a MaybeSource to become the next source to be subscribed to
tillTheEnd – If true, errors from this Flowable or any of the inner MaybeSources are delayed until all of them terminate. If false, an error from this Flowable is delayed until the current inner MaybeSource terminates and only then is it emitted to the downstream.
Type parameters:
<R> – the result type of the inner MaybeSources
See Also:
concatMapMaybe(Function, int)
concatMapMaybeDelayError(Function, boolean, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the
     * other terminates, emits their success value if available and optionally delaying all errors
     * till both this {@code Flowable} and all inner {@code MaybeSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code MaybeSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code MaybeSource} to become the next source to
     *               be subscribed to
     * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the
     *                   inner {@code MaybeSource}s are delayed until all
     *                   of them terminate. If {@code false}, an error from this
     *                   {@code Flowable} is delayed until the current inner
     *                   {@code MaybeSource} terminates and only then is
     *                   it emitted to the downstream.
     * @return a new Flowable instance
     * @see #concatMapMaybe(Function, int)
     * @see #concatMapMaybeDelayError(Function, boolean, int)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapMaybeDelayError(Function<? super T, ? extends MaybeSource<? extends R>> mapper, boolean tillTheEnd) {
        return concatMapMaybeDelayError(mapper, tillTheEnd, 2);
    }

    
Maps the upstream items into MaybeSources and subscribes to them one after the other terminates, emits their success value if available and optionally delaying all errors till both this Flowable and all inner MaybeSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapMaybeDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a MaybeSource to become the next source to be subscribed to
tillTheEnd – If true, errors from this Flowable or any of the inner MaybeSources are delayed until all of them terminate. If false, an error from this Flowable is delayed until the current inner MaybeSource terminates and only then is it emitted to the downstream.
prefetch – The number of upstream items to prefetch so that fresh items are ready to be mapped when a previous MaybeSource terminates. The operator replenishes after half of the prefetch amount has been consumed and turned into MaybeSources.
Type parameters:
<R> – the result type of the inner MaybeSources
See Also:
concatMapMaybe(Function, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link MaybeSource}s and subscribes to them one after the
     * other terminates, emits their success value if available and optionally delaying all errors
     * till both this {@code Flowable} and all inner {@code MaybeSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code MaybeSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code MaybeSource} to become the next source to
     *               be subscribed to
     * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the
     *                   inner {@code MaybeSource}s are delayed until all
     *                   of them terminate. If {@code false}, an error from this
     *                   {@code Flowable} is delayed until the current inner
     *                   {@code MaybeSource} terminates and only then is
     *                   it emitted to the downstream.
     * @param prefetch The number of upstream items to prefetch so that fresh items are
     *                 ready to be mapped when a previous {@code MaybeSource} terminates.
     *                 The operator replenishes after half of the prefetch amount has been consumed
     *                 and turned into {@code MaybeSource}s.
     * @return a new Flowable instance
     * @see #concatMapMaybe(Function, int)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapMaybeDelayError(Function<? super T, ? extends MaybeSource<? extends R>> mapper, boolean tillTheEnd, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapMaybe<T, R>(this, mapper, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY, prefetch));
    }

    
Maps the upstream items into SingleSources and subscribes to them one after the other succeeds, emits their success values or terminates immediately if either this Flowable or the current inner SingleSource fail. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapSingle does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a SingleSource to become the next source to be subscribed to
Type parameters:
<R> – the result type of the inner SingleSources
See Also:
concatMapSingleDelayError(Function)
concatMapSingle(Function, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the
     * other succeeds, emits their success values or terminates immediately if
     * either this {@code Flowable} or the current inner {@code SingleSource} fail.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapSingle} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code SingleSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code SingleSource} to become the next source to
     *               be subscribed to
     * @return a new Flowable instance
     * @see #concatMapSingleDelayError(Function)
     * @see #concatMapSingle(Function, int)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapSingle(Function<? super T, ? extends SingleSource<? extends R>> mapper) {
        return concatMapSingle(mapper, 2);
    }

    
Maps the upstream items into SingleSources and subscribes to them one after the other succeeds, emits their success values or terminates immediately if either this Flowable or the current inner SingleSource fail. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapSingle does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a SingleSource to become the next source to be subscribed to
prefetch – The number of upstream items to prefetch so that fresh items are ready to be mapped when a previous SingleSource terminates. The operator replenishes after half of the prefetch amount has been consumed and turned into SingleSources.
Type parameters:
<R> – the result type of the inner SingleSources
See Also:
concatMapSingle(Function)
concatMapSingleDelayError(Function, boolean, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the
     * other succeeds, emits their success values or terminates immediately if
     * either this {@code Flowable} or the current inner {@code SingleSource} fail.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapSingle} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code SingleSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code SingleSource} to become the next source to
     *               be subscribed to
     * @param prefetch The number of upstream items to prefetch so that fresh items are
     *                 ready to be mapped when a previous {@code SingleSource} terminates.
     *                 The operator replenishes after half of the prefetch amount has been consumed
     *                 and turned into {@code SingleSource}s.
     * @return a new Flowable instance
     * @see #concatMapSingle(Function)
     * @see #concatMapSingleDelayError(Function, boolean, int)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapSingle(Function<? super T, ? extends SingleSource<? extends R>> mapper, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapSingle<T, R>(this, mapper, ErrorMode.IMMEDIATE, prefetch));
    }

    
Maps the upstream items into SingleSources and subscribes to them one after the other succeeds or fails, emits their success values and delays all errors till both this Flowable and all inner SingleSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapSingleDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a SingleSource to become the next source to be subscribed to
Type parameters:
<R> – the result type of the inner SingleSources
See Also:
concatMapSingle(Function)
concatMapSingleDelayError(Function, boolean)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the
     * other succeeds or fails, emits their success values and delays all errors
     * till both this {@code Flowable} and all inner {@code SingleSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code SingleSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code SingleSource} to become the next source to
     *               be subscribed to
     * @return a new Flowable instance
     * @see #concatMapSingle(Function)
     * @see #concatMapSingleDelayError(Function, boolean)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapSingleDelayError(Function<? super T, ? extends SingleSource<? extends R>> mapper) {
        return concatMapSingleDelayError(mapper, true, 2);
    }

    
Maps the upstream items into SingleSources and subscribes to them one after the other succeeds or fails, emits their success values and optionally delays all errors till both this Flowable and all inner SingleSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapSingleDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a SingleSource to become the next source to be subscribed to
tillTheEnd – If true, errors from this Flowable or any of the inner SingleSources are delayed until all of them terminate. If false, an error from this Flowable is delayed until the current inner SingleSource terminates and only then is it emitted to the downstream.
Type parameters:
<R> – the result type of the inner SingleSources
See Also:
concatMapSingle(Function, int)
concatMapSingleDelayError(Function, boolean, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the
     * other succeeds or fails, emits their success values and optionally delays all errors
     * till both this {@code Flowable} and all inner {@code SingleSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code SingleSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code SingleSource} to become the next source to
     *               be subscribed to
     * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the
     *                   inner {@code SingleSource}s are delayed until all
     *                   of them terminate. If {@code false}, an error from this
     *                   {@code Flowable} is delayed until the current inner
     *                   {@code SingleSource} terminates and only then is
     *                   it emitted to the downstream.
     * @return a new Flowable instance
     * @see #concatMapSingle(Function, int)
     * @see #concatMapSingleDelayError(Function, boolean, int)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapSingleDelayError(Function<? super T, ? extends SingleSource<? extends R>> mapper, boolean tillTheEnd) {
        return concatMapSingleDelayError(mapper, tillTheEnd, 2);
    }

    
Maps the upstream items into SingleSources and subscribes to them one after the other succeeds or fails, emits their success values and optionally delays errors till both this Flowable and all inner SingleSources terminate. 



 
Backpressure:
 
The operator expects the upstream to support backpressure and honors the backpressure from downstream. If this Flowable violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
concatMapSingleDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the upstream item and should return a SingleSource to become the next source to be subscribed to
tillTheEnd – If true, errors from this Flowable or any of the inner SingleSources are delayed until all of them terminate. If false, an error from this Flowable is delayed until the current inner SingleSource terminates and only then is it emitted to the downstream.
prefetch – The number of upstream items to prefetch so that fresh items are ready to be mapped when a previous SingleSource terminates. The operator replenishes after half of the prefetch amount has been consumed and turned into SingleSources.
Type parameters:
<R> – the result type of the inner SingleSources
See Also:
concatMapSingle(Function, int)
Returns:
a new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link SingleSource}s and subscribes to them one after the
     * other succeeds or fails, emits their success values and optionally delays errors
     * till both this {@code Flowable} and all inner {@code SingleSource}s terminate.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the upstream to support backpressure and honors
     *  the backpressure from downstream. If this {@code Flowable} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the result type of the inner {@code SingleSource}s
     * @param mapper the function called with the upstream item and should return
     *               a {@code SingleSource} to become the next source to
     *               be subscribed to
     * @param tillTheEnd If {@code true}, errors from this {@code Flowable} or any of the
     *                   inner {@code SingleSource}s are delayed until all
     *                   of them terminate. If {@code false}, an error from this
     *                   {@code Flowable} is delayed until the current inner
     *                   {@code SingleSource} terminates and only then is
     *                   it emitted to the downstream.
     * @param prefetch The number of upstream items to prefetch so that fresh items are
     *                 ready to be mapped when a previous {@code SingleSource} terminates.
     *                 The operator replenishes after half of the prefetch amount has been consumed
     *                 and turned into {@code SingleSource}s.
     * @return a new Flowable instance
     * @see #concatMapSingle(Function, int)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> concatMapSingleDelayError(Function<? super T, ? extends SingleSource<? extends R>> mapper, boolean tillTheEnd, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowableConcatMapSingle<T, R>(this, mapper, tillTheEnd ? ErrorMode.END : ErrorMode.BOUNDARY, prefetch));
    }

    
Returns a Flowable that emits the items emitted from the current Publisher, then the next, one after
the other, without interleaving them.



 
Backpressure:
 
The operator honors backpressure from downstream. Both this and the other Publishers are expected to honor backpressure as well. If any of then violates this rule, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
concatWith does not operate by default on a particular Scheduler.

Params:
other – 
           a Publisher to be concatenated after the current
See Also:
ReactiveX operators documentation: Concat
Returns:
a Flowable that emits items emitted by the two source Publishers, one after the other,
        without interleaving them/**
     * Returns a Flowable that emits the items emitted from the current Publisher, then the next, one after
     * the other, without interleaving them.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both this and the {@code other} {@code Publisher}s
     *  are expected to honor backpressure as well. If any of then violates this rule, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param other
     *            a Publisher to be concatenated after the current
     * @return a Flowable that emits items emitted by the two source Publishers, one after the other,
     *         without interleaving them
     * @see <a href="http://reactivex.io/documentation/operators/concat.html">ReactiveX operators documentation: Concat</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> concatWith(Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return concat(this, other);
    }

    
Returns a Flowable that emits the items from this Flowable followed by the success item or error event of the other SingleSource. 



 
Backpressure:
 
The operator supports backpressure and makes sure the success item of the other SingleSource is only emitted when there is a demand for it.
 
Scheduler:
 
concatWith does not operate by default on a particular Scheduler.

History: 2.1.10 - experimental
Params:
other – the SingleSource whose signal should be emitted after this Flowable completes normally.
Returns:
the new Flowable instance
Since:
2.2/**
     * Returns a {@code Flowable} that emits the items from this {@code Flowable} followed by the success item or error event
     * of the other {@link SingleSource}.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator supports backpressure and makes sure the success item of the other {@code SingleSource}
     *  is only emitted when there is a demand for it.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.10 - experimental
     * @param other the SingleSource whose signal should be emitted after this {@code Flowable} completes normally.
     * @return the new Flowable instance
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> concatWith(@NonNull SingleSource<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableConcatWithSingle<T>(this, other));
    }

    
Returns a Flowable that emits the items from this Flowable followed by the success item or terminal events of the other MaybeSource. 



 
Backpressure:
 
The operator supports backpressure and makes sure the success item of the other MaybeSource is only emitted when there is a demand for it.
 
Scheduler:
 
concatWith does not operate by default on a particular Scheduler.

History: 2.1.10 - experimental
Params:
other – the MaybeSource whose signal should be emitted after this Flowable completes normally.
Returns:
the new Flowable instance
Since:
2.2/**
     * Returns a {@code Flowable} that emits the items from this {@code Flowable} followed by the success item or terminal events
     * of the other {@link MaybeSource}.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator supports backpressure and makes sure the success item of the other {@code MaybeSource}
     *  is only emitted when there is a demand for it.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.10 - experimental
     * @param other the MaybeSource whose signal should be emitted after this Flowable completes normally.
     * @return the new Flowable instance
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> concatWith(@NonNull MaybeSource<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableConcatWithMaybe<T>(this, other));
    }

    
Returns a Flowable that emits items from this Flowable and when it completes normally, the other CompletableSource is subscribed to and the returned Flowable emits its terminal events. 



 
Backpressure:
 
The operator does not interfere with backpressure between the current Flowable and the downstream consumer (i.e., acts as pass-through). When the operator switches to the Completable, backpressure is no longer present because Completable doesn't have items to apply backpressure to.
 
Scheduler:
 
concatWith does not operate by default on a particular Scheduler.

History: 2.1.10 - experimental
Params:
other – the CompletableSource to subscribe to once the current Flowable completes normally
Returns:
the new Flowable instance
Since:
2.2/**
     * Returns a {@code Flowable} that emits items from this {@code Flowable} and when it completes normally, the
     * other {@link CompletableSource} is subscribed to and the returned {@code Flowable} emits its terminal events.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/concat.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator does not interfere with backpressure between the current Flowable and the
     *  downstream consumer (i.e., acts as pass-through). When the operator switches to the
     *  {@code Completable}, backpressure is no longer present because {@code Completable} doesn't
     *  have items to apply backpressure to.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code concatWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.10 - experimental
     * @param other the {@code CompletableSource} to subscribe to once the current {@code Flowable} completes normally
     * @return the new Flowable instance
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> concatWith(@NonNull CompletableSource other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableConcatWithCompletable<T>(this, other));
    }

    
Returns a Single that emits a Boolean that indicates whether the source Publisher emitted a
specified item.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
contains does not operate by default on a particular Scheduler.

Params:
item – 
           the item to search for in the emissions from the source Publisher
See Also:
ReactiveX operators documentation: Contains
Returns:
a Flowable that emits true if the specified item is emitted by the source Publisher, or false if the source Publisher completes without emitting that item/**
     * Returns a Single that emits a Boolean that indicates whether the source Publisher emitted a
     * specified item.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/contains.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code contains} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item
     *            the item to search for in the emissions from the source Publisher
     * @return a Flowable that emits {@code true} if the specified item is emitted by the source Publisher,
     *         or {@code false} if the source Publisher completes without emitting that item
     * @see <a href="http://reactivex.io/documentation/operators/contains.html">ReactiveX operators documentation: Contains</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<Boolean> contains(final Object item) {
        ObjectHelper.requireNonNull(item, "item is null");
        return any(Functions.equalsWith(item));
    }

    
Returns a Single that counts the total number of items emitted by the source Publisher and emits
this count as a 64-bit Long.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
count does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Count
count()
Returns:
a Single that emits a single item: the number of items emitted by the source Publisher as a
        64-bit Long item/**
     * Returns a Single that counts the total number of items emitted by the source Publisher and emits
     * this count as a 64-bit Long.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/longCount.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code count} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Single that emits a single item: the number of items emitted by the source Publisher as a
     *         64-bit Long item
     * @see <a href="http://reactivex.io/documentation/operators/count.html">ReactiveX operators documentation: Count</a>
     * @see #count()
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<Long> count() {
        return RxJavaPlugins.onAssembly(new FlowableCountSingle<T>(this));
    }

    
Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
source Publisher that are followed by another item within a computed debounce duration.


 The delivery of the item happens on the thread of the first onNext or onComplete signal of the generated Publisher sequence, which if takes too long, a newer item may arrive from the upstream, causing the generated sequence to get cancelled, which may also interrupt any downstream blocking operation (yielding an InterruptedException). It is recommended processing items that may take long time to be moved to another thread via observeOn applied after debounce itself. 

 
Backpressure:
 
This operator does not support backpressure as it uses the debounceSelector to mark boundaries.
 
Scheduler:
 
This version of debounce does not operate by default on a particular Scheduler.

Params:
debounceIndicator – 
           function to retrieve a sequence that indicates the throttle duration for each item
Type parameters:
<U> – 
           the debounce value type (ignored)
See Also:
ReactiveX operators documentation: Debounce
RxJava wiki: Backpressure
Returns:
a Flowable that omits items emitted by the source Publisher that are followed by another item
        within a computed debounce duration/**
     * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
     * source Publisher that are followed by another item within a computed debounce duration.
     * <p>
     * <img width="640" height="425" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/debounce.f.png" alt="">
     * <p>
     * The delivery of the item happens on the thread of the first {@code onNext} or {@code onComplete}
     * signal of the generated {@code Publisher} sequence,
     * which if takes too long, a newer item may arrive from the upstream, causing the
     * generated sequence to get cancelled, which may also interrupt any downstream blocking operation
     * (yielding an {@code InterruptedException}). It is recommended processing items
     * that may take long time to be moved to another thread via {@link #observeOn} applied after
     * {@code debounce} itself.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses the {@code debounceSelector} to mark
     *      boundaries.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code debounce} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the debounce value type (ignored)
     * @param debounceIndicator
     *            function to retrieve a sequence that indicates the throttle duration for each item
     * @return a Flowable that omits items emitted by the source Publisher that are followed by another item
     *         within a computed debounce duration
     * @see <a href="http://reactivex.io/documentation/operators/debounce.html">ReactiveX operators documentation: Debounce</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<T> debounce(Function<? super T, ? extends Publisher<U>> debounceIndicator) {
        ObjectHelper.requireNonNull(debounceIndicator, "debounceIndicator is null");
        return RxJavaPlugins.onAssembly(new FlowableDebounce<T, U>(this, debounceIndicator));
    }

    
Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
source Publisher that are followed by newer items before a timeout value expires. The timer resets on
each emission.

Note: If items keep being emitted by the source Publisher faster than the timeout then no items
will be emitted by the resulting Publisher.


 Delivery of the item after the grace period happens on the computation Scheduler's Worker which if takes too long, a newer item may arrive from the upstream, causing the Worker's task to get disposed, which may also interrupt any downstream blocking operation (yielding an InterruptedException). It is recommended processing items that may take long time to be moved to another thread via observeOn applied after debounce itself. 

 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
debounce operates by default on the computation Scheduler.

Params:
timeout – 
           the length of the window of time that must pass after the emission of an item from the source
           Publisher in which that Publisher emits no items in order for the item to be emitted by the
           resulting Publisher
unit –  the unit of time for the specified timeout
See Also:
ReactiveX operators documentation: Debounce
RxJava wiki: Backpressure
throttleWithTimeout(long, TimeUnit)
Returns:
a Flowable that filters out items from the source Publisher that are too quickly followed by
        newer items/**
     * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
     * source Publisher that are followed by newer items before a timeout value expires. The timer resets on
     * each emission.
     * <p>
     * <em>Note:</em> If items keep being emitted by the source Publisher faster than the timeout then no items
     * will be emitted by the resulting Publisher.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/debounce.png" alt="">
     * <p>
     * Delivery of the item after the grace period happens on the {@code computation} {@code Scheduler}'s
     * {@code Worker} which if takes too long, a newer item may arrive from the upstream, causing the
     * {@code Worker}'s task to get disposed, which may also interrupt any downstream blocking operation
     * (yielding an {@code InterruptedException}). It is recommended processing items
     * that may take long time to be moved to another thread via {@link #observeOn} applied after
     * {@code debounce} itself.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code debounce} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timeout
     *            the length of the window of time that must pass after the emission of an item from the source
     *            Publisher in which that Publisher emits no items in order for the item to be emitted by the
     *            resulting Publisher
     * @param unit
     *            the unit of time for the specified {@code timeout}
     * @return a Flowable that filters out items from the source Publisher that are too quickly followed by
     *         newer items
     * @see <a href="http://reactivex.io/documentation/operators/debounce.html">ReactiveX operators documentation: Debounce</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #throttleWithTimeout(long, TimeUnit)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> debounce(long timeout, TimeUnit unit) {
        return debounce(timeout, unit, Schedulers.computation());
    }

    
Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
source Publisher that are followed by newer items before a timeout value expires on a specified
Scheduler. The timer resets on each emission.

Note: If items keep being emitted by the source Publisher faster than the timeout then no items
will be emitted by the resulting Publisher.


 Delivery of the item after the grace period happens on the given Scheduler's Worker which if takes too long, a newer item may arrive from the upstream, causing the Worker's task to get disposed, which may also interrupt any downstream blocking operation (yielding an InterruptedException). It is recommended processing items that may take long time to be moved to another thread via observeOn applied after debounce itself. 

 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timeout – 
           the time each item has to be "the most recent" of those emitted by the source Publisher to
           ensure that it's not dropped
unit –  the unit of time for the specified timeout
scheduler –  the Scheduler to use internally to manage the timers that handle the timeout for each item
See Also:
ReactiveX operators documentation: Debounce
RxJava wiki: Backpressure
throttleWithTimeout(long, TimeUnit, Scheduler)
Returns:
a Flowable that filters out items from the source Publisher that are too quickly followed by
        newer items/**
     * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
     * source Publisher that are followed by newer items before a timeout value expires on a specified
     * Scheduler. The timer resets on each emission.
     * <p>
     * <em>Note:</em> If items keep being emitted by the source Publisher faster than the timeout then no items
     * will be emitted by the resulting Publisher.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/debounce.s.png" alt="">
     * <p>
     * Delivery of the item after the grace period happens on the given {@code Scheduler}'s
     * {@code Worker} which if takes too long, a newer item may arrive from the upstream, causing the
     * {@code Worker}'s task to get disposed, which may also interrupt any downstream blocking operation
     * (yielding an {@code InterruptedException}). It is recommended processing items
     * that may take long time to be moved to another thread via {@link #observeOn} applied after
     * {@code debounce} itself.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timeout
     *            the time each item has to be "the most recent" of those emitted by the source Publisher to
     *            ensure that it's not dropped
     * @param unit
     *            the unit of time for the specified {@code timeout}
     * @param scheduler
     *            the {@link Scheduler} to use internally to manage the timers that handle the timeout for each
     *            item
     * @return a Flowable that filters out items from the source Publisher that are too quickly followed by
     *         newer items
     * @see <a href="http://reactivex.io/documentation/operators/debounce.html">ReactiveX operators documentation: Debounce</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #throttleWithTimeout(long, TimeUnit, Scheduler)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> debounce(long timeout, TimeUnit unit, Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableDebounceTimed<T>(this, timeout, unit, scheduler));
    }

    
Returns a Flowable that emits the items emitted by the source Publisher or a specified default item
if the source Publisher is empty.



 
Backpressure:
 
If the source Publisher is empty, this operator is guaranteed to honor backpressure from downstream. If the source Publisher is non-empty, it is expected to honor backpressure as well; if the rule is violated, a MissingBackpressureException may get signaled somewhere downstream.
 
 
Scheduler:
 
defaultIfEmpty does not operate by default on a particular Scheduler.

Params:
defaultItem – 
           the item to emit if the source Publisher emits no items
See Also:
ReactiveX operators documentation: DefaultIfEmpty
Returns:
a Flowable that emits either the specified default item if the source Publisher emits no
        items, or the items emitted by the source Publisher/**
     * Returns a Flowable that emits the items emitted by the source Publisher or a specified default item
     * if the source Publisher is empty.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/defaultIfEmpty.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>If the source {@code Publisher} is empty, this operator is guaranteed to honor backpressure from downstream.
     *  If the source {@code Publisher} is non-empty, it is expected to honor backpressure as well; if the rule is violated,
     *  a {@code MissingBackpressureException} <em>may</em> get signaled somewhere downstream.
     *  </dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code defaultIfEmpty} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param defaultItem
     *            the item to emit if the source Publisher emits no items
     * @return a Flowable that emits either the specified default item if the source Publisher emits no
     *         items, or the items emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/defaultifempty.html">ReactiveX operators documentation: DefaultIfEmpty</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> defaultIfEmpty(T defaultItem) {
        ObjectHelper.requireNonNull(defaultItem, "defaultItem is null");
        return switchIfEmpty(just(defaultItem));
    }

    
Returns a Flowable that delays the emissions of the source Publisher via another Publisher on a
per-item basis.



Note: the resulting Publisher will immediately propagate any onError notification from the source Publisher. 

 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher. All of the other Publishers supplied by the function are consumed in an unbounded manner (i.e., no backpressure applied to them).
 
Scheduler:
 
This version of delay does not operate by default on a particular Scheduler.

Params:
itemDelayIndicator –  a function that returns a Publisher for each item emitted by the source Publisher, which is then used to delay the emission of that item by the resulting Publisher until the Publisher returned from itemDelay emits an item
Type parameters:
<U> – 
           the item delay value type (ignored)
See Also:
ReactiveX operators documentation: Delay
Returns:
a Flowable that delays the emissions of the source Publisher via another Publisher on a
        per-item basis/**
     * Returns a Flowable that delays the emissions of the source Publisher via another Publisher on a
     * per-item basis.
     * <p>
     * <img width="640" height="450" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delay.o.png" alt="">
     * <p>
     * <em>Note:</em> the resulting Publisher will immediately propagate any {@code onError} notification
     * from the source Publisher.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
     *  All of the other {@code Publisher}s supplied by the function are consumed
     *  in an unbounded manner (i.e., no backpressure applied to them).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code delay} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the item delay value type (ignored)
     * @param itemDelayIndicator
     *            a function that returns a Publisher for each item emitted by the source Publisher, which is
     *            then used to delay the emission of that item by the resulting Publisher until the Publisher
     *            returned from {@code itemDelay} emits an item
     * @return a Flowable that delays the emissions of the source Publisher via another Publisher on a
     *         per-item basis
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<T> delay(final Function<? super T, ? extends Publisher<U>> itemDelayIndicator) {
        ObjectHelper.requireNonNull(itemDelayIndicator, "itemDelayIndicator is null");
        return flatMap(FlowableInternalHelper.itemDelay(itemDelayIndicator));
    }

    
Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a
specified delay. Error notifications from the source Publisher are not delayed.



 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher.
 
Scheduler:
 
This version of delay operates by default on the computation Scheduler.

Params:
delay – 
           the delay to shift the source by
unit –  the TimeUnit in which period is defined
See Also:
ReactiveX operators documentation: Delay
Returns:
the source Publisher shifted in time by the specified delay/**
     * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a
     * specified delay. Error notifications from the source Publisher are not delayed.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delay.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code delay} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param delay
     *            the delay to shift the source by
     * @param unit
     *            the {@link TimeUnit} in which {@code period} is defined
     * @return the source Publisher shifted in time by the specified delay
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> delay(long delay, TimeUnit unit) {
        return delay(delay, unit, Schedulers.computation(), false);
    }

    
Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a specified delay. If delayError is true, error notifications will also be delayed. 



 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher.
 
Scheduler:
 
This version of delay operates by default on the computation Scheduler.

Params:
delay – 
           the delay to shift the source by
unit –  the TimeUnit in which period is defined
delayError – 
           if true, the upstream exception is signaled with the given delay, after all preceding normal elements,
           if false, the upstream exception is signaled immediately
See Also:
ReactiveX operators documentation: Delay
Returns:
the source Publisher shifted in time by the specified delay/**
     * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a
     * specified delay. If {@code delayError} is true, error notifications will also be delayed.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delay.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code delay} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param delay
     *            the delay to shift the source by
     * @param unit
     *            the {@link TimeUnit} in which {@code period} is defined
     * @param delayError
     *            if true, the upstream exception is signaled with the given delay, after all preceding normal elements,
     *            if false, the upstream exception is signaled immediately
     * @return the source Publisher shifted in time by the specified delay
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> delay(long delay, TimeUnit unit, boolean delayError) {
        return delay(delay, unit, Schedulers.computation(), delayError);
    }

    
Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a
specified delay. Error notifications from the source Publisher are not delayed.



 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
delay – 
           the delay to shift the source by
unit –  the time unit of delay
scheduler –  the Scheduler to use for delaying
See Also:
ReactiveX operators documentation: Delay
Returns:
the source Publisher shifted in time by the specified delay/**
     * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a
     * specified delay. Error notifications from the source Publisher are not delayed.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delay.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param delay
     *            the delay to shift the source by
     * @param unit
     *            the time unit of {@code delay}
     * @param scheduler
     *            the {@link Scheduler} to use for delaying
     * @return the source Publisher shifted in time by the specified delay
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> delay(long delay, TimeUnit unit, Scheduler scheduler) {
        return delay(delay, unit, scheduler, false);
    }

    
Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a specified delay. If delayError is true, error notifications will also be delayed. 



 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
delay – 
           the delay to shift the source by
unit –  the time unit of delay
scheduler –  the Scheduler to use for delaying
delayError – 
           if true, the upstream exception is signaled with the given delay, after all preceding normal elements,
           if false, the upstream exception is signaled immediately
See Also:
ReactiveX operators documentation: Delay
Returns:
the source Publisher shifted in time by the specified delay/**
     * Returns a Flowable that emits the items emitted by the source Publisher shifted forward in time by a
     * specified delay. If {@code delayError} is true, error notifications will also be delayed.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delay.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param delay
     *            the delay to shift the source by
     * @param unit
     *            the time unit of {@code delay}
     * @param scheduler
     *            the {@link Scheduler} to use for delaying
     * @param delayError
     *            if true, the upstream exception is signaled with the given delay, after all preceding normal elements,
     *            if false, the upstream exception is signaled immediately
     * @return the source Publisher shifted in time by the specified delay
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> delay(long delay, TimeUnit unit, Scheduler scheduler, boolean delayError) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");

        return RxJavaPlugins.onAssembly(new FlowableDelay<T>(this, Math.max(0L, delay), unit, scheduler, delayError));
    }

    
Returns a Flowable that delays the subscription to and emissions from the source Publisher via another
Publisher on a per-item basis.



Note: the resulting Publisher will immediately propagate any onError notification from the source Publisher. 

 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher. All of the other Publishers supplied by the functions are consumed in an unbounded manner (i.e., no backpressure applied to them).
 
Scheduler:
 
This version of delay does not operate by default on a particular Scheduler.

Params:
subscriptionIndicator – 
           a function that returns a Publisher that triggers the subscription to the source Publisher
           once it emits any item
itemDelayIndicator –  a function that returns a Publisher for each item emitted by the source Publisher, which is then used to delay the emission of that item by the resulting Publisher until the Publisher returned from itemDelay emits an item
Type parameters:
<U> – 
           the subscription delay value type (ignored)
<V> – 
           the item delay value type (ignored)
See Also:
ReactiveX operators documentation: Delay
Returns:
a Flowable that delays the subscription and emissions of the source Publisher via another
        Publisher on a per-item basis/**
     * Returns a Flowable that delays the subscription to and emissions from the source Publisher via another
     * Publisher on a per-item basis.
     * <p>
     * <img width="640" height="450" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delay.oo.png" alt="">
     * <p>
     * <em>Note:</em> the resulting Publisher will immediately propagate any {@code onError} notification
     * from the source Publisher.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.
     *  All of the other {@code Publisher}s supplied by the functions are consumed
     *  in an unbounded manner (i.e., no backpressure applied to them).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code delay} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the subscription delay value type (ignored)
     * @param <V>
     *            the item delay value type (ignored)
     * @param subscriptionIndicator
     *            a function that returns a Publisher that triggers the subscription to the source Publisher
     *            once it emits any item
     * @param itemDelayIndicator
     *            a function that returns a Publisher for each item emitted by the source Publisher, which is
     *            then used to delay the emission of that item by the resulting Publisher until the Publisher
     *            returned from {@code itemDelay} emits an item
     * @return a Flowable that delays the subscription and emissions of the source Publisher via another
     *         Publisher on a per-item basis
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, V> Flowable<T> delay(Publisher<U> subscriptionIndicator,
            Function<? super T, ? extends Publisher<V>> itemDelayIndicator) {
        return delaySubscription(subscriptionIndicator).delay(itemDelayIndicator);
    }

    
Returns a Flowable that delays the subscription to this Publisher
until the other Publisher emits an element or completes normally.

 
Backpressure:
 
The operator forwards the backpressure requests to this Publisher once
 the subscription happens and requests Long.MAX_VALUE from the other Publisher
 
Scheduler:
 
This method does not operate by default on a particular Scheduler.

Params:
subscriptionIndicator – the other Publisher that should trigger the subscription
       to this Publisher.
Type parameters:
<U> – the value type of the other Publisher, irrelevant
Returns:
a Flowable that delays the subscription to this Publisher
        until the other Publisher emits an element or completes normally.
Since:
2.0/**
     * Returns a Flowable that delays the subscription to this Publisher
     * until the other Publisher emits an element or completes normally.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator forwards the backpressure requests to this Publisher once
     *  the subscription happens and requests Long.MAX_VALUE from the other Publisher</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This method does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the value type of the other Publisher, irrelevant
     * @param subscriptionIndicator the other Publisher that should trigger the subscription
     *        to this Publisher.
     * @return a Flowable that delays the subscription to this Publisher
     *         until the other Publisher emits an element or completes normally.
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<T> delaySubscription(Publisher<U> subscriptionIndicator) {
        ObjectHelper.requireNonNull(subscriptionIndicator, "subscriptionIndicator is null");
        return RxJavaPlugins.onAssembly(new FlowableDelaySubscriptionOther<T, U>(this, subscriptionIndicator));
    }

    
Returns a Flowable that delays the subscription to the source Publisher by a given amount of time.



 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher.
 
Scheduler:
 
This version of delaySubscription operates by default on the computation Scheduler.

Params:
delay – 
           the time to delay the subscription
unit –  the time unit of delay
See Also:
ReactiveX operators documentation: Delay
Returns:
a Flowable that delays the subscription to the source Publisher by the given amount/**
     * Returns a Flowable that delays the subscription to the source Publisher by a given amount of time.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delaySubscription.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code delaySubscription} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param delay
     *            the time to delay the subscription
     * @param unit
     *            the time unit of {@code delay}
     * @return a Flowable that delays the subscription to the source Publisher by the given amount
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> delaySubscription(long delay, TimeUnit unit) {
        return delaySubscription(delay, unit, Schedulers.computation());
    }

    
Returns a Flowable that delays the subscription to the source Publisher by a given amount of time,
both waiting and subscribing on a given Scheduler.



 
Backpressure:
 
The operator doesn't interfere with the backpressure behavior which is determined by the source Publisher.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
delay – 
           the time to delay the subscription
unit –  the time unit of delay
scheduler – 
           the Scheduler on which the waiting and subscription will happen
See Also:
ReactiveX operators documentation: Delay
Returns:
a Flowable that delays the subscription to the source Publisher by a given
        amount, waiting and subscribing on the given Scheduler/**
     * Returns a Flowable that delays the subscription to the source Publisher by a given amount of time,
     * both waiting and subscribing on a given Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/delaySubscription.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with the backpressure behavior which is determined by the source {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param delay
     *            the time to delay the subscription
     * @param unit
     *            the time unit of {@code delay}
     * @param scheduler
     *            the Scheduler on which the waiting and subscription will happen
     * @return a Flowable that delays the subscription to the source Publisher by a given
     *         amount, waiting and subscribing on the given Scheduler
     * @see <a href="http://reactivex.io/documentation/operators/delay.html">ReactiveX operators documentation: Delay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> delaySubscription(long delay, TimeUnit unit, Scheduler scheduler) {
        return delaySubscription(timer(delay, unit, scheduler));
    }

    
Returns a Flowable that reverses the effect of materialize by transforming the Notification objects emitted by the source Publisher into the items or notifications they represent. 


 When the upstream signals an onError or onComplete item, the returned Flowable cancels the flow and terminates with that type of terminal event: 

Flowable.just(createOnNext(1), createOnComplete(), createOnNext(2))
.doOnCancel(() -> System.out.println("Cancelled!"));
.dematerialize()
.test()
.assertResult(1);
 If the upstream signals onError or onComplete directly, the flow is terminated with the same event. 

Flowable.just(createOnNext(1), createOnNext(2))
.dematerialize()
.test()
.assertResult(1, 2);
 If this behavior is not desired, the completion can be suppressed by applying concatWith(Publisher) with a never() source. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
dematerialize does not operate by default on a particular Scheduler.

Type parameters:
<T2> – the output value type
See Also:
ReactiveX operators documentation: Dematerialize
dematerialize(Function)
Returns:
a Flowable that emits the items and notifications embedded in the Notification objects emitted by the source Publisher
Deprecated:
in 2.2.4; inherently type-unsafe as it overrides the output generic type. Use dematerialize(Function) instead./**
     * Returns a Flowable that reverses the effect of {@link #materialize materialize} by transforming the
     * {@link Notification} objects emitted by the source Publisher into the items or notifications they
     * represent.
     * <p>
     * <img width="640" height="335" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/dematerialize.png" alt="">
     * <p>
     * When the upstream signals an {@link Notification#createOnError(Throwable) onError} or
     * {@link Notification#createOnComplete() onComplete} item, the
     * returned Flowable cancels the flow and terminates with that type of terminal event:
     * <pre><code>
     * Flowable.just(createOnNext(1), createOnComplete(), createOnNext(2))
     * .doOnCancel(() -&gt; System.out.println("Cancelled!"));
     * .dematerialize()
     * .test()
     * .assertResult(1);
     * </code></pre>
     * If the upstream signals {@code onError} or {@code onComplete} directly, the flow is terminated
     * with the same event.
     * <pre><code>
     * Flowable.just(createOnNext(1), createOnNext(2))
     * .dematerialize()
     * .test()
     * .assertResult(1, 2);
     * </code></pre>
     * If this behavior is not desired, the completion can be suppressed by applying {@link #concatWith(Publisher)}
     * with a {@link #never()} source.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code dematerialize} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T2> the output value type
     * @return a Flowable that emits the items and notifications embedded in the {@link Notification} objects
     *         emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/materialize-dematerialize.html">ReactiveX operators documentation: Dematerialize</a>
     * @see #dematerialize(Function)
     * @deprecated in 2.2.4; inherently type-unsafe as it overrides the output generic type. Use {@link #dematerialize(Function)} instead.
     */
    @CheckReturnValue
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @Deprecated
    @SuppressWarnings({ "unchecked", "rawtypes" })
    public final <T2> Flowable<T2> dematerialize() {
        return RxJavaPlugins.onAssembly(new FlowableDematerialize(this, Functions.identity()));
    }

    
Returns a Flowable that reverses the effect of materialize by transforming the Notification objects extracted from the source items via a selector function into their respective Subscriber signal types. 


 The intended use of the selector function is to perform a type-safe identity mapping (see example) on a source that is already of type Notification<T>. The Java language doesn't allow limiting instance methods to a certain generic argument shape, therefore, a function is used to ensure the conversion remains type safe. 
 When the upstream signals an onError or onComplete item, the returned Flowable cancels of the flow and terminates with that type of terminal event: 

Flowable.just(createOnNext(1), createOnComplete(), createOnNext(2))
.doOnCancel(() -> System.out.println("Canceled!"));
.dematerialize(notification -> notification)
.test()
.assertResult(1);
 If the upstream signals onError or onComplete directly, the flow is terminated with the same event. 

Flowable.just(createOnNext(1), createOnNext(2))
.dematerialize(notification -> notification)
.test()
.assertResult(1, 2);
 If this behavior is not desired, the completion can be suppressed by applying concatWith(Publisher) with a never() source. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
dematerialize does not operate by default on a particular Scheduler.

Params:
selector – function that returns the upstream item and should return a Notification to signal the corresponding Subscriber event to the downstream.
Type parameters:
<R> – the output value type
See Also:
ReactiveX operators documentation: Dematerialize
Returns:
a Flowable that emits the items and notifications embedded in the Notification objects selected from the items emitted by the source Flowable
Since:
2.2.4 - experimental/**
     * Returns a Flowable that reverses the effect of {@link #materialize materialize} by transforming the
     * {@link Notification} objects extracted from the source items via a selector function
     * into their respective {@code Subscriber} signal types.
     * <p>
     * <img width="640" height="335" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/dematerialize.png" alt="">
     * <p>
     * The intended use of the {@code selector} function is to perform a
     * type-safe identity mapping (see example) on a source that is already of type
     * {@code Notification<T>}. The Java language doesn't allow
     * limiting instance methods to a certain generic argument shape, therefore,
     * a function is used to ensure the conversion remains type safe.
     * <p>
     * When the upstream signals an {@link Notification#createOnError(Throwable) onError} or
     * {@link Notification#createOnComplete() onComplete} item, the
     * returned Flowable cancels of the flow and terminates with that type of terminal event:
     * <pre><code>
     * Flowable.just(createOnNext(1), createOnComplete(), createOnNext(2))
     * .doOnCancel(() -&gt; System.out.println("Canceled!"));
     * .dematerialize(notification -&gt; notification)
     * .test()
     * .assertResult(1);
     * </code></pre>
     * If the upstream signals {@code onError} or {@code onComplete} directly, the flow is terminated
     * with the same event.
     * <pre><code>
     * Flowable.just(createOnNext(1), createOnNext(2))
     * .dematerialize(notification -&gt; notification)
     * .test()
     * .assertResult(1, 2);
     * </code></pre>
     * If this behavior is not desired, the completion can be suppressed by applying {@link #concatWith(Publisher)}
     * with a {@link #never()} source.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code dematerialize} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the output value type
     * @param selector function that returns the upstream item and should return a Notification to signal
     * the corresponding {@code Subscriber} event to the downstream.
     * @return a Flowable that emits the items and notifications embedded in the {@link Notification} objects
     *         selected from the items emitted by the source Flowable
     * @see <a href="http://reactivex.io/documentation/operators/materialize-dematerialize.html">ReactiveX operators documentation: Dematerialize</a>
     * @since 2.2.4 - experimental
     */
    @Experimental
    @CheckReturnValue
    @NonNull
    @SchedulerSupport(SchedulerSupport.NONE)
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    public final <R> Flowable<R> dematerialize(Function<? super T, Notification<R>> selector) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        return RxJavaPlugins.onAssembly(new FlowableDematerialize<T, R>(this, selector));
    }

    
Returns a Flowable that emits all items emitted by the source Publisher that are distinct based on Object.equals(Object) comparison. 


 It is recommended the elements' class T in the flow overrides the default Object.equals() and Object.hashCode() to provide a meaningful comparison between items as the default Java implementation only considers reference equivalence. 
 By default, distinct() uses an internal HashSet per Subscriber to remember previously seen items and uses Set.add(Object) returning false as the indicator for duplicates. 
 Note that this internal HashSet may grow unbounded as items won't be removed from it by the operator. Therefore, using very long or infinite upstream (with very distinct elements) may lead to OutOfMemoryError. 
 Customizing the retention policy can happen only by providing a custom Collection implementation to the distinct(Function, Callable) overload. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
distinct does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Distinct
distinct(Function)
distinct(Function, Callable)
Returns:
a Flowable that emits only those items emitted by the source Publisher that are distinct from
        each other/**
     * Returns a Flowable that emits all items emitted by the source Publisher that are distinct
     * based on {@link Object#equals(Object)} comparison.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/distinct.png" alt="">
     * <p>
     * It is recommended the elements' class {@code T} in the flow overrides the default {@code Object.equals()} and {@link Object#hashCode()} to provide
     * a meaningful comparison between items as the default Java implementation only considers reference equivalence.
     * <p>
     * By default, {@code distinct()} uses an internal {@link java.util.HashSet} per Subscriber to remember
     * previously seen items and uses {@link java.util.Set#add(Object)} returning {@code false} as the
     * indicator for duplicates.
     * <p>
     * Note that this internal {@code HashSet} may grow unbounded as items won't be removed from it by
     * the operator. Therefore, using very long or infinite upstream (with very distinct elements) may lead
     * to {@code OutOfMemoryError}.
     * <p>
     * Customizing the retention policy can happen only by providing a custom {@link java.util.Collection} implementation
     * to the {@link #distinct(Function, Callable)} overload.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code distinct} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits only those items emitted by the source Publisher that are distinct from
     *         each other
     * @see <a href="http://reactivex.io/documentation/operators/distinct.html">ReactiveX operators documentation: Distinct</a>
     * @see #distinct(Function)
     * @see #distinct(Function, Callable)
     */
    @SuppressWarnings({ "rawtypes", "unchecked" })
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> distinct() {
        return distinct((Function)Functions.identity(), Functions.<T>createHashSet());
    }

    
Returns a Flowable that emits all items emitted by the source Publisher that are distinct according to a key selector function and based on Object.equals(Object) comparison of the objects returned by the key selector function. 


 It is recommended the keys' class K overrides the default Object.equals() and Object.hashCode() to provide a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence. 
 By default, distinct() uses an internal HashSet per Subscriber to remember previously seen keys and uses Set.add(Object) returning false as the indicator for duplicates. 
 Note that this internal HashSet may grow unbounded as keys won't be removed from it by the operator. Therefore, using very long or infinite upstream (with very distinct keys) may lead to OutOfMemoryError. 
 Customizing the retention policy can happen only by providing a custom Collection implementation to the distinct(Function, Callable) overload. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
distinct does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that projects an emitted item to a key value that is used to decide whether an item
           is distinct from another one or not
Type parameters:
<K> – the key type
See Also:
ReactiveX operators documentation: Distinct
distinct(Function, Callable)
Returns:
a Flowable that emits those items emitted by the source Publisher that have distinct keys/**
     * Returns a Flowable that emits all items emitted by the source Publisher that are distinct according
     * to a key selector function and based on {@link Object#equals(Object)} comparison of the objects
     * returned by the key selector function.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/distinct.key.png" alt="">
     * <p>
     * It is recommended the keys' class {@code K} overrides the default {@code Object.equals()} and {@link Object#hashCode()} to provide
     * a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence.
     * <p>
     * By default, {@code distinct()} uses an internal {@link java.util.HashSet} per Subscriber to remember
     * previously seen keys and uses {@link java.util.Set#add(Object)} returning {@code false} as the
     * indicator for duplicates.
     * <p>
     * Note that this internal {@code HashSet} may grow unbounded as keys won't be removed from it by
     * the operator. Therefore, using very long or infinite upstream (with very distinct keys) may lead
     * to {@code OutOfMemoryError}.
     * <p>
     * Customizing the retention policy can happen only by providing a custom {@link java.util.Collection} implementation
     * to the {@link #distinct(Function, Callable)} overload.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code distinct} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type
     * @param keySelector
     *            a function that projects an emitted item to a key value that is used to decide whether an item
     *            is distinct from another one or not
     * @return a Flowable that emits those items emitted by the source Publisher that have distinct keys
     * @see <a href="http://reactivex.io/documentation/operators/distinct.html">ReactiveX operators documentation: Distinct</a>
     * @see #distinct(Function, Callable)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K> Flowable<T> distinct(Function<? super T, K> keySelector) {
        return distinct(keySelector, Functions.<K>createHashSet());
    }

    
Returns a Flowable that emits all items emitted by the source Publisher that are distinct according to a key selector function and based on Object.equals(Object) comparison of the objects returned by the key selector function. 


 It is recommended the keys' class K overrides the default Object.equals() and Object.hashCode() to provide a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
distinct does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that projects an emitted item to a key value that is used to decide whether an item
           is distinct from another one or not
collectionSupplier – 
           function called for each individual Subscriber to return a Collection subtype for holding the extracted
           keys and whose add() method's return indicates uniqueness.
Type parameters:
<K> – the key type
See Also:
ReactiveX operators documentation: Distinct
Returns:
a Flowable that emits those items emitted by the source Publisher that have distinct keys/**
     * Returns a Flowable that emits all items emitted by the source Publisher that are distinct according
     * to a key selector function and based on {@link Object#equals(Object)} comparison of the objects
     * returned by the key selector function.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/distinct.key.png" alt="">
     * <p>
     * It is recommended the keys' class {@code K} overrides the default {@code Object.equals()} and {@link Object#hashCode()} to provide
     * a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code distinct} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type
     * @param keySelector
     *            a function that projects an emitted item to a key value that is used to decide whether an item
     *            is distinct from another one or not
     * @param collectionSupplier
     *            function called for each individual Subscriber to return a Collection subtype for holding the extracted
     *            keys and whose add() method's return indicates uniqueness.
     * @return a Flowable that emits those items emitted by the source Publisher that have distinct keys
     * @see <a href="http://reactivex.io/documentation/operators/distinct.html">ReactiveX operators documentation: Distinct</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K> Flowable<T> distinct(Function<? super T, K> keySelector,
            Callable<? extends Collection<? super K>> collectionSupplier) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        ObjectHelper.requireNonNull(collectionSupplier, "collectionSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableDistinct<T, K>(this, keySelector, collectionSupplier));
    }

    
Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their immediate predecessors based on Object.equals(Object) comparison. 


 It is recommended the elements' class T in the flow overrides the default Object.equals() to provide a meaningful comparison between items as the default Java implementation only considers reference equivalence. Alternatively, use the distinctUntilChanged(BiPredicate) overload and provide a comparison function in case the class T can't be overridden with custom equals() or the comparison itself should happen on different terms or properties of the class T. 

Note that the operator always retains the latest item from upstream regardless of the comparison result
and uses it in the next comparison with the next upstream item.
 Note that if element type T in the flow is mutable, the comparison of the previous and current item may yield unexpected results if the items are mutated externally. Common cases are mutable CharSequences or Lists where the objects will actually have the same references when they are modified and distinctUntilChanged will evaluate subsequent items as same. To avoid such situation, it is recommended that mutable data is converted to an immutable one, for example using map(CharSequence::toString) or map(list -> Collections.unmodifiableList(new ArrayList<>(list))). 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
distinctUntilChanged does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Distinct
distinctUntilChanged(BiPredicate)
Returns:
a Flowable that emits those items from the source Publisher that are distinct from their
        immediate predecessors/**
     * Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their
     * immediate predecessors based on {@link Object#equals(Object)} comparison.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/distinctUntilChanged.png" alt="">
     * <p>
     * It is recommended the elements' class {@code T} in the flow overrides the default {@code Object.equals()} to provide
     * a meaningful comparison between items as the default Java implementation only considers reference equivalence.
     * Alternatively, use the {@link #distinctUntilChanged(BiPredicate)} overload and provide a comparison function
     * in case the class {@code T} can't be overridden with custom {@code equals()} or the comparison itself
     * should happen on different terms or properties of the class {@code T}.
     * <p>
     * Note that the operator always retains the latest item from upstream regardless of the comparison result
     * and uses it in the next comparison with the next upstream item.
     * <p>
     * Note that if element type {@code T} in the flow is mutable, the comparison of the previous and current
     * item may yield unexpected results if the items are mutated externally. Common cases are mutable
     * {@code CharSequence}s or {@code List}s where the objects will actually have the same
     * references when they are modified and {@code distinctUntilChanged} will evaluate subsequent items as same.
     * To avoid such situation, it is recommended that mutable data is converted to an immutable one,
     * for example using {@code map(CharSequence::toString)} or {@code map(list -> Collections.unmodifiableList(new ArrayList<>(list)))}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code distinctUntilChanged} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits those items from the source Publisher that are distinct from their
     *         immediate predecessors
     * @see <a href="http://reactivex.io/documentation/operators/distinct.html">ReactiveX operators documentation: Distinct</a>
     * @see #distinctUntilChanged(BiPredicate)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> distinctUntilChanged() {
        return distinctUntilChanged(Functions.identity());
    }

    
Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their immediate predecessors, according to a key selector function and based on Object.equals(Object) comparison of those objects returned by the key selector function. 


 It is recommended the keys' class K overrides the default Object.equals() to provide a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence. Alternatively, use the distinctUntilChanged(BiPredicate) overload and provide a comparison function in case the class K can't be overridden with custom equals() or the comparison itself should happen on different terms or properties of the item class T (for which the keys can be derived via a similar selector). 

Note that the operator always retains the latest key from upstream regardless of the comparison result
and uses it in the next comparison with the next key derived from the next upstream item.
 Note that if element type T in the flow is mutable, the comparison of the previous and current item may yield unexpected results if the items are mutated externally. Common cases are mutable CharSequences or Lists where the objects will actually have the same references when they are modified and distinctUntilChanged will evaluate subsequent items as same. To avoid such situation, it is recommended that mutable data is converted to an immutable one, for example using map(CharSequence::toString) or map(list -> Collections.unmodifiableList(new ArrayList<>(list))). 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
distinctUntilChanged does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that projects an emitted item to a key value that is used to decide whether an item
           is distinct from another one or not
Type parameters:
<K> – the key type
See Also:
ReactiveX operators documentation: Distinct
Returns:
a Flowable that emits those items from the source Publisher whose keys are distinct from
        those of their immediate predecessors/**
     * Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their
     * immediate predecessors, according to a key selector function and based on {@link Object#equals(Object)} comparison
     * of those objects returned by the key selector function.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/distinctUntilChanged.key.png" alt="">
     * <p>
     * It is recommended the keys' class {@code K} overrides the default {@code Object.equals()} to provide
     * a meaningful comparison between the key objects as the default Java implementation only considers reference equivalence.
     * Alternatively, use the {@link #distinctUntilChanged(BiPredicate)} overload and provide a comparison function
     * in case the class {@code K} can't be overridden with custom {@code equals()} or the comparison itself
     * should happen on different terms or properties of the item class {@code T} (for which the keys can be
     * derived via a similar selector).
     * <p>
     * Note that the operator always retains the latest key from upstream regardless of the comparison result
     * and uses it in the next comparison with the next key derived from the next upstream item.
     * <p>
     * Note that if element type {@code T} in the flow is mutable, the comparison of the previous and current
     * item may yield unexpected results if the items are mutated externally. Common cases are mutable
     * {@code CharSequence}s or {@code List}s where the objects will actually have the same
     * references when they are modified and {@code distinctUntilChanged} will evaluate subsequent items as same.
     * To avoid such situation, it is recommended that mutable data is converted to an immutable one,
     * for example using {@code map(CharSequence::toString)} or {@code map(list -> Collections.unmodifiableList(new ArrayList<>(list)))}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code distinctUntilChanged} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type
     * @param keySelector
     *            a function that projects an emitted item to a key value that is used to decide whether an item
     *            is distinct from another one or not
     * @return a Flowable that emits those items from the source Publisher whose keys are distinct from
     *         those of their immediate predecessors
     * @see <a href="http://reactivex.io/documentation/operators/distinct.html">ReactiveX operators documentation: Distinct</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K> Flowable<T> distinctUntilChanged(Function<? super T, K> keySelector) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        return RxJavaPlugins.onAssembly(new FlowableDistinctUntilChanged<T, K>(this, keySelector, ObjectHelper.equalsPredicate()));
    }

    
Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their
immediate predecessors when compared with each other via the provided comparator function.



Note that the operator always retains the latest item from upstream regardless of the comparison result
and uses it in the next comparison with the next upstream item.
 Note that if element type T in the flow is mutable, the comparison of the previous and current item may yield unexpected results if the items are mutated externally. Common cases are mutable CharSequences or Lists where the objects will actually have the same references when they are modified and distinctUntilChanged will evaluate subsequent items as same. To avoid such situation, it is recommended that mutable data is converted to an immutable one, for example using map(CharSequence::toString) or map(list -> Collections.unmodifiableList(new ArrayList<>(list))). 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
distinctUntilChanged does not operate by default on a particular Scheduler.

Params:
comparer – the function that receives the previous item and the current item and is
                  expected to return true if the two are equal, thus skipping the current value.
See Also:
ReactiveX operators documentation: Distinct
Returns:
a Flowable that emits those items from the source Publisher that are distinct from their
        immediate predecessors
Since:
2.0/**
     * Returns a Flowable that emits all items emitted by the source Publisher that are distinct from their
     * immediate predecessors when compared with each other via the provided comparator function.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/distinctUntilChanged.png" alt="">
     * <p>
     * Note that the operator always retains the latest item from upstream regardless of the comparison result
     * and uses it in the next comparison with the next upstream item.
     * <p>
     * Note that if element type {@code T} in the flow is mutable, the comparison of the previous and current
     * item may yield unexpected results if the items are mutated externally. Common cases are mutable
     * {@code CharSequence}s or {@code List}s where the objects will actually have the same
     * references when they are modified and {@code distinctUntilChanged} will evaluate subsequent items as same.
     * To avoid such situation, it is recommended that mutable data is converted to an immutable one,
     * for example using {@code map(CharSequence::toString)} or {@code map(list -> Collections.unmodifiableList(new ArrayList<>(list)))}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code distinctUntilChanged} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param comparer the function that receives the previous item and the current item and is
     *                   expected to return true if the two are equal, thus skipping the current value.
     * @return a Flowable that emits those items from the source Publisher that are distinct from their
     *         immediate predecessors
     * @see <a href="http://reactivex.io/documentation/operators/distinct.html">ReactiveX operators documentation: Distinct</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> distinctUntilChanged(BiPredicate<? super T, ? super T> comparer) {
        ObjectHelper.requireNonNull(comparer, "comparer is null");
        return RxJavaPlugins.onAssembly(new FlowableDistinctUntilChanged<T, T>(this, Functions.<T>identity(), comparer));
    }

    
Calls the specified action after this Flowable signals onError or onCompleted or gets canceled by
the downstream.
In case of a race between a terminal event and a cancellation, the provided onFinally action is executed once per subscription. 
Note that the onFinally action is shared between subscriptions and as such should be thread-safe. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doFinally does not operate by default on a particular Scheduler.
 
Operator-fusion:
 
This operator supports normal and conditional Subscribers as well as boundary-limited
 synchronous or asynchronous queue-fusion.

History: 2.0.1 - experimental
Params:
onFinally – the action called when this Flowable terminates or gets canceled
Returns:
the new Flowable instance
Since:
2.1/**
     * Calls the specified action after this Flowable signals onError or onCompleted or gets canceled by
     * the downstream.
     * <p>In case of a race between a terminal event and a cancellation, the provided {@code onFinally} action
     * is executed once per subscription.
     * <p>Note that the {@code onFinally} action is shared between subscriptions and as such
     * should be thread-safe.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doFinally} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Operator-fusion:</b></dt>
     *  <dd>This operator supports normal and conditional Subscribers as well as boundary-limited
     *  synchronous or asynchronous queue-fusion.</dd>
     * </dl>
     * <p>History: 2.0.1 - experimental
     * @param onFinally the action called when this Flowable terminates or gets canceled
     * @return the new Flowable instance
     * @since 2.1
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doFinally(Action onFinally) {
        ObjectHelper.requireNonNull(onFinally, "onFinally is null");
        return RxJavaPlugins.onAssembly(new FlowableDoFinally<T>(this, onFinally));
    }

    
Calls the specified consumer with the current item after this item has been emitted to the downstream.
Note that the onAfterNext action is shared between subscriptions and as such should be thread-safe. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doAfterNext does not operate by default on a particular Scheduler.
 
Operator-fusion:
 
This operator supports normal and conditional Subscribers as well as boundary-limited
 synchronous or asynchronous queue-fusion.

History: 2.0.1 - experimental
Params:
onAfterNext – the Consumer that will be called after emitting an item from upstream to the downstream
Returns:
the new Flowable instance
Since:
2.1/**
     * Calls the specified consumer with the current item after this item has been emitted to the downstream.
     * <p>Note that the {@code onAfterNext} action is shared between subscriptions and as such
     * should be thread-safe.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doAfterNext} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Operator-fusion:</b></dt>
     *  <dd>This operator supports normal and conditional Subscribers as well as boundary-limited
     *  synchronous or asynchronous queue-fusion.</dd>
     * </dl>
     * <p>History: 2.0.1 - experimental
     * @param onAfterNext the Consumer that will be called after emitting an item from upstream to the downstream
     * @return the new Flowable instance
     * @since 2.1
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doAfterNext(Consumer<? super T> onAfterNext) {
        ObjectHelper.requireNonNull(onAfterNext, "onAfterNext is null");
        return RxJavaPlugins.onAssembly(new FlowableDoAfterNext<T>(this, onAfterNext));
    }

    
Registers an Action to be called when this Publisher invokes either onComplete or onError. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doAfterTerminate does not operate by default on a particular Scheduler.

Params:
onAfterTerminate –  an Action to be invoked when the source Publisher finishes
See Also:
ReactiveX operators documentation: Do
doOnTerminate(Action)
Returns:
a Flowable that emits the same items as the source Publisher, then invokes the Action/**
     * Registers an {@link Action} to be called when this Publisher invokes either
     * {@link Subscriber#onComplete onComplete} or {@link Subscriber#onError onError}.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/finallyDo.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doAfterTerminate} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onAfterTerminate
     *            an {@link Action} to be invoked when the source Publisher finishes
     * @return a Flowable that emits the same items as the source Publisher, then invokes the
     *         {@link Action}
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     * @see #doOnTerminate(Action)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doAfterTerminate(Action onAfterTerminate) {
        return doOnEach(Functions.emptyConsumer(), Functions.emptyConsumer(),
                Functions.EMPTY_ACTION, onAfterTerminate);
    }

    
Calls the cancel Action if the downstream cancels the sequence. 

The action is shared between subscriptions and thus may be called concurrently from multiple
threads; the action must be thread-safe.
 If the action throws a runtime exception, that exception is rethrown by the onCancel() call, sometimes as a CompositeException if there were multiple exceptions along the way. 
 Note that terminal events trigger the action unless the Publisher is subscribed to via unsafeSubscribe(). 



 
Backpressure:
 
doOnCancel does not interact with backpressure requests or value delivery; backpressure behavior is preserved between its upstream and its downstream.
 
Scheduler:
 
doOnCancel does not operate by default on a particular Scheduler.

Params:
onCancel –  the action that gets called when the source Publisher's Subscription is canceled
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher modified so as to call this Action when appropriate/**
     * Calls the cancel {@code Action} if the downstream cancels the sequence.
     * <p>
     * The action is shared between subscriptions and thus may be called concurrently from multiple
     * threads; the action must be thread-safe.
     * <p>
     * If the action throws a runtime exception, that exception is rethrown by the {@code onCancel()} call,
     * sometimes as a {@code CompositeException} if there were multiple exceptions along the way.
     * <p>
     * Note that terminal events trigger the action unless the {@code Publisher} is subscribed to via {@code unsafeSubscribe()}.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnUnsubscribe.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>{@code doOnCancel} does not interact with backpressure requests or value delivery; backpressure
     *  behavior is preserved between its upstream and its downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnCancel} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onCancel
     *            the action that gets called when the source {@code Publisher}'s Subscription is canceled
     * @return the source {@code Publisher} modified so as to call this Action when appropriate
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnCancel(Action onCancel) {
        return doOnLifecycle(Functions.emptyConsumer(), Functions.EMPTY_LONG_CONSUMER, onCancel);
    }

    
Modifies the source Publisher so that it invokes an action when it calls onComplete. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnComplete does not operate by default on a particular Scheduler.

Params:
onComplete –  the action to invoke when the source Publisher calls onComplete
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Modifies the source Publisher so that it invokes an action when it calls {@code onComplete}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnComplete.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnComplete} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onComplete
     *            the action to invoke when the source Publisher calls {@code onComplete}
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnComplete(Action onComplete) {
        return doOnEach(Functions.emptyConsumer(), Functions.emptyConsumer(),
                onComplete, Functions.EMPTY_ACTION);
    }

    
Calls the appropriate onXXX consumer (shared between all subscribers) whenever a signal with the same type
passes through, before forwarding them to downstream.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnEach does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Calls the appropriate onXXX consumer (shared between all subscribers) whenever a signal with the same type
     * passes through, before forwarding them to downstream.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnEach.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnEach} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    private Flowable<T> doOnEach(Consumer<? super T> onNext, Consumer<? super Throwable> onError,
            Action onComplete, Action onAfterTerminate) {
        ObjectHelper.requireNonNull(onNext, "onNext is null");
        ObjectHelper.requireNonNull(onError, "onError is null");
        ObjectHelper.requireNonNull(onComplete, "onComplete is null");
        ObjectHelper.requireNonNull(onAfterTerminate, "onAfterTerminate is null");
        return RxJavaPlugins.onAssembly(new FlowableDoOnEach<T>(this, onNext, onError, onComplete, onAfterTerminate));
    }

    
Modifies the source Publisher so that it invokes an action for each item it emits.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnEach does not operate by default on a particular Scheduler.

Params:
onNotification – 
           the action to invoke for each item emitted by the source Publisher
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Modifies the source Publisher so that it invokes an action for each item it emits.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnEach.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnEach} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNotification
     *            the action to invoke for each item emitted by the source Publisher
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnEach(final Consumer<? super Notification<T>> onNotification) {
        ObjectHelper.requireNonNull(onNotification, "onNotification is null");
        return doOnEach(
                Functions.notificationOnNext(onNotification),
                Functions.notificationOnError(onNotification),
                Functions.notificationOnComplete(onNotification),
                Functions.EMPTY_ACTION
            );
    }

    
Modifies the source Publisher so that it notifies a Subscriber for each item and terminal event it emits.
 In case the onError of the supplied Subscriber throws, the downstream will receive a composite exception containing the original exception and the exception thrown by onError. If either the onNext or the onComplete method of the supplied Subscriber throws, the downstream will be terminated and will receive this thrown exception. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnEach does not operate by default on a particular Scheduler.

Params:
subscriber – 
           the Subscriber to be notified about onNext, onError and onComplete events on its
           respective methods before the actual downstream Subscriber gets notified.
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Modifies the source Publisher so that it notifies a Subscriber for each item and terminal event it emits.
     * <p>
     * In case the {@code onError} of the supplied Subscriber throws, the downstream will receive a composite
     * exception containing the original exception and the exception thrown by {@code onError}. If either the
     * {@code onNext} or the {@code onComplete} method of the supplied Subscriber throws, the downstream will be
     * terminated and will receive this thrown exception.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnEach.o.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnEach} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param subscriber
     *            the Subscriber to be notified about onNext, onError and onComplete events on its
     *            respective methods before the actual downstream Subscriber gets notified.
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnEach(final Subscriber<? super T> subscriber) {
        ObjectHelper.requireNonNull(subscriber, "subscriber is null");
        return doOnEach(
                FlowableInternalHelper.subscriberOnNext(subscriber),
                FlowableInternalHelper.subscriberOnError(subscriber),
                FlowableInternalHelper.subscriberOnComplete(subscriber),
                Functions.EMPTY_ACTION);
    }

    
Modifies the source Publisher so that it invokes an action if it calls onError. 
 In case the onError action throws, the downstream will receive a composite exception containing the original exception and the exception thrown by onError. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnError does not operate by default on a particular Scheduler.

Params:
onError –  the action to invoke if the source Publisher calls onError
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Modifies the source Publisher so that it invokes an action if it calls {@code onError}.
     * <p>
     * In case the {@code onError} action throws, the downstream will receive a composite exception containing
     * the original exception and the exception thrown by {@code onError}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnError.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onError
     *            the action to invoke if the source Publisher calls {@code onError}
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnError(Consumer<? super Throwable> onError) {
        return doOnEach(Functions.emptyConsumer(), onError,
                Functions.EMPTY_ACTION, Functions.EMPTY_ACTION);
    }

    
Calls the appropriate onXXX method (shared between all Subscribers) for the lifecycle events of
the sequence (subscription, cancellation, requesting).



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnLifecycle does not operate by default on a particular Scheduler.

Params:
onSubscribe – 
             a Consumer called with the Subscription sent via Subscriber.onSubscribe()
onRequest – 
             a LongConsumer called with the request amount sent via Subscription.request()
onCancel – 
             called when the downstream cancels the Subscription via cancel()
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Calls the appropriate onXXX method (shared between all Subscribers) for the lifecycle events of
     * the sequence (subscription, cancellation, requesting).
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnNext.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnLifecycle} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onSubscribe
     *              a Consumer called with the Subscription sent via Subscriber.onSubscribe()
     * @param onRequest
     *              a LongConsumer called with the request amount sent via Subscription.request()
     * @param onCancel
     *              called when the downstream cancels the Subscription via cancel()
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnLifecycle(final Consumer<? super Subscription> onSubscribe,
            final LongConsumer onRequest, final Action onCancel) {
        ObjectHelper.requireNonNull(onSubscribe, "onSubscribe is null");
        ObjectHelper.requireNonNull(onRequest, "onRequest is null");
        ObjectHelper.requireNonNull(onCancel, "onCancel is null");
        return RxJavaPlugins.onAssembly(new FlowableDoOnLifecycle<T>(this, onSubscribe, onRequest, onCancel));
    }

    
Modifies the source Publisher so that it invokes an action when it calls onNext. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnNext does not operate by default on a particular Scheduler.

Params:
onNext –  the action to invoke when the source Publisher calls onNext
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Modifies the source Publisher so that it invokes an action when it calls {@code onNext}.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnNext.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnNext} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *            the action to invoke when the source Publisher calls {@code onNext}
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnNext(Consumer<? super T> onNext) {
        return doOnEach(onNext, Functions.emptyConsumer(),
                Functions.EMPTY_ACTION, Functions.EMPTY_ACTION);
    }

    
Modifies the source Publisher so that it invokes the given action when it receives a request for more items. 

Note: This operator is for tracing the internal behavior of back-pressure request
patterns and generally intended for debugging use.

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
Scheduler:
doOnRequest does not operate by default on a particular Scheduler.

Params:
onRequest –  the action that gets called when a Subscriber requests items from this Publisher
See Also:
ReactiveX operators
     *      documentation: Do
Returns:
the source Publisher modified so as to call this Action when appropriate
Since:
2.0/**
     * Modifies the source {@code Publisher} so that it invokes the given action when it receives a
     * request for more items.
     * <p>
     * <b>Note:</b> This operator is for tracing the internal behavior of back-pressure request
     * patterns and generally intended for debugging use.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     * <dt><b>Scheduler:</b></dt>
     * <dd>{@code doOnRequest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onRequest
     *            the action that gets called when a Subscriber requests items from this
     *            {@code Publisher}
     * @return the source {@code Publisher} modified so as to call this Action when appropriate
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators
     *      documentation: Do</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnRequest(LongConsumer onRequest) {
        return doOnLifecycle(Functions.emptyConsumer(), onRequest, Functions.EMPTY_ACTION);
    }

    
Modifies the source Publisher so that it invokes the given action when it is subscribed from its subscribers. Each subscription will result in an invocation of the given action except when the source Publisher is reference counted, in which case the source Publisher will invoke the given action for the first subscription. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnSubscribe does not operate by default on a particular Scheduler.

Params:
onSubscribe –  the Consumer that gets called when a Subscriber subscribes to the current Flowable
See Also:
ReactiveX operators documentation: Do
Returns:
the source Publisher modified so as to call this Consumer when appropriate/**
     * Modifies the source {@code Publisher} so that it invokes the given action when it is subscribed from
     * its subscribers. Each subscription will result in an invocation of the given action except when the
     * source {@code Publisher} is reference counted, in which case the source {@code Publisher} will invoke
     * the given action for the first subscription.
     * <p>
     * <img width="640" height="390" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnSubscribe.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onSubscribe
     *            the Consumer that gets called when a Subscriber subscribes to the current {@code Flowable}
     * @return the source {@code Publisher} modified so as to call this Consumer when appropriate
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnSubscribe(Consumer<? super Subscription> onSubscribe) {
        return doOnLifecycle(onSubscribe, Functions.EMPTY_LONG_CONSUMER, Functions.EMPTY_ACTION);
    }

    
Modifies the source Publisher so that it invokes an action when it calls onComplete or onError. 


 This differs from doAfterTerminate in that this happens before the onComplete or onError notification. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
doOnTerminate does not operate by default on a particular Scheduler.

Params:
onTerminate –  the action to invoke when the source Publisher calls onComplete or onError
See Also:
ReactiveX operators documentation: Do
doAfterTerminate(Action)
Returns:
the source Publisher with the side-effecting behavior applied/**
     * Modifies the source Publisher so that it invokes an action when it calls {@code onComplete} or
     * {@code onError}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/doOnTerminate.png" alt="">
     * <p>
     * This differs from {@code doAfterTerminate} in that this happens <em>before</em> the {@code onComplete} or
     * {@code onError} notification.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code doOnTerminate} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onTerminate
     *            the action to invoke when the source Publisher calls {@code onComplete} or {@code onError}
     * @return the source Publisher with the side-effecting behavior applied
     * @see <a href="http://reactivex.io/documentation/operators/do.html">ReactiveX operators documentation: Do</a>
     * @see #doAfterTerminate(Action)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> doOnTerminate(final Action onTerminate) {
        return doOnEach(Functions.emptyConsumer(), Functions.actionConsumer(onTerminate),
                onTerminate, Functions.EMPTY_ACTION);
    }

    
Returns a Maybe that emits the single item at a specified index in a sequence of emissions from
this Flowable or completes if this Flowable sequence has fewer elements than index.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
elementAt does not operate by default on a particular Scheduler.

Params:
index – 
           the zero-based index of the item to retrieve
See Also:
ReactiveX operators documentation: ElementAt
Returns:
a Maybe that emits a single item: the item at the specified position in the sequence of
        those emitted by the source Publisher/**
     * Returns a Maybe that emits the single item at a specified index in a sequence of emissions from
     * this Flowable or completes if this Flowable sequence has fewer elements than index.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/elementAt.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code elementAt} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param index
     *            the zero-based index of the item to retrieve
     * @return a Maybe that emits a single item: the item at the specified position in the sequence of
     *         those emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/elementat.html">ReactiveX operators documentation: ElementAt</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Maybe<T> elementAt(long index) {
        if (index < 0) {
            throw new IndexOutOfBoundsException("index >= 0 required but it was " + index);
        }
        return RxJavaPlugins.onAssembly(new FlowableElementAtMaybe<T>(this, index));
    }

    
Returns a Single that emits the item found at a specified index in a sequence of emissions from
this Flowable, or a default item if that index is out of range.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
elementAt does not operate by default on a particular Scheduler.

Params:
index – 
           the zero-based index of the item to retrieve
defaultItem – 
           the default item
Throws:
IndexOutOfBoundsException –  if index is less than 0
See Also:
ReactiveX operators documentation: ElementAt
Returns:
a Single that emits the item at the specified position in the sequence emitted by the source
        Publisher, or the default item if that index is outside the bounds of the source sequence/**
     * Returns a Single that emits the item found at a specified index in a sequence of emissions from
     * this Flowable, or a default item if that index is out of range.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/elementAtOrDefault.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code elementAt} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param index
     *            the zero-based index of the item to retrieve
     * @param defaultItem
     *            the default item
     * @return a Single that emits the item at the specified position in the sequence emitted by the source
     *         Publisher, or the default item if that index is outside the bounds of the source sequence
     * @throws IndexOutOfBoundsException
     *             if {@code index} is less than 0
     * @see <a href="http://reactivex.io/documentation/operators/elementat.html">ReactiveX operators documentation: ElementAt</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> elementAt(long index, T defaultItem) {
        if (index < 0) {
            throw new IndexOutOfBoundsException("index >= 0 required but it was " + index);
        }
        ObjectHelper.requireNonNull(defaultItem, "defaultItem is null");
        return RxJavaPlugins.onAssembly(new FlowableElementAtSingle<T>(this, index, defaultItem));
    }

    
Returns a Single that emits the item found at a specified index in a sequence of emissions from this Flowable or signals a NoSuchElementException if this Flowable has fewer elements than index. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
elementAtOrError does not operate by default on a particular Scheduler.

Params:
index – 
           the zero-based index of the item to retrieve
Throws:
IndexOutOfBoundsException –  if index is less than 0
See Also:
ReactiveX operators documentation: ElementAt
Returns:
a Single that emits the item at the specified position in the sequence emitted by the source
        Publisher, or the default item if that index is outside the bounds of the source sequence/**
     * Returns a Single that emits the item found at a specified index in a sequence of emissions from
     * this Flowable or signals a {@link NoSuchElementException} if this Flowable has fewer elements than index.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/elementAtOrDefault.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded manner
     *  (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code elementAtOrError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param index
     *            the zero-based index of the item to retrieve
     * @return a Single that emits the item at the specified position in the sequence emitted by the source
     *         Publisher, or the default item if that index is outside the bounds of the source sequence
     * @throws IndexOutOfBoundsException
     *             if {@code index} is less than 0
     * @see <a href="http://reactivex.io/documentation/operators/elementat.html">ReactiveX operators documentation: ElementAt</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> elementAtOrError(long index) {
        if (index < 0) {
            throw new IndexOutOfBoundsException("index >= 0 required but it was " + index);
        }
        return RxJavaPlugins.onAssembly(new FlowableElementAtSingle<T>(this, index, null));
    }

    
Filters items emitted by a Publisher by only emitting those that satisfy a specified predicate.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
filter does not operate by default on a particular Scheduler.

Params:
predicate –  a function that evaluates each item emitted by the source Publisher, returning true if it passes the filter
See Also:
ReactiveX operators documentation: Filter
Returns:
a Flowable that emits only those items emitted by the source Publisher that the filter evaluates as true/**
     * Filters items emitted by a Publisher by only emitting those that satisfy a specified predicate.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/filter.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code filter} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param predicate
     *            a function that evaluates each item emitted by the source Publisher, returning {@code true}
     *            if it passes the filter
     * @return a Flowable that emits only those items emitted by the source Publisher that the filter
     *         evaluates as {@code true}
     * @see <a href="http://reactivex.io/documentation/operators/filter.html">ReactiveX operators documentation: Filter</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> filter(Predicate<? super T> predicate) {
        ObjectHelper.requireNonNull(predicate, "predicate is null");
        return RxJavaPlugins.onAssembly(new FlowableFilter<T>(this, predicate));
    }

    
Returns a Maybe that emits only the very first item emitted by this Flowable or
completes if this Flowable is empty.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
firstElement does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: First
Returns:
the new Maybe instance/**
     * Returns a Maybe that emits only the very first item emitted by this Flowable or
     * completes if this Flowable is empty.
     * <p>
     * <img width="640" height="237" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/firstElement.m.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code firstElement} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the new Maybe instance
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX operators documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL) // take may trigger UNBOUNDED_IN
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Maybe<T> firstElement() {
        return elementAt(0);
    }

    
Returns a Single that emits only the very first item emitted by this Flowable, or a default
item if this Flowable completes without emitting anything.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
first does not operate by default on a particular Scheduler.

Params:
defaultItem – 
           the default item to emit if the source Publisher doesn't emit anything
See Also:
ReactiveX operators documentation: First
Returns:
a Single that emits only the very first item from the source, or a default item if the
        source Publisher completes without emitting any items/**
     * Returns a Single that emits only the very first item emitted by this Flowable, or a default
     * item if this Flowable completes without emitting anything.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/first.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code first} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param defaultItem
     *            the default item to emit if the source Publisher doesn't emit anything
     * @return a Single that emits only the very first item from the source, or a default item if the
     *         source Publisher completes without emitting any items
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX operators documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL) // take may trigger UNBOUNDED_IN
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> first(T defaultItem) {
        return elementAt(0, defaultItem);
    }

    
Returns a Single that emits only the very first item emitted by this Flowable or signals a NoSuchElementException if this Flowable is empty. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
firstOrError does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: First
Returns:
the new Single instance/**
     * Returns a Single that emits only the very first item emitted by this Flowable or
     * signals a {@link NoSuchElementException} if this Flowable is empty.
     * <p>
     * <img width="640" height="237" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/firstOrError.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code firstOrError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the new Single instance
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX operators documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL) // take may trigger UNBOUNDED_IN
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> firstOrError() {
        return elementAtOrError(0);
    }

    
Returns a Flowable that emits items based on applying a function that you supply to each item emitted
by the source Publisher, where that function returns a Publisher, and then merging those resulting
Publishers and emitting the results of this merger.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to bufferSize() outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
Type parameters:
<R> – the value type of the inner Publishers and the output type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the result of applying the transformation function to each item emitted
        by the source Publisher and merging the results of the Publishers obtained from this
        transformation/**
     * Returns a Flowable that emits items based on applying a function that you supply to each item emitted
     * by the source Publisher, where that function returns a Publisher, and then merging those resulting
     * Publishers and emitting the results of this merger.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the value type of the inner Publishers and the output type
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @return a Flowable that emits the result of applying the transformation function to each item emitted
     *         by the source Publisher and merging the results of the Publishers obtained from this
     *         transformation
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends R>> mapper) {
        return flatMap(mapper, false, bufferSize(), bufferSize());
    }

    
Returns a Flowable that emits items based on applying a function that you supply to each item emitted
by the source Publisher, where that function returns a Publisher, and then merging those resulting
Publishers and emitting the results of this merger.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to bufferSize() outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
delayErrors – 
           if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
           if false, the first one signaling an exception will terminate the whole sequence immediately
Type parameters:
<R> – the value type of the inner Publishers and the output type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the result of applying the transformation function to each item emitted
        by the source Publisher and merging the results of the Publishers obtained from this
        transformation/**
     * Returns a Flowable that emits items based on applying a function that you supply to each item emitted
     * by the source Publisher, where that function returns a Publisher, and then merging those resulting
     * Publishers and emitting the results of this merger.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the value type of the inner Publishers and the output type
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @param delayErrors
     *            if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
     *            if false, the first one signaling an exception will terminate the whole sequence immediately
     * @return a Flowable that emits the result of applying the transformation function to each item emitted
     *         by the source Publisher and merging the results of the Publishers obtained from this
     *         transformation
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends R>> mapper, boolean delayErrors) {
        return flatMap(mapper, delayErrors, bufferSize(), bufferSize());
    }

    
Returns a Flowable that emits items based on applying a function that you supply to each item emitted
by the source Publisher, where that function returns a Publisher, and then merging those resulting
Publishers and emitting the results of this merger, while limiting the maximum number of concurrent
subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to maxConcurrency outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
maxConcurrency – 
        the maximum number of Publishers that may be subscribed to concurrently
Type parameters:
<R> – the value type of the inner Publishers and the output type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the result of applying the transformation function to each item emitted
        by the source Publisher and merging the results of the Publishers obtained from this
        transformation
Since:
2.0/**
     * Returns a Flowable that emits items based on applying a function that you supply to each item emitted
     * by the source Publisher, where that function returns a Publisher, and then merging those resulting
     * Publishers and emitting the results of this merger, while limiting the maximum number of concurrent
     * subscriptions to these Publishers.
     * <!-- <p> -->
     * <!-- <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMap.png" alt=""> -->
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the value type of the inner Publishers and the output type
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @param maxConcurrency
     *         the maximum number of Publishers that may be subscribed to concurrently
     * @return a Flowable that emits the result of applying the transformation function to each item emitted
     *         by the source Publisher and merging the results of the Publishers obtained from this
     *         transformation
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends R>> mapper, int maxConcurrency) {
        return flatMap(mapper, false, maxConcurrency, bufferSize());
    }

    
Returns a Flowable that emits items based on applying a function that you supply to each item emitted
by the source Publisher, where that function returns a Publisher, and then merging those resulting
Publishers and emitting the results of this merger, while limiting the maximum number of concurrent
subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to maxConcurrency outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
maxConcurrency – 
        the maximum number of Publishers that may be subscribed to concurrently
delayErrors – 
           if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
           if false, the first one signaling an exception will terminate the whole sequence immediately
Type parameters:
<R> – the value type of the inner Publishers and the output type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the result of applying the transformation function to each item emitted
        by the source Publisher and merging the results of the Publishers obtained from this
        transformation
Since:
2.0/**
     * Returns a Flowable that emits items based on applying a function that you supply to each item emitted
     * by the source Publisher, where that function returns a Publisher, and then merging those resulting
     * Publishers and emitting the results of this merger, while limiting the maximum number of concurrent
     * subscriptions to these Publishers.
     * <!-- <p> -->
     * <!-- <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMap.png" alt=""> -->
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the value type of the inner Publishers and the output type
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @param maxConcurrency
     *         the maximum number of Publishers that may be subscribed to concurrently
     * @param delayErrors
     *            if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
     *            if false, the first one signaling an exception will terminate the whole sequence immediately
     * @return a Flowable that emits the result of applying the transformation function to each item emitted
     *         by the source Publisher and merging the results of the Publishers obtained from this
     *         transformation
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends R>> mapper, boolean delayErrors, int maxConcurrency) {
        return flatMap(mapper, delayErrors, maxConcurrency, bufferSize());
    }

    
Returns a Flowable that emits items based on applying a function that you supply to each item emitted
by the source Publisher, where that function returns a Publisher, and then merging those resulting
Publishers and emitting the results of this merger, while limiting the maximum number of concurrent
subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to maxConcurrency outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
maxConcurrency – 
        the maximum number of Publishers that may be subscribed to concurrently
delayErrors – 
           if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
           if false, the first one signaling an exception will terminate the whole sequence immediately
bufferSize – 
           the number of elements to prefetch from each inner Publisher
Type parameters:
<R> – the value type of the inner Publishers and the output type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the result of applying the transformation function to each item emitted
        by the source Publisher and merging the results of the Publishers obtained from this
        transformation
Since:
2.0/**
     * Returns a Flowable that emits items based on applying a function that you supply to each item emitted
     * by the source Publisher, where that function returns a Publisher, and then merging those resulting
     * Publishers and emitting the results of this merger, while limiting the maximum number of concurrent
     * subscriptions to these Publishers.
     * <!-- <p> -->
     * <!-- <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMap.png" alt=""> -->
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the value type of the inner Publishers and the output type
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @param maxConcurrency
     *         the maximum number of Publishers that may be subscribed to concurrently
     * @param delayErrors
     *            if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
     *            if false, the first one signaling an exception will terminate the whole sequence immediately
     * @param bufferSize
     *            the number of elements to prefetch from each inner Publisher
     * @return a Flowable that emits the result of applying the transformation function to each item emitted
     *         by the source Publisher and merging the results of the Publishers obtained from this
     *         transformation
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends R>> mapper,
            boolean delayErrors, int maxConcurrency, int bufferSize) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        if (this instanceof ScalarCallable) {
            @SuppressWarnings("unchecked")
            T v = ((ScalarCallable<T>)this).call();
            if (v == null) {
                return empty();
            }
            return FlowableScalarXMap.scalarXMap(v, mapper);
        }
        return RxJavaPlugins.onAssembly(new FlowableFlatMap<T, R>(this, mapper, delayErrors, maxConcurrency, bufferSize));
    }

    
Returns a Flowable that applies a function to each item emitted or notification raised by the source
Publisher and then flattens the Publishers returned from these functions and emits the resulting items.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to bufferSize() outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
onNextMapper – 
           a function that returns a Publisher to merge for each item emitted by the source Publisher
onErrorMapper – 
           a function that returns a Publisher to merge for an onError notification from the source
           Publisher
onCompleteSupplier – 
           a function that returns a Publisher to merge for an onComplete notification from the source
           Publisher
Type parameters:
<R> – 
           the result type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of merging the Publishers returned from applying the
        specified functions to the emissions and notifications of the source Publisher/**
     * Returns a Flowable that applies a function to each item emitted or notification raised by the source
     * Publisher and then flattens the Publishers returned from these functions and emits the resulting items.
     * <p>
     * <img width="640" height="410" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeMap.nce.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the result type
     * @param onNextMapper
     *            a function that returns a Publisher to merge for each item emitted by the source Publisher
     * @param onErrorMapper
     *            a function that returns a Publisher to merge for an onError notification from the source
     *            Publisher
     * @param onCompleteSupplier
     *            a function that returns a Publisher to merge for an onComplete notification from the source
     *            Publisher
     * @return a Flowable that emits the results of merging the Publishers returned from applying the
     *         specified functions to the emissions and notifications of the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMap(
            Function<? super T, ? extends Publisher<? extends R>> onNextMapper,
            Function<? super Throwable, ? extends Publisher<? extends R>> onErrorMapper,
            Callable<? extends Publisher<? extends R>> onCompleteSupplier) {
        ObjectHelper.requireNonNull(onNextMapper, "onNextMapper is null");
        ObjectHelper.requireNonNull(onErrorMapper, "onErrorMapper is null");
        ObjectHelper.requireNonNull(onCompleteSupplier, "onCompleteSupplier is null");
        return merge(new FlowableMapNotification<T, Publisher<? extends R>>(this, onNextMapper, onErrorMapper, onCompleteSupplier));
    }

    
Returns a Flowable that applies a function to each item emitted or notification raised by the source
Publisher and then flattens the Publishers returned from these functions and emits the resulting items,
while limiting the maximum number of concurrent subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to maxConcurrency outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
onNextMapper – 
           a function that returns a Publisher to merge for each item emitted by the source Publisher
onErrorMapper – 
           a function that returns a Publisher to merge for an onError notification from the source
           Publisher
onCompleteSupplier – 
           a function that returns a Publisher to merge for an onComplete notification from the source
           Publisher
maxConcurrency – 
        the maximum number of Publishers that may be subscribed to concurrently
Type parameters:
<R> – 
           the result type
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of merging the Publishers returned from applying the
        specified functions to the emissions and notifications of the source Publisher
Since:
2.0/**
     * Returns a Flowable that applies a function to each item emitted or notification raised by the source
     * Publisher and then flattens the Publishers returned from these functions and emits the resulting items,
     * while limiting the maximum number of concurrent subscriptions to these Publishers.
     * <!-- <p> -->
     * <!-- <img width="640" height="410" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeMap.nce.png" alt=""> -->
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the result type
     * @param onNextMapper
     *            a function that returns a Publisher to merge for each item emitted by the source Publisher
     * @param onErrorMapper
     *            a function that returns a Publisher to merge for an onError notification from the source
     *            Publisher
     * @param onCompleteSupplier
     *            a function that returns a Publisher to merge for an onComplete notification from the source
     *            Publisher
     * @param maxConcurrency
     *         the maximum number of Publishers that may be subscribed to concurrently
     * @return a Flowable that emits the results of merging the Publishers returned from applying the
     *         specified functions to the emissions and notifications of the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMap(
            Function<? super T, ? extends Publisher<? extends R>> onNextMapper,
            Function<Throwable, ? extends Publisher<? extends R>> onErrorMapper,
            Callable<? extends Publisher<? extends R>> onCompleteSupplier,
            int maxConcurrency) {
        ObjectHelper.requireNonNull(onNextMapper, "onNextMapper is null");
        ObjectHelper.requireNonNull(onErrorMapper, "onErrorMapper is null");
        ObjectHelper.requireNonNull(onCompleteSupplier, "onCompleteSupplier is null");
        return merge(new FlowableMapNotification<T, Publisher<? extends R>>(
                this, onNextMapper, onErrorMapper, onCompleteSupplier), maxConcurrency);
    }

    
Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
source Publisher and a specified collection Publisher.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to maxConcurrency outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns a Publisher for each item emitted by the source Publisher
combiner – 
           a function that combines one item emitted by each of the source and collection Publishers and
           returns an item to be emitted by the resulting Publisher
Type parameters:
<U> – 
           the type of items emitted by the inner Publishers
<R> – 
           the type of items emitted by the combiner function
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of applying a function to a pair of values emitted by the
        source Publisher and the collection Publisher/**
     * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
     * source Publisher and a specified collection Publisher.
     * <p>
     * <img width="640" height="390" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeMap.r.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the inner Publishers
     * @param <R>
     *            the type of items emitted by the combiner function
     * @param mapper
     *            a function that returns a Publisher for each item emitted by the source Publisher
     * @param combiner
     *            a function that combines one item emitted by each of the source and collection Publishers and
     *            returns an item to be emitted by the resulting Publisher
     * @return a Flowable that emits the results of applying a function to a pair of values emitted by the
     *         source Publisher and the collection Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends U>> mapper,
            BiFunction<? super T, ? super U, ? extends R> combiner) {
        return flatMap(mapper, combiner, false, bufferSize(), bufferSize());
    }

    
Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
source Publisher and a specified collection Publisher.



Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to bufferSize() outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns a Publisher for each item emitted by the source Publisher
combiner – 
           a function that combines one item emitted by each of the source and collection Publishers and
           returns an item to be emitted by the resulting Publisher
delayErrors – 
           if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
           if false, the first one signaling an exception will terminate the whole sequence immediately
Type parameters:
<U> – 
           the type of items emitted by the inner Publishers
<R> – 
           the type of items emitted by the combiner functions
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of applying a function to a pair of values emitted by the
        source Publisher and the collection Publisher/**
     * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
     * source Publisher and a specified collection Publisher.
     * <p>
     * <img width="640" height="390" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeMap.r.png" alt="">
     * <dl>
     * <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the inner Publishers
     * @param <R>
     *            the type of items emitted by the combiner functions
     * @param mapper
     *            a function that returns a Publisher for each item emitted by the source Publisher
     * @param combiner
     *            a function that combines one item emitted by each of the source and collection Publishers and
     *            returns an item to be emitted by the resulting Publisher
     * @param delayErrors
     *            if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
     *            if false, the first one signaling an exception will terminate the whole sequence immediately
     * @return a Flowable that emits the results of applying a function to a pair of values emitted by the
     *         source Publisher and the collection Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends U>> mapper,
            BiFunction<? super T, ? super U, ? extends R> combiner, boolean delayErrors) {
        return flatMap(mapper, combiner, delayErrors, bufferSize(), bufferSize());
    }

    
Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent
subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to maxConcurrency outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns a Publisher for each item emitted by the source Publisher
combiner – 
           a function that combines one item emitted by each of the source and collection Publishers and
           returns an item to be emitted by the resulting Publisher
maxConcurrency – 
        the maximum number of Publishers that may be subscribed to concurrently
delayErrors – 
           if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
           if false, the first one signaling an exception will terminate the whole sequence immediately
Type parameters:
<U> – 
           the type of items emitted by the inner Publishers
<R> – 
           the type of items emitted by the combiner function
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of applying a function to a pair of values emitted by the
        source Publisher and the collection Publisher
Since:
2.0/**
     * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
     * source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent
     * subscriptions to these Publishers.
     * <!-- <p> -->
     * <!-- <img width="640" height="390" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeMap.r.png" alt=""> -->
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the inner Publishers
     * @param <R>
     *            the type of items emitted by the combiner function
     * @param mapper
     *            a function that returns a Publisher for each item emitted by the source Publisher
     * @param combiner
     *            a function that combines one item emitted by each of the source and collection Publishers and
     *            returns an item to be emitted by the resulting Publisher
     * @param maxConcurrency
     *         the maximum number of Publishers that may be subscribed to concurrently
     * @param delayErrors
     *            if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
     *            if false, the first one signaling an exception will terminate the whole sequence immediately
     * @return a Flowable that emits the results of applying a function to a pair of values emitted by the
     *         source Publisher and the collection Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends U>> mapper,
            BiFunction<? super T, ? super U, ? extends R> combiner, boolean delayErrors, int maxConcurrency) {
        return flatMap(mapper, combiner, delayErrors, maxConcurrency, bufferSize());
    }

    
Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent
subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to maxConcurrency outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns a Publisher for each item emitted by the source Publisher
combiner – 
           a function that combines one item emitted by each of the source and collection Publishers and
           returns an item to be emitted by the resulting Publisher
maxConcurrency – 
        the maximum number of Publishers that may be subscribed to concurrently
delayErrors – 
           if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
           if false, the first one signaling an exception will terminate the whole sequence immediately
bufferSize – 
           the number of elements to prefetch from the inner Publishers.
Type parameters:
<U> – 
           the type of items emitted by the inner Publishers
<R> – 
           the type of items emitted by the combiner function
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of applying a function to a pair of values emitted by the
        source Publisher and the collection Publisher
Since:
2.0/**
     * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
     * source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent
     * subscriptions to these Publishers.
     * <!-- <p> -->
     * <!-- <img width="640" height="390" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeMap.r.png" alt=""> -->
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@code maxConcurrency} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the inner Publishers
     * @param <R>
     *            the type of items emitted by the combiner function
     * @param mapper
     *            a function that returns a Publisher for each item emitted by the source Publisher
     * @param combiner
     *            a function that combines one item emitted by each of the source and collection Publishers and
     *            returns an item to be emitted by the resulting Publisher
     * @param maxConcurrency
     *         the maximum number of Publishers that may be subscribed to concurrently
     * @param delayErrors
     *            if true, exceptions from the current Flowable and all inner Publishers are delayed until all of them terminate
     *            if false, the first one signaling an exception will terminate the whole sequence immediately
     * @param bufferSize
     *            the number of elements to prefetch from the inner Publishers.
     * @return a Flowable that emits the results of applying a function to a pair of values emitted by the
     *         source Publisher and the collection Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> flatMap(final Function<? super T, ? extends Publisher<? extends U>> mapper,
            final BiFunction<? super T, ? super U, ? extends R> combiner, boolean delayErrors, int maxConcurrency, int bufferSize) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.requireNonNull(combiner, "combiner is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return flatMap(FlowableInternalHelper.flatMapWithCombiner(mapper, combiner), delayErrors, maxConcurrency, bufferSize);
    }

    
Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent
subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The upstream Flowable is consumed in a bounded manner (up to bufferSize() outstanding request amount for items). The inner Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
flatMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns a Publisher for each item emitted by the source Publisher
combiner – 
           a function that combines one item emitted by each of the source and collection Publishers and
           returns an item to be emitted by the resulting Publisher
maxConcurrency – 
        the maximum number of Publishers that may be subscribed to concurrently
Type parameters:
<U> – 
           the type of items emitted by the inner Publishers
<R> – 
           the type of items emitted by the combiner function
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of applying a function to a pair of values emitted by the
        source Publisher and the collection Publisher
Since:
2.0/**
     * Returns a Flowable that emits the results of a specified function to the pair of values emitted by the
     * source Publisher and a specified collection Publisher, while limiting the maximum number of concurrent
     * subscriptions to these Publishers.
     * <!-- <p> -->
     * <!-- <img width="640" height="390" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/mergeMap.r.png" alt=""> -->
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The upstream Flowable is consumed
     *  in a bounded manner (up to {@link #bufferSize()} outstanding request amount for items).
     *  The inner {@code Publisher}s are expected to honor backpressure; if violated,
     *  the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the inner Publishers
     * @param <R>
     *            the type of items emitted by the combiner function
     * @param mapper
     *            a function that returns a Publisher for each item emitted by the source Publisher
     * @param combiner
     *            a function that combines one item emitted by each of the source and collection Publishers and
     *            returns an item to be emitted by the resulting Publisher
     * @param maxConcurrency
     *         the maximum number of Publishers that may be subscribed to concurrently
     * @return a Flowable that emits the results of applying a function to a pair of values emitted by the
     *         source Publisher and the collection Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> flatMap(Function<? super T, ? extends Publisher<? extends U>> mapper,
            BiFunction<? super T, ? super U, ? extends R> combiner, int maxConcurrency) {
        return flatMap(mapper, combiner, false, maxConcurrency, bufferSize());
    }

    
Maps each element of the upstream Flowable into CompletableSources, subscribes to them and
waits until the upstream and all CompletableSources complete.

 
Backpressure:
 
The operator consumes the upstream in an unbounded manner.
 
Scheduler:
 
flatMapCompletable does not operate by default on a particular Scheduler.

Params:
mapper – the function that received each source value and transforms them into CompletableSources.
Returns:
the new Completable instance/**
     * Maps each element of the upstream Flowable into CompletableSources, subscribes to them and
     * waits until the upstream and all CompletableSources complete.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the upstream in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapCompletable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param mapper the function that received each source value and transforms them into CompletableSources.
     * @return the new Completable instance
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Completable flatMapCompletable(Function<? super T, ? extends CompletableSource> mapper) {
        return flatMapCompletable(mapper, false, Integer.MAX_VALUE);
    }

    
Maps each element of the upstream Flowable into CompletableSources, subscribes to them and
waits until the upstream and all CompletableSources complete, optionally delaying all errors.

 
Backpressure:
 
If maxConcurrency == Integer.MAX_VALUE the operator consumes the upstream in an unbounded manner. Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure the operator behaves as if maxConcurrency == Integer.MAX_VALUE was used.
 
Scheduler:
 
flatMapCompletable does not operate by default on a particular Scheduler.

Params:
mapper – the function that received each source value and transforms them into CompletableSources.
delayErrors – if true errors from the upstream and inner CompletableSources are delayed until each of them
terminates.
maxConcurrency – the maximum number of active subscriptions to the CompletableSources.
Returns:
the new Completable instance/**
     * Maps each element of the upstream Flowable into CompletableSources, subscribes to them and
     * waits until the upstream and all CompletableSources complete, optionally delaying all errors.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>If {@code maxConcurrency == Integer.MAX_VALUE} the operator consumes the upstream in an unbounded manner.
     *  Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure
     *  the operator behaves as if {@code maxConcurrency == Integer.MAX_VALUE} was used.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapCompletable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param mapper the function that received each source value and transforms them into CompletableSources.
     * @param delayErrors if true errors from the upstream and inner CompletableSources are delayed until each of them
     * terminates.
     * @param maxConcurrency the maximum number of active subscriptions to the CompletableSources.
     * @return the new Completable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Completable flatMapCompletable(Function<? super T, ? extends CompletableSource> mapper, boolean delayErrors, int maxConcurrency) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        return RxJavaPlugins.onAssembly(new FlowableFlatMapCompletableCompletable<T>(this, mapper, delayErrors, maxConcurrency));
    }

    
Returns a Flowable that merges each item emitted by the source Publisher with the values in an
Iterable corresponding to that item that is generated by a selector.



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers is expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
flatMapIterable does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns an Iterable sequence of values for when given an item emitted by the
           source Publisher
Type parameters:
<U> – 
           the type of item emitted by the resulting Iterable
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of merging the items emitted by the source Publisher with the values in the Iterables corresponding to those items, as generated by collectionSelector/**
     * Returns a Flowable that merges each item emitted by the source Publisher with the values in an
     * Iterable corresponding to that item that is generated by a selector.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMapIterable.f.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s is
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of item emitted by the resulting Iterable
     * @param mapper
     *            a function that returns an Iterable sequence of values for when given an item emitted by the
     *            source Publisher
     * @return a Flowable that emits the results of merging the items emitted by the source Publisher with
     *         the values in the Iterables corresponding to those items, as generated by {@code collectionSelector}
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<U> flatMapIterable(final Function<? super T, ? extends Iterable<? extends U>> mapper) {
        return flatMapIterable(mapper, bufferSize());
    }

    
Returns a Flowable that merges each item emitted by the source Publisher with the values in an
Iterable corresponding to that item that is generated by a selector.



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers is expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
flatMapIterable does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns an Iterable sequence of values for when given an item emitted by the
           source Publisher
bufferSize – 
           the number of elements to prefetch from the current Flowable
Type parameters:
<U> – 
           the type of item emitted by the resulting Iterable
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of merging the items emitted by the source Publisher with the values in the Iterables corresponding to those items, as generated by collectionSelector/**
     * Returns a Flowable that merges each item emitted by the source Publisher with the values in an
     * Iterable corresponding to that item that is generated by a selector.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMapIterable.f.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s is
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of item emitted by the resulting Iterable
     * @param mapper
     *            a function that returns an Iterable sequence of values for when given an item emitted by the
     *            source Publisher
     * @param bufferSize
     *            the number of elements to prefetch from the current Flowable
     * @return a Flowable that emits the results of merging the items emitted by the source Publisher with
     *         the values in the Iterables corresponding to those items, as generated by {@code collectionSelector}
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<U> flatMapIterable(final Function<? super T, ? extends Iterable<? extends U>> mapper, int bufferSize) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableFlattenIterable<T, U>(this, mapper, bufferSize));
    }

    
Returns a Flowable that emits the results of applying a function to the pair of values from the source
Publisher and an Iterable corresponding to that item that is generated by a selector.



 
Backpressure:
 
The operator honors backpressure from downstream and the source Publishers is consumed in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
flatMapIterable does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns an Iterable sequence of values for each item emitted by the source
           Publisher
resultSelector –  a function that returns an item based on the item emitted by the source Publisher and the Iterable returned for that item by the collectionSelector
Type parameters:
<U> – 
           the collection element type
<V> – 
           the type of item emitted by the resulting Iterable
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the items returned by resultSelector for each item in the source Publisher/**
     * Returns a Flowable that emits the results of applying a function to the pair of values from the source
     * Publisher and an Iterable corresponding to that item that is generated by a selector.
     * <p>
     * <img width="640" height="410" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMapIterable.f.r.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and the source {@code Publisher}s is
     *  consumed in an unbounded manner (i.e., no backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the collection element type
     * @param <V>
     *            the type of item emitted by the resulting Iterable
     * @param mapper
     *            a function that returns an Iterable sequence of values for each item emitted by the source
     *            Publisher
     * @param resultSelector
     *            a function that returns an item based on the item emitted by the source Publisher and the
     *            Iterable returned for that item by the {@code collectionSelector}
     * @return a Flowable that emits the items returned by {@code resultSelector} for each item in the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, V> Flowable<V> flatMapIterable(final Function<? super T, ? extends Iterable<? extends U>> mapper,
            final BiFunction<? super T, ? super U, ? extends V> resultSelector) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.requireNonNull(resultSelector, "resultSelector is null");
        return flatMap(FlowableInternalHelper.flatMapIntoIterable(mapper), resultSelector, false, bufferSize(), bufferSize());
    }

    
Returns a Flowable that merges each item emitted by the source Publisher with the values in an
Iterable corresponding to that item that is generated by a selector, while limiting the number of concurrent
subscriptions to these Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers is expected to honor backpressure as well. If the source Publisher violates the rule, the operator will signal a MissingBackpressureException.
 
Scheduler:
 
flatMapIterable does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that returns an Iterable sequence of values for when given an item emitted by the
           source Publisher
resultSelector –  a function that returns an item based on the item emitted by the source Publisher and the Iterable returned for that item by the collectionSelector
prefetch – 
           the number of elements to prefetch from the current Flowable
Type parameters:
<U> – 
           the element type of the inner Iterable sequences
<V> – 
           the type of item emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the results of merging the items emitted by the source Publisher with the values in the Iterables corresponding to those items, as generated by collectionSelector
Since:
2.0/**
     * Returns a Flowable that merges each item emitted by the source Publisher with the values in an
     * Iterable corresponding to that item that is generated by a selector, while limiting the number of concurrent
     * subscriptions to these Publishers.
     * <p>
     * <img width="640" height="410" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flatMapIterable.f.r.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s is
     *  expected to honor backpressure as well. If the source {@code Publisher} violates the rule, the operator will
     *  signal a {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapIterable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the element type of the inner Iterable sequences
     * @param <V>
     *            the type of item emitted by the resulting Publisher
     * @param mapper
     *            a function that returns an Iterable sequence of values for when given an item emitted by the
     *            source Publisher
     * @param resultSelector
     *            a function that returns an item based on the item emitted by the source Publisher and the
     *            Iterable returned for that item by the {@code collectionSelector}
     * @param prefetch
     *            the number of elements to prefetch from the current Flowable
     * @return a Flowable that emits the results of merging the items emitted by the source Publisher with
     *         the values in the Iterables corresponding to those items, as generated by {@code collectionSelector}
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, V> Flowable<V> flatMapIterable(final Function<? super T, ? extends Iterable<? extends U>> mapper,
            final BiFunction<? super T, ? super U, ? extends V> resultSelector, int prefetch) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.requireNonNull(resultSelector, "resultSelector is null");
        return flatMap(FlowableInternalHelper.flatMapIntoIterable(mapper), resultSelector, false, bufferSize(), prefetch);
    }

    
Maps each element of the upstream Flowable into MaybeSources, subscribes to all of them
and merges their onSuccess values, in no particular order, into a single Flowable sequence.

 
Backpressure:
 
The operator consumes the upstream in an unbounded manner.
 
Scheduler:
 
flatMapMaybe does not operate by default on a particular Scheduler.

Params:
mapper – the function that received each source value and transforms them into MaybeSources.
Type parameters:
<R> – the result value type
Returns:
the new Flowable instance/**
     * Maps each element of the upstream Flowable into MaybeSources, subscribes to all of them
     * and merges their onSuccess values, in no particular order, into a single Flowable sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the upstream in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapMaybe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the result value type
     * @param mapper the function that received each source value and transforms them into MaybeSources.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMapMaybe(Function<? super T, ? extends MaybeSource<? extends R>> mapper) {
        return flatMapMaybe(mapper, false, Integer.MAX_VALUE);
    }

    
Maps each element of the upstream Flowable into MaybeSources, subscribes to at most maxConcurrency MaybeSources at a time and merges their onSuccess values, in no particular order, into a single Flowable sequence, optionally delaying all errors. 

 
Backpressure:
 
If maxConcurrency == Integer.MAX_VALUE the operator consumes the upstream in an unbounded manner. Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure the operator behaves as if maxConcurrency == Integer.MAX_VALUE was used.
 
Scheduler:
 
flatMapMaybe does not operate by default on a particular Scheduler.

Params:
mapper – the function that received each source value and transforms them into MaybeSources.
delayErrors – if true errors from the upstream and inner MaybeSources are delayed until each of them
terminates.
maxConcurrency – the maximum number of active subscriptions to the MaybeSources.
Type parameters:
<R> – the result value type
Returns:
the new Flowable instance/**
     * Maps each element of the upstream Flowable into MaybeSources, subscribes to at most
     * {@code maxConcurrency} MaybeSources at a time and merges their onSuccess values,
     * in no particular order, into a single Flowable sequence, optionally delaying all errors.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>If {@code maxConcurrency == Integer.MAX_VALUE} the operator consumes the upstream in an unbounded manner.
     *  Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure
     *  the operator behaves as if {@code maxConcurrency == Integer.MAX_VALUE} was used.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapMaybe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the result value type
     * @param mapper the function that received each source value and transforms them into MaybeSources.
     * @param delayErrors if true errors from the upstream and inner MaybeSources are delayed until each of them
     * terminates.
     * @param maxConcurrency the maximum number of active subscriptions to the MaybeSources.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMapMaybe(Function<? super T, ? extends MaybeSource<? extends R>> mapper, boolean delayErrors, int maxConcurrency) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        return RxJavaPlugins.onAssembly(new FlowableFlatMapMaybe<T, R>(this, mapper, delayErrors, maxConcurrency));
    }

    
Maps each element of the upstream Flowable into SingleSources, subscribes to all of them
and merges their onSuccess values, in no particular order, into a single Flowable sequence.

 
Backpressure:
 
The operator consumes the upstream in an unbounded manner.
 
Scheduler:
 
flatMapSingle does not operate by default on a particular Scheduler.

Params:
mapper – the function that received each source value and transforms them into SingleSources.
Type parameters:
<R> – the result value type
Returns:
the new Flowable instance/**
     * Maps each element of the upstream Flowable into SingleSources, subscribes to all of them
     * and merges their onSuccess values, in no particular order, into a single Flowable sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the upstream in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapSingle} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the result value type
     * @param mapper the function that received each source value and transforms them into SingleSources.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMapSingle(Function<? super T, ? extends SingleSource<? extends R>> mapper) {
        return flatMapSingle(mapper, false, Integer.MAX_VALUE);
    }

    
Maps each element of the upstream Flowable into SingleSources, subscribes to at most maxConcurrency SingleSources at a time and merges their onSuccess values, in no particular order, into a single Flowable sequence, optionally delaying all errors. 

 
Backpressure:
 
If maxConcurrency == Integer.MAX_VALUE the operator consumes the upstream in an unbounded manner. Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure the operator behaves as if maxConcurrency == Integer.MAX_VALUE was used.
 
Scheduler:
 
flatMapSingle does not operate by default on a particular Scheduler.

Params:
mapper – the function that received each source value and transforms them into SingleSources.
delayErrors – if true errors from the upstream and inner SingleSources are delayed until each of them
terminates.
maxConcurrency – the maximum number of active subscriptions to the SingleSources.
Type parameters:
<R> – the result value type
Returns:
the new Flowable instance/**
     * Maps each element of the upstream Flowable into SingleSources, subscribes to at most
     * {@code maxConcurrency} SingleSources at a time and merges their onSuccess values,
     * in no particular order, into a single Flowable sequence, optionally delaying all errors.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>If {@code maxConcurrency == Integer.MAX_VALUE} the operator consumes the upstream in an unbounded manner.
     *  Otherwise, the operator expects the upstream to honor backpressure. If the upstream doesn't support backpressure
     *  the operator behaves as if {@code maxConcurrency == Integer.MAX_VALUE} was used.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code flatMapSingle} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the result value type
     * @param mapper the function that received each source value and transforms them into SingleSources.
     * @param delayErrors if true errors from the upstream and inner SingleSources are delayed until each of them
     * terminates.
     * @param maxConcurrency the maximum number of active subscriptions to the SingleSources.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> flatMapSingle(Function<? super T, ? extends SingleSource<? extends R>> mapper, boolean delayErrors, int maxConcurrency) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(maxConcurrency, "maxConcurrency");
        return RxJavaPlugins.onAssembly(new FlowableFlatMapSingle<T, R>(this, mapper, delayErrors, maxConcurrency));
    }

    
Subscribes to the Publisher and receives notifications for each element. 
 Alias to subscribe(Consumer) 

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
forEach does not operate by default on a particular Scheduler.

Params:
onNext –  Consumer to execute for each item.
Throws:
NullPointerException –  if onNext is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:

           a Disposable that allows canceling an asynchronous sequence/**
     * Subscribes to the {@link Publisher} and receives notifications for each element.
     * <p>
     * Alias to {@link #subscribe(Consumer)}
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code forEach} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *            {@link Consumer} to execute for each item.
     * @return
     *            a Disposable that allows canceling an asynchronous sequence
     * @throws NullPointerException
     *             if {@code onNext} is null
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.NONE)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable forEach(Consumer<? super T> onNext) {
        return subscribe(onNext);
    }

    
Subscribes to the Publisher and receives notifications for each element until the onNext Predicate returns false. 
 If the Flowable emits an error, it is wrapped into an OnErrorNotImplementedException and routed to the RxJavaPlugins.onError handler. 

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
forEachWhile does not operate by default on a particular Scheduler.

Params:
onNext –  Predicate to execute for each item.
Throws:
NullPointerException –  if onNext is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:
 a Disposable that allows canceling an asynchronous sequence/**
     * Subscribes to the {@link Publisher} and receives notifications for each element until the
     * onNext Predicate returns false.
     * <p>
     * If the Flowable emits an error, it is wrapped into an
     * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException}
     * and routed to the RxJavaPlugins.onError handler.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code forEachWhile} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *            {@link Predicate} to execute for each item.
     * @return
     *            a {@link Disposable} that allows canceling an asynchronous sequence
     * @throws NullPointerException
     *             if {@code onNext} is null
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.NONE)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable forEachWhile(Predicate<? super T> onNext) {
        return forEachWhile(onNext, Functions.ON_ERROR_MISSING, Functions.EMPTY_ACTION);
    }

    
Subscribes to the Publisher and receives notifications for each element and error events until the onNext Predicate returns false. 

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
forEachWhile does not operate by default on a particular Scheduler.

Params:
onNext –  Predicate to execute for each item.
onError –  Consumer to execute when an error is emitted.
Throws:
NullPointerException –  if onNext is null, or if onError is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:
 a Disposable that allows canceling an asynchronous sequence/**
     * Subscribes to the {@link Publisher} and receives notifications for each element and error events until the
     * onNext Predicate returns false.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code forEachWhile} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *            {@link Predicate} to execute for each item.
     * @param onError
     *            {@link Consumer} to execute when an error is emitted.
     * @return
     *            a {@link Disposable} that allows canceling an asynchronous sequence
     * @throws NullPointerException
     *             if {@code onNext} is null, or
     *             if {@code onError} is null
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.NONE)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable forEachWhile(Predicate<? super T> onNext, Consumer<? super Throwable> onError) {
        return forEachWhile(onNext, onError, Functions.EMPTY_ACTION);
    }

    
Subscribes to the Publisher and receives notifications for each element and the terminal events until the onNext Predicate returns false. 

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
forEachWhile does not operate by default on a particular Scheduler.

Params:
onNext –  Predicate to execute for each item.
onError –  Consumer to execute when an error is emitted.
onComplete –  Action to execute when completion is signaled.
Throws:
NullPointerException –  if onNext is null, or if onError is null, or if onComplete is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:
 a Disposable that allows canceling an asynchronous sequence/**
     * Subscribes to the {@link Publisher} and receives notifications for each element and the terminal events until the
     * onNext Predicate returns false.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code forEachWhile} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *            {@link Predicate} to execute for each item.
     * @param onError
     *            {@link Consumer} to execute when an error is emitted.
     * @param onComplete
     *            {@link Action} to execute when completion is signaled.
     * @return
     *            a {@link Disposable} that allows canceling an asynchronous sequence
     * @throws NullPointerException
     *             if {@code onNext} is null, or
     *             if {@code onError} is null, or
     *             if {@code onComplete} is null
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.NONE)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable forEachWhile(final Predicate<? super T> onNext, final Consumer<? super Throwable> onError,
            final Action onComplete) {
        ObjectHelper.requireNonNull(onNext, "onNext is null");
        ObjectHelper.requireNonNull(onError, "onError is null");
        ObjectHelper.requireNonNull(onComplete, "onComplete is null");

        ForEachWhileSubscriber<T> s = new ForEachWhileSubscriber<T>(onNext, onError, onComplete);
        subscribe(s);
        return s;
    }

    
Groups the items emitted by a Publisher according to a specified criterion, and emits these grouped items as GroupedFlowables. The emitted GroupedPublisher allows only a single Subscriber during its lifetime and if this Subscriber cancels before the source terminates, the next emission by the source having the same key will trigger a new GroupedPublisher emission. 



Note: A GroupedFlowable will cache the items it is to emit until such time as it is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those GroupedPublishers that do not concern you. Instead, you can signal to them that they may discard their buffers by applying an operator like ignoreElements to them. 
 Note that the GroupedFlowables should be subscribed to as soon as possible, otherwise, the unconsumed groups may starve other groups due to the internal backpressure coordination of the groupBy operator. Such hangs can be usually avoided by using flatMap(Function, int) or concatMapEager(Function, int, int) and overriding the default maximum concurrency value to be greater or equal to the expected number of groups, possibly using Integer.MAX_VALUE if the number of expected groups is unknown. 

 
Backpressure:
 
Both the returned and its inner Publishers honor backpressure and the source Publisher is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on downstream consumption). Note that both the returned and its inner Publishers use unbounded internal buffers and if the source Publisher doesn't honor backpressure, that may lead to OutOfMemoryError.
 
Scheduler:
 
groupBy does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that extracts the key for each item
Type parameters:
<K> – 
           the key type
See Also:
ReactiveX operators documentation: GroupBy
Returns:
a Publisher that emits GroupedFlowables, each of which corresponds to a unique key value and each of which emits those items from the source Publisher that share that key value/**
     * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these
     * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single
     * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the
     * source terminates, the next emission by the source having the same key will trigger a new
     * {@code GroupedPublisher} emission.
     * <p>
     * <img width="640" height="360" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/groupBy.png" alt="">
     * <p>
     * <em>Note:</em> A {@link GroupedFlowable} will cache the items it is to emit until such time as it
     * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those
     * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may
     * discard their buffers by applying an operator like {@link #ignoreElements} to them.
     * <p>
     * Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise,
     * the unconsumed groups may starve other groups due to the internal backpressure
     * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using
     * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency
     * value to be greater or equal to the expected number of groups, possibly using
     * {@code Integer.MAX_VALUE} if the number of expected groups is unknown.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher}
     *  is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on
     *  downstream consumption). Note that both the returned and its inner {@code Publisher}s use
     *  unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that <em>may</em>
     *  lead to {@code OutOfMemoryError}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code groupBy} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param keySelector
     *            a function that extracts the key for each item
     * @param <K>
     *            the key type
     * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a
     *         unique key value and each of which emits those items from the source Publisher that share that
     *         key value
     * @see <a href="http://reactivex.io/documentation/operators/groupby.html">ReactiveX operators documentation: GroupBy</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K> Flowable<GroupedFlowable<K, T>> groupBy(Function<? super T, ? extends K> keySelector) {
        return groupBy(keySelector, Functions.<T>identity(), false, bufferSize());
    }

    
Groups the items emitted by a Publisher according to a specified criterion, and emits these grouped items as GroupedFlowables. The emitted GroupedPublisher allows only a single Subscriber during its lifetime and if this Subscriber cancels before the source terminates, the next emission by the source having the same key will trigger a new GroupedPublisher emission. 



Note: A GroupedFlowable will cache the items it is to emit until such time as it is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those GroupedPublishers that do not concern you. Instead, you can signal to them that they may discard their buffers by applying an operator like ignoreElements to them. 
 Note that the GroupedFlowables should be subscribed to as soon as possible, otherwise, the unconsumed groups may starve other groups due to the internal backpressure coordination of the groupBy operator. Such hangs can be usually avoided by using flatMap(Function, int) or concatMapEager(Function, int, int) and overriding the default maximum concurrency value to be greater or equal to the expected number of groups, possibly using Integer.MAX_VALUE if the number of expected groups is unknown. 

 
Backpressure:
 
Both the returned and its inner Publishers honor backpressure and the source Publisher is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on downstream consumption). Note that both the returned and its inner Publishers use unbounded internal buffers and if the source Publisher doesn't honor backpressure, that may lead to OutOfMemoryError.
 
Scheduler:
 
groupBy does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that extracts the key for each item
delayError – 
           if true, the exception from the current Flowable is delayed in each group until that specific group emitted
           the normal values; if false, the exception bypasses values in the groups and is reported immediately.
Type parameters:
<K> – 
           the key type
See Also:
ReactiveX operators documentation: GroupBy
Returns:
a Publisher that emits GroupedFlowables, each of which corresponds to a unique key value and each of which emits those items from the source Publisher that share that key value/**
     * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these
     * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single
     * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the
     * source terminates, the next emission by the source having the same key will trigger a new
     * {@code GroupedPublisher} emission.
     * <p>
     * <img width="640" height="360" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/groupBy.png" alt="">
     * <p>
     * <em>Note:</em> A {@link GroupedFlowable} will cache the items it is to emit until such time as it
     * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those
     * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may
     * discard their buffers by applying an operator like {@link #ignoreElements} to them.
     * <p>
     * Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise,
     * the unconsumed groups may starve other groups due to the internal backpressure
     * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using
     * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency
     * value to be greater or equal to the expected number of groups, possibly using
     * {@code Integer.MAX_VALUE} if the number of expected groups is unknown.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher}
     *  is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on
     *  downstream consumption). Note that both the returned and its inner {@code Publisher}s use
     *  unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that <em>may</em>
     *  lead to {@code OutOfMemoryError}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code groupBy} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param keySelector
     *            a function that extracts the key for each item
     * @param <K>
     *            the key type
     * @param delayError
     *            if true, the exception from the current Flowable is delayed in each group until that specific group emitted
     *            the normal values; if false, the exception bypasses values in the groups and is reported immediately.
     * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a
     *         unique key value and each of which emits those items from the source Publisher that share that
     *         key value
     * @see <a href="http://reactivex.io/documentation/operators/groupby.html">ReactiveX operators documentation: GroupBy</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K> Flowable<GroupedFlowable<K, T>> groupBy(Function<? super T, ? extends K> keySelector, boolean delayError) {
        return groupBy(keySelector, Functions.<T>identity(), delayError, bufferSize());
    }

    
Groups the items emitted by a Publisher according to a specified criterion, and emits these grouped items as GroupedFlowables. The emitted GroupedPublisher allows only a single Subscriber during its lifetime and if this Subscriber cancels before the source terminates, the next emission by the source having the same key will trigger a new GroupedPublisher emission. 



Note: A GroupedFlowable will cache the items it is to emit until such time as it is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those GroupedPublishers that do not concern you. Instead, you can signal to them that they may discard their buffers by applying an operator like ignoreElements to them. 
 Note that the GroupedFlowables should be subscribed to as soon as possible, otherwise, the unconsumed groups may starve other groups due to the internal backpressure coordination of the groupBy operator. Such hangs can be usually avoided by using flatMap(Function, int) or concatMapEager(Function, int, int) and overriding the default maximum concurrency value to be greater or equal to the expected number of groups, possibly using Integer.MAX_VALUE if the number of expected groups is unknown. 

 
Backpressure:
 
Both the returned and its inner Publishers honor backpressure and the source Publisher is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on downstream consumption). Note that both the returned and its inner Publishers use unbounded internal buffers and if the source Publisher doesn't honor backpressure, that may lead to OutOfMemoryError.
 
Scheduler:
 
groupBy does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that extracts the key for each item
valueSelector – 
           a function that extracts the return element for each item
Type parameters:
<K> – 
           the key type
<V> – 
           the element type
See Also:
ReactiveX operators documentation: GroupBy
Returns:
a Publisher that emits GroupedFlowables, each of which corresponds to a unique key value and each of which emits those items from the source Publisher that share that key value/**
     * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these
     * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single
     * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the
     * source terminates, the next emission by the source having the same key will trigger a new
     * {@code GroupedPublisher} emission.
     * <p>
     * <img width="640" height="360" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/groupBy.png" alt="">
     * <p>
     * <em>Note:</em> A {@link GroupedFlowable} will cache the items it is to emit until such time as it
     * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those
     * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may
     * discard their buffers by applying an operator like {@link #ignoreElements} to them.
     * <p>
     * Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise,
     * the unconsumed groups may starve other groups due to the internal backpressure
     * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using
     * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency
     * value to be greater or equal to the expected number of groups, possibly using
     * {@code Integer.MAX_VALUE} if the number of expected groups is unknown.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher}
     *  is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on
     *  downstream consumption). Note that both the returned and its inner {@code Publisher}s use
     *  unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that <em>may</em>
     *  lead to {@code OutOfMemoryError}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code groupBy} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param keySelector
     *            a function that extracts the key for each item
     * @param valueSelector
     *            a function that extracts the return element for each item
     * @param <K>
     *            the key type
     * @param <V>
     *            the element type
     * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a
     *         unique key value and each of which emits those items from the source Publisher that share that
     *         key value
     * @see <a href="http://reactivex.io/documentation/operators/groupby.html">ReactiveX operators documentation: GroupBy</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Flowable<GroupedFlowable<K, V>> groupBy(Function<? super T, ? extends K> keySelector,
            Function<? super T, ? extends V> valueSelector) {
        return groupBy(keySelector, valueSelector, false, bufferSize());
    }

    
Groups the items emitted by a Publisher according to a specified criterion, and emits these grouped items as GroupedFlowables. The emitted GroupedPublisher allows only a single Subscriber during its lifetime and if this Subscriber cancels before the source terminates, the next emission by the source having the same key will trigger a new GroupedPublisher emission. 



Note: A GroupedFlowable will cache the items it is to emit until such time as it is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those GroupedPublishers that do not concern you. Instead, you can signal to them that they may discard their buffers by applying an operator like ignoreElements to them. 
 Note that the GroupedFlowables should be subscribed to as soon as possible, otherwise, the unconsumed groups may starve other groups due to the internal backpressure coordination of the groupBy operator. Such hangs can be usually avoided by using flatMap(Function, int) or concatMapEager(Function, int, int) and overriding the default maximum concurrency value to be greater or equal to the expected number of groups, possibly using Integer.MAX_VALUE if the number of expected groups is unknown. 

 
Backpressure:
 
Both the returned and its inner Publishers honor backpressure and the source Publisher is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on downstream consumption). Note that both the returned and its inner Publishers use unbounded internal buffers and if the source Publisher doesn't honor backpressure, that may lead to OutOfMemoryError.
 
Scheduler:
 
groupBy does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that extracts the key for each item
valueSelector – 
           a function that extracts the return element for each item
delayError – 
           if true, the exception from the current Flowable is delayed in each group until that specific group emitted
           the normal values; if false, the exception bypasses values in the groups and is reported immediately.
Type parameters:
<K> – 
           the key type
<V> – 
           the element type
See Also:
ReactiveX operators documentation: GroupBy
Returns:
a Publisher that emits GroupedFlowables, each of which corresponds to a unique key value and each of which emits those items from the source Publisher that share that key value/**
     * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these
     * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single
     * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the
     * source terminates, the next emission by the source having the same key will trigger a new
     * {@code GroupedPublisher} emission.
     * <p>
     * <img width="640" height="360" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/groupBy.png" alt="">
     * <p>
     * <em>Note:</em> A {@link GroupedFlowable} will cache the items it is to emit until such time as it
     * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those
     * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may
     * discard their buffers by applying an operator like {@link #ignoreElements} to them.
     * <p>
     * Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise,
     * the unconsumed groups may starve other groups due to the internal backpressure
     * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using
     * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency
     * value to be greater or equal to the expected number of groups, possibly using
     * {@code Integer.MAX_VALUE} if the number of expected groups is unknown.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher}
     *  is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on
     *  downstream consumption). Note that both the returned and its inner {@code Publisher}s use
     *  unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that <em>may</em>
     *  lead to {@code OutOfMemoryError}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code groupBy} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param keySelector
     *            a function that extracts the key for each item
     * @param valueSelector
     *            a function that extracts the return element for each item
     * @param <K>
     *            the key type
     * @param <V>
     *            the element type
     * @param delayError
     *            if true, the exception from the current Flowable is delayed in each group until that specific group emitted
     *            the normal values; if false, the exception bypasses values in the groups and is reported immediately.
     * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a
     *         unique key value and each of which emits those items from the source Publisher that share that
     *         key value
     * @see <a href="http://reactivex.io/documentation/operators/groupby.html">ReactiveX operators documentation: GroupBy</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Flowable<GroupedFlowable<K, V>> groupBy(Function<? super T, ? extends K> keySelector,
            Function<? super T, ? extends V> valueSelector, boolean delayError) {
        return groupBy(keySelector, valueSelector, delayError, bufferSize());
    }

    
Groups the items emitted by a Publisher according to a specified criterion, and emits these grouped items as GroupedFlowables. The emitted GroupedPublisher allows only a single Subscriber during its lifetime and if this Subscriber cancels before the source terminates, the next emission by the source having the same key will trigger a new GroupedPublisher emission. 



Note: A GroupedFlowable will cache the items it is to emit until such time as it is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those GroupedPublishers that do not concern you. Instead, you can signal to them that they may discard their buffers by applying an operator like ignoreElements to them. 
 Note that the GroupedFlowables should be subscribed to as soon as possible, otherwise, the unconsumed groups may starve other groups due to the internal backpressure coordination of the groupBy operator. Such hangs can be usually avoided by using flatMap(Function, int) or concatMapEager(Function, int, int) and overriding the default maximum concurrency value to be greater or equal to the expected number of groups, possibly using Integer.MAX_VALUE if the number of expected groups is unknown. 

 
Backpressure:
 
Both the returned and its inner Publishers honor backpressure and the source Publisher is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on downstream consumption). Note that both the returned and its inner Publishers use unbounded internal buffers and if the source Publisher doesn't honor backpressure, that may lead to OutOfMemoryError.
 
Scheduler:
 
groupBy does not operate by default on a particular Scheduler.

Params:
keySelector – 
           a function that extracts the key for each item
valueSelector – 
           a function that extracts the return element for each item
delayError – 
           if true, the exception from the current Flowable is delayed in each group until that specific group emitted
           the normal values; if false, the exception bypasses values in the groups and is reported immediately.
bufferSize –  the hint for how many GroupedFlowables and element in each GroupedFlowable should be buffered
Type parameters:
<K> – 
           the key type
<V> – 
           the element type
See Also:
ReactiveX operators documentation: GroupBy
Returns:
a Publisher that emits GroupedFlowables, each of which corresponds to a unique key value and each of which emits those items from the source Publisher that share that key value/**
     * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these
     * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedPublisher} allows only a single
     * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the
     * source terminates, the next emission by the source having the same key will trigger a new
     * {@code GroupedPublisher} emission.
     * <p>
     * <img width="640" height="360" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/groupBy.png" alt="">
     * <p>
     * <em>Note:</em> A {@link GroupedFlowable} will cache the items it is to emit until such time as it
     * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those
     * {@code GroupedPublisher}s that do not concern you. Instead, you can signal to them that they may
     * discard their buffers by applying an operator like {@link #ignoreElements} to them.
     * <p>
     * Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise,
     * the unconsumed groups may starve other groups due to the internal backpressure
     * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using
     * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency
     * value to be greater or equal to the expected number of groups, possibly using
     * {@code Integer.MAX_VALUE} if the number of expected groups is unknown.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Both the returned and its inner {@code Publisher}s honor backpressure and the source {@code Publisher}
     *  is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on
     *  downstream consumption). Note that both the returned and its inner {@code Publisher}s use
     *  unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that <em>may</em>
     *  lead to {@code OutOfMemoryError}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code groupBy} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param keySelector
     *            a function that extracts the key for each item
     * @param valueSelector
     *            a function that extracts the return element for each item
     * @param delayError
     *            if true, the exception from the current Flowable is delayed in each group until that specific group emitted
     *            the normal values; if false, the exception bypasses values in the groups and is reported immediately.
     * @param bufferSize
     *            the hint for how many {@link GroupedFlowable}s and element in each {@link GroupedFlowable} should be buffered
     * @param <K>
     *            the key type
     * @param <V>
     *            the element type
     * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a
     *         unique key value and each of which emits those items from the source Publisher that share that
     *         key value
     * @see <a href="http://reactivex.io/documentation/operators/groupby.html">ReactiveX operators documentation: GroupBy</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Flowable<GroupedFlowable<K, V>> groupBy(Function<? super T, ? extends K> keySelector,
            Function<? super T, ? extends V> valueSelector,
            boolean delayError, int bufferSize) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        ObjectHelper.requireNonNull(valueSelector, "valueSelector is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");

        return RxJavaPlugins.onAssembly(new FlowableGroupBy<T, K, V>(this, keySelector, valueSelector, bufferSize, delayError, null));
    }

    
Groups the items emitted by a Publisher according to a specified criterion, and emits these grouped items as GroupedFlowables. The emitted GroupedFlowable allows only a single Subscriber during its lifetime and if this Subscriber cancels before the source terminates, the next emission by the source having the same key will trigger a new GroupedPublisher emission. The evictingMapFactory is used to create a map that will be used to hold the GroupedFlowables by key. The evicting map created by this factory must notify the provided Consumer<Object> with the entry value (not the key!) when an entry in this map has been evicted. The next source emission will bring about the completion of the evicted GroupedFlowables and the arrival of an item with the same key as a completed GroupedFlowable will prompt the creation and emission of a new GroupedFlowable with that key. 
A use case for specifying an evictingMapFactory is where the source is infinite and fast and over time the number of keys grows enough to be a concern in terms of the memory footprint of the internal hash map containing the GroupedFlowables. 
The map created by an evictingMapFactory must be thread-safe. 
An example of an evictingMapFactory using CacheBuilder from the Guava library is below:

Function<Consumer<Object>, Map<Integer, Object>> evictingMapFactory =
  notify ->
      CacheBuilder
        .newBuilder()
        .maximumSize(3)
        .removalListener(entry -> {
             try {
                 // emit the value not the key!
                 notify.accept(entry.getValue());
             } catch (Exception e) {
                 throw new RuntimeException(e);
             }
           })
        .<Integer, Object> build()
        .asMap();
// Emit 1000 items but ensure that the
// internal map never has more than 3 items in it
Flowable
  .range(1, 1000)
  // note that number of keys is 10
  .groupBy(x -> x % 10, x -> x, true, 16, evictingMapFactory)
  .flatMap(g -> g)
  .forEach(System.out::println);




Note: A GroupedFlowable will cache the items it is to emit until such time as it is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those GroupedFlowables that do not concern you. Instead, you can signal to them that they may discard their buffers by applying an operator like ignoreElements to them. 
 Note that the GroupedFlowables should be subscribed to as soon as possible, otherwise, the unconsumed groups may starve other groups due to the internal backpressure coordination of the groupBy operator. Such hangs can be usually avoided by using flatMap(Function, int) or concatMapEager(Function, int, int) and overriding the default maximum concurrency value to be greater or equal to the expected number of groups, possibly using Integer.MAX_VALUE if the number of expected groups is unknown. 

 
Backpressure:
 
Both the returned and its inner GroupedFlowables honor backpressure and the source Publisher is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on downstream consumption). Note that both the returned and its inner GroupedFlowables use unbounded internal buffers and if the source Publisher doesn't honor backpressure, that may lead to OutOfMemoryError.
 
Scheduler:
 
groupBy does not operate by default on a particular Scheduler.

History: 2.1.10 - beta
Params:
keySelector – 
           a function that extracts the key for each item
valueSelector – 
           a function that extracts the return element for each item
delayError – 
           if true, the exception from the current Flowable is delayed in each group until that specific group emitted
           the normal values; if false, the exception bypasses values in the groups and is reported immediately.
bufferSize –  the hint for how many GroupedFlowables and element in each GroupedFlowable should be buffered
evictingMapFactory –  The factory used to create a map that will be used by the implementation to hold the GroupedFlowables. The evicting map created by this factory must notify the provided Consumer<Object> with the entry value (not the key!) when an entry in this map has been evicted. The next source emission will bring about the completion of the evicted GroupedFlowables. See example above.
Type parameters:
<K> – 
           the key type
<V> – 
           the element type
See Also:
ReactiveX operators documentation: GroupBy
Returns:
a Publisher that emits GroupedFlowables, each of which corresponds to a unique key value and each of which emits those items from the source Publisher that share that key value
Since:
2.2/**
     * Groups the items emitted by a {@code Publisher} according to a specified criterion, and emits these
     * grouped items as {@link GroupedFlowable}s. The emitted {@code GroupedFlowable} allows only a single
     * {@link Subscriber} during its lifetime and if this {@code Subscriber} cancels before the
     * source terminates, the next emission by the source having the same key will trigger a new
     * {@code GroupedPublisher} emission. The {@code evictingMapFactory} is used to create a map that will
     * be used to hold the {@link GroupedFlowable}s by key. The evicting map created by this factory must
     * notify the provided {@code Consumer<Object>} with the entry value (not the key!) when an entry in this
     * map has been evicted. The next source emission will bring about the completion of the evicted
     * {@link GroupedFlowable}s and the arrival of an item with the same key as a completed {@link GroupedFlowable}
     * will prompt the creation and emission of a new {@link GroupedFlowable} with that key.
     * 
     * <p>A use case for specifying an {@code evictingMapFactory} is where the source is infinite and fast and
     * over time the number of keys grows enough to be a concern in terms of the memory footprint of the
     * internal hash map containing the {@link GroupedFlowable}s.
     * 
     * <p>The map created by an {@code evictingMapFactory} must be thread-safe.
     * 
     * <p>An example of an {@code evictingMapFactory} using <a href="https://google.github.io/guava/releases/24.0-jre/api/docs/com/google/common/cache/CacheBuilder.html">CacheBuilder</a> from the Guava library is below:
     * 
     * <pre><code>
     * Function&lt;Consumer&lt;Object&gt;, Map&lt;Integer, Object&gt;&gt; evictingMapFactory =
     *   notify -&gt;
     *       CacheBuilder
     *         .newBuilder()
     *         .maximumSize(3)
     *         .removalListener(entry -&gt; {
     *              try {
     *                  // emit the value not the key!
     *                  notify.accept(entry.getValue());
     *              } catch (Exception e) {
     *                  throw new RuntimeException(e);
     *              }
     *            })
     *         .&lt;Integer, Object&gt; build()
     *         .asMap();
     *
     * // Emit 1000 items but ensure that the
     * // internal map never has more than 3 items in it
     * Flowable
     *   .range(1, 1000)
     *   // note that number of keys is 10
     *   .groupBy(x -&gt; x % 10, x -&gt; x, true, 16, evictingMapFactory)
     *   .flatMap(g -&gt; g)
     *   .forEach(System.out::println);
     * </code></pre>
     * 
     * <p>
     * <img width="640" height="360" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/groupBy.png" alt="">
     * <p>
     * <em>Note:</em> A {@link GroupedFlowable} will cache the items it is to emit until such time as it
     * is subscribed to. For this reason, in order to avoid memory leaks, you should not simply ignore those
     * {@code GroupedFlowable}s that do not concern you. Instead, you can signal to them that they may
     * discard their buffers by applying an operator like {@link #ignoreElements} to them.
     * <p>
     * Note that the {@link GroupedFlowable}s should be subscribed to as soon as possible, otherwise,
     * the unconsumed groups may starve other groups due to the internal backpressure
     * coordination of the {@code groupBy} operator. Such hangs can be usually avoided by using
     * {@link #flatMap(Function, int)} or {@link #concatMapEager(Function, int, int)} and overriding the default maximum concurrency
     * value to be greater or equal to the expected number of groups, possibly using
     * {@code Integer.MAX_VALUE} if the number of expected groups is unknown.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Both the returned and its inner {@code GroupedFlowable}s honor backpressure and the source {@code Publisher}
     *  is consumed in a bounded mode (i.e., requested a fixed amount upfront and replenished based on
     *  downstream consumption). Note that both the returned and its inner {@code GroupedFlowable}s use
     *  unbounded internal buffers and if the source {@code Publisher} doesn't honor backpressure, that <em>may</em>
     *  lead to {@code OutOfMemoryError}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code groupBy} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.10 - beta
     * @param keySelector
     *            a function that extracts the key for each item
     * @param valueSelector
     *            a function that extracts the return element for each item
     * @param delayError
     *            if true, the exception from the current Flowable is delayed in each group until that specific group emitted
     *            the normal values; if false, the exception bypasses values in the groups and is reported immediately.
     * @param bufferSize
     *            the hint for how many {@link GroupedFlowable}s and element in each {@link GroupedFlowable} should be buffered
     * @param evictingMapFactory
     *            The factory used to create a map that will be used by the implementation to hold the
     *            {@link GroupedFlowable}s. The evicting map created by this factory must
     *            notify the provided {@code Consumer<Object>} with the entry value (not the key!) when
     *            an entry in this map has been evicted. The next source emission will bring about the
     *            completion of the evicted {@link GroupedFlowable}s. See example above.
     * @param <K>
     *            the key type
     * @param <V>
     *            the element type
     * @return a {@code Publisher} that emits {@link GroupedFlowable}s, each of which corresponds to a
     *         unique key value and each of which emits those items from the source Publisher that share that
     *         key value
     * @see <a href="http://reactivex.io/documentation/operators/groupby.html">ReactiveX operators documentation: GroupBy</a>
     *
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Flowable<GroupedFlowable<K, V>> groupBy(Function<? super T, ? extends K> keySelector,
            Function<? super T, ? extends V> valueSelector,
            boolean delayError, int bufferSize,
            Function<? super Consumer<Object>, ? extends Map<K, Object>> evictingMapFactory) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        ObjectHelper.requireNonNull(valueSelector, "valueSelector is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        ObjectHelper.requireNonNull(evictingMapFactory, "evictingMapFactory is null");

        return RxJavaPlugins.onAssembly(new FlowableGroupBy<T, K, V>(this, keySelector, valueSelector, bufferSize, delayError, evictingMapFactory));
    }

    
Returns a Flowable that correlates two Publishers when they overlap in time and groups the results.

There are no guarantees in what order the items get combined when multiple
items from one or both source Publishers overlap.



 
Backpressure:
 
The operator doesn't support backpressure and consumes all participating Publishers in an unbounded mode (i.e., not applying any backpressure to them).
 
Scheduler:
 
groupJoin does not operate by default on a particular Scheduler.

Params:
other – 
           the other Publisher to correlate items from the source Publisher with
leftEnd – 
           a function that returns a Publisher whose emissions indicate the duration of the values of
           the source Publisher
rightEnd –  a function that returns a Publisher whose emissions indicate the duration of the values of the right Publisher
resultSelector – 
           a function that takes an item emitted by each Publisher and returns the value to be emitted
           by the resulting Publisher
Type parameters:
<TRight> – the value type of the right Publisher source
<TLeftEnd> – the element type of the left duration Publishers
<TRightEnd> – the element type of the right duration Publishers
<R> – the result type
See Also:
ReactiveX operators documentation: Join
Returns:
a Flowable that emits items based on combining those items emitted by the source Publishers
        whose durations overlap/**
     * Returns a Flowable that correlates two Publishers when they overlap in time and groups the results.
     * <p>
     * There are no guarantees in what order the items get combined when multiple
     * items from one or both source Publishers overlap.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/groupJoin.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure and consumes all participating {@code Publisher}s in
     *  an unbounded mode (i.e., not applying any backpressure to them).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code groupJoin} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <TRight> the value type of the right Publisher source
     * @param <TLeftEnd> the element type of the left duration Publishers
     * @param <TRightEnd> the element type of the right duration Publishers
     * @param <R> the result type
     * @param other
     *            the other Publisher to correlate items from the source Publisher with
     * @param leftEnd
     *            a function that returns a Publisher whose emissions indicate the duration of the values of
     *            the source Publisher
     * @param rightEnd
     *            a function that returns a Publisher whose emissions indicate the duration of the values of
     *            the {@code right} Publisher
     * @param resultSelector
     *            a function that takes an item emitted by each Publisher and returns the value to be emitted
     *            by the resulting Publisher
     * @return a Flowable that emits items based on combining those items emitted by the source Publishers
     *         whose durations overlap
     * @see <a href="http://reactivex.io/documentation/operators/join.html">ReactiveX operators documentation: Join</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <TRight, TLeftEnd, TRightEnd, R> Flowable<R> groupJoin(
            Publisher<? extends TRight> other,
            Function<? super T, ? extends Publisher<TLeftEnd>> leftEnd,
            Function<? super TRight, ? extends Publisher<TRightEnd>> rightEnd,
            BiFunction<? super T, ? super Flowable<TRight>, ? extends R> resultSelector) {
        ObjectHelper.requireNonNull(other, "other is null");
        ObjectHelper.requireNonNull(leftEnd, "leftEnd is null");
        ObjectHelper.requireNonNull(rightEnd, "rightEnd is null");
        ObjectHelper.requireNonNull(resultSelector, "resultSelector is null");
        return RxJavaPlugins.onAssembly(new FlowableGroupJoin<T, TRight, TLeftEnd, TRightEnd, R>(
                this, other, leftEnd, rightEnd, resultSelector));
    }

    
Hides the identity of this Flowable and its Subscription.
Allows hiding extra features such as Processor's Subscriber methods or preventing certain identity-based optimizations (fusion). 

 
Backpressure:
 
The operator is a pass-through for backpressure, the behavior is determined by the upstream's
 backpressure behavior.
 
Scheduler:
 
hide does not operate by default on a particular Scheduler.

Returns:
the new Flowable instance
Since:
2.0/**
     * Hides the identity of this Flowable and its Subscription.
     * <p>Allows hiding extra features such as {@link Processor}'s
     * {@link Subscriber} methods or preventing certain identity-based
     * optimizations (fusion).
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator is a pass-through for backpressure, the behavior is determined by the upstream's
     *  backpressure behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code hide} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @return the new Flowable instance
     *
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> hide() {
        return RxJavaPlugins.onAssembly(new FlowableHide<T>(this));
    }

    
Ignores all items emitted by the source Publisher and only calls onComplete or onError. 



 
Backpressure:
 
This operator ignores backpressure as it doesn't emit any elements and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
ignoreElements does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: IgnoreElements
Returns:
a Completable that only calls onComplete or onError, based on which one is called by the source Publisher/**
     * Ignores all items emitted by the source Publisher and only calls {@code onComplete} or {@code onError}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/ignoreElements.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator ignores backpressure as it doesn't emit any elements and consumes the source {@code Publisher}
     *  in an unbounded manner (i.e., no backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code ignoreElements} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Completable that only calls {@code onComplete} or {@code onError}, based on which one is
     *         called by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/ignoreelements.html">ReactiveX operators documentation: IgnoreElements</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Completable ignoreElements() {
        return RxJavaPlugins.onAssembly(new FlowableIgnoreElementsCompletable<T>(this));
    }

    
Returns a Single that emits true if the source Publisher is empty, otherwise false. 
 In Rx.Net this is negated as the any Subscriber but we renamed this in RxJava to better match Java naming idioms. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
isEmpty does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Contains
Returns:
a Flowable that emits a Boolean/**
     * Returns a Single that emits {@code true} if the source Publisher is empty, otherwise {@code false}.
     * <p>
     * In Rx.Net this is negated as the {@code any} Subscriber but we renamed this in RxJava to better match Java
     * naming idioms.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/isEmpty.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code isEmpty} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits a Boolean
     * @see <a href="http://reactivex.io/documentation/operators/contains.html">ReactiveX operators documentation: Contains</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<Boolean> isEmpty() {
        return all(Functions.alwaysFalse());
    }

    
Correlates the items emitted by two Publishers based on overlapping durations.

There are no guarantees in what order the items get combined when multiple
items from one or both source Publishers overlap.



 
Backpressure:
 
The operator doesn't support backpressure and consumes all participating Publishers in an unbounded mode (i.e., not applying any backpressure to them).
 
Scheduler:
 
join does not operate by default on a particular Scheduler.

Params:
other – 
           the second Publisher to join items from
leftEnd – 
           a function to select a duration for each item emitted by the source Publisher, used to
           determine overlap
rightEnd –  a function to select a duration for each item emitted by the right Publisher, used to determine overlap
resultSelector – 
           a function that computes an item to be emitted by the resulting Publisher for any two
           overlapping items emitted by the two Publishers
Type parameters:
<TRight> – the value type of the right Publisher source
<TLeftEnd> – the element type of the left duration Publishers
<TRightEnd> – the element type of the right duration Publishers
<R> – the result type
See Also:
ReactiveX operators documentation: Join
Returns:
a Flowable that emits items correlating to items emitted by the source Publishers that have
        overlapping durations/**
     * Correlates the items emitted by two Publishers based on overlapping durations.
     * <p>
     * There are no guarantees in what order the items get combined when multiple
     * items from one or both source Publishers overlap.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/join_.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure and consumes all participating {@code Publisher}s in
     *  an unbounded mode (i.e., not applying any backpressure to them).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code join} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <TRight> the value type of the right Publisher source
     * @param <TLeftEnd> the element type of the left duration Publishers
     * @param <TRightEnd> the element type of the right duration Publishers
     * @param <R> the result type
     * @param other
     *            the second Publisher to join items from
     * @param leftEnd
     *            a function to select a duration for each item emitted by the source Publisher, used to
     *            determine overlap
     * @param rightEnd
     *            a function to select a duration for each item emitted by the {@code right} Publisher, used to
     *            determine overlap
     * @param resultSelector
     *            a function that computes an item to be emitted by the resulting Publisher for any two
     *            overlapping items emitted by the two Publishers
     * @return a Flowable that emits items correlating to items emitted by the source Publishers that have
     *         overlapping durations
     * @see <a href="http://reactivex.io/documentation/operators/join.html">ReactiveX operators documentation: Join</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <TRight, TLeftEnd, TRightEnd, R> Flowable<R> join(
            Publisher<? extends TRight> other,
            Function<? super T, ? extends Publisher<TLeftEnd>> leftEnd,
            Function<? super TRight, ? extends Publisher<TRightEnd>> rightEnd,
            BiFunction<? super T, ? super TRight, ? extends R> resultSelector) {
        ObjectHelper.requireNonNull(other, "other is null");
        ObjectHelper.requireNonNull(leftEnd, "leftEnd is null");
        ObjectHelper.requireNonNull(rightEnd, "rightEnd is null");
        ObjectHelper.requireNonNull(resultSelector, "resultSelector is null");
        return RxJavaPlugins.onAssembly(new FlowableJoin<T, TRight, TLeftEnd, TRightEnd, R>(
                this, other, leftEnd, rightEnd, resultSelector));
    }

    
Returns a Maybe that emits the last item emitted by this Flowable or completes if
this Flowable is empty.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
lastElement does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Last
Returns:
a new Maybe instance/**
     * Returns a Maybe that emits the last item emitted by this Flowable or completes if
     * this Flowable is empty.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/last.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code lastElement} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a new Maybe instance
     * @see <a href="http://reactivex.io/documentation/operators/last.html">ReactiveX operators documentation: Last</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Maybe<T> lastElement() {
        return RxJavaPlugins.onAssembly(new FlowableLastMaybe<T>(this));
    }

    
Returns a Single that emits only the last item emitted by this Flowable, or a default item
if this Flowable completes without emitting any items.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
last does not operate by default on a particular Scheduler.

Params:
defaultItem – 
           the default item to emit if the source Publisher is empty
See Also:
ReactiveX operators documentation: Last
Returns:
the new Single instance/**
     * Returns a Single that emits only the last item emitted by this Flowable, or a default item
     * if this Flowable completes without emitting any items.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/lastOrDefault.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code last} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param defaultItem
     *            the default item to emit if the source Publisher is empty
     * @return the new Single instance
     * @see <a href="http://reactivex.io/documentation/operators/last.html">ReactiveX operators documentation: Last</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> last(T defaultItem) {
        ObjectHelper.requireNonNull(defaultItem, "defaultItem");
        return RxJavaPlugins.onAssembly(new FlowableLastSingle<T>(this, defaultItem));
    }

    
Returns a Single that emits only the last item emitted by this Flowable or signals a NoSuchElementException if this Flowable is empty. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
lastOrError does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Last
Returns:
the new Single instance/**
     * Returns a Single that emits only the last item emitted by this Flowable or signals
     * a {@link NoSuchElementException} if this Flowable is empty.
     * <p>
     * <img width="640" height="236" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/lastOrError.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code lastOrError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the new Single instance
     * @see <a href="http://reactivex.io/documentation/operators/last.html">ReactiveX operators documentation: Last</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> lastOrError() {
        return RxJavaPlugins.onAssembly(new FlowableLastSingle<T>(this, null));
    }

    
This method requires advanced knowledge about building operators, please consider
other standard composition methods first; Returns a Flowable which, when subscribed to, invokes the apply(Subscriber) method of the provided FlowableOperator for each individual downstream Subscriber and allows the insertion of a custom operator by accessing the downstream's Subscriber during this subscription phase and providing a new Subscriber, containing the custom operator's intended business logic, that will be used in the subscription process going further upstream. 
 Generally, such a new Subscriber will wrap the downstream's Subscriber and forwards the onNext, onError and onComplete events from the upstream directly or according to the emission pattern the custom operator's business logic requires. In addition, such operator can intercept the flow control calls of cancel and request that would have traveled upstream and perform additional actions depending on the same business logic requirements. 

Example:

// Step 1: Create the consumer type that will be returned by the FlowableOperator.apply():
public final class CustomSubscriber<T> implements FlowableSubscriber<T>, Subscription {
    // The downstream's Subscriber that will receive the onXXX events
    final Subscriber<? super String> downstream;
    // The connection to the upstream source that will call this class' onXXX methods
    Subscription upstream;
    // The constructor takes the downstream subscriber and usually any other parameters
    public CustomSubscriber(Subscriber<? super String> downstream) {
        this.downstream = downstream;
    }
    // In the subscription phase, the upstream sends a Subscription to this class
    // and subsequently this class has to send a Subscription to the downstream.
    // Note that relaying the upstream's Subscription instance directly is not allowed in RxJava
    @Override
    public void onSubscribe(Subscription s) {
        if (upstream != null) {
            s.cancel();
        } else {
            upstream = s;
            downstream.onSubscribe(this);
        }
    }
    // The upstream calls this with the next item and the implementation's
    // responsibility is to emit an item to the downstream based on the intended
    // business logic, or if it can't do so for the particular item,
    // request more from the upstream
    @Override
    public void onNext(T item) {
        String str = item.toString();
        if (str.length() < 2) {
            downstream.onNext(str);
        } else {
            upstream.request(1);
        }
    }
    // Some operators may handle the upstream's error while others
    // could just forward it to the downstream.
    @Override
    public void onError(Throwable throwable) {
        downstream.onError(throwable);
    }
    // When the upstream completes, usually the downstream should complete as well.
    @Override
    public void onComplete() {
        downstream.onComplete();
    }
    // Some operators have to intercept the downstream's request calls to trigger
    // the emission of queued items while others can simply forward the request
    // amount as is.
    @Override
    public void request(long n) {
        upstream.request(n);
    }
    // Some operators may use their own resources which should be cleaned up if
    // the downstream cancels the flow before it completed. Operators without
    // resources can simply forward the cancellation to the upstream.
    // In some cases, a canceled flag may be set by this method so that other parts
    // of this class may detect the cancellation and stop sending events
    // to the downstream.
    @Override
    public void cancel() {
        upstream.cancel();
    }
}
// Step 2: Create a class that implements the FlowableOperator interface and
//         returns the custom consumer type from above in its apply() method.
//         Such class may define additional parameters to be submitted to
//         the custom consumer type.
final class CustomOperator<T> implements FlowableOperator<String> {
    @Override
    public Subscriber<? super String> apply(Subscriber<? super T> upstream) {
        return new CustomSubscriber<T>(upstream);
    }
}
// Step 3: Apply the custom operator via lift() in a flow by creating an instance of it
//         or reusing an existing one.
Flowable.range(5, 10)
.lift(new CustomOperator<Integer>())
.test()
.assertResult("5", "6", "7", "8", "9");


Creating custom operators can be complicated and it is recommended one consults the
RxJava wiki: Writing operators page about
the tools, requirements, rules, considerations and pitfalls of implementing them.
 Note that implementing custom operators via this lift() method adds slightly more overhead by requiring an additional allocation and indirection per assembled flows. Instead, extending the abstract Flowable class and creating a FlowableTransformer with it is recommended. 
 Note also that it is not possible to stop the subscription phase in lift() as the apply() method requires a non-null Subscriber instance to be returned, which is then unconditionally subscribed to the upstream Flowable. For example, if the operator decided there is no reason to subscribe to the upstream source because of some optimization possibility or a failure to prepare the operator, it still has to return a Subscriber that should immediately cancel the upstream's Subscription in its onSubscribe method. Again, using a FlowableTransformer and extending the Flowable is a better option as subscribeActual can decide to not subscribe to its upstream after all. 

 
Backpressure:
 
The Subscriber instance returned by the FlowableOperator is responsible to be backpressure-aware or document the fact that the consumer of the returned Publisher has to apply one of the onBackpressureXXX operators.
 
Scheduler:
 
lift does not operate by default on a particular Scheduler, however, the FlowableOperator may use a Scheduler to support its own asynchronous behavior.

Params:
lifter – the FlowableOperator that receives the downstream's Subscriber and should return a Subscriber with custom behavior to be used as the consumer for the current Flowable.
Type parameters:
<R> – the output value type
See Also:
RxJava wiki: Writing operators
compose(FlowableTransformer)
Returns:
the new Flowable instance/**
     * <strong>This method requires advanced knowledge about building operators, please consider
     * other standard composition methods first;</strong>
     * Returns a {@code Flowable} which, when subscribed to, invokes the {@link FlowableOperator#apply(Subscriber) apply(Subscriber)} method
     * of the provided {@link FlowableOperator} for each individual downstream {@link Subscriber} and allows the
     * insertion of a custom operator by accessing the downstream's {@link Subscriber} during this subscription phase
     * and providing a new {@code Subscriber}, containing the custom operator's intended business logic, that will be
     * used in the subscription process going further upstream.
     * <p>
     * Generally, such a new {@code Subscriber} will wrap the downstream's {@code Subscriber} and forwards the
     * {@code onNext}, {@code onError} and {@code onComplete} events from the upstream directly or according to the
     * emission pattern the custom operator's business logic requires. In addition, such operator can intercept the
     * flow control calls of {@code cancel} and {@code request} that would have traveled upstream and perform
     * additional actions depending on the same business logic requirements.
     * <p>
     * Example:
     * <pre><code>
     * // Step 1: Create the consumer type that will be returned by the FlowableOperator.apply():
     * 
     * public final class CustomSubscriber&lt;T&gt; implements FlowableSubscriber&lt;T&gt;, Subscription {
     *
     *     // The downstream's Subscriber that will receive the onXXX events
     *     final Subscriber&lt;? super String&gt; downstream;
     *
     *     // The connection to the upstream source that will call this class' onXXX methods
     *     Subscription upstream;
     *
     *     // The constructor takes the downstream subscriber and usually any other parameters
     *     public CustomSubscriber(Subscriber&lt;? super String&gt; downstream) {
     *         this.downstream = downstream;
     *     }
     *
     *     // In the subscription phase, the upstream sends a Subscription to this class
     *     // and subsequently this class has to send a Subscription to the downstream.
     *     // Note that relaying the upstream's Subscription instance directly is not allowed in RxJava
     *     &#64;Override
     *     public void onSubscribe(Subscription s) {
     *         if (upstream != null) {
     *             s.cancel();
     *         } else {
     *             upstream = s;
     *             downstream.onSubscribe(this);
     *         }
     *     }
     *
     *     // The upstream calls this with the next item and the implementation's
     *     // responsibility is to emit an item to the downstream based on the intended
     *     // business logic, or if it can't do so for the particular item,
     *     // request more from the upstream
     *     &#64;Override
     *     public void onNext(T item) {
     *         String str = item.toString();
     *         if (str.length() &lt; 2) {
     *             downstream.onNext(str);
     *         } else {
     *             upstream.request(1);
     *         }
     *     }
     *
     *     // Some operators may handle the upstream's error while others
     *     // could just forward it to the downstream.
     *     &#64;Override
     *     public void onError(Throwable throwable) {
     *         downstream.onError(throwable);
     *     }
     *
     *     // When the upstream completes, usually the downstream should complete as well.
     *     &#64;Override
     *     public void onComplete() {
     *         downstream.onComplete();
     *     }
     *
     *     // Some operators have to intercept the downstream's request calls to trigger
     *     // the emission of queued items while others can simply forward the request
     *     // amount as is.
     *     &#64;Override
     *     public void request(long n) {
     *         upstream.request(n);
     *     }
     *
     *     // Some operators may use their own resources which should be cleaned up if
     *     // the downstream cancels the flow before it completed. Operators without
     *     // resources can simply forward the cancellation to the upstream.
     *     // In some cases, a canceled flag may be set by this method so that other parts
     *     // of this class may detect the cancellation and stop sending events
     *     // to the downstream.
     *     &#64;Override
     *     public void cancel() {
     *         upstream.cancel();
     *     }
     * }
     *
     * // Step 2: Create a class that implements the FlowableOperator interface and
     * //         returns the custom consumer type from above in its apply() method.
     * //         Such class may define additional parameters to be submitted to
     * //         the custom consumer type.
     *
     * final class CustomOperator&lt;T&gt; implements FlowableOperator&lt;String&gt; {
     *     &#64;Override
     *     public Subscriber&lt;? super String&gt; apply(Subscriber&lt;? super T&gt; upstream) {
     *         return new CustomSubscriber&lt;T&gt;(upstream);
     *     }
     * }
     *
     * // Step 3: Apply the custom operator via lift() in a flow by creating an instance of it
     * //         or reusing an existing one.
     *
     * Flowable.range(5, 10)
     * .lift(new CustomOperator&lt;Integer&gt;())
     * .test()
     * .assertResult("5", "6", "7", "8", "9");
     * </code></pre>
     * <p>
     * Creating custom operators can be complicated and it is recommended one consults the
     * <a href="https://github.com/ReactiveX/RxJava/wiki/Writing-operators-for-2.0">RxJava wiki: Writing operators</a> page about
     * the tools, requirements, rules, considerations and pitfalls of implementing them.
     * <p>
     * Note that implementing custom operators via this {@code lift()} method adds slightly more overhead by requiring
     * an additional allocation and indirection per assembled flows. Instead, extending the abstract {@code Flowable}
     * class and creating a {@link FlowableTransformer} with it is recommended.
     * <p>
     * Note also that it is not possible to stop the subscription phase in {@code lift()} as the {@code apply()} method
     * requires a non-null {@code Subscriber} instance to be returned, which is then unconditionally subscribed to
     * the upstream {@code Flowable}. For example, if the operator decided there is no reason to subscribe to the
     * upstream source because of some optimization possibility or a failure to prepare the operator, it still has to
     * return a {@code Subscriber} that should immediately cancel the upstream's {@code Subscription} in its
     * {@code onSubscribe} method. Again, using a {@code FlowableTransformer} and extending the {@code Flowable} is
     * a better option as {@link #subscribeActual} can decide to not subscribe to its upstream after all.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The {@code Subscriber} instance returned by the {@link FlowableOperator} is responsible to be
     *  backpressure-aware or document the fact that the consumer of the returned {@code Publisher} has to apply one of
     *  the {@code onBackpressureXXX} operators.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code lift} does not operate by default on a particular {@link Scheduler}, however, the
     *  {@link FlowableOperator} may use a {@code Scheduler} to support its own asynchronous behavior.</dd>
     * </dl>
     *
     * @param <R> the output value type
     * @param lifter the {@link FlowableOperator} that receives the downstream's {@code Subscriber} and should return
     *               a {@code Subscriber} with custom behavior to be used as the consumer for the current
     *               {@code Flowable}.
     * @return the new Flowable instance
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Writing-operators-for-2.0">RxJava wiki: Writing operators</a>
     * @see #compose(FlowableTransformer)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> lift(FlowableOperator<? extends R, ? super T> lifter) {
        ObjectHelper.requireNonNull(lifter, "lifter is null");
        return RxJavaPlugins.onAssembly(new FlowableLift<R, T>(this, lifter));
    }

    
Limits both the number of upstream items (after which the sequence completes)
and the total downstream request amount requested from the upstream to
possibly prevent the creation of excess items by the upstream.
 The operator requests at most the given count of items from upstream even if the downstream requests more than that. For example, given a limit(5), if the downstream requests 1, a request of 1 is submitted to the upstream and the operator remembers that only 4 items can be requested now on. A request of 5 at this point will request 4 from the upstream and any subsequent requests will be ignored. 
 Note that requests are negotiated on an operator boundary and limit's amount may not be preserved further upstream. For example, source.observeOn(Schedulers.computation()).limit(5) will still request the default (128) elements from the given source. 
 The main use of this operator is with sources that are async boundaries that don't interfere with request amounts, such as certain Flowable-based network endpoints that relay downstream request amounts unchanged and are, therefore, prone to trigger excessive item creation/transmission over the network. 

 
Backpressure:
 
The operator requests a total of the given count items from the upstream.
 
Scheduler:
 
limit does not operate by default on a particular Scheduler.

History: 2.1.6 - experimental
Params:
count – the maximum number of items and the total request amount, non-negative.
             Zero will immediately cancel the upstream on subscription and complete
             the downstream.
See Also:
take(long)
rebatchRequests(int)
Returns:
the new Flowable instance
Since:
2.2/**
     * Limits both the number of upstream items (after which the sequence completes)
     * and the total downstream request amount requested from the upstream to
     * possibly prevent the creation of excess items by the upstream.
     * <p>
     * The operator requests at most the given {@code count} of items from upstream even
     * if the downstream requests more than that. For example, given a {@code limit(5)},
     * if the downstream requests 1, a request of 1 is submitted to the upstream
     * and the operator remembers that only 4 items can be requested now on. A request
     * of 5 at this point will request 4 from the upstream and any subsequent requests will
     * be ignored.
     * <p>
     * Note that requests are negotiated on an operator boundary and {@code limit}'s amount
     * may not be preserved further upstream. For example,
     * {@code source.observeOn(Schedulers.computation()).limit(5)} will still request the
     * default (128) elements from the given {@code source}.
     * <p>
     * The main use of this operator is with sources that are async boundaries that
     * don't interfere with request amounts, such as certain {@code Flowable}-based
     * network endpoints that relay downstream request amounts unchanged and are, therefore,
     * prone to trigger excessive item creation/transmission over the network.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator requests a total of the given {@code count} items from the upstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code limit} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.6 - experimental
     * @param count the maximum number of items and the total request amount, non-negative.
     *              Zero will immediately cancel the upstream on subscription and complete
     *              the downstream.
     * @return the new Flowable instance
     * @see #take(long)
     * @see #rebatchRequests(int)
     * @since 2.2
     */
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    public final Flowable<T> limit(long count) {
        if (count < 0) {
            throw new IllegalArgumentException("count >= 0 required but it was " + count);
        }
        return RxJavaPlugins.onAssembly(new FlowableLimit<T>(this, count));
    }

    
Returns a Flowable that applies a specified function to each item emitted by the source Publisher and
emits the results of these function applications.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
map does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function to apply to each item emitted by the Publisher
Type parameters:
<R> – the output type
See Also:
ReactiveX operators documentation: Map
Returns:
a Flowable that emits the items from the source Publisher, transformed by the specified
        function/**
     * Returns a Flowable that applies a specified function to each item emitted by the source Publisher and
     * emits the results of these function applications.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/map.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code map} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the output type
     * @param mapper
     *            a function to apply to each item emitted by the Publisher
     * @return a Flowable that emits the items from the source Publisher, transformed by the specified
     *         function
     * @see <a href="http://reactivex.io/documentation/operators/map.html">ReactiveX operators documentation: Map</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> map(Function<? super T, ? extends R> mapper) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        return RxJavaPlugins.onAssembly(new FlowableMap<T, R>(this, mapper));
    }

    
Returns a Flowable that represents all of the emissions and notifications from the source Publisher into emissions marked with their original types within Notification objects. 



 
Backpressure:
 
The operator honors backpressure from downstream and expects it from the source Publisher. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
materialize does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Materialize
dematerialize(Function)
Returns:
a Flowable that emits items that are the result of materializing the items and notifications
        of the source Publisher/**
     * Returns a Flowable that represents all of the emissions <em>and</em> notifications from the source
     * Publisher into emissions marked with their original types within {@link Notification} objects.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/materialize.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and expects it from the source {@code Publisher}.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code materialize} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits items that are the result of materializing the items and notifications
     *         of the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/materialize-dematerialize.html">ReactiveX operators documentation: Materialize</a>
     * @see #dematerialize(Function)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Notification<T>> materialize() {
        return RxJavaPlugins.onAssembly(new FlowableMaterialize<T>(this));
    }

    
Flattens this and another Publisher into a single Publisher, without any transformation.


 You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by using the mergeWith method. 

 
Backpressure:
 
The operator honors backpressure from downstream. This and the other Publishers are expected to honor backpressure; if violated, the operator may signal MissingBackpressureException.
 
Scheduler:
 
mergeWith does not operate by default on a particular Scheduler.

Params:
other – 
           a Publisher to be merged
See Also:
ReactiveX operators documentation: Merge
Returns:
a Flowable that emits all of the items emitted by the source Publishers/**
     * Flattens this and another Publisher into a single Publisher, without any transformation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * You can combine items emitted by multiple Publishers so that they appear as a single Publisher, by
     * using the {@code mergeWith} method.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. This and the other {@code Publisher}s are expected to honor
     *  backpressure; if violated, the operator <em>may</em> signal {@code MissingBackpressureException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param other
     *            a Publisher to be merged
     * @return a Flowable that emits all of the items emitted by the source Publishers
     * @see <a href="http://reactivex.io/documentation/operators/merge.html">ReactiveX operators documentation: Merge</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> mergeWith(Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return merge(this, other);
    }

    
Merges the sequence of items of this Flowable with the success value of the other SingleSource.


 The success value of the other SingleSource can get interleaved at any point of this Flowable sequence. 

 
Backpressure:
 
The operator honors backpressure from downstream and ensures the success item from the SingleSource is emitted only when there is a downstream demand.
 
Scheduler:
 
mergeWith does not operate by default on a particular Scheduler.

History: 2.1.10 - experimental
Params:
other – the SingleSource whose success value to merge with
Returns:
the new Flowable instance
Since:
2.2/**
     * Merges the sequence of items of this Flowable with the success value of the other SingleSource.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * The success value of the other {@code SingleSource} can get interleaved at any point of this
     * {@code Flowable} sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and ensures the success item from the
     *  {@code SingleSource} is emitted only when there is a downstream demand.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.10 - experimental
     * @param other the {@code SingleSource} whose success value to merge with
     * @return the new Flowable instance
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> mergeWith(@NonNull SingleSource<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableMergeWithSingle<T>(this, other));
    }

    
Merges the sequence of items of this Flowable with the success value of the other MaybeSource
or waits for both to complete normally if the MaybeSource is empty.


 The success value of the other MaybeSource can get interleaved at any point of this Flowable sequence. 

 
Backpressure:
 
The operator honors backpressure from downstream and ensures the success item from the MaybeSource is emitted only when there is a downstream demand.
 
Scheduler:
 
mergeWith does not operate by default on a particular Scheduler.

History: 2.1.10 - experimental
Params:
other – the MaybeSource which provides a success value to merge with or completes
Returns:
the new Flowable instance
Since:
2.2/**
     * Merges the sequence of items of this Flowable with the success value of the other MaybeSource
     * or waits for both to complete normally if the MaybeSource is empty.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <p>
     * The success value of the other {@code MaybeSource} can get interleaved at any point of this
     * {@code Flowable} sequence.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and ensures the success item from the
     *  {@code MaybeSource} is emitted only when there is a downstream demand.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.10 - experimental
     * @param other the {@code MaybeSource} which provides a success value to merge with or completes
     * @return the new Flowable instance
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> mergeWith(@NonNull MaybeSource<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableMergeWithMaybe<T>(this, other));
    }

    
Relays the items of this Flowable and completes only when the other CompletableSource completes
as well.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
mergeWith does not operate by default on a particular Scheduler.

History: 2.1.10 - experimental
Params:
other – the CompletableSource to await for completion
Returns:
the new Flowable instance
Since:
2.2/**
     * Relays the items of this Flowable and completes only when the other CompletableSource completes
     * as well.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/merge.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code mergeWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.10 - experimental
     * @param other the {@code CompletableSource} to await for completion
     * @return the new Flowable instance
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> mergeWith(@NonNull CompletableSource other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableMergeWithCompletable<T>(this, other));
    }

    
Modifies a Publisher to perform its emissions and notifications on a specified Scheduler, asynchronously with a bounded buffer of bufferSize() slots. 
Note that onError notifications will cut ahead of onNext notifications on the emission thread if Scheduler is truly asynchronous. If strict event ordering is required, consider using the observeOn(Scheduler, boolean) overload. 



 
Backpressure:
 
This operator honors backpressure from downstream and expects it from the source Publisher. Violating this expectation will lead to MissingBackpressureException. This is the most common operator where the exception pops up; look for sources up the chain that don't support backpressure, such as interval, timer, {code PublishSubject} or BehaviorSubject and apply any of the onBackpressureXXX operators before applying observeOn itself.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
scheduler –  the Scheduler to notify Subscribers on
See Also:
ReactiveX operators documentation: ObserveOn
RxJava Threading Examples
subscribeOn
observeOn(Scheduler, boolean)
observeOn(Scheduler, boolean, int)
Returns:
the source Publisher modified so that its Subscribers are notified on the specified Scheduler/**
     * Modifies a Publisher to perform its emissions and notifications on a specified {@link Scheduler},
     * asynchronously with a bounded buffer of {@link #bufferSize()} slots.
     *
     * <p>Note that onError notifications will cut ahead of onNext notifications on the emission thread if Scheduler is truly
     * asynchronous. If strict event ordering is required, consider using the {@link #observeOn(Scheduler, boolean)} overload.
     * <p>
     * <img width="640" height="308" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/observeOn.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator honors backpressure from downstream and expects it from the source {@code Publisher}. Violating this
     *  expectation will lead to {@code MissingBackpressureException}. This is the most common operator where the exception
     *  pops up; look for sources up the chain that don't support backpressure,
     *  such as {@code interval}, {@code timer}, {code PublishSubject} or {@code BehaviorSubject} and apply any
     *  of the {@code onBackpressureXXX} operators <strong>before</strong> applying {@code observeOn} itself.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param scheduler
     *            the {@link Scheduler} to notify {@link Subscriber}s on
     * @return the source Publisher modified so that its {@link Subscriber}s are notified on the specified
     *         {@link Scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/observeon.html">ReactiveX operators documentation: ObserveOn</a>
     * @see <a href="http://www.grahamlea.com/2014/07/rxjava-threading-examples/">RxJava Threading Examples</a>
     * @see #subscribeOn
     * @see #observeOn(Scheduler, boolean)
     * @see #observeOn(Scheduler, boolean, int)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> observeOn(Scheduler scheduler) {
        return observeOn(scheduler, false, bufferSize());
    }

    
Modifies a Publisher to perform its emissions and notifications on a specified Scheduler, asynchronously with a bounded buffer and optionally delays onError notifications. 



 
Backpressure:
 
This operator honors backpressure from downstream and expects it from the source Publisher. Violating this expectation will lead to MissingBackpressureException. This is the most common operator where the exception pops up; look for sources up the chain that don't support backpressure, such as interval, timer, {code PublishSubject} or BehaviorSubject and apply any of the onBackpressureXXX operators before applying observeOn itself.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
scheduler –  the Scheduler to notify Subscribers on
delayError – 
           indicates if the onError notification may not cut ahead of onNext notification on the other side of the
           scheduling boundary. If true a sequence ending in onError will be replayed in the same order as was received
           from upstream
See Also:
ReactiveX operators documentation: ObserveOn
RxJava Threading Examples
subscribeOn
observeOn(Scheduler)
observeOn(Scheduler, boolean, int)
Returns:
the source Publisher modified so that its Subscribers are notified on the specified Scheduler/**
     * Modifies a Publisher to perform its emissions and notifications on a specified {@link Scheduler},
     * asynchronously with a bounded buffer and optionally delays onError notifications.
     * <p>
     * <img width="640" height="308" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/observeOn.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator honors backpressure from downstream and expects it from the source {@code Publisher}. Violating this
     *  expectation will lead to {@code MissingBackpressureException}. This is the most common operator where the exception
     *  pops up; look for sources up the chain that don't support backpressure,
     *  such as {@code interval}, {@code timer}, {code PublishSubject} or {@code BehaviorSubject} and apply any
     *  of the {@code onBackpressureXXX} operators <strong>before</strong> applying {@code observeOn} itself.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param scheduler
     *            the {@link Scheduler} to notify {@link Subscriber}s on
     * @param delayError
     *            indicates if the onError notification may not cut ahead of onNext notification on the other side of the
     *            scheduling boundary. If true a sequence ending in onError will be replayed in the same order as was received
     *            from upstream
     * @return the source Publisher modified so that its {@link Subscriber}s are notified on the specified
     *         {@link Scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/observeon.html">ReactiveX operators documentation: ObserveOn</a>
     * @see <a href="http://www.grahamlea.com/2014/07/rxjava-threading-examples/">RxJava Threading Examples</a>
     * @see #subscribeOn
     * @see #observeOn(Scheduler)
     * @see #observeOn(Scheduler, boolean, int)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> observeOn(Scheduler scheduler, boolean delayError) {
        return observeOn(scheduler, delayError, bufferSize());
    }

    
Modifies a Publisher to perform its emissions and notifications on a specified Scheduler, asynchronously with a bounded buffer of configurable size and optionally delays onError notifications. 



 
Backpressure:
 
This operator honors backpressure from downstream and expects it from the source Publisher. Violating this expectation will lead to MissingBackpressureException. This is the most common operator where the exception pops up; look for sources up the chain that don't support backpressure, such as interval, timer, {code PublishSubject} or BehaviorSubject and apply any of the onBackpressureXXX operators before applying observeOn itself.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
scheduler –  the Scheduler to notify Subscribers on
delayError – 
           indicates if the onError notification may not cut ahead of onNext notification on the other side of the
           scheduling boundary. If true a sequence ending in onError will be replayed in the same order as was received
           from upstream
bufferSize – the size of the buffer.
See Also:
ReactiveX operators documentation: ObserveOn
RxJava Threading Examples
subscribeOn
observeOn(Scheduler)
observeOn(Scheduler, boolean)
Returns:
the source Publisher modified so that its Subscribers are notified on the specified Scheduler/**
     * Modifies a Publisher to perform its emissions and notifications on a specified {@link Scheduler},
     * asynchronously with a bounded buffer of configurable size and optionally delays onError notifications.
     * <p>
     * <img width="640" height="308" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/observeOn.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator honors backpressure from downstream and expects it from the source {@code Publisher}. Violating this
     *  expectation will lead to {@code MissingBackpressureException}. This is the most common operator where the exception
     *  pops up; look for sources up the chain that don't support backpressure,
     *  such as {@code interval}, {@code timer}, {code PublishSubject} or {@code BehaviorSubject} and apply any
     *  of the {@code onBackpressureXXX} operators <strong>before</strong> applying {@code observeOn} itself.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param scheduler
     *            the {@link Scheduler} to notify {@link Subscriber}s on
     * @param delayError
     *            indicates if the onError notification may not cut ahead of onNext notification on the other side of the
     *            scheduling boundary. If true a sequence ending in onError will be replayed in the same order as was received
     *            from upstream
     * @param bufferSize the size of the buffer.
     * @return the source Publisher modified so that its {@link Subscriber}s are notified on the specified
     *         {@link Scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/observeon.html">ReactiveX operators documentation: ObserveOn</a>
     * @see <a href="http://www.grahamlea.com/2014/07/rxjava-threading-examples/">RxJava Threading Examples</a>
     * @see #subscribeOn
     * @see #observeOn(Scheduler)
     * @see #observeOn(Scheduler, boolean)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> observeOn(Scheduler scheduler, boolean delayError, int bufferSize) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableObserveOn<T>(this, scheduler, delayError, bufferSize));
    }

    
Filters the items emitted by a Publisher, only emitting those of the specified type.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
ofType does not operate by default on a particular Scheduler.

Params:
clazz – 
           the class type to filter the items emitted by the source Publisher
Type parameters:
<U> – the output type
See Also:
ReactiveX operators documentation: Filter
Returns:
a Flowable that emits items from the source Publisher of type clazz/**
     * Filters the items emitted by a Publisher, only emitting those of the specified type.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/ofClass.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code ofType} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the output type
     * @param clazz
     *            the class type to filter the items emitted by the source Publisher
     * @return a Flowable that emits items from the source Publisher of type {@code clazz}
     * @see <a href="http://reactivex.io/documentation/operators/filter.html">ReactiveX operators documentation: Filter</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<U> ofType(final Class<U> clazz) {
        ObjectHelper.requireNonNull(clazz, "clazz is null");
        return filter(Functions.isInstanceOf(clazz)).cast(clazz);
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer these
items indefinitely until they can be emitted.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to buffer items to the extent system resources allow/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer these
     * items indefinitely until they can be emitted.
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the source Publisher modified to buffer items to the extent system resources allow
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer() {
        return onBackpressureBuffer(bufferSize(), false, true);
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer these
items indefinitely until they can be emitted.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

Params:
delayError – 
               if true, an exception from the current Flowable is delayed until all buffered elements have been
               consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
               any buffered element
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to buffer items to the extent system resources allow/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer these
     * items indefinitely until they can be emitted.
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param delayError
     *                if true, an exception from the current Flowable is delayed until all buffered elements have been
     *                consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
     *                any buffered element
     * @return the source Publisher modified to buffer items to the extent system resources allow
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer(boolean delayError) {
        return onBackpressureBuffer(bufferSize(), delayError, true);
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to a given amount of items until they can be emitted. The resulting Publisher will signal a BufferOverflowException via onError as soon as the buffer's capacity is exceeded, dropping all undelivered items, and canceling the source. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

Params:
capacity – number of slots available in the buffer.
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to buffer items up to the given capacity.
Since:
1.1.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to
     * a given amount of items until they can be emitted. The resulting Publisher will signal
     * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered
     * items, and canceling the source.
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacity number of slots available in the buffer.
     * @return the source {@code Publisher} modified to buffer items up to the given capacity.
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     * @since 1.1.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer(int capacity) {
        return onBackpressureBuffer(capacity, false, false);
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to a given amount of items until they can be emitted. The resulting Publisher will signal a BufferOverflowException via onError as soon as the buffer's capacity is exceeded, dropping all undelivered items, and canceling the source. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

Params:
capacity – number of slots available in the buffer.
delayError – 
               if true, an exception from the current Flowable is delayed until all buffered elements have been
               consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
               any buffered element
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to buffer items up to the given capacity.
Since:
1.1.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to
     * a given amount of items until they can be emitted. The resulting Publisher will signal
     * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered
     * items, and canceling the source.
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacity number of slots available in the buffer.
     * @param delayError
     *                if true, an exception from the current Flowable is delayed until all buffered elements have been
     *                consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
     *                any buffered element
     * @return the source {@code Publisher} modified to buffer items up to the given capacity.
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     * @since 1.1.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer(int capacity, boolean delayError) {
        return onBackpressureBuffer(capacity, delayError, false);
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to a given amount of items until they can be emitted. The resulting Publisher will signal a BufferOverflowException via onError as soon as the buffer's capacity is exceeded, dropping all undelivered items, and canceling the source. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

Params:
capacity – number of slots available in the buffer.
delayError – 
               if true, an exception from the current Flowable is delayed until all buffered elements have been
               consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
               any buffered element
unbounded – 
               if true, the capacity value is interpreted as the internal "island" size of the unbounded buffer
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to buffer items up to the given capacity.
Since:
1.1.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to
     * a given amount of items until they can be emitted. The resulting Publisher will signal
     * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered
     * items, and canceling the source.
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacity number of slots available in the buffer.
     * @param delayError
     *                if true, an exception from the current Flowable is delayed until all buffered elements have been
     *                consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
     *                any buffered element
     * @param unbounded
     *                if true, the capacity value is interpreted as the internal "island" size of the unbounded buffer
     * @return the source {@code Publisher} modified to buffer items up to the given capacity.
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     * @since 1.1.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer(int capacity, boolean delayError, boolean unbounded) {
        ObjectHelper.verifyPositive(capacity, "capacity");
        return RxJavaPlugins.onAssembly(new FlowableOnBackpressureBuffer<T>(this, capacity, unbounded, delayError, Functions.EMPTY_ACTION));
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to a given amount of items until they can be emitted. The resulting Publisher will signal a BufferOverflowException via onError as soon as the buffer's capacity is exceeded, dropping all undelivered items, canceling the source, and notifying the producer with onOverflow. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

Params:
capacity – number of slots available in the buffer.
delayError – 
               if true, an exception from the current Flowable is delayed until all buffered elements have been
               consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
               any buffered element
unbounded – 
               if true, the capacity value is interpreted as the internal "island" size of the unbounded buffer
onOverflow – action to execute if an item needs to be buffered, but there are no available slots.  Null is allowed.
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to buffer items up to the given capacity
Since:
1.1.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to
     * a given amount of items until they can be emitted. The resulting Publisher will signal
     * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered
     * items, canceling the source, and notifying the producer with {@code onOverflow}.
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacity number of slots available in the buffer.
     * @param delayError
     *                if true, an exception from the current Flowable is delayed until all buffered elements have been
     *                consumed by the downstream; if false, an exception is immediately signaled to the downstream, skipping
     *                any buffered element
     * @param unbounded
     *                if true, the capacity value is interpreted as the internal "island" size of the unbounded buffer
     * @param onOverflow action to execute if an item needs to be buffered, but there are no available slots.  Null is allowed.
     * @return the source {@code Publisher} modified to buffer items up to the given capacity
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     * @since 1.1.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer(int capacity, boolean delayError, boolean unbounded,
            Action onOverflow) {
        ObjectHelper.requireNonNull(onOverflow, "onOverflow is null");
        ObjectHelper.verifyPositive(capacity, "capacity");
        return RxJavaPlugins.onAssembly(new FlowableOnBackpressureBuffer<T>(this, capacity, unbounded, delayError, onOverflow));
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to a given amount of items until they can be emitted. The resulting Publisher will signal a BufferOverflowException via onError as soon as the buffer's capacity is exceeded, dropping all undelivered items, canceling the source, and notifying the producer with onOverflow. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

Params:
capacity – number of slots available in the buffer.
onOverflow – action to execute if an item needs to be buffered, but there are no available slots.  Null is allowed.
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to buffer items up to the given capacity
Since:
1.1.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to
     * a given amount of items until they can be emitted. The resulting Publisher will signal
     * a {@code BufferOverflowException} via {@code onError} as soon as the buffer's capacity is exceeded, dropping all undelivered
     * items, canceling the source, and notifying the producer with {@code onOverflow}.
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacity number of slots available in the buffer.
     * @param onOverflow action to execute if an item needs to be buffered, but there are no available slots.  Null is allowed.
     * @return the source {@code Publisher} modified to buffer items up to the given capacity
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     * @since 1.1.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer(int capacity, Action onOverflow) {
        return onBackpressureBuffer(capacity, false, false, onOverflow);
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to a given amount of items until they can be emitted. The resulting Publisher will behave as determined by overflowStrategy if the buffer capacity is exceeded. 

    
BackpressureOverflow.Strategy.ON_OVERFLOW_ERROR (default) will call onError dropping all undelivered items, canceling the source, and notifying the producer with onOverflow. 
    
BackpressureOverflow.Strategy.ON_OVERFLOW_DROP_LATEST will drop any new items emitted by the producer while the buffer is full, without generating any onError. Each drop will, however, invoke onOverflow to signal the overflow to the producer.
    
BackpressureOverflow.Strategy.ON_OVERFLOW_DROP_OLDEST will drop the oldest items in the buffer in order to make room for newly emitted ones. Overflow will not generate anonError, but each drop will invoke onOverflow to signal the overflow to the producer.




 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureBuffer does not operate by default on a particular Scheduler.

Params:
capacity – number of slots available in the buffer.
onOverflow – action to execute if an item needs to be buffered, but there are no available slots.  Null is allowed.
overflowStrategy – how should the Publisher react to buffer overflows. Null is not allowed.
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Flowable modified to buffer items up to the given capacity
Since:
2.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to buffer up to
     * a given amount of items until they can be emitted. The resulting Publisher will behave as determined
     * by {@code overflowStrategy} if the buffer capacity is exceeded.
     *
     * <ul>
     *     <li>{@code BackpressureOverflow.Strategy.ON_OVERFLOW_ERROR} (default) will call {@code onError} dropping all undelivered items,
     *     canceling the source, and notifying the producer with {@code onOverflow}. </li>
     *     <li>{@code BackpressureOverflow.Strategy.ON_OVERFLOW_DROP_LATEST} will drop any new items emitted by the producer while
     *     the buffer is full, without generating any {@code onError}.  Each drop will, however, invoke {@code onOverflow}
     *     to signal the overflow to the producer.</li>
     *     <li>{@code BackpressureOverflow.Strategy.ON_OVERFLOW_DROP_OLDEST} will drop the oldest items in the buffer in order to make
     *     room for newly emitted ones. Overflow will not generate an{@code onError}, but each drop will invoke
     *     {@code onOverflow} to signal the overflow to the producer.</li>
     * </ul>
     *
     * <p>
     * <img width="640" height="300" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.buffer.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureBuffer} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacity number of slots available in the buffer.
     * @param onOverflow action to execute if an item needs to be buffered, but there are no available slots.  Null is allowed.
     * @param overflowStrategy how should the {@code Publisher} react to buffer overflows.  Null is not allowed.
     * @return the source {@code Flowable} modified to buffer items up to the given capacity
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureBuffer(long capacity, Action onOverflow, BackpressureOverflowStrategy overflowStrategy) {
        ObjectHelper.requireNonNull(overflowStrategy, "overflowStrategy is null");
        ObjectHelper.verifyPositive(capacity, "capacity");
        return RxJavaPlugins.onAssembly(new FlowableOnBackpressureBufferStrategy<T>(this, capacity, onOverflow, overflowStrategy));
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to discard,
rather than emit, those items that its Subscriber is not prepared to observe.


 If the downstream request count hits 0 then the Publisher will refrain from calling onNext until the Subscriber invokes request(n) again to increase the request count. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureDrop does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to drop onNext notifications on overflow/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to discard,
     * rather than emit, those items that its Subscriber is not prepared to observe.
     * <p>
     * <img width="640" height="245" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.drop.png" alt="">
     * <p>
     * If the downstream request count hits 0 then the Publisher will refrain from calling {@code onNext} until
     * the Subscriber invokes {@code request(n)} again to increase the request count.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureDrop} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the source Publisher modified to drop {@code onNext} notifications on overflow
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureDrop() {
        return RxJavaPlugins.onAssembly(new FlowableOnBackpressureDrop<T>(this));
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to discard,
rather than emit, those items that its Subscriber is not prepared to observe.


 If the downstream request count hits 0 then the Publisher will refrain from calling onNext until the Subscriber invokes request(n) again to increase the request count. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureDrop does not operate by default on a particular Scheduler.

Params:
onDrop – the action to invoke for each item dropped. onDrop action should be fast and should never block.
See Also:
ReactiveX operators documentation: backpressure operators
Returns:
the source Publisher modified to drop onNext notifications on overflow
Since:
1.1.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to discard,
     * rather than emit, those items that its Subscriber is not prepared to observe.
     * <p>
     * <img width="640" height="245" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.drop.png" alt="">
     * <p>
     * If the downstream request count hits 0 then the Publisher will refrain from calling {@code onNext} until
     * the Subscriber invokes {@code request(n)} again to increase the request count.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureDrop} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onDrop the action to invoke for each item dropped. onDrop action should be fast and should never block.
     * @return the source Publisher modified to drop {@code onNext} notifications on overflow
     * @see <a href="http://reactivex.io/documentation/operators/backpressure.html">ReactiveX operators documentation: backpressure operators</a>
     * @since 1.1.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureDrop(Consumer<? super T> onDrop) {
        ObjectHelper.requireNonNull(onDrop, "onDrop is null");
        return RxJavaPlugins.onAssembly(new FlowableOnBackpressureDrop<T>(this, onDrop));
    }

    
Instructs a Publisher that is emitting items faster than its Subscriber can consume them to
hold onto the latest value and emit that on request.


 Its behavior is logically equivalent to blockingLatest() with the exception that the downstream is not blocking while requesting more values. 

Note that if the upstream Publisher does support backpressure, this operator ignores that capability
and doesn't propagate any backpressure requests from downstream.

Note that due to the nature of how backpressure requests are propagated through subscribeOn/observeOn,
requesting more than 1 from downstream doesn't guarantee a continuous delivery of onNext events.

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., not applying backpressure to it).
 
Scheduler:
 
onBackpressureLatest does not operate by default on a particular Scheduler.

Returns:
the source Publisher modified so that it emits the most recently-received item upon request
Since:
1.1.0/**
     * Instructs a Publisher that is emitting items faster than its Subscriber can consume them to
     * hold onto the latest value and emit that on request.
     * <p>
     * <img width="640" height="245" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/bp.obp.latest.png" alt="">
     * <p>
     * Its behavior is logically equivalent to {@code blockingLatest()} with the exception that
     * the downstream is not blocking while requesting more values.
     * <p>
     * Note that if the upstream Publisher does support backpressure, this operator ignores that capability
     * and doesn't propagate any backpressure requests from downstream.
     * <p>
     * Note that due to the nature of how backpressure requests are propagated through subscribeOn/observeOn,
     * requesting more than 1 from downstream doesn't guarantee a continuous delivery of onNext events.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an unbounded
     *  manner (i.e., not applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onBackpressureLatest} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the source Publisher modified so that it emits the most recently-received item upon request
     * @since 1.1.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onBackpressureLatest() {
        return RxJavaPlugins.onAssembly(new FlowableOnBackpressureLatest<T>(this));
    }

    
Instructs a Publisher to pass control to another Publisher rather than invoking onError if it encounters an error. 


 By default, when a Publisher encounters an error that prevents it from emitting the expected item to its Subscriber, the Publisher invokes its Subscriber's onError method, and then quits without invoking any more of its Subscriber's methods. The onErrorResumeNext method changes this behavior. If you pass a function that returns a Publisher (resumeFunction) to onErrorResumeNext, if the original Publisher encounters an error, instead of invoking its Subscriber's onError method, it will instead relinquish control to the Publisher returned from resumeFunction, which will invoke the Subscriber's onNext method if it is able to do so. In such a case, because no Publisher necessarily invokes onError, the Subscriber may never know that an error happened. 

You can use this to prevent errors from propagating or to supply fallback data should errors be
encountered.

 
Backpressure:
 
The operator honors backpressure from downstream. This and the resuming Publishers are expected to honor backpressure as well. If any of them violate this expectation, the operator may throw an IllegalStateException when the source Publisher completes or a MissingBackpressureException is signaled somewhere downstream.
 
Scheduler:
 
onErrorResumeNext does not operate by default on a particular Scheduler.

Params:
resumeFunction – 
           a function that returns a Publisher that will take over if the source Publisher encounters
           an error
See Also:
ReactiveX operators documentation: Catch
Returns:
the original Publisher, with appropriately modified behavior/**
     * Instructs a Publisher to pass control to another Publisher rather than invoking
     * {@link Subscriber#onError onError} if it encounters an error.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/onErrorResumeNext.png" alt="">
     * <p>
     * By default, when a Publisher encounters an error that prevents it from emitting the expected item to
     * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits
     * without invoking any more of its Subscriber's methods. The {@code onErrorResumeNext} method changes this
     * behavior. If you pass a function that returns a Publisher ({@code resumeFunction}) to
     * {@code onErrorResumeNext}, if the original Publisher encounters an error, instead of invoking its
     * Subscriber's {@code onError} method, it will instead relinquish control to the Publisher returned from
     * {@code resumeFunction}, which will invoke the Subscriber's {@link Subscriber#onNext onNext} method if it is
     * able to do so. In such a case, because no Publisher necessarily invokes {@code onError}, the Subscriber
     * may never know that an error happened.
     * <p>
     * You can use this to prevent errors from propagating or to supply fallback data should errors be
     * encountered.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. This and the resuming {@code Publisher}s
     *  are expected to honor backpressure as well.
     *  If any of them violate this expectation, the operator <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes or
     *  a {@code MissingBackpressureException} is signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onErrorResumeNext} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param resumeFunction
     *            a function that returns a Publisher that will take over if the source Publisher encounters
     *            an error
     * @return the original Publisher, with appropriately modified behavior
     * @see <a href="http://reactivex.io/documentation/operators/catch.html">ReactiveX operators documentation: Catch</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onErrorResumeNext(Function<? super Throwable, ? extends Publisher<? extends T>> resumeFunction) {
        ObjectHelper.requireNonNull(resumeFunction, "resumeFunction is null");
        return RxJavaPlugins.onAssembly(new FlowableOnErrorNext<T>(this, resumeFunction, false));
    }

    
Instructs a Publisher to pass control to another Publisher rather than invoking onError if it encounters an error. 


 By default, when a Publisher encounters an error that prevents it from emitting the expected item to its Subscriber, the Publisher invokes its Subscriber's onError method, and then quits without invoking any more of its Subscriber's methods. The onErrorResumeNext method changes this behavior. If you pass another Publisher (resumeSequence) to a Publisher's onErrorResumeNext method, if the original Publisher encounters an error, instead of invoking its Subscriber's onError method, it will instead relinquish control to resumeSequence which will invoke the Subscriber's onNext method if it is able to do so. In such a case, because no Publisher necessarily invokes onError, the Subscriber may never know that an error happened. 

You can use this to prevent errors from propagating or to supply fallback data should errors be
encountered.

 
Backpressure:
 
The operator honors backpressure from downstream. This and the resuming Publishers are expected to honor backpressure as well. If any of them violate this expectation, the operator may throw an IllegalStateException when the source Publisher completes or MissingBackpressureException is signaled somewhere downstream.
 
Scheduler:
 
onErrorResumeNext does not operate by default on a particular Scheduler.

Params:
next – 
           the next Publisher source that will take over if the source Publisher encounters
           an error
See Also:
ReactiveX operators documentation: Catch
Returns:
the original Publisher, with appropriately modified behavior/**
     * Instructs a Publisher to pass control to another Publisher rather than invoking
     * {@link Subscriber#onError onError} if it encounters an error.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/onErrorResumeNext.png" alt="">
     * <p>
     * By default, when a Publisher encounters an error that prevents it from emitting the expected item to
     * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits
     * without invoking any more of its Subscriber's methods. The {@code onErrorResumeNext} method changes this
     * behavior. If you pass another Publisher ({@code resumeSequence}) to a Publisher's
     * {@code onErrorResumeNext} method, if the original Publisher encounters an error, instead of invoking its
     * Subscriber's {@code onError} method, it will instead relinquish control to {@code resumeSequence} which
     * will invoke the Subscriber's {@link Subscriber#onNext onNext} method if it is able to do so. In such a case,
     * because no Publisher necessarily invokes {@code onError}, the Subscriber may never know that an error
     * happened.
     * <p>
     * You can use this to prevent errors from propagating or to supply fallback data should errors be
     * encountered.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. This and the resuming {@code Publisher}s
     *  are expected to honor backpressure as well.
     *  If any of them violate this expectation, the operator <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes or
     *  {@code MissingBackpressureException} is signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onErrorResumeNext} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param next
     *            the next Publisher source that will take over if the source Publisher encounters
     *            an error
     * @return the original Publisher, with appropriately modified behavior
     * @see <a href="http://reactivex.io/documentation/operators/catch.html">ReactiveX operators documentation: Catch</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onErrorResumeNext(final Publisher<? extends T> next) {
        ObjectHelper.requireNonNull(next, "next is null");
        return onErrorResumeNext(Functions.justFunction(next));
    }

    
Instructs a Publisher to emit an item (returned by a specified function) rather than invoking onError if it encounters an error. 


 By default, when a Publisher encounters an error that prevents it from emitting the expected item to its Subscriber, the Publisher invokes its Subscriber's onError method, and then quits without invoking any more of its Subscriber's methods. The onErrorReturn method changes this behavior. If you pass a function (resumeFunction) to a Publisher's onErrorReturn method, if the original Publisher encounters an error, instead of invoking its Subscriber's onError method, it will instead emit the return value of resumeFunction. 

You can use this to prevent errors from propagating or to supply fallback data should errors be
encountered.

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers is expected to honor backpressure as well. If it this expectation is violated, the operator may throw IllegalStateException when the source Publisher completes or MissingBackpressureException is signaled somewhere downstream.
 
Scheduler:
 
onErrorReturn does not operate by default on a particular Scheduler.

Params:
valueSupplier – 
           a function that returns a single value that will be emitted along with a regular onComplete in case
           the current Flowable signals an onError event
See Also:
ReactiveX operators documentation: Catch
Returns:
the original Publisher with appropriately modified behavior/**
     * Instructs a Publisher to emit an item (returned by a specified function) rather than invoking
     * {@link Subscriber#onError onError} if it encounters an error.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/onErrorReturn.png" alt="">
     * <p>
     * By default, when a Publisher encounters an error that prevents it from emitting the expected item to
     * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits
     * without invoking any more of its Subscriber's methods. The {@code onErrorReturn} method changes this
     * behavior. If you pass a function ({@code resumeFunction}) to a Publisher's {@code onErrorReturn}
     * method, if the original Publisher encounters an error, instead of invoking its Subscriber's
     * {@code onError} method, it will instead emit the return value of {@code resumeFunction}.
     * <p>
     * You can use this to prevent errors from propagating or to supply fallback data should errors be
     * encountered.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s is expected to honor
     *  backpressure as well. If it this expectation is violated, the operator <em>may</em> throw
     *  {@code IllegalStateException} when the source {@code Publisher} completes or
     *  {@code MissingBackpressureException} is signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onErrorReturn} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param valueSupplier
     *            a function that returns a single value that will be emitted along with a regular onComplete in case
     *            the current Flowable signals an onError event
     * @return the original Publisher with appropriately modified behavior
     * @see <a href="http://reactivex.io/documentation/operators/catch.html">ReactiveX operators documentation: Catch</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onErrorReturn(Function<? super Throwable, ? extends T> valueSupplier) {
        ObjectHelper.requireNonNull(valueSupplier, "valueSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableOnErrorReturn<T>(this, valueSupplier));
    }

    
Instructs a Publisher to emit an item (returned by a specified function) rather than invoking onError if it encounters an error. 


 By default, when a Publisher encounters an error that prevents it from emitting the expected item to its Subscriber, the Publisher invokes its Subscriber's onError method, and then quits without invoking any more of its Subscriber's methods. The onErrorReturn method changes this behavior. If you pass a function (resumeFunction) to a Publisher's onErrorReturn method, if the original Publisher encounters an error, instead of invoking its Subscriber's onError method, it will instead emit the return value of resumeFunction. 

You can use this to prevent errors from propagating or to supply fallback data should errors be
encountered.

 
Backpressure:
 
The operator honors backpressure from downstream. The source Publishers is expected to honor backpressure as well. If it this expectation is violated, the operator may throw IllegalStateException when the source Publisher completes or MissingBackpressureException is signaled somewhere downstream.
 
Scheduler:
 
onErrorReturnItem does not operate by default on a particular Scheduler.

Params:
item – 
           the value that is emitted along with a regular onComplete in case the current
           Flowable signals an exception
See Also:
ReactiveX operators documentation: Catch
Returns:
the original Publisher with appropriately modified behavior/**
     * Instructs a Publisher to emit an item (returned by a specified function) rather than invoking
     * {@link Subscriber#onError onError} if it encounters an error.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/onErrorReturn.png" alt="">
     * <p>
     * By default, when a Publisher encounters an error that prevents it from emitting the expected item to
     * its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits
     * without invoking any more of its Subscriber's methods. The {@code onErrorReturn} method changes this
     * behavior. If you pass a function ({@code resumeFunction}) to a Publisher's {@code onErrorReturn}
     * method, if the original Publisher encounters an error, instead of invoking its Subscriber's
     * {@code onError} method, it will instead emit the return value of {@code resumeFunction}.
     * <p>
     * You can use this to prevent errors from propagating or to supply fallback data should errors be
     * encountered.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}s is expected to honor
     *  backpressure as well. If it this expectation is violated, the operator <em>may</em> throw
     *  {@code IllegalStateException} when the source {@code Publisher} completes or
     *  {@code MissingBackpressureException} is signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onErrorReturnItem} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param item
     *            the value that is emitted along with a regular onComplete in case the current
     *            Flowable signals an exception
     * @return the original Publisher with appropriately modified behavior
     * @see <a href="http://reactivex.io/documentation/operators/catch.html">ReactiveX operators documentation: Catch</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onErrorReturnItem(final T item) {
        ObjectHelper.requireNonNull(item, "item is null");
        return onErrorReturn(Functions.justFunction(item));
    }

    
Instructs a Publisher to pass control to another Publisher rather than invoking onError if it encounters an Exception. 
 This differs from onErrorResumeNext in that this one does not handle Throwable or Error but lets those continue through. 


 By default, when a Publisher encounters an exception that prevents it from emitting the expected item to its Subscriber, the Publisher invokes its Subscriber's onError method, and then quits without invoking any more of its Subscriber's methods. The onExceptionResumeNext method changes this behavior. If you pass another Publisher (resumeSequence) to a Publisher's onExceptionResumeNext method, if the original Publisher encounters an exception, instead of invoking its Subscriber's onError method, it will instead relinquish control to resumeSequence which will invoke the Subscriber's onNext method if it is able to do so. In such a case, because no Publisher necessarily invokes onError, the Subscriber may never know that an exception happened. 

You can use this to prevent exceptions from propagating or to supply fallback data should exceptions be
encountered.

 
Backpressure:
 
The operator honors backpressure from downstream. This and the resuming Publishers are expected to honor backpressure as well. If any of them violate this expectation, the operator may throw an IllegalStateException when the source Publisher completes or MissingBackpressureException is signaled somewhere downstream.
 
Scheduler:
 
onExceptionResumeNext does not operate by default on a particular Scheduler.

Params:
next – 
           the next Publisher that will take over if the source Publisher encounters
           an exception
See Also:
ReactiveX operators documentation: Catch
Returns:
the original Publisher, with appropriately modified behavior/**
     * Instructs a Publisher to pass control to another Publisher rather than invoking
     * {@link Subscriber#onError onError} if it encounters an {@link java.lang.Exception}.
     * <p>
     * This differs from {@link #onErrorResumeNext} in that this one does not handle {@link java.lang.Throwable}
     * or {@link java.lang.Error} but lets those continue through.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/onExceptionResumeNextViaPublisher.png" alt="">
     * <p>
     * By default, when a Publisher encounters an exception that prevents it from emitting the expected item
     * to its {@link Subscriber}, the Publisher invokes its Subscriber's {@code onError} method, and then quits
     * without invoking any more of its Subscriber's methods. The {@code onExceptionResumeNext} method changes
     * this behavior. If you pass another Publisher ({@code resumeSequence}) to a Publisher's
     * {@code onExceptionResumeNext} method, if the original Publisher encounters an exception, instead of
     * invoking its Subscriber's {@code onError} method, it will instead relinquish control to
     * {@code resumeSequence} which will invoke the Subscriber's {@link Subscriber#onNext onNext} method if it is
     * able to do so. In such a case, because no Publisher necessarily invokes {@code onError}, the Subscriber
     * may never know that an exception happened.
     * <p>
     * You can use this to prevent exceptions from propagating or to supply fallback data should exceptions be
     * encountered.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. This and the resuming {@code Publisher}s
     *  are expected to honor backpressure as well.
     *  If any of them violate this expectation, the operator <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes or
     *  {@code MissingBackpressureException} is signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onExceptionResumeNext} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param next
     *            the next Publisher that will take over if the source Publisher encounters
     *            an exception
     * @return the original Publisher, with appropriately modified behavior
     * @see <a href="http://reactivex.io/documentation/operators/catch.html">ReactiveX operators documentation: Catch</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onExceptionResumeNext(final Publisher<? extends T> next) {
        ObjectHelper.requireNonNull(next, "next is null");
        return RxJavaPlugins.onAssembly(new FlowableOnErrorNext<T>(this, Functions.justFunction(next), true));
    }

    
Nulls out references to the upstream producer and downstream Subscriber if
the sequence is terminated or downstream cancels.

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
onTerminateDetach does not operate by default on a particular Scheduler.

Returns:
a Flowable which nulls out references to the upstream producer and downstream Subscriber if
the sequence is terminated or downstream cancels
Since:
2.0/**
     * Nulls out references to the upstream producer and downstream Subscriber if
     * the sequence is terminated or downstream cancels.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code onTerminateDetach} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @return a Flowable which nulls out references to the upstream producer and downstream Subscriber if
     * the sequence is terminated or downstream cancels
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> onTerminateDetach() {
        return RxJavaPlugins.onAssembly(new FlowableDetach<T>(this));
    }

    
Parallelizes the flow by creating multiple 'rails' (equal to the number of CPUs)
and dispatches the upstream items to them in a round-robin fashion.
 Note that the rails don't execute in parallel on their own and one needs to apply ParallelFlowable.runOn(Scheduler) to specify the Scheduler where each rail will execute. 
 To merge the parallel 'rails' back into a single sequence, use ParallelFlowable.sequential(). 



 
Backpressure:
 
The operator requires the upstream to honor backpressure and each 'rail' honors backpressure
 as well.
 
Scheduler:
 
parallel does not operate by default on a particular Scheduler.

History: 2.0.5 - experimental; 2.1 - beta
Returns:
the new ParallelFlowable instance
Since:
2.2/**
     * Parallelizes the flow by creating multiple 'rails' (equal to the number of CPUs)
     * and dispatches the upstream items to them in a round-robin fashion.
     * <p>
     * Note that the rails don't execute in parallel on their own and one needs to
     * apply {@link ParallelFlowable#runOn(Scheduler)} to specify the Scheduler where
     * each rail will execute.
     * <p>
     * To merge the parallel 'rails' back into a single sequence, use {@link ParallelFlowable#sequential()}.
     * <p>
     * <img width="640" height="547" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flowable.parallel.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator requires the upstream to honor backpressure and each 'rail' honors backpressure
     *  as well.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code parallel} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.0.5 - experimental; 2.1 - beta
     * @return the new ParallelFlowable instance
     * @since 2.2
     */
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    public final ParallelFlowable<T> parallel() {
        return ParallelFlowable.from(this);
    }

    
Parallelizes the flow by creating the specified number of 'rails'
and dispatches the upstream items to them in a round-robin fashion.
 Note that the rails don't execute in parallel on their own and one needs to apply ParallelFlowable.runOn(Scheduler) to specify the Scheduler where each rail will execute. 
 To merge the parallel 'rails' back into a single sequence, use ParallelFlowable.sequential(). 



 
Backpressure:
 
The operator requires the upstream to honor backpressure and each 'rail' honors backpressure
 as well.
 
Scheduler:
 
parallel does not operate by default on a particular Scheduler.

History: 2.0.5 - experimental; 2.1 - beta
Params:
parallelism – the number of 'rails' to use
Returns:
the new ParallelFlowable instance
Since:
2.2/**
     * Parallelizes the flow by creating the specified number of 'rails'
     * and dispatches the upstream items to them in a round-robin fashion.
     * <p>
     * Note that the rails don't execute in parallel on their own and one needs to
     * apply {@link ParallelFlowable#runOn(Scheduler)} to specify the Scheduler where
     * each rail will execute.
     * <p>
     * To merge the parallel 'rails' back into a single sequence, use {@link ParallelFlowable#sequential()}.
     * <p>
     * <img width="640" height="547" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flowable.parallel.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator requires the upstream to honor backpressure and each 'rail' honors backpressure
     *  as well.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code parallel} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.0.5 - experimental; 2.1 - beta
     * @param parallelism the number of 'rails' to use
     * @return the new ParallelFlowable instance
     * @since 2.2
     */
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    public final ParallelFlowable<T> parallel(int parallelism) {
        ObjectHelper.verifyPositive(parallelism, "parallelism");
        return ParallelFlowable.from(this, parallelism);
    }

    
Parallelizes the flow by creating the specified number of 'rails'
and dispatches the upstream items to them in a round-robin fashion and
uses the defined per-'rail' prefetch amount.
 Note that the rails don't execute in parallel on their own and one needs to apply ParallelFlowable.runOn(Scheduler) to specify the Scheduler where each rail will execute. 
 To merge the parallel 'rails' back into a single sequence, use ParallelFlowable.sequential(). 



 
Backpressure:
 
The operator requires the upstream to honor backpressure and each 'rail' honors backpressure
 as well.
 
Scheduler:
 
parallel does not operate by default on a particular Scheduler.

History: 2.0.5 - experimental; 2.1 - beta
Params:
parallelism – the number of 'rails' to use
prefetch – the number of items each 'rail' should prefetch
Returns:
the new ParallelFlowable instance
Since:
2.2/**
     * Parallelizes the flow by creating the specified number of 'rails'
     * and dispatches the upstream items to them in a round-robin fashion and
     * uses the defined per-'rail' prefetch amount.
     * <p>
     * Note that the rails don't execute in parallel on their own and one needs to
     * apply {@link ParallelFlowable#runOn(Scheduler)} to specify the Scheduler where
     * each rail will execute.
     * <p>
     * To merge the parallel 'rails' back into a single sequence, use {@link ParallelFlowable#sequential()}.
     * <p>
     * <img width="640" height="547" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/flowable.parallel.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator requires the upstream to honor backpressure and each 'rail' honors backpressure
     *  as well.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code parallel} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.0.5 - experimental; 2.1 - beta
     * @param parallelism the number of 'rails' to use
     * @param prefetch the number of items each 'rail' should prefetch
     * @return the new ParallelFlowable instance
     * @since 2.2
     */
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @CheckReturnValue
    public final ParallelFlowable<T> parallel(int parallelism, int prefetch) {
        ObjectHelper.verifyPositive(parallelism, "parallelism");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return ParallelFlowable.from(this, parallelism, prefetch);
    }

    
Returns a ConnectableFlowable, which is a variety of Publisher that waits until its connect method is called before it begins emitting items to those Subscribers that have subscribed to it. 



 
Backpressure:
 
The returned ConnectableFlowable honors backpressure for each of its Subscribers and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator will signal a MissingBackpressureException to its Subscribers and disconnect.
 
Scheduler:
 
publish does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Publish
Returns:
a ConnectableFlowable that upon connection causes the source Publisher to emit items to its Subscribers/**
     * Returns a {@link ConnectableFlowable}, which is a variety of Publisher that waits until its
     * {@link ConnectableFlowable#connect connect} method is called before it begins emitting items to those
     * {@link Subscriber}s that have subscribed to it.
     * <p>
     * <img width="640" height="510" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/publishConnect.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code ConnectableFlowable} honors backpressure for each of its {@code Subscriber}s
     *  and expects the source {@code Publisher} to honor backpressure as well. If this expectation is violated,
     *  the operator will signal a {@code MissingBackpressureException} to its {@code Subscriber}s and disconnect.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code publish} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a {@link ConnectableFlowable} that upon connection causes the source Publisher to emit items
     *         to its {@link Subscriber}s
     * @see <a href="http://reactivex.io/documentation/operators/publish.html">ReactiveX operators documentation: Publish</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final ConnectableFlowable<T> publish() {
        return publish(bufferSize());
    }

    
Returns a Flowable that emits the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the underlying sequence. 



 
Backpressure:
 
The operator expects the source Publisher to honor backpressure and if this expectation is violated, the operator will signal a MissingBackpressureException through the Publisher provided to the function. Since the Publisher returned by the selector may be independent of the provided Publisher to the function, the output's backpressure behavior is determined by this returned Publisher.
 
Scheduler:
 
publish does not operate by default on a particular Scheduler.

Params:
selector – 
           a function that can use the multicasted source sequence as many times as needed, without
           causing multiple subscriptions to the source sequence. Subscribers to the given source will
           receive all notifications of the source from the time of the subscription forward.
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Publish
Returns:
a Flowable that emits the results of invoking the selector on the items emitted by a ConnectableFlowable that shares a single subscription to the underlying sequence/**
     * Returns a Flowable that emits the results of invoking a specified selector on items emitted by a
     * {@link ConnectableFlowable} that shares a single subscription to the underlying sequence.
     * <p>
     * <img width="640" height="510" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/publishConnect.f.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the source {@code Publisher} to honor backpressure and if this expectation is
     *  violated, the operator will signal a {@code MissingBackpressureException} through the {@code Publisher}
     *  provided to the function. Since the {@code Publisher} returned by the {@code selector} may be
     *  independent of the provided {@code Publisher} to the function, the output's backpressure behavior
     *  is determined by this returned {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code publish} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a function that can use the multicasted source sequence as many times as needed, without
     *            causing multiple subscriptions to the source sequence. Subscribers to the given source will
     *            receive all notifications of the source from the time of the subscription forward.
     * @return a Flowable that emits the results of invoking the selector on the items emitted by a {@link ConnectableFlowable} that shares a single subscription to the underlying sequence
     * @see <a href="http://reactivex.io/documentation/operators/publish.html">ReactiveX operators documentation: Publish</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> publish(Function<? super Flowable<T>, ? extends Publisher<R>> selector) {
        return publish(selector, bufferSize());
    }

    
Returns a Flowable that emits the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the underlying sequence. 



 
Backpressure:
 
The operator expects the source Publisher to honor backpressure and if this expectation is violated, the operator will signal a MissingBackpressureException through the Publisher provided to the function. Since the Publisher returned by the selector may be independent of the provided Publisher to the function, the output's backpressure behavior is determined by this returned Publisher.
 
Scheduler:
 
publish does not operate by default on a particular Scheduler.

Params:
selector – 
           a function that can use the multicasted source sequence as many times as needed, without
           causing multiple subscriptions to the source sequence. Subscribers to the given source will
           receive all notifications of the source from the time of the subscription forward.
prefetch – 
           the number of elements to prefetch from the current Flowable
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Publish
Returns:
a Flowable that emits the results of invoking the selector on the items emitted by a ConnectableFlowable that shares a single subscription to the underlying sequence/**
     * Returns a Flowable that emits the results of invoking a specified selector on items emitted by a
     * {@link ConnectableFlowable} that shares a single subscription to the underlying sequence.
     * <p>
     * <img width="640" height="510" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/publishConnect.f.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects the source {@code Publisher} to honor backpressure and if this expectation is
     *  violated, the operator will signal a {@code MissingBackpressureException} through the {@code Publisher}
     *  provided to the function. Since the {@code Publisher} returned by the {@code selector} may be
     *  independent of the provided {@code Publisher} to the function, the output's backpressure behavior
     *  is determined by this returned {@code Publisher}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code publish} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a function that can use the multicasted source sequence as many times as needed, without
     *            causing multiple subscriptions to the source sequence. Subscribers to the given source will
     *            receive all notifications of the source from the time of the subscription forward.
     * @param prefetch
     *            the number of elements to prefetch from the current Flowable
     * @return a Flowable that emits the results of invoking the selector on the items emitted by a {@link ConnectableFlowable} that shares a single subscription to the underlying sequence
     * @see <a href="http://reactivex.io/documentation/operators/publish.html">ReactiveX operators documentation: Publish</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> publish(Function<? super Flowable<T>, ? extends Publisher<? extends R>> selector, int prefetch) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        ObjectHelper.verifyPositive(prefetch, "prefetch");
        return RxJavaPlugins.onAssembly(new FlowablePublishMulticast<T, R>(this, selector, prefetch, false));
    }

    
Returns a ConnectableFlowable, which is a variety of Publisher that waits until its connect method is called before it begins emitting items to those Subscribers that have subscribed to it. 



 
Backpressure:
 
The returned ConnectableFlowable honors backpressure for each of its Subscribers and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator will signal a MissingBackpressureException to its Subscribers and disconnect.
 
Scheduler:
 
publish does not operate by default on a particular Scheduler.

Params:
bufferSize – 
           the number of elements to prefetch from the current Flowable
See Also:
ReactiveX operators documentation: Publish
Returns:
a ConnectableFlowable that upon connection causes the source Publisher to emit items to its Subscribers/**
     * Returns a {@link ConnectableFlowable}, which is a variety of Publisher that waits until its
     * {@link ConnectableFlowable#connect connect} method is called before it begins emitting items to those
     * {@link Subscriber}s that have subscribed to it.
     * <p>
     * <img width="640" height="510" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/publishConnect.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned {@code ConnectableFlowable} honors backpressure for each of its {@code Subscriber}s
     *  and expects the source {@code Publisher} to honor backpressure as well. If this expectation is violated,
     *  the operator will signal a {@code MissingBackpressureException} to its {@code Subscriber}s and disconnect.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code publish} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param bufferSize
     *            the number of elements to prefetch from the current Flowable
     * @return a {@link ConnectableFlowable} that upon connection causes the source Publisher to emit items
     *         to its {@link Subscriber}s
     * @see <a href="http://reactivex.io/documentation/operators/publish.html">ReactiveX operators documentation: Publish</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final ConnectableFlowable<T> publish(int bufferSize) {
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return FlowablePublish.create(this, bufferSize);
    }

    
Requests n initially from the upstream and then 75% of n subsequently after 75% of n values have been emitted to the downstream. 
This operator allows preventing the downstream to trigger unbounded mode via request(Long.MAX_VALUE) or compensate for the per-item overhead of small and frequent requests. 

 
Backpressure:
 
The operator expects backpressure from upstream and honors backpressure from downstream.
 
Scheduler:
 
rebatchRequests does not operate by default on a particular Scheduler.

Params:
n – the initial request amount, further request will happen after 75% of this value
Returns:
the Publisher that rebatches request amounts from downstream
Since:
2.0/**
     * Requests {@code n} initially from the upstream and then 75% of {@code n} subsequently
     * after 75% of {@code n} values have been emitted to the downstream.
     *
     * <p>This operator allows preventing the downstream to trigger unbounded mode via {@code request(Long.MAX_VALUE)}
     * or compensate for the per-item overhead of small and frequent requests.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from upstream and honors backpressure from downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code rebatchRequests} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param n the initial request amount, further request will happen after 75% of this value
     * @return the Publisher that rebatches request amounts from downstream
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> rebatchRequests(int n) {
        return observeOn(ImmediateThinScheduler.INSTANCE, true, n);
    }

    
Returns a Maybe that applies a specified accumulator function to the first item emitted by a source
Publisher, then feeds the result of that function along with the second item emitted by the source
Publisher into the same function, and so on until all items have been emitted by the finite source Publisher,
and emits the final result from the final call to your function as its sole item.


 This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate," "compress," or "inject" in other programming contexts. Groovy, for instance, has an inject method that does a similar operation on lists. 
 Note that this operator requires the upstream to signal onComplete for the accumulator object to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure of its downstream consumer and consumes the
 upstream source in unbounded mode.
 
Scheduler:
 
reduce does not operate by default on a particular Scheduler.

Params:
reducer – 
           an accumulator function to be invoked on each item emitted by the source Publisher, whose
           result will be used in the next accumulator call
See Also:
ReactiveX operators documentation: Reduce
Wikipedia: Fold (higher-order function)
Returns:
a Maybe that emits a single item that is the result of accumulating the items emitted by
        the source Flowable/**
     * Returns a Maybe that applies a specified accumulator function to the first item emitted by a source
     * Publisher, then feeds the result of that function along with the second item emitted by the source
     * Publisher into the same function, and so on until all items have been emitted by the finite source Publisher,
     * and emits the final result from the final call to your function as its sole item.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/reduce.png" alt="">
     * <p>
     * This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate,"
     * "compress," or "inject" in other programming contexts. Groovy, for instance, has an {@code inject} method
     * that does a similar operation on lists.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure of its downstream consumer and consumes the
     *  upstream source in unbounded mode.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code reduce} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param reducer
     *            an accumulator function to be invoked on each item emitted by the source Publisher, whose
     *            result will be used in the next accumulator call
     * @return a Maybe that emits a single item that is the result of accumulating the items emitted by
     *         the source Flowable
     * @see <a href="http://reactivex.io/documentation/operators/reduce.html">ReactiveX operators documentation: Reduce</a>
     * @see <a href="http://en.wikipedia.org/wiki/Fold_(higher-order_function)">Wikipedia: Fold (higher-order function)</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Maybe<T> reduce(BiFunction<T, T, T> reducer) {
        ObjectHelper.requireNonNull(reducer, "reducer is null");
        return RxJavaPlugins.onAssembly(new FlowableReduceMaybe<T>(this, reducer));
    }

    
Returns a Single that applies a specified accumulator function to the first item emitted by a source
Publisher and a specified seed value, then feeds the result of that function along with the second item
emitted by a Publisher into the same function, and so on until all items have been emitted by the
finite source Publisher, emitting the final result from the final call to your function as its sole item.


 This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate," "compress," or "inject" in other programming contexts. Groovy, for instance, has an inject method that does a similar operation on lists. 
 Note that the seed is shared among all subscribers to the resulting Publisher and may cause problems if it is mutable. To make sure each subscriber gets its own value, defer the application of this operator via defer(Callable<? extends Publisher<? extends Object>>): 

Publisher<T> source = ...
Single.defer(() -> source.reduce(new ArrayList<>(), (list, item) -> list.add(item)));
// alternatively, by using compose to stay fluent
source.compose(o ->
    Flowable.defer(() -> o.reduce(new ArrayList<>(), (list, item) -> list.add(item)).toFlowable())
).firstOrError();
// or, by using reduceWith instead of reduce
source.reduceWith(() -> new ArrayList<>(), (list, item) -> list.add(item)));

 Note that this operator requires the upstream to signal onComplete for the accumulator object to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure of its downstream consumer and consumes the
 upstream source in unbounded mode.
 
Scheduler:
 
reduce does not operate by default on a particular Scheduler.

Params:
seed – 
           the initial (seed) accumulator value
reducer – 
           an accumulator function to be invoked on each item emitted by the source Publisher, the
           result of which will be used in the next accumulator call
Type parameters:
<R> – the accumulator and output value type
See Also:
ReactiveX operators documentation: Reduce
Wikipedia: Fold (higher-order function)
reduceWith(Callable, BiFunction)
Returns:
a Single that emits a single item that is the result of accumulating the output from the
        items emitted by the source Publisher/**
     * Returns a Single that applies a specified accumulator function to the first item emitted by a source
     * Publisher and a specified seed value, then feeds the result of that function along with the second item
     * emitted by a Publisher into the same function, and so on until all items have been emitted by the
     * finite source Publisher, emitting the final result from the final call to your function as its sole item.
     * <p>
     * <img width="640" height="325" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/reduceSeed.png" alt="">
     * <p>
     * This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate,"
     * "compress," or "inject" in other programming contexts. Groovy, for instance, has an {@code inject} method
     * that does a similar operation on lists.
     * <p>
     * Note that the {@code seed} is shared among all subscribers to the resulting Publisher
     * and may cause problems if it is mutable. To make sure each subscriber gets its own value, defer
     * the application of this operator via {@link #defer(Callable)}:
     * <pre><code>
     * Publisher&lt;T&gt; source = ...
     * Single.defer(() -&gt; source.reduce(new ArrayList&lt;&gt;(), (list, item) -&gt; list.add(item)));
     *
     * // alternatively, by using compose to stay fluent
     *
     * source.compose(o -&gt;
     *     Flowable.defer(() -&gt; o.reduce(new ArrayList&lt;&gt;(), (list, item) -&gt; list.add(item)).toFlowable())
     * ).firstOrError();
     *
     * // or, by using reduceWith instead of reduce
     *
     * source.reduceWith(() -&gt; new ArrayList&lt;&gt;(), (list, item) -&gt; list.add(item)));
     * </code></pre>
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure of its downstream consumer and consumes the
     *  upstream source in unbounded mode.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code reduce} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the accumulator and output value type
     * @param seed
     *            the initial (seed) accumulator value
     * @param reducer
     *            an accumulator function to be invoked on each item emitted by the source Publisher, the
     *            result of which will be used in the next accumulator call
     * @return a Single that emits a single item that is the result of accumulating the output from the
     *         items emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/reduce.html">ReactiveX operators documentation: Reduce</a>
     * @see <a href="http://en.wikipedia.org/wiki/Fold_(higher-order_function)">Wikipedia: Fold (higher-order function)</a>
     * @see #reduceWith(Callable, BiFunction)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Single<R> reduce(R seed, BiFunction<R, ? super T, R> reducer) {
        ObjectHelper.requireNonNull(seed, "seed is null");
        ObjectHelper.requireNonNull(reducer, "reducer is null");
        return RxJavaPlugins.onAssembly(new FlowableReduceSeedSingle<T, R>(this, seed, reducer));
    }

    
Returns a Single that applies a specified accumulator function to the first item emitted by a source
Publisher and a seed value derived from calling a specified seedSupplier, then feeds the result
of that function along with the second item emitted by a Publisher into the same function, and so on until
all items have been emitted by the finite source Publisher, emitting the final result from the final call to your
function as its sole item.


 This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate," "compress," or "inject" in other programming contexts. Groovy, for instance, has an inject method that does a similar operation on lists. 
 Note that this operator requires the upstream to signal onComplete for the accumulator object to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure of its downstream consumer and consumes the
 upstream source in unbounded mode.
 
Scheduler:
 
reduceWith does not operate by default on a particular Scheduler.

Params:
seedSupplier – 
           the Callable that provides the initial (seed) accumulator value for each individual Subscriber
reducer – 
           an accumulator function to be invoked on each item emitted by the source Publisher, the
           result of which will be used in the next accumulator call
Type parameters:
<R> – the accumulator and output value type
See Also:
ReactiveX operators documentation: Reduce
Wikipedia: Fold (higher-order function)
Returns:
a Single that emits a single item that is the result of accumulating the output from the
        items emitted by the source Publisher/**
     * Returns a Single that applies a specified accumulator function to the first item emitted by a source
     * Publisher and a seed value derived from calling a specified seedSupplier, then feeds the result
     * of that function along with the second item emitted by a Publisher into the same function, and so on until
     * all items have been emitted by the finite source Publisher, emitting the final result from the final call to your
     * function as its sole item.
     * <p>
     * <img width="640" height="325" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/reduceSeed.png" alt="">
     * <p>
     * This technique, which is called "reduce" here, is sometimes called "aggregate," "fold," "accumulate,"
     * "compress," or "inject" in other programming contexts. Groovy, for instance, has an {@code inject} method
     * that does a similar operation on lists.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulator object to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure of its downstream consumer and consumes the
     *  upstream source in unbounded mode.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code reduceWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the accumulator and output value type
     * @param seedSupplier
     *            the Callable that provides the initial (seed) accumulator value for each individual Subscriber
     * @param reducer
     *            an accumulator function to be invoked on each item emitted by the source Publisher, the
     *            result of which will be used in the next accumulator call
     * @return a Single that emits a single item that is the result of accumulating the output from the
     *         items emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/reduce.html">ReactiveX operators documentation: Reduce</a>
     * @see <a href="http://en.wikipedia.org/wiki/Fold_(higher-order_function)">Wikipedia: Fold (higher-order function)</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Single<R> reduceWith(Callable<R> seedSupplier, BiFunction<R, ? super T, R> reducer) {
        ObjectHelper.requireNonNull(seedSupplier, "seedSupplier is null");
        ObjectHelper.requireNonNull(reducer, "reducer is null");
        return RxJavaPlugins.onAssembly(new FlowableReduceWithSingle<T, R>(this, seedSupplier, reducer));
    }

    
Returns a Flowable that repeats the sequence of items emitted by the source Publisher indefinitely.



 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
repeat does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Repeat
Returns:
a Flowable that emits the items emitted by the source Publisher repeatedly and in sequence/**
     * Returns a Flowable that repeats the sequence of items emitted by the source Publisher indefinitely.
     * <p>
     * <img width="640" height="309" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/repeat.o.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code repeat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits the items emitted by the source Publisher repeatedly and in sequence
     * @see <a href="http://reactivex.io/documentation/operators/repeat.html">ReactiveX operators documentation: Repeat</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> repeat() {
        return repeat(Long.MAX_VALUE);
    }

    
Returns a Flowable that repeats the sequence of items emitted by the source Publisher at most count times. 



 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
repeat does not operate by default on a particular Scheduler.

Params:
times – 
           the number of times the source Publisher items are repeated, a count of 0 will yield an empty
           sequence
Throws:
IllegalArgumentException –  if count is less than zero
See Also:
ReactiveX operators documentation: Repeat
Returns:
a Flowable that repeats the sequence of items emitted by the source Publisher at most count times/**
     * Returns a Flowable that repeats the sequence of items emitted by the source Publisher at most
     * {@code count} times.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/repeat.on.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code repeat} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param times
     *            the number of times the source Publisher items are repeated, a count of 0 will yield an empty
     *            sequence
     * @return a Flowable that repeats the sequence of items emitted by the source Publisher at most
     *         {@code count} times
     * @throws IllegalArgumentException
     *             if {@code count} is less than zero
     * @see <a href="http://reactivex.io/documentation/operators/repeat.html">ReactiveX operators documentation: Repeat</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> repeat(long times) {
        if (times < 0) {
            throw new IllegalArgumentException("times >= 0 required but it was " + times);
        }
        if (times == 0) {
            return empty();
        }
        return RxJavaPlugins.onAssembly(new FlowableRepeat<T>(this, times));
    }

    
Returns a Flowable that repeats the sequence of items emitted by the source Publisher until
the provided stop function returns true.



 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
repeatUntil does not operate by default on a particular Scheduler.

Params:
stop – 
               a boolean supplier that is called when the current Flowable completes and unless it returns
               false, the current Flowable is resubscribed
Throws:
NullPointerException –  if stop is null
See Also:
ReactiveX operators documentation: Repeat
Returns:
the new Flowable instance/**
     * Returns a Flowable that repeats the sequence of items emitted by the source Publisher until
     * the provided stop function returns true.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/repeat.on.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code repeatUntil} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param stop
     *                a boolean supplier that is called when the current Flowable completes and unless it returns
     *                false, the current Flowable is resubscribed
     * @return the new Flowable instance
     * @throws NullPointerException
     *             if {@code stop} is null
     * @see <a href="http://reactivex.io/documentation/operators/repeat.html">ReactiveX operators documentation: Repeat</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> repeatUntil(BooleanSupplier stop) {
        ObjectHelper.requireNonNull(stop, "stop is null");
        return RxJavaPlugins.onAssembly(new FlowableRepeatUntil<T>(this, stop));
    }

    
Returns a Flowable that emits the same values as the source Publisher with the exception of an onComplete. An onComplete notification from the source will result in the emission of a void item to the Publisher provided as an argument to the notificationHandler function. If that Publisher calls onComplete or onError then repeatWhen will call onComplete or onError on the child subscription. Otherwise, this Publisher will resubscribe to the source Publisher. 



 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
repeatWhen does not operate by default on a particular Scheduler.

Params:
handler – 
           receives a Publisher of notifications with which a user can complete or error, aborting the repeat.
See Also:
ReactiveX operators documentation: Repeat
Returns:
the source Publisher modified with repeat logic/**
     * Returns a Flowable that emits the same values as the source Publisher with the exception of an
     * {@code onComplete}. An {@code onComplete} notification from the source will result in the emission of
     * a {@code void} item to the Publisher provided as an argument to the {@code notificationHandler}
     * function. If that Publisher calls {@code onComplete} or {@code onError} then {@code repeatWhen} will
     * call {@code onComplete} or {@code onError} on the child subscription. Otherwise, this Publisher will
     * resubscribe to the source Publisher.
     * <p>
     * <img width="640" height="430" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/repeatWhen.f.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code repeatWhen} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param handler
     *            receives a Publisher of notifications with which a user can complete or error, aborting the repeat.
     * @return the source Publisher modified with repeat logic
     * @see <a href="http://reactivex.io/documentation/operators/repeat.html">ReactiveX operators documentation: Repeat</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> repeatWhen(final Function<? super Flowable<Object>, ? extends Publisher<?>> handler) {
        ObjectHelper.requireNonNull(handler, "handler is null");
        return RxJavaPlugins.onAssembly(new FlowableRepeatWhen<T>(this, handler));
    }

    
Returns a ConnectableFlowable that shares a single subscription to the underlying Publisher that will replay all of its items and notifications to any future Subscriber. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that upon connection causes the source Publisher to emit its items to its Subscribers/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the underlying Publisher
     * that will replay all of its items and notifications to any future {@link Subscriber}. A Connectable
     * Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is
     * subscribed to, but only when its {@code connect} method is called.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a {@link ConnectableFlowable} that upon connection causes the source Publisher to emit its
     *         items to its {@link Subscriber}s
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final ConnectableFlowable<T> replay() {
        return FlowableReplay.createFrom(this);
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on the items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay does not operate by default on a particular Scheduler.

Params:
selector – 
           the selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on a ConnectableFlowable that shares a single subscription to the source Publisher/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on the items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher.
     * <p>
     * <img width="640" height="450" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.f.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            the selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @return a Flowable that emits items that are the results of invoking the selector on a
     *         {@link ConnectableFlowable} that shares a single subscription to the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> replay(Function<? super Flowable<T>, ? extends Publisher<R>> selector) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this), selector);
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying bufferSize notifications. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay does not operate by default on a particular Scheduler.

Params:
selector – 
           the selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
bufferSize – 
           the buffer size that limits the number of items the connectable Publisher can replay
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher replaying no more than bufferSize items/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     * replaying {@code bufferSize} notifications.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="440" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.fn.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            the selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @param bufferSize
     *            the buffer size that limits the number of items the connectable Publisher can replay
     * @return a Flowable that emits items that are the results of invoking the selector on items emitted by
     *         a {@link ConnectableFlowable} that shares a single subscription to the source Publisher
     *         replaying no more than {@code bufferSize} items
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> replay(Function<? super Flowable<T>, ? extends Publisher<R>> selector, final int bufferSize) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this, bufferSize), selector);
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying no more than bufferSize items that were emitted within a specified time window. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay operates by default on the computation Scheduler.

Params:
selector – 
           a selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
bufferSize – 
           the buffer size that limits the number of items the connectable Publisher can replay
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, and replays no more than bufferSize items that were emitted within the window defined by time/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     * replaying no more than {@code bufferSize} items that were emitted within a specified time window.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="445" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.fnt.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @param bufferSize
     *            the buffer size that limits the number of items the connectable Publisher can replay
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @return a Flowable that emits items that are the results of invoking the selector on items emitted by
     *         a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, and
     *         replays no more than {@code bufferSize} items that were emitted within the window defined by
     *         {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final <R> Flowable<R> replay(Function<? super Flowable<T>, ? extends Publisher<R>> selector, int bufferSize, long time, TimeUnit unit) {
        return replay(selector, bufferSize, time, unit, Schedulers.computation());
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying no more than bufferSize items that were emitted within a specified time window. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
selector – 
           a selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
bufferSize – 
           the buffer size that limits the number of items the connectable Publisher can replay
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
scheduler – 
           the Scheduler that is the time source for the window
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
Throws:
IllegalArgumentException –  if bufferSize is less than zero
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, and replays no more than bufferSize items that were emitted within the window defined by time/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     * replaying no more than {@code bufferSize} items that were emitted within a specified time window.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="445" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.fnts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @param bufferSize
     *            the buffer size that limits the number of items the connectable Publisher can replay
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the Scheduler that is the time source for the window
     * @return a Flowable that emits items that are the results of invoking the selector on items emitted by
     *         a {@link ConnectableFlowable} that shares a single subscription to the source Publisher, and
     *         replays no more than {@code bufferSize} items that were emitted within the window defined by
     *         {@code time}
     * @throws IllegalArgumentException
     *             if {@code bufferSize} is less than zero
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final <R> Flowable<R> replay(Function<? super Flowable<T>, ? extends Publisher<R>> selector, final int bufferSize, final long time, final TimeUnit unit, final Scheduler scheduler) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return FlowableReplay.multicastSelector(
                FlowableInternalHelper.replayCallable(this, bufferSize, time, unit, scheduler), selector);
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying a maximum of bufferSize items. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
selector – 
           a selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
bufferSize – 
           the buffer size that limits the number of items the connectable Publisher can replay
scheduler – 
           the Scheduler on which the replay is observed
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying no more than bufferSize notifications/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     * replaying a maximum of {@code bufferSize} items.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="440" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.fns.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @param bufferSize
     *            the buffer size that limits the number of items the connectable Publisher can replay
     * @param scheduler
     *            the Scheduler on which the replay is observed
     * @return a Flowable that emits items that are the results of invoking the selector on items emitted by
     *         a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     *         replaying no more than {@code bufferSize} notifications
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final <R> Flowable<R> replay(final Function<? super Flowable<T>, ? extends Publisher<R>> selector, final int bufferSize, final Scheduler scheduler) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this, bufferSize),
                FlowableInternalHelper.replayFunction(selector, scheduler)
        );
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying all items that were emitted within a specified time window. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay operates by default on the computation Scheduler.

Params:
selector – 
           a selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying all items that were emitted within the window defined by time/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     * replaying all items that were emitted within a specified time window.
     * <p>
     * <img width="640" height="435" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.ft.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @return a Flowable that emits items that are the results of invoking the selector on items emitted by
     *         a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     *         replaying all items that were emitted within the window defined by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final <R> Flowable<R> replay(Function<? super Flowable<T>, ? extends Publisher<R>> selector, long time, TimeUnit unit) {
        return replay(selector, time, unit, Schedulers.computation());
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying all items that were emitted within a specified time window. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
selector – 
           a selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
scheduler – 
           the scheduler that is the time source for the window
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying all items that were emitted within the window defined by time/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     * replaying all items that were emitted within a specified time window.
     * <p>
     * <img width="640" height="440" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.fts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the scheduler that is the time source for the window
     * @return a Flowable that emits items that are the results of invoking the selector on items emitted by
     *         a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     *         replaying all items that were emitted within the window defined by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final <R> Flowable<R> replay(Function<? super Flowable<T>, ? extends Publisher<R>> selector, final long time, final TimeUnit unit, final Scheduler scheduler) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this, time, unit, scheduler), selector);
    }

    
Returns a Flowable that emits items that are the results of invoking a specified selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
selector – 
           a selector function, which can use the multicasted sequence as many times as needed, without
           causing multiple subscriptions to the Publisher
scheduler – 
           the Scheduler where the replay is observed
Type parameters:
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Replay
Returns:
a Flowable that emits items that are the results of invoking the selector on items emitted by a ConnectableFlowable that shares a single subscription to the source Publisher, replaying all items/**
     * Returns a Flowable that emits items that are the results of invoking a specified selector on items
     * emitted by a {@link ConnectableFlowable} that shares a single subscription to the source Publisher.
     * <p>
     * <img width="640" height="445" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.fs.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param selector
     *            a selector function, which can use the multicasted sequence as many times as needed, without
     *            causing multiple subscriptions to the Publisher
     * @param scheduler
     *            the Scheduler where the replay is observed
     * @return a Flowable that emits items that are the results of invoking the selector on items emitted by
     *         a {@link ConnectableFlowable} that shares a single subscription to the source Publisher,
     *         replaying all items
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final <R> Flowable<R> replay(final Function<? super Flowable<T>, ? extends Publisher<R>> selector, final Scheduler scheduler) {
        ObjectHelper.requireNonNull(selector, "selector is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return FlowableReplay.multicastSelector(FlowableInternalHelper.replayCallable(this),
                FlowableInternalHelper.replayFunction(selector, scheduler));
    }

    
Returns a ConnectableFlowable that shares a single subscription to the source Publisher that replays at most bufferSize items emitted by that Publisher. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay does not operate by default on a particular Scheduler.

Params:
bufferSize – 
           the buffer size that limits the number of items that can be replayed
See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that shares a single subscription to the source Publisher and replays at most bufferSize items emitted by that Publisher/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher that
     * replays at most {@code bufferSize} items emitted by that Publisher. A Connectable Publisher resembles
     * an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only
     * when its {@code connect} method is called.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.n.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param bufferSize
     *            the buffer size that limits the number of items that can be replayed
     * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     *         replays at most {@code bufferSize} items emitted by that Publisher
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final ConnectableFlowable<T> replay(final int bufferSize) {
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return FlowableReplay.create(this, bufferSize);
    }

    
Returns a ConnectableFlowable that shares a single subscription to the source Publisher and replays at most bufferSize items that were emitted during a specified time window. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay operates by default on the computation Scheduler.

Params:
bufferSize – 
           the buffer size that limits the number of items that can be replayed
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that shares a single subscription to the source Publisher and replays at most bufferSize items that were emitted during the window defined by time/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     * replays at most {@code bufferSize} items that were emitted during a specified time window. A Connectable
     * Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is
     * subscribed to, but only when its {@code connect} method is called.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.nt.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param bufferSize
     *            the buffer size that limits the number of items that can be replayed
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     *         replays at most {@code bufferSize} items that were emitted during the window defined by
     *         {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final ConnectableFlowable<T> replay(int bufferSize, long time, TimeUnit unit) {
        return replay(bufferSize, time, unit, Schedulers.computation());
    }

    
Returns a ConnectableFlowable that shares a single subscription to the source Publisher and that replays a maximum of bufferSize items that are emitted within a specified time window. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
bufferSize – 
           the buffer size that limits the number of items that can be replayed
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
scheduler – 
           the scheduler that is used as a time source for the window
Throws:
IllegalArgumentException –  if bufferSize is less than zero
See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that shares a single subscription to the source Publisher and replays at most bufferSize items that were emitted during the window defined by time/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     * that replays a maximum of {@code bufferSize} items that are emitted within a specified time window. A
     * Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items
     * when it is subscribed to, but only when its {@code connect} method is called.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.nts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param bufferSize
     *            the buffer size that limits the number of items that can be replayed
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the scheduler that is used as a time source for the window
     * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     *         replays at most {@code bufferSize} items that were emitted during the window defined by
     *         {@code time}
     * @throws IllegalArgumentException
     *             if {@code bufferSize} is less than zero
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final ConnectableFlowable<T> replay(final int bufferSize, final long time, final TimeUnit unit, final Scheduler scheduler) {
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return FlowableReplay.create(this, time, unit, scheduler, bufferSize);
    }

    
Returns a ConnectableFlowable that shares a single subscription to the source Publisher and replays at most bufferSize items emitted by that Publisher. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 
 Note that due to concurrency requirements, replay(bufferSize) may hold strong references to more than bufferSize source emissions. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
bufferSize – 
           the buffer size that limits the number of items that can be replayed
scheduler – 
           the scheduler on which the Subscribers will observe the emitted items
See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that shares a single subscription to the source Publisher and replays at most bufferSize items that were emitted by the Publisher/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     * replays at most {@code bufferSize} items emitted by that Publisher. A Connectable Publisher resembles
     * an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only
     * when its {@code connect} method is called.
     * <p>
     * Note that due to concurrency requirements, {@code replay(bufferSize)} may hold strong references to more than
     * {@code bufferSize} source emissions.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.ns.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param bufferSize
     *            the buffer size that limits the number of items that can be replayed
     * @param scheduler
     *            the scheduler on which the Subscribers will observe the emitted items
     * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     *         replays at most {@code bufferSize} items that were emitted by the Publisher
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final ConnectableFlowable<T> replay(final int bufferSize, final Scheduler scheduler) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return FlowableReplay.observeOn(replay(bufferSize), scheduler);
    }

    
Returns a ConnectableFlowable that shares a single subscription to the source Publisher and replays all items emitted by that Publisher within a specified time window. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
This version of replay operates by default on the computation Scheduler.

Params:
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that shares a single subscription to the source Publisher and replays the items that were emitted during the window defined by time/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     * replays all items emitted by that Publisher within a specified time window. A Connectable Publisher
     * resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to,
     * but only when its {@code connect} method is called.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.t.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code replay} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     *         replays the items that were emitted during the window defined by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final ConnectableFlowable<T> replay(long time, TimeUnit unit) {
        return replay(time, unit, Schedulers.computation());
    }

    
Returns a ConnectableFlowable that shares a single subscription to the source Publisher and replays all items emitted by that Publisher within a specified time window. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
time – 
           the duration of the window in which the replayed items must have been emitted
unit –  the time unit of time
scheduler – 
           the Scheduler that is the time source for the window
See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that shares a single subscription to the source Publisher and replays the items that were emitted during the window defined by time/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     * replays all items emitted by that Publisher within a specified time window. A Connectable Publisher
     * resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to,
     * but only when its {@code connect} method is called.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.ts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param time
     *            the duration of the window in which the replayed items must have been emitted
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the Scheduler that is the time source for the window
     * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher and
     *         replays the items that were emitted during the window defined by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final ConnectableFlowable<T> replay(final long time, final TimeUnit unit, final Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return FlowableReplay.create(this, time, unit, scheduler);
    }

    
Returns a ConnectableFlowable that shares a single subscription to the source Publisher that will replay all of its items and notifications to any future Subscriber on the given Scheduler. A Connectable Publisher resembles an ordinary Publisher, except that it does not begin emitting items when it is subscribed to, but only when its connect method is called. 



 
Backpressure:
 
This operator supports backpressure. Note that the upstream requests are determined by the child
 Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
 request 100 elements from the underlying Publisher sequence.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
scheduler – 
           the Scheduler on which the Subscribers will observe the emitted items
See Also:
ReactiveX operators documentation: Replay
Returns:
a ConnectableFlowable that shares a single subscription to the source Publisher that will replay all of its items and notifications to any future Subscriber on the given Scheduler/**
     * Returns a {@link ConnectableFlowable} that shares a single subscription to the source Publisher that
     * will replay all of its items and notifications to any future {@link Subscriber} on the given
     * {@link Scheduler}. A Connectable Publisher resembles an ordinary Publisher, except that it does not
     * begin emitting items when it is subscribed to, but only when its {@code connect} method is called.
     * <p>
     * <img width="640" height="515" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/replay.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator supports backpressure. Note that the upstream requests are determined by the child
     *  Subscriber which requests the largest amount: i.e., two child Subscribers with requests of 10 and 100 will
     *  request 100 elements from the underlying Publisher sequence.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param scheduler
     *            the Scheduler on which the Subscribers will observe the emitted items
     * @return a {@link ConnectableFlowable} that shares a single subscription to the source Publisher that
     *         will replay all of its items and notifications to any future {@link Subscriber} on the given
     *         {@link Scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/replay.html">ReactiveX operators documentation: Replay</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final ConnectableFlowable<T> replay(final Scheduler scheduler) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return FlowableReplay.observeOn(replay(), scheduler);
    }

    
Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls onError (infinite retry count). 


 If the source Publisher calls Subscriber.onError, this method will resubscribe to the source Publisher rather than propagating the onError call. 
 Any and all items emitted by the source Publisher will be emitted by the resulting Publisher, even those emitted during failed subscriptions. For example, if a Publisher fails at first but emits [1, 2] then succeeds the second time and emits [1, 2, 3, 4, 5] then the complete sequence of emissions and notifications would be [1, 2, 1, 2, 3, 4, 5, onComplete]. 

 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
retry does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Retry
Returns:
the source Publisher modified with retry logic/**
     * Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls {@code onError}
     * (infinite retry count).
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/retry.png" alt="">
     * <p>
     * If the source Publisher calls {@link Subscriber#onError}, this method will resubscribe to the source
     * Publisher rather than propagating the {@code onError} call.
     * <p>
     * Any and all items emitted by the source Publisher will be emitted by the resulting Publisher, even
     * those emitted during failed subscriptions. For example, if a Publisher fails at first but emits
     * {@code [1, 2]} then succeeds the second time and emits {@code [1, 2, 3, 4, 5]} then the complete sequence
     * of emissions and notifications would be {@code [1, 2, 1, 2, 3, 4, 5, onComplete]}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code retry} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the source Publisher modified with retry logic
     * @see <a href="http://reactivex.io/documentation/operators/retry.html">ReactiveX operators documentation: Retry</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> retry() {
        return retry(Long.MAX_VALUE, Functions.alwaysTrue());
    }

    
Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls onError and the predicate returns true for that specific exception and retry count. 



 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
retry does not operate by default on a particular Scheduler.

Params:
predicate – 
           the predicate that determines if a resubscription may happen in case of a specific exception
           and retry count
See Also:
retry()
ReactiveX operators documentation: Retry
Returns:
the source Publisher modified with retry logic/**
     * Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls {@code onError}
     * and the predicate returns true for that specific exception and retry count.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/retry.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code retry} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param predicate
     *            the predicate that determines if a resubscription may happen in case of a specific exception
     *            and retry count
     * @return the source Publisher modified with retry logic
     * @see #retry()
     * @see <a href="http://reactivex.io/documentation/operators/retry.html">ReactiveX operators documentation: Retry</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> retry(BiPredicate<? super Integer, ? super Throwable> predicate) {
        ObjectHelper.requireNonNull(predicate, "predicate is null");

        return RxJavaPlugins.onAssembly(new FlowableRetryBiPredicate<T>(this, predicate));
    }

    
Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls onError up to a specified number of retries. 


 If the source Publisher calls Subscriber.onError, this method will resubscribe to the source Publisher for a maximum of count resubscriptions rather than propagating the onError call. 
 Any and all items emitted by the source Publisher will be emitted by the resulting Publisher, even those emitted during failed subscriptions. For example, if a Publisher fails at first but emits [1, 2] then succeeds the second time and emits [1, 2, 3, 4, 5] then the complete sequence of emissions and notifications would be [1, 2, 1, 2, 3, 4, 5, onComplete]. 

 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
retry does not operate by default on a particular Scheduler.

Params:
count – 
           the number of times to resubscribe if the current Flowable fails
See Also:
ReactiveX operators documentation: Retry
Returns:
the source Publisher modified with retry logic/**
     * Returns a Flowable that mirrors the source Publisher, resubscribing to it if it calls {@code onError}
     * up to a specified number of retries.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/retry.png" alt="">
     * <p>
     * If the source Publisher calls {@link Subscriber#onError}, this method will resubscribe to the source
     * Publisher for a maximum of {@code count} resubscriptions rather than propagating the
     * {@code onError} call.
     * <p>
     * Any and all items emitted by the source Publisher will be emitted by the resulting Publisher, even
     * those emitted during failed subscriptions. For example, if a Publisher fails at first but emits
     * {@code [1, 2]} then succeeds the second time and emits {@code [1, 2, 3, 4, 5]} then the complete sequence
     * of emissions and notifications would be {@code [1, 2, 1, 2, 3, 4, 5, onComplete]}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code retry} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the number of times to resubscribe if the current Flowable fails
     * @return the source Publisher modified with retry logic
     * @see <a href="http://reactivex.io/documentation/operators/retry.html">ReactiveX operators documentation: Retry</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> retry(long count) {
        return retry(count, Functions.alwaysTrue());
    }

    
Retries at most times or until the predicate returns false, whichever happens first.

 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
retry does not operate by default on a particular Scheduler.

Params:
times – the number of times to resubscribe if the current Flowable fails
predicate – the predicate called with the failure Throwable and should return true to trigger a retry.
Returns:
the new Flowable instance/**
     * Retries at most times or until the predicate returns false, whichever happens first.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code retry} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param times the number of times to resubscribe if the current Flowable fails
     * @param predicate the predicate called with the failure Throwable and should return true to trigger a retry.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> retry(long times, Predicate<? super Throwable> predicate) {
        if (times < 0) {
            throw new IllegalArgumentException("times >= 0 required but it was " + times);
        }
        ObjectHelper.requireNonNull(predicate, "predicate is null");

        return RxJavaPlugins.onAssembly(new FlowableRetryPredicate<T>(this, times, predicate));
    }

    
Retries the current Flowable if the predicate returns true.

 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
retry does not operate by default on a particular Scheduler.

Params:
predicate – the predicate that receives the failure Throwable and should return true to trigger a retry.
Returns:
the new Flowable instance/**
     * Retries the current Flowable if the predicate returns true.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code retry} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param predicate the predicate that receives the failure Throwable and should return true to trigger a retry.
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> retry(Predicate<? super Throwable> predicate) {
        return retry(Long.MAX_VALUE, predicate);
    }

    
Retries until the given stop function returns true.

 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
retryUntil does not operate by default on a particular Scheduler.

Params:
stop – the function that should return true to stop retrying
Returns:
the new Flowable instance/**
     * Retries until the given stop function returns true.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code retryUntil} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param stop the function that should return true to stop retrying
     * @return the new Flowable instance
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> retryUntil(final BooleanSupplier stop) {
        ObjectHelper.requireNonNull(stop, "stop is null");
        return retry(Long.MAX_VALUE, Functions.predicateReverseFor(stop));
    }

    
Returns a Flowable that emits the same values as the source Publisher with the exception of an onError. An onError notification from the source will result in the emission of a Throwable item to the Publisher provided as an argument to the notificationHandler function. If that Publisher calls onComplete or onError then retry will call onComplete or onError on the child subscription. Otherwise, this Publisher will resubscribe to the source Publisher. 



Example:
This retries 3 times, each time incrementing the number of seconds it waits.

 Flowable.create((FlowableEmitter<? super String> s) -> {
     System.out.println("subscribing");
     s.onError(new RuntimeException("always fails"));
 }, BackpressureStrategy.BUFFER).retryWhen(attempts -> {
     return attempts.zipWith(Flowable.range(1, 3), (n, i) -> i).flatMap(i -> {
         System.out.println("delay retry by " + i + " second(s)");
         return Flowable.timer(i, TimeUnit.SECONDS);
     });
 }).blockingForEach(System.out::println);

Output is:
 
subscribing
delay retry by 1 second(s)
subscribing
delay retry by 2 second(s)
subscribing
delay retry by 3 second(s)
subscribing
 
 Note that the inner Publisher returned by the handler function should signal either onNext, onError or onComplete in response to the received Throwable to indicate the operator should retry or terminate. If the upstream to the operator is asynchronous, signaling onNext followed by onComplete immediately may result in the sequence to be completed immediately. Similarly, if this inner Publisher signals onError or onComplete while the upstream is active, the sequence is terminated with the same signal immediately. 

The following example demonstrates how to retry an asynchronous source with a delay:

Flowable.timer(1, TimeUnit.SECONDS)
    .doOnSubscribe(s -> System.out.println("subscribing"))
    .map(v -> { throw new RuntimeException(); })
    .retryWhen(errors -> {
        AtomicInteger counter = new AtomicInteger();
        return errors
                  .takeWhile(e -> counter.getAndIncrement() != 3)
                  .flatMap(e -> {
                      System.out.println("delay retry by " + counter.get() + " second(s)");
                      return Flowable.timer(counter.get(), TimeUnit.SECONDS);
                  });
    })
    .blockingSubscribe(System.out::println, System.out::println);


 
Backpressure:
 
The operator honors downstream backpressure and expects both the source and inner Publishers to honor backpressure as well. If this expectation is violated, the operator may throw an IllegalStateException.
 
Scheduler:
 
retryWhen does not operate by default on a particular Scheduler.

Params:
handler – 
           receives a Publisher of notifications with which a user can complete or error, aborting the
           retry
See Also:
ReactiveX operators documentation: Retry
Returns:
the source Publisher modified with retry logic/**
     * Returns a Flowable that emits the same values as the source Publisher with the exception of an
     * {@code onError}. An {@code onError} notification from the source will result in the emission of a
     * {@link Throwable} item to the Publisher provided as an argument to the {@code notificationHandler}
     * function. If that Publisher calls {@code onComplete} or {@code onError} then {@code retry} will call
     * {@code onComplete} or {@code onError} on the child subscription. Otherwise, this Publisher will
     * resubscribe to the source Publisher.
     * <p>
     * <img width="640" height="430" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/retryWhen.f.png" alt="">
     * <p>
     * Example:
     *
     * This retries 3 times, each time incrementing the number of seconds it waits.
     *
     * <pre><code>
     *  Flowable.create((FlowableEmitter&lt;? super String&gt; s) -&gt; {
     *      System.out.println("subscribing");
     *      s.onError(new RuntimeException("always fails"));
     *  }, BackpressureStrategy.BUFFER).retryWhen(attempts -&gt; {
     *      return attempts.zipWith(Flowable.range(1, 3), (n, i) -&gt; i).flatMap(i -&gt; {
     *          System.out.println("delay retry by " + i + " second(s)");
     *          return Flowable.timer(i, TimeUnit.SECONDS);
     *      });
     *  }).blockingForEach(System.out::println);
     * </code></pre>
     *
     * Output is:
     *
     * <pre> {@code
     * subscribing
     * delay retry by 1 second(s)
     * subscribing
     * delay retry by 2 second(s)
     * subscribing
     * delay retry by 3 second(s)
     * subscribing
     * } </pre>
     * <p>
     * Note that the inner {@code Publisher} returned by the handler function should signal
     * either {@code onNext}, {@code onError} or {@code onComplete} in response to the received
     * {@code Throwable} to indicate the operator should retry or terminate. If the upstream to
     * the operator is asynchronous, signaling onNext followed by onComplete immediately may
     * result in the sequence to be completed immediately. Similarly, if this inner
     * {@code Publisher} signals {@code onError} or {@code onComplete} while the upstream is
     * active, the sequence is terminated with the same signal immediately.
     * <p>
     * The following example demonstrates how to retry an asynchronous source with a delay:
     * <pre><code>
     * Flowable.timer(1, TimeUnit.SECONDS)
     *     .doOnSubscribe(s -&gt; System.out.println("subscribing"))
     *     .map(v -&gt; { throw new RuntimeException(); })
     *     .retryWhen(errors -&gt; {
     *         AtomicInteger counter = new AtomicInteger();
     *         return errors
     *                   .takeWhile(e -&gt; counter.getAndIncrement() != 3)
     *                   .flatMap(e -&gt; {
     *                       System.out.println("delay retry by " + counter.get() + " second(s)");
     *                       return Flowable.timer(counter.get(), TimeUnit.SECONDS);
     *                   });
     *     })
     *     .blockingSubscribe(System.out::println, System.out::println);
     * </code></pre>
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects both the source
     *  and inner {@code Publisher}s to honor backpressure as well.
     *  If this expectation is violated, the operator <em>may</em> throw an {@code IllegalStateException}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code retryWhen} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param handler
     *            receives a Publisher of notifications with which a user can complete or error, aborting the
     *            retry
     * @return the source Publisher modified with retry logic
     * @see <a href="http://reactivex.io/documentation/operators/retry.html">ReactiveX operators documentation: Retry</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> retryWhen(
            final Function<? super Flowable<Throwable>, ? extends Publisher<?>> handler) {
        ObjectHelper.requireNonNull(handler, "handler is null");

        return RxJavaPlugins.onAssembly(new FlowableRetryWhen<T>(this, handler));
    }

    
Subscribes to the current Flowable and wraps the given Subscriber into a SafeSubscriber
(if not already a SafeSubscriber) that
deals with exceptions thrown by a misbehaving Subscriber (that doesn't follow the
Reactive Streams specification).

 
Backpressure:
 
This operator leaves the reactive world and the backpressure behavior depends on the Subscriber's behavior.
 
Scheduler:
 
safeSubscribe does not operate by default on a particular Scheduler.

Params:
s – the incoming Subscriber instance
Throws:
NullPointerException – if s is null/**
     * Subscribes to the current Flowable and wraps the given Subscriber into a SafeSubscriber
     * (if not already a SafeSubscriber) that
     * deals with exceptions thrown by a misbehaving Subscriber (that doesn't follow the
     * Reactive Streams specification).
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator leaves the reactive world and the backpressure behavior depends on the Subscriber's behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code safeSubscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param s the incoming Subscriber instance
     * @throws NullPointerException if s is null
     */
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void safeSubscribe(Subscriber<? super T> s) {
        ObjectHelper.requireNonNull(s, "s is null");
        if (s instanceof SafeSubscriber) {
            subscribe((SafeSubscriber<? super T>)s);
        } else {
            subscribe(new SafeSubscriber<T>(s));
        }
    }

    
Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
within periodic time intervals.



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
sample operates by default on the computation Scheduler.

Params:
period – 
           the sampling rate
unit –  the TimeUnit in which period is defined
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
throttleLast(long, TimeUnit)
Returns:
a Flowable that emits the results of sampling the items emitted by the source Publisher at
        the specified time interval/**
     * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
     * within periodic time intervals.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sample.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sample} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param period
     *            the sampling rate
     * @param unit
     *            the {@link TimeUnit} in which {@code period} is defined
     * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at
     *         the specified time interval
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #throttleLast(long, TimeUnit)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> sample(long period, TimeUnit unit) {
        return sample(period, unit, Schedulers.computation());
    }

    
Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
within periodic time intervals and optionally emit the very last upstream item when the upstream completes.



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
sample operates by default on the computation Scheduler.

History: 2.0.5 - experimental
Params:
period – 
           the sampling rate
unit –  the TimeUnit in which period is defined
emitLast – 
           if true and the upstream completes while there is still an unsampled item available,
           that item is emitted to downstream before completion
           if false, an unsampled last item is ignored.
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
throttleLast(long, TimeUnit)
Returns:
a Flowable that emits the results of sampling the items emitted by the source Publisher at
        the specified time interval
Since:
2.1/**
     * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
     * within periodic time intervals and optionally emit the very last upstream item when the upstream completes.
     * <p>
     * <img width="640" height="276" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sample.emitlast.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sample} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * <p>History: 2.0.5 - experimental
     * @param period
     *            the sampling rate
     * @param unit
     *            the {@link TimeUnit} in which {@code period} is defined
     * @param emitLast
     *            if true and the upstream completes while there is still an unsampled item available,
     *            that item is emitted to downstream before completion
     *            if false, an unsampled last item is ignored.
     * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at
     *         the specified time interval
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #throttleLast(long, TimeUnit)
     * @since 2.1
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> sample(long period, TimeUnit unit, boolean emitLast) {
        return sample(period, unit, Schedulers.computation(), emitLast);
    }

    
Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
within periodic time intervals, where the intervals are defined on a particular Scheduler.



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
period – 
           the sampling rate
unit –  the TimeUnit in which period is defined
scheduler –  the Scheduler to use when sampling
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
throttleLast(long, TimeUnit, Scheduler)
Returns:
a Flowable that emits the results of sampling the items emitted by the source Publisher at
        the specified time interval/**
     * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
     * within periodic time intervals, where the intervals are defined on a particular Scheduler.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sample.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param period
     *            the sampling rate
     * @param unit
     *            the {@link TimeUnit} in which {@code period} is defined
     * @param scheduler
     *            the {@link Scheduler} to use when sampling
     * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at
     *         the specified time interval
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #throttleLast(long, TimeUnit, Scheduler)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> sample(long period, TimeUnit unit, Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableSampleTimed<T>(this, period, unit, scheduler, false));
    }

    
Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
within periodic time intervals, where the intervals are defined on a particular Scheduler
and optionally emit the very last upstream item when the upstream completes.



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
You specify which Scheduler this operator will use.

History: 2.0.5 - experimental
Params:
period – 
           the sampling rate
unit –  the TimeUnit in which period is defined
scheduler –  the Scheduler to use when sampling
emitLast – 
           if true and the upstream completes while there is still an unsampled item available,
           that item is emitted to downstream before completion
           if false, an unsampled last item is ignored.
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
throttleLast(long, TimeUnit, Scheduler)
Returns:
a Flowable that emits the results of sampling the items emitted by the source Publisher at
        the specified time interval
Since:
2.1/**
     * Returns a Flowable that emits the most recently emitted item (if any) emitted by the source Publisher
     * within periodic time intervals, where the intervals are defined on a particular Scheduler
     * and optionally emit the very last upstream item when the upstream completes.
     * <p>
     * <img width="640" height="276" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sample.s.emitlast.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * <p>History: 2.0.5 - experimental
     * @param period
     *            the sampling rate
     * @param unit
     *            the {@link TimeUnit} in which {@code period} is defined
     * @param scheduler
     *            the {@link Scheduler} to use when sampling
     * @param emitLast
     *            if true and the upstream completes while there is still an unsampled item available,
     *            that item is emitted to downstream before completion
     *            if false, an unsampled last item is ignored.
     * @return a Flowable that emits the results of sampling the items emitted by the source Publisher at
     *         the specified time interval
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #throttleLast(long, TimeUnit, Scheduler)
     * @since 2.1
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> sample(long period, TimeUnit unit, Scheduler scheduler, boolean emitLast) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableSampleTimed<T>(this, period, unit, scheduler, emitLast));
    }

    
Returns a Flowable that, when the specified sampler Publisher emits an item or completes, emits the most recently emitted item (if any) emitted by the source Publisher since the previous emission from the sampler Publisher. 



 
Backpressure:
 
This operator does not support backpressure as it uses the emissions of the sampler Publisher to control data flow.
 
Scheduler:
 
This version of sample does not operate by default on a particular Scheduler.

Params:
sampler – 
           the Publisher to use for sampling the source Publisher
Type parameters:
<U> – the element type of the sampler Publisher
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
Returns:
a Flowable that emits the results of sampling the items emitted by this Publisher whenever the sampler Publisher emits an item or completes/**
     * Returns a Flowable that, when the specified {@code sampler} Publisher emits an item or completes,
     * emits the most recently emitted item (if any) emitted by the source Publisher since the previous
     * emission from the {@code sampler} Publisher.
     * <p>
     * <img width="640" height="289" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sample.o.nolast.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses the emissions of the {@code sampler}
     *      Publisher to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code sample} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the element type of the sampler Publisher
     * @param sampler
     *            the Publisher to use for sampling the source Publisher
     * @return a Flowable that emits the results of sampling the items emitted by this Publisher whenever
     *         the {@code sampler} Publisher emits an item or completes
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<T> sample(Publisher<U> sampler) {
        ObjectHelper.requireNonNull(sampler, "sampler is null");
        return RxJavaPlugins.onAssembly(new FlowableSamplePublisher<T>(this, sampler, false));
    }

    
Returns a Flowable that, when the specified sampler Publisher emits an item or completes, emits the most recently emitted item (if any) emitted by the source Publisher since the previous emission from the sampler Publisher and optionally emit the very last upstream item when the upstream or other Publisher complete. 



 
Backpressure:
 
This operator does not support backpressure as it uses the emissions of the sampler Publisher to control data flow.
 
Scheduler:
 
This version of sample does not operate by default on a particular Scheduler.

History: 2.0.5 - experimental
Params:
sampler – 
           the Publisher to use for sampling the source Publisher
emitLast – 
           if true and the upstream completes while there is still an unsampled item available,
           that item is emitted to downstream before completion
           if false, an unsampled last item is ignored.
Type parameters:
<U> – the element type of the sampler Publisher
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
Returns:
a Flowable that emits the results of sampling the items emitted by this Publisher whenever the sampler Publisher emits an item or completes
Since:
2.1/**
     * Returns a Flowable that, when the specified {@code sampler} Publisher emits an item or completes,
     * emits the most recently emitted item (if any) emitted by the source Publisher since the previous
     * emission from the {@code sampler} Publisher
     * and optionally emit the very last upstream item when the upstream or other Publisher complete.
     * <p>
     * <img width="640" height="289" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/sample.o.emitlast.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses the emissions of the {@code sampler}
     *      Publisher to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code sample} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * <p>History: 2.0.5 - experimental
     * @param <U> the element type of the sampler Publisher
     * @param sampler
     *            the Publisher to use for sampling the source Publisher
     * @param emitLast
     *            if true and the upstream completes while there is still an unsampled item available,
     *            that item is emitted to downstream before completion
     *            if false, an unsampled last item is ignored.
     * @return a Flowable that emits the results of sampling the items emitted by this Publisher whenever
     *         the {@code sampler} Publisher emits an item or completes
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @since 2.1
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<T> sample(Publisher<U> sampler, boolean emitLast) {
        ObjectHelper.requireNonNull(sampler, "sampler is null");
        return RxJavaPlugins.onAssembly(new FlowableSamplePublisher<T>(this, sampler, emitLast));
    }

    
Returns a Flowable that applies a specified accumulator function to the first item emitted by a source
Publisher, then feeds the result of that function along with the second item emitted by the source
Publisher into the same function, and so on until all items have been emitted by the source Publisher,
emitting the result of each of these iterations.



This sort of function is sometimes called an accumulator.

 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. Violating this expectation, a MissingBackpressureException may get signaled somewhere downstream.
 
Scheduler:
 
scan does not operate by default on a particular Scheduler.

Params:
accumulator –  an accumulator function to be invoked on each item emitted by the source Publisher, whose result will be emitted to Subscribers via onNext and used in the next accumulator call
See Also:
ReactiveX operators documentation: Scan
Returns:
a Flowable that emits the results of each call to the accumulator function/**
     * Returns a Flowable that applies a specified accumulator function to the first item emitted by a source
     * Publisher, then feeds the result of that function along with the second item emitted by the source
     * Publisher into the same function, and so on until all items have been emitted by the source Publisher,
     * emitting the result of each of these iterations.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/scan.png" alt="">
     * <p>
     * This sort of function is sometimes called an accumulator.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  Violating this expectation, a {@code MissingBackpressureException} <em>may</em> get signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code scan} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param accumulator
     *            an accumulator function to be invoked on each item emitted by the source Publisher, whose
     *            result will be emitted to {@link Subscriber}s via {@link Subscriber#onNext onNext} and used in the
     *            next accumulator call
     * @return a Flowable that emits the results of each call to the accumulator function
     * @see <a href="http://reactivex.io/documentation/operators/scan.html">ReactiveX operators documentation: Scan</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> scan(BiFunction<T, T, T> accumulator) {
        ObjectHelper.requireNonNull(accumulator, "accumulator is null");
        return RxJavaPlugins.onAssembly(new FlowableScan<T>(this, accumulator));
    }

    
Returns a Flowable that applies a specified accumulator function to the first item emitted by a source
Publisher and a seed value, then feeds the result of that function along with the second item emitted by
the source Publisher into the same function, and so on until all items have been emitted by the source
Publisher, emitting the result of each of these iterations.



This sort of function is sometimes called an accumulator.
 Note that the Publisher that results from this method will emit initialValue as its first emitted item. 
 Note that the initialValue is shared among all subscribers to the resulting Publisher and may cause problems if it is mutable. To make sure each subscriber gets its own value, defer the application of this operator via defer(Callable<? extends Publisher<? extends Object>>): 

Publisher<T> source = ...
Flowable.defer(() -> source.scan(new ArrayList<>(), (list, item) -> list.add(item)));
// alternatively, by using compose to stay fluent
source.compose(o ->
    Flowable.defer(() -> o.scan(new ArrayList<>(), (list, item) -> list.add(item)))
);


 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. Violating this expectation, a MissingBackpressureException may get signaled somewhere downstream.
 
Scheduler:
 
scan does not operate by default on a particular Scheduler.

Params:
initialValue – 
           the initial (seed) accumulator item
accumulator –  an accumulator function to be invoked on each item emitted by the source Publisher, whose result will be emitted to Subscribers via onNext and used in the next accumulator call
Type parameters:
<R> – the initial, accumulator and result type
See Also:
ReactiveX operators documentation: Scan
Returns:
a Flowable that emits initialValue followed by the results of each call to the accumulator function/**
     * Returns a Flowable that applies a specified accumulator function to the first item emitted by a source
     * Publisher and a seed value, then feeds the result of that function along with the second item emitted by
     * the source Publisher into the same function, and so on until all items have been emitted by the source
     * Publisher, emitting the result of each of these iterations.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/scanSeed.png" alt="">
     * <p>
     * This sort of function is sometimes called an accumulator.
     * <p>
     * Note that the Publisher that results from this method will emit {@code initialValue} as its first
     * emitted item.
     * <p>
     * Note that the {@code initialValue} is shared among all subscribers to the resulting Publisher
     * and may cause problems if it is mutable. To make sure each subscriber gets its own value, defer
     * the application of this operator via {@link #defer(Callable)}:
     * <pre><code>
     * Publisher&lt;T&gt; source = ...
     * Flowable.defer(() -&gt; source.scan(new ArrayList&lt;&gt;(), (list, item) -&gt; list.add(item)));
     *
     * // alternatively, by using compose to stay fluent
     *
     * source.compose(o -&gt;
     *     Flowable.defer(() -&gt; o.scan(new ArrayList&lt;&gt;(), (list, item) -&gt; list.add(item)))
     * );
     * </code></pre>
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  Violating this expectation, a {@code MissingBackpressureException} <em>may</em> get signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code scan} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the initial, accumulator and result type
     * @param initialValue
     *            the initial (seed) accumulator item
     * @param accumulator
     *            an accumulator function to be invoked on each item emitted by the source Publisher, whose
     *            result will be emitted to {@link Subscriber}s via {@link Subscriber#onNext onNext} and used in the
     *            next accumulator call
     * @return a Flowable that emits {@code initialValue} followed by the results of each call to the
     *         accumulator function
     * @see <a href="http://reactivex.io/documentation/operators/scan.html">ReactiveX operators documentation: Scan</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> scan(final R initialValue, BiFunction<R, ? super T, R> accumulator) {
        ObjectHelper.requireNonNull(initialValue, "initialValue is null");
        return scanWith(Functions.justCallable(initialValue), accumulator);
    }

    
Returns a Flowable that applies a specified accumulator function to the first item emitted by a source
Publisher and a seed value, then feeds the result of that function along with the second item emitted by
the source Publisher into the same function, and so on until all items have been emitted by the source
Publisher, emitting the result of each of these iterations.



This sort of function is sometimes called an accumulator.
 Note that the Publisher that results from this method will emit the value returned by the seedSupplier as its first item. 

 
Backpressure:
 
The operator honors downstream backpressure and expects the source Publisher to honor backpressure as well. Violating this expectation, a MissingBackpressureException may get signaled somewhere downstream.
 
Scheduler:
 
scanWith does not operate by default on a particular Scheduler.

Params:
seedSupplier – 
           a Callable that returns the initial (seed) accumulator item for each individual Subscriber
accumulator –  an accumulator function to be invoked on each item emitted by the source Publisher, whose result will be emitted to Subscribers via onNext and used in the next accumulator call
Type parameters:
<R> – the initial, accumulator and result type
See Also:
ReactiveX operators documentation: Scan
Returns:
a Flowable that emits initialValue followed by the results of each call to the accumulator function/**
     * Returns a Flowable that applies a specified accumulator function to the first item emitted by a source
     * Publisher and a seed value, then feeds the result of that function along with the second item emitted by
     * the source Publisher into the same function, and so on until all items have been emitted by the source
     * Publisher, emitting the result of each of these iterations.
     * <p>
     * <img width="640" height="320" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/scanSeed.png" alt="">
     * <p>
     * This sort of function is sometimes called an accumulator.
     * <p>
     * Note that the Publisher that results from this method will emit the value returned by
     * the {@code seedSupplier} as its first item.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors downstream backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  Violating this expectation, a {@code MissingBackpressureException} <em>may</em> get signaled somewhere downstream.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code scanWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the initial, accumulator and result type
     * @param seedSupplier
     *            a Callable that returns the initial (seed) accumulator item for each individual Subscriber
     * @param accumulator
     *            an accumulator function to be invoked on each item emitted by the source Publisher, whose
     *            result will be emitted to {@link Subscriber}s via {@link Subscriber#onNext onNext} and used in the
     *            next accumulator call
     * @return a Flowable that emits {@code initialValue} followed by the results of each call to the
     *         accumulator function
     * @see <a href="http://reactivex.io/documentation/operators/scan.html">ReactiveX operators documentation: Scan</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> scanWith(Callable<R> seedSupplier, BiFunction<R, ? super T, R> accumulator) {
        ObjectHelper.requireNonNull(seedSupplier, "seedSupplier is null");
        ObjectHelper.requireNonNull(accumulator, "accumulator is null");
        return RxJavaPlugins.onAssembly(new FlowableScanSeed<T, R>(this, seedSupplier, accumulator));
    }

    
Forces a Publisher's emissions and notifications to be serialized and for it to obey
the Publisher contract in other ways.
 It is possible for a Publisher to invoke its Subscribers' methods asynchronously, perhaps from different threads. This could make such a Publisher poorly-behaved, in that it might try to invoke onComplete or onError before one of its onNext invocations, or it might call onNext from two different threads concurrently. You can force such a Publisher to be well-behaved and sequential by applying the serialize method to it. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
serialize does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Serialize
Returns:
a Publisher that is guaranteed to be well-behaved and to make only serialized calls to its Subscribers/**
     * Forces a Publisher's emissions and notifications to be serialized and for it to obey
     * <a href="http://reactivex.io/documentation/contract.html">the Publisher contract</a> in other ways.
     * <p>
     * It is possible for a Publisher to invoke its Subscribers' methods asynchronously, perhaps from
     * different threads. This could make such a Publisher poorly-behaved, in that it might try to invoke
     * {@code onComplete} or {@code onError} before one of its {@code onNext} invocations, or it might call
     * {@code onNext} from two different threads concurrently. You can force such a Publisher to be
     * well-behaved and sequential by applying the {@code serialize} method to it.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/synchronize.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code serialize} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a {@link Publisher} that is guaranteed to be well-behaved and to make only serialized calls to
     *         its Subscribers
     * @see <a href="http://reactivex.io/documentation/operators/serialize.html">ReactiveX operators documentation: Serialize</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> serialize() {
        return RxJavaPlugins.onAssembly(new FlowableSerialized<T>(this));
    }

    
Returns a new Publisher that multicasts (and shares a single subscription to) the original Publisher. As long as there is at least one Subscriber this Publisher will be subscribed and emitting data. When all subscribers have canceled it will cancel the source Publisher. 
 This is an alias for publish().refCount(). 



 
Backpressure:
 
The operator honors backpressure and expects the source Publisher to honor backpressure as well. If this expectation is violated, the operator will signal a MissingBackpressureException to its Subscribers.
 
Scheduler:
 
share does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: RefCount
Returns:
a Publisher that upon connection causes the source Publisher to emit items to its Subscribers/**
     * Returns a new {@link Publisher} that multicasts (and shares a single subscription to) the original {@link Publisher}. As long as
     * there is at least one {@link Subscriber} this {@link Publisher} will be subscribed and emitting data.
     * When all subscribers have canceled it will cancel the source {@link Publisher}.
     * <p>
     * This is an alias for {@link #publish()}.{@link ConnectableFlowable#refCount() refCount()}.
     * <p>
     * <img width="640" height="510" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/publishRefCount.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure and expects the source {@code Publisher} to honor backpressure as well.
     *  If this expectation is violated, the operator will signal a {@code MissingBackpressureException} to
     *  its {@code Subscriber}s.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code share} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a {@code Publisher} that upon connection causes the source {@code Publisher} to emit items
     *         to its {@link Subscriber}s
     * @see <a href="http://reactivex.io/documentation/operators/refcount.html">ReactiveX operators documentation: RefCount</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> share() {
        return publish().refCount();
    }

    
Returns a Maybe that completes if this Flowable is empty, signals one item if this Flowable signals exactly one item or signals an IllegalArgumentException if this Flowable signals more than one item. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
singleElement does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: First
Returns:
a Maybe that emits the single item emitted by the source Publisher/**
     * Returns a Maybe that completes if this Flowable is empty, signals one item if this Flowable
     * signals exactly one item or signals an {@code IllegalArgumentException} if this Flowable signals
     * more than one item.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/single.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code singleElement} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Maybe that emits the single item emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX operators documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Maybe<T> singleElement() {
        return RxJavaPlugins.onAssembly(new FlowableSingleMaybe<T>(this));
    }

    
Returns a Single that emits the single item emitted by the source Publisher, if that Publisher emits only a single item, or a default item if the source Publisher emits no items. If the source Publisher emits more than one item, an IllegalArgumentException is signaled instead. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
single does not operate by default on a particular Scheduler.

Params:
defaultItem – 
           a default value to emit if the source Publisher emits no item
See Also:
ReactiveX operators documentation: First
Returns:
a Single that emits the single item emitted by the source Publisher, or a default item if
        the source Publisher is empty/**
     * Returns a Single that emits the single item emitted by the source Publisher, if that Publisher
     * emits only a single item, or a default item if the source Publisher emits no items. If the source
     * Publisher emits more than one item, an {@code IllegalArgumentException} is signaled instead.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/singleOrDefault.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code single} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param defaultItem
     *            a default value to emit if the source Publisher emits no item
     * @return a Single that emits the single item emitted by the source Publisher, or a default item if
     *         the source Publisher is empty
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX operators documentation: First</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> single(T defaultItem) {
        ObjectHelper.requireNonNull(defaultItem, "defaultItem is null");
        return RxJavaPlugins.onAssembly(new FlowableSingleSingle<T>(this, defaultItem));
    }

    
Returns a Single that emits the single item emitted by this Flowable, if this Flowable emits only a single item, otherwise if this Flowable completes without emitting any items a NoSuchElementException will be signaled and if this Flowable emits more than one item, an IllegalArgumentException will be signaled. 



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure).
 
Scheduler:
 
singleOrError does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: First
Returns:
the new Single instance/**
     * Returns a Single that emits the single item emitted by this Flowable, if this Flowable
     * emits only a single item, otherwise
     * if this Flowable completes without emitting any items a {@link NoSuchElementException} will be signaled and
     * if this Flowable emits more than one item, an {@code IllegalArgumentException} will be signaled.
     * <p>
     * <img width="640" height="205" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/singleOrError.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code singleOrError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return the new Single instance
     * @see <a href="http://reactivex.io/documentation/operators/first.html">ReactiveX operators documentation: First</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<T> singleOrError() {
        return RxJavaPlugins.onAssembly(new FlowableSingleSingle<T>(this, null));
    }

    
Returns a Flowable that skips the first count items emitted by the source Publisher and emits the remainder. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
This version of skip does not operate by default on a particular Scheduler.

Params:
count – 
           the number of items to skip
See Also:
ReactiveX operators documentation: Skip
Returns:
a Flowable that is identical to the source Publisher except that it does not emit the first count items that the source Publisher emits/**
     * Returns a Flowable that skips the first {@code count} items emitted by the source Publisher and emits
     * the remainder.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skip.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code skip} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the number of items to skip
     * @return a Flowable that is identical to the source Publisher except that it does not emit the first
     *         {@code count} items that the source Publisher emits
     * @see <a href="http://reactivex.io/documentation/operators/skip.html">ReactiveX operators documentation: Skip</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> skip(long count) {
        if (count <= 0L) {
            return RxJavaPlugins.onAssembly(this);
        }
        return RxJavaPlugins.onAssembly(new FlowableSkip<T>(this, count));
    }

    
Returns a Flowable that skips values emitted by the source Publisher before a specified time window
elapses.



 
Backpressure:
 
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and thus has to consume the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
skip does not operate on any particular scheduler but uses the current time from the computation Scheduler.

Params:
time – 
           the length of the time window to skip
unit –  the time unit of time
See Also:
ReactiveX operators documentation: Skip
Returns:
a Flowable that skips values emitted by the source Publisher before the time window defined by time elapses and the emits the remainder/**
     * Returns a Flowable that skips values emitted by the source Publisher before a specified time window
     * elapses.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skip.t.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and
     *  thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code skip} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window to skip
     * @param unit
     *            the time unit of {@code time}
     * @return a Flowable that skips values emitted by the source Publisher before the time window defined
     *         by {@code time} elapses and the emits the remainder
     * @see <a href="http://reactivex.io/documentation/operators/skip.html">ReactiveX operators documentation: Skip</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> skip(long time, TimeUnit unit) {
        return skipUntil(timer(time, unit));
    }

    
Returns a Flowable that skips values emitted by the source Publisher before a specified time window on a specified Scheduler elapses. 



 
Backpressure:
 
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and thus has to consume the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
You specify which Scheduler this operator will use for the timed skipping

Params:
time – 
           the length of the time window to skip
unit –  the time unit of time
scheduler –  the Scheduler on which the timed wait happens
See Also:
ReactiveX operators documentation: Skip
Returns:
a Flowable that skips values emitted by the source Publisher before the time window defined by time and scheduler elapses, and then emits the remainder/**
     * Returns a Flowable that skips values emitted by the source Publisher before a specified time window
     * on a specified {@link Scheduler} elapses.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skip.ts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and
     *  thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use for the timed skipping</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window to skip
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the {@link Scheduler} on which the timed wait happens
     * @return a Flowable that skips values emitted by the source Publisher before the time window defined
     *         by {@code time} and {@code scheduler} elapses, and then emits the remainder
     * @see <a href="http://reactivex.io/documentation/operators/skip.html">ReactiveX operators documentation: Skip</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> skip(long time, TimeUnit unit, Scheduler scheduler) {
        return skipUntil(timer(time, unit, scheduler));
    }

    
Returns a Flowable that drops a specified number of items from the end of the sequence emitted by the
source Publisher.


 This Subscriber accumulates a queue long enough to store the first count items. As more items are received, items are taken from the front of the queue and emitted by the returned Publisher. This causes such items to be delayed. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
This version of skipLast does not operate by default on a particular Scheduler.

Params:
count – 
           number of items to drop from the end of the source sequence
Throws:
IndexOutOfBoundsException –  if count is less than zero
See Also:
ReactiveX operators documentation: SkipLast
Returns:
a Flowable that emits the items emitted by the source Publisher except for the dropped ones
        at the end/**
     * Returns a Flowable that drops a specified number of items from the end of the sequence emitted by the
     * source Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipLast.png" alt="">
     * <p>
     * This Subscriber accumulates a queue long enough to store the first {@code count} items. As more items are
     * received, items are taken from the front of the queue and emitted by the returned Publisher. This causes
     * such items to be delayed.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code skipLast} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            number of items to drop from the end of the source sequence
     * @return a Flowable that emits the items emitted by the source Publisher except for the dropped ones
     *         at the end
     * @throws IndexOutOfBoundsException
     *             if {@code count} is less than zero
     * @see <a href="http://reactivex.io/documentation/operators/skiplast.html">ReactiveX operators documentation: SkipLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> skipLast(int count) {
        if (count < 0) {
            throw new IndexOutOfBoundsException("count >= 0 required but it was " + count);
        }
        if (count == 0) {
            return RxJavaPlugins.onAssembly(this);
        }
        return RxJavaPlugins.onAssembly(new FlowableSkipLast<T>(this, count));
    }

    
Returns a Flowable that drops items emitted by the source Publisher during a specified time window
before the source completes.



Note: this action will cache the latest items arriving in the specified time window.

 
Backpressure:
 
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and thus has to consume the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
skipLast does not operate on any particular scheduler but uses the current time from the computation Scheduler.

Params:
time – 
           the length of the time window
unit –  the time unit of time
See Also:
ReactiveX operators documentation: SkipLast
Returns:
a Flowable that drops those items emitted by the source Publisher in a time window before the source completes defined by time/**
     * Returns a Flowable that drops items emitted by the source Publisher during a specified time window
     * before the source completes.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipLast.t.png" alt="">
     * <p>
     * Note: this action will cache the latest items arriving in the specified time window.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and
     *  thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code skipLast} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @return a Flowable that drops those items emitted by the source Publisher in a time window before the
     *         source completes defined by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/skiplast.html">ReactiveX operators documentation: SkipLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> skipLast(long time, TimeUnit unit) {
        return skipLast(time, unit, Schedulers.computation(), false, bufferSize());
    }

    
Returns a Flowable that drops items emitted by the source Publisher during a specified time window
before the source completes.



Note: this action will cache the latest items arriving in the specified time window.

 
Backpressure:
 
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and thus has to consume the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
skipLast does not operate on any particular scheduler but uses the current time from the computation Scheduler.

Params:
time – 
           the length of the time window
unit –  the time unit of time
delayError – 
           if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
           by the downstream; if false, an exception is immediately signaled and all regular elements dropped
See Also:
ReactiveX operators documentation: SkipLast
Returns:
a Flowable that drops those items emitted by the source Publisher in a time window before the source completes defined by time/**
     * Returns a Flowable that drops items emitted by the source Publisher during a specified time window
     * before the source completes.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipLast.t.png" alt="">
     * <p>
     * Note: this action will cache the latest items arriving in the specified time window.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and
     *  thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code skipLast} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param delayError
     *            if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
     *            by the downstream; if false, an exception is immediately signaled and all regular elements dropped
     * @return a Flowable that drops those items emitted by the source Publisher in a time window before the
     *         source completes defined by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/skiplast.html">ReactiveX operators documentation: SkipLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> skipLast(long time, TimeUnit unit, boolean delayError) {
        return skipLast(time, unit, Schedulers.computation(), delayError, bufferSize());
    }

    
Returns a Flowable that drops items emitted by the source Publisher during a specified time window
(defined on a specified scheduler) before the source completes.



Note: this action will cache the latest items arriving in the specified time window.

 
Backpressure:
 
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and thus has to consume the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
You specify which Scheduler this operator will use for tracking the current time

Params:
time – 
           the length of the time window
unit –  the time unit of time
scheduler – 
           the scheduler used as the time source
See Also:
ReactiveX operators documentation: SkipLast
Returns:
a Flowable that drops those items emitted by the source Publisher in a time window before the source completes defined by time and scheduler/**
     * Returns a Flowable that drops items emitted by the source Publisher during a specified time window
     * (defined on a specified scheduler) before the source completes.
     * <p>
     * <img width="640" height="340" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipLast.ts.png" alt="">
     * <p>
     * Note: this action will cache the latest items arriving in the specified time window.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and
     *  thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use for tracking the current time</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the scheduler used as the time source
     * @return a Flowable that drops those items emitted by the source Publisher in a time window before the
     *         source completes defined by {@code time} and {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/skiplast.html">ReactiveX operators documentation: SkipLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> skipLast(long time, TimeUnit unit, Scheduler scheduler) {
        return skipLast(time, unit, scheduler, false, bufferSize());
    }

    
Returns a Flowable that drops items emitted by the source Publisher during a specified time window
(defined on a specified scheduler) before the source completes.



Note: this action will cache the latest items arriving in the specified time window.

 
Backpressure:
 
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and thus has to consume the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
You specify which Scheduler this operator will use to track the current time

Params:
time – 
           the length of the time window
unit –  the time unit of time
scheduler – 
           the scheduler used as the time source
delayError – 
           if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
           by the downstream; if false, an exception is immediately signaled and all regular elements dropped
See Also:
ReactiveX operators documentation: SkipLast
Returns:
a Flowable that drops those items emitted by the source Publisher in a time window before the source completes defined by time and scheduler/**
     * Returns a Flowable that drops items emitted by the source Publisher during a specified time window
     * (defined on a specified scheduler) before the source completes.
     * <p>
     * <img width="640" height="340" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipLast.ts.png" alt="">
     * <p>
     * Note: this action will cache the latest items arriving in the specified time window.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and
     *  thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use to track the current time</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the scheduler used as the time source
     * @param delayError
     *            if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
     *            by the downstream; if false, an exception is immediately signaled and all regular elements dropped
     * @return a Flowable that drops those items emitted by the source Publisher in a time window before the
     *         source completes defined by {@code time} and {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/skiplast.html">ReactiveX operators documentation: SkipLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> skipLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError) {
        return skipLast(time, unit, scheduler, delayError, bufferSize());
    }

    
Returns a Flowable that drops items emitted by the source Publisher during a specified time window
(defined on a specified scheduler) before the source completes.



Note: this action will cache the latest items arriving in the specified time window.

 
Backpressure:
 
The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and thus has to consume the source Publisher in an unbounded manner (i.e., no backpressure applied to it).
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
time – 
           the length of the time window
unit –  the time unit of time
scheduler – 
           the scheduler used as the time source
delayError – 
           if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
           by the downstream; if false, an exception is immediately signaled and all regular elements dropped
bufferSize – 
           the hint about how many elements to expect to be skipped
See Also:
ReactiveX operators documentation: SkipLast
Returns:
a Flowable that drops those items emitted by the source Publisher in a time window before the source completes defined by time and scheduler/**
     * Returns a Flowable that drops items emitted by the source Publisher during a specified time window
     * (defined on a specified scheduler) before the source completes.
     * <p>
     * <img width="640" height="340" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipLast.ts.png" alt="">
     * <p>
     * Note: this action will cache the latest items arriving in the specified time window.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't support backpressure as it uses time to skip an arbitrary number of elements and
     *  thus has to consume the source {@code Publisher} in an unbounded manner (i.e., no backpressure applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the scheduler used as the time source
     * @param delayError
     *            if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
     *            by the downstream; if false, an exception is immediately signaled and all regular elements dropped
     * @param bufferSize
     *            the hint about how many elements to expect to be skipped
     * @return a Flowable that drops those items emitted by the source Publisher in a time window before the
     *         source completes defined by {@code time} and {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/skiplast.html">ReactiveX operators documentation: SkipLast</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> skipLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError, int bufferSize) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        // the internal buffer holds pairs of (timestamp, value) so double the default buffer size
        int s = bufferSize << 1;
        return RxJavaPlugins.onAssembly(new FlowableSkipLastTimed<T>(this, time, unit, scheduler, s, delayError));
    }

    
Returns a Flowable that skips items emitted by the source Publisher until a second Publisher emits
an item.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
skipUntil does not operate by default on a particular Scheduler.

Params:
other – 
           the second Publisher that has to emit an item before the source Publisher's elements begin
           to be mirrored by the resulting Publisher
Type parameters:
<U> – the element type of the other Publisher
See Also:
ReactiveX operators documentation: SkipUntil
Returns:
a Flowable that skips items from the source Publisher until the second Publisher emits an
        item, then emits the remaining items/**
     * Returns a Flowable that skips items emitted by the source Publisher until a second Publisher emits
     * an item.
     * <p>
     * <img width="640" height="375" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipUntil.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code skipUntil} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the element type of the other Publisher
     * @param other
     *            the second Publisher that has to emit an item before the source Publisher's elements begin
     *            to be mirrored by the resulting Publisher
     * @return a Flowable that skips items from the source Publisher until the second Publisher emits an
     *         item, then emits the remaining items
     * @see <a href="http://reactivex.io/documentation/operators/skipuntil.html">ReactiveX operators documentation: SkipUntil</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<T> skipUntil(Publisher<U> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableSkipUntil<T, U>(this, other));
    }

    
Returns a Flowable that skips all items emitted by the source Publisher as long as a specified
condition holds true, but emits all further source items as soon as the condition becomes false.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
skipWhile does not operate by default on a particular Scheduler.

Params:
predicate – 
           a function to test each item emitted from the source Publisher
See Also:
ReactiveX operators documentation: SkipWhile
Returns:
a Flowable that begins emitting items emitted by the source Publisher when the specified
        predicate becomes false/**
     * Returns a Flowable that skips all items emitted by the source Publisher as long as a specified
     * condition holds true, but emits all further source items as soon as the condition becomes false.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/skipWhile.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code skipWhile} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param predicate
     *            a function to test each item emitted from the source Publisher
     * @return a Flowable that begins emitting items emitted by the source Publisher when the specified
     *         predicate becomes false
     * @see <a href="http://reactivex.io/documentation/operators/skipwhile.html">ReactiveX operators documentation: SkipWhile</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> skipWhile(Predicate<? super T> predicate) {
        ObjectHelper.requireNonNull(predicate, "predicate is null");
        return RxJavaPlugins.onAssembly(new FlowableSkipWhile<T>(this, predicate));
    }
    
Returns a Flowable that emits the events emitted by source Publisher, in a sorted order. Each item emitted by the Publisher must implement Comparable with respect to all other items in the sequence. 
If any item emitted by this Flowable does not implement Comparable with respect to all other items emitted by this Flowable, no items will be emitted and the sequence is terminated with a ClassCastException. 
Note that calling sorted with long, non-terminating or infinite sources might cause OutOfMemoryError 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
sorted does not operate by default on a particular Scheduler.

Returns:
a Flowable that emits the items emitted by the source Publisher in sorted order/**
     * Returns a Flowable that emits the events emitted by source Publisher, in a
     * sorted order. Each item emitted by the Publisher must implement {@link Comparable} with respect to all
     * other items in the sequence.
     *
     * <p>If any item emitted by this Flowable does not implement {@link Comparable} with respect to
     *             all other items emitted by this Flowable, no items will be emitted and the
     *             sequence is terminated with a {@link ClassCastException}.
     * <p>Note that calling {@code sorted} with long, non-terminating or infinite sources
     * might cause {@link OutOfMemoryError}
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sorted} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits the items emitted by the source Publisher in sorted order
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> sorted() {
        return toList().toFlowable().map(Functions.listSorter(Functions.<T>naturalComparator())).flatMapIterable(Functions.<List<T>>identity());
    }

    
Returns a Flowable that emits the events emitted by source Publisher, in a
sorted order based on a specified comparison function.
Note that calling sorted with long, non-terminating or infinite sources might cause OutOfMemoryError 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
sorted does not operate by default on a particular Scheduler.

Params:
sortFunction – 
           a function that compares two items emitted by the source Publisher and returns an Integer
           that indicates their sort order
Returns:
a Flowable that emits the items emitted by the source Publisher in sorted order/**
     * Returns a Flowable that emits the events emitted by source Publisher, in a
     * sorted order based on a specified comparison function.
     *
     * <p>Note that calling {@code sorted} with long, non-terminating or infinite sources
     * might cause {@link OutOfMemoryError}
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code sorted} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param sortFunction
     *            a function that compares two items emitted by the source Publisher and returns an Integer
     *            that indicates their sort order
     * @return a Flowable that emits the items emitted by the source Publisher in sorted order
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> sorted(Comparator<? super T> sortFunction) {
        ObjectHelper.requireNonNull(sortFunction, "sortFunction");
        return toList().toFlowable().map(Functions.listSorter(sortFunction)).flatMapIterable(Functions.<List<T>>identity());
    }

    
Returns a Flowable that emits the items in a specified Iterable before it begins to emit items emitted by the source Publisher. 



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publisher is expected to honor backpressure as well. If it violates this rule, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
startWith does not operate by default on a particular Scheduler.

Params:
items – 
           an Iterable that contains the items you want the modified Publisher to emit first
See Also:
ReactiveX operators documentation: StartWith
Returns:
a Flowable that emits the items in the specified Iterable and then emits the items emitted by the source Publisher/**
     * Returns a Flowable that emits the items in a specified {@link Iterable} before it begins to emit items
     * emitted by the source Publisher.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/startWith.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}
     *  is expected to honor backpressure as well. If it violates this rule, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code startWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param items
     *            an Iterable that contains the items you want the modified Publisher to emit first
     * @return a Flowable that emits the items in the specified {@link Iterable} and then emits the items
     *         emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/startwith.html">ReactiveX operators documentation: StartWith</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> startWith(Iterable<? extends T> items) {
        return concatArray(fromIterable(items), this);
    }

    
Returns a Flowable that emits the items in a specified Publisher before it begins to emit items emitted by the source Publisher. 



 
Backpressure:
 
The operator honors backpressure from downstream. Both this and the other Publishers are expected to honor backpressure as well. If any of then violates this rule, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
startWith does not operate by default on a particular Scheduler.

Params:
other – 
           a Publisher that contains the items you want the modified Publisher to emit first
See Also:
ReactiveX operators documentation: StartWith
Returns:
a Flowable that emits the items in the specified Publisher and then emits the items emitted by the source Publisher/**
     * Returns a Flowable that emits the items in a specified {@link Publisher} before it begins to emit
     * items emitted by the source Publisher.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/startWith.o.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both this and the {@code other} {@code Publisher}s
     *  are expected to honor backpressure as well. If any of then violates this rule, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code startWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param other
     *            a Publisher that contains the items you want the modified Publisher to emit first
     * @return a Flowable that emits the items in the specified {@link Publisher} and then emits the items
     *         emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/startwith.html">ReactiveX operators documentation: StartWith</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> startWith(Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return concatArray(other, this);
    }

    
Returns a Flowable that emits a specified item before it begins to emit items emitted by the source
Publisher.



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publisher is expected to honor backpressure as well. If it violates this rule, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
startWith does not operate by default on a particular Scheduler.

Params:
value – 
           the item to emit first
See Also:
ReactiveX operators documentation: StartWith
Returns:
a Flowable that emits the specified item before it begins to emit items emitted by the source
        Publisher/**
     * Returns a Flowable that emits a specified item before it begins to emit items emitted by the source
     * Publisher.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/startWith.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}
     *  is expected to honor backpressure as well. If it violates this rule, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code startWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param value
     *            the item to emit first
     * @return a Flowable that emits the specified item before it begins to emit items emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/startwith.html">ReactiveX operators documentation: StartWith</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> startWith(T value) {
        ObjectHelper.requireNonNull(value, "value is null");
        return concatArray(just(value), this);
    }

    
Returns a Flowable that emits the specified items before it begins to emit items emitted by the source
Publisher.



 
Backpressure:
 
The operator honors backpressure from downstream. The source Publisher is expected to honor backpressure as well. If it violates this rule, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
startWithArray does not operate by default on a particular Scheduler.

Params:
items – 
           the array of values to emit first
See Also:
ReactiveX operators documentation: StartWith
Returns:
a Flowable that emits the specified items before it begins to emit items emitted by the source
        Publisher/**
     * Returns a Flowable that emits the specified items before it begins to emit items emitted by the source
     * Publisher.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/startWith.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The source {@code Publisher}
     *  is expected to honor backpressure as well. If it violates this rule, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code startWithArray} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param items
     *            the array of values to emit first
     * @return a Flowable that emits the specified items before it begins to emit items emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/startwith.html">ReactiveX operators documentation: StartWith</a>
     */
    @SuppressWarnings("unchecked")
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> startWithArray(T... items) {
        Flowable<T> fromArray = fromArray(items);
        if (fromArray == empty()) {
            return RxJavaPlugins.onAssembly(this);
        }
        return concatArray(fromArray, this);
    }

    
Subscribes to a Publisher and ignores onNext and onComplete emissions. 
 If the Flowable emits an error, it is wrapped into an OnErrorNotImplementedException and routed to the RxJavaPlugins.onError handler. 

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
subscribe does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: Subscribe
Returns:
a Disposable reference with which the caller can stop receiving items before the Publisher has finished sending them/**
     * Subscribes to a Publisher and ignores {@code onNext} and {@code onComplete} emissions.
     * <p>
     * If the Flowable emits an error, it is wrapped into an
     * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException}
     * and routed to the RxJavaPlugins.onError handler.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code subscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a {@link Disposable} reference with which the caller can stop receiving items before
     *         the Publisher has finished sending them
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable subscribe() {
        return subscribe(Functions.emptyConsumer(), Functions.ON_ERROR_MISSING,
                Functions.EMPTY_ACTION, FlowableInternalHelper.RequestMax.INSTANCE);
    }

    
Subscribes to a Publisher and provides a callback to handle the items it emits.
 If the Flowable emits an error, it is wrapped into an OnErrorNotImplementedException and routed to the RxJavaPlugins.onError handler. 

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
subscribe does not operate by default on a particular Scheduler.

Params:
onNext –  the Consumer<T> you have designed to accept emissions from the Publisher
Throws:
NullPointerException –  if onNext is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:
a Disposable reference with which the caller can stop receiving items before the Publisher has finished sending them/**
     * Subscribes to a Publisher and provides a callback to handle the items it emits.
     * <p>
     * If the Flowable emits an error, it is wrapped into an
     * {@link io.reactivex.exceptions.OnErrorNotImplementedException OnErrorNotImplementedException}
     * and routed to the RxJavaPlugins.onError handler.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code subscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *             the {@code Consumer<T>} you have designed to accept emissions from the Publisher
     * @return a {@link Disposable} reference with which the caller can stop receiving items before
     *         the Publisher has finished sending them
     * @throws NullPointerException
     *             if {@code onNext} is null
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable subscribe(Consumer<? super T> onNext) {
        return subscribe(onNext, Functions.ON_ERROR_MISSING,
                Functions.EMPTY_ACTION, FlowableInternalHelper.RequestMax.INSTANCE);
    }

    
Subscribes to a Publisher and provides callbacks to handle the items it emits and any error
notification it issues.

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
subscribe does not operate by default on a particular Scheduler.

Params:
onNext –  the Consumer<T> you have designed to accept emissions from the Publisher
onError –  the Consumer<Throwable> you have designed to accept any error notification from the Publisher
Throws:
NullPointerException –  if onNext is null, or if onError is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:
a Disposable reference with which the caller can stop receiving items before the Publisher has finished sending them/**
     * Subscribes to a Publisher and provides callbacks to handle the items it emits and any error
     * notification it issues.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code subscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *             the {@code Consumer<T>} you have designed to accept emissions from the Publisher
     * @param onError
     *             the {@code Consumer<Throwable>} you have designed to accept any error notification from the
     *             Publisher
     * @return a {@link Disposable} reference with which the caller can stop receiving items before
     *         the Publisher has finished sending them
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     * @throws NullPointerException
     *             if {@code onNext} is null, or
     *             if {@code onError} is null
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable subscribe(Consumer<? super T> onNext, Consumer<? super Throwable> onError) {
        return subscribe(onNext, onError, Functions.EMPTY_ACTION, FlowableInternalHelper.RequestMax.INSTANCE);
    }

    
Subscribes to a Publisher and provides callbacks to handle the items it emits and any error or
completion notification it issues.

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
subscribe does not operate by default on a particular Scheduler.

Params:
onNext –  the Consumer<T> you have designed to accept emissions from the Publisher
onError –  the Consumer<Throwable> you have designed to accept any error notification from the Publisher
onComplete –  the Action you have designed to accept a completion notification from the Publisher
Throws:
NullPointerException –  if onNext is null, or if onError is null, or if onComplete is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:
a Disposable reference with which the caller can stop receiving items before the Publisher has finished sending them/**
     * Subscribes to a Publisher and provides callbacks to handle the items it emits and any error or
     * completion notification it issues.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code subscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *             the {@code Consumer<T>} you have designed to accept emissions from the Publisher
     * @param onError
     *             the {@code Consumer<Throwable>} you have designed to accept any error notification from the
     *             Publisher
     * @param onComplete
     *             the {@code Action} you have designed to accept a completion notification from the
     *             Publisher
     * @return a {@link Disposable} reference with which the caller can stop receiving items before
     *         the Publisher has finished sending them
     * @throws NullPointerException
     *             if {@code onNext} is null, or
     *             if {@code onError} is null, or
     *             if {@code onComplete} is null
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable subscribe(Consumer<? super T> onNext, Consumer<? super Throwable> onError,
            Action onComplete) {
        return subscribe(onNext, onError, onComplete, FlowableInternalHelper.RequestMax.INSTANCE);
    }

    
Subscribes to a Publisher and provides callbacks to handle the items it emits and any error or
completion notification it issues.

 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
subscribe does not operate by default on a particular Scheduler.

Params:
onNext –  the Consumer<T> you have designed to accept emissions from the Publisher
onError –  the Consumer<Throwable> you have designed to accept any error notification from the Publisher
onComplete –  the Action you have designed to accept a completion notification from the Publisher
onSubscribe –  the Consumer that receives the upstream's Subscription
Throws:
NullPointerException –  if onNext is null, or if onError is null, or if onComplete is null, or if onSubscribe is null
See Also:
ReactiveX operators documentation: Subscribe
Returns:
a Disposable reference with which the caller can stop receiving items before the Publisher has finished sending them/**
     * Subscribes to a Publisher and provides callbacks to handle the items it emits and any error or
     * completion notification it issues.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner (i.e., no
     *  backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code subscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param onNext
     *             the {@code Consumer<T>} you have designed to accept emissions from the Publisher
     * @param onError
     *             the {@code Consumer<Throwable>} you have designed to accept any error notification from the
     *             Publisher
     * @param onComplete
     *             the {@code Action} you have designed to accept a completion notification from the
     *             Publisher
     * @param onSubscribe
     *             the {@code Consumer} that receives the upstream's Subscription
     * @return a {@link Disposable} reference with which the caller can stop receiving items before
     *         the Publisher has finished sending them
     * @throws NullPointerException
     *             if {@code onNext} is null, or
     *             if {@code onError} is null, or
     *             if {@code onComplete} is null, or
     *             if {@code onSubscribe} is null
     * @see <a href="http://reactivex.io/documentation/operators/subscribe.html">ReactiveX operators documentation: Subscribe</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Disposable subscribe(Consumer<? super T> onNext, Consumer<? super Throwable> onError,
            Action onComplete, Consumer<? super Subscription> onSubscribe) {
        ObjectHelper.requireNonNull(onNext, "onNext is null");
        ObjectHelper.requireNonNull(onError, "onError is null");
        ObjectHelper.requireNonNull(onComplete, "onComplete is null");
        ObjectHelper.requireNonNull(onSubscribe, "onSubscribe is null");

        LambdaSubscriber<T> ls = new LambdaSubscriber<T>(onNext, onError, onComplete, onSubscribe);

        subscribe(ls);

        return ls;
    }

    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    @Override
    public final void subscribe(Subscriber<? super T> s) {
        if (s instanceof FlowableSubscriber) {
            subscribe((FlowableSubscriber<? super T>)s);
        } else {
            ObjectHelper.requireNonNull(s, "s is null");
            subscribe(new StrictSubscriber<T>(s));
        }
    }

    
Establish a connection between this Flowable and the given FlowableSubscriber and
start streaming events based on the demand of the FlowableSubscriber.
 This is a "factory method" and can be called multiple times, each time starting a new Subscription. 
 Each Subscription will work for only a single FlowableSubscriber. 
 If the same FlowableSubscriber instance is subscribed to multiple Flowables and/or the same Flowable multiple times, it must ensure the serialization over its onXXX methods manually. 
 If the Flowable rejects the subscription attempt or otherwise fails it will signal the error via Subscriber.onError(Throwable). 

This subscribe method relaxes the following Reactive Streams rules:

§1.3: onNext should not be called concurrently until onSubscribe returns.
    FlowableSubscriber.onSubscribe should make sure a sync or async call triggered by request() is safe.
§2.3: onError or onComplete must not call cancel.
    Calling request() or cancel() is NOP at this point.
§2.12: onSubscribe must be called at most once on the same instance.
    FlowableSubscriber reuse is not checked and if happens, it is the responsibility of
    the FlowableSubscriber to ensure proper serialization of its onXXX methods.
§3.9: negative requests should emit an onError(IllegalArgumentException).
    Non-positive requests signal via RxJavaPlugins.onError and the stream is not affected.


 
Backpressure:
 
The backpressure behavior/expectation is determined by the supplied FlowableSubscriber.
 
Scheduler:
 
subscribe does not operate by default on a particular Scheduler.

History: 2.0.7 - experimental; 2.1 - beta
Params:
s – the FlowableSubscriber that will consume signals from this Flowable
Since:
2.2/**
     * Establish a connection between this Flowable and the given FlowableSubscriber and
     * start streaming events based on the demand of the FlowableSubscriber.
     * <p>
     * This is a "factory method" and can be called multiple times, each time starting a new {@link Subscription}.
     * <p>
     * Each {@link Subscription} will work for only a single {@link FlowableSubscriber}.
     * <p>
     * If the same {@link FlowableSubscriber} instance is subscribed to multiple {@link Flowable}s and/or the
     * same {@link Flowable} multiple times, it must ensure the serialization over its {@code onXXX}
     * methods manually.
     * <p>
     * If the {@link Flowable} rejects the subscription attempt or otherwise fails it will signal
     * the error via {@link FlowableSubscriber#onError(Throwable)}.
     * <p>
     * This subscribe method relaxes the following Reactive Streams rules:
     * <ul>
     * <li>§1.3: onNext should not be called concurrently until onSubscribe returns.
     *     <b>FlowableSubscriber.onSubscribe should make sure a sync or async call triggered by request() is safe.</b></li>
     * <li>§2.3: onError or onComplete must not call cancel.
     *     <b>Calling request() or cancel() is NOP at this point.</b></li>
     * <li>§2.12: onSubscribe must be called at most once on the same instance.
     *     <b>FlowableSubscriber reuse is not checked and if happens, it is the responsibility of
     *     the FlowableSubscriber to ensure proper serialization of its onXXX methods.</b></li>
     * <li>§3.9: negative requests should emit an onError(IllegalArgumentException).
     *     <b>Non-positive requests signal via RxJavaPlugins.onError and the stream is not affected.</b></li>
     * </ul>
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The backpressure behavior/expectation is determined by the supplied {@code FlowableSubscriber}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code subscribe} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.0.7 - experimental; 2.1 - beta
     * @param s the FlowableSubscriber that will consume signals from this Flowable
     * @since 2.2
     */
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final void subscribe(FlowableSubscriber<? super T> s) {
        ObjectHelper.requireNonNull(s, "s is null");
        try {
            Subscriber<? super T> z = RxJavaPlugins.onSubscribe(this, s);

            ObjectHelper.requireNonNull(z, "The RxJavaPlugins.onSubscribe hook returned a null FlowableSubscriber. Please check the handler provided to RxJavaPlugins.setOnFlowableSubscribe for invalid null returns. Further reading: https://github.com/ReactiveX/RxJava/wiki/Plugins");

            subscribeActual(z);
        } catch (NullPointerException e) { // NOPMD
            throw e;
        } catch (Throwable e) {
            Exceptions.throwIfFatal(e);
            // can't call onError because no way to know if a Subscription has been set or not
            // can't call onSubscribe because the call might have set a Subscription already
            RxJavaPlugins.onError(e);

            NullPointerException npe = new NullPointerException("Actually not, but can't throw other exceptions due to RS");
            npe.initCause(e);
            throw npe;
        }
    }

    
Operator implementations (both source and intermediate) should implement this method that performs the necessary business logic and handles the incoming Subscribers. 
There is no need to call any of the plugin hooks on the current Flowable instance or the Subscriber; all hooks and basic safeguards have been applied by subscribe(Subscriber) before this method gets called. 
Params:
s – the incoming Subscriber, never null/**
     * Operator implementations (both source and intermediate) should implement this method that
     * performs the necessary business logic and handles the incoming {@link Subscriber}s.
     * <p>There is no need to call any of the plugin hooks on the current {@code Flowable} instance or
     * the {@code Subscriber}; all hooks and basic safeguards have been
     * applied by {@link #subscribe(Subscriber)} before this method gets called.
     * @param s the incoming Subscriber, never null
     */
    protected abstract void subscribeActual(Subscriber<? super T> s);

    
Subscribes a given Subscriber (subclass) to this Flowable and returns the given
Subscriber as is.
Usage example:

Flowable<Integer> source = Flowable.range(1, 10);
CompositeDisposable composite = new CompositeDisposable();
ResourceSubscriber<Integer> rs = new ResourceSubscriber<>() {
    // ...
};
composite.add(source.subscribeWith(rs));


 
Backpressure:
 
The backpressure behavior/expectation is determined by the supplied Subscriber.
 
Scheduler:
 
subscribeWith does not operate by default on a particular Scheduler.

Params:
subscriber – the Subscriber (subclass) to use and return, not null
Type parameters:
<E> – the type of the Subscriber to use and return
Throws:
NullPointerException – if subscriber is null
Returns:
the input subscriber
Since:
2.0/**
     * Subscribes a given Subscriber (subclass) to this Flowable and returns the given
     * Subscriber as is.
     * <p>Usage example:
     * <pre><code>
     * Flowable&lt;Integer&gt; source = Flowable.range(1, 10);
     * CompositeDisposable composite = new CompositeDisposable();
     *
     * ResourceSubscriber&lt;Integer&gt; rs = new ResourceSubscriber&lt;&gt;() {
     *     // ...
     * };
     *
     * composite.add(source.subscribeWith(rs));
     * </code></pre>
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The backpressure behavior/expectation is determined by the supplied {@code Subscriber}.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code subscribeWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <E> the type of the Subscriber to use and return
     * @param subscriber the Subscriber (subclass) to use and return, not null
     * @return the input {@code subscriber}
     * @throws NullPointerException if {@code subscriber} is null
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <E extends Subscriber<? super T>> E subscribeWith(E subscriber) {
        subscribe(subscriber);
        return subscriber;
    }

    
Asynchronously subscribes Subscribers to this Publisher on the specified Scheduler. 
 If there is a create(FlowableOnSubscribe<Object>, BackpressureStrategy) type source up in the chain, it is recommended to use subscribeOn(scheduler, false) instead to avoid same-pool deadlock because requests may pile up behind an eager/blocking emitter. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
scheduler –  the Scheduler to perform subscription actions on
See Also:
ReactiveX operators documentation: SubscribeOn
RxJava Threading Examples
observeOn
subscribeOn(Scheduler, boolean)
Returns:
the source Publisher modified so that its subscriptions happen on the specified Scheduler/**
     * Asynchronously subscribes Subscribers to this Publisher on the specified {@link Scheduler}.
     * <p>
     * If there is a {@link #create(FlowableOnSubscribe, BackpressureStrategy)} type source up in the
     * chain, it is recommended to use {@code subscribeOn(scheduler, false)} instead
     * to avoid same-pool deadlock because requests may pile up behind an eager/blocking emitter.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/subscribeOn.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param scheduler
     *            the {@link Scheduler} to perform subscription actions on
     * @return the source Publisher modified so that its subscriptions happen on the
     *         specified {@link Scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/subscribeon.html">ReactiveX operators documentation: SubscribeOn</a>
     * @see <a href="http://www.grahamlea.com/2014/07/rxjava-threading-examples/">RxJava Threading Examples</a>
     * @see #observeOn
     * @see #subscribeOn(Scheduler, boolean)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> subscribeOn(@NonNull Scheduler scheduler) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return subscribeOn(scheduler, !(this instanceof FlowableCreate));
    }

    
Asynchronously subscribes Subscribers to this Publisher on the specified Scheduler optionally reroutes requests from other threads to the same Scheduler thread. 
 If there is a create(FlowableOnSubscribe<Object>, BackpressureStrategy) type source up in the chain, it is recommended to have requestOn false to avoid same-pool deadlock because requests may pile up behind an eager/blocking emitter. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
You specify which Scheduler this operator will use.

History: 2.1.1 - experimental
Params:
scheduler –  the Scheduler to perform subscription actions on
requestOn – if true, requests are rerouted to the given Scheduler as well (strong pipelining)
                 if false, requests coming from any thread are simply forwarded to
                 the upstream on the same thread (weak pipelining)
See Also:
ReactiveX operators documentation: SubscribeOn
RxJava Threading Examples
observeOn
Returns:
the source Publisher modified so that its subscriptions happen on the specified Scheduler
Since:
2.2/**
     * Asynchronously subscribes Subscribers to this Publisher on the specified {@link Scheduler}
     * optionally reroutes requests from other threads to the same {@link Scheduler} thread.
     * <p>
     * If there is a {@link #create(FlowableOnSubscribe, BackpressureStrategy)} type source up in the
     * chain, it is recommended to have {@code requestOn} false to avoid same-pool deadlock
     * because requests may pile up behind an eager/blocking emitter.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/subscribeOn.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     * <p>History: 2.1.1 - experimental
     * @param scheduler
     *            the {@link Scheduler} to perform subscription actions on
     * @param requestOn if true, requests are rerouted to the given Scheduler as well (strong pipelining)
     *                  if false, requests coming from any thread are simply forwarded to
     *                  the upstream on the same thread (weak pipelining)
     * @return the source Publisher modified so that its subscriptions happen on the
     *         specified {@link Scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/subscribeon.html">ReactiveX operators documentation: SubscribeOn</a>
     * @see <a href="http://www.grahamlea.com/2014/07/rxjava-threading-examples/">RxJava Threading Examples</a>
     * @see #observeOn
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> subscribeOn(@NonNull Scheduler scheduler, boolean requestOn) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableSubscribeOn<T>(this, scheduler, requestOn));
    }

    
Returns a Flowable that emits the items emitted by the source Publisher or the items of an alternate
Publisher if the source Publisher is empty.



 
Backpressure:
 
If the source Publisher is empty, the alternate Publisher is expected to honor backpressure. If the source Publisher is non-empty, it is expected to honor backpressure as instead. In either case, if violated, a MissingBackpressureException may get
 signaled somewhere downstream.
 
 
Scheduler:
 
switchIfEmpty does not operate by default on a particular Scheduler.

Params:
other – 
             the alternate Publisher to subscribe to if the source does not emit any items
Returns:
 a Publisher that emits the items emitted by the source Publisher or the items of an
         alternate Publisher if the source Publisher is empty.
Since:
1.1.0/**
     * Returns a Flowable that emits the items emitted by the source Publisher or the items of an alternate
     * Publisher if the source Publisher is empty.
     * <p>
     * <img width="640" height="255" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchifempty.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>If the source {@code Publisher} is empty, the alternate {@code Publisher} is expected to honor backpressure.
     *  If the source {@code Publisher} is non-empty, it is expected to honor backpressure as instead.
     *  In either case, if violated, a {@code MissingBackpressureException} <em>may</em> get
     *  signaled somewhere downstream.
     *  </dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchIfEmpty} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param other
     *              the alternate Publisher to subscribe to if the source does not emit any items
     * @return  a Publisher that emits the items emitted by the source Publisher or the items of an
     *          alternate Publisher if the source Publisher is empty.
     * @since 1.1.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> switchIfEmpty(Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableSwitchIfEmpty<T>(this, other));
    }

    
Returns a new Publisher by applying a function that you supply to each item emitted by the source
Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
of these Publishers.

The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
If the upstream Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.



 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
Type parameters:
<R> – the element type of the inner Publishers and the output
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the items emitted by the Publisher returned from applying func to the most recently emitted item emitted by the source Publisher/**
     * Returns a new Publisher by applying a function that you supply to each item emitted by the source
     * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
     * of these Publishers.
     * <p>
     * The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
     * If the upstream Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the element type of the inner Publishers and the output
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMap(Function<? super T, ? extends Publisher<? extends R>> mapper) {
        return switchMap(mapper, bufferSize());
    }

    
Returns a new Publisher by applying a function that you supply to each item emitted by the source
Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
of these Publishers.

The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
If the upstream Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.



 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchMap does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
bufferSize – 
           the number of elements to prefetch from the current active inner Publisher
Type parameters:
<R> – the element type of the inner Publishers and the output
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the items emitted by the Publisher returned from applying func to the most recently emitted item emitted by the source Publisher/**
     * Returns a new Publisher by applying a function that you supply to each item emitted by the source
     * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
     * of these Publishers.
     * <p>
     * The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
     * If the upstream Publisher signals an onError, the inner Publisher is canceled and the error delivered in-sequence.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the element type of the inner Publishers and the output
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @param bufferSize
     *            the number of elements to prefetch from the current active inner Publisher
     * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMap(Function<? super T, ? extends Publisher<? extends R>> mapper, int bufferSize) {
        return switchMap0(mapper, bufferSize, false);
    }

    
Maps the upstream values into CompletableSources, subscribes to the newer one while disposing the subscription to the previous CompletableSource, thus keeping at most one active CompletableSource running. 


 Since a CompletableSource doesn't produce any items, the resulting reactive type of this operator is a Completable that can only indicate successful completion or a failure in any of the inner CompletableSources or the failure of the current Flowable. 

 
Backpressure:
 
The operator consumes the current Flowable in an unbounded manner and otherwise does not have backpressure in its return type because no items are ever produced.
 
Scheduler:
 
switchMapCompletable does not operate by default on a particular Scheduler.
 
Error handling:
 
If either this Flowable or the active CompletableSource signals an onError, the resulting Completable is terminated immediately with that Throwable. Use the switchMapCompletableDelayError(Function) to delay such inner failures until every inner CompletableSources and the main Flowable terminates in some fashion. If they fail concurrently, the operator may combine the Throwables into a CompositeException and signal it to the downstream instead. If any inactivated (switched out) CompletableSource signals an onError late, the Throwables will be signaled to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. 

History: 2.1.11 - experimental
Params:
mapper – the function called with each upstream item and should return a CompletableSource to be subscribed to and awaited for (non blockingly) for its terminal event
See Also:
switchMapCompletableDelayError(Function)
Returns:
the new Completable instance
Since:
2.2/**
     * Maps the upstream values into {@link CompletableSource}s, subscribes to the newer one while
     * disposing the subscription to the previous {@code CompletableSource}, thus keeping at most one
     * active {@code CompletableSource} running.
     * <p>
     * <img width="640" height="521" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMapCompletable.f.png" alt="">
     * <p>
     * Since a {@code CompletableSource} doesn't produce any items, the resulting reactive type of
     * this operator is a {@link Completable} that can only indicate successful completion or
     * a failure in any of the inner {@code CompletableSource}s or the failure of the current
     * {@link Flowable}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the current {@link Flowable} in an unbounded manner and otherwise
     *  does not have backpressure in its return type because no items are ever produced.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapCompletable} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>If either this {@code Flowable} or the active {@code CompletableSource} signals an {@code onError},
     *  the resulting {@code Completable} is terminated immediately with that {@code Throwable}.
     *  Use the {@link #switchMapCompletableDelayError(Function)} to delay such inner failures until
     *  every inner {@code CompletableSource}s and the main {@code Flowable} terminates in some fashion.
     *  If they fail concurrently, the operator may combine the {@code Throwable}s into a
     *  {@link io.reactivex.exceptions.CompositeException CompositeException}
     *  and signal it to the downstream instead. If any inactivated (switched out) {@code CompletableSource}
     *  signals an {@code onError} late, the {@code Throwable}s will be signaled to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors.
     *  </dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param mapper the function called with each upstream item and should return a
     *               {@link CompletableSource} to be subscribed to and awaited for
     *               (non blockingly) for its terminal event
     * @return the new Completable instance
     * @see #switchMapCompletableDelayError(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Completable switchMapCompletable(@NonNull Function<? super T, ? extends CompletableSource> mapper) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        return RxJavaPlugins.onAssembly(new FlowableSwitchMapCompletable<T>(this, mapper, false));
    }

    
Maps the upstream values into CompletableSources, subscribes to the newer one while disposing the subscription to the previous CompletableSource, thus keeping at most one active CompletableSource running and delaying any main or inner errors until all of them terminate. 


 Since a CompletableSource doesn't produce any items, the resulting reactive type of this operator is a Completable that can only indicate successful completion or a failure in any of the inner CompletableSources or the failure of the current Flowable. 

 
Backpressure:
 
The operator consumes the current Flowable in an unbounded manner and otherwise does not have backpressure in its return type because no items are ever produced.
 
Scheduler:
 
switchMapCompletableDelayError does not operate by default on a particular Scheduler.
 
Error handling:
 
Errors of this Flowable and all the CompletableSources, who had the chance to run to their completion, are delayed until all of them terminate in some fashion. At this point, if there was only one failure, the respective Throwable is emitted to the downstream. If there was more than one failure, the operator combines all Throwables into a CompositeException and signals that to the downstream. If any inactivated (switched out) CompletableSource signals an onError late, the Throwables will be signaled to the global error handler via RxJavaPlugins.onError(Throwable) method as UndeliverableException errors. 

History: 2.1.11 - experimental
Params:
mapper – the function called with each upstream item and should return a CompletableSource to be subscribed to and awaited for (non blockingly) for its terminal event
See Also:
switchMapCompletableDelayError(Function)
Returns:
the new Completable instance
Since:
2.2/**
     * Maps the upstream values into {@link CompletableSource}s, subscribes to the newer one while
     * disposing the subscription to the previous {@code CompletableSource}, thus keeping at most one
     * active {@code CompletableSource} running and delaying any main or inner errors until all
     * of them terminate.
     * <p>
     * <img width="640" height="453" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMapCompletableDelayError.f.png" alt="">
     * <p>
     * Since a {@code CompletableSource} doesn't produce any items, the resulting reactive type of
     * this operator is a {@link Completable} that can only indicate successful completion or
     * a failure in any of the inner {@code CompletableSource}s or the failure of the current
     * {@link Flowable}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the current {@link Flowable} in an unbounded manner and otherwise
     *  does not have backpressure in its return type because no items are ever produced.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapCompletableDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>Errors of this {@code Flowable} and all the {@code CompletableSource}s, who had the chance
     *  to run to their completion, are delayed until
     *  all of them terminate in some fashion. At this point, if there was only one failure, the respective
     *  {@code Throwable} is emitted to the downstream. If there was more than one failure, the
     *  operator combines all {@code Throwable}s into a {@link io.reactivex.exceptions.CompositeException CompositeException}
     *  and signals that to the downstream.
     *  If any inactivated (switched out) {@code CompletableSource}
     *  signals an {@code onError} late, the {@code Throwable}s will be signaled to the global error handler via
     *  {@link RxJavaPlugins#onError(Throwable)} method as {@code UndeliverableException} errors.
     *  </dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param mapper the function called with each upstream item and should return a
     *               {@link CompletableSource} to be subscribed to and awaited for
     *               (non blockingly) for its terminal event
     * @return the new Completable instance
     * @see #switchMapCompletableDelayError(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Completable switchMapCompletableDelayError(@NonNull Function<? super T, ? extends CompletableSource> mapper) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        return RxJavaPlugins.onAssembly(new FlowableSwitchMapCompletable<T>(this, mapper, true));
    }

    
Returns a new Publisher by applying a function that you supply to each item emitted by the source
Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
of these Publishers and delays any error until all Publishers terminate.

The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
If the upstream Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.



 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchMapDelayError does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
Type parameters:
<R> – the element type of the inner Publishers and the output
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the items emitted by the Publisher returned from applying func to the most recently emitted item emitted by the source Publisher
Since:
2.0/**
     * Returns a new Publisher by applying a function that you supply to each item emitted by the source
     * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
     * of these Publishers and delays any error until all Publishers terminate.
     * <p>
     * The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
     * If the upstream Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
     * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the element type of the inner Publishers and the output
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMapDelayError(Function<? super T, ? extends Publisher<? extends R>> mapper) {
        return switchMapDelayError(mapper, bufferSize());
    }

    
Returns a new Publisher by applying a function that you supply to each item emitted by the source
Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
of these Publishers and delays any error until all Publishers terminate.

The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
If the upstream Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.



 
Backpressure:
 
The operator honors backpressure from downstream. The outer Publisher is consumed in an unbounded manner (i.e., without backpressure) and the inner Publishers are expected to honor backpressure but it is not enforced; the operator won't signal a MissingBackpressureException but the violation may lead to OutOfMemoryError due to internal buffer bloat.
 
Scheduler:
 
switchMapDelayError does not operate by default on a particular Scheduler.

Params:
mapper – 
           a function that, when applied to an item emitted by the source Publisher, returns a
           Publisher
bufferSize – 
           the number of elements to prefetch from the current active inner Publisher
Type parameters:
<R> – the element type of the inner Publishers and the output
See Also:
ReactiveX operators documentation: FlatMap
Returns:
a Flowable that emits the items emitted by the Publisher returned from applying func to the most recently emitted item emitted by the source Publisher
Since:
2.0/**
     * Returns a new Publisher by applying a function that you supply to each item emitted by the source
     * Publisher that returns a Publisher, and then emitting the items emitted by the most recently emitted
     * of these Publishers and delays any error until all Publishers terminate.
     * <p>
     * The resulting Publisher completes if both the upstream Publisher and the last inner Publisher, if any, complete.
     * If the upstream Publisher signals an onError, the termination of the last inner Publisher will emit that error as is
     * or wrapped into a CompositeException along with the other possible errors the former inner Publishers signaled.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The outer {@code Publisher} is consumed in an
     *  unbounded manner (i.e., without backpressure) and the inner {@code Publisher}s are expected to honor
     *  backpressure but it is not enforced; the operator won't signal a {@code MissingBackpressureException}
     *  but the violation <em>may</em> lead to {@code OutOfMemoryError} due to internal buffer bloat.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the element type of the inner Publishers and the output
     * @param mapper
     *            a function that, when applied to an item emitted by the source Publisher, returns a
     *            Publisher
     * @param bufferSize
     *            the number of elements to prefetch from the current active inner Publisher
     * @return a Flowable that emits the items emitted by the Publisher returned from applying {@code func} to the most recently emitted item emitted by the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/flatmap.html">ReactiveX operators documentation: FlatMap</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMapDelayError(Function<? super T, ? extends Publisher<? extends R>> mapper, int bufferSize) {
        return switchMap0(mapper, bufferSize, true);
    }

    <R> Flowable<R> switchMap0(Function<? super T, ? extends Publisher<? extends R>> mapper, int bufferSize, boolean delayError) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        if (this instanceof ScalarCallable) {
            @SuppressWarnings("unchecked")
            T v = ((ScalarCallable<T>)this).call();
            if (v == null) {
                return empty();
            }
            return FlowableScalarXMap.scalarXMap(v, mapper);
        }
        return RxJavaPlugins.onAssembly(new FlowableSwitchMap<T, R>(this, mapper, bufferSize, delayError));
    }

    
Maps the upstream items into MaybeSources and switches (subscribes) to the newer ones while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one if available while failing immediately if this Flowable or any of the active inner MaybeSources fail. 



 
Backpressure:
 
The operator honors backpressure from downstream. The main Flowable is consumed in an unbounded manner (i.e., without backpressure).
 
Scheduler:
 
switchMapMaybe does not operate by default on a particular Scheduler.
 
Error handling:
 
This operator terminates with an onError if this Flowable or any of the inner MaybeSources fail while they are active. When this happens concurrently, their individual Throwable errors may get combined and emitted as a single CompositeException. Otherwise, a late (i.e., inactive or switched out) onError from this Flowable or from any of the inner MaybeSources will be forwarded to the global error handler via RxJavaPlugins.onError(Throwable) as UndeliverableException

History: 2.1.11 - experimental
Params:
mapper – the function called with the current upstream event and should return a MaybeSource to replace the current active inner source and get subscribed to.
Type parameters:
<R> – the output value type
See Also:
switchMapMaybe(Function)
Returns:
the new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link MaybeSource}s and switches (subscribes) to the newer ones
     * while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one if
     * available while failing immediately if this {@code Flowable} or any of the
     * active inner {@code MaybeSource}s fail.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an
     *  unbounded manner (i.e., without backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapMaybe} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>This operator terminates with an {@code onError} if this {@code Flowable} or any of
     *  the inner {@code MaybeSource}s fail while they are active. When this happens concurrently, their
     *  individual {@code Throwable} errors may get combined and emitted as a single
     *  {@link io.reactivex.exceptions.CompositeException CompositeException}. Otherwise, a late
     *  (i.e., inactive or switched out) {@code onError} from this {@code Flowable} or from any of
     *  the inner {@code MaybeSource}s will be forwarded to the global error handler via
     *  {@link io.reactivex.plugins.RxJavaPlugins#onError(Throwable)} as
     *  {@link io.reactivex.exceptions.UndeliverableException UndeliverableException}</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the output value type
     * @param mapper the function called with the current upstream event and should
     *               return a {@code MaybeSource} to replace the current active inner source
     *               and get subscribed to.
     * @return the new Flowable instance
     * @see #switchMapMaybe(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMapMaybe(@NonNull Function<? super T, ? extends MaybeSource<? extends R>> mapper) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        return RxJavaPlugins.onAssembly(new FlowableSwitchMapMaybe<T, R>(this, mapper, false));
    }

    
Maps the upstream items into MaybeSources and switches (subscribes) to the newer ones while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one if available, delaying errors from this Flowable or the inner MaybeSources until all terminate. 



 
Backpressure:
 
The operator honors backpressure from downstream. The main Flowable is consumed in an unbounded manner (i.e., without backpressure).
 
Scheduler:
 
switchMapMaybeDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the current upstream event and should return a MaybeSource to replace the current active inner source and get subscribed to.
Type parameters:
<R> – the output value type
See Also:
switchMapMaybe(Function)
Returns:
the new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link MaybeSource}s and switches (subscribes) to the newer ones
     * while disposing the older ones  (and ignoring their signals) and emits the latest success value of the current one if
     * available, delaying errors from this {@code Flowable} or the inner {@code MaybeSource}s until all terminate.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an
     *  unbounded manner (i.e., without backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapMaybeDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the output value type
     * @param mapper the function called with the current upstream event and should
     *               return a {@code MaybeSource} to replace the current active inner source
     *               and get subscribed to.
     * @return the new Flowable instance
     * @see #switchMapMaybe(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMapMaybeDelayError(@NonNull Function<? super T, ? extends MaybeSource<? extends R>> mapper) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        return RxJavaPlugins.onAssembly(new FlowableSwitchMapMaybe<T, R>(this, mapper, true));
    }

    
Maps the upstream items into SingleSources and switches (subscribes) to the newer ones while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one while failing immediately if this Flowable or any of the active inner SingleSources fail. 



 
Backpressure:
 
The operator honors backpressure from downstream. The main Flowable is consumed in an unbounded manner (i.e., without backpressure).
 
Scheduler:
 
switchMapSingle does not operate by default on a particular Scheduler.
 
Error handling:
 
This operator terminates with an onError if this Flowable or any of the inner SingleSources fail while they are active. When this happens concurrently, their individual Throwable errors may get combined and emitted as a single CompositeException. Otherwise, a late (i.e., inactive or switched out) onError from this Flowable or from any of the inner SingleSources will be forwarded to the global error handler via RxJavaPlugins.onError(Throwable) as UndeliverableException

History: 2.1.11 - experimental
Params:
mapper – the function called with the current upstream event and should return a SingleSource to replace the current active inner source and get subscribed to.
Type parameters:
<R> – the output value type
See Also:
switchMapSingle(Function)
Returns:
the new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link SingleSource}s and switches (subscribes) to the newer ones
     * while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one
     * while failing immediately if this {@code Flowable} or any of the
     * active inner {@code SingleSource}s fail.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an
     *  unbounded manner (i.e., without backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapSingle} does not operate by default on a particular {@link Scheduler}.</dd>
     *  <dt><b>Error handling:</b></dt>
     *  <dd>This operator terminates with an {@code onError} if this {@code Flowable} or any of
     *  the inner {@code SingleSource}s fail while they are active. When this happens concurrently, their
     *  individual {@code Throwable} errors may get combined and emitted as a single
     *  {@link io.reactivex.exceptions.CompositeException CompositeException}. Otherwise, a late
     *  (i.e., inactive or switched out) {@code onError} from this {@code Flowable} or from any of
     *  the inner {@code SingleSource}s will be forwarded to the global error handler via
     *  {@link io.reactivex.plugins.RxJavaPlugins#onError(Throwable)} as
     *  {@link io.reactivex.exceptions.UndeliverableException UndeliverableException}</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the output value type
     * @param mapper the function called with the current upstream event and should
     *               return a {@code SingleSource} to replace the current active inner source
     *               and get subscribed to.
     * @return the new Flowable instance
     * @see #switchMapSingle(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMapSingle(@NonNull Function<? super T, ? extends SingleSource<? extends R>> mapper) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        return RxJavaPlugins.onAssembly(new FlowableSwitchMapSingle<T, R>(this, mapper, false));
    }

    
Maps the upstream items into SingleSources and switches (subscribes) to the newer ones while disposing the older ones (and ignoring their signals) and emits the latest success value of the current one, delaying errors from this Flowable or the inner SingleSources until all terminate. 



 
Backpressure:
 
The operator honors backpressure from downstream. The main Flowable is consumed in an unbounded manner (i.e., without backpressure).
 
Scheduler:
 
switchMapSingleDelayError does not operate by default on a particular Scheduler.

History: 2.1.11 - experimental
Params:
mapper – the function called with the current upstream event and should return a SingleSource to replace the current active inner source and get subscribed to.
Type parameters:
<R> – the output value type
See Also:
switchMapSingle(Function)
Returns:
the new Flowable instance
Since:
2.2/**
     * Maps the upstream items into {@link SingleSource}s and switches (subscribes) to the newer ones
     * while disposing the older ones  (and ignoring their signals) and emits the latest success value of the current one,
     * delaying errors from this {@code Flowable} or the inner {@code SingleSource}s until all terminate.
     * <p>
     * <img width="640" height="350" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/switchMap.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The main {@code Flowable} is consumed in an
     *  unbounded manner (i.e., without backpressure).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code switchMapSingleDelayError} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.11 - experimental
     * @param <R> the output value type
     * @param mapper the function called with the current upstream event and should
     *               return a {@code SingleSource} to replace the current active inner source
     *               and get subscribed to.
     * @return the new Flowable instance
     * @see #switchMapSingle(Function)
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> switchMapSingleDelayError(@NonNull Function<? super T, ? extends SingleSource<? extends R>> mapper) {
        ObjectHelper.requireNonNull(mapper, "mapper is null");
        return RxJavaPlugins.onAssembly(new FlowableSwitchMapSingle<T, R>(this, mapper, true));
    }

    
Returns a Flowable that emits only the first count items emitted by the source Publisher. If the source emits fewer than count items then all of its items are emitted. 


 This method returns a Publisher that will invoke a subscribing Subscriber's onNext function a maximum of count times before invoking onComplete. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior in case the first request is smaller than the count. Otherwise, the source Publisher is consumed in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
This version of take does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum number of items to emit
See Also:
ReactiveX operators documentation: Take
Returns:
a Flowable that emits only the first count items emitted by the source Publisher, or all of the items from the source Publisher if that Publisher emits fewer than count items/**
     * Returns a Flowable that emits only the first {@code count} items emitted by the source Publisher. If the source emits fewer than
     * {@code count} items then all of its items are emitted.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/take.png" alt="">
     * <p>
     * This method returns a Publisher that will invoke a subscribing {@link Subscriber}'s
     * {@link Subscriber#onNext onNext} function a maximum of {@code count} times before invoking
     * {@link Subscriber#onComplete onComplete}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior in case the first request is smaller than the {@code count}. Otherwise, the source {@code Publisher}
     *  is consumed in an unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code take} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum number of items to emit
     * @return a Flowable that emits only the first {@code count} items emitted by the source Publisher, or
     *         all of the items from the source Publisher if that Publisher emits fewer than {@code count} items
     * @see <a href="http://reactivex.io/documentation/operators/take.html">ReactiveX operators documentation: Take</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL) // may trigger UNBOUNDED_IN
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> take(long count) {
        if (count < 0) {
            throw new IllegalArgumentException("count >= 0 required but it was " + count);
        }
        return RxJavaPlugins.onAssembly(new FlowableTake<T>(this, count));
    }

    
Returns a Flowable that emits those items emitted by source Publisher before a specified time runs
out.
 If time runs out before the Flowable completes normally, the onComplete event will be signaled on the default computation Scheduler. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
This version of take operates by default on the computation Scheduler.

Params:
time – 
           the length of the time window
unit –  the time unit of time
See Also:
ReactiveX operators documentation: Take
Returns:
a Flowable that emits those items emitted by the source Publisher before the time runs out/**
     * Returns a Flowable that emits those items emitted by source Publisher before a specified time runs
     * out.
     * <p>
     * If time runs out before the {@code Flowable} completes normally, the {@code onComplete} event will be
     * signaled on the default {@code computation} {@link Scheduler}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/take.t.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code take} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @return a Flowable that emits those items emitted by the source Publisher before the time runs out
     * @see <a href="http://reactivex.io/documentation/operators/take.html">ReactiveX operators documentation: Take</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> take(long time, TimeUnit unit) {
        return takeUntil(timer(time, unit));
    }

    
Returns a Flowable that emits those items emitted by source Publisher before a specified time (on a
specified Scheduler) runs out.
 If time runs out before the Flowable completes normally, the onComplete event will be signaled on the provided Scheduler. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
time – 
           the length of the time window
unit –  the time unit of time
scheduler – 
           the Scheduler used for time source
See Also:
ReactiveX operators documentation: Take
Returns:
a Flowable that emits those items emitted by the source Publisher before the time runs out,
        according to the specified Scheduler/**
     * Returns a Flowable that emits those items emitted by source Publisher before a specified time (on a
     * specified Scheduler) runs out.
     * <p>
     * If time runs out before the {@code Flowable} completes normally, the {@code onComplete} event will be
     * signaled on the provided {@link Scheduler}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/take.ts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the Scheduler used for time source
     * @return a Flowable that emits those items emitted by the source Publisher before the time runs out,
     *         according to the specified Scheduler
     * @see <a href="http://reactivex.io/documentation/operators/take.html">ReactiveX operators documentation: Take</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> take(long time, TimeUnit unit, Scheduler scheduler) {
        return takeUntil(timer(time, unit, scheduler));
    }

    
Returns a Flowable that emits at most the last count items emitted by the source Publisher. If the source emits fewer than count items then all of its items are emitted. 



 
Backpressure:
 
The operator honors backpressure from downstream if the count is non-zero; ignores backpressure if the count is zero as it doesn't signal any values.
 
Scheduler:
 
This version of takeLast does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum number of items to emit from the end of the sequence of items emitted by the source
           Publisher
Throws:
IndexOutOfBoundsException –  if count is less than zero
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits at most the last count items emitted by the source Publisher/**
     * Returns a Flowable that emits at most the last {@code count} items emitted by the source Publisher. If the source emits fewer than
     * {@code count} items then all of its items are emitted.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.n.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream if the {@code count} is non-zero; ignores
     *  backpressure if the {@code count} is zero as it doesn't signal any values.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code takeLast} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum number of items to emit from the end of the sequence of items emitted by the source
     *            Publisher
     * @return a Flowable that emits at most the last {@code count} items emitted by the source Publisher
     * @throws IndexOutOfBoundsException
     *             if {@code count} is less than zero
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> takeLast(int count) {
        if (count < 0) {
            throw new IndexOutOfBoundsException("count >= 0 required but it was " + count);
        } else
        if (count == 0) {
            return RxJavaPlugins.onAssembly(new FlowableIgnoreElements<T>(this));
        } else
        if (count == 1) {
            return RxJavaPlugins.onAssembly(new FlowableTakeLastOne<T>(this));
        }
        return RxJavaPlugins.onAssembly(new FlowableTakeLast<T>(this, count));
    }

    
Returns a Flowable that emits at most a specified number of items from the source Publisher that were
emitted in a specified window of time before the Publisher completed.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
takeLast does not operate on any particular scheduler but uses the current time from the computation Scheduler.

Params:
count – 
           the maximum number of items to emit
time – 
           the length of the time window
unit –  the time unit of time
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits at most count items from the source Publisher that were emitted in a specified window of time before the Publisher completed/**
     * Returns a Flowable that emits at most a specified number of items from the source Publisher that were
     * emitted in a specified window of time before the Publisher completed.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.tn.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code takeLast} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum number of items to emit
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @return a Flowable that emits at most {@code count} items from the source Publisher that were emitted
     *         in a specified window of time before the Publisher completed
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> takeLast(long count, long time, TimeUnit unit) {
        return takeLast(count, time, unit, Schedulers.computation(), false, bufferSize());
    }

    
Returns a Flowable that emits at most a specified number of items from the source Publisher that were
emitted in a specified window of time before the Publisher completed, where the timing information is
provided by a given Scheduler.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
You specify which Scheduler this operator will use for tracking the current time

Params:
count – 
           the maximum number of items to emit
time – 
           the length of the time window
unit –  the time unit of time
scheduler –  the Scheduler that provides the timestamps for the observed items
Throws:
IndexOutOfBoundsException –  if count is less than zero
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits at most count items from the source Publisher that were emitted in a specified window of time before the Publisher completed, where the timing information is provided by the given scheduler/**
     * Returns a Flowable that emits at most a specified number of items from the source Publisher that were
     * emitted in a specified window of time before the Publisher completed, where the timing information is
     * provided by a given Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.tns.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use for tracking the current time</dd>
     * </dl>
     *
     * @param count
     *            the maximum number of items to emit
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the {@link Scheduler} that provides the timestamps for the observed items
     * @return a Flowable that emits at most {@code count} items from the source Publisher that were emitted
     *         in a specified window of time before the Publisher completed, where the timing information is
     *         provided by the given {@code scheduler}
     * @throws IndexOutOfBoundsException
     *             if {@code count} is less than zero
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> takeLast(long count, long time, TimeUnit unit, Scheduler scheduler) {
        return takeLast(count, time, unit, scheduler, false, bufferSize());
    }

    
Returns a Flowable that emits at most a specified number of items from the source Publisher that were
emitted in a specified window of time before the Publisher completed, where the timing information is
provided by a given Scheduler.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it).
 
Scheduler:
 
You specify which Scheduler this operator will use for tracking the current time

Params:
count – 
           the maximum number of items to emit
time – 
           the length of the time window
unit –  the time unit of time
scheduler –  the Scheduler that provides the timestamps for the observed items
delayError – 
           if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
           by the downstream; if false, an exception is immediately signaled and all regular elements dropped
bufferSize – 
           the hint about how many elements to expect to be last
Throws:
IndexOutOfBoundsException –  if count is less than zero
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits at most count items from the source Publisher that were emitted in a specified window of time before the Publisher completed, where the timing information is provided by the given scheduler/**
     * Returns a Flowable that emits at most a specified number of items from the source Publisher that were
     * emitted in a specified window of time before the Publisher completed, where the timing information is
     * provided by a given Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.tns.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use for tracking the current time</dd>
     * </dl>
     *
     * @param count
     *            the maximum number of items to emit
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the {@link Scheduler} that provides the timestamps for the observed items
     * @param delayError
     *            if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
     *            by the downstream; if false, an exception is immediately signaled and all regular elements dropped
     * @param bufferSize
     *            the hint about how many elements to expect to be last
     * @return a Flowable that emits at most {@code count} items from the source Publisher that were emitted
     *         in a specified window of time before the Publisher completed, where the timing information is
     *         provided by the given {@code scheduler}
     * @throws IndexOutOfBoundsException
     *             if {@code count} is less than zero
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> takeLast(long count, long time, TimeUnit unit, Scheduler scheduler, boolean delayError, int bufferSize) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        if (count < 0) {
            throw new IndexOutOfBoundsException("count >= 0 required but it was " + count);
        }
        return RxJavaPlugins.onAssembly(new FlowableTakeLastTimed<T>(this, count, time, unit, scheduler, bufferSize, delayError));
    }

    
Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
window of time before the Publisher completed.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it) but note that this may lead to OutOfMemoryError due to internal buffer bloat. Consider using takeLast(long, long, TimeUnit) in this case.
 
Scheduler:
 
This version of takeLast operates by default on the computation Scheduler.

Params:
time – 
           the length of the time window
unit –  the time unit of time
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits the items from the source Publisher that were emitted in the window of time before the Publisher completed specified by time/**
     * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
     * window of time before the Publisher completed.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.t.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it) but note that this <em>may</em>
     *  lead to {@code OutOfMemoryError} due to internal buffer bloat.
     *  Consider using {@link #takeLast(long, long, TimeUnit)} in this case.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code takeLast} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @return a Flowable that emits the items from the source Publisher that were emitted in the window of
     *         time before the Publisher completed specified by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> takeLast(long time, TimeUnit unit) {
        return takeLast(time, unit, Schedulers.computation(), false, bufferSize());
    }

    
Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
window of time before the Publisher completed.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it) but note that this may lead to OutOfMemoryError due to internal buffer bloat. Consider using takeLast(long, long, TimeUnit) in this case.
 
Scheduler:
 
This version of takeLast operates by default on the computation Scheduler.

Params:
time – 
           the length of the time window
unit –  the time unit of time
delayError – 
           if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
           by the downstream; if false, an exception is immediately signaled and all regular elements dropped
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits the items from the source Publisher that were emitted in the window of time before the Publisher completed specified by time/**
     * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
     * window of time before the Publisher completed.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.t.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it) but note that this <em>may</em>
     *  lead to {@code OutOfMemoryError} due to internal buffer bloat.
     *  Consider using {@link #takeLast(long, long, TimeUnit)} in this case.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code takeLast} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param delayError
     *            if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
     *            by the downstream; if false, an exception is immediately signaled and all regular elements dropped
     * @return a Flowable that emits the items from the source Publisher that were emitted in the window of
     *         time before the Publisher completed specified by {@code time}
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> takeLast(long time, TimeUnit unit, boolean delayError) {
        return takeLast(time, unit, Schedulers.computation(), delayError, bufferSize());
    }

    
Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
window of time before the Publisher completed, where the timing information is provided by a specified
Scheduler.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it) but note that this may lead to OutOfMemoryError due to internal buffer bloat. Consider using takeLast(long, long, TimeUnit, Scheduler) in this case.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
time – 
           the length of the time window
unit –  the time unit of time
scheduler – 
           the Scheduler that provides the timestamps for the Observed items
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits the items from the source Publisher that were emitted in the window of time before the Publisher completed specified by time, where the timing information is provided by scheduler/**
     * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
     * window of time before the Publisher completed, where the timing information is provided by a specified
     * Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.ts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it) but note that this <em>may</em>
     *  lead to {@code OutOfMemoryError} due to internal buffer bloat.
     *  Consider using {@link #takeLast(long, long, TimeUnit, Scheduler)} in this case.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the Scheduler that provides the timestamps for the Observed items
     * @return a Flowable that emits the items from the source Publisher that were emitted in the window of
     *         time before the Publisher completed specified by {@code time}, where the timing information is
     *         provided by {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> takeLast(long time, TimeUnit unit, Scheduler scheduler) {
        return takeLast(time, unit, scheduler, false, bufferSize());
    }

    
Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
window of time before the Publisher completed, where the timing information is provided by a specified
Scheduler.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it) but note that this may lead to OutOfMemoryError due to internal buffer bloat. Consider using takeLast(long, long, TimeUnit, Scheduler) in this case.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
time – 
           the length of the time window
unit –  the time unit of time
scheduler – 
           the Scheduler that provides the timestamps for the Observed items
delayError – 
           if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
           by the downstream; if false, an exception is immediately signaled and all regular elements dropped
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits the items from the source Publisher that were emitted in the window of time before the Publisher completed specified by time, where the timing information is provided by scheduler/**
     * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
     * window of time before the Publisher completed, where the timing information is provided by a specified
     * Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.ts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it) but note that this <em>may</em>
     *  lead to {@code OutOfMemoryError} due to internal buffer bloat.
     *  Consider using {@link #takeLast(long, long, TimeUnit, Scheduler)} in this case.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the Scheduler that provides the timestamps for the Observed items
     * @param delayError
     *            if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
     *            by the downstream; if false, an exception is immediately signaled and all regular elements dropped
     * @return a Flowable that emits the items from the source Publisher that were emitted in the window of
     *         time before the Publisher completed specified by {@code time}, where the timing information is
     *         provided by {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> takeLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError) {
        return takeLast(time, unit, scheduler, delayError, bufferSize());
    }

    
Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
window of time before the Publisher completed, where the timing information is provided by a specified
Scheduler.



 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., no backpressure is applied to it) but note that this may lead to OutOfMemoryError due to internal buffer bloat. Consider using takeLast(long, long, TimeUnit, Scheduler) in this case.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
time – 
           the length of the time window
unit –  the time unit of time
scheduler – 
           the Scheduler that provides the timestamps for the Observed items
delayError – 
           if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
           by the downstream; if false, an exception is immediately signaled and all regular elements dropped
bufferSize – 
           the hint about how many elements to expect to be last
See Also:
ReactiveX operators documentation: TakeLast
Returns:
a Flowable that emits the items from the source Publisher that were emitted in the window of time before the Publisher completed specified by time, where the timing information is provided by scheduler/**
     * Returns a Flowable that emits the items from the source Publisher that were emitted in a specified
     * window of time before the Publisher completed, where the timing information is provided by a specified
     * Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeLast.ts.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., no backpressure is applied to it) but note that this <em>may</em>
     *  lead to {@code OutOfMemoryError} due to internal buffer bloat.
     *  Consider using {@link #takeLast(long, long, TimeUnit, Scheduler)} in this case.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param time
     *            the length of the time window
     * @param unit
     *            the time unit of {@code time}
     * @param scheduler
     *            the Scheduler that provides the timestamps for the Observed items
     * @param delayError
     *            if true, an exception signaled by the current Flowable is delayed until the regular elements are consumed
     *            by the downstream; if false, an exception is immediately signaled and all regular elements dropped
     * @param bufferSize
     *            the hint about how many elements to expect to be last
     * @return a Flowable that emits the items from the source Publisher that were emitted in the window of
     *         time before the Publisher completed specified by {@code time}, where the timing information is
     *         provided by {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/takelast.html">ReactiveX operators documentation: TakeLast</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> takeLast(long time, TimeUnit unit, Scheduler scheduler, boolean delayError, int bufferSize) {
        return takeLast(Long.MAX_VALUE, time, unit, scheduler, delayError, bufferSize);
    }

    
Returns a Flowable that emits items emitted by the source Publisher, checks the specified predicate
for each item, and then completes when the condition is satisfied.


 The difference between this operator and takeWhile(Predicate) is that here, the condition is evaluated after the item is emitted.

 
Backpressure:
 
The operator is a pass-through for backpressure; the backpressure behavior is determined by the upstream
 source and the downstream consumer.
 
Scheduler:
 
takeUntil does not operate by default on a particular Scheduler.

Params:
stopPredicate – 
           a function that evaluates an item emitted by the source Publisher and returns a Boolean
See Also:
ReactiveX operators documentation: TakeUntil
takeWhile(Predicate)
Returns:
a Flowable that first emits items emitted by the source Publisher, checks the specified
        condition after each item, and then completes when the condition is satisfied.
Since:
1.1.0/**
     * Returns a Flowable that emits items emitted by the source Publisher, checks the specified predicate
     * for each item, and then completes when the condition is satisfied.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeUntil.p.png" alt="">
     * <p>
     * The difference between this operator and {@link #takeWhile(Predicate)} is that here, the condition is
     * evaluated <em>after</em> the item is emitted.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator is a pass-through for backpressure; the backpressure behavior is determined by the upstream
     *  source and the downstream consumer.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code takeUntil} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param stopPredicate
     *            a function that evaluates an item emitted by the source Publisher and returns a Boolean
     * @return a Flowable that first emits items emitted by the source Publisher, checks the specified
     *         condition after each item, and then completes when the condition is satisfied.
     * @see <a href="http://reactivex.io/documentation/operators/takeuntil.html">ReactiveX operators documentation: TakeUntil</a>
     * @see Flowable#takeWhile(Predicate)
     * @since 1.1.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> takeUntil(Predicate<? super T> stopPredicate) {
        ObjectHelper.requireNonNull(stopPredicate, "stopPredicate is null");
        return RxJavaPlugins.onAssembly(new FlowableTakeUntilPredicate<T>(this, stopPredicate));
    }

    
Returns a Flowable that emits the items emitted by the source Publisher until a second Publisher
emits an item.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
takeUntil does not operate by default on a particular Scheduler.

Params:
other –  the Publisher whose first emitted item will cause takeUntil to stop emitting items from the source Publisher
Type parameters:
<U> –  the type of items emitted by other
See Also:
ReactiveX operators documentation: TakeUntil
Returns:
a Flowable that emits the items emitted by the source Publisher until such time as other emits its first item/**
     * Returns a Flowable that emits the items emitted by the source Publisher until a second Publisher
     * emits an item.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeUntil.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code takeUntil} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param other
     *            the Publisher whose first emitted item will cause {@code takeUntil} to stop emitting items
     *            from the source Publisher
     * @param <U>
     *            the type of items emitted by {@code other}
     * @return a Flowable that emits the items emitted by the source Publisher until such time as {@code other} emits its first item
     * @see <a href="http://reactivex.io/documentation/operators/takeuntil.html">ReactiveX operators documentation: TakeUntil</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U> Flowable<T> takeUntil(Publisher<U> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return RxJavaPlugins.onAssembly(new FlowableTakeUntil<T, U>(this, other));
    }

    
Returns a Flowable that emits items emitted by the source Publisher so long as each item satisfied a
specified condition, and then completes as soon as this condition is not satisfied.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
takeWhile does not operate by default on a particular Scheduler.

Params:
predicate – 
           a function that evaluates an item emitted by the source Publisher and returns a Boolean
See Also:
ReactiveX operators documentation: TakeWhile
takeUntil(Predicate)
Returns:
a Flowable that emits the items from the source Publisher so long as each item satisfies the condition defined by predicate, then completes/**
     * Returns a Flowable that emits items emitted by the source Publisher so long as each item satisfied a
     * specified condition, and then completes as soon as this condition is not satisfied.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/takeWhile.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code takeWhile} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param predicate
     *            a function that evaluates an item emitted by the source Publisher and returns a Boolean
     * @return a Flowable that emits the items from the source Publisher so long as each item satisfies the
     *         condition defined by {@code predicate}, then completes
     * @see <a href="http://reactivex.io/documentation/operators/takewhile.html">ReactiveX operators documentation: TakeWhile</a>
     * @see Flowable#takeUntil(Predicate)
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<T> takeWhile(Predicate<? super T> predicate) {
        ObjectHelper.requireNonNull(predicate, "predicate is null");
        return RxJavaPlugins.onAssembly(new FlowableTakeWhile<T>(this, predicate));
    }

    
Returns a Flowable that emits only the first item emitted by the source Publisher during sequential
time windows of a specified duration.
 This differs from throttleLast in that this only tracks the passage of time whereas throttleLast ticks at scheduled intervals. 



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
throttleFirst operates by default on the computation Scheduler.

Params:
windowDuration – 
           time to wait before emitting another item after emitting the last item
unit –  the unit of time of windowDuration
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
Returns:
a Flowable that performs the throttle operation/**
     * Returns a Flowable that emits only the first item emitted by the source Publisher during sequential
     * time windows of a specified duration.
     * <p>
     * This differs from {@link #throttleLast} in that this only tracks the passage of time whereas
     * {@link #throttleLast} ticks at scheduled intervals.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleFirst.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code throttleFirst} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param windowDuration
     *            time to wait before emitting another item after emitting the last item
     * @param unit
     *            the unit of time of {@code windowDuration}
     * @return a Flowable that performs the throttle operation
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> throttleFirst(long windowDuration, TimeUnit unit) {
        return throttleFirst(windowDuration, unit, Schedulers.computation());
    }

    
Returns a Flowable that emits only the first item emitted by the source Publisher during sequential
time windows of a specified duration, where the windows are managed by a specified Scheduler.
 This differs from throttleLast in that this only tracks the passage of time whereas throttleLast ticks at scheduled intervals. 



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
skipDuration – 
           time to wait before emitting another item after emitting the last item
unit –  the unit of time of skipDuration
scheduler –  the Scheduler to use internally to manage the timers that handle timeout for each event
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
Returns:
a Flowable that performs the throttle operation/**
     * Returns a Flowable that emits only the first item emitted by the source Publisher during sequential
     * time windows of a specified duration, where the windows are managed by a specified Scheduler.
     * <p>
     * This differs from {@link #throttleLast} in that this only tracks the passage of time whereas
     * {@link #throttleLast} ticks at scheduled intervals.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleFirst.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param skipDuration
     *            time to wait before emitting another item after emitting the last item
     * @param unit
     *            the unit of time of {@code skipDuration}
     * @param scheduler
     *            the {@link Scheduler} to use internally to manage the timers that handle timeout for each
     *            event
     * @return a Flowable that performs the throttle operation
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> throttleFirst(long skipDuration, TimeUnit unit, Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableThrottleFirstTimed<T>(this, skipDuration, unit, scheduler));
    }

    
Returns a Flowable that emits only the last item emitted by the source Publisher during sequential
time windows of a specified duration.
 This differs from throttleFirst in that this ticks along at a scheduled interval whereas throttleFirst does not tick, it just tracks the passage of time. 



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
throttleLast operates by default on the computation Scheduler.

Params:
intervalDuration – 
           duration of windows within which the last item emitted by the source Publisher will be
           emitted
unit –  the unit of time of intervalDuration
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
sample(long, TimeUnit)
Returns:
a Flowable that performs the throttle operation/**
     * Returns a Flowable that emits only the last item emitted by the source Publisher during sequential
     * time windows of a specified duration.
     * <p>
     * This differs from {@link #throttleFirst} in that this ticks along at a scheduled interval whereas
     * {@link #throttleFirst} does not tick, it just tracks the passage of time.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleLast.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code throttleLast} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param intervalDuration
     *            duration of windows within which the last item emitted by the source Publisher will be
     *            emitted
     * @param unit
     *            the unit of time of {@code intervalDuration}
     * @return a Flowable that performs the throttle operation
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #sample(long, TimeUnit)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> throttleLast(long intervalDuration, TimeUnit unit) {
        return sample(intervalDuration, unit);
    }

    
Returns a Flowable that emits only the last item emitted by the source Publisher during sequential
time windows of a specified duration, where the duration is governed by a specified Scheduler.
 This differs from throttleFirst in that this ticks along at a scheduled interval whereas throttleFirst does not tick, it just tracks the passage of time. 



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
intervalDuration – 
           duration of windows within which the last item emitted by the source Publisher will be
           emitted
unit –  the unit of time of intervalDuration
scheduler –  the Scheduler to use internally to manage the timers that handle timeout for each event
See Also:
ReactiveX operators documentation: Sample
RxJava wiki: Backpressure
sample(long, TimeUnit, Scheduler)
Returns:
a Flowable that performs the throttle operation/**
     * Returns a Flowable that emits only the last item emitted by the source Publisher during sequential
     * time windows of a specified duration, where the duration is governed by a specified Scheduler.
     * <p>
     * This differs from {@link #throttleFirst} in that this ticks along at a scheduled interval whereas
     * {@link #throttleFirst} does not tick, it just tracks the passage of time.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleLast.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param intervalDuration
     *            duration of windows within which the last item emitted by the source Publisher will be
     *            emitted
     * @param unit
     *            the unit of time of {@code intervalDuration}
     * @param scheduler
     *            the {@link Scheduler} to use internally to manage the timers that handle timeout for each
     *            event
     * @return a Flowable that performs the throttle operation
     * @see <a href="http://reactivex.io/documentation/operators/sample.html">ReactiveX operators documentation: Sample</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #sample(long, TimeUnit, Scheduler)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> throttleLast(long intervalDuration, TimeUnit unit, Scheduler scheduler) {
        return sample(intervalDuration, unit, scheduler);
    }

    
Throttles items from the upstream Flowable by first emitting the next item from upstream, then periodically emitting the latest item (if any) when the specified timeout elapses between them. 


 Unlike the option with throttleLatest(long, TimeUnit, boolean), the very last item being held back (if any) is not emitted when the upstream completes. 

If no items were emitted from the upstream during this timeout phase, the next
upstream item is emitted immediately and the timeout window starts from then.

 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow. If the downstream is not ready to receive items, a MissingBackpressureException will be signaled.
 
Scheduler:
 
throttleLatest operates by default on the computation Scheduler.

History: 2.1.14 - experimental
Params:
timeout – the time to wait after an item emission towards the downstream
               before trying to emit the latest item from upstream again
unit –    the time unit
See Also:
throttleLatest(long, TimeUnit, boolean)
throttleLatest(long, TimeUnit, Scheduler)
Returns:
the new Flowable instance
Since:
2.2/**
     * Throttles items from the upstream {@code Flowable} by first emitting the next
     * item from upstream, then periodically emitting the latest item (if any) when
     * the specified timeout elapses between them.
     * <p>
     * <img width="640" height="325" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleLatest.png" alt="">
     * <p>
     * Unlike the option with {@link #throttleLatest(long, TimeUnit, boolean)}, the very last item being held back
     * (if any) is not emitted when the upstream completes.
     * <p>
     * If no items were emitted from the upstream during this timeout phase, the next
     * upstream item is emitted immediately and the timeout window starts from then.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.
     *  If the downstream is not ready to receive items, a
     *  {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException}
     *  will be signaled.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code throttleLatest} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.14 - experimental
     * @param timeout the time to wait after an item emission towards the downstream
     *                before trying to emit the latest item from upstream again
     * @param unit    the time unit
     * @return the new Flowable instance
     * @since 2.2
     * @see #throttleLatest(long, TimeUnit, boolean)
     * @see #throttleLatest(long, TimeUnit, Scheduler)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> throttleLatest(long timeout, TimeUnit unit) {
        return throttleLatest(timeout, unit, Schedulers.computation(), false);
    }

    
Throttles items from the upstream Flowable by first emitting the next item from upstream, then periodically emitting the latest item (if any) when the specified timeout elapses between them. 



If no items were emitted from the upstream during this timeout phase, the next
upstream item is emitted immediately and the timeout window starts from then.

 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow. If the downstream is not ready to receive items, a MissingBackpressureException will be signaled.
 
Scheduler:
 
throttleLatest operates by default on the computation Scheduler.

History: 2.1.14 - experimental
Params:
timeout – the time to wait after an item emission towards the downstream
               before trying to emit the latest item from upstream again
unit –    the time unit
emitLast – If true, the very last item from the upstream will be emitted immediately when the upstream completes, regardless if there is a timeout window active or not. If false, the very last upstream item is ignored and the flow terminates.
See Also:
throttleLatest(long, TimeUnit, Scheduler, boolean)
Returns:
the new Flowable instance
Since:
2.2/**
     * Throttles items from the upstream {@code Flowable} by first emitting the next
     * item from upstream, then periodically emitting the latest item (if any) when
     * the specified timeout elapses between them.
     * <p>
     * <img width="640" height="325" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleLatest.e.png" alt="">
     * <p>
     * If no items were emitted from the upstream during this timeout phase, the next
     * upstream item is emitted immediately and the timeout window starts from then.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.
     *  If the downstream is not ready to receive items, a
     *  {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException}
     *  will be signaled.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code throttleLatest} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     * <p>History: 2.1.14 - experimental
     * @param timeout the time to wait after an item emission towards the downstream
     *                before trying to emit the latest item from upstream again
     * @param unit    the time unit
     * @param emitLast If {@code true}, the very last item from the upstream will be emitted
     *                 immediately when the upstream completes, regardless if there is
     *                 a timeout window active or not. If {@code false}, the very last
     *                 upstream item is ignored and the flow terminates.
     * @return the new Flowable instance
     * @see #throttleLatest(long, TimeUnit, Scheduler, boolean)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> throttleLatest(long timeout, TimeUnit unit, boolean emitLast) {
        return throttleLatest(timeout, unit, Schedulers.computation(), emitLast);
    }

    
Throttles items from the upstream Flowable by first emitting the next item from upstream, then periodically emitting the latest item (if any) when the specified timeout elapses between them. 


 Unlike the option with throttleLatest(long, TimeUnit, Scheduler, boolean), the very last item being held back (if any) is not emitted when the upstream completes. 

If no items were emitted from the upstream during this timeout phase, the next
upstream item is emitted immediately and the timeout window starts from then.

 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow. If the downstream is not ready to receive items, a MissingBackpressureException will be signaled.
 
Scheduler:
 
You specify which Scheduler this operator will use.

History: 2.1.14 - experimental
Params:
timeout – the time to wait after an item emission towards the downstream
               before trying to emit the latest item from upstream again
unit –    the time unit
scheduler – the Scheduler where the timed wait and latest item emission will be performed
See Also:
throttleLatest(long, TimeUnit, Scheduler, boolean)
Returns:
the new Flowable instance
Since:
2.2/**
     * Throttles items from the upstream {@code Flowable} by first emitting the next
     * item from upstream, then periodically emitting the latest item (if any) when
     * the specified timeout elapses between them.
     * <p>
     * <img width="640" height="325" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleLatest.s.png" alt="">
     * <p>
     * Unlike the option with {@link #throttleLatest(long, TimeUnit, Scheduler, boolean)}, the very last item being held back
     * (if any) is not emitted when the upstream completes.
     * <p>
     * If no items were emitted from the upstream during this timeout phase, the next
     * upstream item is emitted immediately and the timeout window starts from then.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.
     *  If the downstream is not ready to receive items, a
     *  {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException}
     *  will be signaled.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     * <p>History: 2.1.14 - experimental
     * @param timeout the time to wait after an item emission towards the downstream
     *                before trying to emit the latest item from upstream again
     * @param unit    the time unit
     * @param scheduler the {@link Scheduler} where the timed wait and latest item
     *                  emission will be performed
     * @return the new Flowable instance
     * @see #throttleLatest(long, TimeUnit, Scheduler, boolean)
     * @since 2.2
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> throttleLatest(long timeout, TimeUnit unit, Scheduler scheduler) {
        return throttleLatest(timeout, unit, scheduler, false);
    }

    
Throttles items from the upstream Flowable by first emitting the next item from upstream, then periodically emitting the latest item (if any) when the specified timeout elapses between them. 



If no items were emitted from the upstream during this timeout phase, the next
upstream item is emitted immediately and the timeout window starts from then.

 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow. If the downstream is not ready to receive items, a MissingBackpressureException will be signaled.
 
Scheduler:
 
You specify which Scheduler this operator will use.

History: 2.1.14 - experimental
Params:
timeout – the time to wait after an item emission towards the downstream
               before trying to emit the latest item from upstream again
unit –    the time unit
scheduler – the Scheduler where the timed wait and latest item emission will be performed
emitLast – If true, the very last item from the upstream will be emitted immediately when the upstream completes, regardless if there is a timeout window active or not. If false, the very last upstream item is ignored and the flow terminates.
Returns:
the new Flowable instance
Since:
2.2/**
     * Throttles items from the upstream {@code Flowable} by first emitting the next
     * item from upstream, then periodically emitting the latest item (if any) when
     * the specified timeout elapses between them.
     * <p>
     * <img width="640" height="325" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleLatest.se.png" alt="">
     * <p>
     * If no items were emitted from the upstream during this timeout phase, the next
     * upstream item is emitted immediately and the timeout window starts from then.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.
     *  If the downstream is not ready to receive items, a
     *  {@link io.reactivex.exceptions.MissingBackpressureException MissingBackpressureException}
     *  will be signaled.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     * <p>History: 2.1.14 - experimental
     * @param timeout the time to wait after an item emission towards the downstream
     *                before trying to emit the latest item from upstream again
     * @param unit    the time unit
     * @param scheduler the {@link Scheduler} where the timed wait and latest item
     *                  emission will be performed
     * @param emitLast If {@code true}, the very last item from the upstream will be emitted
     *                 immediately when the upstream completes, regardless if there is
     *                 a timeout window active or not. If {@code false}, the very last
     *                 upstream item is ignored and the flow terminates.
     * @return the new Flowable instance
     * @since 2.2
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> throttleLatest(long timeout, TimeUnit unit, Scheduler scheduler, boolean emitLast) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableThrottleLatest<T>(this, timeout, unit, scheduler, emitLast));
    }

    
Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the source Publisher that are followed by newer items before a timeout value expires. The timer resets on each emission (alias to debounce(long, TimeUnit)). 

Note: If items keep being emitted by the source Publisher faster than the timeout then no items
will be emitted by the resulting Publisher.



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
throttleWithTimeout operates by default on the computation Scheduler.

Params:
timeout – 
           the length of the window of time that must pass after the emission of an item from the source
           Publisher in which that Publisher emits no items in order for the item to be emitted by the
           resulting Publisher
unit –  the unit of time for the specified timeout
See Also:
ReactiveX operators documentation: Debounce
RxJava wiki: Backpressure
debounce(long, TimeUnit)
Returns:
a Flowable that filters out items from the source Publisher that are too quickly followed by
        newer items/**
     * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
     * source Publisher that are followed by newer items before a timeout value expires. The timer resets on
     * each emission (alias to {@link #debounce(long, TimeUnit)}).
     * <p>
     * <em>Note:</em> If items keep being emitted by the source Publisher faster than the timeout then no items
     * will be emitted by the resulting Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleWithTimeout.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code throttleWithTimeout} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timeout
     *            the length of the window of time that must pass after the emission of an item from the source
     *            Publisher in which that Publisher emits no items in order for the item to be emitted by the
     *            resulting Publisher
     * @param unit
     *            the unit of time for the specified {@code timeout}
     * @return a Flowable that filters out items from the source Publisher that are too quickly followed by
     *         newer items
     * @see <a href="http://reactivex.io/documentation/operators/debounce.html">ReactiveX operators documentation: Debounce</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #debounce(long, TimeUnit)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> throttleWithTimeout(long timeout, TimeUnit unit) {
        return debounce(timeout, unit);
    }

    
Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the source Publisher that are followed by newer items before a timeout value expires on a specified Scheduler. The timer resets on each emission (alias to debounce(long, TimeUnit, Scheduler)). 

Note: If items keep being emitted by the source Publisher faster than the timeout then no items
will be emitted by the resulting Publisher.



 
Backpressure:
 
This operator does not support backpressure as it uses time to control data flow.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timeout – 
           the length of the window of time that must pass after the emission of an item from the source
           Publisher in which that Publisher emits no items in order for the item to be emitted by the
           resulting Publisher
unit –  the unit of time for the specified timeout
scheduler –  the Scheduler to use internally to manage the timers that handle the timeout for each item
See Also:
ReactiveX operators documentation: Debounce
RxJava wiki: Backpressure
debounce(long, TimeUnit, Scheduler)
Returns:
a Flowable that filters out items from the source Publisher that are too quickly followed by
        newer items/**
     * Returns a Flowable that mirrors the source Publisher, except that it drops items emitted by the
     * source Publisher that are followed by newer items before a timeout value expires on a specified
     * Scheduler. The timer resets on each emission (alias to {@link #debounce(long, TimeUnit, Scheduler)}).
     * <p>
     * <em>Note:</em> If items keep being emitted by the source Publisher faster than the timeout then no items
     * will be emitted by the resulting Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/throttleWithTimeout.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as it uses time to control data flow.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timeout
     *            the length of the window of time that must pass after the emission of an item from the source
     *            Publisher in which that Publisher emits no items in order for the item to be emitted by the
     *            resulting Publisher
     * @param unit
     *            the unit of time for the specified {@code timeout}
     * @param scheduler
     *            the {@link Scheduler} to use internally to manage the timers that handle the timeout for each
     *            item
     * @return a Flowable that filters out items from the source Publisher that are too quickly followed by
     *         newer items
     * @see <a href="http://reactivex.io/documentation/operators/debounce.html">ReactiveX operators documentation: Debounce</a>
     * @see <a href="https://github.com/ReactiveX/RxJava/wiki/Backpressure">RxJava wiki: Backpressure</a>
     * @see #debounce(long, TimeUnit, Scheduler)
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> throttleWithTimeout(long timeout, TimeUnit unit, Scheduler scheduler) {
        return debounce(timeout, unit, scheduler);
    }

    
Returns a Flowable that emits records of the time interval between consecutive items emitted by the
source Publisher.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
timeInterval does not operate on any particular scheduler but uses the current time from the computation Scheduler.

See Also:
ReactiveX operators documentation: TimeInterval
Returns:
a Flowable that emits time interval information items/**
     * Returns a Flowable that emits records of the time interval between consecutive items emitted by the
     * source Publisher.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeInterval.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code timeInterval} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits time interval information items
     * @see <a href="http://reactivex.io/documentation/operators/timeinterval.html">ReactiveX operators documentation: TimeInterval</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Timed<T>> timeInterval() {
        return timeInterval(TimeUnit.MILLISECONDS, Schedulers.computation());
    }

    
Returns a Flowable that emits records of the time interval between consecutive items emitted by the
source Publisher, where this interval is computed on a specified Scheduler.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
The operator does not operate on any particular scheduler but uses the current time from the specified Scheduler.

Params:
scheduler –  the Scheduler used to compute time intervals
See Also:
ReactiveX operators documentation: TimeInterval
Returns:
a Flowable that emits time interval information items/**
     * Returns a Flowable that emits records of the time interval between consecutive items emitted by the
     * source Publisher, where this interval is computed on a specified Scheduler.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeInterval.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>The operator does not operate on any particular scheduler but uses the current time
     *  from the specified {@link Scheduler}.</dd>
     * </dl>
     *
     * @param scheduler
     *            the {@link Scheduler} used to compute time intervals
     * @return a Flowable that emits time interval information items
     * @see <a href="http://reactivex.io/documentation/operators/timeinterval.html">ReactiveX operators documentation: TimeInterval</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps.
    public final Flowable<Timed<T>> timeInterval(Scheduler scheduler) {
        return timeInterval(TimeUnit.MILLISECONDS, scheduler);
    }

    
Returns a Flowable that emits records of the time interval between consecutive items emitted by the
source Publisher.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
timeInterval does not operate on any particular scheduler but uses the current time from the computation Scheduler.

Params:
unit – the time unit for the current time
See Also:
ReactiveX operators documentation: TimeInterval
Returns:
a Flowable that emits time interval information items/**
     * Returns a Flowable that emits records of the time interval between consecutive items emitted by the
     * source Publisher.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeInterval.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code timeInterval} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param unit the time unit for the current time
     * @return a Flowable that emits time interval information items
     * @see <a href="http://reactivex.io/documentation/operators/timeinterval.html">ReactiveX operators documentation: TimeInterval</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Timed<T>> timeInterval(TimeUnit unit) {
        return timeInterval(unit, Schedulers.computation());
    }

    
Returns a Flowable that emits records of the time interval between consecutive items emitted by the
source Publisher, where this interval is computed on a specified Scheduler.



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
The operator does not operate on any particular scheduler but uses the current time from the specified Scheduler.

Params:
unit – the time unit for the current time
scheduler –  the Scheduler used to compute time intervals
See Also:
ReactiveX operators documentation: TimeInterval
Returns:
a Flowable that emits time interval information items/**
     * Returns a Flowable that emits records of the time interval between consecutive items emitted by the
     * source Publisher, where this interval is computed on a specified Scheduler.
     * <p>
     * <img width="640" height="315" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeInterval.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>The operator does not operate on any particular scheduler but uses the current time
     *  from the specified {@link Scheduler}.</dd>
     * </dl>
     *
     * @param unit the time unit for the current time
     * @param scheduler
     *            the {@link Scheduler} used to compute time intervals
     * @return a Flowable that emits time interval information items
     * @see <a href="http://reactivex.io/documentation/operators/timeinterval.html">ReactiveX operators documentation: TimeInterval</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps.
    public final Flowable<Timed<T>> timeInterval(TimeUnit unit, Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableTimeInterval<T>(this, unit, scheduler));
    }

    
Returns a Flowable that mirrors the source Publisher, but notifies Subscribers of a TimeoutException if an item emitted by the source Publisher doesn't arrive within a window of time after the emission of the previous item, where that period of time is measured by a Publisher that is a function of the previous item. 



Note: The arrival of the first source item is never timed out.

 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
This version of timeout operates by default on the immediate Scheduler.

Params:
itemTimeoutIndicator – 
           a function that returns a Publisher for each item emitted by the source
           Publisher and that determines the timeout window for the subsequent item
Type parameters:
<V> – 
           the timeout value type (ignored)
See Also:
ReactiveX operators documentation: Timeout
Returns:
a Flowable that mirrors the source Publisher, but notifies Subscribers of a TimeoutException if an item emitted by the source Publisher takes longer to arrive than the time window defined by the selector for the previously emitted item/**
     * Returns a Flowable that mirrors the source Publisher, but notifies Subscribers of a
     * {@code TimeoutException} if an item emitted by the source Publisher doesn't arrive within a window of
     * time after the emission of the previous item, where that period of time is measured by a Publisher that
     * is a function of the previous item.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout3.png" alt="">
     * <p>
     * Note: The arrival of the first source item is never timed out.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code timeout} operates by default on the {@code immediate} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <V>
     *            the timeout value type (ignored)
     * @param itemTimeoutIndicator
     *            a function that returns a Publisher for each item emitted by the source
     *            Publisher and that determines the timeout window for the subsequent item
     * @return a Flowable that mirrors the source Publisher, but notifies Subscribers of a
     *         {@code TimeoutException} if an item emitted by the source Publisher takes longer to arrive than
     *         the time window defined by the selector for the previously emitted item
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <V> Flowable<T> timeout(Function<? super T, ? extends Publisher<V>> itemTimeoutIndicator) {
        return timeout0(null, itemTimeoutIndicator, null);
    }

    
Returns a Flowable that mirrors the source Publisher, but that switches to a fallback Publisher if
an item emitted by the source Publisher doesn't arrive within a window of time after the emission of the
previous item, where that period of time is measured by a Publisher that is a function of the previous
item.



Note: The arrival of the first source item is never timed out.

 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
This version of timeout operates by default on the immediate Scheduler.

Params:
itemTimeoutIndicator – 
           a function that returns a Publisher, for each item emitted by the source Publisher, that
           determines the timeout window for the subsequent item
other – 
           the fallback Publisher to switch to if the source Publisher times out
Type parameters:
<V> – 
           the timeout value type (ignored)
See Also:
ReactiveX operators documentation: Timeout
Returns:
a Flowable that mirrors the source Publisher, but switches to mirroring a fallback Publisher
        if an item emitted by the source Publisher takes longer to arrive than the time window defined
        by the selector for the previously emitted item/**
     * Returns a Flowable that mirrors the source Publisher, but that switches to a fallback Publisher if
     * an item emitted by the source Publisher doesn't arrive within a window of time after the emission of the
     * previous item, where that period of time is measured by a Publisher that is a function of the previous
     * item.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout4.png" alt="">
     * <p>
     * Note: The arrival of the first source item is never timed out.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code timeout} operates by default on the {@code immediate} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <V>
     *            the timeout value type (ignored)
     * @param itemTimeoutIndicator
     *            a function that returns a Publisher, for each item emitted by the source Publisher, that
     *            determines the timeout window for the subsequent item
     * @param other
     *            the fallback Publisher to switch to if the source Publisher times out
     * @return a Flowable that mirrors the source Publisher, but switches to mirroring a fallback Publisher
     *         if an item emitted by the source Publisher takes longer to arrive than the time window defined
     *         by the selector for the previously emitted item
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <V> Flowable<T> timeout(Function<? super T, ? extends Publisher<V>> itemTimeoutIndicator, Flowable<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return timeout0(null, itemTimeoutIndicator, other);
    }

    
Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted item. If the next item isn't emitted within the specified timeout duration starting from its predecessor, the resulting Publisher terminates and notifies Subscribers of a TimeoutException. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
This version of timeout operates by default on the computation Scheduler.

Params:
timeout – 
           maximum duration between emitted items before a timeout occurs
timeUnit –  the unit of time that applies to the timeout argument.
See Also:
ReactiveX operators documentation: Timeout
Returns:
the source Publisher modified to notify Subscribers of a TimeoutException in case of a timeout/**
     * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted
     * item. If the next item isn't emitted within the specified timeout duration starting from its predecessor,
     * the resulting Publisher terminates and notifies Subscribers of a {@code TimeoutException}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout.1.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code timeout} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timeout
     *            maximum duration between emitted items before a timeout occurs
     * @param timeUnit
     *            the unit of time that applies to the {@code timeout} argument.
     * @return the source Publisher modified to notify Subscribers of a {@code TimeoutException} in case of a
     *         timeout
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> timeout(long timeout, TimeUnit timeUnit) {
        return timeout0(timeout, timeUnit, null, Schedulers.computation());
    }

    
Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted
item. If the next item isn't emitted within the specified timeout duration starting from its predecessor,
the source Publisher is disposed and resulting Publisher begins instead to mirror a fallback Publisher.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
This version of timeout operates by default on the computation Scheduler.

Params:
timeout – 
           maximum duration between items before a timeout occurs
timeUnit –  the unit of time that applies to the timeout argument
other – 
           the fallback Publisher to use in case of a timeout
See Also:
ReactiveX operators documentation: Timeout
Returns:
the source Publisher modified to switch to the fallback Publisher in case of a timeout/**
     * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted
     * item. If the next item isn't emitted within the specified timeout duration starting from its predecessor,
     * the source Publisher is disposed and resulting Publisher begins instead to mirror a fallback Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout.2.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code timeout} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timeout
     *            maximum duration between items before a timeout occurs
     * @param timeUnit
     *            the unit of time that applies to the {@code timeout} argument
     * @param other
     *            the fallback Publisher to use in case of a timeout
     * @return the source Publisher modified to switch to the fallback Publisher in case of a timeout
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<T> timeout(long timeout, TimeUnit timeUnit, Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return timeout0(timeout, timeUnit, other, Schedulers.computation());
    }

    
Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted
item using a specified Scheduler. If the next item isn't emitted within the specified timeout duration
starting from its predecessor, the source Publisher is disposed and resulting Publisher begins
instead to mirror a fallback Publisher.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timeout – 
           maximum duration between items before a timeout occurs
timeUnit –  the unit of time that applies to the timeout argument
scheduler –  the Scheduler to run the timeout timers on
other – 
           the Publisher to use as the fallback in case of a timeout
See Also:
ReactiveX operators documentation: Timeout
Returns:
the source Publisher modified so that it will switch to the fallback Publisher in case of a
        timeout/**
     * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted
     * item using a specified Scheduler. If the next item isn't emitted within the specified timeout duration
     * starting from its predecessor, the source Publisher is disposed and resulting Publisher begins
     * instead to mirror a fallback Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout.2s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timeout
     *            maximum duration between items before a timeout occurs
     * @param timeUnit
     *            the unit of time that applies to the {@code timeout} argument
     * @param scheduler
     *            the {@link Scheduler} to run the timeout timers on
     * @param other
     *            the Publisher to use as the fallback in case of a timeout
     * @return the source Publisher modified so that it will switch to the fallback Publisher in case of a
     *         timeout
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> timeout(long timeout, TimeUnit timeUnit, Scheduler scheduler, Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(other, "other is null");
        return timeout0(timeout, timeUnit, other, scheduler);
    }

    
Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted item, where this policy is governed by a specified Scheduler. If the next item isn't emitted within the specified timeout duration starting from its predecessor, the resulting Publisher terminates and notifies Subscribers of a TimeoutException. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timeout – 
           maximum duration between items before a timeout occurs
timeUnit –  the unit of time that applies to the timeout argument
scheduler – 
           the Scheduler to run the timeout timers on
See Also:
ReactiveX operators documentation: Timeout
Returns:
the source Publisher modified to notify Subscribers of a TimeoutException in case of a timeout/**
     * Returns a Flowable that mirrors the source Publisher but applies a timeout policy for each emitted
     * item, where this policy is governed by a specified Scheduler. If the next item isn't emitted within the
     * specified timeout duration starting from its predecessor, the resulting Publisher terminates and
     * notifies Subscribers of a {@code TimeoutException}.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout.1s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timeout
     *            maximum duration between items before a timeout occurs
     * @param timeUnit
     *            the unit of time that applies to the {@code timeout} argument
     * @param scheduler
     *            the Scheduler to run the timeout timers on
     * @return the source Publisher modified to notify Subscribers of a {@code TimeoutException} in case of a
     *         timeout
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> timeout(long timeout, TimeUnit timeUnit, Scheduler scheduler) {
        return timeout0(timeout, timeUnit, null, scheduler);
    }

    
Returns a Flowable that mirrors the source Publisher, but notifies Subscribers of a TimeoutException if either the first item emitted by the source Publisher or any subsequent item doesn't arrive within time windows defined by other Publishers. 



 
Backpressure:
 
The operator honors backpressure from downstream. Both this and the returned Publishers are expected to honor backpressure as well. If any of then violates this rule, it may throw an IllegalStateException when the Publisher completes.
 
Scheduler:
 
timeout does not operate by default on any Scheduler.

Params:
firstTimeoutIndicator – 
           a function that returns a Publisher that determines the timeout window for the first source
           item
itemTimeoutIndicator – 
           a function that returns a Publisher for each item emitted by the source Publisher and that
           determines the timeout window in which the subsequent source item must arrive in order to
           continue the sequence
Type parameters:
<U> – 
           the first timeout value type (ignored)
<V> – 
           the subsequent timeout value type (ignored)
See Also:
ReactiveX operators documentation: Timeout
Returns:
a Flowable that mirrors the source Publisher, but notifies Subscribers of a TimeoutException if either the first item or any subsequent item doesn't arrive within the time windows specified by the timeout selectors/**
     * Returns a Flowable that mirrors the source Publisher, but notifies Subscribers of a
     * {@code TimeoutException} if either the first item emitted by the source Publisher or any subsequent item
     * doesn't arrive within time windows defined by other Publishers.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout5.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. Both this and the returned {@code Publisher}s
     *  are expected to honor backpressure as well. If any of then violates this rule, it <em>may</em> throw an
     *  {@code IllegalStateException} when the {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code timeout} does not operate by default on any {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the first timeout value type (ignored)
     * @param <V>
     *            the subsequent timeout value type (ignored)
     * @param firstTimeoutIndicator
     *            a function that returns a Publisher that determines the timeout window for the first source
     *            item
     * @param itemTimeoutIndicator
     *            a function that returns a Publisher for each item emitted by the source Publisher and that
     *            determines the timeout window in which the subsequent source item must arrive in order to
     *            continue the sequence
     * @return a Flowable that mirrors the source Publisher, but notifies Subscribers of a
     *         {@code TimeoutException} if either the first item or any subsequent item doesn't arrive within
     *         the time windows specified by the timeout selectors
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, V> Flowable<T> timeout(Publisher<U> firstTimeoutIndicator,
            Function<? super T, ? extends Publisher<V>> itemTimeoutIndicator) {
        ObjectHelper.requireNonNull(firstTimeoutIndicator, "firstTimeoutIndicator is null");
        return timeout0(firstTimeoutIndicator, itemTimeoutIndicator, null);
    }

    
Returns a Flowable that mirrors the source Publisher, but switches to a fallback Publisher if either
the first item emitted by the source Publisher or any subsequent item doesn't arrive within time windows
defined by other Publishers.



 
Backpressure:
 
The operator honors backpressure from downstream. The Publisher sources are expected to honor backpressure as well. If any of the source Publishers violate this, it may throw an IllegalStateException when the source Publisher completes.
 
Scheduler:
 
timeout does not operate by default on any Scheduler.

Params:
firstTimeoutIndicator – 
           a function that returns a Publisher which determines the timeout window for the first source
           item
itemTimeoutIndicator – 
           a function that returns a Publisher for each item emitted by the source Publisher and that
           determines the timeout window in which the subsequent source item must arrive in order to
           continue the sequence
other – 
           the fallback Publisher to switch to if the source Publisher times out
Type parameters:
<U> – 
           the first timeout value type (ignored)
<V> – 
           the subsequent timeout value type (ignored)
Throws:
NullPointerException –  if itemTimeoutIndicator is null
See Also:
ReactiveX operators documentation: Timeout
Returns:
a Flowable that mirrors the source Publisher, but switches to the other Publisher if either the first item emitted by the source Publisher or any subsequent item doesn't arrive within time windows defined by the timeout selectors/**
     * Returns a Flowable that mirrors the source Publisher, but switches to a fallback Publisher if either
     * the first item emitted by the source Publisher or any subsequent item doesn't arrive within time windows
     * defined by other Publishers.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timeout6.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream. The {@code Publisher}
     *  sources are expected to honor backpressure as well.
     *  If any of the source {@code Publisher}s violate this, it <em>may</em> throw an
     *  {@code IllegalStateException} when the source {@code Publisher} completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code timeout} does not operate by default on any {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the first timeout value type (ignored)
     * @param <V>
     *            the subsequent timeout value type (ignored)
     * @param firstTimeoutIndicator
     *            a function that returns a Publisher which determines the timeout window for the first source
     *            item
     * @param itemTimeoutIndicator
     *            a function that returns a Publisher for each item emitted by the source Publisher and that
     *            determines the timeout window in which the subsequent source item must arrive in order to
     *            continue the sequence
     * @param other
     *            the fallback Publisher to switch to if the source Publisher times out
     * @return a Flowable that mirrors the source Publisher, but switches to the {@code other} Publisher if
     *         either the first item emitted by the source Publisher or any subsequent item doesn't arrive
     *         within time windows defined by the timeout selectors
     * @throws NullPointerException
     *             if {@code itemTimeoutIndicator} is null
     * @see <a href="http://reactivex.io/documentation/operators/timeout.html">ReactiveX operators documentation: Timeout</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, V> Flowable<T> timeout(
            Publisher<U> firstTimeoutIndicator,
            Function<? super T, ? extends Publisher<V>> itemTimeoutIndicator,
                    Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(firstTimeoutIndicator, "firstTimeoutSelector is null");
        ObjectHelper.requireNonNull(other, "other is null");
        return timeout0(firstTimeoutIndicator, itemTimeoutIndicator, other);
    }

    private Flowable<T> timeout0(long timeout, TimeUnit timeUnit, Publisher<? extends T> other,
            Scheduler scheduler) {
        ObjectHelper.requireNonNull(timeUnit, "timeUnit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableTimeoutTimed<T>(this, timeout, timeUnit, scheduler, other));
    }

    private <U, V> Flowable<T> timeout0(
            Publisher<U> firstTimeoutIndicator,
            Function<? super T, ? extends Publisher<V>> itemTimeoutIndicator,
                    Publisher<? extends T> other) {
        ObjectHelper.requireNonNull(itemTimeoutIndicator, "itemTimeoutIndicator is null");
        return RxJavaPlugins.onAssembly(new FlowableTimeout<T, U, V>(this, firstTimeoutIndicator, itemTimeoutIndicator, other));
    }

    
Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a Timed object. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
timestamp does not operate on any particular scheduler but uses the current time from the computation Scheduler.

See Also:
ReactiveX operators documentation: Timestamp
Returns:
a Flowable that emits timestamped items from the source Publisher/**
     * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a
     * {@link Timed} object.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timestamp.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code timestamp} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Flowable that emits timestamped items from the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/timestamp.html">ReactiveX operators documentation: Timestamp</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Timed<T>> timestamp() {
        return timestamp(TimeUnit.MILLISECONDS, Schedulers.computation());
    }

    
Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a Timed object whose timestamps are provided by a specified Scheduler. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
This operator does not operate on any particular scheduler but uses the current time from the specified Scheduler.

Params:
scheduler –  the Scheduler to use as a time source
See Also:
ReactiveX operators documentation: Timestamp
Returns:
a Flowable that emits timestamped items from the source Publisher with timestamps provided by the scheduler/**
     * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a
     * {@link Timed} object whose timestamps are provided by a specified Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timestamp.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator does not operate on any particular scheduler but uses the current time
     *  from the specified {@link Scheduler}.</dd>
     * </dl>
     *
     * @param scheduler
     *            the {@link Scheduler} to use as a time source
     * @return a Flowable that emits timestamped items from the source Publisher with timestamps provided by
     *         the {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/timestamp.html">ReactiveX operators documentation: Timestamp</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps.
    public final Flowable<Timed<T>> timestamp(Scheduler scheduler) {
        return timestamp(TimeUnit.MILLISECONDS, scheduler);
    }

    
Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a Timed object. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
timestamp does not operate on any particular scheduler but uses the current time from the computation Scheduler.

Params:
unit – the time unit for the current time
See Also:
ReactiveX operators documentation: Timestamp
Returns:
a Flowable that emits timestamped items from the source Publisher/**
     * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a
     * {@link Timed} object.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timestamp.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code timestamp} does not operate on any particular scheduler but uses the current time
     *  from the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param unit the time unit for the current time
     * @return a Flowable that emits timestamped items from the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/timestamp.html">ReactiveX operators documentation: Timestamp</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Timed<T>> timestamp(TimeUnit unit) {
        return timestamp(unit, Schedulers.computation());
    }

    
Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a Timed object whose timestamps are provided by a specified Scheduler. 



 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
This operator does not operate on any particular scheduler but uses the current time from the specified Scheduler.

Params:
unit – the time unit for the current time
scheduler –  the Scheduler to use as a time source
See Also:
ReactiveX operators documentation: Timestamp
Returns:
a Flowable that emits timestamped items from the source Publisher with timestamps provided by the scheduler/**
     * Returns a Flowable that emits each item emitted by the source Publisher, wrapped in a
     * {@link Timed} object whose timestamps are provided by a specified Scheduler.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/timestamp.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator does not operate on any particular scheduler but uses the current time
     *  from the specified {@link Scheduler}.</dd>
     * </dl>
     *
     * @param unit the time unit for the current time
     * @param scheduler
     *            the {@link Scheduler} to use as a time source
     * @return a Flowable that emits timestamped items from the source Publisher with timestamps provided by
     *         the {@code scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/timestamp.html">ReactiveX operators documentation: Timestamp</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE) // Supplied scheduler is only used for creating timestamps.
    public final Flowable<Timed<T>> timestamp(final TimeUnit unit, final Scheduler scheduler) {
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return map(Functions.<T>timestampWith(unit, scheduler));
    }

    
Calls the specified converter function during assembly time and returns its resulting value.

This allows fluent conversion to any other type.

 
Backpressure:
 
The backpressure behavior depends on what happens in the converter function.
 
Scheduler:
 
to does not operate by default on a particular Scheduler.

Params:
converter – the function that receives the current Flowable instance and returns a value
Type parameters:
<R> – the resulting object type
Returns:
the value returned by the function/**
     * Calls the specified converter function during assembly time and returns its resulting value.
     * <p>
     * This allows fluent conversion to any other type.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The backpressure behavior depends on what happens in the {@code converter} function.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code to} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param <R> the resulting object type
     * @param converter the function that receives the current Flowable instance and returns a value
     * @return the value returned by the function
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.SPECIAL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> R to(Function<? super Flowable<T>, R> converter) {
        try {
            return ObjectHelper.requireNonNull(converter, "converter is null").apply(this);
        } catch (Throwable ex) {
            Exceptions.throwIfFatal(ex);
            throw ExceptionHelper.wrapOrThrow(ex);
        }
    }

    
Returns a Single that emits a single item, a list composed of all the items emitted by the
finite upstream source Publisher.


 Normally, a Publisher that returns multiple items will do so by invoking its Subscriber's onNext method for each such item. You can change this behavior, instructing the Publisher to compose a list of all of these items and then to invoke the Subscriber's onNext function once, passing it the entire list, by calling the Publisher's toList method prior to calling its subscribe method. 
 Note that this operator requires the upstream to signal onComplete for the accumulated list to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toList does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a List containing all of the items emitted by the source
        Publisher/**
     * Returns a Single that emits a single item, a list composed of all the items emitted by the
     * finite upstream source Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toList.png" alt="">
     * <p>
     * Normally, a Publisher that returns multiple items will do so by invoking its {@link Subscriber}'s
     * {@link Subscriber#onNext onNext} method for each such item. You can change this behavior, instructing the
     * Publisher to compose a list of all of these items and then to invoke the Subscriber's {@code onNext}
     * function once, passing it the entire list, by calling the Publisher's {@code toList} method prior to
     * calling its {@link #subscribe} method.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toList} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @return a Single that emits a single item: a List containing all of the items emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<List<T>> toList() {
        return RxJavaPlugins.onAssembly(new FlowableToListSingle<T, List<T>>(this));
    }

    
Returns a Single that emits a single item, a list composed of all the items emitted by the
finite source Publisher.


 Normally, a Publisher that returns multiple items will do so by invoking its Subscriber's onNext method for each such item. You can change this behavior, instructing the Publisher to compose a list of all of these items and then to invoke the Subscriber's onNext function once, passing it the entire list, by calling the Publisher's toList method prior to calling its subscribe method. 
 Note that this operator requires the upstream to signal onComplete for the accumulated list to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toList does not operate by default on a particular Scheduler.

Params:
capacityHint – 
        the number of elements expected from the current Flowable
See Also:
ReactiveX operators documentation: To
Returns:
a Flowable that emits a single item: a List containing all of the items emitted by the source
        Publisher/**
     * Returns a Single that emits a single item, a list composed of all the items emitted by the
     * finite source Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toList.png" alt="">
     * <p>
     * Normally, a Publisher that returns multiple items will do so by invoking its {@link Subscriber}'s
     * {@link Subscriber#onNext onNext} method for each such item. You can change this behavior, instructing the
     * Publisher to compose a list of all of these items and then to invoke the Subscriber's {@code onNext}
     * function once, passing it the entire list, by calling the Publisher's {@code toList} method prior to
     * calling its {@link #subscribe} method.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toList} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacityHint
     *         the number of elements expected from the current Flowable
     * @return a Flowable that emits a single item: a List containing all of the items emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<List<T>> toList(final int capacityHint) {
        ObjectHelper.verifyPositive(capacityHint, "capacityHint");
        return RxJavaPlugins.onAssembly(new FlowableToListSingle<T, List<T>>(this, Functions.<T>createArrayList(capacityHint)));
    }

    
Returns a Single that emits a single item, a list composed of all the items emitted by the
finite source Publisher.


 Normally, a Publisher that returns multiple items will do so by invoking its Subscriber's onNext method for each such item. You can change this behavior, instructing the Publisher to compose a list of all of these items and then to invoke the Subscriber's onNext function once, passing it the entire list, by calling the Publisher's toList method prior to calling its subscribe method. 
 Note that this operator requires the upstream to signal onComplete for the accumulated collection to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toList does not operate by default on a particular Scheduler.

Params:
collectionSupplier – 
              the Callable returning the collection (for each individual Subscriber) to be filled in
Type parameters:
<U> – the subclass of a collection of Ts
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a List containing all of the items emitted by the source
        Publisher/**
     * Returns a Single that emits a single item, a list composed of all the items emitted by the
     * finite source Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toList.png" alt="">
     * <p>
     * Normally, a Publisher that returns multiple items will do so by invoking its {@link Subscriber}'s
     * {@link Subscriber#onNext onNext} method for each such item. You can change this behavior, instructing the
     * Publisher to compose a list of all of these items and then to invoke the Subscriber's {@code onNext}
     * function once, passing it the entire list, by calling the Publisher's {@code toList} method prior to
     * calling its {@link #subscribe} method.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated collection to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toList} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the subclass of a collection of Ts
     * @param collectionSupplier
     *               the Callable returning the collection (for each individual Subscriber) to be filled in
     * @return a Single that emits a single item: a List containing all of the items emitted by the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U extends Collection<? super T>> Single<U> toList(Callable<U> collectionSupplier) {
        ObjectHelper.requireNonNull(collectionSupplier, "collectionSupplier is null");
        return RxJavaPlugins.onAssembly(new FlowableToListSingle<T, U>(this, collectionSupplier));
    }

    
Returns a Single that emits a single HashMap containing all items emitted by the finite source Publisher, mapped by the keys returned by a specified keySelector function. 



If more than one source item maps to the same key, the HashMap will contain the latest of those items.
 Note that this operator requires the upstream to signal onComplete for the accumulated map to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toMap does not operate by default on a particular Scheduler.

Params:
keySelector – 
           the function that extracts the key from a source item to be used in the HashMap
Type parameters:
<K> – the key type of the Map
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a HashMap containing the mapped items from the source
        Publisher/**
     * Returns a Single that emits a single HashMap containing all items emitted by the finite source Publisher,
     * mapped by the keys returned by a specified {@code keySelector} function.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toMap.png" alt="">
     * <p>
     * If more than one source item maps to the same key, the HashMap will contain the latest of those items.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type of the Map
     * @param keySelector
     *            the function that extracts the key from a source item to be used in the HashMap
     * @return a Single that emits a single item: a HashMap containing the mapped items from the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K> Single<Map<K, T>> toMap(final Function<? super T, ? extends K> keySelector) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        return collect(HashMapSupplier.<K, T>asCallable(), Functions.toMapKeySelector(keySelector));
    }

    
Returns a Single that emits a single HashMap containing values corresponding to items emitted by the finite source Publisher, mapped by the keys returned by a specified keySelector function. 



If more than one source item maps to the same key, the HashMap will contain a single entry that
corresponds to the latest of those items.
 Note that this operator requires the upstream to signal onComplete for the accumulated map to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toMap does not operate by default on a particular Scheduler.

Params:
keySelector – 
           the function that extracts the key from a source item to be used in the HashMap
valueSelector – 
           the function that extracts the value from a source item to be used in the HashMap
Type parameters:
<K> – the key type of the Map
<V> – the value type of the Map
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a HashMap containing the mapped items from the source
        Publisher/**
     * Returns a Single that emits a single HashMap containing values corresponding to items emitted by the
     * finite source Publisher, mapped by the keys returned by a specified {@code keySelector} function.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toMap.png" alt="">
     * <p>
     * If more than one source item maps to the same key, the HashMap will contain a single entry that
     * corresponds to the latest of those items.
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type of the Map
     * @param <V> the value type of the Map
     * @param keySelector
     *            the function that extracts the key from a source item to be used in the HashMap
     * @param valueSelector
     *            the function that extracts the value from a source item to be used in the HashMap
     * @return a Single that emits a single item: a HashMap containing the mapped items from the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Single<Map<K, V>> toMap(final Function<? super T, ? extends K> keySelector, final Function<? super T, ? extends V> valueSelector) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        ObjectHelper.requireNonNull(valueSelector, "valueSelector is null");
        return collect(HashMapSupplier.<K, V>asCallable(), Functions.toMapKeyValueSelector(keySelector, valueSelector));
    }

    
Returns a Single that emits a single Map, returned by a specified mapFactory function, that contains keys and values extracted from the items emitted by the finite source Publisher. 


 Note that this operator requires the upstream to signal onComplete for the accumulated map to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toMap does not operate by default on a particular Scheduler.

Params:
keySelector – 
           the function that extracts the key from a source item to be used in the Map
valueSelector – 
           the function that extracts the value from the source items to be used as value in the Map
mapSupplier – 
           the function that returns a Map instance to be used
Type parameters:
<K> – the key type of the Map
<V> – the value type of the Map
See Also:
ReactiveX operators documentation: To
Returns:
a Flowable that emits a single item: a Map that contains the mapped items emitted by the
        source Publisher/**
     * Returns a Single that emits a single Map, returned by a specified {@code mapFactory} function, that
     * contains keys and values extracted from the items emitted by the finite source Publisher.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toMap.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toMap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type of the Map
     * @param <V> the value type of the Map
     * @param keySelector
     *            the function that extracts the key from a source item to be used in the Map
     * @param valueSelector
     *            the function that extracts the value from the source items to be used as value in the Map
     * @param mapSupplier
     *            the function that returns a Map instance to be used
     * @return a Flowable that emits a single item: a Map that contains the mapped items emitted by the
     *         source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Single<Map<K, V>> toMap(final Function<? super T, ? extends K> keySelector,
            final Function<? super T, ? extends V> valueSelector,
            final Callable<? extends Map<K, V>> mapSupplier) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        ObjectHelper.requireNonNull(valueSelector, "valueSelector is null");
        return collect(mapSupplier, Functions.toMapKeyValueSelector(keySelector, valueSelector));
    }

    
Returns a Single that emits a single HashMap that contains an ArrayList of items emitted by the finite source Publisher keyed by a specified keySelector function. 


 Note that this operator requires the upstream to signal onComplete for the accumulated map to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
This operator does not support backpressure as by intent it is requesting and buffering everything.
 
Scheduler:
 
toMultimap does not operate by default on a particular Scheduler.

Params:
keySelector – 
           the function that extracts the key from the source items to be used as key in the HashMap
Type parameters:
<K> – the key type of the Map
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a HashMap that contains an ArrayList of items mapped from
        the source Publisher/**
     * Returns a Single that emits a single HashMap that contains an ArrayList of items emitted by the
     * finite source Publisher keyed by a specified {@code keySelector} function.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toMultiMap.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator does not support backpressure as by intent it is requesting and buffering everything.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toMultimap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type of the Map
     * @param keySelector
     *            the function that extracts the key from the source items to be used as key in the HashMap
     * @return a Single that emits a single item: a HashMap that contains an ArrayList of items mapped from
     *         the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K> Single<Map<K, Collection<T>>> toMultimap(Function<? super T, ? extends K> keySelector) {
        Function<T, T> valueSelector = Functions.identity();
        Callable<Map<K, Collection<T>>> mapSupplier = HashMapSupplier.asCallable();
        Function<K, List<T>> collectionFactory = ArrayListSupplier.asFunction();
        return toMultimap(keySelector, valueSelector, mapSupplier, collectionFactory);
    }

    
Returns a Single that emits a single HashMap that contains an ArrayList of values extracted by a specified valueSelector function from items emitted by the finite source Publisher, keyed by a specified keySelector function. 


 Note that this operator requires the upstream to signal onComplete for the accumulated map to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toMultimap does not operate by default on a particular Scheduler.

Params:
keySelector – 
           the function that extracts a key from the source items to be used as key in the HashMap
valueSelector – 
           the function that extracts a value from the source items to be used as value in the HashMap
Type parameters:
<K> – the key type of the Map
<V> – the value type of the Map
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a HashMap that contains an ArrayList of items mapped from
        the source Publisher/**
     * Returns a Single that emits a single HashMap that contains an ArrayList of values extracted by a
     * specified {@code valueSelector} function from items emitted by the finite source Publisher, keyed by a
     * specified {@code keySelector} function.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toMultiMap.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toMultimap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type of the Map
     * @param <V> the value type of the Map
     * @param keySelector
     *            the function that extracts a key from the source items to be used as key in the HashMap
     * @param valueSelector
     *            the function that extracts a value from the source items to be used as value in the HashMap
     * @return a Single that emits a single item: a HashMap that contains an ArrayList of items mapped from
     *         the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Single<Map<K, Collection<V>>> toMultimap(Function<? super T, ? extends K> keySelector, Function<? super T, ? extends V> valueSelector) {
        Callable<Map<K, Collection<V>>> mapSupplier = HashMapSupplier.asCallable();
        Function<K, List<V>> collectionFactory = ArrayListSupplier.asFunction();
        return toMultimap(keySelector, valueSelector, mapSupplier, collectionFactory);
    }

    
Returns a Single that emits a single Map, returned by a specified mapFactory function, that contains a custom collection of values, extracted by a specified valueSelector function from items emitted by the finite source Publisher, and keyed by the keySelector function. 


 Note that this operator requires the upstream to signal onComplete for the accumulated map to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toMultimap does not operate by default on a particular Scheduler.

Params:
keySelector – 
           the function that extracts a key from the source items to be used as the key in the Map
valueSelector – 
           the function that extracts a value from the source items to be used as the value in the Map
mapSupplier – 
           the function that returns a Map instance to be used
collectionFactory – 
           the function that returns a Collection instance for a particular key to be used in the Map
Type parameters:
<K> – the key type of the Map
<V> – the value type of the Map
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a Map that contains the collection of mapped items from
        the source Publisher/**
     * Returns a Single that emits a single Map, returned by a specified {@code mapFactory} function, that
     * contains a custom collection of values, extracted by a specified {@code valueSelector} function from
     * items emitted by the finite source Publisher, and keyed by the {@code keySelector} function.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toMultiMap.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toMultimap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type of the Map
     * @param <V> the value type of the Map
     * @param keySelector
     *            the function that extracts a key from the source items to be used as the key in the Map
     * @param valueSelector
     *            the function that extracts a value from the source items to be used as the value in the Map
     * @param mapSupplier
     *            the function that returns a Map instance to be used
     * @param collectionFactory
     *            the function that returns a Collection instance for a particular key to be used in the Map
     * @return a Single that emits a single item: a Map that contains the collection of mapped items from
     *         the source Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Single<Map<K, Collection<V>>> toMultimap(
            final Function<? super T, ? extends K> keySelector,
            final Function<? super T, ? extends V> valueSelector,
            final Callable<? extends Map<K, Collection<V>>> mapSupplier,
            final Function<? super K, ? extends Collection<? super V>> collectionFactory) {
        ObjectHelper.requireNonNull(keySelector, "keySelector is null");
        ObjectHelper.requireNonNull(valueSelector, "valueSelector is null");
        ObjectHelper.requireNonNull(mapSupplier, "mapSupplier is null");
        ObjectHelper.requireNonNull(collectionFactory, "collectionFactory is null");
        return collect(mapSupplier, Functions.toMultimapKeyValueSelector(keySelector, valueSelector, collectionFactory));
    }

    
Returns a Single that emits a single Map, returned by a specified mapFactory function, that contains an ArrayList of values, extracted by a specified valueSelector function from items emitted by the finite source Publisher and keyed by the keySelector function. 


 Note that this operator requires the upstream to signal onComplete for the accumulated map to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toMultimap does not operate by default on a particular Scheduler.

Params:
keySelector – 
           the function that extracts a key from the source items to be used as the key in the Map
valueSelector – 
           the function that extracts a value from the source items to be used as the value in the Map
mapSupplier – 
           the function that returns a Map instance to be used
Type parameters:
<K> – the key type of the Map
<V> – the value type of the Map
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a single item: a Map that contains a list items mapped from the source
        Publisher/**
     * Returns a Single that emits a single Map, returned by a specified {@code mapFactory} function, that
     * contains an ArrayList of values, extracted by a specified {@code valueSelector} function from items
     * emitted by the finite source Publisher and keyed by the {@code keySelector} function.
     * <p>
     * <img width="640" height="305" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toMultiMap.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated map to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toMultimap} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <K> the key type of the Map
     * @param <V> the value type of the Map
     * @param keySelector
     *            the function that extracts a key from the source items to be used as the key in the Map
     * @param valueSelector
     *            the function that extracts a value from the source items to be used as the value in the Map
     * @param mapSupplier
     *            the function that returns a Map instance to be used
     * @return a Single that emits a single item: a Map that contains a list items mapped from the source
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <K, V> Single<Map<K, Collection<V>>> toMultimap(
            Function<? super T, ? extends K> keySelector,
            Function<? super T, ? extends V> valueSelector,
            Callable<Map<K, Collection<V>>> mapSupplier
            ) {
        return toMultimap(keySelector, valueSelector, mapSupplier, ArrayListSupplier.<V, K>asFunction());
    }

    
Converts the current Flowable into a non-backpressured Observable. 

 
Backpressure:
 
Observables don't support backpressure thus the current Flowable is consumed in an unbounded
 manner (by requesting Long.MAX_VALUE).
 
Scheduler:
 
toObservable does not operate by default on a particular Scheduler.

Returns:
the new Observable instance
Since:
2.0/**
     * Converts the current Flowable into a non-backpressured {@link Observable}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>Observables don't support backpressure thus the current Flowable is consumed in an unbounded
     *  manner (by requesting Long.MAX_VALUE).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toObservable} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @return the new Observable instance
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Observable<T> toObservable() {
        return RxJavaPlugins.onAssembly(new ObservableFromPublisher<T>(this));
    }

    
Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a sorted order. Each item emitted by the Publisher must implement Comparable with respect to all other items in the sequence. 
If any item emitted by this Flowable does not implement Comparable with respect to all other items emitted by this Flowable, no items will be emitted and the sequence is terminated with a ClassCastException. 


 Note that this operator requires the upstream to signal onComplete for the accumulated list to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toSortedList does not operate by default on a particular Scheduler.

See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a list that contains the items emitted by the source Publisher in
        sorted order/**
     * Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a
     * sorted order. Each item emitted by the Publisher must implement {@link Comparable} with respect to all
     * other items in the sequence.
     *
     * <p>If any item emitted by this Flowable does not implement {@link Comparable} with respect to
     *             all other items emitted by this Flowable, no items will be emitted and the
     *             sequence is terminated with a {@link ClassCastException}.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toSortedList.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toSortedList} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @return a Single that emits a list that contains the items emitted by the source Publisher in
     *         sorted order
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<List<T>> toSortedList() {
        return toSortedList(Functions.naturalComparator());
    }

    
Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a
sorted order based on a specified comparison function.


 Note that this operator requires the upstream to signal onComplete for the accumulated list to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toSortedList does not operate by default on a particular Scheduler.

Params:
comparator – 
           a function that compares two items emitted by the source Publisher and returns an Integer
           that indicates their sort order
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a list that contains the items emitted by the source Publisher in
        sorted order/**
     * Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a
     * sorted order based on a specified comparison function.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toSortedList.f.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toSortedList} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param comparator
     *            a function that compares two items emitted by the source Publisher and returns an Integer
     *            that indicates their sort order
     * @return a Single that emits a list that contains the items emitted by the source Publisher in
     *         sorted order
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<List<T>> toSortedList(final Comparator<? super T> comparator) {
        ObjectHelper.requireNonNull(comparator, "comparator is null");
        return toList().map(Functions.listSorter(comparator));
    }

    
Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a
sorted order based on a specified comparison function.


 Note that this operator requires the upstream to signal onComplete for the accumulated list to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toSortedList does not operate by default on a particular Scheduler.

Params:
comparator – 
           a function that compares two items emitted by the source Publisher and returns an Integer
           that indicates their sort order
capacityHint – 
            the initial capacity of the ArrayList used to accumulate items before sorting
See Also:
ReactiveX operators documentation: To
Returns:
a Single that emits a list that contains the items emitted by the source Publisher in
        sorted order
Since:
2.0/**
     * Returns a Single that emits a list that contains the items emitted by the finite source Publisher, in a
     * sorted order based on a specified comparison function.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toSortedList.f.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toSortedList} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param comparator
     *            a function that compares two items emitted by the source Publisher and returns an Integer
     *            that indicates their sort order
     * @param capacityHint
     *             the initial capacity of the ArrayList used to accumulate items before sorting
     * @return a Single that emits a list that contains the items emitted by the source Publisher in
     *         sorted order
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<List<T>> toSortedList(final Comparator<? super T> comparator, int capacityHint) {
        ObjectHelper.requireNonNull(comparator, "comparator is null");
        return toList(capacityHint).map(Functions.listSorter(comparator));
    }

    
Returns a Flowable that emits a list that contains the items emitted by the finite source Publisher, in a sorted order. Each item emitted by the Publisher must implement Comparable with respect to all other items in the sequence. 
If any item emitted by this Flowable does not implement Comparable with respect to all other items emitted by this Flowable, no items will be emitted and the sequence is terminated with a ClassCastException. 


 Note that this operator requires the upstream to signal onComplete for the accumulated list to be emitted. Sources that are infinite and never complete will never emit anything through this operator and an infinite source may lead to a fatal OutOfMemoryError. 

 
Backpressure:
 
The operator honors backpressure from downstream and consumes the source Publisher in an unbounded manner (i.e., without applying backpressure to it).
 
Scheduler:
 
toSortedList does not operate by default on a particular Scheduler.

Params:
capacityHint – 
            the initial capacity of the ArrayList used to accumulate items before sorting
See Also:
ReactiveX operators documentation: To
Returns:
a Flowable that emits a list that contains the items emitted by the source Publisher in
        sorted order
Since:
2.0/**
     * Returns a Flowable that emits a list that contains the items emitted by the finite source Publisher, in a
     * sorted order. Each item emitted by the Publisher must implement {@link Comparable} with respect to all
     * other items in the sequence.
     *
     * <p>If any item emitted by this Flowable does not implement {@link Comparable} with respect to
     *             all other items emitted by this Flowable, no items will be emitted and the
     *             sequence is terminated with a {@link ClassCastException}.
     * <p>
     * <img width="640" height="310" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/toSortedList.png" alt="">
     * <p>
     * Note that this operator requires the upstream to signal {@code onComplete} for the accumulated list to
     * be emitted. Sources that are infinite and never complete will never emit anything through this
     * operator and an infinite source may lead to a fatal {@code OutOfMemoryError}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure from downstream and consumes the source {@code Publisher} in an
     *  unbounded manner (i.e., without applying backpressure to it).</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code toSortedList} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param capacityHint
     *             the initial capacity of the ArrayList used to accumulate items before sorting
     * @return a Flowable that emits a list that contains the items emitted by the source Publisher in
     *         sorted order
     * @see <a href="http://reactivex.io/documentation/operators/to.html">ReactiveX operators documentation: To</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Single<List<T>> toSortedList(int capacityHint) {
        return toSortedList(Functions.naturalComparator(), capacityHint);
    }

    
Modifies the source Publisher so that subscribers will cancel it on a specified Scheduler. 

 
Backpressure:
 
The operator doesn't interfere with backpressure which is determined by the source Publisher's backpressure behavior.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
scheduler –  the Scheduler to perform cancellation actions on
See Also:
ReactiveX operators documentation: SubscribeOn
Returns:
the source Publisher modified so that its cancellations happen on the specified Scheduler/**
     * Modifies the source Publisher so that subscribers will cancel it on a specified
     * {@link Scheduler}.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator doesn't interfere with backpressure which is determined by the source {@code Publisher}'s backpressure
     *  behavior.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param scheduler
     *            the {@link Scheduler} to perform cancellation actions on
     * @return the source Publisher modified so that its cancellations happen on the specified
     *         {@link Scheduler}
     * @see <a href="http://reactivex.io/documentation/operators/subscribeon.html">ReactiveX operators documentation: SubscribeOn</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<T> unsubscribeOn(Scheduler scheduler) {
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        return RxJavaPlugins.onAssembly(new FlowableUnsubscribeOn<T>(this, scheduler));
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each containing count items. When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an unbounded buffer that may hold at most count elements.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum size of each window before it should be emitted
Throws:
IllegalArgumentException – if either count is non-positive
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows, each containing at most count items from the source Publisher/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each containing {@code count} items. When the source
     * Publisher completes or encounters an error, the resulting Publisher emits the current window and
     * propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="400" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window3.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an
     *  unbounded buffer that may hold at most {@code count} elements.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum size of each window before it should be emitted
     * @return a Flowable that emits connected, non-overlapping windows, each containing at most
     *         {@code count} items from the source Publisher
     * @throws IllegalArgumentException if either count is non-positive
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Flowable<T>> window(long count) {
        return window(count, count, bufferSize());
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits windows every skip items, each containing no more than count items. When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an unbounded buffer that may hold at most count elements.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum size of each window before it should be emitted
skip –  how many items need to be skipped before starting a new window. Note that if skip and count are equal this is the same operation as window(long).
Throws:
IllegalArgumentException – if either count or skip is non-positive
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits windows every skip items containing at most count items from the source Publisher/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits windows every {@code skip} items, each containing no more than {@code count} items. When
     * the source Publisher completes or encounters an error, the resulting Publisher emits the current window
     * and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="365" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window4.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an
     *  unbounded buffer that may hold at most {@code count} elements.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum size of each window before it should be emitted
     * @param skip
     *            how many items need to be skipped before starting a new window. Note that if {@code skip} and
     *            {@code count} are equal this is the same operation as {@link #window(long)}.
     * @return a Flowable that emits windows every {@code skip} items containing at most {@code count} items
     *         from the source Publisher
     * @throws IllegalArgumentException if either count or skip is non-positive
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Flowable<T>> window(long count, long skip) {
        return window(count, skip, bufferSize());
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits windows every skip items, each containing no more than count items. When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an unbounded buffer that may hold at most count elements.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
count – 
           the maximum size of each window before it should be emitted
skip –  how many items need to be skipped before starting a new window. Note that if skip and count are equal this is the same operation as window(long).
bufferSize – 
           the capacity hint for the buffer in the inner windows
Throws:
IllegalArgumentException – if either count or skip is non-positive
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits windows every skip items containing at most count items from the source Publisher/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits windows every {@code skip} items, each containing no more than {@code count} items. When
     * the source Publisher completes or encounters an error, the resulting Publisher emits the current window
     * and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="365" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window4.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator honors backpressure of its inner and outer subscribers, however, the inner Publisher uses an
     *  unbounded buffer that may hold at most {@code count} elements.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param count
     *            the maximum size of each window before it should be emitted
     * @param skip
     *            how many items need to be skipped before starting a new window. Note that if {@code skip} and
     *            {@code count} are equal this is the same operation as {@link #window(long)}.
     * @param bufferSize
     *            the capacity hint for the buffer in the inner windows
     * @return a Flowable that emits windows every {@code skip} items containing at most {@code count} items
     *         from the source Publisher
     * @throws IllegalArgumentException if either count or skip is non-positive
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final Flowable<Flowable<T>> window(long count, long skip, int bufferSize) {
        ObjectHelper.verifyPositive(skip, "skip");
        ObjectHelper.verifyPositive(count, "count");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableWindow<T>(this, count, skip, bufferSize));
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher starts a new window periodically, as determined by the timeskip argument. It emits each window after a fixed timespan, specified by the timespan argument. When the source Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure but have an unbounded inner buffer that may lead to OutOfMemoryError if left unconsumed.
 
Scheduler:
 
This version of window operates by default on the computation Scheduler.

Params:
timespan – 
           the period of time each window collects items before it should be emitted
timeskip – 
           the period of time after which a new window will be created
unit –  the unit of time that applies to the timespan and timeskip arguments
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits new windows periodically as a fixed timespan elapses/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher starts a new window periodically, as determined by the {@code timeskip} argument. It emits
     * each window after a fixed timespan, specified by the {@code timespan} argument. When the source
     * Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the
     * current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="335" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window7.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure but have an unbounded inner buffer that <em>may</em> lead to {@code OutOfMemoryError}
     *  if left unconsumed.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted
     * @param timeskip
     *            the period of time after which a new window will be created
     * @param unit
     *            the unit of time that applies to the {@code timespan} and {@code timeskip} arguments
     * @return a Flowable that emits new windows periodically as a fixed timespan elapses
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<Flowable<T>> window(long timespan, long timeskip, TimeUnit unit) {
        return window(timespan, timeskip, unit, Schedulers.computation(), bufferSize());
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher starts a new window periodically, as determined by the timeskip argument. It emits each window after a fixed timespan, specified by the timespan argument. When the source Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure but have an unbounded inner buffer that may lead to OutOfMemoryError if left unconsumed.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each window collects items before it should be emitted
timeskip – 
           the period of time after which a new window will be created
unit –  the unit of time that applies to the timespan and timeskip arguments
scheduler –  the Scheduler to use when determining the end and start of a window
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits new windows periodically as a fixed timespan elapses/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher starts a new window periodically, as determined by the {@code timeskip} argument. It emits
     * each window after a fixed timespan, specified by the {@code timespan} argument. When the source
     * Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the
     * current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="335" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window7.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure but have an unbounded inner buffer that <em>may</em> lead to {@code OutOfMemoryError}
     *  if left unconsumed.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted
     * @param timeskip
     *            the period of time after which a new window will be created
     * @param unit
     *            the unit of time that applies to the {@code timespan} and {@code timeskip} arguments
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a window
     * @return a Flowable that emits new windows periodically as a fixed timespan elapses
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<Flowable<T>> window(long timespan, long timeskip, TimeUnit unit, Scheduler scheduler) {
        return window(timespan, timeskip, unit, scheduler, bufferSize());
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher starts a new window periodically, as determined by the timeskip argument. It emits each window after a fixed timespan, specified by the timespan argument. When the source Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure but have an unbounded inner buffer that may lead to OutOfMemoryError if left unconsumed.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each window collects items before it should be emitted
timeskip – 
           the period of time after which a new window will be created
unit –  the unit of time that applies to the timespan and timeskip arguments
scheduler –  the Scheduler to use when determining the end and start of a window
bufferSize – 
           the capacity hint for the buffer in the inner windows
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits new windows periodically as a fixed timespan elapses/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher starts a new window periodically, as determined by the {@code timeskip} argument. It emits
     * each window after a fixed timespan, specified by the {@code timespan} argument. When the source
     * Publisher completes or Publisher completes or encounters an error, the resulting Publisher emits the
     * current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="335" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window7.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure but have an unbounded inner buffer that <em>may</em> lead to {@code OutOfMemoryError}
     *  if left unconsumed.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted
     * @param timeskip
     *            the period of time after which a new window will be created
     * @param unit
     *            the unit of time that applies to the {@code timespan} and {@code timeskip} arguments
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a window
     * @param bufferSize
     *            the capacity hint for the buffer in the inner windows
     * @return a Flowable that emits new windows periodically as a fixed timespan elapses
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<Flowable<T>> window(long timespan, long timeskip, TimeUnit unit, Scheduler scheduler, int bufferSize) {
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        ObjectHelper.verifyPositive(timespan, "timespan");
        ObjectHelper.verifyPositive(timeskip, "timeskip");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.requireNonNull(unit, "unit is null");
        return RxJavaPlugins.onAssembly(new FlowableWindowTimed<T>(this, timespan, timeskip, unit, scheduler, Long.MAX_VALUE, bufferSize, false));
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the timespan argument. When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure and may hold up to count elements at most.
 
Scheduler:
 
This version of window operates by default on the computation Scheduler.

Params:
timespan – 
           the period of time each window collects items before it should be emitted and replaced with a
           new window
unit –  the unit of time that applies to the timespan argument
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows representing items emitted by the
        source Publisher during fixed, consecutive durations/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the
     * {@code timespan} argument. When the source Publisher completes or encounters an error, the resulting
     * Publisher emits the current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="375" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window5.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure and may hold up to {@code count} elements at most.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted and replaced with a
     *            new window
     * @param unit
     *            the unit of time that applies to the {@code timespan} argument
     * @return a Flowable that emits connected, non-overlapping windows representing items emitted by the
     *         source Publisher during fixed, consecutive durations
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<Flowable<T>> window(long timespan, TimeUnit unit) {
        return window(timespan, unit, Schedulers.computation(), Long.MAX_VALUE, false);
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the timespan argument or a maximum size as specified by the count argument (whichever is reached first). When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure and may hold up to count elements at most.
 
Scheduler:
 
This version of window operates by default on the computation Scheduler.

Params:
timespan – 
           the period of time each window collects items before it should be emitted and replaced with a
           new window
unit –  the unit of time that applies to the timespan argument
count – 
           the maximum size of each window before it should be emitted
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows of items from the source Publisher
        that were emitted during a fixed duration of time or when the window has reached maximum capacity
        (whichever occurs first)/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the
     * {@code timespan} argument or a maximum size as specified by the {@code count} argument (whichever is
     * reached first). When the source Publisher completes or encounters an error, the resulting Publisher
     * emits the current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window6.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure and may hold up to {@code count} elements at most.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted and replaced with a
     *            new window
     * @param unit
     *            the unit of time that applies to the {@code timespan} argument
     * @param count
     *            the maximum size of each window before it should be emitted
     * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher
     *         that were emitted during a fixed duration of time or when the window has reached maximum capacity
     *         (whichever occurs first)
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<Flowable<T>> window(long timespan, TimeUnit unit,
            long count) {
        return window(timespan, unit, Schedulers.computation(), count, false);
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the timespan argument or a maximum size as specified by the count argument (whichever is reached first). When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure and may hold up to count elements at most.
 
Scheduler:
 
This version of window operates by default on the computation Scheduler.

Params:
timespan – 
           the period of time each window collects items before it should be emitted and replaced with a
           new window
unit –  the unit of time that applies to the timespan argument
count – 
           the maximum size of each window before it should be emitted
restart – 
           if true, when a window reaches the capacity limit, the timer is restarted as well
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows of items from the source Publisher
        that were emitted during a fixed duration of time or when the window has reached maximum capacity
        (whichever occurs first)/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the
     * {@code timespan} argument or a maximum size as specified by the {@code count} argument (whichever is
     * reached first). When the source Publisher completes or encounters an error, the resulting Publisher
     * emits the current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window6.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure and may hold up to {@code count} elements at most.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} operates by default on the {@code computation} {@link Scheduler}.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted and replaced with a
     *            new window
     * @param unit
     *            the unit of time that applies to the {@code timespan} argument
     * @param count
     *            the maximum size of each window before it should be emitted
     * @param restart
     *            if true, when a window reaches the capacity limit, the timer is restarted as well
     * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher
     *         that were emitted during a fixed duration of time or when the window has reached maximum capacity
     *         (whichever occurs first)
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.COMPUTATION)
    public final Flowable<Flowable<T>> window(long timespan, TimeUnit unit,
            long count, boolean restart) {
        return window(timespan, unit, Schedulers.computation(), count, restart);
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the timespan argument. When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure but have an unbounded inner buffer that may lead to OutOfMemoryError if left unconsumed.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each window collects items before it should be emitted and replaced with a
           new window
unit –  the unit of time which applies to the timespan argument
scheduler –  the Scheduler to use when determining the end and start of a window
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows containing items emitted by the
        source Publisher within a fixed duration/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each of a fixed duration as specified by the
     * {@code timespan} argument. When the source Publisher completes or encounters an error, the resulting
     * Publisher emits the current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="375" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window5.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure but have an unbounded inner buffer that <em>may</em> lead to {@code OutOfMemoryError}
     *  if left unconsumed.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted and replaced with a
     *            new window
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a window
     * @return a Flowable that emits connected, non-overlapping windows containing items emitted by the
     *         source Publisher within a fixed duration
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<Flowable<T>> window(long timespan, TimeUnit unit,
            Scheduler scheduler) {
        return window(timespan, unit, scheduler, Long.MAX_VALUE, false);
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the timespan argument or a maximum size specified by the count argument (whichever is reached first). When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure and may hold up to count elements at most.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each window collects items before it should be emitted and replaced with a
           new window
unit –  the unit of time which applies to the timespan argument
count – 
           the maximum size of each window before it should be emitted
scheduler –  the Scheduler to use when determining the end and start of a window
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows of items from the source Publisher
        that were emitted during a fixed duration of time or when the window has reached maximum capacity
        (whichever occurs first)/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the
     * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached
     * first). When the source Publisher completes or encounters an error, the resulting Publisher emits the
     * current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window6.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure and may hold up to {@code count} elements at most.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted and replaced with a
     *            new window
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param count
     *            the maximum size of each window before it should be emitted
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a window
     * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher
     *         that were emitted during a fixed duration of time or when the window has reached maximum capacity
     *         (whichever occurs first)
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<Flowable<T>> window(long timespan, TimeUnit unit,
            Scheduler scheduler, long count) {
        return window(timespan, unit, scheduler, count, false);
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the timespan argument or a maximum size specified by the count argument (whichever is reached first). When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure and may hold up to count elements at most.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each window collects items before it should be emitted and replaced with a
           new window
unit –  the unit of time which applies to the timespan argument
count – 
           the maximum size of each window before it should be emitted
scheduler –  the Scheduler to use when determining the end and start of a window
restart – 
           if true, when a window reaches the capacity limit, the timer is restarted as well
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows of items from the source Publisher
        that were emitted during a fixed duration of time or when the window has reached maximum capacity
        (whichever occurs first)/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the
     * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached
     * first). When the source Publisher completes or encounters an error, the resulting Publisher emits the
     * current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window6.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure and may hold up to {@code count} elements at most.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted and replaced with a
     *            new window
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param count
     *            the maximum size of each window before it should be emitted
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a window
     * @param restart
     *            if true, when a window reaches the capacity limit, the timer is restarted as well
     * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher
     *         that were emitted during a fixed duration of time or when the window has reached maximum capacity
     *         (whichever occurs first)
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<Flowable<T>> window(long timespan, TimeUnit unit,
            Scheduler scheduler, long count, boolean restart) {
        return window(timespan, unit, scheduler, count, restart, bufferSize());
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the timespan argument or a maximum size specified by the count argument (whichever is reached first). When the source Publisher completes or encounters an error, the resulting Publisher emits the current window and propagates the notification from the source Publisher. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses time to control the creation of windows. The returned inner Publishers honor backpressure and may hold up to count elements at most.
 
Scheduler:
 
You specify which Scheduler this operator will use.

Params:
timespan – 
           the period of time each window collects items before it should be emitted and replaced with a
           new window
unit –  the unit of time which applies to the timespan argument
count – 
           the maximum size of each window before it should be emitted
scheduler –  the Scheduler to use when determining the end and start of a window
restart – 
           if true, when a window reaches the capacity limit, the timer is restarted as well
bufferSize – 
           the capacity hint for the buffer in the inner windows
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows of items from the source Publisher
        that were emitted during a fixed duration of time or when the window has reached maximum capacity
        (whichever occurs first)/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows, each of a fixed duration specified by the
     * {@code timespan} argument or a maximum size specified by the {@code count} argument (whichever is reached
     * first). When the source Publisher completes or encounters an error, the resulting Publisher emits the
     * current window and propagates the notification from the source Publisher.
     * <p>
     * <img width="640" height="370" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window6.s.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  time to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure and may hold up to {@code count} elements at most.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>You specify which {@link Scheduler} this operator will use.</dd>
     * </dl>
     *
     * @param timespan
     *            the period of time each window collects items before it should be emitted and replaced with a
     *            new window
     * @param unit
     *            the unit of time which applies to the {@code timespan} argument
     * @param count
     *            the maximum size of each window before it should be emitted
     * @param scheduler
     *            the {@link Scheduler} to use when determining the end and start of a window
     * @param restart
     *            if true, when a window reaches the capacity limit, the timer is restarted as well
     * @param bufferSize
     *            the capacity hint for the buffer in the inner windows
     * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher
     *         that were emitted during a fixed duration of time or when the window has reached maximum capacity
     *         (whichever occurs first)
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.CUSTOM)
    public final Flowable<Flowable<T>> window(
            long timespan, TimeUnit unit, Scheduler scheduler,
            long count, boolean restart, int bufferSize) {
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        ObjectHelper.requireNonNull(scheduler, "scheduler is null");
        ObjectHelper.requireNonNull(unit, "unit is null");
        ObjectHelper.verifyPositive(count, "count");
        return RxJavaPlugins.onAssembly(new FlowableWindowTimed<T>(this, timespan, timespan, unit, scheduler, count, bufferSize, restart));
    }

    
Returns a Flowable that emits non-overlapping windows of items it collects from the source Publisher
where the boundary of each window is determined by the items emitted from a specified boundary-governing
Publisher.



 
Backpressure:
 
The outer Publisher of this operator does not support backpressure as it uses a boundary Publisher to control data flow. The inner Publishers honor backpressure and buffer everything until the boundary signals the next element.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
boundaryIndicator – 
           a Publisher whose emitted items close and open windows
Type parameters:
<B> – 
           the window element type (ignored)
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits non-overlapping windows of items it collects from the source Publisher where the boundary of each window is determined by the items emitted from the boundary Publisher/**
     * Returns a Flowable that emits non-overlapping windows of items it collects from the source Publisher
     * where the boundary of each window is determined by the items emitted from a specified boundary-governing
     * Publisher.
     * <p>
     * <img width="640" height="475" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window8.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The outer Publisher of this operator does not support backpressure as it uses a {@code boundary} Publisher to control data
     *      flow. The inner Publishers honor backpressure and buffer everything until the boundary signals the next element.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <B>
     *            the window element type (ignored)
     * @param boundaryIndicator
     *            a Publisher whose emitted items close and open windows
     * @return a Flowable that emits non-overlapping windows of items it collects from the source Publisher
     *         where the boundary of each window is determined by the items emitted from the {@code boundary}
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B> Flowable<Flowable<T>> window(Publisher<B> boundaryIndicator) {
        return window(boundaryIndicator, bufferSize());
    }

    
Returns a Flowable that emits non-overlapping windows of items it collects from the source Publisher
where the boundary of each window is determined by the items emitted from a specified boundary-governing
Publisher.



 
Backpressure:
 
The outer Publisher of this operator does not support backpressure as it uses a boundary Publisher to control data flow. The inner Publishers honor backpressure and buffer everything until the boundary signals the next element.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
boundaryIndicator – 
           a Publisher whose emitted items close and open windows
bufferSize – 
           the capacity hint for the buffer in the inner windows
Type parameters:
<B> – 
           the window element type (ignored)
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits non-overlapping windows of items it collects from the source Publisher where the boundary of each window is determined by the items emitted from the boundary Publisher/**
     * Returns a Flowable that emits non-overlapping windows of items it collects from the source Publisher
     * where the boundary of each window is determined by the items emitted from a specified boundary-governing
     * Publisher.
     * <p>
     * <img width="640" height="475" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window8.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The outer Publisher of this operator does not support backpressure as it uses a {@code boundary} Publisher to control data
     *      flow. The inner Publishers honor backpressure and buffer everything until the boundary signals the next element.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <B>
     *            the window element type (ignored)
     * @param boundaryIndicator
     *            a Publisher whose emitted items close and open windows
     * @param bufferSize
     *            the capacity hint for the buffer in the inner windows
     * @return a Flowable that emits non-overlapping windows of items it collects from the source Publisher
     *         where the boundary of each window is determined by the items emitted from the {@code boundary}
     *         Publisher
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B> Flowable<Flowable<T>> window(Publisher<B> boundaryIndicator, int bufferSize) {
        ObjectHelper.requireNonNull(boundaryIndicator, "boundaryIndicator is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableWindowBoundary<T, B>(this, boundaryIndicator, bufferSize));
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits windows that contain those items emitted by the source Publisher between the time when the windowOpenings Publisher emits an item and when the Publisher returned by closingSelector emits an item. 



 
Backpressure:
 
The outer Publisher of this operator doesn't support backpressure because the emission of new inner Publishers are controlled by the windowOpenings Publisher. The inner Publishers honor backpressure and buffer everything until the associated closing Publisher signals or completes.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
openingIndicator – 
           a Publisher that, when it emits an item, causes another window to be created
closingIndicator –  a Function that produces a Publisher for every window created. When this Publisher emits an item, the associated window is closed and emitted
Type parameters:
<U> – the element type of the window-opening Publisher
<V> – the element type of the window-closing Publishers
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits windows of items emitted by the source Publisher that are governed by
        the specified window-governing Publishers/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits windows that contain those items emitted by the source Publisher between the time when
     * the {@code windowOpenings} Publisher emits an item and when the Publisher returned by
     * {@code closingSelector} emits an item.
     * <p>
     * <img width="640" height="550" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window2.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The outer Publisher of this operator doesn't support backpressure because the emission of new
     *  inner Publishers are controlled by the {@code windowOpenings} Publisher.
     *  The inner Publishers honor backpressure and buffer everything until the associated closing
     *  Publisher signals or completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the element type of the window-opening Publisher
     * @param <V> the element type of the window-closing Publishers
     * @param openingIndicator
     *            a Publisher that, when it emits an item, causes another window to be created
     * @param closingIndicator
     *            a {@link Function} that produces a Publisher for every window created. When this Publisher
     *            emits an item, the associated window is closed and emitted
     * @return a Flowable that emits windows of items emitted by the source Publisher that are governed by
     *         the specified window-governing Publishers
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, V> Flowable<Flowable<T>> window(
            Publisher<U> openingIndicator,
            Function<? super U, ? extends Publisher<V>> closingIndicator) {
        return window(openingIndicator, closingIndicator, bufferSize());
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits windows that contain those items emitted by the source Publisher between the time when the windowOpenings Publisher emits an item and when the Publisher returned by closingSelector emits an item. 



 
Backpressure:
 
The outer Publisher of this operator doesn't support backpressure because the emission of new inner Publishers are controlled by the windowOpenings Publisher. The inner Publishers honor backpressure and buffer everything until the associated closing Publisher signals or completes.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
openingIndicator – 
           a Publisher that, when it emits an item, causes another window to be created
closingIndicator –  a Function that produces a Publisher for every window created. When this Publisher emits an item, the associated window is closed and emitted
bufferSize – 
           the capacity hint for the buffer in the inner windows
Type parameters:
<U> – the element type of the window-opening Publisher
<V> – the element type of the window-closing Publishers
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits windows of items emitted by the source Publisher that are governed by
        the specified window-governing Publishers/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits windows that contain those items emitted by the source Publisher between the time when
     * the {@code windowOpenings} Publisher emits an item and when the Publisher returned by
     * {@code closingSelector} emits an item.
     * <p>
     * <img width="640" height="550" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window2.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The outer Publisher of this operator doesn't support backpressure because the emission of new
     *  inner Publishers are controlled by the {@code windowOpenings} Publisher.
     *  The inner Publishers honor backpressure and buffer everything until the associated closing
     *  Publisher signals or completes.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the element type of the window-opening Publisher
     * @param <V> the element type of the window-closing Publishers
     * @param openingIndicator
     *            a Publisher that, when it emits an item, causes another window to be created
     * @param closingIndicator
     *            a {@link Function} that produces a Publisher for every window created. When this Publisher
     *            emits an item, the associated window is closed and emitted
     * @param bufferSize
     *            the capacity hint for the buffer in the inner windows
     * @return a Flowable that emits windows of items emitted by the source Publisher that are governed by
     *         the specified window-governing Publishers
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, V> Flowable<Flowable<T>> window(
            Publisher<U> openingIndicator,
            Function<? super U, ? extends Publisher<V>> closingIndicator, int bufferSize) {
        ObjectHelper.requireNonNull(openingIndicator, "openingIndicator is null");
        ObjectHelper.requireNonNull(closingIndicator, "closingIndicator is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableWindowBoundarySelector<T, U, V>(this, openingIndicator, closingIndicator, bufferSize));
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows. It emits the current window and opens a new one whenever the Publisher produced by the specified closingSelector emits an item. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses the closingSelector to control the creation of windows. The returned inner Publishers honor backpressure but have an unbounded inner buffer that may lead to OutOfMemoryError if left unconsumed.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
boundaryIndicatorSupplier –  a Callable that returns a Publisher that governs the boundary between windows. When the source Publisher emits an item, window emits the current window and begins a new one.
Type parameters:
<B> – the element type of the boundary Publisher
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows of items from the source Publisher whenever closingSelector emits an item/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows. It emits the current window and opens a new one
     * whenever the Publisher produced by the specified {@code closingSelector} emits an item.
     * <p>
     * <img width="640" height="455" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window1.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  the {@code closingSelector} to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure but have an unbounded inner buffer that <em>may</em> lead to {@code OutOfMemoryError}
     *  if left unconsumed.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <B> the element type of the boundary Publisher
     * @param boundaryIndicatorSupplier
     *            a {@link Callable} that returns a {@code Publisher} that governs the boundary between windows.
     *            When the source {@code Publisher} emits an item, {@code window} emits the current window and begins
     *            a new one.
     * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher
     *         whenever {@code closingSelector} emits an item
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B> Flowable<Flowable<T>> window(Callable<? extends Publisher<B>> boundaryIndicatorSupplier) {
        return window(boundaryIndicatorSupplier, bufferSize());
    }

    
Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting Publisher emits connected, non-overlapping windows. It emits the current window and opens a new one whenever the Publisher produced by the specified closingSelector emits an item. 



 
Backpressure:
 
The operator consumes the source Publisher in an unbounded manner. The returned Publisher doesn't support backpressure as it uses the closingSelector to control the creation of windows. The returned inner Publishers honor backpressure but have an unbounded inner buffer that may lead to OutOfMemoryError if left unconsumed.
 
Scheduler:
 
This version of window does not operate by default on a particular Scheduler.

Params:
boundaryIndicatorSupplier –  a Callable that returns a Publisher that governs the boundary between windows. When the source Publisher emits an item, window emits the current window and begins a new one.
bufferSize – 
           the capacity hint for the buffer in the inner windows
Type parameters:
<B> – the element type of the boundary Publisher
See Also:
ReactiveX operators documentation: Window
Returns:
a Flowable that emits connected, non-overlapping windows of items from the source Publisher whenever closingSelector emits an item/**
     * Returns a Flowable that emits windows of items it collects from the source Publisher. The resulting
     * Publisher emits connected, non-overlapping windows. It emits the current window and opens a new one
     * whenever the Publisher produced by the specified {@code closingSelector} emits an item.
     * <p>
     * <img width="640" height="455" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/window1.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator consumes the source {@code Publisher} in an unbounded manner.
     *  The returned {@code Publisher} doesn't support backpressure as it uses
     *  the {@code closingSelector} to control the creation of windows. The returned inner {@code Publisher}s honor
     *  backpressure but have an unbounded inner buffer that <em>may</em> lead to {@code OutOfMemoryError}
     *  if left unconsumed.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This version of {@code window} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <B> the element type of the boundary Publisher
     * @param boundaryIndicatorSupplier
     *            a {@link Callable} that returns a {@code Publisher} that governs the boundary between windows.
     *            When the source {@code Publisher} emits an item, {@code window} emits the current window and begins
     *            a new one.
     * @param bufferSize
     *            the capacity hint for the buffer in the inner windows
     * @return a Flowable that emits connected, non-overlapping windows of items from the source Publisher
     *         whenever {@code closingSelector} emits an item
     * @see <a href="http://reactivex.io/documentation/operators/window.html">ReactiveX operators documentation: Window</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.ERROR)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <B> Flowable<Flowable<T>> window(Callable<? extends Publisher<B>> boundaryIndicatorSupplier, int bufferSize) {
        ObjectHelper.requireNonNull(boundaryIndicatorSupplier, "boundaryIndicatorSupplier is null");
        ObjectHelper.verifyPositive(bufferSize, "bufferSize");
        return RxJavaPlugins.onAssembly(new FlowableWindowBoundarySupplier<T, B>(this, boundaryIndicatorSupplier, bufferSize));
    }

    
Merges the specified Publisher into this Publisher sequence by using the resultSelector function only when the source Publisher (this instance) emits an item. 



 
Backpressure:
 
The operator is a pass-through for backpressure: the backpressure support
 depends on the upstream and downstream's backpressure behavior. The other Publisher
 is consumed in an unbounded fashion.
 
Scheduler:
 
This operator, by default, doesn't run any particular Scheduler.

Params:
other – 
           the other Publisher
combiner – 
           the function to call when this Publisher emits an item and the other Publisher has already
           emitted an item, to generate the item to be emitted by the resulting Publisher
Type parameters:
<U> – the element type of the other Publisher
<R> – the result type of the combination
See Also:
ReactiveX operators documentation: CombineLatest
Returns:
a Flowable that merges the specified Publisher into this Publisher by using the resultSelector function only when the source Publisher sequence (this instance) emits an item
Since:
2.0/**
     * Merges the specified Publisher into this Publisher sequence by using the {@code resultSelector}
     * function only when the source Publisher (this instance) emits an item.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/withLatestFrom.png" alt="">
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator is a pass-through for backpressure: the backpressure support
     *  depends on the upstream and downstream's backpressure behavior. The other Publisher
     *  is consumed in an unbounded fashion.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator, by default, doesn't run any particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U> the element type of the other Publisher
     * @param <R> the result type of the combination
     * @param other
     *            the other Publisher
     * @param combiner
     *            the function to call when this Publisher emits an item and the other Publisher has already
     *            emitted an item, to generate the item to be emitted by the resulting Publisher
     * @return a Flowable that merges the specified Publisher into this Publisher by using the
     *         {@code resultSelector} function only when the source Publisher sequence (this instance) emits an
     *         item
     * @since 2.0
     * @see <a href="http://reactivex.io/documentation/operators/combinelatest.html">ReactiveX operators documentation: CombineLatest</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> withLatestFrom(Publisher<? extends U> other,
            BiFunction<? super T, ? super U, ? extends R> combiner) {
        ObjectHelper.requireNonNull(other, "other is null");
        ObjectHelper.requireNonNull(combiner, "combiner is null");

        return RxJavaPlugins.onAssembly(new FlowableWithLatestFrom<T, U, R>(this, combiner, other));
    }

    
Combines the value emission from this Publisher with the latest emissions from the
other Publishers via a function to produce the output item.
Note that this operator doesn't emit anything until all other sources have produced at
least one value. The resulting emission only happens when this Publisher emits (and
not when any of the other sources emit, unlike combineLatest).
If a source doesn't produce any value and just completes, the sequence is completed immediately.

 
Backpressure:
 
This operator is a pass-through for backpressure behavior between the source Publisher and the downstream Subscriber. The other Publishers are consumed in an unbounded manner.
 
Scheduler:
 
This operator does not operate by default on a particular Scheduler.

Params:
source1 – the first other Publisher
source2 – the second other Publisher
combiner – the function called with an array of values from each participating Publisher
Type parameters:
<T1> – the first other source's value type
<T2> – the second other source's value type
<R> – the result value type
Returns:
the new Publisher instance
Since:
2.0/**
     * Combines the value emission from this Publisher with the latest emissions from the
     * other Publishers via a function to produce the output item.
     *
     * <p>Note that this operator doesn't emit anything until all other sources have produced at
     * least one value. The resulting emission only happens when this Publisher emits (and
     * not when any of the other sources emit, unlike combineLatest).
     * If a source doesn't produce any value and just completes, the sequence is completed immediately.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator is a pass-through for backpressure behavior between the source {@code Publisher}
     *  and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the first other source's value type
     * @param <T2> the second other source's value type
     * @param <R> the result value type
     * @param source1 the first other Publisher
     * @param source2 the second other Publisher
     * @param combiner the function called with an array of values from each participating Publisher
     * @return the new Publisher instance
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <T1, T2, R> Flowable<R> withLatestFrom(Publisher<T1> source1, Publisher<T2> source2,
            Function3<? super T, ? super T1, ? super T2, R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        Function<Object[], R> f = Functions.toFunction(combiner);
        return withLatestFrom(new Publisher[] { source1, source2 }, f);
    }

    
Combines the value emission from this Publisher with the latest emissions from the
other Publishers via a function to produce the output item.
Note that this operator doesn't emit anything until all other sources have produced at
least one value. The resulting emission only happens when this Publisher emits (and
not when any of the other sources emit, unlike combineLatest).
If a source doesn't produce any value and just completes, the sequence is completed immediately.

 
Backpressure:
 
This operator is a pass-through for backpressure behavior between the source Publisher and the downstream Subscriber. The other Publishers are consumed in an unbounded manner.
 
Scheduler:
 
This operator does not operate by default on a particular Scheduler.

Params:
source1 – the first other Publisher
source2 – the second other Publisher
source3 – the third other Publisher
combiner – the function called with an array of values from each participating Publisher
Type parameters:
<T1> – the first other source's value type
<T2> – the second other source's value type
<T3> – the third other source's value type
<R> – the result value type
Returns:
the new Publisher instance
Since:
2.0/**
     * Combines the value emission from this Publisher with the latest emissions from the
     * other Publishers via a function to produce the output item.
     *
     * <p>Note that this operator doesn't emit anything until all other sources have produced at
     * least one value. The resulting emission only happens when this Publisher emits (and
     * not when any of the other sources emit, unlike combineLatest).
     * If a source doesn't produce any value and just completes, the sequence is completed immediately.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator is a pass-through for backpressure behavior between the source {@code Publisher}
     *  and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the first other source's value type
     * @param <T2> the second other source's value type
     * @param <T3> the third other source's value type
     * @param <R> the result value type
     * @param source1 the first other Publisher
     * @param source2 the second other Publisher
     * @param source3 the third other Publisher
     * @param combiner the function called with an array of values from each participating Publisher
     * @return the new Publisher instance
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <T1, T2, T3, R> Flowable<R> withLatestFrom(
            Publisher<T1> source1, Publisher<T2> source2,
            Publisher<T3> source3,
            Function4<? super T, ? super T1, ? super T2, ? super T3, R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        Function<Object[], R> f = Functions.toFunction(combiner);
        return withLatestFrom(new Publisher[] { source1, source2, source3 }, f);
    }

    
Combines the value emission from this Publisher with the latest emissions from the
other Publishers via a function to produce the output item.
Note that this operator doesn't emit anything until all other sources have produced at
least one value. The resulting emission only happens when this Publisher emits (and
not when any of the other sources emit, unlike combineLatest).
If a source doesn't produce any value and just completes, the sequence is completed immediately.

 
Backpressure:
 
This operator is a pass-through for backpressure behavior between the source Publisher and the downstream Subscriber. The other Publishers are consumed in an unbounded manner.
 
Scheduler:
 
This operator does not operate by default on a particular Scheduler.

Params:
source1 – the first other Publisher
source2 – the second other Publisher
source3 – the third other Publisher
source4 – the fourth other Publisher
combiner – the function called with an array of values from each participating Publisher
Type parameters:
<T1> – the first other source's value type
<T2> – the second other source's value type
<T3> – the third other source's value type
<T4> – the fourth other source's value type
<R> – the result value type
Returns:
the new Publisher instance
Since:
2.0/**
     * Combines the value emission from this Publisher with the latest emissions from the
     * other Publishers via a function to produce the output item.
     *
     * <p>Note that this operator doesn't emit anything until all other sources have produced at
     * least one value. The resulting emission only happens when this Publisher emits (and
     * not when any of the other sources emit, unlike combineLatest).
     * If a source doesn't produce any value and just completes, the sequence is completed immediately.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator is a pass-through for backpressure behavior between the source {@code Publisher}
     *  and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <T1> the first other source's value type
     * @param <T2> the second other source's value type
     * @param <T3> the third other source's value type
     * @param <T4> the fourth other source's value type
     * @param <R> the result value type
     * @param source1 the first other Publisher
     * @param source2 the second other Publisher
     * @param source3 the third other Publisher
     * @param source4 the fourth other Publisher
     * @param combiner the function called with an array of values from each participating Publisher
     * @return the new Publisher instance
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <T1, T2, T3, T4, R> Flowable<R> withLatestFrom(
            Publisher<T1> source1, Publisher<T2> source2,
            Publisher<T3> source3, Publisher<T4> source4,
            Function5<? super T, ? super T1, ? super T2, ? super T3, ? super T4, R> combiner) {
        ObjectHelper.requireNonNull(source1, "source1 is null");
        ObjectHelper.requireNonNull(source2, "source2 is null");
        ObjectHelper.requireNonNull(source3, "source3 is null");
        ObjectHelper.requireNonNull(source4, "source4 is null");
        Function<Object[], R> f = Functions.toFunction(combiner);
        return withLatestFrom(new Publisher[] { source1, source2, source3, source4 }, f);
    }

    
Combines the value emission from this Publisher with the latest emissions from the
other Publishers via a function to produce the output item.
Note that this operator doesn't emit anything until all other sources have produced at
least one value. The resulting emission only happens when this Publisher emits (and
not when any of the other sources emit, unlike combineLatest).
If a source doesn't produce any value and just completes, the sequence is completed immediately.

 
Backpressure:
 
This operator is a pass-through for backpressure behavior between the source Publisher and the downstream Subscriber. The other Publishers are consumed in an unbounded manner.
 
Scheduler:
 
This operator does not operate by default on a particular Scheduler.

Params:
others – the array of other sources
combiner – the function called with an array of values from each participating Publisher
Type parameters:
<R> – the result value type
Returns:
the new Publisher instance
Since:
2.0/**
     * Combines the value emission from this Publisher with the latest emissions from the
     * other Publishers via a function to produce the output item.
     *
     * <p>Note that this operator doesn't emit anything until all other sources have produced at
     * least one value. The resulting emission only happens when this Publisher emits (and
     * not when any of the other sources emit, unlike combineLatest).
     * If a source doesn't produce any value and just completes, the sequence is completed immediately.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator is a pass-through for backpressure behavior between the source {@code Publisher}
     *  and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the result value type
     * @param others the array of other sources
     * @param combiner the function called with an array of values from each participating Publisher
     * @return the new Publisher instance
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> withLatestFrom(Publisher<?>[] others, Function<? super Object[], R> combiner) {
        ObjectHelper.requireNonNull(others, "others is null");
        ObjectHelper.requireNonNull(combiner, "combiner is null");
        return RxJavaPlugins.onAssembly(new FlowableWithLatestFromMany<T, R>(this, others, combiner));
    }

    
Combines the value emission from this Publisher with the latest emissions from the
other Publishers via a function to produce the output item.
Note that this operator doesn't emit anything until all other sources have produced at
least one value. The resulting emission only happens when this Publisher emits (and
not when any of the other sources emit, unlike combineLatest).
If a source doesn't produce any value and just completes, the sequence is completed immediately.

 
Backpressure:
 
This operator is a pass-through for backpressure behavior between the source Publisher and the downstream Subscriber. The other Publishers are consumed in an unbounded manner.
 
Scheduler:
 
This operator does not operate by default on a particular Scheduler.

Params:
others – the iterable of other sources
combiner – the function called with an array of values from each participating Publisher
Type parameters:
<R> – the result value type
Returns:
the new Publisher instance
Since:
2.0/**
     * Combines the value emission from this Publisher with the latest emissions from the
     * other Publishers via a function to produce the output item.
     *
     * <p>Note that this operator doesn't emit anything until all other sources have produced at
     * least one value. The resulting emission only happens when this Publisher emits (and
     * not when any of the other sources emit, unlike combineLatest).
     * If a source doesn't produce any value and just completes, the sequence is completed immediately.
     *
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>This operator is a pass-through for backpressure behavior between the source {@code Publisher}
     *  and the downstream Subscriber. The other {@code Publisher}s are consumed in an unbounded manner.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>This operator does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <R> the result value type
     * @param others the iterable of other sources
     * @param combiner the function called with an array of values from each participating Publisher
     * @return the new Publisher instance
     * @since 2.0
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.PASS_THROUGH)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <R> Flowable<R> withLatestFrom(Iterable<? extends Publisher<?>> others, Function<? super Object[], R> combiner) {
        ObjectHelper.requireNonNull(others, "others is null");
        ObjectHelper.requireNonNull(combiner, "combiner is null");
        return RxJavaPlugins.onAssembly(new FlowableWithLatestFromMany<T, R>(this, others, combiner));
    }

    
Returns a Flowable that emits items that are the result of applying a specified function to pairs of
values, one each from the source Publisher and a specified Iterable sequence.


 Note that the other Iterable is evaluated as items are observed from the source Publisher; it is not pre-consumed. This allows you to zip infinite streams on either side. 

 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zipWith does not operate by default on a particular Scheduler.

Params:
other – 
           the Iterable sequence
zipper – 
           a function that combines the pairs of items from the Publisher and the Iterable to generate
           the items to be emitted by the resulting Publisher
Type parameters:
<U> –  the type of items in the other Iterable
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that pairs up values from the source Publisher and the other Iterable sequence and emits the results of zipFunction applied to these pairs/**
     * Returns a Flowable that emits items that are the result of applying a specified function to pairs of
     * values, one each from the source Publisher and a specified Iterable sequence.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.i.png" alt="">
     * <p>
     * Note that the {@code other} Iterable is evaluated as items are observed from the source Publisher; it is
     * not pre-consumed. This allows you to zip infinite streams on either side.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zipWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items in the {@code other} Iterable
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param other
     *            the Iterable sequence
     * @param zipper
     *            a function that combines the pairs of items from the Publisher and the Iterable to generate
     *            the items to be emitted by the resulting Publisher
     * @return a Flowable that pairs up values from the source Publisher and the {@code other} Iterable
     *         sequence and emits the results of {@code zipFunction} applied to these pairs
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> zipWith(Iterable<U> other,  BiFunction<? super T, ? super U, ? extends R> zipper) {
        ObjectHelper.requireNonNull(other, "other is null");
        ObjectHelper.requireNonNull(zipper, "zipper is null");
        return RxJavaPlugins.onAssembly(new FlowableZipIterable<T, U, R>(this, other, zipper));
    }

    
Returns a Flowable that emits items that are the result of applying a specified function to pairs of
values, one each from the source Publisher and another specified Publisher.
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 



 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zipWith does not operate by default on a particular Scheduler.

Params:
other – 
           the other Publisher
zipper – 
           a function that combines the pairs of items from the two Publishers to generate the items to
           be emitted by the resulting Publisher
Type parameters:
<U> –  the type of items emitted by the other Publisher
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that pairs up values from the source Publisher and the other Publisher and emits the results of zipFunction applied to these pairs/**
     * Returns a Flowable that emits items that are the result of applying a specified function to pairs of
     * values, one each from the source Publisher and another specified Publisher.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zipWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the {@code other} Publisher
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param other
     *            the other Publisher
     * @param zipper
     *            a function that combines the pairs of items from the two Publishers to generate the items to
     *            be emitted by the resulting Publisher
     * @return a Flowable that pairs up values from the source Publisher and the {@code other} Publisher
     *         and emits the results of {@code zipFunction} applied to these pairs
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     */
    @CheckReturnValue
    @NonNull
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> zipWith(Publisher<? extends U> other, BiFunction<? super T, ? super U, ? extends R> zipper) {
        ObjectHelper.requireNonNull(other, "other is null");
        return zip(this, other, zipper);
    }

    
Returns a Flowable that emits items that are the result of applying a specified function to pairs of
values, one each from the source Publisher and another specified Publisher.
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 



 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zipWith does not operate by default on a particular Scheduler.

Params:
other – 
           the other Publisher
zipper – 
           a function that combines the pairs of items from the two Publishers to generate the items to
           be emitted by the resulting Publisher
delayError – 
           if true, errors from the current Flowable or the other Publisher is delayed until both terminate
Type parameters:
<U> –  the type of items emitted by the other Publisher
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that pairs up values from the source Publisher and the other Publisher and emits the results of zipFunction applied to these pairs
Since:
2.0/**
     * Returns a Flowable that emits items that are the result of applying a specified function to pairs of
     * values, one each from the source Publisher and another specified Publisher.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zipWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the {@code other} Publisher
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param other
     *            the other Publisher
     * @param zipper
     *            a function that combines the pairs of items from the two Publishers to generate the items to
     *            be emitted by the resulting Publisher
     * @param delayError
     *            if true, errors from the current Flowable or the other Publisher is delayed until both terminate
     * @return a Flowable that pairs up values from the source Publisher and the {@code other} Publisher
     *         and emits the results of {@code zipFunction} applied to these pairs
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> zipWith(Publisher<? extends U> other,
            BiFunction<? super T, ? super U, ? extends R> zipper, boolean delayError) {
        return zip(this, other, zipper, delayError);
    }

    
Returns a Flowable that emits items that are the result of applying a specified function to pairs of
values, one each from the source Publisher and another specified Publisher.
 The operator subscribes to its sources in the order they are specified and completes eagerly if one of the sources is shorter than the rest while canceling the other sources. Therefore, it is possible those other sources will never be able to run to completion (and thus not calling doOnComplete()). This can also happen if the sources are exactly the same length; if source A completes and B has been consumed and is about to complete, the operator detects A won't be sending further values and it will cancel B immediately. For example: 
range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -> a + b)
 action1 will be called but action2 won't. 
To work around this termination property, use doOnCancel(Action) as well or use using() to do cleanup in case of completion or cancellation. 



 
Backpressure:
 
The operator expects backpressure from the sources and honors backpressure from the downstream. (I.e., zipping with interval(long, TimeUnit) may result in MissingBackpressureException, use one of the onBackpressureX to handle similar, backpressure-ignoring sources.
 
Scheduler:
 
zipWith does not operate by default on a particular Scheduler.

Params:
other – 
           the other Publisher
zipper – 
           a function that combines the pairs of items from the two Publishers to generate the items to
           be emitted by the resulting Publisher
bufferSize – 
           the capacity hint for the buffer in the inner windows
delayError – 
           if true, errors from the current Flowable or the other Publisher is delayed until both terminate
Type parameters:
<U> –  the type of items emitted by the other Publisher
<R> – 
           the type of items emitted by the resulting Publisher
See Also:
ReactiveX operators documentation: Zip
Returns:
a Flowable that pairs up values from the source Publisher and the other Publisher and emits the results of zipFunction applied to these pairs
Since:
2.0/**
     * Returns a Flowable that emits items that are the result of applying a specified function to pairs of
     * values, one each from the source Publisher and another specified Publisher.
     * <p>
     * The operator subscribes to its sources in the order they are specified and completes eagerly if
     * one of the sources is shorter than the rest while canceling the other sources. Therefore, it
     * is possible those other sources will never be able to run to completion (and thus not calling
     * {@code doOnComplete()}). This can also happen if the sources are exactly the same length; if
     * source A completes and B has been consumed and is about to complete, the operator detects A won't
     * be sending further values and it will cancel B immediately. For example:
     * <pre><code>range(1, 5).doOnComplete(action1).zipWith(range(6, 5).doOnComplete(action2), (a, b) -&gt; a + b)</code></pre>
     * {@code action1} will be called but {@code action2} won't.
     * <br>To work around this termination property,
     * use {@link #doOnCancel(Action)} as well or use {@code using()} to do cleanup in case of completion
     * or cancellation.
     * <p>
     * <img width="640" height="380" src="https://raw.github.com/wiki/ReactiveX/RxJava/images/rx-operators/zip.png" alt="">
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The operator expects backpressure from the sources and honors backpressure from the downstream.
     *  (I.e., zipping with {@link #interval(long, TimeUnit)} may result in MissingBackpressureException, use
     *  one of the {@code onBackpressureX} to handle similar, backpressure-ignoring sources.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code zipWith} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     *
     * @param <U>
     *            the type of items emitted by the {@code other} Publisher
     * @param <R>
     *            the type of items emitted by the resulting Publisher
     * @param other
     *            the other Publisher
     * @param zipper
     *            a function that combines the pairs of items from the two Publishers to generate the items to
     *            be emitted by the resulting Publisher
     * @param bufferSize
     *            the capacity hint for the buffer in the inner windows
     * @param delayError
     *            if true, errors from the current Flowable or the other Publisher is delayed until both terminate
     * @return a Flowable that pairs up values from the source Publisher and the {@code other} Publisher
     *         and emits the results of {@code zipFunction} applied to these pairs
     * @see <a href="http://reactivex.io/documentation/operators/zip.html">ReactiveX operators documentation: Zip</a>
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final <U, R> Flowable<R> zipWith(Publisher<? extends U> other,
            BiFunction<? super T, ? super U, ? extends R> zipper, boolean delayError, int bufferSize) {
        return zip(this, other, zipper, delayError, bufferSize);
    }

    // -------------------------------------------------------------------------
    // Fluent test support, super handy and reduces test preparation boilerplate
    // -------------------------------------------------------------------------
    
Creates a TestSubscriber that requests Long.MAX_VALUE and subscribes
it to this Flowable.

 
Backpressure:
 
The returned TestSubscriber consumes this Flowable in an unbounded fashion.
 
Scheduler:
 
test does not operate by default on a particular Scheduler.

Returns:
the new TestSubscriber instance
Since:
2.0/**
     * Creates a TestSubscriber that requests Long.MAX_VALUE and subscribes
     * it to this Flowable.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned TestSubscriber consumes this Flowable in an unbounded fashion.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code test} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @return the new TestSubscriber instance
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.UNBOUNDED_IN)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final TestSubscriber<T> test() { // NoPMD
        TestSubscriber<T> ts = new TestSubscriber<T>();
        subscribe(ts);
        return ts;
    }

    
Creates a TestSubscriber with the given initial request amount and subscribes
it to this Flowable.

 
Backpressure:
 
The returned TestSubscriber requests the given initialRequest amount upfront.
 
Scheduler:
 
test does not operate by default on a particular Scheduler.

Params:
initialRequest – the initial request amount, positive
Returns:
the new TestSubscriber instance
Since:
2.0/**
     * Creates a TestSubscriber with the given initial request amount and subscribes
     * it to this Flowable.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned TestSubscriber requests the given {@code initialRequest} amount upfront.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code test} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param initialRequest the initial request amount, positive
     * @return the new TestSubscriber instance
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final TestSubscriber<T> test(long initialRequest) { // NoPMD
        TestSubscriber<T> ts = new TestSubscriber<T>(initialRequest);
        subscribe(ts);
        return ts;
    }

    
Creates a TestSubscriber with the given initial request amount,
optionally cancels it before the subscription and subscribes
it to this Flowable.

 
Backpressure:
 
The returned TestSubscriber requests the given initialRequest amount upfront.
 
Scheduler:
 
test does not operate by default on a particular Scheduler.

Params:
initialRequest – the initial request amount, positive
cancel – should the TestSubscriber be canceled before the subscription?
Returns:
the new TestSubscriber instance
Since:
2.0/**
     * Creates a TestSubscriber with the given initial request amount,
     * optionally cancels it before the subscription and subscribes
     * it to this Flowable.
     * <dl>
     *  <dt><b>Backpressure:</b></dt>
     *  <dd>The returned TestSubscriber requests the given {@code initialRequest} amount upfront.</dd>
     *  <dt><b>Scheduler:</b></dt>
     *  <dd>{@code test} does not operate by default on a particular {@link Scheduler}.</dd>
     * </dl>
     * @param initialRequest the initial request amount, positive
     * @param cancel should the TestSubscriber be canceled before the subscription?
     * @return the new TestSubscriber instance
     * @since 2.0
     */
    @CheckReturnValue
    @BackpressureSupport(BackpressureKind.FULL)
    @SchedulerSupport(SchedulerSupport.NONE)
    public final TestSubscriber<T> test(long initialRequest, boolean cancel) { // NoPMD
        TestSubscriber<T> ts = new TestSubscriber<T>(initialRequest);
        if (cancel) {
            ts.cancel();
        }
        subscribe(ts);
        return ts;
    }

}