/*
* Copyright (c) 2000, 2012, Oracle and/or its affiliates. All rights reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License version 2 only, as
* published by the Free Software Foundation. Oracle designates this
* particular file as subject to the "Classpath" exception as provided
* by Oracle in the LICENSE file that accompanied this code.
*
* This code is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
* version 2 for more details (a copy is included in the LICENSE file that
* accompanied this code).
*
* You should have received a copy of the GNU General Public License version
* 2 along with this work; if not, write to the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
*
* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
* or visit www.oracle.com if you need additional information or have any
* questions.
*/
/*
* @author IBM Corp.
*
* Copyright IBM Corp. 1999-2000. All rights reserved.
*/
package javax.management.modelmbean;
import static com.sun.jmx.defaults.JmxProperties.MODELMBEAN_LOGGER;
import static com.sun.jmx.mbeanserver.Util.cast;
import com.sun.jmx.mbeanserver.GetPropertyAction;
import java.io.IOException;
import java.io.ObjectInputStream;
import java.io.ObjectOutputStream;
import java.io.ObjectStreamField;
import java.lang.reflect.Constructor;
import java.security.AccessController;
import java.util.HashMap;
import java.util.Iterator;
import java.util.Map;
import java.util.Set;
import java.util.SortedMap;
import java.util.StringTokenizer;
import java.util.TreeMap;
import java.util.logging.Level;
import javax.management.Descriptor;
import javax.management.ImmutableDescriptor;
import javax.management.MBeanException;
import javax.management.RuntimeOperationsException;
import sun.reflect.misc.ReflectUtil;
This class represents the metadata set for a ModelMBean element. A
descriptor is part of the ModelMBeanInfo,
ModelMBeanNotificationInfo, ModelMBeanAttributeInfo,
ModelMBeanConstructorInfo, and ModelMBeanParameterInfo.
A descriptor consists of a collection of fields. Each field is in
fieldname=fieldvalue format. Field names are not case sensitive,
case will be preserved on field values.
All field names and values are not predefined. New fields can be
defined and added by any program. Some fields have been predefined
for consistency of implementation and support by the
ModelMBeanInfo, ModelMBeanAttributeInfo, ModelMBeanConstructorInfo,
ModelMBeanNotificationInfo, ModelMBeanOperationInfo and ModelMBean
classes.
The serialVersionUID of this class is -6292969195866300415L
.
Since: 1.5
/**
* This class represents the metadata set for a ModelMBean element. A
* descriptor is part of the ModelMBeanInfo,
* ModelMBeanNotificationInfo, ModelMBeanAttributeInfo,
* ModelMBeanConstructorInfo, and ModelMBeanParameterInfo.
* <P>
* A descriptor consists of a collection of fields. Each field is in
* fieldname=fieldvalue format. Field names are not case sensitive,
* case will be preserved on field values.
* <P>
* All field names and values are not predefined. New fields can be
* defined and added by any program. Some fields have been predefined
* for consistency of implementation and support by the
* ModelMBeanInfo, ModelMBeanAttributeInfo, ModelMBeanConstructorInfo,
* ModelMBeanNotificationInfo, ModelMBeanOperationInfo and ModelMBean
* classes.
*
* <p>The <b>serialVersionUID</b> of this class is <code>-6292969195866300415L</code>.
*
* @since 1.5
*/
@SuppressWarnings("serial") // serialVersionUID not constant
public class DescriptorSupport
implements javax.management.Descriptor
{
// Serialization compatibility stuff:
// Two serial forms are supported in this class. The selected form depends
// on system property "jmx.serial.form":
// - "1.0" for JMX 1.0
// - any other value for JMX 1.1 and higher
//
// Serial version for old serial form
private static final long oldSerialVersionUID = 8071560848919417985L;
//
// Serial version for new serial form
private static final long newSerialVersionUID = -6292969195866300415L;
//
// Serializable fields in old serial form
private static final ObjectStreamField[] oldSerialPersistentFields =
{
new ObjectStreamField("descriptor", HashMap.class),
new ObjectStreamField("currClass", String.class)
};
//
// Serializable fields in new serial form
private static final ObjectStreamField[] newSerialPersistentFields =
{
new ObjectStreamField("descriptor", HashMap.class)
};
//
// Actual serial version and serial form
private static final long serialVersionUID;
@serialField descriptor HashMap The collection of fields representing this descriptor
/**
* @serialField descriptor HashMap The collection of fields representing this descriptor
*/
private static final ObjectStreamField[] serialPersistentFields;
private static final String serialForm;
static {
String form = null;
boolean compat = false;
try {
GetPropertyAction act = new GetPropertyAction("jmx.serial.form");
form = AccessController.doPrivileged(act);
compat = "1.0".equals(form); // form may be null
} catch (Exception e) {
// OK: No compat with 1.0
}
serialForm = form;
if (compat) {
serialPersistentFields = oldSerialPersistentFields;
serialVersionUID = oldSerialVersionUID;
} else {
serialPersistentFields = newSerialPersistentFields;
serialVersionUID = newSerialVersionUID;
}
}
//
// END Serialization compatibility stuff
/* Spec says that field names are case-insensitive, but that case
is preserved. This means that we need to be able to map from a
name that may differ in case to the actual name that is used in
the HashMap. Thus, descriptorMap is a TreeMap with a Comparator
that ignores case.
Previous versions of this class had a field called "descriptor"
of type HashMap where the keys were directly Strings. This is
hard to reconcile with the required semantics, so we fabricate
that field virtually during serialization and deserialization
but keep the real information in descriptorMap.
*/
private transient SortedMap<String, Object> descriptorMap;
private static final String currClass = "DescriptorSupport";
Descriptor default constructor.
Default initial descriptor size is 20. It will grow as needed.
Note that the created empty descriptor is not a valid descriptor (the method isValid
returns false
)
/**
* Descriptor default constructor.
* Default initial descriptor size is 20. It will grow as needed.<br>
* Note that the created empty descriptor is not a valid descriptor
* (the method {@link #isValid isValid} returns <CODE>false</CODE>)
*/
public DescriptorSupport() {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"DescriptorSupport()" , "Constructor");
}
init(null);
}
Descriptor constructor. Takes as parameter the initial
capacity of the Map that stores the descriptor fields.
Capacity will grow as needed.
Note that the created empty descriptor is not a valid descriptor (the method isValid
returns false
).
Params: - initNumFields – The initial capacity of the Map that
stores the descriptor fields.
Throws: - RuntimeOperationsException – for illegal value for
initNumFields (<= 0)
- MBeanException – Wraps a distributed communication Exception.
/**
* Descriptor constructor. Takes as parameter the initial
* capacity of the Map that stores the descriptor fields.
* Capacity will grow as needed.<br> Note that the created empty
* descriptor is not a valid descriptor (the method {@link
* #isValid isValid} returns <CODE>false</CODE>).
*
* @param initNumFields The initial capacity of the Map that
* stores the descriptor fields.
*
* @exception RuntimeOperationsException for illegal value for
* initNumFields (<= 0)
* @exception MBeanException Wraps a distributed communication Exception.
*/
public DescriptorSupport(int initNumFields)
throws MBeanException, RuntimeOperationsException {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(initNumFields = " + initNumFields + ")",
"Constructor");
}
if (initNumFields <= 0) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(initNumFields)",
"Illegal arguments: initNumFields <= 0");
}
final String msg =
"Descriptor field limit invalid: " + initNumFields;
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
init(null);
}
Descriptor constructor taking a Descriptor as parameter.
Creates a new descriptor initialized to the values of the
descriptor passed in parameter.
Params: - inDescr – the descriptor to be used to initialize the
constructed descriptor. If it is null or contains no descriptor
fields, an empty Descriptor will be created.
/**
* Descriptor constructor taking a Descriptor as parameter.
* Creates a new descriptor initialized to the values of the
* descriptor passed in parameter.
*
* @param inDescr the descriptor to be used to initialize the
* constructed descriptor. If it is null or contains no descriptor
* fields, an empty Descriptor will be created.
*/
public DescriptorSupport(DescriptorSupport inDescr) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(Descriptor)", "Constructor");
}
if (inDescr == null)
init(null);
else
init(inDescr.descriptorMap);
}
Descriptor constructor taking an XML String.
The format of the XML string is not defined, but an implementation must ensure that the string returned by toXMLString()
on an existing descriptor can be used to instantiate an equivalent descriptor using this constructor.
In this implementation, all field values will be created
as Strings. If the field values are not Strings, the
programmer will have to reset or convert these fields
correctly.
Params: - inStr – An XML-formatted string used to populate this Descriptor. The format is not defined, but any implementation must ensure that the string returned by method
toXMLString
on an existing descriptor can be used to instantiate an equivalent descriptor when instantiated using this constructor.
Throws: - RuntimeOperationsException – If the String inStr
passed in parameter is null
- XMLParseException – XML parsing problem while parsing
the input String
- MBeanException – Wraps a distributed communication Exception.
/**
* <p>Descriptor constructor taking an XML String.</p>
*
* <p>The format of the XML string is not defined, but an
* implementation must ensure that the string returned by
* {@link #toXMLString() toXMLString()} on an existing
* descriptor can be used to instantiate an equivalent
* descriptor using this constructor.</p>
*
* <p>In this implementation, all field values will be created
* as Strings. If the field values are not Strings, the
* programmer will have to reset or convert these fields
* correctly.</p>
*
* @param inStr An XML-formatted string used to populate this
* Descriptor. The format is not defined, but any
* implementation must ensure that the string returned by
* method {@link #toXMLString toXMLString} on an existing
* descriptor can be used to instantiate an equivalent
* descriptor when instantiated using this constructor.
*
* @exception RuntimeOperationsException If the String inStr
* passed in parameter is null
* @exception XMLParseException XML parsing problem while parsing
* the input String
* @exception MBeanException Wraps a distributed communication Exception.
*/
/* At some stage we should rewrite this code to be cleverer. Using
a StringTokenizer as we do means, first, that we accept a lot of
bogus strings without noticing they are bogus, and second, that we
split the string being parsed at characters like > even if they
occur in the middle of a field value. */
public DescriptorSupport(String inStr)
throws MBeanException, RuntimeOperationsException,
XMLParseException {
/* parse an XML-formatted string and populate internal
* structure with it */
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(String = '" + inStr + "')", "Constructor");
}
if (inStr == null) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(String = null)", "Illegal arguments");
}
final String msg = "String in parameter is null";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
final String lowerInStr = inStr.toLowerCase();
if (!lowerInStr.startsWith("<descriptor>")
|| !lowerInStr.endsWith("</descriptor>")) {
throw new XMLParseException("No <descriptor>, </descriptor> pair");
}
// parse xmlstring into structures
init(null);
// create dummy descriptor: should have same size
// as number of fields in xmlstring
// loop through structures and put them in descriptor
StringTokenizer st = new StringTokenizer(inStr, "<> \t\n\r\f");
boolean inFld = false;
boolean inDesc = false;
String fieldName = null;
String fieldValue = null;
while (st.hasMoreTokens()) { // loop through tokens
String tok = st.nextToken();
if (tok.equalsIgnoreCase("FIELD")) {
inFld = true;
} else if (tok.equalsIgnoreCase("/FIELD")) {
if ((fieldName != null) && (fieldValue != null)) {
fieldName =
fieldName.substring(fieldName.indexOf('"') + 1,
fieldName.lastIndexOf('"'));
final Object fieldValueObject =
parseQuotedFieldValue(fieldValue);
setField(fieldName, fieldValueObject);
}
fieldName = null;
fieldValue = null;
inFld = false;
} else if (tok.equalsIgnoreCase("DESCRIPTOR")) {
inDesc = true;
} else if (tok.equalsIgnoreCase("/DESCRIPTOR")) {
inDesc = false;
fieldName = null;
fieldValue = null;
inFld = false;
} else if (inFld && inDesc) {
// want kw=value, eg, name="myname" value="myvalue"
int eq_separator = tok.indexOf("=");
if (eq_separator > 0) {
String kwPart = tok.substring(0,eq_separator);
String valPart = tok.substring(eq_separator+1);
if (kwPart.equalsIgnoreCase("NAME"))
fieldName = valPart;
else if (kwPart.equalsIgnoreCase("VALUE"))
fieldValue = valPart;
else { // xml parse exception
final String msg =
"Expected `name' or `value', got `" + tok + "'";
throw new XMLParseException(msg);
}
} else { // xml parse exception
final String msg =
"Expected `keyword=value', got `" + tok + "'";
throw new XMLParseException(msg);
}
}
} // while tokens
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(XMLString)", "Exit");
}
}
Constructor taking field names and field values. Neither array
can be null.
Params: - fieldNames – String array of field names. No elements of
this array can be null.
- fieldValues – Object array of the corresponding field
values. Elements of the array can be null. The
fieldValue
must be valid for the
fieldName
(as defined in method
isValid
) Note: array sizes of parameters should match. If both arrays
are empty, then an empty descriptor is created.
Throws: - RuntimeOperationsException – for illegal value for
field Names or field Values. The array lengths must be equal.
If the descriptor construction fails for any reason, this
exception will be thrown.
/**
* Constructor taking field names and field values. Neither array
* can be null.
*
* @param fieldNames String array of field names. No elements of
* this array can be null.
* @param fieldValues Object array of the corresponding field
* values. Elements of the array can be null. The
* <code>fieldValue</code> must be valid for the
* <code>fieldName</code> (as defined in method {@link #isValid
* isValid})
*
* <p>Note: array sizes of parameters should match. If both arrays
* are empty, then an empty descriptor is created.</p>
*
* @exception RuntimeOperationsException for illegal value for
* field Names or field Values. The array lengths must be equal.
* If the descriptor construction fails for any reason, this
* exception will be thrown.
*
*/
public DescriptorSupport(String[] fieldNames, Object[] fieldValues)
throws RuntimeOperationsException {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(fieldNames,fieldObjects)", "Constructor");
}
if ((fieldNames == null) || (fieldValues == null) ||
(fieldNames.length != fieldValues.length)) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(fieldNames,fieldObjects)",
"Illegal arguments");
}
final String msg =
"Null or invalid fieldNames or fieldValues";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
/* populate internal structure with fields */
init(null);
for (int i=0; i < fieldNames.length; i++) {
// setField will throw an exception if a fieldName is be null.
// the fieldName and fieldValue will be validated in setField.
setField(fieldNames[i], fieldValues[i]);
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(fieldNames,fieldObjects)", "Exit");
}
}
Constructor taking fields in the fieldName=fieldValue
format.
Params: - fields – String array with each element containing a
field name and value. If this array is null or empty, then the
default constructor will be executed. Null strings or empty
strings will be ignored.
All field values should be Strings. If the field values are
not Strings, the programmer will have to reset or convert these
fields correctly.
Note: Each string should be of the form
fieldName=fieldValue. The field name ends at the first =
character; for example if the String is a=b=c
then the field name is a
and its value is b=c
.
Throws: - RuntimeOperationsException – for illegal value for
field Names or field Values. The field must contain an
"=". "=fieldValue", "fieldName", and "fieldValue" are illegal.
FieldName cannot be null. "fieldName=" will cause the value to
be null. If the descriptor construction fails for any reason,
this exception will be thrown.
/**
* Constructor taking fields in the <i>fieldName=fieldValue</i>
* format.
*
* @param fields String array with each element containing a
* field name and value. If this array is null or empty, then the
* default constructor will be executed. Null strings or empty
* strings will be ignored.
*
* <p>All field values should be Strings. If the field values are
* not Strings, the programmer will have to reset or convert these
* fields correctly.
*
* <p>Note: Each string should be of the form
* <i>fieldName=fieldValue</i>. The field name
* ends at the first {@code =} character; for example if the String
* is {@code a=b=c} then the field name is {@code a} and its value
* is {@code b=c}.
*
* @exception RuntimeOperationsException for illegal value for
* field Names or field Values. The field must contain an
* "=". "=fieldValue", "fieldName", and "fieldValue" are illegal.
* FieldName cannot be null. "fieldName=" will cause the value to
* be null. If the descriptor construction fails for any reason,
* this exception will be thrown.
*
*/
public DescriptorSupport(String... fields)
{
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(String... fields)", "Constructor");
}
init(null);
if (( fields == null ) || ( fields.length == 0))
return;
init(null);
for (int i=0; i < fields.length; i++) {
if ((fields[i] == null) || (fields[i].equals(""))) {
continue;
}
int eq_separator = fields[i].indexOf("=");
if (eq_separator < 0) {
// illegal if no = or is first character
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(String... fields)",
"Illegal arguments: field does not have " +
"'=' as a name and value separator");
}
final String msg = "Field in invalid format: no equals sign";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
String fieldName = fields[i].substring(0,eq_separator);
String fieldValue = null;
if (eq_separator < fields[i].length()) {
// = is not in last character
fieldValue = fields[i].substring(eq_separator+1);
}
if (fieldName.equals("")) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(String... fields)",
"Illegal arguments: fieldName is empty");
}
final String msg = "Field in invalid format: no fieldName";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
setField(fieldName,fieldValue);
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"Descriptor(String... fields)", "Exit");
}
}
private void init(Map<String, ?> initMap) {
descriptorMap =
new TreeMap<String, Object>(String.CASE_INSENSITIVE_ORDER);
if (initMap != null)
descriptorMap.putAll(initMap);
}
// Implementation of the Descriptor interface
public synchronized Object getFieldValue(String fieldName)
throws RuntimeOperationsException {
if ((fieldName == null) || (fieldName.equals(""))) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldValue(String fieldName)",
"Illegal arguments: null field name");
}
final String msg = "Fieldname requested is null";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
Object retValue = descriptorMap.get(fieldName);
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldValue(String fieldName = " + fieldName + ")",
"Returns '" + retValue + "'");
}
return(retValue);
}
public synchronized void setField(String fieldName, Object fieldValue)
throws RuntimeOperationsException {
// field name cannot be null or empty
if ((fieldName == null) || (fieldName.equals(""))) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"setField(fieldName,fieldValue)",
"Illegal arguments: null or empty field name");
}
final String msg = "Field name to be set is null or empty";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
if (!validateField(fieldName, fieldValue)) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"setField(fieldName,fieldValue)",
"Illegal arguments");
}
final String msg =
"Field value invalid: " + fieldName + "=" + fieldValue;
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"setField(fieldName,fieldValue)", "Entry: setting '"
+ fieldName + "' to '" + fieldValue + "'");
}
// Since we do not remove any existing entry with this name,
// the field will preserve whatever case it had, ignoring
// any difference there might be in fieldName.
descriptorMap.put(fieldName, fieldValue);
}
public synchronized String[] getFields() {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFields()", "Entry");
}
int numberOfEntries = descriptorMap.size();
String[] responseFields = new String[numberOfEntries];
Set returnedSet = descriptorMap.entrySet();
int i = 0;
Object currValue = null;
Map.Entry currElement = null;
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFields()", "Returning " + numberOfEntries + " fields");
}
for (Iterator iter = returnedSet.iterator(); iter.hasNext(); i++) {
currElement = (Map.Entry) iter.next();
if (currElement == null) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFields()", "Element is null");
}
} else {
currValue = currElement.getValue();
if (currValue == null) {
responseFields[i] = currElement.getKey() + "=";
} else {
if (currValue instanceof java.lang.String) {
responseFields[i] =
currElement.getKey() + "=" + currValue.toString();
} else {
responseFields[i] =
currElement.getKey() + "=(" +
currValue.toString() + ")";
}
}
}
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFields()", "Exit");
}
return responseFields;
}
public synchronized String[] getFieldNames() {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldNames()", "Entry");
}
int numberOfEntries = descriptorMap.size();
String[] responseFields = new String[numberOfEntries];
Set returnedSet = descriptorMap.entrySet();
int i = 0;
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldNames()",
"Returning " + numberOfEntries + " fields");
}
for (Iterator iter = returnedSet.iterator(); iter.hasNext(); i++) {
Map.Entry currElement = (Map.Entry) iter.next();
if (( currElement == null ) || (currElement.getKey() == null)) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldNames()", "Field is null");
}
} else {
responseFields[i] = currElement.getKey().toString();
}
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldNames()", "Exit");
}
return responseFields;
}
public synchronized Object[] getFieldValues(String... fieldNames) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldValues(String... fieldNames)", "Entry");
}
// if fieldNames == null return all values
// if fieldNames is String[0] return no values
final int numberOfEntries =
(fieldNames == null) ? descriptorMap.size() : fieldNames.length;
final Object[] responseFields = new Object[numberOfEntries];
int i = 0;
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldValues(String... fieldNames)",
"Returning " + numberOfEntries + " fields");
}
if (fieldNames == null) {
for (Iterator iter = descriptorMap.values().iterator();
iter.hasNext(); i++)
responseFields[i] = iter.next();
} else {
for (i=0; i < fieldNames.length; i++) {
if ((fieldNames[i] == null) || (fieldNames[i].equals(""))) {
responseFields[i] = null;
} else {
responseFields[i] = getFieldValue(fieldNames[i]);
}
}
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"getFieldValues(String... fieldNames)", "Exit");
}
return responseFields;
}
public synchronized void setFields(String[] fieldNames,
Object[] fieldValues)
throws RuntimeOperationsException {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"setFields(fieldNames,fieldValues)", "Entry");
}
if ((fieldNames == null) || (fieldValues == null) ||
(fieldNames.length != fieldValues.length)) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"setFields(fieldNames,fieldValues)",
"Illegal arguments");
}
final String msg = "fieldNames and fieldValues are null or invalid";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
for (int i=0; i < fieldNames.length; i++) {
if (( fieldNames[i] == null) || (fieldNames[i].equals(""))) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"setFields(fieldNames,fieldValues)",
"Null field name encountered at element " + i);
}
final String msg = "fieldNames is null or invalid";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae, msg);
}
setField(fieldNames[i], fieldValues[i]);
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"setFields(fieldNames,fieldValues)", "Exit");
}
}
Returns a new Descriptor which is a duplicate of the Descriptor.
Throws: - RuntimeOperationsException – for illegal value for
field Names or field Values. If the descriptor construction
fails for any reason, this exception will be thrown.
/**
* Returns a new Descriptor which is a duplicate of the Descriptor.
*
* @exception RuntimeOperationsException for illegal value for
* field Names or field Values. If the descriptor construction
* fails for any reason, this exception will be thrown.
*/
public synchronized Object clone() throws RuntimeOperationsException {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"clone()", "Entry");
}
return(new DescriptorSupport(this));
}
public synchronized void removeField(String fieldName) {
if ((fieldName == null) || (fieldName.equals(""))) {
return;
}
descriptorMap.remove(fieldName);
}
Compares this descriptor to the given object. The objects are equal if
the given object is also a Descriptor, and if the two Descriptors have
the same field names (possibly differing in case) and the same
associated values. The respective values for a field in the two
Descriptors are equal if the following conditions hold:
- If one value is null then the other must be too.
- If one value is a primitive array then the other must be a primitive
array of the same type with the same elements.
- If one value is an object array then the other must be too and
Arrays.deepEquals
must return true.
- Otherwise
Object.equals(Object)
must return true.
Params: - o – the object to compare with.
Returns: true
if the objects are the same; false
otherwise.
/**
* Compares this descriptor to the given object. The objects are equal if
* the given object is also a Descriptor, and if the two Descriptors have
* the same field names (possibly differing in case) and the same
* associated values. The respective values for a field in the two
* Descriptors are equal if the following conditions hold:</p>
*
* <ul>
* <li>If one value is null then the other must be too.</li>
* <li>If one value is a primitive array then the other must be a primitive
* array of the same type with the same elements.</li>
* <li>If one value is an object array then the other must be too and
* {@link java.util.Arrays#deepEquals(Object[],Object[]) Arrays.deepEquals}
* must return true.</li>
* <li>Otherwise {@link Object#equals(Object)} must return true.</li>
* </ul>
*
* @param o the object to compare with.
*
* @return {@code true} if the objects are the same; {@code false}
* otherwise.
*
*/
// XXXX TODO: This is not very efficient!
// Note: this Javadoc is copied from javax.management.Descriptor
// due to 6369229.
public synchronized boolean equals(Object o) {
if (o == this)
return true;
return new ImmutableDescriptor(descriptorMap).equals(o);
}
Returns the hash code value for this descriptor. The hash code is computed as the sum of the hash codes for each field in the descriptor. The hash code of a field with name n
and value v
is n.toLowerCase().hashCode() ^ h
. Here h
is the hash code of v
, computed as follows:
- If
v
is null then h
is 0.
- If
v
is a primitive array then h
is computed using the appropriate overloading of java.util.Arrays.hashCode
.
- If
v
is an object array then h
is computed using Arrays.deepHashCode
.
- Otherwise
h
is v.hashCode()
.
Returns: A hash code value for this object.
/**
* <p>Returns the hash code value for this descriptor. The hash
* code is computed as the sum of the hash codes for each field in
* the descriptor. The hash code of a field with name {@code n}
* and value {@code v} is {@code n.toLowerCase().hashCode() ^ h}.
* Here {@code h} is the hash code of {@code v}, computed as
* follows:</p>
*
* <ul>
* <li>If {@code v} is null then {@code h} is 0.</li>
* <li>If {@code v} is a primitive array then {@code h} is computed using
* the appropriate overloading of {@code java.util.Arrays.hashCode}.</li>
* <li>If {@code v} is an object array then {@code h} is computed using
* {@link java.util.Arrays#deepHashCode(Object[]) Arrays.deepHashCode}.</li>
* <li>Otherwise {@code h} is {@code v.hashCode()}.</li>
* </ul>
*
* @return A hash code value for this object.
*
*/
// XXXX TODO: This is not very efficient!
// Note: this Javadoc is copied from javax.management.Descriptor
// due to 6369229.
public synchronized int hashCode() {
return new ImmutableDescriptor(descriptorMap).hashCode();
}
Returns true if all of the fields have legal values given their
names.
This implementation does not support interoperating with a directory
or lookup service. Thus, conforming to the specification, no checking is
done on the "export" field.
Otherwise this implementation returns false if:
- name and descriptorType fieldNames are not defined, or
null, or empty, or not String
- class, role, getMethod, setMethod fieldNames, if defined,
are null or not String
- persistPeriod, currencyTimeLimit, lastUpdatedTimeStamp,
lastReturnedTimeStamp if defined, are null, or not a Numeric
String or not a Numeric Value >= -1
- log fieldName, if defined, is null, or not a Boolean or
not a String with value "t", "f", "true", "false". These String
values must not be case sensitive.
- visibility fieldName, if defined, is null, or not a
Numeric String or a not Numeric Value >= 1 and <= 4
- severity fieldName, if defined, is null, or not a Numeric
String or not a Numeric Value >= 0 and <= 6
- persistPolicy fieldName, if defined, is null, or not one of
the following strings:
"OnUpdate", "OnTimer", "NoMoreOftenThan", "OnUnregister", "Always",
"Never". These String values must not be case sensitive.
Throws: - RuntimeOperationsException – If the validity checking
fails for any reason, this exception will be thrown.
/**
* Returns true if all of the fields have legal values given their
* names.
* <P>
* This implementation does not support interoperating with a directory
* or lookup service. Thus, conforming to the specification, no checking is
* done on the <i>"export"</i> field.
* <P>
* Otherwise this implementation returns false if:
* <P>
* <UL>
* <LI> name and descriptorType fieldNames are not defined, or
* null, or empty, or not String
* <LI> class, role, getMethod, setMethod fieldNames, if defined,
* are null or not String
* <LI> persistPeriod, currencyTimeLimit, lastUpdatedTimeStamp,
* lastReturnedTimeStamp if defined, are null, or not a Numeric
* String or not a Numeric Value >= -1
* <LI> log fieldName, if defined, is null, or not a Boolean or
* not a String with value "t", "f", "true", "false". These String
* values must not be case sensitive.
* <LI> visibility fieldName, if defined, is null, or not a
* Numeric String or a not Numeric Value >= 1 and <= 4
* <LI> severity fieldName, if defined, is null, or not a Numeric
* String or not a Numeric Value >= 0 and <= 6<br>
* <LI> persistPolicy fieldName, if defined, is null, or not one of
* the following strings:<br>
* "OnUpdate", "OnTimer", "NoMoreOftenThan", "OnUnregister", "Always",
* "Never". These String values must not be case sensitive.<br>
* </UL>
*
* @exception RuntimeOperationsException If the validity checking
* fails for any reason, this exception will be thrown.
*/
public synchronized boolean isValid() throws RuntimeOperationsException {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"isValid()", "Entry");
}
// verify that the descriptor is valid, by iterating over each field...
Set returnedSet = descriptorMap.entrySet();
if (returnedSet == null) { // null descriptor, not valid
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"isValid()", "Returns false (null set)");
}
return false;
}
// must have a name and descriptor type field
String thisName = (String)(this.getFieldValue("name"));
String thisDescType = (String)(getFieldValue("descriptorType"));
if ((thisName == null) || (thisDescType == null) ||
(thisName.equals("")) || (thisDescType.equals(""))) {
return false;
}
// According to the descriptor type we validate the fields contained
for (Iterator iter = returnedSet.iterator(); iter.hasNext();) {
Map.Entry currElement = (Map.Entry) iter.next();
if (currElement != null) {
if (currElement.getValue() != null) {
// validate the field valued...
if (validateField((currElement.getKey()).toString(),
(currElement.getValue()).toString())) {
continue;
} else {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"isValid()",
"Field " + currElement.getKey() + "=" +
currElement.getValue() + " is not valid");
}
return false;
}
}
}
}
// fell through, all fields OK
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"isValid()", "Returns true");
}
return true;
}
// worker routine for isValid()
// name is not null
// descriptorType is not null
// getMethod and setMethod are not null
// persistPeriod is numeric
// currencyTimeLimit is numeric
// lastUpdatedTimeStamp is numeric
// visibility is 1-4
// severity is 0-6
// log is T or F
// role is not null
// class is not null
// lastReturnedTimeStamp is numeric
private boolean validateField(String fldName, Object fldValue) {
if ((fldName == null) || (fldName.equals("")))
return false;
String SfldValue = "";
boolean isAString = false;
if ((fldValue != null) && (fldValue instanceof java.lang.String)) {
SfldValue = (String) fldValue;
isAString = true;
}
boolean nameOrDescriptorType =
(fldName.equalsIgnoreCase("Name") ||
fldName.equalsIgnoreCase("DescriptorType"));
if (nameOrDescriptorType ||
fldName.equalsIgnoreCase("SetMethod") ||
fldName.equalsIgnoreCase("GetMethod") ||
fldName.equalsIgnoreCase("Role") ||
fldName.equalsIgnoreCase("Class")) {
if (fldValue == null || !isAString)
return false;
if (nameOrDescriptorType && SfldValue.equals(""))
return false;
return true;
} else if (fldName.equalsIgnoreCase("visibility")) {
long v;
if ((fldValue != null) && (isAString)) {
v = toNumeric(SfldValue);
} else if (fldValue instanceof java.lang.Integer) {
v = ((Integer)fldValue).intValue();
} else return false;
if (v >= 1 && v <= 4)
return true;
else
return false;
} else if (fldName.equalsIgnoreCase("severity")) {
long v;
if ((fldValue != null) && (isAString)) {
v = toNumeric(SfldValue);
} else if (fldValue instanceof java.lang.Integer) {
v = ((Integer)fldValue).intValue();
} else return false;
return (v >= 0 && v <= 6);
} else if (fldName.equalsIgnoreCase("PersistPolicy")) {
return (((fldValue != null) && (isAString)) &&
( SfldValue.equalsIgnoreCase("OnUpdate") ||
SfldValue.equalsIgnoreCase("OnTimer") ||
SfldValue.equalsIgnoreCase("NoMoreOftenThan") ||
SfldValue.equalsIgnoreCase("Always") ||
SfldValue.equalsIgnoreCase("Never") ||
SfldValue.equalsIgnoreCase("OnUnregister")));
} else if (fldName.equalsIgnoreCase("PersistPeriod") ||
fldName.equalsIgnoreCase("CurrencyTimeLimit") ||
fldName.equalsIgnoreCase("LastUpdatedTimeStamp") ||
fldName.equalsIgnoreCase("LastReturnedTimeStamp")) {
long v;
if ((fldValue != null) && (isAString)) {
v = toNumeric(SfldValue);
} else if (fldValue instanceof java.lang.Number) {
v = ((Number)fldValue).longValue();
} else return false;
return (v >= -1);
} else if (fldName.equalsIgnoreCase("log")) {
return ((fldValue instanceof java.lang.Boolean) ||
(isAString &&
(SfldValue.equalsIgnoreCase("T") ||
SfldValue.equalsIgnoreCase("true") ||
SfldValue.equalsIgnoreCase("F") ||
SfldValue.equalsIgnoreCase("false") )));
}
// default to true, it is a field we aren't validating (user etc.)
return true;
}
Returns an XML String representing the descriptor.
The format is not defined, but an implementation must ensure that the string returned by this method can be used to build an equivalent descriptor when instantiated using the constructor
DescriptorSupport(String inStr)
.
Fields which are not String objects will have toString()
called on them to create the value. The value will be
enclosed in parentheses. It is not guaranteed that you can
reconstruct these objects unless they have been
specifically set up to support toString() in a meaningful
format and have a matching constructor that accepts a
String in the same format.
If the descriptor is empty the following String is
returned: <Descriptor></Descriptor>
Throws: - RuntimeOperationsException – for illegal value for
field Names or field Values. If the XML formatted string
construction fails for any reason, this exception will be
thrown.
Returns: the XML string.
/**
* <p>Returns an XML String representing the descriptor.</p>
*
* <p>The format is not defined, but an implementation must
* ensure that the string returned by this method can be
* used to build an equivalent descriptor when instantiated
* using the constructor {@link #DescriptorSupport(String)
* DescriptorSupport(String inStr)}.</p>
*
* <p>Fields which are not String objects will have toString()
* called on them to create the value. The value will be
* enclosed in parentheses. It is not guaranteed that you can
* reconstruct these objects unless they have been
* specifically set up to support toString() in a meaningful
* format and have a matching constructor that accepts a
* String in the same format.</p>
*
* <p>If the descriptor is empty the following String is
* returned: <Descriptor></Descriptor></p>
*
* @return the XML string.
*
* @exception RuntimeOperationsException for illegal value for
* field Names or field Values. If the XML formatted string
* construction fails for any reason, this exception will be
* thrown.
*/
public synchronized String toXMLString() {
final StringBuilder buf = new StringBuilder("<Descriptor>");
Set returnedSet = descriptorMap.entrySet();
for (Iterator iter = returnedSet.iterator(); iter.hasNext(); ) {
final Map.Entry currElement = (Map.Entry) iter.next();
final String name = currElement.getKey().toString();
Object value = currElement.getValue();
String valueString = null;
/* Set valueString to non-null if and only if this is a string that
cannot be confused with the encoding of an object. If it
could be so confused (surrounded by parentheses) then we
call makeFieldValue as for any non-String object and end
up with an encoding like "(java.lang.String/(thing))". */
if (value instanceof String) {
final String svalue = (String) value;
if (!svalue.startsWith("(") || !svalue.endsWith(")"))
valueString = quote(svalue);
}
if (valueString == null)
valueString = makeFieldValue(value);
buf.append("<field name=\"").append(name).append("\" value=\"")
.append(valueString).append("\"></field>");
}
buf.append("</Descriptor>");
return buf.toString();
}
private static final String[] entities = {
"  ",
"\""",
"<<",
">>",
"&&",
"\r ",
"\t	",
"\n ",
"\f",
};
private static final Map<String,Character> entityToCharMap =
new HashMap<String,Character>();
private static final String[] charToEntityMap;
static {
char maxChar = 0;
for (int i = 0; i < entities.length; i++) {
final char c = entities[i].charAt(0);
if (c > maxChar)
maxChar = c;
}
charToEntityMap = new String[maxChar + 1];
for (int i = 0; i < entities.length; i++) {
final char c = entities[i].charAt(0);
final String entity = entities[i].substring(1);
charToEntityMap[c] = entity;
entityToCharMap.put(entity, new Character(c));
}
}
private static boolean isMagic(char c) {
return (c < charToEntityMap.length && charToEntityMap[c] != null);
}
/*
* Quote the string so that it will be acceptable to the (String)
* constructor. Since the parsing code in that constructor is fairly
* stupid, we're obliged to quote apparently innocuous characters like
* space, <, and >. In a future version, we should rewrite the parser
* and only quote " plus either \ or & (depending on the quote syntax).
*/
private static String quote(String s) {
boolean found = false;
for (int i = 0; i < s.length(); i++) {
if (isMagic(s.charAt(i))) {
found = true;
break;
}
}
if (!found)
return s;
final StringBuilder buf = new StringBuilder();
for (int i = 0; i < s.length(); i++) {
char c = s.charAt(i);
if (isMagic(c))
buf.append(charToEntityMap[c]);
else
buf.append(c);
}
return buf.toString();
}
private static String unquote(String s) throws XMLParseException {
if (!s.startsWith("\"") || !s.endsWith("\""))
throw new XMLParseException("Value must be quoted: <" + s + ">");
final StringBuilder buf = new StringBuilder();
final int len = s.length() - 1;
for (int i = 1; i < len; i++) {
final char c = s.charAt(i);
final int semi;
final Character quoted;
if (c == '&'
&& (semi = s.indexOf(';', i + 1)) >= 0
&& ((quoted = entityToCharMap.get(s.substring(i, semi+1)))
!= null)) {
buf.append(quoted);
i = semi;
} else
buf.append(c);
}
return buf.toString();
}
Make the string that will go inside "..." for a value that is not
a plain String.
Throws: - RuntimeOperationsException – if the value cannot be encoded.
/**
* Make the string that will go inside "..." for a value that is not
* a plain String.
* @throws RuntimeOperationsException if the value cannot be encoded.
*/
private static String makeFieldValue(Object value) {
if (value == null)
return "(null)";
Class<?> valueClass = value.getClass();
try {
valueClass.getConstructor(String.class);
} catch (NoSuchMethodException e) {
final String msg =
"Class " + valueClass + " does not have a public " +
"constructor with a single string arg";
final RuntimeException iae = new IllegalArgumentException(msg);
throw new RuntimeOperationsException(iae,
"Cannot make XML descriptor");
} catch (SecurityException e) {
// OK: we'll pretend the constructor is there
// too bad if it's not: we'll find out when we try to
// reconstruct the DescriptorSupport
}
final String quotedValueString = quote(value.toString());
return "(" + valueClass.getName() + "/" + quotedValueString + ")";
}
/*
* Parse a field value from the XML produced by toXMLString().
* Given a descriptor XML containing <field name="nnn" value="vvv">,
* the argument to this method will be "vvv" (a string including the
* containing quote characters). If vvv begins and ends with parentheses,
* then it may contain:
* - the characters "null", in which case the result is null;
* - a value of the form "some.class.name/xxx", in which case the
* result is equivalent to `new some.class.name("xxx")';
* - some other string, in which case the result is that string,
* without the parentheses.
*/
private static Object parseQuotedFieldValue(String s)
throws XMLParseException {
s = unquote(s);
if (s.equalsIgnoreCase("(null)"))
return null;
if (!s.startsWith("(") || !s.endsWith(")"))
return s;
final int slash = s.indexOf('/');
if (slash < 0) {
// compatibility: old code didn't include class name
return s.substring(1, s.length() - 1);
}
final String className = s.substring(1, slash);
final Constructor<?> constr;
try {
ReflectUtil.checkPackageAccess(className);
final ClassLoader contextClassLoader =
Thread.currentThread().getContextClassLoader();
final Class<?> c =
Class.forName(className, false, contextClassLoader);
constr = c.getConstructor(new Class[] {String.class});
} catch (Exception e) {
throw new XMLParseException(e,
"Cannot parse value: <" + s + ">");
}
final String arg = s.substring(slash + 1, s.length() - 1);
try {
return constr.newInstance(new Object[] {arg});
} catch (Exception e) {
final String msg =
"Cannot construct instance of " + className +
" with arg: <" + s + ">";
throw new XMLParseException(e, msg);
}
}
Returns a human readable string representing the
descriptor. The string will be in the format of
"fieldName=fieldValue,fieldName2=fieldValue2,..."
If there are no fields in the descriptor, then an empty String
is returned.
If a fieldValue is an object then the toString() method is
called on it and its returned value is used as the value for
the field enclosed in parenthesis.
Throws: - RuntimeOperationsException – for illegal value for
field Names or field Values. If the descriptor string fails
for any reason, this exception will be thrown.
/**
* Returns <pv>a human readable string representing the
* descriptor</pv>. The string will be in the format of
* "fieldName=fieldValue,fieldName2=fieldValue2,..."<br>
*
* If there are no fields in the descriptor, then an empty String
* is returned.<br>
*
* If a fieldValue is an object then the toString() method is
* called on it and its returned value is used as the value for
* the field enclosed in parenthesis.
*
* @exception RuntimeOperationsException for illegal value for
* field Names or field Values. If the descriptor string fails
* for any reason, this exception will be thrown.
*/
public synchronized String toString() {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"toString()", "Entry");
}
String respStr = "";
String[] fields = getFields();
if ((fields == null) || (fields.length == 0)) {
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"toString()", "Empty Descriptor");
}
return respStr;
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"toString()", "Printing " + fields.length + " fields");
}
for (int i=0; i < fields.length; i++) {
if (i == (fields.length - 1)) {
respStr = respStr.concat(fields[i]);
} else {
respStr = respStr.concat(fields[i] + ", ");
}
}
if (MODELMBEAN_LOGGER.isLoggable(Level.FINEST)) {
MODELMBEAN_LOGGER.logp(Level.FINEST,
DescriptorSupport.class.getName(),
"toString()", "Exit returning " + respStr);
}
return respStr;
}
// utility to convert to int, returns -2 if bogus.
private long toNumeric(String inStr) {
long result = -2;
try {
result = java.lang.Long.parseLong(inStr);
} catch (Exception e) {
return -2;
}
return result;
}
Deserializes a DescriptorSupport
from an ObjectInputStream
. /**
* Deserializes a {@link DescriptorSupport} from an {@link
* ObjectInputStream}.
*/
private void readObject(ObjectInputStream in)
throws IOException, ClassNotFoundException {
ObjectInputStream.GetField fields = in.readFields();
Map<String, Object> descriptor = cast(fields.get("descriptor", null));
init(null);
if (descriptor != null) {
descriptorMap.putAll(descriptor);
}
}
Serializes a DescriptorSupport
to an ObjectOutputStream
. /**
* Serializes a {@link DescriptorSupport} to an {@link ObjectOutputStream}.
*/
/* If you set jmx.serial.form to "1.2.0" or "1.2.1", then we are
bug-compatible with those versions. Specifically, field names
are forced to lower-case before being written. This
contradicts the spec, which, though it does not mention
serialization explicitly, does say that the case of field names
is preserved. But in 1.2.0 and 1.2.1, this requirement was not
met. Instead, field names in the descriptor map were forced to
lower case. Those versions expect this to have happened to a
descriptor they deserialize and e.g. getFieldValue will not
find a field whose name is spelt with a different case.
*/
private void writeObject(ObjectOutputStream out) throws IOException {
ObjectOutputStream.PutField fields = out.putFields();
boolean compat = "1.0".equals(serialForm);
if (compat)
fields.put("currClass", currClass);
/* Purge the field "targetObject" from the DescriptorSupport before
* serializing since the referenced object is typically not
* serializable. We do this here rather than purging the "descriptor"
* variable below because that HashMap doesn't do case-insensitivity.
* See CR 6332962.
*/
SortedMap<String, Object> startMap = descriptorMap;
if (startMap.containsKey("targetObject")) {
startMap = new TreeMap<String, Object>(descriptorMap);
startMap.remove("targetObject");
}
final HashMap<String, Object> descriptor;
if (compat || "1.2.0".equals(serialForm) ||
"1.2.1".equals(serialForm)) {
descriptor = new HashMap<String, Object>();
for (Map.Entry<String, Object> entry : startMap.entrySet())
descriptor.put(entry.getKey().toLowerCase(), entry.getValue());
} else
descriptor = new HashMap<String, Object>(startMap);
fields.put("descriptor", descriptor);
out.writeFields();
}
}