• Home
• com.diffplug.spotless
• spotless-ext-greclipse
• 2.3.0
• source code

×

Project download

Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance.
Project price only 1 $

You can buy this project and download/modify it how often you want.

spotless.spotless-ext-greclipse.2.3.0.source-code.R36_buildnotes_jdt-core.html Maven / Gradle / Ivy

   
   
   
   



  

    

      jdt core - build notes 3.6 stream
    
  
  
Java development tools core
  
 
  

  	

	  
	  Here are the build notes for the Eclipse JDT/Core plug-in project
	  org.eclipse.jdt.core,
	  describing bug resolution and substantial changes in the HEAD branch.
	  For more information on 3.6 planning, please refer to JDT/Core release plan,
	  the next milestone plan,
	  the overall official plan,
	  or the build schedule.
	  This present document covers all changes since Release 3.5 (also see a summary of API changes).
	  
Maintenance of previous releases of JDT/Core is performed in parallel branches:
		  R3.5.x,
		  R3.4.x,
		  R3.3.x,
		  R3.2.x,
		  R3.1.x,
		  R3.0.x,
		  R2.1.x,
		  R2.0.x,
		  R1.0.x.
	  
	
  



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6 - June 3, 2010 - 3.6.0

Project org.eclipse.jdt.core v_A58
(cvs).
What's new in this drop

Problem Reports Fixed
315568
improve Javadoc of SearchPattern#createPattern(String, int, int, int)



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC4 - June 3, 2010 - 3.6.0 RC4

Project org.eclipse.jdt.core v_A57
(cvs).
What's new in this drop

Use default disabling/enabling tags in the samples of the Javadoc comments
of the formatter constants FORMATTER_DISABLING_TAG and
FORMATTER_ENABLING_TAG.
Fixed minor javadoc issues of createStrictHierarchyScope().


Problem Reports Fixed
314709
Clarify -encoding <encoding name> in jdt_api_compile.htm



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC3 - May 27, 2010 - 3.6.0 RC3

Project org.eclipse.jdt.core v_A56
(cvs).
What's new in this drop

Problem Reports Fixed
313890
Migration guide to 3.6 for containers with MANIFEST-referred entries

313965
Breaking change in classpath container API



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC3 - May 25, 2010

Project org.eclipse.jdt.core v_A55
(cvs).
What's new in this drop

Problem Reports Fixed
313706
Replace ie. with i.e. in jdt.core documentation



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC2 - May 20, 2010 - 3.6.0 RC2

Project org.eclipse.jdt.core v_A54
(cvs).
What's new in this drop


Added a new preference to force the formatter to try to keep nested expressions on one line.

This new preference is controlled with the option:
DefaultCodeFormatterConstants.FORMATTER_WRAP_OUTER_EXPRESSIONS_WHEN_NESTED
/**
 * FORMATTER / Option to wrap outer expressions in nested expressions
 *     - option id:         "org.eclipse.jdt.core.formatter.wrap_outer_expressions_when_nested"
 *     - possible values:   { TRUE, FALSE }
 *     - default:           TRUE
 *
 * This option changes the formatter behavior when nested method calls are encountered.
 * Since 3.6, the formatter tries to wrap outermost method calls first to have a better output.
 * For example, let's say we are using the Eclipse built-in profile with a max line width=40+space for tab policy.
 * Then consider the following snippet:
 *
 * public class X01 {
 *     void test() {
 *         foo(bar(1, 2, 3, 4), bar(5, 6, 7, 8));
 *     }
 * }
 *
 * With this new strategy, the formatter will wrap the line earlier, between the arguments of the message call
 * for this example, and then it will allow to keep each nested call on a single line.
 * Hence, the output will be:
 *
 * public class X01 {
 *     void test() {
 *         foo(bar(1, 2, 3, 4),
 *             bar(5, 6, 7, 8));
 *     }
 * }
 *
 * Important notes:
 * 1. This new behavior is automatically activated (i.e. the default value for this preference is {@link #TRUE}).
 *    If the backward compatibility regarding previous versions' formatter behavior (i.e. before 3.6 version) is necessary,
 *    then the preference needs to be set to {@link #FALSE} to retrieve the previous formatter behavior.
 * 2. The new strategy currently only applies to nested method calls, but might be extended to other nested expressions in future versions
 * 
 * @see #TRUE
 * @see #FALSE
 * @since 3.6
 */

See bug 313524 for more details.



Problem Reports Fixed
313524
[formatter] Add preference for improved lines wrapping in nested method calls

313109
@SuppressWarnings on multiple locals is marked unnecessary if any local is never used



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC1 - May 12, 2010 - 3.6.0 RC1

Project org.eclipse.jdt.core v_A53
(cvs).
What's new in this drop

Problem Reports Fixed
312326
IllegalArgumentException using open type dialog

310159
Hang in JavaModel.getExternalTarget(JavaModel.java:333)



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC1 - May 11, 2010

Project org.eclipse.jdt.core v_A52
(cvs).
What's new in this drop

Problem Reports Fixed
157847
NPE in WildcardBinding.computeUniqueKey during code assist

311849
[quick fix] @SuppressWarnings does not work as expected inside a for loop



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC1 - May 8, 2010

Project org.eclipse.jdt.core v_A51
(cvs).
What's new in this drop

Problem Reports Fixed
306170
[perfs] Regression for FullSourceWorkspaceTypeHierarchyTests#testPerfAllTypes()

298844
[formatter] New lines in empty method body wrong behavior

311864
[formatter] NPE with empty {@code }



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC1 - May 6, 2010

Project org.eclipse.jdt.core v_A50
(cvs).
What's new in this drop


Added a new preference to switch on/off the usage of the disabling/enabling tags of the formatter.

This new preference is controlled with the option:
DefaultCodeFormatterConstants.FORMATTER_USE_ON_OFF_TAGS
/**
 * FORMATTER / Option to use the disabling and enabling tags defined respectively by the {@link #FORMATTER_DISABLING_TAG} and the {@link #FORMATTER_ENABLING_TAG} options.
 *     - option id:         "org.eclipse.jdt.core.formatter.use_on_off_tags"
 *     - possible values:   TRUE / FALSE
 *     - default:           FALSE
 * 
 * @since 3.6
 */

See bug 311582 for more details.



Problem Reports Fixed
311617
[formatter] provide default tags to enable/disable formatter

311582
[formatter] Master switch to enable/disable on/off tags

307040
Search Job with HierarchyScope on Object does not cancel

302295
After associating source folder with rt.jar project refresh takes exceedingly long time.

311048
AbortCompilation propagated from CompilationUnitProblemFinder.process()



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6RC1 - May 4, 2010

Project org.eclipse.jdt.core v_A49
(cvs).
What's new in this drop

Problem Reports Fixed
310811
[perfs] Big regression on FullSourceWorkspaceFormatterTests#testFormatDefaultBigFile()



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - April 25, 2010 - 3.6.0 M7

Project org.eclipse.jdt.core v_A48
(cvs).
What's new in this drop

Problem Reports Fixed
310330
Add multiple encoding support for the batch compiler



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - April 25, 2010

Project org.eclipse.jdt.core v_A47
(cvs).
What's new in this drop

Problem Reports Fixed
309835
[formatter] adds blank lines on each run

310213
AIOOBE in IndexSelector.initializeIndexLocations()

309966
IType#getKey() does not work for unresolved local ITypes



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - April 23, 2010

Project org.eclipse.jdt.core v_A46
(cvs).
What's new in this drop

Problem Reports Fixed
59891
[formatter] improve lines wrapping in nested method calls

306172
[perfs] Invalid test duration for FullSourceWorkspaceTypeHierarchyTests#testPerSuperTypes()

267091
[content assist] After 'implements' interface members are not proposed

261534
content assist after instanceof should also work after &&

308980
[content assist]An initializer inside a non-array field declaration confuses content assist

310002
ToolFactory.createScanner(..) should use workspace compliance

291528
Synchronize project warning/error settings to build.properties

309787
Extension point "org.eclipse.jdt.core.codeFormatter" is ignored

244820
Content assist after 'instanceof' should also work in assignment

309706
[formatter] doesn't work when code has three semicolons side by side



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - April 20, 2010

Project org.eclipse.jdt.core v_A45
(cvs).
What's new in this drop

Problem Reports Fixed
305037
missing story for attributes of referenced JARs in classpath containers

305116
[index] Improve performance of indexes results tables

236306
[content assist] for method invocation in variable initializer should not guess variable

302865
Issue with "import" a class and "import static" a method with the same name

309022
[ImportRewrite] Add Import wrongly removes import for nested type



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - April 13, 2010

Project org.eclipse.jdt.core v_A44
(cvs).
What's new in this drop

Problem Reports Fixed
308754
CompilationUnit.rewrite messes up .class-literal in annotation instead of changing class to interface

306519
JavaCore#getReferencedClasspathEntries(IClasspathEntry, IJavaProject) should allow null project

308428
Possible problem to get corrections with surrogate characters

307295
Task tags and task priorities

308476
Test ClasspathTests#testBug308150 fails on all platforms

305043
Internal error during classpath init

307486
DBCS3.6: Fail to propose Ext-B labels with content assist in Java Editor

308256
DiagnosticListener always supplies Diagnostic.getSource()==null

308356
codeSelect(..) doesn't work for local variable with surrogate in name

308245
Valid code fails to compile in 3.6

307885
Error message for instanceof <parameterized type> wrong arguments

249704
[code assist] autocomplete with anonymous classes does stop working

308150
JAR with invalid Class-Path entry in MANIFEST.MF crashes the project



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - April 6, 2010

Project org.eclipse.jdt.core v_A43
(cvs).
What's new in this drop

Problem Reports Fixed
306223
[search] Searching for annotation references report all type references

292087
anonymous class in array member initializer confuses content assist

307337
[content assist] Default constructor should not be proposed for anonymous types

306568
[ImportRewrite] Add Import does not work for nested type when package is on-demand imported



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - March 30, 2010

Project org.eclipse.jdt.core v_A42
(cvs).
What's new in this drop

Problem Reports Fixed
202634
[codeassist] missing super proposal in specific source

304394
IJavaElement#getAttachedJavadoc(IProgressMonitor) should support referenced entries

305122
FUP of 302949

306917
Exception occurred during compilation unit conversion:

306196
[search] NPE while searching for annotation references in rt.jar of JRE 6.0

288658
[compiler][1.5] Annotations visibility issues



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - March 23, 2010

Project org.eclipse.jdt.core v_A41
(cvs).
What's new in this drop

Problem Reports Fixed
305518
[formatter] Line inside <pre> tag is wrongly indented by one space when starting just after the star

295825
[formatter] Commentaries are running away after formatting are used

306477
Indexer(?) fails to recognise enum as a type

305830
[formatter] block comment should not be formatted when a non-nls tag is on the same line

300031
The deprecation warning for a type should not include the package name

306078
Navigate to Inaccessible Field

209479
infinite loop in BindingKey when signatures are invalid

293558
[quick assist] "Invert if statement" fails when comment follows

182459
[compiler] Inconsistent error range for unresolved field

192233
[AST] CompilationUnit.rewrite() removes whitespace between return type and method name

306073
ASTRewrite Javadoc wrongly talks about getTargetSourceRangeComputer

305001
Exception occurred in listener of Java element change notification

305590
Redundant null check false-positive

305755
Remove deprecated API that has been added for 3.6



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M7 - March 16, 2010

Project org.eclipse.jdt.core v_A40
(cvs).
What's new in this drop

Problem Reports Fixed
305371
[formatter] Unexpected indentation of line comment

305281
[formatter] Turning off formatting changes comment's formatting



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M6 - March 9, 2010 - 3.6.0 M6

Project org.eclipse.jdt.core v_A39
(cvs).
What's new in this drop

Problem Reports Fixed
289057
Java Content Assist taking too long

303830
"X cannot be resolved or is not a field" erroneously reported

235658
Valid identifier unrecognized.

304841
[search] NPE in IndexSelector.initializeIndexLocations

295866
FormalParameter in JDT DOM/AST documentation

304817
Review documentation of ASTParser class



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M6 - March 5, 2010

Project org.eclipse.jdt.core v_A38
(cvs).
What's new in this drop


Added two new preferences to allow to disable the formatter in a section of the code.
These two preference define respectively the tag which disables the formatting
and the tag which re-enable it.

These new preferences are controlled with the options:
DefaultCodeFormatterConstants.FORMATTER_DISABLING_TAG
DefaultCodeFormatterConstants.FORMATTER_ENABLING_TAG
/**
 * FORMATTER / Option to define the tag to put in a comment to disable the formatting.
 * See the {@link #FORMATTER_ENABLING_TAG} option to re-enable it.
 *     - option id:         "org.eclipse.jdt.core.formatter.disabling_tag"
 *     - possible values:   String, with constraints mentioned below
 *     - default:           ""
 * 
 * Note that:
 * 
 * 1. The tag name will be trimmed. Hence if it does contain white spaces
 *    at the beginning or at the end, they will not be taken into account while
 *    searching for the tag in the comments
 * 2. If a tag is starting with a letter or digit, then it cannot be leaded by
 *    another letter or digit to be recognized
 *    ("ToDisableFormatter" will not be recognized as a disabling tag
 *    "DisableFormatter", but "Re:DisableFormatter"
 *    will be detected for either tag "DisableFormatter" or
 *    ":DisableFormatter").
 *    Respectively, a tag ending with a letter or digit cannot be followed by a letter
 *    or digit to be recognized ("DisableFormatter1" will not be
 *    recognized as a disabling tag "DisableFormatter", but
 *    "DisableFormatter:1" will be detected either for tag
 *    "DisableFormatter" or "DisableFormatter:")
 * 3. As soon as the formatter encounters the defined disabling tag, it stops to
 *    format the code from the beginning of the comment including this tag. If it
 *    was already disabled, the tag has no special effect.
 *    For example, the second defined enabling tag "disable-formatter"
 *    in the following snippet is not necessary as the formatter was already disabled
 *    since the first one:
 *     class X {
 *     // disable-formatter
 *     void foo1() {}
 *     // disable-formatter
 *     void foo2() {}
 *     void bar1() {}
 *     void bar2() {}
 *     }
 *
 * 4. If no enabling tag is found by the formatter after the disabling tag, then
 *    the end of the snippet won't be formatted.
 *    For example, when a disabling tag is put at the beginning of the code, then
 *    the entire content of a compilation unit is not formatted:
 *     // disable-formatter
 *     class X {
 *     void foo1() {}
 *     void foo2() {}
 *     void bar1() {}
 *     void bar2() {}
 *     }
 * 
 * 5. If a mix of disabling and enabling tags is done in the same comment, then
 *    the formatter will only take into account the last encountered tag in the
 *    comment.
 *    For example, in the following snippet, the formatter will be disabled after
 *    the comment:
 *     class X {
 *     /*
 *      * This is a comment with a mix of disabling and enabling tags:
 *      *  - disable-formatter
 *      *  - enable-formatter
 *      *  - disable-formatter
 *      * The formatter will stop to format from the beginning of this comment...
 *      */
 *     void foo() {}
 *     void bar() {}
 *     }
 * 
 * 6. The tag cannot include newline character (i.e. '\n') but it can have white spaces.
 *    E.g. "format: off" is a valid disabling tag
 *    In the future, newlines may be used to support multiple disabling tags.
 * 
 * @since 3.6
 */

/**
 * FORMATTER / Option to define the tag to put in a comment to re-enable the
 * formatting after it has been disabled (see {@link #FORMATTER_DISABLING_TAG})
 *     - option id:         "org.eclipse.jdt.core.formatter.enabling_tag"
 *     - possible values:   String, with constraints mentioned below
 *     - default:           ""
 * 
 * Note that:
 * 
 * 1. The tag name will be trimmed. Hence if it does contain white spaces
 *    at the beginning or at the end, they will not be taken into account while
 *    searching for the tag in the comments
 * 2. If a tag is starting with a letter or digit, then it cannot be leaded by
 *    another letter or digit to be recognized
 *    ("ReEnableFormatter" will not be recognized as an enabling tag
 *    "EnableFormatter", but "Re:EnableFormatter"
 *    will be detected for either tag "EnableFormatter" or
 *    ":EnableFormatter").
 *    Respectively, a tag ending with a letter or digit cannot be followed by a letter
 *    or digit to be recognized ("EnableFormatter1" will not be
 *    recognized as an enabling tag "EnableFormatter", but
 *    "EnableFormatter:1" will be detected either for tag
 *    "EnableFormatter" or "EnableFormatter:")
 * 3. As soon as the formatter encounters the defined enabling tag, it re-starts
 *    to format the code just after the comment including this tag. If it was already
 *    active, i.e. already re-enabled or never disabled, the tag has no special effect.
 *    For example, the defined enabling tag "enable-formatter"
 *    in the following snippet is not necessary as the formatter has never been
 *    disabled:
 *     class X {
 *     void foo1() {}
 *     void foo2() {}
 *     // enable-formatter
 *     void bar1() {}
 *     void bar2() {}
 *     }
 * 
 *    Or, in the following other snippet, the second enabling tag is not necessary as
 *    the formatting will have been re-enabled by the first one:
 *     class X {
 *     // disable-formatter
 *     void foo1() {}
 *     void foo2() {}
 *     // enable-formatter
 *     void bar1() {}
 *     // enable-formatter
 *     void bar2() {}
 *     }
 * 
 * 4. If a mix of disabling and enabling tags is done in the same comment, then
 *    the formatter will only take into account the last encountered tag in the
 *    comment.
 *    For example, in the following snippet, the formatter will be re-enabled after
 *    the comment:
 *     // disable-formatter
 *     class X {
 *     /*
 *      * This is a comment with a mix of disabling and enabling tags:
 *      *  - enable-formatter
 *      *  - disable-formatter
 *      *  - enable-formatter
 *      * The formatter will restart to format after this comment...
 *      */
 *     void foo() {}
 *     void bar() {}
 *     }
 * 
 * 5. The tag cannot include newline character (i.e. '\n') but it can have white spaces.
 *    E.g. "format: on" is a valid enabling tag
 *    In the future, newlines may be used to support multiple enabling tags.
 * 
 * @since 3.6
 */

For example, the following snippet:
public class Test {
/* disable-formatter */
void     foo(    )      {	
				//      unformatted       area  	  
}
/* enable-formatter */
void     bar(    )      {	
				//      formatted       area  	  
}
}

formatted with disabling tags = "disable-formatter" and enabling tags
= "enable-formatter" produces the following output:
public class Test {

/* disable-formatter *	
void     foo(    )      {
				//      unformatted       area  	  
}
/* enable-formatter *
	void bar() {
		// formatted area
	}
}

See bug 27079 for more details.



Problem Reports Fixed
129804
[dom] Local variable bindings from ASTParser#createASTs(.., String[], .., ..) have no declaring method

304705
[formatter] Unexpected indentation of wrapped line comments when 'Never indent line comment on first column' preference is checked

304656
StringIndexOutOfBoundsException when using JDT dom methods to process sourcefile

304506
Task descriptions always have a space after the tag

304081
IJavaProject#isOnClasspath(IJavaElement) returns false for type from referenced JAR

304122
TypeBindings.getAnnotations() breaks interface

304416
VerifyError after compiling without preserve all locals

304529
[formatter] NPE when either the disabling or the enabling tag is not defined

27079
Tags for disabling/enabling code formatter (feature)

304316
NPE when javadoc URL is invalid



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M6 - March 2, 2010

Project org.eclipse.jdt.core v_A37
(cvs).
What's new in this drop

Added new configurable option to fix bug 295551:

/**
 * Compiler option ID: Further Determining the Effect of @SuppressWarnings if also
 * COMPILER_PB_SUPPRESS_WARNINGS is enabled.
 * When enabled, the @SuppressWarnings annotation can additionally be used to suppress 
 * optional compiler diagnostics that have been configured as ERROR.
 * When disabled, all @SuppressWarnings annotations only affects warnings.
 *
 * Option id: "org.eclipse.jdt.core.compiler.problem.suppressOptionalErrors"
 * Possible values: { "enabled", "disabled" }
 * Default: "disabled"
 * 
 * @since 3.6
 * @category CompilerOptionID
 */
public static final String COMPILER_PB_SUPPRESS_OPTIONAL_ERRORS = PLUGIN_ID + ".compiler.problem.suppressOptionalErrors";



Added a new formatter preferences to align method declaration.

This new preference is controlled with the option:
DefaultCodeFormatterConstants.FORMATTER_ALIGNMENT_FOR_METHOD_DECLARATION
/**
 * FORMATTER / Option for alignment of method declaration
 *     - option id:         "org.eclipse.jdt.core.formatter.alignment_for_method_declaration"
 *     - possible values:   values returned by createAlignmentValue(boolean, int, int) call
 *     - default:           createAlignmentValue(false, WRAP_NO_SPLIT, INDENT_DEFAULT)
 * 
 * @see #createAlignmentValue(boolean, int, int)
 * @since 3.6
 */

For example, the following snippet:
public class Test {
public final synchronized java.lang.String a_method_which_has_a_very_long_name() {
return null;
}
}

formatted with this preference activated as 'Wrap only when necessary', will
produce the following output:
public class Test {
	public final synchronized java.lang.String
			a_method_which_has_a_very_long_name() {
		return null;
	}
}

See bug 284789 for more details.

New API to fix bug 246594. See the bug for details.
org.eclipse.jdt.core.ITypeParameter
	/**
	 * Returns the signatures for this type parameter's bounds. The type parameter may have 
	 * been declared as part of a type or a method. The signatures represent only the individual 
	 * bounds and do not include the type variable name or the extends keyword.  
	 * The signatures may be either unresolved (for source types) or resolved (for binary types). 
	 * See {@link Signature} for details.
	 * 
	 * @return the signatures for the bounds of this formal type parameter
	 * @throws JavaModelException
	 *             if this element does not exist or if an exception occurs while accessing its corresponding resource.
	 * @see Signature
	 * @since 3.6
	 */
	String[] getBoundsSignatures() throws JavaModelException;



Added a new formatter preference to enable or disable the formatting of line
comments that start on the first column.

Note that the indentation of line comments will also be disabled when activating
this option, as otherwise the formatter could not produce stable outputs...

The default is to format these comments to have a backward compatible behavior.

This new preferences is controlled with the options:
DefaultCodeFormatterConstants.FORMATTER_COMMENT_FORMAT_LINE_COMMENT_STARTING_ON_FIRST_COLUMN
/**
 * FORMATTER / Option to format line comments that start on the first column
 *     - option id:         "org.eclipse.jdt.core.formatter.format_line_comment_starting_on_first_column"
 *     - possible values:   { TRUE, FALSE }
 *     - default:           TRUE
 * 
 * Note that this option is ignored if either the
 * {@link #FORMATTER_COMMENT_FORMAT_LINE_COMMENT} option has been set to
 * {@link #FALSE} or the formatter is created with the mode
 * {@link ToolFactory#M_FORMAT_NEW}.
 * 
 * @see #TRUE
 * @see #FALSE
 * @see ToolFactory#createCodeFormatter(Map, int)
 * @since 3.6
 */

For example, the following snippet:
public class X01 {
//    int	a  =   1;
//    int	b  =   2;
}

will be untouched by the formatter if both options are activated.
See bug 251133 for more details.

New API to fix bug 252431. See the bug for details.
org.eclipse.jdt.core.IClasspathEntry
	/**
	 * Returns the classpath entry that is making a reference to this classpath entry. For entry kinds 
	 * {@link #CPE_LIBRARY}, the return value is the entry that is representing the JAR that includes 
	 * this in the MANIFEST.MF file's Class-Path section. For entry kinds other than 
	 * {@link #CPE_LIBRARY}, this returns null. For those entries that are on the raw classpath already, 
	 * this returns null.
	 *
	 * It is possible that multiple library entries refer to the same entry
	 * via the MANIFEST.MF file. In those cases, this method returns the first classpath entry 
	 * that appears in the raw classpath. However, this does not mean that the other referencing 
	 * entries do not relate to their referenced entries. 
	 * See {@link JavaCore#getReferencedClasspathEntries(IClasspathEntry, IJavaProject)} for 
	 * more details.
	 * 
	 * @return the classpath entry that is referencing this entry or null if 
	 * 		not applicable.
	 * @since 3.6
	 */
	IClasspathEntry getReferencingEntry();


org.eclipse.jdt.core.IJavaProject
	/**
	 * Works similar to {@link #setRawClasspath(IClasspathEntry[], IPath, IProgressMonitor)} and 
	 * additionally allows persisting the given array of referenced entries for this project.
	 * The referenced entries and their attributes are stored in the .classpath file of this 
	 * project. For details on referenced entries, see 
	 * {@link JavaCore#getReferencedClasspathEntries(IClasspathEntry, IJavaProject)}
	 * and {@link IClasspathEntry#getReferencingEntry()}.
	 * 
	 * Since the referenced entries are stored in the .classpath file, clients can store additional 
	 * information that belong to these entries and retrieve them across sessions, though the referenced
	 * entries themselves may not be present in the raw classpath. By passing a null
	 * referencedEntries, clients can choose not to modify the already persisted referenced entries,
	 * which is fully equivalent to {@link #setRawClasspath(IClasspathEntry[], IPath, IProgressMonitor)}.
	 * If an empty array is passed as referencedEntries, the already persisted referenced entries, 
	 * if any, will be cleared. 
	 * 
	 * If there are duplicates of a referenced entry or if any of the referencedEntries 
	 * is already present in the raw classpath(entries) those referenced entries will 
	 * be excluded and not be persisted.
	 *
	 * @param entries a list of classpath entries
	 * @param referencedEntries the list of referenced classpath entries to be persisted
	 * @param outputLocation the default output location
	 * @param monitor the given progress monitor
	 * @exception JavaModelException if the classpath could not be set. Reasons include:
	 *  	This Java element does not exist (ELEMENT_DOES_NOT_EXIST)
	 *  	The classpath is being modified during resource change event notification (CORE_EXCEPTION)
	 *  	The classpath failed the validation check as defined by {@link JavaConventions#validateClasspath(IJavaProject, IClasspathEntry[], IPath)}
	 * @see IClasspathEntry
	 * @see #getReferencedClasspathEntries()
	 * @since 3.6
	 */
	void setRawClasspath(IClasspathEntry[] entries, IClasspathEntry[] referencedEntries, IPath outputLocation,
			IProgressMonitor monitor) throws JavaModelException;

	/**
	 * Returns the list of referenced classpath entries stored in the .classpath file of this 
	 * java project. Clients can store the referenced classpath entries using 
	 * {@link #setRawClasspath(IClasspathEntry[], IClasspathEntry[], IPath, IProgressMonitor)}
	 * If the client has not stored any referenced entries for this project, an empty array is returned.
	 *
	 * @throws JavaModelException
	 * @return an array of referenced classpath entries stored for this java project or an empty array if none
	 * 			stored earlier.
	 * @since 3.6
	 */
	IClasspathEntry[] getReferencedClasspathEntries() throws JavaModelException;
	

org.eclipse.jdt.core.IPackageFragmentRoot
	/**
	 * Returns the first resolved classpath entry that corresponds to this package fragment root.
	 * A resolved classpath entry is said to correspond to a root if the path of the resolved
	 * entry is equal to the root's path.
	 * 
	 * @return the first resolved classpath entry that corresponds to this package fragment root
	 * @throws JavaModelException if this element does not exist or if an
	 *		exception occurs while accessing its corresponding resource. 
	 * @since 3.6
	 */
	IClasspathEntry getResolvedClasspathEntry() throws JavaModelException;
	

org.eclipse.jdt.core.JavaCore
	/**
	 * Returns an array of classpath entries that are referenced directly or indirectly 
	 * by a given classpath entry. For the entry kind {@link IClasspathEntry#CPE_LIBRARY}, 
	 * the method returns the libraries that are included in the Class-Path section of 
	 * the MANIFEST.MF file. If a referenced JAR file has further references to other library 
	 * entries, they are processed recursively and added to the list. For entry kinds other 
	 * than {@link IClasspathEntry#CPE_LIBRARY}, this method returns an empty array.
	 *
	 * If a referenced entry has already been stored 
	 * in the given project's .classpath, the stored attributes are populated in the corresponding
	 * referenced entry. For more details on storing referenced entries see
	 * see {@link IJavaProject#setRawClasspath(IClasspathEntry[], IClasspathEntry[], IPath, 
	 * IProgressMonitor)}. 
	 * 
	 * @param libraryEntry the library entry whose referenced entries are sought 
	 * @param project project where the persisted referenced entries to be retrieved from
	 * @return an array of classpath entries that are referenced directly or indirectly by the given entry. 
	 * 			If not applicable, returns an empty array.
	 * @since 3.6
	 */
	public static IClasspathEntry[] getReferencedClasspathEntries(IClasspathEntry libraryEntry, IJavaProject project);	




Problem Reports Fixed
252431
New API is needed to better identify referenced jars in the Class-Path: entry

251133
[formatter] Automatic formatting single line comments is incoherent among tools

248897
[1.5][compiler] Wrong warning 'The local variable 'var' is never read'.

304031
Unused @SuppressWarnings(..) not flagged when suppressed problem is set to Error

295551
Add option to automatically promote all warnings to errors

303810
Compact boolean fields on FlowContext

251227
[compiler] Fup of bug 115814, comparing doubles should not be flagged

268798
[1.5][compiler] Eclipse 3.5M5/6 produces new compiler errors with generics

303448
Wrong code generation optimization when assert condition is false

303776
Member types imports are removed too aggressively

302949
JavaModelManager hangs accessing the nonChainingJars set

246594
[model] API request: ITypeParameter#getBoundsSignatures() or #getSignature()

253896
[compiler][null] wrong "Null comparison always yields false" problem for auto-unboxing

284789
[formatter] Does not line-break method declaration exception with parameters

303480
[1.5][compiler] CCE: org.eclipse.jdt.internal.compiler.parser.RecoveredBlock cannot be cast to org.eclipse.jdt.internal.compiler.parser.RecoveredType



Eclipse Platform Build Notes

Java development tools core
Eclipse SDK 3.6M6 - February 23, 2010

Project org.eclipse.jdt.core v_A36
(cvs).
What's new in this drop


Added a new formatter preferences to align annotation arguments (i.e. element-value pairs).

This new preference is controlled with the option:
DefaultCodeFormatterConstants.FORMATTER_ALIGNMENT_FOR_ARGUMENTS_IN_ANNOTATION
/**
 * FORMATTER / Option for alignment of arguments in annotation
 *     - option id:         "org.eclipse.jdt.core.formatter.alignment_for_arguments_in_annotation"
 *     - possible values:   values returned by createAlignmentValue(boolean, int, int) call
 *     - default:           createAlignmentValue(false, WRAP_NO_SPLIT, INDENT_DEFAULT)
 * 
 * @see #createAlignmentValue(boolean, int, int)
 * @since 3.6
 */

For example, the following snippet:
@MyAnnot(value1 = "this is an example", value2 = "of an annotation", value3 = "with several arguments", value4 = "which may need to be wrapped")
public class Test {
}

formatted with this preference activated, will produce the following output
while using the Eclipse [built-in] profile:
@MyAnnot(value1 = "this is an example", value2 = "of an annotation",
		value3 = "with several arguments",
		value4 = "which may need to be wrapped")
public class Test {
}

See bug 282030 for more details.

In order to get bindings outside the Eclipse environment, the following methods has been added on the ASTParser class.

See bug 206391 for details.

org.eclipse.jdt.core.dom.ASTParser
	/**
	 * Set the environment that can be used when no IJavaProject are available.
	 * 
	 * The user has to be sure to include all required types on the classpaths for binary types
	 * or on the sourcepaths for source types to resolve the given source code.
	 * All classpath and sourcepath entries are absolute paths.
	 * If sourcepaths contain units using a specific encoding (not the platform encoding), then the
	 * given encodings must be set. If the given encodings is set, its length must
	 * match the length of the sourcepaths parameter or an IllegalArgumentException will be thrown.
	 * If encodings is not null, the given sourcepathEntries must not be null.
	 * 
	 * @param classpathEntries the given classpath entries to be used to resolve bindings
	 * @param sourcepathEntries the given sourcepath entries to be used to resolve bindings
	 * @param encodings the encodings of the corresponding sourcepath entries or null if the platform encoding
	 * can be used.
	 * @param includeRunningVMBootclasspath true if the bootclasspath of the running VM must be prepended to the
	 * given classpath and false if the bootclasspath of the running VM should be ignored.
	 * @throws IllegalArgumentException if the size of the given encodings is not equals to the size of the given
	 * sourcepathEntries
	 * @since 3.6
	 */
	public void setEnvironment(String[] classpathEntries, String[] sourcepathEntries, String[] encodings, boolean includeRunningVMBootclasspath);
	
	/**
	 * Creates ASTs for a batch of compilation units. When bindings are being resolved, processing a
	 * batch of compilation units is more efficient because much of the work involved in resolving 
	 * bindings can be shared. 
	 *
	 * When bindings are being resolved, all compilation units are resolved 
	 * using the same environment, which must be set beforehand with 
	 * {@link #setEnvironment(String[], String[], String[], boolean) setEnvironment}.
	 * The compilation units are processed one at a time in no specified order. 
	 * For each of the compilation units in turn,
	 *  - {@link ASTParser#createAST(IProgressMonitor) ASTParser.createAST} is called to parse it 
	 *           and create a corresponding AST. The calls to {@link ASTParser#createAST(IProgressMonitor) ASTParser.createAST} 
	 *           all employ the same settings.
	 *  - {@link FileASTRequestor#acceptAST(String, CompilationUnit) FileASTRequestor.acceptAST} is called passing
	 *           the compilation unit path and the corresponding AST to requestor. The compilation unit path is the same
	 *           path that is passed into the given sourceFilePaths parameter.
	 *
	 * Note only ASTs from the given compilation units are reported
	 * to the requestor. If additional compilation units are required to
	 * resolve the original ones, the corresponding ASTs are not
	 * reported to the requestor.
	 * 
	 * Note also the following parser parameters are used, regardless of what
	 * may have been specified:
	 *  - The {@linkplain #setKind(int) parser kind} is K_COMPILATION_UNIT
	 *  - The {@linkplain #setSourceRange(int,int) source range} is (0, -1)
	 *  - The {@linkplain #setFocalPosition(int) focal position} is not set
	 *
	 * The bindingKeys parameter specifies bindings keys
	 * ({@link IBinding#getKey()}) that are to be looked up. These keys may
	 * be for elements either inside or outside the set of compilation
	 * units being processed. When bindings are being resolved,
	 * the keys and corresponding bindings (or null if none) are
	 * passed to {@link FileASTRequestor#acceptBinding(String, IBinding) FileASTRequestor.acceptBinding}. Note that binding keys
	 * for elements outside the set of compilation units being processed are looked up
	 * after all {@link FileASTRequestor#acceptAST(String, CompilationUnit) ASTRequestor.acceptAST}
	 * callbacks have been made.
	 * Binding keys for elements inside the set of compilation units being processed
	 * are looked up and reported right after the corresponding
	 * {@link FileASTRequestor#acceptAST(String, CompilationUnit) FileASTRequestor.acceptAST} callback has been made.
	 * No {@link FileASTRequestor#acceptBinding(String, IBinding) FileASTRequestor.acceptBinding} callbacks are made unless
	 * bindings are being resolved.
	 *
	 * A successful call to this method returns all settings to their
	 * default values so the object is ready to be reused.
	 * 
	 * The given encodings are used to properly parse the given source units. If the platform encoding is sufficient,
	 * then the given encodings can be set to null.
	 *
	 * @param sourceFilePaths the compilation units to create ASTs for
	 * @param encodings the given encoding for the source units
	 * @param bindingKeys the binding keys to create bindings for
	 * @param requestor the AST requestor that collects abstract syntax trees and bindings
	 * @param monitor the progress monitor used to report progress and request cancellation,
	 *   or null if none
	 * @exception IllegalStateException if the settings provided
	 * are insufficient, contradictory, or otherwise unsupported
	 * @since 3.6
	 */
	 public void createASTs(String[] sourceFilePaths, String[] encodings, String[] bindingKeys, FileASTRequestor requestor, IProgressMonitor monitor)

/**
 * FORMATTER / Option to control whether block comments will have new lines at boundaries
 *     - option id:         "org.eclipse.jdt.core.formatter.comment.new_lines_at_block_boundaries"
 *     - possible values:   { TRUE, FALSE }
 *     - default:           TRUE
 * 
 * @see #TRUE
 * @see #FALSE
 * @since 3.6
 */

/**
 * FORMATTER / Option to control whether javadoc comments will have new lines at boundaries
 *     - option id:         "org.eclipse.jdt.core.formatter.comment.new_lines_at_javadoc_boundaries"
 *     - possible values:   { TRUE, FALSE }
 *     - default:           TRUE
 * 
 * @see #TRUE
 * @see #FALSE
 * @since 3.6
 */
 

public class X {
	/*
	 * This block comment after formatting will no longer use a new line
	 * at the beginning and at the end of the comment...
	 */
	void foo() {
	}
	/**
	 * This javadoc comment after formatting will no longer use a new line
	 * at the beginning and at the end of the comment...
	 */
	void bar() {
	}
}

public class X {
	/* This block comment after formatting will no longer use a new line at the
	 * beginning and at the end of the comment... */
	void foo() {
	}

	/** This javadoc comment after formatting will no longer use a new line at
	 * the beginning and at the end of the comment... */
	void bar() {
	}
}

/**
 * FORMATTER / Option to insert a new line after a label
 *     - option id:         "org.eclipse.jdt.core.formatter.insert_new_line_after_label"
 *     - possible values:   { INSERT, DO_NOT_INSERT }
 *     - default:           DO_NOT_INSERT
 *
 * @see JavaCore#INSERT
 * @see JavaCore#DO_NOT_INSERT
 * @since 3.6
 */

public class X {
	void foo() {
		LABEL:for (int i = 0; i < 10; i++) {
		}
	}
}

public class X {
	void foo() {
		LABEL:
		for (int i = 0; i < 10; i++) {
		}
	}
}