]> source.dussan.org Git - javassist.git/log
javassist.git
6 years agostops a potential leak in JarClassPath openClassfile(). Issues #222.
chibash [Sat, 15 Sep 2018 03:06:25 +0000 (12:06 +0900)]
stops a potential leak in JarClassPath openClassfile().  Issues #222.

6 years agoadds a method taking Lookup to java.util.proxy.
chibash [Sun, 9 Sep 2018 17:39:42 +0000 (02:39 +0900)]
adds a method taking Lookup to java.util.proxy.

6 years agoadds Automatic-Module-Name to manifest
chibash [Sun, 9 Sep 2018 14:07:17 +0000 (23:07 +0900)]
adds Automatic-Module-Name to manifest

6 years agomodifies javassist.util.proxy.DefinePackageHelper to be compatible to Java 11.
chibash [Sat, 8 Sep 2018 07:26:19 +0000 (16:26 +0900)]
modifies javassist.util.proxy.DefinePackageHelper to be compatible to Java 11.
It also adds a javadoc comment to CtClass#toClass() to recommend the users to use #toClass(Class) or #toClass(Lookup).

6 years agochanges CtClass#toClass() and ClassPool#toClass() etc. to support Java 11,
chibash [Sat, 8 Sep 2018 06:47:38 +0000 (15:47 +0900)]
changes CtClass#toClass() and ClassPool#toClass() etc. to support Java 11,
in other words, java.lang.invoke.MethodHandles.Lookup.

6 years agodoes refactoring for javassist.util.proxy.
chibash [Thu, 6 Sep 2018 16:04:04 +0000 (01:04 +0900)]
does refactoring for javassist.util.proxy.

6 years agochanges the version number
chibash [Tue, 4 Sep 2018 16:52:23 +0000 (01:52 +0900)]
changes the version number

6 years agoadds test cases for NestHostAttribute and NestMembersAttribute.
chibash [Sun, 2 Sep 2018 14:50:56 +0000 (23:50 +0900)]
adds test cases for NestHostAttribute and NestMembersAttribute.

6 years agomodifies NestHostAttribute and NestMembersAttribute and adds test code
chibash [Sun, 2 Sep 2018 13:56:45 +0000 (22:56 +0900)]
modifies NestHostAttribute and NestMembersAttribute and adds test code

6 years agoMerge pull request #218 from dmitri-gb/master
Shigeru Chiba [Sun, 2 Sep 2018 10:36:51 +0000 (19:36 +0900)]
Merge pull request #218 from dmitri-gb/master

Add support for Java 11 NestHost and NestMembers attributes (JEP-181)

6 years agoadds a comment for the fix to JIRA JASSIST-267
chibash [Sun, 2 Sep 2018 10:25:12 +0000 (19:25 +0900)]
adds a comment for the fix to JIRA JASSIST-267

6 years agoMerge pull request #210 from NingZhang-Ericsson/JASSIST-267
Shigeru Chiba [Fri, 31 Aug 2018 00:22:59 +0000 (09:22 +0900)]
Merge pull request #210 from NingZhang-Ericsson/JASSIST-267

fixed JASSIST-267 (new )

6 years agoAdd support for Java 11 NestHost and NestMembers attributes 218/head
Dmitri Gabbasov [Thu, 23 Aug 2018 13:16:44 +0000 (16:16 +0300)]
Add support for Java 11 NestHost and NestMembers attributes

6 years agofixes issue #209 (NPE in Exception)
chibash [Wed, 18 Jul 2018 21:57:45 +0000 (06:57 +0900)]
fixes issue #209 (NPE in Exception)

6 years agoMerge pull request #207 from andresluuk/patch-2
Shigeru Chiba [Wed, 18 Jul 2018 21:05:45 +0000 (06:05 +0900)]
Merge pull request #207 from andresluuk/patch-2

Custom URL stream handler in ByteArrayClassPath

6 years agoFix JASSIST-267 210/head
NingZhang-Ericsson [Fri, 13 Jul 2018 08:48:13 +0000 (16:48 +0800)]
Fix JASSIST-267
Clean code according to comments.

6 years agofixed JASSIST-267
NingZhang-Ericsson [Fri, 13 Jul 2018 05:06:20 +0000 (13:06 +0800)]
fixed JASSIST-267

6 years agoCustom URL stream handler in ByteArrayClassPath 207/head
Andres Luuk [Mon, 9 Jul 2018 08:05:25 +0000 (11:05 +0300)]
Custom URL stream handler in ByteArrayClassPath

Otherwise a normal streamhandler lookup is made and if javassist is used early on server startup protses this can lead to a deadlock on rare cases on WebSpheare.
For example. A new StreamHandlerFactory is registered but not jet fully initialized. Now javassist is used at the same time but form another thread and it will be the first one to invoke the new StreamHandlerFactory and now it can deadlock with the other thread:

Snipets from my deadlock:
3XMTHREADINFO      "main" J9VMThread:0x000000000175EC00, omrthread_t:0x00007F4120008100, java/lang/Thread:0x00000000624290B8, state:B, prio=5
3XMTHREADBLOCK     Blocked on: org/osgi/util/tracker/ServiceTracker$Tracked@0x00000000627F0A90 Owned by: "other-thread" (J9VMThread:0x0000000001B26300, java/lang/Thread:0x00000000629F1458)
3XMHEAPALLOC             Heap bytes allocated since last GC cycle=0 (0x0)
3XMTHREADINFO3           Java callstack:
4XESTACKTRACE                at org/osgi/util/tracker/ServiceTracker.getServiceReferences(ServiceTracker.java:525)
4XESTACKTRACE                at org/eclipse/osgi/framework/internal/protocol/StreamHandlerFactory.createInternalURLStreamHandler(StreamHandlerFactory.java:162)
4XESTACKTRACE                at org/eclipse/osgi/framework/internal/protocol/StreamHandlerFactory.createURLStreamHandler(StreamHandlerFactory.java:118)
4XESTACKTRACE                at java/net/URL.getURLStreamHandler(URL.java:1154)
4XESTACKTRACE                at java/net/URL.<init>(URL.java:611(Compiled Code))
4XESTACKTRACE                at java/net/URL.<init>(URL.java:502(Compiled Code))
4XESTACKTRACE                at java/net/URL.<init>(URL.java:451)
4XESTACKTRACE                at javassist/ByteArrayClassPath.find(JRebel:92)
...
4XESTACKTRACE                at java/lang/ClassLoader.loadClassHelper(ClassLoader.java:924(Compiled Code))
5XESTACKTRACE                   (entered lock: org/eclipse/core/launcher/Main$StartupClassLoader@0x0000000062F3AA88, entry count: 1)

3XMTHREADINFO      "other-thread" J9VMThread:0x0000000001B26300, omrthread_t:0x00007F4120A5D060, java/lang/Thread:0x00000000629F1458, state:B, prio=5
3XMTHREADBLOCK     Blocked on: org/eclipse/core/launcher/Main$StartupClassLoader@0x0000000062F3AA88 Owned by: "main" (J9VMThread:0x000000000175EC00, java/lang/Thread:0x00000000624290B8)
3XMHEAPALLOC             Heap bytes allocated since last GC cycle=0 (0x0)
3XMTHREADINFO3           Java callstack:
4XESTACKTRACE                at org/osgi/util/tracker/AbstractTracked.size(AbstractTracked.java:356)
4XESTACKTRACE                at org/osgi/util/tracker/ServiceTracker.getServiceReferences(ServiceTracker.java:526)
5XESTACKTRACE                   (entered lock: org/osgi/util/tracker/ServiceTracker$Tracked@0x00000000627F0A90, entry count: 1)
4XESTACKTRACE                at org/eclipse/osgi/framework/internal/protocol/StreamHandlerFactory.createInternalURLStreamHandler(StreamHandlerFactory.java:162)
4XESTACKTRACE                at org/eclipse/osgi/framework/internal/protocol/StreamHandlerFactory.createURLStreamHandler(StreamHandlerFactory.java:118)
4XESTACKTRACE                at java/net/URL.getURLStreamHandler(URL.java:1154)
4XESTACKTRACE                at java/net/URL.<init>(URL.java:611(Compiled Code))
4XESTACKTRACE                at java/net/URL.<init>(URL.java:502(Compiled Code))
4XESTACKTRACE                at java/net/URL.<init>(URL.java:451)
4XESTACKTRACE                at java/net/URI.toURL(URI.java:1100)
4XESTACKTRACE                at com/ibm/jsse2/av.a(av.java:92)
4XESTACKTRACE                at com/ibm/jsse2/av.engineInit(av.java:7)
4XESTACKTRACE                at javax/net/ssl/TrustManagerFactory.init(TrustManagerFactory.java:25)

6 years agoupdates for release 3.23.1 rel_3_23_1_ga
chibash [Sun, 1 Jul 2018 17:27:13 +0000 (02:27 +0900)]
updates for release 3.23.1

6 years agoMerge pull request #204 from beikov/issues/171
Shigeru Chiba [Sat, 23 Jun 2018 12:23:21 +0000 (21:23 +0900)]
Merge pull request #204 from beikov/issues/171

[#171] Allow running on JDK 8 and possibly on JDK 7 as well. Fixes #171

6 years ago[#171] Allow running on JDK 8 and possibly on JDK 7 as well. Fixes #171 204/head
Christian Beikov [Fri, 22 Jun 2018 07:03:55 +0000 (09:03 +0200)]
[#171] Allow running on JDK 8 and possibly on JDK 7 as well. Fixes #171

6 years agoupdates for release 3.23.0-GA rel_3_23_0_ga
chibash [Wed, 20 Jun 2018 16:07:19 +0000 (01:07 +0900)]
updates for release 3.23.0-GA

6 years agofixes a minor bug in a test case.
chibash [Tue, 8 May 2018 00:24:19 +0000 (09:24 +0900)]
fixes a minor bug in a test case.

6 years agoMerge pull request #198 from beikov/JASSIST-271
Shigeru Chiba [Mon, 7 May 2018 10:14:33 +0000 (19:14 +0900)]
Merge pull request #198 from beikov/JASSIST-271

JASSIST-271 Allow running on Java 11 by falling back to ClassLoader.defineClass for Java > 10

6 years agoMerge pull request #168 from cmelchior/cm/bug/classpath-close
Shigeru Chiba [Mon, 7 May 2018 10:06:13 +0000 (19:06 +0900)]
Merge pull request #168 from cmelchior/cm/bug/classpath-close

[WIP] Fix leaking file handlers

6 years agoJASSIST-271 Allow running on Java 11 by falling back to ClassLoader.defineClass for... 198/head
Christian Beikov [Sun, 29 Apr 2018 14:20:56 +0000 (16:20 +0200)]
JASSIST-271 Allow running on Java 11 by falling back to ClassLoader.defineClass for Java > 10

6 years agofixes a bug reported in issue #155 (Javassist Incorrect Code when wrapping with try...
chibash [Tue, 14 Nov 2017 22:02:05 +0000 (07:02 +0900)]
fixes a bug reported in issue #155 (Javassist Incorrect Code when wrapping with try/catch)

6 years agoadds Modifier#isVarArgs()
chibash [Tue, 14 Nov 2017 19:07:28 +0000 (04:07 +0900)]
adds Modifier#isVarArgs()

6 years agoadds Modifier#isVarArgs()
chibash [Tue, 14 Nov 2017 18:46:02 +0000 (03:46 +0900)]
adds Modifier#isVarArgs()

6 years agochanges getDefinedPackage back to getPackage when Java is < 9.
chibash [Tue, 14 Nov 2017 18:05:05 +0000 (03:05 +0900)]
changes getDefinedPackage back to getPackage when Java is < 9.

6 years agoMerge pull request #159 from nickl-/scoped-pool
Shigeru Chiba [Tue, 14 Nov 2017 15:06:52 +0000 (00:06 +0900)]
Merge pull request #159 from nickl-/scoped-pool

Fixes to javassist.scopepool

6 years agoMerge pull request #158 from nickl-/secure-privileged
Shigeru Chiba [Tue, 14 Nov 2017 15:05:37 +0000 (00:05 +0900)]
Merge pull request #158 from nickl-/secure-privileged

Secure privileged

6 years agoMerge pull request #166 from nickl-/test-ctfield-make-with-this
Shigeru Chiba [Mon, 13 Nov 2017 18:22:25 +0000 (03:22 +0900)]
Merge pull request #166 from nickl-/test-ctfield-make-with-this

Unit Test: CtField.make referencing 'this'.

6 years agoMerge pull request #167 from nickl-/innerclass-modifiers
Shigeru Chiba [Mon, 13 Nov 2017 17:50:41 +0000 (02:50 +0900)]
Merge pull request #167 from nickl-/innerclass-modifiers

Fix for #152 inner class modifiers.

6 years agoupdates build.xml to be compatible to pom.xml
chibash [Mon, 13 Nov 2017 16:57:47 +0000 (01:57 +0900)]
updates build.xml to be compatible to pom.xml

6 years agoAdd missing unit test 168/head
Christian Melchior [Mon, 13 Nov 2017 10:25:16 +0000 (11:25 +0100)]
Add missing unit test

6 years agoFix merge mistake
Christian Melchior [Mon, 13 Nov 2017 10:24:57 +0000 (11:24 +0100)]
Fix merge mistake

6 years agoFix #152 inner class modifiers. 167/head
nickl- [Sat, 11 Nov 2017 20:34:34 +0000 (22:34 +0200)]
Fix #152 inner class modifiers.

Applied fix for #152 "Wrong modifiers returned for nested classes" as suggested by @pietrobraione.
Includes unit tests for all access modifiers and static.

6 years agoUnit Test: CtField.make referencing 'this'. 166/head
nickl- [Fri, 10 Nov 2017 17:13:01 +0000 (19:13 +0200)]
Unit Test: CtField.make referencing 'this'.

As per question raised in #161 passing 'this' reference for CtField value new object instance.
Also added assert for testing referncing 'this' for newly added field.

6 years agoFixes to javassist.scopepool 159/head
nickl- [Sun, 29 Oct 2017 21:42:49 +0000 (23:42 +0200)]
Fixes to javassist.scopepool

Applied parameters and squashed all compiler warnings.
Changed loops to enhanced for where appropriate.
Fixed SoftValueHashMap:
Made it a raw type.
Overwritten implementation of the `get` method correctly unwrapped the reference but the methods
`containsValue` and `entrySet` needed to be fixed.
Implemented the remaining outstanding methods from the `java.util.Map` contract `values`, `putAll`
and `keySet`.
Dropped `AbstractMap` inheritance as we were not inheriting anything anymore.
Changed internal map to `ConcurrentHashMap' to be atomic and parallel safe.

Found original unit tests from jboss retro which are now updated to reflect the javassist
versions of the scoped pool classes. Added tests for the SoftValueHashMap including
test which overloads the heap to ensure the cache cleans itself appropriately.

6 years agoDefine parameters for returned iterator. 158/head
nickl- [Sun, 29 Oct 2017 22:53:08 +0000 (00:53 +0200)]
Define parameters for returned iterator.

6 years agoSome functional unit tests proofs.
nickl- [Fri, 27 Oct 2017 06:33:55 +0000 (08:33 +0200)]
Some functional unit tests proofs.

Proves that you cannot access restricted functionality then goes ahead and
mock test all 3 exposures JAVA_9, JAVA_7+ and OTHERS by changing the
private static final class referenced default chosen on initialise.

6 years agoSome code comments to fill in gaps.
nickl- [Fri, 27 Oct 2017 06:12:07 +0000 (08:12 +0200)]
Some code comments to fill in gaps.

6 years agoEven now Oracle still not happy.
nickl- [Fri, 27 Oct 2017 06:10:22 +0000 (08:10 +0200)]
Even now Oracle still not happy.

Runtime still throws warnings illegal actions even after all that, there's
just no logic to it. We can use Unsafe but don't touch setAccessible. O well
might as well enjoy Unsafe while we still can.

Wrapped the Unsafe and added method cache with varargs method for invoke calls.
Can still do a lot with it but it does what it needs to for now.

6 years agoRefactor definePackage as well.
nickl- [Fri, 27 Oct 2017 05:59:12 +0000 (07:59 +0200)]
Refactor definePackage as well.

Yes it is deprecated but seems the right thing to do

6 years agoGive helper private constructor fwiw.
nickl- [Fri, 27 Oct 2017 05:57:45 +0000 (07:57 +0200)]
Give helper private constructor fwiw.

Wont stop you from getting an instance but it makes it more trouble at least.
Also reduce the visibility of the enum method.

6 years agoAdd caller class checking.
nickl- [Fri, 27 Oct 2017 05:55:08 +0000 (07:55 +0200)]
Add caller class checking.

The only thing remotely inforceable is caller class checking. You cannot
bypass an exception with setAccesible. Oracle went and removed
Reflection.getCallerClass() completely in favour of StackWalker. At least
we can share the SecurityManage ClassContext, SecurityActions seems the
appropriate venue for our stack trace peek to support the older versions

6 years agoMethodHandles for JDK7+.
nickl- [Fri, 27 Oct 2017 05:36:38 +0000 (07:36 +0200)]
MethodHandles for JDK7+.

Java 7 and 8 does support method handles so we can cater for them. When doing
repeated invokes to the same method, keeping a reference to the Method handle
is much faster than reflection. Also the API is cleaner not having to Object[]
args etc. Worth the effort...

6 years agoRefactored SecActs methods and retain 1.6 source.
nickl- [Fri, 27 Oct 2017 05:19:22 +0000 (07:19 +0200)]
Refactored SecActs methods and retain 1.6 source.

Remembered that 1.6 won't know about MethodHandles so must use reflection methods for them still.

6 years agoAdd sun.misc.Unsafe privileged retrieval to SecActions.
nickl- [Fri, 27 Oct 2017 04:59:33 +0000 (06:59 +0200)]
Add sun.misc.Unsafe privileged retrieval to SecActions.

Also do it anonomously so as not to raise the alarms and upset the powers to be.

6 years agoType check and paramatized rawtypes for SecurityActions.
nickl- [Fri, 27 Oct 2017 04:50:16 +0000 (06:50 +0200)]
Type check and paramatized rawtypes for SecurityActions.

Some whitespace got shunted around a bit too.

6 years agoAdd getMethodHandle to SecurityActions.
nickl- [Fri, 27 Oct 2017 04:43:49 +0000 (06:43 +0200)]
Add getMethodHandle to SecurityActions.

The main advantage is that we can do the privileged setAccessible during creation and
then freely invoke via the authorized method handle.

As per the javadocs:
Access checks are applied in the factory methods of Lookup, when a method handle is created.
his is a key difference from the Core Reflection API, since java.lang.reflect.Method.invoke
performs access checking against every caller, on every call.

The performance boost is just a bonus.

6 years agoMake an effort to secure privileged use.
nickl- [Wed, 25 Oct 2017 18:40:04 +0000 (20:40 +0200)]
Make an effort to secure privileged use.

Oracle (or maybe it was still Sun) warns that we should prevent from exposing access we gain to
privileged functionality like the unsafe etc. Before Oracle decides to restrict us even more lets
make an effort at least.

6 years agoCleanup
Christian Melchior [Sun, 12 Nov 2017 14:39:20 +0000 (15:39 +0100)]
Cleanup

6 years agoMerge branch 'master' into cm/bug/classpath-close
Christian Melchior [Sun, 12 Nov 2017 14:37:15 +0000 (15:37 +0100)]
Merge branch 'master' into cm/bug/classpath-close

# Conflicts:
# pom.xml
# src/main/javassist/ByteArrayClassPath.java
# src/main/javassist/ClassClassPath.java
# src/main/javassist/ClassPoolTail.java
# src/main/javassist/LoaderClassPath.java
# src/main/javassist/URLClassPath.java

6 years agoRemove close method on ClassPath. Add unit tests for new behaviour.
Christian Melchior [Sun, 12 Nov 2017 14:17:19 +0000 (15:17 +0100)]
Remove close method on ClassPath. Add unit tests for new behaviour.

6 years agoUpdate test README
Christian Melchior [Sun, 12 Nov 2017 14:11:37 +0000 (15:11 +0100)]
Update test README

6 years agoIgnore IntelliJ files
Christian Melchior [Sun, 12 Nov 2017 11:34:10 +0000 (12:34 +0100)]
Ignore IntelliJ files

6 years agoMerge pull request #160 from nickl-/squash-warnings
Shigeru Chiba [Sun, 12 Nov 2017 06:45:25 +0000 (15:45 +0900)]
Merge pull request #160 from nickl-/squash-warnings

Source walk/Spring clean/Parameterize/Enhance/Squash warnings et.al.

6 years agoOrganize imports to disolve ambiguity. 160/head
nickl- [Mon, 30 Oct 2017 19:27:52 +0000 (21:27 +0200)]
Organize imports to disolve ambiguity.

6 years agoSquash compiler warnings in tests.
nickl- [Mon, 30 Oct 2017 18:43:30 +0000 (20:43 +0200)]
Squash compiler warnings in tests.

With the unit tests warnings were squashed by suppressing them with SuppressWarnings annotations, instead of
reparing the issues as was done with the source. Reasons being...
 * The tests might be relying on the types as they were defined.
 * If the tests were written to test against the correct forms they would've/should've stipulated those.
 * Even unused members might be important to tests so we attempt to change as little as possible.

6 years agoSource walk/Spring clean/Parameterize/Enhance et.al.
nickl- [Mon, 30 Oct 2017 18:37:23 +0000 (20:37 +0200)]
Source walk/Spring clean/Parameterize/Enhance et.al.

The following were applied during multiple itterations through the source.
 * Parameterize raw types.
 * Mark unused members.
 * Annotate override and deprecated methods.
 * Convert loops to enhance for loop.
 * Remove redundant else statements.
 * Widening collection type references.
 * Optimize code for improved readability.
 * Squash compiler warnings.
 * Identify smells.

6 years agoFix deperecated junit.framework.Assert.
nickl- [Mon, 30 Oct 2017 06:07:43 +0000 (08:07 +0200)]
Fix deperecated junit.framework.Assert.

Changed to org.junit.Assert.

6 years agoFix deperecated junit.framework.Assert.
nickl- [Mon, 30 Oct 2017 06:03:47 +0000 (08:03 +0200)]
Fix deperecated junit.framework.Assert.

Changed to org.junit.Assert.

6 years agoImplement hashCode for Annotation.
nickl- [Mon, 30 Oct 2017 04:50:20 +0000 (06:50 +0200)]
Implement hashCode for Annotation.

To honour the Object.equals contract.

6 years agoImplement hashCode for Type.
nickl- [Mon, 30 Oct 2017 04:42:35 +0000 (06:42 +0200)]
Implement hashCode for Type.

To honour the Object.equals contract.

6 years agoImplement hashCode for MultiType.
nickl- [Mon, 30 Oct 2017 04:28:52 +0000 (06:28 +0200)]
Implement hashCode for MultiType.

To honour the Object.equals contract.

6 years agoImplement hashCode for MultiArrayType.
nickl- [Mon, 30 Oct 2017 04:22:57 +0000 (06:22 +0200)]
Implement hashCode for MultiArrayType.

To honour the Object.equals contract.

6 years agoFix deperecated junit.framework.Assert.
nickl- [Sun, 29 Oct 2017 22:39:35 +0000 (00:39 +0200)]
Fix deperecated junit.framework.Assert.

Changed to org.junit.Assert.

6 years agoSuppress deprecation Applet.
nickl- [Sat, 28 Oct 2017 04:28:05 +0000 (06:28 +0200)]
Suppress deprecation Applet.

6 years agoSuppress deprecation ClassPool.makePackage
nickl- [Sat, 28 Oct 2017 04:27:02 +0000 (06:27 +0200)]
Suppress deprecation ClassPool.makePackage

6 years agoFix deprecation getPackage to getDefinedPackage
nickl- [Sat, 28 Oct 2017 04:22:42 +0000 (06:22 +0200)]
Fix deprecation getPackage to getDefinedPackage

6 years agoAdded default serialVersionUID.
nickl- [Sat, 28 Oct 2017 04:18:43 +0000 (06:18 +0200)]
Added default serialVersionUID.

6 years agoUpgrade junit from ver 3 to 4. 157/head
nickl- [Wed, 25 Oct 2017 22:33:39 +0000 (00:33 +0200)]
Upgrade junit from ver 3 to 4.

There were minimal changes:
Some tests had their own runners these were discontinued as IDEs/maven run the tests anyway.
Removed the main methods for these.
2 tests were dependent on the running order so added the test sorter.

All tests succeed on junit 4 with legacy support.

6 years agoAdd version to maven-bundle-plugin.
nickl- [Wed, 25 Oct 2017 19:11:20 +0000 (21:11 +0200)]
Add version to maven-bundle-plugin.

In order to squash the mvn build warnings.

6 years agoConfigure source encoding as UTF-8.
nickl- [Wed, 25 Oct 2017 19:10:26 +0000 (21:10 +0200)]
Configure source encoding as UTF-8.

In order to squash the mvn build warnings.

6 years agoMove runtests to build directory.
nickl- [Wed, 25 Oct 2017 19:08:57 +0000 (21:08 +0200)]
Move runtests to build directory.

Move the test output folder 'runtests' to target or build directory so that mvn clean can clean that.

6 years agochanges ClassPool#appendSystemPath(). The new implementation appends ClassClassPath... 3.22-old-style-source
chibash [Thu, 26 Oct 2017 21:51:38 +0000 (06:51 +0900)]
changes ClassPool#appendSystemPath().  The new implementation appends ClassClassPath when the JVM is earlier than 9.

6 years agofixes a bug of getAnnotations(). Some hidden annotations were not ignored.
chibash [Thu, 26 Oct 2017 21:12:33 +0000 (06:12 +0900)]
fixes a bug of getAnnotations().  Some hidden annotations were not ignored.

7 years agoupgrade maven-javadoc-plugin to 3.0.0-M1 to support publishing of 3.22.0-GA rel_3_22_00_ga
Scott Marlow [Tue, 10 Oct 2017 18:12:58 +0000 (14:12 -0400)]
upgrade maven-javadoc-plugin to 3.0.0-M1 to support publishing of 3.22.0-GA

7 years agoremoves JvstTest4#testJIRA150b because it is obsolete. rel_3_22_1_ga
chibash [Tue, 10 Oct 2017 17:18:14 +0000 (02:18 +0900)]
removes JvstTest4#testJIRA150b because it is obsolete.

7 years ago3.22.0-GA release list rel_3_22_0_ga
chibash [Tue, 10 Oct 2017 02:44:54 +0000 (11:44 +0900)]
3.22.0-GA release

7 years agoupdates ConstPool to support Module and Package
chibash [Mon, 11 Sep 2017 16:17:19 +0000 (01:17 +0900)]
updates ConstPool to support Module and Package

7 years agomodifies CtClass#setModifiers() to correctly support inner classes.
chibash [Thu, 1 Jun 2017 13:46:39 +0000 (22:46 +0900)]
modifies CtClass#setModifiers() to correctly support inner classes.

7 years agodelete testHotswap() because java 9 does not seem to allow a java agent to be attache...
chibash [Wed, 31 May 2017 08:44:26 +0000 (17:44 +0900)]
delete testHotswap() because java 9 does not seem to allow a java agent to be attached after the JVM starts

7 years agoto support Java 9-ea+171
chibash [Tue, 30 May 2017 16:52:09 +0000 (01:52 +0900)]
to support Java 9-ea+171

7 years agomodifies the version number rel_3_22_0_cr2
chibash [Sun, 16 Apr 2017 15:28:27 +0000 (00:28 +0900)]
modifies the version number

7 years agoadds javassist.util.HotSwapAgent
chibash [Sun, 16 Apr 2017 14:42:05 +0000 (23:42 +0900)]
adds javassist.util.HotSwapAgent

7 years agoadds javassist.util.proxy.ProxyFactory#onlyPublicMethods. If true, Javassist uses...
chibash [Sat, 15 Apr 2017 14:57:23 +0000 (23:57 +0900)]
adds javassist.util.proxy.ProxyFactory#onlyPublicMethods.  If true, Javassist uses java.lang.invoke.MethodHandles.Lookup to load a class.

7 years agoMerge pull request #130 from andresluuk/patch-1
Shigeru Chiba [Wed, 12 Apr 2017 18:50:48 +0000 (03:50 +0900)]
Merge pull request #130 from andresluuk/patch-1

A fix for older JDK8 versions: https://bugs.openjdk.java.net/browse/JDK-8041920

7 years agomodifies to use sun.misc.Unsafe#defineClass so that Javassist can run with jigsaw
chibash [Wed, 12 Apr 2017 18:45:51 +0000 (03:45 +0900)]
modifies to use sun.misc.Unsafe#defineClass so that Javassist can run with jigsaw

7 years agoUpdate ClassFile.java 130/head
Andres Luuk [Mon, 10 Apr 2017 14:18:07 +0000 (17:18 +0300)]
Update ClassFile.java

A fix for older JDK8 versions: https://bugs.openjdk.java.net/browse/JDK-8041920
This caused random failures a little time after startup. Reported now so that when we upgrade javassist for JDK9 then we would not have to fork this class.

7 years agoupdates the test code to run on Java 9 EA+149.
chibash [Tue, 20 Dec 2016 09:07:11 +0000 (18:07 +0900)]
updates the test code to run on Java 9 EA+149.
Due to the specification change of jigsaw, ClassLoader#defineClass()
cannot be invoked through reflection API.  As a workaround,
  --add-opens java.base/java.lang=ALL-UNNAMED
must be given to the JVM.  The ant target "runtest9" adds this jvm
argument.

7 years agoremoves javassist.ModuleClassPath.
chibash [Fri, 16 Dec 2016 14:22:18 +0000 (23:22 +0900)]
removes javassist.ModuleClassPath.
See #ClassFilesAsResources in JSR 376.  We don't need to hack the module system to read a class file in java.lang.

http://openjdk.java.net/projects/jigsaw/spec/issues/

8 years agoimplements better annotation handling since Javassist may fail to
chibash [Thu, 13 Oct 2016 17:40:06 +0000 (02:40 +0900)]
implements better annotation handling since Javassist may fail to
create a proxy object due to access controls by jigsaw.

8 years agoupdates Readme.html.
chibash [Tue, 4 Oct 2016 23:47:11 +0000 (08:47 +0900)]
updates Readme.html.

8 years agorelease 3.22.0-CR1 rel_3_22_0_cr1
chibash [Tue, 4 Oct 2016 16:02:21 +0000 (01:02 +0900)]
release 3.22.0-CR1

8 years agoMerge branch 'test/java9-jigsaw'
chibash [Tue, 4 Oct 2016 15:51:28 +0000 (00:51 +0900)]
Merge branch 'test/java9-jigsaw'

8 years agomodifies javadoc typos.
chibash [Tue, 4 Oct 2016 10:42:53 +0000 (19:42 +0900)]
modifies javadoc typos.

8 years agofixes typos in javadoc comments rel_3_21_0_ga
chibash [Mon, 3 Oct 2016 15:44:29 +0000 (00:44 +0900)]
fixes typos in javadoc comments