]>
source.dussan.org Git - aspectj.git/log
wisberg [Mon, 5 May 2003 15:08:30 +0000 (15:08 +0000)]
addressing (Andy's) NPE - null source location is permitted for packages, so I checked all uses for NPE.
Also cleaned up the code a bit. Should be semantics-preserving.
wisberg [Mon, 5 May 2003 15:07:36 +0000 (15:07 +0000)]
redoing root to be kind [java file || config file] --
was java file kind for config files.
wisberg [Mon, 5 May 2003 14:54:34 +0000 (14:54 +0000)]
comment re: using shadow file rather than enclosing node.
wisberg [Mon, 5 May 2003 07:34:52 +0000 (07:34 +0000)]
minimal update to test script to test running, remove debugger tests.
Seems out of date; may try to enhance ajbrowser doc as basic ajde test script
wisberg [Mon, 5 May 2003 07:34:04 +0000 (07:34 +0000)]
out of the run loop - now direct from TopFrame to Ajde
wisberg [Mon, 5 May 2003 07:33:44 +0000 (07:33 +0000)]
- added run-in-other-vm button/menu,
unscattered run and build invocations
- nb: key binding broken, might need run-other icon
wisberg [Mon, 5 May 2003 07:33:17 +0000 (07:33 +0000)]
support both same- and other-vm runs using new using new LangUtil API's for running processes
wisberg [Mon, 5 May 2003 07:32:21 +0000 (07:32 +0000)]
added LangUtil.ProcessController for external processes
wisberg [Mon, 5 May 2003 07:31:48 +0000 (07:31 +0000)]
- disabled logging by default
- added tests for FileUtil.Pipe
wisberg [Mon, 5 May 2003 07:31:30 +0000 (07:31 +0000)]
added FileUtil.Pipe
wisberg [Fri, 2 May 2003 21:03:04 +0000 (21:03 +0000)]
accept "." as base dir in resolving files
jhugunin [Fri, 2 May 2003 20:36:06 +0000 (20:36 +0000)]
fix and better tests for
Bugzilla Bug 37152
java.lang.VerifyError:
wisberg [Fri, 2 May 2003 19:24:31 +0000 (19:24 +0000)]
updated with requirement to use -incremental and sourceroots for inc-compile tests
wisberg [Fri, 2 May 2003 19:09:11 +0000 (19:09 +0000)]
passing incremental tests from testing-drivers/testdata/harness/incremental
wisberg [Fri, 2 May 2003 19:07:01 +0000 (19:07 +0000)]
converted to sourceroots, removed one test case
jhugunin [Fri, 2 May 2003 06:28:16 +0000 (06:28 +0000)]
fix for
Bugzilla Bug 37152
java.lang.VerifyError:
This fix needs a little clean-up and testing before closing out
jhugunin [Thu, 1 May 2003 19:47:43 +0000 (19:47 +0000)]
more progress messages for ajde
jhugunin [Thu, 1 May 2003 18:48:06 +0000 (18:48 +0000)]
added test that deprecation warnings are not shown unless asked for
modifying option setting to make that test pass
jhugunin [Thu, 1 May 2003 18:11:27 +0000 (18:11 +0000)]
Updated expected bytecode to match change for fix for issue of
Class.forName differences between IBM and SUN's JDKs
wisberg [Thu, 1 May 2003 13:17:35 +0000 (13:17 +0000)]
minimal run-in-same-vm feature for ajbrowser
wisberg [Thu, 1 May 2003 13:15:39 +0000 (13:15 +0000)]
- permit empty input to split
- split classpath
- check for array arguments
wisberg [Thu, 1 May 2003 11:13:25 +0000 (11:13 +0000)]
updated tests for [method|execution] binary join point detection.
added narrower knownLimitation test
wisberg [Thu, 1 May 2003 10:21:00 +0000 (10:21 +0000)]
removing 1.4 API
jhugunin [Thu, 1 May 2003 03:25:43 +0000 (03:25 +0000)]
expanded coverage and fix for bug involving around advice with either
assert or <name>.class in the body
the fix for now is just to not inline such advice, but in the future these
should be transformed to inlinable constructs
jhugunin [Thu, 1 May 2003 02:02:56 +0000 (02:02 +0000)]
fix for
Bugzilla Bug 37020
wrong line for method execution join point
(but points to first executable line rather than method declaration)
and
Bugzilla Bug 37021
source path wrong in declare warning on binary files
jhugunin [Thu, 1 May 2003 00:21:51 +0000 (00:21 +0000)]
Expected fix for issue of Class.forName differences
between IBM and SUN's JDKs -- not yet tested on IBMs JDK to confirm
jhugunin [Thu, 1 May 2003 00:21:12 +0000 (00:21 +0000)]
Added easy to understand test for issue of Class.forName differences
between IBM and SUN's JDKs
Moved a nit-picking part of SourceLocationWithExpr to its own test that
is listed as a known limitation.
wisberg [Wed, 30 Apr 2003 02:43:30 +0000 (02:43 +0000)]
cleaner, but no semantic changes
wisberg [Wed, 30 Apr 2003 02:42:57 +0000 (02:42 +0000)]
removed printlns
wisberg [Wed, 30 Apr 2003 02:42:25 +0000 (02:42 +0000)]
not using unimplemented buildStructure() api
wisberg [Wed, 30 Apr 2003 02:41:39 +0000 (02:41 +0000)]
no change, just nicer feedback
wisberg [Wed, 30 Apr 2003 02:40:17 +0000 (02:40 +0000)]
mainly changed setup to avoid reading any existing user properties, which was causing build configuration tests to fail now that all options are significant
wisberg [Wed, 30 Apr 2003 02:39:00 +0000 (02:39 +0000)]
identify source of log message for convenience
wisberg [Wed, 30 Apr 2003 02:38:29 +0000 (02:38 +0000)]
- IMessage-based compiler messages
- better toString for logging message lists
wisberg [Wed, 30 Apr 2003 02:37:39 +0000 (02:37 +0000)]
- IMessage-based compiler messages
include the stack trace as tooltip text if there
is an associated exception
wisberg [Wed, 30 Apr 2003 02:36:45 +0000 (02:36 +0000)]
- IMessage-based compiler messages
wisberg [Wed, 30 Apr 2003 02:36:02 +0000 (02:36 +0000)]
- using IMessage-based API's for displaying messages
- if a source line message contains a throwable
and the user double-clicks, this displays the error dialog
that permits them to copy the trace. That should mean that
multiple exceptions can be in the task list.
wisberg [Wed, 30 Apr 2003 02:32:39 +0000 (02:32 +0000)]
support shift-key to rebuild-all in incremental mode
wisberg [Wed, 30 Apr 2003 02:32:09 +0000 (02:32 +0000)]
slight rewording of error dialog, and correct URL.
wisberg [Wed, 30 Apr 2003 02:31:37 +0000 (02:31 +0000)]
- making default filename static (access by tests)
- closing streams to permit better test cleanup
wisberg [Wed, 30 Apr 2003 02:30:21 +0000 (02:30 +0000)]
incremental flag support
wisberg [Wed, 30 Apr 2003 02:29:52 +0000 (02:29 +0000)]
static utility for detecting shift key down - should move
wisberg [Wed, 30 Apr 2003 02:29:18 +0000 (02:29 +0000)]
binary-incompatible change to IMessage-based API (using ISourceLocation anyway)
wisberg [Wed, 30 Apr 2003 02:26:50 +0000 (02:26 +0000)]
using new IMessage-based API for sourceline tasks
wisberg [Wed, 30 Apr 2003 02:26:16 +0000 (02:26 +0000)]
- incremental support
- buildStructure test API now builds everything
- removed unused code from pre-IMessage days
wisberg [Wed, 30 Apr 2003 02:22:20 +0000 (02:22 +0000)]
- supporting incremental
- handling new fast-fail config errors
- better error display, no messages lost
Implemented new options merging:
- can specify any local config file options
- project properties and build options
now can contain any options and
override local values only if not set
- have method to fixup the options for IDE's
to enforce their invariants, but this needs
to delegate to a new adapted API.
wisberg [Wed, 30 Apr 2003 02:16:03 +0000 (02:16 +0000)]
- added incremental checkbox
- broadened size and meaning of "non-standard compiler options" to "other compiler options"
wisberg [Wed, 30 Apr 2003 02:14:57 +0000 (02:14 +0000)]
added flag for incremental mode
wisberg [Wed, 30 Apr 2003 02:14:32 +0000 (02:14 +0000)]
binary-incompatible API change to add methods for building fresh when in incremental mode
wisberg [Wed, 30 Apr 2003 02:13:21 +0000 (02:13 +0000)]
Making genBuildConfig static so it can be re-used from AJDE.
wisberg [Wed, 30 Apr 2003 02:11:54 +0000 (02:11 +0000)]
added support for adopting global configuration values (without overwriting local values)
wisberg [Wed, 30 Apr 2003 02:10:50 +0000 (02:10 +0000)]
refactored batch and incremental into one method to make it easier to see the differences and do a fix.
wisberg [Wed, 30 Apr 2003 02:07:21 +0000 (02:07 +0000)]
test case can't expect usage message any more
wisberg [Wed, 30 Apr 2003 02:05:00 +0000 (02:05 +0000)]
fixed two tests to expect addition abort message for usage,
and commented out one test because it does not add the file correctly (to a matching task)
wisberg [Wed, 30 Apr 2003 02:03:54 +0000 (02:03 +0000)]
Usage is now emitted as an abort message, so AjcTask now ignores usage when determining whether to throw a BuildException after getting an abort message. This fixes 3 tests.
wisberg [Tue, 29 Apr 2003 17:50:43 +0000 (17:50 +0000)]
error messages for undefined ${module.name}
wisberg [Tue, 29 Apr 2003 17:48:32 +0000 (17:48 +0000)]
fix test title
acolyer [Tue, 29 Apr 2003 10:40:20 +0000 (10:40 +0000)]
With the restructure of the org.eclipse.jdt.core project, it no
longer has a bin dir. Updated the classpath to include
jdtcore-for-aspectj.jar instead.
wisberg [Tue, 29 Apr 2003 00:26:19 +0000 (00:26 +0000)]
track internal API changes
wisberg [Tue, 29 Apr 2003 00:26:02 +0000 (00:26 +0000)]
using Kind.isSameOrLessThan(..) rather than direct compare for inferring streams
wisberg [Tue, 29 Apr 2003 00:25:34 +0000 (00:25 +0000)]
semantics-preserving - still only warn if aspectjrt.jar not found
wisberg [Tue, 29 Apr 2003 00:25:15 +0000 (00:25 +0000)]
boolean hasSources() to validate after parsing
wisberg [Tue, 29 Apr 2003 00:24:53 +0000 (00:24 +0000)]
error-handling cleaned up
- usage printed once by caller
- any arg mistake is an error
(i.e., not ignoring/continuing)
(does not include list entries like classpath)
- refactored commands to share common error-handling code
wisberg [Tue, 29 Apr 2003 00:24:32 +0000 (00:24 +0000)]
- detecting unflagged directories
- !! changed warnings to errors, so now
not compiling after command-line mistakes
wisberg [Tue, 29 Apr 2003 00:24:15 +0000 (00:24 +0000)]
not accepting directories any more
wisberg [Mon, 28 Apr 2003 23:56:48 +0000 (23:56 +0000)]
test failing - node is java file, should be method execution main, so no parent
wisberg [Mon, 28 Apr 2003 23:22:33 +0000 (23:22 +0000)]
Checking in Mik's changes sent by email today for model and incremental support in AJDE.
ajde has the same test failing as it did before this checkin,
perhaps caused by a regression per bug 37020
wisberg [Mon, 28 Apr 2003 20:08:56 +0000 (20:08 +0000)]
test cases for binary warnings (bugs 37020, 37021, 37023)
jhugunin [Sun, 27 Apr 2003 23:31:06 +0000 (23:31 +0000)]
fix and tests for
Bugzilla Bug 30663
lame error message: "negation doesn't allow binding"
and
Bugzilla Bug 36329
The compiler crashes when using aspect libraries created without using -noweave
jhugunin [Sun, 27 Apr 2003 23:30:14 +0000 (23:30 +0000)]
tests for
Bugzilla Bug 36936
Error when introducing members of type Class
jhugunin [Sun, 27 Apr 2003 23:29:27 +0000 (23:29 +0000)]
fix and test for
Bugzilla Bug 34951
NPE compiling without aspectjrt.jar
jhugunin [Sun, 27 Apr 2003 22:48:06 +0000 (22:48 +0000)]
fixed bug caught by unit tests
wisberg [Sat, 26 Apr 2003 21:15:23 +0000 (21:15 +0000)]
moving release testing list to tests/, added some old stuff.
TODO: write test-release.xml ant script
jhugunin [Fri, 25 Apr 2003 18:11:33 +0000 (18:11 +0000)]
fix and tests for
Bugzilla Bug 36803
BCException thrown by AspectJ 1.1rc1
jhugunin [Thu, 24 Apr 2003 21:05:44 +0000 (21:05 +0000)]
expanded tests and fix for
Bugzilla Bug 36778
ClassFormatError due to empty interface supertype
wisberg [Wed, 23 Apr 2003 09:20:31 +0000 (09:20 +0000)]
updated to reflect runtime changes for 1.1 compatibility
wisberg [Wed, 23 Apr 2003 09:20:06 +0000 (09:20 +0000)]
gutted. Should be removed once we've confirmed that it can safely be removed.
(not sure if CVS keeps history across file removal/add).
(tests run fine, no apparant clients)
wisberg [Wed, 23 Apr 2003 09:18:47 +0000 (09:18 +0000)]
1.2 call replaced with equivalent 1.1 call.
wisberg [Wed, 23 Apr 2003 08:35:40 +0000 (08:35 +0000)]
ugh. updating XML .classpath parser to handle any number of attributes
still does not handle quoted spaces or nested entities.
Also updated for new attribute names in 2.1?
wisberg [Wed, 23 Apr 2003 08:33:39 +0000 (08:33 +0000)]
extracted classpath setup method for testing
wisberg [Wed, 23 Apr 2003 00:32:49 +0000 (00:32 +0000)]
@testcase try/finally in around advice (same as ...messy arounds?)
Works in current tree, but not in 1.1rc1
wisberg [Tue, 22 Apr 2003 23:13:18 +0000 (23:13 +0000)]
@testcase PR#36778 advise join points in subclass of empty interface
jhugunin [Tue, 22 Apr 2003 22:29:51 +0000 (22:29 +0000)]
a sketch of a test for a naming issue that doesn't appear to be an
actual bug
jhugunin [Tue, 22 Apr 2003 21:58:02 +0000 (21:58 +0000)]
tests and fixes for
Bugzilla Bug 29665
Inconsistant stack height
jhugunin [Tue, 22 Apr 2003 17:55:13 +0000 (17:55 +0000)]
test and fix for Bugzilla Bug 36564
Internal compiler error
jhugunin [Tue, 22 Apr 2003 17:35:11 +0000 (17:35 +0000)]
fix for Bugzilla Bug 36736
compiler error expected for implemented abstract pointcuts
and for
fix for Bugzilla Bug 36673
VerifyError in privileged aspect
minimally extended coverage for this bug
jhugunin [Tue, 22 Apr 2003 17:31:26 +0000 (17:31 +0000)]
fix for Bugzilla Bug 36673
VerifyError in privileged aspect
jhugunin [Tue, 22 Apr 2003 17:31:00 +0000 (17:31 +0000)]
fix for Bugzilla Bug 36736
compiler error expected for implemented abstract pointcuts
(using declaredModifiers instead of modifiers so interfaces don't cause trouble)
jhugunin [Tue, 22 Apr 2003 17:23:07 +0000 (17:23 +0000)]
fix for Bugzilla Bug 36736
compiler error expected for implemented abstract pointcuts
wisberg [Tue, 22 Apr 2003 00:06:39 +0000 (00:06 +0000)]
@testcase PR#36736 implemented abstract pointcut
wisberg [Sat, 19 Apr 2003 20:47:30 +0000 (20:47 +0000)]
@testcase PR#36673 privileged aspect main verify error
wisberg [Sat, 19 Apr 2003 08:14:00 +0000 (08:14 +0000)]
More readable API than comparator for common floor operation (semantics-preserving)
wisberg [Sat, 19 Apr 2003 07:16:57 +0000 (07:16 +0000)]
prevent div/0, NPE (by inspection)
acolyer [Thu, 17 Apr 2003 19:01:19 +0000 (19:01 +0000)]
Release checklist
wisberg [Sun, 13 Apr 2003 10:36:18 +0000 (10:36 +0000)]
one /
wisberg [Sun, 13 Apr 2003 10:35:58 +0000 (10:35 +0000)]
added fork and maxmem
wisberg [Sun, 13 Apr 2003 10:14:43 +0000 (10:14 +0000)]
instructions on changing default resource-copying behavior, and misc rewordings
wisberg [Sun, 13 Apr 2003 09:42:18 +0000 (09:42 +0000)]
- extdirs and bootclasspath not experimental
- noWeave has correct case
- better handling of list arguments
- better comments
- still largely untested
jhugunin [Sat, 12 Apr 2003 01:00:48 +0000 (01:00 +0000)]
fix and test for Bugzilla Bug 36110
IllegalAccessError while accessing introduced variable / 1.1rc1
jhugunin [Sat, 12 Apr 2003 00:47:38 +0000 (00:47 +0000)]
fix and test for Bugzilla Bug 36110
IllegalAccessError while accessing introduced variable / 1.1rc1