aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* moved knownLimitations from ajcTestsFailing to ajcTests and labeled with keywordjhugunin2003-05-272-34/+44
|
* Ant doc updates wisberg2003-05-271-147/+315
| | | | | | - forkclasspath - fork option classpath and conflicts - added maven section
* tests for devguide script examplewisberg2003-05-274-0/+41
|
* more parts of tests for devguide script examplewisberg2003-05-272-0/+135
|
* muddy caveat to JSR-45 commentwisberg2003-05-271-2/+4
|
* tests for setting message holder by class namewisberg2003-05-271-8/+54
|
* -Xlint:{level} testswisberg2003-05-271-0/+21
|
* test checks that setting warn:constructorName workswisberg2003-05-272-0/+17
|
* fixing fork bugs: wisberg2003-05-272-68/+128
| | | | | | - permitting fork in incremental mode only when using tag file - pulling aspectjtools also from the task classpath also failonerror default was not true as stated in the docs
* addressing different results under 1.4wisberg2003-05-251-11/+22
|
* incorrect log file name for harness testswisberg2003-05-251-2/+2
|
* better harness availability, control -- untestedwisberg2003-05-251-13/+62
|
* PR-based queries were dominating - now do union correctlywisberg2003-05-251-1/+2
|
* documenting optionswisberg2003-05-251-49/+127
|
* harness copyright wrong - was created in 2002wisberg2003-05-251-2/+3
|
* added keyword knownLimitations-ajctaskCompiler to the one command-line test ↵wisberg2003-05-251-1/+5
| | | | that won't work with that wrapper.
* - better logging, temp dir, deletion... wisberg2003-05-251-29/+57
| | | | - using build.config system property
* Fix for failure building with useEclipseCompiles testing-drivers initially ↵wisberg2003-05-251-13/+20
| | | | | | | | | | (or any module that only indirectly depended on jdt.core, which has no sources) - better logging - checking zip results - returns false positive for empty zips - compile(..) API change for adopting classes Also, copyright clearly wrong - file created last fall.
* Fix for failure building with useEclipseCompiles testing-drivers initially ↵wisberg2003-05-251-16/+33
| | | | | | | | (or any module that only indirectly depended on jdt.core, which has no sources) - better logging - checking zip results - returns false positive for empty zips - compile(..) API change for adopting classes
* delete, close, and isEmptywisberg2003-05-251-0/+34
|
* fixing 2 bugs disclosed by running ajcTests.xml with -ajctaskCompiler: (1) ↵wisberg2003-05-251-2/+2
| | | | | | compliance set incorrectly (-1.3 instead of 1.4) and (2) warn:... form was ignored. These would break only when running from a Javac compiler adapter and using <compilerarg>
* better loggingwisberg2003-05-251-1/+1
|
* fixing 2 bugs disclosed by running ajcTests.xml with -ajctaskCompiler: (1) ↵wisberg2003-05-251-2/+9
| | | | | | compliance set incorrectly (-1.3 instead of 1.4) and (2) warn:... form was ignored. These would break only when running from a Javac compiler adapter and using <compilerarg>
* updating harness documentation for new featureswisberg2003-05-252-24/+92
|
* tests for new feature to select tests by titlewisberg2003-05-254-4/+136
|
* suite files now constrained to be readable to avoid confusion with ↵wisberg2003-05-251-1/+2
| | | | -ajctestTitle.. referents.
* adding harness feature to select tests based on their title (description), ↵wisberg2003-05-251-18/+192
| | | | reading from a comma-delimited list or a file containing PASS|FAIL lines (from -traceTestsMin), optionally restricted to FAIL.
* updating build.jar to bootstrap changes to permit ant.jar in a testing modulewisberg2003-05-241-0/+0
|
* no real changes, just better docs and an (untested) "all" target.wisberg2003-05-241-2/+45
|
* -noExit to avoid stopping a wrapping ant scriptwisberg2003-05-241-1/+3
|
* moved a command-line-driven test of classpath variants herewisberg2003-05-241-1/+41
|
* Better BuildModule tests following ant.jar updates. wisberg2003-05-241-18/+116
| | | | | These now build with assemble-all off and on, and tests were updated to check by running classes.
* To support the new testing ICommand wrapper for AjcTask, permit ant.jar in a ↵wisberg2003-05-241-5/+9
| | | | | | testing module (ant is omitted from non-testing modules because it is normally deployed separately in ant/lib/ant.jar). This does not affect regular module or product builds.
* documenting -ajdeCompiler and -ajctaskCompiler selectorswisberg2003-05-241-1/+6
|
* added taskdefs to classpath to support AjcTaskCompileCommandwisberg2003-05-242-0/+2
|
* Tests for AjcTaskCompileCommmand adapterwisberg2003-05-241-0/+270
|
* ICommand adapter for Ant taskdef is simple for the non-incremental case and ↵wisberg2003-05-241-0/+294
| | | | ugly otherwise.
* supporting -ajctaskCompiler option with AjcTaskCompileCommandwisberg2003-05-241-5/+12
|
* awful downcast for AjcTaskCompileCommand - should be remedied after 1.1 with ↵wisberg2003-05-241-1/+5
| | | | ICommand interface extension
* updated implementation for -ajdeCompiler. wisberg2003-05-241-61/+182
| | | | | | replaced reference to null ide properties adapter (defined only in testsrc) and created logging proxy to verify API's used. Unfortunately also formatted code, so it looks like there are more changes than there are.
* updates mainly to support maven but also to support new ICommand wrapper ↵wisberg2003-05-241-91/+268
| | | | | | | | | | | | | | | | | for AjcTask: - findToolsJar now accepts versioned jars - supporting a forkclasspath to permit user to explicitly set - upgrades for testing - so multiple iterations work, - not altering cmd in execute() - only fixing up outjar once - reset() complete - so incremental compile tests work, - passing quit through to main - when forking on windows, use javaw rather than java if possible to avoid popping up a console window
* more/better junit tests for AjcTask.java fork when possiblewisberg2003-05-241-124/+199
|
* setting classpath for out-of-eclipse testingwisberg2003-05-241-1/+5
|
* fix for Bugzilla Bug 37758 jhugunin2003-05-211-12/+35
| | | | Weaving rt.jar results in stack overflow
* test for Bugzilla Bug 37758 jhugunin2003-05-216-3/+314
| | | | Weaving rt.jar results in stack overflow
* forcing target to be 1.1 in case someone builds using 1.4.wisberg2003-05-211-0/+0
|
* flagging with comments that 1.1 options are unimplementedwisberg2003-05-211-3/+3
|
* forcing target to be 1.1 in case someone builds using 1.4.wisberg2003-05-211-0/+1
|
* @testcase PR#37898 advice on handler join points should not throw ↵wisberg2003-05-212-0/+39
| | | | unpermitted checked exceptions
* added paragraph for call site implementationwisberg2003-05-201-1/+11
|