can you document more your platform and VM...
This test is based on freemem() and that makes, of course, the assertion quite heuristic... if we all provide numbers for each platforms, maybe we find a way to adjust the heuristic ?
Also, maybe it should be renamed as OptionalTestCoreMemoryLeak and only run when one wants ? (it is very slow anyways).
paul
Le 28 d�c. 04, � 02:48, Brett Porter a �crit :
The test is failing for me. Is that to be expected? [EMAIL PROTECTED] wrote:Added: jelly/src/test/org/apache/commons/jelly/core
TestCoreMemoryLeak.java BaseMemoryLeakTest.java
Removed: jelly/src/java/org/apache/commons/jelly/impl
WeakReferenceWrapperScript.java
Log:
Tag caching is now done in the JellyContext.
The TestCoreMemoryLeak shows that, at least with some usage of JellyContext.clear(), one avoids leaks.
Removing Tag-caching flag. Also, removing the WeakReferenceWrapperScript, both were in the era of ThreadLocal which is to be finished.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
