[
https://issues.apache.org/jira/browse/LANG-727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13079977#comment-13079977
]
Matt Benson commented on LANG-727:
----------------------------------
I presume the property we'd need to check would be *java.vendor* as opposed to
*java.vm.vendor*, but I'd be glad for a second opinion. Note that the test
also passes on an Apple JDK, so I'd prefer we disable it only on IBM for now
until/unless we encounter another classlib that gives us the same problem.
Joerg, can you give us the java.vendor values for whichever JDKs fail? Or, for
example, does the test pass on an earlier IBM JDK? If so, can you give me
whatever property values I need to distinguish the environment in which we
expect a failure?
Thanks!
> ToStringBuilderTest.testReflectionHierarchyArrayList fails with IBM JDK 6
> -------------------------------------------------------------------------
>
> Key: LANG-727
> URL: https://issues.apache.org/jira/browse/LANG-727
> Project: Commons Lang
> Issue Type: Bug
> Components: lang.builder.*
> Affects Versions: 3.0
> Reporter: Joerg Schaible
> Priority: Trivial
> Fix For: 3.0.2
>
>
> The unit test fails when running with IBM JDK 6:
> {noformat}
> Failed tests:
> testReflectionHierarchyArrayList(org.apache.commons.lang3.builder.ToStringBuilderTest): null
> expected:<....ArrayList@115b115b[[elementData={<null>,<null>,<null>,<null>,<null>,<null>,<null>,null>,null>,null>},size=0],modCount=0]>
> but
> was:<....ArrayList@115b115b[[firstIndex=0,lastIndex=0,array={<null>,<null>,<null>,<null>,<null>,<null>,<null>,null>,null>,null>}],modCount=0]>
> {noformat}
> Actually the test is wrong, because it makes wrong assumptions about the
> implementation of ArrayList in the runtime.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
