[
https://issues.apache.org/jira/browse/DERBY-5301?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13092331#comment-13092331
]
Bryan Pendleton commented on DERBY-5301:
----------------------------------------
For some reason, the test's setup code which is supposed to force table-level
locking isn't working for me.
If I run the test with -Dderby.storage.rowLocking=false on the command line,
then I see the same results
that you do. But if I don't specify that flag when I start junit, the test
cases all fail because I am not doing
the expected table-level locking.
It took me a while to figure this out because when this happens, the test case
gets an assertion about
the table-level locking problem, but then the tearDown() method gets another
assertion, and that exception
hides the test problem.
I think it might be better to move the view-dropping code out of the tearDown
method and into normal
subroutines that are called by each test case, to avoid this exception-hiding
problem. Or maybe the tearDown
method should catch and ignore any exceptions that occur in that method?
> Convert store/TableLockBasic.sql to junit
> ------------------------------------------
>
> Key: DERBY-5301
> URL: https://issues.apache.org/jira/browse/DERBY-5301
> Project: Derby
> Issue Type: Sub-task
> Components: Test
> Reporter: Houx Zhang
> Assignee: Houx Zhang
> Labels: gsoc2011
> Attachments: 5301-1-part.patch, junit.out
>
>
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
