db-jdo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ilan Kirsh (JIRA)" <j...@apache.org>
Subject [jira] Commented: (JDO-372) ConcurrentPersistenceManagersSameClasses - Failed on second run
Date Sun, 23 Apr 2006 17:06:06 GMT
    [ http://issues.apache.org/jira/browse/JDO-372?page=comments#action_12375890 ] 

Ilan Kirsh commented on JDO-372:

Thank you for the clarification. I now understand this is useless if binary compatibility
is not supported. Maybe you can add a check before starting the test that binary compatibility
is supported? Similar check is also needed in the org.apache.jdo.tck.enhancement tests. Deleting
in finally could be good, but I think that deletePersistentAll of all the PCRect and PCPoint
at the beginning of the test might be even better. Anyway, it doesn't make much different.

> ConcurrentPersistenceManagersSameClasses - Failed on second run
> ---------------------------------------------------------------
>          Key: JDO-372
>          URL: http://issues.apache.org/jira/browse/JDO-372
>      Project: JDO
>         Type: Bug

>     Reporter: Ilan Kirsh
>  Attachments: JDO-372.patch, JDO-372.patch
> Thanks to deleteTearDownClasses / deleteTearDownInstances, most test cases can run on
either a clean database or on an existing database. Unfortunately, it seems that org.apache.jdo.tck.ConcurrentPersistenceManagersSameClasses
is different. Old objects are not deleted from the 2nd database that this test case uses,
so findPoint may return more than one result object, and the first result that is checked
might be an old object from a previous run. Therefore, the test case passes the first run
and may fail on any additional run on the same database files.

This message is automatically generated by JIRA.
If you think it was sent incorrectly contact one of the administrators:
For more information on JIRA, see:

View raw message