db-jdo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Peter Dettman (JIRA)" <j...@apache.org>
Subject [jira] Commented: (JDO-653) Extra checks in PMF.close tests
Date Wed, 19 May 2010 05:16:56 GMT

    [ https://issues.apache.org/jira/browse/JDO-653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12868996#action_12868996
] 

Peter Dettman commented on JDO-653:
-----------------------------------

Yes, those tests currently fail against the reference implementation (DN 2.0.x).

I am looking firstly for confirmation that these failures represent violations of the spec;
see "11.4 Close the PersistenceManagerFactory" (http://people.apache.org/~clr/jdo-2010-04-09.pdf).

Then it is my intention to patch the reference implementation to fix these problems once I
have that confirmation.

Obviously there are some interdependencies; I am unsure whether these test cases must remain
in limbo until the RI is fixed, or whether there is a mechanism for known-failing tests to
be added to the TCK.


> Extra checks in PMF.close tests
> -------------------------------
>
>                 Key: JDO-653
>                 URL: https://issues.apache.org/jira/browse/JDO-653
>             Project: JDO
>          Issue Type: Test
>          Components: tck
>    Affects Versions: JDO 2 maintenance release 2
>            Reporter: Peter Dettman
>            Assignee: Michelle Caisse
>         Attachments: JDO-653-2nd.patch, JDO-653.patch
>
>
> Some of the implications of section 11.4 don't appear to be checked for in the case of
an already-closed PersistenceManagerFactory.
> 1. If close() is called on an already-closed PMF (with correct permission i.e. "closePersistenceManagerFactory"),
no exception should be thrown.
> 2. If close() is called on an already-closed PMF (without correct permission), exception
should be thrown (even though it would otherwise do nothing).
> 3. If close() fails (throws JDOUserException) due to active transaction(s), then none
of the PersistenceManager objects belonging to the PMF should have actually been closed.
> 4. Several get/set pairs are not tested for their behaviour after close()

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message