[
https://issues.apache.org/jira/browse/DERBY-5285?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13051835#comment-13051835
]
Knut Anders Hatlen commented on DERBY-5285:
-------------------------------------------
This issue looks like DERBY-3371. According to the analysis there, the code in
question is a no-op and could be removed. The variable conglom is initialized
by calling findConglomerate(). Since conglomId is negative in the case affected
by the patch, findConglomerate() returns tempCongloms.get(new Long(conglomId)).
So when we end up calling tempCongloms.put(new Long(conglomId), conglom) in
addColumnToConglomerate(), we're just replacing one object with itself.
> [patch] avoid npe if tempCongloms is null
> -----------------------------------------
>
> Key: DERBY-5285
> URL: https://issues.apache.org/jira/browse/DERBY-5285
> Project: Derby
> Issue Type: Improvement
> Components: Store
> Affects Versions: 10.8.1.4
> Reporter: Dave Brosius
> Assignee: Mike Matrigali
> Priority: Trivial
> Fix For: 10.8.1.4
>
> Attachments: npe_guard.diff
>
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> if (tempCongloms != null)
> tempCongloms.remove(new Long(conglomId));
> tempCongloms.put(new Long(conglomId), conglom);
> code guards against tempCongloms being null for the remove, but not for the
> put. so put the put in the if statement.
> Then realized that the remove is superfluous, as the put will overright so
> just removed the remove.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
