[
https://issues.apache.org/jira/browse/SOLR-924?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12657702#action_12657702
]
Hoss Man commented on SOLR-924:
-------------------------------
What Kay is suggesting is a pretty well established "best practice" .. even if
you *know* the super class has an empty finalize method it's still a good idea
to call super.finalize() in a finally block because you have no garuntee
someone else won't modify the implementation of the super class in a future
version. even when you subclass Object, super.finalize() is a good idea to
protect yourself against yourself: future refactoring could inject a new super
class (which could then gain a non-trivial finalize method)
i really see no downside to the patch
> Solr: Making finalizers call super.finalize() wrapped in try..finally block
> ----------------------------------------------------------------------------
>
> Key: SOLR-924
> URL: https://issues.apache.org/jira/browse/SOLR-924
> Project: Solr
> Issue Type: Improvement
> Components: contrib - DataImportHandler, replication (java)
> Environment: Tomcat 6, JRE 6
> Reporter: Kay Kay
> Priority: Minor
> Fix For: 1.4
>
> Attachments: SOLR-924.patch
>
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> There are some occurences of finalizers in the code base. While the presence
> of them is debatable and discussed in a separate JIRA - the ones that are
> retained are better off wrapped around a try .. finally block to recursively
> call the finalizer of the super class for proper resource usage unwinding ,
> (in case finalizers get invoked ).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
