[
https://issues.apache.org/jira/browse/SOLR-916?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12703669#action_12703669
]
Kay Kay commented on SOLR-916:
------------------------------
| I could go either way. The name change certainly makes some sense, but even
as is, a core that is returned closed is not very useful. Null could just as
well be returned. But when you ask to returnPrev, a working open core is
guaranteed to be returned if a previous one existed. I wouldnt feel so bad
leaving it as it is - your change almost seems to strengthen the contract - you
have to return the closed core as well. Neither side is a very strong argument
to me and I could go either way.
I do not have an issue with the logic ( closing an existing core if returnPrev
is false ) as it is currently but am more concerned with the name of the
variable and the documentation about the same - just to make sure that it is in
sync with the actual logic as of today.
> CoreContainer :: register(String, SolrCore, boolean) documentation
> clarification about returnPrev argument
> -----------------------------------------------------------------------------------------------------------
>
> Key: SOLR-916
> URL: https://issues.apache.org/jira/browse/SOLR-916
> Project: Solr
> Issue Type: Improvement
> Affects Versions: 1.3
> Environment: Tomcat 6, JRE 6
> Reporter: Kay Kay
> Priority: Minor
> Fix For: 1.4
>
> Attachments: SOLR-916.patch
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> In CoreContainer.java :: register(name, core, returnPrev) - the documentation
> says
> it would return a previous core having the same name if it existed *and
> returnPrev = true*.
> * @return a previous core having the same name if it existed and
> returnPrev==true
> */
> public SolrCore register(String name, SolrCore core, boolean returnPrev) ..
> But as per the code towards the end - the previous core is returned anyway,
> irrespective of the value of returnPrev. The difference, though, seems to be
> that when returnPrev is false, the previous core (of the same name, if
> exists) is closed.
> Which one of them is correct . If the code were correct , would the variable
> be better renamed as closePrevious , as opposed to returnPrevious.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
