[
https://issues.apache.org/jira/browse/LUCENE-8166?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16796036#comment-16796036
]
ASF subversion and git services commented on LUCENE-8166:
---------------------------------------------------------
Commit 1d9f00fd5e7323c0ed082e3c5b31d9c64ffa479d in lucene-solr's branch
refs/heads/branch_8x from Adrien Grand
[ https://gitbox.apache.org/repos/asf?p=lucene-solr.git;h=1d9f00f ]
LUCENE-8166: Require merge instances to be consumed in the thread that created
them.
> Require merge instances to be consumed in the thread that created them
> ----------------------------------------------------------------------
>
> Key: LUCENE-8166
> URL: https://issues.apache.org/jira/browse/LUCENE-8166
> Project: Lucene - Core
> Issue Type: Task
> Reporter: Adrien Grand
> Priority: Minor
>
> I would like to improve AssertingCodec to add the restriction that merge
> instances can only be consumed from the thread that pulled them. This is
> something that I relied on in order to avoid cloning index inputs too much
> when implementing LUCENE-4198, but this is neither documented nor checked.
> For the record, I found that it was already an issue before LUCENE-4198 was
> merged, for instance if you would pull a merge instance of the default stored
> fields reader and clone it (for use in another thread), it would no longer be
> a merge instance. So I think this new restriction makes sense?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
