[
https://issues.apache.org/jira/browse/LUCENE-2641?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robert Muir resolved LUCENE-2641.
---------------------------------
Resolution: Fixed
Committed revision 996357, 996360 (3x).
we can always look out for more tests like this and handle them on a case by
case basis,
In general if it creates a huge index, we should ensure reasonable
maxBufferedDocs etc,
and if it has a ton of methods that don't modify the index, we should consider
creating the index in @BeforeClass
But i think the "crazy" defaults in newIndexWriterConfig are reasonable, given
most tests
only use a tiny amount of documents.
> BaseTestRangeFilter can be extremely slow
> -----------------------------------------
>
> Key: LUCENE-2641
> URL: https://issues.apache.org/jira/browse/LUCENE-2641
> Project: Lucene - Java
> Issue Type: Test
> Components: Tests
> Reporter: Robert Muir
> Fix For: 3.1, 4.0
>
> Attachments: LUCENE-2641.patch, LUCENE-2641.patch, LUCENE-2641.patch
>
>
> The tests that extend BaseTestRangeFilter can sometimes be very slow:
> TestFieldCacheRangeFilter, TestMultiTermConstantScore, TestTermRangeFilter
> for example, TestFieldCacheRangeFilter just ran for 10 minutes on my computer
> before I killed it,
> but i noticed these tests frequently run for over a minute.
> I think at the least we should change these to junit4 so the index is built
> once in @beforeClass
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
