[
https://issues.apache.org/jira/browse/LUCENE-2633?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12907228#action_12907228
]
Simon Willnauer commented on LUCENE-2633:
-----------------------------------------
bq. +1, if the test is problematic, we can always either comment out the new
test, or @Ignore it, and add a note.
+1 - I thought about something like the Assume#assumeTrue(MAX_HEAP_MEM >= 256);
(http://kentbeck.github.com/junit/javadoc/latest/org/junit/Assume.html#assumeTrue(boolean))
which could be set to a certain default value and would exclude the test if
not enough memory is available. That would prevent uncommenting the test - I
guess @Ignore would work too but maybe we want disable / enable at a central
place for tests like that.
simon
> PackedInts does not support structures above 256MB
> --------------------------------------------------
>
> Key: LUCENE-2633
> URL: https://issues.apache.org/jira/browse/LUCENE-2633
> Project: Lucene - Java
> Issue Type: Bug
> Affects Versions: 4.0
> Environment: All
> Reporter: Toke Eskildsen
> Attachments: LUCENE-2633.patch
>
>
> The PackedInts Packed32 and Packed64 fails when the internal structure
> exceeds 256MB. This is due to a missing cast that results in the bit position
> calculation being limited by Integer.MAX_VALUE (256MB * 8 = 2GB).
--
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
