[
https://issues.apache.org/jira/browse/LUCENE-6998?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15120352#comment-15120352
]
Robert Muir commented on LUCENE-6998:
-------------------------------------
+1, especially for this test. These crazy ctors were "manually stared at"
before, this is way better.
after this issue we can look at converting some of the other tests (e.g.
TestIndexWriterExceptions2) to also use Points.
when pushing can you change the comment for the new test
(TestAllFilesDetectTruncation) ?
> We should detect truncation for all index files
> -----------------------------------------------
>
> Key: LUCENE-6998
> URL: https://issues.apache.org/jira/browse/LUCENE-6998
> Project: Lucene - Core
> Issue Type: Bug
> Reporter: Michael McCandless
> Assignee: Michael McCandless
> Fix For: 5.5, Trunk, 6.0
>
> Attachments: LUCENE-6998.patch, LUCENE-6998.patch
>
>
> [~rcmuir] noticed that {{Lucene60PointReader}} does not detect truncation of
> its data file on reader init ... so I added a test to catch this, confirmed
> it caught the bug, and fixed the bug, and fixed a couple other things
> uncovered by the test.
> I also improved the other {{TestAllFiles*}} tests to use {{LineFileDocs}} so
> they index points, and it caught another bug in {{Lucene60PointFormat}}
> (using the same codec name in two files).
> There is more to do here, e.g. we also need a test that catches places where
> we fail to check the index header on init, which was also missing for
> {{Lucene60PointReader}}.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
