lucenenet-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Doug Sale (JIRA)" <>
Subject [jira] Commented: (LUCENENET-143) NUnit test for Index/TestStressIndexing2
Date Wed, 13 Aug 2008 16:44:44 GMT


Doug Sale commented on LUCENENET-143:

Please disregard the last line in my previous comment.  I cannot replicate that condition.

However, I have modified the "debug" code in the Lucene.Net version of the tests to correctly
print out the Fields in each Document when they are found to be unequal.  I have also modified
the code in the Java version of the test to always print out the Fields.

Somehow (in the Java version of the test) even when there are duplicate Fields with the same
name, said Fields are always in the same order for both Lists.  This is odd, considering that
the explicitly defined comparator only looks at  I've checked the other data
structures in which the Fields are stored (ArrayList inside of Document), but have found no
other implicit or explicit ordering that would explain this.

So, I modified the comparator to check both field name and field contents to ensure that each
Field List for the 2 Documents are in order prior to comparison.  Both tests (TestRandom()
and TestMultiConfig()) both clear VerifyEquals(Document d1, Document d2) and now are failing
in VerifyEquals(TermFreqVector[] d1, TermFreqVector[] d2).

I will provide a patch that includes these changes (fixing the "debug" code and updating the
comparator) as well as incorporating Digy's patch, shortly.  

> NUnit test for Index/TestStressIndexing2
> ----------------------------------------
>                 Key: LUCENENET-143
>                 URL:
>             Project: Lucene.Net
>          Issue Type: Bug
>         Environment: Lucene.Net 2.3.1
>            Reporter: Digy
>            Priority: Minor
>         Attachments: TestStressIndexing2.patch
> Following patch resolves a bug, but this test continues to fail.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message