lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Doron Cohen (JIRA)" <>
Subject [jira] Resolved: (LUCENE-766) Two same new field with and without Term vector make an IllegalStateException
Date Thu, 03 Jan 2008 07:35:34 GMT


Doron Cohen resolved LUCENE-766.

       Resolution: Cannot Reproduce
    Lucene Fields: [New, Patch Available]  (was: [Patch Available, New])

Issue was apparently solved along the way.
Test committed.
Thanks Nicolas!

> Two same new field with and without Term vector make an IllegalStateException
> -----------------------------------------------------------------------------
>                 Key: LUCENE-766
>                 URL:
>             Project: Lucene - Java
>          Issue Type: Bug
>          Components: Index
>    Affects Versions: 2.1
>            Reporter: Nicolas Lalevée
>            Assignee: Doron Cohen
>         Attachments: bugfix.patch, LUCENE-766-r608138.patch
> On a empty index, adding a document with two fields with the same name but with different
term vector option fail. The field with TermVector.WITH_POSITIONS_OFFSETS is correctly indexed,
as the offset are correclty extracted. The field with TermVector.NO is not. The TermVectorsWriter
tries to add offset info given to the data of the filedinfo from the "fnm" file, but the DocumentWriter
didn't prepared offset datas as it gets its info from the field itself, not from the fieldinfo.
> Attaching a patch with a test. The test without the fix make this stack trace :
> java.lang.IllegalStateException: Trying to write offsets that are null!
> 	at org.apache.lucene.index.TermVectorsWriter.writeField(
> 	at org.apache.lucene.index.TermVectorsWriter.closeField(
> 	at org.apache.lucene.index.TermVectorsWriter.closeDocument(
> 	at org.apache.lucene.index.TermVectorsWriter.close(
> 	at org.apache.lucene.index.DocumentWriter.writePostings(
> 	at org.apache.lucene.index.DocumentWriter.addDocument(
> 	at org.apache.lucene.index.IndexWriter.addDocument(
> 	at org.apache.lucene.index.IndexWriter.addDocument(
> 	at org.apache.lucene.index.TestDocumentWriter.testTermVector(

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:
For additional commands, e-mail:

View raw message