[ 
https://issues.apache.org/jira/browse/LUCENE-5616?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Robert Muir updated LUCENE-5616:
--------------------------------

    Attachment: LUCENE-5616.patch

New patch. I think this found a bug in DV updates.

I think for DV updates, the wrong fieldinfos is passed to the codec? 

{noformat}
 2> NOTE: reproduce with: ant test  -Dtestcase=TestNumericDocValuesUpdates 
-Dtests.method=testManyReopensAndFields -Dtests.seed=3E2A9FE369B4CC43 
-Dtests.locale=th_TH_TH_#u-nu-thai -Dtests.timezone=NZ-CHAT 
-Dtests.file.encoding=UTF-8
   [junit4] ERROR   0.23s J3 | 
TestNumericDocValuesUpdates.testManyReopensAndFields <<<
   [junit4]    > Throwable #1: org.apache.lucene.index.CorruptIndexException: 
Invalid field number: 2, 
input=BufferedChecksumIndexInput(MockIndexInputWrapper(SlicedIndexInput(SlicedIndexInput(_d.dvm
 in RAMInputStream(name=_d.cfs)) in RAMInputStream(name=_d.cfs) slice=412:584)))
{noformat}

> remove FieldInfos leniency
> --------------------------
>
>                 Key: LUCENE-5616
>                 URL: https://issues.apache.org/jira/browse/LUCENE-5616
>             Project: Lucene - Core
>          Issue Type: Bug
>            Reporter: Robert Muir
>         Attachments: LUCENE-5616.patch, LUCENE-5616.patch
>
>
> The current behavior is sheisty: it always returns null for negative field 
> numbers (without consulting the map).
> This means if a field number somehow became negative, there would be silent 
> data loss.
> Instead, we should throw exception on negative numbers.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to