[
https://issues.apache.org/jira/browse/LUCENE-5398?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Peng Cheng updated LUCENE-5398:
-------------------------------
Attachment: NormValueSource.java
Removed.
Run junit TestValueSources without problem. This thing should be trivial and
doesn't require a test case for non-byte situation
> NormValueSource unable to read long field norm
> ----------------------------------------------
>
> Key: LUCENE-5398
> URL: https://issues.apache.org/jira/browse/LUCENE-5398
> Project: Lucene - Core
> Issue Type: Bug
> Components: core/query/scoring
> Affects Versions: 4.6
> Environment: Ubuntu 12.04
> Reporter: Peng Cheng
> Priority: Trivial
> Fix For: 4.7
>
> Attachments: NormValueSource.java
>
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> Previous Lucene used to store norms in memory, hence float values are encoded
> into byte to avoid memory overflow.
> Recent release no longer have this constraint, as a result, normValue are
> generally encoded to/decoded from long.
> But the legacy NormValueSource still typecast any long encoding into byte, as
> seen in line 74 in the java file, making any TFIDFSimilarity using more
> accurate encoding useless.
> It should be removed for the greater good.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
