[ 
https://issues.apache.org/jira/browse/LUCENE-2912?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Robert Muir updated LUCENE-2912:
--------------------------------

    Attachment: LUCENE-2912.patch

Attached is an initial patch, all tests pass.


> remove field param from computeNorm, scorePayload ; remove UOE'd lengthNorm, 
> switch SweetSpot to per-field 
> -----------------------------------------------------------------------------------------------------------
>
>                 Key: LUCENE-2912
>                 URL: https://issues.apache.org/jira/browse/LUCENE-2912
>             Project: Lucene - Java
>          Issue Type: Improvement
>            Reporter: Robert Muir
>             Fix For: 4.0
>
>         Attachments: LUCENE-2912.patch
>
>
> In LUCENE-2236 we switched sim to per field (SimilarityProvider returns a 
> per-field similarity).
> But we didn't completely cleanup there... I think we should now do this:
> * SweetSpotSimilarity loses all its hashmaps. Instead, just configure one per 
> field and return it in your SimilarityProvider. this means for example, all 
> its TF factors can now be configured per-field too, not just the length 
> normalization factors.
> * computeNorm and scorePayload lose their field parameter, as its redundant 
> and confusing.
> * the UOE'd obselete lengthNorm is removed. I also updated javadocs that were 
> pointing to it (this is bad!).

-- 
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org

Reply via email to