lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paul Elschot <paul.elsc...@xs4all.nl>
Subject Re: svn commit: r332747 - in /lucene/java/trunk: ./ src/java/org/apache/lucene/search/regex/ src/test/org/apache/lucene/search/regex/
Date Tue, 15 Nov 2005 18:53:19 GMT
On Tuesday 15 November 2005 19:35, Yonik Seeley wrote:
> On 11/15/05, Doug Cutting <cutting@apache.org> wrote:
> > Paul Elschot wrote:
> > > I think loosing the field boosts for PrefixQuery and friends would not 
be
> > > advisable. Field boosts have a very big range and from that a very big
> > > influence on the score and the order of the results in Hits.
> >
> > It should not be hard to add these.  If a field name is provided, then
> > it would be a one or two line change to ConstantScoreQuery.java to have
> > it boost scores according to that field's norm values.  Right, Yonik?
> >
> > Doug
> 
> Yep, that sounds like that would do it.
> 
> As far as API goes, I guess there should be a constructor
> ConstantScoreQuery(Filter filter, String field)
> If field is non-null, then the field norm can be multiplied into the score.

That might be called a FieldNormQuery, and it would be useful
on occasions.

Not using the document term frequencies in PrefixQuery would still
leave  these as a surprise factor between PrefixQuery and TermQuery.

Regards,
Paul Elschot


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


Mime
View raw message