lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Erik Hatcher <e...@ehatchersolutions.com>
Subject Re: StandardAnalyzer and numbers indexed as text
Date Wed, 14 Jan 2004 00:10:56 GMT
On Jan 13, 2004, at 6:19 PM, Patrick Kates wrote:
> I have a text field called ACTIVE_YEAR that stores (of course) a year 
> like
> 2003.  When I index this field I can see the number in my index (using 
> Luke)
> but I can't search it.  If I add a text character to the end of the 
> field
> and index it (200x) I can then search and find 'x', but not any 
> portion of
> the number.

Is it possible you're using a Keyword field for indexing, but an 
analyzer at query time that is ripping off the number?  Try the latest 
Luke (0.4) and you'll see the query expression that it got parsed to on 
the Search tab.  The results may tell the tale.

> One finally piece of info, when using Luke to view the terms contained 
> in
> the index I can find my 'x' term, but none of the years are listed as 
> terms.
>
> Any ideas on how to fix this?

Be careful how you index and query.  Field.Keyword sounds like what you 
want for indexing.  But for querying, QueryParser has some nasty habits 
of analyzing everything.  Do some Query.toString debugging if you are 
using QueryParser - therein likely lies the clues to the problem.

	Erik


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


Mime
View raw message