Mailing-List: contact java-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: java-user@lucene.apache.org
Received-SPF: pass (athena.apache.org: local policy)
Message-ID: <4993737B.60005@aconex.com>
Date: Thu, 12 Feb 2009 11:55:23 +1100
From: Paul Cowan <cowan@aconex.com>
Organization: Aconex
User-Agent: Thunderbird 2.0.0.19 (X11/20090105)
MIME-Version: 1.0
To: java-user@lucene.apache.org
Subject: Re: Fields with multiple values...
References: <BLU149-W23B24AB6F4CD2427429754B9BA0@phx.gbl>
In-Reply-To: <BLU149-W23B24AB6F4CD2427429754B9BA0@phx.gbl>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit


Dragon Fly wrote:
> I'd like to get a hit if I do:
>     Field1:A AND Field2:C
> This is fine because that's how Lucene works.  However, I do not want to get a hit if I do:
>     Field1:A AND Field2:D
> The reason that I don't want a hit is because A is the first element in Field1 and D is the second element in Field2.  I only want a hit when both values are at the same array index.  Is there a way to do this? Thank you in advance for your help.

With a bit of fudging, you may be able to use a SpanNearQuery to do this 
-- though you'll need to change your analyzer's positionIncrementGap, 
and depending on how many tokens you'll have for each field (if it 
varies between fields) you may need to patch Lucene to change the way 
the gaps are calculated.

I've done a prototype of exactly this (SpanNearQuery and analyzer 
changes) for a very similar situation here and it works BRILLIANTLY -- 
massively, massively faster for us than indexing a new document for each 
set of 'subfields'.

I've put some patches against 
https://issues.apache.org/jira/browse/LUCENE-1494 -- I think this 
approach might work for you. Let me know if you need clarification.

Cheers,

Paul

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