You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-user@lucene.apache.org by Alan Woodward <al...@romseysoftware.co.uk> on 2012/02/13 12:39:18 UTC
Overriding SloppySimScorer
Hello,
I want to score span queries based on the simple presence or absence of a hit (I'm not interested in Tf or Idf here), with a possible boost on specific spans. I've already extended DefaultSimilarity to deal with single terms. From looking at the code it seems that I want to override TFIDFSimilary#sloppySimScorer to return a custom SloppySimScorer instance. However, this method has been declared final. Am I going about this the wrong way? Or should the SimScorer methods on TDIDFSimilarity be unfinalized?
I'm using Lucene trunk, r1241355.
Thanks,
Alan Woodward
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
Re: Overriding SloppySimScorer
Posted by Alan Woodward <al...@romseysoftware.co.uk>.
On 13 Feb 2012, at 12:16, Robert Muir wrote:
> On Mon, Feb 13, 2012 at 6:39 AM, Alan Woodward
> <al...@romseysoftware.co.uk> wrote:
>> Hello,
>>
>> (I'm not interested in Tf or Idf here)
>> I've already extended DefaultSimilarity
>
> In this case, then extending DefaultSimilarity/TFIDFSimilarity is not
> the best approach.
Ah yes, the clue's in the name, I suppose. :-) I've extended the base Similarity class instead, and that works.
Thanks for your help.
>
>> Or should the SimScorer methods on TDIDFSimilarity be unfinalized?
>
> No, it must be final, because TFIDFSimilarity only exposes its
> historical api: tf(), idf(), etc.
>
>> Am I going about this the wrong way?
>
> Maybe you should just extend Similarity...
>
> --
> lucidimagination.com
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
Re: Overriding SloppySimScorer
Posted by Robert Muir <rc...@gmail.com>.
On Mon, Feb 13, 2012 at 6:39 AM, Alan Woodward
<al...@romseysoftware.co.uk> wrote:
> Hello,
>
> (I'm not interested in Tf or Idf here)
> I've already extended DefaultSimilarity
In this case, then extending DefaultSimilarity/TFIDFSimilarity is not
the best approach.
> Or should the SimScorer methods on TDIDFSimilarity be unfinalized?
No, it must be final, because TFIDFSimilarity only exposes its
historical api: tf(), idf(), etc.
> Am I going about this the wrong way?
Maybe you should just extend Similarity...
--
lucidimagination.com
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org