Thanks, that demonstrates it really nicely.

Now if only dismax did wildcards too... :-)

Cheers,

Andrew.


ANithian wrote:
> 
> The best way to get started with answering this is to pass the
> &debugQuery=true and to scroll down the results page. Here, you will see a
> breakdown of how the query you entered in the q field is being parsed and
> sent to lucene via the pf,qf, and bf. You can also see how the weights
> affect the different score and why one document was ranked higher than
> another.
> 
> The text of the query will be analyzed depending on the set of analyzers
> assigned to that particular field for queries (as opposed to indexing).
> For
> example, if "test" is matched against a "string" vs "text" field,
> different
> analyzers may be applied to "string" or "text"
> 
> Hope that helps
> Amit
> 
> On Thu, Oct 29, 2009 at 4:39 AM, Andrew Clegg
> <andrew.cl...@gmail.com>wrote:
> 
>>
>> Morning,
>>
>> Can someone clarify how dismax queries work under the hood? I couldn't
>> work
>> this particular point out from the documentation...
>>
>> I get that they pretty much issue the user's query against all of the
>> fields
>> in the schema -- or rather, all of the fields you've specified in the qf
>> parameter in the config or the request.
>>
>> But, does each of these 'sub'-queries get analyzed according to the
>> normal
>> analysis rules for the field it's getting sent to? Or are they passed
>> through verbatim?
>>
>> I'm hoping it's the former, as we have a variety of different field types
>> with radically different tokenization and filtering...
>>
>> Also, is there any plan to implement wildcards in dismax, or is this
>> unfeasible?
>>
>> Thanks once again :-)
>>
>> Andrew.
>>
>> --
>> View this message in context:
>> http://www.nabble.com/dismax-and-query-analysis-tp26111465p26111465.html
>> Sent from the Solr - User mailing list archive at Nabble.com.
>>
>>
> 
> 

-- 
View this message in context: 
http://www.nabble.com/dismax-and-query-analysis-tp26111465p26118506.html
Sent from the Solr - User mailing list archive at Nabble.com.

Reply via email to