You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by Scott Smith <ss...@mainstreamdata.com> on 2013/09/23 18:54:54 UTC
Solr query processing
I just want to state a couple of things and hear someone say, "that's right".
1. In a solr query you can have multiple fq's, but only a single q. And yes, I can simply AND the multiple "q"s together. Just want to avoid that if I'm wrong.
2. A subtler issue is that when a full query is executied, Solr must look at the schema to see how each field was tokenized (or not) and the various other filters applied to a field so that it can properly transform fields data (e.g., tokenize the text, but not keywords). As an aside, it would be nice if the queryparser could do the same thing in Lucene (I know, wrong forum :)).
Scott
Re: Solr query processing
Posted by Erick Erickson <er...@gmail.com>.
bq: As an aside, it would be nice if the queryparser could do the same
thing in Lucene
Lucene does not and (probably) will not ever know anything about the
schema. It's
purposely unaware of this higher-level construct. I wish you great good luck
persuading the lucene guys to have anything like a schema, you'll need it ;).
Best,
Erick
On Mon, Sep 23, 2013 at 9:44 PM, Otis Gospodnetic
<ot...@gmail.com> wrote:
> That's right.
>
> Otis
> Solr & ElasticSearch Support
> http://sematext.com/
> On Sep 23, 2013 12:55 PM, "Scott Smith" <ss...@mainstreamdata.com> wrote:
>
>> I just want to state a couple of things and hear someone say, "that's
>> right".
>>
>>
>> 1. In a solr query you can have multiple fq's, but only a single q.
>> And yes, I can simply AND the multiple "q"s together. Just want to avoid
>> that if I'm wrong.
>>
>> 2. A subtler issue is that when a full query is executied, Solr must
>> look at the schema to see how each field was tokenized (or not) and the
>> various other filters applied to a field so that it can properly transform
>> fields data (e.g., tokenize the text, but not keywords). As an aside, it
>> would be nice if the queryparser could do the same thing in Lucene (I know,
>> wrong forum :)).
>> Scott
>>
Re: Solr query processing
Posted by Otis Gospodnetic <ot...@gmail.com>.
That's right.
Otis
Solr & ElasticSearch Support
http://sematext.com/
On Sep 23, 2013 12:55 PM, "Scott Smith" <ss...@mainstreamdata.com> wrote:
> I just want to state a couple of things and hear someone say, "that's
> right".
>
>
> 1. In a solr query you can have multiple fq's, but only a single q.
> And yes, I can simply AND the multiple "q"s together. Just want to avoid
> that if I'm wrong.
>
> 2. A subtler issue is that when a full query is executied, Solr must
> look at the schema to see how each field was tokenized (or not) and the
> various other filters applied to a field so that it can properly transform
> fields data (e.g., tokenize the text, but not keywords). As an aside, it
> would be nice if the queryparser could do the same thing in Lucene (I know,
> wrong forum :)).
> Scott
>