Re: Search for All CAPS words

2015-08-07 Thread rks_lucene
Took me a while but I tried it and its works perfectly. Thanks a lot !!
Ritesh



--
View this message in context: 
http://lucene.472066.n3.nabble.com/Search-for-All-CAPS-words-tp4219893p4221597.html
Sent from the Solr - User mailing list archive at Nabble.com.


Re: Search for All CAPS words

2015-07-30 Thread Alexandre Rafalovitch
So, what you want is to duplicate a specific token, rename one of the
copies, and inject it with the same offset as the original. So GATE =>
gate, _gate but gate=>gate.

That, to me, is a custom token filter. You can probably use
KeywordRepeatFilterFactory as a base:
http://www.solr-start.com/javadoc/solr-lucene/org/apache/lucene/analysis/miscellaneous/KeywordRepeatFilterFactory.html
(you can click through to the Filter and then source from there).

Regards,
   Alex.

Solr Analyzers, Tokenizers, Filters, URPs and even a newsletter:
http://www.solr-start.com/


On 30 July 2015 at 13:53, rks_lucene  wrote:
> Thanks and I did think of the copy field option. So what you are suggesting
> is that I have a copyfield in which I do not keep the lowercase factory
> analyzer in my indexing/query chains.
>
> I am afraid that would not help if my search query is complex with many
> words (say a boolean with proximity operators) because the full search
> string would have go into the copyfield (not having the lowercase). The rest
> of the words other than /GATE/ wouldnt match properly then.
>
> Ritesh
>
>
>
>
>
>
>
> --
> View this message in context: 
> http://lucene.472066.n3.nabble.com/Search-for-All-CAPS-words-tp4219893p4219959.html
> Sent from the Solr - User mailing list archive at Nabble.com.


Re: Search for All CAPS words

2015-07-30 Thread rks_lucene
Thanks and I did think of the copy field option. So what you are suggesting
is that I have a copyfield in which I do not keep the lowercase factory
analyzer in my indexing/query chains.

I am afraid that would not help if my search query is complex with many
words (say a boolean with proximity operators) because the full search
string would have go into the copyfield (not having the lowercase). The rest
of the words other than /GATE/ wouldnt match properly then.

Ritesh







--
View this message in context: 
http://lucene.472066.n3.nabble.com/Search-for-All-CAPS-words-tp4219893p4219959.html
Sent from the Solr - User mailing list archive at Nabble.com.


Re: Search for All CAPS words

2015-07-30 Thread Alexandre Rafalovitch
Have you tried copyField with different field type for different
fields yet? That would be my first step. Make the copied field
indexed-only, not stored for efficiency.

And you can then either search against that copied field directly or
use eDisMax against both fields and give that field a higher priority.

Regards,
   Alex.

Solr Analyzers, Tokenizers, Filters, URPs and even a newsletter:
http://www.solr-start.com/


On 30 July 2015 at 10:00, rks_lucene  wrote:
> Hi,
>
> I need the capability to search for /GATE/ separately from /gate/.
>
> I cannot remove the lowercase filter factory in both my search and analysis
> chains since that will break many other search scenarios.
>
> Is there a way to payload/mark an ALL CAPS word in the index analyzer chain
> before it gets lowercased (by the lowercasefilterfactory) so that I can
> search it with some custom grammar and logic in my query parser.
>
> Say I want:
>
> Field:_gate to match /GATE/ only
>
> Field:gate to match both /GATE/ and /gate/
>
> Any pointers would be helpful.
>
> thanks
> Ritesh
>
>
>
> --
> View this message in context: 
> http://lucene.472066.n3.nabble.com/Search-for-All-CAPS-words-tp4219893.html
> Sent from the Solr - User mailing list archive at Nabble.com.


Search for All CAPS words

2015-07-30 Thread rks_lucene
Hi,

I need the capability to search for /GATE/ separately from /gate/.

I cannot remove the lowercase filter factory in both my search and analysis
chains since that will break many other search scenarios.

Is there a way to payload/mark an ALL CAPS word in the index analyzer chain
before it gets lowercased (by the lowercasefilterfactory) so that I can
search it with some custom grammar and logic in my query parser.

Say I want:

Field:_gate to match /GATE/ only

Field:gate to match both /GATE/ and /gate/

Any pointers would be helpful.

thanks
Ritesh



--
View this message in context: 
http://lucene.472066.n3.nabble.com/Search-for-All-CAPS-words-tp4219893.html
Sent from the Solr - User mailing list archive at Nabble.com.