idea about faceting

[Marc Sturlese <ma...@gmail.com>]

Hey there,

I am faceing a problem doing filed facets and I don't know if there exist
any solution in Solr to solve my problem.
I want to do facets with a field that is very small text. To do that I am
using the KeywordTokenizerfactory to keep all the words of the text in just
one token. I use LowerCaseFilterFactory not to miss cases that doesn't match
due to uppercase and ISOLatin1AccentFilterFactory not to miss cases that
doesn't match because of the accents.

The problem apears here, I would like to show the facets with accents or
uppercase.

In my old Lucene system not using Solr I use to create my facet fields with
accents but at searching time I removed the accents and uppercases manually
with java. So, i did the search without accents and upper case but I was
able to show them later.

I have been playing with the facet solr source code but can't find the way
to solve my problem...

Does anyone have an idea about how could I reach this goal?

Thanks in advance

-- 
View this message in context: http://www.nabble.com/idea-about-faceting-tp20638850p20638850.html
Sent from the Solr - User mailing list archive at Nabble.com.

RE: idea about faceting

[Lance Norskog <go...@gmail.com>]

Index two fields instead of one. One field has ISOLatin1Filter, LowerCase
etc. and one does not. Search the processed, filter on the raw one. For more
searching power, you can even index a third field with the Soundex/Metaphone
phoneme translators.

-----Original Message-----
From: Marc Sturlese [mailto:marc.sturlese@gmail.com] 
Sent: Saturday, November 22, 2008 9:43 AM
To: solr-user@lucene.apache.org
Subject: idea about faceting


Hey there,

I am faceing a problem doing filed facets and I don't know if there exist
any solution in Solr to solve my problem.
I want to do facets with a field that is very small text. To do that I am
using the KeywordTokenizerfactory to keep all the words of the text in just
one token. I use LowerCaseFilterFactory not to miss cases that doesn't match
due to uppercase and ISOLatin1AccentFilterFactory not to miss cases that
doesn't match because of the accents.

The problem apears here, I would like to show the facets with accents or
uppercase.

In my old Lucene system not using Solr I use to create my facet fields with
accents but at searching time I removed the accents and uppercases manually
with java. So, i did the search without accents and upper case but I was
able to show them later.

I have been playing with the facet solr source code but can't find the way
to solve my problem...

Does anyone have an idea about how could I reach this goal?

Thanks in advance

--
View this message in context:
http://www.nabble.com/idea-about-faceting-tp20638850p20638850.html
Sent from the Solr - User mailing list archive at Nabble.com.

Re: idea about faceting

[Yonik Seeley <yo...@apache.org>]

On Sat, Nov 22, 2008 at 12:43 PM, Marc Sturlese <ma...@gmail.com> wrote:
> I am faceing a problem doing filed facets and I don't know if there exist
> any solution in Solr to solve my problem.
> I want to do facets with a field that is very small text. To do that I am
> using the KeywordTokenizerfactory to keep all the words of the text in just
> one token. I use LowerCaseFilterFactory not to miss cases that doesn't match
> due to uppercase and ISOLatin1AccentFilterFactory not to miss cases that
> doesn't match because of the accents.
>
> The problem apears here, I would like to show the facets with accents or
> uppercase.

If multiple entries are normalized to a single one for the purposes of
faceting, how would one decide which original entry to use when
displaying?  It seems like extra mapping information is needed for the
presentation labels of facet values.  Most applications currently
maintain this information outside Solr.

-Yonik

> In my old Lucene system not using Solr I use to create my facet fields with
> accents but at searching time I removed the accents and uppercases manually
> with java. So, i did the search without accents and upper case but I was
> able to show them later.
>
> I have been playing with the facet solr source code but can't find the way
> to solve my problem...
>
> Does anyone have an idea about how could I reach this goal?
>
> Thanks in advance
>
> --
> View this message in context: http://www.nabble.com/idea-about-faceting-tp20638850p20638850.html
> Sent from the Solr - User mailing list archive at Nabble.com.
>
>