Mailing-List: contact solr-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: solr-user@lucene.apache.org
Received-SPF: pass (nike.apache.org: local policy)
From: Jonathan Rochkind <rochkind@jhu.edu>
To: "solr-user@lucene.apache.org" <solr-user@lucene.apache.org>
Subject: RE: fq parameter with partial value
Thread-Topic: fq parameter with partial value
Thread-Index: AQHMBXNa+G1TursF3U2qHA94yCi9sZRzKeuAgAAlXACAACrIgIAAB3yA//+9RrQ=
Date: Thu, 28 Apr 2011 13:06:37 +0000
Message-ID: <665DBC51D0250A47B4F9306CE71E5FB705DF7F@JHEMTEBEX1.win.ad.jhu.edu>
References: <BANLkTikz8rquvV4Bp_L7ubm56HK9NSBvug@mail.gmail.com>
 <BANLkTikZKZD=eNUsEKWw_ejAuXxFGr0AJQ@mail.gmail.com>
 <BANLkTinrn3tJ0jS-bhYe-WRdfSU3fj4gcA@mail.gmail.com>
 <BANLkTi=79HMZ1Tc6t6P7Dj-v7TXapaaMGg@mail.gmail.com>,<BANLkTik2HRb_PX2VWEXWYXKyrzjYVdKLLw@mail.gmail.com>
In-Reply-To: <BANLkTik2HRb_PX2VWEXWYXKyrzjYVdKLLw@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

Yep, what you describe is what I do in similar situations, it works fine. =
=0A=
=0A=
It is certainly possible to facet on a tokenized field... but your individu=
al facet values will be the _tokens_, not the complete values. And they'll =
be the post-analyzed tokens at that.  Which is rarely what you want.  Thus =
the use of two fields, one tokenized and analyzed, one not tokenized and mi=
nimimally analzyed (for instance, not stemmed). =0A=
________________________________________=0A=
From: elisabeth benoit [elisaelisaelisa@gmail.com]=0A=
Sent: Thursday, April 28, 2011 9:03 AM=0A=
To: solr-user@lucene.apache.org=0A=
Subject: Re: fq parameter with partial value=0A=
=0A=
yes, the multivalued field is not broken up into tokens.=0A=
=0A=
so, if I understand well what you mean, I could have=0A=
=0A=
a field CATEGORY with  multiValued=3D"true"=0A=
a field CATEGORY_TOKENIZED with  multiValued=3D" true"=0A=
=0A=
and then some POI=0A=
=0A=
<field name=3D"NAME">POI_Name</field>=0A=
...=0A=
<field name=3D"*CATEGORY*">Restaurant Hotel</field>=0A=
<field name=3D"CATEGORY_TOKENIZED">Restaurant</field>=0A=
<field name=3D"CATEGORY_TOKENIZED">Hotel</field>=0A=
=0A=
do faceting on CATEGORY and fq on CATEGORY_TOKENIZED.=0A=
=0A=
But then, wouldn't it be possible to do faceting on CATEGORY_TOKENIZED?=0A=
=0A=
Best regards=0A=
Elisabeth=0A=
=0A=
=0A=
2011/4/28 Erick Erickson <erickerickson@gmail.com>=0A=
=0A=
> So, I assume your CATEGORY field is multiValued but each value is not=0A=
> broken up into tokens, right? If that's the case, would it work to have a=
=0A=
> second field CATEGORY_TOKENIZED and run your fq against that=0A=
> field instead?=0A=
>=0A=
> You could have this be a multiValued field with an increment gap if you=
=0A=
> wanted=0A=
> to prevent matches across separate entries and have your fq do a proximit=
y=0A=
> search where the proximity was less than the increment gap....=0A=
>=0A=
> Best=0A=
> Erick=0A=
>=0A=
> On Thu, Apr 28, 2011 at 6:03 AM, elisabeth benoit=0A=
> <elisaelisaelisa@gmail.com> wrote:=0A=
> > Hi Stefan,=0A=
> >=0A=
> > Thanks for answering.=0A=
> >=0A=
> > In more details, my problem is the following. I'm working on searching=
=0A=
> > points of interest (POIs), which can be hotels, restaurants, plumbers,=
=0A=
> > psychologists, etc.=0A=
> >=0A=
> > Those POIs can be identified among other things  by categories or by=0A=
> brand.=0A=
> > And a single POIs might have different categories (no maximum number).=
=0A=
> User=0A=
> > might enter a query like=0A=
> >=0A=
> >=0A=
> > McDonald=92s Paris=0A=
> >=0A=
> >=0A=
> > or=0A=
> >=0A=
> >=0A=
> > Restaurant Paris=0A=
> >=0A=
> >=0A=
> > or=0A=
> >=0A=
> >=0A=
> > many other possible queries=0A=
> >=0A=
> >=0A=
> > First I want to do a facet search on brand and categories, to find out=
=0A=
> which=0A=
> > case is the current case.=0A=
> >=0A=
> >=0A=
> > http://localhost:8080/solr /select?q=3Drestaurant  paris=0A=
> > &facet=3Dtrue&facet.field=3DBRAND& facet.field=3DCATEGORY=0A=
> >=0A=
> > and get an answer like=0A=
> >=0A=
> > <lst name=3D"facet_fields">=0A=
> >=0A=
> > <lst name=3D"CATEGORY">=0A=
> >=0A=
> > <int name=3D"Restaurant">598</int>=0A=
> >=0A=
> > <int name=3D"Restaurant Hotel">451</int>=0A=
> >=0A=
> >=0A=
> >=0A=
> > Then I want to send a request with fq=3D CATEGORY: Restaurant and still=
 get=0A=
> > answers with CATEGORY=3D Restaurant Hotel.=0A=
> >=0A=
> >=0A=
> >=0A=
> > One solution would be to modify the data to add a new document every ti=
me=0A=
> we=0A=
> > have a new category, so a POI with three different categories would be=
=0A=
> index=0A=
> > three times, each time with a different category.=0A=
> >=0A=
> >=0A=
> > But I was wondering if there was another way around.=0A=
> >=0A=
> >=0A=
> >=0A=
> > Thanks again,=0A=
> >=0A=
> > Elisabeth=0A=
> >=0A=
> >=0A=
> > 2011/4/28 Stefan Matheis <matheis.stefan@googlemail.com>=0A=
> >=0A=
> >> Hi Elisabeth,=0A=
> >>=0A=
> >> that's not what FilterQueries are made for :) What against using that=
=0A=
> >> Criteria in the Query?=0A=
> >> Perhaps you want to describe your UseCase and we'll see if there's=0A=
> >> another way to solve it?=0A=
> >>=0A=
> >> Regards=0A=
> >> Stefan=0A=
> >>=0A=
> >> On Thu, Apr 28, 2011 at 9:09 AM, elisabeth benoit=0A=
> >> <elisaelisaelisa@gmail.com> wrote:=0A=
> >> > Hello,=0A=
> >> >=0A=
> >> > I would like to know if there is a way to use the fq parameter with =
a=0A=
> >> > partial value.=0A=
> >> >=0A=
> >> > For instance, if I have a request with fq=3DNAME:Joe, and I would li=
ke=0A=
> to=0A=
> >> > retrieve all answers where NAME contains Joe, including those with=
=0A=
> NAME =3D=0A=
> >> > Joe Smith.=0A=
> >> >=0A=
> >> > Thanks,=0A=
> >> > Elisabeth=0A=
> >> >=0A=
> >>=0A=
> >=0A=
>=0A=