Mailing-List: contact users-help@jackrabbit.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@jackrabbit.apache.org
Received-SPF: pass (nike.apache.org: local policy)
From: Robert Haycock <Robert.Haycock@artificial-solutions.com>
To: "users@jackrabbit.apache.org" <users@jackrabbit.apache.org>
Subject: RE: How to make indexing case insensitive
Thread-Topic: How to make indexing case insensitive
Thread-Index: Ac3TteUh5NUjyqs4SAa4ChXjg3gUsgADMb3AAAFjnmA=
Date: Thu, 6 Dec 2012 15:43:47 +0000
Message-ID: 
 <565B17AF7325C54B83DFBD01450462DADCDE@EXCHANGE.artificial-solutions.com>
References: 
 <565B17AF7325C54B83DFBD01450462DADC41@EXCHANGE.artificial-solutions.com>
 <9C0FC4C8E9C29945B01766FC7F9D389817293D54F3@eurmbx01.eur.adobe.com>
In-Reply-To: 
 <9C0FC4C8E9C29945B01766FC7F9D389817293D54F3@eurmbx01.eur.adobe.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

That explains it. I thought the analyzer was also used for indexing, from m=
y Lucene days.

Thanks.

-----Original Message-----
From: Marcel Reutegger [mailto:mreutegg@adobe.com]=20
Sent: 06 December 2012 15:06
To: users@jackrabbit.apache.org
Subject: RE: How to make indexing case insensitive

Hi,

the analyzer is only used for fulltext query statements within a
CONTAINS() in SQL or jcr:contains() in XPath.

in your case you should rather use the built-in upper and lower case functi=
ons available in Jackrabbit. See the corresponding JIRA issue for a number =
of example queries:
https://issues.apache.org/jira/browse/JCR-638

Regards
 Marcel

> -----Original Message-----
> From: Robert Haycock [mailto:Robert.Haycock@artificial-solutions.com]
> Sent: Donnerstag, 6. Dezember 2012 14:40
> To: users@jackrabbit.apache.org
> Subject: How to make indexing case insensitive
>=20
> Hi,
>=20
> We have created our custom analyser and set the analyzer param in the=20
> workspace.xml...
> public class LowerCaseAnalyzer extends Analyzer {
>=20
>     @Override
>     public TokenStream tokenStream(String fieldName, Reader reader) {
>         TokenStream stream =3D new LetterTokenizer(reader);
>         stream =3D new LowerCaseFilter(stream);
>         return stream;
>     }
> }
> And when I debug I can see it is being hit and tokens being converted=20
> to lowercase.
> On a node I have a property called "name" with value "FLOW". Problem=20
> is, when I do a JCR2 SQL search the sql "WHERE name =3D 'FLOW'" returns=20
> the node but "WHERE name =3D 'flow'" doesn't. This is not what I expected=
.
> My assumption was that the LowerCaseFilter would convert all text to=20
> lower case before being indexed.
> Is my assumption wrong? How should I tell Jackrabbit to index=20
> everything as lower case so when I search I can convert all my search ter=
ms to lower case?
>=20
> Thanks,
>=20
> Rob.