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 gouthsmsimhadri <go...@gmail.com> on 2015/01/14 22:10:16 UTC
WordDelimiter Works differently in solr3X vs SolrCloud..?
Problem:
While migrating the solr version from 3.X(schema version is 1.4) to cloud
4.10.0 (schema version 1.5), I see a difference in the way the
worddelimiterfilter works for the below configuration
<filter class="solr.WordDelimiterFilterFactory"
generateWordParts="1"generateNumberParts="1" catenateWords="1"
catenateNumbers="1" catenateAll="0" preserveOriginal="1"
splitOnCaseChange="1"/>
In the current version, the catentateWords is done on the last postion of
the word delimited, but in the cloud the catenateWords always done on the
position 1 as below
EX: for token – “iPad2” at index
Current Version:
<http://lucene.472066.n3.nabble.com/file/n4179647/pic1.jpg>
Cloud Version:
<http://lucene.472066.n3.nabble.com/file/n4179647/pic2.png>
When “ipad2” is searched the parsed query on fieldXX using WDF
+fieldXX:\"(ipad2 ipad) 2\"^10.0" doesn’t find a match on the document
which contains “iPad2” in the cloud but finds a match on solr 3X version.
Did implementation of WDF change from 3x Vs Cloud. Is there any work around
to make “iPad2” match when queried for “ipad2” with WDF setting mentioned as
above.
--
View this message in context: http://lucene.472066.n3.nabble.com/WordDelimiter-Works-differently-in-solr3X-vs-SolrCloud-tp4179647.html
Sent from the Solr - User mailing list archive at Nabble.com.
Re: WordDelimiter Works differently in solr3X vs SolrCloud..?
Posted by gouthsmsimhadri <go...@gmail.com>.
Thanks Ahmet, that works.
-----
-goutham
--
View this message in context: http://lucene.472066.n3.nabble.com/WordDelimiter-Works-differently-in-solr3X-vs-SolrCloud-tp4179647p4179662.html
Sent from the Solr - User mailing list archive at Nabble.com.
Re: WordDelimiter Works differently in solr3X vs SolrCloud..?
Posted by Ahmet Arslan <io...@yahoo.com.INVALID>.
Hi,
You could try passing luceneMatchVersion argument to WordDelimiterFilterFactory and see if it works for you.
Factory returns Lucene47WordDelimiterFilter before LUCENE_4_8_0.
Ahmet
On Wednesday, January 14, 2015 11:10 PM, gouthsmsimhadri <go...@gmail.com> wrote:
Problem:
While migrating the solr version from 3.X(schema version is 1.4) to cloud
4.10.0 (schema version 1.5), I see a difference in the way the
worddelimiterfilter works for the below configuration
<filter class="solr.WordDelimiterFilterFactory"
generateWordParts="1"generateNumberParts="1" catenateWords="1"
catenateNumbers="1" catenateAll="0" preserveOriginal="1"
splitOnCaseChange="1"/>
In the current version, the catentateWords is done on the last postion of
the word delimited, but in the cloud the catenateWords always done on the
position 1 as below
EX: for token – “iPad2” at index
Current Version:
<http://lucene.472066.n3.nabble.com/file/n4179647/pic1.jpg>
Cloud Version:
<http://lucene.472066.n3.nabble.com/file/n4179647/pic2.png>
When “ipad2” is searched the parsed query on fieldXX using WDF
+fieldXX:\"(ipad2 ipad) 2\"^10.0" doesn’t find a match on the document
which contains “iPad2” in the cloud but finds a match on solr 3X version.
Did implementation of WDF change from 3x Vs Cloud. Is there any work around
to make “iPad2” match when queried for “ipad2” with WDF setting mentioned as
above.
--
View this message in context: http://lucene.472066.n3.nabble.com/WordDelimiter-Works-differently-in-solr3X-vs-SolrCloud-tp4179647.html
Sent from the Solr - User mailing list archive at Nabble.com.