[Koha] About problem in Unicode Nepali

[Fischer, Katrin]

Hello Nitesh, 

 

we experienced the same problem with Hebrew records. Our problem was solved by using ICU chains for indexing with zebra

 

In the mail archive you can find files and some more information:

http://markmail.org/search/?q=icu+marc#query:icu%20marc+page:1+mid:hmtfccsjhs4kdfzx+state:results

 

We used that as a starting point to change our configuration to ICU. Not all of the enhanced search features work with ICU. It helped to turn off of most of the search system preferences and turn them back on one after the other while testing. I see in our configuration that we still don’t use QueryStemming and QueryWeightFields – not sure if both cause problems.

 

Good luck, 

 

Katrin

 

From: koha-bounces at lists.katipo.co.nz [mailto:koha-bounces at lists.katipo.co.nz] On Behalf Of Nitesh Rijal
Sent: Friday, August 27, 2010 12:50 PM
To: koha
Subject: [Koha] About problem in Unicode Nepali

 

Hello all..

 

Recently I have come across a problem in KOHA 3.0.3 related to nepali unicode.

 

Whenever I search any word in nepali, in my library site, it lists all the records in nepali. Even those which have no relevance with the search string.

 

Please see the link at   nnl.healthnet.org.np and search for a word. there instead of getting matching result, we get all the available Nepali result.

 

As an example text you can type "karki" or "nepal" that turns to "कार्की" and "नेपाल" in Nepali. All of them return the total number of results.

 

My though is that, it is searching for each letter of the word, instead of the whole word itself. Please see once,

 

what may be the problem, please reply.

 

Thank you.

-- 
Nitesh Rijal
BE IT
rijal.it at gmail.com
http://niteshrijal.com.np
http://facebook.com/openrijal
http://twitter.com/openrijal
+9779841458173

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.katipo.co.nz/pipermail/koha/attachments/20100827/33dc8fca/attachment-0001.htm