Author: Alex Barkov
Email: [EMAIL PROTECTED]
Message:
Mnogosearch itself can't remove accents. However, if Oracle
has a accent independant collation for indexes, it is possible.
I heared thast somebody succesfully configured this with MS SQL.


> I want to implement accent insentive search. At the present time, when I search for 
>"pierre m�tras" on my Web site, I don't obtain the same results as when I 
>use "pierre metras".
> 
> Where should I look at?
> 
> My configuration:
> - mnoGoSearch v3.1.19 with a MultiDB Oracle database
> - a LocalCharset defined as iso-8859-1 in indexer.conf and in our customized search 
>forms.
> - Two default languages: fr and en.
> - and the Oracle database is configured to support accents with 
>NLS_LANG=AMERICAN_AMERICA.WE8ISO8859P1. This variable is also set before running the 
>indexer.
> 
> In database, accents are kept and I can see "m�tras" and 
>"metras" in the DICT6 table. I have also some "mitras" and words 
>where it seems the indexer has lost the 7th bit...


Reply: <http://www.mnogosearch.org/board/message.php?id=3771>

___________________________________________
If you want to unsubscribe send "unsubscribe general"
to [EMAIL PROTECTED]

Reply via email to