[Koha-patches] [PATCH] (bug #4856) fix rebuild zebra to delete NSB/NSE chars

[Nahuel ANGELINETTI]

We have some examples like:

\x{88}Des \x{89}savons pour la vie

If you search for "savons", in title index, you will never find your record.



Le 08/06/2010 12:19, Frederic Demians a écrit :
> What kind of problem do you have with NSB/NSE characters and Zebra? It 
> works for me. With your patch, we loose information.
>
> NSB/NSE is not a UNIMARC specific issue:
>
>    http://www.loc.gov/marc/nonsorting.html
>
> NSB/NSE characters are already removed from biblio records before 
> displaying in several places in Koha (too many?). For me, regarding 
> those characters, there is a display issue and an indexing issue. 
> Display issue is solved by XSLT. Indexing issue should be fixed by 
> improving zebra indexation scheme: zebra phrase indexes should be 
> populated striping out string surrounded by NSB/NSE. I'm sure this 
> could be achieved by using Zebra DOM indexing model and appropriate XSLT.
>
> Regards,


-- 
Nahuel ANGELINETTI