
Hi Boris, On Sat, Jan 17, 2004 at 10:31:50AM +0200, Boris Ratner wrote:
I use OTRS with POSTGRESQL where the otrs database defined as "UNICODE". I need OTRS to support messages in hebrew language (that was the main reason for UNICODE database). When i send a hebrew message encoded in UTF-8 everything is fine. The problem is that (almost) nobody uses that encoding to send e-mail , vast majority of people use "Windows-1255" and minority use "iso-8859-8" (for unkown to me reasons mozilla-mail sends hebrew mail in "iso-8859-8-i" encoding). When one of these encodings (win || iso) hits otrs@mydomain the following happens:
<log extract><snip> procmail: Executing "/opt/otrs//bin/PostMaster.pl" DBD::Pg::db do failed: ERROR: invalid byte sequence for encoding "UNICODE": 0xed20e7 a t /opt/otrs/Kernel/System/DB.pm line 172. ERROR: OTRS-PM-10 Perl: 5.8.2 OS: linux Time: Tue Jan 13 02:21:47 2004 [..]
POSSIBLE SOLUTION
Now i'm working on a script to be run before ../bin/Postmaster.pl that will convert _ALL_ incoming mail to UTF-8. IMHO there is no need for ORTS to handle encodings internally - unicode works great.
OTRS 1.2 (current cvs head) should work with utf-8 as backend charset (will convert all charsets to utf-8). Just set "$Self->{DefaultCharset} = 'utf-8';" in Kernel/Config.pm. PS: OTRS 1.2 beta1 will be released on ~ 2004-01-22.
Kind Regards, Boris Ratner.
Martin Edenhofer -- ((otrs.de)) :: OTRS GmbH :: Norsk-Data-Str. 1 :: 61352 Bad Homburg http://www.otrs.de/ :: Manage your communication!