[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-i18n-doc
Subject:    Re: Tamil translation of kdelibs.po
From:       Thuraiappah Vaseeharan <vasee () ee ! umanitoba ! ca>
Date:       2000-03-18 17:51:43
[Download RAW message or body]



On Sat, 18 Mar 2000, Stephan Kulow wrote:

> Thomas Diehl wrote:
> > 
> > On Fri, 17 Mar 2000 14:00:36 -0800 (PST), Sivakumar Shanmugasundaram
> > wrote:
> > 
> > >Unicode support for Tamil is not complete (atleast
> > >the Tamil language experts complain). So TSCII is
> > >a new encoding standard that is popular among Tamil
> > >web community (do you know that among all Indian
> > >languages, Tamil has the most websites, more than
> > >Hindi).
> > >I am trying to locate a Unicode font with Tamil
> > >glyphs.
> > >Could you let me know what Unicode fonts are being
> > >used by other translation team, like Thai.
> > 
> > Unfortunately, use of Unicode format (utf8) is still not very
> > widespread among KDE translators although we are switching already (eg
> > desktop.po _must_ be in utf8 format). All I know about this stuff is
> > from the Unicode Howto at http://developer.kde.org/kde2arch/KDE-Unicode-Howto.html.
> > But you can always try and contact your fellow coordinators,
> > either via the mailing list (kde-i18n-doc@kde.org) or directly from
> > the addresses at http://i18n.kde.org/teams/.
> > 
> You don't need unicode fonts per se. But the files should be stored
> in unicode. How they are represented on screen is another story.
> Currently KDE mixes both a bit, but you can change the font encoding 
> in the KControlCenter while the charset of the PO files will still
> be read out of the charset file.
> 
> Greetings, Stephan
> 


What Stephan says is correct. I think the problem is that the mapping
between the Unicode Tamil encoding and TSCII is not one-to-one. Apparently
the state govt. of Tamil Nadu (in India), has joined the Unicode
consortium to get the encoding changed in future versions. Until this
happens, we can just "translate" the TSCII code points to the Unicode
space and proceed. When a standard does emerge we can do the conversion
again.

About Recode, KBabel and so on: I'm sure TSCII can be made to work with
these tools. TSCII just maps the tamil glyphs to the high ASCII range
(128-256). So any tools/programs that are 8-bit clean should work with
TSCII.

Vasee

[prev in list] [next in list] [prev in thread] [next in thread]