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

List:       koffice
Subject:    Re: KOffice - Kpresenter - are they bugs?
From:       Thomas Diehl <thd () kde ! org>
Date:       2001-10-03 18:32:29
[Download RAW message or body]

Am Mittwoch, 3. Oktober 2001 15:55 schrieb Duvvuri Venu Gopal:

> Is any body working in KDE group for Indian Languages. I want to know! Much
> work was done by the www.cdac.com/org people in Pune for unit and windows.
> Because if the KDE or Linux to be popular in India it should support Indian
> Languages

Support for Indian languages won't probably there before Qt 3.1 (see attached 
message by Lars Knoll for the details). Any help in this respect is very 
welcome of course.

Only available language in this area is Tamil for the time being. For more 
information see the KDE translation sites at http://i18n.kde.org/ (esp. the 
contact addresses and team pages at http://i18n.kde.org/teams/) and the 
translators' mailing list at http://mail.kde.org/mailman/listinfo/kde-i18n-doc

Regards,

Thomas


Subject: Re: Telugu Language
Date: Mon, 23 Apr 2001 13:53:38 +0200
From: Lars Knoll <lars@trolltech.com>
To: "Thomas Diehl" <thd@kde.org>, "srinu@nol.net.in" <srinu@nol.net.in>
Cc: "Harsh Kumar" <harshku@vsnl.com>, "Karunakar Guntupalli" 
<indlinux@rediffmail.com> "U. Mohan Varma" <mvarma@cfdlab.aero.iisc.ernet.in>


Hi,

Telugu, as all other Indic languages is difficult to support in a complete
way. The main problem is, that all these scripts have very complex rules as
how to render a string of characters. If you once read the section about
Devanagari or Tamil in the Unicode book, you probably know what I mean.

All these languages have very complex shaping behaviour (the shape of the
character varies according to position), reordering (some vocals, in
Devanagrai the i, are moved before the consonant they belong to), and a huge
amount of ligatures.

We still don't have support for these scripts in Qt-3.0, but will try to get
support for them into Qt as soon as possible. There are a lot of problematic
issues to resolve before that however.

The first problematic issue is the difference between characters and glyphs.
Unicode only defines "characters" for Indic languages (as opposed to glyphs,
which are shaped variations of specific characters), so the engine needs to
find out which glyphs to use for a sequence of characters in a string. This
implies, that fonts used for rendering an indic script have to be a lot
bigger than the range of characters defined in Unicode. At the moment, there
is no standard on how to encode the glyphs needed for indic scripts (apart
from Microsofts approach with open type). We will need such a standard (or
resort to open type) to be able to handle indic scripts.

The other issues are more or less internal to Qt, but nevertheless important.
Qt will need a plugin mechanism to be able to load text engines for different
scripts. These engines will need support for quite some things, most
noticeably among them support for shaping, ligatures and glyph reordering.
Other issues are cursor positioning and line breaking.

So you see, that all this is quite a complex issue, and will need a lot of
work, before being able to write some indic language in a text editor.

Best regards,
Lars


-- 
KDE translation: http://i18n.kde.org/
Deutsche KDE-Uebersetzung: http://i18n.kde.org/de/

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