[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
I18n and xim
From: |
Christian Gillot |
Subject: |
I18n and xim |
Date: |
Lun, 26 Nov 2001 21:10:01 +0100 |
Hi,
Rather new to GNUstep, I'd like to say that GNUstep is a great environment
and you made a wonderful work. That's why I'd like to contribute to GNUstep.
I've noted that there is an issue with text input. Right now, text input is
rather
ascii-centric. The way the text input is working, a european user can't type
for instance
dead keys o compose keys. And I guess the issue is worst for Asian users, well
I suppose they just can't type in their language in GNUstep applications.
So in order to improve that situation I'm looking forward implementing the
X11's "Input Method Protocol" or XIM.
That means patching core/xgps/Source/SharedX/XGContextEvent.m by putting
some #ifdef to use XIM rather than bare X11 key handling if we want so.
I've studied XIM and the way it's coded in GTK+, for european languages it
seems
rather simple, *but* for asian languages it's far more difficult. I mean, some
input
methods requires to know all the different text widgets in order to manage it
correctly.
This require to communicate at the XIM code the location and the size of all the
text input widgets when creating them and when they are resized.
My plan is first to implement a compose/dead keys (i.e. only for european
languages)
patch by preprocessors directives in XGContextEvent.m so that it don't break
anything.
Then for asian languages, it means far more changes because of the need to
inform
XIM about text input widgets and so a good coordination.
And now the questions :
- Is there anybody working on this right now ?
- To whom must I send the patches, or just on the list ?
- Do I need to do a copyright assignment to the FSF ?
And of course, any commentary is welcome.
Best regards,
PS : sorry about my bad english...
--
Christian Gillot <cgillot@neo-rousseaux.org>
GNU/Linux developer
- I18n and xim,
Christian Gillot <=