Re: [linuxiran] XFree86 4.3

[Arash Zeini]

On Wednesday 14 May 2003 13:01, Aryan Ameri wrote:

> On Monday 12 May 2003 17:24, Arash Zeini wrote:
> > On Monday 12 May 2003 17:00, Aryan Ameri wrote:
> > > I looked at it, and the patch is pretty simple, it adds the unicode
> > > values of arabic to X. It also does include JEH, PEH, GAF and CHEH.
> >
> > You have to explain! I am not sure how it adds the values to X.
> > Remember that it is not enough to add the vlaues to the xkb map.
> > That's waht we have in our map too. The problem is that these values
> > are not passed to different X components/layers correctly. For KDE
> > this problem has been taken care at the Qt level.
> > The lines that you see in the patch are comments, I guess as kind of
> > a reference.
>
> Hey, I ain't a programmer, but beleive me, I have the ability to
> distiguish comments from code !!
>
> OK, maybe that one patch doesn't add Farsi characters to X, but then
> what is it? you say it adds Farsi to X keyboard map? And what is the
> result of that?

No I am saying that the patch modifies the Arabic keyboard and it adds the 
comments to the map file as a point of reference.
I do not remember the patch now in detail but it was not adding any Farsi 
issues to the keyboard.

> I really want to ask this question, what does it take to add farsi to X?
> What kind of a patch does such a thing? What are we looking for?
>
> You always say "For KDE this problem has been taken care at the Qt
> level." Well, it does seem to me that this taken care of, is just a
> patch to Qt, and that Qt is just a toolkit. What should we do to take
> care of it, in the X level? X is also a toolkit, they patched Qt, and
> added Farsi to it, now, what does it take to patch X and add farsi to
> it?

A good question, probably best asked on the right X mailing list. Behdad had 
patched Qt at the time, maybe he can contribute to X as well!
Behdad?

> > > Although it assumes that these are also arabic letters. I am not
> > > familiar with the Unicode table, and I don't know wether according
> > > to Unicode, if they are considered arabic or not. But anyway, the
> > > patch adds all arabic and farsi characters to X.
> >
> > Yes Farsi is part of the Arabic code page. I.e., it is a "subset" of
> > Arabic. The same applies to all the other languages that use a
> > modified Arabic script.
>
> Good to know
>
> > Agreeing with ArashB I assume that the patch or improvement has
> > happened somewhere in the lines of GTK or the like.
>
> Completely disagree. Mozilla might build on top of Gtk, but I can assure
> you that OOo doesn't have anything to do with Gtk. Sun doesn't use any
> toolkit, not Qt, Not MFC, not .Net, Not Motif, and not Gtk, in building
> OOo. OOo uses it's own (rather inefficent) structure.

It looks now as if RH patched X. We have to take a closer look at the 
avaialbel patches. 
I ask myself why these patches aren't in X???

Greetings,
Arash
-- 
The FarsiKDE Project
www.farsikde.org