Re: gtk and composing characters

[Jens Herden <jens khmeros info>]

> The new version of XOrg (7.0) is about to be published soon, so there
> are chances these changes will not make it immediatelly into the new
> version.
> Afaik, XOrg is in second release candidate, so changes are not easy to
> go in just now.

We are not really in a hurry, because I have a script that patches up an 
installed Xorg or Xfree installation.

> The "compose" file is
> http://cvs.freedesktop.org/xorg/xc/nls/Compose/en_US.UTF-8
> and it is referenced from
> http://cvs.freedesktop.org/xorg/xc/nls/compose.dir?view=markup
> Here you need to add a line such as
>
> the_location_of_the_compose_file:		km_CB.UTF-8

probably km_KH.UTF-8 because the country code of Cambodia is kh.


> which means where the compose file. You will notice that this often
> points to en_US.UTF-8.
>
> Have a look at a similar bug for Kyrghiz, for the full steps:
> https://bugs.freedesktop.org/show_bug.cgi?id=2609

Sorry for my slowness, but you lost me here a little. I have difficulties to 
understand how these different locale specific compose files work together. 

I will explain what I am currently doing in my installation script. I look for 
every locale that matches *.UTF-8 and I append my Khmer compose rules to the 
Compose file of these locale. I am doing this because there was no locale for 
Cambodia (kh) in the past and I want to be sure that everyone, no matter 
which locale setting they are using, is able to use the Khmer keyboard. In 
fact most people will use it in conjunction with an English keyboard because 
there are no Roman characters in it. 

I understand that when I am following what you propose I will create a Compose 
file that gets used only when my system is using a Cambodian locale but not 
in case of en_US.UTF-8
This is not want I want to achieve. 

Thanks for your patience 

Jens