[mdk-re] ru_RU.UTF-8

[Serge Skorokhodov]

Здравствуйте!

Заранее прошу прощения за длинный постинг у тех, кому это не
интересно.

Я прошу разработчиков (AEN в первую очередь) прокомментировать
такой фрагмент переписки из cooker-i18n на linux-mandrake.com (на
вопрос о появившемся языке Russian/UTF-8 отвечает Pablo
Saratxaga).

>>>> Is it possible to have unicode support in console? For
>>>> example if i need to work with Latin, Cyrillic, Hebrew,
>>>> Arabic and if possible Mongolian
> 
> Not currently.
>
> Currently you can use unicode but you are limited to 512
> displayable glyphs. And also you are limited to left to right
> and non combining scripts.
> 
> Support for more displayable chars would needs an extra
> program, as there are used for Japanese and Chinese already. It
> isn't impossible to have one of such based on unicode and
> allowing use of TTF fonts for example, with FB console. I
> simply don't know of such a thing.
> 
> In any case support of RTL and/or complex scripts will reqire
> extra code; that isn't yet supported on X11, which I suppose
> has more priority.
> 
>>>> What do i need to during the instalation or/and
>>>> post-configuration? Same thing for X Window System. Can i
>>>> have this things at the same time?
> 
> On X11 things are much better. You can use utf-8 and have good
> display (limited only to available glyphs of your font); you
> are still limited to left to right and non composing scripts;
> however, there are some beta version libs supporting them.
> There is pango and Gtk2 (beta version is version 1.3; packages
> are included in one of the CDs), you need to recompile and link
> with it, and probably do some conversion of the code, but you
> should be able to display those RTL and complex script
> languages very well (look at http://www.pango.org/ for some
> screenshots).
> 
> There is also a project (not made public yet) to add indic
> languages support to XFree86, it is based, as pango, in utf-8
> encoding. It isn't yet available, so it won't be on MDK 8.1;
> but if you need it you should be able to apply the patches once
> they are publicized (and we will probably include them in our
> cooker vesion at that time too).
> 
>>> Well, for this reason I asked what Russian/UTF-8 in Language
>>> selection was for.
> 
> It sets LC_* to "ru_RU.UTF-8". so it uses unicode encoding.
> 
>>> I presumed it should do that.
> 
> Console support is problematic, the current console-tools are
> ignorant of all the nice i18n improvements that were done in
> glibc. they are a pain to correctly configure; and I haven't
> done it for russian utf-8 on console yet. It is better on X11.
> 
>>> Pixel, what you had in mind when
>>> you added it?
>> 
>> not me, pablo did! I don't understand much those pbs.
> 
> I did.
> 
> Now that utf-8 works (there are still some small problems here
> and there), I set utf-8 by default for some locales (az,
> tg,...) for which previously I have made non standard 8bit
> encodings. I also added an utf-8 choice for cyrillic languages
> as they have already a nice mess of multiple encodings to
> choose from, I suppose they are likely candidates to be
> interested in UTF-8.
> 
> I can remove it from DrakX if you thing it will made too much
> problems (utf-8 support is usable, but has still several
> problems, old 8bit encodings may be better for Russian)
> 
> In any case, it is always possible to edit after install the
> /etc/sysconfig/i18n or $HOME/.i18n to use xx_YY.UTF-8
> definitions for all LC_* entries, and then use unicode locale.

Заранее благодарен.

Serge Skorokhodov aka suralis
11.09.2001 suralis-s на mtu-net.ru