emacs russification

emacs russification

[basileia]

[-- Attachment #1: Type: text/plain, Size: 937 bytes --]

Dear sirs!

As I can see, there are two ways to work with cyrillic letters in Emacs: first based on iso8859-5 and second based on utf-8. But at present all available encoders/decoders and language environments for cyrillic languages are based on iso8859-5. Moreover, there is no support (I mean, in user interface level) for cyrillic codepage cp1251, although it is the only codepage used in Windows and it becames popular in Linux, too. And, as I can see, there are no input methods for utf-8 cyrillics included in distribution. So, I want to ask, if there are some plans to improve cyrillic support in Mule/Emacs. May be, it could be a good idea to replace iso8859 as a base for encoding cyrillic characters with utf-8? If somebody works with such a project? And, if not, can I contribute to Emacs writing new utf8-based encoding/decoding procedures and input methods?

Alexej Kryukov
Moscow State University
basileia@yandex.ru

[-- Attachment #2: Type: text/html, Size: 1596 bytes --]

Re: emacs russification

[Eli Zaretskii]

On Mon, 8 Apr 2002 basileia@yandex.ru wrote:

> As I can see, there are two ways to work with cyrillic letters in Emacs:
> first based on iso8859-5 and second based on utf-8.

Not entirely true: Emacs also supports KOI8-R, Cyrillic-Alt, and several 
codepages, including 1251.

> But at present all
> available encoders/decoders and language environments for cyrillic
> languages are based on iso8859-5.

This is not really relevant from the user point of view: Emacs indeed 
uses iso8859-5 as the basis for its internal Cyrillic character set, but 
you shouldn't be worrying about this in normal use.  If you have specific 
problems with this, please describe them.

> Moreover, there is no support (I mean,
> in user interface level) for cyrillic codepage cp1251, although it is
> the only codepage used in Windows and it becames popular in Linux, too.

Yes, there is; it just isn't turned on by default (except on Windows).  Try 
these commands:

	M-x codepage-setup RET 1251 RET
	C-x RET k cp1251 RET

This loads the tables needed for cp1251 support, and tells Emacs the 
keyboard sends cp1251 codes.

> And, as I can see, there are no input methods for utf-8 cyrillics
> included in distribution.

I think development code includes some of these input methods.  But 
Unicode support for Emacs is work in progress, so it's quite possible 
that many Unicode-based input methods are missing.

> So, I want to ask, if there are some plans to
> improve cyrillic support in Mule/Emacs. May be, it could be a good idea
> to replace iso8859 as a base for encoding cyrillic characters with
> utf-8? If somebody works with such a project?

Work is under way to switch _all_ Emacs characters to be based on UTF-8.  
When that happens, Cyrillic characters will be switched as well.

> And, if not, can I
> contribute to Emacs writing new utf8-based encoding/decoding procedures
> and input methods?

I think a Unicode input method for Cyrillic languages will be most 
welcome, even now.  But please use the anon CVS access to Emacs sources 
first to see what is there already.

Please also continue discussions of this issue on emacs-devel@gnu.org, 
which is the mailing list where Emacs development is discussed.