Patch for some Cyrillic languages

[Anton Zinoviev]

Hi!

Attached you can find a patch that makes several changes to Emacs (I
didn't know how to broke it in smaller pieces).  My comments about it
follow.

-----------------------------------------
 emacs/etc/HELLO

Added Belarusian, Bulgarian, Mongolian, Serbian and Ukrainian
greetings.

-----------------------------------------
 emacs/leim/quail/cyrillic.el

Very small documentation changes to bulgarian-phonetic and
bulgarian-bds.

New keyboard layouts: bashkir, kazakh, mongolian and cyrillic-prefix
(the purpose of the last is to support all Cyrillic languages in an
convenient way).

-----------------------------------------
 emacs/lisp/international/fontset.el

x-font-name-charset-alist: koi8-u and cp1251 fonts can be used also
for mule-unicode-0100-24ff characters.  This is because these Cyrillic
codesets are not fully covered by ISO 8859-5 or any other 8-bit
charset.  If some user requests with `-fn' some koi8-u or cp1251 font,
then he/she probable wants to use it for symbols that are not part of
ISO 8859-5.

-----------------------------------------
 emacs/lisp/international/mule-cmds.el

locale-language-names: Added support for Bashkir, Kazakh, Mongolian.
Changed language environments for Macedonian and Serbian.

-----------------------------------------
 emacs/lisp/language/cyrillic.el

Cyrillic-ISO: 

   * Changed the input method from cyrillic-yawerty to
     cyrillic-translit.  Reason: cyrillic-translit is language
     independent and can generate all letters from ISO 8859-5.

   * The sample-text contains greetings for some ISO 8859-5 languages
     (as this is done with latin-N language environments).  Similar
     change to the documentation of the language environment.

   * New language environments: Macedonian and Serbian.

cyrillic-unify-encoding: I suppose that the change I made is correct,
but I don't completely understand how cyrillic-unify-encoding works:

@@ -158,6 +192,7 @@
              (cond ((eq c ? ) ? )
                    ((eq c ?-) ?-)
                    ((eq c ?S) ?S)
+                   ((eq c ?#) ?#)
                    (t (decode-char 'ucs (+ #x400 i)))))
             (ec (aref table c))        ; encoding of 8859-5
             (uc (aref table u)))       ; encoding of Unicode


KOI8-R:

   * I haven't done any change to the generic language environment for
     KOI8-R.  However this encoding is used exclusively for Russian
     and given a specific language environment for Russian I think
     that the generic environment for KOI8-R is obsolete.

   * Changes for the Russian environment: 1. added utf-8 as an
     alternative coding system to koi8-r; 2. more meaningfull
     sample-text ("Happy wark with Emacs")

KOI8-U:

   * ccl-encode-koi8-u-font: small spelling correction.

   * Changes for the Ukrainian language environment: 1. added utf-8 as
     an alternative coding system to koi8-u; 2. added sample-text (I
     don't know well Ukrainian, but this seams to be "While the
     Ukrainian speach is alive, the Ukrainian people will be alive
     (Ogienko)".

ALTERNATIVNYJ:

   * No changes.  The generic language environment for alternativnyj
     makes no harm.  On the other hand I suppose that nobody will
     complain if you remove it.

Tajik language environment:

   * No changes.  I can say nothing definitely about Tajik language.
     I am not sure wether tajiks are still using Cyrillic alphabet.  I
     have never seen a font for KOI8-T, moreover I think this encoding
     is not supported by XFree.

CP1251:

This coding system was implemented differently that the others.  A
buffer with symbols from iso-8859-5 character set can not be saved
using CP1251 (although iso-8859-5 is a subset of CP1251).  There was
no support for cp1251 fonts and no mime-charset.  I haven't given much
thought about this, I simply reimplemented this coding system the same
way as the other Cyrillic coding systems are implemented.

During this I renamed this coding system from windows-1251 to cp1251
(the first stays still as an alias).  Reason: the name CP1251 is used
by Glibc and GNU utilities (includingly gettext), this name is what
most people are used to and will expect from Emacs.

This patch doesn't include a generic language environment for CP1251.
I will make such an environment latter since CP1251 is widely used for
several languages.

   * Bulgarian and Belarusian language environments: added utf-8 as an
     alternative coding system.  Added sample-text (the Bulgarian is
     "Happy working with the editor Emacs!", I don't know well
     Belarusian, but this seams to be "Lives during the ages the
     Belarusian speach -- spirit and glory of the people".

UTF-8:

I added four UTF-8 language environments: Cyrillic-UTF-8 (a generic
Cyrillic environment for UTF-8), Bashkir, Kazakh and Mongolian.

-----------------------------------------

Anton Zinoviev

emacs-cyrillic.diff.gz
Description: Binary data