The following reply was made to PR lib/33090; it has been noted by GNATS.

From: Takehiko NOZAKI <th-nozaki@netwrk.co.jp>
To: cheusov@tut.by
Cc: gnats-bugs@netbsd.org
Subject: Re: lib/33090: patch for new locales: ru_BY.CP1251, ru_RU.CP1251 and be_BY.CP1251
Date: Thu, 16 Mar 2006 21:27:33 +0900

 hi,
  
 > Can anybody apply a small patch to libc,
 > building a few new locales for Russian and Belarusian
 > languages?
  
 I had already committed your patch with some modification, thanks.
   http://mail-index.netbsd.org/source-changes/2006/03/14/0018.html
 please checkout latest CVS HEAD.
  
 > Actually I just copy subpart of bg_BG.CP1251 file
 > to charset/CP1251 file and "#include" it
 > from new locates ctype/ files.
 
 src/share/locale/ctype/charset/* is for ISO-2022-* locale,
 i think other locale should be independent from those.
 
 > P.S.
 > What do such lines mean?
 
 see man 1 mklocale.
 
 this CHARSET token set bit and mask to localesrc's character index
 to convert from/to wchar_t.
 
 for more detailed information about ISO-2022 wchar_t mapping,
 see src/lib/libc/citrus/modules/cirus_iso2022.c.
 
 > All files in $SRC/share/locale/ctype/charset
 > contains this magic, my patch doesn't add it.
 > man mklocale says
 >    "CHARSET    Controls character set for subsequent runes"
 > The above magic symbols are not clear enough for me :-(
 > 
 
 this token's string means ISO/IEC 2022 escape sequence.
 
 >     CHARSET         ",A"
 
 `,' is 96(0x20-0x7F) charset.
 `A' is final-byte of ISO-2022 escape sequence, means ISO 8859-1:GR
 
 
 >     CHARSET         "(I"
 
 `(' is 94(0x21-0x7E) charset.
 `I' is for japanese JISX0201.
 
 
 >     CHARSET         "$(@"
 
 `$' means this is multibyte charset.
 `@' is for japanese JISX 0208-1978.
 
 for more detailed information about ISO/IEC 2022,
 read following documentation:
 
 http://www.ecma-international.org/publicaions/files/ECMA-ST/Ecma-035.pdf
 http://www.itscj.ipsj.or.jp/ISO-IR/
 
 
 truly yours.
 --
 Takehiko NOZAKI<tnozaki@netbsd.org>