🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.std.internat
Article #1031

Re: Unicode -> CNS11643-1992 table

#1031
From: John H. Jenkins
Date: Fri, 26 May 1995 00:00
73 lines
3722 bytes
In article <3ps8hg$3i7@kudpc.kudpc.kyoto-u.ac.jp>, Koichi Yasuoka said:

>I've finished to write a new mapping table from Unicode to CNS11643-1992.
>But I couldn't find "Four-voice marks for Bopomofo" in Unicode (they are
>1-256C to 1-2570 in CNS11643).  Doesn't Unicode assure us of the round-
>trip compatibility with the 1st plane of CNS11643?

No, Unicode does not guarantee round-trip compatibility with CNS 11643.  

It *does* guarantee round-trip compatibility with the ideographs of planes
1 and 2, but that's it.  In particular, the duplicate set of radicals in
11643 is left out of Unicode.  I quote from Unicode's CNS 11643-1986
mapping table, available by anonymous ftp from unicode.org:

#  WARNING!  It is currently impossible to provide round-trip compatibility
#     between CNS 11643-1986 and Unicode.  
#
#  (1) Some characters are not currently mapped because
#     of conflicts with other mappings.  They include the following:
#
#       CNS         Description                    Comments
#
#       0x12224      SPACING HEAVY OVERSCORE       not in Unicode
#       0x12226      SPACING HEAVY UNDERSCORE      not in Unicode
#
#   (2) Some characters are not currently mapped because CNS 11643-1986 includes
#       duplicate versions of some ideographs which are included in Unicode only
#       once.  They include the following:
#
#       0x1243E through 0x12440, HANGZHOU NUMERAL TEN, TWENTY, and
THIRTY         
#       0x12721 through 0x12939, KangXi radical set
#
#   Entries for these characters are not included in this table.
#
#  Notes:
#
#  1.  There is an uncertainty in the mapping of the CNS 11643-1986 character
#     0x1256D.  This character occurs within the CNS 11643-1986 block of
tone marks
#     for bopomofo and is intended to be the tone mark for the first tone in
#     Mandarin Chinese.  We have selected the mapping U+02C9 MODIFIER LETTER
#     MACRON (Mandarin Chinese first tone) to reflect this semantic.  
#     However, because bopomofo uses the absense of a tone mark to indicate
#     the first Mandarin tone, most implementations of CNS 11643-1986 represent
#     this character with a blank space, and so a mapping such as U+2003
EM SPACE
#     might be preferred.  
#     

As for the bopomofo tone marks, they are in Unicode at

0x1256C  0x02D9   #  DOT ABOVE
0x1256D  0x02C9   #  MODIFIER LETTER MACRON
0x1256E  0x02CA   #  MODIFIER LETTER ACUTE ACCENT
0x1256F  0x02C7   #  CARON
0x12570  0x02CB   #  MODIFIER LETTER GRAVE ACCENT

Unicode does not distinguish the bopomofo tone marks from other modifier
letters.

BTW, this table differs in a number of respects from the table available
from Unicode.org.  

People interested in doing CNS 11643 to Unicode conversions should be
aware of the differences and decide which mapping to use.  I should also
caution that Unicode and SC2/WG2/IRG are working on definitive, official
mappings between plans 4 and above of CNS 11643-1992 and Unicode, and so
any mappings in this table should be considered provisional.

-- 
John H. Jenkins
John_Jenkins@taligent.com
Technical Details
Message-ID: <John_Jenkins-2605951356320001@tseng.taligent.com>
Path: arc1-ca.pugleaf.net!unknown!not-for-mail

Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: f9a59,9ee27d04b2a72428
X-Google-Attributes: gidf9a59,public
X-Google-Thread: 10d93a,9ee27d04b2a72428
X-Google-Attributes: gid10d93a,public
X-Google-Thread: f56e0,9ee27d04b2a72428
X-Google-Attributes: gidf56e0,public
From: John_Jenkins@taligent.com (John H. Jenkins)
Subject: Re: Unicode -> CNS11643-1992 table
Date: 1995/05/26
Message-ID: <John_Jenkins-2605951356320001@tseng.taligent.com>
X-Deja-AN: 103365953
organization: Taligent, Inc.
newsgroups: comp.std.internat,alt.chinese.computing,chinese.comp.software