Re: GB 12345-90

From: John H. Jenkins (
Date: Mon Jan 23 1995 - 14:55:14 EST

At 10:04 AM 1/23/95, Ken Lunde wrote:
> I have been using the CJKXRef.text and GB/GB12345 in the
>pub/MappingTables/EastAsiaMaps directory as valuable resources.
>However, I found some oddities that may turn out to be either errors
>or corrections made to GB 12345-90.
> If you are familiar with GB 12345-90 (which I am sure you
>are), you know that it is essentially GB 2312-80 with traditional
>Chinese characters substituted for the simplified versions. About 2/3
>of the Chinese characters are identical between the two standards.
> The following GB 12345-90 code points (in decimal row-cell
>notation) are cases where the CJKXRef.text file claims that the
>traditional form should be substituted for the GB 2312-80 form, but
>the GB 12345-90 standard (I have it in printed form) does not agree:

[ Actual code points edited out... ]

>There are probably more cases of the above phenomenon.
> My question is: which is correct? GB 12345-90 or CJKXref.text?

If I understand you correctly, our tables indicate that the code points in
different look different in GB2312 and GB12345 and your printed copy of
GB12345 says they look the same.

I haven't looked at all the examples you give, but so far, the copy of
GB12345 we have agrees with the contents of CJKXRef.text.

We have noted some inconsistency on the part of the PRC -- sometimes they
print a given character in a given edition of a standard with the
simplified glyph, and sometimes with the traditional one.

In any event, CJKXRef.text reflects the work of the CJK-JRG, which is a
normative part of 10646. It should therefore be taken as "correct."

>Corrections and additions were made to GB 2312-80 through the GB
>6345.1-86 standard. Is there a similar standard that made corrections
>to GB 12345-90? If so, what is the document, and how can I get a hold
>of a copy? If not, under whose authority were such changes made?


> Also, the CJKXRef.text file claims that there are characters
>in rows 92 (58 hanzi) and 93 (94 hanzi from GB 12052-89) of GB
>12345-90, but the GB 12345-90 has nothing defined for those rows.

The actual version of GB12345 used by the CJK-JRG was an ad hoc extension
-- the extra characters are simply characters the PRC wanted to get in
which weren't a part of any national standard. They were therefore "added"
to GB12345 so that they'd have GB code points.

John H. Jenkins

This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:32 EDT