Re: Variations of mapping from Japanese encodings to Unicode

The DTD was changed in the last version (December last year). The changes
are listed in http://www.unicode.org/unicode/reports/tr22/#Modifications

The major change was the removal of <import>, which the UTC felt made the
validity checking too cumbersome. In the previous version, you could use
imports to override mappings in another version.

Unfortunately, I see that you use <import> extensively, so that you would
have to regenerate the files in their complete form.

[There are three more changes that were approved this year for 2022 and some
other minor cases, but they are backwards compatible.]

Mark
—————

Γνῶθι σαυτόν — Θαλῆς
[http://www.macchiato.com]
----- Original Message -----
From: "Martin Duerst" <duerst@w3.org>
To: "MURATA Makoto" <muraw3c@attglobal.net>; <ietf-charsets@iana.org>
Cc: <mark.davis@us.ibm.com>; "Mark Davis" <mark@macchiato.com>
Sent: Wednesday, August 29, 2001 19:19
Subject: Re: Variations of mapping from Japanese encodings to Unicode


> Hello Makoto,
>
> I have tried to look at the .xml files you cited.
> But I got validation errors with various tools.
> E.g. in http://www.w3.org/TR/japanese-xml/x-sjis-cp932.xml,
> XMLSpy complains that there is no ID on characterMapping
> (required according to
> http://www.unicode.org/unicode/reports/tr22/CharacterMapping.dtd).
> My guess is that the problem is due to some updates in
> the CharacterMapping DTD. I'm coping Mark to check.
>
> I suggest that you investigate the problems. If necessary,
> you can always resubmit the XML Japanese Profile Note
> to fix errors.
>
> Regards,   Martin.
>
>
>
> At 00:33 01/08/30 +0900, MURATA Makoto wrote:
> >Unfortunately, different systems use different mapping tables from
> >Shift-JIS to Unicode.  The same thing applies to ISO-2022-JP and
> >Japanese EUC.  Such mapping tables are very unlikely to disappear
> >soon.  More about this, see XML Japanese Profile at:
> >
> >        http://www.w3.org/TR/japanese-xml
> >
> >Such unstable mapping spoils interoperatibility and is fatal
> >especially for digital signature.  Although we do not have any really
> >good solutions, we can assign different charset names to different
> >mapping tables.
> >
> >For this reason, I am going to propose quite a few charsets for
> >Shift-JIS, Japanese EUC, and ISO-2022-JP.  They are new charsets
> >rather than aliases of existing charsets (i.e., shift_jis, euc-jp,
> >iso-2022-jp), which does not have agreed mapping tables.
> >
> >Cheers,
> >
> >MURATA Makoto
>
>

Received on Thursday, 30 August 2001 10:44:29 UTC