Re: charset independent strlist.c

[Masahiro Sakai]

From: Andrew Clausen <address@hidden>
Subject: Re: charset independent strlist.c
Date: Wed, 18 Oct 2000 06:13:42 +1100

> A problem I thought we might have is mbsrtowcs() might not know
> what charset the translations will be in.  eg: Japanese (as you
> know ;-) has EUC and utf8, IIRC.  Someone (takashi?) 

Yes. mbsrtowcs() doesn't know.
But expecting native charset of the locale is reasonable, I think.

> Is the Right Way to deal with this to set the locale.  i.e. to
> jp_EUC, or whatever?

At the present time, the right way is to prepare parted.mo file
that encoded in the locale's charset for each locale, IMHO.

In the future, gettext will do charset conversion internally,
so the result of gettext() is in the locale's charset.
(e.g. glibc-2.2 and
 ftp://ftp.ilog.fr/pub/Users/haible/utf8/gettext-0.10.35-iconv.tar.gz
 will do such conversion.)

As far as I know, it's not necessary to prepare another parted.mo
file for Japanese. Because ja_JP.UTF-8 (Japanese locale using UTF-8)
is not popular now, and most users of it are using development
version of glibc-2.2. 

--
Masahiro Sakai