[tlug] Re: UTF-8 under Linux

[Mike Fabian <mfabian@example.com>]

"Stephen J. Turnbull" <stephen@example.com> writes:

>>>>>> "Mike" == Mike Fabian <mfabian@example.com> writes:
>
>     Mike> This may look ugly if you edit Chinese in UTF-8 because some
>     Mike> Chinese characters are taken from a Japanese font and some
>     Mike> from a Chinese font, for example:
>
> It's possible to change the precedence of charsets in Mule-UCS.  GNU
> may do this in language environment processing, but we don't yet.

It seems that GNU Emacs doesn't change the precedence of charsets for
Mule-UCS in the language environment either.
'set-language-environment' had no effect on the fonts used for mixed
Japanese-Chinese text in UTF-8.

For mixed Japanese-Chinese text in UTF-8 in (X)Emacs, using a uniform
Unicode fontset is probably a good solution. Small differences between
Chinese and Japanese glyph styles are not that important in a text
editor and at least it looks uniform and readable with a Unicode
fontset. With different fonts used for Japanese and Chinese,
some glyphs will always be taken from the wrong font, no matter
how the preferences are.

To make this work perfectly with different fonts for Japanese and
Chinese in UTF-8, one would need some sort of tagging for the
different languages, for example the Unicode language tags.  But that
is probably overkill in a text editor and would be an extra burden for
the user because he would have to enter additional tags when switching
between Chinese and Japanese.

-- 
Mike Fabian   <mfabian@example.com>   http://www.suse.de/~mfabian
睡眠不足はいい仕事の敵だ。