Thomas Karsten <kath0005@hs-niederrhein.de> さんは書きました:
Why isn't there an encoding like 'ja_JP.iso-2022-jp' available? As far as I know iso-2022-jp is very popular in Japan, so why isn't there such an encoding?
A locale using ISO-2022-JP encoding like ja_JP.ISO-2022-JP would make no sense because '/' characters may appear in regular Japanese text. For example, katakana ku (く) in ISO-2022-JP is encoded as the two bytes "$/": mfabian@magellan:~$ echo く | iconv -f utf-8 -t iso-2022-jp | hex 0000 1b 24 42 24 2f 1b 28 42 0a .$B$/.(B . mfabian@magellan:~$ As '/' is used as the directory separator, it is not allowed in a filename. That means, you could not create a file containing 'く' if you were running in a ja_JP.ISO-2022-JP locale. There are more such problems. ISO-2022-JP is not suitable to be used as the encoding for a Unix locale. This is no real problem as ISO-2022-JP is mainly popular in e-mail. And of course you can create and edit files in ISO-2022-JP encoding if you wish, you don't need such a locale to do that.
My last experience using the eucJP, mlterm and SuSE 9.1 was that the texts are *not* encoded in ISO-2022-JP, but I will check this again.
If you run mlterm in ja_JP.eucJP locale and create a text with let's say Vim with the default settings, the file will be EUC-JP encoded. But if you have a ISO-2022-JP encoded files, you can just use "cat" in an mlterm using EUC-JP encoding and they will be displayed correctly.
contrary to kinput2, xcin *does* care for the locale it is started in. You cannot start xcin in zh_CN.UTF-8. But you can start the clients using xcin in zh_CN.UTF-8 locale.
I used to use this solution before. However, some characters are available in UTF-8 only, so I cannot use an xcin, configured with LANG=gb2312, to write these characters, even if the mlterm is using UTF-8.
Trying to run xcin in UTF-8 won't help because xcin just doesn't support these characters (at least not yet).
By the way, why don't you try SCIM? I believe it is much better especially for simplified Chinese.
Ok, I will try SCIM. Maybe the UTF-8 problems will not appear there.
SCIM is certainly better for you if you need Chinese characters which are available in UTF-8 but not in GB2312. -- Mike FABIAN <mfabian@suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。