Le Lundi 20 Décembre 2004 02:33, [EMAIL PROTECTED] a écrit : > I think the more proper solution is each translation file > contains its encoding. But it may be simpler for us to use UTF-8 as we don't have to ask contributors to give us there local encoding. Moreover, I didn't succeed in displaying polish characters in konsole (polish encoding is cp-1250). Using UTF8, I only have to switch font to "unicode" and it works. note : I just discovered that "echo -ne '\033%@'" disable UTF-8.
===== current system =====
* example
nothing to do
* problems
- difficult for maintainers
===== 1st solution : switch the terminal to UTF-8 (manual) =====
* example
function doInUtf8 () { echo -ne '\033%G' ; $@ ; echo -ne '\033%@' ; }
doInUtf8 diff pl pl.old
* problems
- unicode doesn't work in aterm and rxvt
- if the terminal was already in UTF-8, this command disable it. I've tried
the command "vt-is-UTF8" but it always return "Single-byte char mode."
===== 2nd solution : switch the terminal to UTF-8 (setcode) =====
* example
function echoWithEnc () { LC_CTYPE=UTF-8 setcode; $@ ; setcode ; }
* problems
- unicode doesn't work in aterm and rxvt
- licence of setcode is still unknown
- display annoying characters ("bC") when calling "setcode".
[EMAIL PROTECTED] lang]$ setcode
[EMAIL PROTECTED] lang]$
===== 3rd solution : convert strings to local encoding =====
* example
function doFromUtf8 () { $@ | iconv -fUTF-8 ; }
* problems
- is iconv available in all distribs?
- "--to-code" parameter is not marqued as optionnal in iconv man page (but it
works)
- maybe it should never happened, but if a character is not available in
target encoding, iconv quits returning "illegal input sequence at
position..."
I'd like to use the first solution but it requires us to find how to detect
whether the terminal is already in UTF-8.
_____
Tux
pgp4feQe6lrXP.pgp
Description: PGP signature
