Re: (comp) fbiterm and Japanese (and French...)

Back to that pesky console...

Environment: I have an old iBook on which I just put Debian from an old network install with Sarge as base.

I've generated 3 locales: en_US-utf8/fr_FR-utf-8 and ja_JA-utf-8 and choose the en_US-utf-8 by default.

Now, I can use emacs 22 from the standard console and use the input method French to french-prefix and I can type French.
When I set the method to japanese, I get full mojibake in the form of endless "?????" anywhere I type Japanese.

So, I set terminal-encoding to utf-8 and now my French is all mojibaked and the Japanese is differently mojibaked: it comes is different characters now...

When I load fbiterm with the following fonts (it can't be launched without setting the fonts):
fbiterm -a /usr/lib/X11/fonts/misc/8x16.pcf.gz -m /usr/lib/X11/fonts/uni/unifont.pcf.gz

I can't even use French input in emacs anymore, the accented characters do not appear... And the Japanese is not working either. And if I set terminal-encoding to utf-8 I get no progress.

When I set "prefered-encoding" to UTF-8 all my previously fine French input is suddenly not working anymore...

So, whatever I do it seems emacs 22 wants to use latin-1 for French and shift-jis for Japanese (that it can't even display even though there are supposedly plenty of fonts for that...)

It seems I am missing something trivial.

Jean-Christophe Helary


On 2 oct. 06, at 17:39, David Kuehling wrote:

>>>>>> "Jean-Christophe" == Jean-Christophe Helary <[email protected]> writes:
>> Now I need to figure out a number of things that are not exactly
>> clear:
> 
>> 1) My keyboard is roughly a jp106. How can configure the system to be
>> able to input French with as little hassle as possible ?
> 
>> 2) Second input question: I need Japanese input. What is the input
>> method that works the best with fbiterm ? And will all applications
>> running in fbiterm accept the input. Are there still applications that
>> do not support UTF-8 input/display ?
> 
> You might want to consider using emacs on console to do your work.
> Install emacs21 and mule-ucs.  On Debian Sarge you will need to define
> some environment variable to actually activate mule-ucs (see
> /usr/share/doc/mule-ucs/README.Debian or Changelog)
> 
> In emacs, enter M-x set-terminal-coding-system <Ret> utf-8 <Ret> to
> enable emacs to output japanese characters on your console.
> 
> To input japanese, enter M-x set-input-method <Ret> japanese <Ret>
> 
> To input french, you might try 
> M-x set-input-method <Ret> french-prefix <Ret>
> 
> If you need a console with japanese input, just run a console from
> within emacs:  M-x shell <Ret>  
> or probably better:  M-x eshell <Ret>
> 
> To enter M-x (Meta-x) from the console, you might have to press the
> sequence <Esc> x instead.
> 
> If you need a more evolved input method than the emacs-internal
> "japanese", try to install anthy and anthy-el.  Then configure anthy for
> japanese input.  This should work much better.
> 
> To view the documentation of emacs' internal japanese input method,
> enter
> 
> M-x describe-input-method <Ret> japanese <Ret>
> 
> regards,
> 
> David
> -- 
> GnuPG public key: http://user.cs.tu-berlin.de/~dvdkhlng/dk.gpg
> Fingerprint: B17A DC95 D293 657B 4205  D016 7DEF 5323 C174 7D40
> 

Jean-Christophe Helary
---------------------------------
fun: mac4translators.blogspot.com
work: www.doublet.jp (ja/en > fr)
tweets: @brandelune