Rockbox mail archiveSubject: Re: Lang compilation problem (2)
Re: Lang compilation problem (2)
From: Daniel Stenberg <daniel_at_haxx.se>
Date: Wed, 18 Dec 2002 08:31:34 +0100 (MET)
On Tue, 17 Dec 2002, zOOm wrote:
> > > indeed "é" "à"..... are bad interpreted.....
> > What does "bad interpreted" mean in this context?
> bad interpreted means that in the menu all "é" are transform by something
> like "_at_&" (not exactly this char but it looks like...)
Can this be that the font you're using is not having the correct letters on
the correct position? What font are you using? (This is a Recorder model,
> but if I have songs with such letters there are well interpreted...
... which of course indicates that the font is fine.
> this means that I ve got a compilation problem of the language.... but I
> cannot find where the mistake occurs
> > > the probleme just comes from the default .lng generated because if I use
> > > the default french.lng it is good...
> > What is the problem? When does the problem happen? What are you saying you
> > when the problem does not occur?
> As a proof the problem doesn't occurs when I load the french.fnt on the web
> site.... (of course I haven't compile it...)
french.fnt ? Did you mean the french.lng ?
> So how compiling a .fnt ?
There is no .fnt files, AFAIK.
The .lang files are compiled into .lng files using 'tools/binlang'.
> I can see the .lang is used to generate a lang.c and .h which are used in
> order to generate the .o
Yes, but we usually only use the english lang to create the source code, and
then we load binary .lng files in run-time to fix the localized languages.
> Can this be at the compilation of the .o ? By editing the .c and .h all
> things seem ok
Does the problem look the same when building a simulator?
If so, can you show us a screen dump?
And also, then you can set a break-point in a suitable place to see if the
problem is in the string itself or in how the string is visualized on screen.
-- Daniel "Bagder" Stenberg -- http://rockbox.haxx.se/Received on 2002-12-18