Re: [gnome-print] latin2 (i18n?) problems

[Owen Taylor <otaylor redhat com>]

Tomas Pluskal <plusik@pohoda.cz> writes:

> Thanks for advice, the testprint worked ok, so I started investigation in
> gedit & gnumeric sources, and I found out that the problem is probably in
> FreeBSD implementation of multibyte(3) functions. I have written a sample
> program:
> ------ test.c -----------------------------
> #include <stdlib.h>
> #include <locale.h>
> 
> int main(int argc, char *argv[])
> {
>  char s[1024];
>  int x;
> 
>  setlocale(LC_ALL, "cs_CZ.ISO8859-2");
>  x=wctomb(s, 0x0161);
> 
>  printf("char count: %d\nchar: %c\n",x,s[0]);
> 
> return 0;
> }
> -------------------------------------------
> It should output a "s" with hook (unicode 0x161) character, but it outputs
> just "a". In linux it outputs hooked s correctly!

There is no requirement that wide characters are the same as unicode...
though a lot of the GNOME-1.2 code around has the assumption that is
the case.

(If __STDC_ISO_10646__ is defined, it's a valid assumption.)

This sort of code can be ditched with GNOME-2.0 where GLib-2.0 has
nice and portable encoding conversion functions.

Regards,
                                        Owen