RE: [Doxygen-develop] Problem French accent under linux

[Petr P. <Pr...@sk...>]

Hi Xavier,

Xavier Outhier wrote...
> [...]
> I got a user feedback about a problem for vizualising
> the HTML doc in French under Linux (but it's correct
> under Windows).
>=20
> She made the following settings:
>   USE_WINDOWS_ENCODING set to NO
>   OUTPUT_LANGUAGE set to French
>=20
> I feel the problem arise because I'm under Windows, using
> French ASCII in the translator_fr.h.

My guess is that French encoding for Unix is different
than French encoding for Windows. This means that
your translator should return the strings differently.
based on the USE_WINDOWS_ENCODING. It is=20
only a configuration parameter that _can_ be tested
in your translator_fr.h.

If the encodings for Windows/Unix are not identical,=20
the idLanguageCharset() should test the=20
USE_WINDOWS_ENCODING and return the
appropriate string. In such case, you should
also define something like decode() that
also tests the USE_WINDOWS_ENCODING=20
and converts encoding of strings returned=20
by the translator methods. Have a look=20
at translator_cz.h to see how the things=20
can be done.

> I have two questions:
>   -1 Will it solve the problem if I use the HTML entity
>      in the header file?

I am not sure what you mean by that, but probably no.
If you run doxygen in Linux, then the documented
sources probably use the encoding for unix.=20
The translator adds strings that should be also converted
to unix encoding -- otherwise text with both encodings
would be mixed (the unix text from sources and the windows
texts from doxygen.

>   -2 Will it not disturb the other output: Latex, RTF or
>      whatever.
>      If yes how to solve the problem.

The same problem -- possible mixing of two encodings.
You have to enhance your translator_fr.h ;)  Feel free
to ask for details.

Regards,=20
  Petr


--=20
Petr Prikryl (prikrylp at skil dot cz)=20