Re: Patch for Review: Replace "Unicode to UTF8 conversions" with"Unicode to 'Win32 locale' conversions" when sending/receiving file namesto/from Win32 API.

[Ranjit Mathew <rmathew at hotmail dot com>]

> João> this is for those of us who would like to use char codes above 127
> João> with Win32 file names.
> 
> João> There are four unsupported code pages: Chinese (traditional and
> João> simplified), Japanese and Korean.
> 
> What if, instead of having special code here for Windows, we have new
> charset converters for these things?  Then we could just use generic
> code to convert the representations.
> 
> Does Windows not have these tables built in somewhere?  Instead of
> having these tables built in to libgcj it seems like we could use the
> Windows equivalent of the iconv() charset converter.  (If Windows
> doesn't have this sort of thing, we can add the new converters to
> libgcj easily enough.)
> 
> Anyway, I think this idea is definitely needed.  We should probably
> do the same thing for file names in the posix I/O code.

I'm sorry I didn't say anything on this issue earlier, but
that's simply because I don't know much about
internationalisation, especially variations in how
different systems handle it.

However, I would still like to point out that:

1. libiconv, iconv and gettext are avaiable for MinGW at:

   http://sourceforge.net/project/showfiles.php?group_id=2435

2. I (and very likely Mohan as well) used to always build
   GCC/GCJ with --disable-nls, so you shouldn't expect much
   NLS support from the pre-built binaries anyway.

Ranjit.

-- 
Ranjit Mathew          Email: rmathew AT hotmail DOT com

Bangalore, INDIA.      Web: http://ranjitmathew.tripod.com/