RE: [Openexr-devel] UNICODE support in openexr file I/O

[Luc-Eric Rousseau]

In addition to UTF, it would be best if you provided wchar_t support for 
Windows like libtiff (TiffOpenW) does.  Most users that require unicode, like 
the users of our Japanese versions, run Windows.  (The unix users would 
typically be avoiding unicode names because of lack of support elsewhere.)

> The type wchar_t is compiler implementation dependant - it varies 
> in size from compiler to compiler and should NEVER be used as a type in files.

pretty much eveyrthing in c++ a library is compiler-dependant, because name 
mangling, the object (ex: v-table implementation, exceptions) and data (ex: 
enums size) are compiler-dependant.

> -----Original Message-----
> From: Drew Hess
> Sent: Thursday, January 19, 2006 3:22 PM
> To: address@hidden
> Cc: address@hidden
> Subject: Re: [Openexr-devel] UNICODE support in openexr file I/O
> 
> 
> 
> "Santiago " <address@hidden> writes:
> 
> > All the OpenEXR classes ( like TiledRgbaOutputFile ) only 
> accepts old ASCII
> > "const char*" parameter to read/write files.... That's bad 
> because if I wanna
> > name a file using extended characters like Ñç or even Kanji, I can't
> > .... Please... can you provide too a wchar_t version so we 
> can use plenty the
> > UNICODE characters? thx
> 
> 
> Hi Santiago,
> 
> What if we added support for just the UTF-8 encodings, would that
> suffice?  (UTF-8 is a lot easier to deal with, in my opinion.)
> 
> thanks
> d