[Bf-committers] International Input

[Rob M]

> > 1. use platform specific unicode version of file operation API in the
> > implementation of BLI_fileops and encode the (probably) wchar
> > filename into utf8.
> 
> For Windows file operations we need to change “winstuff” to use
> FindFirstFileW
> and FindNextFileW instead of the ansi versions. The results would have
> to be
> converted from Wide to UTF-8.
> 
> Output already supports UTF-8 so this _should_ be most of the work
> done,
> except for some miscellaneous path-processing routines in path_util and
> winstuff
> that will try to mangle everything.

A lot of the file name routines assume an ascii string when doing things like splitting the file name from the directory path and also making paths relative. It should not be too difficult to make these multibyte character aware on windows without having to use utf-8 by using the windows string apis rather than the c ones but I don't want to do this work until after my patch to get UNC names working is accepted.


Rob.