[ruby-core:19304] 1.9, encoding & win32 wide char support

hello,

I've recently hacked 1.8.7 on win32 to use Wide char CRT & system  
calls.  The general strategy is to assume that all strings are  
represented in utf8 (as on unix platforms) and to convert to/from  
"wide" using MultiByteToWideChar at the lowest possible level.

In general, this has meant that the win32 wrappers that already exist  
in win32.c are used everywhere, and we're using the win32/crt calls  
that accept wchar_t.

After completing this work I took a peek at trunk, and noticed that it  
still looks like ascii versions of these calls are in use.

I was wondering if there's any interest in this patch, and further  
what the strategy is on win32 to allow correct handling of paths with  
non-english characters?

This feature is especially important for our product because we're  
embedding ruby and installing the standard library in a user-scoped  
path.  Given a non-english login name on vista or xp, most paths will  
have non-english characters.  If there is general interest, I'd love  
to contribute some time.

best,
lloyd