unicode filenames

[Andrew Dalke]

Neil Hodgson wrote:
>    The encoding registry uses byte strings.

True.  I meant it mostly as a sketch of a solution.

>    How does os.path.abspath deal with a Unicode string?

Err, ummm, I didn't yet include that wrapper function?  Don't
believe me?  How about that I forgot?  :)

>  > If this makes sense, should it be added to Python's core?
> 
>    There are quite a few calls that need to change - from the file
> constructor to stat ...
> 
>    To be robust it needs to deal with multiple encodings in a path.

Yep, and yep.

I think I have just shown that I'm not the perfect candidate to
do so ;)

BTW, how do I test your assertion the RedHat uses UTF-8 for filename
encoding?  I can't figure that out.  I did figure out one problem
is that I need to say "en_US.UTF-8" instead of "en_US.utf-8".

					Andrew
					dalke at dalkescientific.com