Filename encodings

["Dave Williss" <dwilliss microimages com>]

I have a question about how filenames are stored in Linux.
All the standard C library calls take (char*) and not Unicode,
so I assume that if you have a filename in another language,
it has to be encoded in some way.  I would guess UTF8?
I know in the past, people tried to implement this using 
a locale mentality, but that falls apart if I'm logged in using
one locale and try to view a filename created in another.

I ask here because obviously the GUI has to know so that it
can present the name to the user in some meaningful way.

Mac OS 10.1 uses UTF8, but for some reason, decided that
the strings must be decomposed first.  So, for example, a
u with an umlaut would be encoded as two characters: a "u"
and a combining umlaut.

 -- Dave Williss
------
Meddle not in the affairs of dragons, 
   for you are crunchy and taste good with catsup