Kenneth Pronovici wrote:
I think that I can solve my problem by just converting any unicode
strings from configuration into utf-8 simple strings using encode().
Using this solution, all of my existing regression tests still pass, and
my code seems to make it past the unusual directory.
See my other post. UTF-8 might be the wrong choice: the path might be
encoded as, say, ISO-8859-1 on disk, in which case an UTF-8-encoded
path would not be found.
Regards,
Martin
--
http://mail.python.org/mailman/listinfo/python-list
