My WIP on Unicode

[Nicolas Sebrecht]

On Tue, Feb 10, 2015 at 02:59:13PM +0100, Sebastian Spaeth wrote:

> Yes, having a clean nice Unicode-capable code would be a nice goal. Our
> imap library passes everything pretty raw which leads to surprises when
> you have folder names with accents, Umlaute and other "weird"
> characters. Also, it would lead to a path towards a python3 world :-).

I would not call them "weird". They are expected IMAP-modified UTF-7
encoded. Impalib2 is right to not play with encodings.

The bad with imaplib2 is when it plays with some arguments, mostly to
enlive the debug messages. Calling str() or things like that, should
almost be forbidden in libraries. But while I've found some
inconsitencies in imaplib2, it is globally good. A standard library like
"threading" has more undocumented assumptions.

-- 
Nicolas Sebrecht