Nick Coghlan added the comment: I think the convert to str -> process as str -> convert back to bytes approach is a good one - it's the same one we use in urllib.parse.
In this case, since we explicit need to handle mixed encodings, I also agree with the idea of using surrogate escape to make it possible to tunnel arbitrary bytes through the process, and expose that as a new module level API for Python 3.5. ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue17445> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
