R. David Murray added the comment: Thanks, but the patch is incorrect. The model consistently stores its data as surrogateescaped strings, and this assumption is baked in to other parts of the code. So the correct fix is to do the surrogateescape encoding at the time the payload is set.
It might in fact be better to store a binary payload as binary, but making that kind of change to the model requires much more extensive review and testing. ---------- _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue18324> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com