2013/7/25 Larry Hastings <la...@hastings.org>: > It's about nine days from now. I expect to tag the release late next week. > So if you're doing any major brain surgery, please finish it up in the next > week or so.
I hope that I would have enough free time before the alpĥa2 to: * find a consensus on the file descriptor inheritance (!), finish "the" PEP and the implementation * write a new module (!) using the PEP 445 (malloc hooks) to inject memory allocation failures and track memory allocations (It is maybe already too late !) I also hope that the assertions that I added in the issue #18408 would not break too much applications (when using Python 3.4 compiled in debug mode) :-) I added assertions to ensure that a Python function is not called when an exception is set. The problem is that a Python function can clear or replace the current exception. For example, hasattr(obj, name) replaces the current exception, and later clears the new exceptions, if the attribute does not exist. It doesn't care if it was called with an exception already set (PyErr_Occurred() is not NULL). I fixed all issues related to these assertions in the Python stdlib, but I don't know if it breaks third party code. In my opinion, if these assertions fail, there is a bug that must be fixed. Tell me if it's better to discuss these points on python-dev ;-) Victor _______________________________________________ python-committers mailing list python-committers@python.org http://mail.python.org/mailman/listinfo/python-committers
