It is plainly obvious that testtools code is broken and adjusting our good code to match is not the best course of action. Every code example in their own documentation uses the (actual, expected) order. EVERY SINGLE ONE. And everyone else writes tests in that order, too.
It is plainly obvious that testtools code is broken and adjusting our good code to match is not the best course of action. Every code example in their own documentation uses the (actual, expected) order. EVERY SINGLE ONE. And everyone else writes tests in that order, too.
See http:// docs.python. org/3/library/ unittest. html
Has anyone tried to raise this problem with testtools upstream?