Post #1904264

@BarneyDellar @qcoding Given a function like: F() { a(); b(); if (c()) { d(); } } It makes sense to have test(s) for when `c` is true and other test(s) for when it's false. But don't have some test(s) for the `a` effects and some test(s) for the `b` effects. If the `a` and `b` effects need to be verified, assert them both in the same test. And yes: if this makes your tests undesirable in some way, refactor.