… fork d'oh, looks like 1.3.1 is around the corner
seriously? c'mon Matz! "Principle of Least Surprise" and all that!
seems like a separate method could_acquire? is a better choice
seems a bit silly, but we might customize in the future, and we want to document the differences in conditions assert! is checking for in exclusive and shared [ci skip]
requested by @eric for checking to make sure in some long-running thing that there's a way to check that the assumption that we still have the lock is true
document this, don't enforce it with warnings quite yet