Oh?
What possible combinations are missing?

That's called making a simple problem complex, just for the fun of it.

Is it bad that, years later I look back at this and just laugh at how bored I must have been to write this?

Yes, yes it should be. Corrected

Moved

I'm not 100% sure how I could make the test cases clearer.

Are you asking for some sort of display that would show expected board position if it errored out? that may be possible, but not sure how offhand.
If you mean something else, let me know.

And believe me, I know this is a pain to debug. I spent probably over 20 hrs writing the program myself

Now found a moveset that reaches a forced pass within the first 20 moves. May help test pass rules.

I see some of the confusion. I added that each player takes turns placing pieces, UNLESS they have no valid moves.

Interesting that the 5 originals still fail if the shorter ones pass.

That leads me to think that there is an issue with checking that one player does not have valid moves. I'll have to do some research to see if there are examples of games with a early situation where a player runs out of moves while the other still can move. That's usually a situation that only occurs after (in my experience and observation) around move 40, which means that it will be difficult to debug.

I see the confusion. A player only continues to play if he has valid moves IF the other player has no valid moves.

Cleaned up the 4&5 issue, Now only states returns None if error.

Moved some of the rule descriptions and added details I found I'd missed.

Also added 108 short (8-13) move exercises to the sample and main tests, plus more short random tests.

Black should be playing D3 (flipping E4). Then White G5

Approved

Done

Voile, in this case it is actually a complete set, in that it tests every possibility for up to the 8 variables.

Would random tests be needed when anything in random is already tested for in the static?