[Quantum Chinese Chess] Update flying general check #168
Conversation
madcpf
commented
Oct 30, 2023
•
madcpf
commented
- Update flying general check to include the cases where there are quantum pieces between two kings(aka generals).
- Update Move to not depend on Board to break the cyclic dependency.
madcpf
changed the title
| # Let the general/king fly, i.e. the opposite king will capture the current king. | ||
| current_king = self.board[self.king_locations[self.current_player]] | ||
| oppsite_king = self.board[self.king_locations[1 - self.current_player]] | ||
| Jump(MoveVariant.CLASSICAL)(oppsite_king, current_king) |
There was a problem hiding this comment.
The game ends here, right? So, is it necessary to actually move the king? I guess there's no reason not to, but it seems unnecessary.
There was a problem hiding this comment.
Hi Doug, yes the game ends here. I just chose to capture the other king so that later on we print the updated board as the last print of the game.
| print("==== FLYING GENERAL ! ====") | ||
| return True | ||
| else: | ||
| # TODO(): we are leaving the path pieces unchanged in entangled state. Maybe |
There was a problem hiding this comment.
I think this is fine, right? since we measured the ancilla, it's guaranteed that all states remaining on the board have some piece in between.
There was a problem hiding this comment.
Thanks Doug. Yes I've updated the comment here.
| result = board.flying_general_check() | ||
| # We check the ancilla to learn whether the general/king flies or not. | ||
| captured = world.post_selection[world["ancilla_ancilla_flying_general_check_0_0"]] | ||
| if captured: |
There was a problem hiding this comment.
It would be nice to know that each branch of the if statement happens. (That some of the time it is captured and some of the time it is not).
One way to get one branch deterministically would be to test this:
board = set_board(["a3", "e0", "e9"])
board.king_locations = ["e0", "e9"]
board.current_player = 0 # i.e. RED
alpha.PhasedSplit()(world["a3"], world["e3"], world["e5"])
There was a problem hiding this comment.
Thanks Doug. I just added this case as a separate test, where capture could not happen. But I cannot think of a way to deterministically make sure the capture happens.
