Chess programming make and unmake move

Hi! I'm making a chess engine and I have some problems with the make/ unmake methods.

I have a Piece class that holds the type (pawn, queen, etc..) and position of the piece and a Move class that holds the target square, the moved piece and captured piece.

The problem is that, when I call the makeMove method, it changes the piece's position to target square inside the Piece object. But now, I can't call the unmakeMove with the Move object, because now i have no information about where the move came from, as I have just changed the piece's position. How would you tackle this problem?

Thank you very much!

class Board:
# Previous methods omitted. 

    def makeMove(self, move, player):
        """ Makes a move on the board and changes Piece object. Returns None. """
        self.moved_piece = move.getPiece()
        self.captured_piece = move.getCapturedPiece(self)

        if self.captured_piece:  # Remove captured piece from player's piece dict. 
            player.removePiece(self.captured_piece)

        self.setPiece(move.getTargetSquare(), self.moved_piece)  # Set moved piece on target square.
        self.setPiece(self.moved_piece.getPosition(), EMPTY)  # Make the origin square empty.
        self.moved_piece.changePosition(move.getTargetSquare())  # Change piece object's position.

    def unmakeMove(self, move, player):
        """ Unmakes a move. Returns None. """
        self.moved_piece = move.getPiece()
        self.captured_piece = move.getCapturedPiece(self)

        self.setPiece(self.moved_piece.getPosition(), captured_piece)  # Set captured piece or empty square to target square.
        # Set piece to original square. HOW !?

---

Based on my comment and the link on the Memento pattern that Fred Larson posted, here is an example implementation of what you might want to do:

class Engine(object):
    def __init__(self):
        self.board = Board()
        self.move_list = []

    def make_move(self, from, to):
        #board.makeMove returns a "memento object".
        self.move_list.append(board.makeMove(from, to))
        ...

    def undo_move(self):
        board.undo(self.move_list.pop())
    ...
    ...

And let's say you have a move object of this structure:

class Move(object):
    def __init__(self, from_coords, to_coords, capture=None):
        self.from = from_coords
        self.to = to_coords
        self.capture = capture

You're Board object would implement the following methods:

class Board(object):
    ...
    def make_move(self, from_coords, to_coords):
        #move logic here
        return Move(from_coords, to_coords, capturedPiece)

    def undo_move(self, move_object):
        self.make_move(move_object.to_coords, move_object.from_coords)
        self.uncapture(move_object.capture, move_object.to_coords)

Obviously, the code above is just conceptual. The actual implementation would depend on how the rest of your code is structured.

Note: I used a class for the Move object because attribute access is explicit and easy to follow. In reality, a object as basic as this could simply be a tuple of the form (from, to, capturedpiece) .

链接地址: http://www.djcxy.com/p/14812.html

上一篇: 棋子的未来流动性

下一篇: 国际象棋编程制作和取消制作