How to create pgn from a game play

[literadix]

Hello!

I am a new user of this API. Thank you very much. It is really a great work. I would like to ask a question which I could not solve: how can I create a PGN output from a running game: I have written following simple program as tournment of two compuer chess engines:

##########################################
# a very simple tournment between two uci engines
##########################################
import chess
import chess.uci

board = chess.Board()
engine = chess.uci.popen_engine("stockfish")
engine.uci()

while True:

    board.reset()

    while not board.is_game_over():
        engine.position(board)
        bestmove, ponder = engine.go(movetime=1000)
        # print bestmove
        board.push(bestmove)

    print board
    print board.fen()
    print board.result()

Now I would like not only to have a fen represenation when a game has finished, but also a pgn export. So my question is: how can I record all moves and checks and mates etc during a game play, so that I could have a pgn represantation of all moves when the game has ended?

A simple example would be welcome. You are also welcome to use my example and your pgn extension as general example for other users.

Thank you very much,

Maciej

[niklasf]

Hi.

Example 1: Add game nodes during play.

##########################################
# a very simple tournment between two uci engines
##########################################
import chess
import chess.uci
import chess.pgn

board = chess.Board()
engine = chess.uci.popen_engine("stockfish")
engine.uci()

while True:

    board.reset()

    # Create a game and set headers.
    game = chess.pgn.Game()
    game.headers["White"] = engine.name
    game.headers["Black"] = engine.name
    game.setup(board)  # Not required for the standard
                       # starting position.

    node = game

    while not board.is_game_over():
        engine.position(board)
        bestmove, ponder = engine.go(movetime=100)
        print bestmove
        board.push(bestmove)

        node = node.add_variation(bestmove) # Add game node

    print board
    print board.fen()
    print board.result()

    game.headers["Result"] = board.result()
    print game

Example 2: Create a game from the recorded moves.

def board_to_game(board):
    game = chess.pgn.Game()

    # Undo all moves.
    switchyard = collections.deque()
    while board.move_stack:
        switchyard.append(board.pop())

    game.setup(board)
    node = game

    # Replay all moves.
    while switchyard:
        move = switchyard.pop()
        node = node.add_variation(move)
        board.push(move)

    game.headers["Result"] = board.result()
    return game

Especially the second one should be easier. I'll see how I can improve this and update the documentation.

[literadix]

Niklas,

thank you very much. This is exactly what I have been looking for.