j-chess-lib 0.10.1

Python library for a j-chess bot. Beep Boop

j-chess-lib

Python library for a j-chess bot. This library provides an interface to connect to a j-chess-server and play some games with a self written bot. To see the currently supported j-chess-xsd please see Documentation

• Free software: GNU General Public License v3

• Documentation: https://j-chess-lib.readthedocs.io.

Server and Communication xsd

• Server:

  • 

  • See: https://github.com/JoKrus/j-chess-server/

  • Documentation: https://github.com/JoKrus/j-chess-server/wiki

• Communication protocol:

  • 

  • See: https://github.com/JoKrus/j-chess-xsd/

  • Reference: https://github.com/JoKrus/j-chess-xsd

Features

• Easy connection to server

• Interface to write a bot to play chess

  • Provides different interfaces with different complexity levels

• Pre-implemented bot to play random moves

• Basic generation for legal moves

Install

PyPi package is comeing soon. Until then installation is done via

$ pip install j-chess-lib

See Installation for detailed instructions

Usage

See here some examples to use this library. For more detailed explanation see Usage

Start

See below for an example to start the client with your ai

from j_chess_lib.communication import Connection
from j_chess_lib.client import Client
from j_chess_lib.ai.Sample import SampleAI

with Connection(server_address, server_port) as connection:
    ai = SampleAI()
    tournament_code = # Your optional UUID tournament code
    client = Client(connection=connection, ai=your_ai, tournament_code=tournament_code)
    client.start()
    client.join()

This example shows how to setup the connection and the client. The tournament code is optional and does not have to be passed. If None is passed none will be send to the server. The client is its own thread so you could for example start multiple clients parallel or do some other stuff. Like a gui…

AI

See below for an example to implement a very easy AI. It uses the base-class that automatically stores match and game data when started so you can query it in the move generation when needed

from uuid import UUID
from j_chess_lib.ai import StoreAI
from j_chess_lib.ai.Container import GameState
from j_chess_lib.communication import MoveData

class SampleAI(StoreAI):
    def __init__(self):
        super(SampleAI, self).__init__(name=f"Unique Name of your very good AI")

    def get_move(self, game_id: UUID, match_id: UUID, game_state: GameState) -> MoveData:
        enemy, match_data = self.get_match(match_id=match_id)
        white_player = self.get_game(game_id=game_id, match_id=match_id)

        # Your super intelligent code to generate the best chess move ever generated

        move_data = # result of your code
        return move_data

This example initializes a SampleAI

ToDo

• Error fallbacks

• Some features

Credits

• Idea and server generated by the nice JoKrus

• xml library used to generate classes and serialize data xsdata

• This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

History

0.0.1 (2022-01-23)

• Engage in Project

0.1.0 (2022-01-24)

• Implemented basic api. Still some features to do

0.2.0 (2022-01-24)

• Added more docu

• Added readthedocs build config

0.2.4 (2022-01-24)

• Fixed setup.py so it automatically loads xsd and installs classes based on them

0.7.0 (2022-03-10)

• Added tournament code to client

0.7.3 (2022-03-20)

• Implemented error messages on not supported messages

• Fixed client not loading completed message when delay between packages is too big

0.8.0 (2022-03-20)

• Added method to check if move results in the new board you beeing in chess

0.9.0 (2022-03-21)

• Implemented Example AI that replays a loaded PGN. Automatically selects “path” from pgn matching player color

0.9.1 (2022-03-26)

• Is promotion Method now returns False on non valid moves

0.10.0 (2022-03-26)

• Added methods to predict boards and create fen notation from boards