CHANGELOG-en.md

Change Log

The format is modified from Keep a Changelog.

[9.3.5] - 2022.04.18

Added

• Add toggle_with_bias in game_progress

[9.3.4] - 2022.04.18

Fixed

• Fix main.py in easy_game.

[9.3.3] - 2022.04.14

Added

• Add create_scene_progress_data in view_model.py

Change

• Must return frame inget_scene_progress_data

[9.3.2] - 2022.04.12

Fixed

• Improve the efficiency on drawing.

[9.3.1] - 2022.04.08

Change

• MLGame will transmit keyboard event (A-Z,0-9,Arrow) to AI Client.

[9.2.3] - 2022.02.13

Fix

• Fix game will not stop when AI didn't return RESET .

[9.2.2-beta] - 2022.02.09

Changed

• update blockly.json in easy_game

[9.2.1-beta] - 2022.01.24

Changed

• change the format of command
  • old: python MLGame.py -i ml_play_template.py easy_game 1200 15 10 FF9800
  • now: python MLGame.py -i ml_play_template.py easy_game --score 10 --color FF9800 --time_to_play 1200 --total_point 15

Added

• Add scroll button, we could use I J K L U O H to update the view now.
• Screen will be reset when game is reset.

[9.0.7-beta] - 2021.07.23

Changed

• Change the function name and data format in each game.

Added

• Add Interface PaiaGame every game should implement this interface to run on PAIA.
• Add decorator to validate data format.
• Add view.py in MLGame to draw scene of game.
• Add two game in submodule.

[Beta 8.0.1] - 2020.10.05

Changed

• The recorded game progress for the inactivated ml client will be an empty list.

[Beta 8.0] - 2020.09.29

Added

• Add "record_format_version" field to the log file

Changed

• Change the data structure received from or sent to the game class: use a dict to store the information of each ml client. For example, the game has two players and defines the name of them as "ml_1P" and "ml_2P" in the config.py.
  • The scene information returned from get_player_scene_info() will be:

    {
        "ml_1P": scene_info_for_ml_1P,
        "ml_2P": scene_info_for_ml_2P
    }
    

  • The command sent to the update() will be:

    {
        "ml_1P": command_sent_from_ml_1P,
        "ml_2P": command_sent_from_ml_2P
    }
    

  • And the command returned from get_keyboard_command() should also be:

    {
        "ml_1P": command_for_ml_1P,
        "ml_2P": command_for_ml_2P
    }
    

• Record the scene information and command individually for each ml client in the log file
• Update the version of built-in games
  • arkanoid: 1.0 -> 1.1
  • pingpong: 1.1 -> 1.2
  • snake: 1.0 -> 1.1

[Beta 7.2] - 2020.08.23

Fixed

• The game could not be closed when running non-python client

Changed

• -i/--input-script supports specifying the file in the subdirectory
  • For example, -i foo/ml_play.py means the file is at games/<game_name>/ml/foo/ml_play.py

Removed

• Remove --input-module flag

[Beta 7.1.3] - 2020.06.20

Fixed

• Handle the exception of SystemExit
• Handle the situation of reseting the ml executor before the game ends
  • The game executor will ignore the ready command.

Changed

• Optimize the checking of the received object from either side

[Beta 7.1.2] - 2020.06.19

Fixed

• Use wrong value to chack frame delay

Changed

• Modify the game "snake" for the game development tutorial
  • The ml_play_template.py of the game "snake" contains simple rule-based algorithm.

[Beta 7.1.1] - 2020.06.15

Changed

• ML process doesn't send the command only when the returned value of MLPlay.update() is None.

[Beta 7.1] - 2020.06.01

Fixed

• Handle the exception of BrokenPipeError

Added

• Add "dynamic_ml_clients" to the "GAME_SETUP" of the game config

[Beta 7.0.1] - 2020.05.29

This update is compatible with Beta 7.0.

Fixed

• Hang when the game exits on Linux

Added

• Add errno.py to define the exit code of errors
• Handle the exception occurred in manual mode

Changed

• Change the exit code of errors

[Beta 7.0] - 2020.05.27

Added

• Use executors to control the execution loop
  • The game and the ml script only need to provide "class" for the executor to invoke (like an interface).
  • The game doesn't need to provide manual and ml version. Just one game class.
  • Replace ml_loop() with MLPlay class in ml script
• Add commnuication manager
  • The manager for the ml process has a queue for storing the object received. If the queue has more than 15 objects, the oldest object will be dropped.

Changed

• Change the format of the recording game progress
• Replace PROCESSES with GAME_SETUP in config.py of the game to setup the game and the ml scripts
• Rename GameConfig to ExecutionCommand
• Simplfy the communication package into a module

Removed

• Remove record.py and ml version of the game in the game directory

[Beta 6.1] - 2020.05.06

Changed

• Pingpong - version 1.1
  • Shorten the ball speed increasing interval
  • Randomly set the initial position of the blocker, and speed up the moving speed of it

[Beta 6.0] - 2020.04.28

Added

• Add -l and --list flag for listing available games
• Use config.py in games to set up the game parameters and the game execution
• Use argparse for generating and handling game parameters
  • List game parameters of a game by using python MLGame.py <game> -h
• Exit with non-zero value when an error occurred

Changed

• The game execution flags must be specified before the game name, including -i/--input-script/--input-module flags
• -i/--input-script/--input-module flags carry one script or one module at a time.
  • Specify these flags multiple times for multiple scripts or modules, such as -i script_1P -i script_2P.
• Games: Use dictionary objects as communication objects between game and ml processes for flexibility
  • The record file only contains dictionay objects and built-in types, therefore, it can be read outside the mlgame directory.
• mlgame.gamedev.recorder.RecorderHelper only accepts dictionary object.
• Code refactoring

Removed

• Games
  • Remove main.py (replaced by config.py)
  • Remove communication.py
    • For the ml script, use mlgame.communication.ml module to communicate with the game process. See ml_play_template.py for the example.
• Remove CommandReceiver from the mlgame.communication.game
  • The game has to validate the command recevied by itself.

[Beta 5.0.1] - 2020.03.06

Fixed

• Fix typo in the README of the arkanoid
• Arkanoid: Add additional checking condition for the ball bouncing

[Beta 5.0] - 2020.03.03

Added

• Arkanoid and Pingpong:
  • The serving position and direction of the ball can be decided
  • Add difficulties for different mechanisms
  • Add ball slicing mechanism
• Arkanoid: Add hard bricks
• Pingpong: Add blocker

Changed

• Update the python from 3.5 to 3.6: For the auto() of the custom Enum
• Optimize the output of the error message
• Refactor the game classes: Extract the drawing and recording functions
• Add prefix to the filename of the record files
• Physics: Optimize the ball bouncing algorithm

[Beta 4.1] - 2019.11.06

Added

• New game - Snake
• Add README to the game Arkanoid and Pingpong

Changed

• Update pygame from 1.9.4 to 1.9.6
• Arkanoid and Pingpong (Follow the structure of the game Snake):
  • Move SceneInfo to the gamecore.py
  • Rename GameInstruction to GameCommand
• Arkanoid: Add command member to SceneInfo
  • Trying to load the record files generated before beta 4.1 will get AttributeError: 'SceneInfo' object has no attribute 'command' error.
• Code refactoring

[Beta 4.0] - 2019.08.30

Added

• mlgame - MLGame development API
• --input-module flag for specifying the absolute importing path of user modules

Changed

• Use 4 spaces instead of tab
• Support one shot mode in the manual mode
• Fit the existing games to the new API
• Move the directory of game to "games" directory
• Arkanoid: Wait for the ml process before start new round
• Arkanoid: Change the communication API

[Beta 3.2] - 2019.07.30

Changed

• Pingpong: Exchange the 1P and 2P side
• Code refactoring

[Beta 3.1] - 2019.05.28

Changed

• Pingpong: Set the height of the platform from 10 to 30
• Optimize the collision detection algorithm

[Beta 3.0] - 2019.05.22

Added

• 2P game "pingpong"

Changed

• Optimize the call stack message of ml process
• Use argparse instead of optparse

[Beta 2.2.2] - 2019.04.15

Fixed

• The game doesn't wait for the ready command

[Beta 2.2.1] - 2019.04.12

Fixed

• The game hangs when the exception occurred before ml_ready()

Changed

• Some code refactoring and optimization

[Beta 2.2] - 2019.04.01

Added

• -i and --input-script for specifying the custom ml script

[Beta 2.1.1] - 2019.03.21

Added

• Print the whole call stack when the exception occurred

[Beta 2.1] - 2019.03.18

Fixed

• Quit the game automatically when an exception occurred

Added

• -1 and --one-shot for the one shot mode in ml mode
• Version message

[Beta 2.0] - 2019.02.27

Changed

• Use function call instead of class instance to invoke use code
• Optimize the collision detection algorithm
• Increase the difficulty of the game "arkanoid"

Added

• -r and --record to record the game progress