Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/ivzeng/chinese-chess-xiangqi
My Chinese Chess application.
https://github.com/ivzeng/chinese-chess-xiangqi
chinese-chess-engine cpp game-theory-algorithms
Last synced: 21 days ago
JSON representation
My Chinese Chess application.
- Host: GitHub
- URL: https://github.com/ivzeng/chinese-chess-xiangqi
- Owner: ivzeng
- Created: 2022-12-28T22:05:27.000Z (about 2 years ago)
- Default Branch: master
- Last Pushed: 2023-09-30T23:04:13.000Z (over 1 year ago)
- Last Synced: 2024-12-30T17:55:57.669Z (21 days ago)
- Topics: chinese-chess-engine, cpp, game-theory-algorithms
- Language: C++
- Homepage:
- Size: 3.39 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Hello!
### This is my 象棋 (Chinese Chess) program! Currently, it is a console application that provides a decent interface for Chinese Chess and some computer players based on a backtracking move-searching algorithm. So far, the interface part is complete and it supports many features such as board diplay, histroy, hint, and move-undo. I am working on the 5 computer players right now. Specifically, `Computer0`, `Computer1`, `Computer3`, and `computer4` use the basic move-searching algorithm with different depth, while `Computer3` and `Computer4` may make random move, and `Computer4` has its data base that allows it to learn and improve. `Computer2` uses a difference outcome function that I will test it in the future.
#
## Build
To compile and build the program, please use the `makefile` or the bash shell script `build.sh` under `tools` directory.
To see the usage of `build.sh`, please run
```
tools/build.sh usage
```
By default, the compilation will give a program initially in Chinese, but you can run `build.sh` with `-le` to get a program in English, or alternatively, you can set it on the setting page in the program.
## Start a Game
When you execute the program, you will be in the main page. You can enter certain commands to do something.
For example,
to start a game with an ai as the red side and you as the black side, you can enter:
```
play c1 h
```
or
```
play
c2
h
```
to set the language, you can enter:
```
set setl back
```
to exit, simply enter
```
end
```
### Here is the list of commands:
- `showc`: show all avaliable commands for user at the current program state
- `showb`: display the board
- `play`: start a game (and start setting players if any players is not set)
- `train`: enter training mode
- `set`: to setting
- `setl`: set language to another one (Chinese or English)
- `end`: exit (main page) | stop the game and return to the main page (game)
- `back`: return to the previous page
- `hint`: request a hint (human player only)
- `m`: make a move (computer only)
- `undo`: undo a move
- `moves`: show all valid moves of the play at current round
### and the list of players:
- `h`: human player
- `c[0-4]`: computer player [0-4]
### Training Mode
This mode will let two computer players player for a number of games. It is useful for training the database of the computer player.
An example to train Computer 4 with 10 games
```
train c4 c4 10
```
### Demo
There is a demo for input and output, `demo1.in` and `demo1.out` in the demo directory. To view the demo output with coloured texts, please run
```
cat demo1.out
```
---
# TODO!!!
- Continue to optimize and simplify the code
- Get a better outcome function for Board to make C2 smarter
## Changelogs
### 2023-03-17
#### Added
- Building Tools (`build.sh`)
- `build.sh usage` to see the usage of `build.sh`
- Initial Language Options on Building
- you can set the initial language of the program to English by running `build.sh -le` or `make PREPROCESSOR=-DINIT_LANGUAGE=2`
### 2023-03-24
#### Changed
- DFS Depth control
- the weird stepping function is switched into a better and more clear function that recudes the depth by factoring with number of valid moves
### 2023-03-31
#### Added
- Debugging Option:
- to compile with enabled debugging mode, run `build.sh` with `-d` flag or `make` with `make "DEBUG=-DDEBUG -g"`
### 2023-05-03
#### Changed
- adjusted move analysis functions: `Player::dfsMovesAnalysis` now would calculate the expected payoff of each possible move and store the result in a vector of pair (payoff, move) called `movesEPayoff`, sorted in accending order of the payoff
- removed `XQBoard::outcome1` (the outcome function that analysizes the piece exchange outcome), renamed `XQBoard::outcome2` to `XQBoard::outcome1`
- simplified and optimized code in Player.cpp
### 2023-05-15
#### Added
- computer3, an variation of computer2 with ramdonization on move-selection
- (position) bounus matrix for each XQPiece for a better outcome funtion for XQBoard
#### Changed
- improved XQBoard::outcome and the related outcome function
- reduced the operation for computer2/3
### 2023-05-23
#### Added
- Computer 4
- a computer player with all utilities of Computer 3 and the ability of using a database
- Database for Computer 4
- Training mode
- allows two computer player to player for a number of games
#### Changed
- Text.h/cpp
- slight changes on message
### 2023-09-30
#### Added
- builds of the program
#### Changed
- adjust the depth of searching algorithm for each player to ensure that they find a move in a reasonable time
- add a variable to tell the program about the environment
## Have fun!