import gym
from gym import spaces
import numpy as np

class ChengduMahjongEnv(gym.Env):
    def __init__(self):
        super(ChengduMahjongEnv, self).__init__()
        self.observation_space = spaces.Box(low=0, high=4, shape=(136,), dtype=np.int32)  # 每张牌的状态
        self.action_space = spaces.Discrete(136)  # 可选择打出的牌

    def reset(self):
        # 初始化麻将牌局
        self.state = np.zeros(136)
        return self.state

    def step(self, action):
        # 模拟玩家动作
        reward = 0  # 根据规则计算得分
        done = False
        return self.state, reward, done, {}