import numpy as np
import gym
from gym import spaces
from baselines.common.atari_wrappers import make_atari, wrap_deepmind
from baselines.common.vec_env import VecEnv
from multiprocessing import Process, Pipe

# cf

def make_env(env_name, rank, seed):
    env = make_atari(env_name)
    env.seed(seed + rank)
    env = wrap_deepmind(env)
    return env

def worker(remote, parent_remote, env_fn_wrapper):
    env = env_fn_wrapper.x()
    while True:
        cmd, data = remote.recv()
        if cmd == 'step':
            ob, reward, done, info = env.step(data)
            if done:
                ob = env.reset()
            remote.send((ob, reward, done, info))
        elif cmd == 'reset':
            ob = env.reset()
        elif cmd == 'reset_task':
            ob = env.reset_task()
        elif cmd == 'close':
        elif cmd == 'get_spaces':
            remote.send((env.action_space, env.observation_space))
        elif cmd == 'render':
            raise NotImplementedError

class CloudpickleWrapper(object):
    Uses cloudpickle to serialize contents (otherwise multiprocessing tries to use pickle)
    def __init__(self, x):
        self.x = x
    def __getstate__(self):
        import cloudpickle
        return cloudpickle.dumps(self.x)
    def __setstate__(self, ob):
        import pickle
        self.x = pickle.loads(ob)

class RenderSubprocVecEnv(VecEnv):
    def __init__(self, env_fns, render_interval):
        """ Minor addition to SubprocVecEnv, automatically renders environments

        envs: list of gym environments to run in subprocesses
        self.closed = False
        nenvs = len(env_fns)
        self.remotes, self.work_remotes = zip(*[Pipe() for _ in range(nenvs)]) = [Process(target=worker, args=(work_remote, remote, CloudpickleWrapper(env_fn)))
            for (work_remote, remote, env_fn) in zip(self.work_remotes, self.remotes, env_fns)]
        for p in
            p.daemon = True # if the main process crashes, we should not cause things to hang
        for remote in self.work_remotes:

        self.remotes[0].send(('get_spaces', None))
        self.action_space, self.observation_space = self.remotes[0].recv()

        self.render_interval = render_interval
        self.render_timer = 0

    def step(self, actions):
        for remote, action in zip(self.remotes, actions):
            remote.send(('step', action))
        results = [remote.recv() for remote in self.remotes]
        obs, rews, dones, infos = zip(*results)

        self.render_timer += 1
        if self.render_timer == self.render_interval:
            for remote in self.remotes:
                remote.send(('render', None))
            self.render_timer = 0

        return np.stack(obs), np.stack(rews), np.stack(dones), infos

    def reset(self):
        for remote in self.remotes:
            remote.send(('reset', None))
        return np.stack([remote.recv() for remote in self.remotes])

    def reset_task(self):
        for remote in self.remotes:
            remote.send(('reset_task', None))
        return np.stack([remote.recv() for remote in self.remotes])

    def close(self):
        if self.closed:

        for remote in self.remotes:
            remote.send(('close', None))
        for p in
        self.closed = True

    def num_envs(self):
        return len(self.remotes)