Fix return_info for wrappers (#2612)

* Fix `return_info` for Observation wrappers, Atari (?) and framestack

* Make type checkers and IDEs happier

* Merge in #2454

* Update the info dict based on no-op steps
Some type hints

* Bug fix

* Handle resets during frameskip

gym/core.py

@@ -300,8 +300,11 @@ class Wrapper(Env[ObsType, ActType]):
 
 class ObservationWrapper(Wrapper):
     def reset(self, **kwargs):
-        observation = self.env.reset(**kwargs)
+        if kwargs.get("return_info", False):
-        return self.observation(observation)
+            obs, info = self.env.reset(**kwargs)
+            return self.observation(obs), info
+        else:
+            return self.observation(self.env.reset(**kwargs))
 
     def step(self, action):
         observation, reward, done, info = self.env.step(action)

gym/wrappers/atari_preprocessing.py

@@ -1,9 +1,6 @@
-from typing import Optional
-
 import numpy as np
 import gym
 from gym.spaces import Box
-from gym.wrappers import TimeLimit
 
 try:
     import cv2
@@ -45,14 +42,14 @@ class AtariPreprocessing(gym.Wrapper):
 
     def __init__(
         self,
-        env,
+        env: gym.Env,
-        noop_max=30,
+        noop_max: int = 30,
-        frame_skip=4,
+        frame_skip: int = 4,
-        screen_size=84,
+        screen_size: int = 84,
-        terminal_on_life_loss=False,
+        terminal_on_life_loss: bool = False,
-        grayscale_obs=True,
+        grayscale_obs: bool = True,
-        grayscale_newaxis=False,
+        grayscale_newaxis: bool = False,
-        scale_obs=False,
+        scale_obs: bool = False,
     ):
         super().__init__(env)
         assert (
@@ -62,10 +59,14 @@ class AtariPreprocessing(gym.Wrapper):
         assert screen_size > 0
         assert noop_max >= 0
         if frame_skip > 1:
-            assert "NoFrameskip" in env.spec.id, (
+            if (
-                "disable frame-skipping in the original env. for more than one"
+                "NoFrameskip" not in env.spec.id
-                " frame-skip as it will be done by the wrapper"
+                and getattr(env.unwrapped, "_frameskip", None) != 1
-            )
+            ):
+                raise ValueError(
+                    "Disable frame-skipping in the original env. Otherwise, more than one"
+                    " frame-skip will happen as through this wrapper"
+                )
         self.noop_max = noop_max
         assert env.unwrapped.get_action_meanings()[0] == "NOOP"
 
@@ -131,16 +132,26 @@ class AtariPreprocessing(gym.Wrapper):
 
     def reset(self, **kwargs):
         # NoopReset
-        self.env.reset(**kwargs)
+        if kwargs.get("return_info", False):
+            _, reset_info = self.env.reset(**kwargs)
+        else:
+            _ = self.env.reset(**kwargs)
+            reset_info = {}
+
         noops = (
             self.env.unwrapped.np_random.integers(1, self.noop_max + 1)
             if self.noop_max > 0
             else 0
         )
         for _ in range(noops):
-            _, _, done, _ = self.env.step(0)
+            _, _, done, step_info = self.env.step(0)
+            reset_info.update(step_info)
             if done:
-                self.env.reset(**kwargs)
+                if kwargs.get("return_info", False):
+                    _, reset_info = self.env.reset(**kwargs)
+                else:
+                    _ = self.env.reset(**kwargs)
+                    reset_info = {}
 
         self.lives = self.ale.lives()
         if self.grayscale_obs:
@@ -148,7 +159,11 @@ class AtariPreprocessing(gym.Wrapper):
         else:
             self.ale.getScreenRGB(self.obs_buffer[0])
         self.obs_buffer[1].fill(0)
-        return self._get_obs()
+
+        if kwargs.get("return_info", False):
+            return self._get_obs(), reset_info
+        else:
+            return self._get_obs()
 
     def _get_obs(self):
         if self.frame_skip > 1:  # more efficient in-place pooling

gym/wrappers/frame_stack.py

@@ -119,6 +119,14 @@ class FrameStack(ObservationWrapper):
         return self.observation(), reward, done, info
 
     def reset(self, **kwargs):
-        observation = self.env.reset(**kwargs)
+        if kwargs.get("return_info", False):
-        [self.frames.append(observation) for _ in range(self.num_stack)]
+            obs, info = self.env.reset(**kwargs)
-        return self.observation()
+        else:
+            obs = self.env.reset(**kwargs)
+            info = None  # Unused
+        [self.frames.append(obs) for _ in range(self.num_stack)]
+
+        if kwargs.get("return_info", False):
+            return self.observation(), info
+        else:
+            return self.observation()