Update to latest socket spec

Add get/report for game speed/paused

Author: VirxEC

Cargo.lock

@@ -1,6 +1,6 @@
 [package]
 name = "rlviser-py"
-version = "0.6.5"
+version = "0.6.6"
 edition = "2021"
 description = "Python implementation that manages a UDP connection to RLViser"
 license = "MIT"

Cargo.toml

@@ -21,7 +21,7 @@ def render(self, state: GameState, shared_info: Dict[str, Any]) -> Any:
         ball.pos = rsim.Vec(*state.ball.position)
         ball.vel = rsim.Vec(*state.ball.linear_velocity)
         ball.ang_vel = rsim.Vec(*state.ball.angular_velocity)
-        ball.rot_mat = rsim.RotMat(*state.ball.rotation_mtx.transpose().flatten())
+        # ball.rot_mat = rsim.RotMat(*state.ball.rotation_mtx.transpose().flatten())
 
         car_data = []
         for idx, car in enumerate(state.cars.values()):
@@ -59,7 +59,7 @@ def _get_car_state(self, car: Car):
         car_state.has_double_jumped = car.has_double_jumped
         car_state.air_time_since_jump = car.air_time_since_jump
         car_state.flip_time = car.flip_time
-        car_state.last_rel_dodge_torque = rsim.Vec(*car.flip_torque)
+        car_state.flip_rel_torque = rsim.Vec(*car.flip_torque)
 
         car_state.is_auto_flipping = car.is_autoflipping
         car_state.auto_flip_timer = car.autoflip_timer

gym_renderer.py

@@ -1,43 +1,50 @@
 import time
-import numpy as np
 from itertools import chain
 
+import numpy as np
+import rlviser_py
 from rlgym.api import RLGym
 from rlgym.rocket_league.action_parsers import LookupTableAction, RepeatAction
-from rlgym.rocket_league.done_conditions import GoalCondition, AnyCondition, TimeoutCondition, NoTouchTimeoutCondition
+from rlgym.rocket_league.done_conditions import (
+    AnyCondition,
+    GoalCondition,
+    NoTouchTimeoutCondition,
+    TimeoutCondition,
+)
 from rlgym.rocket_league.obs_builders import DefaultObs
 from rlgym.rocket_league.reward_functions import CombinedReward, GoalReward, TouchReward
 from rlgym.rocket_league.sim import RocketSimEngine
-from rlgym.rocket_league.state_mutators import MutatorSequence, FixedTeamSizeMutator, KickoffMutator
+from rlgym.rocket_league.state_mutators import (
+    FixedTeamSizeMutator,
+    KickoffMutator,
+    MutatorSequence,
+)
 
 from gym_renderer import RLViserRenderer
 
 if __name__ == "__main__":
+    game_speed = 1
+
     env = RLGym(
         state_mutator=MutatorSequence(
-            FixedTeamSizeMutator(blue_size=2, orange_size=2),
-            KickoffMutator()
+            FixedTeamSizeMutator(blue_size=2, orange_size=2), KickoffMutator()
         ),
         obs_builder=DefaultObs(zero_padding=None),
         action_parser=RepeatAction(LookupTableAction(), repeats=1),
-        reward_fn=CombinedReward(
-            (GoalReward(), 10.),
-            (TouchReward(), 0.1)
-        ),
+        reward_fn=CombinedReward((GoalReward(), 10.0), (TouchReward(), 0.1)),
         termination_cond=GoalCondition(),
         truncation_cond=AnyCondition(
-            TimeoutCondition(timeout=300.),
-            NoTouchTimeoutCondition(timeout=30.)
+            TimeoutCondition(timeout=300.0), NoTouchTimeoutCondition(timeout=30.0)
         ),
         transition_engine=RocketSimEngine(),
-        renderer=RLViserRenderer(120)
+        renderer=RLViserRenderer(120),
     )
 
     # simulate 2 episodes
     for _ in range(2):
         obs_dict = env.reset()
         steps = 0
-        ep_reward = {agent_id: 0. for agent_id in env.agents}
+        ep_reward = {agent_id: 0.0 for agent_id in env.agents}
         t0 = time.time()
         while True:
             actions = {}
@@ -46,9 +53,11 @@
                 actions[agent_id] = np.random.randint(action_space, size=(1,))
 
             for _ in range(8):
-                obs_dict, reward_dict, terminated_dict, truncated_dict = env.step(actions)
+                obs_dict, reward_dict, terminated_dict, truncated_dict = env.step(
+                    actions
+                )
                 env.render()
-                time.sleep(max(0, t0 + steps / 1200 - time.time()))
+                time.sleep(max(0, t0 + steps / (120 * game_speed) - time.time()))
                 steps += 1
 
             for agent_id, reward in reward_dict.items():
@@ -57,5 +66,11 @@
             if any(chain(terminated_dict.values(), truncated_dict.values())):
                 break
 
+            game_speed = rlviser_py.get_game_speed()
+
         ep_time = time.time() - t0
-        print(f"Steps per second: {steps / ep_time:.0f} | Episode time: {ep_time:.2f} | Episode Reward: {max(ep_reward.values()):.2f}")
+        print(
+            f"Steps per second: {steps / ep_time:.0f} | Episode time: {ep_time:.2f} | Episode Reward: {max(ep_reward.values()):.2f}"
+        )
+
+    env.close()