game_player.py

class GamePlayer:
    def __init__(self, game_dsl):
        self.game_dsl = game_dsl
        self.state = {}

    def reset(self, game_name, agents):
        self.state = {
            "score": {a.name: 0 for a in agents},
            "positions": {a.name: [0, 0] for a in agents},
            "done": False
        }

    def step(self, game, agents):

        events = []
        rewards = {a.name: 0.0 for a in agents}

        for rule in game["rules"]:

            for agent in agents:

                event = self._evaluate_rule(rule, agent)

                if not event:
                    continue

                events.append(event)

                if event["type"] == "score":
                    rewards[event["agent"]] += float(event.get("delta", 1.0))

                if event["type"] == "reset":
                    rewards[agent.name] -= 0.1

        return events, rewards

    def _evaluate_rule(self, rule, agent):

        if "collision" in rule:
            return {"agent": agent.name, "type": "score", "delta": 1}

        if "reset" in rule:
            return {"type": "reset"}

        return None