from typing import List, Dict, Any

class AdaptiveAgent:
    """
    Agent A: The Coach
    Responsibility: Analyzes user performance history to determine the optimal difficulty level.
    """
    
    DIFFICULTIES = ["Easy", "Intermediate", "Expert"]
    
    def evaluate_user(self, history: List[Dict[str, Any]]) -> Dict[str, Any]:
        """
        Analyzes a list of past match results to determine new difficulty.
        
        Args:
            history: List of dicts, e.g., 
                     [{'result': 'success', 'time_taken': 45, 'hints_used': 0}, ...]
        
        Returns:
            Dict with 'recommended_difficulty' and 'reasoning'.
        """
        if not history:
            return {"difficulty": "Easy", "reason": "New user"}
            
        # Analyze last 5 games
        recent_games = history[-5:]
        success_count = sum(1 for game in recent_games if game.get('result') == 'success')
        avg_time = sum(game.get('time_taken', 0) for game in recent_games) / len(recent_games)
        total_hints = sum(game.get('hints_used', 0) for game in recent_games)
        
        current_difficulty = recent_games[-1].get('difficulty', "Easy")
        
        # Promotion Logic
        if success_count >= 4 and total_hints <= 2:
            next_diff = self._change_difficulty(current_difficulty, 1)
            return {
                "difficulty": next_diff,
                "reason": "Consistently successful with few hints. Promoting!"
            }
            
        # Demotion Logic
        if success_count <= 1:
            next_diff = self._change_difficulty(current_difficulty, -1)
            return {
                "difficulty": next_diff,
                "reason": "Struggling with current tasks. Let's practice basics."
            }
            
        # Maintain
        return {
            "difficulty": current_difficulty,
            "reason": "Steady progress. Maintaining current challenge level."
        }

    def _change_difficulty(self, current: str, step: int) -> str:
        try:
            # Normalize casing just in case
            current = current.capitalize()
            if current not in self.DIFFICULTIES:
                 # Fallback if unknown difficulty comes in
                 return "Easy"
                 
            idx = self.DIFFICULTIES.index(current)
            new_idx = max(0, min(len(self.DIFFICULTIES) - 1, idx + step))
            return self.DIFFICULTIES[new_idx]
        except ValueError:
            return "Easy"

if __name__ == "__main__":
    # Test logic
    agent = AdaptiveAgent()
    history = [
        {'result': 'success', 'time_taken': 30, 'hints_used': 0, 'difficulty': 'Beginner'},
        {'result': 'success', 'time_taken': 25, 'hints_used': 0, 'difficulty': 'Beginner'},
        {'result': 'success', 'time_taken': 40, 'hints_used': 1, 'difficulty': 'Beginner'},
        {'result': 'success', 'time_taken': 35, 'hints_used': 0, 'difficulty': 'Beginner'}
    ]
    print(agent.evaluate_user(history))