"""Service for managing interactions"""
import uuid
from datetime import datetime
from typing import Dict, Any, List, Optional

class InteractionService:
    def __init__(self):
        """Initialize interaction service"""
        pass
    
    def create_interaction(
        self,
        db_service,
        llm_service,
        interaction_data: Dict[str, Any]
    ) -> str:
        """
        Create new interaction with AI analysis
        
        Args:
            db_service: Database service instance
            llm_service: LLM service instance
            interaction_data: Interaction details
            
        Returns:
            str: Interaction ID
        """
        # Generate ID
        interaction_id = str(uuid.uuid4())
        
        # Analyze transcript with LLM
        if interaction_data.get('transcript'):
            analysis = llm_service.analyze_interaction(
                interaction_data['transcript']
            )
            
            interaction_data.update({
                'summary': analysis.get('summary'),
                'sentiment_score': analysis.get('sentiment_score'),
                'metadata': {
                    **interaction_data.get('metadata', {}),
                    'key_points': analysis.get('key_points', []),
                    'action_items': analysis.get('action_items', [])
                }
            })
        
        # Save to database
        return db_service.save_interaction({
            'id': interaction_id,
            **interaction_data
        })
    
    def get_interaction_stats(
        self,
        db_service,
        user_id: str
    ) -> Dict[str, Any]:
        """Get interaction statistics for user"""
        recent = db_service.get_recent_interactions(user_id)
        
        # Calculate statistics
        stats = {
            'total_count': len(recent),
            'avg_sentiment': sum(
                float(i['sentiment_score'] or 0) 
                for i in recent
            ) / len(recent) if recent else 0,
            'type_distribution': {}
        }
        
        # Count interaction types
        for interaction in recent:
            interaction_type = interaction['type']
            stats['type_distribution'][interaction_type] = \
                stats['type_distribution'].get(interaction_type, 0) + 1
        
        return stats
    
    def search_interactions(
        self,
        db_service,
        query: str,
        user_id: Optional[str] = None,
        limit: int = 10
    ) -> List[Dict]:
        """Search interactions"""
        return db_service.search_interactions(query, user_id, limit)