"""
Cache manager for tracking processed articles to avoid reprocessing.
"""

import json
import os
import logging
from datetime import datetime
from typing import Dict, List, Optional

from config import CACHE_FILE

logger = logging.getLogger(__name__)

class CacheManager:
    """Manages caching of processed articles to prevent duplicate processing."""
    
    def __init__(self, cache_file: str = CACHE_FILE):
        self.cache_file = cache_file
        self.cache = self._load_cache()
    
    def _load_cache(self) -> Dict:
        """Load cache from file."""
        try:
            if os.path.exists(self.cache_file):
                with open(self.cache_file, 'r', encoding='utf-8') as f:
                    return json.load(f)
            return {}
        except Exception as e:
            logger.error(f"Error loading cache file {self.cache_file}: {e}")
            return {}
    
    def _save_cache(self) -> None:
        """Save cache to file."""
        try:
            # Create directory if it doesn't exist
            os.makedirs(os.path.dirname(self.cache_file), exist_ok=True)
            
            with open(self.cache_file, 'w', encoding='utf-8') as f:
                json.dump(self.cache, f, indent=2, ensure_ascii=False)
        except Exception as e:
            logger.error(f"Error saving cache file {self.cache_file}: {e}")
    
    def is_processed(self, file_path: str) -> bool:
        """Check if a file has been processed."""
        return file_path in self.cache
    
    def mark_processed(self, file_path: str, status: str = "processed") -> None:
        """Mark a file as processed."""
        self.cache[file_path] = {
            "processed_date": datetime.now().isoformat(),
            "status": status,
            "last_updated": datetime.now().isoformat()
        }
        self._save_cache()
    
    def get_processed_files(self) -> List[str]:
        """Get list of all processed files."""
        return list(self.cache.keys())
    
    def get_cache_stats(self) -> Dict:
        """Get cache statistics."""
        return {
            "total_files": len(self.cache),
            "processed_files": len(self.cache),
            "cache_file": self.cache_file
        }
    
    def clear_cache(self) -> None:
        """Clear the entire cache."""
        self.cache = {}
        self._save_cache()
        logger.info("Cache cleared")