utils/config.py

import yaml
from pathlib import Path
import os

def load_config():
    """Load configuration from YAML file"""
    # Get the directory containing this script
    current_dir = Path(__file__).parent.parent
    config_path = current_dir / "config.yaml"
    
    if not config_path.exists():
        # Default configuration
        config = {
            'models': {
                'spacy': 'en_core_web_sm',
                'sentiment': 'nlptown/bert-base-multilingual-uncased-sentiment'
            },
            'analysis': {
                'batch_size': 1000,
                'min_entity_confidence': 0.8,
                'num_topics': 3,
                'max_text_length': 50000
            },
            'security': {
                'max_file_size': 5242880,  # 5MB
                'allowed_extensions': ['txt']
            },
            'logging': {
                'level': 'INFO',
                'format': '%(asctime)s - %(name)s - %(levelname)s - %(message)s'
            }
        }
        
        # Save default config
        with open(config_path, 'w') as f:
            yaml.dump(config, f, default_flow_style=False)
    
    # Load config
    with open(config_path) as f:
        return yaml.safe_load(f)