#!/usr/bin/env python3
"""
Snippet Summarizer - Generates formatted code snippet reports
Usage: python snippet-summarizer.py <directory> [--output <file>]
"""

import os
import sys
import json
from pathlib import Path
from datetime import datetime
from typing import List, Dict, Any

def get_snippet_info(filepath: Path) -> Dict[str, Any]:
    """Extract snippet metadata from a source file."""
    try:
        with open(filepath, 'r', encoding='utf-8') as f:
            content = f.read()
        
        lines = content.split('\n')
        return {
            'path': str(filepath),
            'filename': filepath.name,
            'extension': filepath.suffix,
            'lines': len(lines),
            'characters': len(content),
            'size_kb': len(content.encode('utf-8')) / 1024,
            'first_line': lines[0] if lines else '',
            'hash': ''
        }
    except Exception as e:
        return {
            'path': str(filepath),
            'error': str(e)
        }

def generate_report(directory: str, output_file: str = None) -> Dict[str, Any]:
    """Generate a code snippet summary report."""
    dir_path = Path(directory)
    
    if not dir_path.exists():
        raise ValueError(f"Directory does not exist: {directory}")
    
    # Find all Python source files
    snippets = []
    for py_file in dir_path.rglob('*.py'):
        if not py_file.name.startswith('__'):
            snippets.append(get_snippet_info(py_file))
    
    report = {
        'generated_at': datetime.now().isoformat(),
        'directory': str(dir_path.absolute()),
        'total_snippets': len(snippets),
        'total_lines': sum(s.get('lines', 0) for s in snippets),
        'total_size_kb': sum(s.get('size_kb', 0) for s in snippets),
        'snippets': snippets,
        'summary': {
            'by_extension': {},
            'avg_lines': sum(s.get('lines', 0) for s in snippets) / len(snippets) if snippets else 0
        }
    }
    
    # Group by extension
    for s in snippets:
        ext = s.get('extension', 'unknown')
        report['summary']['by_extension'][ext] = report['summary']['by_extension'].get(ext, 0) + 1
    
    # Write output
    output = json.dumps(report, indent=2)
    
    if output_file:
        with open(output_file, 'w', encoding='utf-8') as f:
            f.write(output)
        print(f"Report written to: {output_file}")
    
    return report

def main():
    if len(sys.argv) < 2:
        print("Usage: python snippet-summarizer.py <directory> [--output <file>]")
        sys.exit(1)
    
    directory = sys.argv[1]
    output_file = None
    
    if '--output' in sys.argv:
        idx = sys.argv.index('--output')
        if idx + 1 < len(sys.argv):
            output_file = sys.argv[idx + 1]
    
    try:
        report = generate_report(directory, output_file)
        print(f"Processed {report['total_snippets']} snippets")
        print(f"Total lines: {report['total_lines']}")
        print(f"Total size: {report['total_size_kb']:.2f} KB")
    except Exception as e:
        print(f"Error: {e}")
        sys.exit(1)

if __name__ == '__main__':
    main()