cli.py

#!/usr/bin/env python3
"""
Command Line Interface for Turkish News RAG Assistant.
Provides easy access to common operations without the web interface.
"""

import argparse
import sys
from datetime import datetime
import json

def main():
    parser = argparse.ArgumentParser(
        description='Turkish News RAG Assistant CLI',
        formatter_class=argparse.RawDescriptionHelpFormatter,
        epilog="""
Examples:
  %(prog)s test                           # Run system tests
  %(prog)s collect --test                 # Collect news (test mode)
  %(prog)s collect --categories Teknoloji Bilim
  %(prog)s ask "Bugün dolar ne kadar?"    # Ask a question
  %(prog)s summary --hours 24             # Generate daily summary
  %(prog)s search "yapay zeka"            # Search news
  %(prog)s stats                          # Show database statistics
  %(prog)s ui                             # Launch web interface
        """
    )
    
    subparsers = parser.add_subparsers(dest='command', help='Available commands')
    
    # Test command
    test_parser = subparsers.add_parser('test', help='Run system tests')
    
    # Collect command
    collect_parser = subparsers.add_parser('collect', help='Collect news articles')
    collect_parser.add_argument('--categories', nargs='+', help='Specific categories to collect')
    collect_parser.add_argument('--test', action='store_true', help='Run in test mode')
    collect_parser.add_argument('--no-scraping', action='store_true', help='Skip content scraping')
    collect_parser.add_argument('--no-newsapi', action='store_true', help='Skip NewsAPI')
    collect_parser.add_argument('--max-articles', type=int, help='Max articles per category')
    
    # Ask command
    ask_parser = subparsers.add_parser('ask', help='Ask a question')
    ask_parser.add_argument('question', help='Question to ask')
    ask_parser.add_argument('--category', help='Filter by category')
    ask_parser.add_argument('--source', help='Filter by source')
    ask_parser.add_argument('--max-sources', type=int, default=5, help='Max sources to use')
    
    # Summary command
    summary_parser = subparsers.add_parser('summary', help='Generate news summary')
    summary_parser.add_argument('--hours', type=int, default=24, help='Time range in hours')
    summary_parser.add_argument('--output', help='Output file (default: stdout)')
    
    # Search command
    search_parser = subparsers.add_parser('search', help='Search news articles')
    search_parser.add_argument('query', help='Search query')
    search_parser.add_argument('--category', help='Filter by category')
    search_parser.add_argument('--limit', type=int, default=10, help='Number of results')
    search_parser.add_argument('--json', action='store_true', help='Output as JSON')
    
    # Stats command
    stats_parser = subparsers.add_parser('stats', help='Show database statistics')
    stats_parser.add_argument('--json', action='store_true', help='Output as JSON')
    
    # UI command
    ui_parser = subparsers.add_parser('ui', help='Launch web interface')
    ui_parser.add_argument('--port', type=int, default=8501, help='Port number')
    
    args = parser.parse_args()
    
    if not args.command:
        parser.print_help()
        return
    
    # Execute commands
    if args.command == 'test':
        run_tests()
    elif args.command == 'collect':
        run_collection(args)
    elif args.command == 'ask':
        run_question(args)
    elif args.command == 'summary':
        run_summary(args)
    elif args.command == 'search':
        run_search(args)
    elif args.command == 'stats':
        run_stats(args)
    elif args.command == 'ui':
        run_ui(args)

def run_tests():
    """Run system tests."""
    print("🧪 Running system tests...")
    try:
        from test_system import run_all_tests
        success = run_all_tests()
        sys.exit(0 if success else 1)
    except ImportError:
        print("❌ Test system not found. Make sure test_system.py exists.")
        sys.exit(1)

def run_collection(args):
    """Run news collection pipeline."""
    print("📊 Starting news collection...")
    
    try:
        from main import NewsPipeline
        
        pipeline = NewsPipeline()
        
        # Configure arguments
        kwargs = {
            'categories': args.categories,
            'scrape_content': not args.no_scraping,
            'use_newsapi': not args.no_newsapi,
            'max_articles_per_category': args.max_articles
        }
        
        # Test mode
        if args.test:
            kwargs['categories'] = ['Bilim', 'Teknoloji']
            kwargs['max_articles_per_category'] = 5
            print("🧪 Running in test mode (limited data)")
        
        # Run pipeline
        report = pipeline.run_full_pipeline(**kwargs)
        
        # Print summary
        print("\n📋 Collection Summary:")
        print(f"  Articles fetched: {report['articles']['fetched']}")
        print(f"  Articles processed: {report['articles']['processed']}")
        print(f"  Chunks created: {report['chunks']['created']}")
        print(f"  Chunks indexed: {report['chunks']['indexed']}")
        print(f"  Duration: {report['execution_time']['duration']}")
        
        if report['errors']:
            print(f"  Errors: {len(report['errors'])}")
        
    except Exception as e:
        print(f"❌ Collection failed: {e}")
        sys.exit(1)

def run_question(args):
    """Answer a question using RAG."""
    print(f"🤖 Answering: {args.question}")
    
    try:
        from rag_engine import RAGEngine
        
        rag = RAGEngine()
        
        result = rag.answer_question(
            question=args.question,
            category=args.category,
            source=args.source,
            max_chunks=args.max_sources
        )
        
        print(f"\n💬 Answer:")
        print(result['answer'])
        
        print(f"\n📊 Confidence: {result['confidence']:.2%}")
        print(f"📚 Sources used: {result['context_used']}")
        
        if result.get('sources'):
            print(f"\n📰 Top Sources:")
            for i, source in enumerate(result['sources'][:3]):
                print(f"  {i+1}. {source.get('source', 'Unknown')} - {source.get('title', 'No title')[:50]}...")
    
    except Exception as e:
        print(f"❌ Question answering failed: {e}")
        sys.exit(1)

def run_summary(args):
    """Generate news summary."""
    print(f"📋 Generating summary for last {args.hours} hours...")
    
    try:
        from rag_engine import RAGEngine
        
        rag = RAGEngine()
        summary = rag.generate_daily_summary(hours=args.hours)
        
        if args.output:
            with open(args.output, 'w', encoding='utf-8') as f:
                f.write(summary)
            print(f"✅ Summary saved to {args.output}")
        else:
            print("\n" + "="*60)
            print("📰 NEWS SUMMARY")
            print("="*60)
            print(summary)
    
    except Exception as e:
        print(f"❌ Summary generation failed: {e}")
        sys.exit(1)

def run_search(args):
    """Search news articles."""
    print(f"🔍 Searching for: {args.query}")
    
    try:
        from rag_engine import RAGEngine
        
        rag = RAGEngine()
        articles = rag.search_news(
            query=args.query,
            limit=args.limit,
            category=args.category
        )
        
        if args.json:
            print(json.dumps(articles, indent=2, ensure_ascii=False, default=str))
        else:
            print(f"\n📊 Found {len(articles)} articles:")
            for i, article in enumerate(articles):
                print(f"\n{i+1}. {article.get('title', 'No title')}")
                print(f"   Source: {article.get('source', 'Unknown')}")
                print(f"   Category: {article.get('category', 'Unknown')}")
                print(f"   Relevance: {article.get('relevance_score', 0):.2%}")
                if article.get('url'):
                    print(f"   URL: {article['url']}")
    
    except Exception as e:
        print(f"❌ Search failed: {e}")
        sys.exit(1)

def run_stats(args):
    """Show database statistics."""
    print("📊 Database Statistics:")
    
    try:
        from db import NewsDatabase
        
        db = NewsDatabase()
        stats = db.get_statistics()
        
        if args.json:
            print(json.dumps(stats, indent=2, ensure_ascii=False))
        else:
            print(f"\nTotal chunks: {stats.get('total_chunks', 0)}")
            
            if stats.get('categories'):
                print(f"\nCategories:")
                for category, count in sorted(stats['categories'].items(), key=lambda x: x[1], reverse=True):
                    print(f"  {category}: {count}")
            
            if stats.get('sources'):
                print(f"\nTop 10 Sources:")
                sorted_sources = sorted(stats['sources'].items(), key=lambda x: x[1], reverse=True)[:10]
                for source, count in sorted_sources:
                    print(f"  {source}: {count}")
    
    except Exception as e:
        print(f"❌ Statistics failed: {e}")
        sys.exit(1)

def run_ui(args):
    """Launch web interface."""
    print(f"🌐 Launching web interface on port {args.port}...")
    
    try:
        import subprocess
        import sys
        
        cmd = [sys.executable, '-m', 'streamlit', 'run', 'ui.py', '--server.port', str(args.port)]
        subprocess.run(cmd)
    
    except KeyboardInterrupt:
        print("\n👋 Web interface stopped.")
    except Exception as e:
        print(f"❌ Failed to launch web interface: {e}")
        sys.exit(1)

if __name__ == "__main__":
    main()