Update gitignore

.gitignore

@@ -10,3 +10,5 @@ wheels/
 .venv
 
 .python-version
+data
+*.pkl

db.py

@@ -217,7 +217,7 @@ def query(db: Database | Path, text: str, record_count: int = 10) -> list[tuple[
     
     return results
 
-def add_document(db: Database | Path, file: Path, max_workers: int = 1) -> None:
+def add_document(db: Database | Path, file: Path, max_workers: int = 4) -> None:
     """
     Adds a new document to the database. If path is given, do load, add, save.
     Loads PDF with PyMuPDF, splits by pages, and creates records and vectors.

main.py

@@ -179,6 +179,71 @@ def query(db_path: str, query_text: str):
 
 
 
+def start_web_server(db_path: str, host: str = "127.0.0.1", port: int = 5000):
+    """Start a web server for the semantic search tool."""
+    try:
+        from flask import Flask, request, jsonify, render_template
+    except ImportError:
+        print("❌ Flask not found. Please install it first:")
+        print("   pip install flask")
+        sys.exit(1)
+    # Set template_folder to 'templates' directory
+    app = Flask(__name__, template_folder="templates")
+    db_file = Path(db_path)
+    
+    # Check if database exists
+    if not db_file.exists():
+        print(f"❌ Database file not found: {db_file}")
+        print("   Create a database first using: python main.py create")
+        sys.exit(1)
+    
+    @app.route('/')
+    def index():
+        return render_template("index.html", results=None)
+    
+    @app.route('/api/search', methods=['POST'])
+    def search():
+        try:
+            data = request.get_json()
+            if not data or 'query' not in data:
+                return jsonify({'error': 'Missing query parameter'}), 400
+            
+            query_text = data['query'].strip()
+            if not query_text:
+                return jsonify({'error': 'Query cannot be empty'}), 400
+            
+            # Perform the search
+            results = db.query(db_file, query_text)
+            
+            # Format results for JSON response
+            formatted_results = []
+            for distance, record in results:
+                formatted_results.append({
+                    'distance': float(distance),
+                    'document': record.document.name,
+                    'page': record.page,
+                    'chunk': record.chunk,
+                    'text': ' '.join(record.text[:300].split())  # Clean and truncate text
+                })
+            
+            return jsonify({'results': formatted_results})
+            
+        except Exception as e:
+            return jsonify({'error': str(e)}), 500
+    
+    print("🚀 Starting web server...")
+    print(f"   Database: {db_file}")
+    print(f"   URL: http://{host}:{port}")
+    print("   Press Ctrl+C to stop")
+    
+    try:
+        app.run(host=host, port=port, debug=False)
+    except KeyboardInterrupt:
+        print("\n👋 Web server stopped")
+    except Exception as e:
+        print(f"❌ Error starting web server: {e}")
+
+
 def main():
     parser = argparse.ArgumentParser(
         description="Semantic Search Tool",
@@ -203,6 +268,12 @@ def main():
     query_parser.add_argument('db', help='Path to the database file (e.g., db.pkl)')
     query_parser.add_argument('query_text', help='Text to search for')
     
+    # Host command (web server)
+    host_parser = subparsers.add_parser('host', aliases=['h'], help='Start a web server for semantic search')
+    host_parser.add_argument('db', help='Path to the database file (e.g., db.pkl)')
+    host_parser.add_argument('--host', default='127.0.0.1', help='Host address to bind to (default: 127.0.0.1)')
+    host_parser.add_argument('--port', type=int, default=5000, help='Port to listen on (default: 5000)')
+    
     # Test command
     subparsers.add_parser('test', aliases=['t'], help='Test database save/load functionality')
     
@@ -216,6 +287,8 @@ def main():
         add_file(args.db, args.file_paths)
     elif args.command in ['query', 'q']:
         query(args.db, args.query_text)
+    elif args.command in ['host', 'h']:
+        start_web_server(args.db, args.host, args.port)
     elif args.command in ['test', 't']:
         test_database()
     else:

templates/index.html

@@ -0,0 +1,73 @@
+<!DOCTYPE html>
+    <html>
+    <head>
+        <title>Semantic Document Search</title>
+        <style>
+            body { font-family: Arial, sans-serif; max-width: 1200px; margin: 0 auto; padding: 20px; }
+            .search-box { margin-bottom: 20px; }
+            input[type="text"] { width: 70%; padding: 10px; font-size: 16px; }
+            button { padding: 10px 20px; font-size: 16px; background: #007cba; color: white; border: none; cursor: pointer; }
+            button:hover { background: #005c8a; }
+            .result { border: 1px solid #ddd; margin: 10px 0; padding: 15px; border-radius: 5px; }
+            .result-header { font-weight: bold; color: #333; margin-bottom: 10px; }
+            .result-text { background: #f9f9f9; padding: 10px; border-radius: 3px; }
+            .distance { color: #666; font-size: 0.9em; }
+            .no-results { text-align: center; color: #666; margin: 40px 0; }
+            .loading { text-align: center; color: #007cba; margin: 20px 0; }
+        </style>
+    </head>
+    <body>
+        <h1>🔍 Semantic Document Search</h1>
+        <div class="search-box">
+            <form id="searchForm">
+                <input type="text" id="queryInput" placeholder="Enter your search query..." required>
+                <button type="submit">Search</button>
+            </form>
+        </div>
+        <div id="results"></div>
+        
+        <script>
+            document.getElementById('searchForm').addEventListener('submit', async (e) => {
+                e.preventDefault();
+                const query = document.getElementById('queryInput').value;
+                const resultsDiv = document.getElementById('results');
+                
+                resultsDiv.innerHTML = '<div class="loading">Searching...</div>';
+                
+                try {
+                    const response = await fetch('/api/search', {
+                        method: 'POST',
+                        headers: { 'Content-Type': 'application/json' },
+                        body: JSON.stringify({ query: query })
+                    });
+                    
+                    const data = await response.json();
+                    
+                    if (data.error) {
+                        resultsDiv.innerHTML = `<div class="no-results">Error: ${data.error}</div>`;
+                        return;
+                    }
+                    
+                    if (data.results.length === 0) {
+                        resultsDiv.innerHTML = '<div class="no-results">No results found.</div>';
+                        return;
+                    }
+                    
+                    resultsDiv.innerHTML = data.results.map((result, i) => `
+                        <div class="result">
+                            <div class="result-header">
+                                Result ${i + 1} - ${result.document} 
+                                <span class="distance">(Distance: ${result.distance.toFixed(4)})</span>
+                            </div>
+                            <div>Page: ${result.page}, Chunk: ${result.chunk}</div>
+                            <div class="result-text">${result.text}</div>
+                        </div>
+                    `).join('');
+                    
+                } catch (error) {
+                    resultsDiv.innerHTML = `<div class="no-results">Error: ${error.message}</div>`;
+                }
+            });
+        </script>
+    </body>
+    </html>