Update app.py

app.py

@@ -25,6 +25,13 @@ logger = logging.getLogger(__name__)
 # Load environment variables
 load_dotenv()
 
+# These are your "customers". You give them a key.
+# In a real app, this would be in a database.
+VALID_API_KEYS = {
+    "anak_mail_123": "Dr. Amelia Ritahani",
+    "irfan_test_key_456": "Irfan (self)",
+    "sistem_gelap": "Gelap"
+}
 # --- 1. NEW HELPER FUNCTIONS TO FIX 'TypeError' ---
 def hydrate_history(raw_history_list: list) -> list:
     """Converts a list of dicts from session back into LangChain Message objects."""
@@ -371,13 +378,46 @@ def reset_metrics(domain):
         return jsonify({"success": True, "message": f"Metrics reset for {domain}"})
     except Exception as e:
         return jsonify({"error": str(e)}), 500
-
+        
+# Helper function to check API key
+def check_api_key(request_data):
+    api_key = request_data.get("api_key")
+    if not api_key or api_key not in VALID_API_KEYS:
+        return False, {"error": "Invalid or missing API key"}, 401
+    logger.info(f"API request authenticated for user: {VALID_API_KEYS[api_key]}")
+    return True, None, None
+
+# Helper function to save and process uploaded files (Base64)
+def process_base64_file(base64_string, file_type):
+    try:
+        # Decode the base64 string
+        file_bytes = base64.b64decode(base64_string)
+        
+        # Save to a temporary file
+        upload_dir = "Uploads"
+        os.makedirs(upload_dir, exist_ok=True)
+        # Use a unique filename
+        temp_filename = f"{file_type}_{int(time.time())}.tmp"
+        temp_path = os.path.join(upload_dir, temp_filename)
+        
+        with open(temp_path, 'wb') as f:
+            f.write(file_bytes)
+            
+        logger.info(f"Saved temporary {file_type} to {temp_path}")
+        return temp_path
+    except Exception as e:
+        logger.error(f"Error decoding/saving base64 file: {e}")
+        return None
 # --- 3. NEW API-ONLY ROUTES ---
 
 @app.route("/api/medical", methods=["POST"])
 def medical_api():
     try:
         data = request.json
+        is_valid, error_response, status_code = check_api_key(data)
+        if not is_valid:
+            return jsonify(error_response), status_code
+            
         query = data.get("query")
         if not query:
             return jsonify({"error": "No query provided"}), 400
@@ -390,8 +430,43 @@ def medical_api():
         if not agent:
             return jsonify({"error": "Medical RAG system not loaded"}), 500
             
+        # --- Handle File Uploads (Base64) ---
+        enhanced_query = query
+        temp_file_path = None
+        
+        if data.get("document_base64"):
+            logger.info("API: Processing base64 document for Swarm")
+            doc_text = base64.b64decode(data.get("document_base64")).decode('utf-8')
+            swarm_answer = run_medical_swarm(doc_text, query)
+            response_dict = {
+                "answer": markdown_bold_to_html(swarm_answer),
+                "thoughts": "Swarm analysis complete.",
+                "validation": (True, "Swarm output generated."),
+                "source": "Medical Swarm",
+                "response_time": 0 # Not tracked for swarm in this path
+            }
+            return jsonify(response_dict)
+            
+        elif data.get("image_base64"):
+            logger.info("API: Processing base64 image")
+            temp_file_path = process_base64_file(data.get("image_base64"), "image")
+            if not temp_file_path:
+                return jsonify({"error": "Invalid base64 image data"}), 400
+            
+            with open(temp_file_path, "rb") as img_file:
+                img_data = base64.b64encode(img_file.read()).decode("utf-8")
+            
+            vision_prompt = f"Analyze image. Query: '{query}'"
+            message = HumanMessage(content=[{"type": "text", "text": vision_prompt}, {"type": "image_url", "image_url": f"data:image/jpeg;base64,{img_data}"}])
+            visual_prediction = llm.invoke([message]).content
+            enhanced_query = (f'User Query: "{query}" Context from Image: "{visual_prediction}"')
+
         # Run the agent
-        response_dict = agent.answer(query, chat_history=history_for_agent)
+        response_dict = agent.answer(enhanced_query, chat_history=history_for_agent)
+
+        # Clean up temp file
+        if temp_file_path and os.path.exists(temp_file_path):
+            os.remove(temp_file_path)
         
         # Return the full, clean JSON response
         return jsonify(response_dict)
@@ -400,7 +475,73 @@ def medical_api():
         logger.error(f"Error on /api/medical: {e}", exc_info=True)
         return jsonify({"error": str(e)}), 500
 
+@app.route("/api/islamic", methods=["POST"])
+def islamic_api():
+    try:
+        data = request.json
+        is_valid, error_response, status_code = check_api_key(data)
+        if not is_valid: return jsonify(error_response), status_code
+        
+        query = data.get("query")
+        if not query: return jsonify({"error": "No query provided"}), 400
+        
+        raw_history = data.get("history", [])
+        history_for_agent = hydrate_history(raw_history)
+        
+        agent = rag_systems['islamic']
+        if not agent: return jsonify({"error": "Islamic RAG system not loaded"}), 500
+            
+        enhanced_query = query
+        temp_file_path = None
+        
+        if data.get("image_base64"):
+            logger.info("API: Processing base64 image")
+            temp_file_path = process_base64_file(data.get("image_base64"), "image")
+            if not temp_file_path:
+                return jsonify({"error": "Invalid base64 image data"}), 400
+            
+            with open(temp_file_path, "rb") as img_file:
+                img_data = base64.b64encode(img_file.read()).decode("utf-8")
+            
+            vision_prompt = f"Analyze image. Query: '{query}'"
+            message = HumanMessage(content=[{"type": "text", "text": vision_prompt}, {"type": "image_url", "image_url": f"data:image/jpeg;base64,{img_data}"}])
+            visual_prediction = llm.invoke([message]).content
+            enhanced_query = (f'User Query: "{query}" Context from Image: "{visual_prediction}"')
 
+        response_dict = agent.answer(enhanced_query, chat_history=history_for_agent)
+        
+        if temp_file_path and os.path.exists(temp_file_path):
+            os.remove(temp_file_path)
+            
+        return jsonify(response_dict)
+        
+    except Exception as e:
+        logger.error(f"Error on /api/islamic: {e}", exc_info=True)
+        return jsonify({"error": str(e)}), 500
+
+@app.route("/api/insurance", methods=["POST"])
+def insurance_api():
+    try:
+        data = request.json
+        is_valid, error_response, status_code = check_api_key(data)
+        if not is_valid: return jsonify(error_response), status_code
+        
+        query = data.get("query")
+        if not query: return jsonify({"error": "No query provided"}), 400
+        
+        raw_history = data.get("history", [])
+        history_for_agent = hydrate_history(raw_history)
+        
+        agent = rag_systems['insurance']
+        if not agent: return jsonify({"error": "Insurance RAG system not loaded"}), 500
+        
+        response_dict = agent.answer(query, chat_history=history_for_agent)
+        return jsonify(response_dict)
+        
+    except Exception as e:
+        logger.error(f"Error on /api/insurance: {e}", exc_info=True)
+        return jsonify({"error": str(e)}), 500
+        
 if __name__ == "__main__":
     logger.info("Starting Flask app for deployment testing...")
     app.run(host="0.0.0.0", port=7860, debug=False)