Upd logger memory

memory.py

@@ -6,10 +6,12 @@ from collections import defaultdict, deque
 from typing import List
 from sentence_transformers import SentenceTransformer
 from google import genai  # must be configured in app.py and imported globally
+import logging
 
 _LLM = "gemini-2.5-flash-lite-preview-06-17" # Small model for NLP simple tasks
 # Load embedding model
 embedding_model = SentenceTransformer("/app/model_cache", device="cpu").half()
+logger = logging.getLogger("medical-chatbot")
 
 class MemoryManager:
     def __init__(self, max_users=1000, history_per_user=10):
@@ -24,7 +26,7 @@ class MemoryManager:
                 oldest = self.user_queue.popleft()
                 self._drop_user(oldest)
             self.user_queue.append(user_id)
-
+        # Normalize
         self.text_cache[user_id].append((query.strip(), response.strip()))
         # Use Gemini to summarize and chunk smartly
         chunks = self.chunk_response(response, lang)
@@ -90,6 +92,7 @@ class MemoryManager:
                 generation_config={"temperature": 0.4}
             )
             output = result.text.strip()
+            logger.info(f"Reasoned RAG result: {output}")
             return [chunk.strip() for chunk in output.split('---') if chunk.strip()]
         except Exception as e:
             print(f"❌ Gemini chunking failed: {e}")