Update app.py

app.py

@@ -237,6 +237,7 @@ from simple_salesforce import Salesforce, SalesforceLogin
 from langchain_groq import ChatGroq
 from langchain_core.prompts import ChatPromptTemplate
 from llama_index.core import StorageContext, VectorStoreIndex, SimpleDirectoryReader, Settings
+from llama_index.core import load_index_from_storage  # Added missing import
 from llama_index.embeddings.huggingface import HuggingFaceEmbedding
 
 # Configure logging
@@ -250,7 +251,7 @@ class MessageRequest(BaseModel):
 # Initialize FastAPI app
 app = FastAPI()
 
-# Allow CORS requests from any domain (not recommended for production)
+# Allow CORS requests (restrict in production)
 app.add_middleware(
     CORSMiddleware,
     allow_origins=["*"],
@@ -272,7 +273,7 @@ for var in required_env_vars:
 
 # Initialize Groq model
 GROQ_API_KEY = os.getenv("CHATGROQ_API_KEY")
-GROQ_MODEL = "llama3-8b-8192"  # Default to a valid Groq model
+GROQ_MODEL = "llama3-8b-8192"
 try:
     llm = ChatGroq(
         model_name=GROQ_MODEL,
@@ -324,7 +325,7 @@ def data_ingestion_from_directory():
             logger.warning("No documents found in PDF_DIRECTORY")
             return
         storage_context = StorageContext.from_defaults()
-        index = VectorStoreIndex.from_documents(documents)
+        index = VectorStoreIndex.from_documents(documents, storage_context=storage_context)
         index.storage_context.persist(persist_dir=PERSIST_DIR)
         logger.info("Data ingestion completed successfully")
     except Exception as e:
@@ -343,7 +344,7 @@ initialize()  # Run initialization tasks
 def handle_query(query):
     # Prepare context from chat history
     chat_context = ""
-    for past_query, response in reversed(current_chat_history[-10:]):  # Limit context to last 10 exchanges
+    for past_query, response in reversed(current_chat_history[-10:]):  # Limit to last 10 exchanges
         if past_query.strip():
             chat_context += f"User asked: '{past_query}'\nBot answered: '{response}'\n"
 
@@ -351,25 +352,27 @@ def handle_query(query):
     try:
         storage_context = StorageContext.from_defaults(persist_dir=PERSIST_DIR)
         index = load_index_from_storage(storage_context)
-        query_engine = index.as_query_engine(similarity_top_k=2)  # Retrieve top 2 relevant chunks
+        query_engine = index.as_query_engine(similarity_top_k=2)
         retrieved = query_engine.query(query)
         doc_context = retrieved.response if hasattr(retrieved, 'response') else "No relevant documents found."
+        logger.info(f"Retrieved documents for query '{query}': {doc_context[:100]}...")
     except Exception as e:
         logger.error(f"Error retrieving documents: {e}")
         doc_context = "Failed to retrieve documents."
 
-    # Construct the prompt with both chat and document context
+    # Construct the prompt with Redferns Tech focus
     prompt_template = ChatPromptTemplate.from_messages([
         ("system", """
-        You are the Clara Redfernstech chatbot. Provide accurate, professional answers in 10-15 words.
-        Use the document context and chat history to inform your response.
-        
+        You are Clara Redfernstech, a chatbot for Redferns Tech, a leader in data science, machine learning, and AI solutions.
+        Provide accurate, professional answers in 10-15 words based on the provided document context and chat history.
+        Focus on Redferns Tech's expertise in data science and AI.
+
         Document Context:
         {doc_context}
-        
+
         Chat History:
         {chat_context}
-        
+
         Question:
         {query}
         """),
@@ -386,7 +389,7 @@ def handle_query(query):
 
     # Update chat history
     if len(current_chat_history) >= MAX_HISTORY_SIZE:
-        current_chat_history.pop(0)  # Remove oldest entry
+        current_chat_history.pop(0)
     current_chat_history.append((query, response_text))
     return response_text
 
@@ -433,7 +436,7 @@ async def receive_form_data(request: Request):
     data = {
         'FirstName': first_name,
         'LastName': last_name,
-        'Description': 'hii',  # Static description
+        'Description': 'hii',
         'Company': form_data.get('company', ''),
         'Phone': form_data.get('phone', '').strip(),
         'Email': form_data.get('email', ''),
@@ -459,14 +462,25 @@ async def chat(request: MessageRequest):
         "timestamp": datetime.datetime.now().isoformat()
     }
     if len(chat_history) >= MAX_HISTORY_SIZE:
-        chat_history.pop(0)  # Remove oldest entry
+        chat_history.pop(0)
     chat_history.append(message_data)
     logger.info(f"Chat message processed: {message}")
     return {"response": response}
 
+@app.get("/health")
+async def health_check():
+    try:
+        storage_context = StorageContext.from_defaults(persist_dir=PERSIST_DIR)
+        index = load_index_from_storage(storage_context)
+        logger.info("Vector index loaded successfully")
+        return {"status": "healthy", "pdf_ingestion": "successful"}
+    except Exception as e:
+        logger.error(f"Health check failed: {e}")
+        return {"status": "unhealthy", "error": str(e)}
+
 @app.get("/")
 def read_root():
-    return {"message": "Welcome to the API"}
+    return {"message": "Welcome to the Redferns Tech Chatbot API"}
 
 def split_name(full_name):
     words = full_name.strip().split()