Update app.py

app.py

@@ -1,5 +1,5 @@
 import streamlit as st
-from rag_components import load_documents, split_documents, create_embeddings, setup_vector_store, create_qa_chain
+from rag_components import load_documents, split_documents, create_embeddings, setup_vector_store, create_qa_chain, create_streaming_response
 import os
 
 # Ensure cache directories exist
@@ -9,10 +9,22 @@ for cache_dir in cache_dirs:
 
 st.set_page_config(
     page_title="Document Chatbot",
-    page_icon="📚",
-    layout="wide"
+    page_icon="🤖",
+    layout="wide",
+    initial_sidebar_state="collapsed"
 )
-st.title("📚 Chat with your Documents")
+
+st.title("Juma's Assistant")
+st.markdown("---")
+
+# Add some helpful information
+with st.expander("How to use this chatbot"):
+    st.markdown("""
+    - Ask questions about the content in your document
+    - The AI will search through the document to find relevant information
+    - Responses are generated in real-time with streaming
+    - Be specific in your questions for better answers
+    """)
 
 @st.cache_resource
 def initialize_rag_components(file_path="me.txt"):
@@ -36,13 +48,13 @@ def initialize_rag_components(file_path="me.txt"):
             
         with st.spinner("Initializing QA chain..."):
             qa_chain = create_qa_chain(retriever)
-            
-        st.success("✅ RAG system initialized successfully!")
+
+        st.success("Welcome! Ask me anything about Juma.")
         return qa_chain, retriever
         
     except Exception as e:
-        st.error(f"❌ Error initializing RAG components: {e}")
-        st.info("💡 This might be due to model download issues. Please try refreshing the page.")
+        st.error(f"Error: initializing: {e}")
+        st.info("This might be due to model download issues. Please try refreshing the page.")
         return None, None
 
 qa_chain, retriever = initialize_rag_components()
@@ -58,7 +70,7 @@ if qa_chain is not None:
             st.markdown(message["content"])
 
     # React to user input
-    if prompt := st.chat_input("Ask me a question about the document"):
+    if prompt := st.chat_input("Ask me any question..."):
         # Display user message in chat message container
         st.session_state.messages.append({"role": "user", "content": prompt})
         with st.chat_message("user"):
@@ -67,17 +79,14 @@ if qa_chain is not None:
         # Display assistant response in chat message container
         with st.chat_message("assistant"):
             message_placeholder = st.empty()
-            full_response = ""
+            
             try:
-                # Assuming qa_chain.stream() yields dictionaries with a 'result' key
-                for chunk in qa_chain.stream(prompt):
-                    if 'result' in chunk:
-                        full_response += chunk['result']
-                        message_placeholder.markdown(full_response + "▌")
-                message_placeholder.markdown(full_response)
+                # Use the new streaming response function
+                full_response = create_streaming_response(qa_chain, prompt, message_placeholder)
+                
             except Exception as e:
                 st.error(f"An error occurred: {e}")
-                full_response = "Sorry, I could not process your request."
+                full_response = "I apologize, but I encountered an error while processing your question. Please try again."
 
         # Add assistant response to chat history
         st.session_state.messages.append({"role": "assistant", "content": full_response})