Overview
The app.py module is the main entry point for the DeenPAL Streamlit web application. It provides an interactive chat interface for users to ask questions about Islamic hadiths and receive AI-generated responses.
Application Structure
Entry Point
Run the application using Streamlit:
Session State Variables
Streamlit uses session state to persist data across reruns.
Stores the complete chat history for the current session. Each message is a dictionary with role and content keys.
Message Structure
{
"role": "user" | "assistant",
"content": "Message text"
}
Initialization
if "messages" not in st.session_state:
st.session_state.messages = []
UI Components
Page Title
st.title("Deen Pal Chatbot")
Chat History Display
for message in st.session_state.messages:
with st.chat_message(message["role"]):
st.markdown(message["content"])
if prompt := st.chat_input("Please type your question"):
# Handle user input
:=) for assignment and condition checking.
Message Handling Flow
with st.chat_message("user"):
st.markdown(prompt)
2. Store User Message
st.session_state.messages.append({"role": "user", "content": prompt})
3. RAG Chain Invocation
response = rag_chain.invoke({
"input": prompt,
"chat_history": st.session_state.messages
})
The user’s current question.
Complete conversation history including previous questions and answers.
The generated response from the LLM containing hadith citations and explanations.
Retrieved document chunks used to generate the answer (not displayed in UI).
4. Assistant Response Display
with st.chat_message("assistant"):
st.markdown(response["answer"])
5. Store Assistant Response
st.session_state.messages.append({
"role": "assistant",
"content": response["answer"]
})
Complete Application Code
import streamlit as st
from chains import rag_chain
# Streamlit App UI
st.title("Deen Pal Chatbot")
# Chat History State
if "messages" not in st.session_state:
st.session_state.messages = []
# Display Chat History
for message in st.session_state.messages:
with st.chat_message(message["role"]):
st.markdown(message["content"])
# Accept User Input
if prompt := st.chat_input("Please type your question"):
with st.chat_message("user"):
st.markdown(prompt)
# Store User Query
st.session_state.messages.append({"role": "user", "content": prompt})
# Perform Retrieval and Generate Answer
response = rag_chain.invoke({"input": prompt, "chat_history": st.session_state.messages})
with st.chat_message("assistant"):
st.markdown(response["answer"])
# Store Assistant Response
st.session_state.messages.append({"role": "assistant", "content": response["answer"]})
Usage Example
Starting the Application
- Install dependencies:
pip install streamlit langchain langchain-openai langchain-community langchain-chroma langchain-huggingface
- Set up environment variables:
echo "OPENAI_API_KEY=your_openrouter_api_key" > .env
- Run the application:
User Interaction Flow
- User opens the web application in browser
- User types a question: “What does Islam say about prayer?”
- Application displays the question immediately
- RAG chain retrieves relevant hadiths and generates response
- Application displays the AI-generated answer with citations
- Conversation continues with full context maintained
Dependencies
streamlit - Web application frameworkchains.rag_chain - RAG pipeline for question answering
Features
Persistent Chat History - Maintains conversation context throughout the session
Real-time Response - Displays messages instantly as they’re processed
Context-Aware - Sends full chat history to RAG chain for better responses
Clean UI - Uses Streamlit’s native chat components for professional appearance
Configuration Options
Customizing the UI
You can modify the appearance by adding Streamlit configuration:
st.set_page_config(
page_title="DeenPAL - Islamic Q&A",
page_icon="📿",
layout="wide"
)
with st.sidebar:
st.header("About DeenPAL")
st.write("Ask questions about Islamic hadiths from Sahih Al-Bukhari and Sahih Muslim.")
Chat History Management
Add a button to clear chat history:
if st.sidebar.button("Clear Chat History"):
st.session_state.messages = []
st.rerun()
The load_and_prepare_data() function in chains.py is cached with @st.cache_resource, ensuring the vector database is only initialized once per session, significantly improving performance.
Error Handling
For production use, consider adding error handling:
try:
response = rag_chain.invoke({
"input": prompt,
"chat_history": st.session_state.messages
})
with st.chat_message("assistant"):
st.markdown(response["answer"])
except Exception as e:
st.error(f"An error occurred: {str(e)}")
st.info("Please try asking your question again.")
Session Lifecycle
- First Visit: Session state is initialized with empty messages list
- User Interaction: Messages are appended to session state
- Page Refresh: Session state persists within the same browser session
- Browser Close: Session state is cleared
- New Tab: Creates a new independent session
