Update app.py

app.py

@@ -570,21 +570,32 @@ def qa_llm():
     logging.info("QA model setup complete")
     return qa
 
-def process_answer(user_question):
-    """Generate an answer to the user’s question."""
+def process_answer(user_question, full_text):
+    """Generate an answer to the user’s question or summarize the PDF content."""
     try:
         logging.info("Processing user question")
-        qa = qa_llm() 
-
-        tailored_prompt = f"""
-        You are an expert chatbot designed to assist with any topic, providing accurate and detailed answers based on the provided PDFs. 
-        Your goal is to deliver the most relevant information and resources based on the question asked. 
-        User question: {user_question}
-        """
-
+        
+        # Check if the question is related to summarization
+        if "summarize" in user_question.lower() or "summary" in user_question.lower():
+            tailored_prompt = f"""
+            Please provide a summary of the following content extracted from the PDF:
+            {full_text}
+            """
+        else:
+            # Regular Q&A with context from the uploaded PDF
+            tailored_prompt = f"""
+            You are an expert chatbot designed to assist with any topic, providing accurate and detailed answers based on the provided PDFs. 
+            Your goal is to deliver the most relevant information and resources based on the question asked. 
+            User question: {user_question}
+            Content from the uploaded document: {full_text}
+            """
+
+        # Pass the tailored prompt to the question-answering chain (QA) system
+        qa = qa_llm()  # Call your QA LLM setup
         generated_text = qa({"query": tailored_prompt})
         answer = generated_text['result']
 
+        # If the answer contains certain fallback phrases, return a default message
         if "not provide" in answer or "no information" in answer:
             return "The document does not provide sufficient information to answer your question."
 
@@ -595,6 +606,7 @@ def process_answer(user_question):
         logging.error(f"Error during answer generation: {str(e)}")
         return "Error processing the question."
 
+
 # Streamlit UI Setup
 st.sidebar.header("File Upload")
 uploaded_files = st.sidebar.file_uploader("Upload your PDF files", type=["pdf"], accept_multiple_files=True)
@@ -633,9 +645,9 @@ if uploaded_files:
         else:
             st.warning("Failed to extract text from this PDF.")
 
-    # Generate summary option
-    if st.button("Generate Summary of Document"):
-        st.write("Summary: [Provide the generated summary here]")
+    # # Generate summary option
+    # if st.button("Generate Summary of Document"):
+    #     st.write("Summary: [Provide the generated summary here]")
     
     # Run data ingestion when files are uploaded
     data_ingestion()