Update utils/database.py

utils/database.py

@@ -20,6 +20,9 @@ import streamlit as st
 import sqlite3
 import traceback
 import time
+import io
+import tempfile
+from langchain_community.document_loaders import PyPDFLoader
 
 from sqlite3 import Error
 
@@ -142,6 +145,8 @@ def verify_vector_store(vector_store):
         st.error(f"Vector store verification failed: {e}")
         return False
 
+
+
 def handle_document_upload(uploaded_files):
     """Handle document upload with progress tracking."""
     try:
@@ -181,18 +186,28 @@ def handle_document_upload(uploaded_files):
             status_container.info(f"🔄 Processing document {idx + 1}/{len(uploaded_files)}: {file_name}")
             details_container.text(f"📄 Current file: {file_name}")
             
-            # Read file content
-            content = uploaded_file.read()
-            
-            # Store in database
-            details_container.text(f"💾 Storing {file_name} in database...")
-            doc_id = insert_document(st.session_state.db_conn, file_name, content.decode('utf-8'))
-            if not doc_id:
-                status_container.error(f"❌ Failed to store document: {file_name}")
-                continue
+            # Create a temporary file to save the PDF
+            with tempfile.NamedTemporaryFile(delete=False, suffix='.pdf') as tmp_file:
+                # Write the uploaded file content to the temporary file
+                tmp_file.write(uploaded_file.getvalue())
+                tmp_file.flush()
+                
+                # Use PyPDFLoader to load the PDF
+                loader = PyPDFLoader(tmp_file.name)
+                pdf_documents = loader.load()
+                
+                # Extract text content from the PDF
+                content = "\n".join(doc.page_content for doc in pdf_documents)
+                
+                # Store in database
+                details_container.text(f"💾 Storing {file_name} in database...")
+                doc_id = insert_document(st.session_state.db_conn, file_name, content)
+                if not doc_id:
+                    status_container.error(f"❌ Failed to store document: {file_name}")
+                    continue
                 
-            documents.append(content.decode('utf-8'))
-            document_names.append(file_name)
+                documents.append(content)
+                document_names.append(file_name)
             
             # Update progress
             current_progress += progress_per_file
@@ -257,7 +272,7 @@ def handle_document_upload(uploaded_files):
         # Clean up progress display after 5 seconds if successful
         if st.session_state.get('qa_system'):
             time.sleep(5)
-            progress_container.empty()    
+            progress_container.empty()
 
 def display_vector_store_info():
     """Display information about the current vector store state."""