Using pdf workaround

src/streamlit_app.py

@@ -4,6 +4,8 @@ import io
 import base64
 from datetime import datetime
 import json
+import tempfile
+import os
 
 # Page configuration
 st.set_page_config(
@@ -19,7 +21,7 @@ if 'uploaded_documents' not in st.session_state:
 if 'current_user' not in st.session_state:
     st.session_state.current_user = "User"
 
-def save_document_info(filename, file_content, file_type):
+def save_document_info(filename, file_content, file_type, temp_path=None):
     """Save document information to session state"""
     if 'documents' not in st.session_state.uploaded_documents:
         st.session_state.uploaded_documents['documents'] = []
@@ -29,15 +31,29 @@ def save_document_info(filename, file_content, file_type):
         'upload_time': datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
         'file_type': file_type,
         'size': len(file_content),
-        'content': file_content.decode('latin-1') if isinstance(file_content, bytes) else str(file_content)
+        'content': file_content.decode('latin-1') if isinstance(file_content, bytes) else str(file_content),
+        'temp_path': temp_path  # Store temp path for later use
     }
     
     st.session_state.uploaded_documents['documents'].append(document_info)
 
-def extract_pdf_text(pdf_file):
-    """Extract text from PDF file"""
+def extract_pdf_text_from_temp(temp_path):
+    """Extract text from PDF file using temporary file path"""
     try:
-        pdf_reader = PyPDF2.PdfReader(pdf_file)
+        with open(temp_path, "rb") as pdf_file:
+            pdf_reader = PyPDF2.PdfReader(pdf_file)
+            text = ""
+            for page in pdf_reader.pages:
+                text += page.extract_text() + "\n"
+        return text
+    except Exception as e:
+        st.error(f"Error reading PDF: {str(e)}")
+        return ""
+
+def extract_pdf_text_from_memory(uploaded_file):
+    """Extract text from PDF file in memory"""
+    try:
+        pdf_reader = PyPDF2.PdfReader(uploaded_file)
         text = ""
         for page in pdf_reader.pages:
             text += page.extract_text() + "\n"
@@ -46,6 +62,14 @@ def extract_pdf_text(pdf_file):
         st.error(f"Error reading PDF: {str(e)}")
         return ""
 
+def cleanup_temp_file(temp_path):
+    """Clean up temporary file"""
+    try:
+        if temp_path and os.path.exists(temp_path):
+            os.remove(temp_path)
+    except Exception as e:
+        st.warning(f"Could not clean up temporary file: {str(e)}")
+
 def main():
     # Sidebar for navigation
     with st.sidebar:
@@ -124,7 +148,7 @@ def show_upload_documents():
     st.markdown("---")
     
     # Add information about file upload
-    st.info("💡 **Note:** File upload is optimized for Hugging Face Spaces. If you experience issues, try refreshing the page.")
+    st.info("💡 **Note:** File upload uses temporary storage for better compatibility with Hugging Face Spaces.")
     
     uploaded_file = st.file_uploader(
         "Choose a PDF file",
@@ -146,38 +170,53 @@ def show_upload_documents():
             for key, value in file_details.items():
                 st.write(f"- {key}: {value}")
             
-            # Extract and display PDF content
-            pdf_text = extract_pdf_text(uploaded_file)
-            
-            if pdf_text.strip():
-                st.subheader("📄 Document Preview")
-                with st.expander("View extracted text"):
-                    st.text_area("PDF Content", pdf_text, height=300)
-            else:
-                st.warning("⚠️ Could not extract text from this PDF. The file may be image-based or encrypted.")
-            
-            # Upload button
-            if st.button("Upload Document", type="primary"):
-                try:
-                    # Reset file pointer to beginning
-                    uploaded_file.seek(0)
-                    
-                    # Save document info
-                    save_document_info(
-                        uploaded_file.name,
-                        uploaded_file.read(),
-                        "PDF"
-                    )
-                    
-                    st.success(f"✅ Document '{uploaded_file.name}' uploaded successfully!")
-                    st.balloons()
-                    
-                    # Clear the file uploader
-                    st.rerun()
-                    
-                except Exception as e:
-                    st.error(f"❌ Error uploading document: {str(e)}")
-                    st.info("💡 Try uploading a smaller file or refresh the page.")
+            # Create temporary file for better PDF processing
+            temp_path = None
+            try:
+                with tempfile.NamedTemporaryFile(mode="wb", suffix=".pdf", delete=False) as temp:
+                    bytes_data = uploaded_file.getvalue()
+                    temp.write(bytes_data)
+                    temp_path = temp.name
+                
+                st.success(f"📁 File temporarily stored at: {temp_path}")
+                
+                # Extract and display PDF content using temporary file
+                pdf_text = extract_pdf_text_from_temp(temp_path)
+                
+                if pdf_text.strip():
+                    st.subheader("📄 Document Preview")
+                    with st.expander("View extracted text"):
+                        st.text_area("PDF Content", pdf_text, height=300)
+                else:
+                    st.warning("⚠️ Could not extract text from this PDF. The file may be image-based or encrypted.")
+                
+                # Upload button
+                if st.button("Upload Document", type="primary"):
+                    try:
+                        # Save document info with temporary file path
+                        save_document_info(
+                            uploaded_file.name,
+                            bytes_data,
+                            "PDF",
+                            temp_path
+                        )
+                        
+                        st.success(f"✅ Document '{uploaded_file.name}' uploaded successfully!")
+                        st.balloons()
+                        
+                        # Clear the file uploader
+                        st.rerun()
+                        
+                    except Exception as e:
+                        st.error(f"❌ Error uploading document: {str(e)}")
+                        st.info("💡 Try uploading a smaller file or refresh the page.")
+                        # Clean up temp file on error
+                        cleanup_temp_file(temp_path)
+                        
+            except Exception as e:
+                st.error(f"❌ Error creating temporary file: {str(e)}")
+                st.info("💡 Please try uploading a different PDF file.")
+                cleanup_temp_file(temp_path)
                     
         except Exception as e:
             st.error(f"❌ Error processing file: {str(e)}")
@@ -192,6 +231,11 @@ def show_upload_documents():
         - Avoid password-protected PDFs
         - If upload fails, try refreshing the page
         
+        **Technical details:**
+        - Files are temporarily stored on the server
+        - Text extraction uses temporary file processing
+        - Automatic cleanup of temporary files
+        
         **Supported formats:** PDF only
         """)
 
@@ -242,9 +286,26 @@ def show_my_documents():
                     st.write(f"**Uploaded:** {doc['upload_time']}")
                     st.write(f"**Size:** {doc['size']} bytes")
                     
-                    # Display content if available
-                    if 'content' in doc and doc['content']:
-                        st.text_area("Document Content", doc['content'], height=400, key=f"content_{i}")
+                    # Check if we have a temporary file path for better content extraction
+                    if doc.get('temp_path') and os.path.exists(doc['temp_path']):
+                        try:
+                            # Extract fresh content from temporary file
+                            fresh_content = extract_pdf_text_from_temp(doc['temp_path'])
+                            if fresh_content.strip():
+                                st.text_area("Document Content (Fresh Extract)", fresh_content, height=400, key=f"fresh_content_{i}")
+                            else:
+                                # Fall back to stored content
+                                if 'content' in doc and doc['content']:
+                                    st.text_area("Document Content (Stored)", doc['content'], height=400, key=f"content_{i}")
+                        except Exception as e:
+                            st.warning(f"Could not read from temporary file: {str(e)}")
+                            # Fall back to stored content
+                            if 'content' in doc and doc['content']:
+                                st.text_area("Document Content (Stored)", doc['content'], height=400, key=f"content_{i}")
+                    else:
+                        # Display stored content
+                        if 'content' in doc and doc['content']:
+                            st.text_area("Document Content", doc['content'], height=400, key=f"content_{i}")
             
             st.markdown("---")
 
@@ -295,9 +356,26 @@ def show_document_library():
                     st.write(f"**Uploaded:** {doc['upload_time']}")
                     st.write(f"**Size:** {doc['size']} bytes")
                     
-                    # Display content if available
-                    if 'content' in doc and doc['content']:
-                        st.text_area("Document Content", doc['content'], height=400, key=f"lib_content_{i}")
+                    # Check if we have a temporary file path for better content extraction
+                    if doc.get('temp_path') and os.path.exists(doc['temp_path']):
+                        try:
+                            # Extract fresh content from temporary file
+                            fresh_content = extract_pdf_text_from_temp(doc['temp_path'])
+                            if fresh_content.strip():
+                                st.text_area("Document Content (Fresh Extract)", fresh_content, height=400, key=f"lib_fresh_content_{i}")
+                            else:
+                                # Fall back to stored content
+                                if 'content' in doc and doc['content']:
+                                    st.text_area("Document Content (Stored)", doc['content'], height=400, key=f"lib_content_{i}")
+                        except Exception as e:
+                            st.warning(f"Could not read from temporary file: {str(e)}")
+                            # Fall back to stored content
+                            if 'content' in doc and doc['content']:
+                                st.text_area("Document Content (Stored)", doc['content'], height=400, key=f"lib_content_{i}")
+                    else:
+                        # Display stored content
+                        if 'content' in doc and doc['content']:
+                            st.text_area("Document Content", doc['content'], height=400, key=f"lib_content_{i}")
             
             st.markdown("---")
 
@@ -309,7 +387,7 @@ def show_settings():
     st.subheader("🔧 System Information")
     st.write("**Version:** Dev LMS v1.0")
     st.write("**Features:**")
-    st.write("- PDF document upload")
+    st.write("- PDF document upload with temporary storage")
     st.write("- Document search and preview")
     st.write("- Document library")
     st.write("- Session-based storage")
@@ -340,11 +418,48 @@ def show_settings():
     # Clear data option
     if st.button("🗑️ Clear All Data"):
         if st.session_state.uploaded_documents.get('documents'):
+            # Clean up temporary files before clearing data
+            documents = st.session_state.uploaded_documents['documents']
+            for doc in documents:
+                if doc.get('temp_path'):
+                    cleanup_temp_file(doc['temp_path'])
+            
             st.session_state.uploaded_documents['documents'] = []
-            st.success("All documents have been cleared!")
+            st.success("All documents and temporary files have been cleared!")
             st.rerun()
         else:
             st.info("No documents to clear.")
+    
+    st.markdown("---")
+    
+    # Cleanup temporary files option
+    if st.button("🧹 Cleanup Temporary Files"):
+        documents = st.session_state.uploaded_documents.get('documents', [])
+        cleaned_count = 0
+        
+        for doc in documents:
+            if doc.get('temp_path') and not os.path.exists(doc['temp_path']):
+                # Remove temp_path reference if file doesn't exist
+                doc.pop('temp_path', None)
+                cleaned_count += 1
+        
+        if cleaned_count > 0:
+            st.success(f"Cleaned up {cleaned_count} missing temporary file references!")
+        else:
+            st.info("No cleanup needed - all temporary files are properly managed.")
+    
+    st.markdown("---")
+    
+    # System status
+    st.subheader("📊 System Status")
+    documents = st.session_state.uploaded_documents.get('documents', [])
+    temp_files_count = sum(1 for doc in documents if doc.get('temp_path') and os.path.exists(doc['temp_path']))
+    
+    col1, col2 = st.columns(2)
+    with col1:
+        st.metric("Total Documents", len(documents))
+    with col2:
+        st.metric("Active Temp Files", temp_files_count)
 
 if __name__ == "__main__":
     main()