Replace temp PDF paths with original filenames in results

app.py

@@ -753,10 +753,13 @@ with col_input:
 
         if pdf_files:
             input_data = []
+            pdf_name_map = {}  # Map temp paths to original filenames
             for f in pdf_files:
                 with tempfile.NamedTemporaryFile(delete=False, suffix='.pdf') as tmp:
                     tmp.write(f.read())
                     input_data.append(tmp.name)
+                    pdf_name_map[tmp.name] = f.name.replace('.pdf', '')  # Store original name without extension
+            st.session_state.pdf_name_map = pdf_name_map
             description = pdf_description or "document"
             original_filename = "pdf_files"
             st.success(f"Uploaded {len(pdf_files)} PDF file(s)")
@@ -1053,6 +1056,20 @@ with col_input:
                         # Combine results
                         result_df = pd.concat(all_results, ignore_index=True)
 
+                        # For PDFs, replace temp file paths with original filenames
+                        if input_type_selected == "pdf" and 'pdf_input' in result_df.columns:
+                            pdf_name_map = st.session_state.get('pdf_name_map', {})
+                            def replace_temp_path(val):
+                                if pd.isna(val):
+                                    return val
+                                val_str = str(val)
+                                for temp_path, orig_name in pdf_name_map.items():
+                                    if temp_path in val_str:
+                                        # Replace temp path with original name, keep page suffix
+                                        return val_str.replace(temp_path, orig_name)
+                                return val_str
+                            result_df['pdf_input'] = result_df['pdf_input'].apply(replace_temp_path)
+
                         # Save CSV
                         with tempfile.NamedTemporaryFile(mode='w', suffix='_classified.csv', delete=False) as f:
                             result_df.to_csv(f.name, index=False)