Fix PDF progress: show document-based progress since catllm processes pages internally

app.py

@@ -981,16 +981,13 @@ with col_input:
                     model_source = get_model_source(model)
                     items_list = input_data if isinstance(input_data, list) else [input_data]
 
-                    # For PDFs, count total pages; for others, count items
+                    # Progress tracking
+                    total_items = len(items_list)
+
+                    # For PDFs, also get page counts for display
                     if input_type_selected == "pdf":
-                        # Build list of (pdf_path, page_count) and calculate total pages
                         pdf_page_counts = [(pdf_path, count_pdf_pages(pdf_path)) for pdf_path in items_list]
                         total_pages = sum(pc for _, pc in pdf_page_counts)
-                        progress_unit = "page"
-                        total_units = total_pages
-                    else:
-                        total_units = len(items_list)
-                        progress_unit = "item"
 
                     # Progress UI
                     progress_bar = st.progress(0)
@@ -998,33 +995,27 @@ with col_input:
 
                     all_results = []
                     start_time = time.time()
-                    processed_units = 0
 
                     for i, item in enumerate(items_list):
-                        # For PDFs, get page count for this document
-                        if input_type_selected == "pdf":
-                            item_pages = pdf_page_counts[i][1]
-                        else:
-                            item_pages = 1
-
                         # Update progress before processing
-                        progress = processed_units / total_units if total_units > 0 else 0
+                        progress = i / total_items if total_items > 0 else 0
                         progress_bar.progress(min(progress, 1.0))
 
                         # Calculate ETA
                         elapsed = time.time() - start_time
-                        if processed_units > 0:
-                            avg_time_per_unit = elapsed / processed_units
-                            remaining_units = total_units - processed_units
-                            eta_seconds = avg_time_per_unit * remaining_units
+                        if i > 0:
+                            avg_time_per_item = elapsed / i
+                            remaining_items = total_items - i
+                            eta_seconds = avg_time_per_item * remaining_items
                             eta_str = f" | ETA: {eta_seconds:.0f}s" if eta_seconds < 60 else f" | ETA: {eta_seconds/60:.1f}m"
                         else:
                             eta_str = ""
 
                         if input_type_selected == "pdf":
-                            status_text.text(f"Processing page {processed_units + 1} of {total_units} ({progress*100:.0f}%){eta_str}")
+                            doc_pages = pdf_page_counts[i][1]
+                            status_text.text(f"Processing document {i+1} of {total_items} ({doc_pages} pages) ({progress*100:.0f}%){eta_str}")
                         else:
-                            status_text.text(f"Processing {progress_unit} {processed_units + 1} of {total_units} ({progress*100:.0f}%){eta_str}")
+                            status_text.text(f"Processing item {i+1} of {total_items} ({progress*100:.0f}%){eta_str}")
 
                         try:
                             classify_kwargs = {
@@ -1042,25 +1033,21 @@ with col_input:
                             item_result = catllm.classify(**classify_kwargs)
                             all_results.append(item_result)
 
-                            # Update processed count (pages for PDFs, 1 for others)
-                            processed_units += item_pages
-
                             # Update progress after processing
-                            progress = processed_units / total_units if total_units > 0 else 1.0
+                            progress = (i + 1) / total_items if total_items > 0 else 1.0
                             progress_bar.progress(min(progress, 1.0))
 
                         except Exception as e:
-                            st.warning(f"Error on {progress_unit} {i+1}: {str(e)}")
-                            processed_units += item_pages  # Still count as processed
+                            st.warning(f"Error on item {i+1}: {str(e)}")
                             continue
 
                     # Complete progress
                     progress_bar.progress(1.0)
                     processing_time = time.time() - start_time
                     if input_type_selected == "pdf":
-                        status_text.text(f"Completed {total_pages} pages in {processing_time:.1f}s")
+                        status_text.text(f"Completed {total_items} document(s) ({total_pages} pages) in {processing_time:.1f}s")
                     else:
-                        status_text.text(f"Completed {total_units} items in {processing_time:.1f}s")
+                        status_text.text(f"Completed {total_items} items in {processing_time:.1f}s")
 
                     if all_results:
                         # Combine results