Update app.py

app.py

@@ -54,28 +54,20 @@ def run_ocr(uploaded_file):
         else:
             result = parse(path)
 
-        # ------ DEBUG: right after parse() ------
         debug_lines = ["========== DEBUG =========="]
         debug_lines.append(f"1. result type: {type(result).__name__}")
-        debug_lines.append(f"2. result length: {len(result) if isinstance(result, list) else 'N/A (single object)'}")
+        debug_lines.append(f"2. result length: {len(result) if isinstance(result, list) else 'N/A'}")
         if isinstance(result, list) and len(result) > 0:
             first = result[0]
             debug_lines.append(f"3. first element type: {type(first).__name__}")
-            debug_lines.append(f"4. first element dir (attrs): {[x for x in dir(first) if not x.startswith('_')][:20]}")
-            debug_lines.append(f"5. has markdown_result: {hasattr(first, 'markdown_result')}")
-            debug_lines.append(f"6. has json_result: {hasattr(first, 'json_result')}")
+            debug_lines.append(f"4. has markdown_result: {hasattr(first, 'markdown_result')}")
+            debug_lines.append(f"5. has json_result: {hasattr(first, 'json_result')}")
             if hasattr(first, "markdown_result"):
                 md = getattr(first, "markdown_result", None)
-                debug_lines.append(f"7. markdown_result type: {type(md).__name__}, len: {len(md) if md else 0}")
-                if md:
-                    debug_lines.append(f"8. markdown_result snippet (first 200 chars): {repr((md or '')[:200])}")
+                debug_lines.append(f"6. markdown_result len: {len(md) if md else 0}")
             if hasattr(first, "json_result"):
                 j = getattr(first, "json_result", None)
-                debug_lines.append(f"9. json_result type: {type(j).__name__}, len: {len(j) if j else 0}")
-                if j and len(j) > 0:
-                    debug_lines.append(f"10. json_result[0] (first page) type: {type(j[0]).__name__}, len: {len(j[0]) if isinstance(j[0], list) else 'N/A'}")
-                    if isinstance(j[0], list) and len(j[0]) > 0:
-                        debug_lines.append(f"11. first region keys: {list(j[0][0].keys()) if isinstance(j[0][0], dict) else 'not a dict'}")
+                debug_lines.append(f"7. json_result len: {len(j) if j else 0}")
         debug_lines.append("==========================")
 
         if not isinstance(result, list):
@@ -98,8 +90,15 @@ def run_ocr(uploaded_file):
                     pages_data.append(page)
             markdown = "\n\n---\n\n".join(markdown_parts)
 
-            debug_lines.append(f"12. markdown_parts count: {len(markdown_parts)}, total markdown len: {len(markdown)}")
-            debug_lines.append(f"13. pages_data count: {len(pages_data)}")
+            labels_seen = set()
+            for page in pages_data:
+                if not isinstance(page, list):
+                    continue
+                for r in page:
+                    if isinstance(r, dict):
+                        labels_seen.add(r.get("label", ""))
+            debug_lines.append(f"8. ALL labels from API: {sorted(labels_seen)}")
+            debug_lines.append(f"9. markdown_parts: {len(markdown_parts)}, pages_data: {len(pages_data)}")
 
             total_headers = 0
             total_footers = 0
@@ -181,7 +180,7 @@ def run_ocr(uploaded_file):
 
 
 with gr.Blocks(title="GLM-OCR — MaaS") as demo:
-    gr.Markdown("# 🔍 GLM-OCR — Zhipu MaaS\nUpload PDF or image. Headers 🔵 and Footers 🟢 from the API.")
+    gr.Markdown("# 🔍 GLM-OCR — Zhipu MaaS\nUpload PDF or image.")
     file_input = gr.File(label="Upload PDF or Image", file_types=[".pdf", ".png", ".jpg", ".jpeg", ".tiff", ".bmp"])
     run_btn = gr.Button("▶ Run OCR", variant="primary", size="lg")
     with gr.Row():