Update app.py

app.py

@@ -8,7 +8,6 @@ GLMOCR_BASE = os.path.dirname(glmocr.__file__)
 config_path = os.path.join(GLMOCR_BASE, "config.yaml")
 formatter_path = os.path.join(GLMOCR_BASE, "postprocess", "result_formatter.py")
 
-# ── Config: MaaS on, keep header/footer ────────────────────────────────────
 with open(config_path, "r") as f:
     config = yaml.safe_load(f)
 config["pipeline"]["maas"]["enabled"] = True
@@ -20,7 +19,6 @@ config["pipeline"]["enable_layout"] = True
 with open(config_path, "w") as f:
     yaml.dump(config, f, default_flow_style=False, sort_keys=False)
 
-# ── Formatter: don't strip header/footer ───────────────────────────────────
 with open(formatter_path, "r") as f:
     source = f.read()
 for label in ['"header"', "'header'", '"footer"', "'footer'", '"doc_header"', "'doc_header'", '"doc_footer"', "'doc_footer'"]:
@@ -32,32 +30,6 @@ with open(formatter_path, "w") as f:
 
 ABANDON = set(config["pipeline"]["result_formatter"]["abandon"])
 
-def get_regions(page_item):
-    if page_item is None:
-        return []
-    if isinstance(page_item, list):
-        return page_item
-    if isinstance(page_item, dict):
-        for key in ("regions", "blocks", "items", "content_list"):
-            if key in page_item and isinstance(page_item[key], list):
-                return page_item[key]
-        if "content" in page_item or "text" in page_item:
-            return [page_item]
-    if isinstance(page_item, str):
-        return [{"label": "text", "content": page_item}]
-    return []
-
-def get_label(r):
-    if not isinstance(r, dict):
-        return "text"
-    return (r.get("label") or r.get("type") or r.get("block_type") or "text")
-
-def get_content(r):
-    if isinstance(r, str):
-        return r
-    if not isinstance(r, dict):
-        return str(r)
-    return str(r.get("content") or r.get("text") or r.get("raw") or "")
 
 def run_ocr(uploaded_file):
     if uploaded_file is None:
@@ -81,22 +53,102 @@ def run_ocr(uploaded_file):
             result = parse(page_images)
         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)'}")
+        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')}")
+            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])}")
+            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("==========================")
+
+        if not isinstance(result, list):
+            result = [result]
+        if not result:
+            return "(No content)", "\n".join(debug_lines)
+
+        first = result[0]
+        has_markdown = getattr(first, "markdown_result", None) is not None
+        has_json = getattr(first, "json_result", None) is not None
+
+        if has_markdown or has_json:
+            markdown_parts = []
+            pages_data = []
+            for r in result:
+                md = getattr(r, "markdown_result", None) or ""
+                markdown_parts.append(md.strip() if md else "(empty page)")
+                j = getattr(r, "json_result", None) or []
+                for page in j:
+                    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)}")
+
+            total_headers = 0
+            total_footers = 0
+            all_summary = []
+            for page_num, page_regions in enumerate(pages_data):
+                if not isinstance(page_regions, list):
+                    page_regions = [page_regions] if page_regions else []
+                page_summary = [f"── PAGE {page_num + 1} of {len(pages_data)} ──"]
+                for region in page_regions:
+                    if not isinstance(region, dict):
+                        continue
+                    label = region.get("label", region.get("type", "text"))
+                    content = str(region.get("content", region.get("text", "")))
+                    if label in ABANDON:
+                        continue
+                    if label == "header":
+                        total_headers += 1
+                        page_summary.append(f"🔵 HEADER: {content[:100]}")
+                    elif label == "footer":
+                        total_footers += 1
+                        page_summary.append(f"🟢 FOOTER: {content[:100]}")
+                    else:
+                        page_summary.append(f"[{label}]: {content[:100]}")
+                all_summary.extend(page_summary)
+                all_summary.append("")
+            summary_body = (
+                f"Total pages   : {len(pages_data)}\n"
+                f"Headers found : {total_headers}\n"
+                f"Footers found : {total_footers}\n"
+                f"{'─'*40}\n" + "\n".join(all_summary)
+            )
+            summary = "\n".join(debug_lines) + "\n\n" + summary_body
+            return markdown, summary
+
         pages_data = result if isinstance(result, list) else []
         total_headers = 0
         total_footers = 0
         all_markdown = []
         all_summary = []
-        for page_num, page_item in enumerate(pages_data):
-            page_regions = get_regions(page_item)
+        for page_num, page_regions in enumerate(pages_data):
             if not isinstance(page_regions, list):
-                page_regions = [page_regions]
+                page_regions = [page_regions] if page_regions else []
             page_md = []
             page_summary = [f"── PAGE {page_num + 1} of {len(pages_data)} ──"]
             for region in page_regions:
-                label = get_label(region)
-                content = get_content(region)
-                if not content.strip():
+                if not isinstance(region, dict):
                     continue
+                label = region.get("label", "text")
+                content = str(region.get("content", ""))
                 if label in ABANDON:
                     continue
                 if label == "header":
@@ -113,20 +165,23 @@ def run_ocr(uploaded_file):
             all_markdown.append("\n\n".join(page_md))
             all_summary.extend(page_summary)
             all_summary.append("")
-        summary = (
+        summary_body = (
             f"Total pages   : {len(pages_data)}\n"
             f"Headers found : {total_headers}\n"
             f"Footers found : {total_footers}\n"
             f"{'─'*40}\n" + "\n".join(all_summary)
         )
         markdown = "\n\n---\n\n".join(all_markdown) if all_markdown else "(No content)"
+        summary = "\n".join(debug_lines) + "\n\n" + summary_body
         return markdown, summary
+
     except Exception as e:
         import traceback
         return f"Error: {str(e)}\n\n{traceback.format_exc()}", "Failed."
 
+
 with gr.Blocks(title="GLM-OCR — MaaS") as demo:
-    gr.Markdown("# 🔍 GLM-OCR — Zhipu MaaS\nUpload PDF or image. Headers 🔵 and Footers 🟢 when returned by the API.")
+    gr.Markdown("# 🔍 GLM-OCR — Zhipu MaaS\nUpload PDF or image. Headers 🔵 and Footers 🟢 from the API.")
     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():
@@ -134,7 +189,7 @@ with gr.Blocks(title="GLM-OCR — MaaS") as demo:
             gr.Markdown("### 📄 Markdown Output")
             markdown_out = gr.Textbox(lines=30, label="")
         with gr.Column():
-            gr.Markdown("### 🗂️ Detected Regions")
+            gr.Markdown("### 🗂️ Detected Regions (+ Debug)")
             regions_out = gr.Textbox(lines=30, label="")
     run_btn.click(fn=run_ocr, inputs=file_input, outputs=[markdown_out, regions_out])
 demo.launch()