timeout warning

app.py

@@ -1,3 +1,4 @@
+import mlcroissant._src.operation_graph.operations.download as dl_mod
 import requests
 import os
 
@@ -16,41 +17,69 @@ import json
 import time
 import traceback
 from validation import validate_json, validate_croissant, validate_records, generate_validation_report
+import threading
 
 def process_file(file):
     results = []
     json_data = None
+    timer = None
+    warning_text = None  # to be set if timer fires
 
     filename = file.name.split("/")[-1]
 
-    # Check 1: JSON validation
+    # JSON validation
     json_valid, json_message, json_data = validate_json(file.name)
     json_message = json_message.replace("\n✓\n", "\n")
     results.append(("JSON Format Validation", json_valid, json_message, "pass" if json_valid else "error"))
-
     if not json_valid:
-        return results, None
+        return results, None, None
 
-    # Check 2: Croissant validation
+    # Schema validation
     croissant_valid, croissant_message = validate_croissant(json_data)
     croissant_message = croissant_message.replace("\n✓\n", "\n")
     results.append(("Croissant Schema Validation", croissant_valid, croissant_message, "pass" if croissant_valid else "error"))
-
     if not croissant_valid:
-        return results, None
+        return results, None, None
+
+    # Start timer before records validation
+    fired = threading.Event()
+
+    def trigger_warning():
+        nonlocal warning_text
+        warning_text = """
+        ⚠️ <b>This is taking longer than usual</b>. It is possible that this checker is currently being used by a lot of people
+        at the same time, which may trigger rate limiting by the platform hosting your data. The app will then try again and may get into a very long loop.<br><br>
+        In that case, we recommend using any of the following options:
+        <ul style="text-align:left; margin: 0 auto; display:inline-block;">
+            <li>🔁 Duplicate this Space on Hugging Face</li>
+            <li>💻 Run it locally (GitHub or Docker)</li>
+            <li>🥐 Use <code>mlcroissant</code> from <a href="https://github.com/mlcommons/croissant" target="_blank">GitHub</a></li>
+        </ul>
+        """
+        fired.set()
+
+    timer = threading.Timer(0.1, trigger_warning)
+    timer.start()
+
+    try:
+        records_valid, records_message, records_status = validate_records(json_data)
+    finally:
+        timer.cancel()
+
+    if fired.is_set():
+        warning_html_update = gr.update(value=warning_text, visible=True)
+    else:
+        warning_html_update = gr.update(visible=False)
 
-    # Check 3: Records validation (with timeout-safe and error-specific logic)
-    records_valid, records_message, records_status = validate_records(json_data)
     records_message = records_message.replace("\n✓\n", "\n")
     results.append(("Records Generation Test", records_valid, records_message, records_status))
 
-    # Generate final report
     report = generate_validation_report(filename, json_data, results)
-
-    return results, report
+    return results, report, warning_html_update
 
 def create_ui():
     with gr.Blocks(theme=gr.themes.Soft()) as app:
+        delayed_warning_html = gr.HTML("", visible=False)
         gr.HTML("<p align='center'><img src='https://upload.wikimedia.org/wikipedia/en/0/08/Logo_for_Conference_on_Neural_Information_Processing_Systems.svg' alt='NeurIPS Logo' width='400'/></p>")
         gr.Markdown("# 🥐 Croissant Validator for NeurIPS")
         gr.Markdown("""
@@ -481,14 +510,12 @@ def create_ui():
                     gr.update(visible=False),  # validation_results
                     gr.update(visible=False),  # validation_progress
                     gr.update(visible=False),  # report_group
-                    None,  # report_text
-                    None   # report_md
+                    None, None,                # report_text, report_md
+                    gr.update(visible=False)   # delayed_warning_html
                 ]
 
-            # Process the file and get results
-            results, report = process_file(file)
+            results, report, warning_html_update = process_file(file)
 
-            # Extract dataset name from the JSON for the report filename
             try:
                 with open(file.name, 'r') as f:
                     json_data = json.load(f)
@@ -496,19 +523,18 @@ def create_ui():
             except:
                 dataset_name = 'unnamed'
 
-            # Save report to file with new naming convention
             report_filename = f"report_croissant-validation_{dataset_name}.md"
             if report:
                 with open(report_filename, "w") as f:
                     f.write(report)
 
-            # Return final state
             return [
-                build_results_html(results),  # validation_results
-                gr.update(visible=False),  # validation_progress
-                gr.update(visible=True) if report else gr.update(visible=False),  # report_group
-                report if report else None,  # report_text
-                report_filename if report else None  # report_md
+                build_results_html(results),
+                gr.update(visible=False),
+                gr.update(visible=True) if report else gr.update(visible=False),
+                report if report else None,
+                report_filename if report else None,
+                warning_html_update or gr.update(visible=False)
             ]
 
         # Connect UI events to functions with updated outputs
@@ -538,7 +564,7 @@ def create_ui():
                 None,  # report_text
                 None   # report_md
             ]
-
+        
         validate_btn.click(
             fn=show_progress,
             inputs=None,
@@ -547,7 +573,7 @@ def create_ui():
         ).then(
             fn=on_validate,
             inputs=file_input,
-            outputs=[validation_results, validation_progress, report_group, report_text, report_md]
+            outputs=[validation_results, validation_progress, report_group, report_text, report_md, delayed_warning_html]
         )
 
         fetch_btn.click(

validation.py

@@ -1,3 +1,4 @@
+import mlcroissant._src.operation_graph.operations.download as dl_mod
 import requests
 import os