Update app.py

app.py

@@ -3,11 +3,11 @@ import sys
 import tempfile
 from pathlib import Path
 import pandas as pd
-
-# Keep Gradio on client-side rendering. In hosted Gradio 6 runs, SSR starts a
-# Node proxy that can emit noisy asyncio cleanup tracebacks at startup/shutdown.
-os.environ.setdefault("GRADIO_SSR_MODE", "false")
-
+
+# Keep Gradio on client-side rendering. In hosted Gradio 6 runs, SSR starts a
+# Node proxy that can emit noisy asyncio cleanup tracebacks at startup/shutdown.
+os.environ.setdefault("GRADIO_SSR_MODE", "false")
+
 import gradio as gr
 import time
 
@@ -35,41 +35,41 @@ def gpu_task(duration=120):
 
 # Initialize Extractor
 print("Initializing Passport Extractor...")
-EXTRACTOR = PassportExtractor(use_gpu=True)
+EXTRACTOR = PassportExtractor(use_gpu=False)
 print("Extractor initialized.")
 
 
 MAX_FILE_SIZE_MB = 100
 ALLOWED_FILE_TYPES = [
-    ".png",
-    ".jpg",
-    ".jpeg",
-    ".pdf",
-    ".avif",
-    ".webp",
-    ".bmp",
-    ".tiff",
-]
-
-
-def _format_data(good_results, airline):
-    if airline == "Fly Dubai":
-        return format_fly_dubai(good_results)
-    if airline == "Iraqi":
-        return format_iraqi(good_results)
-    if airline == "Fly Baghdad":
-        return format_fly_baghdad(good_results)
-
-    # Default view
-    df = pd.DataFrame(good_results)
-    if not df.empty:
-        from extractor import format_date
-        if "date_of_birth" in df.columns:
-            df["date_of_birth"] = df["date_of_birth"].apply(lambda x: format_date(x, "%d/%m/%Y", is_dob=True))
-        if "expiration_date" in df.columns:
-            df["expiration_date"] = df["expiration_date"].apply(lambda x: format_date(x, "%d/%m/%Y", is_dob=False))
-    return df
-
+    ".png",
+    ".jpg",
+    ".jpeg",
+    ".pdf",
+    ".avif",
+    ".webp",
+    ".bmp",
+    ".tiff",
+]
+
+
+def _format_data(good_results, airline):
+    if airline == "Fly Dubai":
+        return format_fly_dubai(good_results)
+    if airline == "Iraqi":
+        return format_iraqi(good_results)
+    if airline == "Fly Baghdad":
+        return format_fly_baghdad(good_results)
+
+    # Default view
+    df = pd.DataFrame(good_results)
+    if not df.empty:
+        from extractor import format_date
+        if "date_of_birth" in df.columns:
+            df["date_of_birth"] = df["date_of_birth"].apply(lambda x: format_date(x, "%d/%m/%Y", is_dob=True))
+        if "expiration_date" in df.columns:
+            df["expiration_date"] = df["expiration_date"].apply(lambda x: format_date(x, "%d/%m/%Y", is_dob=False))
+    return df
+
 
 @gpu_task(duration=300)
 def process_files(files, airline, export_all_files, progress=gr.Progress()):
@@ -146,7 +146,7 @@ def process_files(files, airline, export_all_files, progress=gr.Progress()):
         return file_results, file_problem
 
     completed_files = 0
-    with ThreadPoolExecutor(max_workers=4) as executor:
+    with ThreadPoolExecutor(max_workers=2) as executor:
         future_to_file = {executor.submit(process_single_file, f): f for f in files}
         for future in as_completed(future_to_file):
             completed_files += 1
@@ -167,14 +167,21 @@ def process_files(files, airline, export_all_files, progress=gr.Progress()):
     end_time = time.time()
     duration = round(end_time - start_time, 2)
 
-    if not good_results:
-        status_msg = f"⚠️ No data could be extracted. (Time: {duration}s)"
-        if problematic_files:
-            status_msg += "\n\nProblems:\n" + "\n".join(
-                [f"- {p['file_name']}: {p['reason']}" for p in problematic_files]
-            )
+    if not good_results and problematic_files:
+        status_msg = f"No data could be extracted. (Time: {duration}s)"
+        status_msg += "\n\nProblems:\n" + "\n".join(
+            [f"- {p['file_name']}: {p['reason']}" for p in problematic_files]
+        )
         return None, status_msg, None, None
 
+    if not good_results:
+        return (
+            None,
+            f"⚠️ No data could be extracted. (Time: {duration}s)",
+            None,
+            None,
+        )
+
     df = _format_data(good_results, airline)
     status_msg = f"Extracted {len(good_results)} passport(s) from {total_files} file(s) in {duration} seconds."
     if problematic_files:
@@ -189,9 +196,9 @@ def process_files(files, airline, export_all_files, progress=gr.Progress()):
     return df, status_msg, csv_path, excel_path
 
 
-def create_demo():
+def main():
     with gr.Blocks(title="Passport OCR Extractor") as demo:
-        gr.Markdown("# Passport OCR Extractor")
+        gr.Markdown("#Passport OCR Extractor")
         with gr.Row():
             with gr.Column(scale=1):
                 airline = gr.Dropdown(
@@ -216,13 +223,6 @@ def create_demo():
             inputs=[file_input, airline, gr.State(False)],
             outputs=[table_output, status_output, csv_download, excel_download],
         )
-    return demo
-
-
-demo = create_demo()
-
-
-def main():
     port = int(os.environ.get("PORT", "7860"))
     demo.launch(server_name="0.0.0.0", server_port=port, show_error=True, ssr_mode=False)