Update app.py

app.py

@@ -4,6 +4,7 @@ import pytesseract
 import requests
 import pandas as pd
 import gradio as gr
+import io
 from io import BytesIO
 
 # ──────────────────────────────────────────────────────────────
@@ -54,10 +55,12 @@ def ocr_on_region(image: np.ndarray, box: tuple):
     Return the raw OCR text.
     """
     x, y, w, h = box
-    cropped = image[y:y + h, x:x + w]
+    cropped = image[y : y + h, x : x + w]
     gray_crop = cv2.cvtColor(cropped, cv2.COLOR_BGR2GRAY)
-    _, thresh_crop = cv2.threshold(gray_crop, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
-    custom_config = r'--oem 3 --psm 6'
+    _, thresh_crop = cv2.threshold(
+        gray_crop, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU
+    )
+    custom_config = r"--oem 3 --psm 6"
     text = pytesseract.image_to_string(thresh_crop, config=custom_config)
     return text.strip()
 
@@ -84,7 +87,7 @@ def query_openlibrary(title_text: str, author_text: str = None):
                 "title": doc.get("title", ""),
                 "author_name": ", ".join(doc.get("author_name", [])),
                 "publisher": ", ".join(doc.get("publisher", [])),
-                "first_publish_year": doc.get("first_publish_year", "")
+                "first_publish_year": doc.get("first_publish_year", ""),
             }
     except Exception as e:
         print(f"OpenLibrary query failed: {e}")
@@ -97,7 +100,7 @@ def query_openlibrary(title_text: str, author_text: str = None):
 def process_image(image_file):
     """
     Gradio passes a PIL image or numpy array. Convert to OpenCV BGR, detect covers → OCR → OpenLibrary.
-    Return a DataFrame and CSV bytes.
+    Return a DataFrame and a (filename, BytesIO) tuple for CSV.
     """
     img = np.array(image_file)[:, :, ::-1].copy()  # PIL to OpenCV BGR
     boxes = detect_book_regions(img)
@@ -116,20 +119,28 @@ def process_image(image_file):
         if meta:
             records.append(meta)
         else:
-            records.append({
-                "title": title_guess,
-                "author_name": author_guess or "",
-                "publisher": "",
-                "first_publish_year": "",
-            })
+            records.append(
+                {
+                    "title": title_guess,
+                    "author_name": author_guess or "",
+                    "publisher": "",
+                    "first_publish_year": "",
+                }
+            )
 
     if not records:
         df_empty = pd.DataFrame(columns=["title", "author_name", "publisher", "first_publish_year"])
-        return df_empty, df_empty.to_csv(index=False).encode()
+        # Build an empty CSV bytes buffer
+        empty_csv = df_empty.to_csv(index=False).encode()
+        buffer = io.BytesIO(empty_csv)
+        buffer.name = "books.csv"
+        return df_empty, buffer
 
     df = pd.DataFrame(records)
     csv_bytes = df.to_csv(index=False).encode()
-    return df, csv_bytes
+    buffer = io.BytesIO(csv_bytes)
+    buffer.name = "books.csv"
+    return df, buffer
 
 # ──────────────────────────────────────────────────────────────
 # 5. Build the Gradio Interface
@@ -151,15 +162,20 @@ def build_interface():
 
         output_table = gr.Dataframe(
             headers=["title", "author_name", "publisher", "first_publish_year"],
-            label="Detected Books with Metadata"
+            label="Detected Books with Metadata",
+            datatype="pandas",
         )
-        download_btn = gr.Download(label="Download CSV")
+        download_file = gr.File(label="Download CSV")
 
         def on_run(image):
-            df, csv_bytes = process_image(image)
-            return df, csv_bytes
+            df, file_buffer = process_image(image)
+            return df, file_buffer
 
-        run_button.click(fn=on_run, inputs=[img_in], outputs=[output_table, download_btn])
+        run_button.click(
+            fn=on_run,
+            inputs=[img_in],
+            outputs=[output_table, download_file],
+        )
 
     return demo