Update app.py

app.py

@@ -83,19 +83,26 @@ def is_image_file(filename):
     return any(filename.lower().endswith(ext) for ext in image_file_extensions)
 
 def process_file_and_generate_csv(input_file):
-    # Check if the uploaded file is an image or needs conversion to images
+    output_csv_path = "output.csv"  # Output CSV file name
+    file_content = BytesIO(input_file.read())  # Read file content into memory for processing
+    file_content.seek(0)  # Go to the start of the file-like object
+
+    object_name = os.path.basename(input_file.name)
+    
+    # Check if the uploaded file is an image or needs conversion
     images = []
-    if is_image_file(input_file.name):
-        input_file.seek(0)  # Go to the start of the file
-        images.append(Image.open(input_file))
+    if is_image_file(object_name):
+        images.append(Image.open(file_content))
+        file_content.seek(0)  # Reset for potential re-use
     else:
-        input_file.seek(0)  # Ensure we're at the start of the file
-        images.extend(convert_from_bytes(input_file.read()))
+        # Convert PDF/TIFF to images
+        images.extend(convert_from_path(file_content))
 
     csv_output = BytesIO()
     writer = csv.writer(csv_output)
 
     for i, image in enumerate(images):
+        # Process each image and upload to S3 for Textract processing
         image_byte_array = BytesIO()
         image.save(image_byte_array, format='JPEG')
         image_byte_array.seek(0)
@@ -107,7 +114,7 @@ def process_file_and_generate_csv(input_file):
             tables = [block for block in blocks if block['BlockType'] == "TABLE"]
             generate_table_csv(tables, blocks_map, writer)
     
-    csv_output.seek(0)
+    csv_output.seek(0)  # Go to the start of the CSV in-memory file
     return csv_output, output_csv_path
 
 # Gradio Interface