Update app.py

app.py

@@ -84,25 +84,26 @@ def is_image_file(filename):
 
 def process_file_and_generate_csv(input_file):
     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
-
+    
+    # Handling different types of input file objects
+    if hasattr(input_file, 'read'):  # If input_file has a read method, it's a file-like object
+        file_content = BytesIO(input_file.read())  # Read file content into memory for processing
+    else:  # Assuming input_file might be a path (string) to the uploaded file
+        file_content = open(input_file.name, 'rb')  # Open the file for reading in binary mode
+    
     object_name = os.path.basename(input_file.name)
     
-    # Check if the uploaded file is an image or needs conversion
     images = []
     if is_image_file(object_name):
         images.append(Image.open(file_content))
         file_content.seek(0)  # Reset for potential re-use
     else:
-        # 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)
@@ -114,7 +115,7 @@ 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)  # Go to the start of the CSV in-memory file
+    csv_output.seek(0)
     return csv_output, output_csv_path
 
 # Gradio Interface