amazon-textract-caller

import os from amazon_textract_caller import get_textract_response, TextractFeatures # Configure AWS credentials and region (e.g., via environment variables or AWS CLI config) # os.environ['AWS_ACCESS_KEY_ID'] = os.environ.get('AWS_ACCESS_KEY_ID', '') # os.environ['AWS_SECRET_ACCESS_KEY'] = os.environ.get('AWS_SECRET_ACCESS_KEY', '') # os.environ['AWS_DEFAULT_REGION'] = os.environ.get('AWS_DEFAULT_REGION', 'us-east-1') # Replace with your actual S3 document URI (e.g., "s3://your-bucket/your-document.pdf") s3_document_uri = "s3://YOUR_BUCKET/YOUR_DOCUMENT.pdf" try: # Call Textract API with specified features # This package returns the raw JSON response from Textract. response = get_textract_response( input_document=s3_document_uri, features=[TextractFeatures.FORMS, TextractFeatures.TABLES] ) print("Textract API call successful. Raw JSON response received:") # print(response) # Uncomment to see the full raw Textract JSON response print(f"Detected {len(response.get('Blocks', []))} blocks.") print("\nNote: For higher-level parsing and object models, consider the `amazon-textract-textractor` library.") except Exception as e: print(f"Error during Textract API call: {e}") print("Ensure valid AWS credentials, correct S3 URI, and appropriate permissions.")