Amazon Textract does not work

I apologize for the difficulty you're experiencing with Amazon Textract. There are a few potential reasons why you might not be getting the expected results:

1. Document format: While you mentioned using a single-page PDF, it's important to ensure that the PDF is text-based and not an image-based PDF. Textract works best with text-based PDFs or image formats like JPEG or PNG.

2. Text recognition: If the text in your document is not clearly printed or has unusual fonts, Textract may have difficulty recognizing it accurately.

3. Query formulation: When using the query feature, make sure your query is precise and matches the exact wording in the document. Textract's query feature is sensitive to exact matches.

4. Document quality: The quality of the document can affect Textract's ability to recognize text. Ensure that the document is clear, well-scanned, and has good contrast.

5. Service limitations: Textract has some known limitations, such as difficulties with certain special characters or formatting like bullet points.

To troubleshoot this issue:

1. Try using the DetectDocumentText API instead of the query feature to see if Textract can detect any text in your document at all.

2. Verify that the document is properly uploaded to S3 and accessible.

3. If possible, try converting your PDF to an image format like JPEG or PNG and analyze that instead.

4. Check if there are any error messages in the Lambda logs that might provide more information about the issue.

5. Consider using the synchronous API (AnalyzeDocument) instead of the asynchronous one if you're not already doing so, as it might provide more immediate feedback.

If these steps don't resolve the issue, you may want to try processing a simpler document to isolate whether the problem is with Textract or specific to your document. If problems persist, consider reaching out to AWS support for further assistance.
Sources
textract is not working as it should | AWS re:Post
Textract Error With Multiples Pages | AWS re:Post
AWS Textract not accepting remote/non-aws document URL ?? | AWS re:Post