Textract rejects local PDFs: `Exception: PDF must be in S3 bucket`

The problem

Running python3 textractor.py --documents document-name.pdf --text --forms --tables on a local PDF fails with Exception: PDF must be in S3 bucket. Can Textract process a document on my local disk, or does it have to be in S3?

The fix

The Textract async API (which is what multi-page PDFs need) only accepts documents stored as S3 objects - local files are rejected with exactly that exception. So the standard answer is: upload the PDF to an S3 bucket first, then pass the S3 object to Textract. If you want to avoid S3, read the file into memory and pass bytes instead (e.g. client = boto3.client('textract'); b = open(file_name,'rb').read() and call Textract with the bytes as the input document), which works for the synchronous APIs. Note that async document processing for PDFs always requires S3 - there is no way around that for multi-page PDF jobs.

The thread asked: Running python3 textractor.py --documents document-name.pdf --text --forms --tables on a local PDF fails with Exception: PDF must be in S3 bucket. Can Textract process a document on my local disk, or does it have to be in S3?

The verified answer: The Textract async API (which is what multi-page PDFs need) only accepts documents stored as S3 objects - local files are rejected with exactly that exception. So the standard answer is: upload the PDF to an S3 bucket first, then pass the S3 object to Textract. If you want to avoid S3, read the file into memory and pass bytes instead (e.g. client = boto3.client('textract'); b = open(file_name,'rb').read() and call Textract with the bytes as the input document), which works for the synchronous APIs. Note that async document processing for PDFs always requires S3 - there is no way around that for multi-page PDF jobs.

Thread: https://vectle.com/threads/thr_QEqVNAX-lDR6LOIYIjeDyA