PDF must be in S3 bucket? about amazon-textract-textractor HOT 7 CLOSED

Thanks @darwaishx ! It would be great to add this to the READ.me to clarify to future users.

from amazon-textract-textractor.

Textract async API expect PDF as an S3 object.

Example below show how to take a PDF on local disk, extract individual images and process them.
https://github.com/aws-samples/amazon-textract-searchable-pdf/blob/master/src/SearchablePDF/src/main/java/DemoPdfFromLocalPdf.java

from amazon-textract-textractor.

Yeah @darwaishx that README is pretty confusing...at least it got me on this same issue as well.

There are a couple sections in particular:

• python3 textractor.py --documents [file|folder|S3Object|S3Folder] --text --forms --tables --region [AWSRegion] --insights --medical-insights --translate [LanguageCode]

Argument	Description
--documents	Name of the document or local folder/S3 bucket

which seem to reference (or at least mistakenly imply) the ability for this script to accept local files directly.

from amazon-textract-textractor.

Probably clarify more between 'documents' and 'pdf'.

Beautifully written samples and textractor local tools

from amazon-textract-textractor.

Ran into this as well, send the PDF as bytes instead:

file_name = 'Invoice_INV300351.pdf'
client = boto3.client('textract', 'us-east-1')
with open(file_name, "rb") as sample_file:
    b = bytearray(sample_file.read())
response = call_textract(input_document=b, boto3_textract_client=client )
pickle.dump(response, open(f'response_{file_name.split(".")[0]}.pk', 'wb')) 

from amazon-textract-textractor.

Thx @grantrosse , can you create a new ticket? In 2019 (original ticket date), Textract did in deed not support a PDF to be passed in from the local filesystem. However, that changed now and we should update that behavior.

from amazon-textract-textractor.

Yeah for sure @schadem #236

from amazon-textract-textractor.