Amazon Textract Create / Get Adapters using SDK - PHP

Hi Max, could you try to upgrade your PHP SDK to the latest version?

Textract Adapter related change should be included in PHP SDK after version 3.283.3: https://github.com/aws/aws-sdk-php/commit/b52bcfa58f4decbd2d0ce2a4c0259b11c62fb1a9#diff-1e907b07deb3e31b05390124e65d9ec52aeae68175fcd84216b8586b8d829496

Please let us know if you need more help, thanks!

Thank you for the reply. The upgrade to the AWS version resolved the previous issue. However, we're encountering new technical problems related to the AP SDK:

1. We've created an Adapter > DataSet > Queries in AWS Textract.

2. Our goal is to analyze a document using the Adapter ID through the API to obtain query answers based on the established queries after scanning the document.

We've utilized the 'analyzeDocument' function ($client->analyzeDocument) to retrieve a response, but we're encountering the following issues:

1. The system doesn't allow us to upload a multiple-page PDF for scanning.

2. Although we've set up queries in the AWS Adapter, the 'QueriesConfig' is mandatory when calling the API, which seems unnecessary as we've already configured queries in the adapter.

I've been following the documentation provided [link], but I'm still facing these issues. Could you please assist me with this problem or ask me any specific questions?

[Link to documentation: https://docs.aws.amazon.com/aws-sdk-php/v3/api/api-textract-2018-06-27.html#analyzedocument]

Thank you, Max

Hi Max, for multi-page PDF, we recommend using StartDocumentAnalysis API to process multi-page PDF in asynchronous fashion, please refer to https://docs.aws.amazon.com/textract/latest/dg/async-analyzing-with-sqs.html for more details.

And for QueriesConfig when running AnalyzeDocument API, unfortunately that is required for running this API. This is for allowing our customer to have the flexibility to change the QueriesConfig between trained Adapter and when running inference request.

Let us know if this helps, thanks!

Hi Thanks for reply We tried using "StartDocumentAnalysis" Function and it works to Analyse multi pages PDF and Textract all documents in array but we are not getting Query Result which we sending in QueriesConfig its ok if it require to setup QueriesConfig but its not provide result based on query. just scan whole documents and textract all words/lines

Can you pls help me in this case then we are good

Thanks

Max

Hi Max, I am glad that you are able to process your multi-page PDF using StartDocumentAnalysis API, however, in order to further assist you about the inference response issue, could you open a customer support case in AWS console, and we can further investigate for you if you can provide job related information through customer support case. Hope this helps, thanks!