Raw response for the API Status code 403 - No AWSAccessKey was presented -

[Edivaldo34268467wm7d]

Hello, Adobe Community,

I'm experiencing an issue when trying to perform a file upload to Adobe's service using Amazon S3. I'm receiving a "403 Access Denied: No AWSAccessKey was presented" error. I have checked to ensure all the authentication parameters are correctly included in the request. However, the error persists, suggesting that an AWS access key is missing or there is an issue with the request's signature.

I would greatly appreciate any guidance or support to resolve this issue.

Thank you!

[Raymond Camden]

I assume you are using the REST APIs with external cloud storage, right? And you generated a Read url for input, and Put for output?

[Edivaldo34268467wm7d]

Hello,

Thank you for your response. Yes, I am using the REST APIs with external cloud storage in the context of the PDF Extract API. I have generated a Read URL for input and a PUT URL for output. However, I'm encountering authentication and access issues, specifically the "403 Access Denied: No AWSAccessKey was presented" error. Any additional guidance on how to resolve this issue would be greatly appreciated.

Thank you for your assistance.

[Raymond Camden]

Well, if you log out the read URL, can you copy and paste that into an incognito browser and confirm it loads? If you can't load the PDF, our API won't be able to. 

[Edivaldo34268467wm7d]

I'm trying to integrate it into an application I'm developing on the Bubble platform, to extract data from PDF documents. I've tried OCR Space, Google Vision, and finally I tried ChatGPT. The first two return the test in an unstructured way, making data extraction difficult, with Regex regular expressions, and ChatGPT returns incorrect information, even after "training". I imagined that the integration with Adobe would be more efficient, for obvious reasons. But, life goes on... Thank you.

[Edivaldo34268467wm7d]

I meant: unstructured text

[Raymond Camden]

Ok, but the question remains the same. 🙂 How are you creating the URLs you send to our service? There are standard AWS libraries that make generating S3 urls simple. You would use them to generate a READ (for input) and PUT (for output) when calling the API. Have you done that?

[Edivaldo34268467wm7d]

Bubble offers the possibility of uploading a file, which is saved in the cloud, generating a url. When I compose the url and open it in the anonymous browser, the pdf opens normally.

[Raymond Camden]

In theory, we only document support for S3, Dropbox, Sharepoint, and Azure. I think that's your issue. 

[Edivaldo34268467wm7d]

Okay, thanks.