OCR API filetype not detected

We’re having an issue with the wrong filetype.

+“OCRExitCode”: 99
+“IsErroredOnProcessing”: true
+“ErrorMessage”: array:1 [▼
0 => “Some internal processing error occurred. Please give us some time, we are looking into this!”
+“ErrorDetails”: “”
+“ProcessingTimeInMilliseconds”: “609”

Bump… still errors, everytime. I’m a pro member, and haven’t had an email response either.

Same thing from postman…

@afx Thanks for posting the details, and for your email with the Form Assembly file link. I checked the request header of your PDF file in Chrome developer tools and it turns out that the MIME type is wrong. For a PDF file, it should be application/pdf.

On our side, the error message “internal error” is very confusing and not helpful at all. We will provide a better error message with the next update.

I don’t have any way of specifying the MIME type form From Assembly, and we are instead (and have been) passing the fileType PDF parameter. But that is not working now either.

We haven’t changed anything on our end since this stopped working. Thanks.

The issue has been fixed on the free API endpoint. You can now overwrite the automatic file type detection (which is based on the MIME type) with FILETYPE=....

Test link (menu120.exe is a PNG file):

Once everything is stable we will roll out the patch to all PRO OCR API endpoints in the next days, too. Meanwhile you can use the free endpoint with your PRO api key, too.