Error 2005: While attempting to GET response from form recognizer

Question

Currently I'm using form recognizer version 2.1 preview to train a custom model. I'm able to test the model in Form Recognizer Labeling Tool and got the output. When I input the same file that I got out in labeling tool in my program I'm getting the error below.

{"status": "failed", "createdDateTime": "2020-09-25T20:03:21Z", "lastUpdatedDateTime": "2020-09-25T20:03:21Z", "analyzeResult": {"version": "2.1.0", "errors": [{"code": "2005", "message": "The file submitted couldn't be parsed. This can be due to one of the following reasons: the file format is not supported ( Supported formats include JPEG, PNG, BMP, PDF and TIFF), the file is corrupted or password protected."}]}}

The GET request code used is:

resp = requests.get(url=get_url,headers={"Ocp-Apim-Subscription-Key":FORM_RECOGNIZER_SUBSCRIPTION_KEY})

Answer 1

What file are you trying to use as input ? Form Recognizer supports PDF, Tiff and images (PNG and JPEG) as file types and inputs to the analyze API. See more details here - https://learn.microsoft.com/en-us/azure/cognitive-services/form-recognizer/quickstarts/python-labeled-data?tabs=v2-0#analyze-forms-for-key-value-pairs-and-tables

Answer 2

I have saved the file to server and then tried to read it from there and pass the read file to form recognizer. This worked for me. But I don't know why did it happen.

Answer 3

I also encountered this exact error message following this article.

The article show 4 steps:

1. Train Model (require SAS to blob - whole folder)
2. Get model result
3. Analyze (require SAS to a single file)
4. Get analyze result
5. Profit

I got this error on step 4.

After monkeying around, I figured that the cause is not actually in step 4 but in step 3 instead. I was providing SAS to the blob instead of SAS to the file.

After I corrected the SAS URL it works perfectly.

Here is how to get SAS to blob:

Here is how to get SAS to a file: