LambdaInvokeFunctionOperator expects wrong type for payload arg

[elad-galili-ka]

Apache Airflow version

2.6.2

What happened

I instantiate LambdaInvokeFunctionOperator in my DAG.

I want to call the lambda function with some payload. After following code example from official documentation, I created a dict, and passed its json string version to the operator:

d = {'key': 'value'}
invoke_lambda_task = LambdaInvokeFunctionOperator(..., payload=json.dumps(d))

When I executed the DAG, this task failed. I got the following error message:

Invalid type for parameter Payload, value: {'key': 'value'}, type: <class 'dict'>, valid types: <class 'bytes'>, <class 'bytearray'>, file-like object

Then I went to official boto3 documentation, and found out that indeed, the payload parameter type is bytes or file. See https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/lambda/client/invoke.html

What you think should happen instead

To preserve backward compatibility, The API should encode payload argument if it is str, but also accept bytes or file, in which case it will be passed-through as is.

How to reproduce

1. Create lambda function in AWS
2. Create a simple DAG with LambdaInvokeFunctionOperator
3. pass str value in the payload parameter; use json.dumps() with a simple dict, as lambda expects json payload
4. Run the DAG; the task is expected to fail

Operating System

ubuntu

Versions of Apache Airflow Providers

apache-airflow-providers-amazon==7.3.0

Deployment

Virtualenv installation

Deployment details

No response

Anything else

No response

Are you willing to submit PR?

• Yes I am willing to submit a PR!

Code of Conduct

• I agree to follow this project's Code of Conduct

[boring-cyborg]

Thanks for opening your first issue here! Be sure to follow the issue template! If you are willing to raise PR to address this issue please do so, no need to wait for approval.

[eladkal]

Feel free to raise PR. Assigned the issue to you