Custom event_handler exceptions

[royassis]

Hey

I want to create a custom exception class when using the APIGatewayRestResolver event_handler, is this available ?

I'll describe my use case:
I'm triggering an async step-functions job using a lambda, I want to return a 4xx error when the same lambda is triggered while
the step-function is running.
Right not im returning a 500 error code, with the InternalServerError exception class, but it doesn't feel right as it's not an internal server error.

Best,
Roy

[leandrodamascena]

Hi @royassis! Thanks for starting this thread.

The short answer is yes, you can do that by extending the ServiceError class and creating your own exception class, but you don't have to. You already can do this by raising an exception using the ServiceError class and informing the HTTP status code and message you want. Let me give you an example of both situations:

Using the ServiceError class.

from aws_lambda_powertools import Logger
from aws_lambda_powertools.event_handler import APIGatewayRestResolver
from aws_lambda_powertools.event_handler.exceptions import (
    ServiceError,
)
from aws_lambda_powertools.logging import correlation_paths
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()
app = APIGatewayRestResolver()

@app.get(rule="/already-running-route")
def conflict_error():
    raise ServiceError(409, "CONFLICT -> Already running")

@logger.inject_lambda_context(correlation_id_path=correlation_paths.API_GATEWAY_REST)
def lambda_handler(event: dict, context: LambdaContext) -> dict:
    return app.resolve(event, context)

Creating your own exception class

from http import HTTPStatus

from aws_lambda_powertools import Logger
from aws_lambda_powertools.event_handler import APIGatewayRestResolver
from aws_lambda_powertools.event_handler.exceptions import (
    ServiceError,
)
from aws_lambda_powertools.logging import correlation_paths
from aws_lambda_powertools.utilities.typing import LambdaContext

logger = Logger()
app = APIGatewayRestResolver()

class ConflictError(ServiceError):
    """HTTP 409 conflict error class"""

    def __init__(self, message: str):
        super().__init__(HTTPStatus.CONFLICT, message)


@app.get(rule="/already-running-route")
def conflict_error():
    raise ConflictError("CONFLICT -> Already running")

@logger.inject_lambda_context(correlation_id_path=correlation_paths.API_GATEWAY_REST)
def lambda_handler(event: dict, context: LambdaContext) -> dict:
    return app.resolve(event, context)

The result for both examples is

I used HTTP Status Code 409 because I think it's the best fit for this situation, but you can use whatever you like.

Please let me know if you still have any questions about this and I can help you.
Thank you