python-sqs-listener

python-sqs-listener copied to clipboard

When creating queues via CloudFormation you can allow the queue names to be generated automatically. For example a template with these resources:

  MailDeadLetterQueue:
    Type: AWS::SQS::Queue

  MailQueue:
    Type: AWS::SQS::Queue
    Properties:
      ReceiveMessageWaitTimeSeconds: 20
      RedrivePolicy:
        deadLetterTargetArn: !Sub ${MailDeadLetterQueue.Arn}
        maxReceiveCount: 3

Generates two queues named StackName-MailQueue-UPCWLGTA34SP and StackName-MailDeadLetterQueue-1TD3W1IE8HV2C.

Currently, the codebase only looks for queues that match the queue name as a prefix.

https://github.com/nimbusscale/python-sqs-listener/blob/master/sqs_listener/init.py#L80

Given we are passing the exact queue names to SqsListener, I'd like to update the code so that it looks for each queue individually. Something like this:

        sqs = self._session.client('sqs', region_name=self._region_name, endpoint_url=self._endpoint_name, use_ssl=ssl)
        try:
            self._queue_url = sqs.get_queue_url(QueueName=self._queue_name)['QueueUrl']
            mainQueueExists = True
        except sqs.exceptions.QueueDoesNotExist:
            mainQueueExists = False

        try:
            self._error_queue_url = sqs.get_queue_url(QueueName=self._error_queue_name)['QueueUrl']
            errorQueueExists = True
        except sqs.exceptions.QueueDoesNotExist:
            errorQueueExists = False

This is just to get the concept apart, I realize that there needs to be support for when AWS_ACCOUNT_ID env var is set and there may be a few other changes to match the rest of the code as well. I'll submit a complete PR assuming this sounds like a direction you are OK with and this project is still active (which it seems to be).

Thanks!

I know it's been a year, but I'm trying to give this project a little attention now. If you don't mind explaining the issue better, it's not clear to me what the problem is you're running into