RunnerBean

A simple tool for creating long-running Python workers listening for Beanstalk jobs.

Jobs can be posted as YAML or JSON objects, with keys mapping to the callable's keyword arguments. Passing parse=False to the __init__ method will disable YAML parsing, and the callable will be provided a single string argument containing the job's body.

Returning True from the callable will be seen as a success and will delete the job from the queue. Returning False or None will be seen as a failure, and the job will be buried for later inspection.

If listening on multiple tubes, add the argument __tubes__ to the method to receive the tube name when the callable is executed.

Usage:

import logging
from RunnerBean import Runner

def print_message(recipient, message, __tube__):
    # accepts a job with the following structure:
    """
    message: Hello world!
    recipient: joe bloggs
    """
    print recipient, message

    print __tube__ #= 'messages'

    return True # this deletes the job from the tube

if __name__ == '__main__':
    runner = Runner(print_message,
                    parse=True, # default; job body should be parsed as YAML
                    tubes="messages", # string or list of tubes to listen on
                    host='0.0.0.0', # beanstalkd host
                    port=11300, # beanstalkd port
                    loglevel=logging.DEBUG, # set log level, default: ERROR
                    logfile='runnerbean.log' # set the logfile
                    )
    runner.run()