timeout

Set a maximum execution time for your functions with a simple decorator.

Motivation

There are some use cases where you might need to limit the amount of time required by a specific function.

• Maybe you have a limited amount of time to execute a task (some said AWS Lambda?)
• Or maybe you just want to make sure tasks don't get blocked by some external resource (like a busy database)

Those were basically the problem I wanted to get resolved. And that's why I embed some collective intelligence (a.k.a. StarOverflow) into this package.

AWS Lambda

AWS Lambda is awesome, but still has a maximum execution time of 5 min. That's usually enough for the kind of tasks it was conceived, but I needed a fail-safe mechanism.

You always have the option to check for the remaining time with context.get_remaining_time_in_millis(), but that might not be applicable in every situation.

Install

This package uses the concurrent.futures package available in the standard library of Python 3.x. For Python 2.6 and 2.7 users please install the backport https://github.com/agronholm/pythonfutures.

$ python setup.py install

Use

from timeout import timeout, TimeoutError

@timeout(60)
def monstrous_task():
    """This task should take less than 1 minute, otherwise the execution will be stopped"""
    pass

def main():
    try:
        monstrous_task()
    except TimeoutError:
        print('Task timed out (> 1min)')