Retry Mechanism for Python

After being slightly appalled by the suggested retry mechanism for requests I doodled around a bit and came up with this:

import retry

requests.get = retry.me(requests.get, trigger_method="raise_for_status")

result = requests.get("https://httpbin.org/status/500", retries=retry.http_retries(2))

Works for normal functions as well:

import random, retry

def compute_something():
    if random.random() > 0.8:
        raise ValueError("bad result")
    return "actual result"

compute_something = retry.me(compute_something)

result = compute_something(retries=3)