Redo provides various means to add seamless ability to retry to any Python callable. Redo includes plain functions (redo.retry, redo.retry_async), decorators (redo.retriable, redo.retriable_async), and a context manager (redo.retrying) to enable you to integrate it in the best possible way for your project. As a bonus, a standalone interface is also included ("retry").

For installing with pip, run following commands

pip install redo

Below is the list of functions available

• retrier
• retry
• retry_async
• retriable
• retriable_async
• retrying (contextmanager)

A generator function that sleeps between retries, handles exponential back off and jitter. The action you are retrying is meant to run after retrier yields. At each iteration, we sleep for sleeptime + random.randint(-jitter, jitter). Afterwards sleeptime is multiplied by sleepscale for the next iteration.

Arguments Detail:

1. attempts (int): maximum number of times to try; defaults to 5
2. sleeptime (float): how many seconds to sleep between tries; defaults to 60s (one minute)
3. max_sleeptime (float): the longest we'll sleep, in seconds; defaults to 300s (five minutes)
4. sleepscale (float): how much to multiply the sleep time by each iteration; defaults to 1.5
5. jitter (int): random jitter to introduce to sleep time each iteration. the amount is chosen at random between [-jitter, +jitter] defaults to 1

Output: None, a maximum of attempts number of times

Example:

>>> n = 0
>>> for _ in retrier(sleeptime=0, jitter=0):
...     if n == 3:
...         # We did the thing!
...         break
...     n += 1
>>> n
3
>>> n = 0
>>> for _ in retrier(sleeptime=0, jitter=0):
...     if n == 6:
...         # We did the thing!
...         break
...     n += 1
... else:
...     print("max tries hit")
max tries hit

Calls an action function until it succeeds, or we give up.

Arguments Detail:

1. action (callable): the function to retry
2. attempts (int): maximum number of times to try; defaults to 5
3. sleeptime (float): how many seconds to sleep between tries; defaults to 60s (one minute)
4. max_sleeptime (float): the longest we'll sleep, in seconds; defaults to 300s (five minutes)
5. sleepscale (float): how much to multiply the sleep time by each iteration; defaults to 1.5
6. jitter (int): random jitter to introduce to sleep time each iteration. The amount is chosen at random between [-jitter, +jitter] defaults to 1
7. retry_exceptions (tuple): tuple of exceptions to be caught. If other exceptions are raised by action(), then these are immediately re-raised to the caller.
8. cleanup (callable): optional; called if one of retry_exceptions is caught. No arguments are passed to the cleanup function; if your cleanup requires arguments, consider using functools.partial or a lambda function.
9. args (tuple): positional arguments to call action with
10. kwargs (dict): keyword arguments to call action with

Output: Whatever action(*args, **kwargs) returns

Output: Whatever action(*args, **kwargs) raises. retry_exceptions are caught up until the last attempt, in which case they are re-raised.

Example:

>>> count = 0
>>> def foo():
...     global count
...     count += 1
...     print(count)
...     if count < 3:
...         raise ValueError("count is too small!")
...     return "success!"
>>> retry(foo, sleeptime=0, jitter=0)
1
2
3
'success!'

An asynchronous function that retries a given async callable.

Arguments Detail:

1. func (function): an awaitable function to retry
2. attempts (int): maximum number of attempts; defaults to 5
3. sleeptime_callback (function): function to determine sleep time after each attempt; defaults to calculateSleepTime
4. retry_exceptions (list or exception): exceptions to retry on; defaults to Exception
5. args (list): arguments to pass to func
6. kwargs (dict): keyword arguments to pass to func
7. sleeptime_kwargs (dict): keyword arguments to pass to sleeptime_callback

Output: The value from a successful func call or raises an exception after exceeding attempts.

Example:

>>> async def async_action():
...     # Your async code here
>>> result = await retry_async(async_action)

A decorator factory for retry(). Wrap your function in @retriable(...) to give it retry powers!

Arguments Detail: Same as for retry, with the exception of action, args, and kwargs, which are left to the normal function definition.

Output: A function decorator

Example:

>>> count = 0
>>> @retriable(sleeptime=0, jitter=0)
... def foo():
...     global count
...     count += 1
...     print(count)
...     if count < 3:
...         raise ValueError("count too small")
...     return "success!"
>>> foo()
1
2
3
'success!'

A decorator for asynchronously retrying a function.

Arguments Detail:

1. retry_exceptions (list or exception): exceptions to retry on; defaults to Exception
2. sleeptime_kwargs (dict): keyword arguments to pass to the sleeptime callback

Output: A function decorator that applies retry_async to the decorated function.

Example:

>>> @retriable_async()
... async def async_action():
...     # Your async code here
>>> result = await async_action()

A context manager for wrapping functions with retry functionality.

Arguments Detail:

1. func (callable): the function to wrap other arguments as per retry

Output: A context manager that returns retriable(func) on __enter__

Example:

>>> count = 0
>>> def foo():
...     global count
...     count += 1
...     print(count)
...     if count < 3:
...         raise ValueError("count too small")
...     return "success!"
>>> with retrying(foo, sleeptime=0, jitter=0) as f:
...     f()
1
2
3
'success!'