backon 3.0.0

Function decoration for backoff and retry

backon

Function decoration for backoff and retry — modern, fast, and zero dependencies.

Why backon?

backon is the evolution of backoff — a zero-dependency Python library for retry with exponential backoff. If you know backoff, you already know backon.

Feature	backoff	tenacity	backon
Python 3.10+ native	❌	❌	✅
Type hints	❌ partial	✅	✅ full
disable() / enable() toggle	❌	❌	✅
Context manager API	❌	✅	✅
Functional retry() API	❌	✅	✅
on_attempt callback	❌	✅	✅
Custom sleep injection	❌	❌	✅
time.monotonic()	❌	✅	✅
PDM / PEP 621 build	❌	❌	✅
Zero dependencies	✅	✅	✅

Quick Start

pip install backon

Retry on exception

import backon

@backon.on_exception(backon.expo, ValueError, max_tries=3)
def fetch_data():
    return api.call()

Retry on predicate

@backon.on_predicate(backon.constant, max_tries=5, interval=0.5)
def poll_status():
    return check_ready()

Functional API

result = backon.retry(fetch_data, backon.expo, exception=ValueError, max_tries=3)

Context manager

with backon.Retrying(backon.expo, exception=ValueError, max_tries=3) as r:
    result = r.call(fetch_data)

Wait Generators

Generator	Description
expo(base=2, factor=1, max_value=None)	Exponential backoff
constant(interval=1)	Constant interval
fibo(max_value=None)	Fibonacci backoff
runtime(value=callable)	Dynamic wait from return value
decay(initial_value=1, decay_factor=1, min_value=None)	Exponential decay

Jitter

@backon.on_exception(backon.expo, ValueError, jitter=backon.full_jitter)

• full_jitter — random between 0 and the wait value
• random_jitter — random ±25% around the wait value
• None — no jitter

Handlers

def log_attempt(details):
    print(f"Attempt {details['tries']} for {details['target'].__name__}")

@backon.on_exception(
    backon.expo, ValueError, max_tries=3,
    on_attempt=log_attempt,
    on_backoff=log_attempt,
    on_success=log_attempt,
    on_giveup=log_attempt,
)
def f():
    ...

Available details keys: target, args, kwargs, tries, elapsed, value (on_success/on_backoff/on_giveup), exception (on_backoff/on_giveup), wait (on_backoff).

Global Toggle

backon.disable()   # skip retry, call function directly
backon.enable()    # re-enable retry

Async Support

Everything works with async def functions — no extra flags needed.

@backon.on_exception(backon.expo, ValueError, max_tries=3)
async def fetch_data():
    return await api.call()

Custom Sleep

@backon.on_exception(backon.expo, ValueError, max_tries=3,
                     sleep=lambda s: print(f"waiting {s}s"))
def f():
    ...

Installation

pip install backon

Requires Python 3.10+.

Migrating from backoff

backon is a drop-in replacement for most backoff users. Just change:

# before
import backoff
@backoff.on_exception(backoff.expo, ValueError, max_tries=3)

# after
import backon
@backon.on_exception(backon.expo, ValueError, max_tries=3)

License

MIT