Advanced concept: callbacks and async code

A callback is a function supplied to receive a future value. It is useful in some APIs, but a callback chain can hide the order of work and make error handling difficult.

from collections.abc import Callable


def load_count(callback: Callable[[int], None]) -> None:
    callback(42)


received: list[int] = []
load_count(received.append)

assert received == [42]

Prefer async and await for asynchronous workflows

async functions return their result normally from the reader’s perspective. The event loop manages the continuation instead of requiring each caller to provide one.

import asyncio


async def load_count_async() -> int:
    await asyncio.sleep(0)
    return 42


async def check_load_count() -> None:
    assert await load_count_async() == 42


await check_load_count()  # noqa: F704 - valid top-level await in a Jupyter cell

Expression’s asynchronous effect builders combine this direct style with explicit Option or Result short-circuiting when an asynchronous workflow can be absent or fail in an expected way.