asyncio.all_tasks()

Concurrency & Async

Meaning

Returns a set of all currently pending asyncio.Task objects in the running event loop.

Primary Function

Provide a snapshot of all active asynchronous tasks for monitoring, debugging, or graceful shutdown.

Communicative Purpose

Inform the programmer about the set of outstanding coroutines so they can decide whether to await, cancel, or inspect them.

Pattern

asyncio.all_tasks()

Core Structure

asyncio.all_tasks()

Função primária

Provide a snapshot of all active asynchronous tasks for monitoring, debugging, or graceful shutdown.

Propósito comunicativo

Inform the programmer about the set of outstanding coroutines so they can decide whether to await, cancel, or inspect them.

Situações de gatilho

Preparing to shut down the event loop Debugging hanging coroutines Implementing graceful cancellation of all background work Logging or monitoring async workload

Contextos

Inside an asyncio event loop, e.g., in a shutdown handler or a debugging routine During testing to verify that no tasks are left dangling In library cleanup code before loop.close()

Padrão

asyncio.all_tasks()

Estrutura central

asyncio.all_tasks()

Slots de substituição

none (function takes no arguments; deprecated loop parameter ignored)

Colocados típicos

asyncio Task gather wait cancel loop shutdown gather await

Substituições comuns

none (the function accepts no meaningful arguments
the optional loop parameter is deprecated and ignored)

Erros comuns

Confusing with asyncio.current_task() which returns only the current task Assuming the returned set includes completed tasks (it only contains pending tasks) Treating the result as a list instead of a set Modifying the returned set while iterating over it

Similar / contraste

Similar: asyncio.current_task() – returns the currently executing task only Similar: asyncio.Task.all_tasks(loop) – deprecated alternative Contrast: asyncio.all_tasks() vs. asyncio.all_tasks(loop=None) – the latter is deprecated

Interferências

Do not confuse with asyncio.current_task(); they serve different purposes Avoid mutating the returned set while iterating; changes to tasks after retrieval are reflected in the set but concurrent modification can raise RuntimeError Remember that completed tasks are automatically removed from the set

Família do chunk

asyncio-task-management

Nuance

The function returns only tasks that are still pending; once a task reaches a finished state (finished or cancelled) it is removed from the set.

Efeito pragmático

Gives the programmer an immediate view of the asynchronous workload, enabling orderly cancellation, cleanup, or diagnostic logging of outstanding coroutines.

Dica de memória

Think ‘all pending tasks’ when you need to cancel or await remaining work before shutting down the loop.

Nota

The returned value is a set; iteration order is undefined. The optional loop parameter has been deprecated since Python 3.7 and is ignored in newer versions.

Upgrade path

No planned changes; the function remains stable; the deprecated loop argument may be removed in a future Python release.

Frequência: MediumFormulaicidade: FixedTipo de construção: function callPrioridade de aquisição: Active recallPrioridade de output: OutputTag de espaçamento: Medium-term

Examples

import asyncio async def worker(name, delay): await asyncio.sleep(delay) print(f"{name} done") async def graceful_shutdown(): tasks = asyncio.all_tasks() current = asyncio.current_task() pending = [t for t in tasks if t is not current] for task in pending: task.cancel() await asyncio.gather(*pending, return_exceptions=True) print(f"Cancelled {len(pending)} remaining tasks") async def main(): asyncio.create_task(worker("a", 5)) asyncio.create_task(worker("b", 10)) await asyncio.sleep(1) await graceful_shutdown() asyncio.run(main())

import asyncio async def fetch(url): await asyncio.sleep(2) return f"data from {url}" async def monitor(): while True: tasks = asyncio.all_tasks() print(f"[{len(tasks)}] pending tasks running") for t in tasks: print(f" - {t.get_name()}: {t.get_coro()}") await asyncio.sleep(5) async def main(): monitor_task = asyncio.create_task(monitor()) results = await asyncio.gather( fetch("https://api.example.com/users"), fetch("https://api.example.com/posts") ) monitor_task.cancel() asyncio.run(main())

import asyncio import pytest async def leaky_operation(): asyncio.create_task(asyncio.sleep(100)) @pytest.mark.asyncio async def test_no_dangling_tasks(): before = len(asyncio.all_tasks()) await leaky_operation() await asyncio.sleep(0.1) after = len(asyncio.all_tasks()) assert after == before, f"{after - before} tasks were leaked after the operation"