Using synchronization primitives

Synchronization primitives are objects that are used by tasks to communicate and coordinate with each other. They are useful for things like distributing workload, notifying other tasks and guarding access to shared resources.

Note

AnyIO primitives are not thread-safe, therefore they should not be used directly from worker threads. Use run_sync() for that.

Events

Events are used to notify tasks that something they’ve been waiting to happen has happened. An event object can have multiple listeners and they are all notified when the event is triggered.

Example:

from anyio import Event, create_task_group, run


async def notify(event):
    event.set()


async def main():
    event = Event()
    async with create_task_group() as tg:
        tg.start_soon(notify, event)
        await event.wait()
        print('Received notification!')

run(main)

Note

Unlike standard library Events, AnyIO events cannot be reused, and must be replaced instead. This practice prevents a class of race conditions, and matches the semantics of the Trio library.

Semaphores

Semaphores are used for limiting access to a shared resource. A semaphore starts with a maximum value, which is decremented each time the semaphore is acquired by a task and incremented when it is released. If the value drops to zero, any attempt to acquire the semaphore will block until another task frees it.

Example:

from anyio import Semaphore, create_task_group, sleep, run


async def use_resource(tasknum, semaphore):
    async with semaphore:
        print('Task number', tasknum, 'is now working with the shared resource')
        await sleep(1)


async def main():
    semaphore = Semaphore(2)
    async with create_task_group() as tg:
        for num in range(10):
            tg.start_soon(use_resource, num, semaphore)

run(main)

Tip

If the performance of semaphores is critical for you, you could pass fast_acquire=True to Semaphore. This has the effect of skipping the cancel_shielded_checkpoint() call in Semaphore.acquire() if there is no contention (acquisition succeeds immediately). This could, in some cases, lead to the task never yielding control back to to the event loop if you use the semaphore in a loop that does not have other yield points.

Locks

Locks are used to guard shared resources to ensure sole access to a single task at once. They function much like semaphores with a maximum value of 1, except that only the task that acquired the lock is allowed to release it.

Example:

from anyio import Lock, create_task_group, sleep, run


async def use_resource(tasknum, lock):
    async with lock:
        print('Task number', tasknum, 'is now working with the shared resource')
        await sleep(1)


async def main():
    lock = Lock()
    async with create_task_group() as tg:
        for num in range(4):
            tg.start_soon(use_resource, num, lock)

run(main)

Tip

If the performance of locks is critical for you, you could pass fast_acquire=True to Lock. This has the effect of skipping the cancel_shielded_checkpoint() call in Lock.acquire() if there is no contention (acquisition succeeds immediately). This could, in some cases, lead to the task never yielding control back to to the event loop if use the lock in a loop that does not have other yield points.

Conditions

A condition is basically a combination of an event and a lock. It first acquires a lock and then waits for a notification from the event. Once the condition receives a notification, it releases the lock. The notifying task can also choose to wake up more than one listener at once, or even all of them.

Like Lock, Condition also requires that the task which locked it also the one to release it.

Example:

from anyio import Condition, create_task_group, sleep, run


async def listen(tasknum, condition):
    async with condition:
        await condition.wait()
        print('Woke up task number', tasknum)


async def main():
    condition = Condition()
    async with create_task_group() as tg:
        for tasknum in range(6):
            tg.start_soon(listen, tasknum, condition)

        await sleep(1)
        async with condition:
            condition.notify(1)

        await sleep(1)
        async with condition:
            condition.notify(2)

        await sleep(1)
        async with condition:
            condition.notify_all()

run(main)

Capacity limiters

Capacity limiters are like semaphores except that a single borrower (the current task by default) can only hold a single token at a time. It is also possible to borrow a token on behalf of any arbitrary object, so long as that object is hashable.

Example:

from anyio import CapacityLimiter, create_task_group, sleep, run


async def use_resource(tasknum, limiter):
    async with limiter:
        print('Task number', tasknum, 'is now working with the shared resource')
        await sleep(1)


async def main():
    limiter = CapacityLimiter(2)
    async with create_task_group() as tg:
        for num in range(10):
            tg.start_soon(use_resource, num, limiter)

run(main)

You can adjust the total number of tokens by setting a different value on the limiter’s total_tokens property.

Resource guards

Some resources, such as sockets, are very sensitive about concurrent use and should not allow even attempts to be used concurrently. For such cases, ResourceGuard is the appropriate solution:

class Resource:
    def __init__(self):
        self._guard = ResourceGuard()

    async def do_something() -> None:
        with self._guard:
            ...

Now, if another task tries calling the do_something() method on the same Resource instance before the first call has finished, that will raise a BusyResourceError.

Queues

In place of queues, AnyIO offers a more powerful construct: memory object streams.