bpo-30490: Allow to the Event.set method pass an exception

[pfreixes]

Having the Event as the way to synchronize 1:N coroutines, the none happy path should be able to be expressed making possible call the set_exception for each future related to each waiter.

As an example the following code trying to implement a way to avoid the dogpile effect for a DNS cache. If the coro that holds the event fails, the original exception is also broadcasted to the waiters.

if key in throttle_dns_events:
    yield from throttle_dns_events[key].wait()
else:
    throttle_dns_events[key] = Event(loop=loop)
    try:
        addrs = yield from \
            resolver.resolve(host, port, family=family)
        cached_hosts.add(key, addrs)
        throttle_dns_events[key].set()
    except Exception as e:
        # any DNS exception, independently of the implementation
        # is set for the waiters to raise the same exception.
        throttle_dns_events[key].set(exc=e)
        raise
    finally:
        throttle_dns_events.pop(key)

[mention-bot]

@pfreixes, thanks for your PR! By analyzing the history of the files in this pull request, we identified @1st1, @asvetlov and @serhiy-storchaka to be potential reviewers.

[pfreixes]

Any thoughts about this proposal @1st1 ?

[1st1]

asyncio.Event is designed to be similar to threading.Event, and the similarity of asyncio synchronization primitives to those of the threading module is explicitly expressed in the docs. So IMO, if we want this in asyncio, we should also add this to the threading module. Would you mind looking through python-ideas archives, and if nothing is there, post this idea to discuss it on the mailing list?

[pfreixes]

I will do that, thanks El 18/07/2017 00:29, "Yury Selivanov" <notifications@github.com> escribió:

…

asyncio.Event is designed to be similar to threading.Event, and the similarity of asyncio synchronization primitives to those of the threading module is explicitly expressed in the docs. So IMO, if we want this in asyncio, we should also add this to the threading module. Would you mind looking through python-ideas archives, and if nothing is there, post this idea to discuss it on the mailing list? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1824 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABK1ieJGBwmd-RUps6c3akjjdFia_v-eks5sO-BRgaJpZM4NoDwx> .

[pfreixes]

@1st1 I'm gonna close this PR and the bug opened. I've ended up using a simple solution [1] via a wrapper that came up in that discussion [2]

[1] aio-libs/aiohttp@11afb32
[2] https://mail.python.org/pipermail/python-ideas/2017-July/046486.html

[1st1]

Thanks! Yeah, it's probably easier to write a wrapper than change both threading and asyncio modules in this case.