gh-134471: Fix asyncio.timeout(0) swallowing an unrelated prior cancellation.
#134472
Conversation
|
The following commit authors need to sign the Contributor License Agreement: |
|
Most changes to Python require a NEWS entry. Add one using the blurb_it web app or the blurb command-line tool. If this change has little impact on Python users, wait for a maintainer to apply the |
| async def test_timeout_immediately_after_cancellation(self): | ||
| asyncio.current_task().cancel() | ||
| with self.assertRaises(asyncio.CancelledError): | ||
| async with asyncio.timeout(0.0): | ||
| await asyncio.sleep(1) |
There was a problem hiding this comment.
this added test fails on all python versions and the proposed fix makes it pass without breaking any other tests
| self._state = _State.ENTERED | ||
| self._task = task | ||
| self._cancelling = self._task.cancelling() | ||
| self._must_cancel = self._task._must_cancel |
There was a problem hiding this comment.
i feel uneasy messing around with this internal _must_cancel but i didn't find any other way to fix this. changing the cancellation counting logic breaks timeout usage while handling CancelledError, where it should indeed disregard the previous cancellation (the one it's handling). I couldn't get both cases to work by only relying on counting (cancelling).
so i needed something more "surgical" to be able to tell whether we're currently handling CancelledError or we're just in an already cancelled task that didn't get thrown a CancelledError yet
1 participant
asyncio.timeout(0)swallows a prior task cancellation #134471