bpo-32972: Async test case #13386
bpo-32972: Async test case #13386
Conversation
| self.tearDown() | ||
|
|
||
| def _callCleanup(self, function, *args, **kwargs): | ||
| self._callMaybeAsync(function, *args, **kwargs) |
There was a problem hiding this comment.
Why don't we have an explicit addAsyncCleanup?
There was a problem hiding this comment.
For example, we have an explicit async method in AsyncExitStack: https://docs.python.org/3/library/contextlib.html#contextlib.AsyncExitStack.push_async_callback
Even though we toyed with the idea of just reusing the push_callback method and checking if the returned value is an awaitable.
I'm more in favour of addAsyncCleanup in this case.
There was a problem hiding this comment.
I can add addAsyncCleanup() method, it makes sense.
The implementation should push a callback into the same cleanup queue as used for addCleanup() though, to keep cleanups order.
I think running all sync cleanups before asyncs and vice versa is a bad idea.
Cleanups should be executed exactly in the reverse adding order.
Also, self.doCleanups() should execute all sync and async cleanups altogether.
Do you agree?
|
this would be great to see in 3.8 |
1st1
left a comment
1st1
left a comment
There was a problem hiding this comment.
Please add addAsyncCleanup() before merging.
|
|
Add explicit `asyncSetUp` and `asyncTearDown` methods. The rest is the same as for python#13228 `AsyncTestCase` create a loop instance for every test for the sake of test isolation. Sometimes a loop shared between all tests can speed up tests execution time a lot but it requires control of closed resources after every test finish. Basically, it requires nested supervisors support that was discussed with @1st1 many times. Sorry, asyncio supervisors have no chance to land on Python 3.8. The PR intentionally does not provide API for changing the used event loop or getting the test loop: use `asyncio.set_event_loop_policy()` and `asyncio.get_event_loop()` instead. The PR adds four overridable methods to base `unittest.TestCase` class: ``` def _callSetUp(self): self.setUp() def _callTestMethod(self, method): method() def _callTearDown(self): self.tearDown() def _callCleanup(self, function, /, *args, **kwargs): function(*args, **kwargs) ``` It allows using asyncio facilities with minimal influence on the unittest code. The last but not least: the PR respects contextvars. The context variable installed by `asyncSetUp` is available on test, `tearDown` and a coroutine scheduled by `addCleanup`. https://bugs.python.org/issue32972
6 participants
Add explicit
asyncSetUpandasyncTearDownmethods.The rest is the same as for #13228
AsyncTestCasecreate a loop instance for every test for the sake of test isolation.Sometimes a loop shared between all tests can speed up tests execution time a lot but it requires control of closed resources after every test finish. Basically, it requires nested supervisors support that was discussed with @1st1 many times. Sorry, asyncio supervisors have no chance to land on Python 3.8.
The PR intentionally does not provide API for changing the used event loop or getting the test loop: use
asyncio.set_event_loop_policy()andasyncio.get_event_loop()instead.The PR adds four overridable methods to base
unittest.TestCaseclass:It allows using asyncio facilities with minimal influence on the unittest code.
The last but not least: the PR respects contextvars. The context variable installed by
asyncSetUpis available on test,tearDownand a coroutine scheduled byaddCleanup.https://bugs.python.org/issue32972