Kirill Pinchuk

hey @paultiplady I'm not sure that approach from `pytest-tipsi-djanjo` fits the usual testing model for `pytest`. The most noticeable difference is the finishing on fixtures: currently `pytest` doesn't explicitly finish...

Regarding finalization: https://github.com/tipsi/pytest-tipsi-testing/blob/master/tests/test_finalization.py This test doesn't work without explicit finalization, same as non-function level database fixtures. And this is about pytest implementation, so there is nothing to do in pytest-django...

@blueyed we're using such approach for more than a year (wrap everything into an additional atomic block) it works pretty well. But you cannot just drop-in something like that because...

@blueyed it definitely helps. We're using https://github.com/tipsi/pytest-tipsi-django/blob/1cd89fdd088ba5822df8bd1e8d8ebcfa60bfa918/pytest_tipsi_django/django_fixtures.py#L106 for several months and we're heavily utilizing tests nesting, scopes and transactions on different scopes. You can see example tests for it here:...

@richtier the only one thing you should keep in mind, that your fixture resets flags *before* every test run, not after. if you want to change its behavior to `after`...

I don't suggest this as a final fix. But we were able to bypass this on django 3.1 + channels 2.4 with such fixture: ```python @pytest.fixture def fix_async_db(): local =...

@vinta friendly ping, looks like this PR has reached the goal and deserves to be merged

@erezsh actually there is contribution guide line: 'Pull requests will be merged when their votes reach 20.' we have only 15 here

yeah, you're right

@vinta pydantic now also has 20, let's get it merged