Chris NeJame
Chris NeJame
I'd be happy to :grin: There's some already existing tests I can make tweaked copies of to cover this. Sorry for causing all the ruckus, btw.
>I'd rather avoid new tests if existing ones can be extended / hardened - the test suite is rather slow already, especially due to all the runpytest_subprocess tests. Not a...
Here's the PR https://github.com/pytest-dev/pytest-django/pull/807
@blueyed Thanks for checking this out! > I am not really sure if that should be changed. While it might not be a typical use case it would be good...
Ah, great catch! I'll add in a fix for it to make them deterministic
I think it's upset about the coverage because I added a yield statement to get it to not complain about the other yield statement haha. But I'm not really sure...
Ah, of course! This ran with pytest 5.3.4, which _doesn't_ have my change in pytest! I'll pin 5.3.3 for this branch temporarily, but while I'm confident I'll hit that line,...
~~Definitely shouldn't be merged as I"m investigating an issue with postgres that's potentially related~~ Edit: all is good now
I'm narrowing down the problem, although it's a slow process. Luckily, I've been able to finally reproduce it on my end so I can test a lot faster (I had...
@blueyed Looks like https://github.com/pytest-dev/pytest/pull/6551 did the trick. After some minor tweaking of this branch, everything is passing, except for certain things, which looks like a change in terms of what's...