Fix call to pytest.mark.usefixtures without arguments

[daria-shaw]

This will raise a warning in an upcoming change in pytest.

[jsa34]

I think we need to either introduce a sort of "ScenarioFunction" protocol for mypy to be happy and expect scenario, or some short term ignoring of the mypy error, unless I am missing something, @youtux ?

[The-Compiler]

It's not quite clear to me how this change leads to mypy suddenly seeing the type of scenario_wrapper. It was Any before, and now mypy correctly sees that it's actually Callable[[FixtureRequest, dict[str, str]], Any] aka

def (request: _pytest.fixtures.FixtureRequest, _pytest_bdd_example: builtins.dict[builtins.str, builtins.str]) -> Any

Given the decorator is typed, this should already have been the case before I believe?

As this isn't really related to this PR and we do a return cast(Callable[P, T], scenario_wrapper) at the end already anyways, in the scope of this PR it seems reasonable to me to ignore the error.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.95%. Comparing base (93e446e) to head (a4bf843).
Report is 126 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #746   +/-   ##
=======================================
  Coverage   95.95%   95.95%           
=======================================
  Files          55       55           
  Lines        2199     2200    +1     
  Branches      242      242           
=======================================
+ Hits         2110     2111    +1     
  Misses         54       54           
  Partials       35       35           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[youtux]

I think we need to either introduce a sort of "ScenarioFunction" protocol for mypy to be happy and expect scenario, or some short term ignoring of the mypy error, unless I am missing something, @youtux ?

Actually I wanted to fix that in a different way. I started the work in #658, but never got to finish it...

[jsa34]

I think we need to either introduce a sort of "ScenarioFunction" protocol for mypy to be happy and expect scenario, or some short term ignoring of the mypy error, unless I am missing something, @youtux ?

Actually I wanted to fix that in a different way. I started the work in #658, but never got to finish it...

Ah, yes! I remember this now!