I am working with researchers who are running a steadily growing test suite, and occasionally they are breaking the tests by having code which damages the environment by modifying something in a way that works when you run a single test but fails in the entire test suite.
What I've built before (in Java) is a test runner which alternates a given test with each test in the test run. This seems like it should be relatively easy in pytest, but figuring out the details has eluded me so far. Anything obvious I'm missing?
Aucun commentaire:
Enregistrer un commentaire