Closed krisgesling closed 2 years ago
Voight Kampff Integration Test Succeeded (Results)
For me this is fine.
I wonder if the start of core also should be CI only, to keep it symmetric? Though it doesn't really make much practical difference since the start will work just fine when already started.
I wonder if the start of core also should be CI only, to keep it symmetric? Though it doesn't really make much practical difference since the start will work just fine when already started.
Yeah I debated that in my head. My reasoning was that:
I totally buy that reasoning
Description
When running tests locally, a reasonable amount of time is spent starting and stopping Mycroft services. Rapidly doing this can also cause broader issues. Shutting down services at the end of a Voight Kampff test run in the CI makes sense, but it doesn't make sense for me personally when running locally.
This is the simplest implementation which provides one behaviour for CI, and the other behaviour for "not-CI". Is this sufficient or do we need to add in one or more arguments? eg:
or
How to test
Run some VK tests locally and see that Mycroft is still running after the tests complete. You can now run the tests repeatedly, much faster than you could previously.
Contributor license agreement signed?