We don't write backends so we shouldn't have anything to unit test. By far the most of the code we write deals with GUI and thus should be tested using Autopilot.
Even if we do have something to *unit* test, the added complexity, higher entry barrier and the time spent writing first the unit testing framework and then the tests is unlikely to pay off. If you disagree with the above, you're welcome to prove me wrong though. 200+ messages to mailing list have not convinced any of the core devs so far. You might have better luck convincing people if you come up with a unit testing framework, cover an app with unit tests (if you manage to find something to *unit* test in the first place) and demonstrate that it really actually saves time in a real-life scenario. But I expect simply helping us figure out Autopilot and write UI-driven tests to be much more productive. Either way, the choice is yours. -- Sergey "Shnatsel" Davidoff OS architect @ elementary
-- Mailing list: https://launchpad.net/~elementary-dev-community Post to : elementary-dev-community@lists.launchpad.net Unsubscribe : https://launchpad.net/~elementary-dev-community More help : https://help.launchpad.net/ListHelp
