I see all the examples that demonstrate unit testing code and mocking the calls to the DB since you are not suppose to touch the DB. But it seems to me having a set up tasks that uses the actually schema, loads the lookup tables and then populates it with data using the methods to be tested… This way it is more real world testing and all the stored procedures are tested as well.. But I never have seen any examples like this… Is there some reason I do not understand why using this technique is not as useful as it seems ?

When doing unit tests the “proper” way, i.e. stubbing every public call and return preset values or mocks, I feel like I’m not actually testing anything. I’m literally looking at my code and creating examples based on the flow of logic through my public methods. And every time the implementation changes, I have to go and change those tests, again, not really feeling that I’m accomplishing anything useful (be it mid- or long-term). I also do integration tests (including non-happy-paths) and I don’t really mind the increased testing times. With those, I feel like I’m actually testing for regressions, because they have caught multiple, while all that unit tests do is show me that the implementation of my public method changed, which I already know.