[UK-Al05]

A good compromise is you write an in-memory version of the dependency and the real version.

You write a comprehensive test of integration tests against the real object.

You then set it up so those same tests can run both the in-memory and real version.

Any differences should show up for the interactions you have specified, and tests should fail.

You can write sociable unit tests against the in-memory version, knowing it matches the same behaviour as closely as you have specified in the tests.

[mumblemumble]

That's what I typically do when I must. I find it yields much more readable tests than mocking does, and they're less likely to accidentally become tautological.

That said, discrepancies can still sink in, so I think it's best to also have some baseline level of tests that run against the real dependency, even if they're typically only run overnight on the CI server.