Unit testing

Unit testing is a method by which individual pieces of source code are tested to verify that the design and implementation for that unit have been correctly implemented. There are four phases executed in sequence in a unit test case are the following:

• Setup: The test case initializes the test fixture, that is the before picture required for the SUT to exhibit the expected behavior.
• Exercise: The test case interacts with the SUT, getting some outcome from it as a result. The SUT usually queries another component, named the Depended-On Component (DOC).
• Verify: The test case determines whether the expected outcome has been obtained using assertions (also known as predicates).
• Teardown: The test case tears down the test fixture to put the SUT back into the initial state.

These phases and its relationship with the SUT and DOC is illustrated as follows:

Unit testing is done with the unit under test in isolation, that is, without interacting its DOCs. To that aim, test doubles are employed to replace any components on which the SUT depends. There are several kinds of test doubles:

• A dummy object simply satisfies the real object API but it is never actually used. The typical use case for dummy objects is when they are passed as parameters to meet the method signature, but then the dummy object is not actually used.
• A fake object replaces the real object with a simpler implementation, for example, an in-memory database.
• A stub object replaces the real object providing hard-coded values as responses.
• A mock object also replaces the real object, but this time with programmed expectations as responses.
• A spy object is a partial mock object, meaning that some of its methods are programmed with expectations, but the others use the real object's implementation.