Testing event-driven systems
Event-driven systems are very flexible and, in certain situations, can be incredibly useful in detaching different elements. But this flexibility and detachment can make them difficult to test to ensure that everything works as expected.
In general, unit tests are the fastest tests to generate, but the detached nature of event-driven systems makes them not very useful to properly test the reception of events. Sure, the events can be simulated, and the general behavior of receiving an event can be tested. But the problem is: how can we ensure that the event has been properly generated? And at the right moment?
The only option is to use integration tests to check the behavior of the system. But these tests are more expensive to design and run.
There's always an endless debate about naming tests, what exactly a unit test is compared to an integration test, system test, acceptance test, and so on. To avoid getting into too deep a discussion...
