How to use AmbiguousWithPrimitive method of org.mockitousage.constructor.CreatingMocksWithConstructorTest class

You can't instantiate an abstract class directly, but you can easily instantiate an anonymous subclass, and that's pretty trivial in this case as you're not forced to define any methods:

Mockito.when(mockDAO.getDatabaseResults()).thenThrow(new ProblemException(){});

Alternatively, you can just use another mock:

Mockito.when(mockDAO.getDatabaseResults()).thenThrow(Mockito.mock(ProblemException.class));

This would make sense if the abstract class forced the definition of other methods you don't care about in your test.

There is no real need for @Mock in Spock, because there is already = Mock(), which can be used everywhere an annotation can be used (and also in other places). There is an open pull request for @InjectMocks, but it hasn't been decided if such a feature will make it into spock-core or spock-guice. (Shipping this feature with spock-guice, or at least requiring Guice on the class path, would allow to delegate injection to Guice, rather than reinventing the wheel.) If not, @InjectMocks could always be shipped as a third-party Spock extension.

any() returns null. This choice was inevitable because the signature of any() is

public static <T> T any()

Since generic types are erased in Java, and since nothing is passed to the method that carries any type information (such as a Class<T>), null is the only reasonable return value.

This creates a problem if the method of the mock has an argument of primitive type because unboxing this null value throws a NullPointerException.

There are two possible solutions. You can either use the primitive versions (such as anyDouble()), or the version accepting a Class (e.g. any(Double.class)). In the latter case, since we are supplying type information to the method, it is possible to use this information to return a sensible non-null value (e.g. 0.0D in the case of double).

If you want the mocked method to call a method on (or otherwise alter) a parameter, you'll need to write an Answer as in this question ("How to mock a void return method affecting an object").

From Kevin Welker's answer there:

doAnswer(new Answer() {
    Object answer(InvocationOnMock invocation) {
        Object[] args = invocation.getArguments();
        ((MyClass)args[0]).myClassSetMyField(NEW_VALUE);
        return null; // void method, so return null
    }
}).when(mock).someMethod();

Note that newer best-practices would have a type parameter for Answer, as in Answer<Void>, and that Java 8's lambdas can compress the syntax further. For example:

doAnswer(invocation -> {
  Object[] args = invocation.getArguments();
  ((MyClass)args[0]).myClassSetMyField(NEW_VALUE);
  return null; // void method in a block-style lambda, so return null
}).when(mock).someMethod();

when(myObject.someMethod()).thenReturn(someMockObject);
// run test
verify(myObject, times(n)).someMethod();

See the documentation for more conversion examples.

Here you want to test EventMonitor.doStuff(param) and while executing this method you want to make it sure whether the correct methods on IEventListener were invoked or not.

So, in order to test doStuff(param), you don't need to have a real implementation of IEventListener: what you need is just a mock implementation of IEventListener and you have to verify the exact number of method invocations on IEventListener while testing doStuff. This can be achieved with Mockito or any other mocking framework.

Here is an example with Mockito:

import static org.junit.Assert.assertEquals;
import static org.mockito.Mockito.times;
import static org.mockito.Mockito.verify;

import org.junit.Before;
import org.junit.Test;
import org.mockito.InjectMocks;
import org.mockito.Mock;
import org.mockito.MockitoAnnotations;

public class EventMonitorTest {

    //This will create a mock of IEventListener
    @Mock
    IEventListener eventListener;

    //This will inject the "eventListener" mock into your "EventMonitor" instance.
    @InjectMocks
    EventMonitor eventMonitor = new EventMonitor();

    @Before
    public void initMocks() {
        //This will initialize the annotated mocks
        MockitoAnnotations.initMocks(this);
    }

    @Test
    public void test() {
        eventMonitor.doStuff(param);
        //Here you can verify whether the methods "doStuff1" and "doStuff2" 
        //were executed while calling "eventMonitor.doStuff". 
        //With "times()" method, you can even verify how many times 
        //a particular method was invoked.
        verify(eventListener, times(1)).doStuff1();
        verify(eventListener, times(0)).doStuff2();
    }

}

If the contract of class A includes the fact that it calls method B of an object of type C, then you should test this by making a mock of type C, and verifying that method B has been called.

This implies that the contract of class A has sufficient detail that it talks about type C (which might be an interface or a class). So yes, we're talking about a level of specification that goes beyond just "system requirements", and goes some way to describing implementation.

This is normal for unit tests. When you are unit testing, you want to ensure that each unit is doing the "right thing", and that will usually include its interactions with other units. "Units" here might mean classes, or larger subsets of your application.

Update:

I feel that this doesn't apply just to verification, but to stubbing as well. As soon as you stub a method of a collaborator class, your unit test has become, in some sense, dependent on implementation. It's kind of in the nature of unit tests to be so. Since Mockito is as much about stubbing as it is about verification, the fact that you're using Mockito at all implies that you're going to run across this kind of dependency.

In my experience, if I change the implementation of a class, I often have to change the implementation of its unit tests to match. Typically, though, I won't have to change the inventory of what unit tests there are for the class; unless of course, the reason for the change was the existence of a condition that I failed to test earlier.

So this is what unit tests are about. A test that doesn't suffer from this kind of dependency on the way collaborator classes are used is really a sub-system test or an integration test. Of course, these are frequently written with JUnit too, and frequently involve the use of mocking. In my opinion, "JUnit" is a terrible name, for a product that lets us produce all different types of test.

How about

when( method-call ).thenReturn( value1, value2, value3 );

You can put as many arguments as you like in the brackets of thenReturn, provided they're all the correct type. The first value will be returned the first time the method is called, then the second answer, and so on. The last value will be returned repeatedly once all the other values are used up.

Two more ways to do it (see my comment on the previous answer by @Tomasz Nurkiewicz):

The first relies on the fact that the compiler simply won't let you pass in something of the wrong type:

when(a.method(any(Class.class))).thenReturn(b);

You lose the exact typing (the Class<? extends A>) but it probably works as you need it to.

The second is a lot more involved but is arguably a better solution if you really want to be sure that the argument to method() is an A or a subclass of A:

when(a.method(Matchers.argThat(new ClassOrSubclassMatcher<A>(A.class)))).thenReturn(b);

Where ClassOrSubclassMatcher is an org.hamcrest.BaseMatcher defined as:

public class ClassOrSubclassMatcher<T> extends BaseMatcher<Class<T>> {

    private final Class<T> targetClass;

    public ClassOrSubclassMatcher(Class<T> targetClass) {
        this.targetClass = targetClass;
    }

    @SuppressWarnings("unchecked")
    public boolean matches(Object obj) {
        if (obj != null) {
            if (obj instanceof Class) {
                return targetClass.isAssignableFrom((Class<T>) obj);
            }
        }
        return false;
    }

    public void describeTo(Description desc) {
        desc.appendText("Matches a class or subclass");
    }       
}

Phew! I'd go with the first option until you really need to get finer control over what method() actually returns :-)