How to use getSourceFile method of org.mockito.internal.invocation.InvocationBuilder class

Try the anyVararg() matcher. This was introduced in 1.8.1.

The thenReturn method will always return what is passed to it. The code new Servicemock() is being executed prior to the call to thenReturn. The created ServiceMock is then being passed to thenReturn. Therefore thenReturn has a absolute instance of ServiceMock not a creation mechanism.

If you need to provide an new instance, use thenAnswer

when(mockedMergeContext.createNewEntityOfType(IService.class))
  .thenAnswer(new Answer<IService>() {
     public IService answer(InvocationOnMock invocation) {
        return new ServiceMock();
     }
   });

Ok, I think I found the answer, but it was a headache. Rudy gave me the final hint with using using a spy, but it was still not trivial (although the solution looks "baby-easy"). Here is the complete solution:

import static org.mockito.Matchers.anyString;
import static org.mockito.Mockito.times;
import static org.powermock.api.mockito.PowerMockito.verifyPrivate;
import static org.powermock.api.mockito.PowerMockito.doNothing;
import static org.powermock.api.mockito.PowerMockito.spy;

@RunWith(PowerMockRunner.class)
@PrepareForTest(Factory.class)
public class Tests extends TestCase {

    public void testFactory() throws Exception {

        Factory factorySpy = spy(new Factory());
        String s = factorySpy.factorObject();
        doNothing().when(factorySpy, "checkString", anyString());
        verifyPrivate(factorySpy, times(1)).invoke("checkString", anyString()); 
    }
}

The message is pretty obvious: You can't mock non-visible and final classes. Short answer : Create a named class of your anonymous one, and test this class instead!

Long answer, let's dig why !

An anonymous class is final

You instantiate an anonymous class of FilterFactory, when the compiler sees an anonymous class, it creates a final and package visible class. So the anonymous class is not mockable through standard mean i.e. through Mockito.

Mocking anonymous class : possible but BRITTLE if not HACKY

OK, now suppose you want to be able to mock this anonymous class through Powermock. Current compilers compile anonymous class with following scheme :

Declaring class + $ + <order of declaration starting with 1>

Mocking anonymous class possible but brittle (And I mean it) So supposing the anonymous class is the eleventh to be declared, it will appear as

InputHelper$11.class

So you could potentially prepare for test the anonymous class:

@RunWith(PowerMockRunner.class)
@PrepareForTest({InputHelper$11.class})
public class InputHelperTest {
    @Test
    public void anonymous_class_mocking works() throws Throwable {
        PowerMockito.spy(InputHelper.BZIP2_FACTORY);  // This line fails
    }
}

This code will compile, BUT will eventually be reported as an error with your IDE. The IDE probably doesn't know about InputHelper$11.class. IntelliJ who doesn't use compiled class to check the code report so.

Also the fact that the anonymous class naming actually depends on the order of the declaration is a problem, when someone adds another anonymous class before, the numbering could change. Anonymous classes are made to stay anonymous, what if the compiler guys decide one day to use letters or even random identifiers!

So mocking anonymous classes through Powermock is possible but brittle, don't ever do that in a real project!

EDITED NOTE : The Eclipse compiler has a different numbering scheme, it always uses a 3 digit number :

Declaring class + $ + <pad with 0> + <order of declaration starting with 1>

Also I don't think the JLS clearly specify how the compilers should name anonymous classes.

You don't reassign the spy to the static field

PowerMockito.spy(InputHelper.BZIP2_FACTORY);  // This line fails
whenNew(BufferedInputStream.class).withArguments(in).thenReturn(buffer);
whenNew(CBZip2InputStream.class).withArguments(buffer).thenReturn(expected);
InputStream observed = InputHelper.BZIP2_FACTORY.makeFilter(in);

PowerMockito.spy returns the spy, it doesn't change the value of InputHelper.BZIP2_FACTORY. So you would need to actually set via reflection this field. You can use the Whiteboxutility that Powermock provide.

Conclusion

Too much trouble to just test with mocks that the anonymous filter uses a BufferedInputStream.

Alternative

I would rather write the following code:

An input helper that will use the named class, I don't use the interface name to make clear to the user what is the intent of this filter!

public class InputHelper {
    public static final BufferedBZIP2FilterFactory BZIP2_FACTORY = new BufferedBZIP2FilterFactory();
}

And now the filter itself :

public class BufferedBZIP2FilterFactory {
    public InputStream makeFilter(InputStream in) {
        BufferedInputStream buffer = new BufferedInputStream(in);
        return new CBZip2InputStream(buffer);
    }
}

Now you can write a test like this :

@RunWith(PowerMockRunner.class)
public class BufferedBZIP2FilterFactoryTest {

    @Test
    @PrepareForTest({BufferedBZIP2FilterFactory.class})
    public void wraps_InputStream_in_BufferedInputStream() throws Exception {
        whenNew(CBZip2InputStream.class).withArguments(isA(BufferedInputStream.class))
                .thenReturn(Mockito.mock(CBZip2InputStream.class));

        new BufferedBZIP2FilterFactory().makeFilter(anInputStream());

        verifyNew(CBZip2InputStream.class).withArguments(isA(BufferedInputStream.class));
    }

    private ByteArrayInputStream anInputStream() {
        return new ByteArrayInputStream(new byte[10]);
    }
}

But could eventually avoid powermock stuff for this test scenario if you force the CBZip2InputStream to only accept BufferedInputStream. Usually using Powermock means something is wrong with the design. In my opinion Powermock is great for legacy softwares, but can blind developers when designing new code; as they are missing the point of OOP's good part, I would even say they are designing legacy code.

Hope that helps !

Mockito/Hamcrest and generic classes

Yes, this is a general problem with Mockito/Hamcrest. Generally using isA() with generic classes produces a warning.

There are predifined Mockito matchers for the most common generic classes: anyList(), anyMap(), anySet() and anyCollection().

Suggestions:

anyIterable() in Mockito 2.1.0

Mockito 2.1.0 added a new anyIterable() method for matching Iterables:

when(client.runTask(anyString(), anyString(), anyIterable()).thenReturn(...)

Ignore in Eclipse

If you just want to get rid of the warning in Eclipse. Option exists since Eclipse Indigo:

Window > Preferences > Java > Compiler > Errors/Warnings > Generic types > Ignore unavoidable generic type problems

Quick Fix with @SuppressWarnings

I suggest you do this if you have the problem only once. I personally don't remember ever needing an isA(Iterable.class).

As Daniel Pryden says, you can limit the @SuppressWarnings to a local variable or a helper method.

Use a generic isA() matcher with TypeToken

This solves the problem for good. But it has two disadvantages:

• The syntax is not too pretty and might confuse some people.
• You have an additional dependency on the library providing the TypeToken class. Here I used the TypeToken class from Guava. There's also a TypeToken class in Gson and a GenericType in JAX-RS.

Using the generic matcher:

import static com.arendvr.matchers.InstanceOfGeneric.isA;
import static org.mockito.ArgumentMatchers.argThat;

// ...

when(client.runTask(anyString(), anyString(), argThat(isA(new TypeToken<Iterable<Integer>>() {}))))
            .thenReturn(...);

Generic matcher class:

package com.arendvr.matchers;

import com.google.common.reflect.TypeToken;
import org.mockito.ArgumentMatcher;

public class InstanceOfGeneric<T> implements ArgumentMatcher<T> {
    private final TypeToken<T> typeToken;

    private InstanceOfGeneric(TypeToken<T> typeToken) {
        this.typeToken = typeToken;
    }

    public static <T> InstanceOfGeneric<T> isA(TypeToken<T> typeToken) {
        return new InstanceOfGeneric<>(typeToken);
    }

    @Override
    public boolean matches(Object item) {
        return item != null && typeToken.getRawType().isAssignableFrom(item.getClass());
    }
}

Yes, you can. The trick here is that you have to get to the instance of the lambda that is passed to the registerMessage and then execute that expression and then you can verify the result.

For the purpose of a meaningful example I created this Handler class that contains the dispatchMessage that you want to test:

public class Handler {

    private Dispatcher dispatcher = new Dispatcher();

    public void dispatchMessage(MessageHandler handler, String argument1, String argument2, Long argument3) {

        handler.registerMessage(() -> {
            dispatcher.dispatch(argument1,
                    argument2,
                    argument3);
        });

    }

    interface MessageHandler {
        void registerMessage(Runnable run);
    }

    static class Dispatcher {
        void dispatch(String a, String b, long c){
            // Do dispatch
        }
    }
}

What you have to remember is that a lambda expression is just a short hand form to pass a function to a method. In this example the function is the run method of a Runnable. Therefore the method registerMessage of the interface for MessageHandler takes a Runnable as it's argument. I also included an implementation for the Dispatcher, which is called from within registerMessage. The test for this looks like this:

@RunWith(MockitoJUnitRunner.class)
public class HandlerTest {
    @Mock
    private Dispatcher dispatcher;
    @InjectMocks
    private Handler classUnderTest;
    @Captor
    private ArgumentCaptor<Runnable> registerMessageLambdaCaptor;

    @Test
    public void shouldCallDispatchMethod() {
        final String a = "foo";
        final String b = "bar";
        final long c = 42L;

        MessageHandler handler = mock(MessageHandler.class);

        classUnderTest.dispatchMessage(handler, a, b, c);

        verify(handler).registerMessage(registerMessageLambdaCaptor.capture());

        Runnable lambda = registerMessageLambdaCaptor.getValue();

        lambda.run();

        verify(dispatcher).dispatch(a, b, c);
    }
}

There is an ArgumentCaptor for the lambda expression which we use in the first verification of the registerMessage. After that verification we can retrieve the lambda expression from the captor. The type of the lambda expression is Runnable, as defined in the MessageHandler interface. Hence we can call the run method on it and then verify that the dispatch method on the Dispatcher was called with all the appropriate arguments.

Since I use Spring Boot 1, I solved it this way:

@WebMvcTest(value = ResourceController.class, secure = false)

Please note that for the newer version of Spring Boot 2 the secure option is not available anymore, so use this:

@WebMvcTest(value = ResourceController.class,  excludeAutoConfiguration = {SecurityAutoConfiguration.class})

New Answer
(October 2012)

There's no need to manually build the classpath list. Java supports a convenient wildcard syntax for directories containing jar files.

java -cp "$LIB/*"

(Notice that the * is inside the quotes.)

Explanation from man java:

As a special convenience, a class path element containing a basename of * is considered equivalent to specifying a list of all the files in the directory with the extension .jar or .JAR (a java program cannot tell the difference between the two invocations).

For example, if directory foo contains a.jar and b.JAR, then the class path element foo/* is expanded to a A.jar:b.JAR, except that the order of jar files is unspecified. All jar files in the specified directory, even hidden ones, are included in the list. A classpath entry consisting simply of * expands to a list of all the jar files in the current directory. The CLASSPATH environment variable, where defined, will be similarly expanded. Any classpath wildcard expansion occurs before the Java virtual machine is started — no Java program will ever see unexpanded wildcards except by querying the environment.

---

Old Answer

Good

Simple but not perfect solution:

CLASSPATH=$(echo "$LIB"/*.jar | tr ' ' ':')

There's a slight flaw in that this will not handle file names with spaces correctly. If that matters try this slightly more complicated version:

Better

CLASSPATH=$(find "$LIB" -name '*.jar' -printf '%p:' | sed 's/:$//')

This only works if your find command supports -printf (as GNU find does).

If you don't have GNU find, as on Mac OS X, you can use xargs instead:

CLASSPATH=$(find "." -name '*.jar' | xargs echo | tr ' ' ':')

Best?

Another (weirder) way to do it is to change the field separator variable $IFS. This is very strange-looking but will behave well with all file names and uses only shell built-ins.

CLASSPATH=$(JARS=("$LIB"/*.jar); IFS=:; echo "${JARS[*]}")

Explanation:

1. JARS is set to an array of file names.
2. IFS is changed to :.
3. The array is echoed, and $IFS is used as the separator between array entries. Meaning the file names are printed with colons between them.

All of this is done in a sub-shell so the change to $IFS isn't permanent (which would be baaaad).

An alternative approach would be to use the 'method to class' pattern.

1. Move the methods out of the customer class into another class/classes, say CustomerSomething eg/CustomerFinances (or whatever it's responsibility is).
2. Add a constructor to Customer.
3. Now you don't need to mock Customer, just the CustomerSomething class! You may not need to mock that either if it has no external dependencies.

Here's a good blog on the topic: https://simpleprogrammer.com/back-to-basics-mock-eliminating-patterns/