I'm trying to verify that the class I'm testing calls the correct dependency class's method. So I'm trying to match the method parameters, but I don't really care about the actual values in this test, because I don't want to make my test brittle.
However, I'm running into trouble setting it up because Mockito has decided that the behaviour I'm expecting is a bug: https://github.com/mockito/mockito/issues/134
So what't the correct way to define an ArgumentMatcher for a parameter that might be null?
With issue #134 "fixed", this code fails because the matchers only match in the first case. How can I define a matcher to work in all 4 cases?
MyClass c = mock(MyClass.class);
c.foo("hello", "world");
c.foo("hello", null);
c.foo(null, "world");
c.foo(null, null);
verify(c, times(4)).foo(anyString(), anyString());
From the javadocs of any()
Since Mockito 2.1.0, only allow non-null String. As this
is a nullable reference, the suggested API to match
null wrapper would be isNull(). We felt this
change would make tests harness much safer that it was with Mockito
1.x.
So, the way to match nullable string arguments is explicit declaration:
nullable(String.class)
I got this to work by switching to any(String.class)
I find this a bit misleading, because the API seems to suggest that anyString() is just an alias for any(String.class) at least up till the 2.0 update. To be fair, the documentation does specify that anyString() only matches non-null strings. It just seems counter-intuitive to me.
How about:
verify(c, times(4)).foo(anyObject(), anyObject());
Does that work for you?
Matchers.anyObject() allows for nulls.
Reference in Mockito docs:
Related
Sorry to all you Groovy dudes if this is a bit of a noob question.
In SOAPUI, i can create a Groovy script where i can define an arbitrary variable to the run context to retrieve at a later time.
context.previouslyUndefinedVariable = 3
def num = context.previouslyUndefinedVariable
What feature of Groovy allows previously undefined variables to be added to an object like this? I would like to learn more about it.
Many thanks in advance!
Groovy has the ability to dynamically add methods to a class through metaprogramming.
To learn more, see:
What is Groovy's MetaClass used for?
Groovy Goodness: Add Methods Dynamically to Classes with ExpandoMetaClass
Runtime and compile-time metaprogramming
The accepted answer is a bit of a poor explanation for how SoapUI is doing it.
In this case, context is always an instance of some SoapUI library java class (such as WsdlTestRunContext), and these are all implementations of Map. You can check context.getClass() and assert context in Map.
When you look up a property on a Map, Groovy uses the getAt and putAt methods. There are various syntaxes you can use. All of these are equivalent:
context.someUndef
context.'someUndef'
context[someUndef]
context['someUndef']
context.getAt('someUndef')
And
context.someUndef = 3
context.'someUndef' = 3
context[someUndef] = 3
context['someUndef'] = 3
context.putAt('someUndef', 3)
I like to use any of the above that include quote marks, so that Groovy-Eclipse doesn't flag it as a missing property.
It's also interesting that Groovy looks for a getAt() method before it checks for a get method being referred to as a property.
For example, consider evaluating "foo".class. The String instance doesn't have a property called class and it also doesn't have a method getAt(String), so the next thing it tries is to look for a "get" method with that name, i.e. getClass(), which it finds, and we get our result: String.
But with a map, ['class':'bar'].class refers to the method call getAt('class') first, which will be 'bar'. If we want to know what type of Map it is, we have to be more specific and write in full: ['class':'bar'].getClass() which will be LinkedHashMap.
We still have to specify getClass() even if that Map doesn't have a matching key, because ['foo':'bar'].class will still mean ['foo':'bar'].getAt('class'), which will be null.
I am mocking an interface array which throws java.lang.IllegalArgumentException: Cannot subclass final class class.
Following are the changes I did.
Added the following annotations at class level in this exact order:
#Runwith(PowerMockRunner.class)
#PrepareForTest({ Array1[].class, Array2[].class })
Inside the class I am doing like this:
Array1[] test1= PowerMockito.mock(Array1[].class);
Array2[] test2= PowerMockito.mock(Array2[].class);
and inside test method:
Mockito.when(staticclass.somemethod()).thenReturn(test1);
Mockito.when(staticclass.somediffmethod()).thenReturn(test2);
Basically I need to mock an array of interfaces.
Any help would be appreciated.
Opening up another perspective on your problem: I think you are getting unit tests wrong.
You only use mocking frameworks in order to control the behavior of individual objects that you provide to your code under test. But there is no sense in mocking an array of something.
When your "class under test" needs to deal with some array, list, map, whatever, then you provide an array, a list, or a map to it - you just make sure that the elements within that array/collection ... are as you need them. Maybe the array is empty for one test, maybe it contains a null for another test, and maybe it contains a mocked object for a third test.
Meaning - you don't do:
SomeInterface[] test1 = PowerMock.mock() ...
Instead you do:
SomeInterface[] test1 = new SomeInterface[] { PowerMock.mock(SomeInterface.class) };
And allow for some notes:
At least in your code, it looks like you called your interface "Array1" and "Array2". That is highly misleading. Give interfaces names that say what their behavior is about. The fact that you later create arrays containing objects of that interface ... doesn't matter at all!
Unless you have good reasons - consider not using PowerMock. PowerMock relies on byte-code manipulation; and can simply cause a lot of problems. In most situations, people wrote untestable code; and then they turn to PowerMock to somehow test that. But the correct answer is to rework that broken design, and to use a mocking framework that comes without "power" in its name. You can watch those videos giving you lengthy explanations how to write testable code!
For a case in which stubbed variable in made inside the actual production code,
How can we pass actual arguments with stubbing e.g using Mockito.when in JUnits?
E.g if a method in production code is like:
1) servicaClass.doSomething(Calendar.getInstance)
2) servicaClass.doSomething(someMethodToCreateActualVariable())
while testing how a actual parameter can be passed?
Like
-> when(mockedServiceClass.doSomething(Calendar.geInstance)).thenReturn("")
But while testing production code will take its own calendar value while executing.
There can be a way to make public setter getter method for the used variable to be stuffed. But that dont seem to be a optimum solution.
Any pointers with this regard will be helpful?
If you know the matching value before the fact, you can use stubbing. Mockito matchers like eq (compare with equals) and same (compare with ==) will help there, or you can get the eq behavior by specifying the value directly. Note that you have to use Matchers for all values if you use any at all; you can't use Matchers for only one argument of a two-argument method call.
// Without matchers
when(yourMock.method(objectToBeComparedWithEquals))
.thenReturn(returnValue);
// With matchers
when(yourMock.method(eq(objectToBeComparedWithEquals)))
.thenReturn(returnValue);
when(yourMock.method(same(objectToBeComparedReferentially)))
.thenReturn(returnValue);
If you don't know the matching value until after you run the method, you might want verification instead. Same rules apply about Matchers.
SomeValue someValue = yourSystemUnderTest.action();
verify(yourMock).initializeValue(someValue);
And if you need to inspect the value after the fact, you can use a Captor:
ArgumentCaptor myCaptor = ArgumentCaptor.forClass(SomeValue.class);
yourSystemUnderTest.action();
verify(yourMock).initializeValue(myCaptor.capture());
SomeValue valueMockWasCalledWith = myCaptor.getValue();
// Now you've retrieved the reference out of the mock, so you can assert as usual.
assertEquals(42, valueMockWasCalledWith.getInteger());
Using Mockito I got in trouble with the following:
Mockito.when(restOperationMock.exchange(
Mockito.anyString(),
Mockito.any(HttpMethod.class),
Mockito.any(HttpEntity.class),
Mockito.eq(CustomerResponse.class),
**Mockito.anyMap()**)).
thenReturn(re);
The problem was the method wasn't intercepted because I was using Mockito.any(Map.class) instead of Mockito.anyMap() and I was passing as a parameter a HashMap. What are the differences between Mockito.any(Map.class) and Mockito.anyMap()?
There is only one small difference between any(Map.class) and anyMap(): Starting with Mockito 2.0, Mockito will treat the any(Map.class) call to mean isA(Map.class) rather than ignoring the parameter entirely. (See the comment from Mockito contributor Brice on this SO answer.) Because restOperationMock.exchange takes an Object vararg, you may need anyMap to catch a case where a non-Map object is being passed, or no object at all is passed.
(I'd previously put that as a "dummy value" to return, Mockito can return an empty Map for calls to anyMap(), but can only return a null for calls to any(Map.class). If restOperationMock.exchange delegates to a real implementation during stubbing, such as if it were a spy or unmockable method (final method, method on a final class, etc), then that dummy value may be used in real code. However, that's only true for any(); anyMap() and any(Map.class) both give Mockito enough information to return a dummy Map implementation, where any() has its generics erased and only knows enough to return null.)
To illustrate the following example I created a litte spock test (but it's about groovy itself, not spock):
void "some spock test"() {
given: String value = null
expect: someMethod(value) == 3
}
int someMethod(String s) {
return 3
}
int someMethod(Map s) {
return 5
}
There are two methods who's signatures only differ by the type of the given parameter. I thought that when I give it a null value that is explicitly typed as a string, the string-method will be called.
But that doesn't happen; the test fails, because the map-method is called! Why?
I guess groovy ignores the type and treats all nulls the same. There seems to be some kind of priority of types: When I use Object instead of Map as the parameter type of the wrong-method, its all the same, but when I for instance use Integer, the test succeeds.
But than again: If groovy really ignores the type of nulls, why can the following fix the original test:
expect: someMethod((String) value) == 3
If you read my answer to the question Tim already mentioned you will see that I talk there about runtime types. The static type plays normally no role in this. I also described there how the distance calculation is used and that for null the distance to Object is used to determine the best fitting method. What I did not mention is that you can force method selection by using a cast. Internally Groovy will use a wrapper for the object, that also transports the type. Then the transported type is used instead. But you surely understand, that this means a one additional object creation per method class, which is very inefficient. Thus it is not the standard. In the future Groovy maybe change to include that static type information, but this requires a change to the MOP as well. And that is difficult