gmock setting default actions / ON_CALL vs. EXPECT_CALL

Here is the "official" explanation about the most important differences between ON_CALL and EXPECT_CALL as explained in the gMock Cookbook.

There are basically two constructs for defining the behavior of a mock object: ON_CALL and EXPECT_CALL.

The difference?

ON_CALL defines what happens when a mock method is called, but doesn't imply any expectation on the method being called.

EXPECT_CALL not only defines the behavior, but also sets an expectation that the method must be called with the given arguments, for the given number of times (and in the given order when you specify the order too).

Since EXPECT_CALL does more, isn't it better than ON_CALL?

Not really. Every EXPECT_CALL adds a constraint on the behavior of the code under test. Having more constraints than necessary is bad - even worse than not having enough constraints.

This may be counter-intuitive. How could tests that verify more be worse than tests that verify less? Isn't verification the whole point of tests?

The answer, lies in what a test should verify. A good test verifies the contract of the code. If a test over-specifies, it doesn't leave enough freedom to the implementation. As a result, changing the implementation without breaking the contract (e.g. refactoring and optimization), which should be perfectly fine to do, can break such tests. Then you have to spend time fixing them, only to see them broken again the next time the implementation is changed.

Keep in mind that one doesn't have to verify more than one property in one test. In fact, it's a good style to verify only one thing in one test. If you do that, a bug will likely break only one or two tests instead of dozens (which case would you rather debug?). If you are also in the habit of giving tests descriptive names that tell what they verify, you can often easily guess what's wrong just from the test log itself.

So use ON_CALL by default, and only use EXPECT_CALL when you actually intend to verify that the call is made.


ON_CALL(mock, methodX(_)).WillByDefault(Return(0x01));
EXPECT_CALL(mock, methodX(_)).WillRepeatedly(Return(0x01));

As you said, these two lines are doing exactly the same thing, therefore there are no differences at all. Use either way to set a default action as you please.

However, there is a logical difference :

  • ON_CALL(mock, methodX(_)).WillByDefault(Return(0x01)); means that the method might be called, and if that happens, every call will return 0x01
  • EXPECT_CALL(mock, methodX(_)).WillRepeatedly(Return(0x01)); means that it is expected that the method will be called, and every call will return 0x01

By the way, there is a Setting default actions in their cheat sheet, which says :

To customize the default action for a particular method, use ON_CALL():

ON_CALL(mock_object, method(matchers))
    .With(multi_argument_matcher)  ?
    .WillByDefault(action);

There are subtle but significant differences between the two statements. EXPECT_CALL sets expectation on a mock calls. Writing

EXPECT_CALL(mock, methodX(_)).WillRepeatedly(do_action);

tells gMock that methodX may be called on mock any number of times with any arguments, and when it is, mock will perform do_action. On the other hand,

ON_CALL(mock, methodX(_)).WillByDefault(do_action);

tells gMock that whenever methodX is invoked on mock, it should perform do_action. That feature is helpful in a scenario where you have to write many expectations on your mock, and most/all of them have to specify the same action -- especially if it's complex. You can specify that action in ON_CALL, and then write EXPECT_CALLs without specifying the action explicitly. E.g.,

ON_CALL(mock, Sign(Eq(0), _))
  .WillByDefault(DoAll(SetArgPointee<1>("argument is zero"), Return(0)));
ON_CALL(mock, Sign(Gt(0), _))
  .WillByDefault(DoAll(SetArgPointee<1>("argument is positive"), Return(1)));
ON_CALL(mock, Sign(Lt(0), _))
  .WillByDefault(DoAll(SetArgPointee<1>("argument is negative"), Return(-1)));

Now, if you have to write a lot of EXPECT_CALLs, you don't have to mock's specify the behavior every time:

EXPECT_CALL(mock, Sign(-4, _));
EXPECT_CALL(mock, Sign(0, _));
EXPECT_CALL(mock, Sign(1, _)).Times(2);
EXPECT_CALL(mock, Sign(2, _));
EXPECT_CALL(mock, Sign(3, _));
EXPECT_CALL(mock, Sign(5, _));

In another example, assuming Sign returns int, if you write

ON_CALL(mock, Sign(Gt(0), _)).WillByDefault(Return(1));
EXPECT_CALL(mock, Sign(10, _));

the call mock.Sign(10) will return 1 as ON_CALL provides default behavior for a call specified by EXPECT_CALL. But if you write

EXPECT_CALL(mock, Sign(Gt(0), _).WillRepeatedly(Return(1));
EXPECT_CALL(mock, Sign(10, _));

the invocation of mock.Sign(10, p) will return 0. It will be matched against the second expectation. That expectation specifies no explicit action and gMock will generate a default action for it. That default action is to return a default value of the return type, which is 0 for int. The first expectation will be totally ignored in this case.

Tags:

C++

Googlemock