3 ### When I call a method on my mock object, the method for the real object is invoked instead. What's the problem?
5 In order for a method to be mocked, it must be *virtual*, unless you use the
6 [high-perf dependency injection technique](gmock_cook_book.md#MockingNonVirtualMethods).
8 ### Can I mock a variadic function?
10 You cannot mock a variadic function (i.e. a function taking ellipsis (`...`)
11 arguments) directly in gMock.
13 The problem is that in general, there is *no way* for a mock object to know how
14 many arguments are passed to the variadic method, and what the arguments' types
15 are. Only the *author of the base class* knows the protocol, and we cannot look
18 Therefore, to mock such a function, the *user* must teach the mock object how to
19 figure out the number of arguments and their types. One way to do it is to
20 provide overloaded versions of the function.
22 Ellipsis arguments are inherited from C and not really a C++ feature. They are
23 unsafe to use and don't work with arguments that have constructors or
24 destructors. Therefore we recommend to avoid them in C++ as much as possible.
26 ### MSVC gives me warning C4301 or C4373 when I define a mock method with a const parameter. Why?
28 If you compile this using Microsoft Visual C++ 2005 SP1:
33 virtual void Bar(const int i) = 0;
36 class MockFoo : public Foo {
38 MOCK_METHOD(void, Bar, (const int i), (override));
42 You may get the following warning:
45 warning C4301: 'MockFoo::Bar': overriding virtual function only differs from 'Foo::Bar' by const/volatile qualifier
48 This is a MSVC bug. The same code compiles fine with gcc, for example. If you
49 use Visual C++ 2008 SP1, you would get the warning:
52 warning C4373: 'MockFoo::Bar': virtual function overrides 'Foo::Bar', previous versions of the compiler did not override when parameters only differed by const/volatile qualifiers
55 In C++, if you *declare* a function with a `const` parameter, the `const`
56 modifier is ignored. Therefore, the `Foo` base class above is equivalent to:
61 virtual void Bar(int i) = 0; // int or const int? Makes no difference.
65 In fact, you can *declare* `Bar()` with an `int` parameter, and define it with a
66 `const int` parameter. The compiler will still match them up.
68 Since making a parameter `const` is meaningless in the method declaration, we
69 recommend to remove it in both `Foo` and `MockFoo`. That should workaround the
72 Note that we are talking about the *top-level* `const` modifier here. If the
73 function parameter is passed by pointer or reference, declaring the pointee or
74 referee as `const` is still meaningful. For example, the following two
75 declarations are *not* equivalent:
78 void Bar(int* p); // Neither p nor *p is const.
79 void Bar(const int* p); // p is not const, but *p is.
82 ### I can't figure out why gMock thinks my expectations are not satisfied. What should I do?
84 You might want to run your test with `--gmock_verbose=info`. This flag lets
85 gMock print a trace of every mock function call it receives. By studying the
86 trace, you'll gain insights on why the expectations you set are not met.
88 If you see the message "The mock function has no default action set, and its
89 return type has no default value set.", then try
90 [adding a default action](gmock_cheat_sheet.md#OnCall). Due to a known issue,
91 unexpected calls on mocks without default actions don't print out a detailed
92 comparison between the actual arguments and the expected arguments.
94 ### My program crashed and `ScopedMockLog` spit out tons of messages. Is it a gMock bug?
96 gMock and `ScopedMockLog` are likely doing the right thing here.
98 When a test crashes, the failure signal handler will try to log a lot of
99 information (the stack trace, and the address map, for example). The messages
100 are compounded if you have many threads with depth stacks. When `ScopedMockLog`
101 intercepts these messages and finds that they don't match any expectations, it
102 prints an error for each of them.
104 You can learn to ignore the errors, or you can rewrite your expectations to make
105 your test more robust, for example, by adding something like:
108 using ::testing::AnyNumber;
109 using ::testing::Not;
111 // Ignores any log not done by us.
112 EXPECT_CALL(log, Log(_, Not(EndsWith("/my_file.cc")), _))
116 ### How can I assert that a function is NEVER called?
121 EXPECT_CALL(foo, Bar(_))
125 ### I have a failed test where gMock tells me TWICE that a particular expectation is not satisfied. Isn't this redundant?
127 When gMock detects a failure, it prints relevant information (the mock function
128 arguments, the state of relevant expectations, and etc) to help the user debug.
129 If another failure is detected, gMock will do the same, including printing the
130 state of relevant expectations.
132 Sometimes an expectation's state didn't change between two failures, and you'll
133 see the same description of the state twice. They are however *not* redundant,
134 as they refer to *different points in time*. The fact they are the same *is*
135 interesting information.
137 ### I get a heapcheck failure when using a mock object, but using a real object is fine. What can be wrong?
139 Does the class (hopefully a pure interface) you are mocking have a virtual
142 Whenever you derive from a base class, make sure its destructor is virtual.
143 Otherwise Bad Things will happen. Consider the following code:
148 // Not virtual, but should be.
153 class Derived : public Base {
161 Base* p = new Derived;
163 delete p; // Surprise! ~Base() will be called, but ~Derived() will not
164 // - value_ is leaked.
167 By changing `~Base()` to virtual, `~Derived()` will be correctly called when
168 `delete p` is executed, and the heap checker will be happy.
170 ### The "newer expectations override older ones" rule makes writing expectations awkward. Why does gMock do that?
172 When people complain about this, often they are referring to code like:
175 using ::testing::Return;
177 // foo.Bar() should be called twice, return 1 the first time, and return
178 // 2 the second time. However, I have to write the expectations in the
179 // reverse order. This sucks big time!!!
180 EXPECT_CALL(foo, Bar())
182 .RetiresOnSaturation();
183 EXPECT_CALL(foo, Bar())
185 .RetiresOnSaturation();
188 The problem, is that they didn't pick the **best** way to express the test's
191 By default, expectations don't have to be matched in *any* particular order. If
192 you want them to match in a certain order, you need to be explicit. This is
193 gMock's (and jMock's) fundamental philosophy: it's easy to accidentally
194 over-specify your tests, and we want to make it harder to do so.
196 There are two better ways to write the test spec. You could either put the
197 expectations in sequence:
200 using ::testing::Return;
202 // foo.Bar() should be called twice, return 1 the first time, and return
203 // 2 the second time. Using a sequence, we can write the expectations
204 // in their natural order.
207 EXPECT_CALL(foo, Bar())
209 .RetiresOnSaturation();
210 EXPECT_CALL(foo, Bar())
212 .RetiresOnSaturation();
216 or you can put the sequence of actions in the same expectation:
219 using ::testing::Return;
221 // foo.Bar() should be called twice, return 1 the first time, and return
222 // 2 the second time.
223 EXPECT_CALL(foo, Bar())
226 .RetiresOnSaturation();
229 Back to the original questions: why does gMock search the expectations (and
230 `ON_CALL`s) from back to front? Because this allows a user to set up a mock's
231 behavior for the common case early (e.g. in the mock's constructor or the test
232 fixture's set-up phase) and customize it with more specific rules later. If
233 gMock searches from front to back, this very useful pattern won't be possible.
235 ### gMock prints a warning when a function without EXPECT_CALL is called, even if I have set its behavior using ON_CALL. Would it be reasonable not to show the warning in this case?
237 When choosing between being neat and being safe, we lean toward the latter. So
238 the answer is that we think it's better to show the warning.
240 Often people write `ON_CALL`s in the mock object's constructor or `SetUp()`, as
241 the default behavior rarely changes from test to test. Then in the test body
242 they set the expectations, which are often different for each test. Having an
243 `ON_CALL` in the set-up part of a test doesn't mean that the calls are expected.
244 If there's no `EXPECT_CALL` and the method is called, it's possibly an error. If
245 we quietly let the call go through without notifying the user, bugs may creep in
248 If, however, you are sure that the calls are OK, you can write
253 EXPECT_CALL(foo, Bar(_))
254 .WillRepeatedly(...);
266 This tells gMock that you do expect the calls and no warning should be printed.
268 Also, you can control the verbosity by specifying `--gmock_verbose=error`. Other
269 values are `info` and `warning`. If you find the output too noisy when
270 debugging, just choose a less verbose level.
272 ### How can I delete the mock function's argument in an action?
274 If your mock function takes a pointer argument and you want to delete that
275 argument, you can use testing::DeleteArg<N>() to delete the N'th (zero-indexed)
281 MOCK_METHOD(void, Bar, (X* x, const Y& y));
283 EXPECT_CALL(mock_foo_, Bar(_, _))
284 .WillOnce(testing::DeleteArg<0>()));
287 ### How can I perform an arbitrary action on a mock function's argument?
289 If you find yourself needing to perform some action that's not supported by
290 gMock directly, remember that you can define your own actions using
291 [`MakeAction()`](#NewMonoActions) or
292 [`MakePolymorphicAction()`](#NewPolyActions), or you can write a stub function
293 and invoke it using [`Invoke()`](#FunctionsAsActions).
297 using ::testing::Invoke;
299 MOCK_METHOD(void, Bar, (X* p));
301 EXPECT_CALL(mock_foo_, Bar(_))
302 .WillOnce(Invoke(MyAction(...)));
305 ### My code calls a static/global function. Can I mock it?
307 You can, but you need to make some changes.
309 In general, if you find yourself needing to mock a static function, it's a sign
310 that your modules are too tightly coupled (and less flexible, less reusable,
311 less testable, etc). You are probably better off defining a small interface and
312 call the function through that interface, which then can be easily mocked. It's
313 a bit of work initially, but usually pays for itself quickly.
315 This Google Testing Blog
316 [post](https://testing.googleblog.com/2008/06/defeat-static-cling.html) says it
317 excellently. Check it out.
319 ### My mock object needs to do complex stuff. It's a lot of pain to specify the actions. gMock sucks!
321 I know it's not a question, but you get an answer for free any way. :-)
323 With gMock, you can create mocks in C++ easily. And people might be tempted to
324 use them everywhere. Sometimes they work great, and sometimes you may find them,
325 well, a pain to use. So, what's wrong in the latter case?
327 When you write a test without using mocks, you exercise the code and assert that
328 it returns the correct value or that the system is in an expected state. This is
329 sometimes called "state-based testing".
331 Mocks are great for what some call "interaction-based" testing: instead of
332 checking the system state at the very end, mock objects verify that they are
333 invoked the right way and report an error as soon as it arises, giving you a
334 handle on the precise context in which the error was triggered. This is often
335 more effective and economical to do than state-based testing.
337 If you are doing state-based testing and using a test double just to simulate
338 the real object, you are probably better off using a fake. Using a mock in this
339 case causes pain, as it's not a strong point for mocks to perform complex
340 actions. If you experience this and think that mocks suck, you are just not
341 using the right tool for your problem. Or, you might be trying to solve the
344 ### I got a warning "Uninteresting function call encountered - default action taken.." Should I panic?
346 By all means, NO! It's just an FYI. :-)
348 What it means is that you have a mock function, you haven't set any expectations
349 on it (by gMock's rule this means that you are not interested in calls to this
350 function and therefore it can be called any number of times), and it is called.
351 That's OK - you didn't say it's not OK to call the function!
353 What if you actually meant to disallow this function to be called, but forgot to
354 write `EXPECT_CALL(foo, Bar()).Times(0)`? While one can argue that it's the
355 user's fault, gMock tries to be nice and prints you a note.
357 So, when you see the message and believe that there shouldn't be any
358 uninteresting calls, you should investigate what's going on. To make your life
359 easier, gMock dumps the stack trace when an uninteresting call is encountered.
360 From that you can figure out which mock function it is, and how it is called.
362 ### I want to define a custom action. Should I use Invoke() or implement the ActionInterface interface?
364 Either way is fine - you want to choose the one that's more convenient for your
367 Usually, if your action is for a particular function type, defining it using
368 `Invoke()` should be easier; if your action can be used in functions of
369 different types (e.g. if you are defining `Return(*value*)`),
370 `MakePolymorphicAction()` is easiest. Sometimes you want precise control on what
371 types of functions the action can be used in, and implementing `ActionInterface`
372 is the way to go here. See the implementation of `Return()` in
373 `testing/base/public/gmock-actions.h` for an example.
375 ### I use SetArgPointee() in WillOnce(), but gcc complains about "conflicting return type specified". What does it mean?
377 You got this error as gMock has no idea what value it should return when the
378 mock method is called. `SetArgPointee()` says what the side effect is, but
379 doesn't say what the return value should be. You need `DoAll()` to chain a
380 `SetArgPointee()` with a `Return()` that provides a value appropriate to the API
383 See this [recipe](gmock_cook_book.md#mocking-side-effects) for more details and
386 ### I have a huge mock class, and Microsoft Visual C++ runs out of memory when compiling it. What can I do?
388 We've noticed that when the `/clr` compiler flag is used, Visual C++ uses 5~6
389 times as much memory when compiling a mock class. We suggest to avoid `/clr`
390 when compiling native C++ mocks.