Mocks, Stubs & Doubles
Last updated
Was this helpful?
Last updated
Was this helpful?
Was this helpful?
Mocking is used for protocol testing - it tests how you'll use an API, and how you'll react when the API reacts accordingly.
Ideally (in many cases at least), that API should be specified as an interface rather than a class - an interface defines a protocol, a class defines at least part of an implementation.
Test doubles turn integration tests into unit/isolation tests
https://8thlight.com/blog/uncle-bob/2014/05/14/TheLittleMocker.html
https://martinfowler.com/articles/mocksArentStubs.html
https://javacodehouse.com/blog/mockito-tutorial/
https://www.codurance.com/publications/2019/04/08/introduction-to-test-doubles
http://xunitpatterns.com/Mocks,%20Fakes,%20Stubs%20and%20Dummies.html
https://stackoverflow.com/questions/22867680/how-does-mockito-mock-interfaces
https://www.toptal.com/java/a-guide-to-everyday-mockito
Cannot mock final class/methods using mockito 1.x, need latest version (use inline dependency)
If class is final but in your scope of change
Add an interface to the class to be controlled, and create a test class that implements that interface, then inject it into the system under test
If class is final but not in your scope of change
Cannot add an interface
Can wrap the dependency (adapter), replace the call of the actual with the adapter in codebase and mock the adapter in the tests
When a object is instantiated in the class under test, and want to verify that this object is used, we can use argumentcaptor
Thus when we it is used (we check the captor is used in the mehtod we verify) then can grab use the object that was captured, use its method and assert on them
https://stackoverflow.com/questions/10598898/mockito-numberformat-mocking-nullpointer-in-when-method
provide canned answers to calls made during the test, usually not responding at all to anything outside what's programmed in for the test
Stubbing is the process of giving behavior to a function that otherwise has no behavior on its own—you specify to the function exactly what values to return (that is, you stub the return values).
ie From mockito: when(...).thenReturn(...)
typically done through mocking frameworks
stubbing is appropriate when you need a function to return a specific value to get the system under test into a certain state
each stubbed function should have a direct relationship with the test’s assertions.
real implementations or fakes are still preferred because they don’t expose implementation details and they give you more guarantees about the correctness of the code compared to stubbing
are passed around but never actually used. Usually they are just used to fill parameter lists.
are stubs that also record some information based on how they were called. One form of this might be an email service that records how many messages it was sent.
A fake is a lightweight implementation of an API that behaves similar to the real implementation but isn’t suitable for production
for example, an inmemory database, wiremock
Using a fake is often the ideal technique when you need to use a test double, but a fake might not exist for an object you need to use in a test, and writing one can be challenging because you need to ensure that it has similar behavior to the real implementation, now and in the future
Normally use an already written on
otherwise it is hard to create, as it must mimic the real implementation and must be maintained
are pre-programmed with expectations which form a specification of the calls they are expected to receive. They can throw an exception if they receive a call they don't expect and are checked during verification to ensure they got all the calls they were expecting
is a way to validate how a function is called without actually calling the implementation of the function. A test should fail if afunction isn’t called the correct way; for example, if the function isn’t called at all, it’s called too many times, or it’s called with the wrong arguments.
interaction testing is typically done through mocking frameworks
Interaction testing is sometimes called mocking
it is preferred to test code through state testing
it can only validate that certain functions are called as expected. It requires you to make an assumption about the behavior of the code; for example, “If database.save(item) is called, we assume the item will be saved to the database."
Leaks details to test
when to use
You cannot perform state testing because you are unable to use a real implementation or a fake
Differences in the number or order of calls to a function would cause undesired behavior
you call the system under test and validate that either the correct value was returned or that some other state in the system under test was properly changed
it actually validates this assumption (such as by saving an item to a database and then querying the database to validate that the item exists)
if cannot state, use interaction testing, but also use integration testing
A mocking framework is a software library that makes it easier to create test doubles within tests;
it allows you to replace an object with a mock, which is a test double whose behavior is specified inline in a test.
The use of mocking frameworks reduces boilerplate because you don’t need to define a new class each time you need a test double.
Essential for testing and not want to call anything which can have consequences (ie state change (internally or externally), production calls, charges for services)
If you want to verify the behavior of the system under test (SUT), using mockito can be more effective as it allows you to verify interactions between the SUT and its collaborators.
When you mock something you’re removing all confidence in the integration between what you’re testing and what’s being mocked.
You will need to do separate tests for integration part, to see that it actually works
mocking is somewhat overused - often you're not really interested in the exact interaction, you really want a stub... but mocking framework can be used to create stubs, and you fall into the trap of creating brittle tests by mocking instead of stubbing. It's a hard balance to get right though.
When using verify (to check a dependency has been called) this ties the test to the implementation/structure of the code, rather than the behaviour
better to have stub injected, and assert on the stub
dependencies are part of the constructor and thus public
Can lead to mocking everything except the SUT
This can make test brittle
SUT is not necessarily a class, but can be a bunch of objects to test the behaviour
Over-reliance on mocks can lead to tests that are not representative of the real-world behavior of the system, resulting in tests that are less effective and less reliable.
SUT with lots of dependencies and interactions, and use of mocks can make tests more complex and unclear with lots of extra code
Leaks implementation details into your test
Ideally, a good test should need to change only if user-facing behavior of an API changes; it should remain unaffected by changes to the API’s implementation
Use of mocking frameworks can lead to false sense of security
MAke tests hard to maintain
makes refactoring harder, as tests will break
bug finding hard
test doubles, does not have exact sameness as the real implementation, so always some form of security that it is give correct evaluations
https://engineering.talkdesk.com/double-trouble-why-we-decided-against-mocking-498c915bbe1c
https://youtu.be/7AGQ9dhWCX0 Test Doubles Are A Scam – Matt Diephouse
Mocking can add additional processing time during the build process, slowing down the overall build time.
makes refactoring easier
known as classical testing
tests are more realistic, more confidence in the system
Using real implementations can cause your test to fail if there is a bug in the real implementation.
A real implementation is preferred if it is fast, deterministic, and has simple dependencies. For example, a real implementation should be used for a value object
for more complex code, using a real implementation often isn’t feasible
ie database or network calls, edge of app. these can be mocked
Can be slow
Need to construct the object graph for the SUT, can be complex
Should use the same mechanism as test, and adjust (ie factories) to replace mocked dependencies
Instead of a real implementation, use an inmemory version
This a fake or stub
If there is a logic which mimics the actual impl in the fake, it needs to be tested to ensure it matches the real impl and be used with accuracy in the tests
https://medium.com/javascript-scene/mocking-is-a-code-smell-944a70c90a6a
https://chemaclass.medium.com/to-mock-or-not-to-mock-af995072b22e
This should not be done, should avoid mocking third party code
https://youtu.be/v6hP2MXoVrI Don't Mock 3rd Party Code
you should perform interaction testing only for functions that are state-changing
ie for nondeterministic code such as external servers, time, randommness