Masonite testing is very simple. You can test very complex parts of your code with ease by just extending your class with a Masonite unit test class.
Although Masonite uses pytest to run tests, Masonite's test suite is based on unittest. So you will use unittest syntax but run the tests with pytest. Because of this, all syntax will be in camelCase instead of PEP 8 lower_case_with_underscores. Just know that all TestCase assertions used during testing is in camelCase form to maintain unittest standards.
Testing Environment¶
When running tests, Masonite will automatically set the environment to testing. You are free to define other testing environment configuration values as necessary.
You can create a .env.testing file. Feel free to load any testing environment variables in here. By default they will not be commited. When pytest runs it will additionally load and override any additional environment variables.
Creating Tests¶
You can simply create a file starting with test_ and then creating a test class inheriting from masonite TestCase class.
You can also directly use the command
to create tests/unit/test_some_feature.py:
from masonite.tests import TestCase
class SomeFeatureTest(TestCase):
def setUp(self):
super().setUp()
def test_something(self):
self.assertTrue(True)
That's it! You're ready to start testing. Read on to learn how to build your test cases and run them.
Running Tests¶
You can run tests by calling
This will automatically discover your tests following pytest automatic tests discovery. You can also run a specific test class
Or a specific test method
Finally you can re-run the last failed tests automatically
Building Tests¶
Test Life Cycle¶
When you run a test class each test method of this test class will be ran following a specific life cycle.
class TestFeatures(TestCase):
@classmethod
def setUpClass(cls):
"""Called once before all tests of this class are executed."""
print("Setting up test class")
@classmethod
def tearDownClass(cls):
"""Called once after all tests of this class are executed."""
print("Cleaning up test class")
def setUp(self):
"""Called once before each test are executed."""
super().setUp()
print("Setting up individual unit test")
def tearDown(self):
"""Called once after each test are executed."""
super().tearDown()
print("Cleaning up individual unit test")
def test_1(self):
print("Running test 1")
def test_2(self):
print("Running test 2")
Running the above test class will create this output:
Setting up test class
Setting up individual unit test
Running test 2
Cleaning up individual unit test
Setting up individual unit test
Running test 1
Cleaning up individual unit test
Cleaning up test class
Warning
Note that tests methods are not always ran in the order specified in the class. Anyway you should not make the assumptions that tests will be run in a given order. You should try to make your tests idempotent.
Chaining Assertions¶
All methods that begin with assert can be chained together to run through many assertions. All other method will return some kind of boolean or value which you can use to do your own assertions.
Asserting Exceptions¶
Sometimes you need to assert that a given piece of code will raise a given exception. To do this you
can use the standard assertRaises() context manager:
with self.assertRaises(ValidationError) as e:
# run some code here
raise ValidationError("An error occured !")
self.assertEqual(str(e.exception), "An error occured !")
Capturing Output¶
Sometimes you need to test the output of a function that prints to the console. To do this in your
tests you can use the captureOutput() context manager:
with self.captureOutput() as output:
# run some code here
print("Hello World !")
self.assertEqual(output.getvalue().strip(), "Hello World !")
Overriding Debug Mode¶
Sometimes you would need to change the debug mode value during the lifetime of a test. To do this
you can use the debugMode() context manager:
# run the code context with DEBUG enabled
with self.debugMode() as output:
self.get("/").assertError()
# run the code context with DEBUG disabled
with self.debugMode(False) as output:
self.get("/").assertError()
Dumping Data¶
During tests execution, print() statements will not be visible. You can use the dump() test
helper to dump data to console during a test:
def test_can_create_user(self):
user = User.find(1)
self.get("/register").assertRedirect()
self.dump("Hello")
self.dump(user, "User with ID 1")
Note that you can provide a second argument to name the dump in console.
Stopping Test¶
If you want to programmatically stop the test execution you can use the stop() helper. You can
even provide a reason.
def test_can_create_user(self):
user = User.find(1)
self.get("/register").assertRedirect()
self.stop("for debugging") #== the test will be stopped here.
self.post("/login", {"email": user.email, "password": "secret"})
Test are stopped by returning a pytest 2 exit code (user interruption).
Test Helpers¶
Masonite comes with different helpers that can ease writing tests. Some of them have already been explained in sections above.
- withExceptionsHandling
- withoutExceptionsHandling
- withCsrf
- withoutCsrf
- withCookies
- withHeaders
- fakeTime
- fakeTimeTomorrow
- fakeTimeYesterday
- fakeTimeInFuture
- fakeTimeInPast
- restoreTime
withExceptionsHandling¶
Enable exceptions handling during testing.
withoutExceptionsHandling¶
Disable exceptions handling during testing.
Warning
Note that exception handling is disabled by default during testing.
withCsrf¶
Enable CSRF protection during testing.
withoutCsrf¶
Disable CSRF protection during testing.
Warning
Note that CSRF protection is disabled by default during testing.
withCookies¶
Add cookies that will be used in the next request. This method accepts a dictionary of name / value pairs. Cookies dict is reset between each test.
withHeaders¶
Add headers that will be used in the next request. This method accepts a dictionary of name / value pairs. Headers dict is reset between each test.
fakeTime¶
Set a given pendulum instance to be returned when a now (or today, tomorrow yesterday) instance is created. It's really useful during tests to check
timestamps logic.
This allow to control which datetime will be returned to be able to always have an expected behaviour in the tests.
given_date = pendulum.datetime(2021, 2, 5)
self.fakeTime(given_date)
self.assertEqual(pendulum.now(), given_date)
Warning
When using those helpers you should not forget to reset the default pendulum behaviour with restoreTime() helper to avoid breaking other tests. It can be done directly in the test or in a tearDown() method.
fakeTimeTomorrow¶
Set the mocked time as tomorrow. (It's a shortcut to avoid doing self.fakeTime(pendulum.tomorrow())).
fakeTimeYesterday¶
Set the mocked time as yesterday.
fakeTimeInFuture¶
Set the mocked time as an offset of a given unit of time in the future. Unit can be specified among pendulum units: seconds, minutes, hours, days (default), weeks, months, years.
real_now = pendulum.now()
self.fakeTimeInFuture(1, "months")
self.assertEqual(pendulum.now().diff(real_now).in_months(), 1)
fakeTimeInPast¶
Set the mocked time as an offset of a given unit of time in the past. Unit can be specified among pendulum units: seconds, minutes, hours, days (default), weeks, months, years.
restoreTime¶
Restore the mocked time behaviour to default behaviour of pendulum. When using fake time helpers you should not forget to call this helper at the end.
It can be done directly in the test or in a tearDown() method.