Testing is important:

Ok. So now you know why testing is so crucial. But what is Unit Testing?

Unit Testing is testing each software component separately like functions and methods.

Unit Testing is key to:

Running pytest without mentioning a filename will run all files of the format test_*.py or *_test.py in the current directory and subdirectories.

The advantage of naming test filenames as %_test.py is that the file that has the test of a module will be close to the file if we list the files alphabetically. If we name the test files as test_%.py, it will list all test files together. Some people prefer to have the test files in a folder structure separated from the codebase, usually in a directory called tests.

Inside the Pytest test files, create functions which names starts with “test_”. Each of those functions will be counted as one item on the report that shows when you run pytest.

Fixtures are Pytest functions that contain data or get data from other files like configuration files. It also can be used to load test data into the database and set environment variables to be used during the tests.

Here is an example of a module pytesting.py that we will use in an example:

Here we start the test file importing what we will use:

Now, we create a fixture that has data of one of the products we will use as an example:

Here we have a fixture with an example of another product:

Then, we create another fixture that uses the other two fixtures:

Finally, we use the fixture in a test:

It is possible to have the fixtures in the same file as the tests or in a conftest.py file. Having them in a conftest.py file, we can use them in all test files of the project. We also can have a conftest.py file in each folder. This way we can have a hierarchy and each folder has its own conftest.py file and fixtures.

Pytest Fixture scopes specify how often a fixture will be executed:

Pytest Markers are used to create categories to select which categories we want to run or exclude when running Pytest.

Running pytest --markers we see all the markers that Pytest is aware of.

To define a marker in a 

pytest -v -m "not marker_name"

With parametrization, we can do several tests using less code.

We will create a test using the same add function that as before:

Here we do one test with 3 test cases without much additional code:

Executing Pytest with the parameter --durations=X, we can see the duration of the X slowest tests.

The English meaning of mocking is to make a replica or imitation of something.

In unit testing, a code block that you want to test may depend on other code. When we unit test, we test only a code block without those dependencies. To isolate the behavior of the code you need to test, you need to replace the other code with mocks that simulate the behavior of the real code. So, mocking is using mocks that simulate real code behavior.

Here is a simple example. We mock an object with a method called “method_example”. We tested:

Here is a simple function run_dice that calls random.randint to simulate a dice roll. And the function returns "Unlucky" if we get a number less than 4 and "Lucky" otherwise. This code is in the file mocking.py.

This is the test. With @patch("testing.mocking.random"), we mock random from the mocking.py file which is the file where run_dice is located and is in the testing directory. The mock_random parameter of the test_mock_function_with_dice method is the name of the mock object we pass to the test method. Then inside the function, we say that randint should return 4. Then we execute the run_dice function and test if it results in "Lucky". Finally, we say that randint should return 3 then we execute the run_dice function and test if it returns "Unlucky".

Now we will mock using pytest-mock. Therefore, for this to work we need to install it:

Here is how we would do the same test shown before using pytest-mock. The fixture mocker was created by pytest-mock.

In this article I explained how to use Pytest and Mock to implement tests in Python. I discussed basic use, fixtures, markers, parametrization and duration. With this you have a comprehensive toolset to test your Python application.

This is part of a series about Python recommended coding practices. The code for this series is also available in this GitHub project.

This article has an index to my other articles.

What are your thoughts on the article? Are there any suggestions, criticisms, or opinions? Let me know in the comments!

#python #pythontests #pytest  #pythonbestpractices #bestpractices #pythonprogramming #codingpractices #codingskills #codingtips