Unit testing

A unit test is a small test, which checks that a single component of your code operates in the right way. A unit test helps you to isolate what is broken in your application and fix it faster.

(Real Python: Getting Started with testing)

Testing frameworks in Python

• unittest

• nosetest

• pytest (We will use this one)

Basic structure of a test

Given:

Put your system in the right state for testing e.g. create test variables.

When:

Execute the feature that you are testing.

Then:

Define the expected result of the test and check that the actual outcome equals the expected result.

Example using pytest

→ The code for testing is contained in a function whose name starts with test_xxx()

def add(a, b):
	return a + b 

def test_add():
	# Test variables
	a = 2
	b = 3
	expected_result = 5

	# Calculate area
	actual_result = add(a=a, b=b)

	# Check if calculated result equals expected result
	assert expected_result == actual_result

→ All test functions are stored in a file called test_xxx.py, e.g. test_geometry.py

What a good test looks like

Good:

• Short and quick to execute

• Easy to read

• Exercise one thing

• Fails fast

Bad:

• Relies on data files

• Messes with “real-life” files, servers, databases

Question:

What would you do if the user executes Rectangle("4", "5")?

a) Use print() to inform the user the arguments need to be numbers and return a new Rectangle object.

b) Use print() to inform the user the arguments need to be numbers and return None.

c) Raise an error.

d) Fix the arguments and return a new Rectangle object.

Test-Driven Development