Testing with pytest

pytest is a framework for testing Python programs. Until now, you tested a function by calling it under if __name__ == "__main__": and reading the output yourself. pytest does the checking for you: you write down the answer you expect, and pytest tells you whether the function got it right. This lab shows how to run pytest in the terminal and in VS Code, and how to write your own test functions.

Installing pytest

Open your CS149 folder in VS Code, and open a new terminal (Terminal > New Terminal). Make sure the command line begins with (.venv), as in Step 5 of the VS Code setup. Then run the following command:

pip install pytest-cov

This command installs pytest along with pytest-cov, which adds coverage. You will use coverage at the end of this lab.

Example code

Create a Week08 folder in your CS149 folder. Download the following files into your Week08 folder. (Right click each link, and "Save link as…")

• triangles.py
• test_triangles.py

Open triangles.py and run the program. You should see the following output:

valid((3, 4, 5)): True
area((3, 4, 5)): 6.0
classify((3, 4, 5)): Scalene

Now open test_triangles.py. Each test function calls a function from triangles.py and uses assert to state what the result should be. If the result is what the assert says, nothing happens. If it isn't, the test fails.

Running pytest in the Terminal

The terminal starts in your CS149 folder. Like ruff, pytest looks at every file under the current folder, so first move into the Week08 folder:

cd Week08

Then run pytest:

pytest

This command runs every module in the current folder whose name begins with test_. Within each module, every function whose name begins with test_ is called. If any assertion fails, pytest displays a detailed error message.

Optional arguments

pytest works the same way as ruff. If you run pytest without any arguments, all test modules are run. Alternatively, you can run a specific test module:

pytest test_triangles.py

You can run a specific test function using the -k option:

pytest -k test_area

The -q (quiet) option makes the output less verbose:

pytest -q

Multiple options can be combined:

pytest -q -k test_area

The approx() function

Because floating-point arithmetic is imprecise, comparison with == can be unreliable. For example, 0.1 + 0.2 == 0.3 is False, because 0.1 + 0.2 is 0.30000000000000004.

pytest includes a function named approx() that checks whether a value is within a small difference of the expected value. So 0.1 + 0.2 == approx(0.3) is True. Notice that test_area() uses approx(), because the area() function returns a float.

0.1 + 0.2

0.1 + 0.2 == 0.3

from pytest import approx

approx(0.3)

0.1 + 0.2 == approx(0.3)

Running pytest in VS Code

You can also run pytest without typing any commands!

1. Click "Testing" (the beaker icon, near the bottom) on the Activity Bar (left sidebar).
2. Expand the list until you see test_triangles.py and its three test functions.
3. Click the buttons to run pytest at any level: one test, one module, or everything.
   • A green check means the test passed, and a red X means it failed.
   • Click a failed test to see the same error message that you saw in the terminal.
4. Open test_triangles.py and notice the buttons next to each test function.
   • You can also right-click a to "Debug Test" or "Run Test with Coverage".

The Testing view finds every test in your CS149 folder, not just the ones in Week08. That's the same reason you cd Week08 before running pytest in the terminal.

Debugging a test

A failed test tells you that a function is wrong, but not where. The debugger lets you pause the program and run it one line at a time, watching the variables change.

• A breakpoint is a line where the program will pause. Click to the left of a line number to set one (a red dot appears), and click the dot again to remove it.
• When the program pauses, the Variables panel (on the left) shows the current value of every variable.
• The toolbar at the top of the window controls the program: Step Over (F10) runs the current line, Step Into (F11) goes inside a function call, and Continue (F5) runs until the next breakpoint.

Writing your own tests

Test functions use Python's assert statement.

• Ex: assert a % 2 == 0, "value was odd, should be even"
• The error message after the comma is optional, but helpful.

Follow this pattern when testing your own programs:

• For every module m, create a module named test_m.
• For every function f, write a function named test_f.
• Write enough assert statements to test every branch.

The provided tests also use raises(). area() and classify() don't return a value for an invalid triangle; they raise an error instead. A with raises(ValueError): block passes only if the code inside it raises that error.

Going further

See pytest's documentation and API Reference for additional details.