Quick tip: Easily identify failing parametrized test in pytest with pytest.param

Pytest is a great tool for writing and running tests in python. One of the features I find myself using a lot is parametrized tests.

Parametrization of arguments e.g. when unit testing a function is great, as it keeps the code of the tests short and you can test a lot of different inputs that will give you more confidence regarding how your code works with different inputs. By the way, a great way to get different sets of inputs is to use property testing (e.g. with a tool like hypothesis).

Let’s assume we have a calculator module, which includes an add method. As you can see below, the add function has a bug when the first number is negative.

# File location: calc/calculator.py (cacl is a package)

def add(x, y):
    if x < 0:
        return x * y

    return x + y

Let’s assume that our test suite uses parametrization:

# File location: calc/test_calculator.py (cacl is a package)

import pytest
from calculator import add


@pytest.mark.parametrize(
    'x,y,result',
    [
        (1, 2, 3),
        (-3, 4, 1)
    ]
)
def test_add(x, y, result):
    assert add(x, y) == result

When running the tests, we’ll get an assertion error:

======================================== short test summary info ========================================
FAILED test_calculator.py::test_add[-3-4-1] - assert -12 == 1

We see that the test fails when the arguments passed to it are (-3, 4, 1). Imagine testing other functions (or even http endpoints) with parametrized tests, wouldn’t some context be helpful, especially to other people?

Here is where pytest.param comes in handy. Let’s change our test’s code:

# File location: calc/test_calculator.py (cacl is a package)

import pytest
from calculator import add


@pytest.mark.parametrize(
    'x,y,result',
    [
        pytest.param(1, 2, 3, id='Addition of positive numbers'),
        pytest.param(-3, 4, 1, id='Addition of positive and negative numbers')
    ]
)
def test_add(x, y, result):
    assert add(x, y) == result

Let’s see the test failure when we run this updated version of the test:

======================================== short test summary info ========================================
FAILED test_calculator.py::test_add[Addition of a positive number to a negative one] - assert -12 == 1

As you can see, with a simple change (using pytest.param and providing an id), we get much more context about what the parametrized test is about.

That’s all for now folks!