Parameterized Testing - Running Tests with Multiple Data Sets

diagram parametrize makes N tests, not one test with a loop mermaid

  flowchart LR
  A["@pytest.mark.parametrize('a,b,want', [(1,1,2), (2,3,5), (0,0,0)])"] --> B["one function"]
  B --> C["test_add[1-1-2]"]
  B --> D["test_add[2-3-5]"]
  B --> E["test_add[0-0-0]"]
  F["a loop inside one test"] --> G["first failure stops the rest"]
  G --> H["one result for all cases"]
  C --> I["3 independent results"]
  D --> I
  E --> I

Each set of arguments is collected as its own test, with its own id and its own result. That is the whole advantage over looping inside a single test: one bad case is reported as one failure, the others still report as passes, and you can re-run just the failing id.

Why parameterize?

Instead of repeating similar tests, parameterize inputs.

Example

test_parametrize.py

import pytest
 
 
def add(a, b):
    return a + b
 
 
@pytest.mark.parametrize(
    "a,b,expected",
    [
        (1, 2, 3),
        (0, 0, 0),
        (-1, 1, 0),
    ],
)
def test_add(a, b, expected):
    assert add(a, b) == expected

Parametrize with ids

test_parametrize_ids.py

import pytest
 
 
@pytest.mark.parametrize(
    "s,expected",
    [("10", 10), ("0", 0)],
    ids=["ten", "zero"],
)
def test_int_parse(s, expected):
    assert int(s) == expected

🧪 Try It Yourself

Exercise 1 – Write a unittest TestCase

Exercise 2 – assertRaises

Exercise 3 – setUp and tearDown

pch.coffeeTagline

pch.feedbackHeading

pch.feedbackSubheading