Parameterization enables executing identical test logic with multiple input datasets, minimizing code duplication while maximizing coverage for similar test scenarios. Common use cases include validating payment processors, search algorithms, or mathematical functions.

The core mechanism stores test cases in collections and iterates through them, supplying distinct parameters during each test execution.

Implementation Techniques

Use @pytest.mark.parametrize decorator for parameterized tests. Consider a numerical addition function:

 def calculate_sum(value1, value2):
     return value1 + value2

List-Based Test Cases

Represent inputs as tuples within a list:

import pytest

def prepare_inputs():
    return [
        (3, 7, 10),
        ('foo', 'bar', 'foobar'),
        (2.5, 3.5, 6.0),
        ((1,2), (3,4), (1,2,3,4))
    ]

@pytest.mark.parametrize('operand1, operand2, expected', prepare_inputs())
def test_calculate_sum(operand1, operand2, expected):
    assert calculate_sum(operand1, operand2) == expected

Dictionary-Based Test Cases

Use named fields for improved readability:

import pytest

def get_test_scenarios():
    return [
        {'first': 3, 'second': 7, 'sum': 10},
        {'first': 'foo', 'second': 'bar', 'sum': 'foobar'},
        {'first': 2.5, 'second': 3.5, 'sum': 6.0}
    ]

@pytest.mark.parametrize('scenario', get_test_scenarios())
def test_calculate_sum(scenario):
    assert calculate_sum(scenario['first'], scenario['second']) == scenario['sum']

Exteranl Data Sources

Load test cases from JSON files:

test_data.json

[
    {"val_a": 3, "val_b": 7, "total": 10},
    {"val_a": 2.5, "val_b": 3.5, "total": 6.0}
]

test_module.py

import pytest
import json

def read_external_data():
    with open('test_data.json') as f:
        return json.load(f)

@pytest.mark.parametrize('dataset', read_external_data())
def test_calculate_sum(dataset):
    assert calculate_sum(dataset['val_a'], dataset['val_b']) == dataset['total']

Parameterization enhances test maintainability for data-driven validation scenarios.