Generates multiple test invocations from a single test function using supplied argument tuples.

Instructs the test runner to execute the decorated test function once per set of argument values.

When writing repetitive test cases that differ only in input values, to reduce boilerplate and improve readability.

Inside a pytest test module, directly above a test function definition that accepts the parametrized arguments.

@pytest.mark.parametrize('<arg_names>', [<tuple_of_values>])

@pytest.mark.parametrize followed by parentheses containing a comma‑separated string of argument names and a list of tuples holding the test values.

arg_names (comma‑separated string of parameter names), test_cases (list of tuples matching the number of arguments)

• pytest
• test
• def test_
• assert
• param
• ids

• arg names: 'x
• expected'
• 'input
• expected'
• test cases: [(1
• 2)
• (2
• 4)]
• [(0
• 0)
• (1
• 1)
• (2
• 4)]
• [('hello'
• 5)
• ('world'
• 5)]

Mismatch between number of argument names and tuple elements, forgetting to import pytest.mark, using incorrect argument names in the test function, assuming lists are required (tuples work but lists are also accepted), mismatched data types.

Similar to @pytest.mark.parametrize(indirect=True) for fixture parametrization; contrasted with @pytest.fixture for providing setup data and with @pytest.mark.parametrize without ids for anonymous test cases.

Confusing @pytest.mark.parametrize with @pytest.mark.parametrize(indirect=...) when using fixtures; mixing up argument order; forgetting that each tuple corresponds to one test invocation.

• pytest parametrize pattern

Enables data‑driven testing while keeping test code concise; each tuple yields a distinct test case reported separately by pytest.

Signals to readers that the test is data‑driven, encouraging concise and maintainable test suites.

Parametrize your test with input‑output pairs to avoid repetitive test functions.

The decorator generates separate test invocations; each tuple produces its own test case visible in pytest output.

Consider adding ids for better readability, or switch to indirect parametrization with fixtures for more complex setups.