pytest-maybe-raises

Pytest fixture for optional exception testing.

Documentation

I like to parametrize tests for functions that could accepts different values. I often find myself writing this type of pattern:

import pytest

@pytest.mark.parametrize(
    ('value', 'expected_result'),
    (
        ('foo', 'bar'),
        ('baz', ValueError)
    )
)
def test_something(value, expected_result):
    if hasattr(expected_result, '__traceback__'):  # is an exception
        with pytest.raises(expected_result):
            my_func(value)
    else:
        result = my_func(value)
        assert result == expected_result
        assert bool(result) 

But this is really long and ugly.

Does pytest provides a fixture that allows me to only uses the pytest.raises context when the provided argument is an exception class? The answer is not (see this question in Stackoverflow), so I've written this convenient fixture:

import contextlib
import pytest

@contextlib.contextmanager
def _maybe_raises(maybe_exception_class, *args, **kwargs):
    if hasattr(maybe_exception_class, '__traceback__'):
        with pytest.raises(maybe_exception_class, *args, **kwargs) as exc_info:
            yield exc_info
    else:
        yield maybe_exception_class

@pytest.fixture()
def maybe_raises():
    return _maybe_raises

This package provides a wrapper fixture over pytest.raises which context only has effect if the passed argument is an exception class, otherwise uses a null context like contextlib.nullcontext.

Using this package I can rewrite my tests as:

import pytest

@pytest.mark.parametrize(
    ('value', 'expected_result'),
    (
        ('foo', 'bar'),
        ('baz', ValueError)
    )
)
def test_something(value, expected_result, maybe_raises):
    with maybe_raises(expected_result):
        result = my_func(value)
        assert result == expected_result
        assert bool(result)

But in order to use it properly you need to know how the magic works:

Note that when an exception raises inside pytest.raises context, the context exits so the later assert ... == expected_result comparison is not executed. If other exception raises would be propagated to your test so the comparison is not executed either. This allows you to write more assertions after the execution of the function for successfull calls.