pytest-case-provider 0.3.0

Advanced pytest parametrization plugin that generates test case instances from sync or async factories.

pytest-case-provider

Case parametrization for pytest with fixture, class, and async support. Provides declarative, typed, on demand case injection for both sync and async test functions.

---

Overview

pytest-case-provider extends pytest’s parametrization system.
It was inspired by pytest-cases, but redesigned from scratch for strict typing, async support, and fixture-native case injection.

It allows attaching case providers directly to test functions or methods via @inject_cases_func and
@inject_cases_method.

Cases can be:

• Synchronous or asynchronous.
• Iterable or async-iterable.
• Fixture-dependent.
• Composable across tests or classes.

---

API Summary

Symbol	Description
inject_cases_func	Decorator/injector for test functions
inject_cases_method	Same as above, for test class methods
CaseStorage[T]	Mutable case storage container
CompositeCaseStorage[T]	Aggregates multiple CaseCollector

---

Installation

pip install pytest-case-provider

---

Usage example

import typing
from dataclasses import dataclass, replace
import pytest
from pytest_case_provider import inject_cases_func, inject_cases_method


@dataclass(frozen=True)
class MyCase:
    foo: int


@pytest.fixture
def number() -> int:
    return 42


# Regular test
def test_without_case_injection() -> None:
    assert True


# Case-enabled test function
@inject_cases_func()
def test_case_injected(case: MyCase) -> None:
    assert isinstance(case, MyCase)


# Cross-inject from another test
@inject_cases_func.include(test_case_injected)
def test_case_increment(case: MyCase, case_foo_inc: MyCase) -> None:
    assert case.foo + 1 == case_foo_inc.foo


# Define case providers
@test_case_injected.case()
def case_one() -> MyCase:
    return MyCase(foo=1)


@test_case_injected.case()
def case_two() -> MyCase:
    return MyCase(foo=2)


# Use other fixtures in case providers
@test_case_injected.case()
def case_number(number: int) -> MyCase:
    return MyCase(foo=number)


# Async case provider
@test_case_injected.case()
async def case_async_generated() -> MyCase:
    return MyCase(foo=999)


# Iterable provider
@test_case_injected.case()
def case_iterable() -> typing.Iterator[MyCase]:
    yield MyCase(foo=10)


# Async iterable provider
@test_case_injected.case()
async def case_async_iterable() -> typing.Iterator[MyCase]:
    yield MyCase(foo=20)


# Access case object from fixture
@pytest.fixture
def case_foo_inc(case: MyCase) -> MyCase:
    return replace(case, foo=case.foo + 1)


# Example class-based usage
class TestClass:
    @inject_cases_method()
    def test_class_cases(self, case: MyCase) -> None:
        assert isinstance(case, MyCase)

    @test_class_cases.case()
    def case_three(self) -> MyCase:
        return MyCase(foo=3)

    @test_class_cases.case()
    def case_four(self) -> MyCase:
        return MyCase(foo=4)

Test Discovery Output

Pytest will expand each injected case as a distinct test variant using pytest's parametrization:

test_example.py::test_case_injected[case_one]
test_example.py::test_case_injected[case_two]
test_example.py::test_case_injected[case_number]
test_example.py::test_case_injected[case_async_generated]
test_example.py::test_case_injected[case_iterable]
test_example.py::test_case_injected[case_async_iterable]
test_example.py::test_case_increment[case_one]
test_example.py::test_case_increment[case_two]
test_example.py::test_case_increment[case_number]
test_example.py::test_case_increment[case_async_generated]
test_example.py::test_case_increment[case_iterable]
test_example.py::test_case_increment[case_async_iterable]
test_example.py::TestClass::test_class_cases[case_three]
test_example.py::TestClass::test_class_cases[case_four]