Better support for asynchronous testing

Oct 13, 2025 by wbrum.org 49 views

Problem Statement

When writing tests for asynchronous functions, it can be challenging to assert that a specific exception is raised under certain conditions. The current implementation of pytest.raises does not support asynchronous context managers, making it difficult to write tests that are both readable and efficient.

Solution Proposal

To address this issue, we propose adding methods __aenter__ and __aexit__ to the RaisesContext class. This will enable the use of pytest.raises as an asynchronous context manager, allowing developers to write tests that are both concise and easy to understand.

Benefits of the Proposal

The proposed solution offers several benefits, including:

Improved readability: With the ability to use pytest.raises as an asynchronous context manager, tests will become more readable and easier to maintain.
Increased efficiency: The proposed solution will reduce the amount of code required to write tests, making it easier to focus on the logic of the test rather than the boilerplate code.
Better support for asynchronous testing: The proposed solution will provide better support for asynchronous testing, enabling developers to write tests that are both efficient and effective.

Alternative Solutions

Currently, developers can use a decorator to create an asynchronous context manager from an instance of RaisesContext. However, this approach requires additional code and can make tests more difficult to read.

Implementation Details

To implement the proposed solution, we will add the following methods to the RaisesContext class:

__aenter__: This method will be responsible for returning the context object when entering the asynchronous context.
__aexit__: This method will be responsible for handling the exit of the asynchronous context.

The implementation of these methods will be trivial, as they will simply delegate to the existing __enter__ and __exit__ methods.

Example Usage

With the proposed solution, tests for asynchronous functions will become more concise and easier to read. For example:

import pytest  # needs plugin pytest-asyncio

async def somefunc():
    await do_something()
    1 / 0  # raise ZeroDivisionError

async def test_somefunc():
    ctx = pytest.raises(ZeroDivisionError)
    async with ctx as ex:
        await somefunc()
    ...

Minimal Implementation

A minimal implementation of the proposed solution would be as follows:

class RaisesContext(ContextManager, AsyncContextManager):
    ...

    async def __aenter__(self):
        return self.__enter__()

    async def __aexit__(self, exc_type, exc_val, exc_tb):
        return self.__exit__(exc_type, exc_val, exc_tb)

Conclusion

Q&A

Q: What is the current issue with pytest.raises?

A: The current implementation of pytest.raises does not support asynchronous context managers, making it difficult to write tests that are both readable and efficient.

Q: How does the proposed solution address this issue?

A: The proposed solution adds methods __aenter__ and __aexit__ to the RaisesContext class, enabling the use of pytest.raises as an asynchronous context manager.

Q: What are the benefits of the proposed solution?

A: The proposed solution offers several benefits, including:

Improved readability: With the ability to use pytest.raises as an asynchronous context manager, tests will become more readable and easier to maintain.
Increased efficiency: The proposed solution will reduce the amount of code required to write tests, making it easier to focus on the logic of the test rather than the boilerplate code.
Better support for asynchronous testing: The proposed solution will provide better support for asynchronous testing, enabling developers to write tests that are both efficient and effective.

Q: What is the alternative solution currently being used?

A: Currently, developers can use a decorator to create an asynchronous context manager from an instance of RaisesContext. However, this approach requires additional code and can make tests more difficult to read.

Q: How does the proposed solution compare to the alternative solution?

A: The proposed solution is more efficient and easier to read than the alternative solution. With the proposed solution, developers can write tests that are both concise and effective, without requiring additional code.

Q: What is the minimal implementation of the proposed solution?

A: A minimal implementation of the proposed solution would be as follows:

class RaisesContext(ContextManager, AsyncContextManager):
    ...

    async def __aenter__(self):
        return self.__enter__()

    async def __aexit__(self, exc_type, exc_val, exc_tb):
        return self.__exit__(exc_type, exc_val, exc_tb)

Q: How can developers use the proposed solution in their tests?

A: Developers can use the proposed solution in their tests by importing pytest.raises and using it as an asynchronous context manager. For example:

import pytest  # needs plugin pytest-asyncio

async def somefunc():
    await do_something()
    1 / 0  # raise ZeroDivisionError

async def test_somefunc():
    ctx = pytest.raises(ZeroDivisionError)
    async with ctx as ex:
        await somefunc()
    ...

Q: What are the next steps for implementing the proposed solution?

A: The next steps for implementing the proposed solution would be to add the necessary code to the pytest.raises class and to test the implementation thoroughly.