Unless there is a reason the async client should not skip the tests (or a particular reason the synchronous one should) is it not better to make the tests test the same thing the same way (or implement #25 and condense that logic into a module that both versions use which would eliminate the need for testing in both areas.
Yes, the skipif should be added to the async test as well, but I'd keep the 2 tests separate. The tests should not know/care if valkey and valkey.asyncio share some implementation detail or not.
Version: 5.1.0b7
Platform: Python 3.12 / MacOS Sonoma in M1 Arm Device
Description:
There is an inconsistent skip on testing if SSL isn't installed between the connection pool in the sync and async tests.
These tests are otherwise identical (pointing to my suggested issue #25)
sync: https://github.com/valkey-io/valkey-py/blob/d116aa6efa03877a0c37598139a556554ad47c13/tests/test_connection_pool.py#L456C1-L494C1
async: https://github.com/valkey-io/valkey-py/blob/d116aa6efa03877a0c37598139a556554ad47c13/tests/test_asyncio/test_connection_pool.py#L543-L572
Unless there is a reason the async client should not skip the tests (or a particular reason the synchronous one should) is it not better to make the tests test the same thing the same way (or implement #25 and condense that logic into a module that both versions use which would eliminate the need for testing in both areas.