René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Improve tests of `typing.override` ","articleBody":"Right now we have several things to improve:\r\n1. This branch is not covered by tests: https://github.com/python/cpython/blob/4c87537efb5fd28b4e4ee9631076ed5953720156/Lib/typing.py#L3490 We don't have a single test that ensures that `@override` will not fail for some weird cases. Since 99.99% of use-cases will simply be `@override def some(...): ...`, it is not very important - but still desirable\r\n2. We don't test `@classmethod`s, I guess it is special enough to be included\r\n3. `@property` is also missing from tests, but is mentioned in PEP: https://peps.python.org/pep-0698/#specification and https://peps.python.org/pep-0698/#limitations-of-setting-override\r\n4. PEP also mentions `@functools.lru_cache` and similar decorators as an example of how wrappers should be nested. Why don't we test it? Right now there are no tests for nested method decorators (except `staticmethod`)\r\n5. It might not hurt to test that parent's methods definitions are not mutated (I think this is a part of the contract)\r\n\r\nPR is on its way :)\n\n\u003c!-- gh-linked-prs --\u003e\n### Linked PRs\n* gh-102325\n\u003c!-- /gh-linked-prs --\u003e\n","author":{"url":"https://github.com/sobolevn","@type":"Person","name":"sobolevn"},"datePublished":"2023-02-28T13:09:27.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":0},"url":"https://github.com/102324/cpython/issues/102324"}