René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Replace lru_cache usage with cache","articleBody":"[`functools.lru_cache`](https://docs.python.org/3/library/functools.html#functools.lru_cache) provides a nice way of caching function returns, and is super useful for things like [`jimport`](https://github.com/scijava/scyjava/blob/af02aaad23a76cf6b85b4b8039fd1e0ae51f3721/scyjava/__init__.py#L307). As @ctrueden noted, however:\r\n\u003e I think [`@cache`](https://docs.python.org/3/library/functools.html#functools.cache) is the nicest. It just implies (rightly so) that function results are cached and returned again the second time. Whereas `lru_cache(maxsize=None)` is an oxymoron to me: if there is no size limit, it's not \"LRU\" anymore.\r\n\r\nThis is a good point. Unfortunately, `functools.cache` is a feature introduced in Python 3.9, while scyjava requires \u003e=3.6 and tests 3.8\r\n\r\nOnce we move on from these Python versions, we should probably reevaulate the usage of `lru_cache` on `jimport`.","author":{"url":"https://github.com/gselzer","@type":"Person","name":"gselzer"},"datePublished":"2022-03-30T15:21:50.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":0},"url":"https://github.com/35/scyjava/issues/35"}