Lru_cache provides caching with least-recently-used eviction policy.
Lru_cache has seen many years of production use in Iron. We intend to make it the preferred LRU cache implementation.
Related functionality (as of 2019-05):
Core.Memo.lru: Its interface is written for transparent memoization of functions and thus exposes no access to the cache itself.Memo_cache: Uses a re-implementation of Hash_queue. Cannot change cache capacity after creation. Its README suggests using Core.Memo instead.