I think that "functools.lru_cache()" should have the ability to "invalidate" a (possibly cached) value. Something like:

@functools.lru_cache()
def func(param):
  ...

func.invalidate(PARAM) # discard this cached call, or ignore if not cached

Perhaps the existing ``cache_clear`` method could take optional arguments?

def cache_clear(self, *args, **kw):
    if not (args or kw):
        # clear the entire cache
    else:
        # clear just the cache entry for *args, **kw

Could cache_clear() clear the entire cache or discard just the cached call without arguments?

This is a duplicate of rejected issue28178.

See also related issue17528, issue18577, issue19859, issue23030, issue26082, and issue28112.

Sorry Jesús, I don't want to add feature creep to the LRU cache.

FWIW, it is very easy to make your own variants from collections.OrderedDict.

If you're interested, here is a starting point for experimenting with any variations you want (invalidate a specific entry, changeable maxsize, pickle/unpickle, expiration of entries after a specific time, inspection of the internal contents, ability to inject known values, testing whether a specific argument tuple is in the cache, logging of cache access, or just about anything you could do with a regular dictionary):

    class LRU(OrderedDict):

        def __init__(self, func, maxsize=128):
            self.maxsize = 128
            self.func = func

        def __call__(self, *args):
            if args in self:
                value = self[args]
                self.move_to_end(args)
                return value
            value = self.func(*args)
            if len(self) >= self.maxsize:
                self.popitem(False)
            self[args] = value
            return value