Optional arguments

They key to writing a decorator like lru_cache, that can be used as @lru_cache() or as @lru_cache(1000) is remembering that decorators are still Python functions. A Python function can accept optional arguments if the parameters are given a default value, which is what you must do when writing a decorator like this.

As far as the caching example goes, you could set 1000 as the default value of maxsize:

def cache_factory(maxsize=1000):
    ...

By doing this, you allow cache_factory to be used as @cache_factory(), in which case the cache will have a default maximum size of 1,000 items.

Alternatively, you could set the default value to None and then tweak the decorator so that @cache_factory() means the cache will have no maximum size:

def cache_factory(maxsize=None):
    def cache(f):
        cache = {}
        def cached_f(*args):
            if args not in cache:
                #  vvvvvvvvvvv
                if maxsize is not None and len(cache) >= maxsize:
                    cache.popitem()
                cache[args] = f(*args)
            return cache[args]

        return cached_f

    return cache

The new conditional statement if maxsize and len(cache) >= maxsize only checks the size of the cache if maxsize is not the default value None. If maxsize is the default value None, then the cache can grow unbounded.

You can now use cache_factory as @cache_factory() or as @cache_factory(1000), but how do you make it so that the decorator works without any parentheses..?