87 – Caching sets and frozen sets

87 – Caching sets and frozen sets#

The built-in type set has a variant frozenset that is immutable and that you can use as a dictionary key, for caching purposes, or in other contexts where you require a hashable value.

As an example, suppose you have a function products_with_tags that searches for products in a store that have the given tags, and since the database of products is fairly stable, you want to cache these lookups.

You write the function with a cache:

from functools import cache

@cache
def products_with_tags(tags):
    ...  # Send an API request,
    ...  # process the results, etc.

Since sets aren’t hashable, you can’t call this function with an argument of the built-in type set:

tags = {"sale", "local"}
print(products_with_tags(tags))  # Exception

Since frozen sets are immutable, they can be made hashable, and thus they can be used here:

tags = frozenset({"sale", "local"})
print(products_with_tags(tags))

Frozen sets support all operations that sets support and, in particular, they support very fast lookups/containment checks.

The only operations that frozen sets do not support are the set operations that modify the set in-place.