Fastest way to remove subsets of lists from a list in Python

Question

Suppose I have a list of lists like the one below (the actual list is much longer):

fruits = [['apple', 'pear'],
          ['apple', 'pear', 'banana'],
          ['banana', 'pear'],
          ['pear', 'pineapple'],
          ['apple', 'pear', 'banana', 'watermelon']]

In this case, all the items in the lists ['banana', 'pear'], ['apple', 'pear'] and ['apple', 'pear', 'banana'] are contained in the list ['apple', 'pear', 'banana', 'watermelon'] (the order of items does not matter), so I would like to remove ['banana', 'pear'], ['apple', 'pear'], and ['apple', 'pear', 'banana'] as they are subsets of ['apple', 'pear', 'banana', 'watermelon'].

My current solution is shown below. I first use ifilter and imap to create a generator for the supersets that each list might have. Then for those cases that do have supersets, I use compress and imap to drop them.

from itertools import imap, ifilter, compress

supersets = imap(lambda a: list(ifilter(lambda x: len(a) < len(x) and set(a).issubset(x), fruits)), fruits)


new_list = list(compress(fruits, imap(lambda x: 0 if x else 1, supersets)))
new_list
#[['pear', 'pineapple'], ['apple', 'pear', 'banana', 'watermelon']]

I wonder if there are more efficient ways to do this?

Answer 1

filter(lambda f: not any(set(f) < set(g) for g in fruits), fruits)