Simple failing test

[Alex-Fischman]

The following test fails:

#[test]
fn subsets_small2() {
  let mut v = SetTrie::new();
  v.insert(&[1, 2], 'a');
  
  let mut s = v.subsets(&[&0, &2]);
  let o = s.next();
  assert_eq!(o, None);
}

The test is correct because we construct a set-trie that only contains the set {1, 2}. We then query for all subsets of {0, 2}. However, since {1, 2} is not a subset of {0, 2}, we should get the empty iterator.

[KaiserKarel]

I'll check this out tomorrow. I use an in-tree fork internally, which has a bugfix for this I reckon.

[KaiserKarel]

@Alex-Fischman could you review and test 6-simple-failing-test and see if it resolves your issue?

[wilcoxjay]

There seems to be a similar problem with supersets:

    #[test]
    fn superset_small2() {
        let mut v = SetTrie::new();
        v.insert(&[0, 2], ());
        assert_eq!(v.supersets(&[&0, &1]).next(), None);
    }

Similar to the original test for subsets, the correct answer here is the empty iterator because {0,2} is not a superset of {0,1}.

[wilcoxjay]

And here's another small test for subsets that still fails on branch 6-simple-failing-test (though it passes on the main branch).

    #[test]
    fn subsets_small3() {
        let mut v = SetTrie::new();
        v.insert([], ());
        assert_eq!(v.subsets(&[0]).next(), Some(&()));
    }

Since the empty set is a subset of {0}, the iterator should not be empty.

[KaiserKarel]

I've identified the issue, booked out this weekend to resolve it.