Yeah, that was my concern with this one as well. Though clever, it's a quadric runtime rather than the more brute force single loop through the list of characters, storing which ones were seen in a hash map, then incrementing a counter whenever you see a letter twice for the first time. It's more code and less clever, but would be linear complexity..

Nice code. Just have to separate with spaces for readability. :)

not readable. :P

What is true time complexity for this approach? I believe filter in js is O(n), n being array size(or in this case text input length). And indexOf or lastIndexOf is O(n). So would it be fair to say that this approach has O(2 *(n^2)) or O(n^2) time complexity?