elixir-lang Finding non-duplicate elements in a list

Question

I'm trying to find non duplicate values from a list e.g.

original list:

iex> list = [2, 3, 4, 4, 5, 6, 6, 6, 7, 8, 8, 8, 9, 9, 10, 10, 10]
[2, 3, 4, 4, 5, 6, 6, 6, 7, 8, 8, 8, 9, 9, 10, 10, 10]

iex> unique = Enum.uniq(list)
[2, 3, 4, 5, 6, 7, 8, 9, 10]

iex> nondupes = unique -- Enum.uniq(list -- unique)
[2, 3, 5, 7]

result: [2, 3, 5, 7]

I was wondering if there was a better way to achieve this in elixir/erlang

Answer 1

Another method (not necessarily better) that might be faster for large data is to build a map from elements to their counts and select the ones where count is 1:

list
|> Enum.reduce(%{}, fn (el, acc) -> Dict.put(acc, el, Dict.get(acc, el, 0) + 1) end)
|> Enum.filter(fn {key, val} -> val == 1 end)
|> Enum.map(fn {key, val} -> key end)

This should have runtime O(n * log(n)) instead of the O(n ^ 2) your solution requires (the subtraction should be quadratic if the whole input is unique already).