`map_put` result depends on representation

[lukaszsamson]

Existing issue

• I have searched existing issues and could not find a duplicate.

Elixir and Erlang/OTP versions

Erlang/OTP 28 [erts-16.4.0.1] [source] [64-bit] [smp:12:12] [ds:12:12:10] [async-threads:1] [jit]

Interactive Elixir (1.21.0-dev)

Operating system

any

Current behavior

Repro 1, :
two different representations of none() return different results

a = none()
a2 = open_map([c: none()])

equal?(a, a2) # true

map_put(a, atom([:a]), integer()) # :badmap, correct
map_put(a2, atom([:a]), integer()) # {:ok, %{map: {41739984, :open, [a: %{bitmap: 8}, c: %{}]}}}, wrong

Repro 2:
two different representations, a and a2 = (a n knife) u (a \ knife), both return {:ok, _} but the results differ (r2 is subtype of r1)

    a = closed_map([a: opt_negation(open_map([{[:port], float()},
          {:a, closed_map([{[:reference], boolean()}, {:b, boolean()}, {:a, boolean()}])},
          {:b, opt_difference(fun(1), tuple())}]))])
    knife = closed_map([{[:map], opt_negation(open_tuple([fun(1)]))}])
    a2 = opt_union(opt_intersection(a, knife), opt_difference(a, knife))
    equal?(a, a2)  # true
    {:ok, r1} = map_put(a, atom(), knife)
    {:ok, r2} = map_put(a2, atom(), knife)
    equal?(r1, r2) # false, wrong
    subtype?(r2, r1) # true

Expected behavior

Result equal regardless of representation

[lukaszsamson]

A simpler repro of b, updated to the new API after optional refactor:

    a = closed_map([])
    c = closed_map([{:a, {term(), false}}]) 
    cover = opt_union(a, c)
    a2 = opt_difference(cover, opt_difference(cover, a))
    equal?(a, a2) # true
    {:ok, r1} = map_put(a, atom(), c)
    {:ok, r2} = map_put(a2, atom(), c)
    equal?(r1, r2) # false