@spec at!(Enum.t(), integer() | [integer()], any()) :: any()

Like Enum.at/3 but raises if index is out of bounds.

@spec compact(Enum.t()) :: Enum.t()

Removes nil elements from enum.

@spec compact_blank(Enum.t()) :: Enum.t()

Removes blank elements (as determined by Moar.Term.blank?1) from enum.

@spec find_indices(Enum.t(), [any()], (any(), any() -> boolean())) :: [integer()]

Returns the indices of elements in enum, using fun for comparisons (defaulting to Kernel.==/2)

iex> Moar.Enum.find_indices(~w[apple banana cherry], ~w[cherry apple])
[2, 0]

iex> Moar.Enum.find_indices(~w[apple banana cherry], ~w[CHERRY APPLE], fn a, b ->
...>   String.downcase(a) == String.downcase(b)
...> end)
[2, 0]

@spec find_indices!(Enum.t(), [any()], (any(), any() -> boolean())) :: [integer()]

Returns the indices of elements in enum, raising if any member of elements is not found.

iex> Moar.Enum.find_indices!(~w[apple banana cherry], ~w[cherry apple])
[2, 0]

iex> Moar.Enum.find_indices!(~w[apple banana], ~w[cherry apple])
** (RuntimeError) Element "cherry" not present in:
["apple", "banana"]

@spec first!(Enum.t()) :: any()

Returns the first item of enum, or raises if it is empty.

@spec index_by(Enum.t(), (any() -> any())) :: map()

Converts an enum into a map of maps indexed by the return value of index_fun. See also the similar map-specific Moar.Map.index_by/2.

iex> Moar.Enum.index_by([%{name: "Alice", tid: "alice"}, %{name: "Billy", tid: "billy"}], & &1.tid)
%{"alice" => %{name: "Alice", tid: "alice"}, "billy" => %{name: "Billy", tid: "billy"}}

@spec into!(nil | Enum.t(), Enum.t()) :: Enum.t()

Like Enum.into but accepts nil as the first argument

@spec is_map_or_keyword(any()) :: boolean()

Deprecated in favor of map_or_keyword?/1.

@spec is_map_or_nonempty_keyword(any()) :: boolean()

Deprecated in favor of map_or_nonempty_keyword?/1.

@spec isort(Enum.t()) :: Enum.t()

Sorts enum case-insensitively. Uses Enum.sort_by/3 under the hood.

@spec isort_by(Enum.t(), (any() -> any())) :: Enum.t()

Sorts enum case-insensitively by mapper function. Uses Enum.sort_by/3 under the hood.

@spec lists_to_maps([[any()]], [any()] | :first_list) :: [map()]

Converts a list of lists to a list of maps with the given keys. The keys can be a list, or can be :first_list which uses the first list in list_of_lists as the keys and the remaining lists in list_of_lists as the values.

iex> Moar.Enum.lists_to_maps([[1, 2], [3, 4]], ["a", "b"])
[%{"a" => 1, "b" => 2}, %{"a" => 3, "b" => 4}]

iex> csv = [["a", "b"], [1, 2], [3, 4]]
iex> [headers | rows] = csv
iex> Moar.Enum.lists_to_maps(rows, headers)
[%{"a" => 1, "b" => 2}, %{"a" => 3, "b" => 4}]

iex> csv = [["a", "b"], [1, 2], [3, 4]]
iex> Moar.Enum.lists_to_maps(csv, :first_list)
[%{"a" => 1, "b" => 2}, %{"a" => 3, "b" => 4}]

@spec map_or_keyword?(any()) :: boolean()

Returns true if the value is a map or a keyword list. This uses standard Elixir functions for determining if a term is a map or a keyword, and therefore counts an empty list as a keyword list. See also map_or_nonempty_keyword?/1.

This cannot be used as a guard because it uses Keyword.keyword? under the hood. Also, because of that, it might scan an entire list to see if it's a keyword list, so it might be expensive.

@spec map_or_nonempty_keyword?(any()) :: boolean()

Like map_or_keyword?/1 but returns false if the term is an empty list.

@spec take_at(Enum.t(), integer() | [integer()] | :all) :: any()

Returns a list of elements at the given indices, in the given order. If :all is given instead of a list of indices, the entire enum is returned.

iex> Moar.Enum.take_at(["A", "B", "C"], [0, 2])
["A", "C"]

iex> Moar.Enum.take_at(["A", "B", "C"], [2, 0])
["C", "A"]

iex> Moar.Enum.take_at(["A", "B", "C"], :all)
["A", "B", "C"]

@spec test_ids(Enum.t(), [test_ids_opts()]) :: list()

Returns :test_id fields from enumerable.

This unusual function exists because the authors of Moar use test_ids extensively in tests. See also tids/2.

@spec tids(Enum.t(), [tids_opts()]) :: list()

Returns :tid fields from enumerable.

This unusual function exists because the authors of Moar use tids (test IDs) extensively in tests. See also test_ids/2.