After checking the source, it appears I was presuming wrong, as `:maps.map`
is not currently optimized like it could be (PR to come one day đŸ˜…)
On Thursday, 13 January 2022 at 11:55:24 UTC-6 Marc-André Lafortune wrote:
> I somehow missed both the conversation to introduce `Map.map` and that
> about deprecating it... đŸ¤”
>
> What I tend to use a lot is applying a 1-arity function to all values:
>
> ```
> Map.new(my_map, fn {key, value} -> {key, do_something(value)} end)
> ```
>
> Probably because I come from Ruby, I miss `transform_values`, both for its
> conciseness and it's expressivity. It makes it clear that the result will
> be a map with the same keys.
> I saw a similar request:
> https://elixirforum.com/t/suggestion-map-map-2/29451
>
> The example above could be written using two short forms that were not
> available:
> ```
> Map.new(my_map, fn {key, value} -> {key, do_something(value)} end)
> # Makes it possible to inline an expression with &(...&1...):
> Map.transform_values(my_map, &do_something(&1))
> # or if transform is actually a function:
> Map.transform_values(my_map, &do_something\1)
> ```
>
> From an efficiency point of view, `Maps.transform_values` can use
> `maps:map`, which will be more efficient (I presume avoids all key
> comparisons, for small maps reuses the list of keys, for large maps avoids
> rehashing, comparing and organizing the keys)
>
> For completedness sake, I'd propose `Map.transform_keys` which I also use,
> although less often. It has no performance advantage, but remains clear in
> its intent.
>
>
--
You received this message because you are subscribed to the Google Groups
"elixir-lang-core" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to elixir-lang-core+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/elixir-lang-core/32012234-b21c-41e3-b5d7-16c421ba9237n%40googlegroups.com.
