Dict get by key is O(log(n)), List get by 'key' is O(n). Thus not really a
difference until you get to a *lot* of items that are key'd but not ordered.
On Saturday, September 3, 2016 at 2:28:04 AM UTC-6, Brian John Farrar wrote:
>
> I am looking for design guidelines on when to model a collection of
> records as a Dict vs. a List.
>
> The TodoMVC uses a List, though at first glance a Dict seems like a more
> natural match:
> - There are several "id based" Msg's (ex. EditingEntry, UpdateEntry,
> Delete, Check)
> - Using Html.Keyed should be (slightly) easier to use
>
> But I think I understand why a List is more pragmatic.
> In (my) practice, things get muddy attempting to implement TodoMVC with a
> Dict.
> Dict.update wants a (Maybe v -> Maybe v) resulting in nested case
> statements which are okay but preferable to avoid.
> (...it could be that my FP-fu is too weak to factor a concise expression.
> Maybe.map doesn't appear useful as record updates aren't compose-able?)
>
> So, the "mapWhen" idiom (via an "updateEntry" helper) does seem to be the
> simplest approach
>
> let
> updateEntry t =
> if ti.id = id then { t | field = newValue } else t
> in
> { model | entries = List.map updateEntry model.entries }
>
> So if Dict isn't the best choice for a use case like TodoMVC...what kinds
> of use cases is it good for?
>
>
>
>
--
You received this message because you are subscribed to the Google Groups "Elm
Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
