Oh yeah, I missed that hashtable ordering is undeterministic. I played a
lot with trees these days.

Anyway, how's the proposal itself?

>
> How is it needed for Map/Set? Those use SameValueZero, and don't have a
sorting method.
>
> On Sun, Sep 18, 2016 at 10:32 PM, Park Hyeonu <nemo1275 at gmail.com> wrote:
>>
>> Well, this proposal is not for check equality. Its purpose is to compare
two object that which object is `greater`.
>>
>> I'm sure this feature is already implemented for every modern js
engines, as it's needed to implement Map/Set. But we don't have js
interface for it yet. Thus proposal is to expose this functionality to
userland application.
>>
>>
>> 2016. 9. 19. 오후 2:26에 "Jordan Harband" <ljharb at gmail.com>님이 작성:
>>
>>> This may have already been proposed as
https://github.com/sebmarkbage/ecmascript-shallow-equal - if so, you'll
want to post your comments there.
>>>
>>> On Sun, Sep 18, 2016 at 10:10 PM, Park Hyeonu <nemo1275 at gmail.com>
wrote:
>>>>
>>>> tl;dr - deterministic function that compares arbitrary objects
>>>>
>>>> I know now we have `Map` and `Set`, but sometimes we need more low
level ones. Especially for library developers.
>>>>
>>>> For the case, say, I want to implement a map whose key can be a
arbitrary number of objects. Currently this can only implemented as nested
map whose structure is `keylen -> key1 -> key2 -> ... -> value`. Not really
elegant. But if we have `Object.compare()`, `(key1, key2, ...) -> value` is
possible.
>>>>
>>>> As its intended purpose is to be used with custom map/set, I don't
think we should define how this function works in spec. Just guaranteed to
be deterministic in each runtime is enough.
>>>>
>>>> For detail, this function follows `(left: any, right: any) => number`
form. When `Object.is(left, right)` is true, `Object.compare(left, right)`
must return `0`. Otherwise, the result must be deterministic, means
whenever(in same runtime) we call this function with same arguments it
should return same value(or at least, always greater than/lesser 0).
>>>>
>>>> Seems good?
>>>>
>>>>
>>>> _______________________________________________
>>>> es-discuss mailing list
>>>> es-discuss at mozilla.org
>>>> https://mail.mozilla.org/listinfo/es-discuss
>>>>
>>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.mozilla.org/pipermail/es-discuss/attachments/20160919/0c4a7adf/attachment.html>

Oh yeah, I missed that hashtable ordering is undeterministic. I played a
lot with trees these days.

Anyway, how's the proposal itself?

(Wow, I got really confused how this mail thread works)