2023-05-03
[05:35:40.0560] <annevk>
littledan: I think https://github.com/whatwg/html/issues/9042 might be of interest to you

[10:01:09.0230] <ptomato>
is there an existing idiom in ECMA-262 spec language for removing duplicates from a List?

[10:01:48.0659] <bakkot>
don't think so

[10:01:58.0423] <bakkot>
for Sets etc we just avoid putting duplicates in the list in the first place

[10:02:53.0304] <bakkot>
the usual idiom for which is like "If `list` does not contain `item`, add `item` to `list`"

[10:03:34.0080] <bakkot>
(Set.prototype.add is written in a slightly different style for... reasons)

[10:04:29.0312] <ptomato>
thanks!

[10:05:02.0071] <shu>
if possible, doing deduplication up front is better yeah

[10:05:13.0354] <shu>
i think it's probably fine to say "remove duplicates from `list`" if you need to do it after the fact, but we'd need to add a bit in the identity section about that language using "is" for comparison

[15:49:36.0220] <ptomato>
we don't have an existing idiom for break/continue in spec language loops either, right?

[16:06:09.0930] <ljharb>
not that i know of

[16:06:18.0327] <ljharb>
usually that would be in an AO that early-returns

[16:06:22.0086] <ljharb>
 * usually that might be in an AO that early-returns


2023-05-04
[10:42:22.0647] <TabAtkins>
We have the idiom in Infra, but not in ES-ese.

[11:16:54.0650] <Michael Ficarra>
ptomato: try to avoid it if you can


2023-05-10
[11:12:39.0194] <bakkot>
```
function f(){}
let sym = Symbol();
f[Symbol.toPrimitive] = () => sym;
let o = { [sym]: 0 };
console.log(o[f]) // 0
```
... hm

[11:13:08.0679] <bakkot>
keying objects by functions

[11:13:09.0878] <bakkot>
kinda fun

[11:13:20.0443] <shu>
cursed

[11:27:20.0748] <bakkot>
```
let functorMapSymbol = Symbol('Functor.map');
let Functor = {};
Functor.map = function(rec, ...args){ return rec[functorMapSymbol](...args); };
Functor.map[Symbol.toPrimitive] = () => functorMapSymbol;

let o = {
  [Functor.map](arg) {
    console.log(arg);
  },
};

// usage:
o[Functor.map]('hi');

// alternative usage:
Functor.map(o, 'hi');
```

[11:29:10.0130] <Francisco Tolmasky>
I actually do something similar to that all the time. Also use it to "carry" information in properties, something like { [property `my-prop` ({ enumerable: false })]: 10 }

[11:29:38.0463] <Francisco Tolmasky>
 * I actually do something similar to that all the time. Also use it to "carry" information in properties, something like `{ [property \`my-prop\` ({ enumerable: false })]: 10 }`

[11:29:51.0241] <Francisco Tolmasky>
 * I actually do something similar to that all the time. Also use it to "carry" information in properties, something like `{ [property 'my-prop' ({ enumerable: false })\]: 10 }\`

[11:30:44.0792] <Francisco Tolmasky>
dont know how to nest backticks, but you get the idea, you can have a function that creates metadata, then creates a custom symbol on the fly, associates the metadata to the custom symbol in a weakmap, then returns the custom symbol, then on the other end you can use it to retrieve the metadata

[11:32:15.0246] <Francisco Tolmasky>
for example like `Schema({ [property 'whatever'.readonly]: 10 })` (Schema then knows how to retrieve the metadata and construct the property descriptor from it to create the property on itself


2023-05-17
[11:22:42.0051] <Michael Ficarra>
is anybody interested in maintaining the queue tool that we use (TCQ)? https://github.com/bterlson/tcq

[11:22:53.0805] <Michael Ficarra>
it's a pretty simple, self-contained Vue app

[11:23:07.0293] <shu>
is brian stepping away?

[11:23:16.0637] <Michael Ficarra>
the committee is *highly* dependent on it

[11:23:45.0053] <Michael Ficarra>
shu: Brian hasn't always been available to maintain it

[11:23:49.0334] <Michael Ficarra>
Yulia took it up for a while

[11:24:18.0048] <Michael Ficarra>
but there's a decent list of issues and feature requests that I'd love to see some progress on: https://github.com/bterlson/tcq/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-desc

[11:24:49.0067] <Michael Ficarra>
if someone from the community is looking for a way that they could contribute, tackling some of these would be very much appreciated

[11:27:31.0332] <Luca Casonato>
I tried contributing to it a few months back, and got stuck on a bunch of Azure dependencies

[11:28:41.0811] <Michael Ficarra>
well, having it support other cloud environments would be a great contribution then 🙂

[11:29:17.0164] <Luca Casonato>
I’ll give it another shot

[11:32:59.0744] <Michael Ficarra>
anything to make it easier to maintain or contribute to, including documentation, would be a great force multiplier

[11:33:18.0567] <Chris de Almeida>
I also tried to fiddle with it some time back -- even something that explains how to run it would be helpful

[11:33:41.0760] <Chris de Almeida>
 * I also tried to fiddle with it some time back -- even something that explains how to run it (dev env) would be helpful

[11:36:56.0419] <Luca Casonato>
I was thinking whether “just” switching the backend to some SQLite db and plopping it onto a VPS somewhere might make it all a bit less complex


2023-05-19
[20:25:40.0197] <sideshowbarker>
https://tc39.es/proposal-array-grouping/#sec-array.prototype.group and https://tc39.es/proposal-array-grouping/#sec-array.prototype.grouptomap seem to have been removed from the Array Grouping proposal?

[20:26:49.0200] <ljharb>
> <@sideshowbarker:matrix.org> https://tc39.es/proposal-array-grouping/#sec-array.prototype.group and https://tc39.es/proposal-array-grouping/#sec-array.prototype.grouptomap seem to have been removed from the Array Grouping proposal?

yes, now they’re Object.groupBy and Map.groupBy

[20:27:08.0217] <ljharb>
also it’d stage 2 now. Hoping to go back to stage 3 next meeting.

[20:27:16.0230] <ljharb>
* also it’s stage 2 now. Hoping to go back to stage 3 next meeting.

[20:27:30.0093] <sideshowbarker>
> <@ljharb:matrix.org> yes, now they’re Object.groupBy and Map.groupBy

Thanks — I’ll update the URLs in MDN

[03:38:16.0076] <littledan>
> <@sideshowbarker:matrix.org> Thanks — I’ll update the URLs in MDN

It might make sense for MDN to wait on including these docs until it’s at stage 3 and/or some browser ships it

[10:13:08.0688] <shu>
i would wait at least until stage 3