| 01:30 | <weinig> | annevk: is there anyway for a non-persistent notiification to know which of its actions, if any, was clicked? Or is that functionality only avaliable for persistent notifications via NotificationEvent? |
| 01:39 | <MikeSmith> | beverloo_: ⬆ |
| 03:06 | <annevk> | weinig: might have been an oversight, best file an issue |
| 03:06 | <weinig> | annevk: ok, will do |
| 03:06 | <weinig> | thanks |
| 03:49 | <annevk> | Domenic: you'll file the remainder of the worker name bugs, right? |
| 03:49 | <Domenic> | annevk: Oh shoot, my plan was to wait until we had implementer commitments before merging... |
| 03:50 | <Domenic> | I guess filing bugs is the way to go now though :-/ |
| 03:50 | <annevk> | Domenic: yeah, I was a little surprised you merged it |
| 03:50 | <Domenic> | That's what I get for working before going to sleep |
| 03:50 | <annevk> | Domenic: maybe mention in the bugs that a "late" objection is reasonable |
| 03:51 | <Domenic> | sg |
| 04:36 | <annevk> | Domenic: do you know more on https://stackoverflow.com/questions/40939857/fetch-with-readablestream/41222366? |
| 06:03 | <MikeSmith> | botie, inform zcorpan please look at https://bugzilla.mozilla.org/show_bug.cgi?id=1364281. <!DOCTYPE html<title>test</title><template><td>fdsa</template> Should we update the spec to say that the </td> closing tag must not be omitted when the td element is a child of a template element? And same for the tr element. And maybe for some other elements? I mean because in the template context, parsing does |
| 06:03 | <MikeSmith> | not result in an implicit </td> being seen there |
| 11:55 | <tobie> | What are the examples, if any, of requesting a trusted event to use an API on the platform? |
| 11:56 | <annevk> | tobie: <input type=file> |
| 11:57 | <tobie> | annevk: ta |
| 11:57 | <annevk> | tobie: https://html.spec.whatwg.org/multipage/interaction.html#triggered-by-user-activation |
| 11:57 | <tobie> | annevk: beautiful! |
| 13:04 | <jochen__> | annevk: any idea about https://github.com/whatwg/html/issues/2622#issuecomment-301061744 ? |
| 13:05 | <annevk> | jochen__: no, I haven't been involved much with download |
| 13:33 | <annevk> | So close, but no sigar: https://github.com/whatwg/html/pull/2671 |
| 14:32 | <beverloo_> | MikeSmith, annevk, weinig - I'll reply to an issue too, but the reason that actions aren't supported on non-persistent notifications is that defining their behaviour when the page goes away is hard |
| 14:32 | <beverloo_> | clicking on a non-persistent notification that doesn't have attached event handlers anymore just opens the page again, but that likely won't make sense for buttons |
| 14:33 | <beverloo_> | the page also has no way of knowing which button (if any) the user clicked on |
| 14:34 | <MikeSmith> | beverloo_: I see yeah that makes sense |
| 15:20 | <annevk> | beverloo_: we could close the non-persistent navigations once the document is unloaded though |
| 15:20 | <annevk> | beverloo_: perhaps silently so |
| 15:21 | <beverloo_> | that'd break Google Calendar |
| 15:21 | <beverloo_> | and presumably other apps too |
| 15:23 | <annevk> | Domenic: re https://github.com/WebAssembly/design/issues/1048 presumably non-enumerable was intentional by TC39 so folks designing new classes might want to have that too (such as you did with Streams) |
| 15:23 | <annevk> | Domenic: taking reasons away to avoid IDL can eventually lead to a lot more sharing of code I think |
| 15:24 | <annevk> | beverloo_: I was thinking only for actions, but it's a little icky |
| 15:24 | <beverloo_> | annevk, that's a bit hard to reason about though |
| 15:24 | <beverloo_> | from a developer point of view |
| 15:25 | <annevk> | beverloo_: yeah |
| 15:32 | <Domenic> | annevk: it's not really intentional in the wasm case though |
| 15:32 | <annevk> | Domenic: it seemed like a nice testcase, but I'm fine with letting it go |
| 15:33 | <annevk> | One day |
| 15:33 | <Domenic> | For TC39... it's important that Array and Object properties not be enumerable, but for everything else it's meh... |
| 15:33 | <Domenic> | We're stuck between consistency with the web platform (enumerable) and Array/Object (non-enumerable) |
| 15:33 | <Domenic> | And there are enough people on the committee who don't care about the web platform that I doubt we'll change for future classes |
| 15:41 | <annevk> | Domenic: right, but the web can change for new classes too |
| 15:51 | <annevk> | I'm really happy with https://github.com/whatwg/html/issues/2671 |
| 15:52 | <annevk> | There's still issues of course, but once that lands it will be possible to solve them |
| 15:53 | <annevk> | Domenic: if you can review the above today and ignore the XXX segment that'd be good |
| 15:55 | <annevk> | Domenic: the document.open/write/writeln refactor too, though that's less pressing navigation-wise, it just bothers me what poor state that whole thing was in |
| 16:05 | <Domenic> | annevk: can do! I've been seeing the emails go by and it seemed kind of depressing, so I'm glad you're feeling like it's going well. |
| 16:06 | <annevk> | Domenic: it's not been an easy week, but I learned a lot and made progress (and I documented most issues I came across this time) |
| 16:07 | <annevk> | Domenic: the main issue is that each time you see something you want to fix, there's something else you have to fix first, and on it goes |
| 16:37 | <annevk> | Domenic: btw, I've got about 8 more working days until I'm gone so if we need to do abort controllers Monday would be a good time to start |
| 16:38 | <Domenic> | annevk: eek, ok, sounds like a plan |
| 16:38 | <annevk> | Domenic: I can go through what needs to be done in DOM still |
| 16:39 | <annevk> | Domenic: Fetch is going to wait until July |
| 16:39 | <annevk> | Domenic: btw, do you know when ReadableStream for requests happens? |
| 16:40 | <Domenic> | annevk: implementation? yhirano__ is working on some refactoring to make it possible, but it's high priority from what i understand |
| 16:40 | <annevk> | Domenic: looking forward to it |
| 16:41 | <annevk> | I really hope that by the end of the summer unit of related browsing contexts is created imperatively |
| 16:41 | <annevk> | Stretch goal would be doing the same for agent / agent clusters |
| 16:42 | <annevk> | Btw, Domenic, I'm hoping you still want to do History at some point, I'm only touching that when necessary |
| 16:43 | annevk | -> weekend |
| 16:45 | <Domenic> | Yeah, at some point... keep getting dragged into more urgent things, SAB and now modules |