| 02:39 | <MikeSmith> | I can’t find that the PromiseRejectionEvent() constructor requirements are specified beyond just what follows from the IDL for the interface at https://html.spec.whatwg.org/multipage/webappapis.html#the-promiserejectionevent-interface |
| 02:40 | <TimothyGu> | MikeSmith: https://dom.spec.whatwg.org/#concept-event-constructor |
| 02:40 | MikeSmith | loks |
| 02:40 | <MikeSmith> | TimothyGu: thanks |
| 02:40 | <TimothyGu> | np |
| 02:41 | <MikeSmith> | https://developer.mozilla.org/en-US/docs/Web/API/PromiseRejectionEvent/PromiseRejectionEvent#Parameters says the type argument to the constructor “must be one of "rejectionhandled" or "unhandledrejection". |
| 02:42 | <MikeSmith> | ... but I can’t see where that’s actually specified or what it follows from |
| 02:43 | <TimothyGu> | MikeSmith: I think they mean that the event fired by browsers can only be either of them |
| 02:59 | <MikeSmith> | TimothyGu: OK, thanks ー so to be clear, in fact there’s nothing that prevents me from giving that constructor any string I want as the first argument, right? |
| 02:59 | <MikeSmith> | there’s nothing specified that says it will throw if the string is something other than "rejectionhandled" or "unhandledrejection" |
| 02:59 | <TimothyGu> | MikeSmith: correct |
| 02:59 | <MikeSmith> | ok, thanks again |
| 05:35 | <TimothyGu> | amazing |
| 05:35 | <TimothyGu> | apparently Chrome only fires beforeunload for HTML documents |
| 05:35 | <TimothyGu> | not XHTML documents |
| 07:13 | <annevk> | Poor XHTML |
| 07:13 | <annevk> | I guess in this case it's actually a win for XHTML |
| 07:13 | <annevk> | But most other things are stacked against it still |
| 08:49 | <yoav> | annevk: FYI https://github.com/w3c/resource-timing/issues/70#issuecomment-414960153 |
| 10:01 | <annevk> | Sigh |
| 14:40 | <DeltaHeavy> | Does HTML5 technically have namespaces as a language? Or does it simply define one for use with XML? |
| 14:45 | <zcorpan> | DeltaHeavy: it doesn't provide a means to specify namespace in text/html, but the HTML parser can create elements in the HTML, SVG and MathML namespaces |
| 14:46 | <DeltaHeavy> | Thank you! |
| 17:19 | <TimothyGu> | … never mind, Chrome does fire beforeunload on XHTML documents |
| 17:19 | <TimothyGu> | however, WPT's /common/dummy.xhtml turns out to not be an actual XHTML document |
| 17:20 | <TimothyGu> | as it doesn't have xmlns="http://www.w3.org/1999/xhtml"; on the html element |
| 17:22 | <TimothyGu> | and Chrome tests for the existence of a body element before trying to fire beforeunload |
| 17:22 | <TimothyGu> | o_O |
| 17:50 | <Domenic> | AMAZING |
| 18:10 | <jugglinmike> | annevk: do you know of any discussion about the trend to define new JavaScript APIs using configurable properties? |
| 19:42 | <annevk> | jugglinmike: not aware of a trend |
| 19:42 | <annevk> | jugglinmike: IDL/JS defaults mismatch? |
| 19:46 | <jugglinmike> | annevk: oh, hm |
| 19:46 | <jugglinmike> | I thought it was intentional |
| 19:47 | <jugglinmike> | that as a web developer, I can generally write, e.g. `navigator.permissions = myMockPermissionsAPI` |
| 19:48 | <Domenic> | That's been true since like 1995 |
| 19:49 | <Domenic> | Well, I guess you'd have to use Object.defineProperty, since it's a getter |