11:36
<annevk>
MikeSmith: are you interested in making a wattsi fix for https://github.com/whatwg/html/pull/5855?
11:50
<zcorpan_>
annevk: about the COEP parsing tests, firefox times out for ["","require-corp"] (2 headers). SFV says "Parsers MAY fail when processing a field value spread across multiple field lines", but Fetch says to join headers with 0x2C 0x20 before passing to SFV parsing
11:50
<zcorpan_>
annevk: i think it times out because the error page doesn't fire a load event on the iframe
11:52
<zcorpan_>
annevk: so... do we say this is a bug (or 2 bugs) in firefox?
12:02
<annevk>
zcorpan_: frames not firing a load event due to network errors is a known Firefox bug
12:03
<annevk>
zcorpan_: so in absence of other data I'd say that's one (known) bug
12:09
<zcorpan_>
annevk: ok, thanks. then we have lack of interop on how to handle multiple COEP headers, where chromium I think does what the fetch spec says, and firefox gives a network error
12:10
<zcorpan_>
I can file in bugzilla
12:19
<MikeSmith>
annevk: yes I can work on it tomorrow my time
12:21
<annevk>
zcorpan_: oh yeah, COEP shouldn't be a network error, it just results in no policy
12:21
<annevk>
zcorpan_: thanks
12:25
<zcorpan_>
annevk: hmm, right. It's not the page that sets the COEP headers that result in a network error. It's the iframe embedded within that lacks CORP that is blocked and doesn't get a 'load' event
12:27
<annevk>
zcorpan_: does that mean COEP is incorrectly enforced?
12:27
<zcorpan_>
annevk: yes
12:28
<zcorpan_>
that is more clearly a bug
12:28
<annevk>
Hopefully the switch in parser will address that, but that seems worth of a bug, yes
12:28
<annevk>
worthy*
12:32
<zcorpan_>
ok https://bugzilla.mozilla.org/show_bug.cgi?id=1661252
15:34
<annevk>
Domenic: did you see the whenDefined PR? It's pretty small and ready to go
15:35
<Domenic>
annevk: I remember it going by, but I was hoping others could handle it since I was weakly -1 on the idea
15:36
<Domenic>
If that feels like a dereliction of duty I can make it work
15:36
<annevk>
Domenic: the main problem is that I made some commits, but I could still merge it
15:36
<Domenic>
Fair enough, I'll review
15:55
<Domenic>
annevk: Ah I remember what I was waiting for. WebReflection definitely works in the field of web technologies; they need their employer to sign up
15:57
<annevk>
Domenic: I checked with the SG and hey said no (and not just myself)
15:59
<annevk>
I might have missed something though
15:59
<Domenic>
Hmm
16:00
<Domenic>
Well they do make the decisions, but he spends so much time trying to influence web standards...