WHATWG
2023-07-07
plaintext logs
prev next
source on github
09:39
<zcorpan (OOO until Aug 14)>
Have a nice summer folks
09:43
<Magnus Jurdal>
The same to you all Have a wonderful summer vacation
14:07
<jgraham>
https://url.spec.whatwg.org/#serialize-an-integer is entertainingly succinct. No idea what data type it returns for example :)
14:17
<jgraham>
I found https://github.com/whatwg/infra/issues/201, so now I know fetch has a very slight improvement on this, which has apparently been copied into various other specs.
16:00
<annevk>
A decimal number is a data type :p
16:02
<annevk>
Anyway, yeah, we should add numbers and ways to serialize them to strings to Infra. Maybe now that BigInt is defined we can actually do so.
20:55
<annevk>
jarhar: mfreed: hey, did you see https://github.com/whatwg/html/pull/9456#issuecomment-1611419095?
20:56
<annevk>
jarhar: mfreed: would be great to have your input; cc ntim
20:57
<sideshowbarker>

I’m looking at https://github.com/w3c/webappsec-credential-management/actions/runs/5465713950/jobs/9949551741

      LINE ~299: No 'dfn' refs found for 'parent browsing context'.
      [=parent browsing context=]
20:58
<sideshowbarker>
I guess parent browsing context got changed to something else at some point before March?
21:02
<sideshowbarker>
aha, parent navigable, it seems. Right?
21:02
<annevk>
Yeah, though I'm not sure it's really a term of art on its own; you'd get the current navigable somehow and then grab its parent member.
21:03
<sideshowbarker>
I see
21:03
<sideshowbarker>
Lemme grab the relevant part of the credential-management spec
21:05
<sideshowbarker>

https://w3c.github.io/webappsec-credential-management/#algorithm-same-origin-with-ancestors

  1. While current has a parent browsing context:

    1. Set current to current’s parent browsing context.
21:06
<sideshowbarker>
So I guess I can’t just s/parent browsing context/parent navigable/ there?
21:07
<sideshowbarker>
Maybe I need to add some earlier step to get the current navigable first?
21:10
<annevk>
current prolly needs to become a navigable and then you check current's parent
21:11
<annevk>
Domenic and Dominic Farolino have a better hold on this and might be able to suggest even better things
21:11
<sideshowbarker>
is there a “get navigable for…” hook?
21:12
<sideshowbarker>
Domenic and Dominic Farolino have a better hold on this and might be able to suggest even better things
Thanks yeah then I’ll wait to hear from them. It’s not urgent. The related issue has been open since March 2nd
21:12
<annevk>
You can grab it from the Window object at least
21:12
<Domenic>
It looks like you want to go from document -> navigable. For that, use "node navigable".
22:10
<sideshowbarker>

Unrelated incidental bikeshed warning:

LINK ERROR: Multiple possible 'fetch()' idl refs.
Arbitrarily chose https://wicg.github.io/background-fetch/#dom-backgroundfetchmanager-fetch
To auto-select one of the following refs, insert one of these lines into a <pre class=link-defaults> block:
spec:background-fetch; type:method; text:fetch(id, requests)
spec:fetch; type:method; text:fetch(input)
{{fetch()}}
22:11
<sideshowbarker>

…but I did already put:

@@ -76,6 +76,7 @@ spec: web-otp; urlPrefix: https://wicg.github.io/web-otp
 <pre class="link-defaults">
 spec:html; type:dfn; for:html-origin-def; text:origin
 spec:html; type:dfn; for:environment settings object; text:global object
+spec:fetch; type:method; text:fetch(input)
 spec:fetch; type:dfn; for:/; text:request
 spec:fetch; type:dictionary; for:/; text:RequestInit
 spec:infra; type:dfn; for:/; text:set
22:12
<sideshowbarker>
…but Bikeshed seems to be ignoring that
22:18
<annevk>
sideshowbarker: I think that's because it lacks input as argument so you either want to remove input in link-defaults or add it where you reference
22:19
<annevk>
sideshowbarker: in fact if you did <a lt=fetch(input)><code>fetch()</code></a> you'd probably not need link-defaults
22:23
<sideshowbarker>
sideshowbarker: in fact if you did <a lt=fetch(input)><code>fetch()</code></a> you'd probably not need link-defaults
Hai. Will do that instead
23:03
<canadahonk>
hi all, edgecase with HTMLStyleElement for you: if the disabled flag is set and the children are updated, should the stylesheet be updated/applied?
23:04
<canadahonk>

example:

const style = document.createElement("style");
document.body.appendChild(style);

style.disabled = true;
style.disabled // true

style.textContent = "new content";
style.disabled // false
23:13
<annevk>
Yeah, see https://html.spec.whatwg.org/#update-a-style-block and caller