| 04:31 | <sideshowbarker> | * jarhar: The “A CSS Animation running on an element within a <details> element is canceled after the <details> element is closed” test at https://wpt.fyi/results/css/css-animations/animation-canceled-by-parent-details-element-being-closed.html times out in Chrome, while passing in Firefox and current Safari. But the patch in https://github.com/WebKit/WebKit/pull/18785 for implementing auto-expanding The reason it times out (both in Blink and patched WebKit) is that the I’m going to make time to investigate it more, but it’s not immediately clear to me why the changes made for auto-expanding |
| 05:04 | <sideshowbarker> | * jarhar: The “A CSS Animation running on an element within a <details> element is canceled after the <details> element is closed” test at https://wpt.fyi/results/css/css-animations/animation-canceled-by-parent-details-element-being-closed.html times out in Chrome, while passing in Firefox and current Safari. But the patch in https://github.com/WebKit/WebKit/pull/18785 for implementing auto-expanding The reason it times out (both in Blink and patched WebKit) is that the I’m going to make time to investigate it more, but it’s not immediately clear to me why the changes made for auto-expanding |
| 09:52 | <axo10tl> | Hi! Can I request some historical information here? 😅 I was wondering what kind of arguments were raised for the current CSS custom property syntax, more specifically; was the SASS-like dollar syntax considered, and why didn't custom properties adopt something similar to that? |
| 09:54 | <axo10tl> | Tried searching the whatwg github issues, but couldn't find any discussion related to this. Apparently some discussions are held in IRC/Matrix, so there would not be public logs if that's the case? |
| 10:05 | <Noam Rosenthal> | Hi! Can I request some historical information here? 😅 |
| 10:06 | <axo10tl> | oof, thanks (and sorry).. :D |
| 10:21 | <annevk> | Panos Astithas: I don't have the Oct 19 WHATNOT meeting in my calendar. I was going to add RitamDas30. Could it be that I'm not on the invite list or did I mess up somehow? |
| 13:09 | <sideshowbarker> | jarhar: The “A CSS Animation running on an element within a <details> element is canceled after the <details> element is closed” test at https://wpt.fyi/results/css/css-animations/animation-canceled-by-parent-details-element-being-closed.html times out in Chrome, and in Safari built from https://github.com/WebKit/WebKit/pull/18785 — apparently because, with the switch to using Per-spec is the If it’s not supposed to fire, then I guess that test needs to be changed. But if it is actually supposed to fire, then I guess that’d mean both Blink and WebKit have a bug in their content-visibility implementations that needs to be fixed? |
| 14:11 | <freddy> | shower thought: Wouldn't it be nice to have a "sameoriginmessage" (name to be bikeshedded) counterpart of the "message" Event type which code could consider implicitly trusted (or not)? |
| 14:20 | <annevk> | freddy: possibly, or some kind of event listener filter option |
| 14:27 | <freddy> | The options bag for addEventListener doesn't seem like a great pattern to support specific options only for certain event types. But sure, I don't have strong feelings. This started off as a shower thought ;) |
| 15:33 | <zcorpan> | zcorpan: did you want to give feedback on https://github.com/whatwg/html/pull/9689? I think Jonathan Kew agrees with it so probably have enough in terms of implementer interest, but always nice to get more review |
| 15:54 | <annevk> | zcorpan: hah, so I just found a flaw, only submit buttons participate in form submission |
| 15:54 | <annevk> | So maybe dirname needs to be more limited than dir=auto is after all |
| 15:55 | <zcorpan> | annevk: good catch. but it's "only" dead code, then, for dirname |
| 15:55 | <annevk> | Ah I guess the form submission algorithm is still fine as the buttons are filtered early on, indeed |
| 15:56 | <annevk> | So it's just a matter of whether dirname applies and is conforming, which it should not be |
| 15:56 | <annevk> | I'll change that |
| 15:57 | <zcorpan> | right |
| 15:58 | <annevk> | I was trying to implement this and the BaseButtonInputType class in WebKit was just like no, we don't handle form data here |
| 16:39 | <jarhar> | i think that animations are supposed to "pause", so it probably makes sense that cancel doesnt fire |
| 16:39 | <jarhar> | https://drafts.csswg.org/css-contain-2/#:~:text=While%20an%20element%20is%20skipped%2C%20CSS%20transitions%20and%20animations%20on%20the%20element%20do%20not%20update |
| 16:39 | <jarhar> | i didnt implement any of the interactions between content-visibility and animations though, so i'm not the most knowledgable |
| 16:50 | <Panos Astithas> | I see your personal email is on the invite. Perhaps I shall add your work email, too? |
| 16:53 | <Panos Astithas> | I just removed and re-added you, hopefully that generated a useful notification. |
| 17:23 | <annevk> | It didn't. Not sure what's going on. It's not in Trash. I can find it using the Google Calendar web interface search, but then it can't open it. It's also not a declined invitation. |
| 17:36 | <Panos Astithas> | That's weird. Shall I add your work email to see if it helps you debug this further? |
| 18:05 | <annevk> | Could you add annevk@annevk.nl perhaps? |
| 18:06 | <annevk> | I don't think my Apple address will help, but I could be mistaken... |
| 18:27 | <Panos Astithas> | That address was already there, but I removed it, saved the event, and then added it again. Maybe you'll get a notification this time |
| 19:53 | <TabAtkins> | Hiya, editor of CSS Variables here. Yes, $foo was an early proposed syntax. It started as var-foo: blue; color: var(foo);, then we discussed using $ and using the same name consistently in https://lists.w3.org/Archives/Public/www-style/2012May/0744.html, and then we explored a bunch of options that wouldn't require parsing changes (for a bit it was an underscore, then eventually the --, see https://lists.w3.org/Archives/Public/www-style/2014Mar/0261.html) |
| 20:00 | <bkardell> | The - was basically keeping it consistent with what we'd already done with vendor prefixes, and we just did that for all the custom things https://lists.w3.org/Archives/Public/www-style/2014Mar/0269.html |