WHATWG
2024-11-29
plaintext logs
prev next
source on github
01:46
<sideshowbarker>
TabAtkins: (or anybody else who might know) is calc() allowed in media queries?
02:18
<Alan Stearns>
sideshowbarker: according to a bunch of mq-calc-* tests, yes: https://wpt.fyi/results/css/mediaqueries?label=experimental&label=master&aligned
02:19
<emilio>
Yes, though relative font stuff compute against the initial font etc
02:19
<sideshowbarker>
sideshowbarker: according to a bunch of mq-calc-* tests, yes: https://wpt.fyi/results/css/mediaqueries?label=experimental&label=master&aligned
Thanks much. Seems like the spec may not be up to date with the tests?
02:22
<sideshowbarker>
https://drafts.csswg.org/mediaqueries/#mq-features and https://drafts.csswg.org/css-values-4/#math-function are the spec parts I mean. I could be that we’re reading those wrong? (we = having internal discussion with Ladybird contributors)
02:26
<Alan Stearns>
I assume that when a spec says the value type is <length> then calc() expressions that evaluate to a length are allowed, but I am not currently finding the spec text that supports my assumption
02:29
<Alan Stearns>
Probably this: https://drafts.csswg.org/css-values-4/#calc-type-checking
02:30
<sideshowbarker>

That looks like it, yeah

A math function can be many possible types, such as <length>, <number>, etc., depending on the calculations it contains, as defined below. It can be used anywhere a value of that type is allowed.

02:34
<sideshowbarker>
Alan Stearns: Thanks much (and emilio thanks for the confirmation — and the caveat)
09:37
<annevk>
Domenic: https://github.com/whatwg/dom/issues/1255#issuecomment-2507354378 is very good. Already thinking about referencing it in the future when there's a similar situation. Thanks! (Still need to think a bit about whether we need a fallback method in v1; curious to see what other people say.)
10:45
<Noam Rosenthal>
I got some offline good responses to it from people who commented on the thread before, suggesting placeBefore as an alternate shorter name for the version with the fallback
11:29
<Jake Archibald>

What are the conditions for move? Is it:

const canMove =
  element.isConnected === target.isConnected &&
  (!target.isConnected || element.ownerDocument === target.ownerDocument);
11:32
<Jake Archibald>
Or is cross-document never a move?
11:54
<Noam Rosenthal>
Or is cross-document never a move?
This
11:54
<Jake Archibald>
Why? If it's disconnected to disconnected, there's no state loss, right?
11:55
<Jake Archibald>
Or is it more that it's clearly not a move if it goes through the adoption process?
11:55
<Noam Rosenthal>
There are all kinds of things you lose when adopting, eg iirc custom elements downgrade
11:56
<Jake Archibald>
ah yeah, that makes sense
11:56
<Noam Rosenthal>
Things that don’t happen on disconnect