Re: [csswg-drafts] [css-view-transitions-1] Flush the callback queue before performing other view-transition operations (#11947)

On Fri, Mar 21, 2025, 12:41 p.m. Noam Rosenthal ***@***.***>
wrote:

> This PR makes it a bit more obvious that the initial PR (#11693
> <https://github.com/w3c/csswg-drafts/pull/11693>) may have had a problem
> in that we're running script in the middle of update the rendering steps.
> Namely after layout but before intersection observers and paint. That seems
> problematic.
>
> Good point
>
> Maybe we should instead frame it as if we need to set up capture and the
> queue is not empty, abort these steps. And then we can flush the queue
> outside of update the rendering (and schedule another animation frame)
>
> Maybe, instead, we should keep rendering surpression until the queue has
> been emptied?
>
> —
> Reply to this email directly, view it on GitHub
> <https://github.com/w3c/csswg-drafts/pull/11947#issuecomment-2743905758>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AAORSCYIJLGAH2DT3WW2GN32VQ6KTAVCNFSM6AAAAABZEAZJROVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONBTHEYDKNZVHA>
> .
> You are receiving this because you commented.Message ID:
> ***@***.***>
> [image: noamr]*noamr* left a comment (w3c/csswg-drafts#11947)
> <https://github.com/w3c/csswg-drafts/pull/11947#issuecomment-2743905758>
>
> This PR makes it a bit more obvious that the initial PR (#11693
> <https://github.com/w3c/csswg-drafts/pull/11693>) may have had a problem
> in that we're running script in the middle of update the rendering steps.
> Namely after layout but before intersection observers and paint. That seems
> problematic.
>
> Good point
>
> Maybe we should instead frame it as if we need to set up capture and the
> queue is not empty, abort these steps. And then we can flush the queue
> outside of update the rendering (and schedule another animation frame)
>
> Maybe, instead, we should keep rendering surpression until the queue has
> been emptied?
>

That's a good question. If we're going to capture the old state after the
queue is flushed, we'd presumably need at least one frame produced for
those pixels. Not sure exactly what the author or user expectations would
be here

> —
> Reply to this email directly, view it on GitHub
> <https://github.com/w3c/csswg-drafts/pull/11947#issuecomment-2743905758>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AAORSCYIJLGAH2DT3WW2GN32VQ6KTAVCNFSM6AAAAABZEAZJROVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDONBTHEYDKNZVHA>
> .
> You are receiving this because you commented.Message ID:
> ***@***.***>
>


-- 
GitHub Notification of comment by vmpstr
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/pull/11947#issuecomment-2745382277 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Saturday, 22 March 2025 17:42:55 UTC