Re: [w3ctag/design-reviews] Declarative Interactions (Issue #1111)

szager-chromium left a comment (w3ctag/design-reviews#1111)

@xiaochengh -- thanks for the review! Some answers...

1. This is a fairly concrete proposal, despite the fact that the CSS syntax is still evolving. I've already done some implementation work in chromium (not landed). So I don't think early design review is quite right.

2. There is no precedent for using CSS selectors within CSS properties, and I was warned off that approach by seasoned CSSWG members whom I respect. Using dashed-identifiers is the idiomatic way to do this.

3. The list of supported events is not yet fully specified, but you're right that they will be a subset of UI Events. One of the driving considerations is where a hit test is required to dispatch the event. If a hit test is required, then we know that ComputedStyle is up-to-date and there is no danger that an `animation-trigger` property has been applied but not yet flushed to ComputedStyle. For event types that do *not* perform a hit test, we either have to do a forced style update prior to dispatching the event (which would be pretty sub-optimal), or just not support that event type.

As for the dependency between CSS Animations and UI Events -- I think the parts of the UI Events spec referenced by this feature are pretty stable so I wouldn't anticipate much maintenance burden, but it probably makes sense to add language and cross-references to both of those specs to hopefully avoid anything being missed in future edits.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3ctag/design-reviews/issues/1111#issuecomment-3137892544
You are receiving this because you are subscribed to this thread.

Message ID: <w3ctag/design-reviews/issues/1111/3137892544@github.com>