You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Q: What is the layout of the ::v-t pseudo in relation to the scope?
Because the scope owns the ::view-transition pseudo-element, it may seem natural for the pseudo to be laid out as an absolute-positioned child of the scope:
However, this becomes problematic when the scope is self-participating (#12319), because
the pseudo is sized to the scope's padding box, which doesn't include borders; and
if the scope is a scroller, the pseudo is placed inside the scrolling contents.
You can see this problem in action here (use Chrome 139+ with experimental web platform features enabled). The scope is a scroller that is scrolled down. During the transition, the "outside" view of the scroller (i.e. its border box) is captured and render-redirected into the ::view-transition... which is at the top of the scrolling contents inside the scroll clip.
An alternative might be to layout the pseudo as a sibling of the scope that is effectively anchor-positioned to the scope:
This issue tracks an open question about the behavior of Scoped View Transitions. Further context appears in the document Self-Participating Scopes.
Q: What is the layout of the ::v-t pseudo in relation to the scope?
Because the scope owns the
::view-transition
pseudo-element, it may seem natural for the pseudo to be laid out as an absolute-positioned child of the scope:However, this becomes problematic when the scope is self-participating (#12319), because
You can see this problem in action here (use Chrome 139+ with experimental web platform features enabled). The scope is a scroller that is scrolled down. During the transition, the "outside" view of the scroller (i.e. its border box) is captured and render-redirected into the
::view-transition
... which is at the top of the scrolling contents inside the scroll clip.An alternative might be to layout the pseudo as a sibling of the scope that is effectively anchor-positioned to the scope:
There is some further discussion on crbug.com/417988089.
cc @noamr @vmpstr @flackr @bramus @jakearchibald
The text was updated successfully, but these errors were encountered: