Re: [whatwg/fullscreen] Remove top layer definitions, now that CSS Position 4 contains them. (PR #223)

> For testing, a think a single new test would cover this PR. It should have two sibling divs, A and B. Request fullscreen once per click in order A-B-A and assert that all requests succeed but the last is a no-op and that `document.fullscreenElement` is still B. Bonus points for exiting fullscreen once and asserting that `document.fullscreenElement` is then A again, and then once more to make `document.fullscreenElement` null.

So I wrote [a test](https://github.com/web-platform-tests/wpt/pull/39828) for this, but I'm getting a bit more concerned about the behavior. As-is, Chromium fails this test, because the current behavior is that A is ([manually](https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/renderer/core/fullscreen/fullscreen.cc;l=203;drc=4a24affd324ec742bfb061cda42e459308e36161)) popped to the top of the top layer stack when `A.requestFullscreen()` is called the second time. Gecko and WebKit both fail before that point (at `B.requestFullscreen()`) because of the descendant check. I'm still supportive of changing this, but I'm worried about compat. Perhaps, if there's agreement with this direction (@annevk @emilio - thoughts?) then I can add a use counter for this situation and see where we stand?

The other way to go is to keep Chromium's current behavior of popping A back to the top on the second call to `requestFullscreen()`. This is definitely web compatible, and I could see an argument for this being useful for developers.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/fullscreen/pull/223#issuecomment-1533527350
You are receiving this because you are subscribed to this thread.

Message ID: <whatwg/fullscreen/pull/223/c1533527350@github.com>