- From: <bugzilla@jessica.w3.org>
- Date: Thu, 15 Jan 2015 05:21:45 +0000
- To: public-webapps-bugzilla@w3.org
https://www.w3.org/Bugs/Public/show_bug.cgi?id=21066 --- Comment #65 from Hayato Ito <hayato@chromium.org> --- (In reply to Koji Ishii from comment #64) > Got it, thank you for the pointer. That's very clear. > > It looks like we're heading to B), correct? > readonly attribute FrozenArray<Gamepad> gamepads; > > Look forward to it being resolved, but in the meantime, what do you prefer? > > We're: > - event.path is nowhere in the spec. Events in shadow is in the shadow spec, > but this is DOM API that could work without shadow DOM involved (though not > useful), and we're stuck. I'm afraid I don't understand what is a real issue here. UA can provide event.path API without any Shadow DOM implementation, can't we? > - Blink still implements NodeList without window included. > - Gecko is trying to follow the Blink behavior by explicitly removing window > from the list. Gecko doesn't have to follow the current Blink behavior. The current spec already includes Window in event path. > > Options for the spec updates: > A) Keep this undocumented until bug 23682 is resolved. > B) Put a tentative one once in Shadow DOM spec with the issue marked, then > move to DOM when bug 23682 is resolved. Shadow DOM spec mentions this issue of IDL as 'ISSUE 1'. Is it different from plan B? > C) Put a tentative one to DOM spec and update when bug 23682 is resolved. > > Since T[] is much closer to FrozenArray<T> than NodeList is, and it looks > like not having window is troubling Gecko, I think impl could update to T[] > to include window then update again when bug 23682 is done. -- You are receiving this mail because: You are the QA Contact for the bug.
Received on Thursday, 15 January 2015 05:21:49 UTC