- From: Simon Pieters <simonp@opera.com>
- Date: Wed, 17 Apr 2013 14:27:32 +0200
- To: Håkon Wium Lie <howcome@opera.com>
- Cc: "www-style@w3.org" <www-style@w3.org>
On Fri, 12 Apr 2013 14:12:57 +0200, Simon Pieters <simonp@opera.com> wrote: > On Fri, 12 Apr 2013 12:55:41 +0200, Håkon Wium Lie <howcome@opera.com> > wrote: > >> Simon Pieters wrote: >> >> > Consider a multicol element with a column rule that extends outside >> the >> > multicol element. If you click on the rule outside the element, what >> > should the click event's target be? In Opera/Chrome/Firefox, the >> click >> > goes through the rule and the target is whatever is behind the rule. >> Maybe >> > the spec should call this out. >> > >> > http://software.hixie.ch/utilities/js/live-dom-viewer/saved/2206 >> >> I can see users being confused when parts of the rule respond >> differently to clicks. OTOH, it seems natural to clip at the edge of >> the element, and implementations seem to do this. I agree that the >> spec should describe the behavior. Do you have a proposed wording? > > I don't. Is there any precedent for something like this? Can it be > defined in terms of the Pointer Events spec? Since hit testing isn't defined yet, I suggest hand-waving the issue like so: "The column rule must be considered to be absent for the purpose of hit testing (e.g. for the elementFromPoint() API) and determining event target for pointer events. [CSSOMVIEW] [POINTEREVENTS]" -- Simon Pieters Opera Software
Received on Wednesday, 17 April 2013 12:28:01 UTC