- From: Alan Stearns <stearns@adobe.com>
- Date: Tue, 31 Jul 2012 12:22:10 -0700
- To: "www-style@w3.org" <www-style@w3.org>
- CC: Adam Barth <abarth@webkit.org>, Sam Weinig <sam@webkit.org>
Hey all, At the beginning of this year we changed part of the Regions CSSOM to move some properties out of a supplemental Element interface into a separate Region interface [1]. This was the result of a discussion about what could be a CSS Region, and was done to support the idea of CSS Regions being made out of boxes defined in CSS (via existing pseudo-elements or some future mechanism). I have updated this change to define Region as a [NoInterfaceObject] supplemental interface that Element implements [2]. At the moment, Element is the only object that implements this interface. The intent is that if in the future other region-capable boxes become exposed to the CSSOM they too will implement the supplemental Region interface, and the Region interface members will not be limited to an Element or Node. We have gotten some feedback [3] that this may not be the best way to implement this interface. There are other supplemental interfaces implemented by multiple objects, but apparently it's a new thing to have a method that returns one of these interfaces, the way that NamedFlow.getRegions() does. That method returns a sequence<Region> whose members can be Elements or any other object that implements Region. Are there other instances of supplemental interfaces being used as a return type? Are there other implementations that would balk at implementing this? Does anyone have a suggestion as to how we could improve this part of the spec, while still leaving the option of what can be a CSS Region open? Thanks, Alan [1] http://lists.w3.org/Archives/Public/www-style/2012Feb/1354.html [2] http://dev.w3.org/csswg/css3-regions/#the-region-interface [3] http://lists.webkit.org/pipermail/webkit-dev/2012-July/021620.html
Received on Tuesday, 31 July 2012 19:22:37 UTC