W3C home > Mailing lists > Public > www-style@w3.org > October 2010

Re: [css3-2d-transforms][css3-3d-transforms] Clarify equivalency with position: relative

From: David Hyatt <hyatt@apple.com>
Date: Fri, 29 Oct 2010 19:28:23 -0500
Cc: www-style list <www-style@w3.org>
Message-id: <C7CF0E02-F30D-4486-9EA2-46838BBCE936@apple.com>
To: Sylvain Galineau <sylvaing@microsoft.com>
On Oct 29, 2010, at 6:21 PM, Sylvain Galineau wrote:

> Related to this is whether transforms should affect CSSOM properties
> such as offsetLeft.
> 

I do think an element that specifies a transform should be an offsetParent.  We don't currently do this in WebKit, but I think we should.  It's conceptually similar to IE's CSS "zoom" property (which WebKit also supports), and that property causes you to become an offsetParent when specified I believe (I did that in WebKit at least, and I assume I did it to match IE).

I don't think offsetLeft etc. should be affected by the transform simply because it can be something other than a simple translation or scale.

dave
(hyatt@apple.com)

> Currently, offsetLeft is unchanged however much you translate or scale
> the element. But it is reflects position:relative.
> 
> JS code that relies on these properties for overlays and other effects 
> will fail if any transform is applied to the element.
> 
>> -----Original Message-----
>> From: Sylvain Galineau
>> Sent: Friday, October 22, 2010 7:56 PM
>> To: www-style list
>> Subject: [css3-2d-transforms][css3-3d-transforms] Clarify equivalency
>> with position: relative
>> 
>> In CSS3 2D Transforms[1], we have:
>> 
>> # The transform property does not affect the flow of the content
>> # surrounding the transformed element. However, the value of the
>> # overflow area takes into account transformed elements. This
>> # behavior is similar to what happens when elements are translated
>> # via relative positioning.
>> 
>> In CSS3 3D Transforms [2] :
>> 
>> # The object acts as though position: relative has been specified, but
>> also acts as a containing block for fixed positioned descendants.
>> 
>> This language is rather stronger e.g. it could be read as saying top
>> and left position the element like they would a position:relative
>> element. However, in the perspective property definition [3], we have:
>> 
>> # ...It also establishes a containing block (somewhat similar to
>> position:relative),
>> 
>> Which would imply it's not supposed to act as though it was
>> position:relative.
>> 
>> Given current implementations, I assume 2D Transforms and 3D's
>> perspective property define the intended behavior ?
>> 
>> [1] http://www.w3.org/TR/css3-2d-transforms/
>> [2] http://www.w3.org/TR/css3-3d-transforms/
>> [3] http://www.w3.org/TR/css3-3d-transforms/#perspective-property
>> 
> 
> 
Received on Saturday, 30 October 2010 00:28:57 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:33 GMT