- From: Erik Dahlstrom <ed@opera.com>
- Date: Tue, 16 Feb 2010 14:33:10 +0100
- To: "Robert Longson" <longsonr@gmail.com>, www-svg@w3.org
On Tue, 16 Feb 2010 11:39:29 +0100, Robert Longson <longsonr@gmail.com> wrote: > Hi, > > This article illustrates how I think clip works... > http://www.ibloomstudios.com/articles/misunderstood_css_clip/ > > It looks to me as if clip changed between CSS2 ( > http://www.w3.org/TR/2008/REC-CSS2-20080411/visufx.html#clipping) and > CSS2.1 > (http://www.w3.org/TR/CSS2/visufx.html#clipping) > > In particular top and bottom are now offsets from the top edge of the box > and left and right offsets from the left edge of the box, at least in the > testcase referenced below. > > Currently SVG 1.1 references CSS2 clipping ( > http://www.w3.org/TR/SVG/masking.html#OverflowAndClipProperties) whereas > I > think it should reference CSS2.1. I wonder if this is something we can really change in SVG 1.1 second edition, it's a rather large change (e.g webfonts were also dropped from CSS21). We should definately consider it for SVG 2.0 however. > I think the following test should also be changed: > http://dev.w3.org/SVG/profiles/1.1F2/test/harness/htmlObject/masking-path-06-b.htmlbecause > clip="rect(10,10,10,10)" in the example should clip the images > completely. Agreed, if CSS21 is the normative reference. That might well be the reason this test is still in draft status. It would probably also be useful to clarify that when the CSS spec says "box" in the definition of 'clip' that means the element's boundingbox for the svg case, except for when an svg element actually has a CSS box (for the outermost svg element). Cheers /Erik -- Erik Dahlstrom, Core Technology Developer, Opera Software Co-Chair, W3C SVG Working Group Personal blog: http://my.opera.com/macdev_ed
Received on Tuesday, 16 February 2010 13:28:40 UTC