16448 2012-03-20 15:26:37 +0000 Should we revisit the decision to not allow SVG path syntax in the shape-inside, shape-outside properties 2015-08-27 09:35:37 +0000 1 1 1 Unclassified CSS Shapes unspecified PC All RESOLVED FIXED P2 normal --- 1 vhardy stearns dschulze eoconnor ratan stearns public-css-bugzilla oldest_to_newest 65838 0 vhardy 2012-03-20 15:26:37 +0000 Support for the SVG path syntax was recently added to the Canvas API: http://www.whatwg.org/specs/web-apps/current-work/multipage/the-canvas-element.html#path-objects Having a path construct as short-hand would be consistent technically because the properties already allow other shapes. The argument that was made in the past is that while it is quite expected people would hand-code basic shapes in their CSS, and it made sense to provide a syntax for that (instead of also requiring pointing to an SVG element), people do not expect most people to hand code SVG path data. So the idea is that SVG path data would anyway be authored by a tool. The other argument is that most SVG shapes are significantly longer and more verbose than the basic shapes and that justifies a different treatment. The new data point we have compared to when we had the discussion is that the HTML5 canvas, when faced with a similar dilemma, decided to support the path syntax right there, and that is not consistent with the decision we made on a similar issue. The proposal is to add a new primitive such as: shape-inside: path(d); where d would follow the SVG path syntax. 87812 1 stearns 2013-05-17 17:20:09 +0000 This could be added in a later CSS Shapes module level. I've added it to the list of future features here: http://wiki.csswg.org/spec/css-shapes 122789 2 stearns 2015-08-27 09:35:37 +0000 added to level 2 diff spec https://hg.csswg.org/drafts/rev/b766fa9c039f