- From: David Dailey <ddailey@zoominternet.net>
- Date: Mon, 9 Sep 2013 20:32:55 -0400
- To: "'Tab Atkins Jr.'" <jackalmage@gmail.com>
- Cc: "'www-style list'" <www-style@w3.org>
Hi, What I meant by the regular <rect> syntax was <rect x=[real number or percentage of screen] y=[ real number or percentage of screen] width=[length or percentage of screen] height=[length or percentage of screen] ... attribute-value list > -- I am assuming the spec talks about real numbers rather than imaginaries or quaternions or something else weird. And yes, sure, why not just allow people to reference a real SVG object complete with fill and stroke and all the rest instead of inventing a thing called rectangle? In the case of inner flow, having text inherit the background of a drawn svg object complete with gradient and filters makes sense to me. Maybe I'm missing some aspect of what the syntax in this draft is referencing though. Cheers D -----Original Message----- From: Tab Atkins Jr. [mailto:jackalmage@gmail.com] Sent: Monday, September 09, 2013 8:13 PM To: David Dailey Cc: www-style list Subject: Re: [[css-shapes]] feedback On Mon, Sep 9, 2013 at 5:03 PM, David Dailey <ddailey@zoominternet.net> wrote: > a) Just allow the SVG versions of these various basic shapes? That is, > instead of, for example, > rectangle([<length>|<percentage>][, [<length>|<percentage>]] why not > allow the regular SVG syntax for a <rect>? What is "the regular SVG syntax for a <rect>"? The XML? The attributes? Something else? > b) Arbitrary <path> elements to flow shape are harder what with all the > microsyntax of paths, but since all browsers support SVG anyhow (and > in particular, they all seem to know how to render paths) why not > define text flow for arbitrary paths? What I found a bit tricky in my > implementation was following the left and right sides of paths, but > compared to the issue of concavities (which you seem to be handling > with the even-odd fill rule – that can get a bit odd at times I’m > sure) that seems relatively simple in comparison. Yes, given that we have polygan(), I'm not sure why we wouldn't support path() as well. It's just a polygon() with curved paths, basically - all the other difficulties of path() are already present and handled in polygon(). ~TJ
Received on Tuesday, 10 September 2013 00:33:26 UTC