W3C home > Mailing lists > Public > www-style@w3.org > December 2012

Re: [css3-exclusions] issue with shape-outside: url() (ISSUE-296)

From: Alan Stearns <stearns@adobe.com>
Date: Mon, 10 Dec 2012 06:32:53 -0800
To: Daniel Glazman <daniel.glazman@disruptive-innovations.com>, Rossen Atanassov <Rossen.Atanassov@microsoft.com>
CC: "www-style@w3.org" <www-style@w3.org>
Message-ID: <CCEB2EC4.184E1%stearns@adobe.com>
On 12/10/12 5:56 AM, "Daniel Glazman"
<daniel.glazman@disruptive-innovations.com> wrote:

>On 10/12/12 02:18, Rossen Atanassov wrote:
>> The idea of being able to re-position a shape was considered before and
>>what we agreed at the time was to introduce a shape-transform. What is
>>this needed and/or different than css transforms - simply because it has
>>to affect layout (as the shape moves there must be a reflow around it
>> We also agreed that this is not something we want to take for the
>>current version of the spec, thus I'm not sure why these changes are
>>needed now.
>Why ? This is simple, Rossen ; see this:
>   #foo {
>     float: right;
>     width: 400px;
>     height: 350px;
>     shape-outside: url(foo.png) /* a 100x150px image */
>   }
>Where is placed the exclusion in #foo's box ? Centered ? Cornered?

I'm wondering when positioning a smaller shape inside a larger box will be
used, and how important it is to have in this level. In the case above,
you have a 100x150px bounding box around the shape that is used for
positioning the float and wrapping content around it. So if the shape is
placed at the top left, whatever is displayed in the right and lower side
of the 400x350px box will overlap with the content wrapping around the
float. Or if the shape is centered, there will be a bit of overlap on all

I have definitely seen instances that look like this (a background
partially intruding into some text) but I expect the usual case will be a
shape that matches a displayed image, or a shape whose bounding box is the
same as the content area.


Received on Monday, 10 December 2012 14:33:21 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:24 UTC