W3C home > Mailing lists > Public > www-style@w3.org > June 2011

CSS Regions / Exclusions

From: Daniel Glazman <daniel.glazman@disruptive-innovations.com>
Date: Fri, 17 Jun 2011 15:54:55 +0200
Message-ID: <4DFB5CAF.60608@disruptive-innovations.com>
To: "www-style@w3.org" <www-style@w3.org>
My idea is to consider an exclusion (to the outside of a path) or
a restriction (to the inside of a path) are precisely similar to a
background + an action.

The background and layout shape of an image are then defined by a stack

   1. a region, determined by a set of masks and a restriction mode
   2. background color (clipped by the region I presume)
   3. stack of background images (clipped by the region I presume)

A region be specified by the following properties, exactly like
a background:

1. region-image: <msk-image>[,<msk-image>]* | inherit | initial

     where <msk-image> = <uri> | none
     with initial value of 'none'

2. region-restriction: inside | outside | inherit | initial

3. region-repeat

    exactly similar to background-repeat for regions but with
    an initial value of 'no-repeat'

4. region-position, region-clip, region-origin exactly
    similar to their background-* counterparts but for regions

5. region-attachment does NOT exist while background-attachment does

6. region-size, counterpart to background-size, should probably not
    exist since it could lead to extremely expensive and deep recursion

7. a 'region' shorthand

I think this scenario is

a. much simpler conceptually for web authors
b. much simpler to implement and manipulate for content authoring
    software, and Adobe does content authoring software...
c. much simpler in terms of layout algo since exclusion or restriction
    are really the same algo applied to a region and the negated
d. avoid all the issues and implementation complexity related to floats
e. allow elements to carry directly their exclusion/restriction areas
    and then be copy-paste-able, while a solution based on float is NOT

Received on Friday, 17 June 2011 13:55:45 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:46 UTC