W3C home > Mailing lists > Public > www-style@w3.org > July 2006

Re: [[css3-layout]] Cascading properties

From: Bert Bos <bert@w3.org>
Date: Tue, 25 Jul 2006 12:11:58 +0200
To: Karl Dubost <karl@w3.org>
Cc: www-style@w3.org
Message-Id: <200607251211.59038.bert@w3.org>

On Tuesday 25 July 2006 02:41, Karl Dubost wrote:
> Le 25 juil. 06 à 03:42, Bert Bos a écrit :
> > On Friday 21 July 2006 07:05, karl@w3.org wrote:
> >> Hi,
> >> This is a QA Review comment for "CSS3 Advanced Layout Module"
> >> http://www.w3.org/TR/2005/WD-css3-layout-20051215/
> >> 2005-12-15
> >> 1st WD
> >>
> >> About http://www.w3.org/TR/2005/WD-css3-layout-20051215/
> >>
> >> What's happening when something like that is written in the CSS?
> >>
> >> 	body > * { display: "ab"
> >>                         "cd" }
> >
> > Not sure what the question is. All children of the body will be
> > template
> > elements. Presumably some of their children in turn will have
> > 'position' properties, but if not, all that happens is that the
> > right half of the window is unused (the 'a' slots in the template
> > get content
> > by virtue of being the default slot, the other slots remain empty.)
>
> Interesting. I was wondering if it would generate a cascade difficult
> to handle by browsers. A kind of fractal cascade, but it seems not
> by your answer. Testing will be interesting when you will start to
> design test cases.

Your example doesn't nest the templates. But even if you do

    * {display: "ab"
                "cd"}

the nesting is still finite, because the source document only has a 
limited nesting depth.

I remember a test for table implementations some years ago that tested 
nested tables. There is nothing wrong with nested tables in theory, but 
some browsers implemented them so poorly that they ran out of memory 
and crashed at the fourth table or so, while others easily handled a 
hundred.

I guess it will be similar here: you can make a browser work hard by 
nesting templates and asking it to shrink-wrap each of them (by setting 
the width of some column to 'intrinsic'). Computing the width of a 
hundred nested templates may take a noticeable time, but there is no 
reason it should crash the browser. Especially as there is nothing new 
here: the effect is the same when you nest tables or floats.

>
> Is there a repository already for CSS 3 Test Cases?

No, but the mailing list public-css-testsuite@w3.org is open for ideas 
and submissions, as always.

http://lists.w3.org/Archives/Public/public-css-testsuite/



Bert
-- 
  Bert Bos                                ( W 3 C ) http://www.w3.org/
  http://www.w3.org/people/bos                               W3C/ERCIM
  bert@w3.org                             2004 Rt des Lucioles / BP 93
  +33 (0)4 92 38 76 92            06902 Sophia Antipolis Cedex, France
Received on Tuesday, 25 July 2006 10:12:18 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:46 GMT