Re: [css3-regions] New CSS Regions editor draft

On Thu, Jun 2, 2011 at 6:50 PM, Alex Mogilevsky <alexmog@microsoft.com>wrote:

>  Let’s together figure out the best way to approach this. Can we start a
> list somewhere of what a good pagination/fragmentation spec needs to have?
> Once we have it we can decide which spec it belongs to.
>

The CSS specs often implicitly assume that certain kinds of elements only
have one box, and that the "containing block" for an element is a unique
ancestor element (or box) that is the same for all the element's boxes, and
that the width of the containing block is the same for all an element's
boxes. We're probably going to have to break some of those assumptions, and
other assumptions that I haven't thought of. Wherever those assumptions are
broken, we'll have to figure out how to update the spec so it's not
ambiguous.

So for starters, if we're going to allow a block's boxes' widths to adapt to
varying container widths, a pagination spec would have to update CSS 2.1
section 10.1 (or whatever it is in the equivalent CSS3 module) to define
"containing block" and/or "containing block width" in a way that permits
that. I expect "containing block" will have to change from being a function
from elements to rectangles to being a function from boxes to rectangles,
which means pretty much every mention of "containing block" will need to be
checked and updated wherever it is applied to an element that can have more
than one box. Hopefully we can find a systematic way to do that.

That's probably not all that would be required in a good pagination spec,
but it'd be a great start.

Rob
-- 
"Now the Bereans were of more noble character than the Thessalonians, for
they received the message with great eagerness and examined the Scriptures
every day to see if what Paul said was true." [Acts 17:11]