W3C home > Mailing lists > Public > public-css-testsuite@w3.org > January 2011

Re: block-in-inline-relpos-002 invalid

From: L. David Baron <dbaron@dbaron.org>
Date: Sat, 29 Jan 2011 09:25:25 -0800
To: Alan Gresley <alan@css-class.com>
Cc: public-css-testsuite@w3.org
Message-ID: <20110129172525.GA19056@pickering.dbaron.org>
On Saturday 2011-01-29 20:13 +1100, Alan Gresley wrote:
> Some of the examples in your test case clearly shows that the inline
> boxes are flowing around their nested floats.

Inline boxes flow around floats, including floats that are
"anchored" inside the inline box itself.

> So if the inline boxes
> position is due to flowing around the floats, then how can the the
> floats in-turn be positioned in respect to their containing blocks?

Because 4 of the 9 rules in
http://www.w3.org/TR/CSS21/visuren.html#float-position that describe
how to position floats mention containing blocks?

> Another way of saying it is, how can the floats be positioned in
> respect to their containing blocks  when these containing blocks are
> positioned due to flowing around the floats?

An inline box never forms the containing block for a float.  Only a
block box does.  See section 10.1.

However, the anonymous block box formed as described in 9.2.1.1
(when an inline element contains a block box) is a block box, and
can establish a containing block for a float.  (There's an exception
in 9.2.1.1 only for percentage heights.)

-David

-- 
L. David Baron                                 http://dbaron.org/
Mozilla Corporation                       http://www.mozilla.com/
Received on Saturday, 29 January 2011 17:25:56 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 29 January 2011 17:26:05 GMT