W3C home > Mailing lists > Public > www-style@w3.org > February 2004

Question on CSS2.1 absolute positioning

From: George Chavchanidze <gch@rmi.acnet.ge>
Date: Fri, 27 Feb 2004 12:52:51 -0400 (GMT)
To: www-style@w3.org
Message-ID: <Pine.GSO.3.96.1040227125013.8909A-100000@rmi-sun.rmi.acnet.ge>

CSS2.1 visual formatting model says that width of absolutely positioned
elements should be determined from the following constraint:

'left' + 'margin-left' + 'border-left-width' + 'padding-left' + 'width' +
'padding-right' + 'border-right-width' + 'margin-right' + 'right' = width
of containing block

In the same time CSS2.1 clearly says that 'left' and 'right' offsets are
calculated from
padding edge of containing block (and not content edge), while under width
of
containing block one usually means content width. Now imagine that all
offsets, margins, padding and borders
of absolutely positioned element vanish (just for simplicity).
Then since offsets are calculated from
padding edge of the containing block it is clear that width of the
positioned element should be equal to
width plus sum of left and right padding of its containing block. From the
other hand
constraint mentioned above gives rise to apparently inconsistent result:
'width' = width of containing block
Is it error in specs or under width of containing block you mean
width of padding box (not content box)? If so I think it is better to
indicate this in specs because in
other parts of spec this term refers to width of content box.
Here is test page that demonstates the problem:
http://geocities.com/csssite/operabugs/bug3.xml
view it in Opera and you will see how browser miscalculates right offset,
due to inconsistency in spec.
Best wishes,
George
Received on Friday, 27 February 2004 03:46:02 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:26 GMT