- From: Andrew n marshall <amarshal@usc.edu>
- Date: Tue, 25 Nov 1997 01:05:46 -0800
- To: "'www-style@w3.org'" <www-style@w3.org>
I haven't figured out a reasonable way to specify that an element should
take up the remaining space, either horizontally or vertically. This is
specifically related to positioning alongside a dynamic element, such as a
sidebar who width is defined by the text it contains.
Does it seem reasonable to have a value 'available' for properties 'height'
and 'width'?
This would solve my immediate problem, but it does explicitly describe what
happens when two or more elements with "height/width:available;" occur in
the same line/column. The simple solution involves splitting the space
evenly. There is also the solution implemented by HTML <FRAME> tags:
define a unit '*' to weight the division of the remaining space. This is
more general, but also more difficult to implement (although most of it is
already required to implement "alignment:justify;").
Any sort of 'remaining/available' setting is apt to have a profound effect
on the rendering/positioning algorithm since it forces the document object
tree to be traversed twice during rendering (first for normal width/height
values, second to allocate remaining space and set final positions).
Has this been thought of/discussed before? What are your the reactions?
On a similar note, what to people think of adding the short-cut values of
'maximum' and 'minimum' to height and width? 'maximum' should mean:
horizontal-visibility==visible ?
max( min( width-max, width available ), width needed ) :
min( width-max, width available );
and 'minimum' should mean:
horizontal-visibility==visible ?
max( width-min, width needed ) :
width-min;
Obviously similar values could be defined for height.
(Note: horizontal-visibility was proposed in my last email. It is
equivalent to the current visibility, but only applies to horizontal
spacing)
Andrew n marshall
student - artist - programmer
http://www.media-electronica.com/anm-bin/anm
"Everyone a mentor, Everyone a pupil"
Received on Tuesday, 25 November 1997 04:05:29 UTC