Re: Inheriting computed or other values

Ian Hickson wrote to <www-style@w3.org> on 27 September
2002 in "Re: Inheriting computed or other values"
(<mid:Pine.LNX.4.21.0209271151200.15733-100000@dhalsim.dreamhost.com>):

> On 26 Sep 2002, Etan Wexler wrote:
> 
> > [...]  If the Working Group decrees that a
> > <number> cascaded value yields a <number>,
> > rather than <length>, computed value, it will
> > be necessary to further compute the computed
> > value before rendition.
> 
> This is already the case with several properties,
> e.g. border-width and the keywords, font-size and
> available fonts, etc.

Why would "border-top-width: thin", for example, yield
anything but an absolute <length> computed value?

As for font sizes, it might (not will) be necessary to
modify the computed value, but only because of external
factors.  What is more, the computed 'font-size' value
should be ready to use as is: that is, if the computed
size is available, it is used.  The case of a <number>
'line-height' value differs in that it always requires
further computation.

> This is why we have the 'actual value' step.

My understanding differs from yours.  I quote CSS2
section 6.1.3, "Actual values"
(<http://www.w3.org/TR/REC-CSS2/cascade.html#actual-value>):

"A computed value is in principle ready to be used, but
a user agent may not be able to make use of the value
in a given environment. For example, a user agent may
only be able to render borders with integer pixel
widths and may therefore have to approximate the
computed width. The actual value is the computed value
after any approximations have been applied."

My understanding is that actual values may differ from
computed values only due to system limitations.

Maybe we need yet another step in the value processing
chain:

1. Declared value (as proposed by me elsewhere [1])

2. Cascaded value (the proposed renaming of "specified
   value" [1])

3. Inheritable value (this is the new step)

4. Computed value (ready to use)

5. Actual value (modified by system limitations)

This model also happens to solve the issue of
compounding 'font-size-adjust' [2] [3].


[1]  Etan Wexler.
"Re: Specified values: what are they?".
12 September 2002. 
<mid:CSS-specified-values@d20020912.etan.wexler> /
<http://lists.w3.org/Archives/Public/www-style/2002Sep/0117.html>.

[2]  Michel Suignard and Chris Lilley, editors.
"3.4 Font size: the 'font-size' and 'font-size-adjust' properties"
   in "CSS3 module: Fonts".
2 August 2002. 
<http://www.w3.org/TR/2002/WD-css3-fonts-20020802/#font-size-props>

[3]  Etan Wexler.
"WD-css3-fonts-20020802 comments". 
26 September 2002. 
<mid:comments-WD-css3-fonts-20020802@d20020926.etan.wexler> /
<http://lists.w3.org/Archives/Public/www-style/2002Sep/0176.html>.

Received on Friday, 27 September 2002 17:48:23 UTC