W3C home > Mailing lists > Public > www-style@w3.org > March 2010

Re: [css3-values] Computed Values definition wrong

From: Øyvind Stenhaug <oyvinds@opera.com>
Date: Fri, 26 Mar 2010 11:39:20 +0100
To: Håkon Wium Lie <howcome@opera.com>
Cc: "www-style@w3.org" <www-style@w3.org>
Message-ID: <op.u957vus0ru61ud@oyvinds-desktop>
On Fri, 26 Mar 2010 06:21:40 +0100, Håkon Wium Lie <howcome@opera.com>  
wrote:

> fantasai wrote:
>
>  > http://www.w3.org/TR/css3-values/#computed0
>  >
>  > The Computed Values section of css3-values talks about resolving  
> percentages.
>  > It shouldn't, because in almost all cases, we don't resolve  
> percentages at
>  > the computed style stage, but at the used style stage.
>
> In many cases, percentages can be resolved without formatting the
> document. For example:
>
>    p { font-size: 120% }
>
>    div { width: 10cm }
>    div div { width: 50% }
>
> These types of expansions should take place in the computed values
> stage where the values are processed as far as possible without
> formatting the document.

So 'width' may be inherited (with the 'inherit' keyword) as an absolute  
value or as a percentage depending on the rest of the stylesheet? Sounds  
icky.

Also:

<div>
   <table>
     <tr>
       <td>
         Filler
         <div><!-- This won't be 5cm --></div>
       </td>
       <td>
         text
       </td>
     </tr>
   </table>
</div>

-- 
Øyvind Stenhaug
Core Norway, Opera Software ASA
Received on Friday, 26 March 2010 10:39:12 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:25 GMT