W3C home > Mailing lists > Public > www-style@w3.org > October 2011

Re: [css21] The 'inherit' keyword in shorthand properties

From: Øyvind Stenhaug <oyvinds@opera.com>
Date: Thu, 13 Oct 2011 17:52:27 +0200
To: "Tab Atkins Jr." <jackalmage@gmail.com>
Cc: "www-style@w3.org" <www-style@w3.org>
Message-ID: <op.v3arppb0bunlto@oyvinds-desktop>
On Thu, 13 Oct 2011 17:26:22 +0200, Tab Atkins Jr. <jackalmage@gmail.com>  

> On Thu, Oct 13, 2011 at 8:13 AM, Øyvind Stenhaug <oyvinds@opera.com>  
> wrote:
>> http://www.w3.org/2011/10/12-css-irc#T16-47-22
>> "RESOLVED: accept TabAtkins and fantasai's proposal such that inherit  
>> turns
>> the specified value into the parent's computed value"
>> I just realized that this doesn't necessarily make any sense for  
>> shorthand
>> properties. In fact, most of chapter 6 seems to operate under an  
>> unstated
>> assumption that each declaration setting a shorthand has been converted  
>> to
>> equivalent declarations for the corresponding longhand properties.
> Do you mean because it's possible that the computed values of the
> longhand properties end up not forming a valid value for the shorthand
> when combined?  Or just that shorthands don't really have computed
> values, and are just mechanisms to pipe values into the longhand
> properties?

Well, both, really. What first brought it to mind was indeed the cases  
where the shorthand can't represent all possible combinations. E.g. for

<div style="border-right-style: solid"><div style="border:  

both elements should end up with a solid right border, but you can't do  
that by getting a valid 'border' value from the outer div.

But more generally, computed values aren't really defined for shorthands  
as far as I can tell, specs mostly just say "see individual properties".

Øyvind Stenhaug
Core Norway, Opera Software ASA
Received on Thursday, 13 October 2011 15:53:06 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:05 UTC