W3C home > Mailing lists > Public > www-style@w3.org > June 2008

Re: WebKit now supports CSS Variables

From: David Hyatt <hyatt@apple.com>
Date: Tue, 24 Jun 2008 12:42:50 -0500
To: Pascal Germroth <pascal@germroth.name>
Cc: Daniel Glazman <daniel.glazman@disruptive-innovations.com>, www-style list <www-style@w3.org>
Message-id: <530B5F48-02A7-4897-8D66-45DA047530BD@apple.com>

On Jun 24, 2008, at 9:12 AM, Pascal Germroth wrote:

> Hello,
>
>>> I'd prefer to throw away invalid declarations at parse time, and not
>>> have to keep them around. Therefore, I suggest that we add text
>>> similar to the @namespace specification:
>>>
>>>  The namespace prefix is declared only within the style sheet in
>>>  which its @namespace rule appears. It is not declared in any style
>>>  sheets importing or imported by that style sheet, nor in any other
>>>  style sheets applying to the document.
>> I see your point but I strongly disagree. The main purpose of CSS
>> Variables for a corporation is unicity of design for all corporate
>> web sites. That means having a single corporateVars.css somewhere  
>> that
>> all other stylesheets can import... If you break at @import  
>> boundaries,
>> the whole thing loses most of its interest.
>
> But the problem of your spec is, that it only allows the @variables- 
> block to appear before any @imports, so an imported  
> corporateVars.css would always override any "locally" defined  
> variables, this might be undesireable. So I suggest that @variables  
> and @imports should be allowed to appear in any order, so that a  
> stylesheet could override some imported variables locally.
>
> Also, what happens when A imports B and C, and C uses a variable  
> thats defined in B?
>

Then C uses B's variable in WebKit's implementation.  Variable  
resolution is delayed until all stylesheets are loaded.

dave
Received on Tuesday, 24 June 2008 17:43:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:07 GMT