Re: [CSS21] Status of defult (intrinsic) style sheet in UA

----- Original Message ----- 
From: "Boris Zbarsky" <bzbarsky@mit.edu>


> Andrew Fedoniouk wrote:
>> Again how then they related to each other?
>
> That's what the section that talks about sorting specifies.
>
>> Let's say we have element with initially empty style.
>>
>> Then sequence of style resolving actions will be
>>
>> 1) Apply UA default styles.
>> 2) Apply author and user styles.
>>
>> #2 will unconditionally override values of #1.
>
> OK.  But how is that different from:
>
> 1)  Apply UA default styles
> 2)  Apply user styles
> 3)  Apply author styles
> 4)  Apply author !important styles
> 5)  Apply user !important styles
>
> which is what the spec currently says?

Where did you read "Apply"?   Spec says "Sort".
Apply assumed after sorting and in the sort order, right?

> That is, what is the benefit of having a 2-step process with step 2 broken 
> up into 4 steps vs just having a 5-step process?

This is exactly what I am trying to explain:

There are five independent style systems.
One of it is UA default style table.
Independency of each system shall be highlighted.
They are orthogonal - "weight" of
particular style definition in one layer is not
conflicting nor even comparable with other weights
in other style layers.

And it is really does not matter where UA styles come
from. And you right, they even can not exist in
pure form as style definitions.


>
>> Why do we need to put them in one table?
>
> It seems to me that it makes it clearer what overrides what -- there is 
> only one place you have to look at to find out.
>
>> ... according to cascading rule
>> "more specific selectors will override more general ones."
>> UA rule will override Author style for second li:
>> <ul><li><li>...</li></li></ul>
>
> See my mail requesting clarification that each subsequent sort is only 
> within things that were "the same" for a previous sort.

Got it, thanks. This is it too.

>
>> If we need all five style systems behave as one table
>> then why just not to add into
>> [http://www.w3.org/TR/CSS21/cascade.html#specificity]
>> one more "digit" into "a-b-c-d" ?
>
> That could indeed be done (and is in fact more or less what the "inline 
> style rule" digit is)...

Yep. For the terms of formal specification this will work better.


Andrew Fedoniouk.
http://terrainformatica.com

Received on Wednesday, 23 November 2005 04:20:01 UTC