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

Re: [css3-text-layout] New editor's draft - margin-before/after/start/end etc.

From: Andrew Fedoniouk <news@terrainformatica.com>
Date: Fri, 28 May 2010 21:13:27 -0700
Message-ID: <C480B5EC3C5C4F72A93794D1A2125F50@terra3>
To: HåkonWium Lie <howcome@opera.com>, "Alan Gresley" <alan@css-class.com>
Cc: <www-style@w3.org>


--------------------------------------------------
From: "HåkonWium Lie" <howcome@opera.com>
Sent: Friday, May 28, 2010 11:26 AM
To: "Alan Gresley" <alan@css-class.com>
Cc: <www-style@w3.org>
Subject: Re: [css3-text-layout] New editor's draft - 
margin-before/after/start/end etc.


>  .start:lang(he), .start:lang(ara) {
>    .start li { margin-right: 100px; padding-right: 100px }
>    .end li { margin-left :100px; padding-left: 100px;}
>  }
>
> Now, there's a bunch of Arabic and Persian language codes, so you would
> have to extend the selector to get complete coverage.
>
> Using language codes as hooks would be an incentive for people to use
> language codes.
>

I've proposed on the list once to introduce :ltr and :rtl pseudo-classes.

:rtl is true if element itself or its nearest parent have @dir defined
and value of that @dir is exactly "rtl"
In the same way :ltr is defined. And :ttb.

Having them all these controversial margin-before/after/start/end are not
needed as anyone can explicitly define

 .start
{
   .start li { margin-right: 50px; padding-right: 50px }
   .end li { margin-left :100px; padding-left: 100px;}
}

 .start:rtl
{
   direction: rtl;
   .start li { margin-right: 100px; padding-right: 100px }
   .end li { margin-left :50px; padding-left: 50px;}
 }

So the @dir is a sole source that defines all aspects
of directionality. Having one CSS property to be dependable
on another is IMO path nowhere.

Anyway your list ( :lang(he), :lang(ara) ) is not full.
And yet even on, say, Arabic pages it could be
"LTR islands" that are marked by dir="ltr".

And think about pages like gmail. You could have
the page itself using :lang(en) but email message will
be :lang(ara) or so. I think it is obvious what may happen
with your selectors in this case.


My two cents.

-- 
Andrew Fedoniouk

http://terrainformatica.com



 
Received on Saturday, 29 May 2010 04:13:59 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:27 GMT