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

RE: selectors in :any

From: Justin Rogers <justrog@microsoft.com>
Date: Tue, 22 Mar 2011 18:03:48 +0000
To: "L. David Baron" <dbaron@dbaron.org>, Ojan Vafai <ojan@chromium.org>
CC: "www-style@w3.org" <www-style@w3.org>, David Hyatt <hyatt@apple.com>, "koivisto@iki.fi" <koivisto@iki.fi>
Message-ID: <5C67CAD50F0B1640AB2F23A115902DA069714DF7@TK5EX14MBXC118.redmond.corp.microsoft.com>

I'm using the above for reference. I'm a little bit concerned here. It seems the standard rules of selectors are broken by the syntax. Namely that a simple selector consisting of a pseudo class :foo is actually the selector *|*:foo, but the example notes that the replacement for :-moz-any is in place and is no longer a "tag rule". This is creative, but if we plan on doing this we should introduce a concept into the grammar which is not a pseudo-class that allows you to root the front of a selector using a new special character so you can disambiguate from a standard tag rule immediately.

Justin Rogers [MSFT]

-----Original Message-----
From: www-style-request@w3.org [mailto:www-style-request@w3.org] On Behalf Of L. David Baron
Sent: Tuesday, March 22, 2011 9:51 AM
To: Ojan Vafai
Cc: www-style@w3.org; David Hyatt; koivisto@iki.fi
Subject: Re: selectors in :any

On Tuesday 2011-03-22 18:18 +1100, Ojan Vafai wrote:
> For now, WebKit's implementation of :any is following Mozilla's lead 
> and only allowing simple selectors. I don't see any need for this restriction.
> Are there any objections to allowing all selectors?
> I see a similar discussion has happend for :not and that there seemed 
> to be agreement that we'd change :not in Selectors 4 to not restrict 
> to simple
> selectors.*

FWIW, the restrictions on the two are actually different.

The arguments to :-moz-any() are what CSS2.1 calls a simple selector and what css3-selectors calls a sequence of simple selectors.

The argument to :not() is what CSS2.1 calls a component of a simple selector and what css3-selectors calls a simple selector.

So there are a bunch of options for what we could change.  We could:
 * change :not() to accept what CSS2.1 calls a simple selector
   and what css3-selectors calls a sequence of simple selectors.
 * change either to accept a selector (which implicitly makes :any()
   take a group of selectors, since it's comma operator is
   essentially the same as selector grouping)
 * change :not() to accept a group of selectors

(I regret not formally objecting to the shifts in terminology.)


L. David Baron                                 http://dbaron.org/
Mozilla Corporation                       http://www.mozilla.com/
Received on Tuesday, 22 March 2011 18:04:22 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:38:44 UTC