- From: Ian Hickson <ianh@netscape.com>
- Date: Mon, 16 Oct 2000 13:55:26 -0700 (Pacific Daylight Time)
- To: fantasai <fantasai@escape.com>
- cc: www-style@w3.org
On Mon, 16 Oct 2000, fantasai wrote: > > From your proposal > (http://lists.w3.org/Archives/Public/www-style/1999Nov/0241.html): > > Proposed syntax: x:matches(y) > > WD syntax: x:subject y > > matches: <x MATCHED> <z> <y/> </z> </x> > > In the syntax I've described, x:matches(y) wouldn't match anything. > > I wrote: >> Redefining X:matches(Y) to mean any X that matches the Y pattern where >> Y is given as a CSS selector. > > Thus, to use :matches for the above situation, you would have to put > x:matches($ y) Ooooooh. I like this. This is much better than :subject and :matches()! > :matches() in this context takes the intersection of the set defined in the > parentheses and the set defined by the simple selector it's modifying. > > In 'x:matches(y)', the intersection of 'x' & 'y' is the empty set - > 'x' matches any element <x>. > 'y' matches any element <y>. > You cannot find an element that matches both conditions. > > For 'x:matches($ y)' - > 'x' matches any element <x> > '$ y' matches any element with a descendant of <y>. > An element in the pattern <x> <z> <y/> </z> </x> will satisfy both > conditions. Yes! This is great. A better syntax for :subject, along with future extensibility for complex selectors. I like it. -- Ian Hickson )\ _. - ._.) fL Netscape, Standards Compliance QA /. `- ' ( `--' +1 650 937 6593 `- , ) - > ) \ irc.mozilla.org:Hixie _________________________ (.' \) (.' -' __________
Received on Monday, 16 October 2000 16:57:06 UTC