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

Re: CSS Selector Complexity : Working draft document

From: Boris Zbarsky <bzbarsky@MIT.EDU>
Date: Thu, 31 Jul 2008 13:58:26 -0700
Message-ID: <48922772.6010502@mit.edu>
To: Francois Remy <fremycompany_pub@yahoo.fr>
CC: CSS 3 W3C Group <www-style@w3.org>

Francois Remy wrote:
> - Loop Complexity (or static complexity: time needed to find matching 
> elements at load time)

As I said in my other mail, this is not an interesting metric for a 
rendering engine (unless you're implementing querySelectorAll).  During 
load time the question that's asked (repeatedly) is not "which elements 
match this selector?" but "which selectors match this element, so that I 
can apply the relevant style rules?"

> - Update Complexity (number of times the matched elements can be 
> reevaluated * time it need)

My previous post broke this down in what I think is a more useful way.

>     * I voluntary excluded the optimizations that can be done by the UA 
> because they are probably not the same

But the space of possible optimizations is in fact key...  The question 
is always what the fastest smallest way is to implement something.

Received on Thursday, 31 July 2008 20:59:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:10 GMT