Anne van Kesteren wrote:
On Mon, 14 Jul 2008 21:30:18 +0200, Boris Zbarsky <[EMAIL PROTECTED]> wrote:
Last I checked, the main envisioned use of the querySelector* APIs was
to speed up existing queries toolkits do. Might be worth asking them
what they think would be useful.
http://krijnhoetmer.nl/irc-logs/whatwg/20080709#l-597
http://api.dojotoolkit.org/jsdoc/dojo/HEAD/dojo.query (It doesn't even
do XML unless all names are lowercase.)
http://www.prototypejs.org/api/utility/dollar-dollar
I agree that asking them explicitly would be good though, but it seems
like so far the need was not compelling enough to implement it. (And
libraries generally have gone to extreme lengths to implement other
things not natively supported by browsers or not available cross browser.)
Given that IE, Webkit and Opera seem to be in favour of dropping
NSResolver, if Mozilla will agree, I'm more than happy to do so and
possibly deferring it till a future version of the spec when we can
really justify its use cases and find a solution that doesn't suffer
from so many problems. If it is dropped, then I'd rather not have an
implementation ship with support for it, since then it would probably
need to be specced anyway Although, maybe it would depend on whether or
not content relies on it.
--
Lachlan Hunt - Opera Software
http://lachy.id.au/
http://www.opera.com/
