" ... If I create a div inside whatever function thanks gosh Sizzle
won't return
that div in a search $("div"), do you follow me? ..."
I "follow" this yes, I am just "wondering" if this FF "feature" might
break "something somewhere" in jQ,
because jQ users are allowed to create dom nodes "on fly"and use them
as a container (aka context), with Sizzle, without ever attaching them
to the curent dom tree:
// example 1
function faraway ( jq_ ){ return jq_.find(".a") ; }
//
$div = $("<div style='color:red!important' ><div><div><div class='a b
c'>ABC</div></div></div></div>") ;
// now pass and search inside this "un-attached" element
$a = faraway( $div ) ;
//
$a.css("color")
/* returns : red */
And , yes, for the above test
Returns is an empty string (aka "") in CHROME :
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/532.3
(KHTML, like Gecko) Chrome/4.0.223.11 Safari/532.3
Return is "red" in IE8 :
Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0;
SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729;
Media Center PC 6.0)
And return is "rgb(255, 0, 0)" (aka "red") in FF :
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-GB; rv:1.9.1.3) Gecko/
20090824 Firefox/3.5.3
This example is not that esoteric jQ usage, and thus it is more likely
to happen. And fail in CHROME ...
Now I do not know is this only this Chrome . is it also the same in
Chromium, or in Opera 10, etc ... ?
For now I attach element to the head (temporarily) which seems to
work.
function example2(color) {
function faraway(jq_) { return jq_.find(".a"); }
var $div = $("<div style='color:" + color + "!important'
><div><div><div class='a b c'>ABC</div></div></div></div>");
// do this for browsers (like CHROME) who do not allow CSS2
properties on non-attached element
$div.appendTo($("head"));
// now pass and search inside this "un-attached" element
var $a = faraway($div);
var retval = $a.css("color")
$div.remove();
return retval;
}
example2("red") returns "red" in IE and "rgb(255,0,0)" in FF and
CHROME ...
AS I said, if jQ team thinks this is necessary to solve before people
start complaining (possibly) about jQ + Chrome, they are wellcome to
use these findings.
--DBJ
On Oct 24, 11:09 am, Andrea Giammarchi <andrea.giammar...@gmail.com>
wrote:
> P.S. for all ... the reason getComputedStyle does not make sense for a node
> is that styles are inherited and you'll never know in which position that
> style would be respected or not.
>
> You append the div in a node with style color:red !important and your div
> had style color:black ... how can you predict a style if you don't know what
> kind of dependencies/inheritance that node is gonna have one appended
> somewhere?
>
> If an engine does it by default it's silly cause lots of properties gonna be
> replaced, changed, modified, due to Cascade Style Sheet inheritance.
>
> Regards
>
> On Sat, Oct 24, 2009 at 11:05 AM, Andrea Giammarchi <
>
> andrea.giammar...@gmail.com> wrote:
> > Sizzle searches over nodes already present in the document.
>
> > If I create a div inside whatever function thanks gosh Sizzle won't return
> > that div in a search $("div"), do you follow me?
>
> > I know you simply brought here yet another inconsistency in the browsers
> > panorama, but to fix this doubtfully useful operation which logically does
> > not make sense I would go for the other way .... if(!node.parentNode)return
> > null;
>
> > To have a computed style for a not in DOM element you need to append it,
> > retrieve the style, remove the node ... no way, since this is gonna be a
> > performances killer and we all take care about performances, don't we?
>
> > Regards
>
> > On Sat, Oct 24, 2009 at 10:14 AM, DBJDBJ <dbj...@gmail.com> wrote:
>
> >> @Andrea : thanks for normal tone, but again somehow you fundamentaly
> >> misunderstood me ? If anything you (probably) followed my code
> >> through ;o)
>
> >> One of the strengths of jQuery is: It smooths the sharply uneven
> >> surface of cross browser javascript.
> >> If I post something that does not mean I want to do it that way and no
> >> other way.
> >> I simply stumbled upon this issue which , I thought, might be
> >> interesting for the jQ team, because this is yet another "feature"
> >> that works in some browsers and does not work in others.
> >> 99% of jQuery are much more concerned to have code that works and that
> >> uses jQuery, than to engage in W3C spec minutiae...
>
> >> My example/finding uses somewhat esoteric code, but the issue of
> >> CHROME not allowing getComputedStyle() on non-attached new elements,
> >> might break some other part of Sizzle or jQ core, so I thought I
> >> report it...
>
> >> --DBJ
>
> >> On Oct 24, 1:23 am, Andrea Giammarchi <andrea.giammar...@gmail.com>
> >> wrote:
> >> > On Sat, Oct 24, 2009 at 12:22 AM, DBJDBJ <dbj...@gmail.com> wrote:
>
> >> > > Well for the FF team this was not meaningless, since it is possible to
> >> > > set/get css value on the non-attached newly created dom element in FF
> >> > > and then get to it through getComputedStyle()
> >> > > You just have to follow in the Firebug and Chrome console, what I have
> >> > > posted initally. Instead of "wondering" ...
>
> >> > DBJ it does not matter what an engine does internally, it maters why on
> >> > earth you would ask for a Computed property when this has never been
> >> > computed ... this would be normal for a JScript engine, not SpiderMonkey
> >> or
> >> > V8 where performances should be a priority.
>
> >> > Accordingly, I would rather ask Firefox (or Firebug plug-in) or Chrome
> >> why
> >> > they provide such information for an element has never been rendered
> >> since
> >> > from *Computation* point of view this is the real waste of time/resurces
> >> and
> >> > probablythe reason Firefox 3.5 is so slow with DOM render compared even
> >> with
> >> > Internet Explorer.
>
> >> > It's nice in any case to see how you trust "applications" more than
> >> logic
> >> > ... do you agree that a picture could never be showed if that fact never
> >> > happened?
>
> >> > Regards
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"jQuery Development" group.
To post to this group, send email to jquery-dev@googlegroups.com
To unsubscribe from this group, send email to
jquery-dev+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/jquery-dev?hl=en
-~----------~----~----~----~------~----~------~--~---
