I think the problem is the cache optimization for previousSibling when
the search match is in the checkSet.
If you need a quick fix, removing the if
if ( sibDir && elem.nodeType === 1 ) { ... }
in dirNode and dirNodeCheck should fix the problem (didn't try it)
--Fabio
On Fri, Mar 6, 2009 at 4:15 PM, d3r1v3d (Gavin Mulligan)
<[email protected]> wrote:
>
> Actually, I found a bit more time to actually attempt the 'patch' I
> suggested in the previous post. I want to tweak my suggestion to:
>
> 1.3.2 jquery.js:
> - 2357: match = checkSet[elem.sizset];
> + 2357: match = true;
>
> That fixes the error found in the test code given in the root of this
> list thread, however I have no idea what the far-reaching implications
> of that change are in the grand scheme of jQuery. :)
>
> - Gavin
>
> On Mar 6, 9:37 am, "d3r1v3d (Gavin Mulligan)" <[email protected]>
> wrote:
>> I did a bit of snooping and think I've tracked down what the problem
>> seems to be. In selector.js, I believe the issues emanates from
>> changes made to the dirCheck function.
>>
>> In 1.3.2, the match variable is set to the 'elem.sizset' index of the
>> checkSet array. This checkSet array contains the 'selector validation
>> status' (my terminology) of a series of sibling divs (in this case).
>> The following code, in 1.3.2 dirCheck(...), should set the match
>> variable equal to the element which matches the given selector based
>> on the current element in checkSet being investigated:
>> if ( elem.sizcache === doneName ) {
>> match = checkSet[elem.sizset];
>> break;
>>
>> }
>>
>> In the test file, the div in checkSet being investigated is the second
>> div in the body and elem, in this case, is the div#prev which matches
>> to the given selector. I believe that match should be set equal to
>> elem here instead of checkSet[elem.sizset] or, in other words:
>> - match = checkSet[elem.sizSet];
>> + match = elem;
>>
>> This is similar to how 1.3.1 handles this case. However, I am nowhere
>> near familiar enough with the selector code to suggest this as a
>> definitive patch. Does someone who is intimately familiar with the
>> selector.js code think this would do the trick?
>>
>> - Gavin
>>
>> On Mar 6, 8:32 am, Cloudream <[email protected]> wrote:
>>
>> > nightly is 1.3.1 ...
>>
>> > On Fri, Mar 6, 2009 at 1:30 PM, Igor <[email protected]> wrote:
>>
>> > > Hi!
>> > > Have been using your library for ages, it is awesome, thank you!
>>
>> > > Recently found some issues, that seems to be related to
>> > > previousSibling ( "~" ) selector
>> > > Can you point me to a place, where it is broken, since i afraid of
>> > > using nightly builds in product
>>
>> > > Sample markup:
>> > > <html>
>> > > <head>
>> > > <!-- 1.3.2. looks like broken -->
>> > > <script type="text/javascript" src="http://ajax.googleapis.com/ajax/
>> > > libs/jquery/1.3.2/jquery.min.js"></script>
>> > > <!-- Nightly. works --> <!--
>> > > <script type="text/javascript" src="http://code.jquery.com/nightlies/
>> > > jquery-nightly.js"></script>
>> > > --><!-- 1.3.1. works --><!--
>> > > <script type="text/javascript" src="http://ajax.googleapis.com/ajax/
>> > > libs/jquery/1.3.1/jquery.min.js"></script>
>> > > -->
>> > > <script type="text/javascript">
>> > > $(document).ready(function(){
>> > > console.log($('#prev ~ div').length);/*expecting 1, got
>> > > zero
>> > > in
>> > > 1.3.2*/
>> > > });
>> > > </script>
>> > > </head>
>> > > <body>
>> > > <div id="prev">1</div>
>> > > <div>2</div>
>> > > </body>
>> > > </html>
> >
>
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"jQuery Development" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/jquery-dev?hl=en
-~----------~----~----~----~------~----~------~--~---
