There is a ticket raised for this - http://dev.jquery.com/ticket/1854
- as a result of a discussion on the jQuery development list.
It would seem there is some doubt about whether or not it is actually
a bug, since this restriction (or feature) was (allegedly!) a
conscious decision taken during implementation of the not()
functionality (method and filter).
[but shsh! ... keep it quiet, because very few people know yet!]
Meantime, what you have to do is change your code to call filter(),
with a function as its argument, which returns the inverse of a test
for what you want to exclude (?!) ...
eg
$('form :input').filter( function(){ return !$
(this).is(':hidden[name=a]'); } );
Neat, huh?
On Oct 31, 7:16 pm, Pete Kruckenberg <[EMAIL PROTECTED]>
wrote:
> I've come across some behavior of not() that doesn't match the docs,
> or I'm not understanding the docs.
>
> Given the following form:
>
> <form>
> <input type=hidden name=a>
> <input type=hidden name=b>
> </form>
>
> I'd expect "$('form :input:not(:hidden[name=a])')" to return "[<input
> type=hidden name=b>]",
> since "$('form :hidden[name=a]')" returns "[<input type=hidden
> name=a>]" and
> and not(selector) should "Filters out all elements matching the given
> selector".
>
> However, "$('form :input:not(:hidden[name=a])')" instead returns
> "[ ]" (empty array), acting more like
> "$('form :input:not(:hidden):not([name=a])')" or "$
> ('form :input:not(:hidden,[name=a])')" (i.e. "or"ing the selector
> parameters instead of "and"ing them).
>
> "$('form :input').not(':hidden[name=a])')" produces the same results
> (empty array). I'd expect it to be the inverse of "$
> ('form :input').filter(':hidden[name=a])')" (which returns "[<input
> type=hidden name=a>]".
>
> Am I misunderstanding the docs, or is this a bug with not()?
>
> I get the same results in both 1.2.1 and 1.1.4. Tested on FF2.0.0.8/
> Win.
>
> Thanks.
> Pete.
