Yeah, as Klaus mentioned, it is a tiny bit faster if you leave out the
'img'.
The real optimization you can do is to avoid using the same selector
repeatedly. In a way, jQuery makes it too easy to write inefficient code:
$('#foo').doSomething();
// ...later...
$('#foo').doSomethingElse();
Replace that with:
var $foo = $('#foo');
$foo.doSomething();
// ...later...
$foo.doSomethingElse();
As you can see, I use $variableName as a convention to remind me that the
variable is a jQuery object.
This will make more of a difference with slower selectors like
$('.someClass'). Even though jQuery is much faster than it used to be, it's
still faster if you can avoid repeating the selector query at all.
-Mike
> From: NeilM
>
> I have just started to use jQuery in my web development and
> have a question concerning optimising element selection.
>
> I find that I create a lot of references to id'd elements:
>
> <img id="myImage" src="todaysimage.gif" width="20" height="20" />
>
> Then, in my JavaScript, I want to obtain a reference to the element.
> Now, even though it has an id, I tend to use the full CSS
> descriptor as, to me anyway, it makes the code more readable:
>
> $("img#myImage").addClass("border");
>
> My question is, apart from the overhead of some extra
> characters in the script (e.g. 'img#'), does this style of
> specification introduce a performance overhead internally
> within jQuery? Would/does jQuery optimise the element
> selection process if it just 'sees' a simple id string
> starting with a #?
>
> $("#myImage").addClass("border");
