getElementsByClass

from

function getElementsByClass(searchClass,node,tag) {
	var classElements = new Array();
	if ( node == null )
		node = document;
	if ( tag == null )
		tag = '*';
	var els = node.getElementsByTagName(tag);
	var elsLen = els.length;
	var pattern = new RegExp("(^|\\s)"+searchClass+"(\\s|$)");
	for (i = 0, j = 0; i

How it works

It’s simple. It works just how you think getElementsByClass would work, except better.

Supply a class name as a string.

(optional) Supply a node. This can be obtained by getElementById, or simply by just throwing in “document� (it will be document if don’t supply a node)). It’s mainly useful if you know your parent and you don’t want to loop through the entire D.O.M.

(optional) Limit your results by adding a tagName. Very useful when you’re toggling checkboxes and etcetera. You could just supply “input“. Or, if you’re like me, and you said Good Bye to IE5, you can use the “*� asterisk as a catch-all (meaning ‘any element).