[Javascript] Wrestling with DOM access (artificial attributes, IE vs Firefox vs Safari)

[Triche Osborne]

Charles Albrecht wrote:
> In my case, the element I'm looking at is something like:
> 
> <tr id="row512" level="2" hidecount="0" expanded="true">  [1]
> 
> . . . Actually, I'd probably prefer to get them
> out of the tag attributes so the page will come closer to validating
> against its doctype.
> 
> I was trying to think of other ways of doing this. Perhaps trying to
> get the values out of the DOM into an object or series of objects (one
> for each of mytable.getElementsByTagName("tr")). But of course, it all
> needs to tie back to the DOM.

I often initialize new instance properties (onload) for form validation. 
I know this works in Moz/FF and IE, though I've not tested in others. 
You could to the same with your TRs, providing it works in the browsers 
you're concerned about. Something like . . .

function instantiate()
{
	var rows = document.getElementsByTagName("tr");
	for(var i=0; i < rows.length; i++)
	{
		rows[i].newProperty = "initialPropertyValue";
	}
}

This attaches the property to the rows without impinging on your XHTML. 
Your remaining functions can alter the value of the property from there.

Triche