DOM access issues in Firefox.

Fi_C**

Hi all,

I have an AJAX/ DOM/XML/Javascript site that is working just fine in IE - however the Javascript that is accessing the DOM does not seem to be working in FF. I have put alert statements all over the code and where everything seems to be failing is in the 2 statements highlighed in red below:

if (http_request.readyState == 4)
{ alert('step2');
if (http_request.status == 200)
{ alert('step3');
xmlDoc = http_request.responseXML;

var info = xmlDoc.getElementsByTagName('para');




for(i=0; i < info.length; i++)
{

alert('here1');
var infoNode = info.item(i);
if(infoNode != null && infoNode.hasChildNodes())
{
var infoPar = infoNode.getElementsByTagName('info');
alert('here2');

for(j=0; j<infoPar.length; j++)
{
alert('here3');
var infoParNode = infoPar.item(j);

var paragraph = infoParNode.getAttribute('paragraph');

var infoBody = infoPar.item(j).text;

//alert(infoPar.item(j));

alert(infoBody);



setInfoBody =''+openFormat+infoBody+closeFormat;


openFormat='';
closeFormat='';

}

heading = infoNode.childNodes(0).childNodes(0).data;

alert(headings);
setHead+='<a href="javascript:showInfo('+i+')" >'+heading+'</a><br>';
setInfo = '<p><img src="symb.gif"> '+heading[0]+'</p><div id="body_container">'+setInfoBody[0]+'</div><br>';


infoContainer.innerHTML = setInfo;

}


I've been trying different code all morning:
i.e. changing .data and .text to .value or .innerHTML
nothing seems to be working -
Has anybody come across this problem before? All the stuff I've been reading contradicts whats happening - seems that IE is the problem for everyone else on the net!

Thanks,
Fi*

}

Hobbes

If you custom install firefox it allows you to install a dom inspector. Might help in the debugging.

ast

I'm guessing here but try replacing

var infoBody = infoPar.item(j).text;
with
var infoBody = infoPar[j].text;

and

heading = infoNode.childNodes(0).childNodes(0).data;
with
heading = infoNode.childNodes[0].childNodes[0].data;

akari no ryu

Also:

// workaround for internet explorer
function getXmlNodeValue(xmlNode){
	try{
		if(XMLHttpRequest){
			return xmlNode.textContent;
		}
	}catch(e){
		try{
			return xmlNode.text;
		}catch(e){
		}
	}
}

The text property is an IE only thing. And what does the data property refer to?

Fi_C**

Thanks akari, using that text / textContent workaround worked well.
I have also resolved the other issue .. just by chance -,
I replaced
heading = infoNode.childNodes(0).childNodes(0).data;

with:
if(window.XMLHttpRequest){
heading = infoNode.firstChild.nextSibling.textContent;
}
else {
heading = infoNode.childNodes(0).childNodes(0).data;
}

I don't know why this works ...
It seems Mozilla sees the tree differently to IE - In Mozilla the node is seen as a sibling but in IE as a childNode?

Oh well I don't care as long as it works
Thanks again for your help - I'm off to fix all the other wierd cross-browser issues that have just been revealed

Regards,
Fi**

akari no ryu

It's not that Mozilla sees the tree differently to IE.

It's that IE ignores the specifications, repeatedly. If you look at www.xulplanet.com you'll see an object reference with all of the specified objects and their methods and properties. It's not much by way of tutorial but it's a fantastic resource.

Exploder just has its own way of doing things, as ever.