This doesn't seem to return anchors properly. That is, on an HTML page with the following in the body:

<p>Here is an anchor:
<br><a id="blar"></a>Blar
</p>

the following javascript returns nothing:

document.onload = alert(document.getElementById('blar'));

This is a sorry state of affairs, is it not?

-- It certainly might be, if the code were doing what you think it is. However, you have a few errors:

• The alert statement itself must be inside a function block; otherwise, you are assigning the return value of the call to the event handler.
• The onload event handler is on the window object, not the document object.

Here is the correct code, which returns the anticipated value of "[object]":

window.onload = function() 
{
   alert(typeof(document.getElementById('blar')));
}

(jsudds)

It should be noted, that this method also matches on the name attribute, thus if you had say:

<meta name="description" content="matching on this is a bug"/>

<!-- All your code... -->

<textarea name="description" id="description">This is information about the bug</textarea>

In W3C compliant browsers, document.getElementById('description'); would match on the form field, not the meta tag!!!

You can make Internet Explorer's document.getElementById() method work according to W3C standards and only return elements with matching id and not name by overriding the native function in JavaScript like this:

if (/msie/i.test (navigator.userAgent)) //only override IE
{
  document.nativeGetElementById = document.getElementById; 
  document.getElementById = function(id)
  {
    var elem = document.nativeGetElementById(id);
    if(elem)
    {
      //make sure that it is a valid match on id
      if(elem.attributes['id'].value == id)
      {
        return elem;
      }
      else
      {
        //otherwise find the correct element
        for(var i=1;i<document.all[id].length;i++)
        {
          if(document.all[id][i].attributes['id'].value == id)
          {
            return document.all[id][i];
          }
        }
      }
    }
    return null;
  }
}

Workaround attributed to http://webbugtrack.blogspot.com/2007/08/bug-152-getelementbyid-returns.html
Workaround revised to work around a different error in IE's implementation of getAttribute() see:
http://www.sixteensmallstones.org/ie-javascript-bugs-overriding-internet-explorers-documentgetelementbyid-to-be-w3c-compliant-exposes-an-additional-bug-in-getattributes

This will generate the message "Object doesn't support this property or method"

<SCRIPT>
function fnGetId(){
   // Returns the first DIV element in the collection.
   var oDiv1=document.getElementById("oDiv1");
}
</SCRIPT>
<DIV ID="oDiv1">Div #1</DIV>
<DIV ID="oDiv2">Div #2</DIV>
<DIV ID="oDiv3">Div #3</DIV>
<INPUT TYPE="button" VALUE="Get Names" onclick="fnGetId()">

[jsudds] This isn't a "working" example of the problem. The script here, which was copied from the topic above, does not generate an error message. I tried to implement the behavior you describe, by using the script name and a global object, but both IE6 and IE7 run it flawlessly.

<script type="text/javascript">
function fnGetId()
{  

   var oDiv1 = document.getElementById("fnGetId");
   alert(oDiv1.innerHTML);

   var oDiv2 = document.getElementById("window");
   alert(oDiv2.innerHTML);
}
</script>
<div id="fnGetId">Div #1</div>
<div id="window">Div #2</div>
<input type="button" value="Get Names" onclick="fnGetId()"/>

Can you provide a repro?

[sglass] There's no doubt this exists. We've been dealing with this issue for the past two days. Turns out we were using this shared library called mootools. Within mootools they had defined a function called getElementById(). Once I renamed this method in mootools, my problem dissappeared. This seems to be plaguing a lot of IE users out there. Hopefully they'll find this and get past it.

[arnoschaefer]

I can reproduce the "Object doesn't support this property or method" error in IE8 when omitting the "var" keyword, thus declaring the variable implicitly in the global context:

<script type="text/javascript">
function doit () {
    a = document.getElementById ("a");
    alert (a);
}
</script>
</head>
<body onload="doit ();">
<h1 id="a">foo</h1>
</body>

Do the following:

<div id="a"></div>
<div id="b"></div>
<script type="text/javascript">
document.getElementById("a").innerHTML = "test";
document.getElementById("b").innerHTML = "test";
</script>

and the following:

<div id="a" />
<div id="b" />
<script type="text/javascript">
document.getElementById("a").innerHTML = "test";
document.getElementById("b").innerHTML = "test";
</script>

They do not work exactly the same way, even though they should, since <tag></tag> is the same as <tag />.


[Scott Trenda]

<tag></tag> is the same as <tag /> in XML, but not in HTML. You need to explicitly open and close the tag in order for the HTML parser to correctly interpret the structure.

There seems to be a problem with IE6 & 7 and the getElementById function

When looking for an Id in a DIV everything works perfectly fine, however if the DIV is inside a form, the DIV is not found...

Ohter browsers such as Opera, Safari and Firefox work fine...

Is there any workaround ?

N.B. #1.

There exists a strange data type, call it, hole,
for which document.getElementById(hole)
generates an error, instead of a null object...

This data type, hole, is the blank value
of ... srcElement.parentElement.id ...
where ... there is no id, in an A-element ...

hole == blank,
hole != null,

It crashes getElementById() and cancels its JScript.

(I'm yet tracking it down... it's running deep.... Ray.)

N.B. #2.

By name, does not work in MSIE-6,7 ... And, returns differently:

A-element returns "" (blank);
P-element returns "null";
Q-element returns "null";
DIV, SPAN, BODY,,,, etc. return "null".

This latter is possibly documentation confusion, as NAME does not show in Parameters here.