- From: Will Levine <wlevine@gmail.com>
- Date: Mon, 12 Jul 2004 14:37:39 -0400
On Mon, 12 Jul 2004 18:22:11 +0000 (UTC), Ian Hickson <ian at hixie.ch> wrote:
> >
> > I'm no expert on the DOM, but I don't really understand what you mean.
> > The following code:
> >
> > <object id="log" classid="pizza" name="mmm"></object>
> > <script>
> > alert(document.getElementById("log").getAttribute("name"));
> > alert(document.getElementsByTagName("object")[0].getAttribute("classid"));
> > </script>
> >
> > works exactly how I would expect it to work in IE 6.
>
> Now change "pizza" to "urn:pizza" (or anything that doesn't resolve in
> DNS) and add something inside the <object>.
Uh huh. That is very odd.
> See also some of the tests here:
>
> http://www.hixie.ch/tests/evil/page-loading/alternative-content/investigation/ieisms/
>
> ...for more fun.
>
> Basically <object> is so incredibly unreliable in IE that it is unusable.
Point taken.
> > Also, I thought someone said something about IE discarding DOM info on
> > unknown elements (e.g. datalist) also.
>
> It doesn't discard them, it keeps enough information in the DOM to
> reconstruct unknown elements. (To be precise, it converts unknown start
> tags to empty elements and unknown end tags to empty elements with the tag
> name prefixed by a slash.)
I don't really understand what you are saying here, but now that I've
done a bit more testing I can certainly understand how a unknown
element is easier to play with than a object element.
Will
Received on Monday, 12 July 2004 11:37:39 UTC