Re: [whatwg] behavior

Fri, 16 Oct 2009 03:48:42 -0700

On Fri, 16 Oct 2009 06:19:04 -0400, Simon Pieters <sim...@opera.com> wrote:

On Fri, 16 Oct 2009 12:10:35 +0200, Michael A. Puls II <shadow2...@gmail.com> wrote:

On Fri, 16 Oct 2009 05:28:46 -0400, Ian Hickson <i...@hixie.ch> wrote:

There was also some discussion of what to do about preventing a plugin
instantiating. It seems to me that authors can do that by not creating the
<object> element ahead of time.

And, if it's desired to specify the <object> via parsed markup (as opposed to doing it all with JS), one can omit @type and @data so things don't load and add them later like so:

<object style="display: none" data-load-on-demand-type="application/x-java-applet" id="test">
     <param name="code" value="MyJavaClass">
     fallback
</object>

<script>
     window.onload = function() {
         var obj = document.getElementById("test");
         obj.style.display = "inline-block";
         obj.type = obj.dataset["load-on-demand-type"];
         alert("Come alive! Hide your fallback! I command you!");
     };
</script>

"One or both of the data and type attributes must be present." says the spec.

Oops. Right. Forgot about that.

<embed> doesn't seem to have the same requirement for src and type. (Also compare with img, iframe, video...)



--
Michael

Reply via email to