- From: Ian Hickson <ian@hixie.ch>
- Date: Wed, 14 Jan 2009 21:07:53 +0000 (UTC)
- To: Travis Leithead <Travis.Leithead@microsoft.com>
- Cc: "public-html@w3.org" <public-html@w3.org>
On Wed, 14 Jan 2009, Travis Leithead wrote:
>
> <style>
> P { border: red solid 1px; }
> P P { border: green solid 1px;}
> </style>
> <p>A paragraph
> <object>
> <p>innerP
> </object>
>
> Issue: P nested inside another P is incorrect.
The HTML parsing rules don't prevent illegal DOMs from being created,
correct. Other examples that generate illegal DOMs:
Bogus content in a table:
<table><input type=hidden><tbody><tr><td>...</table>
P inside SPAN:
<span><p>...
All of the above are also mising the <!DOCTYPE> and the <title> element.
> I thought I read somewhere on this mailing list that per HTML5 we're
> supposed to ignore/treat-as-opaque fallback content from an Object. Can
> someone point out where that might be specified?
I'm not sure what you mean. Scripts, form controls, etc, inside an
<object> continue to work as normal; regardless of whether the fallback is
shown or not. The fallback being shown or not is purely a CSS issue
really. The only things that the spec does or will say about fallback as
far as I am aware is that plugins on elements inside <object>s that aren't
showing their fallback should not themselves be instantiated. (This is
already in the spec for nested <object>s, it's the first step of the
<object> algorithm. I still need to explicitly say it for <applet>-in-
<object> and <embed>-in-<object>, that's bug 6434.)
HTH,
--
Ian Hickson U+1047E )\._.,--....,'``. fL
http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,.
Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'
Received on Wednesday, 14 January 2009 21:08:34 UTC