[whatwg] Adoption Agency Algorithm

Fri Jan 27 15:47:57 PST 2006

Ian Hickson wrote:
> On Sat, 28 Jan 2006, Lachlan Hunt wrote:
>> Why can't it just be defined that noframes and noscript content gets 
>> parsed exactly as regular markup
> 
> Because there are a _lot_ of side-effects of parsing as regular markup.
> 
> e.g.
> 
>    <noframes>
>      <style>...</style>
>      <i> Foo
>    </noframes>
>    Bar
> 
> ...the "Bar" musn't be in italics, the <style> block musn't be used, etc. 
> There's really a LOT of things to catch.

In a frameset document, where all the content is displayed within frames
using external documents, when would "Bar" ever be displayed anyway
(unless the UA doesn't support frames and was rendering the noframes
content).  In which case, why does it matter whether or not "Bar" ends
up within the I element or not?  It should be treated exactly the same
way as this would be for a non-frameset document:

<body>
   <i>Foo
</body>
Bar

Besides, if that was a real problem, could it not be defined as a
special case so that upon encountering </noframes>, all unclosed child
elements are then closed and not reopened?

I'm not sure what the style element in your example is supposed to show,
but I'm sure it could be ignored just like scripts, although it would
have no effect on the documents within the frames anyway.

-- 
Lachlan Hunt
http://lachy.id.au/