[whatwg] <object> behavior

[Ian Hickson]

On Fri, 14 Aug 2009, Andrew Oakley wrote:
> >> - Should the type attribute take precedence over the Content-Type header?
> > 
> > No, I believe what the spec says here is the preferred behaviour. 
> > Unless this is incompatible with legacy content, we should try to move 
> > towards this behaviour.
> 
> OK, should the spec mention the change that Boris Zbarsky said had been 
> made for Gecko?  I guess this use makes up the majority of uses of 
> <object> on the web (although the Content-Type header should be set to 
> the same thing so it wouldn't normally matter).  Presumably this would 
> also change the rules for when to instantiate a new plugin, given that 
> the current rules are essentially "whenever a different plugin or data 
> would be loaded".

I'm not sure exactly what change you mean. The spec already has some of 
Gecko's behaviour (in particular the special-casing of certain MIME types 
to enable sniffing), are there other changes you think we should include? 


> I'm also confused about the criteria for creating a nested browsing context.
> 
> > - If the resource type is an XML MIME type
> > - If the resource type is HTML
> > - If the resource type does not start with "image/"
> 
> Does that mean:
> - xml_mime_type OR html OR not_image
>   this is the same as xml_mime_type OR not_image
> - xml_mime_type AND html AND not_image
>   this would be application/xhtml+xml, application/ce-html+xml etc. but
>   *not* text/html
> - (xml_mime_type OR html) AND not_image
>   makes sense but would certainly need clarification in the spec

The former. I've added an "or" and removed the (redundant) HTML line to 
make this clearer.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'