RE: [cssom] unrecognized rules and properties

Anne van Kesteren wrote:
> On Tue, 24 Nov 2009 09:43:24 +0100, Mike Wilson wrote:
> > One may choose to apply grammars at different levels of
> > detail, so I'm not entirely sure what is the specific problem.
> > Have there been changes in these areas since the last CSSOM
> > spec? (sorry if this is obvious, haven't been following)
> 
> To answer this question specifically. The last CSSOM draft 
> defined almost nothing in detail. Especially when it comes 
> to string representation of CSS constructs. All it said 
> about CSSUnknownRule was "The CSSUnknownRule interface 
> represents an at-rule not supported by this user agent."  
> Considering that the CSS parsing specification dicates 
> throwing away everything you do not recognize without 
> defining implementation details as to what should be thrown 
> away etc. this is highly unuseful.

Right. I still believe a case could be made that it is fairly
straight-forward to map that specific case literally into 
implementations (at-rules that only differ from established
syntax in the at-identifier). Though, considering the WG's
statement from 2003 that you linked to:
http://lists.w3.org/Archives/Public/www-style/2003Oct/0347.html
(where CSSUnknownRule is "deprecated"), I guess this form of
unrecognized rules access is out of the picture. Or did anyone
implement it?

So, regarding unrecognized rules, I'll just finish with 
encouraging the WG to work on the "Replacements will be 
provided at some future point" mentioned in 2003.

> You mention preserving unrecognized properties as well and 
> presumably other constructs than unrecognized at-rules too? 

Preserving unrecognized properties is the killer app at the
moment I think (with so much happening in CSS). As it is 
implemented in all relevant IE versions also makes it part of 
the web.

My primary use case for this (as I mentioned previously) is
being able to use script to implement new CSS features in old
browsers, to ease the transition to new standards. Imagine
libraries like Dean Edwards' IE7.js/IE8.js, but targeted at
more browsers.

It seems to me that retaining unrecognized properties is quite 
trivial, as long as they use existing formats. Or are there
any particular problems in this area? (grammar-wise or other)

Mike Wilson

Received on Monday, 7 December 2009 11:14:06 UTC