[css3-conditional] conditionText serialization [WAS]: Review of the new grammar for @supports

On Thu, 07 Feb 2013 23:30:08 +0100, L. David Baron <dbaron@dbaron.org>  
wrote:

>> * "The conditionText attribute, on getting, must return the result of
>> serializing the associated condition."
>> We had a discussion at the San Diego face to face about what the
>> serialization of the supports_rule should look like, but I cannot find
>> the conclusion we had reached in this draft. [...]
>
> Ah, http://lists.w3.org/Archives/Public/www-style/2012Aug/0749.html
> records that:
>   #  - RESOLVED: .conditionText returns either the token stream or  
> source text
>   #              from the style rule, with no logical simplifications,  
> just
>   #              tokenization ones
>
> I've implemented this resolution [...] in
> http://dev.w3.org/csswg/css3-conditional/#the-csssupportsrule-interface
> https://dvcs.w3.org/hg/csswg/rev/7dace721dff3
>
> Does that seem ok to you?

Yes, this accurately reflects is what we agreed to.

That said, for the benefit of the reader who is reading this without  
knowledge of the discussion we had, I wonder if it would be worth being  
explicit about the fact that collapsing multiple levels of nested  
parentheses to a single one is one of these logical simplifications we  
forbid. Your text already
implies it, but it might also be misinterpreted to only mean forbidding  
converting "(not (a:1)) and (not (b:2))" into "not ((a:1) or (b:2))".

  - Florian

Received on Thursday, 7 February 2013 22:51:57 UTC