Re: @font-feature-values rule syntax

On Mon, 15 Nov 2010 01:49:39 +0100, John Daggett <jdaggett@mozilla.com>  
wrote:
> During the CSS WG discussion at TPAC, the topic of the syntax of
> the @font-feature-values rule was discussed (see original proposal
> here). This rule has been proposed as a way to deal with the
> problem of font-specific numbers used in conjuction with some values
> of font-variant (e.g. styleset, swash).
>
> A example showing various aspects of the existing proposed syntax:
>
>   @font-feature-values Jupiter Sans {
>     swash: swishy 1, flowing 2;
>     stylistic: long-k 2;
>     styleset: alt-g 1, alt-m 3;
>     styleset: curly-quotes 5, code-forms 4 7 9;  /* additional values */
>   }
>
>   body { font-family: Jupiter Sans, GreatJapaneseFont, sans-serif; }
>   h2 { font-variant-alternates: styleset(alt-m, curly-quotes); }
>   h2:first-letter { font-variant-alternates: swash(flowing); }
>
>   code { font-variant-alternates: styleset(code-forms); }
>   p { font-family: Another Lovely Font; }
>
> Here the @font-feature-values rule defines values for font-specific
> alternates of a single font.  When fallback occurs those values are
> ignored unless the same value name has also been defined for the
> fallback fonts.

Why is it important the the second styleset declaration is separate and  
not just merged? I.e. why does

   styleset: alt-g 1, alt-m 3, curly-quotes 5, code-forms 4 7 9;

not work? Or

   styleset: alt-g 1, alt-m 3 / curly-quotes 5, code-forms 4 7 9;

or something similar?


The minutes say the problem is local style sheets wanting to tweak various  
settings, but the example does not illustrate that. The minutes also  
mention a more constrained proposal by Bert, that was not dismissed, and  
seems simpler for authors, but not presented as an option here. Or maybe I  
am missing something?


-- 
Anne van Kesteren
http://annevankesteren.nl/

Received on Monday, 15 November 2010 10:54:26 UTC