- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Wed, 23 Nov 2011 10:36:05 -0800
- To: Håkon Wium Lie <howcome@opera.com>
- Cc: Peter Moulder <peter.moulder@monash.edu>, www-style@w3.org
On Wed, Nov 23, 2011 at 9:48 AM, Håkon Wium Lie <howcome@opera.com> wrote: > Also sprach Tab Atkins Jr.: > > > Do we really need the longhand? Why not just have a space-separated > > > list? > > > > We need the longhand because images are allowed as glyphs. > > Right. How common will it be to combine images and glyphs in the same > list style? One indication that it may not be too common is that there > are no examples of this use in the spec. Not very common, I would think. The spec used to contain an example of mixing an image and text in the *same* marker (I ended up editting it out), but that's actually impossible in the current syntax (the example used the 'content' property to do it). > It seems that all the examples in the spec that use strings can be > rewritten in the shorthand syntax. > > So, I suggest adding the space-separated shorthand syntax and rewrite > examples to use it. I'm still concerned about this for two reasons. The first is just theoretical - CSS currently has no other example of a string containing complex data like this. We always use lists of tokens, and let strings exist as a single opaque unit. The only reason we're trying to use a string here is that the normal tokenization rules don't like the sorts of things we might provide. The second is practical - the space-separated syntax prevents an author from ever having a marker with a space in it. Daniel provided an example of this in the "Mongolian and French lists" thread. What if we used a function as the quotes instead? @counter-style lower-norwegian { type: alphabetic; glyphs: glyphs(a b c d e f g h i j k l m n o p q r s t u v w x y z æ ø å); } We'd tokenize the contents of the glyphs() function by spaces, and only require that the various brace characters ()[]{} be escaped within it. ~TJ
Received on Wednesday, 23 November 2011 18:36:53 UTC