On Tue, Oct 7, 2008 at 7:23 PM, fantasai <fantasai.lists@inkedblade.net>wrote: > > Matt Patenaude wrote: > >> >> Hey Keiji, >> >> On Oct 6, 2008, at 11:56 PM, Keiji Ikari wrote: >> >> Instead, I'd suggest something like this: >>> >>> div.(product, service):first-child > h3 + p >>> >>> So each string (not necessarily a class name) separated by commas in >>> the parentheses would be "expanded out" like mathematical brackets and >>> would be changed by the parser to >>> >>> div.product:first-child > h3 + p, div.service:first-child > h3 + p >>> >>> before being applied. >>> >> >> I actually thought of something like this as well, and I would probably >> use this syntax over the one I proposed for new designs. However, I think >> this only has half of the capability of the alias syntax I proposed, and >> still presents a lot of search-and-replace work for modifying existing code. >> >> Ideally I'd like to see a little of both. :) >> > > Something like > div:matches(.product, .service):first-child > h3 + p > has been proposed before. It is purely syntactic sugar, so should be > very straightforward to implement, and I would support putting it into > Selectors 4. > > The sort of "Aliases" you suggest, not so simple. If the above solves > your use case, I'd much rather we concentrate on that. :) I like it, too. I've had several cases where some simple sugar like that would have saved me from typing a *long* run of selectors. ~TJ
Received on Wednesday, 8 October 2008 13:15:57 UTC