W3C home > Mailing lists > Public > www-style@w3.org > February 2014

Re: [selectors4][css-syntax] Pseudo-elements vs. combinators

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Wed, 5 Feb 2014 17:39:05 -0800
Message-ID: <CAAWBYDDGH8UOrTcPr+bZgBcKxvtnbBVkwc8X+4Bq+eceYhV=1w@mail.gmail.com>
To: Peter Moulder <pjrm@mail.internode.on.net>
Cc: www-style list <www-style@w3.org>
On Wed, Feb 5, 2014 at 5:04 PM, Peter Moulder
<pjrm@mail.internode.on.net> wrote:
> On Wed, Feb 05, 2014 at 08:04:46AM -0800, Tab Atkins Jr. wrote:
>> On Wed, Feb 5, 2014 at 7:52 AM, Brian Kardell <bkardell@gmail.com> wrote:
>> > Crap... Yeah i don't know why i said  tilde i meant to say ` ... TIL Don't
>> > mail the list before you finish your first cup of coffee I guess.
>>
>> Heh, no problem.  Personally, I don't like backtick because it looks
>> like a quote character.
>
> It could actually be used as a quote-like character, as in
> dt `next` dd { break-before: avoid }.
>
> This would match at least one programming language (viz. Mercury) that uses
> paired backticks to allow user-defined infix operators (like A `union` B).

Also: Haskell.  Alphanumeric function names can be wrapped in ` to
turn them into infix operators.

That's... not a bad idea.  Definitely in the running for me, alongside ^foo.

> Is there any support for  dt/following-sibling::dd { ... } ?
>
> Even if the ‘::’ part is considered too reminiscent of pseudo-elements,

So that's the XPath syntax.  I think it's a no-go in CSS, due to it
looking like it's referring to a pseudo-element named ::dd.

> I think
> we should consider
>   dt /following-sibling dd { ... }
>
> (while keeping in mind Tab's comment
>
>> Slashes were used for the ref combinator, though we're punting that
>> and might not do it at all.
>
> ).

Hmm, let's see:

article ^shadow heading ^descendants p

Than in:

article /shadow heading /descendants p

Damn, that's pretty good, actually.  I thought I would like it less
due to it being less visible, but it's really not bad.

> And even if we reject slash, we should consider using the same names as in
> xpath wherever possible (e.g. ^following-sibling or whatever), so long as
> the xpath combinator doesn't have some difference in semantics sufficient to
> cause "false friend" problems.

Yeah, I've got no problem with that, assuming we introduce variants of
the existing combinators.

~TJ
Received on Thursday, 6 February 2014 01:40:00 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:39:18 UTC