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

Re: [css-values] The last grammar combinator - 1+ in order

From: Dirk Schulze <dschulze@adobe.com>
Date: Thu, 24 Apr 2014 06:54:11 +0000
To: Tab Atkins Jr. <jackalmage@gmail.com>
CC: www-style list <www-style@w3.org>
Message-ID: <EB33B763-2393-4721-947C-1D947718C4FA@adobe.com>

> On Apr 24, 2014, at 4:20 AM, "Tab Atkins Jr." <jackalmage@gmail.com> wrote:
> 
>> On Thu, Apr 17, 2014 at 5:27 PM, Tab Atkins Jr. <jackalmage@gmail.com> wrote:
>> We're currently missing a grammar combinator in CSS's property grammar syntax.
>> 
>> We can express "exactly one" as "a | b", and we can express 5 of the 6
>> possibilities in {0+, 1+, all}{in order, any order}* .  The one thing
>> we're missing is "1+, in order", which we can't write without
>> duplication and honestly confusing grammar.
> 
> More syntax possibilities:
> 
> * A // combinator - it looks similar to ||, which is good, because
> they both occupy the "1+" row in the table.

This could be confused with || if a spec uses an italic font style.

Greetings
Dirk

> 
> * A way of indicating that something is "required", which means the
> term inside can't disappear to nothingness.  I thought of using ! as a
> counterpoint to ?, like "[A? B?]!".  fantasai indicated that she
> doesn't like this, because it looks like a multiplier.
> 
> One wrinkle is that a combinator is not compatible with
> comma-separation (or in general, any non-whitespace separator), which
> makes it complicated to use with several of the things that we already
> know we want to use it on, like image() and <picture>'s "sizes"
> attribute.  zcorpan suggested that we could amend the existing
> multi-choice combinators to allow a comma between the two characters,
> to indicate that the choices are comma-separated, like "A |,| B |,| C"
> allowing "A" or "A, C", or "C, A, B", etc.
> 
> Related to this wrinkle, the existing syntax for "0+, in order", which
> is "A? B?", also isn't compatible with non-ws separators - you need a
> comma if you include both, but not if you omit either.  I think we can
> solve this without any new syntax, by just declaring that commas in
> the grammar always get omitted if they're not separating anything.  I
> think this would let us simplify a few grammars as well, where we're
> currently required to do some complicated fiddling with brackets due
> to the presence of commas.
> 
> ~TJ
> 
Received on Thursday, 24 April 2014 06:54:41 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:39:21 UTC