Re: [selectors] :sorted pseudoclass for sorted tables

On Wed, Dec 11, 2013 at 12:03 PM, Linss, Peter <peter.linss@hp.com> wrote:
> On Dec 2, 2013, at 1:21 PM, Tab Atkins Jr. wrote:
>> On Sat, Nov 30, 2013 at 11:44 PM, Daniel Glazman
>> <daniel.glazman@disruptive-innovations.com> wrote:
>>> On 01/12/13 06:33, Peter Linss wrote:
>>>> The number isn’t for numeric vs alphabetic sorting, it’s the ordinality of the sort key, as in, sort by this column first, then this second. When multiple sort keys are being used, it’ll likely be useful to style them so that the ordinality is visible.
>>>
>>> That integer could be extended to ranges:
>>>   1
>>>   -3
>>>   4-8
>>>   9-
>>
>> That's just an an+b term.  (And we can mix an+b with other grammars now, yay!)
>
> Actually, no. I don't think what Daniel is describing is an+b term, but a actual range of indices (as in up to and including the 3rd item, or the 4th through 8th items, or the 9th item and above). If we add a range (which could be quite useful) then we should add it everywhere an+b is allowed (which might be an extension of an+b).

Ah, yeah, to do a closed range with an+b you have to combine two of
them together in a slightly clumsy way; to do 4-8 you have to pair up
(n+4) and (-n+8), which is weird.

I think I'd be okay with extending an+b to accept straight numeric ranges.

~TJ

Received on Wednesday, 11 December 2013 03:06:44 UTC