Re: [counter-styles] i18n-ISSUE-285: Hebrew number converter inadequate for numbers >= 1000

I was making the necessary changes to my tests and the Predefined 
Counter Styles WD when it occurred to me that we are making a mistake 
here to make the 'longer-hebrew' style described below an alternative.

Firefox, Chrome, Safari and Opera all implement hebrew numbering already 
per the longer-hebrew style.  IE and old-Opera don't implement hebrew 
numbering at all.

Run the test here: 
http://www.w3.org/International/tests/repository/run?manifest=predefined-counter-styles&test=list-style-type-116a

See the results here: 
http://www.w3.org/International/tests/repository/predefined-counter-styles/results/results-predefined-counter-styles#hebrew

So why not make the definition of hebrew in the spec be the definition 
provided for longer-hebrew below, and possibly keep the other hanging 
around as the alternative?

If we don't, I doubt that hebrew will get through CR. If we do, it will 
sail through, and if people really want the verbose version that only 
goes up to 2000, they can use a definition in the Predefined Counter 
Styles doc (though I'm not sure what I'd call it).

RI



On 01/11/2013 23:27, Tab Atkins Jr. wrote:
> On Thu, Oct 31, 2013 at 5:59 AM, Richard Ishida <ishida@w3.org> wrote:
>> On 30/10/2013 18:03, Tab Atkins Jr. wrote:
>>> We could also add alternative hebrew numbering to the i18n doc, using
>>> geresh or not.  The whole point of the i18n doc is to be a repository
>>> of bunches of other styles that authors can use.  I'd be happy to add
>>> a note after the 'hebrew' style specifically pointing out the
>>> alternate forms in the i18n doc.
>>>
>>> I'll go ahead and edit hebrew to stop at 2000. Let me know if I should
>>> do anything else.
>>
>> I'd be happy to add an alternative style to the i18n predefined list, but
>> I'm a bit pressed for time just now, and I'm thinking that you might be able
>> to drum up the necessary details much faster than me anyway, Tab. Would you
>> be up for providing me with the @counter-style rule?
>
> Sure!  This one goes up to 10999, and uses the assumptions that
> numbers higher than 1k are indicated by placing a geresh after the <1k
> part, then writing the rest of the number with the same digits as
> before, but *1000.  (In other words, it's what Mati described.)
>
> @counter-style longer-hebrew {
>    system: additive;
>    range: 1 10999;
>    additive-symbols: 10000 \5D9\5F3, 9000 \5D8\5F3, 8000 \5D7\5F3, 7000
> \5D6\5F3, 6000 \5D5\5F3, 5000 \5D4\5F3, 4000 \5D3\5F3, 3000 \5D2\5F3,
> 2000 \5D1\5F3, 1000 \5D0\5F3, 400 \5EA, 300 \5E9, 200 \5E8, 100 \5E7,
> 90 \5E6, 80 \5E4, 70 \5E2, 60 \5E1, 50 \5E0, 40 \5DE, 30 \5DC, 20
> \5DB, 19 \5D9\5D8, 18 \5D9\5D7, 17 \5D9\5D6, 16 \5D8\5D6, 15 \5D8\5D5,
> 10 \5D9, 9 \5D8, 8 \5D7, 7 \5D6, 6 \5D5, 5 \5D4, 4 \5D3, 3 \5D2, 2
> \5D1, 1 \5D0;
> }
>
> ~TJ
>

Received on Thursday, 6 February 2014 18:13:56 UTC