Re: [css-counter-styles] implementation of complex cjk counter styles

On Mon, Nov 4, 2013 at 1:58 PM, Tab Atkins Jr. <jackalmage@gmail.com> wrote:
> On Sun, Nov 3, 2013 at 7:35 PM, Xidorn Quan <quanxunzhen@gmail.com> wrote:
>> I just submitted an implementation of longhand East Asian counter styles for
>> Firefox. You can find it at
>> https://bugzilla.mozilla.org/show_bug.cgi?id=934072 .
>>
>> As described in comment 8 in the page mentioned above, this impl generates a
>> slightly different result with the current draft: for 11,111, it generates
>> "一万千百十一" in japanese-informal and "萬 千百十一" in korean-hanja-informal; and it
>> generates "一千万" for 10,000,000 in japanese-informal. These modifications are
>> based on the discussion in this mailing list and replies from some of my
>> native friends, and I also referred to the result of Google Translate.
>
> Can you please describe what these changes are in terms of the
> algorithms in the spec?

Xidorn replied privately (possibly accidentally), so here's his
feedback publicly:

> I reviewed the answer from my Korean-native friend, and I believe that
> the "drop ones" rule for Korean informal style should be changed.
>
> The second term of "drop ones"
>
>> For the Japanese informal and Korean informal styles, if any of the digit markers
>> are preceded by the digit 1, and that digit is not the first digit of the group, remove
>> the digit (leave the digit marker).
>
> is correct for Japanese informal style, but for Korean informal
> styles, even if the digit is the first digit of the group, it could be
> dropped.
>
> Also result from Google search in Korean: for 1,1111, 만 천백십일 (萬 千百十一)
> has about 10,700 result, but 만 일천백십일 (萬 一千百十一), which is generated
> from the current rule, has only 2 results.

Jungshik, opinions?

~TJ

Received on Thursday, 7 November 2013 19:13:24 UTC