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

On Tue, Aug 20, 2013 at 11:27 AM, Richard Ishida <> wrote:
> Raised by:
>     Matitiahu Allouche
> Opened on:
>     2013-08-04
> Description:
>     This comment relates to
> section 6.1 for
> Hebrew, and also to
> section 12
> Hebrew which is a duplicate of the former.
>     In the text, the additive symbols max up at 400. For numbers larger than
> 799, the symbol for 400 will appear several times. For numbers like the
> Hebrew year (currently 5773), this gives 14 such symbols, which is hardly
> readable and does not conform to common usage.
>     The better usage is to have the number of thousands followed by Geresh
> (\5F3) followed by the rest of the number.
>     For 5773, this would give \5D4 \5F3 \5EA \5E9 \5E2 \5D2.
>     Note that there is no space between the thousands and the rest of the
> number.
>     For 12345, we would have \5D9 \5D1 \5F3 \5E9 \5DE \5D4.

The Hebrew system was originally defined this way, as a custom system,
as it can't be reasonably represented in the current algorithms the
way you describe.  I was told by another Hebrew speaker (Aryeh Gregor)
that the form the spec currently specifies, with repeated tavs, is

I could do the geresh-based system up to 10999 without too much
trouble (I just include the geresh in the "digit" for the thousands or
10k digit), but going past that becomes untenable (it requires me to
define digits for 11k, 12k, etc.).  To handle this, I'd need to add a
new feature for group separators, which isn't currently required for
any other language (though it might be nice for some, like English

I'm somewhat loathe to grow Counter Style's feature set until I get
some implementor indication that the current feature set is planning
to be implemented, but I can poll the WG to see if they want to make
this final addition.

(If we ever need to go any farther than this feature, I'm strongly in
favor of just punting to the already-defined systems for doing
arbitrary number formatting.)


Received on Wednesday, 30 October 2013 00:41:54 UTC