Just to make sure,
We are requesting to add the rule for dual form.. the other rule can be ignored.

0 : n==1 : singular form
1 : n==0||n >2 :plural form
2 : n==2 : dual form

We do not actually need the many form for the translation to be correct.

(Assuming putting dual form on 2 can mainyain some bacward compatibility)

Thanks,
Niv


בתאריך 14 בספט׳ 2016 09:52,‏ "ניב באר" <bloop93@gmail.com> כתב:
>Thanks for the suggestion, take a quick look at the cardinal part, it shows exactly what we're talking about so CLDR has this rule already :)
I guess you are referring to the dual rule..
the rule for the many form there is incorrect[1] and should be fixed (at CLDR at least, we do not really need it for translations)

[1]http://hebrew-academy.org.il/topic/hahlatot/grammardecisions/terminology-ordinance/4-3-%d7%94%d7%a9%d7%99%d7%9e%d7%95%d7%a9-%d7%91%d7%a9%d7%9d-%d7%94%d7%9e%d7%a1%d7%a4%d7%a8/#target-3363
(it's in hebrew, sorry)

2016-09-14 9:37 GMT+03:00 Yaron Shahrabani <sh.yaron@gmail.com>:
Hi Yuri!
Thanks for the suggestion, take a quick look at the cardinal part, it shows exactly what we're talking about so CLDR has this rule already :)

Yaron Shahrabani
<DevOps - Hebrew translator>

On Wed, Sep 14, 2016 at 9:05 AM, Yuri Chornoivan <yurchor@ukr.net> wrote:
середа, 14-вер-2016 08:56:32 ניב באר написано:
> Thank you,
> forth form isn't quiet correct
>
> The actual hebrew plural forms are the following:
>
> one object - use singular form
>
> two objects - use dual form for specific cases, otherwise use plural form
>
> 3-10 objects - use plural form
>
> none or more than 10 objects - can use plural form or many form (optional)
>
>
> also, with both singular and dual form the number should not be written
> (anaconda, for example, does not aloow it).
>
>
> generally, the many form isn't required but it could be nice to have (and
> prefered when speaking of years),
>
> it is not widely use and we can do without him (using the regular plural
> form will sound natural as well).
>
>
> In most cases, the existing plural form can be copied into the other form
> (or two).
>
>
> I will also try to see if I can find a source in english who can support it
> (I could only find in hebrew and not all in one place).
>
>
> I can review the translations on Zanata and I will try to contact hebrew
> translators of another projects
>
>
> Thanks again,
> Niv

Hi,

Just 2 cents+

It is worth to inspect (and change if needed) the plural rules on CLDR.

http://www.unicode.org/cldr/charts/29/supplemental/language_plural_rules.html#he

These rules are widely used in hardcoding the plural forms for many projects
including FOSS.

Hope this helps.

Best regards,
Yuri
>
> 2016-09-14 2:41 GMT+03:00 Rafal Luzynski <digitalfreak@lingonborough.com>:
> > 13.09.2016 10:06 ניב באר <bloop93@gmail.com> wrote:
> > >  Hello,
> > >  The plural form for hebrew is currently set as:
> > >  nplurals=2; plural=(n != 1);
> > >
> > >  Altough plurals form fits in most cases,
> > >  we do have dual form in hebrew which is mostly used when speaking about
> > >
> > > time[1].
> > >
> > >  which makes the use of plural form in those cases seems unnatural.
> > >
> > >  I would like to update hebrew plural form to be:
> > >  nplurals=3; plural=(n==1) ? 0 : 1 + (n==2);
> >
> > This formula looks tricky and I can't confirm it will always work
> > as expected. Also it must be less obfuscated for other human readers.
> >
> > I have grepped over the source code of those several projects
> > from GNOME whose source code I apparently have on my computer
> > and all of them use the simple nplurals=2 system in Hebrew
> >
> > except GNOME Software [2] which uses this expression:
> >     nplurals=3; plural=n==1 ? 0 : (n>2||n==0) ? 1 : 2;
> >
> > Our reference source should be CLDR [3] which mentions
> > four forms for Hebrew:
> >
> > - one (n==1);
> > - two (n==2);
> > - many: dividable by 10 except 10 itself (n%10==0 && n>10)
> > - other (including 0 and 10).
> >
> > Can you confirm you need this fourth form for the numbers
> > dividable by 10?
> >
> > Otherwise, it seems you have spotted a bug in Hebrew language
> > configuration. IMHO the plural rules for Hebrew should be
> > changed to reflect all possible cases and you will have to live
> > with the fact that in most cases two or three forms will be
> > identical.
> >
> > >  or at least to be able to specify a custom plural form to a project on
> > >
> > > Zanata.
> >
> > I don't know if it's possible to change the plural rules for
> > a single project but indeed it should be acceptable if the
> > problem is so rare that it actually appears only in few projects.
> > Otherwise it should be changed globally, for all projects
> > in Zanata. Of course, it will break your existing translations
> > so you will have to review them all. Sounds overwhelming but
> > the satisfaction when you see all plural forms displayed correctly
> > in your computer is priceless.
> >
> > Also note that I know nothing about Hebrew language so I can't
> > confirm what plural rules are correct and whether they should be
> > applied globally or exceptionally for just few projects. I think
> > these rules should be confirmed by other Hebrew speaking people
> > and by references to some external and reliable sources like
> > CLDR or ISO documents or a national language regulator.
> >
> > Best regards,
> >
> > Rafal
> >
> > >  Thanks,
> > >  Niv Baehr
> > >
> > >
> > >  [1]https://en.wikipedia.org/wiki/Dual_(grammatical_number)
> >
> > #Modern_Hebrew
> >
> > [2] https://git.gnome.org/browse/gnome-software/tree/po/he.po
> > [3]
> > http://www.unicode.org/cldr/charts/29/supplemental/
> > language_plural_rules.html


--
trans mailing list
trans@lists.fedoraproject.org
https://lists.fedoraproject.org/admin/lists/trans@lists.fedoraproject.org