Re: gettext "plural forms" in Turkish

[Sertaç Ö . Yıldız]

* Bruno Haible [03.Ara.07 12:42 +0100]:
> Sertaç Ö. Yıldız wrote on 2007-11-26:
> >     msgid "the apple"
> >     msgid_plural "the apples"
> 
> This is not the major usecase of the plural forms in gettext. In fact,
> I'm not aware of any packages using the singular/plural distinction without
> mentioning the number. But...

This kind of usage is documented in section 11.2.6, under ngettext. And
I know that at least seahorse¹ uses like so:–

559    const gchar *title = ngettext("Key Imported", "Keys Imported", keynum);

    [On using “Plural-Forms: nplurals=2; plural=n != 1;”]
> Do you agree to this doc change, and to changing the default plural
> formula for Turkish (as used by 'msginit')?

Yes, that’s better. But [if you won’t call me anal] …

Like in French, we actually use singular for n=0 in Turkish. That was
why I initially proposed “plural=(n>1)”. I’m told that Hungarian also
uses singular for n=0 :–

    “0 apples”  → “0 alma”
    “no apples” → “nincs alma”

As far as I can see, using (n>1) or (n!=1) in the formula will not
matter in practice. But just to be precise, how about moving paragraphs
about both Turkish and Hungarian below French and using (n>1)?

¹ 
http://svn.gnome.org/viewvc/seahorse/trunk/libseahorse/seahorse-notification.c?revision=1812&view=markup

-- 
~sertaç