[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-i18n-doc
Subject:    Re: Different formulas for plurals in Qt and KDE
From:       "Yuri Chornoivan" <yurchor () ukr ! net>
Date:       2014-04-08 4:52:08
Message-ID: op.xdzeg6inl2zvei () ip-4ffd ! proline ! net ! ua
[Download RAW message or body]

Hi,

Please read some answers below.

=D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BD=D0=B5 Tue, 08 Apr 2014 04:07:=
31 +0300, Alexander Potashev  =

<aspotashev@gmail.com>:

> Good night!
>
> More on plurals and KF5...
>
> As you may know, Serbian, Ukrainian and Russian translations of KDE
> software are using a new plural formula with 4 plural forms, while Qt
> knows only about the traditional plural formula and only 3 plural
> forms.
>
> Because, like I was said 4 years ago, scripty pulls the plural formula=

> from kdelibs4.po and scatters it to all files in l10n-kde4/[LANG]/, it=

> is not possible to have some .po with 4 plural forms (for those
> packages using the ki18n framework) and others with 3 plural forms
> (for packages using QObject::tr).
>
> Now I have a two questions:
>  1. Is there any problem at all? For example, what happens if a .po
> file for a Qt-only KF5 module is converted back to .ts - will it
> truncate the 4th plural form?

Guessing from Trojita ans Kst translations, lconvert just skip the whole=
  =

translation and left the messages untranslated.

The error message is:

"Removed plural forms as the target language has less forms.
If this sounds wrong, possibly the target language is not set or  =

recognized."

>  2. In the case there is a problem, what options may there be to  =

> overcome it?

It thought that some addition of sed magic to lconvert wrapper may help.=


> P.S.  I didn't read Qt sources much, therefore I might be wrong in my
> assumption that plural formulas for languages are hardcoded in Qt.

Yes, they are hardcoded.

Best regards,
Yuri
[prev in list] [next in list] [prev in thread] [next in thread]