From kde-i18n-doc  Thu Aug 18 18:26:26 2005
From: Nicolas Goutte <nicolasg () snafu ! de>
Date: Thu, 18 Aug 2005 18:26:26 +0000
To: kde-i18n-doc
Subject: Re: KGeography needs your help
Message-Id: <200508182026.27746.nicolasg () snafu ! de>
X-MARC-Message: https://marc.info/?l=kde-i18n-doc&m=112438969919965

On Thursday 18 August 2005 18:53, Chusslove Illich wrote:
> > [: Nicolas Goutte :]
> > Gettext remains the reference implementation, which continues to be
> > developed, while KDE's support remains at an old level.
> > [...]
> > Personally I do not find this situation satisfactory.
>
> I agree.
>
> > I do not tell that everything has to be done now. Even it cannot be
> > done now, especially as there would be a need of the Gettext plural
> > support in KBabel.
>
> Err, doesn't KBabel support Gettext plurals already? At least when I open a
> PO file with Gettext plurals, it shows plural forms using tabs. Or perhaps
> this feature is still not working well throughout?

As far as I know the support is only partial, especially there is no support 
at save.

>
(...)
>
> > I do not think that completely freeing programmers from thinking about
> > plurals is such a good idea. Less work for the developer could well
> > mean more work for the translators. A balance would be better I think.
>
> Here I also see a possible trouble. Currently, if programmer forgets to use
> a plural handling variant of i18n, some of the translation teams is bound
> to make him aware of that. Then, *all* other teams will be forced to
> acknowledge this, for automatic tools will complain if some translator
> just neglects plurals. In this other approach, there would be no such
> forcing measure, now allowing *translators* to forget about plurals :)
>

> Krzysztof did talk about interface spewing warnings to log file when
> translators don't use plurals, but using this as correction tool seems a
> lot more harder than having automatic tools, like check_po_files and such.

The advantage is that check_po_files does not need to be compiled to work. If 
a test needs KDE to be compiled, it will be harder to test.

>
> There is another point too. It would be nice to settle on Gettext plurals,
> so that translators and translation tools don't have to think about
> particular "dialects" of PO files. 

Well probably not everything will be the same. But the point is valid for 
KBabel too, which needs to support the KDE version and which should at least 
support the version described in Gettext's documentation.

>(On a side note, why wasn't Gettext's
> plural handling used in the first place, it seems to me there are no
> technical reasons?) The lack of functionality could be supplemented with
> scripting (or, with future extensions to Gettext, hmm...)

I had thought that it was a historic reason. However I have never really 
checked if Gettext had or had not plural forms when KDE started to use them.
(In any case, Qt had no translation scheme, that is why it is Gettext and not 
Qt's.)

>
> Having said that, using Gettext plurals doesn't prevent us from *scripting*
> additional plural forms (or whatever else within plural message). With
> Gettext plurals, the script would simply be attached to the msgstr that
> one gets when number is 99 :) (checking the conditions in KDE's plural
> implementation, that would be the last msgstr for any language.)
(...)

Have a nice day!