Re: DON'T update GTK+ po-properties translations

[Yukihiro Nakai <ynakai redhat com>]

Oh, OK.
That was my misunderstanding.
Thanks for your great work.

--
Nakai

On Mon, 19 Jan 2004 17:36:14 +0100
Danilo Segan <danilo gnome org> wrote:

> Hi Owen, Yukihiro,
> 
> Owen Taylor <otaylor redhat com> writes:
> 
> > On Mon, 2004-01-19 at 03:02, Yukihiro Nakai wrote:
> >> Could you commit some appealing messages in the directory?
> >> Something like DONT_TRANSLATE_THIS file.
> >> 
> >> I cannot find any warning messages in the directory.
> >> http://cvs.gnome.org/bonsai/rview.cgi?cvsroot=/cvs/gnome&dir=gtk%2b/po-properties
> >> 
> >> Translators want to translate everything. If there is no big sign, you might
> >> repeat to say this.
> >
> > I think what is meant is 
> >
> >  Don't run intltool-update
> >
> > or something like that. Translating strings should be fine. 
> 
> Yes, sorry for not being clear enough.  I just supposed that those who
> are aware of the real problems won't have problems translating strings
> anyway. Since at least one translation was commited with wrong
> strings, I wanted this to be noticed, since it's no fun reverting more
> of them, and it's especially no fun for translators who may not be so 
> proficient with CVS.
> 
> Otherwise, both translation status pages and intltool-update (two
> most common mechanisms translators use to "refresh" their
> translations) were broken at the time of message, and I think
> translation status pages are still broken; it just doesn't show up
> because they don't update the POT file at all. I can fix that, but I
> don't want to do that before latest intltool from CVS is installed on
> the machine where status pages are generated, otherwise, we would get
> incorrectly updated PO files in the status pages, and we might
> get some translators to commit incorrect translations as well, if they
> don't hold the entire checkout.
> 
> > Running 'make update-po' should work fine.  
> 
> If translators actually ran ./autogen.sh Gtk+ in their checkout, that
> would be viable, but I seriously doubt they have.  Btw, not many know
> about it anyway, and certainly most of them expect intltool-update to
> work correctly. Maybe I'm wrong, but all the l10n guides point people
> at intltool-update, so there's good chance of me not being wrong :)
> 
> > Maybe Danilo can clarify?
> 
> Hope it's a bit more clear now what I meant ;)
> 
> Didn't want to stop anyone from translating, of course, just wanted
> to save translators from wasting some effort. I certainly expected
> this to be resolved in a day or two, so I already posted a message on
> gnome-i18n to use the latest intltool when updating Gtk+.
> 
> 
> > (It looks like if you use the latest intltool CVS, running
> > intltool-update should be fine as well.)
> 
> Yes, bug #131885 for intltool.
> 
> Cheers,
> Danilo