Re: 3.0 branch: POTFILES.in broken

["Zack Weinberg" <zackw at Stanford dot EDU>]

On Tue, Jun 12, 2001 at 08:10:36AM +0200, Dennis Bjorklund wrote:
> On Mon, 11 Jun 2001, Zack Weinberg wrote:
> 
> > > A better way is to apply the patch attached that removes sv from the
> > > variable ALL_LINGUAS which means it will not be built.
> >
> > The thing is so old that the translated error messages you get might
> > be actively misleading.  At that point I think we're better off
> > without having it at all.
> 
> I know what the file in cvs looks like, I wrote it after all :-) The
> translation system work so that there will be no misleading messages, but
> it will be a mixture of english and swedish strings. As sone a string is
> changed it will become fuzy in the po file and not used in the binary.
> It's just the unchanged strings that are left and used. So it can never be
> that we emit a message that is wrong just because the translation is old,
> in that case the message emited will be english.

I had been under the misapprehension that fuzzy strings were still put
into the compiled .mo file.  If they aren't, then there isn't nearly
so much of a problem.

> I was merely hoping that someone would see the translation, get interested
> and help out.

Okay.  I'll just change ALL_LINGUAS on the branch, and leave it alone
on the trunk.  Can you merge your better sv.po against the new gcc.pot
and then apply it to trunk, please?  And on the branch, too, if Mark's
okay with that.  [Note that gcc.pot is different on trunk and branch.]

-- 
zw                I'm on a spaceship full of college students.
                  	-- Martin "PCHammer" Rose