i18n issues: Markup in glade labels?

[Christian Stimming]

Hi,

thanks for pointing to that bug report. *sigh* Unfortunately it seems the 
glade/libglade developers were unable to agree on a common technical solution 
to this problem. And it's back on the translators to simply put up with that 
(albeit small) number of duplicated translations.

As for gnucash, roughly 50 strings are concerned, but only (I just checked) 
about 15 of these did appear in the 1-8-branch. So if there were a technical 
solution to ignoring this markup, then it would save 15 new translations. 

So yes, I have to agree with Andreas: Unfortunately we just have to accept 
that limitation of glade/libglade and translate these additional string.

By the way, the translation file itself grew much larger in HEAD: Instead of 
the 2800 msgids that we had in gnucash-1.8.12 we now have 3400 msgids, of 
which roughly 1100 are untranslated when merging a 100%-translated 1.8 po 
file into HEAD. This means there is quite some work ahead for all 
translators.

Christian

Am Samstag, 12. November 2005 11:46 schrieb Andreas Köhler:
> Hi,
>
> regarding markup in glade files:
>
> http://bugzilla.gnome.org/show_bug.cgi?id=97061
>
> So I guess it is easier not to spend too much time on discussing how
> this could be solved :)
>
> --andi5
>
> On Saturday, 12 Nov 2005, 10:35 CET, Christian Stimming wrote:
> > I'm now starting to review the strings marked for translation (i18n =
> > internationalization) in gnucash. During this process I'll point out
> > different issues that would require to change the marked strings in the
> > gnucash source code.
> >
> > One issue that I've encountered now is that in HEAD there are many text
> > labels in glade files where the label text contains Markup. As in:
> >
> > #: ../src/business/business-gnome/glade/businessprefs.glade.h:1
> > msgid "<b>Bills</b>"
> >
> > which is from the glade file
> > src/business/business-gnome/glade/businessprefs.glade. There are 50 other
> > such messages. The problem here is that the markup contains no relevant
> > information to the translator. And the words in itself (without the
> > markup) are also already translated somewhere in the file -- only the
> > fact that here it is noted with the HTML markup makes them a second
> > msgid. This is rather a nuisance to the translators...