Re: en_GB translations (was Re: GNOME 2.2 Translation Statisticsand Rankings)

[Bastien Nocera <hadess hadess net>]

On Fri, 2003-01-31 at 14:56, Christian Rose wrote:
> fre 2003-01-31 klockan 11.33 skrev Bastien Nocera:
> > > 
> > >     $ msgen --help
> > >     Usage: msgen [OPTION] INPUTFILE
> > >     
> > >     Creates an English translation catalog.  The input file is the last
> > >     created English PO file, or a PO Template file (generally created by
> > >     xgettext).  Untranslated entries are assigned a translation that is
> > >     identical to the msgid, and are marked fuzzy.
> > > 
> > > This way, the en_GB translator doesn't need to manually copy the
> > > contents of the individual msgids into the msgstrs all the time when he
> > > or she translates or verifies the translation of messages; only
> > > un-fuzzying the messages is required to indicate they have been
> > > verified. And when the msgstrs are un-fuzzied and non-empty it's clear
> > > that they have been verified by an en_GB translator. Unfortunately,
> > > currenly that just isn't the case for 98.52% of the en_GB messages.
> > 
> > I fixed the en_GB translation script in gnome-i18n to do that. It just
> > copies the msgid when it doesn't need translating.
> 
> Excellent! It does mark them fuzzy too at the same time, I presume?

No, I don't want to spend my time removing the "fuzzy" lines. Here's how
I proceed when I do en_GB translations:

- I spot something that's not in proper English (American English, say
;)
- I do a cd po && make *.pot
- run the en_GB.pl script on it
- see if what I spotted was translated with the script, and check all
the translations. Update and re-run the script if necessary.
- Update ChangeLog and commit.

So, no, I don't mark stuff as fuzzy...

Cheers

-- 
/Bastien Nocera
http://hadess.net

#2  0x4205a2cc in printf ("Oh my %s\n", preferred_deity) from
/lib/i686/libc.so.6 printf ("Oh my %s\n", preferred_deity);
Segmentation fault