Re: [patches] Italy LC_TIME wrong

[Maxim Kuvyrkov <maxim@xxxxxxxxxxxxxxxx>]

Joseph S. Myers wrote:
> On Tue, 21 Jul 2009, Maxim Kuvyrkov wrote:
>
> > Nicolò Chieffo wrote:
> > > Hello, I would like to inform you that the LC_TIME for Italy is wrong
> > > (so both it_IT and it_CH)
> > >
> > > Using `locale -k LC_TIME` to see the settings there is an error:
> > > date_fmt="%a %b %e %H:%M:%S %Z %Y"
> > > should be changed to
> > > date_fmt="%a %e %b %H:%M:%S %Z %Y"
> > > (so swap %b with %e)
> > > because in Italy we put the day_number before the month
> > Have you posted this as a bug report in FSF GLIBC (at
> > sourceware.org/bugzilla)?  It would be best to have this fixed in FSF GLIBC;
> > EGLIBC will quickly pick up the fix then.
>
> I'd also like to remind people of what I said in 
> <http://www.eglibc.org/archives/patches/msg00667.html>: neither FSF GLIBC 
> nor EGLIBC is really the right place for locale data to be maintained and 
> there's at least one separate project maintaining this data.  If people 
> find it convenient to have a full set of locale sources packaged with the 
> C library, the right approach for that would be as with the timezone data: 
> separately maintained data of which snapshots are imported into the C 
> library sources without modifications (and in such a case, each of EGLIBC 
> and FSF GLIBC could import new snapshots whenever convenient and there 
> would be no particular need for EGLIBC to import only via FSF GLIBC if 
> there were a single source upstream of both).

FWIW, I fully agree with that.  Do you know what is FSF GLIBC's position 
on moving localedata into a separate project?

--
Maxim