Swedish dates [was: Re: test binaries for Beta 5]

Dirk Hohndel dirk at hohndel.org
Wed Aug 6 13:10:02 PDT 2014


On Wed, Aug 06, 2014 at 09:44:21PM +0200, Joakim Bygdell wrote:
> On Wed, Aug 6, 2014 at 9:29 PM, Dirk Hohndel <dirk at hohndel.org> wrote:
> 
> > On Wed, Aug 06, 2014 at 04:24:56PM -0300, Thiago Macieira wrote:
> > > On Wednesday 06 August 2014 10:36:26 Dirk Hohndel wrote:
> > > > > “Sön 6 jul, 2014” will work in all cases.
> > > >
> > > > I don't think I can easily tell Qt not to use ordinals here. Even
> > though
> > > > the wiki above seems to say that that's preferred for dates.
> > > >
> > > > Thiago?
> > >
> > > Looks like the QLocale data is faulty somehow.
> > >
> > > I asked the bot on IRC and it replied:
> > >
> > > 14:40 <+thiago> qout: { QLocale l("sv-SE"); qDebug() <<
> > l.toString(QDate(2014,
> > > 8, 1), QLocale::LongFormat); }
> > > 14:40 <+qout> "fredagen den 1:e augusti 2014"
> > >
> > > I know we process the CLDR data from unicode.org, but a rule like "use
> > this on
> > > day 1 and use that on other days" is definitely not part of Qt. Most
> > likely, we
> > > failed to parse and will need to add an exception.
> >
> > The format appears to try to build all ordinals with ':e' which I have now
> > learned is wrong :-/
> >
> > /D
> >
> A correct parsing rule is a bit tricky as it must include some exeptions.
> Dates ending on 1 or 2 should have ":a" after except if the second to last
> is 1 then they should have ":e" as should all dates ending on any number
> except 1 or 2.

Yeah, I decided to go with the official suggestion (according to the wiki)
to not use ordinals for dates :-)

Did you have a chance to try the DMG that I made for you? Do the dates
look sane?

Linus, I'm sure you get bored merging all these Linux patches... could you
run Subsurface in Swedish and take a look at the dates as well?

Thanks


/D


More information about the subsurface mailing list