Port: astro/sunclock: Malaysia is in the wrong time zone

[raf]

On Sun, Mar 21, 2021 at 07:16:03PM +1100, Greg 'groggy' Lehey <grog at FreeBSD.org> wrote:

> On Sunday, 21 March 2021 at 14:39:41 +0800, Erich Dollansky wrote:
> > On Sun, 21 Mar 2021 17:03:30 +1100
> > raf <freebsd at raf.org> wrote:
> >
> >> On Sun, Mar 21, 2021 at 11:31:34AM +0800, Erich Dollansky
> >> <freebsd.ed.lists at sumeritec.com> wrote:
> >>
> >>> Hi,
> >>>
> >>> I just noticed that Malaysia is in the same time zone as Jawa
> >>> (Indonesia) and Thailand.  Wikipedia says for Malaysia UTC+8 and for
> >>> Jawa and Thailand UTC + 7.
> >>>
> >>> Is there no maintainer for this port as this list is listed as
> >>> maintainer.
> >>
> >> I don't know anything about the sunclock port on
> >> FreeBSD, but I have been using it on Linux and macOS
> >> for decades, so I thought I pipe up.
> >>
> >> Part of Malaysia (the part containing Kuala Lumpur) is
> >> directly South of Thailand, so it wouldn't be
> >> surprising if that part had the same timezone as
> >
> > Malaysia is in the wrong time zone just as Singapore.
> 
> Malaysia and Singapore are in the correct time zones, UTC+8.  So is
> Thailand.  That's what their governments decided.  Time zones have
> only a passing similarity with solar time.  In (West) Malaysia's case,
> they decided to adopt the East Malaysian time to get rid of the
> discrepancy.  Before that, it was UTC+7:30 in the west and UTC+8 in
> the east.  You can find all this in so many places, but the easiest
> might be
> https://www.timetemperature.com/time-zone-maps/large-world-time-zone-map.shtml
> or https://en.wikipedia.org/wiki/Malaysia.
> 
> >> Perhaps the other part (around Brunei) has a different time zone,
> >> and Wikipedia is referring to that part? If so, maybe Wikipedia is
> >> incomplete?
> 
> No, Wikipedia is correct in this case.  So is the tz database at
> /usr/src/contrib/tzdata/asia
> 
> >> I must admit, I don't know where to see timezone information in
> >> sunclock. I can click on places, and it shows the solar time, which
> >> makes sense, and the solar time between the two parts of Malaysia
> >> are about an hour apart.
> >>
> >> Ah, it's in the Sunclockrc file. But it only seems to
> >> store timezones for cities, not countries, and I think
> >> the only Malaysian city it has is Kuala Lumpur.
> >
> > Yes, and the information given for Bangkok, Kuala Lumpur and Jakarta
> > are correct.
> 
> FWIW, since 1 January 1982 Malaysia has only one time zone, with
> offset UTC+8.  Singapore chose to go with Malaysia on the same date.
> 
> >> It contains these lines (for part of Malaysia and
> >> Thailand):
> >>
> >>   addcity 2|KualaLumpur|3.133|101.700|PST-8
> >>   addcity 2|Bangkok|13.833|100.483|TST-7
> >>
> >> You'd think that they'd have their own timezone
> >> abbreviations, and not have to borrow abbreviations
> >> from other countries so many timezones away. I don't
> >> even know what TST is supposed to mean.
> >
> > The time zone abbreviations are also unknown to me.
> 
> Indeed, it doesn't inspire confidence.
> 
> >> The manpage says that the timezone needs to be the name
> >> of a file under /usr/share/zoneinfo (or whatever
> >> directory is used on your system). So it should accept
> >> "Asia/Kuala_Lumpur" instead of "PST-8".
> >>
> >> Can you see if that works? If so, I'd be willing to do
> >> an audit and update all of the timezones in Sunclockrc
> >> to use the modern, geographic names for timezones. It
> >> would take a while, though.
> >
> > We first would have to get an idea of where to look. When showing
> > the map, Malaysia should share the colour with China. But it shares
> > its colour with Thailand an the western part of Indonesia.
> 
> I've lost track of the map, but it's not important.  There's no
> dispute about the time zones in the four countries (except that
> Indonesia has several time zones, and only Waktu Indonesia Tengah
> matches Malaysia.  But Java is in Waktu Indonesia Barat, UTC+7).
> 
> Greg
> --
> Sent from my desktop computer.
> See complete headers for address and phone numbers.
> This message is digitally signed.  If your Microsoft mail program
> reports problems, please read http://lemis.com/broken-MUA

Thanks, Greg. A better map for the purpose or
comparison with sunclock's is:

  https://www.worldtimezone.com/

If I can wrap my head around the vmf format enough to
fix it (I may be some time), I'll try to modernise the
timezones in the sunclockrc file (that's if they're
actually used for anything).

cheers,
raf