Hi,

Now that I live in Chile I was checking the timezone data settings for Chile and noticed that they are wrong. The TZrules.dat file in the KStars sources contains this

#"CL": DST starts 2nd Sun in Oct, reverts 2nd Sun in Mar (4:00)
#(Chile)
CL Aug 2Sun 04:00 May 2Sun 03:00

The first line contains wrong info. Chile DST starts 1st Sunday in September and ends 1st Sunday of April at midnight. The third line doesn't make sense to me at all. What are the items supposed to mean? I suppose that CL means Chile, 2Sun the second Sunday and 04:00 and 03:00 the times at which the change takes place. But Aug and May? As a matter of fact all other entries in the file have those strange months as well.

Does anyone know why this is such? And also, how could I best request a change for the DST start and end dates and times for Chile? Is posting here enough or should I open an RFC somewhere else? If appreciated I could push the fix myself.


Wouter

I updated it to:

CL Sep 1Sun 00:00 Apr 1Sun 00:00

And pushed the fix, can you please verify?

I pulled all code updates for libgphoto2, indi, indi-3rdparty and kstarts and built them in that order. When building KStars I get

/home/wouter/Projects/kstars/kstars/ekos/capture/customproperties.cpp:119: undefined reference to `INDI::Property::getNumber()'

while the build passed last week. Any clue what is wrong?

did you install libindi? maybe you have older libindi in the system?

I simply did a make in the existing kstars build directory and that was my mistake. I removed the directory, created it again, ran cmake and the build is now ongoing. Sorry for the noise.

Jasem, the fix works in the sense that KStars now again displays the correct local time. Not sure yet if the switches to and from DST will work. If there is any issue with that then it will be for at most an hour and the only consequence will be that the local time is displayed incorrectly.

One small issue which may be harder to fix: the DST info now shows
DST starts 1nd Sun in Sep, reverts 1nd Sun in Apr
and of course that should be 1st. It is only semantics, however.


Wouter

Thanks, the typo is fixed.

Thanks Jasem. I pulled the c ode and recompiled and now it looks great!