[Kstars-devel] kdeedu/kstars/kstars

[Jason Harris]

CVS commit by harris: 

Fixing rise/set times.  Joseph North pointed out that the Moon's set 
time was sometimes given for the following day, rather than the current 
day.

This is actually a complicated issue.  To compute the "set time", you 
first have to specify *which* set event you are talking about.  One can 
think of several sensible ways to define it:  the next chronological set 
event; the set event which occurs on the current calendar date; the 
nearest set event going either forward or backward in time.  It isn't 
clear that any of these is obviously the best idea, yet they all behave 
differently.

We used to simply display the previous rise event, and the next set 
event.  Now, we first assess if the object is currently below the 
horizon.  If so, we modify the time by +/- 12 hours to bring the object 
above the horizon, and *then* calculate the previous rise event and next 
set event.  

This gives us the behavior that the rise/set times only change at the 
instant an object crosses the lower meridian (when it reaches its lowest 
point and changes from "setting" to "rising"). 

Also, I improved the behavior in the WUT tool.  It was also giving 
rise/set times for dates which were sometimes off by one.  We now use 
special reference times (midnight, 6:00 in the evening, and Noon the 
following day) to ensure that the various rise/set times are computed 
correctly.

I noticed that the transit times are completely bogus.  I'm looking into 
it.

CCMAIL: kstars-devel at kde.org
CCMAIL: joseph.north at mail.mcgill.ca


  M +1 -1      skymap.cpp   1.142
  M +22 -2     skyobject.cpp   1.43
  M +1 -1      skyobject.h   1.43
  M +63 -44    tools/wutdialog.cpp   1.31
  M +1 -2      tools/wutdialog.h   1.8