Bug 402043

Summary: Local horizon display broken in RA mode
Product: [Applications] kstars Reporter: Pit <P.Suetterlin>
Component: generalAssignee: Jasem Mutlaq <mutlaqja>
Status: RESOLVED FIXED    
Severity: normal    
Priority: NOR    
Version: git   
Target Milestone: ---   
Platform: Other   
OS: Linux   
Latest Commit: Version Fixed In:
Sentry Crash Report:
Attachments: Screenshots of kstars with working and broken horizon display
kstars userdb with horizon definition

Description Pit 2018-12-12 10:42:44 UTC
Created attachment 116873 [details]
Screenshots of kstars with working and broken horizon display

SUMMARY

I have defined a local horizon using 'Settings->Artificial Horizon.
This works nice in Horizontal mode (AltAz), see kstars_az.jpg in the attached archive.  But switching to Equatorial mode (RaDec) the drawn polygon is completely bad, and depending on where I center the view starts tinting the whole star map (kstars_ra1.jpg and kstars_ra2.jpg in the archive; the first one is at the same location as kstars_az.jpg, just pressed 'Space' to switch)

EXPECTED RESULT

The local horizon should be properly displayed also in RaDec mode (or at least only drawn in AltAz mode)

SOFTWARE/OS VERSIONS
Linux/KDE Plasma: openSUSE Tumbleweed
KDE Plasma Version: 5.14.4
KDE Frameworks Version: 5.52.0
Qt Version: 5.11.2
Comment 1 Jasem Mutlaq 2018-12-21 06:18:23 UTC
Can you attach your userdb.sqlite file? or export the horizon data from it? I want to reproduce it here and work on it.
Comment 2 Jasem Mutlaq 2018-12-21 06:18:55 UTC
The file is located under ~/.local/share/kstars
Comment 3 Pit 2018-12-21 09:30:12 UTC
Created attachment 117038 [details]
kstars userdb with horizon definition

Sure, here it is.
The behavior seems to vary.  One time it was just not displaying the horizon data in RA mode at all, sometimes the display was almost correct.  But usually it draws wild patterns across the whole sky, changing when I move the view.
Comment 4 Jasem Mutlaq 2021-09-13 13:24:58 UTC
This was changed to an updated system in 3.5.4, can you please check?
Comment 5 Pit 2021-09-14 12:28:58 UTC
Ah - mea culpa!  Completely forgot about this report.

Yes, indeed in the current version (git) this problem no longer occurs.  The horizon is drawn correctly(*) in all cases, so this can be closed as fixed :)

(*) There's another issue now that AZ points **have** to be monotonic, so no overhangs.  But that is already marked 'wontfix' (design choice) by Hy ;^>