Inappropriate MAP_POLAR_CAP default value for linear projection with polar coordinates (-JP)
|Target version:||Candidate for next bugfix release|
MAP_POLAR_CAP has a default value 85/90. This default value is OK for geographic-related projections.
However, for most linear projection with polar coordinates (-JP), this default value makes radial lines disappear in r<5.
Example below helps explain this bug. When R=0/360/0/10, half of the radial don't show. When R=0/360/0/1, all radial lines disappear.
gmt psbasemap -R0/360/0/90 -JP6c -Bxg30 -Byg30 -B+t'-R=0/360/0/90' -K > test.ps gmt psbasemap -R0/360/0/10 -JP6c -Bxg30 -Byg1 -B+t'-R=0/360/0/10' -K -O -X8c >> test.ps gmt psbasemap -R0/360/0/1 -JP6c -Bxg30 -Byg0.1 -B+t'R=0/360/0/1' -O -X8c >> test.ps
This unexpected behavior may make users confusing, especially for new GMT users.
Although users can modify MAP_POLAR_CAP value to fix this issue, it may be better if it has a more reasonable default value, such as 5%?
- Status changed from New to Feedback
- Assignee set to Paul
We did not consider the polar r,theta projection when MAP_POLAR_CAP was implemented. So there are two options:
- Declare a bug and restrict this default feature to geographic maps only, as it was intended.
- Extend the behavior to also apply to r-theta plots.
If we were to do the latter then clearly the 85/90 default is somewhat meaningless. Also, 5% is not so good either since anyone would want it to be a multiple of some gridline spacing and not just some floating point 5% fraction of the range. So my first suggestion would be we declare a bug and restrict it to geographic maps.