Support images?

[tammojan]

I think it would be great to reproject all-sky images (FITS files) onto a polyhedron. Do you see a possibility to add that to this code? Or would it require a totally different approach?

For inspiration, see how Earth can be projected onto different foldable polyhedra: http://www.progonos.com/furuti/MapProj/Normal/ProjPoly/Foldout/foldout.html

[tammojan]

To answer my own question: I took a map of 720x360 pixels (representing latitude and longitude), from http://vizier.cfa.harvard.edu/viz-bin/Cat?VIII/76 . The following code makes an entry in stars.dat for each pixel in the image:

from astropy.io import fits
import astropy.units as u
from astropy.coordinates import Longitude, Latitude

data=fits.open("lab.fit")[0].data[445,:,:]
(ymax, xmax) = data.shape

f = open('mystars.dat', 'w')
for (y, x), val in np.ndenumerate(data):
    ra_deg = x/(xmax-1)*360 * u.deg
    dec_deg = (y/(ymax-1)*180-90) * u.deg
    ra, dec = Longitude(ra_deg), Latitude(dec_deg)
    print("{},{},{}".format(ra.to_string(sep=" ", unit=u.hour),
                            dec.to_string(sep=" ", alwayssign=True, unit=u.deg),
                            val),
          file=f)

f.close()

Also, I made some changes to color stars according to magnitude, I'll create a PR for that.