Some tools to work with healpix.
Note reader utilities use camelCase to avoid name collision with the healpy readers.
import healpix_util as hu
# create a HealPix object. This will carry around the nside, scheme, npix,
# etc.
hpix=hu.HealPix("ring", 4096)
print(hpix)
scheme: 1
scheme_name: RING
nside: 4096
npix: 201326592
ncap: 33546240
area: 6.24178e-08 square degrees
area: 0.000224704 square arcmin
area: 0.808935 square arcsec
# has many attributes, e.g. .scheme, .nside, .npix, .area, etc.
# find pixel number for input ra,dec arrays
pixnums = hpix.eq2pix(ra, dec)
ra, dec = hpix.pix2eq(pixnums)
# or for theta,phi
pixnums = hpix.ang2pix(theta, phi)
theta, phi = hpix.pix2ang(pixnums)
# query disc
pixnums = hpix.query_disc(ra, dec, radius_degrees)
pixnums = hpix.query_disc(ra, dec, radius_degrees, inclusive=True)
pixnums = hpix.query_disc(theta, phi, radius_radians, system='ang')
# load a healpix.Map object
m = hu.readMap(filename)
m = hu.readMap(filename, column='I')
# load multiple Maps into an ordered dict, keyed by column name
maps=hu.readMaps(filename)
maps=hu.readMaps(filename,columns=["I","Q"])
# convert between schemes
mnest = m.convert("nest")
mring = m.convert("ring")
# find the mapval for the input coordinates
mapvals = m.get_mapval(ra, dec)
mapvals = m.get_mapval(theta, phi, system='ang')
# load a healpix.DensityMap, a special map that represents
# a spatial density
dmap=hu.readDensityMap(filename)
dmaps=hu.readDensityMaps(filename)
# generate random points from the map
ra,dec = dmap.genrand(100000)
# limit the range to a region known to have non-zero density, or
# weight. This will speed it up greatly
ra,dec = dmap.genrand(100000, ra_range=[60.,95.], dec_range=[-62.,-42.])
# generate randoms in theta,phi
theta,phi = dmap.genrand(100000, system='ang', theta_range=tr, phi_range=pr)
# healpy routines are pulled into the healpix_util namespace
npix=hu.nside2npix(nside)
# obscure features
# check quadrants around the input points
# make sure weighted position ellipticity in adjacent quadrants
# less than 0.05
ellip_max=0.05
for i in xrange(ra.size):
maskflags[i] = dmap.check_quad(ra[i], dec[i], radius_degrees[i], ellip_max)
# see docs on quad_check
# checking maskflags > 1 indicates the cluster is "good", but note during
# a lensing measurement you should be careful to always use sources from
# the same pair of quadrants (unless all are good)- numpy
- healpy
- fitsio # for file reading