Installation procedure for Healpy

  • Note: Healpix is included, so you don’t need to get it separately


  • Healpix needs cfitsio. You can either:
  • use an existing installation : Either define the environment variable CFITSIO_EXT_PREFIX where to find the cfitsio library and include file (eg /usr/local, so that /usr/local/include/fitsio.h and /usr/local/lib/libcfitsio.a exists), or define CFITSIO_EXT_INC (with the cfitsio include directory) and CFITSIO_EXT_LIB (with the cfitsio library file name) Please do not define all three of these environment variables simultaneously.
  • compile a specific cfitsio lib: Define EXTERNAL_CFITSIO=no, place the cfitsioXXXX.tar.gz in hpbeta/libcfitsio before installing. The cfitsio version XXXX must match the version in hpbeta/planck.make (or you need to modify it there).
  • the psht spherical transform library is integrated into healpy with the

pshyt cython wrapper. a pregenerated c code is included in the repository, but if you have cython installed it will be run in the build phase.


$ cd healpy $ python build

OR, if you do not want OpenMP support (sometimes, it causes SegFault)

$ python build –without-openmp (alternatively, you can define the environment variable HEALPY_WITHOUT_OPENMP)

If you do not want the “-march=native” flag (if your g++ is too old)

$ python build –without-native (alternatively, you can define the environment variable HEALPY_WITHOUT_NATIVE)

If everything goes fine, you can give a test: $ cd build/lib* $ ipython -pylab >>> import healpy as H >>> H.mollview(arange(12)) >>>

If the plot looks good, you can install: $ sudo python install # install in default location, need root rights or $ python install –install-lib=~/Softs/Python # will install healpy in directory ~/Softs/Python, which then must be in your PYTHONPATH or $ python install –user # will install it in your User python directory (python >= 2.6)

Compile on OSX

Suggested compilation on OSX Lion is installing pyfits, cython and cfitsio using mac ports and run: >>> python –without-openmp

Development install

the best way to install healpy if you plan to develop is to build the C++ extensions in place with:

python build_ext –inplace

the add the healpy/healpy folder to your PYTHONPATH

Project Versions

Table Of Contents

Previous topic

Healpy tutorial

Next topic

pixelfunc – Pixelisation related functions

This Page