Stephen Walton wrote:

Mark's announcement of WCSLIB 3.3 reminds me to ask a generic question
about an issue I'm sure others have faced: how best to switch from an
IRAF environment to a CFITSIO/WCSLIB one. While IRAF has served us well
for 15 years, I don't see a huge effort being undertaken to bring its
WCS library in compliance with the standard, nor to support other map
projections beyond those already in place. As a solar astronomer I have
other needs :-) .

Most of my code is in Fortran, with just a few SPP "wrapper" main
programs to get access to some of the IRAF IMIO routines.

My WCSTools package supports all of Mark's projections (though I am still
in the process of upgrading it from WCSLIB 2.9 to WCSLIB 3.3) and reads
IRAF imh files as well as FITS files. I support the IRAF TNX projection,
but haven't quite got around to implementing the IRAF ZPX projection.
There are Fortran wrapper for the WCS subroutines, too. WCSTools has
its own FITS code--I've been dealing with FITS for over 20 years--but
the WCS subroutines are easily separated from the rest and are currently
used by CFITSIO, as well as many of the popular astronomical image display
programs. I am continuing to make an effort to support the evolving
standard as well as legacy WCS formats, including most drafts of the
FITS WCS papers. Multiple (and even chained) WCS's are supported, and
I have starting implementing FITS-WCS Paper IV by providing distortion
correction for SIRTF images.

Documentation and source code are available at

http://tdc-www.harvard.edu/software/wcstools/

-Doug Mink
Telescope Data Center
Smithsonian Astrophysical Observatory
Cambridge, Massachusetts USA