[fitsbits] INF/NAN as header values?

[Rob Seaman]

On Jun 7, 2013, at 12:44 PM, Tim Pearson <tjp at astro.caltech.edu> wrote:

> Incidentally, it would be nice if the already-great CFITSIO could be modified to make it impossible to write an invalid FITS file (i.e., one that fails FITSVERIFY), but perhaps that is an unreasonable request.

There are competing requirements here.  For instance FPACK (layered on CFITSIO) seeks to compress FITS files in a reversible and idempotent (and lossless where appropriate) way.  If a not strictly conforming file is presented to FPACK, then FUNPACK should later restore that same not strictly conforming file verbatim.

By all means FITSVERIFY could be elaborated into a whole toolbox of commands (and related library routines and arguments) geared to improving conformance with the standard, but the responsibility ultimately lies with the application and workflow programmers.  (Who may even have explicitly chosen to violate particular keyword formats, for instance.)

Gödel's theorem that a system can't be both complete and self-consistent would apply to even a perfected FITS standard and a community of idealized users :-)

Rob Seaman
NOAO
--

http://heasarc.nasa.gov/fitsio/fpack/