[fitsbits] Primary & Alternate WCS Keyword Order

[Mark Calabretta]

On Wed 2012/06/27 09:50:26 -0400, Arnold Rots wrote
in a message to: Mark Calabretta <mark at calabretta.id.au>
and copied to: William Pence <William.Pence at nasa.gov>, fitsbits at nrao.edu

>You stated that it does not make sense for WCSAXES to be less than the
>number of non-degenerate axes in the image.
>However, would it not make sense, if one had a RA-Dec-Doppler cube to
>define a coordinate system that only provides an alternate for the
>spatial axes, in which case WCSAXESA would be 2?

There is no inheritance between coordinate representations.
What to do in such a case is undefined.  WCSLIB takes the
maximum of NAXIS, WCSAXESa, and the highest axis number in any
of the 'A' WCS keywords.  Hence it would ignore WCSAXESA here
and use defaults for the third axis, basically giving it pixel
coordinates.

Regards, Mark