[fitswcs] Keywords for other FITS image representations

[William Pence]

Mark Calabretta wrote:

> Since we now allow projections to be associated with multiple axes we have to
> generalize the auxiliary parameters LONGPOLE, LATPOLE, EQUINOX, RADECSYS, and
> MJD-OBS to include an axis number as well as a version number.

In the vast majority of cases, these parameters are going to be the same
for all the axes, aren't they?  If so, can't we continue to allow the
current keywords to be used, i.e., the software should first look for an
EQNOXiiv keyword, and if it doesn't exist, then look for EQUINOX.  This
would aid backward compatibility in the majority of cases.

> If there are no backwards compatibility issues I would also extend this scheme
> to the remaining keywords:
> 
>    Primary:  CTYPEiiv  CUNITiiv  CRPIXiiv  CRVALiiv  CDii_jjv
> 
>   Bintable:  CTYiivnn  CUNiivnn  CRPiivnn  CRViivnn  Ciijjvnn
>          or  CTYivnnn  CUNivnnn  CRPivnnn  CRVivnnn  Cijvnnn
> 
>    Pixlist:  TCTYnnnv  TCUNnnnv  TCRPnnnv  TCRVnnnv  CDnn_mmv  (for backwards
>                                                                 compatibility)
> If not for backwards compatibility I would have written
> 
>    Pixlist:  CTYPnnnv  CUNInnnv  CRPInnnv  CRVAnnnv  CDnn_mmv

I'm not sure how the rest of the X-ray community will feel, but personally
I prefer the second Pixlist version (e.g., CTYPnnnv).  This serves to
clearly indicate that the file is using the new improved WCS parameters,
and not the old set of keywords, i.e.,

	TCTYPnnn
	TCRPXnnn
	TCRVLnnn
	TCDLTnnn

Note that for current X-ray missions, at least, there is no need for any
rotation correction in the Pixlist case because the pixel coordinates
listed in the table are transformed such that North is up.  In principle,
any distortions from a simple -TAN projection have also been removed.

I am not aware of anyone that is currently using the Bintable
representation for images and also trying to store WCS information, so
there should be few if any backwards compatibility issues with whatever
keyword naming scheme you want to propose.

> If not too much of a shock to the system I suggest the "v" version letter be
> lower-case to distinguish it from the fixed part of the keyword.  Gasp!

Gasp is right!  Most FITS readers and writers can't cope with lowercase
letters in keyword names (not to mention that it violates the FITS
standard).  CFITSIO for instance automaticly converts them to uppercase
before writing the keyword to the file and there is no way for the user to
bypass this.

-Bill