[fitsbits] Update to the FITS Tiled Image Compression Convention

[Erik Bray]

On 07/24/2013 05:33 PM, William Pence wrote:
> This is to announce that a new version (v2.3) of the document
> describing the FITS Tiled Image Compression Convention has been placed
> on the FITS Support office web site at
>
>      http://fits.gsfc.nasa.gov/registry/tilecompression.html
>
> The main changes in this new version of the document are:
>
> * A new variant of the GZIP algorithm, called GZIP_2, is supported, in
> which the bytes in the pixel array are shuffled into decreasing order of
> significance before being compressed with gzip.  This is especially
> effective with floating point arrays.
>
> * The algorithm for dithering floating point values before quantizing
> them into scaled integers has been expanded and improved.  There is a
> new variant in which the pixels that are equal to 0.0 are not dithered,
> and instead the value is exactly preserved.
>
> * A new ZDITHER0 keyword is defined, to record the random seed value
> that is used to generate the dithering pattern.
>
> * If the compressed file is larger than about 2.1 GB, then the '1Q'
> variable length array FITS column format (with 64-bit pointers) must be
> used
> instead of the '1P' format.

That's great news--I've had an issue open in PyFITS [1] for a while about the 
last issue.  It was impossible to compressed larger files without allowing the 
use of Q-type VLAs.

Has CFITSIO been updated to support this too.  If not, is there any sense of 
when to look for that?

Thanks,

Erik

[1] https://trac.assembla.com/pyfits/ticket/159