Matt,
It appears that the problems you were having with FITS file
manipulation may be due to rfits() trying to be too smart. The Right
Way to get a consistent data file and fits header, in scientific
units, is:
$a = rfits($filename);
$hdr = $a->hdr;
That will scale $a into scientific units (applying any BSCALE and
BZERO correction from the header). It also deletes BSCALE and BZERO
from the header, to avoid confusion.
If you want to avoid the scaling, you can do this:
$a = rfits($filename, {bscale=>0 });
(See the rfits documentation for more details).
In your case, you were doing this:
$a = rfits($filename);
$hdr = rfits_hdr($filename);
This yielded a scaled $a, and the original header. You should either
use the rfits-supplied header (as the first example) or tell rfits
not to apply bscaling (as in the second example).
Now, you were having a problem with IRAF not reading your output
files with BITPIX=-64 and BZERO set to 32768; but that appears to be
an IRAF problem: I checked my printed copy of the FITS User's Guide
v4.0 (a bit long in the tooth) and the more recent keyword dictionary
at "http://heasarc.gsfc.nasa.gov/docs/fcg/standard_dict.html";, and
there's no mention of any restriction on BZERO and BSCALE. IRAF may
have been complaining because the erroneous BZERO was shifting your
data by another +32768. The one you had shifts data values from the
range -32768-32767 (unsigned shorts) to scientific values in the
range 0-65535 (unsigned shorts). On the second read, it would shift
again, to the range 32768-98303.
Kindly,
Craig
_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
