iraf.net

From: "Detrick Branston" <branston@noao.edu>
Date: August 12, 2005 3:43:29 PM MST
To: fvaldes@noao.edu
Subject: IRAF error checking
Hello Frank,I am encountering a file truncation error, at times, while running
IMSTACK. Obviously, this is not a problem with IRAF but with some of
my data. I am running my script in batch mode and can't afford these
sporadic errors, especially overnight. I want to know if there is
any type of error checking that can be done. Is there some value
that is returned by IMSTACK that I can intercept and use to make the
scripts more robust and able to move beyond these corrupt files
without crashing?Thanks,DetrickFrom: Frank Valdes <valdes@noao.edu>
Date: August 26, 2005 9:33:24 AM MST
To: Detrick Branston <branston@noao.edu>
Subject: Re: IRAF error checking
Hi Detrick,It may be the truncation is occuring in IRAF. This usually occurs
because something corrupts the FITS header and then the computation
of the data segment is wrong. There have been bugs in the FITS
kernel of IRAF that could cause this but currently all bugs should be
fixed. One possible solution would be to convert to imh format if
this is something to do with the FITS handling. If your data is
corrupted before being touched by IRAF then you are right that you
have to deal with it separately.Your question about running your script without crashing due to
corrupt files sounds like a good example of why the recent release of
the error handling CL was created. This should allow you to trap the
error in you script and then skip the bad data while not stopping the
script. I have not made use of the new features and I don't know the
details and whether something as strong as the file truncation error
can be trapped. The following link will give you information. This
new CL should be installed on the NOAO machines by default. If you
have more questions write back to iraf@noao.edu.http://iraf.noao.edu/iraf/web/new_stuff/ecl_beta.htmlYours,
Frank