Brad,
I have fixed AST so that it reports an informative error in the
situation you are facing instead of core dumping. The changes are checked
into CVS and I've but a new ast.tar into /classic on saturn (AST V3.3-2),
so the changes should be in /stardev tomorrow.
This error is triggered by the values for the PCi_j keywords in the
header:
PC1_1 = 1.0 / Transformation matrix element
PC2_2 = 0.000000E+00 / Transformation matrix element
PC3_2 = 1.000000E+00 / Transformation matrix element
The problem is that this PC matrix cannot be inverted, due to the
value of zero for PC2_2.
David
On Wed, 7 Jul 2004, Brad Cavanagh wrote:
> I would have filed this via Bugzilla but it's currently broken.
>
> Sorry to break into the CVS extravaganza with mundane old Classic
> problems, but I'm getting a segmentation fault from KAPPA/WCSFRAME. It
> appears to happen when a 2D image has a 3D WCS. UIST raw frames are HDS
> containers, with the data in an .I1 component and most of the headers
> (including all of the WCS information) in a .HEADER component. If I
> NDFCOPY the .I1 into an NDF, then merge the FITS headers from the .HEADER
> into that NDF, then do a WCSFRAME, I get a segfault. I'm afraid I can't be
> more specific about the "merge FITS headers" bit, but if you want to look
> at the code that does it it's in
> /star/bin/oracdr/lib/perl5/ORAC/Frame/UIST.pm, in the mergehdr() method.
> KAPPA/NDFTRACE also segfaults on this file.
>
> I have a sample file at
> http://www.jach.hawaii.edu/~bradc/starlink/u20040626_00095_raw.sdf that
> shows the segfault. This is occurring under KAPPA v1.3-1, from /stardev
> that was built on 08 June.
>
> Cheers,
> Brad.
>
----------------------------------------------------------------------
Dr David S. Berry ([log in to unmask])
STARLINK project | Centre for Astrophysics
(http://www.starlink.ac.uk/) | University of Central Lancashire
Rutherford Appleton Laboratory | PRESTON
DIDCOT | United Kingdom
United Kingdom | PR1 2HE
OX11 0QX Tel. 01772 893733
01257 273192
|