Dear Professor Ichikawa,
Thanks for the fast check! It was great to see that the segmentation
faults no longer occur. The remaining FAIL had a very simple fix, which
I was able to implement in this commit:
https://git.savannah.gnu.org/cgit/gnuastro.git/commit/?id=56b7953e0
The latest tarball now contains this fix:
http://akhlaghi.org/gnuastro-latest.tar.lz
If you get a chance to verify that this tarball has no FAILs, it would
be great.
Thanks again for the very useful bug report,
Cheers,
Mohammad
P.S. Just as a side-note, the particular script that crashed enables the
very nice display of the full dynamic range of astronomical images in
publications. See the bottom two panels of the figure in its paper:
https://ui.adsabs.harvard.edu/abs/2024RNAAS...8...10I
It can also be used without libjpeg, just set the output file as PDF. We
have also prepared a tutorial in the book on how to optimize the script
for the particular dataset:
https://www.gnu.org/software/gnuastro/manual/html_node/Color-images-with-full-dynamic-range.html