iraf.net

When I use 64-bit IRAF 2.15.1a, imtranspose returns very strange output. The problem does not happen with IRAF 2.14.1 and 32-bit IRAF 2.15.1a. I'm using IRAF mainly with pyraf on Mac OS X 10.7.3. I've uploaded the outputs on ftp://iraf.noao.edu/pub/test_len_blk_annotated.png (Did I upload correctly?).The input image, "test.fits", was made with following python script:import pyfits
import numpy as np
img = np.array(np.random.randint(2, size=(2048,2048)), dtype=np.int32)
hdu = pyfits.PrimaryHDU(img)
hdu.writeto('test.fits')Then I run imtranspose on pyraf, like:
> imtranspose test.fits tr_test_2048.fits len_blk=2048
When I use len_blk=1, imtranspose worked as expected, but in other cases, the periodic patterns with len_blk/2 lengths appeared. Cheers,
Masato

This problem was fixed just recently for the upcoming release, you can get a patched v2.15 IMAGES binary and library fromftp://iraf.noao.edu/iraf/v215/support/macintel/x_images.e
ftp://iraf.noao.edu/iraf/v215/support/macintel/libex.aDownload these in binary mode and install in your $iraf/bin.macintel directory. The images binary will fix the imtranspose problem immediately, however since this problem affects all tasks you'll also want to get the libex.a file so you can relink your system and external packages.To relink:[code:1:acc48aeeda]
% cd /iraf/iraf/ # go to the iraf root
% make macintel # reset the architecture
% make update # relink
[/code:1:acc48aeeda]External packages will need to be updated as well, let me know if you have questions about how to do that for a particular package (the details will vary).

Thanks a lot! It worked!Apparently there are some more files, especially relating to spectroscopy packages in
ftp://iraf.noao.edu/iraf/v215/support/macintel/
Since I use twodspec/onedspec extensively for the reduction and analysis of my data, do I need to update those files as well? [quote:ff00907ee8="fitz"]External packages will need to be updated as well, let me know if you have questions about how to do that for a particular package (the details will vary).
[/quote:ff00907ee8]I'm using only STSDAS/TABLES packages at this moment.