SETAIRMASS utmiddle format and MKNOBSFILE
subasavage wrote on Jun 05, 2011
I see from the "IRAF V2.14.1-EXPORT Release Notes" that the setairmass utmiddle format is always in date format, rather than the format defined in the UT keyword. This is causing problems when I run mknobsfile and extract various bits from the phot output .mag file. The UT time is always coming back as just the year, rather than the UT time. This is, of course, not particularly useful. Is there some way for mknobsfile to properly read the utmiddle keyword and only return the UT time rather than the first four characters of the date?
Thank You,
John Subasavage
Thank You,
John Subasavage
Mike Fitzpatrick wrote on Jun 05, 2011
Could you please post more information (e.g. an example) and/or upload sample data to the anonftp at ftp://iraf.noao.edu/pub? I'm having trouble figuring out whether the table column is just now the wrong type or something else isn't properly converting the datetime string. Thanks.
subasavage wrote on Jun 05, 2011
No problem! I posted a tarball (subasavage.tar.gz) that consists of an image (subasavage.fits), a .mag file output from phot (subasavage.mag.7), and the output file from MKNOBSFILE (subasavage.obs). In the .obs file, the OTIME column is the problematic one. Please let me know if these files did not transfer correctly (I had trouble with ftp so I'm skeptical).
Many Thanks!
John
Many Thanks!
John
Mike Fitzpatrick wrote on Jun 05, 2011
Sorry for the slow response, but I got to looking at this today and found that the subasavage.tar.gz file appears to be corrupted, i.e. I get most of the .fits file before tar craps out.
I get enough of the image to notice that there is no UT keyword in the header as I expected. Looking through the code I also see that the UTMIDDLE keyword isn't used directly, is it possible you've set the e.g. datapars.obstime parameter to use UTMIDDLE or somesuch?? If so, you could try using the 'TIME-OBS' keyword instead (which appears to be the UT time in your header). It would help if you could update the tar file again in binary mode, or else let me know where I can get them from you.
If changing the keyword helps then the problem may not be in mknobsfile at all, but upstream in the process somewhere.
I get enough of the image to notice that there is no UT keyword in the header as I expected. Looking through the code I also see that the UTMIDDLE keyword isn't used directly, is it possible you've set the e.g. datapars.obstime parameter to use UTMIDDLE or somesuch?? If so, you could try using the 'TIME-OBS' keyword instead (which appears to be the UT time in your header). It would help if you could update the tar file again in binary mode, or else let me know where I can get them from you.
If changing the keyword helps then the problem may not be in mknobsfile at all, but upstream in the process somewhere.
subasavage wrote on Jun 05, 2011
Hi Fitz,
Back at ya on the slow reply (baby arrived a week ago so I'm just now catching up on things!).
You are correct that I was using UTMIDDLE as the value for datapars.obstime. In the past, this worked fine when the UTMIDDLE format was without the date. I tried again using TIME-OBS and, indeed, it fixed the problem. So is it the case that I am unable to use UTMIDDLE? In practice, the two are not significantly different most of the time but UTMIDDLE seems to be the more appropriate.
Just in case, I've uploaded the tar ball (with comparison files using TIME-OBS) to my ftp site (ftp://ftp.ctio.noao.edu/pub/subasavage/IRAF_Test/) if you'd like to have a look.
Thanks again!
John
Back at ya on the slow reply (baby arrived a week ago so I'm just now catching up on things!).
You are correct that I was using UTMIDDLE as the value for datapars.obstime. In the past, this worked fine when the UTMIDDLE format was without the date. I tried again using TIME-OBS and, indeed, it fixed the problem. So is it the case that I am unable to use UTMIDDLE? In practice, the two are not significantly different most of the time but UTMIDDLE seems to be the more appropriate.
Just in case, I've uploaded the tar ball (with comparison files using TIME-OBS) to my ftp site (ftp://ftp.ctio.noao.edu/pub/subasavage/IRAF_Test/) if you'd like to have a look.
Thanks again!
John
Mike Fitzpatrick wrote on Jun 05, 2011
Hi John,
Congrats on the new arrival!
Since you have a workaround for now just use the TIME-OBS keyword. I'll change the code to allow the new UTMIDDLE value to be parsed correctly for the next release. I was able to download your new tarball and everything seems to unpack correctly now, thanks.
-Mike
Congrats on the new arrival!
Since you have a workaround for now just use the TIME-OBS keyword. I'll change the code to allow the new UTMIDDLE value to be parsed correctly for the next release. I was able to download your new tarball and everything seems to unpack correctly now, thanks.
-Mike
subasavage wrote on Jun 05, 2011
Hi Mike,
Will do on the workaround until the next release. Thanks for your help!
Cheers,
John
Will do on the workaround until the next release. Thanks for your help!
Cheers,
John
Last post on Jun 05, 2011