iraf.net

Hello Dugan,The formating is causing a problem. The program is trying to figure
out if the keyword value is a number or a string. It assumes a
string starts in column 11 with a quote mark. Since this is not the
case it is interpreting the value as a number. The program could be
smarter but it is not. A way to fix your data automatically is with
the task HFIX. The following should work:cl> hfix *.fits command="!sed 's/DATE-OBS= */DATE-OBS= /' $fname >
temp; mv temp $fname"I haven't tried to reproduce the segmentation violation but I expect
it may be due to the format problem. If you still get an error you
can just send me a long header listing (imhead <image> l+) and I can
create a dummy image with the same header.Yours,
Frank ValdesOn Oct 20, 2005, at 9:18 AM, Dugan Witherick wrote:> Dear Sir/Madam,
>
> I have been experiencing a problem using the IRAF task asthedit and I
> would be grateful for any feedback you have on a possible solution or
> work around. To begin with, I am using IRAF V2.12.2a on a pentium 4
> computer running ubuntu linux. I have, however, tried the same
> procedure on a similar computer running SUSE linux 9.3 and V2.12.2 of
> IRAF and had the same result.
>
> I have been trying to use asthedit to add the sidereal time keywork
> (st)
> to the headers of several fits spectra (these fits files are from the
> Ritter Observatory archive). The command file that I was using
> contained the line
>
> st = mst (@'date-obs', ut, '83.58')
>
> Running the command on a file (in verbose mode) produces the output
>
> File of commands (st.cmds):
> 930924.018.fits:
> $I = 930924.018.fits
> $D = 20/10/05
> $T = 17:06:08
> $GMD = 2005-10-20
> $GMT = 16:06:08
> $GMDT = 2005-10-20T16:06:08
> ERROR: segmentation violation
>
> I have used the same command (obviously with a look up of the
> different
> observatory's longitude) on fits spectra from other sites without any
> problem. If I enter the date by hand i.e. '24/09/93', it also works
> fine.
>
> I believed I have slightly narrowed down the problem to the way that
> DATE-OBS is stored in the header. In the Ritter observatory file
> it is
> stored as
>
> DATE-OBS= '24/09/93' /(character) date of data
> aquisition
>
> and when you try and access the keyword for example using the command
> file
>
> test = @'date-obs'
>
> you get
>
> test = 24.
>
> In fits spectra from the INT archive, where this task works fine,
> date-obs is stored as
>
> DATE-OBS= '03/09/91' /Date of the observation
>
> and running
>
> test = @'date-obs'
>
> produces
>
> test = 03/09/91
>
> i.e. it works fine. Would you know why the task works with one
> keyword
> format and not the other and, would you by any chance know how I
> can get
> around the problem without having to enter the date into the command
> file for each fits spectrum that I am looking at (there could be
> quite a
> few)? I am happy to send you example files if it would help but I
> decided that this would probably not be a wise option to begin with.
>
> Many thanks for your time and any help you can give me.
>
> Best wishes,
> Dugan Witherick
> School of Physics, Trinity College Dublin
>