Seeing the algorithm of an IRAF program
rohit wrote on Oct 02, 2007
Thanks Fitz for your help in the previous question.
I would like to know how I could find the code written for a particular IRAF task. That is, how FXCOR task is written, that is see the code.
Please let me know.
Thanks!
I would like to know how I could find the code written for a particular IRAF task. That is, how FXCOR task is written, that is see the code.
Please let me know.
Thanks!
Mike Fitzpatrick wrote on Oct 02, 2007
You can find the source for FXCOR in the noao$rv directory. I suspect you'll be most interested in the rvrvcor.x source that does the correction, there is a lot more code that prepares the spectra for the FFT, etc.
-Mike
-Mike
rohit wrote on Oct 02, 2007
thanks, Mike.
I have emailed the support astronomer at KECK for the CDELT and CRVAL values. Suppose I get them, is it possible to change the header file to incorporate them? Is so, how do you add them in the header file? I think these are the ONLY two values I need to run the FXCOR without error, right? I feel I am so close to getting it, yet so far!! :-(
I have emailed the support astronomer at KECK for the CDELT and CRVAL values. Suppose I get them, is it possible to change the header file to incorporate them? Is so, how do you add them in the header file? I think these are the ONLY two values I need to run the FXCOR without error, right? I feel I am so close to getting it, yet so far!! :-(
Mike Fitzpatrick wrote on Oct 02, 2007
CRVAL/CDELT are a minimum WCS, the DC-FLAG is also used to tell IRAF the type of dispersion these represent (log, linear, etc) and things like RA/DEC, exposure time, etc are also used for the RV correction (see the KEYWPARS pset) but appear to be removed by the IDL script.
The HEDIT task can be used to add/edit keywords in an image, however your images still need to somehow be extraced to 1-D spectra before the task will work properly (unless you've got something like 90 orders in this spectrum??).
-Mike
The HEDIT task can be used to add/edit keywords in an image, however your images still need to somehow be extraced to 1-D spectra before the task will work properly (unless you've got something like 90 orders in this spectrum??).
-Mike
Last post on Oct 02, 2007