fit1d
Guest wrote on Apr 05, 2006
Hi - I'm trying to use the images.imfit.fit1d package to fit a function to 1 dimension of a 2d image - and it appears to work very well, throwing up an irafterm displaying the fit. The helpfile advises to 'q' out of the irafterm and then 'return' to create an output image. However at this stage the cursor returns to iraf with an "ERROR: segmentation violation" message and, more importantly, no image. Similar problems occur when attempting to use the program non-interactively and when using noao.imred.generic.flat1d. Does anyone have any clue how to fix this, or another package that could be used instead? Many thanks
Mike Fitzpatrick wrote on Apr 05, 2006
Please post more information such as the parameters used, ideally a link to the image file itself, and especially which platform and IRAF version you are using so we can try to reproduce the bug. A segvio typically indicates a coding error of some kind, but since you see it with both FIT1D and FLAT1D this may be in the underlying ICFIT library used by both tasks. The only know bug I could find is
http://iraf.net/article.php?story=7392&query=icfit
(Buglog 513), let us know if this doesn't apply to you.
Otherwise, the "See Also" section of the help page lists similar tasks or you could browse for tasks by keyword (e.g. 'function' or 'fit') using the Apropos button on the banner bar. Hope this helps.
Cheers,
-Mike
http://iraf.net/article.php?story=7392&query=icfit
(Buglog 513), let us know if this doesn't apply to you.
Otherwise, the "See Also" section of the help page lists similar tasks or you could browse for tasks by keyword (e.g. 'function' or 'fit') using the Apropos button on the banner bar. Hope this helps.
Cheers,
-Mike
Guest wrote on Apr 05, 2006
iraf = 2.12.2 used on suse 9.3, but I think I've pinned down the problem: if the file already exists or the system has a panic writing out the file (it takes a good 100 seconds) then you get a segmentation fault. I hope that helps and thank you.
Mike Fitzpatrick wrote on Apr 05, 2006
I still need the parameters and perhaps the image. Using the default params and a sequence like
ecl> fit1d dev$pix foo fit inter-
ecl> fit1d dev$pix foo fit inter-
(i.e. your comment that the image exists) works fine for me. How is your command any different (order, function, interactive use, samples, etc)? Can you demonstrate the problem using dev$pix as above or is it specific to the image that you're using?
-Mike
ecl> fit1d dev$pix foo fit inter-
ecl> fit1d dev$pix foo fit inter-
(i.e. your comment that the image exists) works fine for me. How is your command any different (order, function, interactive use, samples, etc)? Can you demonstrate the problem using dev$pix as above or is it specific to the image that you're using?
-Mike
Guest wrote on Apr 05, 2006
OK - your sequence worked for me too and now I can't even reproduce my problem with my image so possibly it was a computer hiccup here - my apologies.
Last post on Apr 05, 2006