iraf.net

Hello Stefano,I apologize for the delay in replying. We are not aware of any
specific bug though the pixel mask code is complex enough that there
is the possibility of a problem. Actually, there are two ways to
store masks, in .pl files and in .fits files with the type=mask
attribute. I think you are saying that you are working with the mask
stored in a FITS extension. There was a change in the structure of
the format used when storing masks in FITS extensions between the two
versions you mention. So this may be the source of the problem. Are
you trying to IMREPLACE masks created in the earlier version using
the newer version? In this case it may not work. If instead you
created the mask with 2.12.2a and then you get the error when using
IMREPLACE from 2.12.2a then this is something I would need to
reproduce. In the latter case could you send me or make available by
ftp or http a mask before you run IMREPLACE and tell me the command
you are using. Also what type of OS are you using?As a workaround you could probably convert or use the .pl form and
then IMCOPY it to a FITS file when you are done. Let me know if you
have questions or want my help to resolve this problem.Yours,
Frank Valdes
On Sep 2, 2005, at 6:19 AM, Stefano Berta wrote:> Dear IRAF support group,
>
> We write you because we are experiencing some
> troubles using IMREPLACE.
>
> We have IRAF 2.12.2a currently installed, and apparently Jan97
> version of
> the IMUTIL package.
>
> We are using IMREPLACE for masking, by directly modifying .fits
> object-masks, created with the OBJMASK task in the NPROTO (January
> 1992)
> package.
>
> When run, IMREPLACE seems to work correctly (no error message is
> prompted), but afterwards, when using the resulting masks, we have a
> "segmentation violation" message. Dysplaying the mask,
> it turns out that it is "broken", i.e. no pixels are shown.
> Even statistics fail (IMSTAT).
>
> On a previous installation (2.12.1 IRAF version)
> it all worked properly.
>
> Is this a known bug? Is there a solution?
> We looked on IRAF web page, but we only found a couple of reports
> about imreplacing on .pl images and about the usage of the "radius"
> parameter.
> Note that we are using radius=0 and we are working on .fits frames,
> hence these cases do not apply.
>
> Thank you very much indeed,
> best regards,
>
> Stefano Berta
> Stefano Rubele
>
>
> **********************************************************************
> *
> Stefano BERTA E-mail: berta@pd.astro.it
> Post-Doc,
> Astronomy Department
> University of Padova Phone : (+39) 049 8278250-51
> Vicolo dell'Osservatorio, 2 fax : (+39) 049 8278212
> I-35122 Padova, ITALY
>
> NEW homepage: http://dipastro.pd.astro.it/berta
> ESIS homepage: http://dipastro.pd.astro.it/esis
> **********************************************************************
> *
>
>
>