iraf.net

Hi folks,I just tried working with some 8-bit FITS images (raw IUE data) in
IRAF for the first time in quite a few months, and much to my surprise
tasks like display, minmax, and imstat don't seem to be able to
handle them correctly anymore. Display produces an image that has
a wide band of "wacky" pixel values across the middle of the frame,
and minmax and imstat think that some pixels have values that are
far outside of the 0-255 range theoretically possible for an 8-bit image
(e.g. -30000 to +30000 or so). If I use rfits to convert any of these
images to imh or hhh form, they are fine. So it appears that we have
a FITS kernel problem somewhere in v2.11.3, at least for Solaris platforms
(I haven't tried it yet on any others).Here's an excerpt of the top of the header of one these files:SIMPLE = T / Fits standard
BITPIX = 8 / Bits per pixel
NAXIS = 2 / Number of axes
NAXIS1 = 768 / Axis length
NAXIS2 = 768 / Axis length
EXTEND = F / File may contain extensions
IRAF-TLM= '15:42:56 (17/09/1998)' / Time of last modification
CTYPE1 = 'SAMPLE ' / x-axis
CTYPE2 = 'LINE ' / y-axis
BUNIT = 'DN ' / Data Numbers
TELESCOP= 'IUE ' / International Ultraviolet Explorer
FILENAME= 'LWP01455.RILO' / Filename(camera)(number).RI(disp)
DATE = '21/09/98' / Date file was written
ORIGIN = 'GSFC ' / Institution generating the file
DATAMIN = 0. / Minimum pixel value
DATAMAX = 255. / Maximum pixel value Hope you can help,
Howard

Hello Howard;
Yes, somebody else had the same problem and the problem is already
fix inhere. The only inmediate solution is for you to get an updated
version of the IRAF share library. For this I need to know the
OS version number you are running.Cheers, Nelson

>
> Hi folks,
>
> I just tried working with some 8-bit FITS images (raw IUE data) in
> IRAF for the first time in quite a few months, and much to my surprise
> tasks like display, minmax, and imstat don't seem to be able to
> handle them correctly anymore. Display produces an image that has
> a wide band of "wacky" pixel values across the middle of the frame,
> and minmax and imstat think that some pixels have values that are
> far outside of the 0-255 range theoretically possible for an 8-bit image
> (e.g. -30000 to +30000 or so). If I use rfits to convert any of these
> images to imh or hhh form, they are fine. So it appears that we have
> a FITS kernel problem somewhere in v2.11.3, at least for Solaris platforms
> (I haven't tried it yet on any others).
>
> Here's an excerpt of the top of the header of one these files:
>
> SIMPLE = T / Fits standard Hello again;
In case you are running either SUNSO 5.5 or 5.7 there are 2 S11 files
in iraf.noao.edu/pubS11_FK5.5.e
S11_FK5.7.egrab the file and then rename:mv S11_FK5.5.e S11_5.5.e
or
mv S11_FK5.7.e S11_5.7.eCopy the file for your version to some directory and then declare
a UNIX environment variable:setenv IRAFULIB <pathname_to_dir>/ # Notice the ending '/'Now start the cl and see if the 8 bit problem is gone.When the IRAF patch or release containing this fix you should
undeclared the IRAFULIB.Cheers
Nelson

Hello;
I am inclined to send them only the zfiofxf.x file that takes
care of this specific problem. The file is in /p5/iraft/zfiofxf.xThanks, N
> Hi Nelson,
>
> We would like to install the patches for the 8-bit FITS image problem
> into our institute-wide IRAF distributions. May I just replace one of
> the files in our distribution with your modified S11 file? We have had
> to re-build all of our executables in order to incorporate more IRAF
> nodes than the standard distribution allows, so perhaps the best way is
> to modify the source code and re-link rather than just replace an
> executable?
>
> Will the patch be incorporated in the next IRAF release? When might that be?
>
> Thanks
> Ed Colbert
> STSDAS administrator
>
> > From bushouse@stsci.edu Fri Mar 24 17:01:46 2000
> >
> > ------------- Begin Forwarded Message -------------
> >
> > From zarate@noao.edu Thu Mar 23 18:18 EST 2000
> > Date: Thu, 23 Mar 2000 16:17:47 -0700 (MST)
> > From: Nelson Zarate <zarate@noao.edu>
> > To: bushouse@stsci.edu
> > Subject: Re: 8-bit FITS image problems
> > Cc: sites@tucana.tuc.noao.edu
> >
> > >
> > > Hi folks,
> > >
> > > I just tried working with some 8-bit FITS images (raw IUE data) in
> > > IRAF for the first time in quite a few months, and much to my surprise
> > > tasks like display, minmax, and imstat don't seem to be able to
> > > handle them correctly anymore. Display produces an image that has
> > > a wide band of "wacky" pixel values across the middle of the frame,
> > > and minmax and imstat think that some pixels have values that are
> > > far outside of the 0-255 range theoretically possible for an 8-bit image
> > > (e.g. -30000 to +30000 or so). If I use rfits to convert any of these
> > > images to imh or hhh form, they are fine. So it appears that we have
> > > a FITS kernel problem somewhere in v2.11.3, at least for Solaris platforms
> > > (I haven't tried it yet on any others).
> > >
> > > Here's an excerpt of the top of the header of one these files:
> > >
> > > SIMPLE = T / Fits standard
> >
> > Hello again;
> > In case you are running either SUNSO 5.5 or 5.7 there are 2 S11 files
> > in iraf.noao.edu/pub
> >
> > S11_FK5.5.e
> > S11_FK5.7.e
> >
> > grab the file and then rename:
> >
> > mv S11_FK5.5.e S11_5.5.e
> > or
> > mv S11_FK5.7.e S11_5.7.e
> >
> > Copy the file for your version to some directory and then declare
> > a UNIX environment variable:
> >
> > setenv IRAFULIB <pathname_to_dir>/ # Notice the ending '/'
> >
> > Now start the cl and see if the 8 bit problem is gone.
> >
> > When the IRAF patch or release containing this fix you should
> > undeclared the IRAFULIB.
> >
> > Cheers
> > Nelson
> > ------------- End Forwarded Message -------------
> >
>