iraf.net

Below is a message I got from one of our user's of iraf-2.12--He is
using iraf-2.12 on an SGI and the iraf seems to come up fine-After
the email with the problems, I have included his current "login.cl
and what are other useful info to helo solve his problem.
--------------
Hi Vic --More stuff for the iraf gurus:So with my graphcap modified using the sed command you sent me, and
with a loginuser.cl file containing
reset graphcap=$home/dev/mygraphcap
I can now run saoimage from iraf in a clumsy way. That is, I can type
saoimage blotto
from within iraf, and an saoimage window pops up and shows the file blotto.
The 'display' task doesn't work, however, nor do any of the cursor readback
or subimage display functions that normally go with it.
'display' complains about a bad value
for the stdimage environment variable (which by default is set to some
imtool thing), but it doesn't like stdimage=saoimage either (it complains
that there is no 'saoimage' entry in the graphcap file), so I still don't
know how to make that work properly. Any clues would be appreciated.Also, I need to use the daophot package. This seems to mostly work, except
that with my data (from the 4shooter at Mt. Hopkins, if anybody cares),
all paths through the package (in particular the 'group' and 'allstar'
tasks) die with the complaint "Image header parameter not found (FUNCTION)".
I don't know if this is really a failing of the data headers, or how to
work around it, or where to find out. As I recall, there used to be a
"DAOPHOT in IRAF" manual, years ago. Does something like that exist, to
help with differences between the iraf and the free-standing versions?Thanks -- Tim-------------------------------
# LOGIN.CL -- User login file for the IRAF command language.# Identify login.cl version (checked in images.cl).
if (defpar ("logver"))
logver = "IRAF V2.12.1 July 2002"set home = "/tmp_mnt/home/timbrown/iraf/"
set imdir = "/local/d/timbrown/iraf/"
set uparm = "home$uparm/"
set userid = "timbrown"# Set the terminal type.
if (envget("TERM") == "xterm") {
if (!access (".hushiraf"))
print "setting terminal type to xgterm..."
stty xgterm
} else {
if (!access (".hushiraf"))
print "setting terminal type to winterm..."
stty winterm
} Uncomment and edit to change the defaults.
#set editor = vi
#set printer = lp
#set pspage = "letter"
#set stdimage = imt800
#set stdimcur = stdimage
#set stdplot = lw
#set clobber = no
#set filewait = yes
#set cmbuflen = 512000
#set min_lenuserarea = 64000
#set imtype = "imh"
#set imextn = "oif:imh fxf:fits,fit plf:pl qpf:qp stf:hhh,??h"
# IMTOOL/XIMAGE stuff. Set node to the name of your workstation to
# enable remote image display. The trailing "!" is required.
set node = "nadreck:0.0!"# CL parameters you mighth want to change.
ehinit = "nostandout eol noverify"
#epinit = "standout showall"
showtype = yes# Default USER package; extend or modify as you wish. Note that this can
# be used to call FORTRAN programs from IRAF.package usertask $adb $bc $cal $cat $comm $cp $csh $date $dbx $df $diff = "$foreign"
task $du $find $finger $ftp $grep $lpq $lprm $ls $mail $make = "$foreign"
task $man $mon $mv $nm $od $ps $rcp $rlogin $rsh $ruptime = "$foreign"
task $rwho $sh $spell $sps $strings $su $telnet $tip $top = "$foreign"
task $touch $vi $emacs $w $wc $less $rusers $sync $pwd $gdb = "$foreign"task $xc $mkpkg $generic $rtar $wtar $buglog = "$foreign"
#task $fc = "$xc -h $* -limfort -lsys -lvops -los"
task $fc = ("$" // envget("iraf") // "unix/hlib/fc.csh" //
" -h $* -limfort -lsys -lvops -los")
task $nbugs = ("$(setenv EDITOR 'buglog -e';" //
"less -Cqm +G " // envget ("iraf") // "local/bugs.*)")
task $cls = "$clear;ls"task $clw = "$clear;w"
task $pg = ("$(less -Cqm $*)")
task $saoimage = "$saoimage -idev ~timbrown/iraf/dev/imt1i -odev
~timbrown/iraf/dev/imt1o"if (access ("home$loginuser.cl"))
cl < "home$loginuser.cl"
;keep; clpackageprcache directory
cache directory page type help# Print the message of the day.
if (access (".hushiraf"))
menus = no
else {
clear; type hlib$motd
}# Delete any old MTIO lock (magtape position) files.
if (deftask ("mtclean"))
mtclean
else
delete uparm$mt?.lok,uparm$*.wcs verify-# List any packages you want loaded at login time, ONE PER LINE.
images # general image operators
plot # graphics tasks
dataio # data conversions, import export
lists # list processing# The if(deftask...) is needed for V2.9 compatibility.
if (deftask ("proto"))
proto # prototype or ad hoc taskstv # image display
utilities # miscellaneous utilities
noao # optical astronomy packages
digiphot
daophotkeep
-------------------------------
listing of user's home$iraf/uparm directorycletables.par datdataps.par datpstset.par pltcontor.par sympage.par
daorfits.par datfindps.par imlimhear.par pltprow.par
datdaofid.par datphot.par imlimstas.par symhelp.par-----------------------------------------
listing of user's home$iraf/dev directoryimt1i| imt1o| mygraphcap mygraphcap1-------------------------------------We went ahead and did the following to create the pipes and mygraphcap
file # Rob Seaman, NOAO CCS, November 20, 1990. mkdir dev
cd dev /usr/etc/mknod imt1i p
/usr/etc/mknod imt1o p
chmod 600 imt1o imt1i sed s+/dev/imt1+$cwd/imt1+g $iraf/dev/graphcap > mygraphcap exit
This script does not address the case of a user having mul-
tiple SAOimage windows running on the same machine. In that
case, you must create multiple sets of fifo pipes and modify
a private copy of graphcap more extensively. Contact the
HOTline for assistance with this or additional customiza-
tions.
--------------------
I am sure I am missing some custom stuff somewhere--Anyhow
any help would be great--I only install and don't know how
to use it and fell kinda worthless when comes to custom things.--VicVictor P. Tisone Computer System Management Team (CSMT)
tisone@ucar.edu voice: (303)497-1519 fax: (303)497-1589
High Altitude Observatory P.O. Box 3000 Boulder CO 80307 USA

Hi Vic,I have duplicated a few of the problems concerning the display, and
include my comments in your included email below.However, the fixes I propose work for ximtool, which can be downloaded
through the IRAF website. I tried saoimage, but without success.On Tue, 14 Jan 2003 16:14:26 -0700 (MST)
Vic Tisone <tisone@hao.ucar.edu> wrote:> Below is a message I got from one of our user's of iraf-2.12--He is
> using iraf-2.12 on an SGI and the iraf seems to come up fine-After
> the email with the problems, I have included his current "login.cl
> and what are other useful info to helo solve his problem.
> --------------
> Hi Vic --
>
> More stuff for the iraf gurus:
>
> So with my graphcap modified using the sed command you sent me, and
> with a loginuser.cl file containing
> reset graphcap=$home/dev/mygraphcapReplacing the above "reset graphcap" with the following should work. reset graphcap=home$dev/mygraphcapConcerning the script below, you should diff mygraphcap with
$iraf/dev/graphcap to see if there are differences. If there are
none, as I suspect, then you need to add the path to the devices you
created (below) on line 112-113:change:
imt1|imt512|imtool|Imtool display server:\
:cn#1:LC:BS@:z0#1:zr#200D=node!imtool,,512,512:tc=iism70:to:
imt1|imt512|imtool|Imtool display server:\
:cn#1:LC:BS@:z0#1:zr#200D=node!imtool,<path-to-dev>/imt1,512,512:tc=iism70:after these fixes, ximtool will work. I apologize for not sending a
fix to the saoimage problem.<snip>
> Also, I need to use the daophot package. This seems to mostly work,
> except that with my data (from the 4shooter at Mt. Hopkins, if
> anybody cares), all paths through the package (in particular the
> 'group' and 'allstar' tasks) die with the complaint "Image header
> parameter not found (FUNCTION)". I don't know if this is really a
> failing of the data headers, or how to work around it, or where to
> find out. As I recall, there used to be a"DAOPHOT in IRAF" manual,
> years ago. Does something like that exist, to help with differences
> between the iraf and the free-standing versions?The IRAF/DAOPHOT manual can be retrieved from:http://iraf.noao.edu/iraf/ftp/iraf/docs/daorefman.ps.ZConcerning the problem with "FUNCTION", it appears that it is looking
for one of the following fitting functions:moffat25, penny1, moffat15, penny2, lorentz, or auto(see page 50 of the daorefman.ps manual).I am not sure if this needs to be in the header. I will have to look
into this more.I hope the above fixes help. Let me know if I can be of further
assistance.Best,--Phil

Hi Vic,After further discussion with Mike Fitzpatrick, I thought I would pass
along a bit more information which I thought was very helpful.Mike mentioned that whenever a "reset" occurs, it always needs to be
followed by a "gflush" to reinitialize the graphics system and pick up
the change. Also, the stdimage should be one of the 'imt' entries in
the file, the script should have already edited the graphcap file with
the proper paths.Another issue with saoimage, is that the pipes are in the wrong order.
In Tim's login.cl file, you have:
task $saoimage = "$saoimage \
-idev ~timbrown/iraf/dev/imt1i \
-odev ~timbrown/iraf/dev/imt1o"which should be:task $saoimage = "$saoimage \
-idev ~timbrown/iraf/dev/imt1o \
-odev ~timbrown/iraf/dev/imt1i"as the idev flag takes 'imt1o' pipe and the odev takes the 'imt1i'
pipe. Another issue is the 'node' in the same file. You have:set node = "nadreck:0.0!"which should actually be:set node = "nadreck!"The 'node' is the iraf node name, where you have the DISPLAY setting
for X. It should just be "set node = nadreck!". The NETSTAT command
will say whether iraf networking is working in the first two lines,
otherwise this has no effect.As an alternative to the above modifications, you can just define setenv IMTDEV fifo:/$HOME/iraf/dev/imt1i:/$HOME/iraf/dev/imt1oin his .cshrc and iraf will display to these pipes and no special
graphcap is necessary.I will again mention XImtool, which was mentioned in my discussion
with Mike. XImtool doesn't require any of this special setup. It'll
use a unique connection by default. So, if the use of XImtool will
ease your administrating, you may wish to follow that path.Concerning the FUNCTION keyword, Mike also mentions:The FUNCTION keyword is added by one of the tasks for use by another
later on, most likely he just forgot a step somewhere and the doc
should set him straight.I hope this helps. Let us know if we can help any further.Regards,--Phil

Phil,
We tried the changes you send below and here is the response back
from the user.
Sorry about not being much help on this end--I don't really know
how to use iraf--I just install it and pass things on to you all.
Any help greatly appreciated.

Vic
-------------------------------------
Hi Vic --I've made the changes in login.cl and mygraphcap that were suggested
in the e-mails from Phil Warner on Jan 16, but I'm still missing something.When I try to execute the command 'display image_name', I now getERROR: Cannot open device (node!imtool,<path-to-dev>/imt1,512,512)This happens with or without the
setenv IMTDEV fifo:/$HOME/iraf/dev/imt1i:/$HOME/iraf/dev/imt1o
unix command being executed first.I am not necessarily wedded to saoimage, if it is hard to make it run
under our system. But it isn't clear to me how to make XImtool do its thing.
I note that my system has a working ximtool (spelled exactly that way),
but defaults or no, there's no evidence that IRAF knows anything about it.Sorry to be dense, but there are all sorts of things about how these
system components are supposed to talk to one another that I just don't
understand.Cheers -- Tim
-------------------
> Hi Vic,
>
> After further discussion with Mike Fitzpatrick, I thought I would pass
> along a bit more information which I thought was very helpful.
>
> Mike mentioned that whenever a "reset" occurs, it always needs to be
> followed by a "gflush" to reinitialize the graphics system and pick up
> the change. Also, the stdimage should be one of the 'imt' entries in
> the file, the script should have already edited the graphcap file with
> the proper paths.
>
> Another issue with saoimage, is that the pipes are in the wrong order.
> In Tim's login.cl file, you have:
>
>
> task $saoimage = "$saoimage \
> -idev ~timbrown/iraf/dev/imt1i \
> -odev ~timbrown/iraf/dev/imt1o"
>
> which should be:
>
> task $saoimage = "$saoimage \
> -idev ~timbrown/iraf/dev/imt1o \
> -odev ~timbrown/iraf/dev/imt1i"
>
> as the idev flag takes 'imt1o' pipe and the odev takes the 'imt1i'
> pipe. Another issue is the 'node' in the same file. You have:
>
> set node = "nadreck:0.0!"
>
> which should actually be:
>
> set node = "nadreck!"
>
> The 'node' is the iraf node name, where you have the DISPLAY setting
> for X. It should just be "set node = nadreck!". The NETSTAT command
> will say whether iraf networking is working in the first two lines,
> otherwise this has no effect.
>
> As an alternative to the above modifications, you can just define
>
> setenv IMTDEV fifo:/$HOME/iraf/dev/imt1i:/$HOME/iraf/dev/imt1o
>
> in his .cshrc and iraf will display to these pipes and no special
> graphcap is necessary.
>
> I will again mention XImtool, which was mentioned in my discussion
> with Mike. XImtool doesn't require any of this special setup. It'll
> use a unique connection by default. So, if the use of XImtool will
> ease your administrating, you may wish to follow that path.
>
> Concerning the FUNCTION keyword, Mike also mentions:
>
> The FUNCTION keyword is added by one of the tasks for use by another
> later on, most likely he just forgot a step somewhere and the doc
> should set him straight.
>
> I hope this helps. Let us know if we can help any further.
>
> Regards,
>
> --Phil
> --VicVictor P. Tisone Computer System Management Team (CSMT)
tisone@ucar.edu voice: (303)497-1519 fax: (303)497-1589
High Altitude Observatory P.O. Box 3000 Boulder CO 80307 USA

Hi Vic (and Tim),I can see an immediate problem; my comment is embedded in the quoted
reply:On Tue, 28 Jan 2003 13:50:58 -0700 (MST)
Vic Tisone <tisone@hao.ucar.edu> wrote:> Phil,
> We tried the changes you send below and here is the response
> back from the user.
> Sorry about not being much help on this end--I don't really know
> how to use iraf--I just install it and pass things on to you all.
> Any help greatly appreciated.
>
> Vic
> -------------------------------------
> Hi Vic --
>
> I've made the changes in login.cl and mygraphcap that were suggested
> in the e-mails from Phil Warner on Jan 16, but I'm still missing
> something.
>
> When I try to execute the command 'display image_name', I now get
>
> ERROR: Cannot open device (node!imtool,<path-to-dev>/imt1,512,512)display cannot access the device, since it needs to know where imt1 is
located. In other words, <path-to-dev> should be the actual path to
the "imt1" device. For example, if I create these devices in my home
directory, under "dev", the path would be:/home/pwarner/dev/imt1Hence, the appropriate entry in the "mygraphcap" file (or whatever you
called it) should look similar to:imt1|imt512|imtool|Imtool display server:\
:cn#1:LC:BS@:z0#1:zr#200D=node!imtool,/home/pwarner/dev/imt1,512,512:tc=iism70:This should fix that problem.Now, concerning the decision whether to use saoimage or ximtool, this
is entirely up to the user. As long as the custom configurations are
commented out, ximtool should work as normal.Saoimage requires fifo pipes (the /dev/imt* files) to allow
communication with iraf. When another user opens another saoimage,
the same fifo is used, and hence the image displays in the original
process. Ximtool uses a different method, in that it sets up a fifo
pipe in /tmp, and keeps the user-id in the name. For example, if my
user-id is 504, then the pipes that ximtool set up have "504" appended
to the filename. Hence, any number of users can open ximtool at the
same time, and it will recognize the user, and display the image to
the appropriate process.> This happens with or without the
> setenv IMTDEV fifo:/$HOME/iraf/dev/imt1i:/$HOME/iraf/dev/imt1o
> unix command being executed first.
>
> I am not necessarily wedded to saoimage, if it is hard to make it
> run under our system. But it isn't clear to me how to make XImtool
> do its thing. I note that my system has a working ximtool (spelled
> exactly that way), but defaults or no, there's no evidence that IRAF
> knows anything about it.
>
> Sorry to be dense, but there are all sorts of things about how these
> system components are supposed to talk to one another that I just
> don't understand.
>
> Cheers -- Tim
I hope that helps. Let us know if we can help any further.Cheers,--Phil

Phil,
Soory but we are still having problems--We have mad the changes
that we think we needed to do--Both Tim and I are not real good
with IRAF and its too bad that we don't have someone locally
that look over our shoulder and correct our mistakes or what we
assume needs to be done. Anyhow below is the error message Tim gets when
he trys "display image" is below
ERROR: Cannot open device (node!imtool,/home/timbrown/iraf/dev/imt1,512,512)Tim has the following directory structure /home/timbrown/iraf
/home/timbrown/iraf/dev

The device directory looks like the following:hagar) iraf : cd dev
(hagar) dev : ls -l
total 164
prw------- 1 timbrown afoe 0 Sep 24 1996 imt1i|
prw------- 1 timbrown afoe 0 Sep 24 1996 imt1o|
-rw-rw-r-- 1 timbrown afoe 32539 Sep 25 1996 mygraphcap
-rw-rw-r-- 1 timbrown afoe 50294 Jan 31 09:23 mygraphcap1Tim's login.cl looks like:# LOGIN.CL -- User login file for the IRAF command language.# Identify login.cl version (checked in images.cl).
if (defpar ("logver"))
logver = "IRAF V2.12.1 July 2002"set home = "/tmp_mnt/home/timbrown/iraf/"
set imdir = "/local/d/timbrown/iraf/"
set uparm = "home$uparm/"
set userid = "timbrown"# Set the terminal type.
if (envget("TERM") == "xterm") {
if (!access (".hushiraf"))
print "setting terminal type to xgterm..."
stty xgterm
} else {
if (!access (".hushiraf"))
print "setting terminal type to winterm..."
stty winterm
}# Uncomment and edit to change the defaults.
#set editor = vi
#set printer = lp
#set pspage = "letter"
#set stdimage = imt800
#set stdimcur = stdimage
#set stdplot = lw
#set clobber = no
#set filewait = yes
#set cmbuflen = 512000
#set min_lenuserarea = 64000
#set imtype = "imh"
#set imextn = "oif:imh fxf:fits,fit plf:pl qpf:qp stf:hhh,??h"
# IMTOOL/XIMAGE stuff. Set node to the name of your workstation to
# enable remote image display. The trailing "!" is required.
set node = "nadreck!"# CL parameters you mighth want to change.
#ehinit = "nostandout eol noverify"
#epinit = "standout showall"
showtype = yes# Default USER package; extend or modify as you wish. Note that this can
# be used to call FORTRAN programs from IRAF.package usertask $adb $bc $cal $cat $comm $cp $csh $date $dbx $df $diff = "$foreign"
task $du $find $finger $ftp $grep $lpq $lprm $ls $mail $make = "$foreign"
task $man $mon $mv $nm $od $ps $rcp $rlogin $rsh $ruptime = "$foreign"
task $rwho $sh $spell $sps $strings $su $telnet $tip $top = "$foreign"
task $touch $vi $emacs $w $wc $less $rusers $sync $pwd $gdb = "$foreign"task $xc $mkpkg $generic $rtar $wtar $buglog = "$foreign"
#task $fc = "$xc -h $* -limfort -lsys -lvops -los"
task $fc = ("$" // envget("iraf") // "unix/hlib/fc.csh" //
" -h $* -limfort -lsys -lvops -los")
task $nbugs = ("$(setenv EDITOR 'buglog -e';" //
"less -Cqm +G " // envget ("iraf") // "local/bugs.*)")
task $cls = "$clear;ls"
task $clw = "$clear;w"
task $pg = ("$(less -Cqm $*)")
task $saoimage = "$saoimage -idev ~timbrown/iraf/dev/imt1o -odev
~timbrown/iraf/dev/imt1i"
task $imtool = "$ximtool -idev ~timbrown/iraf/dev/imt1o -odev
~timbrown/iraf/dev/imt1i"if (access ("home$loginuser.cl"))
cl < "home$loginuser.cl"
;keep; clpackageprcache directory
cache directory page type help# Print the message of the day.
if (access (".hushiraf"))
menus = no
else {
clear; type hlib$motd
}# Delete any old MTIO lock (magtape position) files.
if (deftask ("mtclean"))
mtclean
else
delete uparm$mt?.lok,uparm$*.wcs verify-# List any packages you want loaded at login time, ONE PER LINE.
images # general image operators
plot # graphics tasks
dataio # data conversions, import export
lists # list processing# The if(deftask...) is needed for V2.9 compatibility.
if (deftask ("proto"))
proto # prototype or ad hoc taskstv # image display
utilities # miscellaneous utilities
noao # optical astronomy packages
digiphot
daophotkeep-------------------------------
Tim is currently using /home/timbrown/iraf/dev/mygraphcap1
which has the current entries for imt1 imt2
# STDIMAGE devices.imtx|Imtool display server:\
:cn#1:LC:BS@:z0#1:zr#200D=node!imtool,/tmp/imt1,512,512:tc=iism7
0:
imt1|imt512|imtool|Imtool display server:\
:cn#1:LC:BS@:z0#1:zr#200D=node!imtool,/home/timbrown/iraf/dev/imt1,512,5
12:tc=iism70:
# :cn#1:LC:BS@:z0#1:zr#200D=node!imtool,,512,512:tc=iism70:
imt2|imt800|:cn#2r#800:yr#800:\
:LC:BS@:z0#1:zr#200D=node!imtool,,800,800:tc=iism70:


---
The questions are what are we doing wrong in trying to use saoimage to work.
How to we set things up to use ximtool if we go that route--ximtool
is available on the SGIs--Thought we needed to stick it into the
login.cl as a task but looks like we did not do that right. I know
from your email that it runs different than saoimage--Do we just
put it in the task line along with the general UNIX commands like
"vi" without all the bells and whistles that saoimage needs which
looked to be implied.Would like to be able to use either saoimage or ximtool if
possible--Currently don't have ximtool loaded on the SOLARIS platform
here. We really appreicate the help.
Vic
--VicVictor P. Tisone Computer System Management Team (CSMT)
tisone@ucar.edu voice: (303)497-1519 fax: (303)497-1589
High Altitude Observatory P.O. Box 3000 Boulder CO 80307 USA

Hi Vic,On Fri, 31 Jan 2003 10:31:34 -0700 (MST)
Vic Tisone <tisone@hao.ucar.edu> wrote:> Phil,
> Soory but we are still having problems--We have mad the changes
> that we think we needed to do--Both Tim and I are not real good
> with IRAF and its too bad that we don't have someone locally
> that look over our shoulder and correct our mistakes or what we
> assume needs to be done. Anyhow below is the error message Tim gets
> when he trys "display image" is below
>
>
> ERROR: Cannot open device
> (node!imtool,/home/timbrown/iraf/dev/imt1,512,512)I am not sure how to correct this problem, as I have tried the same,
and got the same results.
<snip>> task $imtool = "$ximtool -idev ~timbrown/iraf/dev/imt1o -odev
> ~timbrown/iraf/dev/imt1i"ximtool does not require a specific set of pipes, as these are set up
automatically in the system tmp directory (e.g., /tmp). Hence you can
change the task to: task $imtool = "$!ximtool"which you could also call task $ximtool = "$!ximtool"depending upon the user's preference.
<snip>> ---
> The questions are what are we doing wrong in trying to use saoimage
> to work. How to we set things up to use ximtool if we go that
> route--ximtool is available on the SGIs--Thought we needed to stick
> it into the login.cl as a task but looks like we did not do that
> right. I know from your email that it runs different than
> saoimage--Do we just put it in the task line along with the general
> UNIX commands like"vi" without all the bells and whistles that
> saoimage needs which looked to be implied.Ximtool is a UNIX application, and will run just like saoimage, but
without the FIFO requirements, as it sets up its own, specific to the
user. See the above comment for how to set up the task in IRAF.> Would like to be able to use either saoimage or ximtool if
> possible--Currently don't have ximtool loaded on the SOLARIS
> platform here. We really appreicate the help.You can get the x11iraf (which includes xgterm and ximtool) binaries
from the IRAF FTP site, iraf.noao.edu in the directory /iraf/x11iraf.The package you want (for SOLARIS) is one of eitherx11iraf-v1.3.1-bin.ssun.tar.Zorx11iraf-v1.3.1-bin.ssun.tar.gzI hope that helps.Best,--Phil

Phil,
Here is the latest try -below--Should he take out the line in
his mygrapcap1 file that he has below.


imtx|Imtool display server:\
:cn#1:LC:BS@:z0#1:zr#200D=node!imtool,/tmp/imt1,512,512:tc=iism70:
imt1|imt512|imtool|Imtool display server:\
:cn#1:LC:BS@:z0#1:zr#200D=node!imtool,/home/timbrown/iraf/dev/imt1,512,512:tc=
iism70:In addition when I try "ximtool 0132.aur1.fits" by itself on the SGIS I
get the following message:
Warning: No frame buffer configuration file found.
Warning: cannot open /dev/imt1oalso--Do I need make a /dev/imt1o in the /dev area on the
system that needs to run ximtool?
------------- Begin Forwarded Message -------------From: "Tim Brown" <timbrown@hao.ucar.edu>
Date: Fri, 31 Jan 2003 13:44:06 -0700
To: Vic Tisone <tisone@hao.ucar.edu>
Subject: Re: iraf stuff
Mime-Version: 1.0Hi Vic --So my login.cl file now contains the line
task $imtool = "$!ximtool"I now run cl, and move to a directory with a fits-extension image file
in it (it's 4-CCD mosaic data, called 0132.aur1.fits). To this point, I have
done nothing but the cd. Then I get the following things:display 0132.aur1.fits[1] yields an error:
ERROR: Cannot open device (node!imtool,/home/timbrown/iraf/dev/imt1,512,512)
imtool 0132.aur1.fits[1] can't find the image, because the [1] syntax
isn't recognized:
ximtool: File does not exist: '0132.aur1.fits[1]'
imtool complains about colormaps, but also something to do
with pipes and frame buffers. It pops an ximtool
window, but I can't do anything with it because
my images are not 2D. I don't return to the cl
prompt until I kill the ximtool window. Output:Warning: Cannot allocate colormap entry for "gray63"
Warning: Cannot allocate colormap entry for "royalBlue3"
Warning: Cannot allocate colormap entry for "mediumVioletRed"
Warning: Cannot allocate colormap entry for "blue2"
Warning: Cannot allocate colormap entry for "purple4"
Warning: Cannot allocate colormap entry for "gray65"
Warning: Cannot allocate colormap entry for "Gold"
Warning: No frame buffer configuration file found.
Warning: cannot open /dev/imt1oSo I'm still stumped.Cheers -- Tim
------------- End Forwarded Message -------------
--VicVictor P. Tisone Computer System Management Team (CSMT)
tisone@ucar.edu voice: (303)497-1519 fax: (303)497-1589
High Altitude Observatory P.O. Box 3000 Boulder CO 80307 USA

Hi Vic,I apologize for not forseeing this problem. You can use the same pipes that you create in the user's directory.Change the ximtool command (in IRAF) to: task $imtool = "$!ximtool -fifo /home/timbrown/iraf/dev/imt1"Also, take out the dependence upon the custom graphcap, i.e., either
remove or comment out the lines in loginuser.cl which remap the
graphcap to the custom graphcap.
Let me know if we can do anything more.Cheers,--Phil

Phil,
Here is the forwarded email as to what happen after Tim made
the suggested changes--Below
is now what he has in his login.cl file

task $imtool = "$!ximtool -fifo /home/timbrown/iraf/dev/imt1"

Do have a question--Do we need and any entries in the systems
/dev area for the system--There are none for imt0 or imt1 there.
Also when I try to use ximtool on the image on an SGI, I get the below
warnings and also that the image(brown.fits) is not a 2 dimensional
image--The window does appear.

(samms) tisone : ximtool brown.fits
Warning: No frame buffer configuration file found.
Warning: cannot open /dev/imt1oVic

------------- Begin Forwarded Message -------------From: "Tim Brown" <timbrown@hao.ucar.edu>
Date: Fri, 31 Jan 2003 16:17:16 -0700
To: Vic Tisone <tisone@hao.ucar.edu>
Subject: Re: iraf stuff
Mime-Version: 1.0Hi Vic --Still no good. I have changed the imtool task in my login.cl, and commented
out the reference to mygraphcap1 in the loginuser.cl file. I change
directories as before, do a display, and getcl> display 0132.aur1.fits[2]
ERROR: Cannot open device (node!imtool,,512,512)All the other symptoms look the same, except there were no colortable
complaints this time. I bet this has to do with trying (or not) to run imtool
and IDL at the same time.Cheers -- Tim------------- End Forwarded Message -------------
--VicVictor P. Tisone Computer System Management Team (CSMT)
tisone@ucar.edu voice: (303)497-1519 fax: (303)497-1589
High Altitude Observatory P.O. Box 3000 Boulder CO 80307 USA

Hi Vic,If other users on the system are going to use ximtool, then I would
suggest you create the pipes in /dev. This would remove any
complaints ximtool has about them, and you can run ximtool without
specifying the pipe.In fact, after doing this, you may want to remove the previous pipes,
as they will not be needed.As far as IDL is concerned, the presence of the colortable messages
may have something to do with IDL, as I believe IDL uses the same
GUI toolkit. I could be wrong.Also, be sure ximtool is running before displaying the image, as the
"Cannot open device" message is also sent when no display is detected.I hope this fixes your problems. Let us know.Cheers,--Phil

Phil,
Finally making some progess-_We got both the saoimage and ximtool working
below are some additional questions and comments on getting things to work
or better understand as to what is going on.

Vic
------------- Begin Forwarded Message -------------From: "Tim Brown" <timbrown@hao.ucar.edu>
Date: Wed, 5 Feb 2003 10:45:06 -0700
To: Vic Tisone <tisone@hao.ucar.edu>
Subject: Re: iraf stuff
Mime-Version: 1.0Hi Vic --So you can pass it along to the IRAF folks --The error message I get when starting ximtool is
Warning: No frame buffer configuration file found.Also, both when using ximtool and saoimage, I am evidently able to display
only the central 512 x 512 pixels of the image. The entire image is not
shown in the thumbnail displays, and I cannot pan or zoom in such a way
as to make it display. I suppose this has to do somehow with the graphcap
entry (which seems to define the device as 512 pix square), but maybe not.Finally, the cursor readouts on both ximtool and saoimage report data values
for the pixel under the cursor, but only if the value lies within the
grayscale range that appears on the plot. So if I place the cursor in the
center of a star, I typically get a value read back that says ">566",
or some such, where I infer that data values above 566 map to pure white.
It looks as if IRAF is pre-scaling the image before shipping it to the
image routine, since fiddling with the color table once the image has gotten
to the image tool doesn't change anything. Is there a way to prevent
(or control) this behavior?Thanks -- Tim------------- End Forwarded Message -------------
--VicVictor P. Tisone Computer System Management Team (CSMT)
tisone@ucar.edu voice: (303)497-1519 fax: (303)497-1589
High Altitude Observatory P.O. Box 3000 Boulder CO 80307 USA

Hi Vic (and Tim),My comments are below.On Wed, 5 Feb 2003 10:49:15 -0700 (MST)
Vic Tisone <tisone@hao.ucar.edu> wrote:> Phil,
> Finally making some progess-_We got both the saoimage and ximtool
> working below are some additional questions and comments on getting
> things to work or better understand as to what is going on.
>
> Vic
> ------------- Begin Forwarded Message -------------
>
> From: "Tim Brown" <timbrown@hao.ucar.edu>
> Date: Wed, 5 Feb 2003 10:45:06 -0700
> To: Vic Tisone <tisone@hao.ucar.edu>
> Subject: Re: iraf stuff
> Mime-Version: 1.0
>
> Hi Vic --
>
> So you can pass it along to the IRAF folks --
>
> The error message I get when starting ximtool is
> Warning: No frame buffer configuration file found.That warning means a /usr/local/lib/imtoolrc link (to dev$imtoolrc),
or a ".imtoolrc" in your unix login directory, couldn't be found on
the system. This file is used to set the various frame buffer sizes
other than 512x512. To fix this problem, You can
copy the dev$imtoolrc file to your unix login directory to ".imtoolrc"
and restart ximtool as another workaround for this.You can also fix this error by issuing the command:ln -s $iraf/dev/imtoolrc /usr/local/lib/imtoolrcwhere $iraf is the path to the local iraf installation.This problem was an issue in an earlier IRAF install script. I
believe this was fixed in the v2.12 patch release (16 July 2002).>
> Also, both when using ximtool and saoimage, I am evidently able to
> display only the central 512 x 512 pixels of the image. The entire
> image is not shown in the thumbnail displays, and I cannot pan or
> zoom in such a way as to make it display. I suppose this has to do
> somehow with the graphcap entry (which seems to define the device as
> 512 pix square), but maybe not.If displaying with iraf (i.e., disp xxx.fits), you can set the "fill"
display parameter to "yes", e.g., using "epar disp", (fill = yes) scale image to fit display windowIf displaying with the command-line (i.e., ximtool xxx.fits), you can
resize the main view windows with the yellow square located inside the
thumbnail. Put the cursor over a corner of this square, click and
drag.> Finally, the cursor readouts on both ximtool and saoimage report
> data values for the pixel under the cursor, but only if the value
> lies within the grayscale range that appears on the plot. So if I
> place the cursor in the center of a star, I typically get a value
> read back that says ">566", or some such, where I infer that data
> values above 566 map to pure white. It looks as if IRAF is
> pre-scaling the image before shipping it to the image routine, since
> fiddling with the color table once the image has gotten to the image
> tool doesn't change anything. Is there a way to prevent(or control)
> this behavior?I am not sure how to prevent/control this behavior.You can look at pixel values in ximtool (Options->Control Panel then
the Coords tab, and finally "Pix Table"), but seems to give similar
results.You can also use imexamine to show the pixel values (type "a" over the
pixel when using imexamine).I hope this helps. Sorry I could not offer more help with the pixel
values.Regards,--Phil

Hi Vic and Tim,Concerning the following issue, I have received more information.> Finally, the cursor readouts on both ximtool and saoimage report
> data values for the pixel under the cursor, but only if the value
> lies within the grayscale range that appears on the plot. So if I
> place the cursor in the center of a star, I typically get a value
> read back that says ">566", or some such, where I infer that data
> values above 566 map to pure white. It looks as if IRAF is
> pre-scaling the image before shipping it to the image routine, since
> fiddling with the color table once the image has gotten to the image
> tool doesn't change anything. Is there a way to prevent(or control)
> this behavior?
>
Tim is correct that the scaling is done on the IRAF side and that what
you see in the readout is an approximation based on the z1/z2 linear
transform done to scale the pixels to 8-bits. To get the actual pixel
values the display server would need access to the image itself.
SAOimage and DS9 when used in standalone mode to display the image
will readout with the proper values and coordinates, but will still
give only the approximation when the image is display from iraf. The
latest XImtool V1.3 release along with IRAF V2.12 has changes which
allow ximtool to do the pixel-value and WCS readouts either in
standalone mode or from iraf. This is done using a small plugin to
access the images along with extra information now being passed along
with the WCS when displaying the image. DS9 could easily pick up the
new protocol changes but it might be more difficult to make use of the
new information without implementing the same kind of plugin. Prior
to this release ximtool only ever printed out the approximate value.For the new ximtool to do this you must have the ism_wcspix.e binary
in your path. It is typically installed in the same location as the
ximtool binary. If you are running an old version of ximtool, you
should consider upgrading.We can thank Mike Fitzpatrick yet again for pointing this out.Hope that helps.Cheers,--Phil