I have collected a few scripts and image sets in my Tdata directories. I store this work in the T_runs/ds9+xpa directory. I have some running notes there in:

 

    T_runs/ds9+xpa/ex3_ngc3379/S/README.ds9+xpa

 
 # Zoom to fit an (already displayed) image then plot regions in ifu.reg 
% xpaset -p ds9 zoom to fit                    
% cat ifu.reg | xpaset ds9 regions -format ds9 

# display image 1 and overplot regions file just created
% xpaset -p ds9 frame frameno 1                
% xpaset -p ds9 file $1                        
% xpaset -p ds9 scale limits $2 $3             

# Save marked regions to a file 
% xpaset -p ds9 regions save ds9_save2.reg     

# Delet all markers 
% xpaset -p ds9 regions delete all     

# Be sure you save physical coordinates
% xpaset -p ds9 regions system physical
% xpaset -p ds9 regions save Saved_Coordinates.physical.reg

# Lay out the 2 image viewing frames
% xpaset -p ds9 tile grid layout 2 1           
% xpaset -p ds9 tile yes                       

===========================================================
Here I show a useful example from a script. The closing 
of the window part is really hansdled simply by the 
"xpaset -p ds9 exit" command string. 

# close the ds9 window
printf "\nClose the ds9 window? (Y/n):"
read Final_Answer
if [ "$Final_Answer" = "Y" ]
then
printf "Deleting the ds9 window. Goodbye.\n"
 xpaset -p ds9 exit
else
printf "Goodbye.\n"
fi
===========================================================

# Adjust images to the WCS of frame1 (assumed image 
# in frame 1 has valid WCS, test with wcs_ask) 
wcs_ask $1 > wcs.present
read WCS_There < wcs.present
printf "WCS is present = $WCS_There \n"
if [ $WCS_There = "Y" ]
then
 xpaset -p ds9 frame frameno 1
 xpaset -p ds9 frame match wcs
fi
 

In Apr2016 I began working on my some ways to view multiple images. I want to put different images into different ds9 frames. I also modified the ds9_view script so that I could use a variety of ways to automatically set the image grey scale level. Here is an example of doing this:

 
 
% ds9_open 1000 700
% xpaset -p ds9 tile grid layout 2 1
% xpaset -p ds9 tile yes
% xpaset -p ds9 frame frameno 1
% xpaset -p ds9 frame frameno 2
% ds9_view Rsco2040.fits zscale n 1
% ds9_view Rsco2040.fits minmax n 2

Here is an example of opening a window with two frames. I display the same image using two different scalings. The following set of command lines were used for this: % ds9_open 1000 700 % xpaset -p ds9 tile grid layout 2 1 % xpaset -p ds9 tile yes % xpaset -p ds9 frame frameno 1 % xpaset -p ds9 frame frameno 2 % ds9_view Rsco2040.fits zscale n 1 % ds9_view Rsco2040.fits minmax n 2 Recall that ds9_open is my own little script for opening a ds9 window of a user-specified size and ds9_view is another script I wrote that handles the display of a given image.

I often want to view a portion of a big image. I want to specify the image location, usually by specifying a location in physical coordinates.

 
 

Places a crosshair marker on the position X,Y=870,551
% xpaset -p ds9 crosshair 870 551

The pan command is very useful: 
$xpaget ds9 pan                                   # get current image coords
$xpaset -p ds9 pan to 400 400 physical            # pan to physical coords What I really need!
$xpaset -p ds9 pan to 13:29:55 47:11:50 wcs fk5   # pan to wcs coords

Still have not figured thsi one out!

 
 

These change as I change the part of the image viewed:
% xpaget ds9 pan
 734.6942 576.35785       # tells me cnetr of viewed image
% xpaget ds9 zoom
26.623333                 # tells me the zoom factor

Sometimes the DATASEC keyword can obscure what we want to see.

 
 
Being sure you see ALL of the pixels. 
=====================================

What is the "size" of the image: 
% gethead 20161221T041713.7_066LL_sci.fits NAXIS1 NAXIS2 DATASEC BIASSEC
2128 1032 [1:2064,1:1032] [2065:2128,1:1032]

I simple way to see the image: 
 ds9 20161221T041713.7_066LL_sci.fits

An easy SCO wayto see the image on mcs:
 ds9_open 800 800
 ds9_view 20161221T041713.7_066LL_sci.fits zscale n 1

When I zoom into the top-right corner and measure 
the X,Y position of the corner pixel ====>   X,Y = 2064 1032   

Hence, ds9 does NOT show me the bias region!
Hoow do I see it? I do the following:
ANSWER: 
Select from the menu bar at top:  Scale > DATASEC 
>>> I see the image change size. The bias area is 
    now visible on the right side of the display. 

Soemtimes it is easier to bypass xpa and just run ds9 from the command line. Her are examples:

% ds9 -geometry -scale zscale -match scale *.fits 

*** I could use the "zoom to fit" command, but it is easier to show all of the 
    images, zoom in obe of them, then use the Frame>Match>Image option to re-zoom
    all of the displayed images. 
 

The point region types are useful for plotting regions with sizes that are smaller than an image pixel.

 

% cat ds9.reg
# Region file format: DS9 version 4.1
global color=green dashlist=8 3 width=1 font="helvetica 10 normal roman" select=1 highlite=1 dash=0 fixed=0 edit=1 move=1 delete=1 include=1 source=1
physical
point(425.89899,351.05168) # point=circle
point(428.26533,354.09412) # point=box
point(429.57997,351.23948) # point=cross

One thing I have wanted to know for a long time: waht is the name of the image (the FITS image name) read into a given frame. This seems so obvious, duh.....