This is a useful function for debugging, and exactly like the function
`graph()`, except that it's intended for double precision floating
point numbers. It pops up a graph on the computer screen (using the
graphing program `xmgr`) showing a graph of some array which you
happen to want to look at.

The arguments are:

`array:`Input. The array that you want a graph of.`n:`Input. The number of array elements that you want to graph.`spacing:`Input. The spacing of the array elements that you want to graph. The elements graphed are`array[0]`,`array[spacing]`,`array[2*spacing]`,...,`array[(n-1)*spacing]`.

This function is a handy way to get a quick look at the contents of
some array. It writes the output to a temporary file and then starts
up `xmgr`, reading the input from the file. The values are
evenly spaced integers from to `n-1`. The values are the
(subset of) points in `array[ ]`. If your array contains real
data, you might want to use `spacing=1`. If your array contains
complex data (with real and imaginary parts interleaved) you will use
`spacing=2`, and make separate calls to see the real and imaginary
parts. For example if `complex[0..2047]` contains 1024 complex
numbers, then:
`graph(complex,1024,2)` (view 1024 real values)
`graph(complex+1,1024,2)` (view 1024 imaginary values)

Note that in order not to produce too much garbage on the screen,
any output or error messages from `xmgr` are tossed into
`/dev/null`!

- Authors: Bruce Allen, ballen@dirac.phys.uwm.edu
- Comments: None.