TRIANGULATE says "Points are co-linear, no solution"

[Jonathan Joseph <jj(at)scorpio.tn.cornell.edu>]

TRIANGULATE says "Points are co-linear, no solution"

This really bugs me, because it's just not true.
Well, it may be true for a subset of the points I'm
trying to triangulate, but certainly not all.

Is the IDL triangulate code just really weak?
I wish I had an example of this with a small
number of points.

I thought I should be able to get a triangulation with
just about any set of points, except for one in which
all of the points lie along the same line.  This is
certainly not true for my dataset - in fact, I think
it looks darned reasonable, and I want to
resample onto a regular grid dammit!

Anyway, if anyone wants to take a look, Please!

The data set is a file called test.dat (saved on hp-ux
using writeu.  I hope there are no byte swapping problems
for other platforms)

ftp://scorpio.tn.cornell.edu/jj/idl/

IDL> x=fltarr(9853)
IDL> y=fltarr(9853)
IDL> openr,unit,'test.dat',/get_lun
IDL> readu,unit,x,
IDL> triangulate,x,y,tr,b

% TRIANGULATE: Points are co-linear, no solution.

;; see the data (doesn't look bad to me)
IDL> plot,xx,yy,/device,psym=3

HELP!!

Thanks.

BTW, this was orignally a double array - that didn't work either.

-Jonathan