No worries, glad to see it is working.
Most of the transformations are defined only over the range (-180 -
180) in degrees longitude, that might be your problem. Try:
$coords -= 360*($coords>180);
just before the transformation.
Threading in PDL generally keeps the active dimensions of each
operation first, followed by extra thread dimensions -- so it should
not be too surprising that the 0 index should contain the vectors, and
the 1 index should run across vectors. But that can be hard to
remember.
I just had a look, and the documentation for apply() does indeed call
out "...a collection of N-vectors (with index in the 0th dimension)".
This may be just another case of stuff in the middle of documentation
being invisible -- happens all the time, I'm afraid. I remember being
astonished as my cohort (of graduate students at Stanford in the
1990s) all failed to see the instruction "For a tutorial, type control-
H T" that was right smack dab in the middle of the emacs splashscreen.
Best,
Craig
On Feb 3, 2009, at 12:01 PM, David Whysong wrote:
> Thanks, with that change, the transform appears to do something. Is
> there any documentation on setting up the vectors to be transformed?
>
> My map is still not correct. With input vectors having the range
> (0~360,-90~90), the output range is (0~4,-2~2) and the resulting image
> (see attachment) shows only the right half of an ellipse. The plane of
> the galaxy is near the edge, this does not appear to be correct. I'm
> toying with offsetting the inputs now, maybe that will help.
>
> David Whysong
> <radio2.jpg>
_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
