Re: off by one error in poisson_grid_fill

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Thu May 30 2013 - 14:36:41 MDT

Hi Andrew,

You are correct. The fix has been checked in
and will be part of the next release of NCL (6.2.0).

If you need an updated version let us know.


On 5/29/13 1:20 PM, Andrew Dawson wrote:
> I think there is a small error in the source for the subroutine poisxy2,
> which is called by poisson_grid_fill. The error is only present for cyclic
> grids.
> Line 118 of fill_msg_grid.f reads:
> if (i.eq.1 .and. gtype.eq.1) im1 = il-1
> I think this should be:
> if (i.eq.1 .and. gtype.eq.1) im1 = il
> When determining the longitude points contributing to the Laplacian of a
> point on the left-most edge of a cyclic grid, the correct points are the
> right-most point at the same latitude and the point to the right of the
> current one. The code as it stands uses the second right-most value instead
> of the right-most value, probably just a typing error. I hope that makes
> sense!
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
Received on Thu May 30 14:36:48 2013

This archive was generated by hypermail 2.1.8 : Tue Jun 04 2013 - 09:50:17 MDT