[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: griddata
From: |
John W. Eaton |
Subject: |
Re: griddata |
Date: |
Mon, 10 Nov 2003 19:19:21 -0600 |
On 10-Nov-2003, Paul Kienzle <address@hidden> wrote:
| There is a bug in octave introduced sometime between octave
| 2.1.40 and 2.1.48 where the wrong thing happens if the shape
| of a boolean indexing matrix does not match the shape of the
| matrix being indexed.
|
| E.g.,
|
| octave-2.1.40> x=[1,2;3,4]; x(:)'(x!=4)
| ans =
|
| 1 3 2
|
| octave-2.1.48> x=[1,2;3,4]; x(:)'(x!=4)
| ans =
|
| 1 2
| 3 5
|
| It is still present in octave-2.1.50.
I just checked in the following change to fix this problem. Because
of the extensive changes that have been made recently to support N-d
arrays, this patch will not apply directly to the 2.1.50 sources, but
it should give you an idea of what to do if you want to fix 2.1.50.
For that version, I think the function to look at should be
Array2<T>::index in Array2-idx.h.
jwe
2003-11-10 John W. Eaton <address@hidden>
* Array.cc (Array<T>::index2): Return value has orientation of
indexed value if indexing a vector with a bool matrix.
Index: liboctave/Array.cc
===================================================================
RCS file: /usr/local/cvsroot/octave/liboctave/Array.cc,v
retrieving revision 1.70
retrieving revision 1.71
diff -u -r1.70 -r1.71
--- liboctave/Array.cc 10 Nov 2003 17:18:48 -0000 1.70
+++ liboctave/Array.cc 11 Nov 2003 01:13:09 -0000 1.71
@@ -1656,7 +1656,8 @@
}
else
{
- if (idx_orig_rows == 1 || idx_orig_columns == 1)
+ if (idx_arg.one_zero_only ()
+ || idx_orig_rows == 1 || idx_orig_columns == 1)
{
if (nr == 1)
retval = Array<T> (tmp, dim_vector (1, len));
-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.
Octave's home on the web: http://www.octave.org
How to fund new projects: http://www.octave.org/funding.html
Subscription information: http://www.octave.org/archive.html
-------------------------------------------------------------
- griddata, Claudio Belotti, 2003/11/10
- Re: griddata, Paul Kienzle, 2003/11/10
- Re: griddata,
John W. Eaton <=