[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: num2cell
From: |
David Bateman |
Subject: |
Re: num2cell |
Date: |
Tue, 13 Sep 2005 14:34:52 +0200 |
User-agent: |
Mozilla Thunderbird 1.0.2 (Windows/20050317) |
Søren Hauberg a écrit :
After John's mail I've tried num2cell.cc and it works just fine for my
uses, but it has some oddities (is that a word?). I don't have access
to matlab (nor do I want it) so I don't know what the "correct"
behaviour is, but
num2cell("hello")
gives me a 1x5 cell of empty matrices ([]'s), which doesn't seem
right. Also if num2cell is called with a struct as argument, the
answer will be a cell containing an empty matrix.
I hope I'll be able to find some time this week to have a look at the
code...
Thanks,
Søren
Ok, then how about the attached that deals with everything in a generic
manner. The result however is that
a.b = 1; a.c = 2; num2cell(a)
error: can't perform indexing operations for struct type
which makes sense to me... I think matlab's behaviour is in error..
Another interesting odd-ball behaviour with matlab is
num2cell(gf([0,1;2,3],4))
which gives 3 empty cells!!! I'd call that a bug..
D.
/*
Copyright (C) 2005 David Bateman
This program is free software; you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by the
Free Software Foundation; either version 2, or (at your option) any
later version.
This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
for more details.
You should have received a copy of the GNU General Public License
along with this program; see the file COPYING. If not, write to the Free
Software Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
In addition to the terms of the GPL, you are permitted to link
this program with any Open Source program, as defined by the
Open Source Initiative (www.opensource.org)
*/
#include <octave/config.h>
#include <octave/oct.h>
#include <octave/Cell.h>
#include <octave/ov-str-mat.h>
#include <octave/ov-colon.h>
DEFUN_DLD (num2cell, args, ,
"-*- texinfo -*-\n\
@deftypefn {Loadable Function} address@hidden =} num2cell (@var{m})\n\
@deftypefnx {Loadable Function} address@hidden =} num2cell (@var{m}, @var{d})\n\
Convert to matrix @var{m} into a cell array. If @var{d} is defined the\n\
value @var{c} is of dimension 1 in this dimension and the elements of\n\
@var{m} are placed in slices in @var{c}.\n\
@end deftypefn\n\
@seealso{mat2cell}")
{
int nargin = args.length();
octave_value retval;
if (nargin < 1 || nargin > 2)
usage ("num2cell");
else
{
dim_vector dv = args(0).dims ();
Array<int> sings;
if (nargin == 2)
{
ColumnVector dsings = ColumnVector (args(1).vector_value
(false, true));
sings.resize (dsings.length());
if (!error_state)
for (int i = 0; i < dsings.length(); i++)
if (dsings(i) > dv.length() || dsings(i) < 1 ||
D_NINT(dsings(i)) != dsings(i))
{
error ("invalid dimension specified");
break;
}
else
sings(i) = NINT(dsings(i)) - 1;
}
if (! error_state)
{
Array<bool> idx_colon (dv.length());
dim_vector new_dv (dv);
octave_value_list lst (new_dv.length(), octave_value());
for (int i = 0; i < dv.length(); i++)
{
idx_colon(i) = false;
for (int j = 0; j < sings.length(); j++)
{
if (sings(j) == i)
{
new_dv(i) = 1;
idx_colon(i) = true;
lst(i) = octave_value (octave_value::magic_colon_t);
break;
}
}
}
Cell ret (new_dv);
octave_idx_type nel = new_dv.numel();
octave_idx_type ntot = 1;
for (int j = 0; j < new_dv.length()-1; j++)
ntot *= new_dv(j);
for (octave_idx_type i = 0; i < nel; i++)
{
octave_idx_type n = ntot;
octave_idx_type ii = i;
for (int j = new_dv.length() - 1; j >= 0 ; j--)
{
if (! idx_colon(j))
lst (j) = ii/n + 1;
ii = ii % n;
if (j != 0)
n /= new_dv(j-1);
}
ret(i) = args(0).do_index_op(lst, 0);
}
retval = ret;
}
}
return retval;
}
/*
;;; Local Variables: ***
;;; mode: C++ ***
;;; End: ***
*/