--- a/scripts/ChangeLog	Tue Jan 26 23:10:24 2010 +0200
+++ b/scripts/ChangeLog	Tue Jan 26 23:27:21 2010 -0500
@@ -1,3 +1,7 @@
+2010-01-26  John W. Eaton  <jwe@octave.org>
+
+	* general/chop.m: New function.
+
 2010-01-26  Jaroslav Hajek  <highegg@gmail.com>
 
 	* optimization/fsolve.m: Support TypicalX, autoscale only if

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scripts/general/chop.m	Tue Jan 26 23:27:21 2010 -0500
@@ -0,0 +1,79 @@
+## Copyright (C) 2010 John W. Eaton
+##
+## This file is part of Octave.
+##
+## Octave 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 3 of the License, or (at
+## your option) any later version.
+##
+## Octave 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 Octave; see the file COPYING.  If not, see
+## <http://www.gnu.org/licenses/>.
+
+## -*- texinfo -*-
+## @deftypefn {Function File} chop (@var{x}, @var{d}, @var{u})
+## Truncate elements of @var{x} to @var{d} digits such that the
+## resulting digits are exactly divisible by @var{u}.  If @var{u} is not
+## specified it defaults to 10.
+##
+## @example
+## @group
+## chop (-pi, 5, 10)
+##      @result{} -3.14200000000000
+## chop (-pi, 5, 5)
+##      @result{} -3.14150000000000
+## @end group
+## @end example
+## @end deftypefn
+
+function retval = chop (x, digits, units = 10)
+
+  if (nargin == 2 || nargin == 3)
+    tmp = abs (x);
+
+    ## Avoid computing log (0).
+    tmp(x == 0) = 1;
+
+    ## Digits to the left of the decimal.
+    tmp = floor (log10 (tmp) + 1);
+
+    ## The expression
+    ##
+    ##   round (x .* inflate)
+    ##
+    ## produces an integer that contains the digits we want to keep.
+    ## Multiplying by deflate puts the decimal back where it belngs.
+    ##
+    ## Further scaling and rounding with the units factor produces a
+    ## value with digits exactly divisible by units.  We skip that step
+    ## unless units was explicitly provided.
+
+    inflate = 10 .^ (digits - tmp);
+    deflate = 10 .^ (tmp - digits);
+    if (nargin == 2)
+      retval = deflate .* round (x .* inflate);
+    else
+      retval = units .* deflate .* round (round (x .* inflate) ./ units);
+    endif
+  else
+    print_usage ();
+  endif
+
+endfunction
+
+%!assert (chop (e, 3), 2.72)
+%!assert (chop (e, 4), 2.718)
+%!assert (chop (e, 4, 5), 2.72)
+%!assert (chop (e, 4, 7), 2.716)
+%!assert (chop (-e, 3), -2.72)
+%!assert (chop (-e, 4), -2.718)
+%!assert (chop (-e, 4, 5), -2.72)
+%!assert (chop (-e, 4, 7), -2.716)
+%!assert (chop (hilb (3), 3), [1,.5,.333;.5,.333,.25;.333,.25,.2])
+%!assert (chop (hilb (3), 2, 7), [.7,.49,.35;.49,.35,.28;.35,.28,.21], 2*eps)