All calls to "find" use the same generic implementation (bug #42408, 42421) * find.cc: Rewrite. Move generic "find" logic to find.h (Ffind) : Changed calls to find_nonzero_elem_idx to find_templated Added unit test for bug #42421 * Array.cc (and .h) (Array::find): Deleted function. Replaced with find::find(Array) from find.h * Array.h: Added typedef for array_iterator (in nz-iterators.h) as Array::iter_type * DiagArray2.h: Added typedef for diag_iterator (in nz-iterators.h) as DiagArray2::iter_type * PermMatrix.h: Added typedef for perm_iterator (in nz-iterators.h) as PermMatrix::iter_type Also added typedef for bool as PermMatrix::element_type (not octave_idx_type) Added an nnz() function (which is an alias for perm_length) and a perm_elem(i) function for retrieving the ith element of the permutation * Sparse.h: Added typedef for sparse_iterator (in nz-iterators.h) as Sparse::iter_type Added a short comment documenting the the argument to the numel function * idx-vector.cc (idx_vector::idx_mask_rep::as_array): Changed Array.find to find::find(Array) (in find.h) * (new file) find.h * (new file) interp-idx.h: Simple methods for converting between interpreter index type and internal octave_idx_type/row-col pair * (new file) min-with-nnz.h: Fast methods for taking an arbitrary matrix M and an octave_idx_type n and finding min(M.nnz(), n) * (new file) nz-iterators.h: Iterators for traversing (in column-major order) the nonzero elements of any array or matrix backwards or forwards * (new file) direction.h: Generic methods for simplifying code has to deal with a "backwards or forwards" template argument * build-sparse-tests.sh: Removed 5-return-value calls to "find" in unit-tests; Admittedly this commit breaks this "feature" which was undocumented and only partially supported to begin with (ie never worked for full matrices, permutation matrices, or diagonal matrices)

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/liboctave/util/interp-idx.h	Tue Jun 17 16:41:11 2014 -0600
@@ -0,0 +1,44 @@
+/*
+
+Copyright (C) 2014 David Spies
+
+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/>.
+
+*/
+#if !defined (octave_interp_idx_h)
+#define octave_interp_idx_h 1
+
+// Simple method for converting between C++ octave_idx_type and
+// Octave data-types (convert to double and add one).
+inline double
+to_interp_idx (octave_idx_type idx)
+{
+  return idx + 1.L;
+}
+
+// Simple method for taking a row-column pair together with the matrix
+// height and returning the corresponding index as an octave data-type
+// (note that for large heights, there's a risk of losing precision.
+// This method will not overflow or throw a bad alloc, it will simply
+// choose the nearest possible double-value to the proper index).
+inline double
+to_interp_idx (octave_idx_type row, octave_idx_type col, octave_idx_type height)
+{
+  return col * static_cast<double> (height) + row + 1;
+}
+
+#endif