Mercurial > octave-nkf

changeset 18883:aa9ca67f09fb

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
make all permutation matrices column permutations (bug #42418) Making all permutation matrices column permutations allows permutation matrices to be iterated over in column-major order just like any other matrix. * PermMatrix.h, PermMatrix.cc (PermMatrix::_colp): Delete member variable and all uses. (PermMatrix::transpose): Create new matrix with internal representation flipped. (PermMatrix::pos_power): New function. (PermMatrix::eye): Call the one-argument constructor which already defaults to the identity matrix. (PermMatrix::is_col_perm): Unconditionally return true. (PermMatrix::is_row_perm): Unconditionally return false. (PermMatrix::data, PermMatrix::fortran_vec, PermMatrix::pvec): Delete. (PermMatrix::col_perm_vec): New function. * lu.cc: New test. * base-lu.cc (base_lu<lu_type>::base_lu): Call transpose in ipvt initialization. * find.cc, kron.cc, pr-output.cc, ov-perm.cc, PermMatrix.cc, PermMatrix.h, Sparse.cc, dMatrix.cc, fMatrix.cc, CmplxQRP.cc, base-lu.cc, dbleQRP.cc, fCmplxQRP.cc, floatQRP.cc, Sparse-perm-op-defs.h, mx-op-defs.h: Adapt to PermMatrix changes.
author David Spies <dnspies@gmail.com>
date Wed, 18 Jun 2014 19:38:40 -0600
parents 3d33fe79816c
children 479d1d3cb5c3
files libinterp/corefcn/find.cc libinterp/corefcn/kron.cc libinterp/corefcn/lu.cc libinterp/corefcn/pr-output.cc libinterp/octave-value/ov-perm.cc liboctave/array/PermMatrix.cc liboctave/array/PermMatrix.h liboctave/array/Sparse.cc liboctave/array/dMatrix.cc liboctave/array/fMatrix.cc liboctave/numeric/CmplxQRP.cc liboctave/numeric/base-lu.cc liboctave/numeric/dbleQRP.cc liboctave/numeric/fCmplxQRP.cc liboctave/numeric/floatQRP.cc liboctave/operators/Sparse-perm-op-defs.h liboctave/operators/mx-op-defs.h
diffstat 17 files changed, 143 insertions(+), 193 deletions(-) [+]
line wrap: on
line diff
--- a/libinterp/corefcn/find.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/libinterp/corefcn/find.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -263,33 +263,15 @@
 
   if (count > 0)
     {
-      const octave_idx_type* p = v.data ();
-      if (v.is_col_perm ())
+      const Array<octave_idx_type>& p = v.col_perm_vec ();
+      for (octave_idx_type k = 0; k < count; k++)
         {
-          for (octave_idx_type k = 0; k < count; k++)
-            {
-              OCTAVE_QUIT;
-              const octave_idx_type j = start_nc + k;
-              const octave_idx_type i = p[j];
-              i_idx(k) = static_cast<double> (1+i);
-              j_idx(k) = static_cast<double> (1+j);
-              idx(k) = j * nc + i + 1;
-            }
-        }
-      else
-        {
-          for (octave_idx_type k = 0; k < count; k++)
-            {
-              OCTAVE_QUIT;
-              const octave_idx_type i = start_nc + k;
-              const octave_idx_type j = p[i];
-              // Scatter into the index arrays according to
-              // j adjusted by the start point.
-              const octave_idx_type koff = j - start_nc;
-              i_idx(koff) = static_cast<double> (1+i);
-              j_idx(koff) = static_cast<double> (1+j);
-              idx(koff) = j * nc + i + 1;
-            }
+          OCTAVE_QUIT;
+          const octave_idx_type j = start_nc + k;
+          const octave_idx_type i = p(j);
+          i_idx(k) = static_cast<double> (1+i);
+          j_idx(k) = static_cast<double> (1+j);
+          idx(k) = j * nc + i + 1;
         }
     }
   else
--- a/libinterp/corefcn/kron.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/libinterp/corefcn/kron.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -136,39 +136,18 @@
 {
   octave_idx_type na = a.rows ();
   octave_idx_type nb = b.rows ();
-  const octave_idx_type *pa = a.data ();
-  const octave_idx_type *pb = b.data ();
-  PermMatrix c(na*nb); // Row permutation.
-  octave_idx_type *pc = c.fortran_vec ();
-
-  bool cola = a.is_col_perm ();
-  bool colb = b.is_col_perm ();
-  if (cola && colb)
-    {
-      for (octave_idx_type i = 0; i < na; i++)
-        for (octave_idx_type j = 0; j < nb; j++)
-          pc[pa[i]*nb+pb[j]] = i*nb+j;
-    }
-  else if (cola)
+  const Array<octave_idx_type>& pa = a.col_perm_vec ();
+  const Array<octave_idx_type>& pb = b.col_perm_vec ();
+  Array<octave_idx_type> res_perm;
+  octave_idx_type rescol = 0;
+  for (octave_idx_type i = 0; i < na; i++)
     {
-      for (octave_idx_type i = 0; i < na; i++)
-        for (octave_idx_type j = 0; j < nb; j++)
-          pc[pa[i]*nb+j] = i*nb+pb[j];
-    }
-  else if (colb)
-    {
-      for (octave_idx_type i = 0; i < na; i++)
-        for (octave_idx_type j = 0; j < nb; j++)
-          pc[i*nb+pb[j]] = pa[i]*nb+j;
-    }
-  else
-    {
-      for (octave_idx_type i = 0; i < na; i++)
-        for (octave_idx_type j = 0; j < nb; j++)
-          pc[i*nb+j] = pa[i]*nb+pb[j];
+      octave_idx_type a_add = pa(i) * nb;
+      for (octave_idx_type j = 0; j < nb; j++)
+        res_perm.xelem (rescol++) = a_add + pb(j);
     }
 
-  return c;
+  return PermMatrix (res_perm, true);
 }
 
 template <class MTA, class MTB>
--- a/libinterp/corefcn/lu.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/libinterp/corefcn/lu.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -842,6 +842,20 @@
 %! assert (norm (vec (P'*L*U - A - u*v.'), Inf) < norm (A)*1e1*eps);
 %!
 %!testif HAVE_QRUPDATE_LUU
+%! [L,U,P] = lu (A);
+%! [~,ordcols] = max (P,[],1);
+%! [~,ordrows] = max (P,[],2);
+%! P1 = eye (size(P))(:,ordcols);
+%! P2 = eye (size(P))(ordrows,:);
+%! assert(P1 == P);
+%! assert(P2 == P);
+%! [L,U,P] = luupdate (L,U,P,u,v);
+%! [L,U,P1] = luupdate (L,U,P1,u,v);
+%! [L,U,P2] = luupdate (L,U,P2,u,v);
+%! assert(P1 == P);
+%! assert(P2 == P);
+%!
+%!testif HAVE_QRUPDATE_LUU
 %! [L,U,P] = lu (Ac);
 %! [L,U,P] = luupdate (L,U,P,uc,vc);
 %! assert (norm (vec (tril (L)-L), Inf) == 0);
--- a/libinterp/corefcn/pr-output.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/libinterp/corefcn/pr-output.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -2485,11 +2485,10 @@
 
       if (pr_as_read_syntax)
         {
-          Array<octave_idx_type> pvec = m.pvec ();
-          bool colp = m.is_col_perm ();
+          Array<octave_idx_type> pvec = m.col_perm_vec ();
 
           os << "eye (";
-          if (colp) os << ":, ";
+          os << ":, ";
 
           octave_idx_type col = 0;
           while (col < nc)
@@ -2520,7 +2519,6 @@
               else
                 os << " ...\n";
             }
-          if (! colp) os << ", :";
           os << ")";
         }
       else
--- a/libinterp/octave-value/ov-perm.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/libinterp/octave-value/ov-perm.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -261,9 +261,9 @@
 octave_perm_matrix::save_ascii (std::ostream& os)
 {
   os << "# size: " << matrix.rows () << "\n";
-  os << "# orient: " << (matrix.is_col_perm () ? 'c' : 'r') << '\n';
+  os << "# orient: c\n";
 
-  Array<octave_idx_type> pvec = matrix.pvec ();
+  Array<octave_idx_type> pvec = matrix.col_perm_vec ();
   octave_idx_type n = pvec.length ();
   ColumnVector tmp (n);
   for (octave_idx_type i = 0; i < n; i++) tmp(i) = pvec(i) + 1;
@@ -314,11 +314,12 @@
 {
 
   int32_t sz = matrix.rows ();
-  bool colp = matrix.is_col_perm ();
+  bool colp = true;
   os.write (reinterpret_cast<char *> (&sz), 4);
   os.write (reinterpret_cast<char *> (&colp), 1);
-  os.write (reinterpret_cast<const char *> (matrix.data ()),
-                                            matrix.byte_size ());
+  const Array<octave_idx_type>& col_perm = matrix.col_perm_vec ();
+  os.write (reinterpret_cast<const char *> (col_perm.data ()),
+                                            col_perm.byte_size ());
 
   return true;
 }
--- a/liboctave/array/PermMatrix.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/array/PermMatrix.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -36,8 +36,8 @@
     ("PermMatrix: invalid permutation vector");
 }
 
-PermMatrix::PermMatrix (const Array<octave_idx_type>& p, bool colp, bool check)
-  : Array<octave_idx_type> (p), _colp(colp)
+void
+PermMatrix::setup (const Array<octave_idx_type>& p, bool colp, bool check)
 {
   if (check)
     {
@@ -47,12 +47,28 @@
           Array<octave_idx_type>::operator = (Array<octave_idx_type> ());
         }
     }
+
+  if (!colp)
+    *this = this->transpose ();
+}
+
+PermMatrix::PermMatrix (const Array<octave_idx_type>& p)
+  : Array<octave_idx_type> (p)
+{
+  setup (p, false, true);
 }
 
-PermMatrix::PermMatrix (const idx_vector& idx, bool colp, octave_idx_type n)
-  : Array<octave_idx_type> (), _colp(colp)
+PermMatrix::PermMatrix (const Array<octave_idx_type>& p, bool colp, bool check)
+  : Array<octave_idx_type> (p)
+{
+  setup (p, colp, check);
+}
+
+void
+PermMatrix::setup (const idx_vector& idx, bool colp, octave_idx_type n)
 {
   octave_idx_type len = idx.length (n);
+
   if (! idx.is_permutation (len))
     gripe_invalid_permutation ();
   else
@@ -61,12 +77,28 @@
       for (octave_idx_type i = 0; i < len; i++) idxa(i) = idx(i);
       Array<octave_idx_type>::operator = (idxa);
     }
+
+  if (!colp)
+    *this = this->transpose ();
+}
+
+PermMatrix::PermMatrix (const idx_vector& idx)
+  : Array<octave_idx_type> ()
+{
+    setup (idx, false, 0);
+}
+
+PermMatrix::PermMatrix (const idx_vector& idx, bool colp, octave_idx_type n)
+  : Array<octave_idx_type> ()
+{
+    setup (idx, colp, n);
 }
 
 PermMatrix::PermMatrix (octave_idx_type n)
-  : Array<octave_idx_type> (dim_vector (n, 1)), _colp (false)
+  : Array<octave_idx_type> (dim_vector (n, 1))
 {
-  for (octave_idx_type i = 0; i < n; i++) xelem (i) = i;
+  for (octave_idx_type i = 0; i < n; i++)
+    xelem (i) = i;
 }
 
 octave_idx_type
@@ -86,8 +118,13 @@
 PermMatrix
 PermMatrix::transpose (void) const
 {
-  PermMatrix retval (*this);
-  retval._colp = ! retval._colp;
+  octave_idx_type len = Array<octave_idx_type>::length ();
+
+  PermMatrix retval (len);
+
+  for (octave_idx_type i = 0; i < len; ++i)
+    retval.xelem (xelem (i)) = i;
+
   return retval;
 }
 
@@ -133,17 +170,20 @@
 }
 
 PermMatrix
-PermMatrix::power (octave_idx_type m) const
+PermMatrix::power(octave_idx_type m) const
+{
+  if (m < 0)
+    return inverse ().pos_power (-m);
+  else if (m > 0)
+    return pos_power (m);
+  else
+    return PermMatrix (rows ());
+}
+
+PermMatrix
+PermMatrix::pos_power (octave_idx_type m) const
 {
   octave_idx_type n = rows ();
-  bool res_colp = _colp;
-  if (m < 0)
-    {
-      res_colp = ! res_colp;
-      m = -m;
-    }
-  else if (m == 0)
-    return PermMatrix (n);
 
   const octave_idx_type *p = data ();
   Array<octave_idx_type> res_pvec (dim_vector (n, 1), -1);
@@ -176,43 +216,29 @@
 
     }
 
-  return PermMatrix (res_pvec, res_colp, false);
+  return PermMatrix (res_pvec, true, false);
 }
 
 PermMatrix
 PermMatrix::eye (octave_idx_type n)
 {
-  Array<octave_idx_type> p (dim_vector (n, 1));
-  for (octave_idx_type i = 0; i < n; i++)
-    p(i) = i;
-
-  return PermMatrix (p, false, false);
+  return PermMatrix (n);
 }
 
 PermMatrix
 operator *(const PermMatrix& a, const PermMatrix& b)
 {
-  const Array<octave_idx_type> ia = a.pvec ();
-  const Array<octave_idx_type> ib = b.pvec ();
   PermMatrix r;
+
+  const Array<octave_idx_type> ia = a.col_perm_vec ();
+  const Array<octave_idx_type> ib = b.col_perm_vec ();
+
   octave_idx_type n = a.columns ();
+
   if (n != b.rows ())
     gripe_nonconformant ("operator *", n, n, b.rows (), b.rows ());
-  else if (a._colp == b._colp)
-    {
-      r = PermMatrix ((a._colp
-                       ? ia.index (idx_vector (ib))
-                       : ib.index (idx_vector (ia))), a._colp, false);
-    }
   else
-    {
-      Array<octave_idx_type> ra (dim_vector (n, 1));
-      if (a._colp)
-        ra.assign (idx_vector (ia), ib);
-      else
-        ra.assign (idx_vector (ib), ia);
-      r = PermMatrix (ra, a._colp, false);
-    }
+    r = PermMatrix (ia.index (idx_vector (ib)), true, false);
 
   return r;
 }
--- a/liboctave/array/PermMatrix.h	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/array/PermMatrix.h	Wed Jun 18 19:38:40 2014 -0600
@@ -31,20 +31,21 @@
 
 class OCTAVE_API PermMatrix : protected Array<octave_idx_type>
 {
-
 public:
 
-  PermMatrix (void) : Array<octave_idx_type> (), _colp (false) { }
+  PermMatrix (void) : Array<octave_idx_type> () { }
 
   PermMatrix (octave_idx_type n);
 
-  PermMatrix (const Array<octave_idx_type>& p, bool colp = false,
-              bool check = true);
+  PermMatrix (const Array<octave_idx_type>& p) GCC_ATTR_DEPRECATED;
+
+  PermMatrix (const Array<octave_idx_type>& p, bool colp, bool check = true);
 
-  PermMatrix (const PermMatrix& m)
-    : Array<octave_idx_type> (m), _colp(m._colp) { }
+  PermMatrix (const PermMatrix& m) : Array<octave_idx_type> (m) { }
 
-  PermMatrix (const idx_vector& idx, bool colp = false, octave_idx_type n = 0);
+  PermMatrix (const idx_vector& idx) GCC_ATTR_DEPRECATED;
+
+  PermMatrix (const idx_vector& idx, bool colp, octave_idx_type n = 0);
 
   octave_idx_type dim1 (void) const
   { return Array<octave_idx_type>::length (); }
@@ -68,15 +69,13 @@
 
   dim_vector dims (void) const { return dim_vector (dim1 (), dim2 ()); }
 
-  Array<octave_idx_type> pvec (void) const
+  const Array<octave_idx_type>& col_perm_vec (void) const
   { return *this; }
 
   octave_idx_type
   elem (octave_idx_type i, octave_idx_type j) const
   {
-    return (_colp
-            ? ((Array<octave_idx_type>::elem (j) == i) ? 1 : 0)
-              : ((Array<octave_idx_type>::elem (i) == j) ? 1 : 0));
+    return (Array<octave_idx_type>::elem (j) == i) ? 1 : 0;
   }
 
   octave_idx_type
@@ -102,20 +101,8 @@
   // Efficient integer power of a permutation.
   PermMatrix power (octave_idx_type n) const;
 
-  bool is_col_perm (void) const { return _colp; }
-  bool is_row_perm (void) const { return !_colp; }
-
-  friend OCTAVE_API PermMatrix operator *(const PermMatrix& a,
-                                          const PermMatrix& b);
-
-  const octave_idx_type *data (void) const
-  { return Array<octave_idx_type>::data (); }
-
-  const octave_idx_type *fortran_vec (void) const
-  { return Array<octave_idx_type>::fortran_vec (); }
-
-  octave_idx_type *fortran_vec (void)
-  { return Array<octave_idx_type>::fortran_vec (); }
+  bool is_col_perm (void) const { return true; }
+  bool is_row_perm (void) const { return false; }
 
   void print_info (std::ostream& os, const std::string& prefix) const
   { Array<octave_idx_type>::print_info (os, prefix); }
@@ -123,12 +110,17 @@
   static PermMatrix eye (octave_idx_type n);
 
 private:
-  bool _colp;
+
+  PermMatrix pos_power (octave_idx_type m) const;
+
+  void setup (const Array<octave_idx_type>& p, bool colp, bool check);
+
+  void setup (const idx_vector& idx, bool colp, octave_idx_type n);
 };
 
 // Multiplying permutations together.
 PermMatrix
 OCTAVE_API
-operator *(const PermMatrix& a, const PermMatrix& b);
+operator * (const PermMatrix& a, const PermMatrix& b);
 
 #endif
--- a/liboctave/array/Sparse.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/array/Sparse.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -61,18 +61,10 @@
   for (octave_idx_type i = 0; i <= n; i++)
     cidx (i) = i;
 
-  const Array<octave_idx_type> pv = a.pvec ();
-
-  if (a.is_row_perm ())
-    {
-      for (octave_idx_type i = 0; i < n; i++)
-        ridx (pv(i)) = i;
-    }
-  else
-    {
-      for (octave_idx_type i = 0; i < n; i++)
-        ridx (i) = pv(i);
-    }
+  const Array<octave_idx_type> pv = a.col_perm_vec ();
+
+  for (octave_idx_type i = 0; i < n; i++)
+    ridx (i) = pv(i);
 
   for (octave_idx_type i = 0; i < n; i++)
     data (i) = 1.0;
--- a/liboctave/array/dMatrix.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/array/dMatrix.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -259,14 +259,10 @@
 Matrix::Matrix (const PermMatrix& a)
   : MArray<double> (a.dims (), 0.0)
 {
-  const Array<octave_idx_type> ia (a.pvec ());
+  const Array<octave_idx_type> ia (a.col_perm_vec ());
   octave_idx_type len = a.rows ();
-  if (a.is_col_perm ())
-    for (octave_idx_type i = 0; i < len; i++)
-      elem (ia(i), i) = 1.0;
-  else
-    for (octave_idx_type i = 0; i < len; i++)
-      elem (i, ia(i)) = 1.0;
+  for (octave_idx_type i = 0; i < len; i++)
+    elem (ia(i), i) = 1.0;
 }
 
 // FIXME: could we use a templated mixed-type copy function here?
--- a/liboctave/array/fMatrix.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/array/fMatrix.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -259,14 +259,10 @@
 FloatMatrix::FloatMatrix (const PermMatrix& a)
   : MArray<float> (a.dims (), 0.0)
 {
-  const Array<octave_idx_type> ia (a.pvec ());
+  const Array<octave_idx_type> ia (a.col_perm_vec ());
   octave_idx_type len = a.rows ();
-  if (a.is_col_perm ())
-    for (octave_idx_type i = 0; i < len; i++)
-      elem (ia(i), i) = 1.0;
-  else
-    for (octave_idx_type i = 0; i < len; i++)
-      elem (i, ia(i)) = 1.0;
+  for (octave_idx_type i = 0; i < len; i++)
+    elem (ia(i), i) = 1.0;
 }
 
 // FIXME: could we use a templated mixed-type copy function here?
--- a/liboctave/numeric/CmplxQRP.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/numeric/CmplxQRP.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -103,7 +103,7 @@
 RowVector
 ComplexQRP::Pvec (void) const
 {
-  Array<double> pa (p.pvec ());
+  Array<double> pa (p.col_perm_vec ());
   RowVector pv (MArray<double> (pa) + 1.0);
   return pv;
 }
--- a/liboctave/numeric/base-lu.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/numeric/base-lu.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -30,7 +30,7 @@
 template <class lu_type>
 base_lu<lu_type>::base_lu (const lu_type& l, const lu_type& u,
                            const PermMatrix& p)
-  : a_fact (u), l_fact (l), ipvt (p.pvec ())
+  : a_fact (u), l_fact (l), ipvt (p.transpose ().col_perm_vec ())
 {
   if (l.columns () != u.rows ())
     (*current_liboctave_error_handler) ("lu: dimension mismatch");
--- a/liboctave/numeric/dbleQRP.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/numeric/dbleQRP.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -100,7 +100,7 @@
 RowVector
 QRP::Pvec (void) const
 {
-  Array<double> pa (p.pvec ());
+  Array<double> pa (p.col_perm_vec ());
   RowVector pv (MArray<double> (pa) + 1.0);
   return pv;
 }
--- a/liboctave/numeric/fCmplxQRP.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/numeric/fCmplxQRP.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -103,7 +103,7 @@
 FloatRowVector
 FloatComplexQRP::Pvec (void) const
 {
-  Array<float> pa (p.pvec ());
+  Array<float> pa (p.col_perm_vec ());
   FloatRowVector pv (MArray<float> (pa) + 1.0f);
   return pv;
 }
--- a/liboctave/numeric/floatQRP.cc	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/numeric/floatQRP.cc	Wed Jun 18 19:38:40 2014 -0600
@@ -100,7 +100,7 @@
 FloatRowVector
 FloatQRP::Pvec (void) const
 {
-  Array<float> pa (p.pvec ());
+  Array<float> pa (p.col_perm_vec ());
   FloatRowVector pv (MArray<float> (pa) + 1.0f);
   return pv;
 }
--- a/liboctave/operators/Sparse-perm-op-defs.h	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/operators/Sparse-perm-op-defs.h	Wed Jun 18 19:38:40 2014 -0600
@@ -67,17 +67,7 @@
       return SM ();
     }
 
-  if (p.is_row_perm ())
-    {
-      // Form the column permutation and then call the colpm_sm routine.
-      const octave_idx_type *prow = p.pvec ().data ();
-      OCTAVE_LOCAL_BUFFER (octave_idx_type, pcol, nr);
-      for (octave_idx_type i = 0; i < nr; ++i)
-        pcol[prow[i]] = i;
-      return octinternal_do_mul_colpm_sm (pcol, a);
-    }
-  else
-    return octinternal_do_mul_colpm_sm (p.pvec ().data (), a);
+  return octinternal_do_mul_colpm_sm (p.col_perm_vec ().data (), a);
 }
 
 template <typename SM>
@@ -163,10 +153,7 @@
       return SM ();
     }
 
-  if (p.is_row_perm ())
-    return octinternal_do_mul_sm_rowpm (a, p.pvec ().data ());
-  else
-    return octinternal_do_mul_sm_colpm (a, p.pvec ().data ());
+  return octinternal_do_mul_sm_colpm (a, p.col_perm_vec ().data ());
 }
 
 #endif // octave_Sparse_perm_op_defs_h
--- a/liboctave/operators/mx-op-defs.h	Thu Jun 19 10:35:39 2014 -0700
+++ b/liboctave/operators/mx-op-defs.h	Wed Jun 18 19:38:40 2014 -0600
@@ -606,13 +606,8 @@
     gripe_nonconformant ("operator *", p.rows (), p.columns (), nr, nc); \
   else \
     { \
-      if (p.is_col_perm ()) \
-        { \
-          result = M (nr, nc); \
-          result.assign (p.pvec (), idx_vector::colon, x); \
-        } \
-      else \
-        result = x.index (p.pvec (), idx_vector::colon); \
+      result = M (nr, nc); \
+      result.assign (p.col_perm_vec (), idx_vector::colon, x); \
     } \
   \
   return result; \
@@ -627,15 +622,7 @@
   if (p.rows () != nc) \
     gripe_nonconformant ("operator *", nr, nc, p.rows (), p.columns ()); \
   else \
-    { \
-      if (p.is_col_perm ()) \
-        result = x.index (idx_vector::colon, p.pvec ()); \
-      else \
-        { \
-          result = M (nr, nc); \
-          result.assign (idx_vector::colon, p.pvec (), x); \
-        } \
-    } \
+    result = x.index (idx_vector::colon, p.col_perm_vec ()); \
   \
   return result; \
 }