struct2cell creates trailing singleton dimension

[David Bateman]

The struct2cell function creates a trailing singleton dimension when the 
struct array it is call with is scalar or a column vector For example

a.fld1 = 1;
a.fld2 = 2;
b = struct2cell(a);
size (b)
a(2).fld1 = 3;
a(2).fld2 = 4;
c = struct2cell(a);
size(c)
d = struct2cell(a.');
size(d)

The attached patch addresses this, though I can't easily apply the patch 
at the moment, so if someone else wants to  I'd appreciate it..

D.

--
David Bateman                                address@hidden
35 rue Gambetta                              +33 1 46 04 02 18 (Home)
92100 Boulogne-Billancourt FRANCE            +33 6 72 01 06 33 (Mob)

diff --git a/src/ChangeLog b/src/ChangeLog
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,3 +1,8 @@
+2008-04-08  David Bateman  <address@hidden>
+
+       * ov-cell.cc (Fstruct2cell): Treat possible trailing singleton for
+       creation of cell array from column structure vector.
+
 2008-04-03  David Bateman  <address@hidden>
 
        * DLD-FUNCTIONS/max.cc (MINMAX_SPARSE_BODY): Allow sparse boolean 
diff --git a/src/ov-cell.cc b/src/ov-cell.cc
--- a/src/ov-cell.cc
+++ b/src/ov-cell.cc
@@ -1357,9 +1357,13 @@
 
          // The resulting dim_vector should have dimensions:
          // [numel(fields) size(struct)]
+         // except if the struct is a column vector.
 
          dim_vector result_dv;
-         result_dv.resize (m_dv.length () + 1); // Add 1 for the fields.
+         if (m_dv (m_dv.length () - 1) == 1)
+             result_dv.resize (m_dv.length ());
+         else
+             result_dv.resize (m_dv.length () + 1); // Add 1 for the fields.
 
          result_dv(0) = num_fields;