Taken care of code review comments partially

theano/sparse/basic.py

@@ -8,9 +8,11 @@ http://www-users.cs.umn.edu/~saad/software/SPARSKIT/paper.ps
 # Automatic methods for determining best sparse format?
 
 import sys
+
 import numpy
 import theano
 import scipy.sparse
+
 from theano import gof, tensor, compile, scalar, config
 from theano.gof.python25 import all
 from theano.gradient import DisconnectedType
@@ -1754,8 +1756,8 @@ class AddSD(gof.op.Op):
                   %(z)s = (PyArrayObject *) PyArray_NewCopy(%(y)s, NPY_CORDER);
                 }else{
                   %(z)s = %(y)s;
+                  Py_XINCREF(%(z)s);
                 }
-                Py_XINCREF(%(z)s);
                 
                 npy_intp N =  PyArray_DIMS(%(_indptr)s)[0]-1;
                 const npy_int32 * __restrict__ indptr = (npy_int32 *)%(_indptr)s->data;

theano/tensor/basic.py

@@ -6593,11 +6593,7 @@ class ConstructSparse(Op):
 
     def connection_pattern(self, node):
 
-        rval = [[True], [True]]
-
-        for ipt in node.inputs[2:]:
-            rval.append([False])
-
+        rval = [[True], [True], [False]]
         return rval
 
     def grad(self, inputs, grads):
@@ -6700,11 +6696,7 @@ class AdvancedIncSubtensor1(Op):
 
     def connection_pattern(self, node):
 
-        rval = [[True], [True]]
-
-        for ipt in node.inputs[2:]:
-            rval.append([False])
-
+        rval = [[True], [True], [False]]
         return rval
 
     def grad(self, inputs, grads):