Doc about new sparse __getitem__.

NEWS.txt

@@ -120,7 +120,8 @@ New features:
  * Added a_tensor.transpose(axes) axes is optional (James)
     * theano.tensor.transpose(a_tensor, kwargs) We where ignoring kwargs, now it is used as the axes.
  * a_CudaNdarray_object[*] = int, now work (Frederic)
-
+ * sparse_variable[N, N] now work (Li Yao, Frederic)
+ * sparse_variable[M:N, O:P] now work (Li Yao, Frederic)
 
 New optimizations:
  * AdvancedSubtensor1 reuses preallocated memory if available (scan, c|py_nogc linker) (Frederic)

doc/library/sparse/index.txt

@@ -53,9 +53,12 @@ grad?
     - When the operation has the form dot(csr_matrix, dense) the gradient of
       this operation can be performed inplace by UsmmCscDense. This leads to
       significant speed-ups.
-
-Subtensor selection (aka. square-bracket notation, aka indexing) is not implemented, but the
-CSR and CSC datastructures support effecient implementations.
+- Subtensor
+    - sparse_varible[N, N], return a tensor scalar
+    - sparse_varible[M:N, O:P], return a sparse matrix
+    - Don't support [M, N:O] and [M:N, O] as we don't support sparse vector
+      and returning a sparse matrix would break the numpy interface.
+      Use [M:M+1, N:O] and [M:N, O:O+1] instead.
 
 There are no GPU implementations for sparse matrices implemented in Theano.