Show screen output only on rank 0

Otherwise distributed computations lead to a lot of cluttering.

src/finite-strain-elasticity.cc

@@ -185,8 +185,7 @@ int main (int argc, char *argv[]) try
   BoundaryPatch<GridView> dirichletBoundary(gridView, dirichletVertices);
   BoundaryPatch<GridView> neumannBoundary(gridView, neumannVertices);
 
-  if (mpiHelper.rank()==0)
-    std::cout << "Neumann boundary has " << neumannBoundary.numFaces() << " faces\n";
+  std::cout << "On process " << mpiHelper.rank() << ": Neumann boundary has " << neumannBoundary.numFaces() << " faces\n";
 
 
   BitSetVector<1> dirichletNodes(feBasis.size(), false);
@@ -257,6 +256,7 @@ int main (int argc, char *argv[]) try
       neumannFunction = std::make_shared<NeumannFunction>(parameterSet.get<FieldVector<double,dim> >("neumannValues"),
                                                      homotopyParameter);
 
+    if (mpiHelper.rank()==0)
       std::cout << "Neumann values: " << parameterSet.get<FieldVector<double,dim> >("neumannValues") << std::endl;
     }
 
@@ -267,6 +267,7 @@ int main (int argc, char *argv[]) try
     }
 
     // Assembler using ADOL-C
+    if (mpiHelper.rank()==0)
     std::cout << "Selected energy is: " << parameterSet.get<std::string>("energy") << std::endl;
     std::shared_ptr<Elasticity::LocalEnergy<GridView,
                                      FEBasis::LocalView::Tree::FiniteElement,