From: Peter Munch <peterrmuench@gmail.com>
Date: Wed, 24 Jun 2020 05:23:54 +0000 (+0200)
Subject: Fix parallel::distributed::SolutionTransfer for FENothing
X-Git-Tag: v9.3.0-rc1~1383^2
X-Git-Url: https://gitweb.dealii.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=refs%2Fpull%2F10592%2Fhead;p=dealii.git

Fix parallel::distributed::SolutionTransfer for FENothing
---

diff --git a/doc/news/changes/minor/20200624SoldnerMunch b/doc/news/changes/minor/20200624SoldnerMunch
new file mode 100644
index 0000000000..3354fd86ec
--- /dev/null
+++ b/doc/news/changes/minor/20200624SoldnerMunch
@@ -0,0 +1,4 @@
+Fixed: The class parallel::distributed::SolutionTransfer can now
+also handle FE_Nothing.
+<br>
+(Dominic Soldner, Peter Munch, 2020/06/24)
diff --git a/source/distributed/solution_transfer.cc b/source/distributed/solution_transfer.cc
index 52ca51e57f..d4b7672851 100644
--- a/source/distributed/solution_transfer.cc
+++ b/source/distributed/solution_transfer.cc
@@ -365,6 +365,9 @@ namespace parallel
       const unsigned int dofs_per_cell =
         dof_handler->get_fe(fe_index).dofs_per_cell;
 
+      if (dofs_per_cell == 0)
+        return std::vector<char>(); // nothing to do for FE_Nothing
+
       auto it_input  = input_vectors.cbegin();
       auto it_output = dof_values.begin();
       for (; it_input != input_vectors.cend(); ++it_input, ++it_output)
@@ -437,6 +440,9 @@ namespace parallel
       const unsigned int dofs_per_cell =
         dof_handler->get_fe(fe_index).dofs_per_cell;
 
+      if (dofs_per_cell == 0)
+        return; // nothing to do for FE_Nothing
+
       const std::vector<::dealii::Vector<typename VectorType::value_type>>
         dof_values =
           unpack_dof_values<typename VectorType::value_type>(data_range,
diff --git a/tests/hp/distributed_solution_transfer_01.cc b/tests/hp/distributed_solution_transfer_01.cc
new file mode 100644
index 0000000000..bda8020149
--- /dev/null
+++ b/tests/hp/distributed_solution_transfer_01.cc
@@ -0,0 +1,116 @@
+// ---------------------------------------------------------------------
+//
+// Copyright (C) 2020 by the deal.II authors
+//
+// This file is part of the deal.II library.
+//
+// The deal.II library is free software; you can use it, redistribute
+// it, and/or modify it under the terms of the GNU Lesser General
+// Public License as published by the Free Software Foundation; either
+// version 2.1 of the License, or (at your option) any later version.
+// The full text of the license can be found in the file LICENSE.md at
+// the top level directory of deal.II.
+//
+// ---------------------------------------------------------------------
+
+
+
+// Test parallel::distributed::SolutionTransfer for FE_Nothing
+// (see also https://github.com/dealii/dealii/issues/10570).
+
+#include <deal.II/base/function.h>
+
+#include <deal.II/distributed/solution_transfer.h>
+#include <deal.II/distributed/tria.h>
+
+#include <deal.II/dofs/dof_accessor.h>
+#include <deal.II/dofs/dof_handler.h>
+#include <deal.II/dofs/dof_tools.h>
+
+#include <deal.II/fe/fe_nothing.h>
+#include <deal.II/fe/fe_q.h>
+
+#include <deal.II/grid/grid_generator.h>
+#include <deal.II/grid/grid_refinement.h>
+#include <deal.II/grid/tria.h>
+#include <deal.II/grid/tria_accessor.h>
+#include <deal.II/grid/tria_iterator.h>
+
+#include <deal.II/lac/la_parallel_vector.h>
+
+#include <iostream>
+#include <vector>
+
+#include "../tests.h"
+
+
+template <int dim>
+void
+transfer(const MPI_Comm &comm)
+{
+  AssertDimension(Utilities::MPI::n_mpi_processes(comm), 1);
+
+  parallel::distributed::Triangulation<dim> tria(comm);
+  GridGenerator::hyper_cube(tria);
+  tria.refine_global(1);
+
+  hp::FECollection<dim> fe;
+  fe.push_back(FE_Q<dim>(1));
+  fe.push_back(FE_Nothing<dim>());
+
+  // create a DoFHandler on which we have both cells with FE_Q as well as
+  // FE_Nothing
+  DoFHandler<dim> dof_handler(tria, true);
+  dof_handler.begin(0)->child(0)->set_active_fe_index(1);
+
+  IndexSet locally_relevant_dofs;
+  DoFTools::extract_locally_relevant_dofs(dof_handler, locally_relevant_dofs);
+
+  LinearAlgebra::distributed::Vector<double> solution(
+    dof_handler.locally_owned_dofs(), locally_relevant_dofs, comm);
+  AffineConstraints<double> cm;
+  cm.close();
+
+  dof_handler.distribute_dofs(fe);
+  solution.reinit(dof_handler.n_dofs());
+
+  for (unsigned int i = 0; i < solution.size(); ++i)
+    solution(i) = i;
+
+  parallel::distributed::
+    SolutionTransfer<dim, LinearAlgebra::distributed::Vector<double>>
+      soltrans(dof_handler);
+
+  for (const auto &cell : tria.active_cell_iterators())
+    cell->set_refine_flag();
+
+  LinearAlgebra::distributed::Vector<double> old_solution = solution;
+
+  tria.prepare_coarsening_and_refinement();
+  soltrans.prepare_for_coarsening_and_refinement(old_solution);
+  tria.execute_coarsening_and_refinement();
+  dof_handler.distribute_dofs(fe);
+  solution.reinit(dof_handler.n_dofs());
+
+  soltrans.interpolate(solution);
+
+  deallog << "OK" << std::endl;
+}
+
+
+int
+main(int argc, char *argv[])
+{
+  Utilities::MPI::MPI_InitFinalize mpi_initialization(argc, argv, 1);
+
+  const MPI_Comm comm = MPI_COMM_WORLD;
+  initlog();
+
+  deallog.push("2D solution transfer");
+  transfer<2>(comm);
+  deallog.pop();
+
+  deallog.push("3D solution transfer");
+  transfer<3>(comm);
+  deallog.pop();
+}
diff --git a/tests/hp/distributed_solution_transfer_01.mpirun=1.output b/tests/hp/distributed_solution_transfer_01.mpirun=1.output
new file mode 100644
index 0000000000..f368bb0bb4
--- /dev/null
+++ b/tests/hp/distributed_solution_transfer_01.mpirun=1.output
@@ -0,0 +1,3 @@
+
+DEAL:2D solution transfer::OK
+DEAL:3D solution transfer::OK