Public Member Functions | Protected Member Functions | Protected Attributes | List of all members
oomph::RefineableTetgenMesh< ELEMENT > Class Template Reference

#include <refineable_tetgen_mesh.template.h>

+ Inheritance diagram for oomph::RefineableTetgenMesh< ELEMENT >:

Public Member Functions

 RefineableTetgenMesh (TetMeshFacetedClosedSurface *const &outer_boundary_pt, Vector< TetMeshFacetedSurface * > &internal_closed_surface_pt, const double &element_volume, TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper, const bool &use_attributes=false, const bool &split_corner_elements=false, Vector< double > *const &target_element_volume_in_region_pt=nullptr)
 Build mesh, based on a TetMeshFacetedClosedSurface that specifies the outer boundary of the domain and any number of internal closed curves, specified by TetMeshFacetedSurfaces. Also specify target volume for uniform element size. More...
 
virtual ~RefineableTetgenMesh ()
 Empty Destructor. More...
 
void refine_uniformly (DocInfo &doc_info)
 Refine mesh uniformly and doc process. More...
 
unsigned unrefine_uniformly ()
 Unrefine mesh uniformly: Return 0 for success, 1 for failure (if unrefinement has reached the coarsest permitted level) More...
 
void adapt (const Vector< double > &elem_error)
 Adapt mesh, based on elemental error provided. More...
 
bool projection_is_disabled ()
 Is projection of old solution onto new mesh disabled? More...
 
void disable_projection ()
 Disable projection of old solution onto new mesh. More...
 
void enable_projection ()
 Disable projection of old solution onto new mesh. More...
 
- Public Member Functions inherited from oomph::TetgenMesh< ELEMENT >
 TetgenMesh ()
 Empty constructor. More...
 
 TetgenMesh (const std::string &node_file_name, const std::string &element_file_name, const std::string &face_file_name, TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper, const bool &use_attributes=false)
 Constructor with the input files. More...
 
 TetgenMesh (tetgenio &tetgen_data, TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper, const bool &use_attributes=false)
 Constructor with tetgenio data structure. More...
 
 TetgenMesh (const std::string &node_file_name, const std::string &element_file_name, const std::string &face_file_name, const bool &split_corner_elements, TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper, const bool &use_attributes=false)
 Constructor with the input files. Setting the boolean flag to true splits "corner" elements, i.e. elements that that have at least three faces on a domain boundary. The relevant elements are split without introducing hanging nodes so the sons have a "worse" shape than their fathers. However, this step avoids otherwise-hard-to-diagnose problems in fluids problems where the application of boundary conditions at such "corner" elements can overconstrain the solution. More...
 
 TetgenMesh (tetgenio &tetgen_data, const bool &split_corner_elements, TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper, const bool &use_attributes=false)
 Constructor with tetgen data structure Setting the boolean flag to true splits "corner" elements, i.e. elements that that have at least three faces on a domain boundary. The relevant elements are split without introducing hanging nodes so the sons have a "worse" shape than their fathers. However, this step avoids otherwise-hard-to-diagnose problems in fluids problems where the application of boundary conditions at such "corner" elements can overconstrain the solution. More...
 
 TetgenMesh (TetMeshFacetedClosedSurface *const &outer_boundary_pt, Vector< TetMeshFacetedSurface * > &internal_surface_pt, const double &element_volume, TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper, const bool &use_attributes=false, const bool &split_corner_elements=false, Vector< double > *const &target_element_volume_in_region_pt=nullptr)
 Build mesh, based on a TetgenMeshFactedClosedSurface that specifies the outer boundary of the domain and any number of internal boundaries, specified by TetMeshFacetedSurfaces. Also specify target size for uniform element size. Optionally specify the target element volume in each region. More...
 
void build_tetgenio (TetMeshFacetedSurface *const &outer_boundary_pt, Vector< TetMeshFacetedSurface * > &internal_surface_pt, Vector< double > *const &target_element_volume_in_region_pt, tetgenio &tetgen_io)
 Build tetgenio object from the TetMeshFacetedSurfaces. More...
 
 ~TetgenMesh ()
 Empty destructor. More...
 
void set_mesh_level_time_stepper (TimeStepper *const &time_stepper_pt, const bool &preserve_existing_data)
 Overload set_mesh_level_time_stepper so that the stored time stepper now corresponds to the new timestepper. More...
 
bool tetgenio_exists () const
 Boolen defining whether tetgenio object has been built or not. More...
 
tetgenio *& tetgenio_pt ()
 Access to the triangulateio representation of the mesh. More...
 
void set_deep_copy_tetgenio_pt (tetgenio *const &tetgenio_pt)
 Set the tetgen pointer by a deep copy. More...
 
void deep_copy_of_tetgenio (tetgenio *const &input_pt, tetgenio *&output_pt)
 Transfer tetgenio data from the input to the output The output is assumed to have been constructed and "empty". More...
 
- Public Member Functions inherited from oomph::TetMeshBase
 TetMeshBase ()
 Constructor. More...
 
 TetMeshBase (const TetMeshBase &node)=delete
 Broken copy constructor. More...
 
void operator= (const TetMeshBase &)=delete
 Broken assignment operator. More...
 
virtual ~TetMeshBase ()
 Destructor (empty) More...
 
void assess_mesh_quality (std::ofstream &some_file)
 Assess mesh quality: Ratio of max. edge length to min. height, so if it's very large it's BAAAAAD. More...
 
template<class ELEMENT >
void setup_boundary_coordinates (const unsigned &b)
 Setup boundary coordinate on boundary b which is assumed to be planar. Boundary coordinates are the x-y coordinates in the plane of that boundary, with the x-axis along the line from the (lexicographically) "lower left" to the "upper right" node. The y axis is obtained by taking the cross-product of the positive x direction with the outer unit normal computed by the face elements (or its negative if switch_normal is set to true). Doc faces in output file (if it's open). More...
 
template<class ELEMENT >
void setup_boundary_coordinates (const unsigned &b, const bool &switch_normal)
 Setup boundary coordinate on boundary b which is assumed to be planar. Boundary coordinates are the x-y coordinates in the plane of that boundary, with the x-axis along the line from the (lexicographically) "lower left" to the "upper right" node. The y axis is obtained by taking the cross-product of the positive x direction with the outer unit normal computed by the face elements (or its negative if switch_normal is set to true). Doc faces in output file (if it's open). More...
 
template<class ELEMENT >
void setup_boundary_coordinates (const unsigned &b, const bool &switch_normal, std::ofstream &outfile)
 Setup boundary coordinate on boundary b which is assumed to be planar. Boundary coordinates are the x-y coordinates in the plane of that boundary, with the x-axis along the line from the (lexicographically) "lower left" to the "upper right" node. The y axis is obtained by taking the cross-product of the positive x direction with the outer unit normal computed by the face elements (or its negative if switch_normal is set to true). Doc faces in output file (if it's open). More...
 
template<class ELEMENT >
void setup_boundary_coordinates (const unsigned &b, std::ofstream &outfile)
 Setup boundary coordinate on boundary b which is assumed to be planar. Boundary coordinates are the x-y coordinates in the plane of that boundary, with the x-axis along the line from the (lexicographically) "lower left" to the "upper right" node. The y axis is obtained by taking the cross-product of the positive x direction with the outer unit normal computed by the face elements (or its negative if switch_normal is set to true). Doc faces in output file (if it's open). More...
 
unsigned nboundary_element_in_region (const unsigned &b, const unsigned &r) const
 Return the number of elements adjacent to boundary b in region r. More...
 
FiniteElementboundary_element_in_region_pt (const unsigned &b, const unsigned &r, const unsigned &e) const
 Return pointer to the e-th element adjacent to boundary b in region r. More...
 
int face_index_at_boundary_in_region (const unsigned &b, const unsigned &r, const unsigned &e) const
 Return face index of the e-th element adjacent to boundary b in region r. More...
 
unsigned nregion ()
 Return the number of regions specified by attributes. More...
 
unsigned nregion_element (const unsigned &r)
 Return the number of elements in region r. More...
 
double region_attribute (const unsigned &i)
 Return the i-th region attribute (here only used as the (assumed to be unsigned) region id. More...
 
FiniteElementregion_element_pt (const unsigned &r, const unsigned &e)
 Return the e-th element in the r-th region. More...
 
template<class ELEMENT >
void snap_to_quadratic_surface (const Vector< unsigned > &boundary_id, const std::string &quadratic_surface_file_name, const bool &switch_normal, DocInfo &doc_info)
 Snap boundaries specified by the IDs listed in boundary_id to a quadratric surface, specified in the file quadratic_surface_file_name. This is usually used with vmtk-based meshes for which oomph-lib's xda to poly conversion code produces the files "quadratic_fsi_boundary.dat" and "quadratic_outer_solid_boundary.dat" which specify the quadratic FSI boundary (for the fluid and the solid) and the quadratic representation of the outer boundary of the solid. When used with these files, the flag switch_normal should be set to true when calling the function for the outer boundary of the solid. The DocInfo object can be used to label optional output files. (Uses directory and label). More...
 
template<class ELEMENT >
void snap_to_quadratic_surface (const Vector< unsigned > &boundary_id, const std::string &quadratic_surface_file_name, const bool &switch_normal)
 Snap boundaries specified by the IDs listed in boundary_id to a quadratric surface, specified in the file quadratic_surface_file_name. This is usually used with vmtk-based meshes for which oomph-lib's xda to poly conversion code produces the files "quadratic_fsi_boundary.dat" and "quadratic_outer_solid_boundary.dat" which specify the quadratic FSI boundary (for the fluid and the solid) and the quadratic representation of the outer boundary of the solid. When used with these files, the flag switch_normal should be set to true when calling the function for the outer boundary of the solid. More...
 
void snap_nodes_onto_geometric_objects ()
 Move the nodes on boundaries with associated GeomObjects so that they exactly coincide with the geometric object. This requires that the boundary coordinates are set up consistently. More...
 
template<class ELEMENT >
void split_elements_in_corners (TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper)
 Non-Delaunay split elements that have three faces on a boundary into sons. Timestepper species timestepper for new nodes; defaults to to steady timestepper. More...
 
void setup_boundary_element_info ()
 Setup lookup schemes which establish which elements are located next to mesh's boundaries (wrapper to suppress doc). More...
 
void setup_boundary_element_info (std::ostream &outfile)
 Setup lookup schemes which establish which elements are located next to mesh's boundaries. Doc in outfile (if it's open). More...
 
- Public Member Functions inherited from oomph::Mesh
void resize_halo_nodes ()
 Helper function that resizes halo nodes to the same size as their non-halo counterparts if required. (A discrepancy can arise if a FaceElement that introduces additional unknowns are attached to a bulk element that shares a node with a haloed element. In that case the joint node between haloed and non-haloed element is resized on that processor but not on the one that holds the halo counterpart (because no FaceElement is attached to the halo element) More...
 
 Mesh ()
 Default constructor. More...
 
 Mesh (const Vector< Mesh * > &sub_mesh_pt)
 Constructor builds combined mesh from the meshes specified. Note: This simply merges the meshes' elements and nodes (ignoring duplicates; no boundary information etc. is created). More...
 
void merge_meshes (const Vector< Mesh * > &sub_mesh_pt)
 Merge meshes. Note: This simply merges the meshes' elements and nodes (ignoring duplicates; no boundary information etc. is created). More...
 
virtual void reset_boundary_element_info (Vector< unsigned > &ntmp_boundary_elements, Vector< Vector< unsigned >> &ntmp_boundary_elements_in_region, Vector< FiniteElement * > &deleted_elements)
 Virtual function to perform the reset boundary elements info rutines. More...
 
template<class BULK_ELEMENT >
void doc_boundary_coordinates (const unsigned &b, std::ofstream &the_file)
 Output boundary coordinates on boundary b – template argument specifies the bulk element type (needed to create FaceElement of appropriate type on mesh boundary). More...
 
virtual void scale_mesh (const double &factor)
 Scale all nodal coordinates by given factor. Virtual so it can be overloaded in SolidMesh class where it also re-assigns the Lagrangian coordinates. More...
 
 Mesh (const Mesh &dummy)=delete
 Broken copy constructor. More...
 
void operator= (const Mesh &)=delete
 Broken assignment operator. More...
 
virtual ~Mesh ()
 Virtual Destructor to clean up all memory. More...
 
void flush_element_and_node_storage ()
 Flush storage for elements and nodes by emptying the vectors that store the pointers to them. This is useful if a particular mesh is only built to generate a small part of a bigger mesh. Once the elements and nodes have been created, they are typically copied into the new mesh and the auxiliary mesh can be deleted. However, if we simply call the destructor of the auxiliary mesh, it will also wipe out the nodes and elements, because it still "thinks" it's in charge of these... More...
 
void flush_element_storage ()
 Flush storage for elements (only) by emptying the vectors that store the pointers to them. This is useful if a particular mesh is only built to generate a small part of a bigger mesh. Once the elements and nodes have been created, they are typically copied into the new mesh and the auxiliary mesh can be deleted. However, if we simply call the destructor of the auxiliary mesh, it will also wipe out the nodes and elements, because it still "thinks" it's in charge of these... More...
 
void flush_node_storage ()
 Flush storage for nodes (only) by emptying the vectors that store the pointers to them. More...
 
Node *& node_pt (const unsigned long &n)
 Return pointer to global node n. More...
 
Nodenode_pt (const unsigned long &n) const
 Return pointer to global node n (const version) More...
 
GeneralisedElement *& element_pt (const unsigned long &e)
 Return pointer to element e. More...
 
GeneralisedElementelement_pt (const unsigned long &e) const
 Return pointer to element e (const version) More...
 
const Vector< GeneralisedElement * > & element_pt () const
 Return reference to the Vector of elements. More...
 
Vector< GeneralisedElement * > & element_pt ()
 Return reference to the Vector of elements. More...
 
FiniteElementfinite_element_pt (const unsigned &e) const
 Upcast (downcast?) to FiniteElement (needed to access FiniteElement member functions). More...
 
Node *& boundary_node_pt (const unsigned &b, const unsigned &n)
 Return pointer to node n on boundary b. More...
 
Nodeboundary_node_pt (const unsigned &b, const unsigned &n) const
 Return pointer to node n on boundary b. More...
 
void set_nboundary (const unsigned &nbound)
 Set the number of boundaries in the mesh. More...
 
void remove_boundary_nodes ()
 Clear all pointers to boundary nodes. More...
 
void remove_boundary_nodes (const unsigned &b)
 Remove all information about nodes stored on the b-th boundary of the mesh. More...
 
void remove_boundary_node (const unsigned &b, Node *const &node_pt)
 Remove a node from the boundary b. More...
 
void add_boundary_node (const unsigned &b, Node *const &node_pt)
 Add a (pointer to) a node to the b-th boundary. More...
 
void copy_boundary_node_data_from_nodes ()
 Replace existing boundary node lookup schemes with new schemes created using the boundary data stored in the nodes. More...
 
bool boundary_coordinate_exists (const unsigned &i) const
 Indicate whether the i-th boundary has an intrinsic coordinate. More...
 
unsigned long nelement () const
 Return number of elements in the mesh. More...
 
unsigned long nnode () const
 Return number of nodes in the mesh. More...
 
unsigned ndof_types () const
 Return number of dof types in mesh. More...
 
unsigned elemental_dimension () const
 Return number of elemental dimension in mesh. More...
 
unsigned nodal_dimension () const
 Return number of nodal dimension in mesh. More...
 
void add_node_pt (Node *const &node_pt)
 Add a (pointer to a) node to the mesh. More...
 
void add_element_pt (GeneralisedElement *const &element_pt)
 Add a (pointer to) an element to the mesh. More...
 
virtual void node_update (const bool &update_all_solid_nodes=false)
 Update nodal positions in response to changes in the domain shape. Uses the FiniteElement::get_x(...) function for FiniteElements and doesn't do anything for other element types. If a MacroElement pointer has been set for a FiniteElement, the MacroElement representation is used to update the nodal positions; if not get_x(...) uses the FE interpolation and thus leaves the nodal positions unchanged. Virtual, so it can be overloaded by specific meshes, such as AlgebraicMeshes or SpineMeshes. Generally, this function updates the position of all nodes in response to changes in the boundary position. However, we ignore all SolidNodes since their position is computed as part of the solution – unless the bool flag is set to true. Such calls are typically made when the initial mesh is created and/or after a mesh has been refined repeatedly before the start of the computation. More...
 
virtual void reorder_nodes (const bool &use_old_ordering=true)
 Re-order nodes in the order in which they appear in elements – can be overloaded for more efficient re-ordering. More...
 
virtual void get_node_reordering (Vector< Node * > &reordering, const bool &use_old_ordering=true) const
 Get a reordering of the nodes in the order in which they appear in elements – can be overloaded for more efficient re-ordering. More...
 
template<class BULK_ELEMENT , template< class > class FACE_ELEMENT>
void build_face_mesh (const unsigned &b, Mesh *const &face_mesh_pt)
 Constuct a Mesh of FACE_ELEMENTs along the b-th boundary of the mesh (which contains elements of type BULK_ELEMENT) More...
 
unsigned self_test ()
 Self-test: Check elements and nodes. Return 0 for OK. More...
 
void max_and_min_element_size (double &max_size, double &min_size)
 Determine max and min area for all FiniteElements in the mesh (non-FiniteElements are ignored) More...
 
double total_size ()
 Determine the sum of all "sizes" of the FiniteElements in the mesh (non-FiniteElements are ignored). This gives the length/area/volume occupied by the mesh. More...
 
void check_inverted_elements (bool &mesh_has_inverted_elements, std::ofstream &inverted_element_file)
 Check for inverted elements and report outcome in boolean variable. This visits all elements at their integration points and checks if the Jacobian of the mapping between local and global coordinates is positive – using the same test that would be carried out (but only in PARANOID mode) during the assembly of the elements' Jacobian matrices. Inverted elements are output in inverted_element_file (if the stream is open). More...
 
void check_inverted_elements (bool &mesh_has_inverted_elements)
 Check for inverted elements and report outcome in boolean variable. This visits all elements at their integration points and checks if the Jacobian of the mapping between local and global coordinates is positive – using the same test that would be carried out (but only in PARANOID mode) during the assembly of the elements' Jacobian matrices. More...
 
unsigned check_for_repeated_nodes (const double &epsilon=1.0e-12)
 Check for repeated nodes within a given spatial tolerance. Return (0/1) for (pass/fail). More...
 
Vector< Node * > prune_dead_nodes ()
 Prune nodes. Nodes that have been marked as obsolete are removed from the mesh (and its boundary-node scheme). Returns vector of pointers to deleted nodes. More...
 
unsigned nboundary () const
 Return number of boundaries. More...
 
unsigned long nboundary_node (const unsigned &ibound) const
 Return number of nodes on a particular boundary. More...
 
FiniteElementboundary_element_pt (const unsigned &b, const unsigned &e) const
 Return pointer to e-th finite element on boundary b. More...
 
Nodeget_some_non_boundary_node () const
 Find a node not on any boundary in mesh_pt (useful for pinning a single node in a purely Neumann problem so that it is fully determined). More...
 
unsigned nboundary_element (const unsigned &b) const
 Return number of finite elements that are adjacent to boundary b. More...
 
int face_index_at_boundary (const unsigned &b, const unsigned &e) const
 For the e-th finite element on boundary b, return int to indicate the face_index of the face adjacent to the boundary. This is consistent with input required during the generation of FaceElements. More...
 
virtual void dump (std::ofstream &dump_file, const bool &use_old_ordering=true) const
 Dump the data in the mesh into a file for restart. More...
 
void dump (const std::string &dump_file_name, const bool &use_old_ordering=true) const
 Dump the data in the mesh into a file for restart. More...
 
virtual void read (std::ifstream &restart_file)
 Read solution from restart file. More...
 
void output_paraview (std::ofstream &file_out, const unsigned &nplot) const
 Output in paraview format into specified file. Breaks up each element into sub-elements for plotting purposes. We assume that all elements are of the same type (fct will break break (in paranoid mode) if paraview output fcts of the elements are inconsistent). More...
 
void output_fct_paraview (std::ofstream &file_out, const unsigned &nplot, FiniteElement::SteadyExactSolutionFctPt exact_soln_pt) const
 Output in paraview format into specified file. Breaks up each element into sub-elements for plotting purposes. We assume that all elements are of the same type (fct will break break (in paranoid mode) if paraview output fcts of the elements are inconsistent). More...
 
void output_fct_paraview (std::ofstream &file_out, const unsigned &nplot, const double &time, FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt) const
 Output in paraview format into specified file. Breaks up each element into sub-elements for plotting purposes. We assume that all elements are of the same type (fct will break break (in paranoid mode) if paraview output fcts of the elements are inconsistent). More...
 
void output (std::ostream &outfile)
 Output for all elements. More...
 
void output (std::ostream &outfile, const unsigned &n_plot)
 Output at f(n_plot) points in each element. More...
 
void output (FILE *file_pt)
 Output for all elements (C-style output) More...
 
void output (FILE *file_pt, const unsigned &nplot)
 Output at f(n_plot) points in each element (C-style output) More...
 
void output (const std::string &output_filename)
 Output for all elements. More...
 
void output (const std::string &output_filename, const unsigned &n_plot)
 Output at f(n_plot) points in each element. More...
 
void output_fct (std::ostream &outfile, const unsigned &n_plot, FiniteElement::SteadyExactSolutionFctPt)
 Output a given Vector function at f(n_plot) points in each element. More...
 
void output_fct (std::ostream &outfile, const unsigned &n_plot, const double &time, FiniteElement::UnsteadyExactSolutionFctPt)
 Output a given time-dep. Vector function at f(n_plot) points in each element. More...
 
void output_boundaries (std::ostream &outfile)
 Output the nodes on the boundaries (into separate tecplot zones) More...
 
void output_boundaries (const std::string &output_filename)
 Output the nodes on the boundaries (into separate tecplot zones). Specify filename. More...
 
void assign_initial_values_impulsive ()
 Assign initial values for an impulsive start. More...
 
void shift_time_values ()
 Shift time-dependent data along for next timestep: Deal with nodal Data/positions and the element's internal Data. More...
 
void calculate_predictions ()
 Calculate predictions for all Data and positions associated with the mesh, usually used in adaptive time-stepping. More...
 
void set_nodal_and_elemental_time_stepper (TimeStepper *const &time_stepper_pt, const bool &preserve_existing_data)
 Set the timestepper associated with all nodal and elemental data stored in the mesh. More...
 
void set_consistent_pinned_values_for_continuation (ContinuationStorageScheme *const &continuation_stepper_pt)
 Set consistent values for pinned data in continuation. More...
 
bool does_pointer_correspond_to_mesh_data (double *const &parameter_pt)
 Does the double pointer correspond to any mesh data. More...
 
void set_nodal_time_stepper (TimeStepper *const &time_stepper_pt, const bool &preserve_existing_data)
 Set the timestepper associated with the nodal data in the mesh. More...
 
void set_elemental_internal_time_stepper (TimeStepper *const &time_stepper_pt, const bool &preserve_existing_data)
 Set the timestepper associated with the internal data stored within elements in the meah. More...
 
virtual void compute_norm (double &norm)
 Compute norm of solution by summing contributions of compute_norm(...) for all constituent elements in the mesh. What that norm means depends on what's defined in the element's function; may need to take the square root afterwards if the elements compute the square of the L2 norm, say. More...
 
virtual void compute_norm (Vector< double > &norm)
 Compute norm of solution by summing contributions of compute_norm(...) for all constituent elements in the mesh. What that norm means depends on what's defined in the element's function; may need to take the square root afterwards if the elements compute the square of the L2 norm, say. More...
 
virtual void compute_error (std::ostream &outfile, FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt, const double &time, double &error, double &norm)
 Plot error when compared against a given exact solution. Also returns the norm of the error and that of the exact solution. More...
 
virtual void compute_error (std::ostream &outfile, FiniteElement::SteadyExactSolutionFctPt exact_soln_pt, double &error, double &norm)
 Plot error when compared against a given time-depdendent exact solution. Also returns the norm of the error and that of the exact solution. More...
 
virtual void compute_error (FiniteElement::SteadyExactSolutionFctPt exact_soln_pt, double &error, double &norm)
 Plot error when compared against a given time-dependent exact solution. Also returns the norm of the error and that of the exact solution. More...
 
virtual void compute_error (FiniteElement::SteadyExactSolutionFctPt exact_soln_pt, Vector< double > &error, Vector< double > &norm)
 Plot error when compared against a given time-dependent exact solution. Also returns the norm of the error and that of the exact solution. More...
 
virtual void compute_error (std::ostream &outfile, FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt, const double &time, Vector< double > &error, Vector< double > &norm)
 Plot error when compared against a given time-depdendent exact solution. Also returns the norm of the error and that of the exact solution. Version with vectors of norms and errors so that different variables' norms and errors can be returned individually. More...
 
virtual void compute_error (std::ostream &outfile, FiniteElement::SteadyExactSolutionFctPt exact_soln_pt, Vector< double > &error, Vector< double > &norm)
 Plot error when compared against a given time-depdendent exact solution. Also returns the norm of the error and that of the exact solution. Version with vectors of norms and errors so that different variables' norms and errors can be returned individually. More...
 
virtual void compute_error (FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt, const double &time, double &error, double &norm)
 Returns the norm of the error and that of the exact solution. More...
 
virtual void compute_error (FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt, const double &time, Vector< double > &error, Vector< double > &norm)
 Returns the norm of the error and that of the exact solution. Version with vectors of norms and errors so that different variables' norms and errors can be returned individually. More...
 
bool is_mesh_distributed () const
 Boolean to indicate if Mesh has been distributed. More...
 
OomphCommunicatorcommunicator_pt () const
 Read-only access fct to communicator (Null if mesh is not distributed, i.e. if we don't have mpi). More...
 
void set_communicator_pt (OomphCommunicator *comm_pt)
 Function to set communicator (mesh is assumed to be distributed if the communicator pointer is non-null). Only defined if mpi is enabled becaus Comm_pt itself is only defined when mpi is enabled. More...
 
void set_keep_all_elements_as_halos ()
 Call this function to keep all the elements as halo elements. More...
 
void unset_keep_all_elements_as_halos ()
 Calll this function to unset the flag that keeps all elements in the mesh as halo elements. More...
 
virtual void distribute (OomphCommunicator *comm_pt, const Vector< unsigned > &element_domain, Vector< GeneralisedElement * > &deleted_element_pt, DocInfo &doc_info, const bool &report_stats, const bool &overrule_keep_as_halo_element_status)
 Distribute the problem and doc; make this virtual to allow overloading for particular meshes where further work is required. Add to vector of pointers to deleted elements. More...
 
void distribute (OomphCommunicator *comm_pt, const Vector< unsigned > &element_domain, Vector< GeneralisedElement * > &deleted_element_pt, const bool &report_stats=false)
 Distribute the problem Add to vector of pointers to deleted elements. More...
 
void prune_halo_elements_and_nodes (Vector< GeneralisedElement * > &deleted_element_pt, const bool &report_stats=false)
 (Irreversibly) prune halo(ed) elements and nodes, usually after another round of refinement, to get rid of excessively wide halo layers. Note that the current mesh will be now regarded as the base mesh and no unrefinement relative to it will be possible once this function has been called. More...
 
void prune_halo_elements_and_nodes (Vector< GeneralisedElement * > &deleted_element_pt, DocInfo &doc_info, const bool &report_stats)
 (Irreversibly) prune halo(ed) elements and nodes, usually after another round of refinement, to get rid of excessively wide halo layers. Note that the current mesh will be now regarded as the base mesh and no unrefinement relative to it will be possible once this function has been called. More...
 
void get_efficiency_of_mesh_distribution (double &av_efficiency, double &max_efficiency, double &min_efficiency)
 Get efficiency of mesh distribution: In an ideal distribution without halo overhead, each processor would only hold its own elements. Efficieny per processor = (number of non-halo elements)/ (total number of elements). More...
 
void doc_mesh_distribution (DocInfo &doc_info)
 Doc the mesh distribution, to be processed with tecplot macros. More...
 
void check_halo_schemes (DocInfo &doc_info, double &max_permitted_error_for_halo_check)
 Check halo and shared schemes on the mesh. More...
 
virtual void classify_halo_and_haloed_nodes (DocInfo &doc_info, const bool &report_stats)
 Classify the halo and haloed nodes in the mesh. Virtual so it can be overloaded to perform additional functionality (such as synchronising hanging nodes) in refineable meshes, say. More...
 
virtual void classify_halo_and_haloed_nodes (const bool &report_stats=false)
 Classify the halo and haloed nodes in the mesh. Virtual so it can be overloaded to perform additional functionality (such as synchronising hanging nodes) in refineable meshes, say. More...
 
void synchronise_shared_nodes (const bool &report_stats)
 Synchronise shared node lookup schemes to cater for the the case where: (1) a certain node on the current processor is halo with proc p (i.e. its non-halo counterpart lives on processor p) (2) that node is also exists (also as a halo) on another processor (q, say) where its non-halo counter part is also known to be on processor p. However, without calling this function the current processor does not necessarily know that it shares a node with processor q. This information can be required, e.g. when synchronising hanging node schemes over all processors. More...
 
void get_all_halo_data (std::map< unsigned, double * > &map_of_halo_data)
 Get all the halo data stored in the mesh and add pointers to the data to the map, indexed by global equation number. More...
 
Vector< GeneralisedElement * > halo_element_pt (const unsigned &p)
 Return vector of halo elements in this Mesh whose non-halo counterpart is held on processor p. More...
 
Vector< GeneralisedElement * > haloed_element_pt (const unsigned &p)
 Return vector of haloed elements in this Mesh whose haloing counterpart is held on processor p. More...
 
unsigned nnon_halo_element ()
 Total number of non-halo elements in this mesh (Costly call computes result on the fly) More...
 
unsigned nroot_halo_element ()
 Total number of root halo elements in this Mesh. More...
 
unsigned nroot_halo_element (const unsigned &p)
 Number of root halo elements in this Mesh whose non-halo counterpart is held on processor p. More...
 
Vector< GeneralisedElement * > root_halo_element_pt (const unsigned &p)
 Vector of pointers to root halo elements in this Mesh whose non-halo counterpart is held on processor p. More...
 
GeneralisedElement *& root_halo_element_pt (const unsigned &p, const unsigned &e)
 Access fct to the e-th root halo element in this Mesh whose non-halo counterpart is held on processor p. More...
 
void add_root_halo_element_pt (const unsigned &p, GeneralisedElement *&el_pt)
 Add root halo element whose non-halo counterpart is held on processor p to this Mesh. More...
 
unsigned nhalo_node ()
 Total number of halo nodes in this Mesh. More...
 
unsigned nhalo_node (const unsigned &p)
 Number of halo nodes in this Mesh whose non-halo counterpart is held on processor p. More...
 
void add_halo_node_pt (const unsigned &p, Node *&nod_pt)
 Add halo node whose non-halo counterpart is held on processor p to the storage scheme for halo nodes. More...
 
Nodehalo_node_pt (const unsigned &p, const unsigned &j)
 Access fct to the j-th halo node in this Mesh whose non-halo counterpart is held on processor p. More...
 
unsigned nroot_haloed_element ()
 Total number of root haloed elements in this Mesh. More...
 
unsigned nroot_haloed_element (const unsigned &p)
 Number of root haloed elements in this Mesh whose non-halo counterpart is held on processor p. More...
 
Vector< GeneralisedElement * > root_haloed_element_pt (const unsigned &p)
 Vector of pointers to root haloed elements in this Mesh whose non-halo counterpart is held on processor p. More...
 
GeneralisedElement *& root_haloed_element_pt (const unsigned &p, const unsigned &e)
 Access fct to the e-th root haloed element in this Mesh whose non-halo counterpart is held on processor p. More...
 
void add_root_haloed_element_pt (const unsigned &p, GeneralisedElement *&el_pt)
 Add root haloed element whose non-halo counterpart is held on processor p to the storage scheme for haloed elements. Note: This does not add the element to the storage scheme for elements as it's understood to naturally live on this processor anyway! More...
 
unsigned nhaloed_node ()
 Total number of haloed nodes in this Mesh. More...
 
unsigned nhaloed_node (const unsigned &p)
 Number of haloed nodes in this Mesh whose haloed counterpart is held on processor p. More...
 
Nodehaloed_node_pt (const unsigned &p, const unsigned &j)
 Access fct to the j-th haloed node in this Mesh whose halo counterpart is held on processor p. More...
 
void add_haloed_node_pt (const unsigned &p, Node *&nod_pt)
 Add haloed node whose halo counterpart is held on processor p to the storage scheme for haloed nodes. More...
 
void disable_resizing_of_halo_nodes ()
 Function to suppress resizing of halo nodes – optmisation but call it at your own risk! More...
 
void enable_resizing_of_halo_nodes ()
 Function to (re-)enable resizing of halo nodes – this returns things to the default behaviour. More...
 
void disable_output_of_halo_elements ()
 Function to disable halo element output. More...
 
void enable_output_of_halo_elements ()
 Function to enable halo element output. More...
 
unsigned nshared_node ()
 Total number of shared nodes in this Mesh. More...
 
void doc_shared_nodes ()
 Doc shared nodes. More...
 
unsigned nshared_node (const unsigned &p)
 Number of shared nodes in this Mesh who have a counterpart on processor p. More...
 
Nodeshared_node_pt (const unsigned &p, const unsigned &j)
 Access fct to the j-th shared node in this Mesh who has a counterpart on processor p. More...
 
void get_shared_node_pt (const unsigned &p, Vector< Node * > &shared_node_pt)
 Get vector of pointers to shared nodes with processor p. Required for faster search in Missing_masters_functions::add_external_haloed_node_helper() and Missing_masters_functions::add_external_haloed_master_node_helper() More...
 
void add_shared_node_pt (const unsigned &p, Node *&nod_pt)
 Add shared node whose counterpart is held on processor p to the storage scheme for shared nodes. (NB: ensure that this routine is called twice, once for each process) More...
 
void get_halo_node_stats (double &av_number, unsigned &max_number, unsigned &min_number)
 Get halo node stats for this distributed mesh: Average/max/min number of halo nodes over all processors. Careful: Involves MPI Broadcasts and must therefore be called on all processors! More...
 
void get_haloed_node_stats (double &av_number, unsigned &max_number, unsigned &min_number)
 Get haloed node stats for this distributed mesh: Average/max/min number of haloed nodes over all processors. Careful: Involves MPI Broadcasts and must therefore be called on all processors! More...
 
void output_external_halo_elements (std::ostream &outfile, const unsigned &n_plot=5)
 Output all external halo elements. More...
 
void output_external_halo_elements (const unsigned &p, std::ostream &outfile, const unsigned &n_plot=5)
 Output all external halo elements with processor p. More...
 
void output_external_haloed_elements (std::ostream &outfile, const unsigned &n_plot=5)
 Output all external haloed elements. More...
 
void output_external_haloed_elements (const unsigned &p, std::ostream &outfile, const unsigned &n_plot=5)
 Output all external haloed elements with processor p. More...
 
unsigned nexternal_halo_element ()
 Total number of external halo elements in this Mesh. More...
 
unsigned nexternal_halo_element (const unsigned &p)
 Number of external halo elements in this Mesh whose non-halo counterpart is held on processor p. More...
 
GeneralisedElement *& external_halo_element_pt (const unsigned &p, const unsigned &e)
 Access fct to the e-th external halo element in this Mesh whose non-halo counterpart is held on processor p. More...
 
void add_external_halo_element_pt (const unsigned &p, GeneralisedElement *&el_pt)
 Add external halo element whose non-halo counterpart is held on processor p to this Mesh. More...
 
unsigned nexternal_haloed_element ()
 Total number of external haloed elements in this Mesh. More...
 
unsigned nexternal_haloed_element (const unsigned &p)
 Number of external haloed elements in this Mesh whose non-halo counterpart is held on processor p. More...
 
GeneralisedElement *& external_haloed_element_pt (const unsigned &p, const unsigned &e)
 Access fct to the e-th external haloed element in this Mesh whose non-halo counterpart is held on processor p. More...
 
unsigned add_external_haloed_element_pt (const unsigned &p, GeneralisedElement *&el_pt)
 Add external haloed element whose non-halo counterpart is held on processor p to the storage scheme for haloed elements. More...
 
unsigned nexternal_halo_node ()
 Total number of external halo nodes in this Mesh. More...
 
void get_external_halo_node_pt (Vector< Node * > &external_halo_node_pt)
 Get vector of pointers to all external halo nodes. More...
 
unsigned nexternal_halo_node (const unsigned &p)
 Number of external halo nodes in this Mesh whose non-halo (external) counterpart is held on processor p. More...
 
void add_external_halo_node_pt (const unsigned &p, Node *&nod_pt)
 Add external halo node whose non-halo (external) counterpart is held on processor p to the storage scheme for halo nodes. More...
 
Node *& external_halo_node_pt (const unsigned &p, const unsigned &j)
 Access fct to the j-th external halo node in this Mesh whose non-halo external counterpart is held on processor p. More...
 
Vector< Node * > external_halo_node_pt (const unsigned &p)
 Access fct to vector of external halo node in this Mesh whose non-halo external counterpart is held on processor p. (read only) More...
 
void set_external_halo_node_pt (const unsigned &p, const Vector< Node * > &external_halo_node_pt)
 Set vector of external halo node in this Mesh whose non-halo external counterpart is held on processor p. More...
 
void null_external_halo_node (const unsigned &p, Node *nod_pt)
 Null out specified external halo node (used when deleting duplicates) More...
 
void remove_null_pointers_from_external_halo_node_storage ()
 Consolidate external halo node storage by removing nulled out pointes in external halo and haloed schemes. More...
 
unsigned nexternal_haloed_node ()
 Total number of external haloed nodes in this Mesh. More...
 
unsigned nexternal_haloed_node (const unsigned &p)
 Number of external haloed nodes in this Mesh whose halo (external) counterpart is held on processor p. More...
 
Node *& external_haloed_node_pt (const unsigned &p, const unsigned &j)
 Access fct to the j-th external haloed node in this Mesh whose halo external counterpart is held on processor p. More...
 
unsigned add_external_haloed_node_pt (const unsigned &p, Node *&nod_pt)
 Add external haloed node whose halo (external) counterpart is held on processor p to the storage scheme for haloed nodes. More...
 
Vector< Node * > external_haloed_node_pt (const unsigned &p)
 Access fct to vector of external haloed node in this Mesh whose halo external counterpart is held on processor p. (read only) More...
 
void set_external_haloed_node_pt (const unsigned &p, const Vector< Node * > &external_haloed_node_pt)
 Set vector of external haloed node in this Mesh whose halo external counterpart is held on processor p. More...
 
std::set< int > external_halo_proc ()
 Return the set of processors that hold external halo nodes. This is required to avoid having to pass a communicator into the node_update functions for Algebraic-based and MacroElement-based Meshes. More...
 
virtual void create_shared_boundaries (OomphCommunicator *comm_pt, const Vector< unsigned > &element_domain, const Vector< GeneralisedElement * > &backed_up_el_pt, const Vector< FiniteElement * > &backed_up_f_el_pt, std::map< Data *, std::set< unsigned >> &processors_associated_with_data, const bool &overrule_keep_as_halo_element_status)
 Creates the shared boundaries, only used in unstructured meshes In this case with the "TriangleMesh" class. More...
 
virtual unsigned try_to_add_root_haloed_element_pt (const unsigned &p, GeneralisedElement *&el_pt)
 
virtual unsigned try_to_add_haloed_node_pt (const unsigned &p, Node *&nod_pt)
 
void delete_all_external_storage ()
 Wipe the storage for all externally-based elements. More...
 
- Public Member Functions inherited from oomph::RefineableTetMeshBase
double & max_element_size ()
 Max element size allowed during adaptation. More...
 
double & min_element_size ()
 Min element size allowed during adaptation. More...
 
double & max_permitted_edge_ratio ()
 Min edge ratio before remesh gets triggered. More...
 
void doc_adaptivity_targets (std::ostream &outfile)
 Doc the targets for mesh adaptation. More...
 
double compute_volume_target (const Vector< double > &elem_error, Vector< double > &target_volume)
 Compute target volume based on the elements' error and the error target; return max edge ratio. More...
 
- Public Member Functions inherited from oomph::RefineableMeshBase
bool adapt_flag ()
 
 RefineableMeshBase ()
 Constructor sets default values for refinement targets etc. and initialises pointer to spatial error estimator to NULL. More...
 
 RefineableMeshBase (const RefineableMeshBase &dummy)=delete
 Broken copy constructor. More...
 
void operator= (const RefineableMeshBase &)=delete
 Broken assignment operator. More...
 
virtual ~RefineableMeshBase ()
 Empty Destructor: More...
 
unsigned nrefined ()
 Access fct for number of elements that were refined. More...
 
unsigned nunrefined ()
 Access fct for number of elements that were unrefined. More...
 
unsigned & nrefinement_overruled ()
 Number of elements that would have liked to be refined further but can't because they've reached the max. refinement level. More...
 
unsigned & max_keep_unrefined ()
 Max. number of elements that we allow to remain unrefined if no other mesh adaptation is required (to avoid mesh-adaptations that would only unrefine a few elements and then force a new solve – this can't be worth our while!) More...
 
ErrorEstimator *& spatial_error_estimator_pt ()
 Access to spatial error estimator. More...
 
ErrorEstimatorspatial_error_estimator_pt () const
 Access to spatial error estimator (const version. More...
 
double & min_permitted_error ()
 Access fct for min. error (i.e. (try to) merge elements if their error is smaller) More...
 
double & max_permitted_error ()
 Access fct for max. error (i.e. split elements if their error is larger) More...
 
double & min_error ()
 Access fct for min. actual error in present solution (i.e. before re-solve on adapted mesh) More...
 
double & max_error ()
 Access fct for max. actual error in present solution (i.e. before re-solve on adapted mesh) More...
 
DocInfo *& doc_info_pt ()
 Access fct for pointer to DocInfo. More...
 
void enable_adaptation ()
 Enable adaptation. More...
 
void disable_adaptation ()
 Disable adaptation. More...
 
void enable_p_adaptation ()
 Enable adaptation. More...
 
void disable_p_adaptation ()
 Disable adaptation. More...
 
void enable_additional_synchronisation_of_hanging_nodes ()
 Enable additional synchronisation of hanging nodes. More...
 
void disable_additional_synchronisation_of_hanging_nodes ()
 Disable additional synchronisation of hanging nodes. More...
 
bool is_adaptation_enabled () const
 Return whether the mesh is to be adapted. More...
 
bool is_p_adaptation_enabled () const
 Return whether the mesh is to be adapted. More...
 
bool is_additional_synchronisation_of_hanging_nodes_disabled () const
 Return whether additional synchronisation is enabled. More...
 
DocInfo doc_info ()
 Access fct for DocInfo. More...
 
virtual void p_adapt (const Vector< double > &elemental_error)
 p-adapt mesh: Refine elements whose error is lager than err_max and (try to) unrefine those whose error is smaller than err_min More...
 
virtual void refine_uniformly ()
 Refine mesh uniformly. More...
 
virtual void p_refine_uniformly (DocInfo &doc_info)
 p-refine mesh uniformly and doc process More...
 
virtual void p_refine_uniformly ()
 p-refine mesh uniformly More...
 
void p_unrefine_uniformly (DocInfo &doc_info)
 p-unrefine mesh uniformly More...
 

Protected Member Functions

 RefineableTetgenMesh (const Vector< double > &target_volume, tetgenio *const &tetgen_io_pt, TetMeshFacetedClosedSurface *const &outer_boundary_pt, Vector< TetMeshFacetedSurface * > &internal_surface_pt, TimeStepper *time_stepper_pt=&Mesh::Default_TimeStepper, const bool &use_attributes=false)
 Specialised constructor used during adaptation only. Element sizes are specified by vector tetgen_io is passed in from previous mesh (is then modified to build new mesh) Ditto with use_attributes, which comes from the previous mesh. More...
 
void initialise_adaptation_data ()
 Helper function to initialise data associated with adaptation. More...
 
void update_faceted_surface_using_face_mesh (TetMeshFacetedSurface *&faceted_surface_pt)
 Helper function that updates the input faceted surface by using the flattened elements from FaceMesh(es) that are constructed for the boundaries associated with the segments of the polygon. More...
 
void surface_remesh_for_inner_hole_boundaries ()
 Generate a new faceted representation of the inner hole boundaries. More...
 
void snap_nodes_onto_boundary (RefineableTetgenMesh< ELEMENT > *&new_mesh_pt, const unsigned &b)
 Snap the boundary nodes onto any curvilinear boundaries. More...
 
- Protected Member Functions inherited from oomph::TetgenMesh< ELEMENT >
void build_from_scaffold (TimeStepper *time_stepper_pt, const bool &use_attributes)
 Build mesh from scaffold. More...
 
void setup_reverse_lookup_schemes_for_faceted_surface (TetMeshFacetedSurface *const &faceted_surface_pt)
 Function to setup the reverse look-up schemes. More...
 
- Protected Member Functions inherited from oomph::Mesh
void setup_shared_node_scheme ()
 Setup shared node scheme. More...
 
unsigned long assign_global_eqn_numbers (Vector< double * > &Dof_pt)
 Assign the global equation numbers in the Data stored at the nodes and also internal element Data. Also, build (via push_back) the Vector of pointers to the dofs (variables). More...
 
void describe_dofs (std::ostream &out, const std::string &current_string) const
 Function to describe the dofs of the Mesh. The ostream specifies the output stream to which the description is written; the string stores the currently assembled output that is ultimately written to the output stream by Data::describe_dofs(...); it is typically built up incrementally as we descend through the call hierarchy of this function when called from Problem::describe_dofs(...) More...
 
void describe_local_dofs (std::ostream &out, const std::string &current_string) const
 Function to describe the local dofs of the elements. The ostream specifies the output stream to which the description is written; the string stores the currently assembled output that is ultimately written to the output stream by Data::describe_dofs(...); it is typically built up incrementally as we descend through the call hierarchy of this function when called from Problem::describe_dofs(...) More...
 
void assign_local_eqn_numbers (const bool &store_local_dof_pt)
 Assign the local equation numbers in all elements If the boolean argument is true then also store pointers to dofs. More...
 
void convert_to_boundary_node (Node *&node_pt, const Vector< FiniteElement * > &finite_element_pt)
 A function that upgrades an ordinary node to a boundary node We shouldn't ever really use this, but it does make life that bit easier for the lazy mesh writer. The pointer to the node is replaced by a pointer to the new boundary node in all element look-up schemes and in the mesh's Node_pt vector. The new node is also addressed by node_pt on return from the function. More...
 
void convert_to_boundary_node (Node *&node_pt)
 A function that upgrades an ordinary node to a boundary node. All pointers to the node from the mesh's elements are found. and replaced by pointers to the new boundary node. If the node is present in the mesh's list of nodes, that pointer is also replaced. Finally, the pointer argument node_pt addresses the new node on return from the function. We shouldn't ever really use this, but it does make life that bit easier for the lazy mesh writer. More...
 

Protected Attributes

bool Projection_is_disabled
 Disable projection of solution onto new mesh during adaptation. More...
 
bool Corner_elements_must_be_split
 Corner elements which have all of their nodes on the outer boundary are to be split into elements which have some non-boundary nodes. More...
 
- Protected Attributes inherited from oomph::TetgenMesh< ELEMENT >
TetgenScaffoldMeshTmp_mesh_pt
 Temporary scaffold mesh. More...
 
bool Tetgenio_exists
 Boolean to indicate whether a tetgenio representation of the mesh exists. More...
 
tetgenio * Tetgenio_pt
 Tetgen representation of mesh. More...
 
bool Use_attributes
 Boolean flag to indicate whether to use attributes or not (required for multidomain meshes) More...
 
- Protected Attributes inherited from oomph::TetMeshBase
Vector< Vector< FiniteElement * > > Region_element_pt
 Vectors of vectors of elements in each region (note: this just stores them; the region IDs are contained in Region_attribute!) More...
 
Vector< double > Region_attribute
 Vector of attributes associated with the elements in each region NOTE: double is enforced on us by tetgen. We use it as an unsigned to indicate the actual (zero-based) region ID. More...
 
Vector< std::map< unsigned, Vector< FiniteElement * > > > Boundary_region_element_pt
 Storage for elements adjacent to a boundary in a particular region. More...
 
Vector< std::map< unsigned, Vector< int > > > Face_index_region_at_boundary
 Storage for the face index adjacent to a boundary in a particular region. More...
 
TetMeshFacetedClosedSurfaceOuter_boundary_pt
 Faceted surface that defines outer boundaries. More...
 
Vector< TetMeshFacetedSurface * > Internal_surface_pt
 Vector to faceted surfaces that define internal boundaries. More...
 
std::map< unsigned, TetMeshFacetedSurface * > Tet_mesh_faceted_surface_pt
 Reverse lookup scheme: Pointer to faceted surface (if any!) associated with boundary b. More...
 
std::map< unsigned, TetMeshFacet * > Tet_mesh_facet_pt
 Reverse lookup scheme: Pointer to facet (if any!) associated with boundary b. More...
 
std::map< unsigned, Vector< Vector< double > > > Triangular_facet_vertex_boundary_coordinate
 Boundary coordinates of vertices in triangular facets associated with given boundary. Is only set up for triangular facets! More...
 
TimeStepperTime_stepper_pt
 Timestepper used to build nodes. More...
 
- Protected Attributes inherited from oomph::Mesh
Vector< Vector< Node * > > Boundary_node_pt
 Vector of Vector of pointers to nodes on the boundaries: Boundary_node_pt(b,n). Note that this is private to force the use of the add_boundary_node() function, which ensures that the reverse look-up schemes for the nodes are set up. More...
 
bool Lookup_for_elements_next_boundary_is_setup
 Flag to indicate that the lookup schemes for elements that are adjacent to the boundaries has been set up. More...
 
Vector< Vector< FiniteElement * > > Boundary_element_pt
 Vector of Vector of pointers to elements on the boundaries: Boundary_element_pt(b,e) More...
 
Vector< Vector< int > > Face_index_at_boundary
 For the e-th finite element on boundary b, this is the index of the face that lies along that boundary. More...
 
std::map< unsigned, Vector< GeneralisedElement * > > Root_halo_element_pt
 Map of vectors holding the pointers to the root halo elements. More...
 
std::map< unsigned, Vector< GeneralisedElement * > > Root_haloed_element_pt
 Map of vectors holding the pointers to the root haloed elements. More...
 
std::map< unsigned, Vector< Node * > > Halo_node_pt
 Map of vectors holding the pointers to the halo nodes. More...
 
std::map< unsigned, Vector< Node * > > Haloed_node_pt
 Map of vectors holding the pointers to the haloed nodes. More...
 
std::map< unsigned, Vector< Node * > > Shared_node_pt
 Map of vectors holding the pointers to the shared nodes. These are all the nodes that are on two "neighbouring" processes (the halo(ed) lookup scheme depends upon which processor is in charge. More...
 
OomphCommunicatorComm_pt
 Pointer to communicator – set to NULL if mesh is not distributed. More...
 
std::map< unsigned, Vector< GeneralisedElement * > > External_halo_element_pt
 External halo(ed) elements are created as and when they are needed to act as source elements for the particular process's mesh. The storage is wiped and rebuilt every time the mesh is refined. More...
 
std::map< unsigned, Vector< GeneralisedElement * > > External_haloed_element_pt
 Map of vectors holding the pointers to the external haloed elements. More...
 
std::map< unsigned, Vector< Node * > > External_halo_node_pt
 Map of vectors holding the pointers to the external halo nodes. More...
 
std::map< unsigned, Vector< Node * > > External_haloed_node_pt
 Map of vectors holding the pointers to the external haloed nodes. More...
 
bool Keep_all_elements_as_halos
 bool to indicate whether to keep all elements in a mesh as halos or not More...
 
bool Resize_halo_nodes_not_required
 Set this to true to suppress resizing of halo nodes (at your own risk!) More...
 
Vector< Node * > Node_pt
 Vector of pointers to nodes. More...
 
Vector< GeneralisedElement * > Element_pt
 Vector of pointers to generalised elements. More...
 
std::vector< bool > Boundary_coordinate_exists
 Vector of boolean data that indicates whether the boundary coordinates have been set for the boundary. More...
 
- Protected Attributes inherited from oomph::RefineableMeshBase
ErrorEstimatorSpatial_error_estimator_pt
 Pointer to spatial error estimator. More...
 
double Max_permitted_error
 Max. error (i.e. split elements if their error is larger) More...
 
double Min_permitted_error
 Min. error (i.e. (try to) merge elements if their error is smaller) More...
 
double Min_error
 Min.actual error. More...
 
double Max_error
 Max. actual error. More...
 
unsigned Nrefined
 Stats: Number of elements that were refined. More...
 
unsigned Nunrefined
 Stats: Number of elements that were unrefined. More...
 
bool Adapt_flag
 Flag that requests adaptation. More...
 
bool P_adapt_flag
 Flag that requests p-adaptation. More...
 
bool Additional_synchronisation_of_hanging_nodes_not_required
 Flag that disables additional synchronisation of hanging nodes. More...
 
DocInfoDoc_info_pt
 Pointer to DocInfo. More...
 
unsigned Max_keep_unrefined
 Max. number of elements that can remain unrefined if no other mesh adaptation is required (to avoid mesh-adaptations that would only unrefine a few elements and then force a new solve – this can't be worth our while!) More...
 
unsigned Nrefinement_overruled
 Number of elements that would like to be refined further but can't because they've reached the max. refinement level. More...
 

Additional Inherited Members

- Public Types inherited from oomph::Mesh
typedef void(FiniteElement::* SteadyExactSolutionFctPt) (const Vector< double > &x, Vector< double > &soln)
 Typedef for function pointer to function that computes steady exact solution. More...
 
typedef void(FiniteElement::* UnsteadyExactSolutionFctPt) (const double &time, const Vector< double > &x, Vector< double > &soln)
 Typedef for function pointer to function that computes unsteady exact solution. More...
 
- Public Attributes inherited from oomph::Mesh
bool Output_halo_elements
 Bool for output of halo elements. More...
 
- Public Attributes inherited from oomph::RefineableTetMeshBase
double Max_element_size
 Max permitted element size. More...
 
double Min_element_size
 Min permitted element size. More...
 
double Max_permitted_edge_ratio
 Max edge ratio before remesh gets triggered. More...
 
- Static Public Attributes inherited from oomph::TetMeshBase
static double Tolerance_for_boundary_finding = 1.0e-5
 Global static data that specifies the permitted error in the setup of the boundary coordinates. More...
 
- Static Public Attributes inherited from oomph::Mesh
static Steady< 0 > Default_TimeStepper
 Default Steady Timestepper, to be used in default arguments to Mesh constructors. More...
 
static bool Suppress_warning_about_empty_mesh_level_time_stepper_function
 Boolean used to control warning about empty mesh level timestepper function. More...
 

Detailed Description

template<class ELEMENT>
class oomph::RefineableTetgenMesh< ELEMENT >

Definition at line 47 of file refineable_tetgen_mesh.template.h.

Constructor & Destructor Documentation

◆ RefineableTetgenMesh() [1/2]

template<class ELEMENT >
oomph::RefineableTetgenMesh< ELEMENT >::RefineableTetgenMesh ( TetMeshFacetedClosedSurface *const &  outer_boundary_pt,
Vector< TetMeshFacetedSurface * > &  internal_closed_surface_pt,
const double &  element_volume,
TimeStepper time_stepper_pt = &Mesh::Default_TimeStepper,
const bool &  use_attributes = false,
const bool &  split_corner_elements = false,
Vector< double > *const &  target_element_volume_in_region_pt = nullptr 
)
inline

Build mesh, based on a TetMeshFacetedClosedSurface that specifies the outer boundary of the domain and any number of internal closed curves, specified by TetMeshFacetedSurfaces. Also specify target volume for uniform element size.

Definition at line 55 of file refineable_tetgen_mesh.template.h.

References oomph::RefineableTetgenMesh< ELEMENT >::initialise_adaptation_data().

◆ RefineableTetgenMesh() [2/2]

template<class ELEMENT >
oomph::RefineableTetgenMesh< ELEMENT >::RefineableTetgenMesh ( const Vector< double > &  target_volume,
tetgenio *const &  tetgen_io_pt,
TetMeshFacetedClosedSurface *const &  outer_boundary_pt,
Vector< TetMeshFacetedSurface * > &  internal_surface_pt,
TimeStepper time_stepper_pt = &Mesh::Default_TimeStepper,
const bool &  use_attributes = false 
)
inlineprotected

◆ ~RefineableTetgenMesh()

template<class ELEMENT >
virtual oomph::RefineableTetgenMesh< ELEMENT >::~RefineableTetgenMesh ( )
inlinevirtual

Empty Destructor.

Definition at line 183 of file refineable_tetgen_mesh.template.h.

Member Function Documentation

◆ adapt()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::adapt ( const Vector< double > &  elem_error)
virtual

◆ disable_projection()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::disable_projection ( )
inline

Disable projection of old solution onto new mesh.

Definition at line 218 of file refineable_tetgen_mesh.template.h.

References oomph::RefineableTetgenMesh< ELEMENT >::Projection_is_disabled.

◆ enable_projection()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::enable_projection ( )
inline

Disable projection of old solution onto new mesh.

Definition at line 224 of file refineable_tetgen_mesh.template.h.

References oomph::RefineableTetgenMesh< ELEMENT >::Projection_is_disabled.

◆ initialise_adaptation_data()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::initialise_adaptation_data ( )
inlineprotected

◆ projection_is_disabled()

template<class ELEMENT >
bool oomph::RefineableTetgenMesh< ELEMENT >::projection_is_disabled ( )
inline

Is projection of old solution onto new mesh disabled?

Definition at line 212 of file refineable_tetgen_mesh.template.h.

References oomph::RefineableTetgenMesh< ELEMENT >::Projection_is_disabled.

◆ refine_uniformly()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::refine_uniformly ( DocInfo doc_info)
inlinevirtual

Refine mesh uniformly and doc process.

Implements oomph::RefineableMeshBase.

Definition at line 186 of file refineable_tetgen_mesh.template.h.

◆ snap_nodes_onto_boundary()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::snap_nodes_onto_boundary ( RefineableTetgenMesh< ELEMENT > *&  new_mesh_pt,
const unsigned &  b 
)
protected

◆ surface_remesh_for_inner_hole_boundaries()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::surface_remesh_for_inner_hole_boundaries
protected

Generate a new faceted representation of the inner hole boundaries.

Definition at line 227 of file refineable_tetgen_mesh.template.cc.

◆ unrefine_uniformly()

template<class ELEMENT >
unsigned oomph::RefineableTetgenMesh< ELEMENT >::unrefine_uniformly ( )
inlinevirtual

Unrefine mesh uniformly: Return 0 for success, 1 for failure (if unrefinement has reached the coarsest permitted level)

Implements oomph::RefineableMeshBase.

Definition at line 197 of file refineable_tetgen_mesh.template.h.

◆ update_faceted_surface_using_face_mesh()

template<class ELEMENT >
void oomph::RefineableTetgenMesh< ELEMENT >::update_faceted_surface_using_face_mesh ( TetMeshFacetedSurface *&  faceted_surface_pt)
protected

Helper function that updates the input faceted surface by using the flattened elements from FaceMesh(es) that are constructed for the boundaries associated with the segments of the polygon.

List of vertex nodes

Definition at line 49 of file refineable_tetgen_mesh.template.cc.

References e, oomph::Mesh::element_pt(), oomph::Mesh::finite_element_pt(), i, oomph::Mesh::nelement(), oomph::TetMeshFacetedSurface::nfacet(), oomph::FiniteElement::node_pt(), oomph::TetMeshFacetedSurface::one_based_facet_boundary_id(), oomph::oomph_info, and oomph::FaceElement::set_boundary_number_in_bulk_mesh().

Member Data Documentation

◆ Corner_elements_must_be_split

template<class ELEMENT >
bool oomph::RefineableTetgenMesh< ELEMENT >::Corner_elements_must_be_split
protected

Corner elements which have all of their nodes on the outer boundary are to be split into elements which have some non-boundary nodes.

Definition at line 260 of file refineable_tetgen_mesh.template.h.

◆ Projection_is_disabled

template<class ELEMENT >
bool oomph::RefineableTetgenMesh< ELEMENT >::Projection_is_disabled
protected

The documentation for this class was generated from the following files: