Reference documentation for deal.II version 9.5.0
|
Typedefs | |
using | subdomain_id = unsigned int |
using | global_vertex_index = std::uint64_t |
using | fe_index = unsigned short int |
using | global_dof_index = unsigned int |
using | signed_global_dof_index = std::make_signed_t< global_dof_index > |
using | global_cell_index = unsigned int |
using | coarse_cell_id = global_cell_index |
using | boundary_id = unsigned int |
using | manifold_id = unsigned int |
using | material_id = unsigned int |
using | geometric_entity_type = std::uint8_t |
using | blas_int = int |
using | suitesparse_index = long int |
using | particle_index = unsigned int |
A namespace in which we declare alias for types used in deal.II, as well as special values for these types.
using types::subdomain_id = typedef unsigned int |
The type used to denote subdomain_ids of cells.
See the glossary for more information.
There is a special value, numbers::invalid_subdomain_id that is used to indicate an invalid value of this type.
using types::global_vertex_index = typedef std::uint64_t |
using types::fe_index = typedef unsigned short int |
using types::global_dof_index = typedef unsigned int |
The type used to denote the global index of degrees of freedom. This type is then also used for querying the global number of degrees of freedom, since the number is simply the largest index plus one.
While in sequential computations the 4 billion indices of 32-bit unsigned integers is plenty, parallel computations using (for example) the parallel::distributed::Triangulation class can overflow this number and consequently, deal.II chooses a larger integer type when configured to use 64-bit indices.
The data type always corresponds to an unsigned integer type.
See the When to use types::global_dof_index instead of unsigned int page for guidance on when this type should or should not be used.
using types::signed_global_dof_index = typedef std::make_signed_t<global_dof_index> |
using types::global_cell_index = typedef unsigned int |
The type used to denote the global index of a cell. This type is then also used for querying the global number of cells in a triangulation since the number is simply the largest index plus one.
While in sequential computations the 4 billion indices of 32-bit unsigned integers is plenty, parallel computations using (for example) the parallel::distributed::Triangulation class can overflow this number and consequently, deal.II chooses a larger integer type when configured to use 64-bit indices.
The data type always corresponds to an unsigned integer type.
using types::coarse_cell_id = typedef global_cell_index |
The type used for coarse-cell ids. See the glossary entry on coarse cell IDs for more information.
using types::boundary_id = typedef unsigned int |
The type used to denote boundary indicators associated with every piece of the boundary of the domain.
There is a special value, numbers::internal_face_boundary_id that is used to indicate an invalid value of this type and that is used as the boundary indicator for faces that are in the interior of the domain and therefore not part of any addressable boundary component.
using types::manifold_id = typedef unsigned int |
The type used to denote manifold indicators associated with every object of the mesh.
There is a special value, numbers::flat_manifold_id that is used to indicate the standard cartesian manifold.
using types::material_id = typedef unsigned int |
The type used to denote material indicators associated with every cell.
There is a special value, numbers::invalid_material_id that is used to indicate an invalid value of this type.
using types::geometric_entity_type = typedef std::uint8_t |
using types::blas_int = typedef int |
Integer type in BLAS.
Definition at line 37 of file lapack_support.h.
using types::suitesparse_index = typedef long int |
Index type for UMFPACK. SuiteSparse_long has to be used here for the Windows 64 build.
Definition at line 45 of file sparse_direct.h.
using types::particle_index = typedef unsigned int |
The type used for indices of particles. While in sequential computations the 4 billion indices of 32-bit unsigned integers is plenty, parallel computations using hundreds of processes can overflow this number and we need a bigger index space. We here utilize the same build variable that controls the dof indices because the number of degrees of freedom and the number of particles are typically on the same order of magnitude.
The data type always indicates an unsigned integer type.
Definition at line 65 of file property_pool.h.