Classes | |
class | DataSetDescriptor |
Public Types | |
typedef Quadrature< dim-1 > | SubQuadrature |
Static Public Member Functions | |
static void | project_to_face (const SubQuadrature &quadrature, const unsigned int face_no, std::vector< Point< dim > > &q_points) |
static Quadrature< dim > | project_to_face (const SubQuadrature &quadrature, const unsigned int face_no) |
static void | project_to_subface (const SubQuadrature &quadrature, const unsigned int face_no, const unsigned int subface_no, std::vector< Point< dim > > &q_points, const RefinementCase< dim-1 > &ref_case=RefinementCase< dim-1 >::isotropic_refinement) |
static Quadrature< dim > | project_to_subface (const SubQuadrature &quadrature, const unsigned int face_no, const unsigned int subface_no, const RefinementCase< dim-1 > &ref_case=RefinementCase< dim-1 >::isotropic_refinement) |
static Quadrature< dim > | project_to_all_faces (const SubQuadrature &quadrature) |
static Quadrature< dim > | project_to_all_subfaces (const SubQuadrature &quadrature) |
static Quadrature< dim > | project_to_child (const Quadrature< dim > &quadrature, const unsigned int child_no) |
static Quadrature< dim > | project_to_line (const Quadrature< 1 > &quadrature, const Point< dim > &p1, const Point< dim > &p2) |
Static Private Member Functions | |
static Quadrature< 2 > | reflect (const Quadrature< 2 > &q) |
static Quadrature< 2 > | rotate (const Quadrature< 2 > &q, const unsigned int n_times) |
This class is a helper class to facilitate the usage of quadrature formulae on faces or subfaces of cells. It computes the locations of quadrature points on the unit cell from a quadrature object for a manifold of one dimension less than that of the cell and the number of the face. For example, giving the Simpson rule in one dimension and using the project_to_face() function with face number 1, the returned points will be (1,0), (1,0.5) and (1,1). Note that faces have an orientation, so when projecting to face 3, you will get (0,0), (0,0.5) and (0,1), which is in clockwise sense, while for face 1 the points were in counterclockwise sense.
For the projection to subfaces (i.e. to the children of a face of the unit cell), the same applies as above. Note the order in which the children of a face are numbered, which in two dimensions coincides with the orientation of the face.
The second set of functions generates a quadrature formula by projecting a given quadrature rule on all faces and subfaces. This is used in the FEFaceValues and FESubfaceValues classes. Since we now have the quadrature points of all faces and subfaces in one array, we need to have a way to find the starting index of the points and weights corresponding to one face or subface within this array. This is done through the DataSetDescriptor member class.
The different functions are grouped into a common class to avoid putting them into global namespace. However, since they have no local data, all functions are declared static
and can be called without creating an object of this class.
For the 3d case, you should note that the orientation of faces is even more intricate than for two dimensions. Quadrature formulae are projected upon the faces in their standard orientation, not to the inside or outside of the hexahedron. To make things more complicated, in 3d we allow faces in two orientations (which can be identified using cell->face_orientation(face)
), so we have to project quadrature formula onto faces and subfaces in two orientations. (Refer to the documentation of the Triangulation class for a description of the orientation of the different faces, as well as to the glossary entry on face orientation for more information on this.) The DataSetDescriptor member class is used to identify where each dataset starts.
typedef Quadrature<dim-1> QProjector< dim >::SubQuadrature |
Define a typedef for a quadrature that acts on an object of one dimension less. For cells, this would then be a face quadrature.
static void QProjector< dim >::project_to_face | ( | const SubQuadrature & | quadrature, | |
const unsigned int | face_no, | |||
std::vector< Point< dim > > & | q_points | |||
) | [static] |
Compute the quadrature points on the cell if the given quadrature formula is used on face face_no
. For further details, see the general doc for this class.
static Quadrature<dim> QProjector< dim >::project_to_face | ( | const SubQuadrature & | quadrature, | |
const unsigned int | face_no | |||
) | [static] |
Compute the cell quadrature formula corresponding to using quadrature
on face face_no
. For further details, see the general doc for this class.
static void QProjector< dim >::project_to_subface | ( | const SubQuadrature & | quadrature, | |
const unsigned int | face_no, | |||
const unsigned int | subface_no, | |||
std::vector< Point< dim > > & | q_points, | |||
const RefinementCase< dim-1 > & | ref_case = RefinementCase< dim-1 >::isotropic_refinement | |||
) | [static] |
Compute the quadrature points on the cell if the given quadrature formula is used on face face_no
, subface number subface_no
corresponding to RefineCase::Type ref_case
. The last argument is only used in 3D.
static Quadrature<dim> QProjector< dim >::project_to_subface | ( | const SubQuadrature & | quadrature, | |
const unsigned int | face_no, | |||
const unsigned int | subface_no, | |||
const RefinementCase< dim-1 > & | ref_case = RefinementCase< dim-1 >::isotropic_refinement | |||
) | [static] |
Compute the cell quadrature formula corresponding to using quadrature
on subface subface_no
of face face_no
with RefinementCase<dim-1> ref_case
. The last argument is only used in 3D.
static Quadrature<dim> QProjector< dim >::project_to_all_faces | ( | const SubQuadrature & | quadrature | ) | [static] |
Take a face quadrature formula and generate a cell quadrature formula from it where the quadrature points of the given argument are projected on all faces.
The weights of the new rule are replications of the original weights. This is not a proper handling, in that the sum of weights does not equal one, but it is consistent with the use of this function, namely to generate sets of face quadrature points on a cell, one set of which will then be selected at each time. This is used in the FEFaceValues class, where we initialize the values, derivatives, etc on all faces at once, while selecting the data of one particular face only happens later.
static Quadrature<dim> QProjector< dim >::project_to_all_subfaces | ( | const SubQuadrature & | quadrature | ) | [static] |
This function is alike the previous one, but projects the given face quadrature formula to the subfaces of a cell, i.e. to the children of the faces of the unit cell.
static Quadrature<dim> QProjector< dim >::project_to_child | ( | const Quadrature< dim > & | quadrature, | |
const unsigned int | child_no | |||
) | [static] |
Project a give quadrature formula to a child of a cell. You may want to use this function in case you want to extend an integral only over the area which a potential child would occupy. The child numbering is the same as the children would be numbered upon refinement of the cell.
As integration using this quadrature formula now only extends over a fraction of the cell, the weights of the resulting object are divided by GeometryInfo<dim>::children_per_cell.
static Quadrature<dim> QProjector< dim >::project_to_line | ( | const Quadrature< 1 > & | quadrature, | |
const Point< dim > & | p1, | |||
const Point< dim > & | p2 | |||
) | [static] |
Project the onedimensional rule quadrature
to the straight line connecting the points p1
and p2
.
static Quadrature<2> QProjector< dim >::reflect | ( | const Quadrature< 2 > & | q | ) | [static, private] |
Given a quadrature object in 2d, reflect all quadrature points at the main diagonal and return them with their original weights.
This function is necessary for projecting a 2d quadrature rule onto the faces of a 3d cube, since there we need both orientations.
static Quadrature<2> QProjector< dim >::rotate | ( | const Quadrature< 2 > & | q, | |
const unsigned int | n_times | |||
) | [static, private] |
Given a quadrature object in 2d, rotate all quadrature points by n_times
* 90 degrees counterclockwise and return them with their original weights.
This function is necessary for projecting a 2d quadrature rule onto the faces of a 3d cube, since there we need all rotations to account for face_flip and face_rotation of non-standard faces.