Parent class(es)
This class represents a user-defined expression. Expressions can
be used as coefficients in variational forms or interpolated
into finite element spaces.
An expression is defined by overloading the eval() method. Users
may choose to overload either a simple version of eval(), in the
case of expressions only depending on the coordinate x, or an
optional version for expressions depending on x and mesh data
like cell indices or facet normals.
The geometric dimension (the size of x) and the value rank and
dimensions of an expression must supplied as arguments to the
constructor.
-
Expression()
Create scalar expression.
-
explicit Expression(std::size_t dim)
Create vector-valued expression with given dimension.
- Arguments
- dim (std::size_t)
- Dimension of the vector-valued expression.
-
Expression(std::size_t dim0, std::size_t dim1)
Create matrix-valued expression with given dimensions.
- Arguments
- dim0 (std::size_t)
- Dimension (rows).
- dim1 (std::size_t)
- Dimension (columns).
-
explicit Expression(std::vector<std::size_t> value_shape)
Create tensor-valued expression with given shape.
- Arguments
- value_shape (std::vector<std::size_t>)
- Shape of expression.
-
Expression(const Expression& expression)
Copy constructor
- Arguments
- expression (Expression)
- Object to be copied.
-
void eval(Array<double>& values, const Array<double>& x, const ufc::cell& cell) const
Note: The reimplementation of eval is needed for the Python interface.
Evaluate at given point in given cell.
- Arguments
- values (Array <double>)
- The values at the point.
- x (Array <double>)
- The coordinates of the point.
- cell (ufc::cell)
- The cell which contains the given point.
-
void eval(Array<double>& values, const Array<double>& x) const
Evaluate at given point.
- Arguments
- values (Array <double>)
- The values at the point.
- x (Array <double>)
- The coordinates of the point.
-
std::size_t value_rank() const
Return value rank.
- Returns
- std::size_t
- The value rank.
-
std::size_t value_dimension(std::size_t i) const
Return value dimension for given axis.
- Arguments
- i (std::size_t)
- Integer denoting the axis to use.
- Returns
- std::size_t
- The value dimension (for the given axis).
-
void restrict(double* w, const FiniteElement& element, const Cell& dolfin_cell, const double* vertex_coordinates, const ufc::cell& ufc_cell) const
Restrict function to local cell (compute expansion coefficients w).
- Arguments
- w (list of doubles)
- Expansion coefficients.
- element (FiniteElement)
- The element.
- dolfin_cell (Cell)
- The cell.
- ufc_cell (ufc::cell)
- The ufc::cell.
-
void compute_vertex_values(std::vector<double>& vertex_values, const Mesh& mesh) const
Compute values at all mesh vertices.
- Arguments
- vertex_values (Array <double>)
- The values at all vertices.
- mesh (Mesh)
- The mesh.