Discrete linear operator stored as a H-matrix. More...

#include </home/wojtek/Projects/BEM/bempp-sven/bempp/lib/assembly/discrete_aca_boundary_operator.hpp>

Inheritance diagram for Bempp::DiscreteAcaBoundaryOperator< ValueType >:

Public Types
typedef Fiber::ScalarTraits < ValueType >::RealType	CoordinateType

typedef AhmedDofWrapper < CoordinateType >	AhmedDofType

typedef bbxbemblcluster < AhmedDofType, AhmedDofType >	AhmedBemBlcluster

typedef mblock< typename AhmedTypeTraits< ValueType > ::Type >	AhmedMblock

typedef boost::shared_array < AhmedMblock * >	AhmedMblockArray

typedef boost::shared_array < AhmedMblock * >	AhmedConstMblockArray

Public Member Functions
	DiscreteAcaBoundaryOperator (unsigned int rowCount, unsigned int columnCount, double epsUsedInAssembly, int maximumRankUsedInAssembly, int symmetry, const shared_ptr< const AhmedBemBlcluster > &blockCluster_, const AhmedMblockArray &blocks_, const IndexPermutation &domainPermutation_, const IndexPermutation &rangePermutation_, const ParallelizationOptions &parallelizationOptions_, const std::vector< AhmedConstMblockArray > &sharedBlocks_=std::vector< AhmedConstMblockArray >())
	Constructor. More...

	DiscreteAcaBoundaryOperator (unsigned int rowCount, unsigned int columnCount, int maximumRankUsedInAssembly, int symmetry, std::auto_ptr< const AhmedBemBlcluster > blockCluster_, AhmedMblockArray blocks_, const IndexPermutation &domainPermutation_, const IndexPermutation &rangePermutation_, const ParallelizationOptions &parallelizationOptions_, const std::vector< AhmedConstMblockArray > &sharedBlocks_=std::vector< AhmedConstMblockArray >()) BEMPP_DEPRECATED
	Constructor. More...

BEMPP_DEPRECATED	DiscreteAcaBoundaryOperator (unsigned int rowCount, unsigned int columnCount, int maximumRankUsedInAssembly, int symmetry, const shared_ptr< const AhmedBemBlcluster > &blockCluster_, const AhmedMblockArray &blocks_, const IndexPermutation &domainPermutation_, const IndexPermutation &rangePermutation_, const ParallelizationOptions &parallelizationOptions_, const std::vector< AhmedConstMblockArray > &sharedBlocks_=std::vector< AhmedConstMblockArray >())
	Constructor. More...

virtual arma::Mat< ValueType >	asMatrix () const
	Matrix representation of the operator. More...

virtual unsigned int	rowCount () const
	Number of rows of the operator.

virtual unsigned int	columnCount () const
	Number of columns of the operator.

virtual void	addBlock (const std::vector< int > &rows, const std::vector< int > &cols, const ValueType alpha, arma::Mat< ValueType > &block) const
	Add a subblock of this operator to a matrix. More...

virtual shared_ptr< const DiscreteBoundaryOperator < ValueType > >	asDiscreteAcaBoundaryOperator (double eps=-1, int maximumRank=-1, bool interleave=false) const
	Return a representation that can be cast to a DiscreteAcaBoundaryOperator. More...

void	makeAllMblocksDense ()
	Uncompress all blocks of the H-matrix and store them as dense matrices. More...

int	maximumRank () const
	Return the upper bound for the rank of low-rank mblocks specified during H-matrix construction.

double	eps () const
	Return the value of the epsilon parameter specified during H-matrix construction.

int	actualMaximumRank () const
	Return the actual maximum rank of low-rank mblocks.

int	symmetry () const
	Return a flag describing the symmetry of this operator.

shared_ptr< const AhmedBemBlcluster >	blockCluster () const
	Return the block cluster.

AhmedMblockArray	blocks () const
	Return the mblocks making up the H-matrix represented by this operator. More...

size_t	blockCount () const
	Return the number of mblocks making up this operator.

const IndexPermutation &	domainPermutation () const
	Return the domain index permutation.

const IndexPermutation &	rangePermutation () const
	Return the range index permutation.

const ParallelizationOptions &	parallelizationOptions () const
	Return the parallelization options used in the matrix-vector multiply.

std::vector < AhmedConstMblockArray >	sharedBlocks () const
	Return the vector of mblock arrays that this operator implicitly depends on.

virtual Teuchos::RCP< const Thyra::VectorSpaceBase < ValueType > >	domain () const

virtual Teuchos::RCP< const Thyra::VectorSpaceBase < ValueType > >	range () const

Public Member Functions inherited from Bempp::DiscreteBoundaryOperator< ValueType >
virtual	~DiscreteBoundaryOperator ()
	Destructor.

void	apply (const Thyra::EOpTransp M_trans, const Thyra::MultiVectorBase< ValueType > &x_in, const Thyra::Ptr< Thyra::MultiVectorBase< ValueType > > &y_inout, const ValueType alpha, const ValueType beta) const
	Apply the linear operator to a multivector. More...

void	apply (const TranspositionMode trans, const arma::Mat< ValueType > &x_in, arma::Mat< ValueType > &y_inout, const ValueType alpha, const ValueType beta) const
	Apply the operator to a vector. More...

virtual void	dump () const
	Write a textual representation of the operator to standard output. More...

Static Public Member Functions
static const DiscreteAcaBoundaryOperator < ValueType > &	castToAca (const DiscreteBoundaryOperator< ValueType > &discreteOperator)
	Downcast a reference to a DiscreteBoundaryOperator object to DiscreteAcaBoundaryOperator. More...

static shared_ptr< const DiscreteAcaBoundaryOperator < ValueType > >	castToAca (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &discreteOperator)
	Downcast a shared pointer to a DiscreteBoundaryOperator object to a shared pointer to a DiscreteAcaBoundaryOperator. More...

Protected Member Functions
virtual bool	opSupportedImpl (Thyra::EOpTransp M_trans) const

Protected Member Functions inherited from Bempp::DiscreteBoundaryOperator< ValueType >
virtual void	applyImpl (const Thyra::EOpTransp M_trans, const Thyra::MultiVectorBase< ValueType > &X_in, const Teuchos::Ptr< Thyra::MultiVectorBase< ValueType > > &Y_inout, const ValueType alpha, const ValueType beta) const

Private Member Functions
virtual void	applyBuiltInImpl (const TranspositionMode trans, const arma::Col< ValueType > &x_in, arma::Col< ValueType > &y_inout, const ValueType alpha, const ValueType beta) const

Friends
class	AcaApproximateLuInverse< ValueType >

shared_ptr< const DiscreteBoundaryOperator < ValueType > >	acaOperatorSum (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op1, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op2, double eps, int maximumRank)

shared_ptr< const DiscreteBoundaryOperator < ValueType > >	scaledAcaOperator (const ValueType &multiplier, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)

shared_ptr< const DiscreteBoundaryOperator < ValueType > >	scaledAcaOperator (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op, const ValueType &multiplier)

Related Functions
(Note that these are not member functions.)
template<typename ValueType >
shared_ptr< const DiscreteBoundaryOperator < ValueType > >	acaOperatorSum (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op1, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op2, double eps, int maximumRank)
	Add two discrete boundary operators stored as H-matrices. More...

template<typename ValueType >
shared_ptr< const DiscreteBoundaryOperator < ValueType > >	scaledAcaOperator (const ValueType &multiplier, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Multiply the H-matrix representation of a discrete boundary operator by a scalar and wrap the result in a new discrete boundary operator. More...

template<typename ValueType >
shared_ptr< const DiscreteBoundaryOperator < ValueType > >	scaledAcaOperator (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op, const ValueType &multiplier)
	Multiply the H-matrix representation of a discrete boundary operator by a scalar and wrap the result in a new discrete boundary operator. More...

template<typename ValueType >
shared_ptr< const DiscreteBoundaryOperator < ValueType > >	acaOperatorApproximateLuInverse (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op, double delta)
	LU inverse of a discrete boundary operator stored as a H-matrix. More...

Related Functions inherited from Bempp::DiscreteBoundaryOperator< ValueType >
template<typename ValueType >
shared_ptr< const DiscreteBoundaryOperator < ValueType > >	operator+ (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Unary plus: return a copy of the argument.

template<typename ValueType >
shared_ptr< const DiscreteBoundaryOperator < ValueType > >	operator- (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Return a shared pointer to a DiscreteBoundaryOperator representing the operand multiplied by -1. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	operator+ (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op1, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op2)
	Return a shared pointer to a DiscreteBoundaryOperator representing the sum of the operands. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	operator- (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op1, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op2)
	Return a shared pointer to a DiscreteBoundaryOperator representing the difference of the operands. More...

template<typename ValueType , typename ScalarType >
boost::enable_if< typename boost::mpl::has_key < boost::mpl::set< float, double, std::complex< float > , std::complex< double > >, ScalarType >, shared_ptr < DiscreteBoundaryOperator < ValueType > > >::type	operator* (ScalarType scalar, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Return a shared pointer to a DiscreteBoundaryOperator representing the operator `*op` multiplied by `scalar`. More...

template<typename ValueType , typename ScalarType >
boost::enable_if< typename boost::mpl::has_key < boost::mpl::set< float, double, std::complex< float > , std::complex< double > >, ScalarType >, shared_ptr < DiscreteBoundaryOperator < ValueType > > >::type	operator* (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op, ScalarType scalar)
	Return a shared pointer to a DiscreteBoundaryOperator representing the operator `*op` multiplied by `scalar`. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	operator* (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op1, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op2)
	Return a shared pointer to a DiscreteBoundaryOperator representing a composition of operators, `(op1) (*op2)`. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	mul (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op1, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op2)
	Return a shared pointer to a DiscreteBoundaryOperator representing a composition of operators, `(op1) (*op2)`. More...

template<typename ValueType , typename ScalarType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	operator/ (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op, ScalarType scalar)
	Return a shared pointer to a DiscreteBoundaryOperator representing the operator `*op` divided by `scalar`. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	transpose (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Return a shared pointer to a DiscreteBoundaryOperator representing the transposed operator `*op`. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	conjugate (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Return a shared pointer to a DiscreteBoundaryOperator representing the conjugated operator `*op`. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	conjugateTranspose (const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Return a shared pointer to a DiscreteBoundaryOperator representing the conjugate-transposed operator `*op`. More...

template<typename ValueType >
shared_ptr < DiscreteBoundaryOperator < ValueType > >	transpose (TranspositionMode trans, const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &op)
	Return a shared pointer to a DiscreteBoundaryOperator representing the transposed and/or conjugated (depending on the value of the argument `trans`) operator `*op`. More...

template<typename RealType >
shared_ptr < DiscreteBoundaryOperator < std::complex< RealType > > >	complexify (const shared_ptr< const DiscreteBoundaryOperator< RealType > > &op)
	Return a shared pointer to a DiscreteBoundaryOperator representing the "complexified" operator `*op`. More...

Detailed Description

template<typename ValueType>
class Bempp::DiscreteAcaBoundaryOperator< ValueType >

Discrete linear operator stored as a H-matrix.

Constructor & Destructor Documentation

template<typename ValueType >

Bempp::DiscreteAcaBoundaryOperator< ValueType >::DiscreteAcaBoundaryOperator	(	unsigned int	rowCount,
		unsigned int	columnCount,
		double	epsUsedInAssembly,
		int	maximumRankUsedInAssembly,
		int	symmetry,
		const shared_ptr< const AhmedBemBlcluster > &	blockCluster_,
		const AhmedMblockArray &	blocks_,
		const IndexPermutation &	domainPermutation_,
		const IndexPermutation &	rangePermutation_,
		const ParallelizationOptions &	parallelizationOptions_,
		const std::vector< AhmedConstMblockArray > &	sharedBlocks_ = `std::vector<AhmedConstMblockArray>()`
	)

Constructor.

Parameters

[in]	rowCount	Number of rows.
[in]	columnCount	Number of columns.
[in]	epsUsedInAssembly	The epsilon parameter used during assembly of the H-matrix.
[in]	maximumRankUsedInAssembly	The limit on block rank used during assembly of the H-matrix. This is used in particular when generating the approximate H-matrix LU.
[in]	symmetry	H-matrix symmetry. Can be any combination of the flags defined in the Symmetry enumeration type.
[in]	blockCluster_	Block cluster defining the structure of the H-matrix.
[in]	blocks_	Array containing the H-matrix blocks.
[in]	domainPermutation_	Mapping from original to permuted column indices.
[in]	rangePermutation_	Mapping from original to permuted row indices.
[in]	parallelizationOptions_	Options determining the maximum number of threads used in the apply() routine for the H-matrix-vector product.
[in]	sharedBlocks_	Vector of arrays of mblocks on which this operator implicitly depends and which therefore must stay alive for the lifetime of this operator. Useful for constructing ACA operators that combine mblocks of several other operators.

Note: Currently the apply() routine is only parallelized for non-Hermitian H-matrices.

template<typename ValueType >

Bempp::DiscreteAcaBoundaryOperator< ValueType >::DiscreteAcaBoundaryOperator	(	unsigned int	rowCount,
		unsigned int	columnCount,
		int	maximumRankUsedInAssembly,
		int	symmetry,
		std::auto_ptr< const AhmedBemBlcluster >	blockCluster_,
		AhmedMblockArray	blocks_,
		const IndexPermutation &	domainPermutation_,
		const IndexPermutation &	rangePermutation_,
		const ParallelizationOptions &	parallelizationOptions_,
		const std::vector< AhmedConstMblockArray > &	sharedBlocks_ = `std::vector<AhmedConstMblockArray>()`
	)

Constructor.

Parameters

[in]	rowCount	Number of rows.
[in]	columnCount	Number of columns.
[in]	maximumRankUsedInAssembly	The limit on block rank used during assembly of the H-matrix. This is used in particular when generating the approximate H-matrix LU.
[in]	symmetry	H-matrix symmetry. Can be any combination of the flags defined in the Symmetry enumeration type.
[in]	blockCluster_	Block cluster defining the structure of the H-matrix.
[in]	blocks_	Array containing the H-matrix blocks.
[in]	domainPermutation_	Mapping from original to permuted column indices.
[in]	rangePermutation_	Mapping from original to permuted row indices.
[in]	parallelizationOptions_	Options determining the maximum number of threads used in the apply() routine for the H-matrix-vector product.
[in]	sharedBlocks_	Vector of arrays of mblocks on which this operator implicitly depends and which therefore must stay alive for the lifetime of this operator. Useful for constructing ACA operators that combine mblocks of several other operators.

Note: Currently the apply() routine is only parallelized for non-Hermitian H-matrices.

Deprecated:: This constructor is deprecated. Use the non-deprecated constructor.

template<typename ValueType >

Bempp::DiscreteAcaBoundaryOperator< ValueType >::DiscreteAcaBoundaryOperator	(	unsigned int	rowCount,
		unsigned int	columnCount,
		int	maximumRankUsedInAssembly,
		int	symmetry,
		const shared_ptr< const AhmedBemBlcluster > &	blockCluster_,
		const AhmedMblockArray &	blocks_,
		const IndexPermutation &	domainPermutation_,
		const IndexPermutation &	rangePermutation_,
		const ParallelizationOptions &	parallelizationOptions_,
		const std::vector< AhmedConstMblockArray > &	sharedBlocks_ = `std::vector<AhmedConstMblockArray>()`
	)

Constructor.

Parameters

[in]	rowCount	Number of rows.
[in]	columnCount	Number of columns.
[in]	maximumRankUsedInAssembly	The limit on block rank used during assembly of the H-matrix. This is used in particular when generating the approximate H-matrix LU.
[in]	symmetry	H-matrix symmetry. Can be any combination of the flags defined in the Symmetry enumeration type.
[in]	blockCluster_	Block cluster defining the structure of the H-matrix.
[in]	blocks_	Array containing the H-matrix blocks.
[in]	domainPermutation_	Mapping from original to permuted column indices.
[in]	rangePermutation_	Mapping from original to permuted row indices.
[in]	parallelizationOptions_	Options determining the maximum number of threads used in the apply() routine for the H-matrix-vector product.
[in]	sharedBlocks_	Vector of arrays of mblocks on which this operator implicitly depends and which therefore must stay alive for the lifetime of this operator. Useful for constructing ACA operators that combine mblocks of several other operators.

Note: Currently the apply() routine is only parallelized for non-Hermitian H-matrices.

Deprecated:: This constructor is deprecated. Use the non-deprecated constructor.

Member Function Documentation

template<typename ValueType>

void Bempp::DiscreteAcaBoundaryOperator< ValueType >::addBlock	(	const std::vector< int > &	rows,
		const std::vector< int > &	cols,
		const ValueType	alpha,
		arma::Mat< ValueType > &	block
	)		const

virtual

Add a subblock of this operator to a matrix.

Perform the operation block += alpha * L[rows, cols], where block is a matrix, alpha a scalar and L[rows, cols] a subblock of this discrete linear operator.

Parameters

[in]	rows	Vector of row indices.
[in]	cols	Vector of column indices.
[in]	alpha	Multiplier.
[in,out]	block	On entry, matrix of size (`rows.size()`, `cols.size()`). On exit, each element (i, j) of this matrix will be augmented by the element (`rows[i]`, `cols[j]`) of this operator, multiplied by `alpha`.

Row and column indices may be unsorted.

This method need not be supported by all subclasses. It is mainly intended for internal use during weak-form assembly.

Implements Bempp::DiscreteBoundaryOperator< ValueType >.

template<typename ValueType >

shared_ptr< const DiscreteBoundaryOperator< ValueType > > Bempp::DiscreteAcaBoundaryOperator< ValueType >::asDiscreteAcaBoundaryOperator	(	double	eps = `-1`,
		int	maximumRank = `-1`,
		bool	interleave = `false`
	)		const

virtual

Return a representation that can be cast to a DiscreteAcaBoundaryOperator.

The conversion only succeeds if all members of the DiscreteBoundaryOperator itself can be cast to a DiscreteAcaBoundaryOperator or if it is a linear combination of DiscreteOperators that can be cast to type DiscreteAcaBoundaryOperator. Operator compositions are not yet supported by this function.

Parameters

[in]	eps	The $\epsilon$ parameter controlling the accuracy of H-matrix addition in case this operation is needed to construct the H-matrix representation of the operator. If `eps` is negative (default), it is set to the smaller of the values of $\epsilon$ specified during the creation of the two H-matrices to be added.
[in]	maximumRank	Maximum rank of blocks to be considered low rank when adding two H-Matrices. If `maximumRank` is negative, it is set to the larger of the maximum ranks specified during the creation of the two H-matrices to be added.

Returns: A pointer to a DiscreteBoundaryOperator object castable to DiscreteAcaBoundaryOperator.

Note: This function throws an exception if BEM++ has been compiled without AHMED.

Reimplemented from Bempp::DiscreteBoundaryOperator< ValueType >.

template<typename ValueType >

arma::Mat< ValueType > Bempp::DiscreteAcaBoundaryOperator< ValueType >::asMatrix ( ) const

virtual

Matrix representation of the operator.

The default implementation is slow and should be overridden where possible.

Reimplemented from Bempp::DiscreteBoundaryOperator< ValueType >.

template<typename ValueType >

DiscreteAcaBoundaryOperator< ValueType >::AhmedMblockArray Bempp::DiscreteAcaBoundaryOperator< ValueType >::blocks ( ) const

Return the mblocks making up the H-matrix represented by this operator.

Note: This function returns a shared array of pointers to non-constant blocks. However, you must not modify it! This is just a workaround for AHMED's lack of const-correctness.

template<typename ValueType>

const DiscreteAcaBoundaryOperator< ValueType > & Bempp::DiscreteAcaBoundaryOperator< ValueType >::castToAca ( const DiscreteBoundaryOperator< ValueType > & discreteOperator )

static

Downcast a reference to a DiscreteBoundaryOperator object to DiscreteAcaBoundaryOperator.

If the object referenced by discreteOperator is not in fact a DiscreteAcaBoundaryOperator, a std::bad_cast exception is thrown.

Referenced by Bempp::DiscreteAcaBoundaryOperator< ValueType >::acaOperatorApproximateLuInverse(), Bempp::DiscreteAcaBoundaryOperator< ValueType >::acaOperatorSum(), and Bempp::DiscreteAcaBoundaryOperator< ValueType >::scaledAcaOperator().

template<typename ValueType>

shared_ptr< const DiscreteAcaBoundaryOperator< ValueType > > Bempp::DiscreteAcaBoundaryOperator< ValueType >::castToAca ( const shared_ptr< const DiscreteBoundaryOperator< ValueType > > & discreteOperator )

static

Downcast a shared pointer to a DiscreteBoundaryOperator object to a shared pointer to a DiscreteAcaBoundaryOperator.

If the object referenced by discreteOperator is not in fact a DiscreteAcaBoundaryOperator, a std::bad_cast exception is thrown.

template<typename ValueType >

void Bempp::DiscreteAcaBoundaryOperator< ValueType >::makeAllMblocksDense ( )

Uncompress all blocks of the H-matrix and store them as dense matrices.

Sometimes useful for debugging.

Friends And Related Function Documentation

template<typename ValueType >

shared_ptr< const DiscreteBoundaryOperator< ValueType > > acaOperatorApproximateLuInverse	(	const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &	op,
		double	delta
	)

Parameters

[in]	op	Discrete boundary operator for which to compute the LU inverse.
[in]	delta	Approximation accuracy of the inverse.

Returns: A shared pointer to a newly allocated discrete boundary operator representing the (approximate) LU inverse of op and stored as an (approximate) LU decomposition of op.

References Bempp::DiscreteAcaBoundaryOperator< ValueType >::castToAca().

template<typename ValueType >

shared_ptr< const DiscreteBoundaryOperator< ValueType > > acaOperatorSum	(	const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &	op1,
		const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &	op2,
		double	eps,
		int	maximumRank
	)

A std::bad_cast exception is thrown if the input operators can not be cast to DiscreteAcaBoundaryOperator.

Parameters

[in]	op1	First operand.
[in]	op2	Second operand.
[in]	eps	???

Todo:: look into M. Bebendorf's book.

Parameters

[in] maximumRank Maximum rank of blocks that should be considered low-rank in the H-matrix to be constructed.

Returns: A shared pointer to a newly allocated discrete boundary operator representing the sum of the operands op1 and op2 stored as a single H-matrix.

References Bempp::DiscreteAcaBoundaryOperator< ValueType >::castToAca().

template<typename ValueType >

shared_ptr< const DiscreteBoundaryOperator< ValueType > > scaledAcaOperator	(	const ValueType &	multiplier,
		const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &	op
	)

A std::bad_cast exception is thrown if the input operator can not be cast to DiscreteAcaBoundaryOperator

Parameters

[in]	multiplier	Scalar multiplier.
[in]	op	Discrete boundary operator to be multiplied.

Returns: A shared pointer to a newly allocated discrete boundary operator representing the operand op multiplied by multiplier and stored as a H-matrix.

References Bempp::DiscreteAcaBoundaryOperator< ValueType >::castToAca().

template<typename ValueType >

shared_ptr< const DiscreteBoundaryOperator< ValueType > > scaledAcaOperator	(	const shared_ptr< const DiscreteBoundaryOperator< ValueType > > &	op,
		const ValueType &	multiplier
	)

A std::bad_cast exception is thrown if the input operator can not be cast to DiscreteAcaBoundaryOperator

Parameters

[in]	op	Discrete boundary operator to be multiplied.
[in]	multiplier	Scalar multiplier.

Returns: A shared pointer to a newly allocated discrete boundary operator representing the operand op multiplied by multiplier and stored as a H-matrix.

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

/home/wojtek/Projects/BEM/bempp-sven/bempp/lib/assembly/discrete_aca_boundary_operator.hpp
/home/wojtek/Projects/BEM/bempp-sven/bempp/lib/assembly/discrete_aca_boundary_operator.cpp

Public Types

Public Member Functions

Static Public Member Functions

Protected Member Functions

Private Member Functions

Friends

Related Functions

Detailed Description

template<typename ValueType> class Bempp::DiscreteAcaBoundaryOperator< ValueType >

Constructor & Destructor Documentation

Member Function Documentation

Friends And Related Function Documentation

template<typename ValueType>
class Bempp::DiscreteAcaBoundaryOperator< ValueType >