oomph::Matrix< T, MATRIX_TYPE > Class Template Reference

#include <matrices.h>

Inheritance diagram for oomph::Matrix< T, MATRIX_TYPE >:

Public Member Functions
	Matrix ()
	(Empty) constructor More...
	Matrix (const Matrix &matrix)=delete
	Broken copy constructor. More...
void	operator= (const Matrix &)=delete
	Broken assignment operator. More...
virtual	~Matrix ()
	Virtual (empty) destructor. More...
virtual unsigned long	nrow () const =0
	Return the number of rows of the matrix. More...
virtual unsigned long	ncol () const =0
	Return the number of columns of the matrix. More...
T	operator() (const unsigned long &i, const unsigned long &j) const
T &	operator() (const unsigned long &i, const unsigned long &j)
virtual void	output (std::ostream &outfile) const
virtual void	output_bottom_right_zero_helper (std::ostream &outfile) const =0
virtual void	sparse_indexed_output_helper (std::ostream &outfile) const =0
void	sparse_indexed_output (std::ostream &outfile, const unsigned &precision=0, const bool &output_bottom_right_zero=false) const
void	sparse_indexed_output (std::string filename, const unsigned &precision=0, const bool &output_bottom_right_zero=false) const

Protected Member Functions
void	range_check (const unsigned long &i, const unsigned long &j) const

Detailed Description

template<class T, class MATRIX_TYPE>
class oomph::Matrix< T, MATRIX_TYPE >

Abstract base class for matrices, templated by the type of object that is stored in them and the type of matrix. The MATRIX_TYPE template argument is used as part of the Curiously Recurring Template Pattern, see http://en.wikipedia.org/wiki/Curiously_Recurring_Template_Pattern The pattern is used to force the inlining of the round bracket access functions by ensuring that they are NOT virtual functions.

Constructor & Destructor Documentation

Matrix() [1/2]

template<class T , class MATRIX_TYPE >

oomph::Matrix< T, MATRIX_TYPE >::Matrix ( )

inline

(Empty) constructor

{}

Matrix() [2/2]

template<class T , class MATRIX_TYPE >

oomph::Matrix< T, MATRIX_TYPE >::Matrix ( const Matrix< T, MATRIX_TYPE > &  matrix )

delete

Broken copy constructor.

~Matrix()

template<class T , class MATRIX_TYPE >

virtual oomph::Matrix< T, MATRIX_TYPE >::~Matrix ( )

inlinevirtual

Virtual (empty) destructor.

{}

Member Function Documentation

ncol()

template<class T , class MATRIX_TYPE >

virtual unsigned long oomph::Matrix< T, MATRIX_TYPE >::ncol ( ) const

pure virtual

Return the number of columns of the matrix.

Implemented in oomph::SumOfMatrices, oomph::CCDoubleMatrix, oomph::DenseDoubleMatrix, oomph::CRDoubleMatrix, oomph::SparseMatrix< T, MATRIX_TYPE >, oomph::SparseMatrix< T, CCMatrix< T > >, oomph::SparseMatrix< T, CRMatrix< T > >, oomph::DenseMatrix< T >, oomph::DenseMatrix< oomph::MatrixVectorProduct * >, oomph::DenseMatrix< double >, oomph::DenseMatrix< int >, oomph::DenseMatrix< double * >, oomph::DenseMatrix< oomph::CRDoubleMatrix * >, oomph::DenseMatrix< std::complex< double > >, oomph::DenseMatrix< bool >, oomph::CCComplexMatrix, oomph::CRComplexMatrix, and oomph::DenseComplexMatrix.

Referenced by oomph::Matrix< T, MATRIX_TYPE >::range_check(), and oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output().

nrow()

template<class T , class MATRIX_TYPE >

virtual unsigned long oomph::Matrix< T, MATRIX_TYPE >::nrow ( ) const

pure virtual

Return the number of rows of the matrix.

Implemented in oomph::SumOfMatrices, oomph::CCDoubleMatrix, oomph::DenseDoubleMatrix, oomph::CRDoubleMatrix, oomph::SparseMatrix< T, MATRIX_TYPE >, oomph::SparseMatrix< T, CCMatrix< T > >, oomph::SparseMatrix< T, CRMatrix< T > >, oomph::DenseMatrix< T >, oomph::DenseMatrix< oomph::MatrixVectorProduct * >, oomph::DenseMatrix< double >, oomph::DenseMatrix< int >, oomph::DenseMatrix< double * >, oomph::DenseMatrix< oomph::CRDoubleMatrix * >, oomph::DenseMatrix< std::complex< double > >, oomph::DenseMatrix< bool >, oomph::CCComplexMatrix, oomph::CRComplexMatrix, and oomph::DenseComplexMatrix.

Referenced by oomph::AxisymmetricNavierStokesEquations::fill_in_contribution_to_hessian_vector_products(), oomph::GeneralisedNewtonianAxisymmetricNavierStokesEquations::fill_in_contribution_to_hessian_vector_products(), oomph::Matrix< T, MATRIX_TYPE >::range_check(), and oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output().

operator()() [1/2]

template<class T , class MATRIX_TYPE >

T& oomph::Matrix< T, MATRIX_TYPE >::operator() ( const unsigned long &  i, const unsigned long &  j  )

inline

Round brackets to give access as a(i,j) for read-write access. The function uses the MATRIX_TYPE template parameter to call the entry() function which must be defined in all derived classes that are to be fully instantiated. If the particular Matrix does not allow write access, the function should break with an error message.

    {
      return static_cast<MATRIX_TYPE*>(this)->entry(i, j);
    }

References i, and j.

operator()() [2/2]

template<class T , class MATRIX_TYPE >

T oomph::Matrix< T, MATRIX_TYPE >::operator() ( const unsigned long &  i, const unsigned long &  j  ) const

inline

Round brackets to give access as a(i,j) for read only (we're not providing a general interface for component-wise write access since not all matrix formats allow efficient direct access!) The function uses the MATRIX_TYPE template parameter to call the get_entry() function which must be defined in all derived classes that are to be fully instantiated.

    {
      return static_cast<MATRIX_TYPE const*>(this)->get_entry(i, j);
    }

References i, and j.

Referenced by oomph::CRMatrix< T >::output_bottom_right_zero_helper(), and oomph::CCMatrix< T >::output_bottom_right_zero_helper().

operator=()

template<class T , class MATRIX_TYPE >

void oomph::Matrix< T, MATRIX_TYPE >::operator= ( const Matrix< T, MATRIX_TYPE > &  )

delete

Broken assignment operator.

output()

template<class T , class MATRIX_TYPE >

virtual void oomph::Matrix< T, MATRIX_TYPE >::output ( std::ostream &  outfile ) const

inlinevirtual

Output function to print a matrix row-by-row, in the form a(0,0) a(0,1) ... a(1,0) a(1,1) ... ... to the stream outfile. Broken virtual since it might not be sensible to implement this for some sparse matrices.

Reimplemented in oomph::DenseMatrix< T >, oomph::DenseMatrix< oomph::MatrixVectorProduct * >, oomph::DenseMatrix< double >, oomph::DenseMatrix< int >, oomph::DenseMatrix< double * >, oomph::DenseMatrix< oomph::CRDoubleMatrix * >, oomph::DenseMatrix< std::complex< double > >, and oomph::DenseMatrix< bool >.

    {
      throw OomphLibError(
        "Output function is not implemented for this matrix class",
        OOMPH_CURRENT_FUNCTION,
        OOMPH_EXCEPTION_LOCATION);
    }

References OOMPH_CURRENT_FUNCTION, and OOMPH_EXCEPTION_LOCATION.

output_bottom_right_zero_helper()

template<class T , class MATRIX_TYPE >

virtual void oomph::Matrix< T, MATRIX_TYPE >::output_bottom_right_zero_helper ( std::ostream &  outfile ) const

pure virtual

Output the "bottom right" entry regardless of it being zero or not (this allows automatic detection of matrix size in e.g. matlab, python). This functionality was moved from the function sparse_indexed_output(...) because at the moment, generalisation of this functionality does not work in parallel. CRDoubleMatrix has an nrow() function but it should it should use nrow_local() - which is the N variable in the underlaying CRMatrix.

Implemented in oomph::SumOfMatrices, oomph::CCMatrix< T >, oomph::CCMatrix< double >, oomph::CCMatrix< std::complex< double > >, oomph::CRDoubleMatrix, oomph::CRMatrix< T >, oomph::CRMatrix< double >, oomph::CRMatrix< std::complex< double > >, oomph::SparseMatrix< T, MATRIX_TYPE >, oomph::SparseMatrix< T, CCMatrix< T > >, oomph::SparseMatrix< T, CRMatrix< T > >, oomph::DenseMatrix< T >, oomph::DenseMatrix< oomph::MatrixVectorProduct * >, oomph::DenseMatrix< double >, oomph::DenseMatrix< int >, oomph::DenseMatrix< double * >, oomph::DenseMatrix< oomph::CRDoubleMatrix * >, oomph::DenseMatrix< std::complex< double > >, and oomph::DenseMatrix< bool >.

Referenced by oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output().

range_check()

template<class T , class MATRIX_TYPE >

void oomph::Matrix< T, MATRIX_TYPE >::range_check ( const unsigned long &  i, const unsigned long &  j  ) const

inlineprotected

Range check to catch when an index is out of bounds, if so, it issues a warning message and dies by throwing an OomphLibError

    {
      if (i >= nrow())
      {
        std::ostringstream error_message;
        error_message << "Range Error: i=" << i << " is not in the range (0,"
                      << nrow() - 1 << ")." << std::endl;
 
        throw OomphLibError(error_message.str(),
                            OOMPH_CURRENT_FUNCTION,
                            OOMPH_EXCEPTION_LOCATION);
      }
      else if (j >= ncol())
      {
        std::ostringstream error_message;
        error_message << "Range Error: j=" << j << " is not in the range (0,"
                      << ncol() - 1 << ")." << std::endl;
 
        throw OomphLibError(error_message.str(),
                            OOMPH_CURRENT_FUNCTION,
                            OOMPH_EXCEPTION_LOCATION);
      }
    }

References i, j, oomph::Matrix< T, MATRIX_TYPE >::ncol(), oomph::Matrix< T, MATRIX_TYPE >::nrow(), OOMPH_CURRENT_FUNCTION, and OOMPH_EXCEPTION_LOCATION.

Referenced by oomph::CRMatrix< T >::get_entry(), and oomph::CCMatrix< T >::get_entry().

sparse_indexed_output() [1/2]

template<class T , class MATRIX_TYPE >

void oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output ( std::ostream &  outfile, const unsigned &  precision = 0, const bool &  output_bottom_right_zero = false  ) const

inline

Indexed output function to print a matrix to the stream outfile as i,j,a(i,j) for a(i,j)!=0 only with specified precision (if precision=0 then nothing is changed). If optional boolean flag is set to true we also output the "bottom right" entry regardless of it being zero or not (this allows automatic detection of matrix size in e.g. matlab, python).

    {
      // Implemented as a wrapper around "sparse_indexed_output(std::ostream)"
      // so that only one output helper function is needed in derived classes.
 
      // We can't have separate functions for only "output_bottom_right_zero"
      // because people often write false as "0" and then C++ would pick the
      // wrong function.
 
      // If requested set the new precision and store the previous value.
      unsigned old_precision = 0;
      if (precision != 0)
      {
        old_precision = outfile.precision();
        outfile.precision(precision);
      }
 
      // Output as normal using the helper function defined in each matrix class
      sparse_indexed_output_helper(outfile);
 
      // If requested and there is no output for the last entry then output a
      // zero entry.
      if (output_bottom_right_zero && ncol() > 0 && nrow() > 0)
      {
        // Output as normal using the helper function defined
        // in each matrix class
        output_bottom_right_zero_helper(outfile);
      }
 
      // Restore the old value of the precision if we changed it
      if (precision != 0)
      {
        outfile.precision(old_precision);
      }
    }

References oomph::Matrix< T, MATRIX_TYPE >::ncol(), oomph::Matrix< T, MATRIX_TYPE >::nrow(), oomph::Matrix< T, MATRIX_TYPE >::output_bottom_right_zero_helper(), and oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output_helper().

Referenced by oomph::BlackBoxFDNewtonSolver::black_box_fd_newton_solve(), BiharmonicTestProblem1::dump_jacobian(), main(), print_elemental_jacobian(), and oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output().

sparse_indexed_output() [2/2]

template<class T , class MATRIX_TYPE >

void oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output ( std::string  filename, const unsigned &  precision = 0, const bool &  output_bottom_right_zero = false  ) const

inline

Indexed output function to print a matrix to the file named filename as i,j,a(i,j) for a(i,j)!=0 only with specified precision. If optional boolean flag is set to true we also output the "bottom right" entry regardless of it being zero or not (this allows automatic detection of matrix size in e.g. matlab, python).

    {
      // Implemented as a wrapper around "sparse_indexed_output(std::ostream)"
      // so that only one output function needs to be written in matrix
      // subclasses.
 
      // Open file
      std::ofstream some_file(filename.c_str());
 
      // Output as normal
      sparse_indexed_output(some_file, precision, output_bottom_right_zero);
 
      // Close file
      some_file.close();
    }

References MergeRestartFiles::filename, and oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output().

sparse_indexed_output_helper()

template<class T , class MATRIX_TYPE >

virtual void oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output_helper ( std::ostream &  outfile ) const

pure virtual

Indexed output function to print a matrix to the stream outfile as i,j,a(i,j) for a(i,j)!=0 only.

Implemented in oomph::SumOfMatrices, oomph::CCMatrix< T >, oomph::CCMatrix< double >, oomph::CCMatrix< std::complex< double > >, oomph::CRDoubleMatrix, oomph::CRMatrix< T >, oomph::CRMatrix< double >, oomph::CRMatrix< std::complex< double > >, oomph::SparseMatrix< T, MATRIX_TYPE >, oomph::SparseMatrix< T, CCMatrix< T > >, oomph::SparseMatrix< T, CRMatrix< T > >, oomph::DenseMatrix< T >, oomph::DenseMatrix< oomph::MatrixVectorProduct * >, oomph::DenseMatrix< double >, oomph::DenseMatrix< int >, oomph::DenseMatrix< double * >, oomph::DenseMatrix< oomph::CRDoubleMatrix * >, oomph::DenseMatrix< std::complex< double > >, and oomph::DenseMatrix< bool >.

Referenced by oomph::Matrix< T, MATRIX_TYPE >::sparse_indexed_output().

---

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

• matrices.h