field_elementwise.hh

Go to the documentation of this file.

/*!
 *
 * Copyright (C) 2015 Technical University of Liberec.  All rights reserved.
 * 
 * This program is free software; you can redistribute it and/or modify it under
 * the terms of the GNU General Public License version 3 as published by the
 * Free Software Foundation. (http://www.gnu.org/licenses/gpl-3.0.en.html)
 * 
 * This program is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
 * FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more details.
 *
 * 
 * @file    field_elementwise.hh
 * @brief   
 */

#ifndef FIELD_ELEMENTWISE_HH_
#define FIELD_ELEMENTWISE_HH_

/**
 * The simplest approximative Field. In future we want to replace it by pair: DofHandler + vector of dof values.
 *
 * In current setting:
 * This should implement mapping: (ElementAccessor.INDEX) -> value
 * Currently INDEX is global index of element (any dimension, both bulk and boundary part). In future INDEX should
 * probably consist of MESH_LEVEL (unique mesh, dimension, bulk/boundary, level of refinement) and INDEX within this level.
 * We want to make memory optimization since usually field lives either on boundary or on bulk part and some bulk fields live only on some dimension(s).
 * This can be achieved by two level indirection table of mesh_levelscontaining tables for indexes. We should test if the performance penalty is not to big.
 *
 * Currently, we just use one vector for bulk and one for boundary elements.
 *
 * TODO:
 * - move raw access resolution functions from FieldValues_ into FieldElementwise
 * - allow elementwise int or FieldEnum data with optimal storage buffer, this needs
 *   templated GMSH reader
 * - After this do following cleanup:
 *   Partitioning::subdomain_id_field_data should return vector<int>
 *   pertitioning_test.cpp should make correct test.
 *
 * - allow initialization of multiple fields by one reader
 * - allow common storage for more elementwise fields to have values for one element on one place
 */

#include "system/system.hh"
#include "fields/field_algo_base.hh"
#include "input/factory.hh"

class GmshMeshReader;

template <int spacedim, class Value>
class FieldElementwise : public FieldAlgorithmBase<spacedim, Value>
{
public:
    typedef typename FieldAlgorithmBase<spacedim, Value>::Point Point;
    typedef FieldAlgorithmBase<spacedim, Value> FactoryBaseType;

    FieldElementwise(unsigned int n_comp=0);

    /**
     * Alternative to previous constructor.
     */
    FieldElementwise(std::shared_ptr< std::vector<typename Value::element_type> > data, unsigned int n_components);

    static const Input::Type::Record & get_input_type();

    virtual void init_from_input(const Input::Record &rec, const struct FieldAlgoBaseInitData& init_data) override;

    /**
     * Set row of boundary data. Used to implement old BC input.
     */
    //void set_data_row(unsigned int boundary_idx, typename Value::return_type &value);

    /**
     * Update time and possibly update data from GMSH file.
     */
    bool set_time(const TimeStep &time) override;

    /**
     * Has to be set before calling init_from_input. This also
     * allocate and initialize internal buffer. Do nothing if mesh is already set.
     *
     * See also description of the FieldBase<...>::set_mesh.
     */
    void set_mesh(const Mesh *mesh, bool boundary_domain) override;


    /**
     * Returns one value in one given point. ResultType can be used to avoid some costly calculation if the result is trivial.
     */
    typename Value::return_type const &value(const Point &p, const ElementAccessor<spacedim> &elm) override;

    /**
     * Returns std::vector of scalar values in several points at once.
     */
    void value_list (const std::vector< Point >  &point_list, const ElementAccessor<spacedim> &elm,
                       std::vector<typename Value::return_type>  &value_list) override;


    virtual ~FieldElementwise();

private:
    /// Multiply @p data_ with @p unit_conversion_coefficient_ and check limits of field
    void scale_and_check_limits();
    /// Is flase whne the data vector is provided at construction. Then, we disallow initialization form input
    /// and do not delete data pointer in destructor.
    bool internal_raw_data;
    /**
     * Is set in set_mesh method. Value true means, that we accept only boundary element accessors in the @p value method.
     * TODO: temporary solution until we have separate mesh for the boundary part
     */
    bool boundary_domain_;
    /// Raw buffer of n_entities rows each containing Value::size() doubles.
    std::shared_ptr< std::vector<typename Value::element_type> > data_;
    /// Number of rows in @p data_ buffer.
    unsigned int n_entities_;
    /// Size of Value
    unsigned int n_components_;

    FilePath reader_file_;
    const Mesh *mesh_;
    std::string field_name_;

    /// Accessor to Input::Record
    Input::Record in_rec_;

    /**
     * Initialization data of field. Necessary for check limits.
     *
     * TODO: Temporary solution will be replaced with shared_ptr to field data in FieldAlgoBase
     */
    UnitSI &unit_si_;
    std::pair<double, double> limits_;

    /// Registrar of class to factory
    static const int registrar;
};


#endif /* FIELD_ELEMENTWISE_HH_ */