treebank_processor.hpp

/*********************************************************************
 *
 *  Linear Arrangement Library - A library that implements a collection
 *  algorithms for linear arrangments of graphs.
 *
 *  Copyright (C) 2019 - 2021
 *
 *  This file is part of Linear Arrangement Library. To see the full code
 *  visit the webpage:
 *      https://github.com/lluisalemanypuig/linear-arrangement-library.git
 *
 *  Linear Arrangement Library is free software: you can redistribute it
 *  and/or modify it under the terms of the GNU Affero General Public License
 *  as published by the Free Software Foundation, either version 3 of the
 *  License, or (at your option) any later version.
 *
 *  Linear Arrangement Library 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 Affero General Public License for more details.
 *
 *  You should have received a copy of the GNU Affero General Public License
 *  along with Linear Arrangement Library.  If not, see <http://www.gnu.org/licenses/>.
 *
 *  Contact:
 *
 *      Lluís Alemany Puig (lalemany@cs.upc.edu)
 *          LARCA (Laboratory for Relational Algorithmics, Complexity and Learning)
 *          CQL (Complexity and Quantitative Linguistics Lab)
 *          Jordi Girona St 1-3, Campus Nord UPC, 08034 Barcelona.   CATALONIA, SPAIN
 *          Webpage: https://cqllab.upc.edu/people/lalemany/
 *
 *      Ramon Ferrer i Cancho (rferrericancho@cs.upc.edu)
 *          LARCA (Laboratory for Relational Algorithmics, Complexity and Learning)
 *          CQL (Complexity and Quantitative Linguistics Lab)
 *          Office S124, Omega building
 *          Jordi Girona St 1-3, Campus Nord UPC, 08034 Barcelona.   CATALONIA, SPAIN
 *          Webpage: https://cqllab.upc.edu/people/rferrericancho/
 *
 ********************************************************************/
 
#pragma once
 
// C++ includes
#if defined DEBUG
#include <cassert>
#endif
#include <vector>
#include <string>
 
// lal includes
#include <lal/io/treebank_error.hpp>
#include <lal/io/process_treebank_base.hpp>
 
namespace lal {
namespace io {
 
class treebank_processor : public _process_treebank_base {
public:
    // PROCESS THE TREEBANK collection
 
    treebank_error init(
        const std::string& treebank_input_file,
        const std::string& output_file,
        const std::string& treebank_id = ""
    )
    noexcept;
 
    treebank_error process() noexcept;
 
private:
    template<class TREE, class OUT_STREAM>
    void process_tree
    (const TREE& rT, double *props, char *prop_set, OUT_STREAM& out_lab_file)
    const noexcept;
 
    // HEADER
 
    template<class OUT_STREAM>
    void output_tree_type_header
    (OUT_STREAM& out_lab_file)
    const noexcept;
 
    template<class OUT_STREAM>
    void output_syndepstruct_type_header
    (OUT_STREAM& out_lab_file)
    const noexcept;
 
    // VALUES
 
    template<class TREE_TYPE, class OUT_STREAM>
    void output_tree_type_values
    (TREE_TYPE& t, OUT_STREAM& out_lab_file)
    const noexcept;
 
    template<class TREE_TYPE, class OUT_STREAM>
    void output_syndepstruct_type_values
    (const TREE_TYPE& t, OUT_STREAM& out_lab_file)
    const noexcept;
 
private:
    std::string m_treebank_filename = "none";
    std::string m_output_file = "none";
    std::string m_treebank_id = "";
};
 
inline
treebank_error process_treebank(
    const std::string& treebank_file,
    const std::string& output_file
)
noexcept
{
    treebank_processor tbproc;
    auto err = tbproc.init(treebank_file, output_file);
    if (err.get_error_type() != treebank_error_type::no_error) {
        return err;
    }
    return tbproc.process();
}
 
} // -- namespace io
} // -- namespace lal