treebank_collection_processor.hpp

/*********************************************************************
 *
 * Linear Arrangement Library - A library that implements a collection
 * algorithms for linear arrangments of graphs.
 *
 * Copyright (C) 2019 - 2024
 *
 * This file is part of Linear Arrangement Library. The full code is available
 * at:
 *      https://github.com/LAL-project/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 (lluis.alemany.puig@upc.edu)
 *         LQMC (Quantitative, Mathematical, and Computational Linguisitcs)
 *         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)
 *         LQMC (Quantitative, Mathematical, and Computational Linguisitcs)
 *         CQL (Complexity and Quantitative Linguistics Lab)
 *         Office 220, 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>
#include <tuple>
 
// lal includes
#include <lal/io/treebank_file_error.hpp>
#include <lal/io/treebank_processor_base.hpp>
 
namespace lal {
namespace io {
 
class treebank_collection_processor : public _treebank_processor_base {
public:
    // SETTERS
 
    void set_join_files(const bool v) noexcept {
        m_join_files = v;
    }
 
    void set_number_threads(const std::size_t n_threads) noexcept {
#if defined DEBUG
        assert(n_threads != 0);
#endif
        m_num_threads = n_threads;
    }
 
    // GETTERS
 
    [[nodiscard]] std::size_t get_num_errors() const noexcept {
        return m_errors_from_processing.size();
    }
 
    [[nodiscard]] const treebank_file_error& get_error_type(const std::size_t i)
    const noexcept
    {
        return std::get<0>(m_errors_from_processing[i]);
    }
 
    [[nodiscard]] const std::string& get_error_treebank_filename(const std::size_t i)
    const noexcept
    {
        return std::get<1>(m_errors_from_processing[i]);
    }
 
    [[nodiscard]] const std::string& get_error_treebank_name(const std::size_t i)
    const noexcept
    {
        return std::get<2>(m_errors_from_processing[i]);
    }
 
    void set_join_to_file_name(const std::string& join_to) noexcept {
        m_join_to_file = join_to;
    }
 
    void set_treebank_column_name(const std::string& name) noexcept {
        m_treebank_column_name = name;
    }
 
    // PROCESS THE TREEBANK COLLECTION
 
    [[nodiscard]] treebank_file_error init
    (const std::string& main_file, const std::string& output_directory)
    noexcept;
 
    [[nodiscard]] treebank_file_error process() noexcept;
 
private:
    [[nodiscard]] treebank_file_error join_all_files() const noexcept;
 
private:
    std::vector<std::string> m_all_individual_treebank_ids;
    std::string m_join_to_file = "";
    bool m_join_files = true;
    std::string m_treebank_column_name = "treebank";
    std::size_t m_num_threads = 1;
    std::string m_column_join_name = "";
 
    std::vector<std::tuple<treebank_file_error, std::string, std::string>>
    m_errors_from_processing;
 
    std::string m_out_dir = "none";
    std::string m_main_file = "none";
};
 
[[nodiscard]] inline treebank_file_error process_treebank_collection
(
    const std::string& treebank_collection_main_file,
    const std::string& output_directory,
    const std::size_t num_threads = 1
)
noexcept
{
    treebank_collection_processor tbcolproc;
    auto err = tbcolproc.init(treebank_collection_main_file, output_directory);
    if (not err.is_error()) { return err; }
    tbcolproc.set_number_threads(num_threads);
    return tbcolproc.process();
}
 
} // -- namespace io
} // -- namespace lal