UsenetSearch/src/Configuration.cpp

/*
    Copyright© 2021 John Sennesael

    UsenetSearch is Free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.

    UsenetSearch 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.

    You should have received a copy of the GNU General Public License
    along with UsenetSearch.  If not, see <https://www.gnu.org/licenses/>.
*/

#include "usenetsearch/Configuration.h"

#include "usenetsearch/Logger.h"
#include "usenetsearch/ScopeExit.h"
#include "usenetsearch/StringUtils.h"

#include <filesystem>
#include <fstream>
#include <regex>
#include <string>

namespace usenetsearch {

std::uint16_t Configuration::BatchSize() const
{
    return m_batchSize;
}

std::filesystem::path Configuration::DatabasePath() const
{
    return m_databasePath;
}

std::vector<std::string>& Configuration::FilterEraseSubtoken()
{
    return m_filterEraseSubtoken;
}

std::vector<std::regex>& Configuration::FilterNewsgroupBlacklist()
{
    return m_filterNewsgroupBlacklist;
}

std::vector<std::regex>& Configuration::FilterNewsgroupWhitelist()
{
    return m_filterNewsgroupWhitelist;
}

std::vector<std::string>& Configuration::FilterWordsNoSubtoken()
{
    return m_filterWordsNoSubtoken;
}

std::uint16_t Configuration::MaxThreads() const
{
    return m_maxThreads;
}

std::uint8_t Configuration::MaxTreeDepth() const
{
    return m_maxTreeDepth;
}

std::uint16_t Configuration::MinSubtokenWords() const
{
    return m_minSubtokenWords;
}

std::string Configuration::NNTPServerHost() const
{
    return m_nntpServerHost;
}

std::string Configuration::NNTPServerPassword() const
{
    return m_nntpServerPassword;
}

int Configuration::NNTPServerPort() const
{
    return m_nntpServerPort;
}

bool Configuration::NNTPServerSSL() const
{
    return m_nntpServerSSL;
}

std::string Configuration::NNTPServerUser() const
{
    return m_nntpServerUser;
}

void Configuration::Open(const std::string& filename)
{
    std::string line;
    std::ifstream fin(filename.c_str());
    if (!fin.is_open())
    {
        Logger::Get().Fatal<ConfigurationException>(
            LOGID("Configuration"),
            "Could not open configuration file: " + filename
        );
    }
    ScopeExit finCloser([&fin](){ fin.close(); });
    int line_nr = 0;
    while(std::getline(fin,line))
    {
        line_nr++;
        line = StringTrim(line);
        // Immediately skip blank lines.
        if (line == "") continue;
        // Skip comments.
        if (StringStartsWith("#",line)==true) continue;
        // Split line in key-value pair.
        const auto kvp = StringSplit(line, std::string{":"}, 2);
        if (kvp.size() != 2)
        {
            Logger::Get().Fatal<ConfigurationException>(
                LOGID("Configuration"),
                std::string("Invalid configuration in ")
                + filename + std::string(" line ")
                + std::to_string(line_nr)
            );
        }
        const std::string key = StringToLower(kvp[0]);
        const std::string value = StringTrim(kvp[1]);
        if (key == "batch_size")
        {
            m_batchSize = stoi(value);
        }
        else if (key == "database_path")
        {
            m_databasePath = value;
        }
        else if (key == "filter_erase_subtoken")
        {
            const auto tokens = StringSplit(value, std::string{","});
            for (const auto& token: tokens)
            {
                const auto trimmedToken = StringToLower(StringTrim(token));
                if (trimmedToken != "")
                {
                    m_filterEraseSubtoken.emplace_back(trimmedToken);
                }
            }
        }
        else if (key == "filter_newsgroup_blacklist")
        {
            try 
            {
                std::regex re(value);
                m_filterNewsgroupBlacklist.emplace_back(re);
            }
            catch (const std::regex_error& e)
            {
                Logger::Get().Fatal<ConfigurationException>(
                    LOGID("Configuration"),
                    std::string("Invalid configuration in ")
                    + filename + std::string(" line ")
                    + std::to_string(line_nr) + " : Regular expression \""
                    + value + "\" did not parse: " + e.what()
                );
            }
        }
        else if (key == "filter_newsgroup_whitelist")
        {
            try 
            {
                std::regex re(value);
                m_filterNewsgroupWhitelist.emplace_back(re);
            }
            catch (const std::regex_error& e)
            {
                Logger::Get().Fatal<ConfigurationException>(
                    LOGID("Configuration"),
                    std::string("Invalid configuration in ")
                    + filename + std::string(" line ")
                    + std::to_string(line_nr) + " : Regular expression \""
                    + value + "\" did not parse: " + e.what()
                );
            }
        }
        else if (key == "filter_no_subtoken")
        {
            const auto tokens = StringSplit(value, std::string{","});
            for (const auto& token: tokens)
            {
                const auto trimmedToken = StringToLower(StringTrim(token));
                if (trimmedToken != "")
                {
                    m_filterWordsNoSubtoken.emplace_back(trimmedToken);
                }
            }
        }
        else if (key == "max_threads")
        {
            m_maxThreads = stoi(value);
        }
        else if (key == "max_tree_depth")
        {
            m_maxTreeDepth = stoi(value);
        }
        else if (key == "minimum_subtoken_words")
        {
            m_minSubtokenWords = stoi(value);
        }
        else if (key == "nntp_server_host")
        {
            m_nntpServerHost = value;
        }
        else if (key == "nntp_server_pass")
        {
            m_nntpServerPassword = value;
        }
        else if (key == "nntp_server_port")
        {
            m_nntpServerPort = stoi(value);
        }
        else if (key == "nntp_server_use_ssl")
        {
            try
            {
                m_nntpServerSSL = StringToBoolean(value);
            }
            catch (const StringException& e)
            {
                fin.close();
                Logger::Get().Fatal<ConfigurationException>(
                    LOGID("Configuration"),
                    std::string("Invalid configuration in ")
                    + filename + std::string(" line ")
                    + std::to_string(line_nr) + " - " + e.what()
                );
            }
        }
        else if (key == "nntp_server_user")
        {
            m_nntpServerUser = value;
        }
        else
        {
            Logger::Get().Fatal<ConfigurationException>(
                LOGID("Configuration"),
                std::string("Invalid configuration in ")
                + filename + std::string(" line ")
                + std::to_string(line_nr)
            );
        }
    }
}

} // namespace usenetsearch