UsenetSearch/src/Database.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 <filesystem>
#include <fstream>
#include <string>
#include <vector>

#include "usenetsearch/StringUtils.h"
#include "usenetsearch/UsenetClient.h"

#include "usenetsearch/Database.h"

namespace usenetsearch {

Database::~Database()
{
    if (m_newsGroupFileInput.is_open())
    {
        m_newsGroupFileInput.close();
    }
    if (m_newsGroupFileOutput.is_open())
    {
        m_newsGroupFileOutput.close();
    }
}

std::filesystem::path Database::GetArticleFilePath(
    const std::wstring& newsgroup)
{
    const auto groupFile = StringHash(m_conv.to_bytes(newsgroup)) + ".db";
    return m_databasePath / groupFile;
}

std::unique_ptr<std::vector<std::wstring>> Database::LoadArticleList(
    const std::wstring& newsgroup)
{
    const auto articleFile = GetArticleFilePath(newsgroup);
    if (!std::filesystem::exists(articleFile))
    {
        throw DatabaseException(ENOTFOUND,
            "No article list found for newsgroup " + m_conv.to_bytes(newsgroup)
        );
    }
    std::ifstream io;
    io.open(articleFile, std::ios::binary);
    std::uint64_t articleCount;
    io.read(
        reinterpret_cast<char*>(&articleCount),
        sizeof(articleCount)
    );
    auto result = std::make_unique<std::vector<std::wstring>>();
    for (std::uint64_t i = 0; i != articleCount; ++i)
    {
        std::wstring articleId;
        io >> articleId;
        result->emplace_back(articleId);
    }
    io.close();
    return result;
}

std::unique_ptr<std::vector<NntpListEntry>> Database::LoadNewsgroupList()
{
    OpenNewsGroupFile();

    std::uint64_t dbVersion{0};
    m_newsGroupFileInput.read(
        reinterpret_cast<char*>(&dbVersion),
        sizeof(dbVersion)
    );
    if (dbVersion != m_databaseVersion)
    {
        throw DatabaseException(EINVAL,
            "The loaded database version (" + std::to_string(dbVersion) 
            + ") does not match the current database version ("
            + std::to_string(m_databaseVersion) + ")");
    }

    size_t newsGroupCount{0};
    m_newsGroupFileInput.read(
        reinterpret_cast<char*>(&newsGroupCount),
        sizeof(newsGroupCount)
    );

    auto result = std::make_unique<std::vector<NntpListEntry>>();
    for (size_t numLoaded = 0; numLoaded != newsGroupCount; ++numLoaded)
    {
        NntpListEntry entry;
        m_newsGroupFileInput >> entry;
        result->emplace_back(entry);
    }
    return result;
}

void Database::Open(std::filesystem::path dbPath)
{
    m_databasePath = dbPath;
    if (!std::filesystem::exists(dbPath))
    {
        std::filesystem::create_directory(dbPath);
    }
    OpenNewsGroupFile();
}

void Database::OpenNewsGroupFile()
{
    if (m_newsGroupFileInput.is_open() && m_newsGroupFileOutput.is_open())
    {
        return;
    }
    const std::filesystem::path newsGroupFilePath =
        m_databasePath / "newsgroups.db";
    if (!m_newsGroupFileInput.is_open())
    {
        m_newsGroupFileInput.open(newsGroupFilePath, std::ios::binary);
    }
    if (!m_newsGroupFileOutput.is_open())
    {
        m_newsGroupFileOutput.open(newsGroupFilePath, std::ios::binary);
    }
}

void Database::UpdateArticleList(
    const std::wstring& newsgroup,
    const std::vector<std::wstring>& articleIds)
{
    const auto articleFile = GetArticleFilePath(newsgroup);
    std::ofstream io;
    io.open(articleFile, std::ios::binary);
    const std::uint64_t articleCount = articleIds.size();
    io.write(
        reinterpret_cast<const char*>(&articleCount),
        sizeof(articleCount)
    );
    for (const auto& id: articleIds)
    {
        io << id;
    }
    io.close();
}

void Database::UpdateNewsgroupList(const std::vector<NntpListEntry>& list)
{
    OpenNewsGroupFile();

    m_newsGroupFileOutput.write(
        reinterpret_cast<const char*>(&m_databaseVersion),
        sizeof(m_databaseVersion)
    );

    const std::uint64_t newsGroupCount = list.size();
    m_newsGroupFileOutput.write(
        reinterpret_cast<const char*>(&newsGroupCount),
        sizeof(newsGroupCount)
    );

    for (const auto& entry: list)
    {
        m_newsGroupFileOutput << entry;
    }
    m_newsGroupFileOutput.flush();
}

} // namespace usenetsearch