ClickHouse/dbms/src/Dictionaries/Embedded/RegionsNames.h

#pragma once

#include <string>
#include <vector>
#include <Poco/Exception.h>
#include <common/StringRef.h>
#include <common/Types.h>
#include "GeodataProviders/INamesProvider.h"


/** A class that allows you to recognize by region id its text name in one of the supported languages: ru, en, ua, by, kz, tr.
  *
  * Information about region names loads from text files with the following format names:
  *     regions_names_xx.txt,
  * where xx is one of the two letters of the following supported languages:
  *     ru, en, ua, by, kz, tr.
  *
  * Can on request update the data.
  */
class RegionsNames
{
public:
    enum class Language : size_t
    {
        RU = 0,
        EN,
        UA,
        BY,
        KZ,
        TR,

        END
    };

private:
    static inline constexpr const char * supported_languages[] =
    {
        "ru",
        "en",
        "ua",
        "by",
        "kz",
        "tr"
    };

    static inline constexpr std::pair<const char *, Language> language_aliases[] =
    {
        {"ru", Language::RU},
        {"en", Language::EN},
        {"ua", Language::UA},
        {"uk", Language::UA},
        {"by", Language::BY},
        {"kz", Language::KZ},
        {"tr", Language::TR}
    };

    static constexpr size_t ROOT_LANGUAGE = 0;
    static constexpr size_t SUPPORTED_LANGUAGES_COUNT = size_t(Language::END);
    static constexpr size_t LANGUAGE_ALIASES_COUNT = sizeof(language_aliases);

    using NamesSources = std::vector<std::shared_ptr<ILanguageRegionsNamesDataSource>>;

    using Chars = std::vector<char>;
    using CharsForLanguageID = std::vector<Chars>;
    using StringRefs = std::vector<StringRef>; /// Lookup table RegionID -> StringRef
    using StringRefsForLanguageID = std::vector<StringRefs>;

public:
    RegionsNames(IRegionsNamesDataProviderPtr data_provider);

    StringRef getRegionName(RegionID region_id, Language language = Language::RU) const
    {
        size_t language_id = static_cast<size_t>(language);

        if (region_id >= names_refs[language_id].size())
            return StringRef("", 0);

        StringRef ref = names_refs[language_id][region_id];

        while (ref.size == 0 && language_id != ROOT_LANGUAGE)
        {
            static const size_t FALLBACK[] = {0, 0, 0, 0, 0, 1};
            language_id = FALLBACK[language_id];
            ref = names_refs[language_id][region_id];
        }

        return ref;
    }

    static Language getLanguageEnum(const std::string & language)
    {
        if (language.size() == 2)
        {
            for (size_t i = 0; i < LANGUAGE_ALIASES_COUNT; ++i)
            {
                const auto & alias = language_aliases[i];
                if (language[0] == alias.first[0] && language[1] == alias.first[1])
                    return alias.second;
            }
        }
        throw Poco::Exception("Unsupported language for region name. Supported languages are: " + dumpSupportedLanguagesNames() + ".");
    }

    void reload();

private:
    static std::string dumpSupportedLanguagesNames();

    NamesSources names_sources = NamesSources(SUPPORTED_LANGUAGES_COUNT);

    /// Bytes of names for each language, laid out in a row, separated by zeros
    CharsForLanguageID chars = CharsForLanguageID(SUPPORTED_LANGUAGES_COUNT);

    /// Mapping for each language from the region id into a pointer to the byte range of the name
    StringRefsForLanguageID names_refs = StringRefsForLanguageID(SUPPORTED_LANGUAGES_COUNT);
};