// // Copyright (c) 2009-2011 Artyom Beilis (Tonkikh) // // Distributed under the Boost Software License, Version 1.0. // https://www.boost.org/LICENSE_1_0.txt #ifndef BOOST_LOCALE_CONVERTER_HPP_INCLUDED #define BOOST_LOCALE_CONVERTER_HPP_INCLUDED #include #include #include #ifdef BOOST_MSVC # pragma warning(push) # pragma warning(disable : 4275 4251 4231 4660) #endif namespace boost { namespace locale { /// \defgroup convert Text Conversions /// /// This module provides various function for string manipulation like Unicode normalization, case conversion etc. /// @{ /// \brief This class provides base flags for text manipulation. It is used as base for converter facet. class converter_base { public: /// The flag used for facet - the type of operation to perform enum conversion_type { normalization, ///< Apply Unicode normalization on the text upper_case, ///< Convert text to upper case lower_case, ///< Convert text to lower case case_folding, ///< Fold case in the text title_case ///< Convert text to title case }; }; template class converter; #ifdef BOOST_LOCALE_DOXYGEN /// /// \brief The facet that implements text manipulation /// /// It is used to performs text conversion operations defined by \ref converter_base::conversion_type. /// It is specialized for four types of characters \c char, \c wchar_t, \c char16_t, \c char32_t template class BOOST_LOCALE_DECL converter : public converter_base, public std::locale::facet { public: /// Locale identification static std::locale::id id; /// Standard constructor converter(size_t refs = 0) : std::locale::facet(refs) {} /// Convert text in range [\a begin, \a end) according to conversion method \a how. Parameter /// \a flags is used for specification of normalization method like nfd, nfc etc. virtual std::basic_string convert(conversion_type how, const Char* begin, const Char* end, int flags = 0) const = 0; }; #else template<> class BOOST_LOCALE_DECL converter : public converter_base, public std::locale::facet { public: static std::locale::id id; converter(size_t refs = 0) : std::locale::facet(refs) {} ~converter(); virtual std::string convert(conversion_type how, const char* begin, const char* end, int flags = 0) const = 0; }; template<> class BOOST_LOCALE_DECL converter : public converter_base, public std::locale::facet { public: static std::locale::id id; converter(size_t refs = 0) : std::locale::facet(refs) {} ~converter(); virtual std::wstring convert(conversion_type how, const wchar_t* begin, const wchar_t* end, int flags = 0) const = 0; }; # ifdef BOOST_LOCALE_ENABLE_CHAR16_T template<> class BOOST_LOCALE_DECL converter : public converter_base, public std::locale::facet { public: static std::locale::id id; converter(size_t refs = 0) : std::locale::facet(refs) {} ~converter(); virtual std::u16string convert(conversion_type how, const char16_t* begin, const char16_t* end, int flags = 0) const = 0; }; # endif # ifdef BOOST_LOCALE_ENABLE_CHAR32_T template<> class BOOST_LOCALE_DECL converter : public converter_base, public std::locale::facet { public: static std::locale::id id; converter(size_t refs = 0) : std::locale::facet(refs) {} ~converter(); virtual std::u32string convert(conversion_type how, const char32_t* begin, const char32_t* end, int flags = 0) const = 0; }; # endif #endif /// The type that defined normalization form enum norm_type { norm_nfd, ///< Canonical decomposition norm_nfc, ///< Canonical decomposition followed by canonical composition norm_nfkd, ///< Compatibility decomposition norm_nfkc, ///< Compatibility decomposition followed by canonical composition. norm_default = norm_nfc, ///< Default normalization - canonical decomposition followed by canonical composition }; /// Normalize Unicode string in range [begin,end) according to \ref norm_type "normalization form" \a n /// /// Note: This function receives only Unicode strings, i.e.: UTF-8, UTF-16 or UTF-32. It does not take /// in account the locale encoding, because Unicode decomposition and composition are meaningless outside /// of a Unicode character set. /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string normalize(const CharType* begin, const CharType* end, norm_type n = norm_default, const std::locale& loc = std::locale()) { return std::use_facet>(loc).convert(converter_base::normalization, begin, end, n); } /// Normalize Unicode string \a str according to \ref norm_type "normalization form" \a n /// /// Note: This function receives only Unicode strings, i.e.: UTF-8, UTF-16 or UTF-32. It does not take /// in account the locale encoding, because Unicode decomposition and composition are meaningless outside /// of a Unicode character set. /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string normalize(const std::basic_string& str, norm_type n = norm_default, const std::locale& loc = std::locale()) { return normalize(str.data(), str.data() + str.size(), n, loc); } /// Normalize NULL terminated Unicode string \a str according to \ref norm_type "normalization form" \a n /// /// Note: This function receives only Unicode strings, i.e.: UTF-8, UTF-16 or UTF-32. It does not take /// in account the locale encoding, because Unicode decomposition and composition are meaningless outside /// of a Unicode character set. /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string normalize(const CharType* str, norm_type n = norm_default, const std::locale& loc = std::locale()) { return normalize(str, util::str_end(str), n, loc); } /////////////////////////////////////////////////// /// Convert a string in range [begin,end) to upper case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_upper(const CharType* begin, const CharType* end, const std::locale& loc = std::locale()) { return std::use_facet>(loc).convert(converter_base::upper_case, begin, end); } /// Convert a string \a str to upper case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_upper(const std::basic_string& str, const std::locale& loc = std::locale()) { return to_upper(str.data(), str.data() + str.size(), loc); } /// Convert a NULL terminated string \a str to upper case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_upper(const CharType* str, const std::locale& loc = std::locale()) { return to_upper(str, util::str_end(str), loc); } /////////////////////////////////////////////////// /// Convert a string in range [begin,end) to lower case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_lower(const CharType* begin, const CharType* end, const std::locale& loc = std::locale()) { return std::use_facet>(loc).convert(converter_base::lower_case, begin, end); } /// Convert a string \a str to lower case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_lower(const std::basic_string& str, const std::locale& loc = std::locale()) { return to_lower(str.data(), str.data() + str.size(), loc); } /// Convert a NULL terminated string \a str to lower case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_lower(const CharType* str, const std::locale& loc = std::locale()) { return to_lower(str, util::str_end(str), loc); } /////////////////////////////////////////////////// /// Convert a string in range [begin,end) to title case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_title(const CharType* begin, const CharType* end, const std::locale& loc = std::locale()) { return std::use_facet>(loc).convert(converter_base::title_case, begin, end); } /// Convert a string \a str to title case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_title(const std::basic_string& str, const std::locale& loc = std::locale()) { return to_title(str.data(), str.data() + str.size(), loc); } /// Convert a NULL terminated string \a str to title case according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string to_title(const CharType* str, const std::locale& loc = std::locale()) { return to_title(str, util::str_end(str), loc); } /////////////////////////////////////////////////// /// Fold case of a string in range [begin,end) according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string fold_case(const CharType* begin, const CharType* end, const std::locale& loc = std::locale()) { return std::use_facet>(loc).convert(converter_base::case_folding, begin, end); } /// Fold case of a string \a str according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string fold_case(const std::basic_string& str, const std::locale& loc = std::locale()) { return fold_case(str.data(), str.data() + str.size(), loc); } /// Fold case of a NULL terminated string \a str according to locale \a loc /// /// \note throws std::bad_cast if loc does not have \ref converter facet installed template std::basic_string fold_case(const CharType* str, const std::locale& loc = std::locale()) { return fold_case(str, util::str_end(str), loc); } ///@} }} // namespace boost::locale #ifdef BOOST_MSVC # pragma warning(pop) #endif /// \example conversions.cpp /// /// Example of using various text conversion functions. /// /// \example wconversions.cpp /// /// Example of using various text conversion functions with wide strings. #endif