From 0b7d58968b37dc9d36a90d3fb439891a57b81a18 Mon Sep 17 00:00:00 2001 From: "Matthias P. Braendli" Date: Wed, 17 Jan 2024 09:48:56 +0100 Subject: Update ClockTAI and RemoteControl --- lib/charset/utf8/checked.h | 176 ++++++++++++++++----------- lib/charset/utf8/core.h | 275 ++++++++++++++++++++++++++++++++++--------- lib/charset/utf8/cpp11.h | 70 +++++++++++ lib/charset/utf8/cpp17.h | 96 +++++++++++++++ lib/charset/utf8/cpp20.h | 124 +++++++++++++++++++ lib/charset/utf8/unchecked.h | 163 +++++++++++++++++-------- 6 files changed, 724 insertions(+), 180 deletions(-) create mode 100644 lib/charset/utf8/cpp11.h create mode 100644 lib/charset/utf8/cpp17.h create mode 100644 lib/charset/utf8/cpp20.h (limited to 'lib/charset/utf8') diff --git a/lib/charset/utf8/checked.h b/lib/charset/utf8/checked.h index 1331155..98949f8 100644 --- a/lib/charset/utf8/checked.h +++ b/lib/charset/utf8/checked.h @@ -1,4 +1,4 @@ -// Copyright 2006 Nemanja Trifunovic +// Copyright 2006-2016 Nemanja Trifunovic /* Permission is hereby granted, free of charge, to any person or organization @@ -39,64 +39,62 @@ namespace utf8 // Exceptions that may be thrown from the library functions. class invalid_code_point : public exception { - uint32_t cp; + utfchar32_t cp; public: - invalid_code_point(uint32_t cp) : cp(cp) {} - virtual const char* what() const throw() { return "Invalid code point"; } - uint32_t code_point() const {return cp;} + invalid_code_point(utfchar32_t codepoint) : cp(codepoint) {} + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid code point"; } + utfchar32_t code_point() const {return cp;} }; class invalid_utf8 : public exception { - uint8_t u8; + utfchar8_t u8; public: - invalid_utf8 (uint8_t u) : u8(u) {} - virtual const char* what() const throw() { return "Invalid UTF-8"; } - uint8_t utf8_octet() const {return u8;} + invalid_utf8 (utfchar8_t u) : u8(u) {} + invalid_utf8 (char c) : u8(static_cast(c)) {} + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid UTF-8"; } + utfchar8_t utf8_octet() const {return u8;} }; class invalid_utf16 : public exception { - uint16_t u16; + utfchar16_t u16; public: - invalid_utf16 (uint16_t u) : u16(u) {} - virtual const char* what() const throw() { return "Invalid UTF-16"; } - uint16_t utf16_word() const {return u16;} + invalid_utf16 (utfchar16_t u) : u16(u) {} + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid UTF-16"; } + utfchar16_t utf16_word() const {return u16;} }; class not_enough_room : public exception { public: - virtual const char* what() const throw() { return "Not enough space"; } + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Not enough space"; } }; /// The library API - functions intended to be called by the users template - octet_iterator append(uint32_t cp, octet_iterator result) + octet_iterator append(utfchar32_t cp, octet_iterator result) { if (!utf8::internal::is_code_point_valid(cp)) throw invalid_code_point(cp); - if (cp < 0x80) // one octet - *(result++) = static_cast(cp); - else if (cp < 0x800) { // two octets - *(result++) = static_cast((cp >> 6) | 0xc0); - *(result++) = static_cast((cp & 0x3f) | 0x80); - } - else if (cp < 0x10000) { // three octets - *(result++) = static_cast((cp >> 12) | 0xe0); - *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); - *(result++) = static_cast((cp & 0x3f) | 0x80); - } - else { // four octets - *(result++) = static_cast((cp >> 18) | 0xf0); - *(result++) = static_cast(((cp >> 12) & 0x3f) | 0x80); - *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); - *(result++) = static_cast((cp & 0x3f) | 0x80); - } - return result; + return internal::append(cp, result); + } + + inline void append(utfchar32_t cp, std::string& s) + { + append(cp, std::back_inserter(s)); + } + + template + word_iterator append16(utfchar32_t cp, word_iterator result) + { + if (!utf8::internal::is_code_point_valid(cp)) + throw invalid_code_point(cp); + + return internal::append16(cp, result); } template - output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_t replacement) + output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, utfchar32_t replacement) { while (start != end) { octet_iterator sequence_start = start; @@ -107,7 +105,9 @@ namespace utf8 *out++ = *it; break; case internal::NOT_ENOUGH_ROOM: - throw not_enough_room(); + out = utf8::append (replacement, out); + start = end; + break; case internal::INVALID_LEAD: out = utf8::append (replacement, out); ++start; @@ -129,14 +129,28 @@ namespace utf8 template inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out) { - static const uint32_t replacement_marker = utf8::internal::mask16(0xfffd); + static const utfchar32_t replacement_marker = utf8::internal::mask16(0xfffd); return utf8::replace_invalid(start, end, out, replacement_marker); } + inline std::string replace_invalid(const std::string& s, utfchar32_t replacement) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::string replace_invalid(const std::string& s) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + template - uint32_t next(octet_iterator& it, octet_iterator end) + utfchar32_t next(octet_iterator& it, octet_iterator end) { - uint32_t cp = 0; + utfchar32_t cp = 0; internal::utf_error err_code = utf8::internal::validate_next(it, end, cp); switch (err_code) { case internal::UTF8_OK : @@ -146,21 +160,31 @@ namespace utf8 case internal::INVALID_LEAD : case internal::INCOMPLETE_SEQUENCE : case internal::OVERLONG_SEQUENCE : - throw invalid_utf8(*it); + throw invalid_utf8(static_cast(*it)); case internal::INVALID_CODE_POINT : throw invalid_code_point(cp); } return cp; } + template + utfchar32_t next16(word_iterator& it, word_iterator end) + { + utfchar32_t cp = 0; + internal::utf_error err_code = utf8::internal::validate_next16(it, end, cp); + if (err_code == internal::NOT_ENOUGH_ROOM) + throw not_enough_room(); + return cp; + } + template - uint32_t peek_next(octet_iterator it, octet_iterator end) + utfchar32_t peek_next(octet_iterator it, octet_iterator end) { return utf8::next(it, end); } template - uint32_t prior(octet_iterator& it, octet_iterator start) + utfchar32_t prior(octet_iterator& it, octet_iterator start) { // can't do much if it == start if (it == start) @@ -174,23 +198,19 @@ namespace utf8 return utf8::peek_next(it, end); } - /// Deprecated in versions that include "prior" - template - uint32_t previous(octet_iterator& it, octet_iterator pass_start) - { - octet_iterator end = it; - while (utf8::internal::is_trail(*(--it))) - if (it == pass_start) - throw invalid_utf8(*it); // error - no lead byte in the sequence - octet_iterator temp = it; - return utf8::next(temp, end); - } - template void advance (octet_iterator& it, distance_type n, octet_iterator end) { - for (distance_type i = 0; i < n; ++i) - utf8::next(it, end); + const distance_type zero(0); + if (n < zero) { + // backward + for (distance_type i = n; i < zero; ++i) + utf8::prior(it, end); + } else { + // forward + for (distance_type i = zero; i < n; ++i) + utf8::next(it, end); + } } template @@ -207,23 +227,23 @@ namespace utf8 octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) { while (start != end) { - uint32_t cp = utf8::internal::mask16(*start++); + utfchar32_t cp = utf8::internal::mask16(*start++); // Take care of surrogate pairs first if (utf8::internal::is_lead_surrogate(cp)) { if (start != end) { - uint32_t trail_surrogate = utf8::internal::mask16(*start++); + const utfchar32_t trail_surrogate = utf8::internal::mask16(*start++); if (utf8::internal::is_trail_surrogate(trail_surrogate)) cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; else - throw invalid_utf16(static_cast(trail_surrogate)); + throw invalid_utf16(static_cast(trail_surrogate)); } else - throw invalid_utf16(static_cast(cp)); + throw invalid_utf16(static_cast(cp)); } // Lone trail surrogate else if (utf8::internal::is_trail_surrogate(cp)) - throw invalid_utf16(static_cast(cp)); + throw invalid_utf16(static_cast(cp)); result = utf8::append(cp, result); } @@ -233,14 +253,14 @@ namespace utf8 template u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) { - while (start != end) { - uint32_t cp = utf8::next(start, end); + while (start < end) { + const utfchar32_t cp = utf8::next(start, end); if (cp > 0xffff) { //make a surrogate pair - *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); - *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); } else - *result++ = static_cast(cp); + *result++ = static_cast(cp); } return result; } @@ -257,7 +277,7 @@ namespace utf8 template u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) { - while (start != end) + while (start < end) (*result++) = utf8::next(start, end); return result; @@ -265,23 +285,28 @@ namespace utf8 // The iterator class template - class iterator : public std::iterator { + class iterator { octet_iterator it; octet_iterator range_start; octet_iterator range_end; public: + typedef utfchar32_t value_type; + typedef utfchar32_t* pointer; + typedef utfchar32_t& reference; + typedef std::ptrdiff_t difference_type; + typedef std::bidirectional_iterator_tag iterator_category; iterator () {} explicit iterator (const octet_iterator& octet_it, - const octet_iterator& range_start, - const octet_iterator& range_end) : - it(octet_it), range_start(range_start), range_end(range_end) + const octet_iterator& rangestart, + const octet_iterator& rangeend) : + it(octet_it), range_start(rangestart), range_end(rangeend) { if (it < range_start || it > range_end) throw std::out_of_range("Invalid utf-8 iterator position"); } // the default "big three" are OK octet_iterator base () const { return it; } - uint32_t operator * () const + utfchar32_t operator * () const { octet_iterator temp = it; return utf8::next(temp, range_end); @@ -322,6 +347,13 @@ namespace utf8 } // namespace utf8 -#endif //header guard +#if UTF_CPP_CPLUSPLUS >= 202002L // C++ 20 or later +#include "cpp20.h" +#elif UTF_CPP_CPLUSPLUS >= 201703L // C++ 17 or later +#include "cpp17.h" +#elif UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later +#include "cpp11.h" +#endif // C++ 11 or later +#endif //header guard diff --git a/lib/charset/utf8/core.h b/lib/charset/utf8/core.h index 693d388..4494c53 100644 --- a/lib/charset/utf8/core.h +++ b/lib/charset/utf8/core.h @@ -29,15 +29,42 @@ DEALINGS IN THE SOFTWARE. #define UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 #include +#include +#include + +// Determine the C++ standard version. +// If the user defines UTF_CPP_CPLUSPLUS, use that. +// Otherwise, trust the unreliable predefined macro __cplusplus + +#if !defined UTF_CPP_CPLUSPLUS + #define UTF_CPP_CPLUSPLUS __cplusplus +#endif + +#if UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later + #define UTF_CPP_OVERRIDE override + #define UTF_CPP_NOEXCEPT noexcept +#else // C++ 98/03 + #define UTF_CPP_OVERRIDE + #define UTF_CPP_NOEXCEPT throw() +#endif // C++ 11 or later + namespace utf8 { - // The typedefs for 8-bit, 16-bit and 32-bit unsigned integers - // You may need to change them to match your system. - // These typedefs have the same names as ones from cstdint, or boost/cstdint - typedef unsigned char uint8_t; - typedef unsigned short uint16_t; - typedef unsigned int uint32_t; +// The typedefs for 8-bit, 16-bit and 32-bit code units +#if UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later + #if UTF_CPP_CPLUSPLUS >= 202002L // C++ 20 or later + typedef char8_t utfchar8_t; + #else // C++ 11/14/17 + typedef unsigned char utfchar8_t; + #endif + typedef char16_t utfchar16_t; + typedef char32_t utfchar32_t; +#else // C++ 98/03 + typedef unsigned char utfchar8_t; + typedef unsigned short utfchar16_t; + typedef unsigned int utfchar32_t; +#endif // C++ 11 or later // Helper code - not intended to be directly called by the library users. May be changed at any time namespace internal @@ -45,61 +72,62 @@ namespace internal // Unicode constants // Leading (high) surrogates: 0xd800 - 0xdbff // Trailing (low) surrogates: 0xdc00 - 0xdfff - const uint16_t LEAD_SURROGATE_MIN = 0xd800u; - const uint16_t LEAD_SURROGATE_MAX = 0xdbffu; - const uint16_t TRAIL_SURROGATE_MIN = 0xdc00u; - const uint16_t TRAIL_SURROGATE_MAX = 0xdfffu; - const uint16_t LEAD_OFFSET = LEAD_SURROGATE_MIN - (0x10000 >> 10); - const uint32_t SURROGATE_OFFSET = 0x10000u - (LEAD_SURROGATE_MIN << 10) - TRAIL_SURROGATE_MIN; + const utfchar16_t LEAD_SURROGATE_MIN = 0xd800u; + const utfchar16_t LEAD_SURROGATE_MAX = 0xdbffu; + const utfchar16_t TRAIL_SURROGATE_MIN = 0xdc00u; + const utfchar16_t TRAIL_SURROGATE_MAX = 0xdfffu; + const utfchar16_t LEAD_OFFSET = 0xd7c0u; // LEAD_SURROGATE_MIN - (0x10000 >> 10) + const utfchar32_t SURROGATE_OFFSET = 0xfca02400u; // 0x10000u - (LEAD_SURROGATE_MIN << 10) - TRAIL_SURROGATE_MIN // Maximum valid value for a Unicode code point - const uint32_t CODE_POINT_MAX = 0x0010ffffu; + const utfchar32_t CODE_POINT_MAX = 0x0010ffffu; template - inline uint8_t mask8(octet_type oc) + inline utfchar8_t mask8(octet_type oc) { - return static_cast(0xff & oc); + return static_cast(0xff & oc); } template - inline uint16_t mask16(u16_type oc) + inline utfchar16_t mask16(u16_type oc) { - return static_cast(0xffff & oc); + return static_cast(0xffff & oc); } + template inline bool is_trail(octet_type oc) { return ((utf8::internal::mask8(oc) >> 6) == 0x2); } - template - inline bool is_lead_surrogate(u16 cp) + inline bool is_lead_surrogate(utfchar32_t cp) { return (cp >= LEAD_SURROGATE_MIN && cp <= LEAD_SURROGATE_MAX); } - template - inline bool is_trail_surrogate(u16 cp) + inline bool is_trail_surrogate(utfchar32_t cp) { return (cp >= TRAIL_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); } - template - inline bool is_surrogate(u16 cp) + inline bool is_surrogate(utfchar32_t cp) { return (cp >= LEAD_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); } - template - inline bool is_code_point_valid(u32 cp) + inline bool is_code_point_valid(utfchar32_t cp) { return (cp <= CODE_POINT_MAX && !utf8::internal::is_surrogate(cp)); } + inline bool is_in_bmp(utfchar32_t cp) + { + return cp < utfchar32_t(0x10000); + } + template - inline typename std::iterator_traits::difference_type - sequence_length(octet_iterator lead_it) + int sequence_length(octet_iterator lead_it) { - uint8_t lead = utf8::internal::mask8(*lead_it); + const utfchar8_t lead = utf8::internal::mask8(*lead_it); if (lead < 0x80) return 1; else if ((lead >> 5) == 0x6) @@ -112,8 +140,7 @@ namespace internal return 0; } - template - inline bool is_overlong_sequence(uint32_t cp, octet_difference_type length) + inline bool is_overlong_sequence(utfchar32_t cp, int length) { if (cp < 0x80) { if (length != 1) @@ -127,7 +154,6 @@ namespace internal if (length != 3) return true; } - return false; } @@ -135,22 +161,22 @@ namespace internal /// Helper for get_sequence_x template - utf_error increase_safely(octet_iterator& it, octet_iterator end) + utf_error increase_safely(octet_iterator& it, const octet_iterator end) { if (++it == end) return NOT_ENOUGH_ROOM; if (!utf8::internal::is_trail(*it)) return INCOMPLETE_SEQUENCE; - + return UTF8_OK; } - #define UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(IT, END) {utf_error ret = increase_safely(IT, END); if (ret != UTF8_OK) return ret;} + #define UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(IT, END) {utf_error ret = increase_safely(IT, END); if (ret != UTF8_OK) return ret;} /// get_sequence_x functions decode utf-8 sequences of the length x template - utf_error get_sequence_1(octet_iterator& it, octet_iterator end, uint32_t& code_point) + utf_error get_sequence_1(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) { if (it == end) return NOT_ENOUGH_ROOM; @@ -161,11 +187,11 @@ namespace internal } template - utf_error get_sequence_2(octet_iterator& it, octet_iterator end, uint32_t& code_point) + utf_error get_sequence_2(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) { if (it == end) return NOT_ENOUGH_ROOM; - + code_point = utf8::internal::mask8(*it); UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) @@ -176,7 +202,7 @@ namespace internal } template - utf_error get_sequence_3(octet_iterator& it, octet_iterator end, uint32_t& code_point) + utf_error get_sequence_3(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) { if (it == end) return NOT_ENOUGH_ROOM; @@ -195,7 +221,7 @@ namespace internal } template - utf_error get_sequence_4(octet_iterator& it, octet_iterator end, uint32_t& code_point) + utf_error get_sequence_4(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) { if (it == end) return NOT_ENOUGH_ROOM; @@ -220,21 +246,23 @@ namespace internal #undef UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR template - utf_error validate_next(octet_iterator& it, octet_iterator end, uint32_t& code_point) + utf_error validate_next(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) { + if (it == end) + return NOT_ENOUGH_ROOM; + // Save the original value of it so we can go back in case of failure // Of course, it does not make much sense with i.e. stream iterators octet_iterator original_it = it; - uint32_t cp = 0; + utfchar32_t cp = 0; // Determine the sequence length based on the lead octet - typedef typename std::iterator_traits::difference_type octet_difference_type; - const octet_difference_type length = utf8::internal::sequence_length(it); + const int length = utf8::internal::sequence_length(it); // Get trail octets and calculate the code point utf_error err = UTF8_OK; switch (length) { - case 0: + case 0: return INVALID_LEAD; case 1: err = utf8::internal::get_sequence_1(it, end, cp); @@ -273,16 +301,133 @@ namespace internal template inline utf_error validate_next(octet_iterator& it, octet_iterator end) { - uint32_t ignored; + utfchar32_t ignored; return utf8::internal::validate_next(it, end, ignored); } + template + utf_error validate_next16(word_iterator& it, word_iterator end, utfchar32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + // Save the original value of it so we can go back in case of failure + // Of course, it does not make much sense with i.e. stream iterators + word_iterator original_it = it; + + utf_error err = UTF8_OK; + + const utfchar16_t first_word = *it++; + if (!is_surrogate(first_word)) { + code_point = first_word; + return UTF8_OK; + } + else { + if (it == end) + err = NOT_ENOUGH_ROOM; + else if (is_lead_surrogate(first_word)) { + const utfchar16_t second_word = *it++; + if (is_trail_surrogate(second_word)) { + code_point = (first_word << 10) + second_word + SURROGATE_OFFSET; + return UTF8_OK; + } else + err = INCOMPLETE_SEQUENCE; + + } else { + err = INVALID_LEAD; + } + } + // error branch + it = original_it; + return err; + } + + // Internal implementation of both checked and unchecked append() function + // This function will be invoked by the overloads below, as they will know + // the octet_type. + template + octet_iterator append(utfchar32_t cp, octet_iterator result) { + if (cp < 0x80) // one octet + *(result++) = static_cast(cp); + else if (cp < 0x800) { // two octets + *(result++) = static_cast((cp >> 6) | 0xc0); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else if (cp < 0x10000) { // three octets + *(result++) = static_cast((cp >> 12) | 0xe0); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else { // four octets + *(result++) = static_cast((cp >> 18) | 0xf0); + *(result++) = static_cast(((cp >> 12) & 0x3f)| 0x80); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + return result; + } + + // One of the following overloads will be invoked from the API calls + + // A simple (but dangerous) case: the caller appends byte(s) to a char array + inline char* append(utfchar32_t cp, char* result) { + return append(cp, result); + } + + // Hopefully, most common case: the caller uses back_inserter + // i.e. append(cp, std::back_inserter(str)); + template + std::back_insert_iterator append + (utfchar32_t cp, std::back_insert_iterator result) { + return append, + typename container_type::value_type>(cp, result); + } + + // The caller uses some other kind of output operator - not covered above + // Note that in this case we are not able to determine octet_type + // so we assume it's utfchar8_t; that can cause a conversion warning if we are wrong. + template + octet_iterator append(utfchar32_t cp, octet_iterator result) { + return append(cp, result); + } + + // Internal implementation of both checked and unchecked append16() function + // This function will be invoked by the overloads below, as they will know + // the word_type. + template + word_iterator append16(utfchar32_t cp, word_iterator result) { + if (is_in_bmp(cp)) + *(result++) = static_cast(cp); + else { + // Code points from the supplementary planes are encoded via surrogate pairs + *(result++) = static_cast(LEAD_OFFSET + (cp >> 10)); + *(result++) = static_cast(TRAIL_SURROGATE_MIN + (cp & 0x3FF)); + } + return result; + } + + // Hopefully, most common case: the caller uses back_inserter + // i.e. append16(cp, std::back_inserter(str)); + template + std::back_insert_iterator append16 + (utfchar32_t cp, std::back_insert_iterator result) { + return append16, + typename container_type::value_type>(cp, result); + } + + // The caller uses some other kind of output operator - not covered above + // Note that in this case we are not able to determine word_type + // so we assume it's utfchar16_t; that can cause a conversion warning if we are wrong. + template + word_iterator append16(utfchar32_t cp, word_iterator result) { + return append16(cp, result); + } + } // namespace internal /// The library API - functions intended to be called by the users // Byte order mark - const uint8_t bom[] = {0xef, 0xbb, 0xbf}; + const utfchar8_t bom[] = {0xef, 0xbb, 0xbf}; template octet_iterator find_invalid(octet_iterator start, octet_iterator end) @@ -296,12 +441,36 @@ namespace internal return result; } + inline const char* find_invalid(const char* str) + { + const char* end = str + std::strlen(str); + return find_invalid(str, end); + } + + inline std::size_t find_invalid(const std::string& s) + { + std::string::const_iterator invalid = find_invalid(s.begin(), s.end()); + return (invalid == s.end()) ? std::string::npos : static_cast(invalid - s.begin()); + } + template inline bool is_valid(octet_iterator start, octet_iterator end) { return (utf8::find_invalid(start, end) == end); } + inline bool is_valid(const char* str) + { + return (*(utf8::find_invalid(str)) == '\0'); + } + + inline bool is_valid(const std::string& s) + { + return is_valid(s.begin(), s.end()); + } + + + template inline bool starts_with_bom (octet_iterator it, octet_iterator end) { @@ -311,17 +480,11 @@ namespace internal ((it != end) && (utf8::internal::mask8(*it)) == bom[2]) ); } - - //Deprecated in release 2.3 - template - inline bool is_bom (octet_iterator it) + + inline bool starts_with_bom(const std::string& s) { - return ( - (utf8::internal::mask8(*it++)) == bom[0] && - (utf8::internal::mask8(*it++)) == bom[1] && - (utf8::internal::mask8(*it)) == bom[2] - ); - } + return starts_with_bom(s.begin(), s.end()); + } } // namespace utf8 #endif // header guard diff --git a/lib/charset/utf8/cpp11.h b/lib/charset/utf8/cpp11.h new file mode 100644 index 0000000..691633c --- /dev/null +++ b/lib/charset/utf8/cpp11.h @@ -0,0 +1,70 @@ +// Copyright 2018 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_a184c22c_d012_11e8_a8d5_f2801f1b9fd1 +#define UTF8_FOR_CPP_a184c22c_d012_11e8_a8d5_f2801f1b9fd1 + +#include "checked.h" + +namespace utf8 +{ + inline void append16(utfchar32_t cp, std::u16string& s) + { + append16(cp, std::back_inserter(s)); + } + + inline std::string utf16to8(const std::u16string& s) + { + std::string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(const std::string& s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::string utf32to8(const std::u32string& s) + { + std::string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(const std::string& s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } +} // namespace utf8 + +#endif // header guard + diff --git a/lib/charset/utf8/cpp17.h b/lib/charset/utf8/cpp17.h new file mode 100644 index 0000000..6e2fcc2 --- /dev/null +++ b/lib/charset/utf8/cpp17.h @@ -0,0 +1,96 @@ +// Copyright 2018 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_7e906c01_03a3_4daf_b420_ea7ea952b3c9 +#define UTF8_FOR_CPP_7e906c01_03a3_4daf_b420_ea7ea952b3c9 + +#include "cpp11.h" + +namespace utf8 +{ + inline std::string utf16to8(std::u16string_view s) + { + std::string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(std::string_view s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::string utf32to8(std::u32string_view s) + { + std::string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(std::string_view s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::size_t find_invalid(std::string_view s) + { + std::string_view::const_iterator invalid = find_invalid(s.begin(), s.end()); + return (invalid == s.end()) ? std::string_view::npos : static_cast(invalid - s.begin()); + } + + inline bool is_valid(std::string_view s) + { + return is_valid(s.begin(), s.end()); + } + + inline std::string replace_invalid(std::string_view s, char32_t replacement) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::string replace_invalid(std::string_view s) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline bool starts_with_bom(std::string_view s) + { + return starts_with_bom(s.begin(), s.end()); + } + +} // namespace utf8 + +#endif // header guard + diff --git a/lib/charset/utf8/cpp20.h b/lib/charset/utf8/cpp20.h new file mode 100644 index 0000000..07b61d0 --- /dev/null +++ b/lib/charset/utf8/cpp20.h @@ -0,0 +1,124 @@ +// Copyright 2022 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_207e906c01_03a3_4daf_b420_ea7ea952b3c9 +#define UTF8_FOR_CPP_207e906c01_03a3_4daf_b420_ea7ea952b3c9 + +#include "cpp17.h" + +namespace utf8 +{ + inline std::u8string utf16tou8(const std::u16string& s) + { + std::u8string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u8string utf16tou8(std::u16string_view s) + { + std::u8string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(const std::u8string& s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(const std::u8string_view& s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u8string utf32tou8(const std::u32string& s) + { + std::u8string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u8string utf32tou8(const std::u32string_view& s) + { + std::u8string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(const std::u8string& s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(const std::u8string_view& s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::size_t find_invalid(const std::u8string& s) + { + std::u8string::const_iterator invalid = find_invalid(s.begin(), s.end()); + return (invalid == s.end()) ? std::string_view::npos : static_cast(invalid - s.begin()); + } + + inline bool is_valid(const std::u8string& s) + { + return is_valid(s.begin(), s.end()); + } + + inline std::u8string replace_invalid(const std::u8string& s, char32_t replacement) + { + std::u8string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::u8string replace_invalid(const std::u8string& s) + { + std::u8string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline bool starts_with_bom(const std::u8string& s) + { + return starts_with_bom(s.begin(), s.end()); + } + +} // namespace utf8 + +#endif // header guard + diff --git a/lib/charset/utf8/unchecked.h b/lib/charset/utf8/unchecked.h index cb24271..65d4948 100644 --- a/lib/charset/utf8/unchecked.h +++ b/lib/charset/utf8/unchecked.h @@ -32,37 +32,79 @@ DEALINGS IN THE SOFTWARE. namespace utf8 { - namespace unchecked + namespace unchecked { template - octet_iterator append(uint32_t cp, octet_iterator result) + octet_iterator append(utfchar32_t cp, octet_iterator result) { - if (cp < 0x80) // one octet - *(result++) = static_cast(cp); - else if (cp < 0x800) { // two octets - *(result++) = static_cast((cp >> 6) | 0xc0); - *(result++) = static_cast((cp & 0x3f) | 0x80); - } - else if (cp < 0x10000) { // three octets - *(result++) = static_cast((cp >> 12) | 0xe0); - *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); - *(result++) = static_cast((cp & 0x3f) | 0x80); - } - else { // four octets - *(result++) = static_cast((cp >> 18) | 0xf0); - *(result++) = static_cast(((cp >> 12) & 0x3f)| 0x80); - *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); - *(result++) = static_cast((cp & 0x3f) | 0x80); + return internal::append(cp, result); + } + + template + word_iterator append16(utfchar32_t cp, word_iterator result) + { + return internal::append16(cp, result); + } + + template + output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, utfchar32_t replacement) + { + while (start != end) { + octet_iterator sequence_start = start; + internal::utf_error err_code = utf8::internal::validate_next(start, end); + switch (err_code) { + case internal::UTF8_OK : + for (octet_iterator it = sequence_start; it != start; ++it) + *out++ = *it; + break; + case internal::NOT_ENOUGH_ROOM: + out = utf8::unchecked::append(replacement, out); + start = end; + break; + case internal::INVALID_LEAD: + out = utf8::unchecked::append(replacement, out); + ++start; + break; + case internal::INCOMPLETE_SEQUENCE: + case internal::OVERLONG_SEQUENCE: + case internal::INVALID_CODE_POINT: + out = utf8::unchecked::append(replacement, out); + ++start; + // just one replacement mark for the sequence + while (start != end && utf8::internal::is_trail(*start)) + ++start; + break; + } } + return out; + } + + template + inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out) + { + static const utfchar32_t replacement_marker = utf8::internal::mask16(0xfffd); + return utf8::unchecked::replace_invalid(start, end, out, replacement_marker); + } + + inline std::string replace_invalid(const std::string& s, utfchar32_t replacement) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::string replace_invalid(const std::string& s) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); return result; } template - uint32_t next(octet_iterator& it) + utfchar32_t next(octet_iterator& it) { - uint32_t cp = utf8::internal::mask8(*it); - typename std::iterator_traits::difference_type length = utf8::internal::sequence_length(it); - switch (length) { + utfchar32_t cp = utf8::internal::mask8(*it); + switch (utf8::internal::sequence_length(it)) { case 1: break; case 2: @@ -85,40 +127,50 @@ namespace utf8 break; } ++it; - return cp; + return cp; } template - uint32_t peek_next(octet_iterator it) + utfchar32_t peek_next(octet_iterator it) { - return utf8::unchecked::next(it); + return utf8::unchecked::next(it); } - template - uint32_t prior(octet_iterator& it) + template + utfchar32_t next16(word_iterator& it) { - while (utf8::internal::is_trail(*(--it))) ; - octet_iterator temp = it; - return utf8::unchecked::next(temp); + utfchar32_t cp = utf8::internal::mask16(*it++); + if (utf8::internal::is_lead_surrogate(cp)) + return (cp << 10) + *it++ + utf8::internal::SURROGATE_OFFSET; + return cp; } - // Deprecated in versions that include prior, but only for the sake of consistency (see utf8::previous) template - inline uint32_t previous(octet_iterator& it) + utfchar32_t prior(octet_iterator& it) { - return utf8::unchecked::prior(it); + while (utf8::internal::is_trail(*(--it))) ; + octet_iterator temp = it; + return utf8::unchecked::next(temp); } template - void advance (octet_iterator& it, distance_type n) + void advance(octet_iterator& it, distance_type n) { - for (distance_type i = 0; i < n; ++i) - utf8::unchecked::next(it); + const distance_type zero(0); + if (n < zero) { + // backward + for (distance_type i = n; i < zero; ++i) + utf8::unchecked::prior(it); + } else { + // forward + for (distance_type i = zero; i < n; ++i) + utf8::unchecked::next(it); + } } template typename std::iterator_traits::difference_type - distance (octet_iterator first, octet_iterator last) + distance(octet_iterator first, octet_iterator last) { typename std::iterator_traits::difference_type dist; for (dist = 0; first < last; ++dist) @@ -127,37 +179,39 @@ namespace utf8 } template - octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) - { + octet_iterator utf16to8(u16bit_iterator start, u16bit_iterator end, octet_iterator result) + { while (start != end) { - uint32_t cp = utf8::internal::mask16(*start++); - // Take care of surrogate pairs first + utfchar32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first if (utf8::internal::is_lead_surrogate(cp)) { - uint32_t trail_surrogate = utf8::internal::mask16(*start++); + if (start == end) + return result; + utfchar32_t trail_surrogate = utf8::internal::mask16(*start++); cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; } result = utf8::unchecked::append(cp, result); } - return result; + return result; } template - u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) + u16bit_iterator utf8to16(octet_iterator start, octet_iterator end, u16bit_iterator result) { while (start < end) { - uint32_t cp = utf8::unchecked::next(start); + utfchar32_t cp = utf8::unchecked::next(start); if (cp > 0xffff) { //make a surrogate pair - *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); - *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); } else - *result++ = static_cast(cp); + *result++ = static_cast(cp); } return result; } template - octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) + octet_iterator utf32to8(u32bit_iterator start, u32bit_iterator end, octet_iterator result) { while (start != end) result = utf8::unchecked::append(*(start++), result); @@ -166,7 +220,7 @@ namespace utf8 } template - u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) + u32bit_iterator utf8to32(octet_iterator start, octet_iterator end, u32bit_iterator result) { while (start < end) (*result++) = utf8::unchecked::next(start); @@ -176,14 +230,19 @@ namespace utf8 // The iterator class template - class iterator : public std::iterator { + class iterator { octet_iterator it; public: + typedef utfchar32_t value_type; + typedef utfchar32_t* pointer; + typedef utfchar32_t& reference; + typedef std::ptrdiff_t difference_type; + typedef std::bidirectional_iterator_tag iterator_category; iterator () {} explicit iterator (const octet_iterator& octet_it): it(octet_it) {} // the default "big three" are OK octet_iterator base () const { return it; } - uint32_t operator * () const + utfchar32_t operator * () const { octet_iterator temp = it; return utf8::unchecked::next(temp); -- cgit v1.2.3