// Copyright (c) 2006 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_2675DCD0_9480_4c0c_B92A_CC14C027B731 #define UTF8_FOR_CPP_2675DCD0_9480_4c0c_B92A_CC14C027B731 #include #include 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; // Exceptions that may be thrown from the library functions. class invalid_code_point : public std::exception { uint32_t cp; public: invalid_code_point(uint32_t cp) : cp(cp) {} const char* what() { return "Invalid code point"; } uint32_t code_point() const {return cp;} }; class invalid_utf8 : public std::exception { uint8_t u8; public: invalid_utf8 (uint8_t u) : u8(u) {} const char* what() { return "Invalid UTF-8"; } uint8_t utf8_octet() const {return u8;} }; class invalid_utf16 : public std::exception { uint16_t u16; public: invalid_utf16 (uint16_t u) : u16(u) {} const char* what() { return "Invalid UTF-16"; } uint16_t utf16_word() const {return u16;} }; class not_enough_room : public std::exception { public: const char* what() { return "Not enough space"; } }; // Helper code - not intended to be directly called by the library users. May be changed at any time 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; // Maximum valid value for a Unicode code point const uint32_t CODE_POINT_MAX = 0x0010ffffu; template inline uint8_t mask8(octet_type oc) { return static_cast(0xff & oc); } template inline uint16_t mask16(u16_type oc) { return static_cast(0xffff & oc); } template inline bool is_trail(octet_type oc) { return ((mask8(oc) >> 6) == 0x2); } template inline bool is_surrogate(u16 cp) { return (cp >= LEAD_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); } template inline bool is_code_point_valid(u32 cp) { return (cp <= CODE_POINT_MAX && !is_surrogate(cp) && cp != 0xfffe && cp != 0xffff); } enum utf_error {OK, NOT_ENOUGH_ROOM, INVALID_LEAD, INCOMPLETE_SEQUENCE, OVERLONG_SEQUENCE, INVALID_CODE_POINT}; template utf_error validate_next(octet_iterator& it, octet_iterator end, uint32_t* code_point = 0) { uint32_t cp = mask8(*it); // Check the lead octet typedef typename std::iterator_traits::difference_type octet_difference_type; octet_difference_type sequence_length; if (cp < 0x80) sequence_length = 1; else if ((cp >> 5) == 0x6) sequence_length = 2; else if ((cp >> 4) == 0xe) sequence_length = 3; else if ((cp >> 3) == 0x1e) sequence_length = 4; else return INVALID_LEAD; // Do we have enough memory? if (end - it < sequence_length) return NOT_ENOUGH_ROOM; // Check trail octets and calculate the code point switch (sequence_length) { case 2: if (is_trail(*(++it))) { cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f); } else { --it; return INCOMPLETE_SEQUENCE; } break; case 3: if (is_trail(*(++it))) { cp = ((cp << 12) & 0xffff) + ((mask8(*it) << 6) & 0xfff); if (is_trail(*(++it))) { cp += (*it) & 0x3f; } else { --it; --it; return INCOMPLETE_SEQUENCE; } } else { --it; return INCOMPLETE_SEQUENCE; } break; case 4: if (is_trail(*(++it))) { cp = ((cp << 18) & 0x1fffff) + ((mask8(*it) << 12) & 0x3ffff); if (is_trail(*(++it))) { cp += (mask8(*it) << 6) & 0xfff; if (is_trail(*(++it))) { cp += (*it) & 0x3f; } else { --it; --it; --it; return INCOMPLETE_SEQUENCE; } } else { --it; --it; return INCOMPLETE_SEQUENCE; } } else { --it; return INCOMPLETE_SEQUENCE; } break; } // Is the code point valid? if (!is_code_point_valid(cp)) { for (octet_difference_type i = 0; i < sequence_length - 1; ++i) --it; return INVALID_CODE_POINT; } if (code_point) *code_point = cp; if (cp < 0x80) { if (sequence_length != 1) { for (octet_difference_type i = 0; i < sequence_length - 1; ++i) --it; return OVERLONG_SEQUENCE; } } else if (cp < 0x800) { if (sequence_length != 2) { for (octet_difference_type i = 0; i < sequence_length - 1; ++i) --it; return OVERLONG_SEQUENCE; } } else if (cp < 0x10000) { if (sequence_length != 3) { for (octet_difference_type i = 0; i < sequence_length - 1; ++i) --it; return OVERLONG_SEQUENCE; } } ++it; return OK; } } // namespace internal /// The library API - functions intended to be called by the users // Byte order mark const uint8_t bom[] = {0xef, 0xbb, 0xbf}; template octet_iterator find_invalid(octet_iterator start, octet_iterator end) { octet_iterator result = start; while (result != end) { internal::utf_error err_code = internal::validate_next(result, end); if (err_code != internal::OK) return result; } return result; } template bool is_valid(octet_iterator start, octet_iterator end) { return (find_invalid(start, end) == end); } template bool is_bom (octet_iterator it) { return ( (internal::mask8(*it++)) == bom[0] && (internal::mask8(*it++)) == bom[1] && (internal::mask8(*it)) == bom[2] ); } template octet_iterator append(uint32_t cp, octet_iterator result) { if (!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 if (cp <= internal::CODE_POINT_MAX) { // 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); } else throw invalid_code_point(cp); return result; } template uint32_t next(octet_iterator& it, octet_iterator end) { uint32_t cp = 0; internal::utf_error err_code = internal::validate_next(it, end, &cp); switch (err_code) { case internal::OK : break; case internal::NOT_ENOUGH_ROOM : throw not_enough_room(); case internal::INVALID_LEAD : case internal::INCOMPLETE_SEQUENCE : case internal::OVERLONG_SEQUENCE : throw invalid_utf8(*it); case internal::INVALID_CODE_POINT : throw invalid_code_point(cp); } return cp; } template uint32_t previous(octet_iterator& it, octet_iterator pass_start) { octet_iterator end = it; while (internal::is_trail(*(--it))) if (it == pass_start) throw invalid_utf8(*it); // error - no lead byte in the sequence octet_iterator temp = it; return next(temp, end); } template void advance (octet_iterator& it, distance_type n, octet_iterator end) { for (distance_type i = 0; i < n; ++i) next(it, end); } template typename std::iterator_traits::difference_type distance (octet_iterator first, octet_iterator last) { typename std::iterator_traits::difference_type dist; for (dist = 0; first < last; ++dist) next(first, last); return dist; } template octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) { while (start != end) { uint32_t cp = internal::mask16(*start++); // Take care of surrogate pairs first if (cp >= internal::LEAD_SURROGATE_MIN && cp <= internal::LEAD_SURROGATE_MAX) { if (start != end) { uint32_t trail_surrogate = internal::mask16(*start++); if (trail_surrogate >= internal::TRAIL_SURROGATE_MIN && trail_surrogate <= internal::TRAIL_SURROGATE_MAX) cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; else throw invalid_utf16(static_cast(trail_surrogate)); } else throw invalid_utf16(static_cast(*start)); } result = append(cp, result); } return result; } template u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) { while (start != end) { uint32_t cp = 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); } else *result++ = static_cast(cp); } return result; } template octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) { while (start != end) result = append(*(start++), result); return result; } template u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) { while (start < end) (*result++) = next(start, end); return result; } namespace unchecked { template octet_iterator append(uint32_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; } template uint32_t next(octet_iterator& it) { uint32_t cp = internal::mask8(*it); if (cp < 0x80) ; else if ((internal::mask8(*it) >> 5) == 0x6) { it++; cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f); } else if ((internal::mask8(*it) >> 4) == 0xe) { ++it; cp = ((cp << 12) & 0xffff) + ((internal::mask8(*it) << 6) & 0xfff); ++it; cp += (*it) & 0x3f; } else if ((internal::mask8(*it) >> 3) == 0x1e) { ++it; cp = ((cp << 18) & 0x1fffff) + ((internal::mask8(*it) << 12) & 0x3ffff); ++it; cp += (internal::mask8(*it) << 6) & 0xfff; ++it; cp += (*it) & 0x3f; } ++it; return cp; } template uint32_t previous(octet_iterator& it) { while (internal::is_trail(*(--it))) ; octet_iterator temp = it; return next(temp); } template void advance (octet_iterator& it, distance_type n) { for (distance_type i = 0; i < n; ++i) next(it); } template uint32_t get(octet_iterator it) { return next(it); } template typename std::iterator_traits::difference_type distance (octet_iterator first, octet_iterator last) { typename std::iterator_traits::difference_type dist; for (dist = 0; first < last; ++dist) next(first); return dist; } template octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) { while (start != end) { uint32_t cp = internal::mask16(*start++); // Take care of surrogate pairs first if (cp >= internal::LEAD_SURROGATE_MIN && cp <= internal::LEAD_SURROGATE_MAX) { uint32_t trail_surrogate = internal::mask16(*start++); cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; } result = append(cp, result); } return result; } template u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) { while (start != end) { uint32_t cp = 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); } else *result++ = static_cast(cp); } return result; } template octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) { while (start != end) result = append(*(start++), result); return result; } template u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) { while (start < end) (*result++) = next(start); return result; } } // namespace utf8::unchecked } // namespace utf8 #endif // header guard