1 // Copyright 2006 Nemanja Trifunovic 2 3 /* 4 Permission is hereby granted, free of charge, to any person or organization 5 obtaining a copy of the software and accompanying documentation covered by 6 this license (the "Software") to use, reproduce, display, distribute, 7 execute, and transmit the Software, and to prepare derivative works of the 8 Software, and to permit third-parties to whom the Software is furnished to 9 do so, all subject to the following: 10 11 The copyright notices in the Software and this entire statement, including 12 the above license grant, this restriction and the following disclaimer, 13 must be included in all copies of the Software, in whole or in part, and 14 all derivative works of the Software, unless such copies or derivative 15 works are solely in the form of machine-executable object code generated by 16 a source language processor. 17 18 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 19 IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 20 FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT 21 SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE 22 FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, 23 ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER 24 DEALINGS IN THE SOFTWARE. 25 */ 26 27 28 #ifndef UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 29 #define UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 30 31 #include "core.h" 32 33 namespace utf8 34 { 35 namespace unchecked 36 { 37 template <typename octet_iterator> append(uint32_t cp,octet_iterator result)38 octet_iterator append(uint32_t cp, octet_iterator result) 39 { 40 if (cp < 0x80) // one octet 41 *(result++) = static_cast<uint8_t>(cp); 42 else if (cp < 0x800) { // two octets 43 *(result++) = static_cast<uint8_t>((cp >> 6) | 0xc0); 44 *(result++) = static_cast<uint8_t>((cp & 0x3f) | 0x80); 45 } 46 else if (cp < 0x10000) { // three octets 47 *(result++) = static_cast<uint8_t>((cp >> 12) | 0xe0); 48 *(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f) | 0x80); 49 *(result++) = static_cast<uint8_t>((cp & 0x3f) | 0x80); 50 } 51 else { // four octets 52 *(result++) = static_cast<uint8_t>((cp >> 18) | 0xf0); 53 *(result++) = static_cast<uint8_t>(((cp >> 12) & 0x3f)| 0x80); 54 *(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f) | 0x80); 55 *(result++) = static_cast<uint8_t>((cp & 0x3f) | 0x80); 56 } 57 return result; 58 } 59 60 template <typename octet_iterator> next(octet_iterator & it)61 uint32_t next(octet_iterator& it) 62 { 63 uint32_t cp = utf8::internal::mask8(*it); 64 typename std::iterator_traits<octet_iterator>::difference_type length = utf8::internal::sequence_length(it); 65 switch (length) { 66 case 1: 67 break; 68 case 2: 69 it++; 70 cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f); 71 break; 72 case 3: 73 ++it; 74 cp = ((cp << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff); 75 ++it; 76 cp += (*it) & 0x3f; 77 break; 78 case 4: 79 ++it; 80 cp = ((cp << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff); 81 ++it; 82 cp += (utf8::internal::mask8(*it) << 6) & 0xfff; 83 ++it; 84 cp += (*it) & 0x3f; 85 break; 86 } 87 ++it; 88 return cp; 89 } 90 91 template <typename octet_iterator> peek_next(octet_iterator it)92 uint32_t peek_next(octet_iterator it) 93 { 94 return utf8::unchecked::next(it); 95 } 96 97 template <typename octet_iterator> prior(octet_iterator & it)98 uint32_t prior(octet_iterator& it) 99 { 100 while (utf8::internal::is_trail(*(--it))) ; 101 octet_iterator temp = it; 102 return utf8::unchecked::next(temp); 103 } 104 105 // Deprecated in versions that include prior, but only for the sake of consistency (see utf8::previous) 106 template <typename octet_iterator> previous(octet_iterator & it)107 inline uint32_t previous(octet_iterator& it) 108 { 109 return utf8::unchecked::prior(it); 110 } 111 112 template <typename octet_iterator, typename distance_type> advance(octet_iterator & it,distance_type n)113 void advance (octet_iterator& it, distance_type n) 114 { 115 for (distance_type i = 0; i < n; ++i) 116 utf8::unchecked::next(it); 117 } 118 119 template <typename octet_iterator> 120 typename std::iterator_traits<octet_iterator>::difference_type distance(octet_iterator first,octet_iterator last)121 distance (octet_iterator first, octet_iterator last) 122 { 123 typename std::iterator_traits<octet_iterator>::difference_type dist; 124 for (dist = 0; first < last; ++dist) 125 utf8::unchecked::next(first); 126 return dist; 127 } 128 129 template <typename u16bit_iterator, typename octet_iterator> utf16to8(u16bit_iterator start,u16bit_iterator end,octet_iterator result)130 octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) 131 { 132 while (start != end) { 133 uint32_t cp = utf8::internal::mask16(*start++); 134 // Take care of surrogate pairs first 135 if (utf8::internal::is_lead_surrogate(cp)) { 136 uint32_t trail_surrogate = utf8::internal::mask16(*start++); 137 cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; 138 } 139 result = utf8::unchecked::append(cp, result); 140 } 141 return result; 142 } 143 144 template <typename u16bit_iterator, typename octet_iterator> utf8to16(octet_iterator start,octet_iterator end,u16bit_iterator result)145 u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) 146 { 147 while (start < end) { 148 uint32_t cp = utf8::unchecked::next(start); 149 if (cp > 0xffff) { //make a surrogate pair 150 *result++ = static_cast<uint16_t>((cp >> 10) + internal::LEAD_OFFSET); 151 *result++ = static_cast<uint16_t>((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); 152 } 153 else 154 *result++ = static_cast<uint16_t>(cp); 155 } 156 return result; 157 } 158 159 template <typename octet_iterator, typename u32bit_iterator> utf32to8(u32bit_iterator start,u32bit_iterator end,octet_iterator result)160 octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) 161 { 162 while (start != end) 163 result = utf8::unchecked::append(*(start++), result); 164 165 return result; 166 } 167 168 template <typename octet_iterator, typename u32bit_iterator> utf8to32(octet_iterator start,octet_iterator end,u32bit_iterator result)169 u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) 170 { 171 while (start < end) 172 (*result++) = utf8::unchecked::next(start); 173 174 return result; 175 } 176 177 // The iterator class 178 template <typename octet_iterator> 179 class iterator : public std::iterator <std::bidirectional_iterator_tag, uint32_t> { 180 octet_iterator it; 181 public: iterator()182 iterator () {} iterator(const octet_iterator & octet_it)183 explicit iterator (const octet_iterator& octet_it): it(octet_it) {} 184 // the default "big three" are OK base()185 octet_iterator base () const { return it; } 186 uint32_t operator * () const 187 { 188 octet_iterator temp = it; 189 return utf8::unchecked::next(temp); 190 } 191 bool operator == (const iterator& rhs) const 192 { 193 return (it == rhs.it); 194 } 195 bool operator != (const iterator& rhs) const 196 { 197 return !(operator == (rhs)); 198 } 199 iterator& operator ++ () 200 { 201 ::std::advance(it, utf8::internal::sequence_length(it)); 202 return *this; 203 } 204 iterator operator ++ (int) 205 { 206 iterator temp = *this; 207 ::std::advance(it, utf8::internal::sequence_length(it)); 208 return temp; 209 } 210 iterator& operator -- () 211 { 212 utf8::unchecked::prior(it); 213 return *this; 214 } 215 iterator operator -- (int) 216 { 217 iterator temp = *this; 218 utf8::unchecked::prior(it); 219 return temp; 220 } 221 }; // class iterator 222 223 } // namespace utf8::unchecked 224 } // namespace utf8 225 226 227 #endif // header guard 228 229