| 1 |
// Copyright 2006-2016 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_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 |
| 29 |
#define UTF8_FOR_CPP_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 |
| 30 |
|
| 31 |
#include "core.h" |
| 32 |
#include <stdexcept> |
| 33 |
|
| 34 |
namespace utf8 |
| 35 |
{ |
| 36 |
// Base for the exceptions that may be thrown from the library |
| 37 |
class exception : public ::std::exception { |
| 38 |
}; |
| 39 |
|
| 40 |
// Exceptions that may be thrown from the library functions. |
| 41 |
class invalid_code_point : public exception { |
| 42 |
uint32_t cp; |
| 43 |
public: |
| 44 |
invalid_code_point(uint32_t codepoint) : cp(codepoint) {} |
| 45 |
virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid code point"; } |
| 46 |
uint32_t code_point() const {return cp;} |
| 47 |
}; |
| 48 |
|
| 49 |
class invalid_utf8 : public exception { |
| 50 |
uint8_t u8; |
| 51 |
public: |
| 52 |
invalid_utf8 (uint8_t u) : u8(u) {} |
| 53 |
invalid_utf8 (char c) : u8(static_cast<uint8_t>(c)) {} |
| 54 |
virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid UTF-8"; } |
| 55 |
uint8_t utf8_octet() const {return u8;} |
| 56 |
}; |
| 57 |
|
| 58 |
class invalid_utf16 : public exception { |
| 59 |
uint16_t u16; |
| 60 |
public: |
| 61 |
invalid_utf16 (uint16_t u) : u16(u) {} |
| 62 |
virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid UTF-16"; } |
| 63 |
uint16_t utf16_word() const {return u16;} |
| 64 |
}; |
| 65 |
|
| 66 |
class not_enough_room : public exception { |
| 67 |
public: |
| 68 |
virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Not enough space"; } |
| 69 |
}; |
| 70 |
|
| 71 |
/// The library API - functions intended to be called by the users |
| 72 |
|
| 73 |
template <typename octet_iterator> |
| 74 |
octet_iterator append(uint32_t cp, octet_iterator result) |
| 75 |
{ |
| 76 |
if (!utf8::internal::is_code_point_valid(cp)) |
| 77 |
throw invalid_code_point(cp); |
| 78 |
|
| 79 |
return internal::append(cp, result); |
| 80 |
} |
| 81 |
|
| 82 |
template <typename octet_iterator, typename output_iterator> |
| 83 |
output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_t replacement) |
| 84 |
{ |
| 85 |
while (start != end) { |
| 86 |
octet_iterator sequence_start = start; |
| 87 |
internal::utf_error err_code = utf8::internal::validate_next(start, end); |
| 88 |
switch (err_code) { |
| 89 |
case internal::UTF8_OK : |
| 90 |
for (octet_iterator it = sequence_start; it != start; ++it) |
| 91 |
*out++ = *it; |
| 92 |
break; |
| 93 |
case internal::NOT_ENOUGH_ROOM: |
| 94 |
out = utf8::append (replacement, out); |
| 95 |
start = end; |
| 96 |
break; |
| 97 |
case internal::INVALID_LEAD: |
| 98 |
out = utf8::append (replacement, out); |
| 99 |
++start; |
| 100 |
break; |
| 101 |
case internal::INCOMPLETE_SEQUENCE: |
| 102 |
case internal::OVERLONG_SEQUENCE: |
| 103 |
case internal::INVALID_CODE_POINT: |
| 104 |
out = utf8::append (replacement, out); |
| 105 |
++start; |
| 106 |
// just one replacement mark for the sequence |
| 107 |
while (start != end && utf8::internal::is_trail(*start)) |
| 108 |
++start; |
| 109 |
break; |
| 110 |
} |
| 111 |
} |
| 112 |
return out; |
| 113 |
} |
| 114 |
|
| 115 |
template <typename octet_iterator, typename output_iterator> |
| 116 |
inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out) |
| 117 |
{ |
| 118 |
static const uint32_t replacement_marker = utf8::internal::mask16(0xfffd); |
| 119 |
return utf8::replace_invalid(start, end, out, replacement_marker); |
| 120 |
} |
| 121 |
|
| 122 |
template <typename octet_iterator> |
| 123 |
uint32_t next(octet_iterator& it, octet_iterator end) |
| 124 |
{ |
| 125 |
uint32_t cp = 0; |
| 126 |
internal::utf_error err_code = utf8::internal::validate_next(it, end, cp); |
| 127 |
switch (err_code) { |
| 128 |
case internal::UTF8_OK : |
| 129 |
break; |
| 130 |
case internal::NOT_ENOUGH_ROOM : |
| 131 |
throw not_enough_room(); |
| 132 |
case internal::INVALID_LEAD : |
| 133 |
case internal::INCOMPLETE_SEQUENCE : |
| 134 |
case internal::OVERLONG_SEQUENCE : |
| 135 |
throw invalid_utf8(static_cast<uint8_t>(*it)); |
| 136 |
case internal::INVALID_CODE_POINT : |
| 137 |
throw invalid_code_point(cp); |
| 138 |
} |
| 139 |
return cp; |
| 140 |
} |
| 141 |
|
| 142 |
template <typename octet_iterator> |
| 143 |
uint32_t peek_next(octet_iterator it, octet_iterator end) |
| 144 |
{ |
| 145 |
return utf8::next(it, end); |
| 146 |
} |
| 147 |
|
| 148 |
template <typename octet_iterator> |
| 149 |
uint32_t prior(octet_iterator& it, octet_iterator start) |
| 150 |
{ |
| 151 |
// can't do much if it == start |
| 152 |
if (it == start) |
| 153 |
throw not_enough_room(); |
| 154 |
|
| 155 |
octet_iterator end = it; |
| 156 |
// Go back until we hit either a lead octet or start |
| 157 |
while (utf8::internal::is_trail(*(--it))) |
| 158 |
if (it == start) |
| 159 |
throw invalid_utf8(*it); // error - no lead byte in the sequence |
| 160 |
return utf8::peek_next(it, end); |
| 161 |
} |
| 162 |
|
| 163 |
template <typename octet_iterator, typename distance_type> |
| 164 |
void advance (octet_iterator& it, distance_type n, octet_iterator end) |
| 165 |
{ |
| 166 |
const distance_type zero(0); |
| 167 |
if (n < zero) { |
| 168 |
// backward |
| 169 |
for (distance_type i = n; i < zero; ++i) |
| 170 |
utf8::prior(it, end); |
| 171 |
} else { |
| 172 |
// forward |
| 173 |
for (distance_type i = zero; i < n; ++i) |
| 174 |
utf8::next(it, end); |
| 175 |
} |
| 176 |
} |
| 177 |
|
| 178 |
template <typename octet_iterator> |
| 179 |
typename std::iterator_traits<octet_iterator>::difference_type |
| 180 |
distance (octet_iterator first, octet_iterator last) |
| 181 |
{ |
| 182 |
typename std::iterator_traits<octet_iterator>::difference_type dist; |
| 183 |
for (dist = 0; first < last; ++dist) |
| 184 |
utf8::next(first, last); |
| 185 |
return dist; |
| 186 |
} |
| 187 |
|
| 188 |
template <typename u16bit_iterator, typename octet_iterator> |
| 189 |
octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) |
| 190 |
{ |
| 191 |
while (start != end) { |
| 192 |
uint32_t cp = utf8::internal::mask16(*start++); |
| 193 |
// Take care of surrogate pairs first |
| 194 |
if (utf8::internal::is_lead_surrogate(cp)) { |
| 195 |
if (start != end) { |
| 196 |
uint32_t trail_surrogate = utf8::internal::mask16(*start++); |
| 197 |
if (utf8::internal::is_trail_surrogate(trail_surrogate)) |
| 198 |
cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; |
| 199 |
else |
| 200 |
throw invalid_utf16(static_cast<uint16_t>(trail_surrogate)); |
| 201 |
} |
| 202 |
else |
| 203 |
throw invalid_utf16(static_cast<uint16_t>(cp)); |
| 204 |
|
| 205 |
} |
| 206 |
// Lone trail surrogate |
| 207 |
else if (utf8::internal::is_trail_surrogate(cp)) |
| 208 |
throw invalid_utf16(static_cast<uint16_t>(cp)); |
| 209 |
|
| 210 |
result = utf8::append(cp, result); |
| 211 |
} |
| 212 |
return result; |
| 213 |
} |
| 214 |
|
| 215 |
template <typename u16bit_iterator, typename octet_iterator> |
| 216 |
u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) |
| 217 |
{ |
| 218 |
while (start < end) { |
| 219 |
uint32_t cp = utf8::next(start, end); |
| 220 |
if (cp > 0xffff) { //make a surrogate pair |
| 221 |
*result++ = static_cast<uint16_t>((cp >> 10) + internal::LEAD_OFFSET); |
| 222 |
*result++ = static_cast<uint16_t>((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); |
| 223 |
} |
| 224 |
else |
| 225 |
*result++ = static_cast<uint16_t>(cp); |
| 226 |
} |
| 227 |
return result; |
| 228 |
} |
| 229 |
|
| 230 |
template <typename octet_iterator, typename u32bit_iterator> |
| 231 |
octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) |
| 232 |
{ |
| 233 |
while (start != end) |
| 234 |
result = utf8::append(*(start++), result); |
| 235 |
|
| 236 |
return result; |
| 237 |
} |
| 238 |
|
| 239 |
template <typename octet_iterator, typename u32bit_iterator> |
| 240 |
u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) |
| 241 |
{ |
| 242 |
while (start < end) |
| 243 |
(*result++) = utf8::next(start, end); |
| 244 |
|
| 245 |
return result; |
| 246 |
} |
| 247 |
|
| 248 |
// The iterator class |
| 249 |
template <typename octet_iterator> |
| 250 |
class iterator { |
| 251 |
octet_iterator it; |
| 252 |
octet_iterator range_start; |
| 253 |
octet_iterator range_end; |
| 254 |
public: |
| 255 |
typedef uint32_t value_type; |
| 256 |
typedef uint32_t* pointer; |
| 257 |
typedef uint32_t& reference; |
| 258 |
typedef std::ptrdiff_t difference_type; |
| 259 |
typedef std::bidirectional_iterator_tag iterator_category; |
| 260 |
iterator () {} |
| 261 |
explicit iterator (const octet_iterator& octet_it, |
| 262 |
const octet_iterator& rangestart, |
| 263 |
const octet_iterator& rangeend) : |
| 264 |
it(octet_it), range_start(rangestart), range_end(rangeend) |
| 265 |
{ |
| 266 |
if (it < range_start || it > range_end) |
| 267 |
throw std::out_of_range("Invalid utf-8 iterator position"); |
| 268 |
} |
| 269 |
// the default "big three" are OK |
| 270 |
octet_iterator base () const { return it; } |
| 271 |
uint32_t operator * () const |
| 272 |
{ |
| 273 |
octet_iterator temp = it; |
| 274 |
return utf8::next(temp, range_end); |
| 275 |
} |
| 276 |
bool operator == (const iterator& rhs) const |
| 277 |
{ |
| 278 |
if (range_start != rhs.range_start || range_end != rhs.range_end) |
| 279 |
throw std::logic_error("Comparing utf-8 iterators defined with different ranges"); |
| 280 |
return (it == rhs.it); |
| 281 |
} |
| 282 |
bool operator != (const iterator& rhs) const |
| 283 |
{ |
| 284 |
return !(operator == (rhs)); |
| 285 |
} |
| 286 |
iterator& operator ++ () |
| 287 |
{ |
| 288 |
utf8::next(it, range_end); |
| 289 |
return *this; |
| 290 |
} |
| 291 |
iterator operator ++ (int) |
| 292 |
{ |
| 293 |
iterator temp = *this; |
| 294 |
utf8::next(it, range_end); |
| 295 |
return temp; |
| 296 |
} |
| 297 |
iterator& operator -- () |
| 298 |
{ |
| 299 |
utf8::prior(it, range_start); |
| 300 |
return *this; |
| 301 |
} |
| 302 |
iterator operator -- (int) |
| 303 |
{ |
| 304 |
iterator temp = *this; |
| 305 |
utf8::prior(it, range_start); |
| 306 |
return temp; |
| 307 |
} |
| 308 |
}; // class iterator |
| 309 |
|
| 310 |
} // namespace utf8 |
| 311 |
|
| 312 |
#if UTF_CPP_CPLUSPLUS >= 201703L // C++ 17 or later |
| 313 |
#include "cpp17.h" |
| 314 |
#elif UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later |
| 315 |
#include "cpp11.h" |
| 316 |
#endif // C++ 11 or later |
| 317 |
|
| 318 |
#endif //header guard |
| 319 |
|