| 1 | /**************************************************************************** |
| 2 | ** |
| 3 | ** Copyright (C) 2020 The Qt Company Ltd. |
| 4 | ** Copyright (C) 2020 Intel Corporation. |
| 5 | ** Contact: https://www.qt.io/licensing/ |
| 6 | ** |
| 7 | ** This file is part of the QtCore module of the Qt Toolkit. |
| 8 | ** |
| 9 | ** $QT_BEGIN_LICENSE:LGPL$ |
| 10 | ** Commercial License Usage |
| 11 | ** Licensees holding valid commercial Qt licenses may use this file in |
| 12 | ** accordance with the commercial license agreement provided with the |
| 13 | ** Software or, alternatively, in accordance with the terms contained in |
| 14 | ** a written agreement between you and The Qt Company. For licensing terms |
| 15 | ** and conditions see https://www.qt.io/terms-conditions. For further |
| 16 | ** information use the contact form at https://www.qt.io/contact-us. |
| 17 | ** |
| 18 | ** GNU Lesser General Public License Usage |
| 19 | ** Alternatively, this file may be used under the terms of the GNU Lesser |
| 20 | ** General Public License version 3 as published by the Free Software |
| 21 | ** Foundation and appearing in the file LICENSE.LGPL3 included in the |
| 22 | ** packaging of this file. Please review the following information to |
| 23 | ** ensure the GNU Lesser General Public License version 3 requirements |
| 24 | ** will be met: https://www.gnu.org/licenses/lgpl-3.0.html. |
| 25 | ** |
| 26 | ** GNU General Public License Usage |
| 27 | ** Alternatively, this file may be used under the terms of the GNU |
| 28 | ** General Public License version 2.0 or (at your option) the GNU General |
| 29 | ** Public license version 3 or any later version approved by the KDE Free |
| 30 | ** Qt Foundation. The licenses are as published by the Free Software |
| 31 | ** Foundation and appearing in the file LICENSE.GPL2 and LICENSE.GPL3 |
| 32 | ** included in the packaging of this file. Please review the following |
| 33 | ** information to ensure the GNU General Public License requirements will |
| 34 | ** be met: https://www.gnu.org/licenses/gpl-2.0.html and |
| 35 | ** https://www.gnu.org/licenses/gpl-3.0.html. |
| 36 | ** |
| 37 | ** $QT_END_LICENSE$ |
| 38 | ** |
| 39 | ****************************************************************************/ |
| 40 | |
| 41 | #ifndef QSTRINGCONVERTER_P_H |
| 42 | #define QSTRINGCONVERTER_P_H |
| 43 | |
| 44 | // |
| 45 | // W A R N I N G |
| 46 | // ------------- |
| 47 | // |
| 48 | // This file is not part of the Qt API. It exists purely as an |
| 49 | // implementation detail. This header file may change from version to |
| 50 | // version without notice, or even be removed. |
| 51 | // |
| 52 | // We mean it. |
| 53 | // |
| 54 | |
| 55 | #include <QtCore/qstring.h> |
| 56 | #include <QtCore/qendian.h> |
| 57 | #include <QtCore/qstringconverter.h> |
| 58 | |
| 59 | QT_BEGIN_NAMESPACE |
| 60 | |
| 61 | #ifndef __cpp_char8_t |
| 62 | enum char8_t : uchar {}; |
| 63 | #endif |
| 64 | |
| 65 | struct QUtf8BaseTraits |
| 66 | { |
| 67 | static const bool isTrusted = false; |
| 68 | static const bool allowNonCharacters = true; |
| 69 | static const bool skipAsciiHandling = false; |
| 70 | static const int Error = -1; |
| 71 | static const int EndOfString = -2; |
| 72 | |
| 73 | static bool isValidCharacter(uint u) |
| 74 | { return int(u) >= 0; } |
| 75 | |
| 76 | static void appendByte(uchar *&ptr, uchar b) |
| 77 | { *ptr++ = b; } |
| 78 | |
| 79 | static void appendByte(char8_t *&ptr, char8_t b) |
| 80 | { *ptr++ = b; } |
| 81 | |
| 82 | static uchar peekByte(const uchar *ptr, qsizetype n = 0) |
| 83 | { return ptr[n]; } |
| 84 | |
| 85 | static uchar peekByte(const char8_t *ptr, int n = 0) |
| 86 | { return ptr[n]; } |
| 87 | |
| 88 | static qptrdiff availableBytes(const uchar *ptr, const uchar *end) |
| 89 | { return end - ptr; } |
| 90 | |
| 91 | static qptrdiff availableBytes(const char8_t *ptr, const char8_t *end) |
| 92 | { return end - ptr; } |
| 93 | |
| 94 | static void advanceByte(const uchar *&ptr, qsizetype n = 1) |
| 95 | { ptr += n; } |
| 96 | |
| 97 | static void advanceByte(const char8_t *&ptr, int n = 1) |
| 98 | { ptr += n; } |
| 99 | |
| 100 | static void appendUtf16(ushort *&ptr, ushort uc) |
| 101 | { *ptr++ = uc; } |
| 102 | |
| 103 | static void appendUtf16(char16_t *&ptr, ushort uc) |
| 104 | { *ptr++ = char16_t(uc); } |
| 105 | |
| 106 | static void appendUcs4(ushort *&ptr, uint uc) |
| 107 | { |
| 108 | appendUtf16(ptr, QChar::highSurrogate(uc)); |
| 109 | appendUtf16(ptr, QChar::lowSurrogate(uc)); |
| 110 | } |
| 111 | |
| 112 | static void appendUcs4(char16_t *&ptr, char32_t uc) |
| 113 | { |
| 114 | appendUtf16(ptr, QChar::highSurrogate(uc)); |
| 115 | appendUtf16(ptr, QChar::lowSurrogate(uc)); |
| 116 | } |
| 117 | |
| 118 | static ushort peekUtf16(const ushort *ptr, qsizetype n = 0) |
| 119 | { return ptr[n]; } |
| 120 | |
| 121 | static ushort peekUtf16(const char16_t *ptr, int n = 0) |
| 122 | { return ptr[n]; } |
| 123 | |
| 124 | static qptrdiff availableUtf16(const ushort *ptr, const ushort *end) |
| 125 | { return end - ptr; } |
| 126 | |
| 127 | static qptrdiff availableUtf16(const char16_t *ptr, const char16_t *end) |
| 128 | { return end - ptr; } |
| 129 | |
| 130 | static void advanceUtf16(const ushort *&ptr, qsizetype n = 1) |
| 131 | { ptr += n; } |
| 132 | |
| 133 | static void advanceUtf16(const char16_t *&ptr, int n = 1) |
| 134 | { ptr += n; } |
| 135 | |
| 136 | // it's possible to output to UCS-4 too |
| 137 | static void appendUtf16(uint *&ptr, ushort uc) |
| 138 | { *ptr++ = uc; } |
| 139 | |
| 140 | static void appendUtf16(char32_t *&ptr, ushort uc) |
| 141 | { *ptr++ = char32_t(uc); } |
| 142 | |
| 143 | static void appendUcs4(uint *&ptr, uint uc) |
| 144 | { *ptr++ = uc; } |
| 145 | |
| 146 | static void appendUcs4(char32_t *&ptr, uint uc) |
| 147 | { *ptr++ = char32_t(uc); } |
| 148 | }; |
| 149 | |
| 150 | struct QUtf8BaseTraitsNoAscii : public QUtf8BaseTraits |
| 151 | { |
| 152 | static const bool skipAsciiHandling = true; |
| 153 | }; |
| 154 | |
| 155 | namespace QUtf8Functions |
| 156 | { |
| 157 | /// returns 0 on success; errors can only happen if \a u is a surrogate: |
| 158 | /// Error if \a u is a low surrogate; |
| 159 | /// if \a u is a high surrogate, Error if the next isn't a low one, |
| 160 | /// EndOfString if we run into the end of the string. |
| 161 | template <typename Traits, typename OutputPtr, typename InputPtr> inline |
| 162 | int toUtf8(ushort u, OutputPtr &dst, InputPtr &src, InputPtr end) |
| 163 | { |
| 164 | if (!Traits::skipAsciiHandling && u < 0x80) { |
| 165 | // U+0000 to U+007F (US-ASCII) - one byte |
| 166 | Traits::appendByte(dst, uchar(u)); |
| 167 | return 0; |
| 168 | } else if (u < 0x0800) { |
| 169 | // U+0080 to U+07FF - two bytes |
| 170 | // first of two bytes |
| 171 | Traits::appendByte(dst, 0xc0 | uchar(u >> 6)); |
| 172 | } else { |
| 173 | if (!QChar::isSurrogate(u)) { |
| 174 | // U+0800 to U+FFFF (except U+D800-U+DFFF) - three bytes |
| 175 | if (!Traits::allowNonCharacters && QChar::isNonCharacter(u)) |
| 176 | return Traits::Error; |
| 177 | |
| 178 | // first of three bytes |
| 179 | Traits::appendByte(dst, 0xe0 | uchar(u >> 12)); |
| 180 | } else { |
| 181 | // U+10000 to U+10FFFF - four bytes |
| 182 | // need to get one extra codepoint |
| 183 | if (Traits::availableUtf16(src, end) == 0) |
| 184 | return Traits::EndOfString; |
| 185 | |
| 186 | ushort low = Traits::peekUtf16(src); |
| 187 | if (!QChar::isHighSurrogate(u)) |
| 188 | return Traits::Error; |
| 189 | if (!QChar::isLowSurrogate(low)) |
| 190 | return Traits::Error; |
| 191 | |
| 192 | Traits::advanceUtf16(src); |
| 193 | uint ucs4 = QChar::surrogateToUcs4(u, low); |
| 194 | |
| 195 | if (!Traits::allowNonCharacters && QChar::isNonCharacter(ucs4)) |
| 196 | return Traits::Error; |
| 197 | |
| 198 | // first byte |
| 199 | Traits::appendByte(dst, 0xf0 | (uchar(ucs4 >> 18) & 0xf)); |
| 200 | |
| 201 | // second of four bytes |
| 202 | Traits::appendByte(dst, 0x80 | (uchar(ucs4 >> 12) & 0x3f)); |
| 203 | |
| 204 | // for the rest of the bytes |
| 205 | u = ushort(ucs4); |
| 206 | } |
| 207 | |
| 208 | // second to last byte |
| 209 | Traits::appendByte(dst, 0x80 | (uchar(u >> 6) & 0x3f)); |
| 210 | } |
| 211 | |
| 212 | // last byte |
| 213 | Traits::appendByte(dst, 0x80 | (u & 0x3f)); |
| 214 | return 0; |
| 215 | } |
| 216 | |
| 217 | inline bool isContinuationByte(uchar b) |
| 218 | { |
| 219 | return (b & 0xc0) == 0x80; |
| 220 | } |
| 221 | |
| 222 | /// returns the number of characters consumed (including \a b) in case of success; |
| 223 | /// returns negative in case of error: Traits::Error or Traits::EndOfString |
| 224 | template <typename Traits, typename OutputPtr, typename InputPtr> inline |
| 225 | qsizetype fromUtf8(uchar b, OutputPtr &dst, InputPtr &src, InputPtr end) |
| 226 | { |
| 227 | qsizetype charsNeeded; |
| 228 | uint min_uc; |
| 229 | uint uc; |
| 230 | |
| 231 | if (!Traits::skipAsciiHandling && b < 0x80) { |
| 232 | // US-ASCII |
| 233 | Traits::appendUtf16(dst, b); |
| 234 | return 1; |
| 235 | } |
| 236 | |
| 237 | if (!Traits::isTrusted && Q_UNLIKELY(b <= 0xC1)) { |
| 238 | // an UTF-8 first character must be at least 0xC0 |
| 239 | // however, all 0xC0 and 0xC1 first bytes can only produce overlong sequences |
| 240 | return Traits::Error; |
| 241 | } else if (b < 0xe0) { |
| 242 | charsNeeded = 2; |
| 243 | min_uc = 0x80; |
| 244 | uc = b & 0x1f; |
| 245 | } else if (b < 0xf0) { |
| 246 | charsNeeded = 3; |
| 247 | min_uc = 0x800; |
| 248 | uc = b & 0x0f; |
| 249 | } else if (b < 0xf5) { |
| 250 | charsNeeded = 4; |
| 251 | min_uc = 0x10000; |
| 252 | uc = b & 0x07; |
| 253 | } else { |
| 254 | // the last Unicode character is U+10FFFF |
| 255 | // it's encoded in UTF-8 as "\xF4\x8F\xBF\xBF" |
| 256 | // therefore, a byte higher than 0xF4 is not the UTF-8 first byte |
| 257 | return Traits::Error; |
| 258 | } |
| 259 | |
| 260 | qptrdiff bytesAvailable = Traits::availableBytes(src, end); |
| 261 | if (Q_UNLIKELY(bytesAvailable < charsNeeded - 1)) { |
| 262 | // it's possible that we have an error instead of just unfinished bytes |
| 263 | if (bytesAvailable > 0 && !isContinuationByte(Traits::peekByte(src, 0))) |
| 264 | return Traits::Error; |
| 265 | if (bytesAvailable > 1 && !isContinuationByte(Traits::peekByte(src, 1))) |
| 266 | return Traits::Error; |
| 267 | return Traits::EndOfString; |
| 268 | } |
| 269 | |
| 270 | // first continuation character |
| 271 | b = Traits::peekByte(src, 0); |
| 272 | if (!isContinuationByte(b)) |
| 273 | return Traits::Error; |
| 274 | uc <<= 6; |
| 275 | uc |= b & 0x3f; |
| 276 | |
| 277 | if (charsNeeded > 2) { |
| 278 | // second continuation character |
| 279 | b = Traits::peekByte(src, 1); |
| 280 | if (!isContinuationByte(b)) |
| 281 | return Traits::Error; |
| 282 | uc <<= 6; |
| 283 | uc |= b & 0x3f; |
| 284 | |
| 285 | if (charsNeeded > 3) { |
| 286 | // third continuation character |
| 287 | b = Traits::peekByte(src, 2); |
| 288 | if (!isContinuationByte(b)) |
| 289 | return Traits::Error; |
| 290 | uc <<= 6; |
| 291 | uc |= b & 0x3f; |
| 292 | } |
| 293 | } |
| 294 | |
| 295 | // we've decoded something; safety-check it |
| 296 | if (!Traits::isTrusted) { |
| 297 | if (uc < min_uc) |
| 298 | return Traits::Error; |
| 299 | if (QChar::isSurrogate(uc) || uc > QChar::LastValidCodePoint) |
| 300 | return Traits::Error; |
| 301 | if (!Traits::allowNonCharacters && QChar::isNonCharacter(uc)) |
| 302 | return Traits::Error; |
| 303 | } |
| 304 | |
| 305 | // write the UTF-16 sequence |
| 306 | if (!QChar::requiresSurrogates(uc)) { |
| 307 | // UTF-8 decoded and no surrogates are required |
| 308 | // detach if necessary |
| 309 | Traits::appendUtf16(dst, ushort(uc)); |
| 310 | } else { |
| 311 | // UTF-8 decoded to something that requires a surrogate pair |
| 312 | Traits::appendUcs4(dst, uc); |
| 313 | } |
| 314 | |
| 315 | Traits::advanceByte(src, charsNeeded - 1); |
| 316 | return charsNeeded; |
| 317 | } |
| 318 | } |
| 319 | |
| 320 | enum DataEndianness |
| 321 | { |
| 322 | DetectEndianness, |
| 323 | BigEndianness, |
| 324 | LittleEndianness |
| 325 | }; |
| 326 | |
| 327 | struct QUtf8 |
| 328 | { |
| 329 | Q_CORE_EXPORT static QChar *convertToUnicode(QChar *buffer, QByteArrayView in) noexcept; |
| 330 | static QString convertToUnicode(QByteArrayView in); |
| 331 | Q_CORE_EXPORT static QString convertToUnicode(QByteArrayView in, QStringConverter::State *state); |
| 332 | static QChar *convertToUnicode(QChar *out, QByteArrayView in, QStringConverter::State *state); |
| 333 | Q_CORE_EXPORT static QByteArray convertFromUnicode(QStringView in); |
| 334 | Q_CORE_EXPORT static QByteArray convertFromUnicode(QStringView in, QStringConverterBase::State *state); |
| 335 | static char *convertFromUnicode(char *out, QStringView in, QStringConverter::State *state); |
| 336 | struct ValidUtf8Result { |
| 337 | bool isValidUtf8; |
| 338 | bool isValidAscii; |
| 339 | }; |
| 340 | static ValidUtf8Result isValidUtf8(QByteArrayView in); |
| 341 | static int compareUtf8(QByteArrayView utf8, QStringView utf16) noexcept; |
| 342 | static int compareUtf8(QByteArrayView utf8, QLatin1String s); |
| 343 | }; |
| 344 | |
| 345 | struct QUtf16 |
| 346 | { |
| 347 | Q_CORE_EXPORT static QString convertToUnicode(QByteArrayView, QStringConverter::State *, DataEndianness = DetectEndianness); |
| 348 | static QChar *convertToUnicode(QChar *out, QByteArrayView, QStringConverter::State *state, DataEndianness endian); |
| 349 | Q_CORE_EXPORT static QByteArray convertFromUnicode(QStringView, QStringConverter::State *, DataEndianness = DetectEndianness); |
| 350 | static char *convertFromUnicode(char *out, QStringView in, QStringConverter::State *state, DataEndianness endian); |
| 351 | }; |
| 352 | |
| 353 | struct QUtf32 |
| 354 | { |
| 355 | static QChar *convertToUnicode(QChar *out, QByteArrayView, QStringConverter::State *state, DataEndianness endian); |
| 356 | Q_CORE_EXPORT static QString convertToUnicode(QByteArrayView, QStringConverter::State *, DataEndianness = DetectEndianness); |
| 357 | Q_CORE_EXPORT static QByteArray convertFromUnicode(QStringView, QStringConverter::State *, DataEndianness = DetectEndianness); |
| 358 | static char *convertFromUnicode(char *out, QStringView in, QStringConverter::State *state, DataEndianness endian); |
| 359 | }; |
| 360 | |
| 361 | struct Q_CORE_EXPORT QLocal8Bit |
| 362 | { |
| 363 | #if !defined(Q_OS_WIN) || defined(QT_BOOTSTRAPPED) |
| 364 | static QString convertToUnicode(QByteArrayView in, QStringConverter::State *state) |
| 365 | { return QUtf8::convertToUnicode(in, state); } |
| 366 | static QByteArray convertFromUnicode(QStringView in, QStringConverter::State *state) |
| 367 | { return QUtf8::convertFromUnicode(in, state); } |
| 368 | #else |
| 369 | static QString convertToUnicode(QByteArrayView, QStringConverter::State *); |
| 370 | static QByteArray convertFromUnicode(QStringView, QStringConverter::State *); |
| 371 | #endif |
| 372 | }; |
| 373 | |
| 374 | QT_END_NAMESPACE |
| 375 | |
| 376 | #endif // QSTRINGCONVERTER_P_H |
| 377 | |