| 1 | // |
| 2 | // UTF16Encoding.cpp |
| 3 | // |
| 4 | // Library: Foundation |
| 5 | // Package: Text |
| 6 | // Module: UTF16Encoding |
| 7 | // |
| 8 | // Copyright (c) 2004-2007, Applied Informatics Software Engineering GmbH. |
| 9 | // and Contributors. |
| 10 | // |
| 11 | // SPDX-License-Identifier: BSL-1.0 |
| 12 | // |
| 13 | |
| 14 | |
| 15 | #include "Poco/UTF16Encoding.h" |
| 16 | #include "Poco/ByteOrder.h" |
| 17 | #include "Poco/String.h" |
| 18 | |
| 19 | |
| 20 | namespace Poco { |
| 21 | |
| 22 | |
| 23 | const char* UTF16Encoding::_names[] = |
| 24 | { |
| 25 | "UTF-16" , |
| 26 | "UTF16" , |
| 27 | NULL |
| 28 | }; |
| 29 | |
| 30 | |
| 31 | const TextEncoding::CharacterMap UTF16Encoding::_charMap = |
| 32 | { |
| 33 | /* 00 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 34 | /* 10 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 35 | /* 20 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 36 | /* 30 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 37 | /* 40 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 38 | /* 50 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 39 | /* 60 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 40 | /* 70 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 41 | /* 80 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 42 | /* 90 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 43 | /* a0 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 44 | /* b0 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 45 | /* c0 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 46 | /* d0 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 47 | /* e0 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 48 | /* f0 */ -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, -2, |
| 49 | }; |
| 50 | |
| 51 | |
| 52 | UTF16Encoding::UTF16Encoding(ByteOrderType byteOrder) |
| 53 | { |
| 54 | setByteOrder(byteOrder); |
| 55 | } |
| 56 | |
| 57 | |
| 58 | UTF16Encoding::UTF16Encoding(int byteOrderMark) |
| 59 | { |
| 60 | setByteOrder(byteOrderMark); |
| 61 | } |
| 62 | |
| 63 | |
| 64 | UTF16Encoding::~UTF16Encoding() |
| 65 | { |
| 66 | } |
| 67 | |
| 68 | |
| 69 | UTF16Encoding::ByteOrderType UTF16Encoding::getByteOrder() const |
| 70 | { |
| 71 | #if defined(POCO_ARCH_BIG_ENDIAN) |
| 72 | return _flipBytes ? LITTLE_ENDIAN_BYTE_ORDER : BIG_ENDIAN_BYTE_ORDER; |
| 73 | #else |
| 74 | return _flipBytes ? BIG_ENDIAN_BYTE_ORDER : LITTLE_ENDIAN_BYTE_ORDER; |
| 75 | #endif |
| 76 | } |
| 77 | |
| 78 | |
| 79 | void UTF16Encoding::setByteOrder(ByteOrderType byteOrder) |
| 80 | { |
| 81 | #if defined(POCO_ARCH_BIG_ENDIAN) |
| 82 | _flipBytes = byteOrder == LITTLE_ENDIAN_BYTE_ORDER; |
| 83 | #else |
| 84 | _flipBytes = byteOrder == BIG_ENDIAN_BYTE_ORDER;; |
| 85 | #endif |
| 86 | } |
| 87 | |
| 88 | |
| 89 | void UTF16Encoding::setByteOrder(int byteOrderMark) |
| 90 | { |
| 91 | _flipBytes = byteOrderMark != 0xFEFF; |
| 92 | } |
| 93 | |
| 94 | |
| 95 | const char* UTF16Encoding::canonicalName() const |
| 96 | { |
| 97 | return _names[0]; |
| 98 | } |
| 99 | |
| 100 | |
| 101 | bool UTF16Encoding::isA(const std::string& encodingName) const |
| 102 | { |
| 103 | for (const char** name = _names; *name; ++name) |
| 104 | { |
| 105 | if (Poco::icompare(encodingName, *name) == 0) |
| 106 | return true; |
| 107 | } |
| 108 | return false; |
| 109 | } |
| 110 | |
| 111 | |
| 112 | const TextEncoding::CharacterMap& UTF16Encoding::characterMap() const |
| 113 | { |
| 114 | return _charMap; |
| 115 | } |
| 116 | |
| 117 | |
| 118 | int UTF16Encoding::convert(const unsigned char* bytes) const |
| 119 | { |
| 120 | UInt16 uc; |
| 121 | unsigned char* p = (unsigned char*) &uc; |
| 122 | *p++ = *bytes++; |
| 123 | *p++ = *bytes++; |
| 124 | |
| 125 | if (_flipBytes) |
| 126 | { |
| 127 | ByteOrder::flipBytes(uc); |
| 128 | } |
| 129 | |
| 130 | if (uc >= 0xd800 && uc < 0xdc00) |
| 131 | { |
| 132 | UInt16 uc2; |
| 133 | p = (unsigned char*) &uc2; |
| 134 | *p++ = *bytes++; |
| 135 | *p++ = *bytes++; |
| 136 | |
| 137 | if (_flipBytes) |
| 138 | { |
| 139 | ByteOrder::flipBytes(uc2); |
| 140 | } |
| 141 | if (uc2 >= 0xdc00 && uc2 < 0xe000) |
| 142 | { |
| 143 | return ((uc & 0x3ff) << 10) + (uc2 & 0x3ff) + 0x10000; |
| 144 | } |
| 145 | else |
| 146 | { |
| 147 | return -1; |
| 148 | } |
| 149 | } |
| 150 | else |
| 151 | { |
| 152 | return uc; |
| 153 | } |
| 154 | } |
| 155 | |
| 156 | |
| 157 | int UTF16Encoding::convert(int ch, unsigned char* bytes, int length) const |
| 158 | { |
| 159 | if (ch <= 0xFFFF) |
| 160 | { |
| 161 | if (bytes && length >= 2) |
| 162 | { |
| 163 | UInt16 ch1 = _flipBytes ? ByteOrder::flipBytes((UInt16) ch) : (UInt16) ch; |
| 164 | unsigned char* p = (unsigned char*) &ch1; |
| 165 | *bytes++ = *p++; |
| 166 | *bytes++ = *p++; |
| 167 | } |
| 168 | return 2; |
| 169 | } |
| 170 | else |
| 171 | { |
| 172 | if (bytes && length >= 4) |
| 173 | { |
| 174 | int ch1 = ch - 0x10000; |
| 175 | UInt16 w1 = 0xD800 + ((ch1 >> 10) & 0x3FF); |
| 176 | UInt16 w2 = 0xDC00 + (ch1 & 0x3FF); |
| 177 | if (_flipBytes) |
| 178 | { |
| 179 | w1 = ByteOrder::flipBytes(w1); |
| 180 | w2 = ByteOrder::flipBytes(w2); |
| 181 | } |
| 182 | unsigned char* p = (unsigned char*) &w1; |
| 183 | *bytes++ = *p++; |
| 184 | *bytes++ = *p++; |
| 185 | p = (unsigned char*) &w2; |
| 186 | *bytes++ = *p++; |
| 187 | *bytes++ = *p++; |
| 188 | } |
| 189 | return 4; |
| 190 | } |
| 191 | } |
| 192 | |
| 193 | |
| 194 | int UTF16Encoding::queryConvert(const unsigned char* bytes, int length) const |
| 195 | { |
| 196 | int ret = -2; |
| 197 | |
| 198 | if (length >= 2) |
| 199 | { |
| 200 | UInt16 uc; |
| 201 | unsigned char* p = (unsigned char*) &uc; |
| 202 | *p++ = *bytes++; |
| 203 | *p++ = *bytes++; |
| 204 | if (_flipBytes) |
| 205 | ByteOrder::flipBytes(uc); |
| 206 | if (uc >= 0xd800 && uc < 0xdc00) |
| 207 | { |
| 208 | if (length >= 4) |
| 209 | { |
| 210 | UInt16 uc2; |
| 211 | p = (unsigned char*) &uc2; |
| 212 | *p++ = *bytes++; |
| 213 | *p++ = *bytes++; |
| 214 | if (_flipBytes) |
| 215 | ByteOrder::flipBytes(uc2); |
| 216 | if (uc2 >= 0xdc00 && uc < 0xe000) |
| 217 | { |
| 218 | ret = ((uc & 0x3ff) << 10) + (uc2 & 0x3ff) + 0x10000; |
| 219 | } |
| 220 | else |
| 221 | { |
| 222 | ret = -1; // Malformed sequence |
| 223 | } |
| 224 | } |
| 225 | else |
| 226 | { |
| 227 | ret = -4; // surrogate pair, four bytes needed |
| 228 | } |
| 229 | } |
| 230 | else |
| 231 | { |
| 232 | ret = uc; |
| 233 | } |
| 234 | } |
| 235 | |
| 236 | return ret; |
| 237 | } |
| 238 | |
| 239 | |
| 240 | int UTF16Encoding::sequenceLength(const unsigned char* bytes, int length) const |
| 241 | { |
| 242 | int ret = -2; |
| 243 | |
| 244 | if (_flipBytes) |
| 245 | { |
| 246 | if (length >= 1) |
| 247 | { |
| 248 | unsigned char c = *bytes; |
| 249 | if (c >= 0xd8 && c < 0xdc) |
| 250 | ret = 4; |
| 251 | else |
| 252 | ret = 2; |
| 253 | } |
| 254 | } |
| 255 | else |
| 256 | { |
| 257 | if (length >= 2) |
| 258 | { |
| 259 | UInt16 uc; |
| 260 | unsigned char* p = (unsigned char*) &uc; |
| 261 | *p++ = *bytes++; |
| 262 | *p++ = *bytes++; |
| 263 | if (uc >= 0xd800 && uc < 0xdc00) |
| 264 | ret = 4; |
| 265 | else |
| 266 | ret = 2; |
| 267 | } |
| 268 | } |
| 269 | return ret; |
| 270 | } |
| 271 | |
| 272 | |
| 273 | } // namespace Poco |
| 274 | |