| OLD | NEW |
|---|
| (Empty) |
| 1 /* | |
| 2 * Copyright (C) 2004, 2006, 2008 Apple Inc. All rights reserved. | |
| 3 * | |
| 4 * Redistribution and use in source and binary forms, with or without | |
| 5 * modification, are permitted provided that the following conditions | |
| 6 * are met: | |
| 7 * 1. Redistributions of source code must retain the above copyright | |
| 8 * notice, this list of conditions and the following disclaimer. | |
| 9 * 2. Redistributions in binary form must reproduce the above copyright | |
| 10 * notice, this list of conditions and the following disclaimer in the | |
| 11 * documentation and/or other materials provided with the distribution. | |
| 12 * | |
| 13 * THIS SOFTWARE IS PROVIDED BY APPLE COMPUTER, INC. ``AS IS'' AND ANY | |
| 14 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE | |
| 15 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR | |
| 16 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE COMPUTER, INC. OR | |
| 17 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, | |
| 18 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, | |
| 19 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR | |
| 20 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY | |
| 21 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT | |
| 22 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE | |
| 23 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. | |
| 24 */ | |
| 25 | |
| 26 #include "wtf/text/TextCodecLatin1.h" | |
| 27 | |
| 28 #include "wtf/PtrUtil.h" | |
| 29 #include "wtf/text/CString.h" | |
| 30 #include "wtf/text/StringBuffer.h" | |
| 31 #include "wtf/text/TextCodecASCIIFastPath.h" | |
| 32 #include "wtf/text/WTFString.h" | |
| 33 #include <memory> | |
| 34 | |
| 35 namespace WTF { | |
| 36 | |
| 37 static const UChar table[256] = { | |
| 38 0x0000, 0x0001, 0x0002, 0x0003, 0x0004, 0x0005, 0x0006, 0x0007, // 00-07 | |
| 39 0x0008, 0x0009, 0x000A, 0x000B, 0x000C, 0x000D, 0x000E, 0x000F, // 08-0F | |
| 40 0x0010, 0x0011, 0x0012, 0x0013, 0x0014, 0x0015, 0x0016, 0x0017, // 10-17 | |
| 41 0x0018, 0x0019, 0x001A, 0x001B, 0x001C, 0x001D, 0x001E, 0x001F, // 18-1F | |
| 42 0x0020, 0x0021, 0x0022, 0x0023, 0x0024, 0x0025, 0x0026, 0x0027, // 20-27 | |
| 43 0x0028, 0x0029, 0x002A, 0x002B, 0x002C, 0x002D, 0x002E, 0x002F, // 28-2F | |
| 44 0x0030, 0x0031, 0x0032, 0x0033, 0x0034, 0x0035, 0x0036, 0x0037, // 30-37 | |
| 45 0x0038, 0x0039, 0x003A, 0x003B, 0x003C, 0x003D, 0x003E, 0x003F, // 38-3F | |
| 46 0x0040, 0x0041, 0x0042, 0x0043, 0x0044, 0x0045, 0x0046, 0x0047, // 40-47 | |
| 47 0x0048, 0x0049, 0x004A, 0x004B, 0x004C, 0x004D, 0x004E, 0x004F, // 48-4F | |
| 48 0x0050, 0x0051, 0x0052, 0x0053, 0x0054, 0x0055, 0x0056, 0x0057, // 50-57 | |
| 49 0x0058, 0x0059, 0x005A, 0x005B, 0x005C, 0x005D, 0x005E, 0x005F, // 58-5F | |
| 50 0x0060, 0x0061, 0x0062, 0x0063, 0x0064, 0x0065, 0x0066, 0x0067, // 60-67 | |
| 51 0x0068, 0x0069, 0x006A, 0x006B, 0x006C, 0x006D, 0x006E, 0x006F, // 68-6F | |
| 52 0x0070, 0x0071, 0x0072, 0x0073, 0x0074, 0x0075, 0x0076, 0x0077, // 70-77 | |
| 53 0x0078, 0x0079, 0x007A, 0x007B, 0x007C, 0x007D, 0x007E, 0x007F, // 78-7F | |
| 54 0x20AC, 0x0081, 0x201A, 0x0192, 0x201E, 0x2026, 0x2020, 0x2021, // 80-87 | |
| 55 0x02C6, 0x2030, 0x0160, 0x2039, 0x0152, 0x008D, 0x017D, 0x008F, // 88-8F | |
| 56 0x0090, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014, // 90-97 | |
| 57 0x02DC, 0x2122, 0x0161, 0x203A, 0x0153, 0x009D, 0x017E, 0x0178, // 98-9F | |
| 58 0x00A0, 0x00A1, 0x00A2, 0x00A3, 0x00A4, 0x00A5, 0x00A6, 0x00A7, // A0-A7 | |
| 59 0x00A8, 0x00A9, 0x00AA, 0x00AB, 0x00AC, 0x00AD, 0x00AE, 0x00AF, // A8-AF | |
| 60 0x00B0, 0x00B1, 0x00B2, 0x00B3, 0x00B4, 0x00B5, 0x00B6, 0x00B7, // B0-B7 | |
| 61 0x00B8, 0x00B9, 0x00BA, 0x00BB, 0x00BC, 0x00BD, 0x00BE, 0x00BF, // B8-BF | |
| 62 0x00C0, 0x00C1, 0x00C2, 0x00C3, 0x00C4, 0x00C5, 0x00C6, 0x00C7, // C0-C7 | |
| 63 0x00C8, 0x00C9, 0x00CA, 0x00CB, 0x00CC, 0x00CD, 0x00CE, 0x00CF, // C8-CF | |
| 64 0x00D0, 0x00D1, 0x00D2, 0x00D3, 0x00D4, 0x00D5, 0x00D6, 0x00D7, // D0-D7 | |
| 65 0x00D8, 0x00D9, 0x00DA, 0x00DB, 0x00DC, 0x00DD, 0x00DE, 0x00DF, // D8-DF | |
| 66 0x00E0, 0x00E1, 0x00E2, 0x00E3, 0x00E4, 0x00E5, 0x00E6, 0x00E7, // E0-E7 | |
| 67 0x00E8, 0x00E9, 0x00EA, 0x00EB, 0x00EC, 0x00ED, 0x00EE, 0x00EF, // E8-EF | |
| 68 0x00F0, 0x00F1, 0x00F2, 0x00F3, 0x00F4, 0x00F5, 0x00F6, 0x00F7, // F0-F7 | |
| 69 0x00F8, 0x00F9, 0x00FA, 0x00FB, 0x00FC, 0x00FD, 0x00FE, 0x00FF // F8-FF | |
| 70 }; | |
| 71 | |
| 72 void TextCodecLatin1::registerEncodingNames(EncodingNameRegistrar registrar) { | |
| 73 // Taken from the alias table at https://encoding.spec.whatwg.org/ | |
| 74 registrar("windows-1252", "windows-1252"); | |
| 75 registrar("ANSI_X3.4-1968", "windows-1252"); | |
| 76 registrar("ASCII", "windows-1252"); | |
| 77 registrar("cp1252", "windows-1252"); | |
| 78 registrar("cp819", "windows-1252"); | |
| 79 registrar("csISOLatin1", "windows-1252"); | |
| 80 registrar("IBM819", "windows-1252"); | |
| 81 registrar("ISO-8859-1", "windows-1252"); | |
| 82 registrar("iso-ir-100", "windows-1252"); | |
| 83 registrar("iso8859-1", "windows-1252"); | |
| 84 registrar("iso88591", "windows-1252"); | |
| 85 registrar("iso_8859-1", "windows-1252"); | |
| 86 registrar("iso_8859-1:1987", "windows-1252"); | |
| 87 registrar("l1", "windows-1252"); | |
| 88 registrar("latin1", "windows-1252"); | |
| 89 registrar("US-ASCII", "windows-1252"); | |
| 90 registrar("x-cp1252", "windows-1252"); | |
| 91 } | |
| 92 | |
| 93 static std::unique_ptr<TextCodec> newStreamingTextDecoderWindowsLatin1( | |
| 94 const TextEncoding&, | |
| 95 const void*) { | |
| 96 return WTF::wrapUnique(new TextCodecLatin1); | |
| 97 } | |
| 98 | |
| 99 void TextCodecLatin1::registerCodecs(TextCodecRegistrar registrar) { | |
| 100 registrar("windows-1252", newStreamingTextDecoderWindowsLatin1, 0); | |
| 101 | |
| 102 // ASCII and Latin-1 both decode as Windows Latin-1 although they retain | |
| 103 // unique identities. | |
| 104 registrar("ISO-8859-1", newStreamingTextDecoderWindowsLatin1, 0); | |
| 105 registrar("US-ASCII", newStreamingTextDecoderWindowsLatin1, 0); | |
| 106 } | |
| 107 | |
| 108 String TextCodecLatin1::decode(const char* bytes, | |
| 109 size_t length, | |
| 110 FlushBehavior, | |
| 111 bool, | |
| 112 bool&) { | |
| 113 LChar* characters; | |
| 114 if (!length) | |
| 115 return emptyString; | |
| 116 String result = String::createUninitialized(length, characters); | |
| 117 | |
| 118 const uint8_t* source = reinterpret_cast<const uint8_t*>(bytes); | |
| 119 const uint8_t* end = reinterpret_cast<const uint8_t*>(bytes + length); | |
| 120 const uint8_t* alignedEnd = alignToMachineWord(end); | |
| 121 LChar* destination = characters; | |
| 122 | |
| 123 while (source < end) { | |
| 124 if (isASCII(*source)) { | |
| 125 // Fast path for ASCII. Most Latin-1 text will be ASCII. | |
| 126 if (isAlignedToMachineWord(source)) { | |
| 127 while (source < alignedEnd) { | |
| 128 MachineWord chunk = *reinterpret_cast_ptr<const MachineWord*>(source); | |
| 129 | |
| 130 if (!isAllASCII<LChar>(chunk)) | |
| 131 goto useLookupTable; | |
| 132 | |
| 133 copyASCIIMachineWord(destination, source); | |
| 134 source += sizeof(MachineWord); | |
| 135 destination += sizeof(MachineWord); | |
| 136 } | |
| 137 | |
| 138 if (source == end) | |
| 139 break; | |
| 140 } | |
| 141 *destination = *source; | |
| 142 } else { | |
| 143 useLookupTable: | |
| 144 if (table[*source] > 0xff) | |
| 145 goto upConvertTo16Bit; | |
| 146 | |
| 147 *destination = static_cast<LChar>(table[*source]); | |
| 148 } | |
| 149 | |
| 150 ++source; | |
| 151 ++destination; | |
| 152 } | |
| 153 | |
| 154 return result; | |
| 155 | |
| 156 upConvertTo16Bit: | |
| 157 UChar* characters16; | |
| 158 String result16 = String::createUninitialized(length, characters16); | |
| 159 | |
| 160 UChar* destination16 = characters16; | |
| 161 | |
| 162 // Zero extend and copy already processed 8 bit data | |
| 163 LChar* ptr8 = characters; | |
| 164 LChar* endPtr8 = destination; | |
| 165 | |
| 166 while (ptr8 < endPtr8) | |
| 167 *destination16++ = *ptr8++; | |
| 168 | |
| 169 // Handle the character that triggered the 16 bit path | |
| 170 *destination16 = table[*source]; | |
| 171 ++source; | |
| 172 ++destination16; | |
| 173 | |
| 174 while (source < end) { | |
| 175 if (isASCII(*source)) { | |
| 176 // Fast path for ASCII. Most Latin-1 text will be ASCII. | |
| 177 if (isAlignedToMachineWord(source)) { | |
| 178 while (source < alignedEnd) { | |
| 179 MachineWord chunk = *reinterpret_cast_ptr<const MachineWord*>(source); | |
| 180 | |
| 181 if (!isAllASCII<LChar>(chunk)) | |
| 182 goto useLookupTable16; | |
| 183 | |
| 184 copyASCIIMachineWord(destination16, source); | |
| 185 source += sizeof(MachineWord); | |
| 186 destination16 += sizeof(MachineWord); | |
| 187 } | |
| 188 | |
| 189 if (source == end) | |
| 190 break; | |
| 191 } | |
| 192 *destination16 = *source; | |
| 193 } else { | |
| 194 useLookupTable16: | |
| 195 *destination16 = table[*source]; | |
| 196 } | |
| 197 | |
| 198 ++source; | |
| 199 ++destination16; | |
| 200 } | |
| 201 | |
| 202 return result16; | |
| 203 } | |
| 204 | |
| 205 template <typename CharType> | |
| 206 static CString encodeComplexWindowsLatin1(const CharType* characters, | |
| 207 size_t length, | |
| 208 UnencodableHandling handling) { | |
| 209 size_t targetLength = length; | |
| 210 Vector<char> result(targetLength); | |
| 211 char* bytes = result.data(); | |
| 212 | |
| 213 size_t resultLength = 0; | |
| 214 for (size_t i = 0; i < length;) { | |
| 215 UChar32 c; | |
| 216 // If CharType is LChar the U16_NEXT call reads a byte and increments; | |
| 217 // since the convention is that LChar is already latin1 this is safe. | |
| 218 U16_NEXT(characters, i, length, c); | |
| 219 // If input was a surrogate pair (non-BMP character) then we overestimated | |
| 220 // the length. | |
| 221 if (c > 0xffff) | |
| 222 --targetLength; | |
| 223 unsigned char b = static_cast<unsigned char>(c); | |
| 224 // Do an efficient check to detect characters other than 00-7F and A0-FF. | |
| 225 if (b != c || (c & 0xE0) == 0x80) { | |
| 226 // Look for a way to encode this with Windows Latin-1. | |
| 227 for (b = 0x80; b < 0xA0; ++b) { | |
| 228 if (table[b] == c) | |
| 229 goto gotByte; | |
| 230 } | |
| 231 // No way to encode this character with Windows Latin-1. | |
| 232 UnencodableReplacementArray replacement; | |
| 233 int replacementLength = | |
| 234 TextCodec::getUnencodableReplacement(c, handling, replacement); | |
| 235 DCHECK_GT(replacementLength, 0); | |
| 236 // Only one char was initially reserved per input character, so grow if | |
| 237 // necessary. Note that the case of surrogate pairs and | |
| 238 // QuestionMarksForUnencodables the result length may be shorter than | |
| 239 // the input length. | |
| 240 targetLength += replacementLength - 1; | |
| 241 if (targetLength > result.size()) { | |
| 242 result.grow(targetLength); | |
| 243 bytes = result.data(); | |
| 244 } | |
| 245 memcpy(bytes + resultLength, replacement, replacementLength); | |
| 246 resultLength += replacementLength; | |
| 247 continue; | |
| 248 } | |
| 249 gotByte: | |
| 250 bytes[resultLength++] = b; | |
| 251 } | |
| 252 | |
| 253 return CString(bytes, resultLength); | |
| 254 } | |
| 255 | |
| 256 template <typename CharType> | |
| 257 CString TextCodecLatin1::encodeCommon(const CharType* characters, | |
| 258 size_t length, | |
| 259 UnencodableHandling handling) { | |
| 260 { | |
| 261 char* bytes; | |
| 262 CString string = CString::createUninitialized(length, bytes); | |
| 263 | |
| 264 // Convert the string a fast way and simultaneously do an efficient check to | |
| 265 // see if it's all ASCII. | |
| 266 UChar ored = 0; | |
| 267 for (size_t i = 0; i < length; ++i) { | |
| 268 UChar c = characters[i]; | |
| 269 bytes[i] = static_cast<char>(c); | |
| 270 ored |= c; | |
| 271 } | |
| 272 | |
| 273 if (!(ored & 0xFF80)) | |
| 274 return string; | |
| 275 } | |
| 276 | |
| 277 // If it wasn't all ASCII, call the function that handles more-complex cases. | |
| 278 return encodeComplexWindowsLatin1(characters, length, handling); | |
| 279 } | |
| 280 | |
| 281 CString TextCodecLatin1::encode(const UChar* characters, | |
| 282 size_t length, | |
| 283 UnencodableHandling handling) { | |
| 284 return encodeCommon(characters, length, handling); | |
| 285 } | |
| 286 | |
| 287 CString TextCodecLatin1::encode(const LChar* characters, | |
| 288 size_t length, | |
| 289 UnencodableHandling handling) { | |
| 290 return encodeCommon(characters, length, handling); | |
| 291 } | |
| 292 | |
| 293 } // namespace WTF | |
| OLD | NEW |
|---|
Chromium Code Reviews
Issue 2764283002:
Move files in wtf/ to platform/wtf/ (Part 10). (Closed)
