| OLD | NEW |
|---|
| (Empty) |
| 1 // Copyright 2014 PDFium Authors. All rights reserved. | |
| 2 // Use of this source code is governed by a BSD-style license that can be | |
| 3 // found in the LICENSE file. | |
| 4 | |
| 5 // Original code copyright 2014 Foxit Software Inc. http://www.foxitsoftware.com | |
| 6 | |
| 7 #include "core/fpdfapi/fpdf_font/font_int.h" | |
| 8 | |
| 9 #include "core/fpdfapi/cmaps/cmap_int.h" | |
| 10 #include "core/fpdfapi/cpdf_modulemgr.h" | |
| 11 #include "core/fpdfapi/fpdf_font/ttgsubtable.h" | |
| 12 #include "core/fpdfapi/fpdf_page/cpdf_pagemodule.h" | |
| 13 #include "core/fpdfapi/fpdf_parser/cpdf_array.h" | |
| 14 #include "core/fpdfapi/fpdf_parser/cpdf_dictionary.h" | |
| 15 #include "core/fpdfapi/fpdf_parser/cpdf_simple_parser.h" | |
| 16 #include "core/fxcrt/fx_ext.h" | |
| 17 #include "core/fxge/fx_freetype.h" | |
| 18 | |
| 19 namespace { | |
| 20 | |
| 21 const FX_CHAR* const g_CharsetNames[CIDSET_NUM_SETS] = { | |
| 22 nullptr, "GB1", "CNS1", "Japan1", "Korea1", "UCS"}; | |
| 23 | |
| 24 | |
| 25 class CPDF_PredefinedCMap { | |
| 26 public: | |
| 27 const FX_CHAR* m_pName; | |
| 28 CIDSet m_Charset; | |
| 29 CIDCoding m_Coding; | |
| 30 CPDF_CMap::CodingScheme m_CodingScheme; | |
| 31 uint8_t m_LeadingSegCount; | |
| 32 uint8_t m_LeadingSegs[4]; | |
| 33 }; | |
| 34 | |
| 35 const CPDF_PredefinedCMap g_PredefinedCMaps[] = { | |
| 36 {"GB-EUC", | |
| 37 CIDSET_GB1, | |
| 38 CIDCODING_GB, | |
| 39 CPDF_CMap::MixedTwoBytes, | |
| 40 1, | |
| 41 {0xa1, 0xfe}}, | |
| 42 {"GBpc-EUC", | |
| 43 CIDSET_GB1, | |
| 44 CIDCODING_GB, | |
| 45 CPDF_CMap::MixedTwoBytes, | |
| 46 1, | |
| 47 {0xa1, 0xfc}}, | |
| 48 {"GBK-EUC", | |
| 49 CIDSET_GB1, | |
| 50 CIDCODING_GB, | |
| 51 CPDF_CMap::MixedTwoBytes, | |
| 52 1, | |
| 53 {0x81, 0xfe}}, | |
| 54 {"GBKp-EUC", | |
| 55 CIDSET_GB1, | |
| 56 CIDCODING_GB, | |
| 57 CPDF_CMap::MixedTwoBytes, | |
| 58 1, | |
| 59 {0x81, 0xfe}}, | |
| 60 {"GBK2K-EUC", | |
| 61 CIDSET_GB1, | |
| 62 CIDCODING_GB, | |
| 63 CPDF_CMap::MixedTwoBytes, | |
| 64 1, | |
| 65 {0x81, 0xfe}}, | |
| 66 {"GBK2K", | |
| 67 CIDSET_GB1, | |
| 68 CIDCODING_GB, | |
| 69 CPDF_CMap::MixedTwoBytes, | |
| 70 1, | |
| 71 {0x81, 0xfe}}, | |
| 72 {"UniGB-UCS2", CIDSET_GB1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 73 {"UniGB-UTF16", CIDSET_GB1, CIDCODING_UTF16, CPDF_CMap::TwoBytes, 0, {}}, | |
| 74 {"B5pc", | |
| 75 CIDSET_CNS1, | |
| 76 CIDCODING_BIG5, | |
| 77 CPDF_CMap::MixedTwoBytes, | |
| 78 1, | |
| 79 {0xa1, 0xfc}}, | |
| 80 {"HKscs-B5", | |
| 81 CIDSET_CNS1, | |
| 82 CIDCODING_BIG5, | |
| 83 CPDF_CMap::MixedTwoBytes, | |
| 84 1, | |
| 85 {0x88, 0xfe}}, | |
| 86 {"ETen-B5", | |
| 87 CIDSET_CNS1, | |
| 88 CIDCODING_BIG5, | |
| 89 CPDF_CMap::MixedTwoBytes, | |
| 90 1, | |
| 91 {0xa1, 0xfe}}, | |
| 92 {"ETenms-B5", | |
| 93 CIDSET_CNS1, | |
| 94 CIDCODING_BIG5, | |
| 95 CPDF_CMap::MixedTwoBytes, | |
| 96 1, | |
| 97 {0xa1, 0xfe}}, | |
| 98 {"UniCNS-UCS2", CIDSET_CNS1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 99 {"UniCNS-UTF16", CIDSET_CNS1, CIDCODING_UTF16, CPDF_CMap::TwoBytes, 0, {}}, | |
| 100 {"83pv-RKSJ", | |
| 101 CIDSET_JAPAN1, | |
| 102 CIDCODING_JIS, | |
| 103 CPDF_CMap::MixedTwoBytes, | |
| 104 2, | |
| 105 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 106 {"90ms-RKSJ", | |
| 107 CIDSET_JAPAN1, | |
| 108 CIDCODING_JIS, | |
| 109 CPDF_CMap::MixedTwoBytes, | |
| 110 2, | |
| 111 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 112 {"90msp-RKSJ", | |
| 113 CIDSET_JAPAN1, | |
| 114 CIDCODING_JIS, | |
| 115 CPDF_CMap::MixedTwoBytes, | |
| 116 2, | |
| 117 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 118 {"90pv-RKSJ", | |
| 119 CIDSET_JAPAN1, | |
| 120 CIDCODING_JIS, | |
| 121 CPDF_CMap::MixedTwoBytes, | |
| 122 2, | |
| 123 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 124 {"Add-RKSJ", | |
| 125 CIDSET_JAPAN1, | |
| 126 CIDCODING_JIS, | |
| 127 CPDF_CMap::MixedTwoBytes, | |
| 128 2, | |
| 129 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 130 {"EUC", | |
| 131 CIDSET_JAPAN1, | |
| 132 CIDCODING_JIS, | |
| 133 CPDF_CMap::MixedTwoBytes, | |
| 134 2, | |
| 135 {0x8e, 0x8e, 0xa1, 0xfe}}, | |
| 136 {"H", CIDSET_JAPAN1, CIDCODING_JIS, CPDF_CMap::TwoBytes, 1, {0x21, 0x7e}}, | |
| 137 {"V", CIDSET_JAPAN1, CIDCODING_JIS, CPDF_CMap::TwoBytes, 1, {0x21, 0x7e}}, | |
| 138 {"Ext-RKSJ", | |
| 139 CIDSET_JAPAN1, | |
| 140 CIDCODING_JIS, | |
| 141 CPDF_CMap::MixedTwoBytes, | |
| 142 2, | |
| 143 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 144 {"UniJIS-UCS2", CIDSET_JAPAN1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 145 {"UniJIS-UCS2-HW", | |
| 146 CIDSET_JAPAN1, | |
| 147 CIDCODING_UCS2, | |
| 148 CPDF_CMap::TwoBytes, | |
| 149 0, | |
| 150 {}}, | |
| 151 {"UniJIS-UTF16", | |
| 152 CIDSET_JAPAN1, | |
| 153 CIDCODING_UTF16, | |
| 154 CPDF_CMap::TwoBytes, | |
| 155 0, | |
| 156 {}}, | |
| 157 {"KSC-EUC", | |
| 158 CIDSET_KOREA1, | |
| 159 CIDCODING_KOREA, | |
| 160 CPDF_CMap::MixedTwoBytes, | |
| 161 1, | |
| 162 {0xa1, 0xfe}}, | |
| 163 {"KSCms-UHC", | |
| 164 CIDSET_KOREA1, | |
| 165 CIDCODING_KOREA, | |
| 166 CPDF_CMap::MixedTwoBytes, | |
| 167 1, | |
| 168 {0x81, 0xfe}}, | |
| 169 {"KSCms-UHC-HW", | |
| 170 CIDSET_KOREA1, | |
| 171 CIDCODING_KOREA, | |
| 172 CPDF_CMap::MixedTwoBytes, | |
| 173 1, | |
| 174 {0x81, 0xfe}}, | |
| 175 {"KSCpc-EUC", | |
| 176 CIDSET_KOREA1, | |
| 177 CIDCODING_KOREA, | |
| 178 CPDF_CMap::MixedTwoBytes, | |
| 179 1, | |
| 180 {0xa1, 0xfd}}, | |
| 181 {"UniKS-UCS2", CIDSET_KOREA1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 182 {"UniKS-UTF16", CIDSET_KOREA1, CIDCODING_UTF16, CPDF_CMap::TwoBytes, 0, {}}, | |
| 183 }; | |
| 184 | |
| 185 CIDSet CIDSetFromSizeT(size_t index) { | |
| 186 if (index >= CIDSET_NUM_SETS) { | |
| 187 NOTREACHED(); | |
| 188 return CIDSET_UNKNOWN; | |
| 189 } | |
| 190 return static_cast<CIDSet>(index); | |
| 191 } | |
| 192 | |
| 193 CFX_ByteStringC CMap_GetString(const CFX_ByteStringC& word) { | |
| 194 if (word.GetLength() <= 2) | |
| 195 return CFX_ByteStringC(); | |
| 196 return CFX_ByteStringC(&word[1], word.GetLength() - 2); | |
| 197 } | |
| 198 | |
| 199 int CompareDWORD(const void* data1, const void* data2) { | |
| 200 return (*(uint32_t*)data1) - (*(uint32_t*)data2); | |
| 201 } | |
| 202 | |
| 203 int CompareCID(const void* key, const void* element) { | |
| 204 if ((*(uint32_t*)key) < (*(uint32_t*)element)) { | |
| 205 return -1; | |
| 206 } | |
| 207 if ((*(uint32_t*)key) > | |
| 208 (*(uint32_t*)element) + ((uint32_t*)element)[1] / 65536) { | |
| 209 return 1; | |
| 210 } | |
| 211 return 0; | |
| 212 } | |
| 213 | |
| 214 int CheckCodeRange(uint8_t* codes, | |
| 215 int size, | |
| 216 CMap_CodeRange* pRanges, | |
| 217 int nRanges) { | |
| 218 int iSeg = nRanges - 1; | |
| 219 while (iSeg >= 0) { | |
| 220 if (pRanges[iSeg].m_CharSize < size) { | |
| 221 --iSeg; | |
| 222 continue; | |
| 223 } | |
| 224 int iChar = 0; | |
| 225 while (iChar < size) { | |
| 226 if (codes[iChar] < pRanges[iSeg].m_Lower[iChar] || | |
| 227 codes[iChar] > pRanges[iSeg].m_Upper[iChar]) { | |
| 228 break; | |
| 229 } | |
| 230 ++iChar; | |
| 231 } | |
| 232 if (iChar == pRanges[iSeg].m_CharSize) | |
| 233 return 2; | |
| 234 | |
| 235 if (iChar) | |
| 236 return (size == pRanges[iSeg].m_CharSize) ? 2 : 1; | |
| 237 iSeg--; | |
| 238 } | |
| 239 return 0; | |
| 240 } | |
| 241 | |
| 242 int GetCharSizeImpl(uint32_t charcode, | |
| 243 CMap_CodeRange* pRanges, | |
| 244 int iRangesSize) { | |
| 245 if (!iRangesSize) | |
| 246 return 1; | |
| 247 | |
| 248 uint8_t codes[4]; | |
| 249 codes[0] = codes[1] = 0x00; | |
| 250 codes[2] = (uint8_t)(charcode >> 8 & 0xFF); | |
| 251 codes[3] = (uint8_t)charcode; | |
| 252 int offset = 0; | |
| 253 int size = 4; | |
| 254 for (int i = 0; i < 4; ++i) { | |
| 255 int iSeg = iRangesSize - 1; | |
| 256 while (iSeg >= 0) { | |
| 257 if (pRanges[iSeg].m_CharSize < size) { | |
| 258 --iSeg; | |
| 259 continue; | |
| 260 } | |
| 261 int iChar = 0; | |
| 262 while (iChar < size) { | |
| 263 if (codes[offset + iChar] < pRanges[iSeg].m_Lower[iChar] || | |
| 264 codes[offset + iChar] > pRanges[iSeg].m_Upper[iChar]) { | |
| 265 break; | |
| 266 } | |
| 267 ++iChar; | |
| 268 } | |
| 269 if (iChar == pRanges[iSeg].m_CharSize) | |
| 270 return size; | |
| 271 --iSeg; | |
| 272 } | |
| 273 --size; | |
| 274 ++offset; | |
| 275 } | |
| 276 return 1; | |
| 277 } | |
| 278 | |
| 279 } // namespace | |
| 280 | |
| 281 CPDF_CMapManager::CPDF_CMapManager() { | |
| 282 FXSYS_memset(m_CID2UnicodeMaps, 0, sizeof m_CID2UnicodeMaps); | |
| 283 } | |
| 284 CPDF_CMapManager::~CPDF_CMapManager() { | |
| 285 for (const auto& pair : m_CMaps) { | |
| 286 delete pair.second; | |
| 287 } | |
| 288 m_CMaps.clear(); | |
| 289 for (size_t i = 0; i < FX_ArraySize(m_CID2UnicodeMaps); ++i) { | |
| 290 delete m_CID2UnicodeMaps[i]; | |
| 291 } | |
| 292 } | |
| 293 CPDF_CMap* CPDF_CMapManager::GetPredefinedCMap(const CFX_ByteString& name, | |
| 294 bool bPromptCJK) { | |
| 295 auto it = m_CMaps.find(name); | |
| 296 if (it != m_CMaps.end()) { | |
| 297 return it->second; | |
| 298 } | |
| 299 CPDF_CMap* pCMap = LoadPredefinedCMap(name, bPromptCJK); | |
| 300 if (!name.IsEmpty()) { | |
| 301 m_CMaps[name] = pCMap; | |
| 302 } | |
| 303 return pCMap; | |
| 304 } | |
| 305 CPDF_CMap* CPDF_CMapManager::LoadPredefinedCMap(const CFX_ByteString& name, | |
| 306 bool bPromptCJK) { | |
| 307 CPDF_CMap* pCMap = new CPDF_CMap; | |
| 308 const FX_CHAR* pname = name.c_str(); | |
| 309 if (*pname == '/') { | |
| 310 pname++; | |
| 311 } | |
| 312 pCMap->LoadPredefined(this, pname, bPromptCJK); | |
| 313 return pCMap; | |
| 314 } | |
| 315 | |
| 316 CPDF_CID2UnicodeMap* CPDF_CMapManager::GetCID2UnicodeMap(CIDSet charset, | |
| 317 bool bPromptCJK) { | |
| 318 if (!m_CID2UnicodeMaps[charset]) | |
| 319 m_CID2UnicodeMaps[charset] = LoadCID2UnicodeMap(charset, bPromptCJK); | |
| 320 return m_CID2UnicodeMaps[charset]; | |
| 321 } | |
| 322 CPDF_CID2UnicodeMap* CPDF_CMapManager::LoadCID2UnicodeMap(CIDSet charset, | |
| 323 bool bPromptCJK) { | |
| 324 CPDF_CID2UnicodeMap* pMap = new CPDF_CID2UnicodeMap(); | |
| 325 pMap->Load(this, charset, bPromptCJK); | |
| 326 return pMap; | |
| 327 } | |
| 328 | |
| 329 CPDF_CMapParser::CPDF_CMapParser() | |
| 330 : m_pCMap(nullptr), m_Status(0), m_CodeSeq(0) {} | |
| 331 | |
| 332 CPDF_CMapParser::~CPDF_CMapParser() {} | |
| 333 | |
| 334 void CPDF_CMapParser::Initialize(CPDF_CMap* pCMap) { | |
| 335 m_pCMap = pCMap; | |
| 336 m_Status = 0; | |
| 337 m_CodeSeq = 0; | |
| 338 m_AddMaps.EstimateSize(0, 10240); | |
| 339 } | |
| 340 | |
| 341 void CPDF_CMapParser::ParseWord(const CFX_ByteStringC& word) { | |
| 342 if (word.IsEmpty()) { | |
| 343 return; | |
| 344 } | |
| 345 if (word == "begincidchar") { | |
| 346 m_Status = 1; | |
| 347 m_CodeSeq = 0; | |
| 348 } else if (word == "begincidrange") { | |
| 349 m_Status = 2; | |
| 350 m_CodeSeq = 0; | |
| 351 } else if (word == "endcidrange" || word == "endcidchar") { | |
| 352 m_Status = 0; | |
| 353 } else if (word == "/WMode") { | |
| 354 m_Status = 6; | |
| 355 } else if (word == "/Registry") { | |
| 356 m_Status = 3; | |
| 357 } else if (word == "/Ordering") { | |
| 358 m_Status = 4; | |
| 359 } else if (word == "/Supplement") { | |
| 360 m_Status = 5; | |
| 361 } else if (word == "begincodespacerange") { | |
| 362 m_Status = 7; | |
| 363 m_CodeSeq = 0; | |
| 364 } else if (word == "usecmap") { | |
| 365 } else if (m_Status == 1 || m_Status == 2) { | |
| 366 m_CodePoints[m_CodeSeq] = CMap_GetCode(word); | |
| 367 m_CodeSeq++; | |
| 368 uint32_t StartCode, EndCode; | |
| 369 uint16_t StartCID; | |
| 370 if (m_Status == 1) { | |
| 371 if (m_CodeSeq < 2) { | |
| 372 return; | |
| 373 } | |
| 374 EndCode = StartCode = m_CodePoints[0]; | |
| 375 StartCID = (uint16_t)m_CodePoints[1]; | |
| 376 } else { | |
| 377 if (m_CodeSeq < 3) { | |
| 378 return; | |
| 379 } | |
| 380 StartCode = m_CodePoints[0]; | |
| 381 EndCode = m_CodePoints[1]; | |
| 382 StartCID = (uint16_t)m_CodePoints[2]; | |
| 383 } | |
| 384 if (EndCode < 0x10000) { | |
| 385 for (uint32_t code = StartCode; code <= EndCode; code++) { | |
| 386 m_pCMap->m_pMapping[code] = (uint16_t)(StartCID + code - StartCode); | |
| 387 } | |
| 388 } else { | |
| 389 uint32_t buf[2]; | |
| 390 buf[0] = StartCode; | |
| 391 buf[1] = ((EndCode - StartCode) << 16) + StartCID; | |
| 392 m_AddMaps.AppendBlock(buf, sizeof buf); | |
| 393 } | |
| 394 m_CodeSeq = 0; | |
| 395 } else if (m_Status == 3) { | |
| 396 m_Status = 0; | |
| 397 } else if (m_Status == 4) { | |
| 398 m_pCMap->m_Charset = CharsetFromOrdering(CMap_GetString(word)); | |
| 399 m_Status = 0; | |
| 400 } else if (m_Status == 5) { | |
| 401 m_Status = 0; | |
| 402 } else if (m_Status == 6) { | |
| 403 m_pCMap->m_bVertical = CMap_GetCode(word) != 0; | |
| 404 m_Status = 0; | |
| 405 } else if (m_Status == 7) { | |
| 406 if (word == "endcodespacerange") { | |
| 407 int nSegs = m_CodeRanges.GetSize(); | |
| 408 if (nSegs > 1) { | |
| 409 m_pCMap->m_CodingScheme = CPDF_CMap::MixedFourBytes; | |
| 410 m_pCMap->m_nCodeRanges = nSegs; | |
| 411 FX_Free(m_pCMap->m_pLeadingBytes); | |
| 412 m_pCMap->m_pLeadingBytes = | |
| 413 FX_Alloc2D(uint8_t, nSegs, sizeof(CMap_CodeRange)); | |
| 414 FXSYS_memcpy(m_pCMap->m_pLeadingBytes, m_CodeRanges.GetData(), | |
| 415 nSegs * sizeof(CMap_CodeRange)); | |
| 416 } else if (nSegs == 1) { | |
| 417 m_pCMap->m_CodingScheme = (m_CodeRanges[0].m_CharSize == 2) | |
| 418 ? CPDF_CMap::TwoBytes | |
| 419 : CPDF_CMap::OneByte; | |
| 420 } | |
| 421 m_Status = 0; | |
| 422 } else { | |
| 423 if (word.GetLength() == 0 || word.GetAt(0) != '<') { | |
| 424 return; | |
| 425 } | |
| 426 if (m_CodeSeq % 2) { | |
| 427 CMap_CodeRange range; | |
| 428 if (CMap_GetCodeRange(range, m_LastWord.AsStringC(), word)) { | |
| 429 m_CodeRanges.Add(range); | |
| 430 } | |
| 431 } | |
| 432 m_CodeSeq++; | |
| 433 } | |
| 434 } | |
| 435 m_LastWord = word; | |
| 436 } | |
| 437 | |
| 438 // Static. | |
| 439 uint32_t CPDF_CMapParser::CMap_GetCode(const CFX_ByteStringC& word) { | |
| 440 pdfium::base::CheckedNumeric<uint32_t> num = 0; | |
| 441 if (word.GetAt(0) == '<') { | |
| 442 for (int i = 1; i < word.GetLength() && std::isxdigit(word.GetAt(i)); ++i) { | |
| 443 num = num * 16 + FXSYS_toHexDigit(word.GetAt(i)); | |
| 444 if (!num.IsValid()) | |
| 445 return 0; | |
| 446 } | |
| 447 return num.ValueOrDie(); | |
| 448 } | |
| 449 | |
| 450 for (int i = 0; i < word.GetLength() && std::isdigit(word.GetAt(i)); ++i) { | |
| 451 num = num * 10 + FXSYS_toDecimalDigit(static_cast<FX_WCHAR>(word.GetAt(i))); | |
| 452 if (!num.IsValid()) | |
| 453 return 0; | |
| 454 } | |
| 455 return num.ValueOrDie(); | |
| 456 } | |
| 457 | |
| 458 // Static. | |
| 459 bool CPDF_CMapParser::CMap_GetCodeRange(CMap_CodeRange& range, | |
| 460 const CFX_ByteStringC& first, | |
| 461 const CFX_ByteStringC& second) { | |
| 462 if (first.GetLength() == 0 || first.GetAt(0) != '<') | |
| 463 return false; | |
| 464 | |
| 465 int i; | |
| 466 for (i = 1; i < first.GetLength(); ++i) { | |
| 467 if (first.GetAt(i) == '>') { | |
| 468 break; | |
| 469 } | |
| 470 } | |
| 471 range.m_CharSize = (i - 1) / 2; | |
| 472 if (range.m_CharSize > 4) | |
| 473 return false; | |
| 474 | |
| 475 for (i = 0; i < range.m_CharSize; ++i) { | |
| 476 uint8_t digit1 = first.GetAt(i * 2 + 1); | |
| 477 uint8_t digit2 = first.GetAt(i * 2 + 2); | |
| 478 range.m_Lower[i] = FXSYS_toHexDigit(digit1) * 16 + FXSYS_toHexDigit(digit2); | |
| 479 } | |
| 480 | |
| 481 uint32_t size = second.GetLength(); | |
| 482 for (i = 0; i < range.m_CharSize; ++i) { | |
| 483 uint8_t digit1 = ((uint32_t)i * 2 + 1 < size) | |
| 484 ? second.GetAt((FX_STRSIZE)i * 2 + 1) | |
| 485 : '0'; | |
| 486 uint8_t digit2 = ((uint32_t)i * 2 + 2 < size) | |
| 487 ? second.GetAt((FX_STRSIZE)i * 2 + 2) | |
| 488 : '0'; | |
| 489 range.m_Upper[i] = FXSYS_toHexDigit(digit1) * 16 + FXSYS_toHexDigit(digit2); | |
| 490 } | |
| 491 return true; | |
| 492 } | |
| 493 | |
| 494 CPDF_CMap::CPDF_CMap() { | |
| 495 m_Charset = CIDSET_UNKNOWN; | |
| 496 m_Coding = CIDCODING_UNKNOWN; | |
| 497 m_CodingScheme = TwoBytes; | |
| 498 m_bVertical = false; | |
| 499 m_bLoaded = false; | |
| 500 m_pMapping = nullptr; | |
| 501 m_pLeadingBytes = nullptr; | |
| 502 m_pAddMapping = nullptr; | |
| 503 m_pEmbedMap = nullptr; | |
| 504 m_nCodeRanges = 0; | |
| 505 } | |
| 506 CPDF_CMap::~CPDF_CMap() { | |
| 507 FX_Free(m_pMapping); | |
| 508 FX_Free(m_pAddMapping); | |
| 509 FX_Free(m_pLeadingBytes); | |
| 510 } | |
| 511 | |
| 512 bool CPDF_CMap::IsLoaded() const { | |
| 513 return m_bLoaded; | |
| 514 } | |
| 515 | |
| 516 bool CPDF_CMap::IsVertWriting() const { | |
| 517 return m_bVertical; | |
| 518 } | |
| 519 | |
| 520 void CPDF_CMap::LoadPredefined(CPDF_CMapManager* pMgr, | |
| 521 const CFX_ByteString& bsName, | |
| 522 bool bPromptCJK) { | |
| 523 m_PredefinedCMap = bsName; | |
| 524 if (m_PredefinedCMap == "Identity-H" || m_PredefinedCMap == "Identity-V") { | |
| 525 m_Coding = CIDCODING_CID; | |
| 526 m_bVertical = bsName[9] == 'V'; | |
| 527 m_bLoaded = true; | |
| 528 return; | |
| 529 } | |
| 530 CFX_ByteString cmapid = m_PredefinedCMap; | |
| 531 m_bVertical = cmapid.Right(1) == "V"; | |
| 532 if (cmapid.GetLength() > 2) { | |
| 533 cmapid = cmapid.Left(cmapid.GetLength() - 2); | |
| 534 } | |
| 535 const CPDF_PredefinedCMap* map = nullptr; | |
| 536 for (size_t i = 0; i < FX_ArraySize(g_PredefinedCMaps); ++i) { | |
| 537 if (cmapid == CFX_ByteStringC(g_PredefinedCMaps[i].m_pName)) { | |
| 538 map = &g_PredefinedCMaps[i]; | |
| 539 break; | |
| 540 } | |
| 541 } | |
| 542 if (!map) | |
| 543 return; | |
| 544 | |
| 545 m_Charset = map->m_Charset; | |
| 546 m_Coding = map->m_Coding; | |
| 547 m_CodingScheme = map->m_CodingScheme; | |
| 548 if (m_CodingScheme == MixedTwoBytes) { | |
| 549 m_pLeadingBytes = FX_Alloc(uint8_t, 256); | |
| 550 for (uint32_t i = 0; i < map->m_LeadingSegCount; ++i) { | |
| 551 const uint8_t* segs = map->m_LeadingSegs; | |
| 552 for (int b = segs[i * 2]; b <= segs[i * 2 + 1]; ++b) { | |
| 553 m_pLeadingBytes[b] = 1; | |
| 554 } | |
| 555 } | |
| 556 } | |
| 557 FPDFAPI_FindEmbeddedCMap(bsName, m_Charset, m_Coding, m_pEmbedMap); | |
| 558 if (!m_pEmbedMap) | |
| 559 return; | |
| 560 | |
| 561 m_bLoaded = true; | |
| 562 } | |
| 563 | |
| 564 void CPDF_CMap::LoadEmbedded(const uint8_t* pData, uint32_t size) { | |
| 565 m_pMapping = FX_Alloc(uint16_t, 65536); | |
| 566 CPDF_CMapParser parser; | |
| 567 parser.Initialize(this); | |
| 568 CPDF_SimpleParser syntax(pData, size); | |
| 569 while (1) { | |
| 570 CFX_ByteStringC word = syntax.GetWord(); | |
| 571 if (word.IsEmpty()) { | |
| 572 break; | |
| 573 } | |
| 574 parser.ParseWord(word); | |
| 575 } | |
| 576 if (m_CodingScheme == MixedFourBytes && parser.m_AddMaps.GetSize()) { | |
| 577 m_pAddMapping = FX_Alloc(uint8_t, parser.m_AddMaps.GetSize() + 4); | |
| 578 *(uint32_t*)m_pAddMapping = parser.m_AddMaps.GetSize() / 8; | |
| 579 FXSYS_memcpy(m_pAddMapping + 4, parser.m_AddMaps.GetBuffer(), | |
| 580 parser.m_AddMaps.GetSize()); | |
| 581 FXSYS_qsort(m_pAddMapping + 4, parser.m_AddMaps.GetSize() / 8, 8, | |
| 582 CompareDWORD); | |
| 583 } | |
| 584 } | |
| 585 | |
| 586 uint16_t CPDF_CMap::CIDFromCharCode(uint32_t charcode) const { | |
| 587 if (m_Coding == CIDCODING_CID) { | |
| 588 return (uint16_t)charcode; | |
| 589 } | |
| 590 if (m_pEmbedMap) { | |
| 591 return FPDFAPI_CIDFromCharCode(m_pEmbedMap, charcode); | |
| 592 } | |
| 593 if (!m_pMapping) { | |
| 594 return (uint16_t)charcode; | |
| 595 } | |
| 596 if (charcode >> 16) { | |
| 597 if (m_pAddMapping) { | |
| 598 void* found = FXSYS_bsearch(&charcode, m_pAddMapping + 4, | |
| 599 *(uint32_t*)m_pAddMapping, 8, CompareCID); | |
| 600 if (!found) | |
| 601 return 0; | |
| 602 return (uint16_t)(((uint32_t*)found)[1] % 65536 + charcode - | |
| 603 *(uint32_t*)found); | |
| 604 } | |
| 605 return 0; | |
| 606 } | |
| 607 return m_pMapping[charcode]; | |
| 608 } | |
| 609 | |
| 610 uint32_t CPDF_CMap::GetNextChar(const FX_CHAR* pString, | |
| 611 int nStrLen, | |
| 612 int& offset) const { | |
| 613 switch (m_CodingScheme) { | |
| 614 case OneByte: | |
| 615 return ((uint8_t*)pString)[offset++]; | |
| 616 case TwoBytes: | |
| 617 offset += 2; | |
| 618 return ((uint8_t*)pString)[offset - 2] * 256 + | |
| 619 ((uint8_t*)pString)[offset - 1]; | |
| 620 case MixedTwoBytes: { | |
| 621 uint8_t byte1 = ((uint8_t*)pString)[offset++]; | |
| 622 if (!m_pLeadingBytes[byte1]) { | |
| 623 return byte1; | |
| 624 } | |
| 625 uint8_t byte2 = ((uint8_t*)pString)[offset++]; | |
| 626 return byte1 * 256 + byte2; | |
| 627 } | |
| 628 case MixedFourBytes: { | |
| 629 uint8_t codes[4]; | |
| 630 int char_size = 1; | |
| 631 codes[0] = ((uint8_t*)pString)[offset++]; | |
| 632 CMap_CodeRange* pRanges = (CMap_CodeRange*)m_pLeadingBytes; | |
| 633 while (1) { | |
| 634 int ret = CheckCodeRange(codes, char_size, pRanges, m_nCodeRanges); | |
| 635 if (ret == 0) { | |
| 636 return 0; | |
| 637 } | |
| 638 if (ret == 2) { | |
| 639 uint32_t charcode = 0; | |
| 640 for (int i = 0; i < char_size; i++) { | |
| 641 charcode = (charcode << 8) + codes[i]; | |
| 642 } | |
| 643 return charcode; | |
| 644 } | |
| 645 if (char_size == 4 || offset == nStrLen) { | |
| 646 return 0; | |
| 647 } | |
| 648 codes[char_size++] = ((uint8_t*)pString)[offset++]; | |
| 649 } | |
| 650 break; | |
| 651 } | |
| 652 } | |
| 653 return 0; | |
| 654 } | |
| 655 int CPDF_CMap::GetCharSize(uint32_t charcode) const { | |
| 656 switch (m_CodingScheme) { | |
| 657 case OneByte: | |
| 658 return 1; | |
| 659 case TwoBytes: | |
| 660 return 2; | |
| 661 case MixedTwoBytes: | |
| 662 case MixedFourBytes: | |
| 663 if (charcode < 0x100) { | |
| 664 return 1; | |
| 665 } | |
| 666 if (charcode < 0x10000) { | |
| 667 return 2; | |
| 668 } | |
| 669 if (charcode < 0x1000000) { | |
| 670 return 3; | |
| 671 } | |
| 672 return 4; | |
| 673 } | |
| 674 return 1; | |
| 675 } | |
| 676 int CPDF_CMap::CountChar(const FX_CHAR* pString, int size) const { | |
| 677 switch (m_CodingScheme) { | |
| 678 case OneByte: | |
| 679 return size; | |
| 680 case TwoBytes: | |
| 681 return (size + 1) / 2; | |
| 682 case MixedTwoBytes: { | |
| 683 int count = 0; | |
| 684 for (int i = 0; i < size; i++) { | |
| 685 count++; | |
| 686 if (m_pLeadingBytes[((uint8_t*)pString)[i]]) { | |
| 687 i++; | |
| 688 } | |
| 689 } | |
| 690 return count; | |
| 691 } | |
| 692 case MixedFourBytes: { | |
| 693 int count = 0, offset = 0; | |
| 694 while (offset < size) { | |
| 695 GetNextChar(pString, size, offset); | |
| 696 count++; | |
| 697 } | |
| 698 return count; | |
| 699 } | |
| 700 } | |
| 701 return size; | |
| 702 } | |
| 703 | |
| 704 int CPDF_CMap::AppendChar(FX_CHAR* str, uint32_t charcode) const { | |
| 705 switch (m_CodingScheme) { | |
| 706 case OneByte: | |
| 707 str[0] = (uint8_t)charcode; | |
| 708 return 1; | |
| 709 case TwoBytes: | |
| 710 str[0] = (uint8_t)(charcode / 256); | |
| 711 str[1] = (uint8_t)(charcode % 256); | |
| 712 return 2; | |
| 713 case MixedTwoBytes: | |
| 714 case MixedFourBytes: | |
| 715 if (charcode < 0x100) { | |
| 716 CMap_CodeRange* pRanges = (CMap_CodeRange*)m_pLeadingBytes; | |
| 717 int iSize = GetCharSizeImpl(charcode, pRanges, m_nCodeRanges); | |
| 718 if (iSize == 0) { | |
| 719 iSize = 1; | |
| 720 } | |
| 721 if (iSize > 1) { | |
| 722 FXSYS_memset(str, 0, sizeof(uint8_t) * iSize); | |
| 723 } | |
| 724 str[iSize - 1] = (uint8_t)charcode; | |
| 725 return iSize; | |
| 726 } | |
| 727 if (charcode < 0x10000) { | |
| 728 str[0] = (uint8_t)(charcode >> 8); | |
| 729 str[1] = (uint8_t)charcode; | |
| 730 return 2; | |
| 731 } | |
| 732 if (charcode < 0x1000000) { | |
| 733 str[0] = (uint8_t)(charcode >> 16); | |
| 734 str[1] = (uint8_t)(charcode >> 8); | |
| 735 str[2] = (uint8_t)charcode; | |
| 736 return 3; | |
| 737 } | |
| 738 str[0] = (uint8_t)(charcode >> 24); | |
| 739 str[1] = (uint8_t)(charcode >> 16); | |
| 740 str[2] = (uint8_t)(charcode >> 8); | |
| 741 str[3] = (uint8_t)charcode; | |
| 742 return 4; | |
| 743 } | |
| 744 return 0; | |
| 745 } | |
| 746 | |
| 747 CPDF_CID2UnicodeMap::CPDF_CID2UnicodeMap() { | |
| 748 m_EmbeddedCount = 0; | |
| 749 } | |
| 750 | |
| 751 CPDF_CID2UnicodeMap::~CPDF_CID2UnicodeMap() {} | |
| 752 | |
| 753 bool CPDF_CID2UnicodeMap::IsLoaded() { | |
| 754 return m_EmbeddedCount != 0; | |
| 755 } | |
| 756 | |
| 757 FX_WCHAR CPDF_CID2UnicodeMap::UnicodeFromCID(uint16_t CID) { | |
| 758 if (m_Charset == CIDSET_UNICODE) { | |
| 759 return CID; | |
| 760 } | |
| 761 if (CID < m_EmbeddedCount) { | |
| 762 return m_pEmbeddedMap[CID]; | |
| 763 } | |
| 764 return 0; | |
| 765 } | |
| 766 | |
| 767 void CPDF_CID2UnicodeMap::Load(CPDF_CMapManager* pMgr, | |
| 768 CIDSet charset, | |
| 769 bool bPromptCJK) { | |
| 770 m_Charset = charset; | |
| 771 | |
| 772 CPDF_FontGlobals* pFontGlobals = | |
| 773 CPDF_ModuleMgr::Get()->GetPageModule()->GetFontGlobals(); | |
| 774 m_pEmbeddedMap = pFontGlobals->m_EmbeddedToUnicodes[charset].m_pMap; | |
| 775 m_EmbeddedCount = pFontGlobals->m_EmbeddedToUnicodes[charset].m_Count; | |
| 776 } | |
| 777 | |
| 778 CIDSet CharsetFromOrdering(const CFX_ByteStringC& ordering) { | |
| 779 for (size_t charset = 1; charset < FX_ArraySize(g_CharsetNames); ++charset) { | |
| 780 if (ordering == g_CharsetNames[charset]) | |
| 781 return CIDSetFromSizeT(charset); | |
| 782 } | |
| 783 return CIDSET_UNKNOWN; | |
| 784 } | |
| OLD | NEW |
|---|
Chromium Code Reviews
Issue 2392773003:
Move core/fpdfapi/fpdf_font to core/fpdfapi/font (Closed)
