| 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/src/fpdfapi/fpdf_font/font_int.h" | |
| 8 | |
| 9 #include "core/include/fpdfapi/cpdf_array.h" | |
| 10 #include "core/include/fpdfapi/cpdf_dictionary.h" | |
| 11 #include "core/include/fpdfapi/cpdf_simple_parser.h" | |
| 12 #include "core/include/fpdfapi/fpdf_module.h" | |
| 13 #include "core/include/fpdfapi/fpdf_page.h" | |
| 14 #include "core/include/fpdfapi/fpdf_resource.h" | |
| 15 #include "core/include/fxcrt/fx_ext.h" | |
| 16 #include "core/include/fxge/fx_freetype.h" | |
| 17 #include "core/include/fxge/fx_ge.h" | |
| 18 #include "core/src/fpdfapi/fpdf_cmaps/cmap_int.h" | |
| 19 #include "core/src/fpdfapi/fpdf_font/ttgsubtable.h" | |
| 20 | |
| 21 namespace { | |
| 22 | |
| 23 const FX_CHAR* const g_CharsetNames[CIDSET_NUM_SETS] = | |
| 24 {nullptr, "GB1", "CNS1", "Japan1", "Korea1", "UCS"}; | |
| 25 | |
| 26 const int g_CharsetCPs[CIDSET_NUM_SETS] = {0, 936, 950, 932, 949, 1200}; | |
| 27 | |
| 28 class CPDF_PredefinedCMap { | |
| 29 public: | |
| 30 const FX_CHAR* m_pName; | |
| 31 CIDSet m_Charset; | |
| 32 int m_Coding; | |
| 33 CPDF_CMap::CodingScheme m_CodingScheme; | |
| 34 FX_DWORD m_LeadingSegCount; | |
| 35 uint8_t m_LeadingSegs[4]; | |
| 36 }; | |
| 37 | |
| 38 const CPDF_PredefinedCMap g_PredefinedCMaps[] = { | |
| 39 {"GB-EUC", | |
| 40 CIDSET_GB1, | |
| 41 CIDCODING_GB, | |
| 42 CPDF_CMap::MixedTwoBytes, | |
| 43 1, | |
| 44 {0xa1, 0xfe}}, | |
| 45 {"GBpc-EUC", | |
| 46 CIDSET_GB1, | |
| 47 CIDCODING_GB, | |
| 48 CPDF_CMap::MixedTwoBytes, | |
| 49 1, | |
| 50 {0xa1, 0xfc}}, | |
| 51 {"GBK-EUC", | |
| 52 CIDSET_GB1, | |
| 53 CIDCODING_GB, | |
| 54 CPDF_CMap::MixedTwoBytes, | |
| 55 1, | |
| 56 {0x81, 0xfe}}, | |
| 57 {"GBKp-EUC", | |
| 58 CIDSET_GB1, | |
| 59 CIDCODING_GB, | |
| 60 CPDF_CMap::MixedTwoBytes, | |
| 61 1, | |
| 62 {0x81, 0xfe}}, | |
| 63 {"GBK2K-EUC", | |
| 64 CIDSET_GB1, | |
| 65 CIDCODING_GB, | |
| 66 CPDF_CMap::MixedTwoBytes, | |
| 67 1, | |
| 68 {0x81, 0xfe}}, | |
| 69 {"GBK2K", | |
| 70 CIDSET_GB1, | |
| 71 CIDCODING_GB, | |
| 72 CPDF_CMap::MixedTwoBytes, | |
| 73 1, | |
| 74 {0x81, 0xfe}}, | |
| 75 {"UniGB-UCS2", CIDSET_GB1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 76 {"UniGB-UTF16", CIDSET_GB1, CIDCODING_UTF16, CPDF_CMap::TwoBytes, 0, {}}, | |
| 77 {"B5pc", | |
| 78 CIDSET_CNS1, | |
| 79 CIDCODING_BIG5, | |
| 80 CPDF_CMap::MixedTwoBytes, | |
| 81 1, | |
| 82 {0xa1, 0xfc}}, | |
| 83 {"HKscs-B5", | |
| 84 CIDSET_CNS1, | |
| 85 CIDCODING_BIG5, | |
| 86 CPDF_CMap::MixedTwoBytes, | |
| 87 1, | |
| 88 {0x88, 0xfe}}, | |
| 89 {"ETen-B5", | |
| 90 CIDSET_CNS1, | |
| 91 CIDCODING_BIG5, | |
| 92 CPDF_CMap::MixedTwoBytes, | |
| 93 1, | |
| 94 {0xa1, 0xfe}}, | |
| 95 {"ETenms-B5", | |
| 96 CIDSET_CNS1, | |
| 97 CIDCODING_BIG5, | |
| 98 CPDF_CMap::MixedTwoBytes, | |
| 99 1, | |
| 100 {0xa1, 0xfe}}, | |
| 101 {"UniCNS-UCS2", CIDSET_CNS1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 102 {"UniCNS-UTF16", CIDSET_CNS1, CIDCODING_UTF16, CPDF_CMap::TwoBytes, 0, {}}, | |
| 103 {"83pv-RKSJ", | |
| 104 CIDSET_JAPAN1, | |
| 105 CIDCODING_JIS, | |
| 106 CPDF_CMap::MixedTwoBytes, | |
| 107 2, | |
| 108 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 109 {"90ms-RKSJ", | |
| 110 CIDSET_JAPAN1, | |
| 111 CIDCODING_JIS, | |
| 112 CPDF_CMap::MixedTwoBytes, | |
| 113 2, | |
| 114 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 115 {"90msp-RKSJ", | |
| 116 CIDSET_JAPAN1, | |
| 117 CIDCODING_JIS, | |
| 118 CPDF_CMap::MixedTwoBytes, | |
| 119 2, | |
| 120 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 121 {"90pv-RKSJ", | |
| 122 CIDSET_JAPAN1, | |
| 123 CIDCODING_JIS, | |
| 124 CPDF_CMap::MixedTwoBytes, | |
| 125 2, | |
| 126 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 127 {"Add-RKSJ", | |
| 128 CIDSET_JAPAN1, | |
| 129 CIDCODING_JIS, | |
| 130 CPDF_CMap::MixedTwoBytes, | |
| 131 2, | |
| 132 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 133 {"EUC", | |
| 134 CIDSET_JAPAN1, | |
| 135 CIDCODING_JIS, | |
| 136 CPDF_CMap::MixedTwoBytes, | |
| 137 2, | |
| 138 {0x8e, 0x8e, 0xa1, 0xfe}}, | |
| 139 {"H", CIDSET_JAPAN1, CIDCODING_JIS, CPDF_CMap::TwoBytes, 1, {0x21, 0x7e}}, | |
| 140 {"V", CIDSET_JAPAN1, CIDCODING_JIS, CPDF_CMap::TwoBytes, 1, {0x21, 0x7e}}, | |
| 141 {"Ext-RKSJ", | |
| 142 CIDSET_JAPAN1, | |
| 143 CIDCODING_JIS, | |
| 144 CPDF_CMap::MixedTwoBytes, | |
| 145 2, | |
| 146 {0x81, 0x9f, 0xe0, 0xfc}}, | |
| 147 {"UniJIS-UCS2", CIDSET_JAPAN1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 148 {"UniJIS-UCS2-HW", | |
| 149 CIDSET_JAPAN1, | |
| 150 CIDCODING_UCS2, | |
| 151 CPDF_CMap::TwoBytes, | |
| 152 0, | |
| 153 {}}, | |
| 154 {"UniJIS-UTF16", | |
| 155 CIDSET_JAPAN1, | |
| 156 CIDCODING_UTF16, | |
| 157 CPDF_CMap::TwoBytes, | |
| 158 0, | |
| 159 {}}, | |
| 160 {"KSC-EUC", | |
| 161 CIDSET_KOREA1, | |
| 162 CIDCODING_KOREA, | |
| 163 CPDF_CMap::MixedTwoBytes, | |
| 164 1, | |
| 165 {0xa1, 0xfe}}, | |
| 166 {"KSCms-UHC", | |
| 167 CIDSET_KOREA1, | |
| 168 CIDCODING_KOREA, | |
| 169 CPDF_CMap::MixedTwoBytes, | |
| 170 1, | |
| 171 {0x81, 0xfe}}, | |
| 172 {"KSCms-UHC-HW", | |
| 173 CIDSET_KOREA1, | |
| 174 CIDCODING_KOREA, | |
| 175 CPDF_CMap::MixedTwoBytes, | |
| 176 1, | |
| 177 {0x81, 0xfe}}, | |
| 178 {"KSCpc-EUC", | |
| 179 CIDSET_KOREA1, | |
| 180 CIDCODING_KOREA, | |
| 181 CPDF_CMap::MixedTwoBytes, | |
| 182 1, | |
| 183 {0xa1, 0xfd}}, | |
| 184 {"UniKS-UCS2", CIDSET_KOREA1, CIDCODING_UCS2, CPDF_CMap::TwoBytes, 0, {}}, | |
| 185 {"UniKS-UTF16", CIDSET_KOREA1, CIDCODING_UTF16, CPDF_CMap::TwoBytes, 0, {}}, | |
| 186 }; | |
| 187 | |
| 188 CIDSet CIDSetFromSizeT(size_t index) { | |
| 189 if (index >= CIDSET_NUM_SETS) { | |
| 190 NOTREACHED(); | |
| 191 return CIDSET_UNKNOWN; | |
| 192 } | |
| 193 return static_cast<CIDSet>(index); | |
| 194 } | |
| 195 | |
| 196 CIDSet CharsetFromOrdering(const CFX_ByteString& ordering) { | |
| 197 for (size_t charset = 1; charset < FX_ArraySize(g_CharsetNames); ++charset) { | |
| 198 if (ordering == CFX_ByteStringC(g_CharsetNames[charset])) | |
| 199 return CIDSetFromSizeT(charset); | |
| 200 } | |
| 201 return CIDSET_UNKNOWN; | |
| 202 } | |
| 203 | |
| 204 CFX_ByteString CMap_GetString(const CFX_ByteStringC& word) { | |
| 205 return word.Mid(1, word.GetLength() - 2); | |
| 206 } | |
| 207 | |
| 208 int CompareDWORD(const void* data1, const void* data2) { | |
| 209 return (*(FX_DWORD*)data1) - (*(FX_DWORD*)data2); | |
| 210 } | |
| 211 | |
| 212 int CompareCID(const void* key, const void* element) { | |
| 213 if ((*(FX_DWORD*)key) < (*(FX_DWORD*)element)) { | |
| 214 return -1; | |
| 215 } | |
| 216 if ((*(FX_DWORD*)key) > | |
| 217 (*(FX_DWORD*)element) + ((FX_DWORD*)element)[1] / 65536) { | |
| 218 return 1; | |
| 219 } | |
| 220 return 0; | |
| 221 } | |
| 222 | |
| 223 int CheckCodeRange(uint8_t* codes, | |
| 224 int size, | |
| 225 CMap_CodeRange* pRanges, | |
| 226 int nRanges) { | |
| 227 int iSeg = nRanges - 1; | |
| 228 while (iSeg >= 0) { | |
| 229 if (pRanges[iSeg].m_CharSize < size) { | |
| 230 --iSeg; | |
| 231 continue; | |
| 232 } | |
| 233 int iChar = 0; | |
| 234 while (iChar < size) { | |
| 235 if (codes[iChar] < pRanges[iSeg].m_Lower[iChar] || | |
| 236 codes[iChar] > pRanges[iSeg].m_Upper[iChar]) { | |
| 237 break; | |
| 238 } | |
| 239 ++iChar; | |
| 240 } | |
| 241 if (iChar == pRanges[iSeg].m_CharSize) | |
| 242 return 2; | |
| 243 | |
| 244 if (iChar) | |
| 245 return (size == pRanges[iSeg].m_CharSize) ? 2 : 1; | |
| 246 iSeg--; | |
| 247 } | |
| 248 return 0; | |
| 249 } | |
| 250 | |
| 251 int GetCharSizeImpl(FX_DWORD charcode, | |
| 252 CMap_CodeRange* pRanges, | |
| 253 int iRangesSize) { | |
| 254 if (!iRangesSize) | |
| 255 return 1; | |
| 256 | |
| 257 uint8_t codes[4]; | |
| 258 codes[0] = codes[1] = 0x00; | |
| 259 codes[2] = (uint8_t)(charcode >> 8 & 0xFF); | |
| 260 codes[3] = (uint8_t)charcode; | |
| 261 int offset = 0; | |
| 262 int size = 4; | |
| 263 for (int i = 0; i < 4; ++i) { | |
| 264 int iSeg = iRangesSize - 1; | |
| 265 while (iSeg >= 0) { | |
| 266 if (pRanges[iSeg].m_CharSize < size) { | |
| 267 --iSeg; | |
| 268 continue; | |
| 269 } | |
| 270 int iChar = 0; | |
| 271 while (iChar < size) { | |
| 272 if (codes[offset + iChar] < pRanges[iSeg].m_Lower[iChar] || | |
| 273 codes[offset + iChar] > pRanges[iSeg].m_Upper[iChar]) { | |
| 274 break; | |
| 275 } | |
| 276 ++iChar; | |
| 277 } | |
| 278 if (iChar == pRanges[iSeg].m_CharSize) | |
| 279 return size; | |
| 280 --iSeg; | |
| 281 } | |
| 282 --size; | |
| 283 ++offset; | |
| 284 } | |
| 285 return 1; | |
| 286 } | |
| 287 | |
| 288 bool IsValidEmbeddedCharcodeFromUnicodeCharset(CIDSet charset) { | |
| 289 switch (charset) { | |
| 290 case CIDSET_GB1: | |
| 291 case CIDSET_CNS1: | |
| 292 case CIDSET_JAPAN1: | |
| 293 case CIDSET_KOREA1: | |
| 294 return true; | |
| 295 | |
| 296 default: | |
| 297 return false; | |
| 298 } | |
| 299 } | |
| 300 | |
| 301 #if _FXM_PLATFORM_ != _FXM_PLATFORM_WINDOWS_ | |
| 302 FX_DWORD EmbeddedCharcodeFromUnicode(const FXCMAP_CMap* pEmbedMap, | |
| 303 CIDSet charset, | |
| 304 FX_WCHAR unicode) { | |
| 305 if (!IsValidEmbeddedCharcodeFromUnicodeCharset(charset)) | |
| 306 return 0; | |
| 307 | |
| 308 CPDF_FontGlobals* pFontGlobals = | |
| 309 CPDF_ModuleMgr::Get()->GetPageModule()->GetFontGlobals(); | |
| 310 const FX_WORD* pCodes = pFontGlobals->m_EmbeddedToUnicodes[charset].m_pMap; | |
| 311 if (!pCodes) | |
| 312 return 0; | |
| 313 | |
| 314 int nCodes = pFontGlobals->m_EmbeddedToUnicodes[charset].m_Count; | |
| 315 for (int i = 0; i < nCodes; ++i) { | |
| 316 if (pCodes[i] == unicode) { | |
| 317 FX_DWORD CharCode = FPDFAPI_CharCodeFromCID(pEmbedMap, i); | |
| 318 if (CharCode != 0) { | |
| 319 return CharCode; | |
| 320 } | |
| 321 } | |
| 322 } | |
| 323 return 0; | |
| 324 } | |
| 325 #endif // _FXM_PLATFORM_ != _FXM_PLATFORM_WINDOWS_ | |
| 326 | |
| 327 FX_WCHAR EmbeddedUnicodeFromCharcode(const FXCMAP_CMap* pEmbedMap, | |
| 328 CIDSet charset, | |
| 329 FX_DWORD charcode) { | |
| 330 if (!IsValidEmbeddedCharcodeFromUnicodeCharset(charset)) | |
| 331 return 0; | |
| 332 | |
| 333 FX_WORD cid = FPDFAPI_CIDFromCharCode(pEmbedMap, charcode); | |
| 334 if (cid == 0) | |
| 335 return 0; | |
| 336 | |
| 337 CPDF_FontGlobals* pFontGlobals = | |
| 338 CPDF_ModuleMgr::Get()->GetPageModule()->GetFontGlobals(); | |
| 339 const FX_WORD* pCodes = pFontGlobals->m_EmbeddedToUnicodes[charset].m_pMap; | |
| 340 if (!pCodes) | |
| 341 return 0; | |
| 342 | |
| 343 if (cid < pFontGlobals->m_EmbeddedToUnicodes[charset].m_Count) | |
| 344 return pCodes[cid]; | |
| 345 return 0; | |
| 346 } | |
| 347 | |
| 348 void FT_UseCIDCharmap(FXFT_Face face, int coding) { | |
| 349 int encoding; | |
| 350 switch (coding) { | |
| 351 case CIDCODING_GB: | |
| 352 encoding = FXFT_ENCODING_GB2312; | |
| 353 break; | |
| 354 case CIDCODING_BIG5: | |
| 355 encoding = FXFT_ENCODING_BIG5; | |
| 356 break; | |
| 357 case CIDCODING_JIS: | |
| 358 encoding = FXFT_ENCODING_SJIS; | |
| 359 break; | |
| 360 case CIDCODING_KOREA: | |
| 361 encoding = FXFT_ENCODING_JOHAB; | |
| 362 break; | |
| 363 default: | |
| 364 encoding = FXFT_ENCODING_UNICODE; | |
| 365 } | |
| 366 int err = FXFT_Select_Charmap(face, encoding); | |
| 367 if (err) { | |
| 368 err = FXFT_Select_Charmap(face, FXFT_ENCODING_UNICODE); | |
| 369 } | |
| 370 if (err && FXFT_Get_Face_Charmaps(face)) { | |
| 371 FXFT_Set_Charmap(face, *FXFT_Get_Face_Charmaps(face)); | |
| 372 } | |
| 373 } | |
| 374 | |
| 375 const struct CIDTransform { | |
| 376 FX_WORD CID; | |
| 377 uint8_t a, b, c, d, e, f; | |
| 378 } g_Japan1_VertCIDs[] = { | |
| 379 {97, 129, 0, 0, 127, 55, 0}, | |
| 380 {7887, 127, 0, 0, 127, 76, 89}, | |
| 381 {7888, 127, 0, 0, 127, 79, 94}, | |
| 382 {7889, 0, 129, 127, 0, 17, 127}, | |
| 383 {7890, 0, 129, 127, 0, 17, 127}, | |
| 384 {7891, 0, 129, 127, 0, 17, 127}, | |
| 385 {7892, 0, 129, 127, 0, 17, 127}, | |
| 386 {7893, 0, 129, 127, 0, 17, 127}, | |
| 387 {7894, 0, 129, 127, 0, 17, 127}, | |
| 388 {7895, 0, 129, 127, 0, 17, 127}, | |
| 389 {7896, 0, 129, 127, 0, 17, 127}, | |
| 390 {7897, 0, 129, 127, 0, 17, 127}, | |
| 391 {7898, 0, 129, 127, 0, 17, 127}, | |
| 392 {7899, 0, 129, 127, 0, 17, 104}, | |
| 393 {7900, 0, 129, 127, 0, 17, 127}, | |
| 394 {7901, 0, 129, 127, 0, 17, 104}, | |
| 395 {7902, 0, 129, 127, 0, 17, 127}, | |
| 396 {7903, 0, 129, 127, 0, 17, 127}, | |
| 397 {7904, 0, 129, 127, 0, 17, 127}, | |
| 398 {7905, 0, 129, 127, 0, 17, 114}, | |
| 399 {7906, 0, 129, 127, 0, 17, 127}, | |
| 400 {7907, 0, 129, 127, 0, 17, 127}, | |
| 401 {7908, 0, 129, 127, 0, 17, 127}, | |
| 402 {7909, 0, 129, 127, 0, 17, 127}, | |
| 403 {7910, 0, 129, 127, 0, 17, 127}, | |
| 404 {7911, 0, 129, 127, 0, 17, 127}, | |
| 405 {7912, 0, 129, 127, 0, 17, 127}, | |
| 406 {7913, 0, 129, 127, 0, 17, 127}, | |
| 407 {7914, 0, 129, 127, 0, 17, 127}, | |
| 408 {7915, 0, 129, 127, 0, 17, 114}, | |
| 409 {7916, 0, 129, 127, 0, 17, 127}, | |
| 410 {7917, 0, 129, 127, 0, 17, 127}, | |
| 411 {7918, 127, 0, 0, 127, 18, 25}, | |
| 412 {7919, 127, 0, 0, 127, 18, 25}, | |
| 413 {7920, 127, 0, 0, 127, 18, 25}, | |
| 414 {7921, 127, 0, 0, 127, 18, 25}, | |
| 415 {7922, 127, 0, 0, 127, 18, 25}, | |
| 416 {7923, 127, 0, 0, 127, 18, 25}, | |
| 417 {7924, 127, 0, 0, 127, 18, 25}, | |
| 418 {7925, 127, 0, 0, 127, 18, 25}, | |
| 419 {7926, 127, 0, 0, 127, 18, 25}, | |
| 420 {7927, 127, 0, 0, 127, 18, 25}, | |
| 421 {7928, 127, 0, 0, 127, 18, 25}, | |
| 422 {7929, 127, 0, 0, 127, 18, 25}, | |
| 423 {7930, 127, 0, 0, 127, 18, 25}, | |
| 424 {7931, 127, 0, 0, 127, 18, 25}, | |
| 425 {7932, 127, 0, 0, 127, 18, 25}, | |
| 426 {7933, 127, 0, 0, 127, 18, 25}, | |
| 427 {7934, 127, 0, 0, 127, 18, 25}, | |
| 428 {7935, 127, 0, 0, 127, 18, 25}, | |
| 429 {7936, 127, 0, 0, 127, 18, 25}, | |
| 430 {7937, 127, 0, 0, 127, 18, 25}, | |
| 431 {7938, 127, 0, 0, 127, 18, 25}, | |
| 432 {7939, 127, 0, 0, 127, 18, 25}, | |
| 433 {8720, 0, 129, 127, 0, 19, 102}, | |
| 434 {8721, 0, 129, 127, 0, 13, 127}, | |
| 435 {8722, 0, 129, 127, 0, 19, 108}, | |
| 436 {8723, 0, 129, 127, 0, 19, 102}, | |
| 437 {8724, 0, 129, 127, 0, 19, 102}, | |
| 438 {8725, 0, 129, 127, 0, 19, 102}, | |
| 439 {8726, 0, 129, 127, 0, 19, 102}, | |
| 440 {8727, 0, 129, 127, 0, 19, 102}, | |
| 441 {8728, 0, 129, 127, 0, 19, 114}, | |
| 442 {8729, 0, 129, 127, 0, 19, 114}, | |
| 443 {8730, 0, 129, 127, 0, 38, 108}, | |
| 444 {8731, 0, 129, 127, 0, 13, 108}, | |
| 445 {8732, 0, 129, 127, 0, 19, 108}, | |
| 446 {8733, 0, 129, 127, 0, 19, 108}, | |
| 447 {8734, 0, 129, 127, 0, 19, 108}, | |
| 448 {8735, 0, 129, 127, 0, 19, 108}, | |
| 449 {8736, 0, 129, 127, 0, 19, 102}, | |
| 450 {8737, 0, 129, 127, 0, 19, 102}, | |
| 451 {8738, 0, 129, 127, 0, 19, 102}, | |
| 452 {8739, 0, 129, 127, 0, 19, 102}, | |
| 453 {8740, 0, 129, 127, 0, 19, 102}, | |
| 454 {8741, 0, 129, 127, 0, 19, 102}, | |
| 455 {8742, 0, 129, 127, 0, 19, 102}, | |
| 456 {8743, 0, 129, 127, 0, 19, 102}, | |
| 457 {8744, 0, 129, 127, 0, 19, 102}, | |
| 458 {8745, 0, 129, 127, 0, 19, 102}, | |
| 459 {8746, 0, 129, 127, 0, 19, 114}, | |
| 460 {8747, 0, 129, 127, 0, 19, 114}, | |
| 461 {8748, 0, 129, 127, 0, 19, 102}, | |
| 462 {8749, 0, 129, 127, 0, 19, 102}, | |
| 463 {8750, 0, 129, 127, 0, 19, 102}, | |
| 464 {8751, 0, 129, 127, 0, 19, 102}, | |
| 465 {8752, 0, 129, 127, 0, 19, 102}, | |
| 466 {8753, 0, 129, 127, 0, 19, 102}, | |
| 467 {8754, 0, 129, 127, 0, 19, 102}, | |
| 468 {8755, 0, 129, 127, 0, 19, 102}, | |
| 469 {8756, 0, 129, 127, 0, 19, 102}, | |
| 470 {8757, 0, 129, 127, 0, 19, 102}, | |
| 471 {8758, 0, 129, 127, 0, 19, 102}, | |
| 472 {8759, 0, 129, 127, 0, 19, 102}, | |
| 473 {8760, 0, 129, 127, 0, 19, 102}, | |
| 474 {8761, 0, 129, 127, 0, 19, 102}, | |
| 475 {8762, 0, 129, 127, 0, 19, 102}, | |
| 476 {8763, 0, 129, 127, 0, 19, 102}, | |
| 477 {8764, 0, 129, 127, 0, 19, 102}, | |
| 478 {8765, 0, 129, 127, 0, 19, 102}, | |
| 479 {8766, 0, 129, 127, 0, 19, 102}, | |
| 480 {8767, 0, 129, 127, 0, 19, 102}, | |
| 481 {8768, 0, 129, 127, 0, 19, 102}, | |
| 482 {8769, 0, 129, 127, 0, 19, 102}, | |
| 483 {8770, 0, 129, 127, 0, 19, 102}, | |
| 484 {8771, 0, 129, 127, 0, 19, 102}, | |
| 485 {8772, 0, 129, 127, 0, 19, 102}, | |
| 486 {8773, 0, 129, 127, 0, 19, 102}, | |
| 487 {8774, 0, 129, 127, 0, 19, 102}, | |
| 488 {8775, 0, 129, 127, 0, 19, 102}, | |
| 489 {8776, 0, 129, 127, 0, 19, 102}, | |
| 490 {8777, 0, 129, 127, 0, 19, 102}, | |
| 491 {8778, 0, 129, 127, 0, 19, 102}, | |
| 492 {8779, 0, 129, 127, 0, 19, 114}, | |
| 493 {8780, 0, 129, 127, 0, 19, 108}, | |
| 494 {8781, 0, 129, 127, 0, 19, 114}, | |
| 495 {8782, 0, 129, 127, 0, 13, 114}, | |
| 496 {8783, 0, 129, 127, 0, 19, 108}, | |
| 497 {8784, 0, 129, 127, 0, 13, 114}, | |
| 498 {8785, 0, 129, 127, 0, 19, 108}, | |
| 499 {8786, 0, 129, 127, 0, 19, 108}, | |
| 500 {8787, 0, 129, 127, 0, 19, 108}, | |
| 501 {8788, 0, 129, 127, 0, 19, 108}, | |
| 502 {8789, 0, 129, 127, 0, 19, 108}, | |
| 503 {8790, 0, 129, 127, 0, 19, 108}, | |
| 504 {8791, 0, 129, 127, 0, 19, 108}, | |
| 505 {8792, 0, 129, 127, 0, 19, 108}, | |
| 506 {8793, 0, 129, 127, 0, 19, 108}, | |
| 507 {8794, 0, 129, 127, 0, 19, 108}, | |
| 508 {8795, 0, 129, 127, 0, 19, 108}, | |
| 509 {8796, 0, 129, 127, 0, 19, 108}, | |
| 510 {8797, 0, 129, 127, 0, 19, 108}, | |
| 511 {8798, 0, 129, 127, 0, 19, 108}, | |
| 512 {8799, 0, 129, 127, 0, 19, 108}, | |
| 513 {8800, 0, 129, 127, 0, 19, 108}, | |
| 514 {8801, 0, 129, 127, 0, 19, 108}, | |
| 515 {8802, 0, 129, 127, 0, 19, 108}, | |
| 516 {8803, 0, 129, 127, 0, 19, 108}, | |
| 517 {8804, 0, 129, 127, 0, 19, 108}, | |
| 518 {8805, 0, 129, 127, 0, 19, 108}, | |
| 519 {8806, 0, 129, 127, 0, 19, 108}, | |
| 520 {8807, 0, 129, 127, 0, 19, 108}, | |
| 521 {8808, 0, 129, 127, 0, 19, 108}, | |
| 522 {8809, 0, 129, 127, 0, 19, 108}, | |
| 523 {8810, 0, 129, 127, 0, 19, 108}, | |
| 524 {8811, 0, 129, 127, 0, 19, 114}, | |
| 525 {8812, 0, 129, 127, 0, 19, 102}, | |
| 526 {8813, 0, 129, 127, 0, 19, 114}, | |
| 527 {8814, 0, 129, 127, 0, 76, 102}, | |
| 528 {8815, 0, 129, 127, 0, 13, 121}, | |
| 529 {8816, 0, 129, 127, 0, 19, 114}, | |
| 530 {8817, 0, 129, 127, 0, 19, 127}, | |
| 531 {8818, 0, 129, 127, 0, 19, 114}, | |
| 532 {8819, 0, 129, 127, 0, 218, 108}, | |
| 533 }; | |
| 534 | |
| 535 int CompareCIDTransform(const void* key, const void* element) { | |
| 536 FX_WORD CID = *static_cast<const FX_WORD*>(key); | |
| 537 return CID - static_cast<const struct CIDTransform*>(element)->CID; | |
| 538 } | |
| 539 | |
| 540 } // namespace | |
| 541 | |
| 542 CPDF_CMapManager::CPDF_CMapManager() { | |
| 543 m_bPrompted = FALSE; | |
| 544 FXSYS_memset(m_CID2UnicodeMaps, 0, sizeof m_CID2UnicodeMaps); | |
| 545 } | |
| 546 CPDF_CMapManager::~CPDF_CMapManager() { | |
| 547 for (const auto& pair : m_CMaps) { | |
| 548 delete pair.second; | |
| 549 } | |
| 550 m_CMaps.clear(); | |
| 551 for (size_t i = 0; i < FX_ArraySize(m_CID2UnicodeMaps); ++i) { | |
| 552 delete m_CID2UnicodeMaps[i]; | |
| 553 } | |
| 554 } | |
| 555 CPDF_CMap* CPDF_CMapManager::GetPredefinedCMap(const CFX_ByteString& name, | |
| 556 FX_BOOL bPromptCJK) { | |
| 557 auto it = m_CMaps.find(name); | |
| 558 if (it != m_CMaps.end()) { | |
| 559 return it->second; | |
| 560 } | |
| 561 CPDF_CMap* pCMap = LoadPredefinedCMap(name, bPromptCJK); | |
| 562 if (!name.IsEmpty()) { | |
| 563 m_CMaps[name] = pCMap; | |
| 564 } | |
| 565 return pCMap; | |
| 566 } | |
| 567 CPDF_CMap* CPDF_CMapManager::LoadPredefinedCMap(const CFX_ByteString& name, | |
| 568 FX_BOOL bPromptCJK) { | |
| 569 CPDF_CMap* pCMap = new CPDF_CMap; | |
| 570 const FX_CHAR* pname = name; | |
| 571 if (*pname == '/') { | |
| 572 pname++; | |
| 573 } | |
| 574 pCMap->LoadPredefined(this, pname, bPromptCJK); | |
| 575 return pCMap; | |
| 576 } | |
| 577 | |
| 578 void CPDF_CMapManager::ReloadAll() { | |
| 579 for (const auto& pair : m_CMaps) { | |
| 580 CPDF_CMap* pCMap = pair.second; | |
| 581 pCMap->LoadPredefined(this, pair.first, FALSE); | |
| 582 } | |
| 583 for (size_t i = 0; i < FX_ArraySize(m_CID2UnicodeMaps); ++i) { | |
| 584 if (CPDF_CID2UnicodeMap* pMap = m_CID2UnicodeMaps[i]) { | |
| 585 pMap->Load(this, CIDSetFromSizeT(i), FALSE); | |
| 586 } | |
| 587 } | |
| 588 } | |
| 589 CPDF_CID2UnicodeMap* CPDF_CMapManager::GetCID2UnicodeMap(CIDSet charset, | |
| 590 FX_BOOL bPromptCJK) { | |
| 591 if (!m_CID2UnicodeMaps[charset]) | |
| 592 m_CID2UnicodeMaps[charset] = LoadCID2UnicodeMap(charset, bPromptCJK); | |
| 593 return m_CID2UnicodeMaps[charset]; | |
| 594 } | |
| 595 CPDF_CID2UnicodeMap* CPDF_CMapManager::LoadCID2UnicodeMap(CIDSet charset, | |
| 596 FX_BOOL bPromptCJK) { | |
| 597 CPDF_CID2UnicodeMap* pMap = new CPDF_CID2UnicodeMap(); | |
| 598 if (!pMap->Initialize()) { | |
| 599 delete pMap; | |
| 600 return NULL; | |
| 601 } | |
| 602 pMap->Load(this, charset, bPromptCJK); | |
| 603 return pMap; | |
| 604 } | |
| 605 CPDF_CMapParser::CPDF_CMapParser() { | |
| 606 m_pCMap = NULL; | |
| 607 m_Status = 0; | |
| 608 m_CodeSeq = 0; | |
| 609 } | |
| 610 FX_BOOL CPDF_CMapParser::Initialize(CPDF_CMap* pCMap) { | |
| 611 m_pCMap = pCMap; | |
| 612 m_Status = 0; | |
| 613 m_CodeSeq = 0; | |
| 614 m_AddMaps.EstimateSize(0, 10240); | |
| 615 return TRUE; | |
| 616 } | |
| 617 | |
| 618 void CPDF_CMapParser::ParseWord(const CFX_ByteStringC& word) { | |
| 619 if (word.IsEmpty()) { | |
| 620 return; | |
| 621 } | |
| 622 if (word == "begincidchar") { | |
| 623 m_Status = 1; | |
| 624 m_CodeSeq = 0; | |
| 625 } else if (word == "begincidrange") { | |
| 626 m_Status = 2; | |
| 627 m_CodeSeq = 0; | |
| 628 } else if (word == "endcidrange" || word == "endcidchar") { | |
| 629 m_Status = 0; | |
| 630 } else if (word == "/WMode") { | |
| 631 m_Status = 6; | |
| 632 } else if (word == "/Registry") { | |
| 633 m_Status = 3; | |
| 634 } else if (word == "/Ordering") { | |
| 635 m_Status = 4; | |
| 636 } else if (word == "/Supplement") { | |
| 637 m_Status = 5; | |
| 638 } else if (word == "begincodespacerange") { | |
| 639 m_Status = 7; | |
| 640 m_CodeSeq = 0; | |
| 641 } else if (word == "usecmap") { | |
| 642 } else if (m_Status == 1 || m_Status == 2) { | |
| 643 m_CodePoints[m_CodeSeq] = CMap_GetCode(word); | |
| 644 m_CodeSeq++; | |
| 645 FX_DWORD StartCode, EndCode; | |
| 646 FX_WORD StartCID; | |
| 647 if (m_Status == 1) { | |
| 648 if (m_CodeSeq < 2) { | |
| 649 return; | |
| 650 } | |
| 651 EndCode = StartCode = m_CodePoints[0]; | |
| 652 StartCID = (FX_WORD)m_CodePoints[1]; | |
| 653 } else { | |
| 654 if (m_CodeSeq < 3) { | |
| 655 return; | |
| 656 } | |
| 657 StartCode = m_CodePoints[0]; | |
| 658 EndCode = m_CodePoints[1]; | |
| 659 StartCID = (FX_WORD)m_CodePoints[2]; | |
| 660 } | |
| 661 if (EndCode < 0x10000) { | |
| 662 for (FX_DWORD code = StartCode; code <= EndCode; code++) { | |
| 663 m_pCMap->m_pMapping[code] = (FX_WORD)(StartCID + code - StartCode); | |
| 664 } | |
| 665 } else { | |
| 666 FX_DWORD buf[2]; | |
| 667 buf[0] = StartCode; | |
| 668 buf[1] = ((EndCode - StartCode) << 16) + StartCID; | |
| 669 m_AddMaps.AppendBlock(buf, sizeof buf); | |
| 670 } | |
| 671 m_CodeSeq = 0; | |
| 672 } else if (m_Status == 3) { | |
| 673 CMap_GetString(word); | |
| 674 m_Status = 0; | |
| 675 } else if (m_Status == 4) { | |
| 676 m_pCMap->m_Charset = CharsetFromOrdering(CMap_GetString(word)); | |
| 677 m_Status = 0; | |
| 678 } else if (m_Status == 5) { | |
| 679 CMap_GetCode(word); | |
| 680 m_Status = 0; | |
| 681 } else if (m_Status == 6) { | |
| 682 m_pCMap->m_bVertical = CMap_GetCode(word); | |
| 683 m_Status = 0; | |
| 684 } else if (m_Status == 7) { | |
| 685 if (word == "endcodespacerange") { | |
| 686 int nSegs = m_CodeRanges.GetSize(); | |
| 687 if (nSegs > 1) { | |
| 688 m_pCMap->m_CodingScheme = CPDF_CMap::MixedFourBytes; | |
| 689 m_pCMap->m_nCodeRanges = nSegs; | |
| 690 m_pCMap->m_pLeadingBytes = | |
| 691 FX_Alloc2D(uint8_t, nSegs, sizeof(CMap_CodeRange)); | |
| 692 FXSYS_memcpy(m_pCMap->m_pLeadingBytes, m_CodeRanges.GetData(), | |
| 693 nSegs * sizeof(CMap_CodeRange)); | |
| 694 } else if (nSegs == 1) { | |
| 695 m_pCMap->m_CodingScheme = (m_CodeRanges[0].m_CharSize == 2) | |
| 696 ? CPDF_CMap::TwoBytes | |
| 697 : CPDF_CMap::OneByte; | |
| 698 } | |
| 699 m_Status = 0; | |
| 700 } else { | |
| 701 if (word.GetLength() == 0 || word.GetAt(0) != '<') { | |
| 702 return; | |
| 703 } | |
| 704 if (m_CodeSeq % 2) { | |
| 705 CMap_CodeRange range; | |
| 706 if (CMap_GetCodeRange(range, m_LastWord, word)) { | |
| 707 m_CodeRanges.Add(range); | |
| 708 } | |
| 709 } | |
| 710 m_CodeSeq++; | |
| 711 } | |
| 712 } | |
| 713 m_LastWord = word; | |
| 714 } | |
| 715 | |
| 716 // Static. | |
| 717 FX_DWORD CPDF_CMapParser::CMap_GetCode(const CFX_ByteStringC& word) { | |
| 718 int num = 0; | |
| 719 if (word.GetAt(0) == '<') { | |
| 720 for (int i = 1; i < word.GetLength() && std::isxdigit(word.GetAt(i)); ++i) | |
| 721 num = num * 16 + FXSYS_toHexDigit(word.GetAt(i)); | |
| 722 return num; | |
| 723 } | |
| 724 | |
| 725 for (int i = 0; i < word.GetLength() && std::isdigit(word.GetAt(i)); ++i) | |
| 726 num = num * 10 + FXSYS_toDecimalDigit(static_cast<FX_WCHAR>(word.GetAt(i))); | |
| 727 return num; | |
| 728 } | |
| 729 | |
| 730 // Static. | |
| 731 bool CPDF_CMapParser::CMap_GetCodeRange(CMap_CodeRange& range, | |
| 732 const CFX_ByteStringC& first, | |
| 733 const CFX_ByteStringC& second) { | |
| 734 if (first.GetLength() == 0 || first.GetAt(0) != '<') | |
| 735 return false; | |
| 736 | |
| 737 int i; | |
| 738 for (i = 1; i < first.GetLength(); ++i) { | |
| 739 if (first.GetAt(i) == '>') { | |
| 740 break; | |
| 741 } | |
| 742 } | |
| 743 range.m_CharSize = (i - 1) / 2; | |
| 744 if (range.m_CharSize > 4) | |
| 745 return false; | |
| 746 | |
| 747 for (i = 0; i < range.m_CharSize; ++i) { | |
| 748 uint8_t digit1 = first.GetAt(i * 2 + 1); | |
| 749 uint8_t digit2 = first.GetAt(i * 2 + 2); | |
| 750 range.m_Lower[i] = FXSYS_toHexDigit(digit1) * 16 + FXSYS_toHexDigit(digit2); | |
| 751 } | |
| 752 | |
| 753 FX_DWORD size = second.GetLength(); | |
| 754 for (i = 0; i < range.m_CharSize; ++i) { | |
| 755 uint8_t digit1 = ((FX_DWORD)i * 2 + 1 < size) | |
| 756 ? second.GetAt((FX_STRSIZE)i * 2 + 1) | |
| 757 : '0'; | |
| 758 uint8_t digit2 = ((FX_DWORD)i * 2 + 2 < size) | |
| 759 ? second.GetAt((FX_STRSIZE)i * 2 + 2) | |
| 760 : '0'; | |
| 761 range.m_Upper[i] = FXSYS_toHexDigit(digit1) * 16 + FXSYS_toHexDigit(digit2); | |
| 762 } | |
| 763 return true; | |
| 764 } | |
| 765 | |
| 766 CPDF_CMap::CPDF_CMap() { | |
| 767 m_Charset = CIDSET_UNKNOWN; | |
| 768 m_Coding = CIDCODING_UNKNOWN; | |
| 769 m_CodingScheme = TwoBytes; | |
| 770 m_bVertical = 0; | |
| 771 m_bLoaded = FALSE; | |
| 772 m_pMapping = NULL; | |
| 773 m_pLeadingBytes = NULL; | |
| 774 m_pAddMapping = NULL; | |
| 775 m_pEmbedMap = NULL; | |
| 776 m_pUseMap = NULL; | |
| 777 m_nCodeRanges = 0; | |
| 778 } | |
| 779 CPDF_CMap::~CPDF_CMap() { | |
| 780 FX_Free(m_pMapping); | |
| 781 FX_Free(m_pAddMapping); | |
| 782 FX_Free(m_pLeadingBytes); | |
| 783 delete m_pUseMap; | |
| 784 } | |
| 785 void CPDF_CMap::Release() { | |
| 786 if (m_PredefinedCMap.IsEmpty()) { | |
| 787 delete this; | |
| 788 } | |
| 789 } | |
| 790 | |
| 791 FX_BOOL CPDF_CMap::LoadPredefined(CPDF_CMapManager* pMgr, | |
| 792 const FX_CHAR* pName, | |
| 793 FX_BOOL bPromptCJK) { | |
| 794 m_PredefinedCMap = pName; | |
| 795 if (m_PredefinedCMap == "Identity-H" || m_PredefinedCMap == "Identity-V") { | |
| 796 m_Coding = CIDCODING_CID; | |
| 797 m_bVertical = pName[9] == 'V'; | |
| 798 m_bLoaded = TRUE; | |
| 799 return TRUE; | |
| 800 } | |
| 801 CFX_ByteString cmapid = m_PredefinedCMap; | |
| 802 m_bVertical = cmapid.Right(1) == "V"; | |
| 803 if (cmapid.GetLength() > 2) { | |
| 804 cmapid = cmapid.Left(cmapid.GetLength() - 2); | |
| 805 } | |
| 806 const CPDF_PredefinedCMap* map = nullptr; | |
| 807 for (size_t i = 0; i < FX_ArraySize(g_PredefinedCMaps); ++i) { | |
| 808 if (cmapid == CFX_ByteStringC(g_PredefinedCMaps[i].m_pName)) { | |
| 809 map = &g_PredefinedCMaps[i]; | |
| 810 break; | |
| 811 } | |
| 812 } | |
| 813 if (!map) | |
| 814 return FALSE; | |
| 815 | |
| 816 m_Charset = map->m_Charset; | |
| 817 m_Coding = map->m_Coding; | |
| 818 m_CodingScheme = map->m_CodingScheme; | |
| 819 if (m_CodingScheme == MixedTwoBytes) { | |
| 820 m_pLeadingBytes = FX_Alloc(uint8_t, 256); | |
| 821 for (FX_DWORD i = 0; i < map->m_LeadingSegCount; ++i) { | |
| 822 const uint8_t* segs = map->m_LeadingSegs; | |
| 823 for (int b = segs[i * 2]; b <= segs[i * 2 + 1]; ++b) { | |
| 824 m_pLeadingBytes[b] = 1; | |
| 825 } | |
| 826 } | |
| 827 } | |
| 828 FPDFAPI_FindEmbeddedCMap(pName, m_Charset, m_Coding, m_pEmbedMap); | |
| 829 if (m_pEmbedMap) { | |
| 830 m_bLoaded = TRUE; | |
| 831 return TRUE; | |
| 832 } | |
| 833 return FALSE; | |
| 834 } | |
| 835 FX_BOOL CPDF_CMap::LoadEmbedded(const uint8_t* pData, FX_DWORD size) { | |
| 836 m_pMapping = FX_Alloc(FX_WORD, 65536); | |
| 837 CPDF_CMapParser parser; | |
| 838 parser.Initialize(this); | |
| 839 CPDF_SimpleParser syntax(pData, size); | |
| 840 while (1) { | |
| 841 CFX_ByteStringC word = syntax.GetWord(); | |
| 842 if (word.IsEmpty()) { | |
| 843 break; | |
| 844 } | |
| 845 parser.ParseWord(word); | |
| 846 } | |
| 847 if (m_CodingScheme == MixedFourBytes && parser.m_AddMaps.GetSize()) { | |
| 848 m_pAddMapping = FX_Alloc(uint8_t, parser.m_AddMaps.GetSize() + 4); | |
| 849 *(FX_DWORD*)m_pAddMapping = parser.m_AddMaps.GetSize() / 8; | |
| 850 FXSYS_memcpy(m_pAddMapping + 4, parser.m_AddMaps.GetBuffer(), | |
| 851 parser.m_AddMaps.GetSize()); | |
| 852 FXSYS_qsort(m_pAddMapping + 4, parser.m_AddMaps.GetSize() / 8, 8, | |
| 853 CompareDWORD); | |
| 854 } | |
| 855 return TRUE; | |
| 856 } | |
| 857 | |
| 858 FX_WORD CPDF_CMap::CIDFromCharCode(FX_DWORD charcode) const { | |
| 859 if (m_Coding == CIDCODING_CID) { | |
| 860 return (FX_WORD)charcode; | |
| 861 } | |
| 862 if (m_pEmbedMap) { | |
| 863 return FPDFAPI_CIDFromCharCode(m_pEmbedMap, charcode); | |
| 864 } | |
| 865 if (!m_pMapping) { | |
| 866 return (FX_WORD)charcode; | |
| 867 } | |
| 868 if (charcode >> 16) { | |
| 869 if (m_pAddMapping) { | |
| 870 void* found = FXSYS_bsearch(&charcode, m_pAddMapping + 4, | |
| 871 *(FX_DWORD*)m_pAddMapping, 8, CompareCID); | |
| 872 if (!found) { | |
| 873 if (m_pUseMap) { | |
| 874 return m_pUseMap->CIDFromCharCode(charcode); | |
| 875 } | |
| 876 return 0; | |
| 877 } | |
| 878 return (FX_WORD)(((FX_DWORD*)found)[1] % 65536 + charcode - | |
| 879 *(FX_DWORD*)found); | |
| 880 } | |
| 881 if (m_pUseMap) | |
| 882 return m_pUseMap->CIDFromCharCode(charcode); | |
| 883 return 0; | |
| 884 } | |
| 885 FX_DWORD CID = m_pMapping[charcode]; | |
| 886 if (!CID && m_pUseMap) | |
| 887 return m_pUseMap->CIDFromCharCode(charcode); | |
| 888 return (FX_WORD)CID; | |
| 889 } | |
| 890 | |
| 891 FX_DWORD CPDF_CMap::GetNextChar(const FX_CHAR* pString, | |
| 892 int nStrLen, | |
| 893 int& offset) const { | |
| 894 switch (m_CodingScheme) { | |
| 895 case OneByte: | |
| 896 return ((uint8_t*)pString)[offset++]; | |
| 897 case TwoBytes: | |
| 898 offset += 2; | |
| 899 return ((uint8_t*)pString)[offset - 2] * 256 + | |
| 900 ((uint8_t*)pString)[offset - 1]; | |
| 901 case MixedTwoBytes: { | |
| 902 uint8_t byte1 = ((uint8_t*)pString)[offset++]; | |
| 903 if (!m_pLeadingBytes[byte1]) { | |
| 904 return byte1; | |
| 905 } | |
| 906 uint8_t byte2 = ((uint8_t*)pString)[offset++]; | |
| 907 return byte1 * 256 + byte2; | |
| 908 } | |
| 909 case MixedFourBytes: { | |
| 910 uint8_t codes[4]; | |
| 911 int char_size = 1; | |
| 912 codes[0] = ((uint8_t*)pString)[offset++]; | |
| 913 CMap_CodeRange* pRanges = (CMap_CodeRange*)m_pLeadingBytes; | |
| 914 while (1) { | |
| 915 int ret = CheckCodeRange(codes, char_size, pRanges, m_nCodeRanges); | |
| 916 if (ret == 0) { | |
| 917 return 0; | |
| 918 } | |
| 919 if (ret == 2) { | |
| 920 FX_DWORD charcode = 0; | |
| 921 for (int i = 0; i < char_size; i++) { | |
| 922 charcode = (charcode << 8) + codes[i]; | |
| 923 } | |
| 924 return charcode; | |
| 925 } | |
| 926 if (char_size == 4 || offset == nStrLen) { | |
| 927 return 0; | |
| 928 } | |
| 929 codes[char_size++] = ((uint8_t*)pString)[offset++]; | |
| 930 } | |
| 931 break; | |
| 932 } | |
| 933 } | |
| 934 return 0; | |
| 935 } | |
| 936 int CPDF_CMap::GetCharSize(FX_DWORD charcode) const { | |
| 937 switch (m_CodingScheme) { | |
| 938 case OneByte: | |
| 939 return 1; | |
| 940 case TwoBytes: | |
| 941 return 2; | |
| 942 case MixedTwoBytes: | |
| 943 case MixedFourBytes: | |
| 944 if (charcode < 0x100) { | |
| 945 return 1; | |
| 946 } | |
| 947 if (charcode < 0x10000) { | |
| 948 return 2; | |
| 949 } | |
| 950 if (charcode < 0x1000000) { | |
| 951 return 3; | |
| 952 } | |
| 953 return 4; | |
| 954 } | |
| 955 return 1; | |
| 956 } | |
| 957 int CPDF_CMap::CountChar(const FX_CHAR* pString, int size) const { | |
| 958 switch (m_CodingScheme) { | |
| 959 case OneByte: | |
| 960 return size; | |
| 961 case TwoBytes: | |
| 962 return (size + 1) / 2; | |
| 963 case MixedTwoBytes: { | |
| 964 int count = 0; | |
| 965 for (int i = 0; i < size; i++) { | |
| 966 count++; | |
| 967 if (m_pLeadingBytes[((uint8_t*)pString)[i]]) { | |
| 968 i++; | |
| 969 } | |
| 970 } | |
| 971 return count; | |
| 972 } | |
| 973 case MixedFourBytes: { | |
| 974 int count = 0, offset = 0; | |
| 975 while (offset < size) { | |
| 976 GetNextChar(pString, size, offset); | |
| 977 count++; | |
| 978 } | |
| 979 return count; | |
| 980 } | |
| 981 } | |
| 982 return size; | |
| 983 } | |
| 984 | |
| 985 int CPDF_CMap::AppendChar(FX_CHAR* str, FX_DWORD charcode) const { | |
| 986 switch (m_CodingScheme) { | |
| 987 case OneByte: | |
| 988 str[0] = (uint8_t)charcode; | |
| 989 return 1; | |
| 990 case TwoBytes: | |
| 991 str[0] = (uint8_t)(charcode / 256); | |
| 992 str[1] = (uint8_t)(charcode % 256); | |
| 993 return 2; | |
| 994 case MixedTwoBytes: | |
| 995 case MixedFourBytes: | |
| 996 if (charcode < 0x100) { | |
| 997 CMap_CodeRange* pRanges = (CMap_CodeRange*)m_pLeadingBytes; | |
| 998 int iSize = GetCharSizeImpl(charcode, pRanges, m_nCodeRanges); | |
| 999 if (iSize == 0) { | |
| 1000 iSize = 1; | |
| 1001 } | |
| 1002 if (iSize > 1) { | |
| 1003 FXSYS_memset(str, 0, sizeof(uint8_t) * iSize); | |
| 1004 } | |
| 1005 str[iSize - 1] = (uint8_t)charcode; | |
| 1006 return iSize; | |
| 1007 } | |
| 1008 if (charcode < 0x10000) { | |
| 1009 str[0] = (uint8_t)(charcode >> 8); | |
| 1010 str[1] = (uint8_t)charcode; | |
| 1011 return 2; | |
| 1012 } | |
| 1013 if (charcode < 0x1000000) { | |
| 1014 str[0] = (uint8_t)(charcode >> 16); | |
| 1015 str[1] = (uint8_t)(charcode >> 8); | |
| 1016 str[2] = (uint8_t)charcode; | |
| 1017 return 3; | |
| 1018 } | |
| 1019 str[0] = (uint8_t)(charcode >> 24); | |
| 1020 str[1] = (uint8_t)(charcode >> 16); | |
| 1021 str[2] = (uint8_t)(charcode >> 8); | |
| 1022 str[3] = (uint8_t)charcode; | |
| 1023 return 4; | |
| 1024 } | |
| 1025 return 0; | |
| 1026 } | |
| 1027 CPDF_CID2UnicodeMap::CPDF_CID2UnicodeMap() { | |
| 1028 m_EmbeddedCount = 0; | |
| 1029 } | |
| 1030 CPDF_CID2UnicodeMap::~CPDF_CID2UnicodeMap() {} | |
| 1031 FX_BOOL CPDF_CID2UnicodeMap::Initialize() { | |
| 1032 return TRUE; | |
| 1033 } | |
| 1034 FX_BOOL CPDF_CID2UnicodeMap::IsLoaded() { | |
| 1035 return m_EmbeddedCount != 0; | |
| 1036 } | |
| 1037 FX_WCHAR CPDF_CID2UnicodeMap::UnicodeFromCID(FX_WORD CID) { | |
| 1038 if (m_Charset == CIDSET_UNICODE) { | |
| 1039 return CID; | |
| 1040 } | |
| 1041 if (CID < m_EmbeddedCount) { | |
| 1042 return m_pEmbeddedMap[CID]; | |
| 1043 } | |
| 1044 return 0; | |
| 1045 } | |
| 1046 | |
| 1047 void CPDF_CID2UnicodeMap::Load(CPDF_CMapManager* pMgr, | |
| 1048 CIDSet charset, | |
| 1049 FX_BOOL bPromptCJK) { | |
| 1050 m_Charset = charset; | |
| 1051 FPDFAPI_LoadCID2UnicodeMap(charset, m_pEmbeddedMap, m_EmbeddedCount); | |
| 1052 } | |
| 1053 | |
| 1054 CPDF_CIDFont::CPDF_CIDFont() | |
| 1055 : m_pCMap(nullptr), | |
| 1056 m_pAllocatedCMap(nullptr), | |
| 1057 m_pCID2UnicodeMap(nullptr), | |
| 1058 m_pCIDToGIDMap(nullptr), | |
| 1059 m_bCIDIsGID(FALSE), | |
| 1060 m_pAnsiWidths(nullptr), | |
| 1061 m_bAdobeCourierStd(FALSE), | |
| 1062 m_pTTGSUBTable(nullptr) { | |
| 1063 } | |
| 1064 | |
| 1065 CPDF_CIDFont::~CPDF_CIDFont() { | |
| 1066 if (m_pAnsiWidths) { | |
| 1067 FX_Free(m_pAnsiWidths); | |
| 1068 } | |
| 1069 delete m_pAllocatedCMap; | |
| 1070 delete m_pCIDToGIDMap; | |
| 1071 delete m_pTTGSUBTable; | |
| 1072 } | |
| 1073 | |
| 1074 bool CPDF_CIDFont::IsCIDFont() const { | |
| 1075 return true; | |
| 1076 } | |
| 1077 | |
| 1078 const CPDF_CIDFont* CPDF_CIDFont::AsCIDFont() const { | |
| 1079 return this; | |
| 1080 } | |
| 1081 | |
| 1082 CPDF_CIDFont* CPDF_CIDFont::AsCIDFont() { | |
| 1083 return this; | |
| 1084 } | |
| 1085 | |
| 1086 FX_WORD CPDF_CIDFont::CIDFromCharCode(FX_DWORD charcode) const { | |
| 1087 if (!m_pCMap) { | |
| 1088 return (FX_WORD)charcode; | |
| 1089 } | |
| 1090 return m_pCMap->CIDFromCharCode(charcode); | |
| 1091 } | |
| 1092 | |
| 1093 FX_BOOL CPDF_CIDFont::IsVertWriting() const { | |
| 1094 return m_pCMap ? m_pCMap->IsVertWriting() : FALSE; | |
| 1095 } | |
| 1096 | |
| 1097 CFX_WideString CPDF_CIDFont::UnicodeFromCharCode(FX_DWORD charcode) const { | |
| 1098 CFX_WideString str = CPDF_Font::UnicodeFromCharCode(charcode); | |
| 1099 if (!str.IsEmpty()) | |
| 1100 return str; | |
| 1101 FX_WCHAR ret = GetUnicodeFromCharCode(charcode); | |
| 1102 if (ret == 0) | |
| 1103 return CFX_WideString(); | |
| 1104 return ret; | |
| 1105 } | |
| 1106 | |
| 1107 FX_WCHAR CPDF_CIDFont::GetUnicodeFromCharCode(FX_DWORD charcode) const { | |
| 1108 switch (m_pCMap->m_Coding) { | |
| 1109 case CIDCODING_UCS2: | |
| 1110 case CIDCODING_UTF16: | |
| 1111 return (FX_WCHAR)charcode; | |
| 1112 case CIDCODING_CID: | |
| 1113 if (!m_pCID2UnicodeMap || !m_pCID2UnicodeMap->IsLoaded()) { | |
| 1114 return 0; | |
| 1115 } | |
| 1116 return m_pCID2UnicodeMap->UnicodeFromCID((FX_WORD)charcode); | |
| 1117 } | |
| 1118 if (!m_pCMap->IsLoaded() || !m_pCID2UnicodeMap || | |
| 1119 !m_pCID2UnicodeMap->IsLoaded()) { | |
| 1120 #if _FXM_PLATFORM_ == _FXM_PLATFORM_WINDOWS_ | |
| 1121 FX_WCHAR unicode; | |
| 1122 int charsize = 1; | |
| 1123 if (charcode > 255) { | |
| 1124 charcode = (charcode % 256) * 256 + (charcode / 256); | |
| 1125 charsize = 2; | |
| 1126 } | |
| 1127 int ret = FXSYS_MultiByteToWideChar(g_CharsetCPs[m_pCMap->m_Coding], 0, | |
| 1128 (const FX_CHAR*)&charcode, charsize, | |
| 1129 &unicode, 1); | |
| 1130 if (ret != 1) { | |
| 1131 return 0; | |
| 1132 } | |
| 1133 return unicode; | |
| 1134 #endif | |
| 1135 if (m_pCMap->m_pEmbedMap) { | |
| 1136 return EmbeddedUnicodeFromCharcode(m_pCMap->m_pEmbedMap, | |
| 1137 m_pCMap->m_Charset, charcode); | |
| 1138 } | |
| 1139 return 0; | |
| 1140 } | |
| 1141 return m_pCID2UnicodeMap->UnicodeFromCID(CIDFromCharCode(charcode)); | |
| 1142 } | |
| 1143 | |
| 1144 FX_DWORD CPDF_CIDFont::CharCodeFromUnicode(FX_WCHAR unicode) const { | |
| 1145 FX_DWORD charcode = CPDF_Font::CharCodeFromUnicode(unicode); | |
| 1146 if (charcode) | |
| 1147 return charcode; | |
| 1148 switch (m_pCMap->m_Coding) { | |
| 1149 case CIDCODING_UNKNOWN: | |
| 1150 return 0; | |
| 1151 case CIDCODING_UCS2: | |
| 1152 case CIDCODING_UTF16: | |
| 1153 return unicode; | |
| 1154 case CIDCODING_CID: { | |
| 1155 if (!m_pCID2UnicodeMap || !m_pCID2UnicodeMap->IsLoaded()) { | |
| 1156 return 0; | |
| 1157 } | |
| 1158 FX_DWORD CID = 0; | |
| 1159 while (CID < 65536) { | |
| 1160 FX_WCHAR this_unicode = m_pCID2UnicodeMap->UnicodeFromCID((FX_WORD)CID); | |
| 1161 if (this_unicode == unicode) { | |
| 1162 return CID; | |
| 1163 } | |
| 1164 CID++; | |
| 1165 } | |
| 1166 break; | |
| 1167 } | |
| 1168 } | |
| 1169 | |
| 1170 if (unicode < 0x80) { | |
| 1171 return static_cast<FX_DWORD>(unicode); | |
| 1172 } | |
| 1173 if (m_pCMap->m_Coding == CIDCODING_CID) { | |
| 1174 return 0; | |
| 1175 } | |
| 1176 #if _FXM_PLATFORM_ == _FXM_PLATFORM_WINDOWS_ | |
| 1177 uint8_t buffer[32]; | |
| 1178 int ret = | |
| 1179 FXSYS_WideCharToMultiByte(g_CharsetCPs[m_pCMap->m_Coding], 0, &unicode, 1, | |
| 1180 (char*)buffer, 4, NULL, NULL); | |
| 1181 if (ret == 1) { | |
| 1182 return buffer[0]; | |
| 1183 } | |
| 1184 if (ret == 2) { | |
| 1185 return buffer[0] * 256 + buffer[1]; | |
| 1186 } | |
| 1187 #else | |
| 1188 if (m_pCMap->m_pEmbedMap) { | |
| 1189 return EmbeddedCharcodeFromUnicode(m_pCMap->m_pEmbedMap, m_pCMap->m_Charset, | |
| 1190 unicode); | |
| 1191 } | |
| 1192 #endif | |
| 1193 return 0; | |
| 1194 } | |
| 1195 | |
| 1196 FX_BOOL CPDF_CIDFont::Load() { | |
| 1197 if (m_pFontDict->GetStringBy("Subtype") == "TrueType") { | |
| 1198 return LoadGB2312(); | |
| 1199 } | |
| 1200 CPDF_Array* pFonts = m_pFontDict->GetArrayBy("DescendantFonts"); | |
| 1201 if (!pFonts) { | |
| 1202 return FALSE; | |
| 1203 } | |
| 1204 if (pFonts->GetCount() != 1) { | |
| 1205 return FALSE; | |
| 1206 } | |
| 1207 CPDF_Dictionary* pCIDFontDict = pFonts->GetDictAt(0); | |
| 1208 if (!pCIDFontDict) { | |
| 1209 return FALSE; | |
| 1210 } | |
| 1211 m_BaseFont = pCIDFontDict->GetStringBy("BaseFont"); | |
| 1212 if ((m_BaseFont.Compare("CourierStd") == 0 || | |
| 1213 m_BaseFont.Compare("CourierStd-Bold") == 0 || | |
| 1214 m_BaseFont.Compare("CourierStd-BoldOblique") == 0 || | |
| 1215 m_BaseFont.Compare("CourierStd-Oblique") == 0) && | |
| 1216 !IsEmbedded()) { | |
| 1217 m_bAdobeCourierStd = TRUE; | |
| 1218 } | |
| 1219 CPDF_Dictionary* pFontDesc = pCIDFontDict->GetDictBy("FontDescriptor"); | |
| 1220 if (pFontDesc) { | |
| 1221 LoadFontDescriptor(pFontDesc); | |
| 1222 } | |
| 1223 CPDF_Object* pEncoding = m_pFontDict->GetElementValue("Encoding"); | |
| 1224 if (!pEncoding) { | |
| 1225 return FALSE; | |
| 1226 } | |
| 1227 CFX_ByteString subtype = pCIDFontDict->GetStringBy("Subtype"); | |
| 1228 m_bType1 = (subtype == "CIDFontType0"); | |
| 1229 | |
| 1230 if (pEncoding->IsName()) { | |
| 1231 CFX_ByteString cmap = pEncoding->GetString(); | |
| 1232 m_pCMap = | |
| 1233 CPDF_ModuleMgr::Get() | |
| 1234 ->GetPageModule() | |
| 1235 ->GetFontGlobals() | |
| 1236 ->m_CMapManager.GetPredefinedCMap(cmap, m_pFontFile && m_bType1); | |
| 1237 } else if (CPDF_Stream* pStream = pEncoding->AsStream()) { | |
| 1238 m_pAllocatedCMap = m_pCMap = new CPDF_CMap; | |
| 1239 CPDF_StreamAcc acc; | |
| 1240 acc.LoadAllData(pStream, FALSE); | |
| 1241 m_pCMap->LoadEmbedded(acc.GetData(), acc.GetSize()); | |
| 1242 } else { | |
| 1243 return FALSE; | |
| 1244 } | |
| 1245 if (!m_pCMap) { | |
| 1246 return FALSE; | |
| 1247 } | |
| 1248 m_Charset = m_pCMap->m_Charset; | |
| 1249 if (m_Charset == CIDSET_UNKNOWN) { | |
| 1250 CPDF_Dictionary* pCIDInfo = pCIDFontDict->GetDictBy("CIDSystemInfo"); | |
| 1251 if (pCIDInfo) { | |
| 1252 m_Charset = CharsetFromOrdering(pCIDInfo->GetStringBy("Ordering")); | |
| 1253 } | |
| 1254 } | |
| 1255 if (m_Charset != CIDSET_UNKNOWN) | |
| 1256 m_pCID2UnicodeMap = | |
| 1257 CPDF_ModuleMgr::Get() | |
| 1258 ->GetPageModule() | |
| 1259 ->GetFontGlobals() | |
| 1260 ->m_CMapManager.GetCID2UnicodeMap( | |
| 1261 m_Charset, | |
| 1262 !m_pFontFile && (m_pCMap->m_Coding == CIDCODING_CID || | |
| 1263 pCIDFontDict->KeyExist("W"))); | |
| 1264 if (m_Font.GetFace()) { | |
| 1265 if (m_bType1) { | |
| 1266 FXFT_Select_Charmap(m_Font.GetFace(), FXFT_ENCODING_UNICODE); | |
| 1267 } else { | |
| 1268 FT_UseCIDCharmap(m_Font.GetFace(), m_pCMap->m_Coding); | |
| 1269 } | |
| 1270 } | |
| 1271 m_DefaultWidth = pCIDFontDict->GetIntegerBy("DW", 1000); | |
| 1272 CPDF_Array* pWidthArray = pCIDFontDict->GetArrayBy("W"); | |
| 1273 if (pWidthArray) { | |
| 1274 LoadMetricsArray(pWidthArray, m_WidthList, 1); | |
| 1275 } | |
| 1276 if (!IsEmbedded()) { | |
| 1277 LoadSubstFont(); | |
| 1278 } | |
| 1279 if (1) { | |
| 1280 if (m_pFontFile || (GetSubstFont()->m_SubstFlags & FXFONT_SUBST_EXACT)) { | |
| 1281 CPDF_Object* pmap = pCIDFontDict->GetElementValue("CIDToGIDMap"); | |
| 1282 if (pmap) { | |
| 1283 if (CPDF_Stream* pStream = pmap->AsStream()) { | |
| 1284 m_pCIDToGIDMap = new CPDF_StreamAcc; | |
| 1285 m_pCIDToGIDMap->LoadAllData(pStream, FALSE); | |
| 1286 } else if (pmap->GetString() == "Identity") { | |
| 1287 #if _FXM_PLATFORM_ == _FXM_PLATFORM_APPLE_ | |
| 1288 if (m_pFontFile) { | |
| 1289 m_bCIDIsGID = TRUE; | |
| 1290 } | |
| 1291 #else | |
| 1292 m_bCIDIsGID = TRUE; | |
| 1293 #endif | |
| 1294 } | |
| 1295 } | |
| 1296 } | |
| 1297 } | |
| 1298 CheckFontMetrics(); | |
| 1299 if (IsVertWriting()) { | |
| 1300 pWidthArray = pCIDFontDict->GetArrayBy("W2"); | |
| 1301 if (pWidthArray) { | |
| 1302 LoadMetricsArray(pWidthArray, m_VertMetrics, 3); | |
| 1303 } | |
| 1304 CPDF_Array* pDefaultArray = pCIDFontDict->GetArrayBy("DW2"); | |
| 1305 if (pDefaultArray) { | |
| 1306 m_DefaultVY = pDefaultArray->GetIntegerAt(0); | |
| 1307 m_DefaultW1 = pDefaultArray->GetIntegerAt(1); | |
| 1308 } else { | |
| 1309 m_DefaultVY = 880; | |
| 1310 m_DefaultW1 = -1000; | |
| 1311 } | |
| 1312 } | |
| 1313 return TRUE; | |
| 1314 } | |
| 1315 | |
| 1316 FX_RECT CPDF_CIDFont::GetCharBBox(FX_DWORD charcode, int level) { | |
| 1317 if (charcode < 256 && m_CharBBox[charcode].right != FX_SMALL_RECT::kInvalid) | |
| 1318 return FX_RECT(m_CharBBox[charcode]); | |
| 1319 | |
| 1320 FX_RECT rect; | |
| 1321 FX_BOOL bVert = FALSE; | |
| 1322 int glyph_index = GlyphFromCharCode(charcode, &bVert); | |
| 1323 FXFT_Face face = m_Font.GetFace(); | |
| 1324 if (face) { | |
| 1325 if (FXFT_Is_Face_Tricky(face)) { | |
| 1326 int err = FXFT_Load_Glyph(face, glyph_index, | |
| 1327 FXFT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH); | |
| 1328 if (!err) { | |
| 1329 FXFT_BBox cbox; | |
| 1330 FXFT_Glyph glyph; | |
| 1331 err = FXFT_Get_Glyph(((FXFT_Face)face)->glyph, &glyph); | |
| 1332 if (!err) { | |
| 1333 FXFT_Glyph_Get_CBox(glyph, FXFT_GLYPH_BBOX_PIXELS, &cbox); | |
| 1334 int pixel_size_x = ((FXFT_Face)face)->size->metrics.x_ppem; | |
| 1335 int pixel_size_y = ((FXFT_Face)face)->size->metrics.y_ppem; | |
| 1336 if (pixel_size_x == 0 || pixel_size_y == 0) { | |
| 1337 rect = FX_RECT(cbox.xMin, cbox.yMax, cbox.xMax, cbox.yMin); | |
| 1338 } else { | |
| 1339 rect = FX_RECT(cbox.xMin * 1000 / pixel_size_x, | |
| 1340 cbox.yMax * 1000 / pixel_size_y, | |
| 1341 cbox.xMax * 1000 / pixel_size_x, | |
| 1342 cbox.yMin * 1000 / pixel_size_y); | |
| 1343 } | |
| 1344 if (rect.top > FXFT_Get_Face_Ascender(face)) { | |
| 1345 rect.top = FXFT_Get_Face_Ascender(face); | |
| 1346 } | |
| 1347 if (rect.bottom < FXFT_Get_Face_Descender(face)) { | |
| 1348 rect.bottom = FXFT_Get_Face_Descender(face); | |
| 1349 } | |
| 1350 FXFT_Done_Glyph(glyph); | |
| 1351 } | |
| 1352 } | |
| 1353 } else { | |
| 1354 int err = FXFT_Load_Glyph(face, glyph_index, FXFT_LOAD_NO_SCALE); | |
| 1355 if (err == 0) { | |
| 1356 rect = FX_RECT(TT2PDF(FXFT_Get_Glyph_HoriBearingX(face), face), | |
| 1357 TT2PDF(FXFT_Get_Glyph_HoriBearingY(face), face), | |
| 1358 TT2PDF(FXFT_Get_Glyph_HoriBearingX(face) + | |
| 1359 FXFT_Get_Glyph_Width(face), | |
| 1360 face), | |
| 1361 TT2PDF(FXFT_Get_Glyph_HoriBearingY(face) - | |
| 1362 FXFT_Get_Glyph_Height(face), | |
| 1363 face)); | |
| 1364 rect.top += rect.top / 64; | |
| 1365 } | |
| 1366 } | |
| 1367 } | |
| 1368 if (!m_pFontFile && m_Charset == CIDSET_JAPAN1) { | |
| 1369 FX_WORD CID = CIDFromCharCode(charcode); | |
| 1370 const uint8_t* pTransform = GetCIDTransform(CID); | |
| 1371 if (pTransform && !bVert) { | |
| 1372 CFX_Matrix matrix(CIDTransformToFloat(pTransform[0]), | |
| 1373 CIDTransformToFloat(pTransform[1]), | |
| 1374 CIDTransformToFloat(pTransform[2]), | |
| 1375 CIDTransformToFloat(pTransform[3]), | |
| 1376 CIDTransformToFloat(pTransform[4]) * 1000, | |
| 1377 CIDTransformToFloat(pTransform[5]) * 1000); | |
| 1378 CFX_FloatRect rect_f(rect); | |
| 1379 rect_f.Transform(&matrix); | |
| 1380 rect = rect_f.GetOutterRect(); | |
| 1381 } | |
| 1382 } | |
| 1383 if (charcode < 256) | |
| 1384 m_CharBBox[charcode] = rect.ToSmallRect(); | |
| 1385 | |
| 1386 return rect; | |
| 1387 } | |
| 1388 int CPDF_CIDFont::GetCharWidthF(FX_DWORD charcode, int level) { | |
| 1389 if (m_pAnsiWidths && charcode < 0x80) { | |
| 1390 return m_pAnsiWidths[charcode]; | |
| 1391 } | |
| 1392 FX_WORD cid = CIDFromCharCode(charcode); | |
| 1393 int size = m_WidthList.GetSize(); | |
| 1394 FX_DWORD* list = m_WidthList.GetData(); | |
| 1395 for (int i = 0; i < size; i += 3) { | |
| 1396 if (cid >= list[i] && cid <= list[i + 1]) { | |
| 1397 return (int)list[i + 2]; | |
| 1398 } | |
| 1399 } | |
| 1400 return m_DefaultWidth; | |
| 1401 } | |
| 1402 short CPDF_CIDFont::GetVertWidth(FX_WORD CID) const { | |
| 1403 FX_DWORD vertsize = m_VertMetrics.GetSize() / 5; | |
| 1404 if (vertsize == 0) { | |
| 1405 return m_DefaultW1; | |
| 1406 } | |
| 1407 const FX_DWORD* pTable = m_VertMetrics.GetData(); | |
| 1408 for (FX_DWORD i = 0; i < vertsize; i++) | |
| 1409 if (pTable[i * 5] <= CID && pTable[i * 5 + 1] >= CID) { | |
| 1410 return (short)(int)pTable[i * 5 + 2]; | |
| 1411 } | |
| 1412 return m_DefaultW1; | |
| 1413 } | |
| 1414 void CPDF_CIDFont::GetVertOrigin(FX_WORD CID, short& vx, short& vy) const { | |
| 1415 FX_DWORD vertsize = m_VertMetrics.GetSize() / 5; | |
| 1416 if (vertsize) { | |
| 1417 const FX_DWORD* pTable = m_VertMetrics.GetData(); | |
| 1418 for (FX_DWORD i = 0; i < vertsize; i++) | |
| 1419 if (pTable[i * 5] <= CID && pTable[i * 5 + 1] >= CID) { | |
| 1420 vx = (short)(int)pTable[i * 5 + 3]; | |
| 1421 vy = (short)(int)pTable[i * 5 + 4]; | |
| 1422 return; | |
| 1423 } | |
| 1424 } | |
| 1425 FX_DWORD dwWidth = m_DefaultWidth; | |
| 1426 int size = m_WidthList.GetSize(); | |
| 1427 const FX_DWORD* list = m_WidthList.GetData(); | |
| 1428 for (int i = 0; i < size; i += 3) { | |
| 1429 if (CID >= list[i] && CID <= list[i + 1]) { | |
| 1430 dwWidth = (FX_WORD)list[i + 2]; | |
| 1431 break; | |
| 1432 } | |
| 1433 } | |
| 1434 vx = (short)dwWidth / 2; | |
| 1435 vy = (short)m_DefaultVY; | |
| 1436 } | |
| 1437 int CPDF_CIDFont::GetGlyphIndex(FX_DWORD unicode, FX_BOOL* pVertGlyph) { | |
| 1438 if (pVertGlyph) { | |
| 1439 *pVertGlyph = FALSE; | |
| 1440 } | |
| 1441 FXFT_Face face = m_Font.GetFace(); | |
| 1442 int index = FXFT_Get_Char_Index(face, unicode); | |
| 1443 if (unicode == 0x2502) { | |
| 1444 return index; | |
| 1445 } | |
| 1446 if (index && IsVertWriting()) { | |
| 1447 if (m_pTTGSUBTable) { | |
| 1448 uint32_t vindex = 0; | |
| 1449 m_pTTGSUBTable->GetVerticalGlyph(index, &vindex); | |
| 1450 if (vindex) { | |
| 1451 index = vindex; | |
| 1452 if (pVertGlyph) { | |
| 1453 *pVertGlyph = TRUE; | |
| 1454 } | |
| 1455 } | |
| 1456 return index; | |
| 1457 } | |
| 1458 if (!m_Font.GetSubData()) { | |
| 1459 unsigned long length = 0; | |
| 1460 int error = FXFT_Load_Sfnt_Table(face, FT_MAKE_TAG('G', 'S', 'U', 'B'), 0, | |
| 1461 NULL, &length); | |
| 1462 if (!error) { | |
| 1463 m_Font.SetSubData(FX_Alloc(uint8_t, length)); | |
| 1464 } | |
| 1465 } | |
| 1466 int error = FXFT_Load_Sfnt_Table(face, FT_MAKE_TAG('G', 'S', 'U', 'B'), 0, | |
| 1467 m_Font.GetSubData(), NULL); | |
| 1468 if (!error && m_Font.GetSubData()) { | |
| 1469 m_pTTGSUBTable = new CFX_CTTGSUBTable; | |
| 1470 m_pTTGSUBTable->LoadGSUBTable((FT_Bytes)m_Font.GetSubData()); | |
| 1471 uint32_t vindex = 0; | |
| 1472 m_pTTGSUBTable->GetVerticalGlyph(index, &vindex); | |
| 1473 if (vindex) { | |
| 1474 index = vindex; | |
| 1475 if (pVertGlyph) { | |
| 1476 *pVertGlyph = TRUE; | |
| 1477 } | |
| 1478 } | |
| 1479 } | |
| 1480 return index; | |
| 1481 } | |
| 1482 if (pVertGlyph) { | |
| 1483 *pVertGlyph = FALSE; | |
| 1484 } | |
| 1485 return index; | |
| 1486 } | |
| 1487 int CPDF_CIDFont::GlyphFromCharCode(FX_DWORD charcode, FX_BOOL* pVertGlyph) { | |
| 1488 if (pVertGlyph) { | |
| 1489 *pVertGlyph = FALSE; | |
| 1490 } | |
| 1491 if (!m_pFontFile && !m_pCIDToGIDMap) { | |
| 1492 FX_WORD cid = CIDFromCharCode(charcode); | |
| 1493 FX_WCHAR unicode = 0; | |
| 1494 if (m_bCIDIsGID) { | |
| 1495 #if _FXM_PLATFORM_ != _FXM_PLATFORM_APPLE_ | |
| 1496 return cid; | |
| 1497 #else | |
| 1498 if (m_Flags & PDFFONT_SYMBOLIC) { | |
| 1499 return cid; | |
| 1500 } | |
| 1501 CFX_WideString uni_str = UnicodeFromCharCode(charcode); | |
| 1502 if (uni_str.IsEmpty()) { | |
| 1503 return cid; | |
| 1504 } | |
| 1505 unicode = uni_str.GetAt(0); | |
| 1506 #endif | |
| 1507 } else { | |
| 1508 if (cid && m_pCID2UnicodeMap && m_pCID2UnicodeMap->IsLoaded()) { | |
| 1509 unicode = m_pCID2UnicodeMap->UnicodeFromCID(cid); | |
| 1510 } | |
| 1511 if (unicode == 0) { | |
| 1512 unicode = GetUnicodeFromCharCode(charcode); | |
| 1513 } | |
| 1514 if (unicode == 0 && !(m_Flags & PDFFONT_SYMBOLIC)) { | |
| 1515 unicode = UnicodeFromCharCode(charcode).GetAt(0); | |
| 1516 } | |
| 1517 } | |
| 1518 FXFT_Face face = m_Font.GetFace(); | |
| 1519 if (unicode == 0) { | |
| 1520 if (!m_bAdobeCourierStd) { | |
| 1521 return charcode == 0 ? -1 : (int)charcode; | |
| 1522 } | |
| 1523 charcode += 31; | |
| 1524 int index = 0, iBaseEncoding; | |
| 1525 FX_BOOL bMSUnicode = FT_UseTTCharmap(face, 3, 1); | |
| 1526 FX_BOOL bMacRoman = FALSE; | |
| 1527 if (!bMSUnicode) { | |
| 1528 bMacRoman = FT_UseTTCharmap(face, 1, 0); | |
| 1529 } | |
| 1530 iBaseEncoding = PDFFONT_ENCODING_STANDARD; | |
| 1531 if (bMSUnicode) { | |
| 1532 iBaseEncoding = PDFFONT_ENCODING_WINANSI; | |
| 1533 } else if (bMacRoman) { | |
| 1534 iBaseEncoding = PDFFONT_ENCODING_MACROMAN; | |
| 1535 } | |
| 1536 const FX_CHAR* name = GetAdobeCharName(iBaseEncoding, NULL, charcode); | |
| 1537 if (!name) { | |
| 1538 return charcode == 0 ? -1 : (int)charcode; | |
| 1539 } | |
| 1540 FX_WORD unicode = PDF_UnicodeFromAdobeName(name); | |
| 1541 if (unicode) { | |
| 1542 if (bMSUnicode) { | |
| 1543 index = FXFT_Get_Char_Index(face, unicode); | |
| 1544 } else if (bMacRoman) { | |
| 1545 FX_DWORD maccode = | |
| 1546 FT_CharCodeFromUnicode(FXFT_ENCODING_APPLE_ROMAN, unicode); | |
| 1547 index = !maccode ? FXFT_Get_Name_Index(face, (char*)name) | |
| 1548 : FXFT_Get_Char_Index(face, maccode); | |
| 1549 } else { | |
| 1550 return FXFT_Get_Char_Index(face, unicode); | |
| 1551 } | |
| 1552 } else { | |
| 1553 return charcode == 0 ? -1 : (int)charcode; | |
| 1554 } | |
| 1555 if (index == 0 || index == 0xffff) { | |
| 1556 return charcode == 0 ? -1 : (int)charcode; | |
| 1557 } | |
| 1558 return index; | |
| 1559 } | |
| 1560 if (m_Charset == CIDSET_JAPAN1) { | |
| 1561 if (unicode == '\\') { | |
| 1562 unicode = '/'; | |
| 1563 #if _FXM_PLATFORM_ != _FXM_PLATFORM_APPLE_ | |
| 1564 } else if (unicode == 0xa5) { | |
| 1565 unicode = 0x5c; | |
| 1566 #endif | |
| 1567 } | |
| 1568 } | |
| 1569 if (!face) | |
| 1570 return unicode; | |
| 1571 | |
| 1572 int err = FXFT_Select_Charmap(face, FXFT_ENCODING_UNICODE); | |
| 1573 if (err != 0) { | |
| 1574 int i; | |
| 1575 for (i = 0; i < FXFT_Get_Face_CharmapCount(face); i++) { | |
| 1576 FX_DWORD ret = FT_CharCodeFromUnicode( | |
| 1577 FXFT_Get_Charmap_Encoding(FXFT_Get_Face_Charmaps(face)[i]), | |
| 1578 (FX_WCHAR)charcode); | |
| 1579 if (ret == 0) { | |
| 1580 continue; | |
| 1581 } | |
| 1582 FXFT_Set_Charmap(face, FXFT_Get_Face_Charmaps(face)[i]); | |
| 1583 unicode = (FX_WCHAR)ret; | |
| 1584 break; | |
| 1585 } | |
| 1586 if (i == FXFT_Get_Face_CharmapCount(face) && i) { | |
| 1587 FXFT_Set_Charmap(face, FXFT_Get_Face_Charmaps(face)[0]); | |
| 1588 unicode = (FX_WCHAR)charcode; | |
| 1589 } | |
| 1590 } | |
| 1591 if (FXFT_Get_Face_Charmap(face)) { | |
| 1592 int index = GetGlyphIndex(unicode, pVertGlyph); | |
| 1593 if (index == 0) | |
| 1594 return -1; | |
| 1595 return index; | |
| 1596 } | |
| 1597 return unicode; | |
| 1598 } | |
| 1599 if (!m_Font.GetFace()) | |
| 1600 return -1; | |
| 1601 | |
| 1602 FX_WORD cid = CIDFromCharCode(charcode); | |
| 1603 if (m_bType1) { | |
| 1604 if (!m_pCIDToGIDMap) { | |
| 1605 return cid; | |
| 1606 } | |
| 1607 } else { | |
| 1608 if (!m_pCIDToGIDMap) { | |
| 1609 if (m_pFontFile && !m_pCMap->m_pMapping) | |
| 1610 return cid; | |
| 1611 if (m_pCMap->m_Coding == CIDCODING_UNKNOWN || | |
| 1612 !FXFT_Get_Face_Charmap(m_Font.GetFace())) { | |
| 1613 return cid; | |
| 1614 } | |
| 1615 if (FXFT_Get_Charmap_Encoding(FXFT_Get_Face_Charmap(m_Font.GetFace())) == | |
| 1616 FXFT_ENCODING_UNICODE) { | |
| 1617 CFX_WideString unicode_str = UnicodeFromCharCode(charcode); | |
| 1618 if (unicode_str.IsEmpty()) { | |
| 1619 return -1; | |
| 1620 } | |
| 1621 charcode = unicode_str.GetAt(0); | |
| 1622 } | |
| 1623 return GetGlyphIndex(charcode, pVertGlyph); | |
| 1624 } | |
| 1625 } | |
| 1626 FX_DWORD byte_pos = cid * 2; | |
| 1627 if (byte_pos + 2 > m_pCIDToGIDMap->GetSize()) | |
| 1628 return -1; | |
| 1629 | |
| 1630 const uint8_t* pdata = m_pCIDToGIDMap->GetData() + byte_pos; | |
| 1631 return pdata[0] * 256 + pdata[1]; | |
| 1632 } | |
| 1633 FX_DWORD CPDF_CIDFont::GetNextChar(const FX_CHAR* pString, | |
| 1634 int nStrLen, | |
| 1635 int& offset) const { | |
| 1636 return m_pCMap->GetNextChar(pString, nStrLen, offset); | |
| 1637 } | |
| 1638 int CPDF_CIDFont::GetCharSize(FX_DWORD charcode) const { | |
| 1639 return m_pCMap->GetCharSize(charcode); | |
| 1640 } | |
| 1641 int CPDF_CIDFont::CountChar(const FX_CHAR* pString, int size) const { | |
| 1642 return m_pCMap->CountChar(pString, size); | |
| 1643 } | |
| 1644 int CPDF_CIDFont::AppendChar(FX_CHAR* str, FX_DWORD charcode) const { | |
| 1645 return m_pCMap->AppendChar(str, charcode); | |
| 1646 } | |
| 1647 FX_BOOL CPDF_CIDFont::IsUnicodeCompatible() const { | |
| 1648 if (!m_pCMap->IsLoaded() || !m_pCID2UnicodeMap || | |
| 1649 !m_pCID2UnicodeMap->IsLoaded()) { | |
| 1650 return m_pCMap->m_Coding != CIDCODING_UNKNOWN; | |
| 1651 } | |
| 1652 return TRUE; | |
| 1653 } | |
| 1654 FX_BOOL CPDF_CIDFont::IsFontStyleFromCharCode(FX_DWORD charcode) const { | |
| 1655 return TRUE; | |
| 1656 } | |
| 1657 void CPDF_CIDFont::LoadSubstFont() { | |
| 1658 m_Font.LoadSubst(m_BaseFont, !m_bType1, m_Flags, m_StemV * 5, m_ItalicAngle, | |
| 1659 g_CharsetCPs[m_Charset], IsVertWriting()); | |
| 1660 } | |
| 1661 void CPDF_CIDFont::LoadMetricsArray(CPDF_Array* pArray, | |
| 1662 CFX_DWordArray& result, | |
| 1663 int nElements) { | |
| 1664 int width_status = 0; | |
| 1665 int iCurElement = 0; | |
| 1666 int first_code = 0, last_code; | |
| 1667 FX_DWORD count = pArray->GetCount(); | |
| 1668 for (FX_DWORD i = 0; i < count; i++) { | |
| 1669 CPDF_Object* pObj = pArray->GetElementValue(i); | |
| 1670 if (!pObj) | |
| 1671 continue; | |
| 1672 | |
| 1673 if (CPDF_Array* pArray = pObj->AsArray()) { | |
| 1674 if (width_status != 1) | |
| 1675 return; | |
| 1676 | |
| 1677 FX_DWORD count = pArray->GetCount(); | |
| 1678 for (FX_DWORD j = 0; j < count; j += nElements) { | |
| 1679 result.Add(first_code); | |
| 1680 result.Add(first_code); | |
| 1681 for (int k = 0; k < nElements; k++) { | |
| 1682 result.Add(pArray->GetIntegerAt(j + k)); | |
| 1683 } | |
| 1684 first_code++; | |
| 1685 } | |
| 1686 width_status = 0; | |
| 1687 } else { | |
| 1688 if (width_status == 0) { | |
| 1689 first_code = pObj->GetInteger(); | |
| 1690 width_status = 1; | |
| 1691 } else if (width_status == 1) { | |
| 1692 last_code = pObj->GetInteger(); | |
| 1693 width_status = 2; | |
| 1694 iCurElement = 0; | |
| 1695 } else { | |
| 1696 if (!iCurElement) { | |
| 1697 result.Add(first_code); | |
| 1698 result.Add(last_code); | |
| 1699 } | |
| 1700 result.Add(pObj->GetInteger()); | |
| 1701 iCurElement++; | |
| 1702 if (iCurElement == nElements) { | |
| 1703 width_status = 0; | |
| 1704 } | |
| 1705 } | |
| 1706 } | |
| 1707 } | |
| 1708 } | |
| 1709 | |
| 1710 // static | |
| 1711 FX_FLOAT CPDF_CIDFont::CIDTransformToFloat(uint8_t ch) { | |
| 1712 if (ch < 128) { | |
| 1713 return ch * 1.0f / 127; | |
| 1714 } | |
| 1715 return (-255 + ch) * 1.0f / 127; | |
| 1716 } | |
| 1717 | |
| 1718 FX_BOOL CPDF_CIDFont::LoadGB2312() { | |
| 1719 m_BaseFont = m_pFontDict->GetStringBy("BaseFont"); | |
| 1720 CPDF_Dictionary* pFontDesc = m_pFontDict->GetDictBy("FontDescriptor"); | |
| 1721 if (pFontDesc) { | |
| 1722 LoadFontDescriptor(pFontDesc); | |
| 1723 } | |
| 1724 m_Charset = CIDSET_GB1; | |
| 1725 m_bType1 = FALSE; | |
| 1726 m_pCMap = CPDF_ModuleMgr::Get() | |
| 1727 ->GetPageModule() | |
| 1728 ->GetFontGlobals() | |
| 1729 ->m_CMapManager.GetPredefinedCMap("GBK-EUC-H", FALSE); | |
| 1730 m_pCID2UnicodeMap = CPDF_ModuleMgr::Get() | |
| 1731 ->GetPageModule() | |
| 1732 ->GetFontGlobals() | |
| 1733 ->m_CMapManager.GetCID2UnicodeMap(m_Charset, FALSE); | |
| 1734 if (!IsEmbedded()) { | |
| 1735 LoadSubstFont(); | |
| 1736 } | |
| 1737 CheckFontMetrics(); | |
| 1738 m_DefaultWidth = 1000; | |
| 1739 m_pAnsiWidths = FX_Alloc(FX_WORD, 128); | |
| 1740 for (int i = 32; i < 127; i++) { | |
| 1741 m_pAnsiWidths[i] = 500; | |
| 1742 } | |
| 1743 return TRUE; | |
| 1744 } | |
| 1745 | |
| 1746 const uint8_t* CPDF_CIDFont::GetCIDTransform(FX_WORD CID) const { | |
| 1747 if (m_Charset != CIDSET_JAPAN1 || m_pFontFile) | |
| 1748 return nullptr; | |
| 1749 | |
| 1750 const struct CIDTransform* found = (const struct CIDTransform*)FXSYS_bsearch( | |
| 1751 &CID, g_Japan1_VertCIDs, FX_ArraySize(g_Japan1_VertCIDs), | |
| 1752 sizeof(g_Japan1_VertCIDs[0]), CompareCIDTransform); | |
| 1753 return found ? &found->a : nullptr; | |
| 1754 } | |
| OLD | NEW |