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 "fpdfsdk/include/pdfwindow/PWL_FontMap.h" | |
8 | |
9 #include "core/include/fpdfapi/cpdf_document.h" | |
10 #include "core/include/fpdfapi/fpdf_module.h" | |
11 #include "fpdfsdk/include/pdfwindow/PWL_Wnd.h" | |
12 | |
13 namespace { | |
14 | |
15 const char kDefaultFontName[] = "Helvetica"; | |
16 | |
17 const char* const g_sDEStandardFontName[] = {"Courier", | |
18 "Courier-Bold", | |
19 "Courier-BoldOblique", | |
20 "Courier-Oblique", | |
21 "Helvetica", | |
22 "Helvetica-Bold", | |
23 "Helvetica-BoldOblique", | |
24 "Helvetica-Oblique", | |
25 "Times-Roman", | |
26 "Times-Bold", | |
27 "Times-Italic", | |
28 "Times-BoldItalic", | |
29 "Symbol", | |
30 "ZapfDingbats"}; | |
31 | |
32 } // namespace | |
33 | |
34 CPWL_FontMap::CPWL_FontMap(IFX_SystemHandler* pSystemHandler) | |
35 : m_pPDFDoc(NULL), m_pSystemHandler(pSystemHandler) { | |
36 ASSERT(m_pSystemHandler); | |
37 } | |
38 | |
39 CPWL_FontMap::~CPWL_FontMap() { | |
40 delete m_pPDFDoc; | |
41 m_pPDFDoc = NULL; | |
42 | |
43 Empty(); | |
44 } | |
45 | |
46 void CPWL_FontMap::SetSystemHandler(IFX_SystemHandler* pSystemHandler) { | |
47 m_pSystemHandler = pSystemHandler; | |
48 } | |
49 | |
50 CPDF_Document* CPWL_FontMap::GetDocument() { | |
51 if (!m_pPDFDoc) { | |
52 if (CPDF_ModuleMgr::Get()) { | |
53 m_pPDFDoc = new CPDF_Document; | |
54 m_pPDFDoc->CreateNewDoc(); | |
55 } | |
56 } | |
57 | |
58 return m_pPDFDoc; | |
59 } | |
60 | |
61 CPDF_Font* CPWL_FontMap::GetPDFFont(int32_t nFontIndex) { | |
62 if (nFontIndex >= 0 && nFontIndex < m_aData.GetSize()) { | |
63 if (CPWL_FontMap_Data* pData = m_aData.GetAt(nFontIndex)) { | |
64 return pData->pFont; | |
65 } | |
66 } | |
67 | |
68 return NULL; | |
69 } | |
70 | |
71 CFX_ByteString CPWL_FontMap::GetPDFFontAlias(int32_t nFontIndex) { | |
72 if (nFontIndex >= 0 && nFontIndex < m_aData.GetSize()) { | |
73 if (CPWL_FontMap_Data* pData = m_aData.GetAt(nFontIndex)) { | |
74 return pData->sFontName; | |
75 } | |
76 } | |
77 | |
78 return ""; | |
79 } | |
80 | |
81 FX_BOOL CPWL_FontMap::KnowWord(int32_t nFontIndex, FX_WORD word) { | |
82 if (nFontIndex >= 0 && nFontIndex < m_aData.GetSize()) { | |
83 if (m_aData.GetAt(nFontIndex)) { | |
84 return CharCodeFromUnicode(nFontIndex, word) >= 0; | |
85 } | |
86 } | |
87 | |
88 return FALSE; | |
89 } | |
90 | |
91 int32_t CPWL_FontMap::GetWordFontIndex(FX_WORD word, | |
92 int32_t nCharset, | |
93 int32_t nFontIndex) { | |
94 if (nFontIndex > 0) { | |
95 if (KnowWord(nFontIndex, word)) | |
96 return nFontIndex; | |
97 } else { | |
98 if (const CPWL_FontMap_Data* pData = GetFontMapData(0)) { | |
99 if (nCharset == DEFAULT_CHARSET || pData->nCharset == SYMBOL_CHARSET || | |
100 nCharset == pData->nCharset) { | |
101 if (KnowWord(0, word)) | |
102 return 0; | |
103 } | |
104 } | |
105 } | |
106 | |
107 int32_t nNewFontIndex = | |
108 GetFontIndex(GetNativeFontName(nCharset), nCharset, TRUE); | |
109 if (nNewFontIndex >= 0) { | |
110 if (KnowWord(nNewFontIndex, word)) | |
111 return nNewFontIndex; | |
112 } | |
113 nNewFontIndex = GetFontIndex("Arial Unicode MS", DEFAULT_CHARSET, FALSE); | |
114 if (nNewFontIndex >= 0) { | |
115 if (KnowWord(nNewFontIndex, word)) | |
116 return nNewFontIndex; | |
117 } | |
118 return -1; | |
119 } | |
120 | |
121 int32_t CPWL_FontMap::CharCodeFromUnicode(int32_t nFontIndex, FX_WORD word) { | |
122 if (CPWL_FontMap_Data* pData = m_aData.GetAt(nFontIndex)) { | |
123 if (pData->pFont) { | |
124 if (pData->pFont->IsUnicodeCompatible()) { | |
125 int nCharCode = pData->pFont->CharCodeFromUnicode(word); | |
126 pData->pFont->GlyphFromCharCode(nCharCode); | |
127 return nCharCode; | |
128 } | |
129 if (word < 0xFF) | |
130 return word; | |
131 } | |
132 } | |
133 return -1; | |
134 } | |
135 | |
136 CFX_ByteString CPWL_FontMap::GetNativeFontName(int32_t nCharset) { | |
137 // searching native font is slow, so we must save time | |
138 for (int32_t i = 0, sz = m_aNativeFont.GetSize(); i < sz; i++) { | |
139 if (CPWL_FontMap_Native* pData = m_aNativeFont.GetAt(i)) { | |
140 if (pData->nCharset == nCharset) | |
141 return pData->sFontName; | |
142 } | |
143 } | |
144 | |
145 CFX_ByteString sNew = GetNativeFont(nCharset); | |
146 | |
147 if (!sNew.IsEmpty()) { | |
148 CPWL_FontMap_Native* pNewData = new CPWL_FontMap_Native; | |
149 pNewData->nCharset = nCharset; | |
150 pNewData->sFontName = sNew; | |
151 | |
152 m_aNativeFont.Add(pNewData); | |
153 } | |
154 | |
155 return sNew; | |
156 } | |
157 | |
158 void CPWL_FontMap::Empty() { | |
159 { | |
160 for (int32_t i = 0, sz = m_aData.GetSize(); i < sz; i++) | |
161 delete m_aData.GetAt(i); | |
162 | |
163 m_aData.RemoveAll(); | |
164 } | |
165 { | |
166 for (int32_t i = 0, sz = m_aNativeFont.GetSize(); i < sz; i++) | |
167 delete m_aNativeFont.GetAt(i); | |
168 | |
169 m_aNativeFont.RemoveAll(); | |
170 } | |
171 } | |
172 | |
173 void CPWL_FontMap::Initialize() { | |
174 GetFontIndex(kDefaultFontName, ANSI_CHARSET, FALSE); | |
175 } | |
176 | |
177 FX_BOOL CPWL_FontMap::IsStandardFont(const CFX_ByteString& sFontName) { | |
178 for (int32_t i = 0; i < FX_ArraySize(g_sDEStandardFontName); ++i) { | |
179 if (sFontName == g_sDEStandardFontName[i]) | |
180 return TRUE; | |
181 } | |
182 | |
183 return FALSE; | |
184 } | |
185 | |
186 int32_t CPWL_FontMap::FindFont(const CFX_ByteString& sFontName, | |
187 int32_t nCharset) { | |
188 for (int32_t i = 0, sz = m_aData.GetSize(); i < sz; i++) { | |
189 if (CPWL_FontMap_Data* pData = m_aData.GetAt(i)) { | |
190 if (nCharset == DEFAULT_CHARSET || nCharset == pData->nCharset) { | |
191 if (sFontName.IsEmpty() || pData->sFontName == sFontName) | |
192 return i; | |
193 } | |
194 } | |
195 } | |
196 | |
197 return -1; | |
198 } | |
199 | |
200 int32_t CPWL_FontMap::GetFontIndex(const CFX_ByteString& sFontName, | |
201 int32_t nCharset, | |
202 FX_BOOL bFind) { | |
203 int32_t nFontIndex = FindFont(EncodeFontAlias(sFontName, nCharset), nCharset); | |
204 if (nFontIndex >= 0) | |
205 return nFontIndex; | |
206 | |
207 CFX_ByteString sAlias; | |
208 CPDF_Font* pFont = NULL; | |
209 if (bFind) | |
210 pFont = FindFontSameCharset(sAlias, nCharset); | |
211 | |
212 if (!pFont) { | |
213 CFX_ByteString sTemp = sFontName; | |
214 pFont = AddFontToDocument(GetDocument(), sTemp, nCharset); | |
215 sAlias = EncodeFontAlias(sTemp, nCharset); | |
216 } | |
217 AddedFont(pFont, sAlias); | |
218 return AddFontData(pFont, sAlias, nCharset); | |
219 } | |
220 | |
221 int32_t CPWL_FontMap::GetPWLFontIndex(FX_WORD word, int32_t nCharset) { | |
222 int32_t nFind = -1; | |
223 | |
224 for (int32_t i = 0, sz = m_aData.GetSize(); i < sz; i++) { | |
225 if (CPWL_FontMap_Data* pData = m_aData.GetAt(i)) { | |
226 if (pData->nCharset == nCharset) { | |
227 nFind = i; | |
228 break; | |
229 } | |
230 } | |
231 } | |
232 | |
233 CPDF_Font* pNewFont = GetPDFFont(nFind); | |
234 | |
235 if (!pNewFont) | |
236 return -1; | |
237 | |
238 CFX_ByteString sAlias = EncodeFontAlias("Arial_Chrome", nCharset); | |
239 AddedFont(pNewFont, sAlias); | |
240 | |
241 return AddFontData(pNewFont, sAlias, nCharset); | |
242 } | |
243 | |
244 CPDF_Font* CPWL_FontMap::FindFontSameCharset(CFX_ByteString& sFontAlias, | |
245 int32_t nCharset) { | |
246 return NULL; | |
247 } | |
248 | |
249 int32_t CPWL_FontMap::AddFontData(CPDF_Font* pFont, | |
250 const CFX_ByteString& sFontAlias, | |
251 int32_t nCharset) { | |
252 CPWL_FontMap_Data* pNewData = new CPWL_FontMap_Data; | |
253 pNewData->pFont = pFont; | |
254 pNewData->sFontName = sFontAlias; | |
255 pNewData->nCharset = nCharset; | |
256 | |
257 m_aData.Add(pNewData); | |
258 | |
259 return m_aData.GetSize() - 1; | |
260 } | |
261 | |
262 void CPWL_FontMap::AddedFont(CPDF_Font* pFont, | |
263 const CFX_ByteString& sFontAlias) {} | |
264 | |
265 CFX_ByteString CPWL_FontMap::GetFontName(int32_t nFontIndex) { | |
266 if (nFontIndex >= 0 && nFontIndex < m_aData.GetSize()) { | |
267 if (CPWL_FontMap_Data* pData = m_aData.GetAt(nFontIndex)) { | |
268 return pData->sFontName; | |
269 } | |
270 } | |
271 | |
272 return ""; | |
273 } | |
274 | |
275 CFX_ByteString CPWL_FontMap::GetNativeFont(int32_t nCharset) { | |
276 if (nCharset == DEFAULT_CHARSET) | |
277 nCharset = GetNativeCharset(); | |
278 | |
279 CFX_ByteString sFontName = GetDefaultFontByCharset(nCharset); | |
280 if (m_pSystemHandler) { | |
281 if (m_pSystemHandler->FindNativeTrueTypeFont(nCharset, sFontName)) | |
282 return sFontName; | |
283 | |
284 sFontName = m_pSystemHandler->GetNativeTrueTypeFont(nCharset); | |
285 } | |
286 return sFontName; | |
287 } | |
288 | |
289 CPDF_Font* CPWL_FontMap::AddFontToDocument(CPDF_Document* pDoc, | |
290 CFX_ByteString& sFontName, | |
291 uint8_t nCharset) { | |
292 if (IsStandardFont(sFontName)) | |
293 return AddStandardFont(pDoc, sFontName); | |
294 | |
295 return AddSystemFont(pDoc, sFontName, nCharset); | |
296 } | |
297 | |
298 CPDF_Font* CPWL_FontMap::AddStandardFont(CPDF_Document* pDoc, | |
299 CFX_ByteString& sFontName) { | |
300 if (!pDoc) | |
301 return NULL; | |
302 | |
303 CPDF_Font* pFont = NULL; | |
304 | |
305 if (sFontName == "ZapfDingbats") { | |
306 pFont = pDoc->AddStandardFont(sFontName, NULL); | |
307 } else { | |
308 CPDF_FontEncoding fe(PDFFONT_ENCODING_WINANSI); | |
309 pFont = pDoc->AddStandardFont(sFontName, &fe); | |
310 } | |
311 | |
312 return pFont; | |
313 } | |
314 | |
315 CPDF_Font* CPWL_FontMap::AddSystemFont(CPDF_Document* pDoc, | |
316 CFX_ByteString& sFontName, | |
317 uint8_t nCharset) { | |
318 if (!pDoc) | |
319 return NULL; | |
320 | |
321 if (sFontName.IsEmpty()) | |
322 sFontName = GetNativeFont(nCharset); | |
323 if (nCharset == DEFAULT_CHARSET) | |
324 nCharset = GetNativeCharset(); | |
325 | |
326 if (m_pSystemHandler) | |
327 return m_pSystemHandler->AddNativeTrueTypeFontToPDF(pDoc, sFontName, | |
328 nCharset); | |
329 | |
330 return NULL; | |
331 } | |
332 | |
333 CFX_ByteString CPWL_FontMap::EncodeFontAlias(const CFX_ByteString& sFontName, | |
334 int32_t nCharset) { | |
335 CFX_ByteString sPostfix; | |
336 sPostfix.Format("_%02X", nCharset); | |
337 return EncodeFontAlias(sFontName) + sPostfix; | |
338 } | |
339 | |
340 CFX_ByteString CPWL_FontMap::EncodeFontAlias(const CFX_ByteString& sFontName) { | |
341 CFX_ByteString sRet = sFontName; | |
342 sRet.Remove(' '); | |
343 return sRet; | |
344 } | |
345 | |
346 int32_t CPWL_FontMap::GetFontMapCount() const { | |
347 return m_aData.GetSize(); | |
348 } | |
349 | |
350 const CPWL_FontMap_Data* CPWL_FontMap::GetFontMapData(int32_t nIndex) const { | |
351 if (nIndex >= 0 && nIndex < m_aData.GetSize()) { | |
352 return m_aData.GetAt(nIndex); | |
353 } | |
354 | |
355 return NULL; | |
356 } | |
357 | |
358 int32_t CPWL_FontMap::GetNativeCharset() { | |
359 uint8_t nCharset = ANSI_CHARSET; | |
360 int32_t iCodePage = FXSYS_GetACP(); | |
361 switch (iCodePage) { | |
362 case 932: // Japan | |
363 nCharset = SHIFTJIS_CHARSET; | |
364 break; | |
365 case 936: // Chinese (PRC, Singapore) | |
366 nCharset = GB2312_CHARSET; | |
367 break; | |
368 case 950: // Chinese (Taiwan; Hong Kong SAR, PRC) | |
369 nCharset = GB2312_CHARSET; | |
370 break; | |
371 case 1252: // Windows 3.1 Latin 1 (US, Western Europe) | |
372 nCharset = ANSI_CHARSET; | |
373 break; | |
374 case 874: // Thai | |
375 nCharset = THAI_CHARSET; | |
376 break; | |
377 case 949: // Korean | |
378 nCharset = HANGUL_CHARSET; | |
379 break; | |
380 case 1200: // Unicode (BMP of ISO 10646) | |
381 nCharset = ANSI_CHARSET; | |
382 break; | |
383 case 1250: // Windows 3.1 Eastern European | |
384 nCharset = EASTEUROPE_CHARSET; | |
385 break; | |
386 case 1251: // Windows 3.1 Cyrillic | |
387 nCharset = RUSSIAN_CHARSET; | |
388 break; | |
389 case 1253: // Windows 3.1 Greek | |
390 nCharset = GREEK_CHARSET; | |
391 break; | |
392 case 1254: // Windows 3.1 Turkish | |
393 nCharset = TURKISH_CHARSET; | |
394 break; | |
395 case 1255: // Hebrew | |
396 nCharset = HEBREW_CHARSET; | |
397 break; | |
398 case 1256: // Arabic | |
399 nCharset = ARABIC_CHARSET; | |
400 break; | |
401 case 1257: // Baltic | |
402 nCharset = BALTIC_CHARSET; | |
403 break; | |
404 case 1258: // Vietnamese | |
405 nCharset = VIETNAMESE_CHARSET; | |
406 break; | |
407 case 1361: // Korean(Johab) | |
408 nCharset = JOHAB_CHARSET; | |
409 break; | |
410 } | |
411 return nCharset; | |
412 } | |
413 | |
414 const CPWL_FontMap::CharsetFontMap CPWL_FontMap::defaultTTFMap[] = { | |
415 {ANSI_CHARSET, "Helvetica"}, {GB2312_CHARSET, "SimSun"}, | |
416 {CHINESEBIG5_CHARSET, "MingLiU"}, {SHIFTJIS_CHARSET, "MS Gothic"}, | |
417 {HANGUL_CHARSET, "Batang"}, {RUSSIAN_CHARSET, "Arial"}, | |
418 #if _FXM_PLATFORM_ == _FXM_PLATFORM_LINUX_ || \ | |
419 _FXM_PLATFORM_ == _FXM_PLATFORM_APPLE_ | |
420 {EASTEUROPE_CHARSET, "Arial"}, | |
421 #else | |
422 {EASTEUROPE_CHARSET, "Tahoma"}, | |
423 #endif | |
424 {ARABIC_CHARSET, "Arial"}, {-1, NULL}}; | |
425 | |
426 CFX_ByteString CPWL_FontMap::GetDefaultFontByCharset(int32_t nCharset) { | |
427 int i = 0; | |
428 while (defaultTTFMap[i].charset != -1) { | |
429 if (nCharset == defaultTTFMap[i].charset) | |
430 return defaultTTFMap[i].fontname; | |
431 ++i; | |
432 } | |
433 return ""; | |
434 } | |
435 | |
436 int32_t CPWL_FontMap::CharSetFromUnicode(FX_WORD word, int32_t nOldCharset) { | |
437 if (m_pSystemHandler && (-1 != m_pSystemHandler->GetCharSet())) | |
438 return m_pSystemHandler->GetCharSet(); | |
439 // to avoid CJK Font to show ASCII | |
440 if (word < 0x7F) | |
441 return ANSI_CHARSET; | |
442 // follow the old charset | |
443 if (nOldCharset != DEFAULT_CHARSET) | |
444 return nOldCharset; | |
445 | |
446 // find new charset | |
447 if ((word >= 0x4E00 && word <= 0x9FA5) || | |
448 (word >= 0xE7C7 && word <= 0xE7F3) || | |
449 (word >= 0x3000 && word <= 0x303F) || | |
450 (word >= 0x2000 && word <= 0x206F)) { | |
451 return GB2312_CHARSET; | |
452 } | |
453 | |
454 if (((word >= 0x3040) && (word <= 0x309F)) || | |
455 ((word >= 0x30A0) && (word <= 0x30FF)) || | |
456 ((word >= 0x31F0) && (word <= 0x31FF)) || | |
457 ((word >= 0xFF00) && (word <= 0xFFEF))) { | |
458 return SHIFTJIS_CHARSET; | |
459 } | |
460 | |
461 if (((word >= 0xAC00) && (word <= 0xD7AF)) || | |
462 ((word >= 0x1100) && (word <= 0x11FF)) || | |
463 ((word >= 0x3130) && (word <= 0x318F))) { | |
464 return HANGUL_CHARSET; | |
465 } | |
466 | |
467 if (word >= 0x0E00 && word <= 0x0E7F) | |
468 return THAI_CHARSET; | |
469 | |
470 if ((word >= 0x0370 && word <= 0x03FF) || (word >= 0x1F00 && word <= 0x1FFF)) | |
471 return GREEK_CHARSET; | |
472 | |
473 if ((word >= 0x0600 && word <= 0x06FF) || (word >= 0xFB50 && word <= 0xFEFC)) | |
474 return ARABIC_CHARSET; | |
475 | |
476 if (word >= 0x0590 && word <= 0x05FF) | |
477 return HEBREW_CHARSET; | |
478 | |
479 if (word >= 0x0400 && word <= 0x04FF) | |
480 return RUSSIAN_CHARSET; | |
481 | |
482 if (word >= 0x0100 && word <= 0x024F) | |
483 return EASTEUROPE_CHARSET; | |
484 | |
485 if (word >= 0x1E00 && word <= 0x1EFF) | |
486 return VIETNAMESE_CHARSET; | |
487 | |
488 return ANSI_CHARSET; | |
489 } | |
490 | |
491 CPWL_DocFontMap::CPWL_DocFontMap(IFX_SystemHandler* pSystemHandler, | |
492 CPDF_Document* pAttachedDoc) | |
493 : CPWL_FontMap(pSystemHandler), m_pAttachedDoc(pAttachedDoc) {} | |
494 | |
495 CPWL_DocFontMap::~CPWL_DocFontMap() {} | |
496 | |
497 CPDF_Document* CPWL_DocFontMap::GetDocument() { | |
498 return m_pAttachedDoc; | |
499 } | |
OLD | NEW |