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/font/font_int.h"
8
9 #include <memory>
10 #include <utility>
11
12 #include "core/fpdfapi/cpdf_modulemgr.h"
13 #include "core/fpdfapi/page/cpdf_form.h"
14 #include "core/fpdfapi/page/cpdf_pagemodule.h"
15 #include "core/fpdfapi/page/pageint.h"
16 #include "core/fpdfapi/parser/cpdf_array.h"
17 #include "core/fpdfapi/parser/cpdf_dictionary.h"
18 #include "core/fpdfapi/parser/cpdf_document.h"
19 #include "core/fpdfapi/parser/cpdf_name.h"
20 #include "core/fpdfapi/parser/cpdf_number.h"
21 #include "core/fpdfapi/parser/cpdf_simple_parser.h"
22 #include "core/fpdfapi/parser/cpdf_stream_acc.h"
23 #include "core/fxcrt/fx_ext.h"
24 #include "core/fxcrt/fx_safe_types.h"
25 #include "core/fxge/fx_freetype.h"
26 #include "third_party/base/numerics/safe_conversions.h"
27 #include "third_party/base/ptr_util.h"
28 #include "third_party/base/stl_util.h"
29
TT2PDF(int m,FXFT_Face face)30 int TT2PDF(int m, FXFT_Face face) {
31 int upm = FXFT_Get_Face_UnitsPerEM(face);
32 if (upm == 0)
33 return m;
34 return pdfium::base::checked_cast<int>(
35 (static_cast<double>(m) * 1000 + upm / 2) / upm);
36 }
37
FT_UseTTCharmap(FXFT_Face face,int platform_id,int encoding_id)38 bool FT_UseTTCharmap(FXFT_Face face, int platform_id, int encoding_id) {
39 auto* pCharMap = FXFT_Get_Face_Charmaps(face);
40 for (int i = 0; i < FXFT_Get_Face_CharmapCount(face); i++) {
41 if (FXFT_Get_Charmap_PlatformID(pCharMap[i]) == platform_id &&
42 FXFT_Get_Charmap_EncodingID(pCharMap[i]) == encoding_id) {
43 FXFT_Set_Charmap(face, pCharMap[i]);
44 return true;
45 }
46 }
47 return false;
48 }
49
CFX_StockFontArray()50 CFX_StockFontArray::CFX_StockFontArray() {}
51
~CFX_StockFontArray()52 CFX_StockFontArray::~CFX_StockFontArray() {
53 for (size_t i = 0; i < FX_ArraySize(m_StockFonts); ++i) {
54 if (m_StockFonts[i])
55 delete m_StockFonts[i]->GetFontDict();
56 }
57 }
58
GetFont(uint32_t index) const59 CPDF_Font* CFX_StockFontArray::GetFont(uint32_t index) const {
60 if (index >= FX_ArraySize(m_StockFonts))
61 return nullptr;
62 return m_StockFonts[index].get();
63 }
64
SetFont(uint32_t index,std::unique_ptr<CPDF_Font> pFont)65 CPDF_Font* CFX_StockFontArray::SetFont(uint32_t index,
66 std::unique_ptr<CPDF_Font> pFont) {
67 CPDF_Font* result = pFont.get();
68 if (index < FX_ArraySize(m_StockFonts))
69 m_StockFonts[index] = std::move(pFont);
70 return result;
71 }
72
CPDF_FontGlobals()73 CPDF_FontGlobals::CPDF_FontGlobals() {
74 FXSYS_memset(m_EmbeddedCharsets, 0, sizeof(m_EmbeddedCharsets));
75 FXSYS_memset(m_EmbeddedToUnicodes, 0, sizeof(m_EmbeddedToUnicodes));
76 }
77
~CPDF_FontGlobals()78 CPDF_FontGlobals::~CPDF_FontGlobals() {}
79
Find(CPDF_Document * pDoc,uint32_t index)80 CPDF_Font* CPDF_FontGlobals::Find(CPDF_Document* pDoc, uint32_t index) {
81 auto it = m_StockMap.find(pDoc);
82 if (it == m_StockMap.end())
83 return nullptr;
84 return it->second ? it->second->GetFont(index) : nullptr;
85 }
86
Set(CPDF_Document * pDoc,uint32_t index,std::unique_ptr<CPDF_Font> pFont)87 CPDF_Font* CPDF_FontGlobals::Set(CPDF_Document* pDoc,
88 uint32_t index,
89 std::unique_ptr<CPDF_Font> pFont) {
90 if (!pdfium::ContainsKey(m_StockMap, pDoc))
91 m_StockMap[pDoc] = pdfium::MakeUnique<CFX_StockFontArray>();
92 return m_StockMap[pDoc]->SetFont(index, std::move(pFont));
93 }
94
Clear(CPDF_Document * pDoc)95 void CPDF_FontGlobals::Clear(CPDF_Document* pDoc) {
96 m_StockMap.erase(pDoc);
97 }
98
Lookup(uint32_t charcode) const99 CFX_WideString CPDF_ToUnicodeMap::Lookup(uint32_t charcode) const {
100 auto it = m_Map.find(charcode);
101 if (it != m_Map.end()) {
102 uint32_t value = it->second;
103 FX_WCHAR unicode = (FX_WCHAR)(value & 0xffff);
104 if (unicode != 0xffff) {
105 return unicode;
106 }
107 const FX_WCHAR* buf = m_MultiCharBuf.GetBuffer();
108 uint32_t buf_len = m_MultiCharBuf.GetLength();
109 if (!buf || buf_len == 0) {
110 return CFX_WideString();
111 }
112 uint32_t index = value >> 16;
113 if (index >= buf_len) {
114 return CFX_WideString();
115 }
116 uint32_t len = buf[index];
117 if (index + len < index || index + len >= buf_len) {
118 return CFX_WideString();
119 }
120 return CFX_WideString(buf + index + 1, len);
121 }
122 if (m_pBaseMap) {
123 return m_pBaseMap->UnicodeFromCID((uint16_t)charcode);
124 }
125 return CFX_WideString();
126 }
127
ReverseLookup(FX_WCHAR unicode) const128 uint32_t CPDF_ToUnicodeMap::ReverseLookup(FX_WCHAR unicode) const {
129 for (const auto& pair : m_Map) {
130 if (pair.second == static_cast<uint32_t>(unicode))
131 return pair.first;
132 }
133 return 0;
134 }
135
136 // Static.
StringToCode(const CFX_ByteStringC & str)137 uint32_t CPDF_ToUnicodeMap::StringToCode(const CFX_ByteStringC& str) {
138 int len = str.GetLength();
139 if (len == 0)
140 return 0;
141
142 uint32_t result = 0;
143 if (str[0] == '<') {
144 for (int i = 1; i < len && std::isxdigit(str[i]); ++i)
145 result = result * 16 + FXSYS_toHexDigit(str.CharAt(i));
146 return result;
147 }
148
149 for (int i = 0; i < len && std::isdigit(str[i]); ++i)
150 result = result * 10 + FXSYS_toDecimalDigit(str.CharAt(i));
151
152 return result;
153 }
154
StringDataAdd(CFX_WideString str)155 static CFX_WideString StringDataAdd(CFX_WideString str) {
156 CFX_WideString ret;
157 int len = str.GetLength();
158 FX_WCHAR value = 1;
159 for (int i = len - 1; i >= 0; --i) {
160 FX_WCHAR ch = str[i] + value;
161 if (ch < str[i]) {
162 ret.Insert(0, 0);
163 } else {
164 ret.Insert(0, ch);
165 value = 0;
166 }
167 }
168 if (value) {
169 ret.Insert(0, value);
170 }
171 return ret;
172 }
173
174 // Static.
StringToWideString(const CFX_ByteStringC & str)175 CFX_WideString CPDF_ToUnicodeMap::StringToWideString(
176 const CFX_ByteStringC& str) {
177 int len = str.GetLength();
178 if (len == 0)
179 return CFX_WideString();
180
181 CFX_WideString result;
182 if (str[0] == '<') {
183 int byte_pos = 0;
184 FX_WCHAR ch = 0;
185 for (int i = 1; i < len && std::isxdigit(str[i]); ++i) {
186 ch = ch * 16 + FXSYS_toHexDigit(str[i]);
187 byte_pos++;
188 if (byte_pos == 4) {
189 result += ch;
190 byte_pos = 0;
191 ch = 0;
192 }
193 }
194 return result;
195 }
196 return result;
197 }
198
CPDF_ToUnicodeMap()199 CPDF_ToUnicodeMap::CPDF_ToUnicodeMap() : m_pBaseMap(nullptr) {}
200
~CPDF_ToUnicodeMap()201 CPDF_ToUnicodeMap::~CPDF_ToUnicodeMap() {}
202
GetUnicode()203 uint32_t CPDF_ToUnicodeMap::GetUnicode() {
204 FX_SAFE_UINT32 uni = m_MultiCharBuf.GetLength();
205 uni = uni * 0x10000 + 0xffff;
206 return uni.ValueOrDefault(0);
207 }
208
Load(CPDF_Stream * pStream)209 void CPDF_ToUnicodeMap::Load(CPDF_Stream* pStream) {
210 CIDSet cid_set = CIDSET_UNKNOWN;
211 CPDF_StreamAcc stream;
212 stream.LoadAllData(pStream, false);
213 CPDF_SimpleParser parser(stream.GetData(), stream.GetSize());
214 while (1) {
215 CFX_ByteStringC word = parser.GetWord();
216 if (word.IsEmpty()) {
217 break;
218 }
219 if (word == "beginbfchar") {
220 while (1) {
221 word = parser.GetWord();
222 if (word.IsEmpty() || word == "endbfchar") {
223 break;
224 }
225 uint32_t srccode = StringToCode(word);
226 word = parser.GetWord();
227 CFX_WideString destcode = StringToWideString(word);
228 int len = destcode.GetLength();
229 if (len == 0) {
230 continue;
231 }
232 if (len == 1) {
233 m_Map[srccode] = destcode.GetAt(0);
234 } else {
235 m_Map[srccode] = GetUnicode();
236 m_MultiCharBuf.AppendChar(destcode.GetLength());
237 m_MultiCharBuf << destcode;
238 }
239 }
240 } else if (word == "beginbfrange") {
241 while (1) {
242 CFX_ByteString low, high;
243 low = parser.GetWord();
244 if (low.IsEmpty() || low == "endbfrange") {
245 break;
246 }
247 high = parser.GetWord();
248 uint32_t lowcode = StringToCode(low.AsStringC());
249 uint32_t highcode =
250 (lowcode & 0xffffff00) | (StringToCode(high.AsStringC()) & 0xff);
251 if (highcode == (uint32_t)-1) {
252 break;
253 }
254 CFX_ByteString start(parser.GetWord());
255 if (start == "[") {
256 for (uint32_t code = lowcode; code <= highcode; code++) {
257 CFX_ByteString dest(parser.GetWord());
258 CFX_WideString destcode = StringToWideString(dest.AsStringC());
259 int len = destcode.GetLength();
260 if (len == 0) {
261 continue;
262 }
263 if (len == 1) {
264 m_Map[code] = destcode.GetAt(0);
265 } else {
266 m_Map[code] = GetUnicode();
267 m_MultiCharBuf.AppendChar(destcode.GetLength());
268 m_MultiCharBuf << destcode;
269 }
270 }
271 parser.GetWord();
272 } else {
273 CFX_WideString destcode = StringToWideString(start.AsStringC());
274 int len = destcode.GetLength();
275 uint32_t value = 0;
276 if (len == 1) {
277 value = StringToCode(start.AsStringC());
278 for (uint32_t code = lowcode; code <= highcode; code++) {
279 m_Map[code] = value++;
280 }
281 } else {
282 for (uint32_t code = lowcode; code <= highcode; code++) {
283 CFX_WideString retcode;
284 if (code == lowcode) {
285 retcode = destcode;
286 } else {
287 retcode = StringDataAdd(destcode);
288 }
289 m_Map[code] = GetUnicode();
290 m_MultiCharBuf.AppendChar(retcode.GetLength());
291 m_MultiCharBuf << retcode;
292 destcode = retcode;
293 }
294 }
295 }
296 }
297 } else if (word == "/Adobe-Korea1-UCS2") {
298 cid_set = CIDSET_KOREA1;
299 } else if (word == "/Adobe-Japan1-UCS2") {
300 cid_set = CIDSET_JAPAN1;
301 } else if (word == "/Adobe-CNS1-UCS2") {
302 cid_set = CIDSET_CNS1;
303 } else if (word == "/Adobe-GB1-UCS2") {
304 cid_set = CIDSET_GB1;
305 }
306 }
307 if (cid_set) {
308 m_pBaseMap = CPDF_ModuleMgr::Get()
309 ->GetPageModule()
310 ->GetFontGlobals()
311 ->m_CMapManager.GetCID2UnicodeMap(cid_set, false);
312 } else {
313 m_pBaseMap = nullptr;
314 }
315 }
316