1 /*
2 * Copyright 2013 Google Inc.
3 *
4 * Use of this source code is governed by a BSD-style license that can be
5 * found in the LICENSE file.
6 */
7
8 #include "SkOTTable_name.h"
9
10 #include "SkEndian.h"
11 #include "SkString.h"
12 #include "SkTSearch.h"
13 #include "SkTemplates.h"
14 #include "SkUtils.h"
15
SkUTF16BE_NextUnichar(const uint16_t ** srcPtr)16 static SkUnichar SkUTF16BE_NextUnichar(const uint16_t** srcPtr) {
17 SkASSERT(srcPtr && *srcPtr);
18
19 const uint16_t* src = *srcPtr;
20 SkUnichar c = SkEndian_SwapBE16(*src++);
21
22 SkASSERT(!SkUTF16_IsLowSurrogate(c));
23 if (SkUTF16_IsHighSurrogate(c)) {
24 unsigned c2 = SkEndian_SwapBE16(*src++);
25 SkASSERT(SkUTF16_IsLowSurrogate(c2));
26
27 c = (c << 10) + c2 + (0x10000 - (0xD800 << 10) - 0xDC00);
28 }
29 *srcPtr = src;
30 return c;
31 }
32
SkStringFromUTF16BE(const uint16_t * utf16be,size_t length,SkString & utf8)33 static void SkStringFromUTF16BE(const uint16_t* utf16be, size_t length, SkString& utf8) {
34 SkASSERT(utf16be != NULL);
35
36 utf8.reset();
37 size_t numberOf16BitValues = length / 2;
38 const uint16_t* end = utf16be + numberOf16BitValues;
39 while (utf16be < end) {
40 utf8.appendUnichar(SkUTF16BE_NextUnichar(&utf16be));
41 }
42 }
43
44 /** UnicodeFromMacRoman[macRomanPoint - 0x80] -> unicodeCodePoint.
45 * Derived from http://www.unicode.org/Public/MAPPINGS/VENDORS/APPLE/ROMAN.TXT .
46 * In MacRoman the first 128 code points match ASCII code points.
47 * This maps the second 128 MacRoman code points to unicode code points.
48 */
49 static uint16_t UnicodeFromMacRoman[0x80] = {
50 0x00C4, 0x00C5, 0x00C7, 0x00C9, 0x00D1, 0x00D6, 0x00DC, 0x00E1,
51 0x00E0, 0x00E2, 0x00E4, 0x00E3, 0x00E5, 0x00E7, 0x00E9, 0x00E8,
52 0x00EA, 0x00EB, 0x00ED, 0x00EC, 0x00EE, 0x00EF, 0x00F1, 0x00F3,
53 0x00F2, 0x00F4, 0x00F6, 0x00F5, 0x00FA, 0x00F9, 0x00FB, 0x00FC,
54 0x2020, 0x00B0, 0x00A2, 0x00A3, 0x00A7, 0x2022, 0x00B6, 0x00DF,
55 0x00AE, 0x00A9, 0x2122, 0x00B4, 0x00A8, 0x2260, 0x00C6, 0x00D8,
56 0x221E, 0x00B1, 0x2264, 0x2265, 0x00A5, 0x00B5, 0x2202, 0x2211,
57 0x220F, 0x03C0, 0x222B, 0x00AA, 0x00BA, 0x03A9, 0x00E6, 0x00F8,
58 0x00BF, 0x00A1, 0x00AC, 0x221A, 0x0192, 0x2248, 0x2206, 0x00AB,
59 0x00BB, 0x2026, 0x00A0, 0x00C0, 0x00C3, 0x00D5, 0x0152, 0x0153,
60 0x2013, 0x2014, 0x201C, 0x201D, 0x2018, 0x2019, 0x00F7, 0x25CA,
61 0x00FF, 0x0178, 0x2044, 0x20AC, 0x2039, 0x203A, 0xFB01, 0xFB02,
62 0x2021, 0x00B7, 0x201A, 0x201E, 0x2030, 0x00C2, 0x00CA, 0x00C1,
63 0x00CB, 0x00C8, 0x00CD, 0x00CE, 0x00CF, 0x00CC, 0x00D3, 0x00D4,
64 0xF8FF, 0x00D2, 0x00DA, 0x00DB, 0x00D9, 0x0131, 0x02C6, 0x02DC,
65 0x00AF, 0x02D8, 0x02D9, 0x02DA, 0x00B8, 0x02DD, 0x02DB, 0x02C7,
66 };
67
SkStringFromMacRoman(const uint8_t * macRoman,size_t length,SkString & utf8)68 static void SkStringFromMacRoman(const uint8_t* macRoman, size_t length, SkString& utf8) {
69 utf8.reset();
70 for (size_t i = 0; i < length; ++i) {
71 utf8.appendUnichar(macRoman[i] < 0x80 ? macRoman[i]
72 : UnicodeFromMacRoman[macRoman[i] - 0x80]);
73 }
74 }
75
76 static struct BCP47FromLanguageId {
77 uint16_t languageID;
78 const char* bcp47;
79 }
80 /** The Mac and Windows values do not conflict, so this is currently one single table. */
81 BCP47FromLanguageID[] = {
82 /** A mapping from Mac Language Designators to BCP 47 codes.
83 * The following list was constructed more or less manually.
84 * Apple now uses BCP 47 (post OSX10.4), so there will be no new entries.
85 */
86 {0, "en"}, //English
87 {1, "fr"}, //French
88 {2, "de"}, //German
89 {3, "it"}, //Italian
90 {4, "nl"}, //Dutch
91 {5, "sv"}, //Swedish
92 {6, "es"}, //Spanish
93 {7, "da"}, //Danish
94 {8, "pt"}, //Portuguese
95 {9, "nb"}, //Norwegian
96 {10, "he"}, //Hebrew
97 {11, "ja"}, //Japanese
98 {12, "ar"}, //Arabic
99 {13, "fi"}, //Finnish
100 {14, "el"}, //Greek
101 {15, "is"}, //Icelandic
102 {16, "mt"}, //Maltese
103 {17, "tr"}, //Turkish
104 {18, "hr"}, //Croatian
105 {19, "zh-Hant"}, //Chinese (Traditional)
106 {20, "ur"}, //Urdu
107 {21, "hi"}, //Hindi
108 {22, "th"}, //Thai
109 {23, "ko"}, //Korean
110 {24, "lt"}, //Lithuanian
111 {25, "pl"}, //Polish
112 {26, "hu"}, //Hungarian
113 {27, "et"}, //Estonian
114 {28, "lv"}, //Latvian
115 {29, "se"}, //Sami
116 {30, "fo"}, //Faroese
117 {31, "fa"}, //Farsi (Persian)
118 {32, "ru"}, //Russian
119 {33, "zh-Hans"}, //Chinese (Simplified)
120 {34, "nl"}, //Dutch
121 {35, "ga"}, //Irish(Gaelic)
122 {36, "sq"}, //Albanian
123 {37, "ro"}, //Romanian
124 {38, "cs"}, //Czech
125 {39, "sk"}, //Slovak
126 {40, "sl"}, //Slovenian
127 {41, "yi"}, //Yiddish
128 {42, "sr"}, //Serbian
129 {43, "mk"}, //Macedonian
130 {44, "bg"}, //Bulgarian
131 {45, "uk"}, //Ukrainian
132 {46, "be"}, //Byelorussian
133 {47, "uz"}, //Uzbek
134 {48, "kk"}, //Kazakh
135 {49, "az-Cyrl"}, //Azerbaijani (Cyrillic)
136 {50, "az-Arab"}, //Azerbaijani (Arabic)
137 {51, "hy"}, //Armenian
138 {52, "ka"}, //Georgian
139 {53, "mo"}, //Moldavian
140 {54, "ky"}, //Kirghiz
141 {55, "tg"}, //Tajiki
142 {56, "tk"}, //Turkmen
143 {57, "mn-Mong"}, //Mongolian (Traditional)
144 {58, "mn-Cyrl"}, //Mongolian (Cyrillic)
145 {59, "ps"}, //Pashto
146 {60, "ku"}, //Kurdish
147 {61, "ks"}, //Kashmiri
148 {62, "sd"}, //Sindhi
149 {63, "bo"}, //Tibetan
150 {64, "ne"}, //Nepali
151 {65, "sa"}, //Sanskrit
152 {66, "mr"}, //Marathi
153 {67, "bn"}, //Bengali
154 {68, "as"}, //Assamese
155 {69, "gu"}, //Gujarati
156 {70, "pa"}, //Punjabi
157 {71, "or"}, //Oriya
158 {72, "ml"}, //Malayalam
159 {73, "kn"}, //Kannada
160 {74, "ta"}, //Tamil
161 {75, "te"}, //Telugu
162 {76, "si"}, //Sinhalese
163 {77, "my"}, //Burmese
164 {78, "km"}, //Khmer
165 {79, "lo"}, //Lao
166 {80, "vi"}, //Vietnamese
167 {81, "id"}, //Indonesian
168 {82, "tl"}, //Tagalog
169 {83, "ms-Latn"}, //Malay (Roman)
170 {84, "ms-Arab"}, //Malay (Arabic)
171 {85, "am"}, //Amharic
172 {86, "ti"}, //Tigrinya
173 {87, "om"}, //Oromo
174 {88, "so"}, //Somali
175 {89, "sw"}, //Swahili
176 {90, "rw"}, //Kinyarwanda/Ruanda
177 {91, "rn"}, //Rundi
178 {92, "ny"}, //Nyanja/Chewa
179 {93, "mg"}, //Malagasy
180 {94, "eo"}, //Esperanto
181 {128, "cy"}, //Welsh
182 {129, "eu"}, //Basque
183 {130, "ca"}, //Catalan
184 {131, "la"}, //Latin
185 {132, "qu"}, //Quechua
186 {133, "gn"}, //Guarani
187 {134, "ay"}, //Aymara
188 {135, "tt"}, //Tatar
189 {136, "ug"}, //Uighur
190 {137, "dz"}, //Dzongkha
191 {138, "jv-Latn"}, //Javanese (Roman)
192 {139, "su-Latn"}, //Sundanese (Roman)
193 {140, "gl"}, //Galician
194 {141, "af"}, //Afrikaans
195 {142, "br"}, //Breton
196 {143, "iu"}, //Inuktitut
197 {144, "gd"}, //Scottish (Gaelic)
198 {145, "gv"}, //Manx (Gaelic)
199 {146, "ga"}, //Irish (Gaelic with Lenition)
200 {147, "to"}, //Tongan
201 {148, "el"}, //Greek (Polytonic) Note: ISO 15924 does not have an equivalent script name.
202 {149, "kl"}, //Greenlandic
203 {150, "az-Latn"}, //Azerbaijani (Roman)
204 {151, "nn"}, //Nynorsk
205
206 /** A mapping from Windows LCID to BCP 47 codes.
207 * This list is the sorted, curated output of tools/win_lcid.cpp.
208 * Note that these are sorted by value for quick binary lookup, and not logically by lsb.
209 * The 'bare' language ids (e.g. 0x0001 for Arabic) are ommitted
210 * as they do not appear as valid language ids in the OpenType specification.
211 */
212 { 0x0401, "ar-SA" }, //Arabic
213 { 0x0402, "bg-BG" }, //Bulgarian
214 { 0x0403, "ca-ES" }, //Catalan
215 { 0x0404, "zh-TW" }, //Chinese (Traditional)
216 { 0x0405, "cs-CZ" }, //Czech
217 { 0x0406, "da-DK" }, //Danish
218 { 0x0407, "de-DE" }, //German
219 { 0x0408, "el-GR" }, //Greek
220 { 0x0409, "en-US" }, //English
221 { 0x040a, "es-ES_tradnl" }, //Spanish
222 { 0x040b, "fi-FI" }, //Finnish
223 { 0x040c, "fr-FR" }, //French
224 { 0x040d, "he-IL" }, //Hebrew
225 { 0x040d, "he" }, //Hebrew
226 { 0x040e, "hu-HU" }, //Hungarian
227 { 0x040e, "hu" }, //Hungarian
228 { 0x040f, "is-IS" }, //Icelandic
229 { 0x0410, "it-IT" }, //Italian
230 { 0x0411, "ja-JP" }, //Japanese
231 { 0x0412, "ko-KR" }, //Korean
232 { 0x0413, "nl-NL" }, //Dutch
233 { 0x0414, "nb-NO" }, //Norwegian (Bokm�l)
234 { 0x0415, "pl-PL" }, //Polish
235 { 0x0416, "pt-BR" }, //Portuguese
236 { 0x0417, "rm-CH" }, //Romansh
237 { 0x0418, "ro-RO" }, //Romanian
238 { 0x0419, "ru-RU" }, //Russian
239 { 0x041a, "hr-HR" }, //Croatian
240 { 0x041b, "sk-SK" }, //Slovak
241 { 0x041c, "sq-AL" }, //Albanian
242 { 0x041d, "sv-SE" }, //Swedish
243 { 0x041e, "th-TH" }, //Thai
244 { 0x041f, "tr-TR" }, //Turkish
245 { 0x0420, "ur-PK" }, //Urdu
246 { 0x0421, "id-ID" }, //Indonesian
247 { 0x0422, "uk-UA" }, //Ukrainian
248 { 0x0423, "be-BY" }, //Belarusian
249 { 0x0424, "sl-SI" }, //Slovenian
250 { 0x0425, "et-EE" }, //Estonian
251 { 0x0426, "lv-LV" }, //Latvian
252 { 0x0427, "lt-LT" }, //Lithuanian
253 { 0x0428, "tg-Cyrl-TJ" }, //Tajik (Cyrillic)
254 { 0x0429, "fa-IR" }, //Persian
255 { 0x042a, "vi-VN" }, //Vietnamese
256 { 0x042b, "hy-AM" }, //Armenian
257 { 0x042c, "az-Latn-AZ" }, //Azeri (Latin)
258 { 0x042d, "eu-ES" }, //Basque
259 { 0x042e, "hsb-DE" }, //Upper Sorbian
260 { 0x042f, "mk-MK" }, //Macedonian (FYROM)
261 { 0x0432, "tn-ZA" }, //Setswana
262 { 0x0434, "xh-ZA" }, //isiXhosa
263 { 0x0435, "zu-ZA" }, //isiZulu
264 { 0x0436, "af-ZA" }, //Afrikaans
265 { 0x0437, "ka-GE" }, //Georgian
266 { 0x0438, "fo-FO" }, //Faroese
267 { 0x0439, "hi-IN" }, //Hindi
268 { 0x043a, "mt-MT" }, //Maltese
269 { 0x043b, "se-NO" }, //Sami (Northern)
270 { 0x043e, "ms-MY" }, //Malay
271 { 0x043f, "kk-KZ" }, //Kazakh
272 { 0x0440, "ky-KG" }, //Kyrgyz
273 { 0x0441, "sw-KE" }, //Kiswahili
274 { 0x0442, "tk-TM" }, //Turkmen
275 { 0x0443, "uz-Latn-UZ" }, //Uzbek (Latin)
276 { 0x0443, "uz" }, //Uzbek
277 { 0x0444, "tt-RU" }, //Tatar
278 { 0x0445, "bn-IN" }, //Bengali
279 { 0x0446, "pa-IN" }, //Punjabi
280 { 0x0447, "gu-IN" }, //Gujarati
281 { 0x0448, "or-IN" }, //Oriya
282 { 0x0449, "ta-IN" }, //Tamil
283 { 0x044a, "te-IN" }, //Telugu
284 { 0x044b, "kn-IN" }, //Kannada
285 { 0x044c, "ml-IN" }, //Malayalam
286 { 0x044d, "as-IN" }, //Assamese
287 { 0x044e, "mr-IN" }, //Marathi
288 { 0x044f, "sa-IN" }, //Sanskrit
289 { 0x0450, "mn-Cyrl" }, //Mongolian (Cyrillic)
290 { 0x0451, "bo-CN" }, //Tibetan
291 { 0x0452, "cy-GB" }, //Welsh
292 { 0x0453, "km-KH" }, //Khmer
293 { 0x0454, "lo-LA" }, //Lao
294 { 0x0456, "gl-ES" }, //Galician
295 { 0x0457, "kok-IN" }, //Konkani
296 { 0x045a, "syr-SY" }, //Syriac
297 { 0x045b, "si-LK" }, //Sinhala
298 { 0x045d, "iu-Cans-CA" }, //Inuktitut (Syllabics)
299 { 0x045e, "am-ET" }, //Amharic
300 { 0x0461, "ne-NP" }, //Nepali
301 { 0x0462, "fy-NL" }, //Frisian
302 { 0x0463, "ps-AF" }, //Pashto
303 { 0x0464, "fil-PH" }, //Filipino
304 { 0x0465, "dv-MV" }, //Divehi
305 { 0x0468, "ha-Latn-NG" }, //Hausa (Latin)
306 { 0x046a, "yo-NG" }, //Yoruba
307 { 0x046b, "quz-BO" }, //Quechua
308 { 0x046c, "nso-ZA" }, //Sesotho sa Leboa
309 { 0x046d, "ba-RU" }, //Bashkir
310 { 0x046e, "lb-LU" }, //Luxembourgish
311 { 0x046f, "kl-GL" }, //Greenlandic
312 { 0x0470, "ig-NG" }, //Igbo
313 { 0x0478, "ii-CN" }, //Yi
314 { 0x047a, "arn-CL" }, //Mapudungun
315 { 0x047c, "moh-CA" }, //Mohawk
316 { 0x047e, "br-FR" }, //Breton
317 { 0x0480, "ug-CN" }, //Uyghur
318 { 0x0481, "mi-NZ" }, //Maori
319 { 0x0482, "oc-FR" }, //Occitan
320 { 0x0483, "co-FR" }, //Corsican
321 { 0x0484, "gsw-FR" }, //Alsatian
322 { 0x0485, "sah-RU" }, //Yakut
323 { 0x0486, "qut-GT" }, //K'iche
324 { 0x0487, "rw-RW" }, //Kinyarwanda
325 { 0x0488, "wo-SN" }, //Wolof
326 { 0x048c, "prs-AF" }, //Dari
327 { 0x0491, "gd-GB" }, //Scottish Gaelic
328 { 0x0801, "ar-IQ" }, //Arabic
329 { 0x0804, "zh-Hans" }, //Chinese (Simplified)
330 { 0x0807, "de-CH" }, //German
331 { 0x0809, "en-GB" }, //English
332 { 0x080a, "es-MX" }, //Spanish
333 { 0x080c, "fr-BE" }, //French
334 { 0x0810, "it-CH" }, //Italian
335 { 0x0813, "nl-BE" }, //Dutch
336 { 0x0814, "nn-NO" }, //Norwegian (Nynorsk)
337 { 0x0816, "pt-PT" }, //Portuguese
338 { 0x081a, "sr-Latn-CS" }, //Serbian (Latin)
339 { 0x081d, "sv-FI" }, //Swedish
340 { 0x082c, "az-Cyrl-AZ" }, //Azeri (Cyrillic)
341 { 0x082e, "dsb-DE" }, //Lower Sorbian
342 { 0x082e, "dsb" }, //Lower Sorbian
343 { 0x083b, "se-SE" }, //Sami (Northern)
344 { 0x083c, "ga-IE" }, //Irish
345 { 0x083e, "ms-BN" }, //Malay
346 { 0x0843, "uz-Cyrl-UZ" }, //Uzbek (Cyrillic)
347 { 0x0845, "bn-BD" }, //Bengali
348 { 0x0850, "mn-Mong-CN" }, //Mongolian (Traditional Mongolian)
349 { 0x085d, "iu-Latn-CA" }, //Inuktitut (Latin)
350 { 0x085f, "tzm-Latn-DZ" }, //Tamazight (Latin)
351 { 0x086b, "quz-EC" }, //Quechua
352 { 0x0c01, "ar-EG" }, //Arabic
353 { 0x0c04, "zh-Hant" }, //Chinese (Traditional)
354 { 0x0c07, "de-AT" }, //German
355 { 0x0c09, "en-AU" }, //English
356 { 0x0c0a, "es-ES" }, //Spanish
357 { 0x0c0c, "fr-CA" }, //French
358 { 0x0c1a, "sr-Cyrl-CS" }, //Serbian (Cyrillic)
359 { 0x0c3b, "se-FI" }, //Sami (Northern)
360 { 0x0c6b, "quz-PE" }, //Quechua
361 { 0x1001, "ar-LY" }, //Arabic
362 { 0x1004, "zh-SG" }, //Chinese (Simplified)
363 { 0x1007, "de-LU" }, //German
364 { 0x1009, "en-CA" }, //English
365 { 0x100a, "es-GT" }, //Spanish
366 { 0x100c, "fr-CH" }, //French
367 { 0x101a, "hr-BA" }, //Croatian (Latin)
368 { 0x103b, "smj-NO" }, //Sami (Lule)
369 { 0x1401, "ar-DZ" }, //Arabic
370 { 0x1404, "zh-MO" }, //Chinese (Traditional)
371 { 0x1407, "de-LI" }, //German
372 { 0x1409, "en-NZ" }, //English
373 { 0x140a, "es-CR" }, //Spanish
374 { 0x140c, "fr-LU" }, //French
375 { 0x141a, "bs-Latn-BA" }, //Bosnian (Latin)
376 { 0x141a, "bs" }, //Bosnian
377 { 0x143b, "smj-SE" }, //Sami (Lule)
378 { 0x143b, "smj" }, //Sami (Lule)
379 { 0x1801, "ar-MA" }, //Arabic
380 { 0x1809, "en-IE" }, //English
381 { 0x180a, "es-PA" }, //Spanish
382 { 0x180c, "fr-MC" }, //French
383 { 0x181a, "sr-Latn-BA" }, //Serbian (Latin)
384 { 0x183b, "sma-NO" }, //Sami (Southern)
385 { 0x1c01, "ar-TN" }, //Arabic
386 { 0x1c09, "en-ZA" }, //English
387 { 0x1c0a, "es-DO" }, //Spanish
388 { 0x1c1a, "sr-Cyrl-BA" }, //Serbian (Cyrillic)
389 { 0x1c3b, "sma-SE" }, //Sami (Southern)
390 { 0x1c3b, "sma" }, //Sami (Southern)
391 { 0x2001, "ar-OM" }, //Arabic
392 { 0x2009, "en-JM" }, //English
393 { 0x200a, "es-VE" }, //Spanish
394 { 0x201a, "bs-Cyrl-BA" }, //Bosnian (Cyrillic)
395 { 0x201a, "bs-Cyrl" }, //Bosnian (Cyrillic)
396 { 0x203b, "sms-FI" }, //Sami (Skolt)
397 { 0x203b, "sms" }, //Sami (Skolt)
398 { 0x2401, "ar-YE" }, //Arabic
399 { 0x2409, "en-029" }, //English
400 { 0x240a, "es-CO" }, //Spanish
401 { 0x241a, "sr-Latn-RS" }, //Serbian (Latin)
402 { 0x243b, "smn-FI" }, //Sami (Inari)
403 { 0x2801, "ar-SY" }, //Arabic
404 { 0x2809, "en-BZ" }, //English
405 { 0x280a, "es-PE" }, //Spanish
406 { 0x281a, "sr-Cyrl-RS" }, //Serbian (Cyrillic)
407 { 0x2c01, "ar-JO" }, //Arabic
408 { 0x2c09, "en-TT" }, //English
409 { 0x2c0a, "es-AR" }, //Spanish
410 { 0x2c1a, "sr-Latn-ME" }, //Serbian (Latin)
411 { 0x3001, "ar-LB" }, //Arabic
412 { 0x3009, "en-ZW" }, //English
413 { 0x300a, "es-EC" }, //Spanish
414 { 0x301a, "sr-Cyrl-ME" }, //Serbian (Cyrillic)
415 { 0x3401, "ar-KW" }, //Arabic
416 { 0x3409, "en-PH" }, //English
417 { 0x340a, "es-CL" }, //Spanish
418 { 0x3801, "ar-AE" }, //Arabic
419 { 0x380a, "es-UY" }, //Spanish
420 { 0x3c01, "ar-BH" }, //Arabic
421 { 0x3c0a, "es-PY" }, //Spanish
422 { 0x4001, "ar-QA" }, //Arabic
423 { 0x4009, "en-IN" }, //English
424 { 0x400a, "es-BO" }, //Spanish
425 { 0x4409, "en-MY" }, //English
426 { 0x440a, "es-SV" }, //Spanish
427 { 0x4809, "en-SG" }, //English
428 { 0x480a, "es-HN" }, //Spanish
429 { 0x4c0a, "es-NI" }, //Spanish
430 { 0x500a, "es-PR" }, //Spanish
431 { 0x540a, "es-US" }, //Spanish
432 };
433
434 namespace {
BCP47FromLanguageIdLess(const BCP47FromLanguageId & a,const BCP47FromLanguageId & b)435 bool BCP47FromLanguageIdLess(const BCP47FromLanguageId& a, const BCP47FromLanguageId& b) {
436 return a.languageID < b.languageID;
437 }
438 }
439
next(SkOTTableName::Iterator::Record & record)440 bool SkOTTableName::Iterator::next(SkOTTableName::Iterator::Record& record) {
441 const size_t nameRecordsCount = SkEndian_SwapBE16(fName.count);
442 const SkOTTableName::Record* nameRecords = SkTAfter<const SkOTTableName::Record>(&fName);
443 const SkOTTableName::Record* nameRecord;
444
445 // Find the next record which matches the requested type.
446 do {
447 if (fIndex >= nameRecordsCount) {
448 return false;
449 }
450
451 nameRecord = &nameRecords[fIndex];
452 ++fIndex;
453 } while (fType != -1 && nameRecord->nameID.fontSpecific != fType);
454
455 record.type = nameRecord->nameID.fontSpecific;
456
457 const uint16_t stringTableOffset = SkEndian_SwapBE16(fName.stringOffset);
458 const char* stringTable = SkTAddOffset<const char>(&fName, stringTableOffset);
459
460 // Decode the name into UTF-8.
461 const uint16_t nameOffset = SkEndian_SwapBE16(nameRecord->offset);
462 const uint16_t nameLength = SkEndian_SwapBE16(nameRecord->length);
463 const char* nameString = SkTAddOffset<const char>(stringTable, nameOffset);
464 switch (nameRecord->platformID.value) {
465 case SkOTTableName::Record::PlatformID::Windows:
466 if (SkOTTableName::Record::EncodingID::Windows::UnicodeBMPUCS2
467 != nameRecord->encodingID.windows.value
468 && SkOTTableName::Record::EncodingID::Windows::UnicodeUCS4
469 != nameRecord->encodingID.windows.value
470 && SkOTTableName::Record::EncodingID::Windows::Symbol
471 != nameRecord->encodingID.windows.value)
472 {
473 record.name.reset();
474 break;
475 }
476 case SkOTTableName::Record::PlatformID::Unicode:
477 case SkOTTableName::Record::PlatformID::ISO:
478 SkStringFromUTF16BE((const uint16_t*)nameString, nameLength, record.name);
479 break;
480
481 case SkOTTableName::Record::PlatformID::Macintosh:
482 // TODO: need better decoding, especially on Mac.
483 if (SkOTTableName::Record::EncodingID::Macintosh::Roman
484 != nameRecord->encodingID.macintosh.value)
485 {
486 record.name.reset();
487 break;
488 }
489 SkStringFromMacRoman((const uint8_t*)nameString, nameLength, record.name);
490 break;
491
492 case SkOTTableName::Record::PlatformID::Custom:
493 // These should never appear in a 'name' table.
494 default:
495 SkASSERT(false);
496 record.name.reset();
497 break;
498 }
499
500 // Determine the language.
501 const uint16_t languageID = SkEndian_SwapBE16(nameRecord->languageID.languageTagID);
502
503 // Handle format 1 languages.
504 if (SkOTTableName::format_1 == fName.format && languageID >= 0x8000) {
505 const uint16_t languageTagRecordIndex = languageID - 0x8000;
506
507 const SkOTTableName::Format1Ext* format1ext =
508 SkTAfter<const SkOTTableName::Format1Ext>(nameRecords, nameRecordsCount);
509
510 if (languageTagRecordIndex < SkEndian_SwapBE16(format1ext->langTagCount)) {
511 const SkOTTableName::Format1Ext::LangTagRecord* languageTagRecord =
512 SkTAfter<const SkOTTableName::Format1Ext::LangTagRecord>(format1ext);
513
514 uint16_t offset = SkEndian_SwapBE16(languageTagRecord[languageTagRecordIndex].offset);
515 uint16_t length = SkEndian_SwapBE16(languageTagRecord[languageTagRecordIndex].length);
516 const uint16_t* string = SkTAddOffset<const uint16_t>(stringTable, offset);
517 SkStringFromUTF16BE(string, length, record.language);
518 return true;
519 }
520 }
521
522 // Handle format 0 languages, translating them into BCP 47.
523 const BCP47FromLanguageId target = { languageID, "" };
524 int languageIndex = SkTSearch<BCP47FromLanguageId, BCP47FromLanguageIdLess>(
525 BCP47FromLanguageID, SK_ARRAY_COUNT(BCP47FromLanguageID), target, sizeof(target));
526 if (languageIndex >= 0) {
527 record.language = BCP47FromLanguageID[languageIndex].bcp47;
528 return true;
529 }
530
531 // Unknown language, return the BCP 47 code 'und' for 'undetermined'.
532 record.language = "und";
533 return true;
534 }
535