Line data Source code
1 : /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
2 : /* This Source Code Form is subject to the terms of the Mozilla Public
3 : * License, v. 2.0. If a copy of the MPL was not distributed with this
4 : * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
5 :
6 : #include "nsUnicodeRange.h"
7 :
8 : /**********************************************************************
9 : * Unicode subranges as defined in unicode 3.0
10 : * x-western -> latin
11 : * 0000 - 036f
12 : * 1e00 - 1eff
13 : * 2000 - 206f (general punctuation)
14 : * 20a0 - 20cf (currency symbols)
15 : * 2100 - 214f (letterlike symbols)
16 : * 2150 - 218f (Number Forms)
17 : * el -> greek
18 : * 0370 - 03ff
19 : * 1f00 - 1fff
20 : * x-cyrillic -> cyrillic
21 : * 0400 - 04ff
22 : * he -> hebrew
23 : * 0590 - 05ff
24 : * ar -> arabic
25 : * 0600 - 06ff
26 : * fb50 - fdff (arabic presentation forms)
27 : * fe70 - feff (arabic presentation forms b)
28 : * th - thai
29 : * 0e00 - 0e7f
30 : * ko -> korean
31 : * ac00 - d7af (hangul Syllables)
32 : * 1100 - 11ff (jamo)
33 : * 3130 - 318f (hangul compatibility jamo)
34 : * ja
35 : * 3040 - 309f (hiragana)
36 : * 30a0 - 30ff (katakana)
37 : * zh-CN
38 : * zh-TW
39 : *
40 : * CJK
41 : * 3100 - 312f (bopomofo)
42 : * 31a0 - 31bf (bopomofo extended)
43 : * 3000 - 303f (CJK Symbols and Punctuation)
44 : * 2e80 - 2eff (CJK radicals supplement)
45 : * 2f00 - 2fdf (Kangxi Radicals)
46 : * 2ff0 - 2fff (Ideographic Description Characters)
47 : * 3190 - 319f (kanbun)
48 : * 3200 - 32ff (Enclosed CJK letters and Months)
49 : * 3300 - 33ff (CJK compatibility)
50 : * 3400 - 4dbf (CJK Unified Ideographs Extension A)
51 : * 4e00 - 9faf (CJK Unified Ideographs)
52 : * f900 - fa5f (CJK Compatibility Ideographs)
53 : * fe30 - fe4f (CJK compatibility Forms)
54 : * ff00 - ffef (halfwidth and fullwidth forms)
55 : *
56 : * Armenian
57 : * 0530 - 058f
58 : * Sriac
59 : * 0700 - 074f
60 : * Thaana
61 : * 0780 - 07bf
62 : * Devanagari
63 : * 0900 - 097f
64 : * Bengali
65 : * 0980 - 09ff
66 : * Gurmukhi
67 : * 0a00 - 0a7f
68 : * Gujarati
69 : * 0a80 - 0aff
70 : * Oriya
71 : * 0b00 - 0b7f
72 : * Tamil
73 : * 0b80 - 0bff
74 : * Telugu
75 : * 0c00 - 0c7f
76 : * Kannada
77 : * 0c80 - 0cff
78 : * Malayalam
79 : * 0d00 - 0d7f
80 : * Sinhala
81 : * 0d80 - 0def
82 : * Lao
83 : * 0e80 - 0eff
84 : * Tibetan
85 : * 0f00 - 0fbf
86 : * Myanmar
87 : * 1000 - 109f
88 : * Georgian
89 : * 10a0 - 10ff
90 : * Ethiopic
91 : * 1200 - 137f
92 : * Cherokee
93 : * 13a0 - 13ff
94 : * Canadian Aboriginal Syllabics
95 : * 1400 - 167f
96 : * Ogham
97 : * 1680 - 169f
98 : * Runic
99 : * 16a0 - 16ff
100 : * Khmer
101 : * 1780 - 17ff
102 : * Mongolian
103 : * 1800 - 18af
104 : * Misc - superscripts and subscripts
105 : * 2070 - 209f
106 : * Misc - Combining Diacritical Marks for Symbols
107 : * 20d0 - 20ff
108 : * Misc - Arrows
109 : * 2190 - 21ff
110 : * Misc - Mathematical Operators
111 : * 2200 - 22ff
112 : * Misc - Miscellaneous Technical
113 : * 2300 - 23ff
114 : * Misc - Control picture
115 : * 2400 - 243f
116 : * Misc - Optical character recognition
117 : * 2440 - 2450
118 : * Misc - Enclose Alphanumerics
119 : * 2460 - 24ff
120 : * Misc - Box Drawing
121 : * 2500 - 257f
122 : * Misc - Block Elements
123 : * 2580 - 259f
124 : * Misc - Geometric Shapes
125 : * 25a0 - 25ff
126 : * Misc - Miscellaneous Symbols
127 : * 2600 - 267f
128 : * Misc - Dingbats
129 : * 2700 - 27bf
130 : * Misc - Braille Patterns
131 : * 2800 - 28ff
132 : * Yi Syllables
133 : * a000 - a48f
134 : * Yi radicals
135 : * a490 - a4cf
136 : * Alphabetic Presentation Forms
137 : * fb00 - fb4f
138 : * Misc - Combining half Marks
139 : * fe20 - fe2f
140 : * Misc - small form variants
141 : * fe50 - fe6f
142 : * Misc - Specials
143 : * fff0 - ffff
144 : *********************************************************************/
145 :
146 :
147 :
148 : #define NUM_OF_SUBTABLES 10
149 : #define SUBTABLE_SIZE 16
150 :
151 : static const uint8_t gUnicodeSubrangeTable[NUM_OF_SUBTABLES][SUBTABLE_SIZE] =
152 : {
153 : { // table for X---
154 : kRangeTableBase+1, //u0xxx
155 : kRangeTableBase+2, //u1xxx
156 : kRangeTableBase+3, //u2xxx
157 : kRangeSetCJK, //u3xxx
158 : kRangeSetCJK, //u4xxx
159 : kRangeSetCJK, //u5xxx
160 : kRangeSetCJK, //u6xxx
161 : kRangeSetCJK, //u7xxx
162 : kRangeSetCJK, //u8xxx
163 : kRangeSetCJK, //u9xxx
164 : kRangeTableBase+4, //uaxxx
165 : kRangeKorean, //ubxxx
166 : kRangeKorean, //ucxxx
167 : kRangeTableBase+5, //udxxx
168 : kRangePrivate, //uexxx
169 : kRangeTableBase+6 //ufxxx
170 : },
171 : { //table for 0X--
172 : kRangeSetLatin, //u00xx
173 : kRangeSetLatin, //u01xx
174 : kRangeSetLatin, //u02xx
175 : kRangeGreek, //u03xx XXX 0300-036f is in fact kRangeCombiningDiacriticalMarks
176 : kRangeCyrillic, //u04xx
177 : kRangeTableBase+7, //u05xx, includes Cyrillic supplement, Hebrew, and Armenian
178 : kRangeArabic, //u06xx
179 : kRangeTertiaryTable, //u07xx
180 : kRangeUnassigned, //u08xx
181 : kRangeTertiaryTable, //u09xx
182 : kRangeTertiaryTable, //u0axx
183 : kRangeTertiaryTable, //u0bxx
184 : kRangeTertiaryTable, //u0cxx
185 : kRangeTertiaryTable, //u0dxx
186 : kRangeTertiaryTable, //u0exx
187 : kRangeTibetan //u0fxx
188 : },
189 : { //table for 1x--
190 : kRangeTertiaryTable, //u10xx
191 : kRangeKorean, //u11xx
192 : kRangeEthiopic, //u12xx
193 : kRangeTertiaryTable, //u13xx
194 : kRangeCanadian, //u14xx
195 : kRangeCanadian, //u15xx
196 : kRangeTertiaryTable, //u16xx
197 : kRangeKhmer, //u17xx
198 : kRangeMongolian, //u18xx
199 : kRangeUnassigned, //u19xx
200 : kRangeUnassigned, //u1axx
201 : kRangeUnassigned, //u1bxx
202 : kRangeUnassigned, //u1cxx
203 : kRangeUnassigned, //u1dxx
204 : kRangeSetLatin, //u1exx
205 : kRangeGreek //u1fxx
206 : },
207 : { //table for 2x--
208 : kRangeSetLatin, //u20xx
209 : kRangeSetLatin, //u21xx
210 : kRangeMathOperators, //u22xx
211 : kRangeMiscTechnical, //u23xx
212 : kRangeControlOpticalEnclose, //u24xx
213 : kRangeBoxBlockGeometrics, //u25xx
214 : kRangeMiscSymbols, //u26xx
215 : kRangeDingbats, //u27xx
216 : kRangeBraillePattern, //u28xx
217 : kRangeUnassigned, //u29xx
218 : kRangeUnassigned, //u2axx
219 : kRangeUnassigned, //u2bxx
220 : kRangeUnassigned, //u2cxx
221 : kRangeUnassigned, //u2dxx
222 : kRangeSetCJK, //u2exx
223 : kRangeSetCJK //u2fxx
224 : },
225 : { //table for ax--
226 : kRangeYi, //ua0xx
227 : kRangeYi, //ua1xx
228 : kRangeYi, //ua2xx
229 : kRangeYi, //ua3xx
230 : kRangeYi, //ua4xx
231 : kRangeUnassigned, //ua5xx
232 : kRangeUnassigned, //ua6xx
233 : kRangeUnassigned, //ua7xx
234 : kRangeUnassigned, //ua8xx
235 : kRangeUnassigned, //ua9xx
236 : kRangeUnassigned, //uaaxx
237 : kRangeUnassigned, //uabxx
238 : kRangeKorean, //uacxx
239 : kRangeKorean, //uadxx
240 : kRangeKorean, //uaexx
241 : kRangeKorean //uafxx
242 : },
243 : { //table for dx--
244 : kRangeKorean, //ud0xx
245 : kRangeKorean, //ud1xx
246 : kRangeKorean, //ud2xx
247 : kRangeKorean, //ud3xx
248 : kRangeKorean, //ud4xx
249 : kRangeKorean, //ud5xx
250 : kRangeKorean, //ud6xx
251 : kRangeKorean, //ud7xx
252 : kRangeSurrogate, //ud8xx
253 : kRangeSurrogate, //ud9xx
254 : kRangeSurrogate, //udaxx
255 : kRangeSurrogate, //udbxx
256 : kRangeSurrogate, //udcxx
257 : kRangeSurrogate, //uddxx
258 : kRangeSurrogate, //udexx
259 : kRangeSurrogate //udfxx
260 : },
261 : { // table for fx--
262 : kRangePrivate, //uf0xx
263 : kRangePrivate, //uf1xx
264 : kRangePrivate, //uf2xx
265 : kRangePrivate, //uf3xx
266 : kRangePrivate, //uf4xx
267 : kRangePrivate, //uf5xx
268 : kRangePrivate, //uf6xx
269 : kRangePrivate, //uf7xx
270 : kRangePrivate, //uf8xx
271 : kRangeSetCJK, //uf9xx
272 : kRangeSetCJK, //ufaxx
273 : kRangeArabic, //ufbxx, includes alphabic presentation form
274 : kRangeArabic, //ufcxx
275 : kRangeArabic, //ufdxx
276 : kRangeTableBase+8, //ufexx
277 : kRangeTableBase+9 //uffxx, halfwidth and fullwidth forms, includes Specials
278 : },
279 : { //table for 0x0500 - 0x05ff
280 : kRangeCyrillic, //u050x
281 : kRangeCyrillic, //u051x
282 : kRangeCyrillic, //u052x
283 : kRangeArmenian, //u053x
284 : kRangeArmenian, //u054x
285 : kRangeArmenian, //u055x
286 : kRangeArmenian, //u056x
287 : kRangeArmenian, //u057x
288 : kRangeArmenian, //u058x
289 : kRangeHebrew, //u059x
290 : kRangeHebrew, //u05ax
291 : kRangeHebrew, //u05bx
292 : kRangeHebrew, //u05cx
293 : kRangeHebrew, //u05dx
294 : kRangeHebrew, //u05ex
295 : kRangeHebrew //u05fx
296 : },
297 : { //table for 0xfe00 - 0xfeff
298 : kRangeSetCJK, //ufe0x
299 : kRangeSetCJK, //ufe1x
300 : kRangeSetCJK, //ufe2x
301 : kRangeSetCJK, //ufe3x
302 : kRangeSetCJK, //ufe4x
303 : kRangeSetCJK, //ufe5x
304 : kRangeSetCJK, //ufe6x
305 : kRangeArabic, //ufe7x
306 : kRangeArabic, //ufe8x
307 : kRangeArabic, //ufe9x
308 : kRangeArabic, //ufeax
309 : kRangeArabic, //ufebx
310 : kRangeArabic, //ufecx
311 : kRangeArabic, //ufedx
312 : kRangeArabic, //ufeex
313 : kRangeArabic //ufefx
314 : },
315 : { //table for 0xff00 - 0xffff
316 : kRangeSetCJK, //uff0x, fullwidth latin
317 : kRangeSetCJK, //uff1x, fullwidth latin
318 : kRangeSetCJK, //uff2x, fullwidth latin
319 : kRangeSetCJK, //uff3x, fullwidth latin
320 : kRangeSetCJK, //uff4x, fullwidth latin
321 : kRangeSetCJK, //uff5x, fullwidth latin
322 : kRangeSetCJK, //uff6x, halfwidth katakana
323 : kRangeSetCJK, //uff7x, halfwidth katakana
324 : kRangeSetCJK, //uff8x, halfwidth katakana
325 : kRangeSetCJK, //uff9x, halfwidth katakana
326 : kRangeSetCJK, //uffax, halfwidth hangul jamo
327 : kRangeSetCJK, //uffbx, halfwidth hangul jamo
328 : kRangeSetCJK, //uffcx, halfwidth hangul jamo
329 : kRangeSetCJK, //uffdx, halfwidth hangul jamo
330 : kRangeSetCJK, //uffex, fullwidth symbols
331 : kRangeSpecials, //ufffx, Specials
332 : },
333 : };
334 :
335 : // Most scripts between U+0700 and U+16FF are assigned a chunk of 128 (0x80)
336 : // code points so that the number of entries in the tertiary range
337 : // table for that range is obtained by dividing (0x1700 - 0x0700) by 128.
338 : // Exceptions: Ethiopic, Tibetan, Hangul Jamo and Canadian aboriginal
339 : // syllabaries take multiple chunks and Ogham and Runic share a single chunk.
340 : #define TERTIARY_TABLE_SIZE ((0x1700 - 0x0700) / 0x80)
341 :
342 : static const uint8_t gUnicodeTertiaryRangeTable[TERTIARY_TABLE_SIZE] =
343 : { //table for 0x0700 - 0x1600
344 : kRangeSyriac, //u070x
345 : kRangeThaana, //u078x
346 : kRangeUnassigned, //u080x place holder(resolved in the 2ndary tab.)
347 : kRangeUnassigned, //u088x place holder(resolved in the 2ndary tab.)
348 : kRangeDevanagari, //u090x
349 : kRangeBengali, //u098x
350 : kRangeGurmukhi, //u0a0x
351 : kRangeGujarati, //u0a8x
352 : kRangeOriya, //u0b0x
353 : kRangeTamil, //u0b8x
354 : kRangeTelugu, //u0c0x
355 : kRangeKannada, //u0c8x
356 : kRangeMalayalam, //u0d0x
357 : kRangeSinhala, //u0d8x
358 : kRangeThai, //u0e0x
359 : kRangeLao, //u0e8x
360 : kRangeTibetan, //u0f0x place holder(resolved in the 2ndary tab.)
361 : kRangeTibetan, //u0f8x place holder(resolved in the 2ndary tab.)
362 : kRangeMyanmar, //u100x
363 : kRangeGeorgian, //u108x
364 : kRangeKorean, //u110x place holder(resolved in the 2ndary tab.)
365 : kRangeKorean, //u118x place holder(resolved in the 2ndary tab.)
366 : kRangeEthiopic, //u120x place holder(resolved in the 2ndary tab.)
367 : kRangeEthiopic, //u128x place holder(resolved in the 2ndary tab.)
368 : kRangeEthiopic, //u130x
369 : kRangeCherokee, //u138x
370 : kRangeCanadian, //u140x place holder(resolved in the 2ndary tab.)
371 : kRangeCanadian, //u148x place holder(resolved in the 2ndary tab.)
372 : kRangeCanadian, //u150x place holder(resolved in the 2ndary tab.)
373 : kRangeCanadian, //u158x place holder(resolved in the 2ndary tab.)
374 : kRangeCanadian, //u160x
375 : kRangeOghamRunic //u168x this contains two scripts, Ogham & Runic
376 : };
377 :
378 : // A two level index is almost enough for locating a range, with the
379 : // exception of u03xx and u05xx. Since we don't really care about range for
380 : // combining diacritical marks in our font application, they are
381 : // not discriminated further. But future adoption of this module for other use
382 : // should be aware of this limitation. The implementation can be extended if
383 : // there is such a need.
384 : // For Indic, Southeast Asian scripts and some other scripts between
385 : // U+0700 and U+16FF, it's extended to the third level.
386 0 : uint32_t FindCharUnicodeRange(uint32_t ch)
387 : {
388 : uint32_t range;
389 :
390 : // aggregate ranges for non-BMP codepoints
391 0 : if (ch > 0xFFFF) {
392 0 : uint32_t p = (ch >> 16);
393 0 : if (p == 1) {
394 0 : return kRangeSMP;
395 0 : } else if (p == 2) {
396 0 : return kRangeSetCJK;
397 : }
398 0 : return kRangeHigherPlanes;
399 : }
400 :
401 : // lookup explicit range for BMP codepoints
402 : // first general range
403 0 : range = gUnicodeSubrangeTable[0][ch >> 12];
404 :
405 : // if general range is good enough, return that
406 0 : if (range < kRangeTableBase)
407 : // we try to get a specific range
408 0 : return range;
409 :
410 : // otherwise, use subrange tables
411 0 : range = gUnicodeSubrangeTable[range - kRangeTableBase][(ch & 0x0f00) >> 8];
412 0 : if (range < kRangeTableBase)
413 0 : return range;
414 0 : if (range < kRangeTertiaryTable)
415 0 : return gUnicodeSubrangeTable[range - kRangeTableBase][(ch & 0x00f0) >> 4];
416 :
417 : // Yet another table to look at : U+0700 - U+16FF : 128 code point blocks
418 0 : return gUnicodeTertiaryRangeTable[(ch - 0x0700) >> 7];
419 9 : }
|
