1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/gfx/thebes/nsUnicodeRange.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,464 @@
1.4 +/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
1.5 +/* This Source Code Form is subject to the terms of the Mozilla Public
1.6 + * License, v. 2.0. If a copy of the MPL was not distributed with this
1.7 + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
1.8 +
1.9 +#include "nsUnicodeRange.h"
1.10 +#include "nsGkAtoms.h"
1.11 +#include "mozilla/NullPtr.h"
1.12 +
1.13 +// This table depends on unicode range definitions.
1.14 +// Each item's index must correspond unicode range value
1.15 +// eg. x-cyrillic = LangGroupTable[kRangeCyrillic]
1.16 +static nsIAtom **gUnicodeRangeToLangGroupAtomTable[] =
1.17 +{
1.18 + &nsGkAtoms::x_cyrillic,
1.19 + &nsGkAtoms::el_,
1.20 + &nsGkAtoms::tr,
1.21 + &nsGkAtoms::he,
1.22 + &nsGkAtoms::ar,
1.23 + &nsGkAtoms::x_baltic,
1.24 + &nsGkAtoms::th,
1.25 + &nsGkAtoms::ko,
1.26 + &nsGkAtoms::Japanese,
1.27 + &nsGkAtoms::zh_cn,
1.28 + &nsGkAtoms::zh_tw,
1.29 + &nsGkAtoms::x_devanagari,
1.30 + &nsGkAtoms::x_tamil,
1.31 + &nsGkAtoms::x_armn,
1.32 + &nsGkAtoms::x_beng,
1.33 + &nsGkAtoms::x_cans,
1.34 + &nsGkAtoms::x_ethi,
1.35 + &nsGkAtoms::x_geor,
1.36 + &nsGkAtoms::x_gujr,
1.37 + &nsGkAtoms::x_guru,
1.38 + &nsGkAtoms::x_khmr,
1.39 + &nsGkAtoms::x_mlym,
1.40 + &nsGkAtoms::x_orya,
1.41 + &nsGkAtoms::x_telu,
1.42 + &nsGkAtoms::x_knda,
1.43 + &nsGkAtoms::x_sinh,
1.44 + &nsGkAtoms::x_tibt
1.45 +};
1.46 +
1.47 +/**********************************************************************
1.48 + * Unicode subranges as defined in unicode 3.0
1.49 + * x-western, x-central-euro, tr, x-baltic -> latin
1.50 + * 0000 - 036f
1.51 + * 1e00 - 1eff
1.52 + * 2000 - 206f (general punctuation)
1.53 + * 20a0 - 20cf (currency symbols)
1.54 + * 2100 - 214f (letterlike symbols)
1.55 + * 2150 - 218f (Number Forms)
1.56 + * el -> greek
1.57 + * 0370 - 03ff
1.58 + * 1f00 - 1fff
1.59 + * x-cyrillic -> cyrillic
1.60 + * 0400 - 04ff
1.61 + * he -> hebrew
1.62 + * 0590 - 05ff
1.63 + * ar -> arabic
1.64 + * 0600 - 06ff
1.65 + * fb50 - fdff (arabic presentation forms)
1.66 + * fe70 - feff (arabic presentation forms b)
1.67 + * th - thai
1.68 + * 0e00 - 0e7f
1.69 + * ko -> korean
1.70 + * ac00 - d7af (hangul Syllables)
1.71 + * 1100 - 11ff (jamo)
1.72 + * 3130 - 318f (hangul compatibility jamo)
1.73 + * ja
1.74 + * 3040 - 309f (hiragana)
1.75 + * 30a0 - 30ff (katakana)
1.76 + * zh-CN
1.77 + * zh-TW
1.78 + *
1.79 + * CJK
1.80 + * 3100 - 312f (bopomofo)
1.81 + * 31a0 - 31bf (bopomofo extended)
1.82 + * 3000 - 303f (CJK Symbols and Punctuation)
1.83 + * 2e80 - 2eff (CJK radicals supplement)
1.84 + * 2f00 - 2fdf (Kangxi Radicals)
1.85 + * 2ff0 - 2fff (Ideographic Description Characters)
1.86 + * 3190 - 319f (kanbun)
1.87 + * 3200 - 32ff (Enclosed CJK letters and Months)
1.88 + * 3300 - 33ff (CJK compatibility)
1.89 + * 3400 - 4dbf (CJK Unified Ideographs Extension A)
1.90 + * 4e00 - 9faf (CJK Unified Ideographs)
1.91 + * f900 - fa5f (CJK Compatibility Ideographs)
1.92 + * fe30 - fe4f (CJK compatibility Forms)
1.93 + * ff00 - ffef (halfwidth and fullwidth forms)
1.94 + *
1.95 + * Armenian
1.96 + * 0530 - 058f
1.97 + * Sriac
1.98 + * 0700 - 074f
1.99 + * Thaana
1.100 + * 0780 - 07bf
1.101 + * Devanagari
1.102 + * 0900 - 097f
1.103 + * Bengali
1.104 + * 0980 - 09ff
1.105 + * Gurmukhi
1.106 + * 0a00 - 0a7f
1.107 + * Gujarati
1.108 + * 0a80 - 0aff
1.109 + * Oriya
1.110 + * 0b00 - 0b7f
1.111 + * Tamil
1.112 + * 0b80 - 0bff
1.113 + * Telugu
1.114 + * 0c00 - 0c7f
1.115 + * Kannada
1.116 + * 0c80 - 0cff
1.117 + * Malayalam
1.118 + * 0d00 - 0d7f
1.119 + * Sinhala
1.120 + * 0d80 - 0def
1.121 + * Lao
1.122 + * 0e80 - 0eff
1.123 + * Tibetan
1.124 + * 0f00 - 0fbf
1.125 + * Myanmar
1.126 + * 1000 - 109f
1.127 + * Georgian
1.128 + * 10a0 - 10ff
1.129 + * Ethiopic
1.130 + * 1200 - 137f
1.131 + * Cherokee
1.132 + * 13a0 - 13ff
1.133 + * Canadian Aboriginal Syllabics
1.134 + * 1400 - 167f
1.135 + * Ogham
1.136 + * 1680 - 169f
1.137 + * Runic
1.138 + * 16a0 - 16ff
1.139 + * Khmer
1.140 + * 1780 - 17ff
1.141 + * Mongolian
1.142 + * 1800 - 18af
1.143 + * Misc - superscripts and subscripts
1.144 + * 2070 - 209f
1.145 + * Misc - Combining Diacritical Marks for Symbols
1.146 + * 20d0 - 20ff
1.147 + * Misc - Arrows
1.148 + * 2190 - 21ff
1.149 + * Misc - Mathematical Operators
1.150 + * 2200 - 22ff
1.151 + * Misc - Miscellaneous Technical
1.152 + * 2300 - 23ff
1.153 + * Misc - Control picture
1.154 + * 2400 - 243f
1.155 + * Misc - Optical character recognition
1.156 + * 2440 - 2450
1.157 + * Misc - Enclose Alphanumerics
1.158 + * 2460 - 24ff
1.159 + * Misc - Box Drawing
1.160 + * 2500 - 257f
1.161 + * Misc - Block Elements
1.162 + * 2580 - 259f
1.163 + * Misc - Geometric Shapes
1.164 + * 25a0 - 25ff
1.165 + * Misc - Miscellaneous Symbols
1.166 + * 2600 - 267f
1.167 + * Misc - Dingbats
1.168 + * 2700 - 27bf
1.169 + * Misc - Braille Patterns
1.170 + * 2800 - 28ff
1.171 + * Yi Syllables
1.172 + * a000 - a48f
1.173 + * Yi radicals
1.174 + * a490 - a4cf
1.175 + * Alphabetic Presentation Forms
1.176 + * fb00 - fb4f
1.177 + * Misc - Combining half Marks
1.178 + * fe20 - fe2f
1.179 + * Misc - small form variants
1.180 + * fe50 - fe6f
1.181 + * Misc - Specials
1.182 + * fff0 - ffff
1.183 + *********************************************************************/
1.184 +
1.185 +
1.186 +
1.187 +#define NUM_OF_SUBTABLES 10
1.188 +#define SUBTABLE_SIZE 16
1.189 +
1.190 +static const uint8_t gUnicodeSubrangeTable[NUM_OF_SUBTABLES][SUBTABLE_SIZE] =
1.191 +{
1.192 + { // table for X---
1.193 + kRangeTableBase+1, //u0xxx
1.194 + kRangeTableBase+2, //u1xxx
1.195 + kRangeTableBase+3, //u2xxx
1.196 + kRangeSetCJK, //u3xxx
1.197 + kRangeSetCJK, //u4xxx
1.198 + kRangeSetCJK, //u5xxx
1.199 + kRangeSetCJK, //u6xxx
1.200 + kRangeSetCJK, //u7xxx
1.201 + kRangeSetCJK, //u8xxx
1.202 + kRangeSetCJK, //u9xxx
1.203 + kRangeTableBase+4, //uaxxx
1.204 + kRangeKorean, //ubxxx
1.205 + kRangeKorean, //ucxxx
1.206 + kRangeTableBase+5, //udxxx
1.207 + kRangePrivate, //uexxx
1.208 + kRangeTableBase+6 //ufxxx
1.209 + },
1.210 + { //table for 0X--
1.211 + kRangeSetLatin, //u00xx
1.212 + kRangeSetLatin, //u01xx
1.213 + kRangeSetLatin, //u02xx
1.214 + kRangeGreek, //u03xx XXX 0300-036f is in fact kRangeCombiningDiacriticalMarks
1.215 + kRangeCyrillic, //u04xx
1.216 + kRangeTableBase+7, //u05xx, includes Cyrillic supplement, Hebrew, and Armenian
1.217 + kRangeArabic, //u06xx
1.218 + kRangeTertiaryTable, //u07xx
1.219 + kRangeUnassigned, //u08xx
1.220 + kRangeTertiaryTable, //u09xx
1.221 + kRangeTertiaryTable, //u0axx
1.222 + kRangeTertiaryTable, //u0bxx
1.223 + kRangeTertiaryTable, //u0cxx
1.224 + kRangeTertiaryTable, //u0dxx
1.225 + kRangeTertiaryTable, //u0exx
1.226 + kRangeTibetan //u0fxx
1.227 + },
1.228 + { //table for 1x--
1.229 + kRangeTertiaryTable, //u10xx
1.230 + kRangeKorean, //u11xx
1.231 + kRangeEthiopic, //u12xx
1.232 + kRangeTertiaryTable, //u13xx
1.233 + kRangeCanadian, //u14xx
1.234 + kRangeCanadian, //u15xx
1.235 + kRangeTertiaryTable, //u16xx
1.236 + kRangeKhmer, //u17xx
1.237 + kRangeMongolian, //u18xx
1.238 + kRangeUnassigned, //u19xx
1.239 + kRangeUnassigned, //u1axx
1.240 + kRangeUnassigned, //u1bxx
1.241 + kRangeUnassigned, //u1cxx
1.242 + kRangeUnassigned, //u1dxx
1.243 + kRangeSetLatin, //u1exx
1.244 + kRangeGreek //u1fxx
1.245 + },
1.246 + { //table for 2x--
1.247 + kRangeSetLatin, //u20xx
1.248 + kRangeSetLatin, //u21xx
1.249 + kRangeMathOperators, //u22xx
1.250 + kRangeMiscTechnical, //u23xx
1.251 + kRangeControlOpticalEnclose, //u24xx
1.252 + kRangeBoxBlockGeometrics, //u25xx
1.253 + kRangeMiscSymbols, //u26xx
1.254 + kRangeDingbats, //u27xx
1.255 + kRangeBraillePattern, //u28xx
1.256 + kRangeUnassigned, //u29xx
1.257 + kRangeUnassigned, //u2axx
1.258 + kRangeUnassigned, //u2bxx
1.259 + kRangeUnassigned, //u2cxx
1.260 + kRangeUnassigned, //u2dxx
1.261 + kRangeSetCJK, //u2exx
1.262 + kRangeSetCJK //u2fxx
1.263 + },
1.264 + { //table for ax--
1.265 + kRangeYi, //ua0xx
1.266 + kRangeYi, //ua1xx
1.267 + kRangeYi, //ua2xx
1.268 + kRangeYi, //ua3xx
1.269 + kRangeYi, //ua4xx
1.270 + kRangeUnassigned, //ua5xx
1.271 + kRangeUnassigned, //ua6xx
1.272 + kRangeUnassigned, //ua7xx
1.273 + kRangeUnassigned, //ua8xx
1.274 + kRangeUnassigned, //ua9xx
1.275 + kRangeUnassigned, //uaaxx
1.276 + kRangeUnassigned, //uabxx
1.277 + kRangeKorean, //uacxx
1.278 + kRangeKorean, //uadxx
1.279 + kRangeKorean, //uaexx
1.280 + kRangeKorean //uafxx
1.281 + },
1.282 + { //table for dx--
1.283 + kRangeKorean, //ud0xx
1.284 + kRangeKorean, //ud1xx
1.285 + kRangeKorean, //ud2xx
1.286 + kRangeKorean, //ud3xx
1.287 + kRangeKorean, //ud4xx
1.288 + kRangeKorean, //ud5xx
1.289 + kRangeKorean, //ud6xx
1.290 + kRangeKorean, //ud7xx
1.291 + kRangeSurrogate, //ud8xx
1.292 + kRangeSurrogate, //ud9xx
1.293 + kRangeSurrogate, //udaxx
1.294 + kRangeSurrogate, //udbxx
1.295 + kRangeSurrogate, //udcxx
1.296 + kRangeSurrogate, //uddxx
1.297 + kRangeSurrogate, //udexx
1.298 + kRangeSurrogate //udfxx
1.299 + },
1.300 + { // table for fx--
1.301 + kRangePrivate, //uf0xx
1.302 + kRangePrivate, //uf1xx
1.303 + kRangePrivate, //uf2xx
1.304 + kRangePrivate, //uf3xx
1.305 + kRangePrivate, //uf4xx
1.306 + kRangePrivate, //uf5xx
1.307 + kRangePrivate, //uf6xx
1.308 + kRangePrivate, //uf7xx
1.309 + kRangePrivate, //uf8xx
1.310 + kRangeSetCJK, //uf9xx
1.311 + kRangeSetCJK, //ufaxx
1.312 + kRangeArabic, //ufbxx, includes alphabic presentation form
1.313 + kRangeArabic, //ufcxx
1.314 + kRangeArabic, //ufdxx
1.315 + kRangeTableBase+8, //ufexx
1.316 + kRangeTableBase+9 //uffxx, halfwidth and fullwidth forms, includes Specials
1.317 + },
1.318 + { //table for 0x0500 - 0x05ff
1.319 + kRangeCyrillic, //u050x
1.320 + kRangeCyrillic, //u051x
1.321 + kRangeCyrillic, //u052x
1.322 + kRangeArmenian, //u053x
1.323 + kRangeArmenian, //u054x
1.324 + kRangeArmenian, //u055x
1.325 + kRangeArmenian, //u056x
1.326 + kRangeArmenian, //u057x
1.327 + kRangeArmenian, //u058x
1.328 + kRangeHebrew, //u059x
1.329 + kRangeHebrew, //u05ax
1.330 + kRangeHebrew, //u05bx
1.331 + kRangeHebrew, //u05cx
1.332 + kRangeHebrew, //u05dx
1.333 + kRangeHebrew, //u05ex
1.334 + kRangeHebrew //u05fx
1.335 + },
1.336 + { //table for 0xfe00 - 0xfeff
1.337 + kRangeSetCJK, //ufe0x
1.338 + kRangeSetCJK, //ufe1x
1.339 + kRangeSetCJK, //ufe2x
1.340 + kRangeSetCJK, //ufe3x
1.341 + kRangeSetCJK, //ufe4x
1.342 + kRangeSetCJK, //ufe5x
1.343 + kRangeSetCJK, //ufe6x
1.344 + kRangeArabic, //ufe7x
1.345 + kRangeArabic, //ufe8x
1.346 + kRangeArabic, //ufe9x
1.347 + kRangeArabic, //ufeax
1.348 + kRangeArabic, //ufebx
1.349 + kRangeArabic, //ufecx
1.350 + kRangeArabic, //ufedx
1.351 + kRangeArabic, //ufeex
1.352 + kRangeArabic //ufefx
1.353 + },
1.354 + { //table for 0xff00 - 0xffff
1.355 + kRangeSetCJK, //uff0x, fullwidth latin
1.356 + kRangeSetCJK, //uff1x, fullwidth latin
1.357 + kRangeSetCJK, //uff2x, fullwidth latin
1.358 + kRangeSetCJK, //uff3x, fullwidth latin
1.359 + kRangeSetCJK, //uff4x, fullwidth latin
1.360 + kRangeSetCJK, //uff5x, fullwidth latin
1.361 + kRangeSetCJK, //uff6x, halfwidth katakana
1.362 + kRangeSetCJK, //uff7x, halfwidth katakana
1.363 + kRangeSetCJK, //uff8x, halfwidth katakana
1.364 + kRangeSetCJK, //uff9x, halfwidth katakana
1.365 + kRangeSetCJK, //uffax, halfwidth hangul jamo
1.366 + kRangeSetCJK, //uffbx, halfwidth hangul jamo
1.367 + kRangeSetCJK, //uffcx, halfwidth hangul jamo
1.368 + kRangeSetCJK, //uffdx, halfwidth hangul jamo
1.369 + kRangeSetCJK, //uffex, fullwidth symbols
1.370 + kRangeSpecials, //ufffx, Specials
1.371 + },
1.372 +};
1.373 +
1.374 +// Most scripts between U+0700 and U+16FF are assigned a chunk of 128 (0x80)
1.375 +// code points so that the number of entries in the tertiary range
1.376 +// table for that range is obtained by dividing (0x1700 - 0x0700) by 128.
1.377 +// Exceptions: Ethiopic, Tibetan, Hangul Jamo and Canadian aboriginal
1.378 +// syllabaries take multiple chunks and Ogham and Runic share a single chunk.
1.379 +#define TERTIARY_TABLE_SIZE ((0x1700 - 0x0700) / 0x80)
1.380 +
1.381 +static const uint8_t gUnicodeTertiaryRangeTable[TERTIARY_TABLE_SIZE] =
1.382 +{ //table for 0x0700 - 0x1600
1.383 + kRangeSyriac, //u070x
1.384 + kRangeThaana, //u078x
1.385 + kRangeUnassigned, //u080x place holder(resolved in the 2ndary tab.)
1.386 + kRangeUnassigned, //u088x place holder(resolved in the 2ndary tab.)
1.387 + kRangeDevanagari, //u090x
1.388 + kRangeBengali, //u098x
1.389 + kRangeGurmukhi, //u0a0x
1.390 + kRangeGujarati, //u0a8x
1.391 + kRangeOriya, //u0b0x
1.392 + kRangeTamil, //u0b8x
1.393 + kRangeTelugu, //u0c0x
1.394 + kRangeKannada, //u0c8x
1.395 + kRangeMalayalam, //u0d0x
1.396 + kRangeSinhala, //u0d8x
1.397 + kRangeThai, //u0e0x
1.398 + kRangeLao, //u0e8x
1.399 + kRangeTibetan, //u0f0x place holder(resolved in the 2ndary tab.)
1.400 + kRangeTibetan, //u0f8x place holder(resolved in the 2ndary tab.)
1.401 + kRangeMyanmar, //u100x
1.402 + kRangeGeorgian, //u108x
1.403 + kRangeKorean, //u110x place holder(resolved in the 2ndary tab.)
1.404 + kRangeKorean, //u118x place holder(resolved in the 2ndary tab.)
1.405 + kRangeEthiopic, //u120x place holder(resolved in the 2ndary tab.)
1.406 + kRangeEthiopic, //u128x place holder(resolved in the 2ndary tab.)
1.407 + kRangeEthiopic, //u130x
1.408 + kRangeCherokee, //u138x
1.409 + kRangeCanadian, //u140x place holder(resolved in the 2ndary tab.)
1.410 + kRangeCanadian, //u148x place holder(resolved in the 2ndary tab.)
1.411 + kRangeCanadian, //u150x place holder(resolved in the 2ndary tab.)
1.412 + kRangeCanadian, //u158x place holder(resolved in the 2ndary tab.)
1.413 + kRangeCanadian, //u160x
1.414 + kRangeOghamRunic //u168x this contains two scripts, Ogham & Runic
1.415 +};
1.416 +
1.417 +// A two level index is almost enough for locating a range, with the
1.418 +// exception of u03xx and u05xx. Since we don't really care about range for
1.419 +// combining diacritical marks in our font application, they are
1.420 +// not discriminated further. But future adoption of this module for other use
1.421 +// should be aware of this limitation. The implementation can be extended if
1.422 +// there is such a need.
1.423 +// For Indic, Southeast Asian scripts and some other scripts between
1.424 +// U+0700 and U+16FF, it's extended to the third level.
1.425 +uint32_t FindCharUnicodeRange(uint32_t ch)
1.426 +{
1.427 + uint32_t range;
1.428 +
1.429 + // aggregate ranges for non-BMP codepoints
1.430 + if (ch > 0xFFFF) {
1.431 + uint32_t p = (ch >> 16);
1.432 + if (p == 1) {
1.433 + return kRangeSMP;
1.434 + } else if (p == 2) {
1.435 + return kRangeSetCJK;
1.436 + }
1.437 + return kRangeHigherPlanes;
1.438 + }
1.439 +
1.440 + // lookup explicit range for BMP codepoints
1.441 + // first general range
1.442 + range = gUnicodeSubrangeTable[0][ch >> 12];
1.443 +
1.444 + // if general range is good enough, return that
1.445 + if (range < kRangeTableBase)
1.446 + // we try to get a specific range
1.447 + return range;
1.448 +
1.449 + // otherwise, use subrange tables
1.450 + range = gUnicodeSubrangeTable[range - kRangeTableBase][(ch & 0x0f00) >> 8];
1.451 + if (range < kRangeTableBase)
1.452 + return range;
1.453 + if (range < kRangeTertiaryTable)
1.454 + return gUnicodeSubrangeTable[range - kRangeTableBase][(ch & 0x00f0) >> 4];
1.455 +
1.456 + // Yet another table to look at : U+0700 - U+16FF : 128 code point blocks
1.457 + return gUnicodeTertiaryRangeTable[(ch - 0x0700) >> 7];
1.458 +}
1.459 +
1.460 +nsIAtom *LangGroupFromUnicodeRange(uint8_t unicodeRange)
1.461 +{
1.462 + if (kRangeSpecificItemNum > unicodeRange) {
1.463 + nsIAtom **atom = gUnicodeRangeToLangGroupAtomTable[unicodeRange];
1.464 + return *atom;
1.465 + }
1.466 + return nullptr;
1.467 +}
