i18nutil/source/utility/unicode.cxx

   1 /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
   2 /*
   3  * This file is part of the LibreOffice project.
   4  *
   5  * This Source Code Form is subject to the terms of the Mozilla Public
   6  * License, v. 2.0. If a copy of the MPL was not distributed with this
   7  * file, You can obtain one at http://mozilla.org/MPL/2.0/.
   8  *
   9  * This file incorporates work covered by the following license notice:
  10  *
  11  *   Licensed to the Apache Software Foundation (ASF) under one or more
  12  *   contributor license agreements. See the NOTICE file distributed
  13  *   with this work for additional information regarding copyright
  14  *   ownership. The ASF licenses this file to you under the Apache
  15  *   License, Version 2.0 (the "License"); you may not use this file
  16  *   except in compliance with the License. You may obtain a copy of
  17  *   the License at http://www.apache.org/licenses/LICENSE-2.0 .
  18  */
  19
  20 #include <com/sun/star/i18n/UnicodeType.hpp>
  21 #include <com/sun/star/i18n/KCharacterType.hpp>
  22 #include <com/sun/star/i18n/ScriptType.hpp>
  23 #include <i18nlangtag/languagetag.hxx>
  24 #include <i18nlangtag/languagetagicu.hxx>
  25 #include <i18nutil/unicode.hxx>
  26 #include <sal/log.hxx>
  27 #include <unicode/numfmt.h>
  28 #include "unicode_data.h"
  29 #include <com/sun/star/i18n/UnicodeType.hpp>
  30 #include <rtl/character.hxx>
  31 #include <memory>
  32
  33 // Workaround for glibc braindamage:
  34 // glibc 2.4's langinfo.h does "#define CURRENCY_SYMBOL __CURRENCY_SYMBOL"
  35 // which (obviously) breaks UnicodeType::CURRENCY_SYMBOL
  36 #undef CURRENCY_SYMBOL
  37
  38 using namespace ::com::sun::star::i18n;
  39
  40 static const ScriptTypeList defaultTypeList[] = {
  41     { UnicodeScript_kBasicLatin,
  42       UnicodeScript_kBasicLatin,
  43       UnicodeScript_kBasicLatin },      // 0,
  44     { UnicodeScript_kLatin1Supplement,
  45       UnicodeScript_kLatin1Supplement,
  46       UnicodeScript_kLatin1Supplement },// 1,
  47     { UnicodeScript_kLatinExtendedA,
  48       UnicodeScript_kLatinExtendedA,
  49       UnicodeScript_kLatinExtendedA }, // 2,
  50     { UnicodeScript_kLatinExtendedB,
  51       UnicodeScript_kLatinExtendedB,
  52       UnicodeScript_kLatinExtendedB }, // 3,
  53     { UnicodeScript_kIPAExtension,
  54       UnicodeScript_kIPAExtension,
  55       UnicodeScript_kIPAExtension }, // 4,
  56     { UnicodeScript_kSpacingModifier,
  57       UnicodeScript_kSpacingModifier,
  58       UnicodeScript_kSpacingModifier }, // 5,
  59     { UnicodeScript_kCombiningDiacritical,
  60       UnicodeScript_kCombiningDiacritical,
  61       UnicodeScript_kCombiningDiacritical }, // 6,
  62     { UnicodeScript_kGreek,
  63       UnicodeScript_kGreek,
  64       UnicodeScript_kGreek }, // 7,
  65     { UnicodeScript_kCyrillic,
  66       UnicodeScript_kCyrillic,
  67       UnicodeScript_kCyrillic }, // 8,
  68     { UnicodeScript_kArmenian,
  69       UnicodeScript_kArmenian,
  70       UnicodeScript_kArmenian }, // 9,
  71     { UnicodeScript_kHebrew,
  72       UnicodeScript_kHebrew,
  73       UnicodeScript_kHebrew }, // 10,
  74     { UnicodeScript_kArabic,
  75       UnicodeScript_kArabic,
  76       UnicodeScript_kArabic }, // 11,
  77     { UnicodeScript_kSyriac,
  78       UnicodeScript_kSyriac,
  79       UnicodeScript_kSyriac }, // 12,
  80     { UnicodeScript_kThaana,
  81       UnicodeScript_kThaana,
  82       UnicodeScript_kThaana }, // 13,
  83     { UnicodeScript_kDevanagari,
  84       UnicodeScript_kDevanagari,
  85       UnicodeScript_kDevanagari }, // 14,
  86     { UnicodeScript_kBengali,
  87       UnicodeScript_kBengali,
  88       UnicodeScript_kBengali }, // 15,
  89     { UnicodeScript_kGurmukhi,
  90       UnicodeScript_kGurmukhi,
  91       UnicodeScript_kGurmukhi }, // 16,
  92     { UnicodeScript_kGujarati,
  93       UnicodeScript_kGujarati,
  94       UnicodeScript_kGujarati }, // 17,
  95     { UnicodeScript_kOriya,
  96       UnicodeScript_kOriya,
  97       UnicodeScript_kOriya }, // 18,
  98     { UnicodeScript_kTamil,
  99       UnicodeScript_kTamil,
 100       UnicodeScript_kTamil }, // 19,
 101     { UnicodeScript_kTelugu,
 102       UnicodeScript_kTelugu,
 103       UnicodeScript_kTelugu }, // 20,
 104     { UnicodeScript_kKannada,
 105       UnicodeScript_kKannada,
 106       UnicodeScript_kKannada }, // 21,
 107     { UnicodeScript_kMalayalam,
 108       UnicodeScript_kMalayalam,
 109       UnicodeScript_kMalayalam }, // 22,
 110     { UnicodeScript_kSinhala,
 111       UnicodeScript_kSinhala,
 112       UnicodeScript_kSinhala }, // 23,
 113     { UnicodeScript_kThai,
 114       UnicodeScript_kThai,
 115       UnicodeScript_kThai }, // 24,
 116     { UnicodeScript_kLao,
 117       UnicodeScript_kLao,
 118       UnicodeScript_kLao }, // 25,
 119     { UnicodeScript_kTibetan,
 120       UnicodeScript_kTibetan,
 121       UnicodeScript_kTibetan }, // 26,
 122     { UnicodeScript_kMyanmar,
 123       UnicodeScript_kMyanmar,
 124       UnicodeScript_kMyanmar }, // 27,
 125     { UnicodeScript_kGeorgian,
 126       UnicodeScript_kGeorgian,
 127       UnicodeScript_kGeorgian }, // 28,
 128     { UnicodeScript_kHangulJamo,
 129       UnicodeScript_kHangulJamo,
 130       UnicodeScript_kHangulJamo }, // 29,
 131     { UnicodeScript_kEthiopic,
 132       UnicodeScript_kEthiopic,
 133       UnicodeScript_kEthiopic }, // 30,
 134     { UnicodeScript_kCherokee,
 135       UnicodeScript_kCherokee,
 136       UnicodeScript_kCherokee }, // 31,
 137     { UnicodeScript_kUnifiedCanadianAboriginalSyllabics,
 138       UnicodeScript_kUnifiedCanadianAboriginalSyllabics,
 139       UnicodeScript_kUnifiedCanadianAboriginalSyllabics }, // 32,
 140     { UnicodeScript_kOgham,
 141       UnicodeScript_kOgham,
 142       UnicodeScript_kOgham }, // 33,
 143     { UnicodeScript_kRunic,
 144       UnicodeScript_kRunic,
 145       UnicodeScript_kRunic }, // 34,
 146     { UnicodeScript_kKhmer,
 147       UnicodeScript_kKhmer,
 148       UnicodeScript_kKhmer }, // 35,
 149     { UnicodeScript_kMongolian,
 150       UnicodeScript_kMongolian,
 151       UnicodeScript_kMongolian }, // 36,
 152     { UnicodeScript_kLatinExtendedAdditional,
 153       UnicodeScript_kLatinExtendedAdditional,
 154       UnicodeScript_kLatinExtendedAdditional }, // 37,
 155     { UnicodeScript_kGreekExtended,
 156       UnicodeScript_kGreekExtended,
 157       UnicodeScript_kGreekExtended }, // 38,
 158     { UnicodeScript_kGeneralPunctuation,
 159       UnicodeScript_kGeneralPunctuation,
 160       UnicodeScript_kGeneralPunctuation }, // 39,
 161     { UnicodeScript_kSuperSubScript,
 162       UnicodeScript_kSuperSubScript,
 163       UnicodeScript_kSuperSubScript }, // 40,
 164     { UnicodeScript_kCurrencySymbolScript,
 165       UnicodeScript_kCurrencySymbolScript,
 166       UnicodeScript_kCurrencySymbolScript }, // 41,
 167     { UnicodeScript_kSymbolCombiningMark,
 168       UnicodeScript_kSymbolCombiningMark,
 169       UnicodeScript_kSymbolCombiningMark }, // 42,
 170     { UnicodeScript_kLetterlikeSymbol,
 171       UnicodeScript_kLetterlikeSymbol,
 172       UnicodeScript_kLetterlikeSymbol }, // 43,
 173     { UnicodeScript_kNumberForm,
 174       UnicodeScript_kNumberForm,
 175       UnicodeScript_kNumberForm }, // 44,
 176     { UnicodeScript_kArrow,
 177       UnicodeScript_kArrow,
 178       UnicodeScript_kArrow }, // 45,
 179     { UnicodeScript_kMathOperator,
 180       UnicodeScript_kMathOperator,
 181       UnicodeScript_kMathOperator }, // 46,
 182     { UnicodeScript_kMiscTechnical,
 183       UnicodeScript_kMiscTechnical,
 184       UnicodeScript_kMiscTechnical }, // 47,
 185     { UnicodeScript_kControlPicture,
 186       UnicodeScript_kControlPicture,
 187       UnicodeScript_kControlPicture }, // 48,
 188     { UnicodeScript_kOpticalCharacter,
 189       UnicodeScript_kOpticalCharacter,
 190       UnicodeScript_kOpticalCharacter }, // 49,
 191     { UnicodeScript_kEnclosedAlphanumeric,
 192       UnicodeScript_kEnclosedAlphanumeric,
 193       UnicodeScript_kEnclosedAlphanumeric }, // 50,
 194     { UnicodeScript_kBoxDrawing,
 195       UnicodeScript_kBoxDrawing,
 196       UnicodeScript_kBoxDrawing }, // 51,
 197     { UnicodeScript_kBlockElement,
 198       UnicodeScript_kBlockElement,
 199       UnicodeScript_kBlockElement }, // 52,
 200     { UnicodeScript_kGeometricShape,
 201       UnicodeScript_kGeometricShape,
 202       UnicodeScript_kGeometricShape }, // 53,
 203     { UnicodeScript_kMiscSymbol,
 204       UnicodeScript_kMiscSymbol,
 205       UnicodeScript_kMiscSymbol }, // 54,
 206     { UnicodeScript_kDingbat,
 207       UnicodeScript_kDingbat,
 208       UnicodeScript_kDingbat }, // 55,
 209     { UnicodeScript_kBraillePatterns,
 210       UnicodeScript_kBraillePatterns,
 211       UnicodeScript_kBraillePatterns }, // 56,
 212     { UnicodeScript_kCJKRadicalsSupplement,
 213       UnicodeScript_kCJKRadicalsSupplement,
 214       UnicodeScript_kCJKRadicalsSupplement }, // 57,
 215     { UnicodeScript_kKangxiRadicals,
 216       UnicodeScript_kKangxiRadicals,
 217       UnicodeScript_kKangxiRadicals }, // 58,
 218     { UnicodeScript_kIdeographicDescriptionCharacters,
 219       UnicodeScript_kIdeographicDescriptionCharacters,
 220       UnicodeScript_kIdeographicDescriptionCharacters }, // 59,
 221     { UnicodeScript_kCJKSymbolPunctuation,
 222       UnicodeScript_kCJKSymbolPunctuation,
 223       UnicodeScript_kCJKSymbolPunctuation }, // 60,
 224     { UnicodeScript_kHiragana,
 225       UnicodeScript_kHiragana,
 226       UnicodeScript_kHiragana }, // 61,
 227     { UnicodeScript_kKatakana,
 228       UnicodeScript_kKatakana,
 229       UnicodeScript_kKatakana }, // 62,
 230     { UnicodeScript_kBopomofo,
 231       UnicodeScript_kBopomofo,
 232       UnicodeScript_kBopomofo }, // 63,
 233     { UnicodeScript_kHangulCompatibilityJamo,
 234       UnicodeScript_kHangulCompatibilityJamo,
 235       UnicodeScript_kHangulCompatibilityJamo }, // 64,
 236     { UnicodeScript_kKanbun,
 237       UnicodeScript_kKanbun,
 238       UnicodeScript_kKanbun }, // 65,
 239     { UnicodeScript_kBopomofoExtended,
 240       UnicodeScript_kBopomofoExtended,
 241       UnicodeScript_kBopomofoExtended }, // 66,
 242     { UnicodeScript_kEnclosedCJKLetterMonth,
 243       UnicodeScript_kEnclosedCJKLetterMonth,
 244       UnicodeScript_kEnclosedCJKLetterMonth }, // 67,
 245     { UnicodeScript_kCJKCompatibility,
 246       UnicodeScript_kCJKCompatibility,
 247       UnicodeScript_kCJKCompatibility }, // 68,
 248     { UnicodeScript_k_CJKUnifiedIdeographsExtensionA,
 249       UnicodeScript_k_CJKUnifiedIdeographsExtensionA,
 250       UnicodeScript_k_CJKUnifiedIdeographsExtensionA }, // 69,
 251     { UnicodeScript_kCJKUnifiedIdeograph,
 252       UnicodeScript_kCJKUnifiedIdeograph,
 253       UnicodeScript_kCJKUnifiedIdeograph }, // 70,
 254     { UnicodeScript_kYiSyllables,
 255       UnicodeScript_kYiSyllables,
 256       UnicodeScript_kYiSyllables }, // 71,
 257     { UnicodeScript_kYiRadicals,
 258       UnicodeScript_kYiRadicals,
 259       UnicodeScript_kYiRadicals }, // 72,
 260     { UnicodeScript_kHangulSyllable,
 261       UnicodeScript_kHangulSyllable,
 262       UnicodeScript_kHangulSyllable }, // 73,
 263     { UnicodeScript_kHighSurrogate,
 264       UnicodeScript_kHighSurrogate,
 265       UnicodeScript_kHighSurrogate }, // 74,
 266     { UnicodeScript_kHighPrivateUseSurrogate,
 267       UnicodeScript_kHighPrivateUseSurrogate,
 268       UnicodeScript_kHighPrivateUseSurrogate }, // 75,
 269     { UnicodeScript_kLowSurrogate,
 270       UnicodeScript_kLowSurrogate,
 271       UnicodeScript_kLowSurrogate }, // 76,
 272     { UnicodeScript_kPrivateUse,
 273       UnicodeScript_kPrivateUse,
 274       UnicodeScript_kPrivateUse }, // 77,
 275     { UnicodeScript_kCJKCompatibilityIdeograph,
 276       UnicodeScript_kCJKCompatibilityIdeograph,
 277       UnicodeScript_kCJKCompatibilityIdeograph }, // 78,
 278     { UnicodeScript_kAlphabeticPresentation,
 279       UnicodeScript_kAlphabeticPresentation,
 280       UnicodeScript_kAlphabeticPresentation }, // 79,
 281     { UnicodeScript_kArabicPresentationA,
 282       UnicodeScript_kArabicPresentationA,
 283       UnicodeScript_kArabicPresentationA }, // 80,
 284     { UnicodeScript_kCombiningHalfMark,
 285       UnicodeScript_kCombiningHalfMark,
 286       UnicodeScript_kCombiningHalfMark }, // 81,
 287     { UnicodeScript_kCJKCompatibilityForm,
 288       UnicodeScript_kCJKCompatibilityForm,
 289       UnicodeScript_kCJKCompatibilityForm }, // 82,
 290     { UnicodeScript_kSmallFormVariant,
 291       UnicodeScript_kSmallFormVariant,
 292       UnicodeScript_kSmallFormVariant }, // 83,
 293     { UnicodeScript_kArabicPresentationB,
 294       UnicodeScript_kArabicPresentationB,
 295       UnicodeScript_kArabicPresentationB }, // 84,
 296     { UnicodeScript_kNoScript,
 297       UnicodeScript_kNoScript,
 298       UnicodeScript_kNoScript }, // 85,
 299     { UnicodeScript_kHalfwidthFullwidthForm,
 300       UnicodeScript_kHalfwidthFullwidthForm,
 301       UnicodeScript_kHalfwidthFullwidthForm }, // 86,
 302     { UnicodeScript_kScriptCount,
 303       UnicodeScript_kScriptCount,
 304       UnicodeScript_kNoScript } // 87,
 305 };
 306
 307 sal_Int16 SAL_CALL
 308 unicode::getUnicodeScriptType( const sal_Unicode ch, const ScriptTypeList* typeList, sal_Int16 unknownType ) {
 309
 310     if (!typeList) {
 311         typeList = defaultTypeList;
 312         unknownType = UnicodeScript_kNoScript;
 313     }
 314
 315     sal_Int16 i = 0, type = typeList[0].to;
 316     while (type < UnicodeScript_kScriptCount && ch > UnicodeScriptType[type][UnicodeScriptTypeTo]) {
 317         type = typeList[++i].to;
 318     }
 319
 320     return (type < UnicodeScript_kScriptCount &&
 321             ch >= UnicodeScriptType[typeList[i].from][UnicodeScriptTypeFrom]) ?
 322             typeList[i].value : unknownType;
 323 }
 324
 325 sal_Unicode SAL_CALL
 326 unicode::getUnicodeScriptStart( UnicodeScript type) {
 327     return UnicodeScriptType[type][UnicodeScriptTypeFrom];
 328 }
 329
 330 sal_Unicode SAL_CALL
 331 unicode::getUnicodeScriptEnd( UnicodeScript type) {
 332     return UnicodeScriptType[type][UnicodeScriptTypeTo];
 333 }
 334
 335 sal_Int16 SAL_CALL
 336 unicode::getUnicodeType( const sal_Unicode ch ) {
 337     static sal_Unicode c = 0x00;
 338     static sal_Int16 r = 0x00;
 339
 340     if (ch == c) return r;
 341     else c = ch;
 342
 343     sal_Int16 address = UnicodeTypeIndex[ch >> 8];
 344     return r = (sal_Int16)((address < UnicodeTypeNumberBlock) ? UnicodeTypeBlockValue[address] :
 345         UnicodeTypeValue[((address - UnicodeTypeNumberBlock) << 8) + (ch & 0xff)]);
 346 }
 347
 348 sal_uInt8 SAL_CALL
 349 unicode::getUnicodeDirection( const sal_Unicode ch ) {
 350     static sal_Unicode c = 0x00;
 351     static sal_uInt8 r = 0x00;
 352
 353     if (ch == c) return r;
 354     else c = ch;
 355
 356     sal_Int16 address = UnicodeDirectionIndex[ch >> 8];
 357     return r = ((address < UnicodeDirectionNumberBlock) ? UnicodeDirectionBlockValue[address] :
 358         UnicodeDirectionValue[((address - UnicodeDirectionNumberBlock) << 8) + (ch & 0xff)]);
 359
 360 }
 361
 362 #define bit(name)   (1U << name)
 363
 364 #define UPPERMASK   bit(UnicodeType::UPPERCASE_LETTER)
 365
 366 #define LOWERMASK   bit(UnicodeType::LOWERCASE_LETTER)
 367
 368 #define TITLEMASK   bit(UnicodeType::TITLECASE_LETTER)
 369
 370 #define ALPHAMASK   UPPERMASK|LOWERMASK|TITLEMASK|\
 371             bit(UnicodeType::MODIFIER_LETTER)|\
 372             bit(UnicodeType::OTHER_LETTER)
 373
 374 #define SPACEMASK   bit(UnicodeType::SPACE_SEPARATOR)|\
 375             bit(UnicodeType::LINE_SEPARATOR)|\
 376             bit(UnicodeType::PARAGRAPH_SEPARATOR)
 377
 378 #define CONTROLMASK bit(UnicodeType::CONTROL)|\
 379             bit(UnicodeType::FORMAT)|\
 380             bit(UnicodeType::LINE_SEPARATOR)|\
 381             bit(UnicodeType::PARAGRAPH_SEPARATOR)
 382
 383 #define IsType(func, mask)  \
 384 bool SAL_CALL func( const sal_Unicode ch) {\
 385     return (bit(getUnicodeType(ch)) & (mask)) != 0;\
 386 }
 387
 388 IsType(unicode::isControl, CONTROLMASK)
 389 IsType(unicode::isAlpha, ALPHAMASK)
 390 IsType(unicode::isSpace, SPACEMASK)
 391
 392 #define CONTROLSPACE    bit(0x09)|bit(0x0a)|bit(0x0b)|bit(0x0c)|bit(0x0d)|\
 393             bit(0x1c)|bit(0x1d)|bit(0x1e)|bit(0x1f)
 394
 395 bool SAL_CALL unicode::isWhiteSpace( const sal_Unicode ch) {
 396     return (ch != 0xa0 && isSpace(ch)) || (ch <= 0x1F && (bit(ch) & (CONTROLSPACE)));
 397 }
 398
 399 sal_Int16 SAL_CALL unicode::getScriptClassFromUScriptCode(UScriptCode eScript)
 400 {
 401     //See unicode/uscript.h
 402     static const sal_Int16 scriptTypes[] =
 403     {
 404         ScriptType::WEAK, ScriptType::WEAK, ScriptType::COMPLEX, ScriptType::LATIN, ScriptType::COMPLEX,
 405         ScriptType::ASIAN, ScriptType::LATIN, ScriptType::LATIN, ScriptType::LATIN, ScriptType::COMPLEX,
 406         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::LATIN, ScriptType::LATIN, ScriptType::LATIN,
 407     // 15
 408         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::ASIAN, ScriptType::ASIAN, ScriptType::COMPLEX,
 409         ScriptType::ASIAN, ScriptType::COMPLEX, ScriptType::ASIAN, ScriptType::COMPLEX, ScriptType::COMPLEX,
 410         ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::LATIN,
 411     // 30
 412         ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::COMPLEX,
 413         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 414         ScriptType::LATIN, ScriptType::ASIAN, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 415     // 45
 416         ScriptType::COMPLEX, ScriptType::LATIN, ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::COMPLEX,
 417         ScriptType::LATIN, ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::LATIN,
 418         ScriptType::COMPLEX, ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 419     // 60
 420         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 421         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::LATIN, ScriptType::LATIN, ScriptType::COMPLEX,
 422         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::ASIAN, ScriptType::ASIAN,
 423     // 75
 424         ScriptType::COMPLEX, ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 425         ScriptType::LATIN, ScriptType::LATIN, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 426         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 427     // 90
 428         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 429         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 430         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::WEAK, ScriptType::WEAK, ScriptType::COMPLEX,
 431     // 105
 432         ScriptType::ASIAN, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 433         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 434         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::ASIAN,
 435     // 120
 436         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 437         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::WEAK, ScriptType::WEAK,
 438         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 439     // 135
 440         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 441         ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX, ScriptType::COMPLEX,
 442         ScriptType::COMPLEX,
 443         ScriptType::WEAK
 444     };
 445
 446     sal_Int16 nRet;
 447     if (eScript < USCRIPT_COMMON)
 448         nRet = ScriptType::WEAK;
 449     else if (static_cast<size_t>(eScript) >= SAL_N_ELEMENTS(scriptTypes))
 450         nRet = ScriptType::COMPLEX;         // anything new is going to be pretty wild
 451     else
 452         nRet = scriptTypes[eScript];
 453     return nRet;
 454 }
 455
 456 OString SAL_CALL unicode::getExemplarLanguageForUScriptCode(UScriptCode eScript)
 457 {
 458     OString sRet;
 459     switch (eScript)
 460     {
 461         case USCRIPT_CODE_LIMIT:
 462         case USCRIPT_INVALID_CODE:
 463             sRet = "zxx";
 464             break;
 465         case USCRIPT_COMMON:
 466         case USCRIPT_INHERITED:
 467             sRet = "und";
 468             break;
 469         case USCRIPT_MATHEMATICAL_NOTATION:
 470         case USCRIPT_SYMBOLS:
 471             sRet = "zxx";
 472             break;
 473         case USCRIPT_UNWRITTEN_LANGUAGES:
 474         case USCRIPT_UNKNOWN:
 475             sRet = "und";
 476             break;
 477         case USCRIPT_ARABIC:
 478             sRet = "ar";
 479             break;
 480         case USCRIPT_ARMENIAN:
 481             sRet = "hy";
 482             break;
 483         case USCRIPT_BENGALI:
 484             sRet = "bn";
 485             break;
 486         case USCRIPT_BOPOMOFO:
 487             sRet = "zh";
 488             break;
 489         case USCRIPT_CHEROKEE:
 490             sRet = "chr";
 491             break;
 492         case USCRIPT_COPTIC:
 493             sRet = "cop";
 494             break;
 495         case USCRIPT_CYRILLIC:
 496             sRet = "ru";
 497             break;
 498         case USCRIPT_DESERET:
 499             sRet = "en";
 500             break;
 501         case USCRIPT_DEVANAGARI:
 502             sRet = "hi";
 503             break;
 504         case USCRIPT_ETHIOPIC:
 505             sRet = "am";
 506             break;
 507         case USCRIPT_GEORGIAN:
 508             sRet = "ka";
 509             break;
 510         case USCRIPT_GOTHIC:
 511             sRet = "got";
 512             break;
 513         case USCRIPT_GREEK:
 514             sRet = "el";
 515             break;
 516         case USCRIPT_GUJARATI:
 517             sRet = "gu";
 518             break;
 519         case USCRIPT_GURMUKHI:
 520             sRet = "pa";
 521             break;
 522         case USCRIPT_HAN:
 523             sRet = "zh";
 524             break;
 525         case USCRIPT_HANGUL:
 526             sRet = "ko";
 527             break;
 528         case USCRIPT_HEBREW:
 529             sRet = "hr";
 530             break;
 531         case USCRIPT_HIRAGANA:
 532             sRet = "ja";
 533             break;
 534         case USCRIPT_KANNADA:
 535             sRet = "kn";
 536             break;
 537         case USCRIPT_KATAKANA:
 538             sRet = "ja";
 539             break;
 540         case USCRIPT_KHMER:
 541             sRet = "km";
 542             break;
 543         case USCRIPT_LAO:
 544             sRet = "lo";
 545             break;
 546         case USCRIPT_LATIN:
 547             sRet = "en";
 548             break;
 549         case USCRIPT_MALAYALAM:
 550             sRet = "ml";
 551             break;
 552         case USCRIPT_MONGOLIAN:
 553             sRet = "mn";
 554             break;
 555         case USCRIPT_MYANMAR:
 556             sRet = "my";
 557             break;
 558         case USCRIPT_OGHAM:
 559             sRet = "pgl";
 560             break;
 561         case USCRIPT_OLD_ITALIC:
 562             sRet = "osc";
 563             break;
 564         case USCRIPT_ORIYA:
 565             sRet = "or";
 566             break;
 567         case USCRIPT_RUNIC:
 568             sRet = "ang";
 569             break;
 570         case USCRIPT_SINHALA:
 571             sRet = "si";
 572             break;
 573         case USCRIPT_SYRIAC:
 574             sRet = "syr";
 575             break;
 576         case USCRIPT_TAMIL:
 577             sRet = "ta";
 578             break;
 579         case USCRIPT_TELUGU:
 580             sRet = "te";
 581             break;
 582         case USCRIPT_THAANA:
 583             sRet = "dv";
 584             break;
 585         case USCRIPT_THAI:
 586             sRet = "th";
 587             break;
 588         case USCRIPT_TIBETAN:
 589             sRet = "bo";
 590             break;
 591         case USCRIPT_CANADIAN_ABORIGINAL:
 592             sRet = "iu";
 593             break;
 594         case USCRIPT_YI:
 595             sRet = "ii";
 596             break;
 597         case USCRIPT_TAGALOG:
 598             sRet = "tl";
 599             break;
 600         case USCRIPT_HANUNOO:
 601             sRet = "hnn";
 602             break;
 603         case USCRIPT_BUHID:
 604             sRet = "bku";
 605             break;
 606         case USCRIPT_TAGBANWA:
 607             sRet = "tbw";
 608             break;
 609         case USCRIPT_BRAILLE:
 610             sRet = "en";
 611             break;
 612         case USCRIPT_CYPRIOT:
 613             sRet = "ecy";
 614             break;
 615         case USCRIPT_LIMBU:
 616             sRet = "lif";
 617             break;
 618         case USCRIPT_LINEAR_B:
 619             sRet = "gmy";
 620             break;
 621         case USCRIPT_OSMANYA:
 622             sRet = "so";
 623             break;
 624         case USCRIPT_SHAVIAN:
 625             sRet = "en";
 626             break;
 627         case USCRIPT_TAI_LE:
 628             sRet = "tdd";
 629             break;
 630         case USCRIPT_UGARITIC:
 631             sRet = "uga";
 632             break;
 633         case USCRIPT_KATAKANA_OR_HIRAGANA:
 634             sRet = "ja";
 635             break;
 636         case USCRIPT_BUGINESE:
 637             sRet = "bug";
 638             break;
 639         case USCRIPT_GLAGOLITIC:
 640             sRet = "ch";
 641             break;
 642         case USCRIPT_KHAROSHTHI:
 643             sRet = "pra";
 644             break;
 645         case USCRIPT_SYLOTI_NAGRI:
 646             sRet = "syl";
 647             break;
 648         case USCRIPT_NEW_TAI_LUE:
 649             sRet = "khb";
 650             break;
 651         case USCRIPT_TIFINAGH:
 652             sRet = "tmh";
 653             break;
 654         case USCRIPT_OLD_PERSIAN:
 655             sRet = "peo";
 656             break;
 657         case USCRIPT_BALINESE:
 658             sRet = "ban";
 659             break;
 660         case USCRIPT_BATAK:
 661             sRet = "btk";
 662             break;
 663         case USCRIPT_BLISSYMBOLS:
 664             sRet = "en";
 665             break;
 666         case USCRIPT_BRAHMI:
 667             sRet = "pra";
 668             break;
 669         case USCRIPT_CHAM:
 670             sRet = "cja";
 671             break;
 672         case USCRIPT_CIRTH:
 673             sRet = "sjn";
 674             break;
 675         case USCRIPT_OLD_CHURCH_SLAVONIC_CYRILLIC:
 676             sRet = "cu";
 677             break;
 678         case USCRIPT_DEMOTIC_EGYPTIAN:
 679         case USCRIPT_HIERATIC_EGYPTIAN:
 680         case USCRIPT_EGYPTIAN_HIEROGLYPHS:
 681             sRet = "egy";
 682             break;
 683         case USCRIPT_KHUTSURI:
 684             sRet = "ka";
 685             break;
 686         case USCRIPT_SIMPLIFIED_HAN:
 687             sRet = "zh";
 688             break;
 689         case USCRIPT_TRADITIONAL_HAN:
 690             sRet = "zh";
 691             break;
 692         case USCRIPT_PAHAWH_HMONG:
 693             sRet = "blu";
 694             break;
 695         case USCRIPT_OLD_HUNGARIAN:
 696             sRet = "ohu";
 697             break;
 698         case USCRIPT_HARAPPAN_INDUS:
 699             sRet = "xiv";
 700             break;
 701         case USCRIPT_JAVANESE:
 702             sRet = "kaw";
 703             break;
 704         case USCRIPT_KAYAH_LI:
 705             sRet = "eky";
 706             break;
 707         case USCRIPT_LATIN_FRAKTUR:
 708             sRet = "de";
 709             break;
 710         case USCRIPT_LATIN_GAELIC:
 711             sRet = "ga";
 712             break;
 713         case USCRIPT_LEPCHA:
 714             sRet = "lep";
 715             break;
 716         case USCRIPT_LINEAR_A:
 717             sRet = "ecr";
 718             break;
 719         case USCRIPT_MAYAN_HIEROGLYPHS:
 720             sRet = "myn";
 721             break;
 722         case USCRIPT_MEROITIC:
 723             sRet = "xmr";
 724             break;
 725         case USCRIPT_NKO:
 726             sRet = "nqo";
 727             break;
 728         case USCRIPT_ORKHON:
 729             sRet = "otk";
 730             break;
 731         case USCRIPT_OLD_PERMIC:
 732             sRet = "kv";
 733             break;
 734         case USCRIPT_PHAGS_PA:
 735             sRet = "xng";
 736             break;
 737         case USCRIPT_PHOENICIAN:
 738             sRet = "phn";
 739             break;
 740         case USCRIPT_PHONETIC_POLLARD:
 741             sRet = "hmd";
 742             break;
 743         case USCRIPT_RONGORONGO:
 744             sRet = "rap";
 745             break;
 746         case USCRIPT_SARATI:
 747             sRet = "qya";
 748             break;
 749         case USCRIPT_ESTRANGELO_SYRIAC:
 750             sRet = "syr";
 751             break;
 752         case USCRIPT_WESTERN_SYRIAC:
 753             sRet = "tru";
 754             break;
 755         case USCRIPT_EASTERN_SYRIAC:
 756             sRet = "aii";
 757             break;
 758         case USCRIPT_TENGWAR:
 759             sRet = "sjn";
 760             break;
 761         case USCRIPT_VAI:
 762             sRet = "vai";
 763             break;
 764         case USCRIPT_VISIBLE_SPEECH:
 765             sRet = "en";
 766             break;
 767         case USCRIPT_CUNEIFORM:
 768             sRet = "akk";
 769             break;
 770         case USCRIPT_CARIAN:
 771             sRet = "xcr";
 772             break;
 773         case USCRIPT_JAPANESE:
 774             sRet = "ja";
 775             break;
 776         case USCRIPT_LANNA:
 777             sRet = "nod";
 778             break;
 779         case USCRIPT_LYCIAN:
 780             sRet = "xlc";
 781             break;
 782         case USCRIPT_LYDIAN:
 783             sRet = "xld";
 784             break;
 785         case USCRIPT_OL_CHIKI:
 786             sRet = "sat";
 787             break;
 788         case USCRIPT_REJANG:
 789             sRet = "rej";
 790             break;
 791         case USCRIPT_SAURASHTRA:
 792             sRet = "saz";
 793             break;
 794         case USCRIPT_SIGN_WRITING:
 795             sRet = "en";
 796             break;
 797         case USCRIPT_SUNDANESE:
 798             sRet = "su";
 799             break;
 800         case USCRIPT_MOON:
 801             sRet = "en";
 802             break;
 803         case USCRIPT_MEITEI_MAYEK:
 804             sRet = "mni";
 805             break;
 806         case USCRIPT_IMPERIAL_ARAMAIC:
 807             sRet = "arc";
 808             break;
 809         case USCRIPT_AVESTAN:
 810             sRet = "ae";
 811             break;
 812         case USCRIPT_CHAKMA:
 813             sRet = "ccp";
 814             break;
 815         case USCRIPT_KOREAN:
 816             sRet = "ko";
 817             break;
 818         case USCRIPT_KAITHI:
 819             sRet = "awa";
 820             break;
 821         case USCRIPT_MANICHAEAN:
 822             sRet = "xmn";
 823             break;
 824         case USCRIPT_INSCRIPTIONAL_PAHLAVI:
 825         case USCRIPT_PSALTER_PAHLAVI:
 826         case USCRIPT_BOOK_PAHLAVI:
 827         case USCRIPT_INSCRIPTIONAL_PARTHIAN:
 828             sRet = "xpr";
 829             break;
 830         case USCRIPT_SAMARITAN:
 831             sRet = "heb";
 832             break;
 833         case USCRIPT_TAI_VIET:
 834             sRet = "blt";
 835             break;
 836         case USCRIPT_MANDAEAN: /* Aliased to USCRIPT_MANDAIC in icu 4.6. */
 837             sRet = "mic";
 838             break;
 839 #if (U_ICU_VERSION_MAJOR_NUM > 4) || (U_ICU_VERSION_MAJOR_NUM == 4 && U_ICU_VERSION_MINOR_NUM >= 4)
 840         case USCRIPT_NABATAEAN: //no language with an assigned code yet
 841             sRet = "mis";
 842             break;
 843         case USCRIPT_PALMYRENE: //no language with an assigned code yet
 844             sRet = "mis";
 845             break;
 846         case USCRIPT_BAMUM:
 847             sRet = "bax";
 848             break;
 849         case USCRIPT_LISU:
 850             sRet = "lis";
 851             break;
 852         case USCRIPT_NAKHI_GEBA:
 853             sRet = "nxq";
 854             break;
 855         case USCRIPT_OLD_SOUTH_ARABIAN:
 856             sRet = "xsa";
 857             break;
 858         case USCRIPT_BASSA_VAH:
 859             sRet = "bsq";
 860             break;
 861         case USCRIPT_DUPLOYAN_SHORTAND:
 862             sRet = "fr";
 863             break;
 864         case USCRIPT_ELBASAN:
 865             sRet = "sq";
 866             break;
 867         case USCRIPT_GRANTHA:
 868             sRet = "ta";
 869             break;
 870         case USCRIPT_KPELLE:
 871             sRet = "kpe";
 872             break;
 873         case USCRIPT_LOMA:
 874             sRet = "lom";
 875             break;
 876         case USCRIPT_MENDE:
 877             sRet = "men";
 878             break;
 879         case USCRIPT_MEROITIC_CURSIVE:
 880             sRet = "xmr";
 881             break;
 882         case USCRIPT_OLD_NORTH_ARABIAN:
 883             sRet = "xna";
 884             break;
 885         case USCRIPT_SINDHI:
 886             sRet = "sd";
 887             break;
 888         case USCRIPT_WARANG_CITI:
 889             sRet = "hoc";
 890             break;
 891 #endif
 892 #if (U_ICU_VERSION_MAJOR_NUM > 4) || (U_ICU_VERSION_MAJOR_NUM == 4 && U_ICU_VERSION_MINOR_NUM >= 8)
 893         case USCRIPT_AFAKA:
 894             sRet = "djk";
 895             break;
 896         case USCRIPT_JURCHEN:
 897             sRet = "juc";
 898             break;
 899         case USCRIPT_MRO:
 900             sRet = "cmr";
 901             break;
 902         case USCRIPT_NUSHU: //no language with an assigned code yet
 903             sRet = "mis";
 904             break;
 905         case USCRIPT_SHARADA:
 906             sRet = "sa";
 907             break;
 908         case USCRIPT_SORA_SOMPENG:
 909             sRet = "srb";
 910             break;
 911         case USCRIPT_TAKRI:
 912             sRet = "doi";
 913             break;
 914         case USCRIPT_TANGUT:
 915             sRet = "txg";
 916             break;
 917         case USCRIPT_WOLEAI:
 918             sRet = "woe";
 919             break;
 920 #endif
 921 #if (U_ICU_VERSION_MAJOR_NUM >= 49)
 922         case USCRIPT_ANATOLIAN_HIEROGLYPHS:
 923             sRet = "hlu";
 924             break;
 925         case USCRIPT_KHOJKI:
 926             sRet = "gu";
 927             break;
 928         case USCRIPT_TIRHUTA:
 929             sRet = "mai";
 930             break;
 931 #endif
 932 #if (U_ICU_VERSION_MAJOR_NUM >= 52)
 933         case USCRIPT_CAUCASIAN_ALBANIAN:
 934             sRet = "xag";
 935             break;
 936         case USCRIPT_MAHAJANI:
 937             sRet = "mwr";
 938             break;
 939 #endif
 940 #if (U_ICU_VERSION_MAJOR_NUM >= 54)
 941         case USCRIPT_AHOM:
 942             sRet = "aho";
 943             break;
 944         case USCRIPT_HATRAN:
 945             sRet = "qly-Hatr";
 946             break;
 947         case USCRIPT_MODI:
 948             sRet = "mr-Modi";
 949             break;
 950         case USCRIPT_MULTANI:
 951             sRet = "skr-Mutl";
 952             break;
 953         case USCRIPT_PAU_CIN_HAU:
 954             sRet = "ctd-Pauc";
 955             break;
 956         case USCRIPT_SIDDHAM:
 957             sRet = "sa-Sidd";
 958             break;
 959 #endif
 960     }
 961     return sRet;
 962 }
 963
 964 //Format a number as a percentage according to the rules of the given
 965 //language, e.g. 100 -> "100%" for en-US vs "100 %" for de-DE
 966 OUString SAL_CALL unicode::formatPercent(double dNumber,
 967     const LanguageTag &rLangTag)
 968 {
 969     // get a currency formatter for this locale ID
 970     UErrorCode errorCode=U_ZERO_ERROR;
 971
 972     LanguageTag aLangTag(rLangTag);
 973
 974     // As of CLDR Version 24 these languages were not listed as using spacing
 975     // between number and % but are reported as such by our l10n groups
 976     // http://www.unicode.org/cldr/charts/24/by_type/numbers.number_formatting_patterns.html
 977     // so format using French which has the desired rules
 978     if (aLangTag.getLanguage() == "es" || aLangTag.getLanguage() == "sl")
 979         aLangTag = LanguageTag("fr-FR");
 980
 981     icu::Locale aLocale = LanguageTagIcu::getIcuLocale(aLangTag);
 982
 983     std::unique_ptr<NumberFormat> xF(
 984         NumberFormat::createPercentInstance(aLocale, errorCode));
 985     if(U_FAILURE(errorCode))
 986     {
 987         SAL_WARN("i18n", "NumberFormat::createPercentInstance failed");
 988         return OUString::number(dNumber) + "%";
 989     }
 990
 991     UnicodeString output;
 992     xF->format(dNumber/100, output);
 993     OUString aRet(reinterpret_cast<const sal_Unicode *>(output.getBuffer()),
 994         output.length());
 995     if (rLangTag.getLanguage() == "de")
 996     {
 997         //narrow no-break space instead of (normal) no-break space
 998         return aRet.replace(0x00A0, 0x202F);
 999     }
1000     return aRet;
1001 }
1002
1003 ToggleUnicodeCodepoint::ToggleUnicodeCodepoint ()
1004 {
1005     maInput = OUStringBuffer();
1006     maOutput = OUStringBuffer();
1007     maUtf16 = OUStringBuffer();
1008     maCombining = OUStringBuffer();
1009 }
1010
1011 bool ToggleUnicodeCodepoint::AllowMoreInput(sal_Unicode uChar)
1012 {
1013     //arbitrarily chosen maximum length allowed - normal max usage would be around 30.
1014     if( maInput.getLength() > 255 )
1015         mbAllowMoreChars = false;
1016
1017     if( !mbAllowMoreChars )
1018         return false;
1019
1020     bool bPreventNonHex = false;
1021     if( maInput.indexOf("U+") != -1 )
1022         bPreventNonHex = true;
1023
1024     switch ( unicode::getUnicodeType(uChar) )
1025     {
1026         case css::i18n::UnicodeType::SURROGATE:
1027             if( bPreventNonHex )
1028             {
1029                 mbAllowMoreChars = false;
1030                 return false;
1031             }
1032
1033             if( rtl::isLowSurrogate(uChar) && maUtf16.isEmpty() && maInput.isEmpty()  )
1034             {
1035                 maUtf16.append(uChar);
1036                 return true;
1037             }
1038             if( rtl::isHighSurrogate(uChar) && maInput.isEmpty() )
1039                 maUtf16.insert(0, uChar );
1040             //end of hex strings, or unexpected order of high/low, so don't accept more
1041             if( !maUtf16.isEmpty() )
1042                 maInput.append(maUtf16);
1043             if( !maCombining.isEmpty() )
1044                 maInput.append(maCombining);
1045             mbAllowMoreChars = false;
1046             break;
1047
1048         case css::i18n::UnicodeType::NON_SPACING_MARK:
1049         case css::i18n::UnicodeType::COMBINING_SPACING_MARK:
1050             if( bPreventNonHex )
1051             {
1052                 mbAllowMoreChars = false;
1053                 return false;
1054             }
1055
1056             //extreme edge case: already invalid high/low surrogates with preceding combining chars, and now an extra combining mark.
1057             if( !maUtf16.isEmpty() )
1058             {
1059                 maInput = maUtf16;
1060                 if( !maCombining.isEmpty() )
1061                     maInput.append(maCombining);
1062                 mbAllowMoreChars = false;
1063                 return false;
1064             }
1065             maCombining.insert(0, uChar);
1066             break;
1067
1068         default:
1069             //extreme edge case: already invalid high/low surrogates with preceding combining chars, and now an extra character.
1070             if( !maUtf16.isEmpty() )
1071             {
1072                 maInput = maUtf16;
1073                 if( !maCombining.isEmpty() )
1074                     maInput.append(maCombining);
1075                 mbAllowMoreChars = false;
1076                 return false;
1077             }
1078
1079             if( !maCombining.isEmpty() )
1080             {
1081                 maCombining.insert(0, uChar);
1082                 maInput = maCombining;
1083                 mbAllowMoreChars = false;
1084                 return false;
1085             }
1086
1087             // 0 - 1f are control characters.  Do not process those.
1088             if( uChar < 0x20 )
1089             {
1090                 mbAllowMoreChars = false;
1091                 return false;
1092             }
1093
1094             switch( uChar )
1095             {
1096                 case 'u':
1097                 case 'U':
1098                     // U+ notation found.  Continue looking for another one.
1099                     if( mbRequiresU )
1100                     {
1101                         mbRequiresU = false;
1102                         maInput.insert(0,"U+");
1103                     }
1104                     // treat as a normal character
1105                     else
1106                     {
1107                         mbAllowMoreChars = false;
1108                         if( !bPreventNonHex )
1109                             maInput.insertUtf32(0, uChar);
1110                     }
1111                     break;
1112                 case '+':
1113                     // + already found: skip when not U, or edge case of +U+xxxx
1114                     if( mbRequiresU || (maInput.indexOf("U+") == 0) )
1115                         mbAllowMoreChars = false;
1116                     // hex chars followed by '+' - now require a 'U'
1117                     else if ( !maInput.isEmpty() )
1118                         mbRequiresU = true;
1119                     // treat as a normal character
1120                     else
1121                     {
1122                         mbAllowMoreChars = false;
1123                         if( !bPreventNonHex )
1124                             maInput.insertUtf32(0, uChar);
1125                     }
1126                     break;
1127                 default:
1128                     // + already found. Since not U, cancel further input
1129                     if( mbRequiresU )
1130                         mbAllowMoreChars = false;
1131                     // maximum digits per notation is 8: only one notation
1132                     else if( maInput.indexOf("U+") == -1 && maInput.getLength() == 8 )
1133                         mbAllowMoreChars = false;
1134                     // maximum digits per notation is 8: previous notation found
1135                     else if( maInput.indexOf("U+") == 8 )
1136                         mbAllowMoreChars = false;
1137                     // a hex character. Add to string.
1138                     else if( isxdigit(uChar) )
1139                     {
1140                         mbIsHexString = true;
1141                         maInput.insertUtf32(0, uChar);
1142                     }
1143                     // not a hex character: stop input. keep if it is the first input provided
1144                     else
1145                     {
1146                         mbAllowMoreChars = false;
1147                         if( maInput.isEmpty() )
1148                             maInput.insertUtf32(0, uChar);
1149                     }
1150             }
1151     }
1152     return mbAllowMoreChars;
1153 }
1154
1155 OUString ToggleUnicodeCodepoint::StringToReplace()
1156 {
1157     if( maInput.isEmpty() )
1158     {
1159         //edge case - input finished with incomplete low surrogate or combining characters without a base
1160         if( mbAllowMoreChars )
1161         {
1162             if( !maUtf16.isEmpty() )
1163                 maInput = maUtf16;
1164             if( !maCombining.isEmpty() )
1165                 maInput.append(maCombining);
1166         }
1167         return maInput.toString();
1168     }
1169
1170     if( !mbIsHexString )
1171         return maInput.toString();
1172
1173     //this function potentially modifies the input string.  Prevent addition of further characters
1174     mbAllowMoreChars = false;
1175
1176     //validate unicode notation.
1177     OUStringBuffer sIn;
1178     sal_uInt32 nUnicode = 0;
1179     sal_Int32 nUPlus = maInput.indexOf("U+");
1180     //if U+ notation used, strip off all extra chars added not in U+ notation
1181     if( nUPlus != -1 )
1182     {
1183         maInput = maInput.copy(nUPlus);
1184         sIn = maInput.copy(2);
1185         nUPlus = sIn.indexOf("U+");
1186     }
1187     else
1188         sIn = maInput;
1189     while( nUPlus != -1 )
1190     {
1191         nUnicode = sIn.copy(0, nUPlus).toString().toUInt32(16);
1192         //prevent creating control characters or invalid Unicode values
1193         if( !rtl::isUnicodeCodePoint(nUnicode) || nUnicode < 0x20  )
1194             maInput = sIn.copy(nUPlus);
1195         sIn = sIn.copy(nUPlus+2);
1196         nUPlus =  sIn.indexOf("U+");
1197     }
1198
1199     nUnicode = sIn.toString().toUInt32(16);
1200     if( !rtl::isUnicodeCodePoint(nUnicode) || nUnicode < 0x20 )
1201        maInput.truncate().append( sIn[sIn.getLength()-1] );
1202     return maInput.toString();
1203 }
1204
1205 sal_uInt32 ToggleUnicodeCodepoint::CharsToDelete()
1206 {
1207     OUString sIn = StringToReplace();
1208     sal_Int32 nPos = 0;
1209     sal_uInt32 counter = 0;
1210     while( nPos < sIn.getLength() )
1211     {
1212         sIn.iterateCodePoints(&nPos);
1213         ++counter;
1214     }
1215     return counter;
1216 }
1217
1218 OUString ToggleUnicodeCodepoint::ReplacementString()
1219 {
1220     OUString sIn = StringToReplace();
1221     maOutput = "";
1222     sal_Int32 nUPlus = sIn.indexOf("U+");
1223     // convert from hex notation to glyph
1224     if( nUPlus != -1 || (sIn.getLength() > 1 && mbIsHexString) )
1225     {
1226         sal_uInt32 nUnicode = 0;
1227         if( nUPlus == 0)
1228         {
1229             sIn = sIn.copy(2);
1230             nUPlus = sIn.indexOf("U+");
1231         }
1232         while( nUPlus > 0 )
1233         {
1234             nUnicode = sIn.copy(0, nUPlus).toUInt32(16);
1235             maOutput.appendUtf32( nUnicode );
1236
1237             sIn = sIn.copy(nUPlus+2);
1238             nUPlus = sIn.indexOf("U+");
1239         }
1240         nUnicode = sIn.toUInt32(16);
1241         maOutput.appendUtf32( nUnicode );
1242     }
1243     // convert from glyph to hex notation
1244     else
1245     {
1246         sal_Int32 nPos = 0;
1247         while( nPos < sIn.getLength() )
1248         {
1249             OUStringBuffer aTmp = OUString::number(sIn.iterateCodePoints(&nPos),16);
1250             //pad with zeros - minimum length of 4.
1251             for( sal_Int32 i = 4 - aTmp.getLength(); i > 0; --i )
1252                 aTmp.insert( 0,"0" );
1253             maOutput.append( "U+" );
1254             maOutput.append( aTmp );
1255         }
1256     }
1257     return maOutput.toString();
1258 }
1259
1260 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */