1 /* 2 * Copyright (C) 2006 George Staikos <staikos (at) kde.org> 3 * Copyright (C) 2006 Alexey Proskuryakov <ap (at) nypop.com> 4 * Copyright (C) 2006, 2007, 2008, 2009 Apple Inc. All rights reserved. 5 * 6 * This library is free software; you can redistribute it and/or 7 * modify it under the terms of the GNU Library General Public 8 * License as published by the Free Software Foundation; either 9 * version 2 of the License, or (at your option) any later version. 10 * 11 * This library is distributed in the hope that it will be useful, 12 * but WITHOUT ANY WARRANTY; without even the implied warranty of 13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 14 * Library General Public License for more details. 15 * 16 * You should have received a copy of the GNU Library General Public License 17 * along with this library; see the file COPYING.LIB. If not, write to 18 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, 19 * Boston, MA 02110-1301, USA. 20 * 21 */ 22 23 #ifndef WTF_UNICODE_ICU_H 24 #define WTF_UNICODE_ICU_H 25 26 #include <unicode/uchar.h> 27 #include <unicode/ustring.h> 28 29 namespace WTF { 30 31 namespace Unicode { 32 33 enum Direction { 34 LeftToRight = U_LEFT_TO_RIGHT, 35 RightToLeft = U_RIGHT_TO_LEFT, 36 EuropeanNumber = U_EUROPEAN_NUMBER, 37 EuropeanNumberSeparator = U_EUROPEAN_NUMBER_SEPARATOR, 38 EuropeanNumberTerminator = U_EUROPEAN_NUMBER_TERMINATOR, 39 ArabicNumber = U_ARABIC_NUMBER, 40 CommonNumberSeparator = U_COMMON_NUMBER_SEPARATOR, 41 BlockSeparator = U_BLOCK_SEPARATOR, 42 SegmentSeparator = U_SEGMENT_SEPARATOR, 43 WhiteSpaceNeutral = U_WHITE_SPACE_NEUTRAL, 44 OtherNeutral = U_OTHER_NEUTRAL, 45 LeftToRightEmbedding = U_LEFT_TO_RIGHT_EMBEDDING, 46 LeftToRightOverride = U_LEFT_TO_RIGHT_OVERRIDE, 47 RightToLeftArabic = U_RIGHT_TO_LEFT_ARABIC, 48 RightToLeftEmbedding = U_RIGHT_TO_LEFT_EMBEDDING, 49 RightToLeftOverride = U_RIGHT_TO_LEFT_OVERRIDE, 50 PopDirectionalFormat = U_POP_DIRECTIONAL_FORMAT, 51 NonSpacingMark = U_DIR_NON_SPACING_MARK, 52 BoundaryNeutral = U_BOUNDARY_NEUTRAL 53 }; 54 55 enum DecompositionType { 56 DecompositionNone = U_DT_NONE, 57 DecompositionCanonical = U_DT_CANONICAL, 58 DecompositionCompat = U_DT_COMPAT, 59 DecompositionCircle = U_DT_CIRCLE, 60 DecompositionFinal = U_DT_FINAL, 61 DecompositionFont = U_DT_FONT, 62 DecompositionFraction = U_DT_FRACTION, 63 DecompositionInitial = U_DT_INITIAL, 64 DecompositionIsolated = U_DT_ISOLATED, 65 DecompositionMedial = U_DT_MEDIAL, 66 DecompositionNarrow = U_DT_NARROW, 67 DecompositionNoBreak = U_DT_NOBREAK, 68 DecompositionSmall = U_DT_SMALL, 69 DecompositionSquare = U_DT_SQUARE, 70 DecompositionSub = U_DT_SUB, 71 DecompositionSuper = U_DT_SUPER, 72 DecompositionVertical = U_DT_VERTICAL, 73 DecompositionWide = U_DT_WIDE, 74 }; 75 76 enum CharCategory { 77 NoCategory = 0, 78 Other_NotAssigned = U_MASK(U_GENERAL_OTHER_TYPES), 79 Letter_Uppercase = U_MASK(U_UPPERCASE_LETTER), 80 Letter_Lowercase = U_MASK(U_LOWERCASE_LETTER), 81 Letter_Titlecase = U_MASK(U_TITLECASE_LETTER), 82 Letter_Modifier = U_MASK(U_MODIFIER_LETTER), 83 Letter_Other = U_MASK(U_OTHER_LETTER), 84 85 Mark_NonSpacing = U_MASK(U_NON_SPACING_MARK), 86 Mark_Enclosing = U_MASK(U_ENCLOSING_MARK), 87 Mark_SpacingCombining = U_MASK(U_COMBINING_SPACING_MARK), 88 89 Number_DecimalDigit = U_MASK(U_DECIMAL_DIGIT_NUMBER), 90 Number_Letter = U_MASK(U_LETTER_NUMBER), 91 Number_Other = U_MASK(U_OTHER_NUMBER), 92 93 Separator_Space = U_MASK(U_SPACE_SEPARATOR), 94 Separator_Line = U_MASK(U_LINE_SEPARATOR), 95 Separator_Paragraph = U_MASK(U_PARAGRAPH_SEPARATOR), 96 97 Other_Control = U_MASK(U_CONTROL_CHAR), 98 Other_Format = U_MASK(U_FORMAT_CHAR), 99 Other_PrivateUse = U_MASK(U_PRIVATE_USE_CHAR), 100 Other_Surrogate = U_MASK(U_SURROGATE), 101 102 Punctuation_Dash = U_MASK(U_DASH_PUNCTUATION), 103 Punctuation_Open = U_MASK(U_START_PUNCTUATION), 104 Punctuation_Close = U_MASK(U_END_PUNCTUATION), 105 Punctuation_Connector = U_MASK(U_CONNECTOR_PUNCTUATION), 106 Punctuation_Other = U_MASK(U_OTHER_PUNCTUATION), 107 108 Symbol_Math = U_MASK(U_MATH_SYMBOL), 109 Symbol_Currency = U_MASK(U_CURRENCY_SYMBOL), 110 Symbol_Modifier = U_MASK(U_MODIFIER_SYMBOL), 111 Symbol_Other = U_MASK(U_OTHER_SYMBOL), 112 113 Punctuation_InitialQuote = U_MASK(U_INITIAL_PUNCTUATION), 114 Punctuation_FinalQuote = U_MASK(U_FINAL_PUNCTUATION) 115 }; 116 117 inline UChar32 foldCase(UChar32 c) 118 { 119 return u_foldCase(c, U_FOLD_CASE_DEFAULT); 120 } 121 122 inline int foldCase(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error) 123 { 124 UErrorCode status = U_ZERO_ERROR; 125 int realLength = u_strFoldCase(result, resultLength, src, srcLength, U_FOLD_CASE_DEFAULT, &status); 126 *error = !U_SUCCESS(status); 127 return realLength; 128 } 129 130 inline int toLower(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error) 131 { 132 UErrorCode status = U_ZERO_ERROR; 133 int realLength = u_strToLower(result, resultLength, src, srcLength, "", &status); 134 *error = !!U_FAILURE(status); 135 return realLength; 136 } 137 138 inline UChar32 toLower(UChar32 c) 139 { 140 return u_tolower(c); 141 } 142 143 inline UChar32 toUpper(UChar32 c) 144 { 145 return u_toupper(c); 146 } 147 148 inline int toUpper(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error) 149 { 150 UErrorCode status = U_ZERO_ERROR; 151 int realLength = u_strToUpper(result, resultLength, src, srcLength, "", &status); 152 *error = !!U_FAILURE(status); 153 return realLength; 154 } 155 156 inline UChar32 toTitleCase(UChar32 c) 157 { 158 return u_totitle(c); 159 } 160 161 inline bool isArabicChar(UChar32 c) 162 { 163 return ublock_getCode(c) == UBLOCK_ARABIC; 164 } 165 166 inline bool isAlphanumeric(UChar32 c) 167 { 168 return u_isalnum(c); 169 } 170 171 inline bool isSeparatorSpace(UChar32 c) 172 { 173 return u_charType(c) == U_SPACE_SEPARATOR; 174 } 175 176 inline bool isPrintableChar(UChar32 c) 177 { 178 return !!u_isprint(c); 179 } 180 181 inline bool isPunct(UChar32 c) 182 { 183 return !!u_ispunct(c); 184 } 185 186 inline bool hasLineBreakingPropertyComplexContext(UChar32 c) 187 { 188 return u_getIntPropertyValue(c, UCHAR_LINE_BREAK) == U_LB_COMPLEX_CONTEXT; 189 } 190 191 inline UChar32 mirroredChar(UChar32 c) 192 { 193 return u_charMirror(c); 194 } 195 196 inline CharCategory category(UChar32 c) 197 { 198 return static_cast<CharCategory>(U_GET_GC_MASK(c)); 199 } 200 201 inline Direction direction(UChar32 c) 202 { 203 return static_cast<Direction>(u_charDirection(c)); 204 } 205 206 inline bool isLower(UChar32 c) 207 { 208 return !!u_islower(c); 209 } 210 211 inline uint8_t combiningClass(UChar32 c) 212 { 213 return u_getCombiningClass(c); 214 } 215 216 inline DecompositionType decompositionType(UChar32 c) 217 { 218 return static_cast<DecompositionType>(u_getIntPropertyValue(c, UCHAR_DECOMPOSITION_TYPE)); 219 } 220 221 inline int umemcasecmp(const UChar* a, const UChar* b, int len) 222 { 223 return u_memcasecmp(a, b, len, U_FOLD_CASE_DEFAULT); 224 } 225 226 } // namespace Unicode 227 228 } // namespace WTF 229 230 #endif // WTF_UNICODE_ICU_H 231
