OLD | NEW |
| (Empty) |
1 /* | |
2 * Copyright (C) 2006 George Staikos <staikos@kde.org> | |
3 * Copyright (C) 2006 Alexey Proskuryakov <ap@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 #if USE(ICU_UNICODE) | |
27 | |
28 #include <stdlib.h> | |
29 #include <unicode/uchar.h> | |
30 #include <unicode/uscript.h> | |
31 #include <unicode/ustring.h> | |
32 #include <unicode/utf16.h> | |
33 | |
34 namespace WTF { | |
35 | |
36 namespace Unicode { | |
37 | |
38 enum Direction { | |
39 LeftToRight = U_LEFT_TO_RIGHT, | |
40 RightToLeft = U_RIGHT_TO_LEFT, | |
41 EuropeanNumber = U_EUROPEAN_NUMBER, | |
42 EuropeanNumberSeparator = U_EUROPEAN_NUMBER_SEPARATOR, | |
43 EuropeanNumberTerminator = U_EUROPEAN_NUMBER_TERMINATOR, | |
44 ArabicNumber = U_ARABIC_NUMBER, | |
45 CommonNumberSeparator = U_COMMON_NUMBER_SEPARATOR, | |
46 BlockSeparator = U_BLOCK_SEPARATOR, | |
47 SegmentSeparator = U_SEGMENT_SEPARATOR, | |
48 WhiteSpaceNeutral = U_WHITE_SPACE_NEUTRAL, | |
49 OtherNeutral = U_OTHER_NEUTRAL, | |
50 LeftToRightEmbedding = U_LEFT_TO_RIGHT_EMBEDDING, | |
51 LeftToRightOverride = U_LEFT_TO_RIGHT_OVERRIDE, | |
52 RightToLeftArabic = U_RIGHT_TO_LEFT_ARABIC, | |
53 RightToLeftEmbedding = U_RIGHT_TO_LEFT_EMBEDDING, | |
54 RightToLeftOverride = U_RIGHT_TO_LEFT_OVERRIDE, | |
55 PopDirectionalFormat = U_POP_DIRECTIONAL_FORMAT, | |
56 NonSpacingMark = U_DIR_NON_SPACING_MARK, | |
57 BoundaryNeutral = U_BOUNDARY_NEUTRAL | |
58 }; | |
59 | |
60 enum DecompositionType { | |
61 DecompositionNone = U_DT_NONE, | |
62 DecompositionCanonical = U_DT_CANONICAL, | |
63 DecompositionCompat = U_DT_COMPAT, | |
64 DecompositionCircle = U_DT_CIRCLE, | |
65 DecompositionFinal = U_DT_FINAL, | |
66 DecompositionFont = U_DT_FONT, | |
67 DecompositionFraction = U_DT_FRACTION, | |
68 DecompositionInitial = U_DT_INITIAL, | |
69 DecompositionIsolated = U_DT_ISOLATED, | |
70 DecompositionMedial = U_DT_MEDIAL, | |
71 DecompositionNarrow = U_DT_NARROW, | |
72 DecompositionNoBreak = U_DT_NOBREAK, | |
73 DecompositionSmall = U_DT_SMALL, | |
74 DecompositionSquare = U_DT_SQUARE, | |
75 DecompositionSub = U_DT_SUB, | |
76 DecompositionSuper = U_DT_SUPER, | |
77 DecompositionVertical = U_DT_VERTICAL, | |
78 DecompositionWide = U_DT_WIDE, | |
79 }; | |
80 | |
81 enum CharCategory { | |
82 NoCategory = 0, | |
83 Other_NotAssigned = U_MASK(U_GENERAL_OTHER_TYPES), | |
84 Letter_Uppercase = U_MASK(U_UPPERCASE_LETTER), | |
85 Letter_Lowercase = U_MASK(U_LOWERCASE_LETTER), | |
86 Letter_Titlecase = U_MASK(U_TITLECASE_LETTER), | |
87 Letter_Modifier = U_MASK(U_MODIFIER_LETTER), | |
88 Letter_Other = U_MASK(U_OTHER_LETTER), | |
89 | |
90 Mark_NonSpacing = U_MASK(U_NON_SPACING_MARK), | |
91 Mark_Enclosing = U_MASK(U_ENCLOSING_MARK), | |
92 Mark_SpacingCombining = U_MASK(U_COMBINING_SPACING_MARK), | |
93 | |
94 Number_DecimalDigit = U_MASK(U_DECIMAL_DIGIT_NUMBER), | |
95 Number_Letter = U_MASK(U_LETTER_NUMBER), | |
96 Number_Other = U_MASK(U_OTHER_NUMBER), | |
97 | |
98 Separator_Space = U_MASK(U_SPACE_SEPARATOR), | |
99 Separator_Line = U_MASK(U_LINE_SEPARATOR), | |
100 Separator_Paragraph = U_MASK(U_PARAGRAPH_SEPARATOR), | |
101 | |
102 Other_Control = U_MASK(U_CONTROL_CHAR), | |
103 Other_Format = U_MASK(U_FORMAT_CHAR), | |
104 Other_PrivateUse = U_MASK(U_PRIVATE_USE_CHAR), | |
105 Other_Surrogate = U_MASK(U_SURROGATE), | |
106 | |
107 Punctuation_Dash = U_MASK(U_DASH_PUNCTUATION), | |
108 Punctuation_Open = U_MASK(U_START_PUNCTUATION), | |
109 Punctuation_Close = U_MASK(U_END_PUNCTUATION), | |
110 Punctuation_Connector = U_MASK(U_CONNECTOR_PUNCTUATION), | |
111 Punctuation_Other = U_MASK(U_OTHER_PUNCTUATION), | |
112 | |
113 Symbol_Math = U_MASK(U_MATH_SYMBOL), | |
114 Symbol_Currency = U_MASK(U_CURRENCY_SYMBOL), | |
115 Symbol_Modifier = U_MASK(U_MODIFIER_SYMBOL), | |
116 Symbol_Other = U_MASK(U_OTHER_SYMBOL), | |
117 | |
118 Punctuation_InitialQuote = U_MASK(U_INITIAL_PUNCTUATION), | |
119 Punctuation_FinalQuote = U_MASK(U_FINAL_PUNCTUATION) | |
120 }; | |
121 | |
122 inline UChar32 foldCase(UChar32 c) | |
123 { | |
124 return u_foldCase(c, U_FOLD_CASE_DEFAULT); | |
125 } | |
126 | |
127 inline int foldCase(UChar* result, int resultLength, const UChar* src, int srcLe
ngth, bool* error) | |
128 { | |
129 UErrorCode status = U_ZERO_ERROR; | |
130 int realLength = u_strFoldCase(result, resultLength, src, srcLength, U_FOLD_
CASE_DEFAULT, &status); | |
131 *error = !U_SUCCESS(status); | |
132 return realLength; | |
133 } | |
134 | |
135 inline int toLower(UChar* result, int resultLength, const UChar* src, int srcLen
gth, bool* error) | |
136 { | |
137 UErrorCode status = U_ZERO_ERROR; | |
138 int realLength = u_strToLower(result, resultLength, src, srcLength, "", &sta
tus); | |
139 *error = !!U_FAILURE(status); | |
140 return realLength; | |
141 } | |
142 | |
143 inline UChar32 toLower(UChar32 c) | |
144 { | |
145 return u_tolower(c); | |
146 } | |
147 | |
148 inline UChar32 toUpper(UChar32 c) | |
149 { | |
150 return u_toupper(c); | |
151 } | |
152 | |
153 inline int toUpper(UChar* result, int resultLength, const UChar* src, int srcLen
gth, bool* error) | |
154 { | |
155 UErrorCode status = U_ZERO_ERROR; | |
156 int realLength = u_strToUpper(result, resultLength, src, srcLength, "", &sta
tus); | |
157 *error = !!U_FAILURE(status); | |
158 return realLength; | |
159 } | |
160 | |
161 inline UChar32 toTitleCase(UChar32 c) | |
162 { | |
163 return u_totitle(c); | |
164 } | |
165 | |
166 inline bool isArabicChar(UChar32 c) | |
167 { | |
168 return ublock_getCode(c) == UBLOCK_ARABIC; | |
169 } | |
170 | |
171 inline bool isAlphanumeric(UChar32 c) | |
172 { | |
173 return u_isalnum(c); | |
174 } | |
175 | |
176 inline bool isSeparatorSpace(UChar32 c) | |
177 { | |
178 return u_charType(c) == U_SPACE_SEPARATOR; | |
179 } | |
180 | |
181 inline bool isPrintableChar(UChar32 c) | |
182 { | |
183 return !!u_isprint(c); | |
184 } | |
185 | |
186 inline bool isPunct(UChar32 c) | |
187 { | |
188 return !!u_ispunct(c); | |
189 } | |
190 | |
191 inline bool hasLineBreakingPropertyComplexContext(UChar32 c) | |
192 { | |
193 return u_getIntPropertyValue(c, UCHAR_LINE_BREAK) == U_LB_COMPLEX_CONTEXT; | |
194 } | |
195 | |
196 inline UChar32 mirroredChar(UChar32 c) | |
197 { | |
198 return u_charMirror(c); | |
199 } | |
200 | |
201 inline CharCategory category(UChar32 c) | |
202 { | |
203 return static_cast<CharCategory>(U_GET_GC_MASK(c)); | |
204 } | |
205 | |
206 inline Direction direction(UChar32 c) | |
207 { | |
208 return static_cast<Direction>(u_charDirection(c)); | |
209 } | |
210 | |
211 inline bool isLower(UChar32 c) | |
212 { | |
213 return !!u_islower(c); | |
214 } | |
215 | |
216 inline uint8_t combiningClass(UChar32 c) | |
217 { | |
218 return u_getCombiningClass(c); | |
219 } | |
220 | |
221 inline DecompositionType decompositionType(UChar32 c) | |
222 { | |
223 return static_cast<DecompositionType>(u_getIntPropertyValue(c, UCHAR_DECOMPO
SITION_TYPE)); | |
224 } | |
225 | |
226 inline int umemcasecmp(const UChar* a, const UChar* b, int len) | |
227 { | |
228 return u_memcasecmp(a, b, len, U_FOLD_CASE_DEFAULT); | |
229 } | |
230 | |
231 } // namespace Unicode | |
232 | |
233 } // namespace WTF | |
234 | |
235 #endif // USE(ICU_UNICODE) | |
236 | |
237 #endif // WTF_UNICODE_ICU_H | |
OLD | NEW |