1 /*
2 * Copyright (C) 2008 Jürg Billeter <j@bitron.ch>
3 * Copyright (C) 2008 Dominik Röttsches <dominik.roettsches@access-company.com>
4 *
5 * This library is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU Library General Public
7 * License as published by the Free Software Foundation; either
8 * version 2 of the License, or (at your option) any later version.
9 *
10 * This library is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 * Library General Public License for more details.
14 *
15 * You should have received a copy of the GNU Library General Public License
16 * along with this library; see the file COPYING.LIB. If not, write to
17 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
18 * Boston, MA 02110-1301, USA.
19 *
20 */
21
22 #include "config.h"
23 #include "UnicodeGLib.h"
24
25 namespace WTF {
26 namespace Unicode {
27
foldCase(UChar32 ch)28 UChar32 foldCase(UChar32 ch)
29 {
30 GOwnPtr<GError> gerror;
31
32 GOwnPtr<char> utf8char;
33 utf8char.set(g_ucs4_to_utf8(reinterpret_cast<gunichar*>(&ch), 1, 0, 0, &gerror.outPtr()));
34 if (gerror)
35 return ch;
36
37 GOwnPtr<char> utf8caseFolded;
38 utf8caseFolded.set(g_utf8_casefold(utf8char.get(), -1));
39
40 GOwnPtr<gunichar> ucs4Result;
41 ucs4Result.set(g_utf8_to_ucs4_fast(utf8caseFolded.get(), -1, 0));
42
43 return *ucs4Result;
44 }
45
foldCase(UChar * result,int resultLength,const UChar * src,int srcLength,bool * error)46 int foldCase(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error)
47 {
48 *error = false;
49 GOwnPtr<GError> gerror;
50
51 GOwnPtr<char> utf8src;
52 utf8src.set(g_utf16_to_utf8(src, srcLength, 0, 0, &gerror.outPtr()));
53 if (gerror) {
54 *error = true;
55 return -1;
56 }
57
58 GOwnPtr<char> utf8result;
59 utf8result.set(g_utf8_casefold(utf8src.get(), -1));
60
61 long utf16resultLength = -1;
62 GOwnPtr<UChar> utf16result;
63 utf16result.set(g_utf8_to_utf16(utf8result.get(), -1, 0, &utf16resultLength, &gerror.outPtr()));
64 if (gerror) {
65 *error = true;
66 return -1;
67 }
68
69 if (utf16resultLength > resultLength) {
70 *error = true;
71 return utf16resultLength;
72 }
73 memcpy(result, utf16result.get(), utf16resultLength * sizeof(UChar));
74
75 return utf16resultLength;
76 }
77
toLower(UChar * result,int resultLength,const UChar * src,int srcLength,bool * error)78 int toLower(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error)
79 {
80 *error = false;
81 GOwnPtr<GError> gerror;
82
83 GOwnPtr<char> utf8src;
84 utf8src.set(g_utf16_to_utf8(src, srcLength, 0, 0, &gerror.outPtr()));
85 if (gerror) {
86 *error = true;
87 return -1;
88 }
89
90 GOwnPtr<char> utf8result;
91 utf8result.set(g_utf8_strdown(utf8src.get(), -1));
92
93 long utf16resultLength = -1;
94 GOwnPtr<UChar> utf16result;
95 utf16result.set(g_utf8_to_utf16(utf8result.get(), -1, 0, &utf16resultLength, &gerror.outPtr()));
96 if (gerror) {
97 *error = true;
98 return -1;
99 }
100
101 if (utf16resultLength > resultLength) {
102 *error = true;
103 return utf16resultLength;
104 }
105 memcpy(result, utf16result.get(), utf16resultLength * sizeof(UChar));
106
107 return utf16resultLength;
108 }
109
toUpper(UChar * result,int resultLength,const UChar * src,int srcLength,bool * error)110 int toUpper(UChar* result, int resultLength, const UChar* src, int srcLength, bool* error)
111 {
112 *error = false;
113 GOwnPtr<GError> gerror;
114
115 GOwnPtr<char> utf8src;
116 utf8src.set(g_utf16_to_utf8(src, srcLength, 0, 0, &gerror.outPtr()));
117 if (gerror) {
118 *error = true;
119 return -1;
120 }
121
122 GOwnPtr<char> utf8result;
123 utf8result.set(g_utf8_strup(utf8src.get(), -1));
124
125 long utf16resultLength = -1;
126 GOwnPtr<UChar> utf16result;
127 utf16result.set(g_utf8_to_utf16(utf8result.get(), -1, 0, &utf16resultLength, &gerror.outPtr()));
128 if (gerror) {
129 *error = true;
130 return -1;
131 }
132
133 if (utf16resultLength > resultLength) {
134 *error = true;
135 return utf16resultLength;
136 }
137 memcpy(result, utf16result.get(), utf16resultLength * sizeof(UChar));
138
139 return utf16resultLength;
140 }
141
direction(UChar32 c)142 Direction direction(UChar32 c)
143 {
144 PangoBidiType type = pango_bidi_type_for_unichar(c);
145 switch (type) {
146 case PANGO_BIDI_TYPE_L:
147 return LeftToRight;
148 case PANGO_BIDI_TYPE_R:
149 return RightToLeft;
150 case PANGO_BIDI_TYPE_AL:
151 return RightToLeftArabic;
152 case PANGO_BIDI_TYPE_LRE:
153 return LeftToRightEmbedding;
154 case PANGO_BIDI_TYPE_RLE:
155 return RightToLeftEmbedding;
156 case PANGO_BIDI_TYPE_LRO:
157 return LeftToRightOverride;
158 case PANGO_BIDI_TYPE_RLO:
159 return RightToLeftOverride;
160 case PANGO_BIDI_TYPE_PDF:
161 return PopDirectionalFormat;
162 case PANGO_BIDI_TYPE_EN:
163 return EuropeanNumber;
164 case PANGO_BIDI_TYPE_AN:
165 return ArabicNumber;
166 case PANGO_BIDI_TYPE_ES:
167 return EuropeanNumberSeparator;
168 case PANGO_BIDI_TYPE_ET:
169 return EuropeanNumberTerminator;
170 case PANGO_BIDI_TYPE_CS:
171 return CommonNumberSeparator;
172 case PANGO_BIDI_TYPE_NSM:
173 return NonSpacingMark;
174 case PANGO_BIDI_TYPE_BN:
175 return BoundaryNeutral;
176 case PANGO_BIDI_TYPE_B:
177 return BlockSeparator;
178 case PANGO_BIDI_TYPE_S:
179 return SegmentSeparator;
180 case PANGO_BIDI_TYPE_WS:
181 return WhiteSpaceNeutral;
182 default:
183 return OtherNeutral;
184 }
185 }
186
umemcasecmp(const UChar * a,const UChar * b,int len)187 int umemcasecmp(const UChar* a, const UChar* b, int len)
188 {
189 GOwnPtr<char> utf8a;
190 GOwnPtr<char> utf8b;
191
192 utf8a.set(g_utf16_to_utf8(a, len, 0, 0, 0));
193 utf8b.set(g_utf16_to_utf8(b, len, 0, 0, 0));
194
195 GOwnPtr<char> foldedA;
196 GOwnPtr<char> foldedB;
197
198 foldedA.set(g_utf8_casefold(utf8a.get(), -1));
199 foldedB.set(g_utf8_casefold(utf8b.get(), -1));
200
201 // FIXME: umemcasecmp needs to mimic u_memcasecmp of icu
202 // from the ICU docs:
203 // "Compare two strings case-insensitively using full case folding.
204 // his is equivalent to u_strcmp(u_strFoldCase(s1, n, options), u_strFoldCase(s2, n, options))."
205 //
206 // So it looks like we don't need the full g_utf8_collate here,
207 // but really a bitwise comparison of casefolded unicode chars (not utf-8 bytes).
208 // As there is no direct equivalent to this icu function in GLib, for now
209 // we'll use g_utf8_collate():
210
211 return g_utf8_collate(foldedA.get(), foldedB.get());
212 }
213
214 }
215 }
216