1 | /**********************************************************************
|
---|
2 | unicode.c - Oniguruma (regular expression library)
|
---|
3 | **********************************************************************/
|
---|
4 | /*-
|
---|
5 | * Copyright (c) 2002-2013 K.Kosako <sndgk393 AT ybb DOT ne DOT jp>
|
---|
6 | * All rights reserved.
|
---|
7 | *
|
---|
8 | * Redistribution and use in source and binary forms, with or without
|
---|
9 | * modification, are permitted provided that the following conditions
|
---|
10 | * are met:
|
---|
11 | * 1. Redistributions of source code must retain the above copyright
|
---|
12 | * notice, this list of conditions and the following disclaimer.
|
---|
13 | * 2. Redistributions in binary form must reproduce the above copyright
|
---|
14 | * notice, this list of conditions and the following disclaimer in the
|
---|
15 | * documentation and/or other materials provided with the distribution.
|
---|
16 | *
|
---|
17 | * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
---|
18 | * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
---|
19 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
---|
20 | * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
---|
21 | * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
---|
22 | * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
---|
23 | * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
---|
24 | * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
---|
25 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
---|
26 | * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
---|
27 | * SUCH DAMAGE.
|
---|
28 | */
|
---|
29 |
|
---|
30 | #include "regint.h"
|
---|
31 |
|
---|
32 | #define ONIGENC_IS_UNICODE_ISO_8859_1_CTYPE(code,ctype) \
|
---|
33 | ((EncUNICODE_ISO_8859_1_CtypeTable[code] & CTYPE_TO_BIT(ctype)) != 0)
|
---|
34 | #if 0
|
---|
35 | #define ONIGENC_IS_UNICODE_ISO_8859_1_BIT_CTYPE(code,cbit) \
|
---|
36 | ((EncUNICODE_ISO_8859_1_CtypeTable[code] & (cbit)) != 0)
|
---|
37 | #endif
|
---|
38 |
|
---|
39 | static const unsigned short EncUNICODE_ISO_8859_1_CtypeTable[256] = {
|
---|
40 | 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008,
|
---|
41 | 0x4008, 0x420c, 0x4209, 0x4208, 0x4208, 0x4208, 0x4008, 0x4008,
|
---|
42 | 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008,
|
---|
43 | 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008,
|
---|
44 | 0x4284, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0,
|
---|
45 | 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0,
|
---|
46 | 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0,
|
---|
47 | 0x78b0, 0x78b0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0,
|
---|
48 | 0x41a0, 0x7ca2, 0x7ca2, 0x7ca2, 0x7ca2, 0x7ca2, 0x7ca2, 0x74a2,
|
---|
49 | 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2,
|
---|
50 | 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2,
|
---|
51 | 0x74a2, 0x74a2, 0x74a2, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x51a0,
|
---|
52 | 0x41a0, 0x78e2, 0x78e2, 0x78e2, 0x78e2, 0x78e2, 0x78e2, 0x70e2,
|
---|
53 | 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2,
|
---|
54 | 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2,
|
---|
55 | 0x70e2, 0x70e2, 0x70e2, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x4008,
|
---|
56 | 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0288, 0x0008, 0x0008,
|
---|
57 | 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
|
---|
58 | 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
|
---|
59 | 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
|
---|
60 | 0x0284, 0x01a0, 0x00a0, 0x00a0, 0x00a0, 0x00a0, 0x00a0, 0x00a0,
|
---|
61 | 0x00a0, 0x00a0, 0x30e2, 0x01a0, 0x00a0, 0x00a8, 0x00a0, 0x00a0,
|
---|
62 | 0x00a0, 0x00a0, 0x10a0, 0x10a0, 0x00a0, 0x30e2, 0x00a0, 0x01a0,
|
---|
63 | 0x00a0, 0x10a0, 0x30e2, 0x01a0, 0x10a0, 0x10a0, 0x10a0, 0x01a0,
|
---|
64 | 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
|
---|
65 | 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
|
---|
66 | 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x00a0,
|
---|
67 | 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x30e2,
|
---|
68 | 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
|
---|
69 | 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
|
---|
70 | 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x00a0,
|
---|
71 | 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2
|
---|
72 | };
|
---|
73 |
|
---|
74 | typedef struct {
|
---|
75 | int n;
|
---|
76 | OnigCodePoint code[3];
|
---|
77 | } CodePointList3;
|
---|
78 |
|
---|
79 | typedef struct {
|
---|
80 | OnigCodePoint from;
|
---|
81 | CodePointList3 to;
|
---|
82 | } CaseFold_11_Type;
|
---|
83 |
|
---|
84 | typedef struct {
|
---|
85 | OnigCodePoint from;
|
---|
86 | CodePointList3 to;
|
---|
87 | } CaseUnfold_11_Type;
|
---|
88 |
|
---|
89 | typedef struct {
|
---|
90 | int n;
|
---|
91 | OnigCodePoint code[2];
|
---|
92 | } CodePointList2;
|
---|
93 |
|
---|
94 | typedef struct {
|
---|
95 | OnigCodePoint from[2];
|
---|
96 | CodePointList2 to;
|
---|
97 | } CaseUnfold_12_Type;
|
---|
98 |
|
---|
99 | typedef struct {
|
---|
100 | OnigCodePoint from[3];
|
---|
101 | CodePointList2 to;
|
---|
102 | } CaseUnfold_13_Type;
|
---|
103 |
|
---|
104 | static inline int
|
---|
105 | bits_of(const OnigCodePoint c, const int n)
|
---|
106 | {
|
---|
107 | return (c >> (2 - n) * 7) & 127;
|
---|
108 | }
|
---|
109 |
|
---|
110 | static inline int
|
---|
111 | bits_at(const OnigCodePoint *c, const int n)
|
---|
112 | {
|
---|
113 | return bits_of(c[n / 3], n % 3);
|
---|
114 | }
|
---|
115 |
|
---|
116 | static int
|
---|
117 | code1_equal(const OnigCodePoint x, const OnigCodePoint y)
|
---|
118 | {
|
---|
119 | if (x != y) return 0;
|
---|
120 | return 1;
|
---|
121 | }
|
---|
122 |
|
---|
123 | static int
|
---|
124 | code2_equal(const OnigCodePoint *x, const OnigCodePoint *y)
|
---|
125 | {
|
---|
126 | if (x[0] != y[0]) return 0;
|
---|
127 | if (x[1] != y[1]) return 0;
|
---|
128 | return 1;
|
---|
129 | }
|
---|
130 |
|
---|
131 | static int
|
---|
132 | code3_equal(const OnigCodePoint *x, const OnigCodePoint *y)
|
---|
133 | {
|
---|
134 | if (x[0] != y[0]) return 0;
|
---|
135 | if (x[1] != y[1]) return 0;
|
---|
136 | if (x[2] != y[2]) return 0;
|
---|
137 | return 1;
|
---|
138 | }
|
---|
139 |
|
---|
140 | /* macros related to ONIGENC_CASE flags */
|
---|
141 | /* defined here because not used in other files */
|
---|
142 | #define ONIGENC_CASE_SPECIALS (ONIGENC_CASE_TITLECASE | ONIGENC_CASE_IS_TITLECASE | ONIGENC_CASE_UP_SPECIAL | ONIGENC_CASE_DOWN_SPECIAL)
|
---|
143 |
|
---|
144 | /* macros for length in CaseMappingSpecials array in enc/unicode/casefold.h */
|
---|
145 | #define SpecialsLengthOffset 25 /* needs to be higher than the 22 bits used for Unicode codepoints */
|
---|
146 | #define SpecialsLengthExtract(n) ((n) >> SpecialsLengthOffset)
|
---|
147 | #define SpecialsCodepointExtract(n) ((n) & ((1 << SpecialsLengthOffset) - 1))
|
---|
148 | #define SpecialsLengthEncode(n) ((n) << SpecialsLengthOffset)
|
---|
149 |
|
---|
150 | #define OnigSpecialIndexMask (((1 << OnigSpecialIndexWidth) - 1) << OnigSpecialIndexShift)
|
---|
151 | #define OnigSpecialIndexEncode(n) ((n) << OnigSpecialIndexShift)
|
---|
152 | #define OnigSpecialIndexDecode(n) (((n) & OnigSpecialIndexMask) >> OnigSpecialIndexShift)
|
---|
153 |
|
---|
154 | /* macros to shorten "enc/unicode/casefold.h", undefined immediately after including the file */
|
---|
155 | #define U ONIGENC_CASE_UPCASE
|
---|
156 | #define D ONIGENC_CASE_DOWNCASE
|
---|
157 | #define F ONIGENC_CASE_FOLD
|
---|
158 | #define ST ONIGENC_CASE_TITLECASE
|
---|
159 | #define SU ONIGENC_CASE_UP_SPECIAL
|
---|
160 | #define SL ONIGENC_CASE_DOWN_SPECIAL
|
---|
161 | #define IT ONIGENC_CASE_IS_TITLECASE
|
---|
162 | #define I(n) OnigSpecialIndexEncode(n)
|
---|
163 | #define L(n) SpecialsLengthEncode(n)
|
---|
164 |
|
---|
165 | #include "casefold.h"
|
---|
166 |
|
---|
167 | #undef U
|
---|
168 | #undef D
|
---|
169 | #undef F
|
---|
170 | #undef ST
|
---|
171 | #undef SU
|
---|
172 | #undef SL
|
---|
173 | #undef IT
|
---|
174 | #undef I
|
---|
175 | #undef L
|
---|
176 |
|
---|
177 | #include "name2ctype.h"
|
---|
178 |
|
---|
179 | #define CODE_RANGES_NUM numberof(CodeRanges)
|
---|
180 |
|
---|
181 | extern int
|
---|
182 | onigenc_unicode_is_code_ctype(OnigCodePoint code, unsigned int ctype, OnigEncoding enc ARG_UNUSED)
|
---|
183 | {
|
---|
184 | if (
|
---|
185 | #ifdef USE_UNICODE_PROPERTIES
|
---|
186 | ctype <= ONIGENC_MAX_STD_CTYPE &&
|
---|
187 | #endif
|
---|
188 | code < 256) {
|
---|
189 | return ONIGENC_IS_UNICODE_ISO_8859_1_CTYPE(code, ctype);
|
---|
190 | }
|
---|
191 |
|
---|
192 | if (ctype >= CODE_RANGES_NUM) {
|
---|
193 | return ONIGERR_TYPE_BUG;
|
---|
194 | }
|
---|
195 |
|
---|
196 | return onig_is_in_code_range((UChar* )CodeRanges[ctype], code);
|
---|
197 | }
|
---|
198 |
|
---|
199 |
|
---|
200 | extern int
|
---|
201 | onigenc_unicode_ctype_code_range(int ctype, const OnigCodePoint* ranges[])
|
---|
202 | {
|
---|
203 | if (ctype >= CODE_RANGES_NUM) {
|
---|
204 | return ONIGERR_TYPE_BUG;
|
---|
205 | }
|
---|
206 |
|
---|
207 | *ranges = CodeRanges[ctype];
|
---|
208 |
|
---|
209 | return 0;
|
---|
210 | }
|
---|
211 |
|
---|
212 | extern int
|
---|
213 | onigenc_utf16_32_get_ctype_code_range(OnigCtype ctype, OnigCodePoint* sb_out,
|
---|
214 | const OnigCodePoint* ranges[],
|
---|
215 | OnigEncoding enc ARG_UNUSED)
|
---|
216 | {
|
---|
217 | *sb_out = 0x00;
|
---|
218 | return onigenc_unicode_ctype_code_range(ctype, ranges);
|
---|
219 | }
|
---|
220 |
|
---|
221 | #define PROPERTY_NAME_MAX_SIZE (MAX_WORD_LENGTH + 1)
|
---|
222 |
|
---|
223 | extern int
|
---|
224 | onigenc_unicode_property_name_to_ctype(OnigEncoding enc, const UChar* name, const UChar* end)
|
---|
225 | {
|
---|
226 | int len;
|
---|
227 | int ctype;
|
---|
228 | UChar buf[PROPERTY_NAME_MAX_SIZE];
|
---|
229 | const UChar *p;
|
---|
230 | OnigCodePoint code;
|
---|
231 |
|
---|
232 | len = 0;
|
---|
233 | for (p = name; p < end; p += enclen(enc, p, end)) {
|
---|
234 | code = ONIGENC_MBC_TO_CODE(enc, p, end);
|
---|
235 | if (code == ' ' || code == '-' || code == '_')
|
---|
236 | continue;
|
---|
237 | if (code >= 0x80)
|
---|
238 | return ONIGERR_INVALID_CHAR_PROPERTY_NAME;
|
---|
239 |
|
---|
240 | buf[len++] = ONIGENC_ASCII_CODE_TO_LOWER_CASE(code);
|
---|
241 | if (len >= PROPERTY_NAME_MAX_SIZE)
|
---|
242 | return ONIGERR_INVALID_CHAR_PROPERTY_NAME;
|
---|
243 | }
|
---|
244 |
|
---|
245 | buf[len] = 0;
|
---|
246 |
|
---|
247 | if ((ctype = uniname2ctype(buf, len)) < 0) {
|
---|
248 | return ONIGERR_INVALID_CHAR_PROPERTY_NAME;
|
---|
249 | }
|
---|
250 |
|
---|
251 | return ctype;
|
---|
252 | }
|
---|
253 |
|
---|
254 | #define onigenc_unicode_fold_lookup onigenc_unicode_CaseFold_11_lookup
|
---|
255 | #define onigenc_unicode_unfold1_lookup onigenc_unicode_CaseUnfold_11_lookup
|
---|
256 | #define onigenc_unicode_unfold2_lookup onigenc_unicode_CaseUnfold_12_lookup
|
---|
257 | #define onigenc_unicode_unfold3_lookup onigenc_unicode_CaseUnfold_13_lookup
|
---|
258 |
|
---|
259 | enum {
|
---|
260 | I_WITH_DOT_ABOVE = 0x0130,
|
---|
261 | DOTLESS_i = 0x0131,
|
---|
262 | DOT_ABOVE = 0x0307
|
---|
263 | };
|
---|
264 |
|
---|
265 | extern int
|
---|
266 | onigenc_unicode_mbc_case_fold(OnigEncoding enc,
|
---|
267 | OnigCaseFoldType flag ARG_UNUSED, const UChar** pp, const UChar* end,
|
---|
268 | UChar* fold)
|
---|
269 | {
|
---|
270 | const CodePointList3 *to;
|
---|
271 | OnigCodePoint code;
|
---|
272 | int i, len, rlen;
|
---|
273 | const UChar *p = *pp;
|
---|
274 |
|
---|
275 | code = ONIGENC_MBC_TO_CODE(enc, p, end);
|
---|
276 | len = enclen(enc, p, end);
|
---|
277 | *pp += len;
|
---|
278 |
|
---|
279 | #ifdef USE_UNICODE_CASE_FOLD_TURKISH_AZERI
|
---|
280 | if ((flag & ONIGENC_CASE_FOLD_TURKISH_AZERI) != 0) {
|
---|
281 | if (code == 'I') {
|
---|
282 | return ONIGENC_CODE_TO_MBC(enc, DOTLESS_i, fold);
|
---|
283 | }
|
---|
284 | else if (code == I_WITH_DOT_ABOVE) {
|
---|
285 | return ONIGENC_CODE_TO_MBC(enc, 'i', fold);
|
---|
286 | }
|
---|
287 | }
|
---|
288 | #endif
|
---|
289 |
|
---|
290 | if ((to = onigenc_unicode_fold_lookup(code)) != 0) {
|
---|
291 | if (OnigCodePointCount(to->n) == 1) {
|
---|
292 | return ONIGENC_CODE_TO_MBC(enc, to->code[0], fold);
|
---|
293 | }
|
---|
294 | #if 0
|
---|
295 | /* NO NEEDS TO CHECK */
|
---|
296 | else if ((flag & INTERNAL_ONIGENC_CASE_FOLD_MULTI_CHAR) != 0)
|
---|
297 | #else
|
---|
298 | else
|
---|
299 | #endif
|
---|
300 | {
|
---|
301 | rlen = 0;
|
---|
302 | for (i = 0; i < OnigCodePointCount(to->n); i++) {
|
---|
303 | len = ONIGENC_CODE_TO_MBC(enc, to->code[i], fold);
|
---|
304 | fold += len;
|
---|
305 | rlen += len;
|
---|
306 | }
|
---|
307 | return rlen;
|
---|
308 | }
|
---|
309 | }
|
---|
310 |
|
---|
311 | for (i = 0; i < len; i++) {
|
---|
312 | *fold++ = *p++;
|
---|
313 | }
|
---|
314 | return len;
|
---|
315 | }
|
---|
316 |
|
---|
317 | extern int
|
---|
318 | onigenc_unicode_apply_all_case_fold(OnigCaseFoldType flag,
|
---|
319 | OnigApplyAllCaseFoldFunc f, void* arg,
|
---|
320 | OnigEncoding enc ARG_UNUSED)
|
---|
321 | {
|
---|
322 | const CaseUnfold_11_Type* p11;
|
---|
323 | OnigCodePoint code;
|
---|
324 | int i, j, k, r;
|
---|
325 |
|
---|
326 | for (i = 0; i < numberof(CaseUnfold_11); i++) {
|
---|
327 | p11 = &CaseUnfold_11[i];
|
---|
328 | for (j = 0; j < OnigCodePointCount(p11->to.n); j++) {
|
---|
329 | code = p11->from;
|
---|
330 | r = (*f)(p11->to.code[j], &code, 1, arg);
|
---|
331 | if (r != 0) return r;
|
---|
332 |
|
---|
333 | code = p11->to.code[j];
|
---|
334 | r = (*f)(p11->from, &code, 1, arg);
|
---|
335 | if (r != 0) return r;
|
---|
336 |
|
---|
337 | for (k = 0; k < j; k++) {
|
---|
338 | r = (*f)(p11->to.code[j], (OnigCodePoint* )(&p11->to.code[k]), 1, arg);
|
---|
339 | if (r != 0) return r;
|
---|
340 |
|
---|
341 | r = (*f)(p11->to.code[k], (OnigCodePoint* )(&p11->to.code[j]), 1, arg);
|
---|
342 | if (r != 0) return r;
|
---|
343 | }
|
---|
344 | }
|
---|
345 | }
|
---|
346 |
|
---|
347 | #ifdef USE_UNICODE_CASE_FOLD_TURKISH_AZERI
|
---|
348 | if ((flag & ONIGENC_CASE_FOLD_TURKISH_AZERI) != 0) {
|
---|
349 | code = DOTLESS_i;
|
---|
350 | r = (*f)('I', &code, 1, arg);
|
---|
351 | if (r != 0) return r;
|
---|
352 | code = 'I';
|
---|
353 | r = (*f)(DOTLESS_i, &code, 1, arg);
|
---|
354 | if (r != 0) return r;
|
---|
355 |
|
---|
356 | code = I_WITH_DOT_ABOVE;
|
---|
357 | r = (*f)('i', &code, 1, arg);
|
---|
358 | if (r != 0) return r;
|
---|
359 | code = 'i';
|
---|
360 | r = (*f)(I_WITH_DOT_ABOVE, &code, 1, arg);
|
---|
361 | if (r != 0) return r;
|
---|
362 | }
|
---|
363 | else {
|
---|
364 | #endif
|
---|
365 | for (i = 0; i < numberof(CaseUnfold_11_Locale); i++) {
|
---|
366 | p11 = &CaseUnfold_11_Locale[i];
|
---|
367 | for (j = 0; j < OnigCodePointCount(p11->to.n); j++) {
|
---|
368 | code = p11->from;
|
---|
369 | r = (*f)(p11->to.code[j], &code, 1, arg);
|
---|
370 | if (r != 0) return r;
|
---|
371 |
|
---|
372 | code = p11->to.code[j];
|
---|
373 | r = (*f)(p11->from, &code, 1, arg);
|
---|
374 | if (r != 0) return r;
|
---|
375 |
|
---|
376 | for (k = 0; k < j; k++) {
|
---|
377 | r = (*f)(p11->to.code[j], (OnigCodePoint* )(&p11->to.code[k]),
|
---|
378 | 1, arg);
|
---|
379 | if (r != 0) return r;
|
---|
380 |
|
---|
381 | r = (*f)(p11->to.code[k], (OnigCodePoint* )(&p11->to.code[j]),
|
---|
382 | 1, arg);
|
---|
383 | if (r != 0) return r;
|
---|
384 | }
|
---|
385 | }
|
---|
386 | }
|
---|
387 | #ifdef USE_UNICODE_CASE_FOLD_TURKISH_AZERI
|
---|
388 | }
|
---|
389 | #endif
|
---|
390 |
|
---|
391 | if ((flag & INTERNAL_ONIGENC_CASE_FOLD_MULTI_CHAR) != 0) {
|
---|
392 | for (i = 0; i < numberof(CaseUnfold_12); i++) {
|
---|
393 | for (j = 0; j < OnigCodePointCount(CaseUnfold_12[i].to.n); j++) {
|
---|
394 | r = (*f)(CaseUnfold_12[i].to.code[j],
|
---|
395 | (OnigCodePoint* )CaseUnfold_12[i].from, 2, arg);
|
---|
396 | if (r != 0) return r;
|
---|
397 |
|
---|
398 | for (k = 0; k < OnigCodePointCount(CaseUnfold_12[i].to.n); k++) {
|
---|
399 | if (k == j) continue;
|
---|
400 |
|
---|
401 | r = (*f)(CaseUnfold_12[i].to.code[j],
|
---|
402 | (OnigCodePoint* )(&CaseUnfold_12[i].to.code[k]), 1, arg);
|
---|
403 | if (r != 0) return r;
|
---|
404 | }
|
---|
405 | }
|
---|
406 | }
|
---|
407 |
|
---|
408 | #ifdef USE_UNICODE_CASE_FOLD_TURKISH_AZERI
|
---|
409 | if ((flag & ONIGENC_CASE_FOLD_TURKISH_AZERI) == 0) {
|
---|
410 | #endif
|
---|
411 | for (i = 0; i < numberof(CaseUnfold_12_Locale); i++) {
|
---|
412 | for (j = 0; j < OnigCodePointCount(CaseUnfold_12_Locale[i].to.n); j++) {
|
---|
413 | r = (*f)(CaseUnfold_12_Locale[i].to.code[j],
|
---|
414 | (OnigCodePoint* )CaseUnfold_12_Locale[i].from, 2, arg);
|
---|
415 | if (r != 0) return r;
|
---|
416 |
|
---|
417 | for (k = 0; k < OnigCodePointCount(CaseUnfold_12_Locale[i].to.n); k++) {
|
---|
418 | if (k == j) continue;
|
---|
419 |
|
---|
420 | r = (*f)(CaseUnfold_12_Locale[i].to.code[j],
|
---|
421 | (OnigCodePoint* )(&CaseUnfold_12_Locale[i].to.code[k]),
|
---|
422 | 1, arg);
|
---|
423 | if (r != 0) return r;
|
---|
424 | }
|
---|
425 | }
|
---|
426 | }
|
---|
427 | #ifdef USE_UNICODE_CASE_FOLD_TURKISH_AZERI
|
---|
428 | }
|
---|
429 | #endif
|
---|
430 |
|
---|
431 | for (i = 0; i < numberof(CaseUnfold_13); i++) {
|
---|
432 | for (j = 0; j < OnigCodePointCount(CaseUnfold_13[i].to.n); j++) {
|
---|
433 | r = (*f)(CaseUnfold_13[i].to.code[j],
|
---|
434 | (OnigCodePoint* )CaseUnfold_13[i].from, 3, arg);
|
---|
435 | if (r != 0) return r;
|
---|
436 |
|
---|
437 | for (k = 0; k < OnigCodePointCount(CaseUnfold_13[i].to.n); k++) {
|
---|
438 | if (k == j) continue;
|
---|
439 |
|
---|
440 | r = (*f)(CaseUnfold_13[i].to.code[j],
|
---|
441 | (OnigCodePoint* )(&CaseUnfold_13[i].to.code[k]), 1, arg);
|
---|
442 | if (r != 0) return r;
|
---|
443 | }
|
---|
444 | }
|
---|
445 | }
|
---|
446 | }
|
---|
447 |
|
---|
448 | return 0;
|
---|
449 | }
|
---|
450 |
|
---|
451 | #define CodePointListValidP(x) (OnigCodePointCount((x)->n) <= numberof((x)->code))
|
---|
452 |
|
---|
453 | extern int
|
---|
454 | onigenc_unicode_get_case_fold_codes_by_str(OnigEncoding enc,
|
---|
455 | OnigCaseFoldType flag, const OnigUChar* p, const OnigUChar* end,
|
---|
456 | OnigCaseFoldCodeItem items[])
|
---|
457 | {
|
---|
458 | int n, i, j, k, len;
|
---|
459 | OnigCodePoint code, codes[3];
|
---|
460 | const CodePointList3 *to, *z3;
|
---|
461 | const CodePointList2 *z2;
|
---|
462 |
|
---|
463 | n = 0;
|
---|
464 |
|
---|
465 | code = ONIGENC_MBC_TO_CODE(enc, p, end);
|
---|
466 | len = enclen(enc, p, end);
|
---|
467 |
|
---|
468 | #ifdef USE_UNICODE_CASE_FOLD_TURKISH_AZERI
|
---|
469 | if ((flag & ONIGENC_CASE_FOLD_TURKISH_AZERI) != 0) {
|
---|
470 | switch (code) {
|
---|
471 | case 'I':
|
---|
472 | items[0].byte_len = len;
|
---|
473 | items[0].code_len = 1;
|
---|
474 | items[0].code[0] = DOTLESS_i;
|
---|
475 | return 1;
|
---|
476 | case I_WITH_DOT_ABOVE:
|
---|
477 | items[0].byte_len = len;
|
---|
478 | items[0].code_len = 1;
|
---|
479 | items[0].code[0] = 'i';
|
---|
480 | return 1;
|
---|
481 | case DOTLESS_i:
|
---|
482 | items[0].byte_len = len;
|
---|
483 | items[0].code_len = 1;
|
---|
484 | items[0].code[0] = 'I';
|
---|
485 | return 1;
|
---|
486 | case 'i':
|
---|
487 | items[0].byte_len = len;
|
---|
488 | items[0].code_len = 1;
|
---|
489 | items[0].code[0] = I_WITH_DOT_ABOVE;
|
---|
490 | return 1;
|
---|
491 | }
|
---|
492 | }
|
---|
493 | #endif
|
---|
494 |
|
---|
495 | if ((to = onigenc_unicode_fold_lookup(code)) != 0) {
|
---|
496 | if (OnigCodePointCount(to->n) == 1) {
|
---|
497 | OnigCodePoint orig_code = code;
|
---|
498 |
|
---|
499 | items[0].byte_len = len;
|
---|
500 | items[0].code_len = 1;
|
---|
501 | items[0].code[0] = to->code[0];
|
---|
502 | n++;
|
---|
503 |
|
---|
504 | code = to->code[0];
|
---|
505 | if ((to = onigenc_unicode_unfold1_lookup(code)) != 0 &&
|
---|
506 | CodePointListValidP(to)) {
|
---|
507 | for (i = 0; i < OnigCodePointCount(to->n); i++) {
|
---|
508 | if (to->code[i] != orig_code) {
|
---|
509 | items[n].byte_len = len;
|
---|
510 | items[n].code_len = 1;
|
---|
511 | items[n].code[0] = to->code[i];
|
---|
512 | n++;
|
---|
513 | }
|
---|
514 | }
|
---|
515 | }
|
---|
516 | }
|
---|
517 | else if ((flag & INTERNAL_ONIGENC_CASE_FOLD_MULTI_CHAR) != 0) {
|
---|
518 | OnigCodePoint cs[3][4];
|
---|
519 | int fn, ncs[3];
|
---|
520 |
|
---|
521 | for (fn = 0; fn < OnigCodePointCount(to->n); fn++) {
|
---|
522 | cs[fn][0] = to->code[fn];
|
---|
523 | if ((z3 = onigenc_unicode_unfold1_lookup(cs[fn][0])) != 0) {
|
---|
524 | for (i = 0; i < OnigCodePointCount(z3->n); i++) {
|
---|
525 | cs[fn][i+1] = z3->code[i];
|
---|
526 | }
|
---|
527 | ncs[fn] = OnigCodePointCount(z3->n) + 1;
|
---|
528 | }
|
---|
529 | else
|
---|
530 | ncs[fn] = 1;
|
---|
531 | }
|
---|
532 |
|
---|
533 | if (fn == 2) {
|
---|
534 | for (i = 0; i < ncs[0]; i++) {
|
---|
535 | for (j = 0; j < ncs[1]; j++) {
|
---|
536 | items[n].byte_len = len;
|
---|
537 | items[n].code_len = 2;
|
---|
538 | items[n].code[0] = cs[0][i];
|
---|
539 | items[n].code[1] = cs[1][j];
|
---|
540 | n++;
|
---|
541 | }
|
---|
542 | }
|
---|
543 |
|
---|
544 | if ((z2 = onigenc_unicode_unfold2_lookup(to->code)) != 0 &&
|
---|
545 | CodePointListValidP(z2)) {
|
---|
546 | for (i = 0; i < OnigCodePointCount(z2->n); i++) {
|
---|
547 | if (z2->code[i] == code) continue;
|
---|
548 |
|
---|
549 | items[n].byte_len = len;
|
---|
550 | items[n].code_len = 1;
|
---|
551 | items[n].code[0] = z2->code[i];
|
---|
552 | n++;
|
---|
553 | }
|
---|
554 | }
|
---|
555 | }
|
---|
556 | else {
|
---|
557 | for (i = 0; i < ncs[0]; i++) {
|
---|
558 | for (j = 0; j < ncs[1]; j++) {
|
---|
559 | for (k = 0; k < ncs[2]; k++) {
|
---|
560 | items[n].byte_len = len;
|
---|
561 | items[n].code_len = 3;
|
---|
562 | items[n].code[0] = cs[0][i];
|
---|
563 | items[n].code[1] = cs[1][j];
|
---|
564 | items[n].code[2] = cs[2][k];
|
---|
565 | n++;
|
---|
566 | }
|
---|
567 | }
|
---|
568 | }
|
---|
569 |
|
---|
570 | if ((z2 = onigenc_unicode_unfold3_lookup(to->code)) != 0 &&
|
---|
571 | CodePointListValidP(z2)) {
|
---|
572 | for (i = 0; i < OnigCodePointCount(z2->n); i++) {
|
---|
573 | if (z2->code[i] == code) continue;
|
---|
574 |
|
---|
575 | items[n].byte_len = len;
|
---|
576 | items[n].code_len = 1;
|
---|
577 | items[n].code[0] = z2->code[i];
|
---|
578 | n++;
|
---|
579 | }
|
---|
580 | }
|
---|
581 | }
|
---|
582 |
|
---|
583 | /* multi char folded code is not head of another folded multi char */
|
---|
584 | flag = 0; /* DISABLE_CASE_FOLD_MULTI_CHAR(flag); */
|
---|
585 | }
|
---|
586 | }
|
---|
587 | else {
|
---|
588 | if ((to = onigenc_unicode_unfold1_lookup(code)) != 0 &&
|
---|
589 | CodePointListValidP(to)) {
|
---|
590 | for (i = 0; i < OnigCodePointCount(to->n); i++) {
|
---|
591 | items[n].byte_len = len;
|
---|
592 | items[n].code_len = 1;
|
---|
593 | items[n].code[0] = to->code[i];
|
---|
594 | n++;
|
---|
595 | }
|
---|
596 | }
|
---|
597 | }
|
---|
598 |
|
---|
599 |
|
---|
600 | if ((flag & INTERNAL_ONIGENC_CASE_FOLD_MULTI_CHAR) != 0) {
|
---|
601 | p += len;
|
---|
602 | if (p < end) {
|
---|
603 | int clen;
|
---|
604 |
|
---|
605 | codes[0] = code;
|
---|
606 | code = ONIGENC_MBC_TO_CODE(enc, p, end);
|
---|
607 | if ((to = onigenc_unicode_fold_lookup(code)) != 0
|
---|
608 | && OnigCodePointCount(to->n) == 1) {
|
---|
609 | codes[1] = to->code[0];
|
---|
610 | }
|
---|
611 | else
|
---|
612 | codes[1] = code;
|
---|
613 |
|
---|
614 | clen = enclen(enc, p, end);
|
---|
615 | len += clen;
|
---|
616 | if ((z2 = onigenc_unicode_unfold2_lookup(codes)) != 0 &&
|
---|
617 | CodePointListValidP(z2)) {
|
---|
618 | for (i = 0; i < OnigCodePointCount(z2->n); i++) {
|
---|
619 | items[n].byte_len = len;
|
---|
620 | items[n].code_len = 1;
|
---|
621 | items[n].code[0] = z2->code[i];
|
---|
622 | n++;
|
---|
623 | }
|
---|
624 | }
|
---|
625 |
|
---|
626 | p += clen;
|
---|
627 | if (p < end) {
|
---|
628 | code = ONIGENC_MBC_TO_CODE(enc, p, end);
|
---|
629 | if ((to = onigenc_unicode_fold_lookup(code)) != 0
|
---|
630 | && OnigCodePointCount(to->n) == 1) {
|
---|
631 | codes[2] = to->code[0];
|
---|
632 | }
|
---|
633 | else
|
---|
634 | codes[2] = code;
|
---|
635 |
|
---|
636 | clen = enclen(enc, p, end);
|
---|
637 | len += clen;
|
---|
638 | if ((z2 = onigenc_unicode_unfold3_lookup(codes)) != 0 &&
|
---|
639 | CodePointListValidP(z2)) {
|
---|
640 | for (i = 0; i < OnigCodePointCount(z2->n); i++) {
|
---|
641 | items[n].byte_len = len;
|
---|
642 | items[n].code_len = 1;
|
---|
643 | items[n].code[0] = z2->code[i];
|
---|
644 | n++;
|
---|
645 | }
|
---|
646 | }
|
---|
647 | }
|
---|
648 | }
|
---|
649 | }
|
---|
650 |
|
---|
651 | return n;
|
---|
652 | }
|
---|
653 |
|
---|
654 | /* length in bytes for three characters in UTF-32; e.g. needed for ffi (U+FB03) */
|
---|
655 | #define CASE_MAPPING_SLACK 12
|
---|
656 | #define MODIFIED (flags |= ONIGENC_CASE_MODIFIED)
|
---|
657 | extern int
|
---|
658 | onigenc_unicode_case_map(OnigCaseFoldType* flagP,
|
---|
659 | const OnigUChar** pp, const OnigUChar* end,
|
---|
660 | OnigUChar* to, OnigUChar* to_end,
|
---|
661 | const struct OnigEncodingTypeST* enc)
|
---|
662 | {
|
---|
663 | OnigCodePoint code;
|
---|
664 | OnigUChar *to_start = to;
|
---|
665 | OnigCaseFoldType flags = *flagP;
|
---|
666 | int codepoint_length;
|
---|
667 |
|
---|
668 | to_end -= CASE_MAPPING_SLACK;
|
---|
669 | /* copy flags ONIGENC_CASE_UPCASE and ONIGENC_CASE_DOWNCASE over to
|
---|
670 | * ONIGENC_CASE_UP_SPECIAL and ONIGENC_CASE_DOWN_SPECIAL */
|
---|
671 | flags |= (flags & (ONIGENC_CASE_UPCASE | ONIGENC_CASE_DOWNCASE)) << ONIGENC_CASE_SPECIAL_OFFSET;
|
---|
672 |
|
---|
673 | while (*pp < end && to <= to_end) {
|
---|
674 | codepoint_length = ONIGENC_PRECISE_MBC_ENC_LEN(enc, *pp, end);
|
---|
675 | if (codepoint_length < 0)
|
---|
676 | return codepoint_length; /* encoding invalid */
|
---|
677 | code = ONIGENC_MBC_TO_CODE(enc, *pp, end);
|
---|
678 | *pp += codepoint_length;
|
---|
679 |
|
---|
680 | if (code <= 'z') { /* ASCII comes first */
|
---|
681 | if (code >= 'a' && code <= 'z') {
|
---|
682 | if (flags & ONIGENC_CASE_UPCASE) {
|
---|
683 | MODIFIED;
|
---|
684 | if (flags & ONIGENC_CASE_FOLD_TURKISH_AZERI && code == 'i')
|
---|
685 | code = I_WITH_DOT_ABOVE;
|
---|
686 | else
|
---|
687 | code += 'A' - 'a';
|
---|
688 | }
|
---|
689 | }
|
---|
690 | else if (code >= 'A' && code <= 'Z') {
|
---|
691 | if (flags & (ONIGENC_CASE_DOWNCASE | ONIGENC_CASE_FOLD)) {
|
---|
692 | MODIFIED;
|
---|
693 | if (flags & ONIGENC_CASE_FOLD_TURKISH_AZERI && code == 'I')
|
---|
694 | code = DOTLESS_i;
|
---|
695 | else
|
---|
696 | code += 'a' - 'A';
|
---|
697 | }
|
---|
698 | }
|
---|
699 | }
|
---|
700 | else if (!(flags & ONIGENC_CASE_ASCII_ONLY) && code >= 0x00B5) { /* deal with non-ASCII; micron sign (U+00B5) is lowest affected */
|
---|
701 | const CodePointList3 *folded;
|
---|
702 |
|
---|
703 | if (code == I_WITH_DOT_ABOVE) {
|
---|
704 | if (flags & (ONIGENC_CASE_DOWNCASE | ONIGENC_CASE_FOLD)) {
|
---|
705 | MODIFIED;
|
---|
706 | code = 'i';
|
---|
707 | if (!(flags & ONIGENC_CASE_FOLD_TURKISH_AZERI)) { /* make dot above explicit */
|
---|
708 | to += ONIGENC_CODE_TO_MBC(enc, code, to);
|
---|
709 | code = DOT_ABOVE;
|
---|
710 | }
|
---|
711 | }
|
---|
712 | }
|
---|
713 | else if (code == DOTLESS_i) { /* handle this manually, because it isn't involved in folding */
|
---|
714 | if (flags & ONIGENC_CASE_UPCASE) {
|
---|
715 | MODIFIED;
|
---|
716 | code = 'I';
|
---|
717 | }
|
---|
718 | }
|
---|
719 | else if ((folded = onigenc_unicode_fold_lookup(code)) != 0) { /* data about character found in CaseFold_11_Table */
|
---|
720 | if ((flags & ONIGENC_CASE_TITLECASE) /* Titlecase needed, */
|
---|
721 | && (OnigCaseFoldFlags(folded->n) & ONIGENC_CASE_IS_TITLECASE)) { /* but already Titlecase */
|
---|
722 | /* already Titlecase, no changes needed */
|
---|
723 | }
|
---|
724 | else if (flags & OnigCaseFoldFlags(folded->n)) { /* needs and data availability match */
|
---|
725 | const OnigCodePoint *next;
|
---|
726 | int count;
|
---|
727 |
|
---|
728 | MODIFIED;
|
---|
729 | if (flags & OnigCaseFoldFlags(folded->n) & ONIGENC_CASE_SPECIALS) { /* special */
|
---|
730 | const OnigCodePoint *SpecialsStart = CaseMappingSpecials + OnigSpecialIndexDecode(folded->n);
|
---|
731 |
|
---|
732 | if (OnigCaseFoldFlags(folded->n) & ONIGENC_CASE_IS_TITLECASE) { /* swapCASE available */
|
---|
733 | if ((flags & (ONIGENC_CASE_UPCASE | ONIGENC_CASE_DOWNCASE))
|
---|
734 | == (ONIGENC_CASE_UPCASE | ONIGENC_CASE_DOWNCASE)) /* swapCASE needed */
|
---|
735 | goto SpecialsCopy;
|
---|
736 | else /* swapCASE not needed */
|
---|
737 | SpecialsStart += SpecialsLengthExtract(*SpecialsStart);
|
---|
738 | }
|
---|
739 | if (OnigCaseFoldFlags(folded->n) & ONIGENC_CASE_TITLECASE) { /* Titlecase available */
|
---|
740 | if (flags & ONIGENC_CASE_TITLECASE) /* Titlecase needed, but not yet Titlecase */
|
---|
741 | goto SpecialsCopy;
|
---|
742 | else /* Titlecase not needed */
|
---|
743 | SpecialsStart += SpecialsLengthExtract(*SpecialsStart);
|
---|
744 | }
|
---|
745 | if (OnigCaseFoldFlags(folded->n) & ONIGENC_CASE_DOWN_SPECIAL) {
|
---|
746 | if (!(flags & ONIGENC_CASE_DOWN_SPECIAL))
|
---|
747 | SpecialsStart += SpecialsLengthExtract(*SpecialsStart);
|
---|
748 | }
|
---|
749 | /* here, we know we use ONIGENC_CASE_UP_SPECIAL, and the position is right */
|
---|
750 | SpecialsCopy:
|
---|
751 | count = SpecialsLengthExtract(*SpecialsStart);
|
---|
752 | next = SpecialsStart;
|
---|
753 | code = SpecialsCodepointExtract(*next++);
|
---|
754 | }
|
---|
755 | else { /* no specials */
|
---|
756 | count = OnigCodePointCount(folded->n);
|
---|
757 | next = folded->code;
|
---|
758 | code = *next++;
|
---|
759 | }
|
---|
760 | if (count == 1)
|
---|
761 | ;
|
---|
762 | else if (count == 2) {
|
---|
763 | to += ONIGENC_CODE_TO_MBC(enc, code, to);
|
---|
764 | code = *next;
|
---|
765 | }
|
---|
766 | else { /* count == 3 */
|
---|
767 | to += ONIGENC_CODE_TO_MBC(enc, code, to);
|
---|
768 | to += ONIGENC_CODE_TO_MBC(enc, *next++, to);
|
---|
769 | code = *next;
|
---|
770 | }
|
---|
771 | }
|
---|
772 | }
|
---|
773 | else if ((folded = onigenc_unicode_unfold1_lookup(code)) != 0 /* data about character found in CaseUnfold_11_Table */
|
---|
774 | && flags & OnigCaseFoldFlags(folded->n)) { /* needs and data availability match */
|
---|
775 | MODIFIED;
|
---|
776 | code = folded->code[(flags & OnigCaseFoldFlags(folded->n) & ONIGENC_CASE_TITLECASE) ? 1 : 0];
|
---|
777 | }
|
---|
778 | }
|
---|
779 | to += ONIGENC_CODE_TO_MBC(enc, code, to);
|
---|
780 | /* switch from titlecase to lowercase for capitalize */
|
---|
781 | if (flags & ONIGENC_CASE_TITLECASE)
|
---|
782 | flags ^= (ONIGENC_CASE_UPCASE | ONIGENC_CASE_DOWNCASE | ONIGENC_CASE_TITLECASE |
|
---|
783 | ONIGENC_CASE_UP_SPECIAL | ONIGENC_CASE_DOWN_SPECIAL);
|
---|
784 | }
|
---|
785 | *flagP = flags;
|
---|
786 | return (int )(to - to_start);
|
---|
787 | }
|
---|
788 |
|
---|
789 | #if 0
|
---|
790 | const char onigenc_unicode_version_string[] =
|
---|
791 | #ifdef ONIG_UNICODE_VERSION_STRING
|
---|
792 | ONIG_UNICODE_VERSION_STRING
|
---|
793 | #endif
|
---|
794 | "";
|
---|
795 |
|
---|
796 | const int onigenc_unicode_version_number[3] = {
|
---|
797 | #ifdef ONIG_UNICODE_VERSION_MAJOR
|
---|
798 | ONIG_UNICODE_VERSION_MAJOR,
|
---|
799 | ONIG_UNICODE_VERSION_MINOR,
|
---|
800 | ONIG_UNICODE_VERSION_TEENY,
|
---|
801 | #else
|
---|
802 | 0
|
---|
803 | #endif
|
---|
804 | };
|
---|
805 | #endif
|
---|