]>
Commit | Line | Data |
---|---|---|
374ca955 A |
1 | /* |
2 | ******************************************************************************* | |
57a6839d | 3 | * Copyright (C) 2004-2013, International Business Machines |
374ca955 A |
4 | * Corporation and others. All Rights Reserved. |
5 | ******************************************************************************* | |
57a6839d | 6 | * file name: uregex.cpp |
374ca955 A |
7 | */ |
8 | ||
9 | #include "unicode/utypes.h" | |
10 | ||
11 | #if !UCONFIG_NO_REGULAR_EXPRESSIONS | |
12 | ||
13 | #include "unicode/regex.h" | |
14 | #include "unicode/uregex.h" | |
15 | #include "unicode/unistr.h" | |
16 | #include "unicode/ustring.h" | |
17 | #include "unicode/uchar.h" | |
18 | #include "unicode/uobject.h" | |
4388f060 | 19 | #include "unicode/utf16.h" |
374ca955 A |
20 | #include "umutex.h" |
21 | #include "uassert.h" | |
22 | #include "cmemory.h" | |
23 | ||
729e4ab9 A |
24 | #include "regextxt.h" |
25 | ||
26 | #include <stdio.h> | |
27 | ||
28 | U_NAMESPACE_BEGIN | |
46f4442e | 29 | |
729e4ab9 A |
30 | #define REMAINING_CAPACITY(idx,len) ((((len)-(idx))>0)?((len)-(idx)):0) |
31 | ||
32 | struct RegularExpression: public UMemory { | |
374ca955 | 33 | public: |
729e4ab9 A |
34 | RegularExpression(); |
35 | ~RegularExpression(); | |
374ca955 A |
36 | int32_t fMagic; |
37 | RegexPattern *fPat; | |
57a6839d | 38 | u_atomic_int32_t *fPatRefCount; |
374ca955 A |
39 | UChar *fPatString; |
40 | int32_t fPatStringLen; | |
41 | RegexMatcher *fMatcher; | |
42 | const UChar *fText; // Text from setText() | |
43 | int32_t fTextLength; // Length provided by user with setText(), which | |
44 | // may be -1. | |
729e4ab9 | 45 | UBool fOwnsText; |
374ca955 A |
46 | }; |
47 | ||
48 | static const int32_t REXP_MAGIC = 0x72657870; // "rexp" in ASCII | |
49 | ||
729e4ab9 | 50 | RegularExpression::RegularExpression() { |
374ca955 A |
51 | fMagic = REXP_MAGIC; |
52 | fPat = NULL; | |
53 | fPatRefCount = NULL; | |
54 | fPatString = NULL; | |
55 | fPatStringLen = 0; | |
56 | fMatcher = NULL; | |
57 | fText = NULL; | |
58 | fTextLength = 0; | |
729e4ab9 | 59 | fOwnsText = FALSE; |
374ca955 A |
60 | } |
61 | ||
729e4ab9 | 62 | RegularExpression::~RegularExpression() { |
374ca955 A |
63 | delete fMatcher; |
64 | fMatcher = NULL; | |
65 | if (fPatRefCount!=NULL && umtx_atomic_dec(fPatRefCount)==0) { | |
66 | delete fPat; | |
67 | uprv_free(fPatString); | |
57a6839d | 68 | uprv_free((void *)fPatRefCount); |
374ca955 | 69 | } |
729e4ab9 A |
70 | if (fOwnsText && fText!=NULL) { |
71 | uprv_free((void *)fText); | |
72 | } | |
374ca955 A |
73 | fMagic = 0; |
74 | } | |
75 | ||
729e4ab9 A |
76 | U_NAMESPACE_END |
77 | ||
78 | U_NAMESPACE_USE | |
79 | ||
374ca955 A |
80 | //---------------------------------------------------------------------------------------- |
81 | // | |
82 | // validateRE Do boilerplate style checks on API function parameters. | |
83 | // Return TRUE if they look OK. | |
84 | //---------------------------------------------------------------------------------------- | |
4388f060 | 85 | static UBool validateRE(const RegularExpression *re, UBool requiresText, UErrorCode *status) { |
374ca955 A |
86 | if (U_FAILURE(*status)) { |
87 | return FALSE; | |
88 | } | |
89 | if (re == NULL || re->fMagic != REXP_MAGIC) { | |
374ca955 A |
90 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
91 | return FALSE; | |
92 | } | |
729e4ab9 A |
93 | // !!! Not sure how to update this with the new UText backing, which is stored in re->fMatcher anyway |
94 | if (requiresText && re->fText == NULL && !re->fOwnsText) { | |
374ca955 A |
95 | *status = U_REGEX_INVALID_STATE; |
96 | return FALSE; | |
97 | } | |
98 | return TRUE; | |
99 | } | |
100 | ||
101 | //---------------------------------------------------------------------------------------- | |
102 | // | |
103 | // uregex_open | |
104 | // | |
105 | //---------------------------------------------------------------------------------------- | |
106 | U_CAPI URegularExpression * U_EXPORT2 | |
107 | uregex_open( const UChar *pattern, | |
108 | int32_t patternLength, | |
109 | uint32_t flags, | |
110 | UParseError *pe, | |
111 | UErrorCode *status) { | |
112 | ||
113 | if (U_FAILURE(*status)) { | |
114 | return NULL; | |
115 | } | |
116 | if (pattern == NULL || patternLength < -1 || patternLength == 0) { | |
117 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
118 | return NULL; | |
119 | } | |
120 | int32_t actualPatLen = patternLength; | |
121 | if (actualPatLen == -1) { | |
122 | actualPatLen = u_strlen(pattern); | |
123 | } | |
124 | ||
57a6839d A |
125 | RegularExpression *re = new RegularExpression; |
126 | u_atomic_int32_t *refC = (u_atomic_int32_t *)uprv_malloc(sizeof(int32_t)); | |
374ca955 A |
127 | UChar *patBuf = (UChar *)uprv_malloc(sizeof(UChar)*(actualPatLen+1)); |
128 | if (re == NULL || refC == NULL || patBuf == NULL) { | |
129 | *status = U_MEMORY_ALLOCATION_ERROR; | |
130 | delete re; | |
57a6839d | 131 | uprv_free((void *)refC); |
374ca955 A |
132 | uprv_free(patBuf); |
133 | return NULL; | |
134 | } | |
135 | re->fPatRefCount = refC; | |
136 | *re->fPatRefCount = 1; | |
137 | ||
138 | // | |
139 | // Make a copy of the pattern string, so we can return it later if asked. | |
729e4ab9 A |
140 | // For compiling the pattern, we will use a UText wrapper around |
141 | // this local copy, to avoid making even more copies. | |
374ca955 A |
142 | // |
143 | re->fPatString = patBuf; | |
144 | re->fPatStringLen = patternLength; | |
145 | u_memcpy(patBuf, pattern, actualPatLen); | |
146 | patBuf[actualPatLen] = 0; | |
729e4ab9 A |
147 | |
148 | UText patText = UTEXT_INITIALIZER; | |
149 | utext_openUChars(&patText, patBuf, patternLength, status); | |
374ca955 A |
150 | |
151 | // | |
152 | // Compile the pattern | |
153 | // | |
154 | if (pe != NULL) { | |
729e4ab9 | 155 | re->fPat = RegexPattern::compile(&patText, flags, *pe, *status); |
374ca955 | 156 | } else { |
729e4ab9 | 157 | re->fPat = RegexPattern::compile(&patText, flags, *status); |
374ca955 | 158 | } |
729e4ab9 A |
159 | utext_close(&patText); |
160 | ||
374ca955 A |
161 | if (U_FAILURE(*status)) { |
162 | goto ErrorExit; | |
163 | } | |
164 | ||
165 | // | |
166 | // Create the matcher object | |
167 | // | |
168 | re->fMatcher = re->fPat->matcher(*status); | |
169 | if (U_SUCCESS(*status)) { | |
729e4ab9 | 170 | return (URegularExpression*)re; |
374ca955 A |
171 | } |
172 | ||
173 | ErrorExit: | |
174 | delete re; | |
175 | return NULL; | |
176 | ||
177 | } | |
178 | ||
729e4ab9 A |
179 | //---------------------------------------------------------------------------------------- |
180 | // | |
181 | // uregex_openUText | |
182 | // | |
183 | //---------------------------------------------------------------------------------------- | |
184 | U_CAPI URegularExpression * U_EXPORT2 | |
185 | uregex_openUText(UText *pattern, | |
186 | uint32_t flags, | |
187 | UParseError *pe, | |
188 | UErrorCode *status) { | |
189 | ||
190 | if (U_FAILURE(*status)) { | |
191 | return NULL; | |
192 | } | |
193 | if (pattern == NULL) { | |
194 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
195 | return NULL; | |
196 | } | |
197 | ||
198 | int64_t patternNativeLength = utext_nativeLength(pattern); | |
199 | ||
200 | if (patternNativeLength == 0) { | |
201 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
202 | return NULL; | |
203 | } | |
204 | ||
205 | RegularExpression *re = new RegularExpression; | |
206 | ||
207 | UErrorCode lengthStatus = U_ZERO_ERROR; | |
208 | int32_t pattern16Length = utext_extract(pattern, 0, patternNativeLength, NULL, 0, &lengthStatus); | |
209 | ||
57a6839d | 210 | u_atomic_int32_t *refC = (u_atomic_int32_t *)uprv_malloc(sizeof(int32_t)); |
729e4ab9 A |
211 | UChar *patBuf = (UChar *)uprv_malloc(sizeof(UChar)*(pattern16Length+1)); |
212 | if (re == NULL || refC == NULL || patBuf == NULL) { | |
213 | *status = U_MEMORY_ALLOCATION_ERROR; | |
214 | delete re; | |
57a6839d | 215 | uprv_free((void *)refC); |
729e4ab9 A |
216 | uprv_free(patBuf); |
217 | return NULL; | |
218 | } | |
219 | re->fPatRefCount = refC; | |
220 | *re->fPatRefCount = 1; | |
221 | ||
222 | // | |
223 | // Make a copy of the pattern string, so we can return it later if asked. | |
224 | // For compiling the pattern, we will use a read-only UText wrapper | |
225 | // around this local copy, to avoid making even more copies. | |
226 | // | |
227 | re->fPatString = patBuf; | |
228 | re->fPatStringLen = pattern16Length; | |
229 | utext_extract(pattern, 0, patternNativeLength, patBuf, pattern16Length+1, status); | |
230 | ||
231 | UText patText = UTEXT_INITIALIZER; | |
232 | utext_openUChars(&patText, patBuf, pattern16Length, status); | |
233 | ||
234 | // | |
235 | // Compile the pattern | |
236 | // | |
237 | if (pe != NULL) { | |
238 | re->fPat = RegexPattern::compile(&patText, flags, *pe, *status); | |
239 | } else { | |
240 | re->fPat = RegexPattern::compile(&patText, flags, *status); | |
241 | } | |
242 | utext_close(&patText); | |
243 | ||
244 | if (U_FAILURE(*status)) { | |
245 | goto ErrorExit; | |
246 | } | |
247 | ||
248 | // | |
249 | // Create the matcher object | |
250 | // | |
251 | re->fMatcher = re->fPat->matcher(*status); | |
252 | if (U_SUCCESS(*status)) { | |
253 | return (URegularExpression*)re; | |
254 | } | |
255 | ||
256 | ErrorExit: | |
257 | delete re; | |
258 | return NULL; | |
259 | ||
260 | } | |
261 | ||
374ca955 A |
262 | //---------------------------------------------------------------------------------------- |
263 | // | |
264 | // uregex_close | |
265 | // | |
266 | //---------------------------------------------------------------------------------------- | |
267 | U_CAPI void U_EXPORT2 | |
729e4ab9 A |
268 | uregex_close(URegularExpression *re2) { |
269 | RegularExpression *re = (RegularExpression*)re2; | |
374ca955 | 270 | UErrorCode status = U_ZERO_ERROR; |
4388f060 | 271 | if (validateRE(re, FALSE, &status) == FALSE) { |
374ca955 A |
272 | return; |
273 | } | |
274 | delete re; | |
275 | } | |
276 | ||
277 | ||
278 | //---------------------------------------------------------------------------------------- | |
279 | // | |
280 | // uregex_clone | |
281 | // | |
282 | //---------------------------------------------------------------------------------------- | |
283 | U_CAPI URegularExpression * U_EXPORT2 | |
729e4ab9 A |
284 | uregex_clone(const URegularExpression *source2, UErrorCode *status) { |
285 | RegularExpression *source = (RegularExpression*)source2; | |
4388f060 | 286 | if (validateRE(source, FALSE, status) == FALSE) { |
374ca955 A |
287 | return NULL; |
288 | } | |
289 | ||
729e4ab9 | 290 | RegularExpression *clone = new RegularExpression; |
374ca955 A |
291 | if (clone == NULL) { |
292 | *status = U_MEMORY_ALLOCATION_ERROR; | |
293 | return NULL; | |
294 | } | |
295 | ||
296 | clone->fMatcher = source->fPat->matcher(*status); | |
297 | if (U_FAILURE(*status)) { | |
298 | delete clone; | |
299 | return NULL; | |
300 | } | |
374ca955 A |
301 | |
302 | clone->fPat = source->fPat; | |
303 | clone->fPatRefCount = source->fPatRefCount; | |
304 | clone->fPatString = source->fPatString; | |
305 | clone->fPatStringLen = source->fPatStringLen; | |
306 | umtx_atomic_inc(source->fPatRefCount); | |
307 | // Note: fText is not cloned. | |
308 | ||
729e4ab9 | 309 | return (URegularExpression*)clone; |
73c04bcf | 310 | } |
374ca955 A |
311 | |
312 | ||
313 | ||
314 | ||
73c04bcf | 315 | //------------------------------------------------------------------------------ |
374ca955 A |
316 | // |
317 | // uregex_pattern | |
318 | // | |
73c04bcf | 319 | //------------------------------------------------------------------------------ |
374ca955 | 320 | U_CAPI const UChar * U_EXPORT2 |
729e4ab9 A |
321 | uregex_pattern(const URegularExpression *regexp2, |
322 | int32_t *patLength, | |
323 | UErrorCode *status) { | |
324 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
374ca955 | 325 | |
4388f060 | 326 | if (validateRE(regexp, FALSE, status) == FALSE) { |
374ca955 A |
327 | return NULL; |
328 | } | |
329 | if (patLength != NULL) { | |
330 | *patLength = regexp->fPatStringLen; | |
331 | } | |
332 | return regexp->fPatString; | |
73c04bcf | 333 | } |
374ca955 A |
334 | |
335 | ||
729e4ab9 A |
336 | //------------------------------------------------------------------------------ |
337 | // | |
338 | // uregex_patternUText | |
339 | // | |
340 | //------------------------------------------------------------------------------ | |
341 | U_CAPI UText * U_EXPORT2 | |
342 | uregex_patternUText(const URegularExpression *regexp2, | |
343 | UErrorCode *status) { | |
344 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
345 | return regexp->fPat->patternText(*status); | |
346 | } | |
347 | ||
348 | ||
73c04bcf | 349 | //------------------------------------------------------------------------------ |
374ca955 A |
350 | // |
351 | // uregex_flags | |
352 | // | |
73c04bcf | 353 | //------------------------------------------------------------------------------ |
374ca955 | 354 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 A |
355 | uregex_flags(const URegularExpression *regexp2, UErrorCode *status) { |
356 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 357 | if (validateRE(regexp, FALSE, status) == FALSE) { |
374ca955 A |
358 | return 0; |
359 | } | |
360 | int32_t flags = regexp->fPat->flags(); | |
361 | return flags; | |
73c04bcf | 362 | } |
374ca955 A |
363 | |
364 | ||
73c04bcf | 365 | //------------------------------------------------------------------------------ |
374ca955 A |
366 | // |
367 | // uregex_setText | |
368 | // | |
73c04bcf | 369 | //------------------------------------------------------------------------------ |
374ca955 | 370 | U_CAPI void U_EXPORT2 |
729e4ab9 | 371 | uregex_setText(URegularExpression *regexp2, |
374ca955 A |
372 | const UChar *text, |
373 | int32_t textLength, | |
374 | UErrorCode *status) { | |
729e4ab9 | 375 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 376 | if (validateRE(regexp, FALSE, status) == FALSE) { |
374ca955 A |
377 | return; |
378 | } | |
379 | if (text == NULL || textLength < -1) { | |
380 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
381 | return; | |
382 | } | |
729e4ab9 A |
383 | |
384 | if (regexp->fOwnsText && regexp->fText != NULL) { | |
385 | uprv_free((void *)regexp->fText); | |
386 | } | |
387 | ||
374ca955 A |
388 | regexp->fText = text; |
389 | regexp->fTextLength = textLength; | |
729e4ab9 A |
390 | regexp->fOwnsText = FALSE; |
391 | ||
392 | UText input = UTEXT_INITIALIZER; | |
393 | utext_openUChars(&input, text, textLength, status); | |
394 | regexp->fMatcher->reset(&input); | |
395 | utext_close(&input); // reset() made a shallow clone, so we don't need this copy | |
396 | } | |
397 | ||
374ca955 | 398 | |
729e4ab9 A |
399 | //------------------------------------------------------------------------------ |
400 | // | |
401 | // uregex_setUText | |
402 | // | |
403 | //------------------------------------------------------------------------------ | |
404 | U_CAPI void U_EXPORT2 | |
405 | uregex_setUText(URegularExpression *regexp2, | |
406 | UText *text, | |
407 | UErrorCode *status) { | |
408 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 409 | if (validateRE(regexp, FALSE, status) == FALSE) { |
729e4ab9 A |
410 | return; |
411 | } | |
412 | if (text == NULL) { | |
413 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
414 | return; | |
415 | } | |
416 | ||
417 | if (regexp->fOwnsText && regexp->fText != NULL) { | |
418 | uprv_free((void *)regexp->fText); | |
419 | } | |
420 | ||
421 | regexp->fText = NULL; // only fill it in on request | |
422 | regexp->fTextLength = -1; | |
423 | regexp->fOwnsText = TRUE; | |
424 | regexp->fMatcher->reset(text); | |
73c04bcf | 425 | } |
374ca955 A |
426 | |
427 | ||
428 | ||
73c04bcf | 429 | //------------------------------------------------------------------------------ |
374ca955 A |
430 | // |
431 | // uregex_getText | |
432 | // | |
73c04bcf | 433 | //------------------------------------------------------------------------------ |
374ca955 | 434 | U_CAPI const UChar * U_EXPORT2 |
729e4ab9 | 435 | uregex_getText(URegularExpression *regexp2, |
374ca955 A |
436 | int32_t *textLength, |
437 | UErrorCode *status) { | |
729e4ab9 | 438 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 439 | if (validateRE(regexp, FALSE, status) == FALSE) { |
374ca955 A |
440 | return NULL; |
441 | } | |
729e4ab9 A |
442 | |
443 | if (regexp->fText == NULL) { | |
444 | // need to fill in the text | |
445 | UText *inputText = regexp->fMatcher->inputText(); | |
446 | int64_t inputNativeLength = utext_nativeLength(inputText); | |
447 | if (UTEXT_FULL_TEXT_IN_CHUNK(inputText, inputNativeLength)) { | |
448 | regexp->fText = inputText->chunkContents; | |
449 | regexp->fTextLength = (int32_t)inputNativeLength; | |
450 | regexp->fOwnsText = FALSE; // because the UText owns it | |
451 | } else { | |
452 | UErrorCode lengthStatus = U_ZERO_ERROR; | |
453 | regexp->fTextLength = utext_extract(inputText, 0, inputNativeLength, NULL, 0, &lengthStatus); // buffer overflow error | |
454 | UChar *inputChars = (UChar *)uprv_malloc(sizeof(UChar)*(regexp->fTextLength+1)); | |
455 | ||
456 | utext_extract(inputText, 0, inputNativeLength, inputChars, regexp->fTextLength+1, status); | |
457 | regexp->fText = inputChars; | |
458 | regexp->fOwnsText = TRUE; // should already be set but just in case | |
459 | } | |
460 | } | |
461 | ||
374ca955 A |
462 | if (textLength != NULL) { |
463 | *textLength = regexp->fTextLength; | |
464 | } | |
465 | return regexp->fText; | |
73c04bcf | 466 | } |
374ca955 A |
467 | |
468 | ||
729e4ab9 A |
469 | //------------------------------------------------------------------------------ |
470 | // | |
471 | // uregex_getUText | |
472 | // | |
473 | //------------------------------------------------------------------------------ | |
474 | U_CAPI UText * U_EXPORT2 | |
475 | uregex_getUText(URegularExpression *regexp2, | |
476 | UText *dest, | |
477 | UErrorCode *status) { | |
478 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 479 | if (validateRE(regexp, FALSE, status) == FALSE) { |
729e4ab9 A |
480 | return dest; |
481 | } | |
482 | return regexp->fMatcher->getInput(dest, *status); | |
483 | } | |
484 | ||
485 | ||
4388f060 A |
486 | //------------------------------------------------------------------------------ |
487 | // | |
488 | // uregex_refreshUText | |
489 | // | |
490 | //------------------------------------------------------------------------------ | |
491 | U_CAPI void U_EXPORT2 | |
492 | uregex_refreshUText(URegularExpression *regexp2, | |
493 | UText *text, | |
494 | UErrorCode *status) { | |
495 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
496 | if (validateRE(regexp, FALSE, status) == FALSE) { | |
497 | return; | |
498 | } | |
499 | regexp->fMatcher->refreshInputText(text, *status); | |
500 | } | |
501 | ||
502 | ||
73c04bcf | 503 | //------------------------------------------------------------------------------ |
374ca955 A |
504 | // |
505 | // uregex_matches | |
506 | // | |
73c04bcf | 507 | //------------------------------------------------------------------------------ |
374ca955 | 508 | U_CAPI UBool U_EXPORT2 |
729e4ab9 A |
509 | uregex_matches(URegularExpression *regexp2, |
510 | int32_t startIndex, | |
511 | UErrorCode *status) { | |
512 | return uregex_matches64( regexp2, (int64_t)startIndex, status); | |
513 | } | |
514 | ||
515 | U_CAPI UBool U_EXPORT2 | |
516 | uregex_matches64(URegularExpression *regexp2, | |
517 | int64_t startIndex, | |
518 | UErrorCode *status) { | |
519 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
46f4442e | 520 | UBool result = FALSE; |
4388f060 | 521 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
522 | return result; |
523 | } | |
524 | if (startIndex == -1) { | |
525 | result = regexp->fMatcher->matches(*status); | |
526 | } else { | |
527 | result = regexp->fMatcher->matches(startIndex, *status); | |
374ca955 | 528 | } |
374ca955 | 529 | return result; |
73c04bcf | 530 | } |
374ca955 A |
531 | |
532 | ||
73c04bcf | 533 | //------------------------------------------------------------------------------ |
374ca955 A |
534 | // |
535 | // uregex_lookingAt | |
536 | // | |
73c04bcf | 537 | //------------------------------------------------------------------------------ |
374ca955 | 538 | U_CAPI UBool U_EXPORT2 |
729e4ab9 | 539 | uregex_lookingAt(URegularExpression *regexp2, |
374ca955 A |
540 | int32_t startIndex, |
541 | UErrorCode *status) { | |
729e4ab9 A |
542 | return uregex_lookingAt64( regexp2, (int64_t)startIndex, status); |
543 | } | |
544 | ||
545 | U_CAPI UBool U_EXPORT2 | |
546 | uregex_lookingAt64(URegularExpression *regexp2, | |
547 | int64_t startIndex, | |
548 | UErrorCode *status) { | |
549 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
46f4442e | 550 | UBool result = FALSE; |
4388f060 | 551 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
552 | return result; |
553 | } | |
554 | if (startIndex == -1) { | |
555 | result = regexp->fMatcher->lookingAt(*status); | |
556 | } else { | |
557 | result = regexp->fMatcher->lookingAt(startIndex, *status); | |
374ca955 | 558 | } |
374ca955 | 559 | return result; |
73c04bcf | 560 | } |
374ca955 A |
561 | |
562 | ||
563 | ||
73c04bcf | 564 | //------------------------------------------------------------------------------ |
374ca955 A |
565 | // |
566 | // uregex_find | |
567 | // | |
73c04bcf | 568 | //------------------------------------------------------------------------------ |
374ca955 | 569 | U_CAPI UBool U_EXPORT2 |
729e4ab9 | 570 | uregex_find(URegularExpression *regexp2, |
374ca955 A |
571 | int32_t startIndex, |
572 | UErrorCode *status) { | |
729e4ab9 A |
573 | return uregex_find64( regexp2, (int64_t)startIndex, status); |
574 | } | |
575 | ||
576 | U_CAPI UBool U_EXPORT2 | |
577 | uregex_find64(URegularExpression *regexp2, | |
578 | int64_t startIndex, | |
579 | UErrorCode *status) { | |
580 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
46f4442e | 581 | UBool result = FALSE; |
4388f060 | 582 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
583 | return result; |
584 | } | |
585 | if (startIndex == -1) { | |
586 | regexp->fMatcher->resetPreserveRegion(); | |
587 | result = regexp->fMatcher->find(); | |
588 | } else { | |
589 | result = regexp->fMatcher->find(startIndex, *status); | |
374ca955 | 590 | } |
374ca955 | 591 | return result; |
73c04bcf | 592 | } |
374ca955 | 593 | |
729e4ab9 | 594 | |
73c04bcf | 595 | //------------------------------------------------------------------------------ |
374ca955 A |
596 | // |
597 | // uregex_findNext | |
598 | // | |
73c04bcf | 599 | //------------------------------------------------------------------------------ |
374ca955 | 600 | U_CAPI UBool U_EXPORT2 |
729e4ab9 | 601 | uregex_findNext(URegularExpression *regexp2, |
374ca955 | 602 | UErrorCode *status) { |
729e4ab9 | 603 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 604 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
605 | return FALSE; |
606 | } | |
607 | UBool result = regexp->fMatcher->find(); | |
608 | return result; | |
73c04bcf | 609 | } |
374ca955 | 610 | |
73c04bcf | 611 | //------------------------------------------------------------------------------ |
374ca955 A |
612 | // |
613 | // uregex_groupCount | |
614 | // | |
73c04bcf | 615 | //------------------------------------------------------------------------------ |
374ca955 | 616 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 | 617 | uregex_groupCount(URegularExpression *regexp2, |
374ca955 | 618 | UErrorCode *status) { |
729e4ab9 | 619 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 620 | if (validateRE(regexp, FALSE, status) == FALSE) { |
374ca955 A |
621 | return 0; |
622 | } | |
623 | int32_t result = regexp->fMatcher->groupCount(); | |
624 | return result; | |
73c04bcf | 625 | } |
374ca955 A |
626 | |
627 | ||
73c04bcf | 628 | //------------------------------------------------------------------------------ |
374ca955 A |
629 | // |
630 | // uregex_group | |
631 | // | |
73c04bcf | 632 | //------------------------------------------------------------------------------ |
374ca955 | 633 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 | 634 | uregex_group(URegularExpression *regexp2, |
374ca955 A |
635 | int32_t groupNum, |
636 | UChar *dest, | |
637 | int32_t destCapacity, | |
638 | UErrorCode *status) { | |
729e4ab9 | 639 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 640 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
641 | return 0; |
642 | } | |
643 | if (destCapacity < 0 || (destCapacity > 0 && dest == NULL)) { | |
644 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
645 | return 0; | |
646 | } | |
729e4ab9 A |
647 | |
648 | if (destCapacity == 0 || regexp->fText != NULL) { | |
649 | // If preflighting or if we already have the text as UChars, | |
650 | // this is a little cheaper than going through uregex_groupUTextDeep() | |
651 | ||
652 | // | |
653 | // Pick up the range of characters from the matcher | |
654 | // | |
655 | int32_t startIx = regexp->fMatcher->start(groupNum, *status); | |
656 | int32_t endIx = regexp->fMatcher->end (groupNum, *status); | |
657 | if (U_FAILURE(*status)) { | |
658 | return 0; | |
659 | } | |
374ca955 | 660 | |
729e4ab9 A |
661 | // |
662 | // Trim length based on buffer capacity | |
663 | // | |
664 | int32_t fullLength = endIx - startIx; | |
665 | int32_t copyLength = fullLength; | |
666 | if (copyLength < destCapacity) { | |
667 | dest[copyLength] = 0; | |
668 | } else if (copyLength == destCapacity) { | |
669 | *status = U_STRING_NOT_TERMINATED_WARNING; | |
670 | } else { | |
671 | copyLength = destCapacity; | |
672 | *status = U_BUFFER_OVERFLOW_ERROR; | |
673 | } | |
674 | ||
675 | // | |
676 | // Copy capture group to user's buffer | |
677 | // | |
678 | if (copyLength > 0) { | |
679 | u_memcpy(dest, ®exp->fText[startIx], copyLength); | |
680 | } | |
681 | return fullLength; | |
374ca955 | 682 | } else { |
57a6839d | 683 | int32_t result = 0; |
729e4ab9 | 684 | UText *groupText = uregex_groupUTextDeep(regexp2, groupNum, NULL, status); |
57a6839d A |
685 | if (U_SUCCESS(*status)) { |
686 | result = utext_extract(groupText, 0, utext_nativeLength(groupText), dest, destCapacity, status); | |
687 | } | |
729e4ab9 A |
688 | utext_close(groupText); |
689 | return result; | |
374ca955 | 690 | } |
729e4ab9 | 691 | } |
374ca955 | 692 | |
729e4ab9 A |
693 | |
694 | //------------------------------------------------------------------------------ | |
695 | // | |
696 | // uregex_groupUText | |
697 | // | |
698 | //------------------------------------------------------------------------------ | |
699 | U_CAPI UText * U_EXPORT2 | |
700 | uregex_groupUText(URegularExpression *regexp2, | |
701 | int32_t groupNum, | |
702 | UText *dest, | |
703 | int64_t *groupLength, | |
704 | UErrorCode *status) { | |
705 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 706 | if (validateRE(regexp, TRUE, status) == FALSE) { |
729e4ab9 A |
707 | UErrorCode emptyTextStatus = U_ZERO_ERROR; |
708 | return (dest ? dest : utext_openUChars(NULL, NULL, 0, &emptyTextStatus)); | |
374ca955 | 709 | } |
729e4ab9 A |
710 | |
711 | return regexp->fMatcher->group(groupNum, dest, *groupLength, *status); | |
73c04bcf | 712 | } |
374ca955 | 713 | |
729e4ab9 A |
714 | //------------------------------------------------------------------------------ |
715 | // | |
716 | // uregex_groupUTextDeep | |
717 | // | |
718 | //------------------------------------------------------------------------------ | |
719 | U_CAPI UText * U_EXPORT2 | |
720 | uregex_groupUTextDeep(URegularExpression *regexp2, | |
721 | int32_t groupNum, | |
722 | UText *dest, | |
723 | UErrorCode *status) { | |
724 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 725 | if (validateRE(regexp, TRUE, status) == FALSE) { |
729e4ab9 A |
726 | UErrorCode emptyTextStatus = U_ZERO_ERROR; |
727 | return (dest ? dest : utext_openUChars(NULL, NULL, 0, &emptyTextStatus)); | |
728 | } | |
729 | ||
730 | if (regexp->fText != NULL) { | |
731 | // | |
732 | // Pick up the range of characters from the matcher | |
733 | // and use our already-extracted characters | |
734 | // | |
735 | int32_t startIx = regexp->fMatcher->start(groupNum, *status); | |
736 | int32_t endIx = regexp->fMatcher->end (groupNum, *status); | |
737 | if (U_FAILURE(*status)) { | |
738 | UErrorCode emptyTextStatus = U_ZERO_ERROR; | |
739 | return (dest ? dest : utext_openUChars(NULL, NULL, 0, &emptyTextStatus)); | |
740 | } | |
741 | ||
742 | if (dest) { | |
743 | utext_replace(dest, 0, utext_nativeLength(dest), ®exp->fText[startIx], endIx - startIx, status); | |
744 | } else { | |
745 | UText groupText = UTEXT_INITIALIZER; | |
746 | utext_openUChars(&groupText, ®exp->fText[startIx], endIx - startIx, status); | |
747 | dest = utext_clone(NULL, &groupText, TRUE, FALSE, status); | |
748 | utext_close(&groupText); | |
749 | } | |
750 | ||
751 | return dest; | |
752 | } else { | |
753 | return regexp->fMatcher->group(groupNum, dest, *status); | |
754 | } | |
755 | } | |
374ca955 | 756 | |
73c04bcf | 757 | //------------------------------------------------------------------------------ |
374ca955 A |
758 | // |
759 | // uregex_start | |
760 | // | |
73c04bcf | 761 | //------------------------------------------------------------------------------ |
374ca955 | 762 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 | 763 | uregex_start(URegularExpression *regexp2, |
374ca955 A |
764 | int32_t groupNum, |
765 | UErrorCode *status) { | |
729e4ab9 A |
766 | return (int32_t)uregex_start64( regexp2, groupNum, status); |
767 | } | |
768 | ||
769 | U_CAPI int64_t U_EXPORT2 | |
770 | uregex_start64(URegularExpression *regexp2, | |
771 | int32_t groupNum, | |
772 | UErrorCode *status) { | |
773 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 774 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
775 | return 0; |
776 | } | |
777 | int32_t result = regexp->fMatcher->start(groupNum, *status); | |
778 | return result; | |
73c04bcf | 779 | } |
374ca955 | 780 | |
73c04bcf | 781 | //------------------------------------------------------------------------------ |
374ca955 A |
782 | // |
783 | // uregex_end | |
784 | // | |
73c04bcf | 785 | //------------------------------------------------------------------------------ |
374ca955 | 786 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 | 787 | uregex_end(URegularExpression *regexp2, |
374ca955 A |
788 | int32_t groupNum, |
789 | UErrorCode *status) { | |
729e4ab9 A |
790 | return (int32_t)uregex_end64( regexp2, groupNum, status); |
791 | } | |
792 | ||
793 | U_CAPI int64_t U_EXPORT2 | |
794 | uregex_end64(URegularExpression *regexp2, | |
795 | int32_t groupNum, | |
796 | UErrorCode *status) { | |
797 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 798 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
799 | return 0; |
800 | } | |
801 | int32_t result = regexp->fMatcher->end(groupNum, *status); | |
802 | return result; | |
73c04bcf | 803 | } |
374ca955 | 804 | |
73c04bcf | 805 | //------------------------------------------------------------------------------ |
374ca955 A |
806 | // |
807 | // uregex_reset | |
808 | // | |
73c04bcf | 809 | //------------------------------------------------------------------------------ |
374ca955 | 810 | U_CAPI void U_EXPORT2 |
729e4ab9 | 811 | uregex_reset(URegularExpression *regexp2, |
374ca955 A |
812 | int32_t index, |
813 | UErrorCode *status) { | |
729e4ab9 A |
814 | uregex_reset64( regexp2, (int64_t)index, status); |
815 | } | |
816 | ||
817 | U_CAPI void U_EXPORT2 | |
818 | uregex_reset64(URegularExpression *regexp2, | |
819 | int64_t index, | |
820 | UErrorCode *status) { | |
821 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 822 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
823 | return; |
824 | } | |
825 | regexp->fMatcher->reset(index, *status); | |
73c04bcf | 826 | } |
374ca955 A |
827 | |
828 | ||
46f4442e A |
829 | //------------------------------------------------------------------------------ |
830 | // | |
831 | // uregex_setRegion | |
832 | // | |
833 | //------------------------------------------------------------------------------ | |
834 | U_CAPI void U_EXPORT2 | |
729e4ab9 | 835 | uregex_setRegion(URegularExpression *regexp2, |
46f4442e A |
836 | int32_t regionStart, |
837 | int32_t regionLimit, | |
838 | UErrorCode *status) { | |
729e4ab9 A |
839 | uregex_setRegion64( regexp2, (int64_t)regionStart, (int64_t)regionLimit, status); |
840 | } | |
841 | ||
842 | U_CAPI void U_EXPORT2 | |
843 | uregex_setRegion64(URegularExpression *regexp2, | |
844 | int64_t regionStart, | |
845 | int64_t regionLimit, | |
846 | UErrorCode *status) { | |
847 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 848 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
849 | return; |
850 | } | |
851 | regexp->fMatcher->region(regionStart, regionLimit, *status); | |
852 | } | |
853 | ||
854 | ||
729e4ab9 A |
855 | //------------------------------------------------------------------------------ |
856 | // | |
857 | // uregex_setRegionAndStart | |
858 | // | |
859 | //------------------------------------------------------------------------------ | |
51004dcb | 860 | U_CAPI void U_EXPORT2 |
729e4ab9 A |
861 | uregex_setRegionAndStart(URegularExpression *regexp2, |
862 | int64_t regionStart, | |
863 | int64_t regionLimit, | |
864 | int64_t startIndex, | |
865 | UErrorCode *status) { | |
866 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 867 | if (validateRE(regexp, TRUE, status) == FALSE) { |
729e4ab9 A |
868 | return; |
869 | } | |
870 | regexp->fMatcher->region(regionStart, regionLimit, startIndex, *status); | |
871 | } | |
872 | ||
46f4442e A |
873 | //------------------------------------------------------------------------------ |
874 | // | |
875 | // uregex_regionStart | |
876 | // | |
877 | //------------------------------------------------------------------------------ | |
878 | U_CAPI int32_t U_EXPORT2 | |
729e4ab9 | 879 | uregex_regionStart(const URegularExpression *regexp2, |
46f4442e | 880 | UErrorCode *status) { |
729e4ab9 A |
881 | return (int32_t)uregex_regionStart64(regexp2, status); |
882 | } | |
883 | ||
884 | U_CAPI int64_t U_EXPORT2 | |
885 | uregex_regionStart64(const URegularExpression *regexp2, | |
886 | UErrorCode *status) { | |
887 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 888 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
889 | return 0; |
890 | } | |
891 | return regexp->fMatcher->regionStart(); | |
892 | } | |
893 | ||
894 | ||
895 | //------------------------------------------------------------------------------ | |
896 | // | |
897 | // uregex_regionEnd | |
898 | // | |
899 | //------------------------------------------------------------------------------ | |
900 | U_CAPI int32_t U_EXPORT2 | |
729e4ab9 | 901 | uregex_regionEnd(const URegularExpression *regexp2, |
46f4442e | 902 | UErrorCode *status) { |
729e4ab9 A |
903 | return (int32_t)uregex_regionEnd64(regexp2, status); |
904 | } | |
905 | ||
906 | U_CAPI int64_t U_EXPORT2 | |
907 | uregex_regionEnd64(const URegularExpression *regexp2, | |
908 | UErrorCode *status) { | |
909 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 910 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
911 | return 0; |
912 | } | |
913 | return regexp->fMatcher->regionEnd(); | |
914 | } | |
915 | ||
916 | ||
917 | //------------------------------------------------------------------------------ | |
918 | // | |
919 | // uregex_hasTransparentBounds | |
920 | // | |
921 | //------------------------------------------------------------------------------ | |
922 | U_CAPI UBool U_EXPORT2 | |
729e4ab9 | 923 | uregex_hasTransparentBounds(const URegularExpression *regexp2, |
46f4442e | 924 | UErrorCode *status) { |
729e4ab9 | 925 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 926 | if (validateRE(regexp, FALSE, status) == FALSE) { |
46f4442e A |
927 | return FALSE; |
928 | } | |
929 | return regexp->fMatcher->hasTransparentBounds(); | |
930 | } | |
931 | ||
932 | ||
933 | //------------------------------------------------------------------------------ | |
934 | // | |
935 | // uregex_useTransparentBounds | |
936 | // | |
937 | //------------------------------------------------------------------------------ | |
938 | U_CAPI void U_EXPORT2 | |
729e4ab9 A |
939 | uregex_useTransparentBounds(URegularExpression *regexp2, |
940 | UBool b, | |
941 | UErrorCode *status) { | |
942 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 943 | if (validateRE(regexp, FALSE, status) == FALSE) { |
46f4442e A |
944 | return; |
945 | } | |
946 | regexp->fMatcher->useTransparentBounds(b); | |
947 | } | |
948 | ||
949 | ||
950 | //------------------------------------------------------------------------------ | |
951 | // | |
952 | // uregex_hasAnchoringBounds | |
953 | // | |
954 | //------------------------------------------------------------------------------ | |
955 | U_CAPI UBool U_EXPORT2 | |
729e4ab9 A |
956 | uregex_hasAnchoringBounds(const URegularExpression *regexp2, |
957 | UErrorCode *status) { | |
958 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 959 | if (validateRE(regexp, FALSE, status) == FALSE) { |
46f4442e A |
960 | return FALSE; |
961 | } | |
962 | return regexp->fMatcher->hasAnchoringBounds(); | |
963 | } | |
964 | ||
965 | ||
966 | //------------------------------------------------------------------------------ | |
967 | // | |
968 | // uregex_useAnchoringBounds | |
969 | // | |
970 | //------------------------------------------------------------------------------ | |
971 | U_CAPI void U_EXPORT2 | |
729e4ab9 A |
972 | uregex_useAnchoringBounds(URegularExpression *regexp2, |
973 | UBool b, | |
974 | UErrorCode *status) { | |
975 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 976 | if (validateRE(regexp, FALSE, status) == FALSE) { |
46f4442e A |
977 | return; |
978 | } | |
979 | regexp->fMatcher->useAnchoringBounds(b); | |
980 | } | |
981 | ||
982 | ||
983 | //------------------------------------------------------------------------------ | |
984 | // | |
985 | // uregex_hitEnd | |
986 | // | |
987 | //------------------------------------------------------------------------------ | |
988 | U_CAPI UBool U_EXPORT2 | |
729e4ab9 | 989 | uregex_hitEnd(const URegularExpression *regexp2, |
46f4442e | 990 | UErrorCode *status) { |
729e4ab9 | 991 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 992 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
993 | return FALSE; |
994 | } | |
995 | return regexp->fMatcher->hitEnd(); | |
996 | } | |
997 | ||
998 | ||
999 | //------------------------------------------------------------------------------ | |
1000 | // | |
1001 | // uregex_requireEnd | |
1002 | // | |
1003 | //------------------------------------------------------------------------------ | |
1004 | U_CAPI UBool U_EXPORT2 | |
729e4ab9 | 1005 | uregex_requireEnd(const URegularExpression *regexp2, |
46f4442e | 1006 | UErrorCode *status) { |
729e4ab9 | 1007 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1008 | if (validateRE(regexp, TRUE, status) == FALSE) { |
46f4442e A |
1009 | return FALSE; |
1010 | } | |
1011 | return regexp->fMatcher->requireEnd(); | |
1012 | } | |
1013 | ||
1014 | ||
1015 | //------------------------------------------------------------------------------ | |
1016 | // | |
1017 | // uregex_setTimeLimit | |
1018 | // | |
1019 | //------------------------------------------------------------------------------ | |
1020 | U_CAPI void U_EXPORT2 | |
729e4ab9 | 1021 | uregex_setTimeLimit(URegularExpression *regexp2, |
46f4442e A |
1022 | int32_t limit, |
1023 | UErrorCode *status) { | |
729e4ab9 | 1024 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1025 | if (validateRE(regexp, FALSE, status)) { |
46f4442e A |
1026 | regexp->fMatcher->setTimeLimit(limit, *status); |
1027 | } | |
1028 | } | |
1029 | ||
1030 | ||
1031 | ||
1032 | //------------------------------------------------------------------------------ | |
1033 | // | |
1034 | // uregex_getTimeLimit | |
1035 | // | |
1036 | //------------------------------------------------------------------------------ | |
1037 | U_CAPI int32_t U_EXPORT2 | |
729e4ab9 | 1038 | uregex_getTimeLimit(const URegularExpression *regexp2, |
46f4442e A |
1039 | UErrorCode *status) { |
1040 | int32_t retVal = 0; | |
729e4ab9 | 1041 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1042 | if (validateRE(regexp, FALSE, status)) { |
46f4442e A |
1043 | retVal = regexp->fMatcher->getTimeLimit(); |
1044 | } | |
1045 | return retVal; | |
1046 | } | |
1047 | ||
1048 | ||
1049 | ||
1050 | //------------------------------------------------------------------------------ | |
1051 | // | |
1052 | // uregex_setStackLimit | |
1053 | // | |
1054 | //------------------------------------------------------------------------------ | |
1055 | U_CAPI void U_EXPORT2 | |
729e4ab9 A |
1056 | uregex_setStackLimit(URegularExpression *regexp2, |
1057 | int32_t limit, | |
1058 | UErrorCode *status) { | |
1059 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 1060 | if (validateRE(regexp, FALSE, status)) { |
46f4442e A |
1061 | regexp->fMatcher->setStackLimit(limit, *status); |
1062 | } | |
1063 | } | |
1064 | ||
1065 | ||
1066 | ||
1067 | //------------------------------------------------------------------------------ | |
1068 | // | |
1069 | // uregex_getStackLimit | |
1070 | // | |
1071 | //------------------------------------------------------------------------------ | |
1072 | U_CAPI int32_t U_EXPORT2 | |
729e4ab9 A |
1073 | uregex_getStackLimit(const URegularExpression *regexp2, |
1074 | UErrorCode *status) { | |
46f4442e | 1075 | int32_t retVal = 0; |
729e4ab9 | 1076 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1077 | if (validateRE(regexp, FALSE, status)) { |
46f4442e A |
1078 | retVal = regexp->fMatcher->getStackLimit(); |
1079 | } | |
1080 | return retVal; | |
1081 | } | |
1082 | ||
1083 | ||
1084 | //------------------------------------------------------------------------------ | |
1085 | // | |
1086 | // uregex_setMatchCallback | |
1087 | // | |
1088 | //------------------------------------------------------------------------------ | |
1089 | U_CAPI void U_EXPORT2 | |
729e4ab9 | 1090 | uregex_setMatchCallback(URegularExpression *regexp2, |
46f4442e A |
1091 | URegexMatchCallback *callback, |
1092 | const void *context, | |
1093 | UErrorCode *status) { | |
729e4ab9 | 1094 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1095 | if (validateRE(regexp, FALSE, status)) { |
729e4ab9 | 1096 | regexp->fMatcher->setMatchCallback(callback, context, *status); |
46f4442e A |
1097 | } |
1098 | } | |
1099 | ||
1100 | ||
1101 | //------------------------------------------------------------------------------ | |
1102 | // | |
1103 | // uregex_getMatchCallback | |
1104 | // | |
1105 | //------------------------------------------------------------------------------ | |
1106 | U_CAPI void U_EXPORT2 | |
729e4ab9 | 1107 | uregex_getMatchCallback(const URegularExpression *regexp2, |
46f4442e A |
1108 | URegexMatchCallback **callback, |
1109 | const void **context, | |
1110 | UErrorCode *status) { | |
729e4ab9 | 1111 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1112 | if (validateRE(regexp, FALSE, status)) { |
46f4442e A |
1113 | regexp->fMatcher->getMatchCallback(*callback, *context, *status); |
1114 | } | |
1115 | } | |
1116 | ||
1117 | ||
729e4ab9 A |
1118 | //------------------------------------------------------------------------------ |
1119 | // | |
1120 | // uregex_setMatchProgressCallback | |
1121 | // | |
1122 | //------------------------------------------------------------------------------ | |
1123 | U_CAPI void U_EXPORT2 | |
1124 | uregex_setFindProgressCallback(URegularExpression *regexp2, | |
1125 | URegexFindProgressCallback *callback, | |
1126 | const void *context, | |
1127 | UErrorCode *status) { | |
1128 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 1129 | if (validateRE(regexp, FALSE, status)) { |
729e4ab9 A |
1130 | regexp->fMatcher->setFindProgressCallback(callback, context, *status); |
1131 | } | |
1132 | } | |
1133 | ||
1134 | ||
1135 | //------------------------------------------------------------------------------ | |
1136 | // | |
1137 | // uregex_getMatchCallback | |
1138 | // | |
1139 | //------------------------------------------------------------------------------ | |
1140 | U_CAPI void U_EXPORT2 | |
1141 | uregex_getFindProgressCallback(const URegularExpression *regexp2, | |
1142 | URegexFindProgressCallback **callback, | |
1143 | const void **context, | |
1144 | UErrorCode *status) { | |
1145 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 1146 | if (validateRE(regexp, FALSE, status)) { |
729e4ab9 A |
1147 | regexp->fMatcher->getFindProgressCallback(*callback, *context, *status); |
1148 | } | |
1149 | } | |
1150 | ||
1151 | ||
73c04bcf | 1152 | //------------------------------------------------------------------------------ |
374ca955 A |
1153 | // |
1154 | // uregex_replaceAll | |
1155 | // | |
73c04bcf | 1156 | //------------------------------------------------------------------------------ |
374ca955 | 1157 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 | 1158 | uregex_replaceAll(URegularExpression *regexp2, |
73c04bcf | 1159 | const UChar *replacementText, |
374ca955 A |
1160 | int32_t replacementLength, |
1161 | UChar *destBuf, | |
1162 | int32_t destCapacity, | |
1163 | UErrorCode *status) { | |
729e4ab9 | 1164 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1165 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
1166 | return 0; |
1167 | } | |
1168 | if (replacementText == NULL || replacementLength < -1 || | |
729e4ab9 | 1169 | (destBuf == NULL && destCapacity > 0) || |
374ca955 A |
1170 | destCapacity < 0) { |
1171 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
1172 | return 0; | |
1173 | } | |
1174 | ||
1175 | int32_t len = 0; | |
729e4ab9 A |
1176 | |
1177 | uregex_reset(regexp2, 0, status); | |
1178 | ||
1179 | // Note: Seperate error code variables for findNext() and appendReplacement() | |
1180 | // are used so that destination buffer overflow errors | |
1181 | // in appendReplacement won't stop findNext() from working. | |
1182 | // appendReplacement() and appendTail() special case incoming buffer | |
1183 | // overflow errors, continuing to return the correct length. | |
1184 | UErrorCode findStatus = *status; | |
1185 | while (uregex_findNext(regexp2, &findStatus)) { | |
1186 | len += uregex_appendReplacement(regexp2, replacementText, replacementLength, | |
374ca955 A |
1187 | &destBuf, &destCapacity, status); |
1188 | } | |
729e4ab9 A |
1189 | len += uregex_appendTail(regexp2, &destBuf, &destCapacity, status); |
1190 | ||
1191 | if (U_FAILURE(findStatus)) { | |
1192 | // If anything went wrong with the findNext(), make that error trump | |
1193 | // whatever may have happened with the append() operations. | |
1194 | // Errors in findNext() are not expected. | |
1195 | *status = findStatus; | |
1196 | } | |
374ca955 A |
1197 | |
1198 | return len; | |
73c04bcf | 1199 | } |
374ca955 A |
1200 | |
1201 | ||
729e4ab9 A |
1202 | //------------------------------------------------------------------------------ |
1203 | // | |
1204 | // uregex_replaceAllUText | |
1205 | // | |
1206 | //------------------------------------------------------------------------------ | |
1207 | U_CAPI UText * U_EXPORT2 | |
1208 | uregex_replaceAllUText(URegularExpression *regexp2, | |
1209 | UText *replacementText, | |
1210 | UText *dest, | |
1211 | UErrorCode *status) { | |
1212 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 1213 | if (validateRE(regexp, TRUE, status) == FALSE) { |
729e4ab9 A |
1214 | return 0; |
1215 | } | |
1216 | if (replacementText == NULL) { | |
1217 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
1218 | return 0; | |
1219 | } | |
1220 | ||
1221 | dest = regexp->fMatcher->replaceAll(replacementText, dest, *status); | |
1222 | return dest; | |
1223 | } | |
1224 | ||
1225 | ||
73c04bcf | 1226 | //------------------------------------------------------------------------------ |
374ca955 A |
1227 | // |
1228 | // uregex_replaceFirst | |
1229 | // | |
73c04bcf | 1230 | //------------------------------------------------------------------------------ |
374ca955 | 1231 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 | 1232 | uregex_replaceFirst(URegularExpression *regexp2, |
73c04bcf | 1233 | const UChar *replacementText, |
374ca955 A |
1234 | int32_t replacementLength, |
1235 | UChar *destBuf, | |
1236 | int32_t destCapacity, | |
1237 | UErrorCode *status) { | |
729e4ab9 | 1238 | RegularExpression *regexp = (RegularExpression*)regexp2; |
4388f060 | 1239 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
1240 | return 0; |
1241 | } | |
1242 | if (replacementText == NULL || replacementLength < -1 || | |
729e4ab9 | 1243 | (destBuf == NULL && destCapacity > 0) || |
374ca955 A |
1244 | destCapacity < 0) { |
1245 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
1246 | return 0; | |
1247 | } | |
1248 | ||
1249 | int32_t len = 0; | |
1250 | UBool findSucceeded; | |
729e4ab9 A |
1251 | uregex_reset(regexp2, 0, status); |
1252 | findSucceeded = uregex_find(regexp2, 0, status); | |
374ca955 | 1253 | if (findSucceeded) { |
729e4ab9 | 1254 | len = uregex_appendReplacement(regexp2, replacementText, replacementLength, |
374ca955 A |
1255 | &destBuf, &destCapacity, status); |
1256 | } | |
729e4ab9 | 1257 | len += uregex_appendTail(regexp2, &destBuf, &destCapacity, status); |
374ca955 A |
1258 | |
1259 | return len; | |
73c04bcf | 1260 | } |
374ca955 A |
1261 | |
1262 | ||
73c04bcf | 1263 | //------------------------------------------------------------------------------ |
374ca955 | 1264 | // |
729e4ab9 | 1265 | // uregex_replaceFirstUText |
374ca955 | 1266 | // |
73c04bcf | 1267 | //------------------------------------------------------------------------------ |
729e4ab9 A |
1268 | U_CAPI UText * U_EXPORT2 |
1269 | uregex_replaceFirstUText(URegularExpression *regexp2, | |
1270 | UText *replacementText, | |
1271 | UText *dest, | |
1272 | UErrorCode *status) { | |
1273 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 1274 | if (validateRE(regexp, TRUE, status) == FALSE) { |
729e4ab9 A |
1275 | return 0; |
1276 | } | |
1277 | if (replacementText == NULL) { | |
1278 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
1279 | return 0; | |
1280 | } | |
1281 | ||
1282 | dest = regexp->fMatcher->replaceFirst(replacementText, dest, *status); | |
1283 | return dest; | |
1284 | } | |
1285 | ||
374ca955 | 1286 | |
729e4ab9 A |
1287 | //------------------------------------------------------------------------------ |
1288 | // | |
1289 | // uregex_appendReplacement | |
1290 | // | |
1291 | //------------------------------------------------------------------------------ | |
374ca955 | 1292 | |
729e4ab9 | 1293 | U_NAMESPACE_BEGIN |
374ca955 A |
1294 | // |
1295 | // Dummy class, because these functions need to be friends of class RegexMatcher, | |
1296 | // and stand-alone C functions don't work as friends | |
1297 | // | |
374ca955 A |
1298 | class RegexCImpl { |
1299 | public: | |
729e4ab9 | 1300 | inline static int32_t appendReplacement(RegularExpression *regexp, |
73c04bcf | 1301 | const UChar *replacementText, |
374ca955 A |
1302 | int32_t replacementLength, |
1303 | UChar **destBuf, | |
1304 | int32_t *destCapacity, | |
1305 | UErrorCode *status); | |
1306 | ||
729e4ab9 A |
1307 | inline static int32_t appendTail(RegularExpression *regexp, |
1308 | UChar **destBuf, | |
1309 | int32_t *destCapacity, | |
1310 | UErrorCode *status); | |
1311 | ||
1312 | inline static int32_t split(RegularExpression *regexp, | |
1313 | UChar *destBuf, | |
1314 | int32_t destCapacity, | |
1315 | int32_t *requiredCapacity, | |
1316 | UChar *destFields[], | |
1317 | int32_t destFieldsCapacity, | |
1318 | UErrorCode *status); | |
374ca955 | 1319 | }; |
374ca955 | 1320 | |
729e4ab9 | 1321 | U_NAMESPACE_END |
374ca955 | 1322 | |
374ca955 A |
1323 | |
1324 | ||
1325 | static const UChar BACKSLASH = 0x5c; | |
1326 | static const UChar DOLLARSIGN = 0x24; | |
1327 | ||
1328 | // | |
1329 | // Move a character to an output buffer, with bounds checking on the index. | |
1330 | // Index advances even if capacity is exceeded, for preflight size computations. | |
1331 | // This little sequence is used a LOT. | |
1332 | // | |
1333 | static inline void appendToBuf(UChar c, int32_t *idx, UChar *buf, int32_t bufCapacity) { | |
1334 | if (*idx < bufCapacity) { | |
1335 | buf[*idx] = c; | |
1336 | } | |
1337 | (*idx)++; | |
1338 | } | |
1339 | ||
1340 | ||
1341 | // | |
1342 | // appendReplacement, the actual implementation. | |
1343 | // | |
729e4ab9 A |
1344 | int32_t RegexCImpl::appendReplacement(RegularExpression *regexp, |
1345 | const UChar *replacementText, | |
1346 | int32_t replacementLength, | |
1347 | UChar **destBuf, | |
1348 | int32_t *destCapacity, | |
1349 | UErrorCode *status) { | |
374ca955 A |
1350 | |
1351 | // If we come in with a buffer overflow error, don't suppress the operation. | |
1352 | // A series of appendReplacements, appendTail need to correctly preflight | |
1353 | // the buffer size when an overflow happens somewhere in the middle. | |
1354 | UBool pendingBufferOverflow = FALSE; | |
729e4ab9 | 1355 | if (*status == U_BUFFER_OVERFLOW_ERROR && destCapacity != NULL && *destCapacity == 0) { |
374ca955 A |
1356 | pendingBufferOverflow = TRUE; |
1357 | *status = U_ZERO_ERROR; | |
1358 | } | |
1359 | ||
1360 | // | |
1361 | // Validate all paramters | |
1362 | // | |
4388f060 | 1363 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
1364 | return 0; |
1365 | } | |
1366 | if (replacementText == NULL || replacementLength < -1 || | |
1367 | destCapacity == NULL || destBuf == NULL || | |
729e4ab9 | 1368 | (*destBuf == NULL && *destCapacity > 0) || |
374ca955 A |
1369 | *destCapacity < 0) { |
1370 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
1371 | return 0; | |
1372 | } | |
1373 | ||
1374 | RegexMatcher *m = regexp->fMatcher; | |
1375 | if (m->fMatch == FALSE) { | |
1376 | *status = U_REGEX_INVALID_STATE; | |
1377 | return 0; | |
1378 | } | |
1379 | ||
1380 | UChar *dest = *destBuf; | |
1381 | int32_t capacity = *destCapacity; | |
1382 | int32_t destIdx = 0; | |
1383 | int32_t i; | |
1384 | ||
1385 | // If it wasn't supplied by the caller, get the length of the replacement text. | |
1386 | // TODO: slightly smarter logic in the copy loop could watch for the NUL on | |
1387 | // the fly and avoid this step. | |
1388 | if (replacementLength == -1) { | |
1389 | replacementLength = u_strlen(replacementText); | |
1390 | } | |
1391 | ||
1392 | // Copy input string from the end of previous match to start of current match | |
729e4ab9 A |
1393 | if (regexp->fText != NULL) { |
1394 | int32_t matchStart; | |
1395 | int32_t lastMatchEnd; | |
1396 | if (UTEXT_USES_U16(m->fInputText)) { | |
1397 | lastMatchEnd = (int32_t)m->fLastMatchEnd; | |
1398 | matchStart = (int32_t)m->fMatchStart; | |
1399 | } else { | |
1400 | // !!!: Would like a better way to do this! | |
1401 | UErrorCode status = U_ZERO_ERROR; | |
1402 | lastMatchEnd = utext_extract(m->fInputText, 0, m->fLastMatchEnd, NULL, 0, &status); | |
1403 | status = U_ZERO_ERROR; | |
1404 | matchStart = lastMatchEnd + utext_extract(m->fInputText, m->fLastMatchEnd, m->fMatchStart, NULL, 0, &status); | |
1405 | } | |
1406 | for (i=lastMatchEnd; i<matchStart; i++) { | |
1407 | appendToBuf(regexp->fText[i], &destIdx, dest, capacity); | |
1408 | } | |
1409 | } else { | |
1410 | UErrorCode possibleOverflowError = U_ZERO_ERROR; // ignore | |
1411 | destIdx += utext_extract(m->fInputText, m->fLastMatchEnd, m->fMatchStart, | |
4388f060 A |
1412 | dest==NULL?NULL:&dest[destIdx], REMAINING_CAPACITY(destIdx, capacity), |
1413 | &possibleOverflowError); | |
374ca955 | 1414 | } |
4388f060 | 1415 | U_ASSERT(destIdx >= 0); |
374ca955 A |
1416 | |
1417 | // scan the replacement text, looking for substitutions ($n) and \escapes. | |
1418 | int32_t replIdx = 0; | |
1419 | while (replIdx < replacementLength) { | |
1420 | UChar c = replacementText[replIdx]; | |
1421 | replIdx++; | |
1422 | if (c != DOLLARSIGN && c != BACKSLASH) { | |
1423 | // Common case, no substitution, no escaping, | |
1424 | // just copy the char to the dest buf. | |
1425 | appendToBuf(c, &destIdx, dest, capacity); | |
1426 | continue; | |
1427 | } | |
1428 | ||
1429 | if (c == BACKSLASH) { | |
1430 | // Backslash Escape. Copy the following char out without further checks. | |
1431 | // Note: Surrogate pairs don't need any special handling | |
1432 | // The second half wont be a '$' or a '\', and | |
1433 | // will move to the dest normally on the next | |
1434 | // loop iteration. | |
1435 | if (replIdx >= replacementLength) { | |
1436 | break; | |
1437 | } | |
1438 | c = replacementText[replIdx]; | |
1439 | ||
1440 | if (c==0x55/*U*/ || c==0x75/*u*/) { | |
1441 | // We have a \udddd or \Udddddddd escape sequence. | |
1442 | UChar32 escapedChar = | |
729e4ab9 | 1443 | u_unescapeAt(uregex_ucstr_unescape_charAt, |
374ca955 A |
1444 | &replIdx, // Index is updated by unescapeAt |
1445 | replacementLength, // Length of replacement text | |
73c04bcf | 1446 | (void *)replacementText); |
374ca955 A |
1447 | |
1448 | if (escapedChar != (UChar32)0xFFFFFFFF) { | |
1449 | if (escapedChar <= 0xffff) { | |
1450 | appendToBuf((UChar)escapedChar, &destIdx, dest, capacity); | |
1451 | } else { | |
1452 | appendToBuf(U16_LEAD(escapedChar), &destIdx, dest, capacity); | |
1453 | appendToBuf(U16_TRAIL(escapedChar), &destIdx, dest, capacity); | |
1454 | } | |
1455 | continue; | |
1456 | } | |
1457 | // Note: if the \u escape was invalid, just fall through and | |
1458 | // treat it as a plain \<anything> escape. | |
1459 | } | |
1460 | ||
1461 | // Plain backslash escape. Just put out the escaped character. | |
1462 | appendToBuf(c, &destIdx, dest, capacity); | |
1463 | ||
1464 | replIdx++; | |
1465 | continue; | |
1466 | } | |
1467 | ||
1468 | ||
1469 | ||
1470 | // We've got a $. Pick up a capture group number if one follows. | |
1471 | // Consume at most the number of digits necessary for the largest capture | |
1472 | // number that is valid for this pattern. | |
1473 | ||
1474 | int32_t numDigits = 0; | |
1475 | int32_t groupNum = 0; | |
1476 | UChar32 digitC; | |
1477 | for (;;) { | |
1478 | if (replIdx >= replacementLength) { | |
1479 | break; | |
1480 | } | |
1481 | U16_GET(replacementText, 0, replIdx, replacementLength, digitC); | |
1482 | if (u_isdigit(digitC) == FALSE) { | |
1483 | break; | |
1484 | } | |
1485 | ||
1486 | U16_FWD_1(replacementText, replIdx, replacementLength); | |
1487 | groupNum=groupNum*10 + u_charDigitValue(digitC); | |
1488 | numDigits++; | |
1489 | if (numDigits >= m->fPattern->fMaxCaptureDigits) { | |
1490 | break; | |
1491 | } | |
1492 | } | |
1493 | ||
1494 | ||
1495 | if (numDigits == 0) { | |
1496 | // The $ didn't introduce a group number at all. | |
1497 | // Treat it as just part of the substitution text. | |
1498 | appendToBuf(DOLLARSIGN, &destIdx, dest, capacity); | |
1499 | continue; | |
1500 | } | |
1501 | ||
1502 | // Finally, append the capture group data to the destination. | |
4388f060 A |
1503 | destIdx += uregex_group((URegularExpression*)regexp, groupNum, |
1504 | dest==NULL?NULL:&dest[destIdx], REMAINING_CAPACITY(destIdx, capacity), status); | |
374ca955 A |
1505 | if (*status == U_BUFFER_OVERFLOW_ERROR) { |
1506 | // Ignore buffer overflow when extracting the group. We need to | |
1507 | // continue on to get full size of the untruncated result. We will | |
1508 | // raise our own buffer overflow error at the end. | |
1509 | *status = U_ZERO_ERROR; | |
1510 | } | |
1511 | ||
1512 | if (U_FAILURE(*status)) { | |
1513 | // Can fail if group number is out of range. | |
1514 | break; | |
1515 | } | |
1516 | ||
1517 | } | |
1518 | ||
1519 | // | |
1520 | // Nul Terminate the dest buffer if possible. | |
1521 | // Set the appropriate buffer overflow or not terminated error, if needed. | |
1522 | // | |
1523 | if (destIdx < capacity) { | |
1524 | dest[destIdx] = 0; | |
1525 | } else if (destIdx == *destCapacity) { | |
1526 | *status = U_STRING_NOT_TERMINATED_WARNING; | |
1527 | } else { | |
1528 | *status = U_BUFFER_OVERFLOW_ERROR; | |
1529 | } | |
1530 | ||
1531 | // | |
1532 | // Return an updated dest buffer and capacity to the caller. | |
1533 | // | |
1534 | if (destIdx > 0 && *destCapacity > 0) { | |
1535 | if (destIdx < capacity) { | |
1536 | *destBuf += destIdx; | |
1537 | *destCapacity -= destIdx; | |
1538 | } else { | |
1539 | *destBuf += capacity; | |
1540 | *destCapacity = 0; | |
1541 | } | |
1542 | } | |
1543 | ||
1544 | // If we came in with a buffer overflow, make sure we go out with one also. | |
1545 | // (A zero length match right at the end of the previous match could | |
1546 | // make this function succeed even though a previous call had overflowed the buf) | |
1547 | if (pendingBufferOverflow && U_SUCCESS(*status)) { | |
1548 | *status = U_BUFFER_OVERFLOW_ERROR; | |
1549 | } | |
1550 | ||
1551 | return destIdx; | |
1552 | } | |
1553 | ||
1554 | // | |
729e4ab9 | 1555 | // appendReplacement the actual API function, |
374ca955 A |
1556 | // |
1557 | U_CAPI int32_t U_EXPORT2 | |
729e4ab9 A |
1558 | uregex_appendReplacement(URegularExpression *regexp2, |
1559 | const UChar *replacementText, | |
1560 | int32_t replacementLength, | |
1561 | UChar **destBuf, | |
1562 | int32_t *destCapacity, | |
1563 | UErrorCode *status) { | |
1564 | ||
1565 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
374ca955 A |
1566 | return RegexCImpl::appendReplacement( |
1567 | regexp, replacementText, replacementLength,destBuf, destCapacity, status); | |
1568 | } | |
1569 | ||
729e4ab9 A |
1570 | // |
1571 | // uregex_appendReplacementUText...can just use the normal C++ method | |
1572 | // | |
1573 | U_CAPI void U_EXPORT2 | |
1574 | uregex_appendReplacementUText(URegularExpression *regexp2, | |
1575 | UText *replText, | |
1576 | UText *dest, | |
1577 | UErrorCode *status) { | |
1578 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
1579 | regexp->fMatcher->appendReplacement(dest, replText, *status); | |
1580 | } | |
1581 | ||
374ca955 | 1582 | |
73c04bcf | 1583 | //------------------------------------------------------------------------------ |
374ca955 A |
1584 | // |
1585 | // uregex_appendTail | |
1586 | // | |
73c04bcf | 1587 | //------------------------------------------------------------------------------ |
729e4ab9 A |
1588 | int32_t RegexCImpl::appendTail(RegularExpression *regexp, |
1589 | UChar **destBuf, | |
1590 | int32_t *destCapacity, | |
1591 | UErrorCode *status) | |
46f4442e | 1592 | { |
374ca955 A |
1593 | |
1594 | // If we come in with a buffer overflow error, don't suppress the operation. | |
1595 | // A series of appendReplacements, appendTail need to correctly preflight | |
1596 | // the buffer size when an overflow happens somewhere in the middle. | |
1597 | UBool pendingBufferOverflow = FALSE; | |
729e4ab9 | 1598 | if (*status == U_BUFFER_OVERFLOW_ERROR && destCapacity != NULL && *destCapacity == 0) { |
374ca955 A |
1599 | pendingBufferOverflow = TRUE; |
1600 | *status = U_ZERO_ERROR; | |
1601 | } | |
1602 | ||
4388f060 | 1603 | if (validateRE(regexp, TRUE, status) == FALSE) { |
374ca955 A |
1604 | return 0; |
1605 | } | |
729e4ab9 A |
1606 | |
1607 | if (destCapacity == NULL || destBuf == NULL || | |
1608 | (*destBuf == NULL && *destCapacity > 0) || | |
1609 | *destCapacity < 0) | |
1610 | { | |
1611 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
1612 | return 0; | |
374ca955 A |
1613 | } |
1614 | ||
729e4ab9 A |
1615 | RegexMatcher *m = regexp->fMatcher; |
1616 | ||
374ca955 A |
1617 | int32_t destIdx = 0; |
1618 | int32_t destCap = *destCapacity; | |
1619 | UChar *dest = *destBuf; | |
729e4ab9 A |
1620 | |
1621 | if (regexp->fText != NULL) { | |
1622 | int32_t srcIdx; | |
1623 | int64_t nativeIdx = (m->fMatch ? m->fMatchEnd : m->fLastMatchEnd); | |
1624 | if (nativeIdx == -1) { | |
1625 | srcIdx = 0; | |
1626 | } else if (UTEXT_USES_U16(m->fInputText)) { | |
1627 | srcIdx = (int32_t)nativeIdx; | |
374ca955 | 1628 | } else { |
729e4ab9 A |
1629 | UErrorCode status = U_ZERO_ERROR; |
1630 | srcIdx = utext_extract(m->fInputText, 0, nativeIdx, NULL, 0, &status); | |
1631 | } | |
1632 | ||
1633 | for (;;) { | |
4388f060 A |
1634 | U_ASSERT(destIdx >= 0); |
1635 | ||
729e4ab9 | 1636 | if (srcIdx == regexp->fTextLength) { |
374ca955 A |
1637 | break; |
1638 | } | |
729e4ab9 A |
1639 | UChar c = regexp->fText[srcIdx]; |
1640 | if (c == 0 && regexp->fTextLength == -1) { | |
1641 | regexp->fTextLength = srcIdx; | |
1642 | break; | |
1643 | } | |
4388f060 | 1644 | |
729e4ab9 A |
1645 | if (destIdx < destCap) { |
1646 | dest[destIdx] = c; | |
1647 | } else { | |
1648 | // We've overflowed the dest buffer. | |
1649 | // If the total input string length is known, we can | |
1650 | // compute the total buffer size needed without scanning through the string. | |
1651 | if (regexp->fTextLength > 0) { | |
1652 | destIdx += (regexp->fTextLength - srcIdx); | |
1653 | break; | |
1654 | } | |
1655 | } | |
1656 | srcIdx++; | |
1657 | destIdx++; | |
1658 | } | |
1659 | } else { | |
1660 | int64_t srcIdx; | |
1661 | if (m->fMatch) { | |
1662 | // The most recent call to find() succeeded. | |
1663 | srcIdx = m->fMatchEnd; | |
1664 | } else { | |
1665 | // The last call to find() on this matcher failed(). | |
1666 | // Look back to the end of the last find() that succeeded for src index. | |
1667 | srcIdx = m->fLastMatchEnd; | |
1668 | if (srcIdx == -1) { | |
1669 | // There has been no successful match with this matcher. | |
1670 | // We want to copy the whole string. | |
1671 | srcIdx = 0; | |
1672 | } | |
374ca955 | 1673 | } |
729e4ab9 A |
1674 | |
1675 | destIdx = utext_extract(m->fInputText, srcIdx, m->fInputLength, dest, destCap, status); | |
374ca955 A |
1676 | } |
1677 | ||
1678 | // | |
1679 | // NUL terminate the output string, if possible, otherwise issue the | |
1680 | // appropriate error or warning. | |
1681 | // | |
1682 | if (destIdx < destCap) { | |
1683 | dest[destIdx] = 0; | |
1684 | } else if (destIdx == destCap) { | |
1685 | *status = U_STRING_NOT_TERMINATED_WARNING; | |
1686 | } else { | |
1687 | *status = U_BUFFER_OVERFLOW_ERROR; | |
1688 | } | |
1689 | ||
1690 | // | |
1691 | // Update the user's buffer ptr and capacity vars to reflect the | |
1692 | // amount used. | |
1693 | // | |
1694 | if (destIdx < destCap) { | |
1695 | *destBuf += destIdx; | |
1696 | *destCapacity -= destIdx; | |
4388f060 | 1697 | } else if (*destBuf != NULL) { |
374ca955 A |
1698 | *destBuf += destCap; |
1699 | *destCapacity = 0; | |
1700 | } | |
1701 | ||
1702 | if (pendingBufferOverflow && U_SUCCESS(*status)) { | |
1703 | *status = U_BUFFER_OVERFLOW_ERROR; | |
1704 | } | |
1705 | ||
1706 | return destIdx; | |
73c04bcf | 1707 | } |
374ca955 A |
1708 | |
1709 | ||
729e4ab9 A |
1710 | // |
1711 | // appendTail the actual API function | |
1712 | // | |
374ca955 | 1713 | U_CAPI int32_t U_EXPORT2 |
729e4ab9 | 1714 | uregex_appendTail(URegularExpression *regexp2, |
374ca955 A |
1715 | UChar **destBuf, |
1716 | int32_t *destCapacity, | |
1717 | UErrorCode *status) { | |
729e4ab9 | 1718 | RegularExpression *regexp = (RegularExpression*)regexp2; |
374ca955 A |
1719 | return RegexCImpl::appendTail(regexp, destBuf, destCapacity, status); |
1720 | } | |
1721 | ||
1722 | ||
729e4ab9 A |
1723 | // |
1724 | // uregex_appendTailUText...can just use the normal C++ method | |
1725 | // | |
1726 | U_CAPI UText * U_EXPORT2 | |
1727 | uregex_appendTailUText(URegularExpression *regexp2, | |
1728 | UText *dest, | |
1729 | UErrorCode *status) { | |
1730 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
1731 | return regexp->fMatcher->appendTail(dest, *status); | |
1732 | } | |
1733 | ||
1734 | ||
73c04bcf | 1735 | //------------------------------------------------------------------------------ |
374ca955 A |
1736 | // |
1737 | // copyString Internal utility to copy a string to an output buffer, | |
1738 | // while managing buffer overflow and preflight size | |
1739 | // computation. NUL termination is added to destination, | |
1740 | // and the NUL is counted in the output size. | |
1741 | // | |
73c04bcf | 1742 | //------------------------------------------------------------------------------ |
729e4ab9 | 1743 | #if 0 |
374ca955 A |
1744 | static void copyString(UChar *destBuffer, // Destination buffer. |
1745 | int32_t destCapacity, // Total capacity of dest buffer | |
1746 | int32_t *destIndex, // Index into dest buffer. Updated on return. | |
1747 | // Update not clipped to destCapacity. | |
1748 | const UChar *srcPtr, // Pointer to source string | |
1749 | int32_t srcLen) // Source string len. | |
1750 | { | |
1751 | int32_t si; | |
1752 | int32_t di = *destIndex; | |
1753 | UChar c; | |
1754 | ||
1755 | for (si=0; si<srcLen; si++) { | |
1756 | c = srcPtr[si]; | |
1757 | if (di < destCapacity) { | |
1758 | destBuffer[di] = c; | |
1759 | di++; | |
1760 | } else { | |
1761 | di += srcLen - si; | |
1762 | break; | |
1763 | } | |
1764 | } | |
73c04bcf A |
1765 | if (di<destCapacity) { |
1766 | destBuffer[di] = 0; | |
1767 | } | |
1768 | di++; | |
374ca955 A |
1769 | *destIndex = di; |
1770 | } | |
729e4ab9 | 1771 | #endif |
374ca955 | 1772 | |
73c04bcf | 1773 | //------------------------------------------------------------------------------ |
374ca955 A |
1774 | // |
1775 | // uregex_split | |
1776 | // | |
73c04bcf | 1777 | //------------------------------------------------------------------------------ |
729e4ab9 A |
1778 | int32_t RegexCImpl::split(RegularExpression *regexp, |
1779 | UChar *destBuf, | |
1780 | int32_t destCapacity, | |
1781 | int32_t *requiredCapacity, | |
1782 | UChar *destFields[], | |
1783 | int32_t destFieldsCapacity, | |
1784 | UErrorCode *status) { | |
374ca955 A |
1785 | // |
1786 | // Reset for the input text | |
1787 | // | |
1788 | regexp->fMatcher->reset(); | |
729e4ab9 A |
1789 | UText *inputText = regexp->fMatcher->fInputText; |
1790 | int64_t nextOutputStringStart = 0; | |
1791 | int64_t inputLen = regexp->fMatcher->fInputLength; | |
374ca955 A |
1792 | if (inputLen == 0) { |
1793 | return 0; | |
1794 | } | |
1795 | ||
374ca955 A |
1796 | // |
1797 | // Loop through the input text, searching for the delimiter pattern | |
1798 | // | |
1799 | int32_t i; // Index of the field being processed. | |
1800 | int32_t destIdx = 0; // Next available position in destBuf; | |
1801 | int32_t numCaptureGroups = regexp->fMatcher->groupCount(); | |
729e4ab9 | 1802 | UErrorCode tStatus = U_ZERO_ERROR; // Want to ignore any buffer overflow errors so that the strings are still counted |
374ca955 A |
1803 | for (i=0; ; i++) { |
1804 | if (i>=destFieldsCapacity-1) { | |
729e4ab9 | 1805 | // There are one or zero output strings left. |
374ca955 A |
1806 | // Fill the last output string with whatever is left from the input, then exit the loop. |
1807 | // ( i will be == destFieldsCapacity if we filled the output array while processing | |
1808 | // capture groups of the delimiter expression, in which case we will discard the | |
1809 | // last capture group saved in favor of the unprocessed remainder of the | |
1810 | // input string.) | |
729e4ab9 A |
1811 | if (inputLen > nextOutputStringStart) { |
1812 | if (i != destFieldsCapacity-1) { | |
1813 | // No fields are left. Recycle the last one for holding the trailing part of | |
1814 | // the input string. | |
1815 | i = destFieldsCapacity-1; | |
1816 | destIdx = (int32_t)(destFields[i] - destFields[0]); | |
1817 | } | |
1818 | ||
1819 | destFields[i] = &destBuf[destIdx]; | |
1820 | destIdx += 1 + utext_extract(inputText, nextOutputStringStart, inputLen, | |
1821 | &destBuf[destIdx], REMAINING_CAPACITY(destIdx, destCapacity), status); | |
374ca955 | 1822 | } |
374ca955 A |
1823 | break; |
1824 | } | |
1825 | ||
1826 | if (regexp->fMatcher->find()) { | |
1827 | // We found another delimiter. Move everything from where we started looking | |
1828 | // up until the start of the delimiter into the next output string. | |
374ca955 | 1829 | destFields[i] = &destBuf[destIdx]; |
729e4ab9 A |
1830 | |
1831 | destIdx += 1 + utext_extract(inputText, nextOutputStringStart, regexp->fMatcher->fMatchStart, | |
1832 | &destBuf[destIdx], REMAINING_CAPACITY(destIdx, destCapacity), &tStatus); | |
1833 | if (tStatus == U_BUFFER_OVERFLOW_ERROR) { | |
1834 | tStatus = U_ZERO_ERROR; | |
1835 | } else { | |
1836 | *status = tStatus; | |
1837 | } | |
1838 | nextOutputStringStart = regexp->fMatcher->fMatchEnd; | |
374ca955 A |
1839 | |
1840 | // If the delimiter pattern has capturing parentheses, the captured | |
1841 | // text goes out into the next n destination strings. | |
1842 | int32_t groupNum; | |
1843 | for (groupNum=1; groupNum<=numCaptureGroups; groupNum++) { | |
1844 | // If we've run out of output string slots, bail out. | |
1845 | if (i==destFieldsCapacity-1) { | |
1846 | break; | |
1847 | } | |
1848 | i++; | |
1849 | ||
1850 | // Set up to extract the capture group contents into the dest buffer. | |
374ca955 | 1851 | destFields[i] = &destBuf[destIdx]; |
729e4ab9 | 1852 | tStatus = U_ZERO_ERROR; |
4388f060 A |
1853 | int32_t t = uregex_group((URegularExpression*)regexp, |
1854 | groupNum, | |
1855 | destFields[i], | |
1856 | REMAINING_CAPACITY(destIdx, destCapacity), | |
1857 | &tStatus); | |
374ca955 A |
1858 | destIdx += t + 1; // Record the space used in the output string buffer. |
1859 | // +1 for the NUL that terminates the string. | |
729e4ab9 A |
1860 | if (tStatus == U_BUFFER_OVERFLOW_ERROR) { |
1861 | tStatus = U_ZERO_ERROR; | |
1862 | } else { | |
1863 | *status = tStatus; | |
1864 | } | |
374ca955 A |
1865 | } |
1866 | ||
1867 | if (nextOutputStringStart == inputLen) { | |
4388f060 A |
1868 | // The delimiter was at the end of the string. |
1869 | // Output an empty string, and then we are done. | |
1870 | if (destIdx < destCapacity) { | |
1871 | destBuf[destIdx] = 0; | |
1872 | } | |
1873 | if (i < destFieldsCapacity-1) { | |
1874 | ++i; | |
1875 | } | |
1876 | if (destIdx < destCapacity) { | |
1877 | destFields[i] = destBuf + destIdx; | |
1878 | } | |
1879 | ++destIdx; | |
374ca955 A |
1880 | break; |
1881 | } | |
1882 | ||
1883 | } | |
1884 | else | |
1885 | { | |
1886 | // We ran off the end of the input while looking for the next delimiter. | |
1887 | // All the remaining text goes into the current output string. | |
1888 | destFields[i] = &destBuf[destIdx]; | |
729e4ab9 A |
1889 | destIdx += 1 + utext_extract(inputText, nextOutputStringStart, inputLen, |
1890 | &destBuf[destIdx], REMAINING_CAPACITY(destIdx, destCapacity), status); | |
374ca955 A |
1891 | break; |
1892 | } | |
1893 | } | |
1894 | ||
1895 | // Zero out any unused portion of the destFields array | |
1896 | int j; | |
1897 | for (j=i+1; j<destFieldsCapacity; j++) { | |
1898 | destFields[j] = NULL; | |
1899 | } | |
1900 | ||
1901 | if (requiredCapacity != NULL) { | |
1902 | *requiredCapacity = destIdx; | |
1903 | } | |
73c04bcf | 1904 | if (destIdx > destCapacity) { |
374ca955 A |
1905 | *status = U_BUFFER_OVERFLOW_ERROR; |
1906 | } | |
1907 | return i+1; | |
1908 | } | |
1909 | ||
729e4ab9 A |
1910 | // |
1911 | // uregex_split The actual API function | |
1912 | // | |
1913 | U_CAPI int32_t U_EXPORT2 | |
1914 | uregex_split(URegularExpression *regexp2, | |
1915 | UChar *destBuf, | |
1916 | int32_t destCapacity, | |
1917 | int32_t *requiredCapacity, | |
1918 | UChar *destFields[], | |
1919 | int32_t destFieldsCapacity, | |
1920 | UErrorCode *status) { | |
1921 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
4388f060 | 1922 | if (validateRE(regexp, TRUE, status) == FALSE) { |
729e4ab9 A |
1923 | return 0; |
1924 | } | |
1925 | if ((destBuf == NULL && destCapacity > 0) || | |
1926 | destCapacity < 0 || | |
1927 | destFields == NULL || | |
1928 | destFieldsCapacity < 1 ) { | |
1929 | *status = U_ILLEGAL_ARGUMENT_ERROR; | |
1930 | return 0; | |
1931 | } | |
1932 | ||
1933 | return RegexCImpl::split(regexp, destBuf, destCapacity, requiredCapacity, destFields, destFieldsCapacity, status); | |
1934 | } | |
1935 | ||
1936 | ||
1937 | // | |
1938 | // uregex_splitUText...can just use the normal C++ method | |
1939 | // | |
1940 | U_CAPI int32_t U_EXPORT2 | |
1941 | uregex_splitUText(URegularExpression *regexp2, | |
1942 | UText *destFields[], | |
1943 | int32_t destFieldsCapacity, | |
1944 | UErrorCode *status) { | |
1945 | RegularExpression *regexp = (RegularExpression*)regexp2; | |
1946 | return regexp->fMatcher->split(regexp->fMatcher->inputText(), destFields, destFieldsCapacity, *status); | |
1947 | } | |
1948 | ||
374ca955 | 1949 | |
374ca955 | 1950 | #endif // !UCONFIG_NO_REGULAR_EXPRESSIONS |
73c04bcf | 1951 |