1 /* 2 ********************************************************************** 3 * Copyright (C) 2004-2016, International Business Machines 4 * Corporation and others. All Rights Reserved. 5 ********************************************************************** 6 * file name: uregex.h 7 * encoding: US-ASCII 8 * indentation:4 9 * 10 * created on: 2004mar09 11 * created by: Andy Heninger 12 * 13 * ICU Regular Expressions, API for C 14 */ 15 16 /** 17 * \file 18 * \brief C API: Regular Expressions 19 * 20 * <p>This is a C wrapper around the C++ RegexPattern and RegexMatcher classes.</p> 21 */ 22 23 #ifndef UREGEX_H 24 #define UREGEX_H 25 26 #include "unicode/utext.h" 27 #include "unicode/utypes.h" 28 29 #if !UCONFIG_NO_REGULAR_EXPRESSIONS 30 31 #include "unicode/localpointer.h" 32 #include "unicode/parseerr.h" 33 34 struct URegularExpression; 35 /** 36 * Structure representing a compiled regular expression, plus the results 37 * of a match operation. 38 * @stable ICU 3.0 39 */ 40 typedef struct URegularExpression URegularExpression; 41 42 43 /** 44 * Constants for Regular Expression Match Modes. 45 * @stable ICU 2.4 46 */ 47 typedef enum URegexpFlag{ 48 49 #ifndef U_HIDE_DRAFT_API 50 /** Forces normalization of pattern and strings. 51 Not implemented yet, just a placeholder, hence draft. 52 @draft ICU 2.4 */ 53 UREGEX_CANON_EQ = 128, 54 #endif /* U_HIDE_DRAFT_API */ 55 /** Enable case insensitive matching. @stable ICU 2.4 */ 56 UREGEX_CASE_INSENSITIVE = 2, 57 58 /** Allow white space and comments within patterns @stable ICU 2.4 */ 59 UREGEX_COMMENTS = 4, 60 61 /** If set, '.' matches line terminators, otherwise '.' matching stops at line end. 62 * @stable ICU 2.4 */ 63 UREGEX_DOTALL = 32, 64 65 /** If set, treat the entire pattern as a literal string. 66 * Metacharacters or escape sequences in the input sequence will be given 67 * no special meaning. 68 * 69 * The flag UREGEX_CASE_INSENSITIVE retains its impact 70 * on matching when used in conjunction with this flag. 71 * The other flags become superfluous. 72 * 73 * @stable ICU 4.0 74 */ 75 UREGEX_LITERAL = 16, 76 77 /** Control behavior of "$" and "^" 78 * If set, recognize line terminators within string, 79 * otherwise, match only at start and end of input string. 80 * @stable ICU 2.4 */ 81 UREGEX_MULTILINE = 8, 82 83 /** Unix-only line endings. 84 * When this mode is enabled, only \\u000a is recognized as a line ending 85 * in the behavior of ., ^, and $. 86 * @stable ICU 4.0 87 */ 88 UREGEX_UNIX_LINES = 1, 89 90 /** Unicode word boundaries. 91 * If set, \b uses the Unicode TR 29 definition of word boundaries. 92 * Warning: Unicode word boundaries are quite different from 93 * traditional regular expression word boundaries. See 94 * http://unicode.org/reports/tr29/#Word_Boundaries 95 * @stable ICU 2.8 96 */ 97 UREGEX_UWORD = 256, 98 99 /** Error on Unrecognized backslash escapes. 100 * If set, fail with an error on patterns that contain 101 * backslash-escaped ASCII letters without a known special 102 * meaning. If this flag is not set, these 103 * escaped letters represent themselves. 104 * @stable ICU 4.0 105 */ 106 UREGEX_ERROR_ON_UNKNOWN_ESCAPES = 512 107 108 } URegexpFlag; 109 110 /** 111 * Open (compile) an ICU regular expression. Compiles the regular expression in 112 * string form into an internal representation using the specified match mode flags. 113 * The resulting regular expression handle can then be used to perform various 114 * matching operations. 115 * 116 * 117 * @param pattern The Regular Expression pattern to be compiled. 118 * @param patternLength The length of the pattern, or -1 if the pattern is 119 * NUL terminated. 120 * @param flags Flags that alter the default matching behavior for 121 * the regular expression, UREGEX_CASE_INSENSITIVE, for 122 * example. For default behavior, set this parameter to zero. 123 * See <code>enum URegexpFlag</code>. All desired flags 124 * are bitwise-ORed together. 125 * @param pe Receives the position (line and column numbers) of any syntax 126 * error within the source regular expression string. If this 127 * information is not wanted, pass NULL for this parameter. 128 * @param status Receives error detected by this function. 129 * @stable ICU 3.0 130 * 131 */ 132 U_STABLE URegularExpression * U_EXPORT2 133 uregex_open( const UChar *pattern, 134 int32_t patternLength, 135 uint32_t flags, 136 UParseError *pe, 137 UErrorCode *status); 138 139 /** 140 * Open (compile) an ICU regular expression. Compiles the regular expression in 141 * string form into an internal representation using the specified match mode flags. 142 * The resulting regular expression handle can then be used to perform various 143 * matching operations. 144 * <p> 145 * The contents of the pattern UText will be extracted and saved. Ownership of the 146 * UText struct itself remains with the caller. This is to match the behavior of 147 * uregex_open(). 148 * 149 * @param pattern The Regular Expression pattern to be compiled. 150 * @param flags Flags that alter the default matching behavior for 151 * the regular expression, UREGEX_CASE_INSENSITIVE, for 152 * example. For default behavior, set this parameter to zero. 153 * See <code>enum URegexpFlag</code>. All desired flags 154 * are bitwise-ORed together. 155 * @param pe Receives the position (line and column numbers) of any syntax 156 * error within the source regular expression string. If this 157 * information is not wanted, pass NULL for this parameter. 158 * @param status Receives error detected by this function. 159 * 160 * @stable ICU 4.6 161 */ 162 U_STABLE URegularExpression * U_EXPORT2 163 uregex_openUText(UText *pattern, 164 uint32_t flags, 165 UParseError *pe, 166 UErrorCode *status); 167 168 /** 169 * Open (compile) an ICU regular expression. The resulting regular expression 170 * handle can then be used to perform various matching operations. 171 * <p> 172 * This function is the same as uregex_open, except that the pattern 173 * is supplied as an 8 bit char * string in the default code page. 174 * 175 * @param pattern The Regular Expression pattern to be compiled, 176 * NUL terminated. 177 * @param flags Flags that alter the default matching behavior for 178 * the regular expression, UREGEX_CASE_INSENSITIVE, for 179 * example. For default behavior, set this parameter to zero. 180 * See <code>enum URegexpFlag</code>. All desired flags 181 * are bitwise-ORed together. 182 * @param pe Receives the position (line and column numbers) of any syntax 183 * error within the source regular expression string. If this 184 * information is not wanted, pass NULL for this parameter. 185 * @param status Receives errors detected by this function. 186 * @return The URegularExpression object representing the compiled 187 * pattern. 188 * 189 * @stable ICU 3.0 190 */ 191 #if !UCONFIG_NO_CONVERSION 192 U_STABLE URegularExpression * U_EXPORT2 193 uregex_openC( const char *pattern, 194 uint32_t flags, 195 UParseError *pe, 196 UErrorCode *status); 197 #endif 198 199 200 201 /** 202 * Close the regular expression, recovering all resources (memory) it 203 * was holding. 204 * 205 * @param regexp The regular expression to be closed. 206 * @stable ICU 3.0 207 */ 208 U_STABLE void U_EXPORT2 209 uregex_close(URegularExpression *regexp); 210 211 #if U_SHOW_CPLUSPLUS_API 212 213 U_NAMESPACE_BEGIN 214 215 /** 216 * \class LocalURegularExpressionPointer 217 * "Smart pointer" class, closes a URegularExpression via uregex_close(). 218 * For most methods see the LocalPointerBase base class. 219 * 220 * @see LocalPointerBase 221 * @see LocalPointer 222 * @stable ICU 4.4 223 */ 224 U_DEFINE_LOCAL_OPEN_POINTER(LocalURegularExpressionPointer, URegularExpression, uregex_close); 225 226 U_NAMESPACE_END 227 228 #endif 229 230 /** 231 * Make a copy of a compiled regular expression. Cloning a regular 232 * expression is faster than opening a second instance from the source 233 * form of the expression, and requires less memory. 234 * <p> 235 * Note that the current input string and the position of any matched text 236 * within it are not cloned; only the pattern itself and the 237 * match mode flags are copied. 238 * <p> 239 * Cloning can be particularly useful to threaded applications that perform 240 * multiple match operations in parallel. Each concurrent RE 241 * operation requires its own instance of a URegularExpression. 242 * 243 * @param regexp The compiled regular expression to be cloned. 244 * @param status Receives indication of any errors encountered 245 * @return the cloned copy of the compiled regular expression. 246 * @stable ICU 3.0 247 */ 248 U_STABLE URegularExpression * U_EXPORT2 249 uregex_clone(const URegularExpression *regexp, UErrorCode *status); 250 251 /** 252 * Returns a pointer to the source form of the pattern for this regular expression. 253 * This function will work even if the pattern was originally specified as a UText. 254 * 255 * @param regexp The compiled regular expression. 256 * @param patLength This output parameter will be set to the length of the 257 * pattern string. A NULL pointer may be used here if the 258 * pattern length is not needed, as would be the case if 259 * the pattern is known in advance to be a NUL terminated 260 * string. 261 * @param status Receives errors detected by this function. 262 * @return a pointer to the pattern string. The storage for the string is 263 * owned by the regular expression object, and must not be 264 * altered or deleted by the application. The returned string 265 * will remain valid until the regular expression is closed. 266 * @stable ICU 3.0 267 */ 268 U_STABLE const UChar * U_EXPORT2 269 uregex_pattern(const URegularExpression *regexp, 270 int32_t *patLength, 271 UErrorCode *status); 272 273 /** 274 * Returns the source text of the pattern for this regular expression. 275 * This function will work even if the pattern was originally specified as a UChar string. 276 * 277 * @param regexp The compiled regular expression. 278 * @param status Receives errors detected by this function. 279 * @return the pattern text. The storage for the text is owned by the regular expression 280 * object, and must not be altered or deleted. 281 * 282 * @stable ICU 4.6 283 */ 284 U_STABLE UText * U_EXPORT2 285 uregex_patternUText(const URegularExpression *regexp, 286 UErrorCode *status); 287 288 /** 289 * Get the match mode flags that were specified when compiling this regular expression. 290 * @param status Receives errors detected by this function. 291 * @param regexp The compiled regular expression. 292 * @return The match mode flags 293 * @see URegexpFlag 294 * @stable ICU 3.0 295 */ 296 U_STABLE int32_t U_EXPORT2 297 uregex_flags(const URegularExpression *regexp, 298 UErrorCode *status); 299 300 301 /** 302 * Set the subject text string upon which the regular expression will look for matches. 303 * This function may be called any number of times, allowing the regular 304 * expression pattern to be applied to different strings. 305 * <p> 306 * Regular expression matching operations work directly on the application's 307 * string data. No copy is made. The subject string data must not be 308 * altered after calling this function until after all regular expression 309 * operations involving this string data are completed. 310 * <p> 311 * Zero length strings are permitted. In this case, no subsequent match 312 * operation will dereference the text string pointer. 313 * 314 * @param regexp The compiled regular expression. 315 * @param text The subject text string. 316 * @param textLength The length of the subject text, or -1 if the string 317 * is NUL terminated. 318 * @param status Receives errors detected by this function. 319 * @stable ICU 3.0 320 */ 321 U_STABLE void U_EXPORT2 322 uregex_setText(URegularExpression *regexp, 323 const UChar *text, 324 int32_t textLength, 325 UErrorCode *status); 326 327 328 /** 329 * Set the subject text string upon which the regular expression will look for matches. 330 * This function may be called any number of times, allowing the regular 331 * expression pattern to be applied to different strings. 332 * <p> 333 * Regular expression matching operations work directly on the application's 334 * string data; only a shallow clone is made. The subject string data must not be 335 * altered after calling this function until after all regular expression 336 * operations involving this string data are completed. 337 * 338 * @param regexp The compiled regular expression. 339 * @param text The subject text string. 340 * @param status Receives errors detected by this function. 341 * 342 * @stable ICU 4.6 343 */ 344 U_STABLE void U_EXPORT2 345 uregex_setUText(URegularExpression *regexp, 346 UText *text, 347 UErrorCode *status); 348 349 /** 350 * Get the subject text that is currently associated with this 351 * regular expression object. If the input was supplied using uregex_setText(), 352 * that pointer will be returned. Otherwise, the characters in the input will 353 * be extracted to a buffer and returned. In either case, ownership remains 354 * with the regular expression object. 355 * 356 * This function will work even if the input was originally specified as a UText. 357 * 358 * @param regexp The compiled regular expression. 359 * @param textLength The length of the string is returned in this output parameter. 360 * A NULL pointer may be used here if the 361 * text length is not needed, as would be the case if 362 * the text is known in advance to be a NUL terminated 363 * string. 364 * @param status Receives errors detected by this function. 365 * @return Pointer to the subject text string currently associated with 366 * this regular expression. 367 * @stable ICU 3.0 368 */ 369 U_STABLE const UChar * U_EXPORT2 370 uregex_getText(URegularExpression *regexp, 371 int32_t *textLength, 372 UErrorCode *status); 373 374 /** 375 * Get the subject text that is currently associated with this 376 * regular expression object. 377 * 378 * This function will work even if the input was originally specified as a UChar string. 379 * 380 * @param regexp The compiled regular expression. 381 * @param dest A mutable UText in which to store the current input. 382 * If NULL, a new UText will be created as an immutable shallow clone 383 * of the actual input string. 384 * @param status Receives errors detected by this function. 385 * @return The subject text currently associated with this regular expression. 386 * If a pre-allocated UText was provided, it will always be used and returned. 387 * 388 * @stable ICU 4.6 389 */ 390 U_STABLE UText * U_EXPORT2 391 uregex_getUText(URegularExpression *regexp, 392 UText *dest, 393 UErrorCode *status); 394 395 /** 396 * Set the subject text string upon which the regular expression is looking for matches 397 * without changing any other aspect of the matching state. 398 * The new and previous text strings must have the same content. 399 * 400 * This function is intended for use in environments where ICU is operating on 401 * strings that may move around in memory. It provides a mechanism for notifying 402 * ICU that the string has been relocated, and providing a new UText to access the 403 * string in its new position. 404 * 405 * Note that the regular expression implementation never copies the underlying text 406 * of a string being matched, but always operates directly on the original text 407 * provided by the user. Refreshing simply drops the references to the old text 408 * and replaces them with references to the new. 409 * 410 * Caution: this function is normally used only by very specialized 411 * system-level code. One example use case is with garbage collection 412 * that moves the text in memory. 413 * 414 * @param regexp The compiled regular expression. 415 * @param text The new (moved) text string. 416 * @param status Receives errors detected by this function. 417 * 418 * @stable ICU 4.8 419 */ 420 U_STABLE void U_EXPORT2 421 uregex_refreshUText(URegularExpression *regexp, 422 UText *text, 423 UErrorCode *status); 424 425 /** 426 * Attempts to match the input string against the pattern. 427 * To succeed, the match must extend to the end of the string, 428 * or cover the complete match region. 429 * 430 * If startIndex >= zero the match operation starts at the specified 431 * index and must extend to the end of the input string. Any region 432 * that has been specified is reset. 433 * 434 * If startIndex == -1 the match must cover the input region, or the entire 435 * input string if no region has been set. This directly corresponds to 436 * Matcher.matches() in Java 437 * 438 * @param regexp The compiled regular expression. 439 * @param startIndex The input string (native) index at which to begin matching, or -1 440 * to match the input Region. 441 * @param status Receives errors detected by this function. 442 * @return TRUE if there is a match 443 * @stable ICU 3.0 444 */ 445 U_STABLE UBool U_EXPORT2 446 uregex_matches(URegularExpression *regexp, 447 int32_t startIndex, 448 UErrorCode *status); 449 450 /** 451 * 64bit version of uregex_matches. 452 * Attempts to match the input string against the pattern. 453 * To succeed, the match must extend to the end of the string, 454 * or cover the complete match region. 455 * 456 * If startIndex >= zero the match operation starts at the specified 457 * index and must extend to the end of the input string. Any region 458 * that has been specified is reset. 459 * 460 * If startIndex == -1 the match must cover the input region, or the entire 461 * input string if no region has been set. This directly corresponds to 462 * Matcher.matches() in Java 463 * 464 * @param regexp The compiled regular expression. 465 * @param startIndex The input string (native) index at which to begin matching, or -1 466 * to match the input Region. 467 * @param status Receives errors detected by this function. 468 * @return TRUE if there is a match 469 * @stable ICU 4.6 470 */ 471 U_STABLE UBool U_EXPORT2 472 uregex_matches64(URegularExpression *regexp, 473 int64_t startIndex, 474 UErrorCode *status); 475 476 /** 477 * Attempts to match the input string, starting from the specified index, against the pattern. 478 * The match may be of any length, and is not required to extend to the end 479 * of the input string. Contrast with uregex_matches(). 480 * 481 * <p>If startIndex is >= 0 any input region that was set for this 482 * URegularExpression is reset before the operation begins. 483 * 484 * <p>If the specified starting index == -1 the match begins at the start of the input 485 * region, or at the start of the full string if no region has been specified. 486 * This corresponds directly with Matcher.lookingAt() in Java. 487 * 488 * <p>If the match succeeds then more information can be obtained via the 489 * <code>uregexp_start()</code>, <code>uregexp_end()</code>, 490 * and <code>uregex_group()</code> functions.</p> 491 * 492 * @param regexp The compiled regular expression. 493 * @param startIndex The input string (native) index at which to begin matching, or 494 * -1 to match the Input Region 495 * @param status A reference to a UErrorCode to receive any errors. 496 * @return TRUE if there is a match. 497 * @stable ICU 3.0 498 */ 499 U_STABLE UBool U_EXPORT2 500 uregex_lookingAt(URegularExpression *regexp, 501 int32_t startIndex, 502 UErrorCode *status); 503 504 /** 505 * 64bit version of uregex_lookingAt. 506 * Attempts to match the input string, starting from the specified index, against the pattern. 507 * The match may be of any length, and is not required to extend to the end 508 * of the input string. Contrast with uregex_matches(). 509 * 510 * <p>If startIndex is >= 0 any input region that was set for this 511 * URegularExpression is reset before the operation begins. 512 * 513 * <p>If the specified starting index == -1 the match begins at the start of the input 514 * region, or at the start of the full string if no region has been specified. 515 * This corresponds directly with Matcher.lookingAt() in Java. 516 * 517 * <p>If the match succeeds then more information can be obtained via the 518 * <code>uregexp_start()</code>, <code>uregexp_end()</code>, 519 * and <code>uregex_group()</code> functions.</p> 520 * 521 * @param regexp The compiled regular expression. 522 * @param startIndex The input string (native) index at which to begin matching, or 523 * -1 to match the Input Region 524 * @param status A reference to a UErrorCode to receive any errors. 525 * @return TRUE if there is a match. 526 * @stable ICU 4.6 527 */ 528 U_STABLE UBool U_EXPORT2 529 uregex_lookingAt64(URegularExpression *regexp, 530 int64_t startIndex, 531 UErrorCode *status); 532 533 /** 534 * Find the first matching substring of the input string that matches the pattern. 535 * If startIndex is >= zero the search for a match begins at the specified index, 536 * and any match region is reset. This corresponds directly with 537 * Matcher.find(startIndex) in Java. 538 * 539 * If startIndex == -1 the search begins at the start of the input region, 540 * or at the start of the full string if no region has been specified. 541 * 542 * If a match is found, <code>uregex_start(), uregex_end()</code>, and 543 * <code>uregex_group()</code> will provide more information regarding the match. 544 * 545 * @param regexp The compiled regular expression. 546 * @param startIndex The position (native) in the input string to begin the search, or 547 * -1 to search within the Input Region. 548 * @param status A reference to a UErrorCode to receive any errors. 549 * @return TRUE if a match is found. 550 * @stable ICU 3.0 551 */ 552 U_STABLE UBool U_EXPORT2 553 uregex_find(URegularExpression *regexp, 554 int32_t startIndex, 555 UErrorCode *status); 556 557 /** 558 * 64bit version of uregex_find. 559 * Find the first matching substring of the input string that matches the pattern. 560 * If startIndex is >= zero the search for a match begins at the specified index, 561 * and any match region is reset. This corresponds directly with 562 * Matcher.find(startIndex) in Java. 563 * 564 * If startIndex == -1 the search begins at the start of the input region, 565 * or at the start of the full string if no region has been specified. 566 * 567 * If a match is found, <code>uregex_start(), uregex_end()</code>, and 568 * <code>uregex_group()</code> will provide more information regarding the match. 569 * 570 * @param regexp The compiled regular expression. 571 * @param startIndex The position (native) in the input string to begin the search, or 572 * -1 to search within the Input Region. 573 * @param status A reference to a UErrorCode to receive any errors. 574 * @return TRUE if a match is found. 575 * @stable ICU 4.6 576 */ 577 U_STABLE UBool U_EXPORT2 578 uregex_find64(URegularExpression *regexp, 579 int64_t startIndex, 580 UErrorCode *status); 581 582 /** 583 * Find the next pattern match in the input string. Begin searching 584 * the input at the location following the end of he previous match, 585 * or at the start of the string (or region) if there is no 586 * previous match. If a match is found, <code>uregex_start(), uregex_end()</code>, and 587 * <code>uregex_group()</code> will provide more information regarding the match. 588 * 589 * @param regexp The compiled regular expression. 590 * @param status A reference to a UErrorCode to receive any errors. 591 * @return TRUE if a match is found. 592 * @see uregex_reset 593 * @stable ICU 3.0 594 */ 595 U_STABLE UBool U_EXPORT2 596 uregex_findNext(URegularExpression *regexp, 597 UErrorCode *status); 598 599 /** 600 * Get the number of capturing groups in this regular expression's pattern. 601 * @param regexp The compiled regular expression. 602 * @param status A reference to a UErrorCode to receive any errors. 603 * @return the number of capture groups 604 * @stable ICU 3.0 605 */ 606 U_STABLE int32_t U_EXPORT2 607 uregex_groupCount(URegularExpression *regexp, 608 UErrorCode *status); 609 610 /** 611 * Get the group number corresponding to a named capture group. 612 * The returned number can be used with any function that access 613 * capture groups by number. 614 * 615 * The function returns an error status if the specified name does not 616 * appear in the pattern. 617 * 618 * @param regexp The compiled regular expression. 619 * @param groupName The capture group name. 620 * @param nameLength The length of the name, or -1 if the name is a 621 * nul-terminated string. 622 * @param status A pointer to a UErrorCode to receive any errors. 623 * 624 * @stable ICU 55 625 */ 626 U_STABLE int32_t U_EXPORT2 627 uregex_groupNumberFromName(URegularExpression *regexp, 628 const UChar *groupName, 629 int32_t nameLength, 630 UErrorCode *status); 631 632 633 /** 634 * Get the group number corresponding to a named capture group. 635 * The returned number can be used with any function that access 636 * capture groups by number. 637 * 638 * The function returns an error status if the specified name does not 639 * appear in the pattern. 640 * 641 * @param regexp The compiled regular expression. 642 * @param groupName The capture group name, 643 * platform invariant characters only. 644 * @param nameLength The length of the name, or -1 if the name is 645 * nul-terminated. 646 * @param status A pointer to a UErrorCode to receive any errors. 647 * 648 * @stable ICU 55 649 */ 650 U_STABLE int32_t U_EXPORT2 651 uregex_groupNumberFromCName(URegularExpression *regexp, 652 const char *groupName, 653 int32_t nameLength, 654 UErrorCode *status); 655 656 /** Extract the string for the specified matching expression or subexpression. 657 * Group #0 is the complete string of matched text. 658 * Group #1 is the text matched by the first set of capturing parentheses. 659 * 660 * @param regexp The compiled regular expression. 661 * @param groupNum The capture group to extract. Group 0 is the complete 662 * match. The value of this parameter must be 663 * less than or equal to the number of capture groups in 664 * the pattern. 665 * @param dest Buffer to receive the matching string data 666 * @param destCapacity Capacity of the dest buffer. 667 * @param status A reference to a UErrorCode to receive any errors. 668 * @return Length of matching data, 669 * or -1 if no applicable match. 670 * @stable ICU 3.0 671 */ 672 U_STABLE int32_t U_EXPORT2 673 uregex_group(URegularExpression *regexp, 674 int32_t groupNum, 675 UChar *dest, 676 int32_t destCapacity, 677 UErrorCode *status); 678 679 /** Returns a shallow immutable clone of the entire input string with the current index set 680 * to the beginning of the requested capture group. The capture group length is also 681 * returned via groupLength. 682 * Group #0 is the complete string of matched text. 683 * Group #1 is the text matched by the first set of capturing parentheses. 684 * 685 * @param regexp The compiled regular expression. 686 * @param groupNum The capture group to extract. Group 0 is the complete 687 * match. The value of this parameter must be 688 * less than or equal to the number of capture groups in 689 * the pattern. 690 * @param dest A mutable UText in which to store the current input. 691 * If NULL, a new UText will be created as an immutable shallow clone 692 * of the entire input string. 693 * @param groupLength The group length of the desired capture group. Output parameter. 694 * @param status A reference to a UErrorCode to receive any errors. 695 * @return The subject text currently associated with this regular expression. 696 * If a pre-allocated UText was provided, it will always be used and returned. 697 698 * 699 * @stable ICU 4.6 700 */ 701 U_STABLE UText * U_EXPORT2 702 uregex_groupUText(URegularExpression *regexp, 703 int32_t groupNum, 704 UText *dest, 705 int64_t *groupLength, 706 UErrorCode *status); 707 708 /** 709 * Returns the index in the input string of the start of the text matched by the 710 * specified capture group during the previous match operation. Return -1 if 711 * the capture group was not part of the last match. 712 * Group #0 refers to the complete range of matched text. 713 * Group #1 refers to the text matched by the first set of capturing parentheses. 714 * 715 * @param regexp The compiled regular expression. 716 * @param groupNum The capture group number 717 * @param status A reference to a UErrorCode to receive any errors. 718 * @return the starting (native) position in the input of the text matched 719 * by the specified group. 720 * @stable ICU 3.0 721 */ 722 U_STABLE int32_t U_EXPORT2 723 uregex_start(URegularExpression *regexp, 724 int32_t groupNum, 725 UErrorCode *status); 726 727 /** 728 * 64bit version of uregex_start. 729 * Returns the index in the input string of the start of the text matched by the 730 * specified capture group during the previous match operation. Return -1 if 731 * the capture group was not part of the last match. 732 * Group #0 refers to the complete range of matched text. 733 * Group #1 refers to the text matched by the first set of capturing parentheses. 734 * 735 * @param regexp The compiled regular expression. 736 * @param groupNum The capture group number 737 * @param status A reference to a UErrorCode to receive any errors. 738 * @return the starting (native) position in the input of the text matched 739 * by the specified group. 740 * @stable ICU 4.6 741 */ 742 U_STABLE int64_t U_EXPORT2 743 uregex_start64(URegularExpression *regexp, 744 int32_t groupNum, 745 UErrorCode *status); 746 747 /** 748 * Returns the index in the input string of the position following the end 749 * of the text matched by the specified capture group. 750 * Return -1 if the capture group was not part of the last match. 751 * Group #0 refers to the complete range of matched text. 752 * Group #1 refers to the text matched by the first set of capturing parentheses. 753 * 754 * @param regexp The compiled regular expression. 755 * @param groupNum The capture group number 756 * @param status A reference to a UErrorCode to receive any errors. 757 * @return the (native) index of the position following the last matched character. 758 * @stable ICU 3.0 759 */ 760 U_STABLE int32_t U_EXPORT2 761 uregex_end(URegularExpression *regexp, 762 int32_t groupNum, 763 UErrorCode *status); 764 765 /** 766 * 64bit version of uregex_end. 767 * Returns the index in the input string of the position following the end 768 * of the text matched by the specified capture group. 769 * Return -1 if the capture group was not part of the last match. 770 * Group #0 refers to the complete range of matched text. 771 * Group #1 refers to the text matched by the first set of capturing parentheses. 772 * 773 * @param regexp The compiled regular expression. 774 * @param groupNum The capture group number 775 * @param status A reference to a UErrorCode to receive any errors. 776 * @return the (native) index of the position following the last matched character. 777 * @stable ICU 4.6 778 */ 779 U_STABLE int64_t U_EXPORT2 780 uregex_end64(URegularExpression *regexp, 781 int32_t groupNum, 782 UErrorCode *status); 783 784 /** 785 * Reset any saved state from the previous match. Has the effect of 786 * causing uregex_findNext to begin at the specified index, and causing 787 * uregex_start(), uregex_end() and uregex_group() to return an error 788 * indicating that there is no match information available. Clears any 789 * match region that may have been set. 790 * 791 * @param regexp The compiled regular expression. 792 * @param index The position (native) in the text at which a 793 * uregex_findNext() should begin searching. 794 * @param status A reference to a UErrorCode to receive any errors. 795 * @stable ICU 3.0 796 */ 797 U_STABLE void U_EXPORT2 798 uregex_reset(URegularExpression *regexp, 799 int32_t index, 800 UErrorCode *status); 801 802 /** 803 * 64bit version of uregex_reset. 804 * Reset any saved state from the previous match. Has the effect of 805 * causing uregex_findNext to begin at the specified index, and causing 806 * uregex_start(), uregex_end() and uregex_group() to return an error 807 * indicating that there is no match information available. Clears any 808 * match region that may have been set. 809 * 810 * @param regexp The compiled regular expression. 811 * @param index The position (native) in the text at which a 812 * uregex_findNext() should begin searching. 813 * @param status A reference to a UErrorCode to receive any errors. 814 * @stable ICU 4.6 815 */ 816 U_STABLE void U_EXPORT2 817 uregex_reset64(URegularExpression *regexp, 818 int64_t index, 819 UErrorCode *status); 820 821 /** 822 * Sets the limits of the matching region for this URegularExpression. 823 * The region is the part of the input string that will be considered when matching. 824 * Invoking this method resets any saved state from the previous match, 825 * then sets the region to start at the index specified by the start parameter 826 * and end at the index specified by the end parameter. 827 * 828 * Depending on the transparency and anchoring being used (see useTransparentBounds 829 * and useAnchoringBounds), certain constructs such as anchors may behave differently 830 * at or around the boundaries of the region 831 * 832 * The function will fail if start is greater than limit, or if either index 833 * is less than zero or greater than the length of the string being matched. 834 * 835 * @param regexp The compiled regular expression. 836 * @param regionStart The (native) index to begin searches at. 837 * @param regionLimit The (native) index to end searches at (exclusive). 838 * @param status A pointer to a UErrorCode to receive any errors. 839 * @stable ICU 4.0 840 */ 841 U_STABLE void U_EXPORT2 842 uregex_setRegion(URegularExpression *regexp, 843 int32_t regionStart, 844 int32_t regionLimit, 845 UErrorCode *status); 846 847 /** 848 * 64bit version of uregex_setRegion. 849 * Sets the limits of the matching region for this URegularExpression. 850 * The region is the part of the input string that will be considered when matching. 851 * Invoking this method resets any saved state from the previous match, 852 * then sets the region to start at the index specified by the start parameter 853 * and end at the index specified by the end parameter. 854 * 855 * Depending on the transparency and anchoring being used (see useTransparentBounds 856 * and useAnchoringBounds), certain constructs such as anchors may behave differently 857 * at or around the boundaries of the region 858 * 859 * The function will fail if start is greater than limit, or if either index 860 * is less than zero or greater than the length of the string being matched. 861 * 862 * @param regexp The compiled regular expression. 863 * @param regionStart The (native) index to begin searches at. 864 * @param regionLimit The (native) index to end searches at (exclusive). 865 * @param status A pointer to a UErrorCode to receive any errors. 866 * @stable ICU 4.6 867 */ 868 U_STABLE void U_EXPORT2 869 uregex_setRegion64(URegularExpression *regexp, 870 int64_t regionStart, 871 int64_t regionLimit, 872 UErrorCode *status); 873 874 /** 875 * Set the matching region and the starting index for subsequent matches 876 * in a single operation. 877 * This is useful because the usual function for setting the starting 878 * index, urgex_reset(), also resets any region limits. 879 * 880 * @param regexp The compiled regular expression. 881 * @param regionStart The (native) index to begin searches at. 882 * @param regionLimit The (native) index to end searches at (exclusive). 883 * @param startIndex The index in the input text at which the next 884 * match operation should begin. 885 * @param status A pointer to a UErrorCode to receive any errors. 886 * @stable ICU 4.6 887 */ 888 U_STABLE void U_EXPORT2 889 uregex_setRegionAndStart(URegularExpression *regexp, 890 int64_t regionStart, 891 int64_t regionLimit, 892 int64_t startIndex, 893 UErrorCode *status); 894 895 /** 896 * Reports the start index of the matching region. Any matches found are limited to 897 * to the region bounded by regionStart (inclusive) and regionEnd (exclusive). 898 * 899 * @param regexp The compiled regular expression. 900 * @param status A pointer to a UErrorCode to receive any errors. 901 * @return The starting (native) index of this matcher's region. 902 * @stable ICU 4.0 903 */ 904 U_STABLE int32_t U_EXPORT2 905 uregex_regionStart(const URegularExpression *regexp, 906 UErrorCode *status); 907 908 /** 909 * 64bit version of uregex_regionStart. 910 * Reports the start index of the matching region. Any matches found are limited to 911 * to the region bounded by regionStart (inclusive) and regionEnd (exclusive). 912 * 913 * @param regexp The compiled regular expression. 914 * @param status A pointer to a UErrorCode to receive any errors. 915 * @return The starting (native) index of this matcher's region. 916 * @stable ICU 4.6 917 */ 918 U_STABLE int64_t U_EXPORT2 919 uregex_regionStart64(const URegularExpression *regexp, 920 UErrorCode *status); 921 922 /** 923 * Reports the end index (exclusive) of the matching region for this URegularExpression. 924 * Any matches found are limited to to the region bounded by regionStart (inclusive) 925 * and regionEnd (exclusive). 926 * 927 * @param regexp The compiled regular expression. 928 * @param status A pointer to a UErrorCode to receive any errors. 929 * @return The ending point (native) of this matcher's region. 930 * @stable ICU 4.0 931 */ 932 U_STABLE int32_t U_EXPORT2 933 uregex_regionEnd(const URegularExpression *regexp, 934 UErrorCode *status); 935 936 /** 937 * 64bit version of uregex_regionEnd. 938 * Reports the end index (exclusive) of the matching region for this URegularExpression. 939 * Any matches found are limited to to the region bounded by regionStart (inclusive) 940 * and regionEnd (exclusive). 941 * 942 * @param regexp The compiled regular expression. 943 * @param status A pointer to a UErrorCode to receive any errors. 944 * @return The ending point (native) of this matcher's region. 945 * @stable ICU 4.6 946 */ 947 U_STABLE int64_t U_EXPORT2 948 uregex_regionEnd64(const URegularExpression *regexp, 949 UErrorCode *status); 950 951 /** 952 * Queries the transparency of region bounds for this URegularExpression. 953 * See useTransparentBounds for a description of transparent and opaque bounds. 954 * By default, matching boundaries are opaque. 955 * 956 * @param regexp The compiled regular expression. 957 * @param status A pointer to a UErrorCode to receive any errors. 958 * @return TRUE if this matcher is using opaque bounds, false if it is not. 959 * @stable ICU 4.0 960 */ 961 U_STABLE UBool U_EXPORT2 962 uregex_hasTransparentBounds(const URegularExpression *regexp, 963 UErrorCode *status); 964 965 966 /** 967 * Sets the transparency of region bounds for this URegularExpression. 968 * Invoking this function with an argument of TRUE will set matches to use transparent bounds. 969 * If the boolean argument is FALSE, then opaque bounds will be used. 970 * 971 * Using transparent bounds, the boundaries of the matching region are transparent 972 * to lookahead, lookbehind, and boundary matching constructs. Those constructs can 973 * see text beyond the boundaries of the region while checking for a match. 974 * 975 * With opaque bounds, no text outside of the matching region is visible to lookahead, 976 * lookbehind, and boundary matching constructs. 977 * 978 * By default, opaque bounds are used. 979 * 980 * @param regexp The compiled regular expression. 981 * @param b TRUE for transparent bounds; FALSE for opaque bounds 982 * @param status A pointer to a UErrorCode to receive any errors. 983 * @stable ICU 4.0 984 **/ 985 U_STABLE void U_EXPORT2 986 uregex_useTransparentBounds(URegularExpression *regexp, 987 UBool b, 988 UErrorCode *status); 989 990 991 /** 992 * Return true if this URegularExpression is using anchoring bounds. 993 * By default, anchoring region bounds are used. 994 * 995 * @param regexp The compiled regular expression. 996 * @param status A pointer to a UErrorCode to receive any errors. 997 * @return TRUE if this matcher is using anchoring bounds. 998 * @stable ICU 4.0 999 */ 1000 U_STABLE UBool U_EXPORT2 1001 uregex_hasAnchoringBounds(const URegularExpression *regexp, 1002 UErrorCode *status); 1003 1004 1005 /** 1006 * Set whether this URegularExpression is using Anchoring Bounds for its region. 1007 * With anchoring bounds, pattern anchors such as ^ and $ will match at the start 1008 * and end of the region. Without Anchoring Bounds, anchors will only match at 1009 * the positions they would in the complete text. 1010 * 1011 * Anchoring Bounds are the default for regions. 1012 * 1013 * @param regexp The compiled regular expression. 1014 * @param b TRUE if to enable anchoring bounds; FALSE to disable them. 1015 * @param status A pointer to a UErrorCode to receive any errors. 1016 * @stable ICU 4.0 1017 */ 1018 U_STABLE void U_EXPORT2 1019 uregex_useAnchoringBounds(URegularExpression *regexp, 1020 UBool b, 1021 UErrorCode *status); 1022 1023 /** 1024 * Return TRUE if the most recent matching operation touched the 1025 * end of the text being processed. In this case, additional input text could 1026 * change the results of that match. 1027 * 1028 * @param regexp The compiled regular expression. 1029 * @param status A pointer to a UErrorCode to receive any errors. 1030 * @return TRUE if the most recent match hit the end of input 1031 * @stable ICU 4.0 1032 */ 1033 U_STABLE UBool U_EXPORT2 1034 uregex_hitEnd(const URegularExpression *regexp, 1035 UErrorCode *status); 1036 1037 /** 1038 * Return TRUE the most recent match succeeded and additional input could cause 1039 * it to fail. If this function returns false and a match was found, then more input 1040 * might change the match but the match won't be lost. If a match was not found, 1041 * then requireEnd has no meaning. 1042 * 1043 * @param regexp The compiled regular expression. 1044 * @param status A pointer to a UErrorCode to receive any errors. 1045 * @return TRUE if more input could cause the most recent match to no longer match. 1046 * @stable ICU 4.0 1047 */ 1048 U_STABLE UBool U_EXPORT2 1049 uregex_requireEnd(const URegularExpression *regexp, 1050 UErrorCode *status); 1051 1052 1053 1054 1055 1056 /** 1057 * Replaces every substring of the input that matches the pattern 1058 * with the given replacement string. This is a convenience function that 1059 * provides a complete find-and-replace-all operation. 1060 * 1061 * This method scans the input string looking for matches of the pattern. 1062 * Input that is not part of any match is copied unchanged to the 1063 * destination buffer. Matched regions are replaced in the output 1064 * buffer by the replacement string. The replacement string may contain 1065 * references to capture groups; these take the form of $1, $2, etc. 1066 * 1067 * @param regexp The compiled regular expression. 1068 * @param replacementText A string containing the replacement text. 1069 * @param replacementLength The length of the replacement string, or 1070 * -1 if it is NUL terminated. 1071 * @param destBuf A (UChar *) buffer that will receive the result. 1072 * @param destCapacity The capacity of the destination buffer. 1073 * @param status A reference to a UErrorCode to receive any errors. 1074 * @return The length of the string resulting from the find 1075 * and replace operation. In the event that the 1076 * destination capacity is inadequate, the return value 1077 * is still the full length of the untruncated string. 1078 * @stable ICU 3.0 1079 */ 1080 U_STABLE int32_t U_EXPORT2 1081 uregex_replaceAll(URegularExpression *regexp, 1082 const UChar *replacementText, 1083 int32_t replacementLength, 1084 UChar *destBuf, 1085 int32_t destCapacity, 1086 UErrorCode *status); 1087 1088 /** 1089 * Replaces every substring of the input that matches the pattern 1090 * with the given replacement string. This is a convenience function that 1091 * provides a complete find-and-replace-all operation. 1092 * 1093 * This method scans the input string looking for matches of the pattern. 1094 * Input that is not part of any match is copied unchanged to the 1095 * destination buffer. Matched regions are replaced in the output 1096 * buffer by the replacement string. The replacement string may contain 1097 * references to capture groups; these take the form of $1, $2, etc. 1098 * 1099 * @param regexp The compiled regular expression. 1100 * @param replacement A string containing the replacement text. 1101 * @param dest A mutable UText that will receive the result. 1102 * If NULL, a new UText will be created (which may not be mutable). 1103 * @param status A reference to a UErrorCode to receive any errors. 1104 * @return A UText containing the results of the find and replace. 1105 * If a pre-allocated UText was provided, it will always be used and returned. 1106 * 1107 * @stable ICU 4.6 1108 */ 1109 U_STABLE UText * U_EXPORT2 1110 uregex_replaceAllUText(URegularExpression *regexp, 1111 UText *replacement, 1112 UText *dest, 1113 UErrorCode *status); 1114 1115 /** 1116 * Replaces the first substring of the input that matches the pattern 1117 * with the given replacement string. This is a convenience function that 1118 * provides a complete find-and-replace operation. 1119 * 1120 * This method scans the input string looking for a match of the pattern. 1121 * All input that is not part of the match is copied unchanged to the 1122 * destination buffer. The matched region is replaced in the output 1123 * buffer by the replacement string. The replacement string may contain 1124 * references to capture groups; these take the form of $1, $2, etc. 1125 * 1126 * @param regexp The compiled regular expression. 1127 * @param replacementText A string containing the replacement text. 1128 * @param replacementLength The length of the replacement string, or 1129 * -1 if it is NUL terminated. 1130 * @param destBuf A (UChar *) buffer that will receive the result. 1131 * @param destCapacity The capacity of the destination buffer. 1132 * @param status a reference to a UErrorCode to receive any errors. 1133 * @return The length of the string resulting from the find 1134 * and replace operation. In the event that the 1135 * destination capacity is inadequate, the return value 1136 * is still the full length of the untruncated string. 1137 * @stable ICU 3.0 1138 */ 1139 U_STABLE int32_t U_EXPORT2 1140 uregex_replaceFirst(URegularExpression *regexp, 1141 const UChar *replacementText, 1142 int32_t replacementLength, 1143 UChar *destBuf, 1144 int32_t destCapacity, 1145 UErrorCode *status); 1146 1147 /** 1148 * Replaces the first substring of the input that matches the pattern 1149 * with the given replacement string. This is a convenience function that 1150 * provides a complete find-and-replace operation. 1151 * 1152 * This method scans the input string looking for a match of the pattern. 1153 * All input that is not part of the match is copied unchanged to the 1154 * destination buffer. The matched region is replaced in the output 1155 * buffer by the replacement string. The replacement string may contain 1156 * references to capture groups; these take the form of $1, $2, etc. 1157 * 1158 * @param regexp The compiled regular expression. 1159 * @param replacement A string containing the replacement text. 1160 * @param dest A mutable UText that will receive the result. 1161 * If NULL, a new UText will be created (which may not be mutable). 1162 * @param status A reference to a UErrorCode to receive any errors. 1163 * @return A UText containing the results of the find and replace. 1164 * If a pre-allocated UText was provided, it will always be used and returned. 1165 * 1166 * @stable ICU 4.6 1167 */ 1168 U_STABLE UText * U_EXPORT2 1169 uregex_replaceFirstUText(URegularExpression *regexp, 1170 UText *replacement, 1171 UText *dest, 1172 UErrorCode *status); 1173 1174 /** 1175 * Implements a replace operation intended to be used as part of an 1176 * incremental find-and-replace. 1177 * 1178 * <p>The input string, starting from the end of the previous match and ending at 1179 * the start of the current match, is appended to the destination string. Then the 1180 * replacement string is appended to the output string, 1181 * including handling any substitutions of captured text.</p> 1182 * 1183 * <p>A note on preflight computation of buffersize and error handling: 1184 * Calls to uregex_appendReplacement() and uregex_appendTail() are 1185 * designed to be chained, one after another, with the destination 1186 * buffer pointer and buffer capacity updated after each in preparation 1187 * to for the next. If the destination buffer is exhausted partway through such a 1188 * sequence, a U_BUFFER_OVERFLOW_ERROR status will be returned. Normal 1189 * ICU conventions are for a function to perform no action if it is 1190 * called with an error status, but for this one case, uregex_appendRepacement() 1191 * will operate normally so that buffer size computations will complete 1192 * correctly. 1193 * 1194 * <p>For simple, prepackaged, non-incremental find-and-replace 1195 * operations, see replaceFirst() or replaceAll().</p> 1196 * 1197 * @param regexp The regular expression object. 1198 * @param replacementText The string that will replace the matched portion of the 1199 * input string as it is copied to the destination buffer. 1200 * The replacement text may contain references ($1, for 1201 * example) to capture groups from the match. 1202 * @param replacementLength The length of the replacement text string, 1203 * or -1 if the string is NUL terminated. 1204 * @param destBuf The buffer into which the results of the 1205 * find-and-replace are placed. On return, this pointer 1206 * will be updated to refer to the beginning of the 1207 * unused portion of buffer, leaving it in position for 1208 * a subsequent call to this function. 1209 * @param destCapacity The size of the output buffer, On return, this 1210 * parameter will be updated to reflect the space remaining 1211 * unused in the output buffer. 1212 * @param status A reference to a UErrorCode to receive any errors. 1213 * @return The length of the result string. In the event that 1214 * destCapacity is inadequate, the full length of the 1215 * untruncated output string is returned. 1216 * 1217 * @stable ICU 3.0 1218 * 1219 */ 1220 U_STABLE int32_t U_EXPORT2 1221 uregex_appendReplacement(URegularExpression *regexp, 1222 const UChar *replacementText, 1223 int32_t replacementLength, 1224 UChar **destBuf, 1225 int32_t *destCapacity, 1226 UErrorCode *status); 1227 1228 /** 1229 * Implements a replace operation intended to be used as part of an 1230 * incremental find-and-replace. 1231 * 1232 * <p>The input string, starting from the end of the previous match and ending at 1233 * the start of the current match, is appended to the destination string. Then the 1234 * replacement string is appended to the output string, 1235 * including handling any substitutions of captured text.</p> 1236 * 1237 * <p>For simple, prepackaged, non-incremental find-and-replace 1238 * operations, see replaceFirst() or replaceAll().</p> 1239 * 1240 * @param regexp The regular expression object. 1241 * @param replacementText The string that will replace the matched portion of the 1242 * input string as it is copied to the destination buffer. 1243 * The replacement text may contain references ($1, for 1244 * example) to capture groups from the match. 1245 * @param dest A mutable UText that will receive the result. Must not be NULL. 1246 * @param status A reference to a UErrorCode to receive any errors. 1247 * 1248 * @stable ICU 4.6 1249 */ 1250 U_STABLE void U_EXPORT2 1251 uregex_appendReplacementUText(URegularExpression *regexp, 1252 UText *replacementText, 1253 UText *dest, 1254 UErrorCode *status); 1255 1256 /** 1257 * As the final step in a find-and-replace operation, append the remainder 1258 * of the input string, starting at the position following the last match, 1259 * to the destination string. <code>uregex_appendTail()</code> is intended 1260 * to be invoked after one or more invocations of the 1261 * <code>uregex_appendReplacement()</code> function. 1262 * 1263 * @param regexp The regular expression object. This is needed to 1264 * obtain the input string and with the position 1265 * of the last match within it. 1266 * @param destBuf The buffer in which the results of the 1267 * find-and-replace are placed. On return, the pointer 1268 * will be updated to refer to the beginning of the 1269 * unused portion of buffer. 1270 * @param destCapacity The size of the output buffer, On return, this 1271 * value will be updated to reflect the space remaining 1272 * unused in the output buffer. 1273 * @param status A reference to a UErrorCode to receive any errors. 1274 * @return The length of the result string. In the event that 1275 * destCapacity is inadequate, the full length of the 1276 * untruncated output string is returned. 1277 * 1278 * @stable ICU 3.0 1279 */ 1280 U_STABLE int32_t U_EXPORT2 1281 uregex_appendTail(URegularExpression *regexp, 1282 UChar **destBuf, 1283 int32_t *destCapacity, 1284 UErrorCode *status); 1285 1286 /** 1287 * As the final step in a find-and-replace operation, append the remainder 1288 * of the input string, starting at the position following the last match, 1289 * to the destination string. <code>uregex_appendTailUText()</code> is intended 1290 * to be invoked after one or more invocations of the 1291 * <code>uregex_appendReplacementUText()</code> function. 1292 * 1293 * @param regexp The regular expression object. This is needed to 1294 * obtain the input string and with the position 1295 * of the last match within it. 1296 * @param dest A mutable UText that will receive the result. Must not be NULL. 1297 * 1298 * @param status Error code 1299 * 1300 * @return The destination UText. 1301 * 1302 * @stable ICU 4.6 1303 */ 1304 U_STABLE UText * U_EXPORT2 1305 uregex_appendTailUText(URegularExpression *regexp, 1306 UText *dest, 1307 UErrorCode *status); 1308 1309 /** 1310 * Split a string into fields. Somewhat like split() from Perl. 1311 * The pattern matches identify delimiters that separate the input 1312 * into fields. The input data between the matches becomes the 1313 * fields themselves. 1314 * 1315 * Each of the fields is copied from the input string to the destination 1316 * buffer, and NUL terminated. The position of each field within 1317 * the destination buffer is returned in the destFields array. 1318 * 1319 * If the delimiter pattern includes capture groups, the captured text will 1320 * also appear in the destination array of output strings, interspersed 1321 * with the fields. This is similar to Perl, but differs from Java, 1322 * which ignores the presence of capture groups in the pattern. 1323 * 1324 * Trailing empty fields will always be returned, assuming sufficient 1325 * destination capacity. This differs from the default behavior for Java 1326 * and Perl where trailing empty fields are not returned. 1327 * 1328 * The number of strings produced by the split operation is returned. 1329 * This count includes the strings from capture groups in the delimiter pattern. 1330 * This behavior differs from Java, which ignores capture groups. 1331 * 1332 * @param regexp The compiled regular expression. 1333 * @param destBuf A (UChar *) buffer to receive the fields that 1334 * are extracted from the input string. These 1335 * field pointers will refer to positions within the 1336 * destination buffer supplied by the caller. Any 1337 * extra positions within the destFields array will be 1338 * set to NULL. 1339 * @param destCapacity The capacity of the destBuf. 1340 * @param requiredCapacity The actual capacity required of the destBuf. 1341 * If destCapacity is too small, requiredCapacity will return 1342 * the total capacity required to hold all of the output, and 1343 * a U_BUFFER_OVERFLOW_ERROR will be returned. 1344 * @param destFields An array to be filled with the position of each 1345 * of the extracted fields within destBuf. 1346 * @param destFieldsCapacity The number of elements in the destFields array. 1347 * If the number of fields found is less than destFieldsCapacity, 1348 * the extra destFields elements are set to zero. 1349 * If destFieldsCapacity is too small, the trailing part of the 1350 * input, including any field delimiters, is treated as if it 1351 * were the last field - it is copied to the destBuf, and 1352 * its position is in the destBuf is stored in the last element 1353 * of destFields. This behavior mimics that of Perl. It is not 1354 * an error condition, and no error status is returned when all destField 1355 * positions are used. 1356 * @param status A reference to a UErrorCode to receive any errors. 1357 * @return The number of fields into which the input string was split. 1358 * @stable ICU 3.0 1359 */ 1360 U_STABLE int32_t U_EXPORT2 1361 uregex_split( URegularExpression *regexp, 1362 UChar *destBuf, 1363 int32_t destCapacity, 1364 int32_t *requiredCapacity, 1365 UChar *destFields[], 1366 int32_t destFieldsCapacity, 1367 UErrorCode *status); 1368 1369 /** 1370 * Split a string into fields. Somewhat like split() from Perl. 1371 * The pattern matches identify delimiters that separate the input 1372 * into fields. The input data between the matches becomes the 1373 * fields themselves. 1374 * <p> 1375 * The behavior of this function is not very closely aligned with uregex_split(); 1376 * instead, it is based on (and implemented directly on top of) the C++ split method. 1377 * 1378 * @param regexp The compiled regular expression. 1379 * @param destFields An array of mutable UText structs to receive the results of the split. 1380 * If a field is NULL, a new UText is allocated to contain the results for 1381 * that field. This new UText is not guaranteed to be mutable. 1382 * @param destFieldsCapacity The number of elements in the destination array. 1383 * If the number of fields found is less than destCapacity, the 1384 * extra strings in the destination array are not altered. 1385 * If the number of destination strings is less than the number 1386 * of fields, the trailing part of the input string, including any 1387 * field delimiters, is placed in the last destination string. 1388 * This behavior mimics that of Perl. It is not an error condition, and no 1389 * error status is returned when all destField positions are used. 1390 * @param status A reference to a UErrorCode to receive any errors. 1391 * @return The number of fields into which the input string was split. 1392 * 1393 * @stable ICU 4.6 1394 */ 1395 U_STABLE int32_t U_EXPORT2 1396 uregex_splitUText(URegularExpression *regexp, 1397 UText *destFields[], 1398 int32_t destFieldsCapacity, 1399 UErrorCode *status); 1400 1401 /** 1402 * Set a processing time limit for match operations with this URegularExpression. 1403 * 1404 * Some patterns, when matching certain strings, can run in exponential time. 1405 * For practical purposes, the match operation may appear to be in an 1406 * infinite loop. 1407 * When a limit is set a match operation will fail with an error if the 1408 * limit is exceeded. 1409 * <p> 1410 * The units of the limit are steps of the match engine. 1411 * Correspondence with actual processor time will depend on the speed 1412 * of the processor and the details of the specific pattern, but will 1413 * typically be on the order of milliseconds. 1414 * <p> 1415 * By default, the matching time is not limited. 1416 * <p> 1417 * 1418 * @param regexp The compiled regular expression. 1419 * @param limit The limit value, or 0 for no limit. 1420 * @param status A reference to a UErrorCode to receive any errors. 1421 * @stable ICU 4.0 1422 */ 1423 U_STABLE void U_EXPORT2 1424 uregex_setTimeLimit(URegularExpression *regexp, 1425 int32_t limit, 1426 UErrorCode *status); 1427 1428 /** 1429 * Get the time limit for for matches with this URegularExpression. 1430 * A return value of zero indicates that there is no limit. 1431 * 1432 * @param regexp The compiled regular expression. 1433 * @param status A reference to a UErrorCode to receive any errors. 1434 * @return the maximum allowed time for a match, in units of processing steps. 1435 * @stable ICU 4.0 1436 */ 1437 U_STABLE int32_t U_EXPORT2 1438 uregex_getTimeLimit(const URegularExpression *regexp, 1439 UErrorCode *status); 1440 1441 /** 1442 * Set the amount of heap storage available for use by the match backtracking stack. 1443 * <p> 1444 * ICU uses a backtracking regular expression engine, with the backtrack stack 1445 * maintained on the heap. This function sets the limit to the amount of memory 1446 * that can be used for this purpose. A backtracking stack overflow will 1447 * result in an error from the match operation that caused it. 1448 * <p> 1449 * A limit is desirable because a malicious or poorly designed pattern can use 1450 * excessive memory, potentially crashing the process. A limit is enabled 1451 * by default. 1452 * <p> 1453 * @param regexp The compiled regular expression. 1454 * @param limit The maximum size, in bytes, of the matching backtrack stack. 1455 * A value of zero means no limit. 1456 * The limit must be greater than or equal to zero. 1457 * @param status A reference to a UErrorCode to receive any errors. 1458 * 1459 * @stable ICU 4.0 1460 */ 1461 U_STABLE void U_EXPORT2 1462 uregex_setStackLimit(URegularExpression *regexp, 1463 int32_t limit, 1464 UErrorCode *status); 1465 1466 /** 1467 * Get the size of the heap storage available for use by the back tracking stack. 1468 * 1469 * @return the maximum backtracking stack size, in bytes, or zero if the 1470 * stack size is unlimited. 1471 * @stable ICU 4.0 1472 */ 1473 U_STABLE int32_t U_EXPORT2 1474 uregex_getStackLimit(const URegularExpression *regexp, 1475 UErrorCode *status); 1476 1477 1478 /** 1479 * Function pointer for a regular expression matching callback function. 1480 * When set, a callback function will be called periodically during matching 1481 * operations. If the call back function returns FALSE, the matching 1482 * operation will be terminated early. 1483 * 1484 * Note: the callback function must not call other functions on this 1485 * URegularExpression. 1486 * 1487 * @param context context pointer. The callback function will be invoked 1488 * with the context specified at the time that 1489 * uregex_setMatchCallback() is called. 1490 * @param steps the accumulated processing time, in match steps, 1491 * for this matching operation. 1492 * @return TRUE to continue the matching operation. 1493 * FALSE to terminate the matching operation. 1494 * @stable ICU 4.0 1495 */ 1496 U_CDECL_BEGIN 1497 typedef UBool U_CALLCONV URegexMatchCallback ( 1498 const void *context, 1499 int32_t steps); 1500 U_CDECL_END 1501 1502 /** 1503 * Set a callback function for this URegularExpression. 1504 * During matching operations the function will be called periodically, 1505 * giving the application the opportunity to terminate a long-running 1506 * match. 1507 * 1508 * @param regexp The compiled regular expression. 1509 * @param callback A pointer to the user-supplied callback function. 1510 * @param context User context pointer. The value supplied at the 1511 * time the callback function is set will be saved 1512 * and passed to the callback each time that it is called. 1513 * @param status A reference to a UErrorCode to receive any errors. 1514 * @stable ICU 4.0 1515 */ 1516 U_STABLE void U_EXPORT2 1517 uregex_setMatchCallback(URegularExpression *regexp, 1518 URegexMatchCallback *callback, 1519 const void *context, 1520 UErrorCode *status); 1521 1522 1523 /** 1524 * Get the callback function for this URegularExpression. 1525 * 1526 * @param regexp The compiled regular expression. 1527 * @param callback Out parameter, receives a pointer to the user-supplied 1528 * callback function. 1529 * @param context Out parameter, receives the user context pointer that 1530 * was set when uregex_setMatchCallback() was called. 1531 * @param status A reference to a UErrorCode to receive any errors. 1532 * @stable ICU 4.0 1533 */ 1534 U_STABLE void U_EXPORT2 1535 uregex_getMatchCallback(const URegularExpression *regexp, 1536 URegexMatchCallback **callback, 1537 const void **context, 1538 UErrorCode *status); 1539 1540 /** 1541 * Function pointer for a regular expression find callback function. 1542 * 1543 * When set, a callback function will be called during a find operation 1544 * and for operations that depend on find, such as findNext, split and some replace 1545 * operations like replaceFirst. 1546 * The callback will usually be called after each attempt at a match, but this is not a 1547 * guarantee that the callback will be invoked at each character. For finds where the 1548 * match engine is invoked at each character, this may be close to true, but less likely 1549 * for more optimized loops where the pattern is known to only start, and the match 1550 * engine invoked, at certain characters. 1551 * When invoked, this callback will specify the index at which a match operation is about 1552 * to be attempted, giving the application the opportunity to terminate a long-running 1553 * find operation. 1554 * 1555 * If the call back function returns FALSE, the find operation will be terminated early. 1556 * 1557 * Note: the callback function must not call other functions on this 1558 * URegularExpression 1559 * 1560 * @param context context pointer. The callback function will be invoked 1561 * with the context specified at the time that 1562 * uregex_setFindProgressCallback() is called. 1563 * @param matchIndex the next index at which a match attempt will be attempted for this 1564 * find operation. If this callback interrupts the search, this is the 1565 * index at which a find/findNext operation may be re-initiated. 1566 * @return TRUE to continue the matching operation. 1567 * FALSE to terminate the matching operation. 1568 * @stable ICU 4.6 1569 */ 1570 U_CDECL_BEGIN 1571 typedef UBool U_CALLCONV URegexFindProgressCallback ( 1572 const void *context, 1573 int64_t matchIndex); 1574 U_CDECL_END 1575 1576 1577 /** 1578 * Set the find progress callback function for this URegularExpression. 1579 * 1580 * @param regexp The compiled regular expression. 1581 * @param callback A pointer to the user-supplied callback function. 1582 * @param context User context pointer. The value supplied at the 1583 * time the callback function is set will be saved 1584 * and passed to the callback each time that it is called. 1585 * @param status A reference to a UErrorCode to receive any errors. 1586 * @stable ICU 4.6 1587 */ 1588 U_STABLE void U_EXPORT2 1589 uregex_setFindProgressCallback(URegularExpression *regexp, 1590 URegexFindProgressCallback *callback, 1591 const void *context, 1592 UErrorCode *status); 1593 1594 /** 1595 * Get the find progress callback function for this URegularExpression. 1596 * 1597 * @param regexp The compiled regular expression. 1598 * @param callback Out parameter, receives a pointer to the user-supplied 1599 * callback function. 1600 * @param context Out parameter, receives the user context pointer that 1601 * was set when uregex_setFindProgressCallback() was called. 1602 * @param status A reference to a UErrorCode to receive any errors. 1603 * @stable ICU 4.6 1604 */ 1605 U_STABLE void U_EXPORT2 1606 uregex_getFindProgressCallback(const URegularExpression *regexp, 1607 URegexFindProgressCallback **callback, 1608 const void **context, 1609 UErrorCode *status); 1610 1611 #endif /* !UCONFIG_NO_REGULAR_EXPRESSIONS */ 1612 #endif /* UREGEX_H */ 1613