usearch.h (40153B)
1 // © 2016 and later: Unicode, Inc. and others. 2 // License & terms of use: http://www.unicode.org/copyright.html 3 /* 4 ********************************************************************** 5 * Copyright (C) 2001-2011,2014 IBM and others. All rights reserved. 6 ********************************************************************** 7 * Date Name Description 8 * 06/28/2001 synwee Creation. 9 ********************************************************************** 10 */ 11 #ifndef USEARCH_H 12 #define USEARCH_H 13 14 #include "unicode/utypes.h" 15 16 #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION 17 18 #include "unicode/ucol.h" 19 #include "unicode/ucoleitr.h" 20 #include "unicode/ubrk.h" 21 22 #if U_SHOW_CPLUSPLUS_API 23 #include "unicode/localpointer.h" 24 #endif // U_SHOW_CPLUSPLUS_API 25 26 /** 27 * \file 28 * \brief C API: StringSearch 29 * 30 * C APIs for an engine that provides language-sensitive text searching based 31 * on the comparison rules defined in a <code>UCollator</code> data struct, 32 * see <code>ucol.h</code>. This ensures that language eccentricity can be 33 * handled, e.g. for the German collator, characters ß and SS will be matched 34 * if case is chosen to be ignored. 35 * See the <a href="https://htmlpreview.github.io/?https://github.com/unicode-org/icu-docs/blob/main/design/collation/ICU_collation_design.htm"> 36 * "ICU Collation Design Document"</a> for more information. 37 * <p> 38 * As of ICU4C 4.0 / ICU4J 53, the implementation uses a linear search. In previous versions, 39 * a modified form of the Boyer-Moore searching algorithm was used. For more information 40 * on the modified Boyer-Moore algorithm see 41 * <a href="http://icu-project.org/docs/papers/efficient_text_searching_in_java.html"> 42 * "Efficient Text Searching in Java"</a>, published in <i>Java Report</i> 43 * in February, 1999. 44 * <p> 45 * There are 2 match options for selection:<br> 46 * Let S' be the sub-string of a text string S between the offsets start and 47 * end <start, end>. 48 * <br> 49 * A pattern string P matches a text string S at the offsets <start, end> 50 * if 51 * <pre> 52 * option 1. Some canonical equivalent of P matches some canonical equivalent 53 * of S' 54 * option 2. P matches S' and if P starts or ends with a combining mark, 55 * there exists no non-ignorable combining mark before or after S' 56 * in S respectively. 57 * </pre> 58 * Option 2. will be the default. 59 * <p> 60 * This search has APIs similar to that of other text iteration mechanisms 61 * such as the break iterators in <code>ubrk.h</code>. Using these 62 * APIs, it is easy to scan through text looking for all occurrences of 63 * a given pattern. This search iterator allows changing of direction by 64 * calling a <code>reset</code> followed by a <code>next</code> or <code>previous</code>. 65 * Though a direction change can occur without calling <code>reset</code> first, 66 * this operation comes with some speed penalty. 67 * Generally, match results in the forward direction will match the result 68 * matches in the backwards direction in the reverse order 69 * <p> 70 * <code>usearch.h</code> provides APIs to specify the starting position 71 * within the text string to be searched, e.g. <code>usearch_setOffset</code>, 72 * <code>usearch_preceding</code> and <code>usearch_following</code>. Since the 73 * starting position will be set as it is specified, please take note that 74 * there are some dangerous positions which the search may render incorrect 75 * results: 76 * <ul> 77 * <li> The midst of a substring that requires normalization. 78 * <li> If the following match is to be found, the position should not be the 79 * second character which requires to be swapped with the preceding 80 * character. Vice versa, if the preceding match is to be found, 81 * position to search from should not be the first character which 82 * requires to be swapped with the next character. E.g certain Thai and 83 * Lao characters require swapping. 84 * <li> If a following pattern match is to be found, any position within a 85 * contracting sequence except the first will fail. Vice versa if a 86 * preceding pattern match is to be found, a invalid starting point 87 * would be any character within a contracting sequence except the last. 88 * </ul> 89 * <p> 90 * A breakiterator can be used if only matches at logical breaks are desired. 91 * Using a breakiterator will only give you results that exactly matches the 92 * boundaries given by the breakiterator. For instance the pattern "e" will 93 * not be found in the string "\u00e9" if a character break iterator is used. 94 * <p> 95 * Options are provided to handle overlapping matches. 96 * E.g. In English, overlapping matches produces the result 0 and 2 97 * for the pattern "abab" in the text "ababab", where else mutually 98 * exclusive matches only produce the result of 0. 99 * <p> 100 * Options are also provided to implement "asymmetric search" as described in 101 * <a href="http://www.unicode.org/reports/tr10/#Asymmetric_Search"> 102 * UTS #10 Unicode Collation Algorithm</a>, specifically the USearchAttribute 103 * USEARCH_ELEMENT_COMPARISON and its values. 104 * <p> 105 * Though collator attributes will be taken into consideration while 106 * performing matches, there are no APIs here for setting and getting the 107 * attributes. These attributes can be set by getting the collator 108 * from <code>usearch_getCollator</code> and using the APIs in <code>ucol.h</code>. 109 * Lastly to update String Search to the new collator attributes, 110 * usearch_reset() has to be called. 111 * <p> 112 * Restriction: <br> 113 * Currently there are no composite characters that consists of a 114 * character with combining class > 0 before a character with combining 115 * class == 0. However, if such a character exists in the future, the 116 * search mechanism does not guarantee the results for option 1. 117 * 118 * <p> 119 * Example of use:<br> 120 * <pre><code> 121 * char *tgtstr = "The quick brown fox jumped over the lazy fox"; 122 * char *patstr = "fox"; 123 * UChar target[64]; 124 * UChar pattern[16]; 125 * UErrorCode status = U_ZERO_ERROR; 126 * u_uastrcpy(target, tgtstr); 127 * u_uastrcpy(pattern, patstr); 128 * 129 * UStringSearch *search = usearch_open(pattern, -1, target, -1, "en_US", 130 * NULL, &status); 131 * if (U_SUCCESS(status)) { 132 * for (int pos = usearch_first(search, &status); 133 * pos != USEARCH_DONE; 134 * pos = usearch_next(search, &status)) 135 * { 136 * printf("Found match at %d pos, length is %d\n", pos, 137 * usearch_getMatchedLength(search)); 138 * } 139 * } 140 * 141 * usearch_close(search); 142 * </code></pre> 143 * @stable ICU 2.4 144 */ 145 146 /** 147 * DONE is returned by previous() and next() after all valid matches have 148 * been returned, and by first() and last() if there are no matches at all. 149 * @stable ICU 2.4 150 */ 151 #define USEARCH_DONE -1 152 153 /** 154 * Data structure for searching 155 * @stable ICU 2.4 156 */ 157 struct UStringSearch; 158 /** 159 * Data structure for searching 160 * @stable ICU 2.4 161 */ 162 typedef struct UStringSearch UStringSearch; 163 164 /** 165 * @stable ICU 2.4 166 */ 167 typedef enum { 168 /** 169 * Option for overlapping matches 170 * @stable ICU 2.4 171 */ 172 USEARCH_OVERLAP = 0, 173 #ifndef U_HIDE_DEPRECATED_API 174 /** 175 * Option for canonical matches; option 1 in header documentation. 176 * The default value will be USEARCH_OFF. 177 * Note: Setting this option to USEARCH_ON currently has no effect on 178 * search behavior, and this option is deprecated. Instead, to control 179 * canonical match behavior, you must set UCOL_NORMALIZATION_MODE 180 * appropriately (to UCOL_OFF or UCOL_ON) in the UCollator used by 181 * the UStringSearch object. 182 * @see usearch_openFromCollator 183 * @see usearch_getCollator 184 * @see usearch_setCollator 185 * @see ucol_getAttribute 186 * @deprecated ICU 53 187 */ 188 USEARCH_CANONICAL_MATCH = 1, 189 #endif /* U_HIDE_DEPRECATED_API */ 190 /** 191 * Option to control how collation elements are compared. 192 * The default value will be USEARCH_STANDARD_ELEMENT_COMPARISON. 193 * @stable ICU 4.4 194 */ 195 USEARCH_ELEMENT_COMPARISON = 2, 196 197 #ifndef U_HIDE_DEPRECATED_API 198 /** 199 * One more than the highest normal USearchAttribute value. 200 * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420. 201 */ 202 USEARCH_ATTRIBUTE_COUNT = 3 203 #endif /* U_HIDE_DEPRECATED_API */ 204 } USearchAttribute; 205 206 /** 207 * @stable ICU 2.4 208 */ 209 typedef enum { 210 /** 211 * Default value for any USearchAttribute 212 * @stable ICU 2.4 213 */ 214 USEARCH_DEFAULT = -1, 215 /** 216 * Value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH 217 * @stable ICU 2.4 218 */ 219 USEARCH_OFF, 220 /** 221 * Value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH 222 * @stable ICU 2.4 223 */ 224 USEARCH_ON, 225 /** 226 * Value (default) for USEARCH_ELEMENT_COMPARISON; 227 * standard collation element comparison at the specified collator 228 * strength. 229 * @stable ICU 4.4 230 */ 231 USEARCH_STANDARD_ELEMENT_COMPARISON, 232 /** 233 * Value for USEARCH_ELEMENT_COMPARISON; 234 * collation element comparison is modified to effectively provide 235 * behavior between the specified strength and strength - 1. Collation 236 * elements in the pattern that have the base weight for the specified 237 * strength are treated as "wildcards" that match an element with any 238 * other weight at that collation level in the searched text. For 239 * example, with a secondary-strength English collator, a plain 'e' in 240 * the pattern will match a plain e or an e with any diacritic in the 241 * searched text, but an e with diacritic in the pattern will only 242 * match an e with the same diacritic in the searched text. 243 * 244 * This supports "asymmetric search" as described in 245 * <a href="http://www.unicode.org/reports/tr10/#Asymmetric_Search"> 246 * UTS #10 Unicode Collation Algorithm</a>. 247 * 248 * @stable ICU 4.4 249 */ 250 USEARCH_PATTERN_BASE_WEIGHT_IS_WILDCARD, 251 /** 252 * Value for USEARCH_ELEMENT_COMPARISON. 253 * collation element comparison is modified to effectively provide 254 * behavior between the specified strength and strength - 1. Collation 255 * elements in either the pattern or the searched text that have the 256 * base weight for the specified strength are treated as "wildcards" 257 * that match an element with any other weight at that collation level. 258 * For example, with a secondary-strength English collator, a plain 'e' 259 * in the pattern will match a plain e or an e with any diacritic in the 260 * searched text, but an e with diacritic in the pattern will only 261 * match an e with the same diacritic or a plain e in the searched text. 262 * 263 * This option is similar to "asymmetric search" as described in 264 * [UTS #10 Unicode Collation Algorithm](http://www.unicode.org/reports/tr10/#Asymmetric_Search), 265 * but also allows unmarked characters in the searched text to match 266 * marked or unmarked versions of that character in the pattern. 267 * 268 * @stable ICU 4.4 269 */ 270 USEARCH_ANY_BASE_WEIGHT_IS_WILDCARD, 271 272 #ifndef U_HIDE_DEPRECATED_API 273 /** 274 * One more than the highest normal USearchAttributeValue value. 275 * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420. 276 */ 277 USEARCH_ATTRIBUTE_VALUE_COUNT 278 #endif /* U_HIDE_DEPRECATED_API */ 279 } USearchAttributeValue; 280 281 /* open and close ------------------------------------------------------ */ 282 283 /** 284 * Creates a String Search iterator data struct using the argument locale language 285 * rule set. A collator will be created in the process, which will be owned by 286 * this String Search and will be deleted in <code>usearch_close</code>. 287 * 288 * The UStringSearch retains a pointer to both the pattern and text strings. 289 * The caller must not modify or delete them while using the UStringSearch. 290 * 291 * @param pattern for matching 292 * @param patternlength length of the pattern, -1 for null-termination 293 * @param text text string 294 * @param textlength length of the text string, -1 for null-termination 295 * @param locale name of locale for the rules to be used 296 * @param breakiter A BreakIterator that will be used to restrict the points 297 * at which matches are detected. If a match is found, but 298 * the match's start or end index is not a boundary as 299 * determined by the <code>BreakIterator</code>, the match will 300 * be rejected and another will be searched for. 301 * If this parameter is <code>NULL</code>, no break detection is 302 * attempted. 303 * @param status for errors if it occurs. If pattern or text is NULL, or if 304 * patternlength or textlength is 0 then an 305 * U_ILLEGAL_ARGUMENT_ERROR is returned. 306 * @return search iterator data structure, or NULL if there is an error. 307 * @stable ICU 2.4 308 */ 309 U_CAPI UStringSearch * U_EXPORT2 usearch_open(const UChar *pattern, 310 int32_t patternlength, 311 const UChar *text, 312 int32_t textlength, 313 const char *locale, 314 UBreakIterator *breakiter, 315 UErrorCode *status); 316 317 /** 318 * Creates a String Search iterator data struct using the argument collator language 319 * rule set. Note, user retains the ownership of this collator, thus the 320 * responsibility of deletion lies with the user. 321 322 * NOTE: String Search cannot be instantiated from a collator that has 323 * collate digits as numbers (CODAN) turned on (UCOL_NUMERIC_COLLATION). 324 * 325 * The UStringSearch retains a pointer to both the pattern and text strings. 326 * The caller must not modify or delete them while using the UStringSearch. 327 * 328 * @param pattern for matching 329 * @param patternlength length of the pattern, -1 for null-termination 330 * @param text text string 331 * @param textlength length of the text string, -1 for null-termination 332 * @param collator used for the language rules 333 * @param breakiter A BreakIterator that will be used to restrict the points 334 * at which matches are detected. If a match is found, but 335 * the match's start or end index is not a boundary as 336 * determined by the <code>BreakIterator</code>, the match will 337 * be rejected and another will be searched for. 338 * If this parameter is <code>NULL</code>, no break detection is 339 * attempted. 340 * @param status for errors if it occurs. If collator, pattern or text is NULL, 341 * or if patternlength or textlength is 0 then an 342 * U_ILLEGAL_ARGUMENT_ERROR is returned. 343 * @return search iterator data structure, or NULL if there is an error. 344 * @stable ICU 2.4 345 */ 346 U_CAPI UStringSearch * U_EXPORT2 usearch_openFromCollator( 347 const UChar *pattern, 348 int32_t patternlength, 349 const UChar *text, 350 int32_t textlength, 351 const UCollator *collator, 352 UBreakIterator *breakiter, 353 UErrorCode *status); 354 355 /** 356 * Destroys and cleans up the String Search iterator data struct. 357 * If a collator was created in <code>usearch_open</code>, then it will be destroyed here. 358 * @param searchiter The UStringSearch to clean up 359 * @stable ICU 2.4 360 */ 361 U_CAPI void U_EXPORT2 usearch_close(UStringSearch *searchiter); 362 363 #if U_SHOW_CPLUSPLUS_API 364 365 U_NAMESPACE_BEGIN 366 367 /** 368 * \class LocalUStringSearchPointer 369 * "Smart pointer" class, closes a UStringSearch via usearch_close(). 370 * For most methods see the LocalPointerBase base class. 371 * 372 * @see LocalPointerBase 373 * @see LocalPointer 374 * @stable ICU 4.4 375 */ 376 U_DEFINE_LOCAL_OPEN_POINTER(LocalUStringSearchPointer, UStringSearch, usearch_close); 377 378 U_NAMESPACE_END 379 380 #endif 381 382 /* get and set methods -------------------------------------------------- */ 383 384 /** 385 * Sets the current position in the text string which the next search will 386 * start from. Clears previous states. 387 * This method takes the argument index and sets the position in the text 388 * string accordingly without checking if the index is pointing to a 389 * valid starting point to begin searching. 390 * Search positions that may render incorrect results are highlighted in the 391 * header comments 392 * @param strsrch search iterator data struct 393 * @param position position to start next search from. If position is less 394 * than or greater than the text range for searching, 395 * an U_INDEX_OUTOFBOUNDS_ERROR will be returned 396 * @param status error status if any. 397 * @stable ICU 2.4 398 */ 399 U_CAPI void U_EXPORT2 usearch_setOffset(UStringSearch *strsrch, 400 int32_t position, 401 UErrorCode *status); 402 403 /** 404 * Return the current index in the string text being searched. 405 * If the iteration has gone past the end of the text (or past the beginning 406 * for a backwards search), <code>USEARCH_DONE</code> is returned. 407 * @param strsrch search iterator data struct 408 * @see #USEARCH_DONE 409 * @stable ICU 2.4 410 */ 411 U_CAPI int32_t U_EXPORT2 usearch_getOffset(const UStringSearch *strsrch); 412 413 /** 414 * Sets the text searching attributes located in the enum USearchAttribute 415 * with values from the enum USearchAttributeValue. 416 * <code>USEARCH_DEFAULT</code> can be used for all attributes for resetting. 417 * @param strsrch search iterator data struct 418 * @param attribute text attribute to be set 419 * @param value text attribute value 420 * @param status for errors if it occurs 421 * @see #usearch_getAttribute 422 * @stable ICU 2.4 423 */ 424 U_CAPI void U_EXPORT2 usearch_setAttribute(UStringSearch *strsrch, 425 USearchAttribute attribute, 426 USearchAttributeValue value, 427 UErrorCode *status); 428 429 /** 430 * Gets the text searching attributes. 431 * @param strsrch search iterator data struct 432 * @param attribute text attribute to be retrieve 433 * @return text attribute value 434 * @see #usearch_setAttribute 435 * @stable ICU 2.4 436 */ 437 U_CAPI USearchAttributeValue U_EXPORT2 usearch_getAttribute( 438 const UStringSearch *strsrch, 439 USearchAttribute attribute); 440 441 /** 442 * Returns the index to the match in the text string that was searched. 443 * This call returns a valid result only after a successful call to 444 * <code>usearch_first</code>, <code>usearch_next</code>, <code>usearch_previous</code>, 445 * or <code>usearch_last</code>. 446 * Just after construction, or after a searching method returns 447 * <code>USEARCH_DONE</code>, this method will return <code>USEARCH_DONE</code>. 448 * <p> 449 * Use <code>usearch_getMatchedLength</code> to get the matched string length. 450 * @param strsrch search iterator data struct 451 * @return index to a substring within the text string that is being 452 * searched. 453 * @see #usearch_first 454 * @see #usearch_next 455 * @see #usearch_previous 456 * @see #usearch_last 457 * @see #USEARCH_DONE 458 * @stable ICU 2.4 459 */ 460 U_CAPI int32_t U_EXPORT2 usearch_getMatchedStart( 461 const UStringSearch *strsrch); 462 463 /** 464 * Returns the length of text in the string which matches the search pattern. 465 * This call returns a valid result only after a successful call to 466 * <code>usearch_first</code>, <code>usearch_next</code>, <code>usearch_previous</code>, 467 * or <code>usearch_last</code>. 468 * Just after construction, or after a searching method returns 469 * <code>USEARCH_DONE</code>, this method will return 0. 470 * @param strsrch search iterator data struct 471 * @return The length of the match in the string text, or 0 if there is no 472 * match currently. 473 * @see #usearch_first 474 * @see #usearch_next 475 * @see #usearch_previous 476 * @see #usearch_last 477 * @see #USEARCH_DONE 478 * @stable ICU 2.4 479 */ 480 U_CAPI int32_t U_EXPORT2 usearch_getMatchedLength( 481 const UStringSearch *strsrch); 482 483 /** 484 * Returns the text that was matched by the most recent call to 485 * <code>usearch_first</code>, <code>usearch_next</code>, <code>usearch_previous</code>, 486 * or <code>usearch_last</code>. 487 * If the iterator is not pointing at a valid match (e.g. just after 488 * construction or after <code>USEARCH_DONE</code> has been returned, returns 489 * an empty string. If result is not large enough to store the matched text, 490 * result will be filled with the partial text and an U_BUFFER_OVERFLOW_ERROR 491 * will be returned in status. result will be null-terminated whenever 492 * possible. If the buffer fits the matched text exactly, a null-termination 493 * is not possible, then a U_STRING_NOT_TERMINATED_ERROR set in status. 494 * Pre-flighting can be either done with length = 0 or the API 495 * <code>usearch_getMatchedLength</code>. 496 * @param strsrch search iterator data struct 497 * @param result UChar buffer to store the matched string 498 * @param resultCapacity length of the result buffer 499 * @param status error returned if result is not large enough 500 * @return exact length of the matched text, not counting the null-termination 501 * @see #usearch_first 502 * @see #usearch_next 503 * @see #usearch_previous 504 * @see #usearch_last 505 * @see #USEARCH_DONE 506 * @stable ICU 2.4 507 */ 508 U_CAPI int32_t U_EXPORT2 usearch_getMatchedText(const UStringSearch *strsrch, 509 UChar *result, 510 int32_t resultCapacity, 511 UErrorCode *status); 512 513 #if !UCONFIG_NO_BREAK_ITERATION 514 515 /** 516 * Set the BreakIterator that will be used to restrict the points at which 517 * matches are detected. 518 * @param strsrch search iterator data struct 519 * @param breakiter A BreakIterator that will be used to restrict the points 520 * at which matches are detected. If a match is found, but 521 * the match's start or end index is not a boundary as 522 * determined by the <code>BreakIterator</code>, the match will 523 * be rejected and another will be searched for. 524 * If this parameter is <code>NULL</code>, no break detection is 525 * attempted. 526 * @param status for errors if it occurs 527 * @see #usearch_getBreakIterator 528 * @stable ICU 2.4 529 */ 530 U_CAPI void U_EXPORT2 usearch_setBreakIterator(UStringSearch *strsrch, 531 UBreakIterator *breakiter, 532 UErrorCode *status); 533 534 /** 535 * Returns the BreakIterator that is used to restrict the points at which 536 * matches are detected. This will be the same object that was passed to the 537 * constructor or to <code>usearch_setBreakIterator</code>. Note that 538 * <code>NULL</code> 539 * is a legal value; it means that break detection should not be attempted. 540 * @param strsrch search iterator data struct 541 * @return break iterator used 542 * @see #usearch_setBreakIterator 543 * @stable ICU 2.4 544 */ 545 U_CAPI const UBreakIterator * U_EXPORT2 usearch_getBreakIterator( 546 const UStringSearch *strsrch); 547 548 #endif 549 550 /** 551 * Set the string text to be searched. Text iteration will hence begin at the 552 * start of the text string. This method is useful if you want to re-use an 553 * iterator to search for the same pattern within a different body of text. 554 * 555 * The UStringSearch retains a pointer to the text string. The caller must not 556 * modify or delete the string while using the UStringSearch. 557 * 558 * @param strsrch search iterator data struct 559 * @param text new string to look for match 560 * @param textlength length of the new string, -1 for null-termination 561 * @param status for errors if it occurs. If text is NULL, or textlength is 0 562 * then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change 563 * done to strsrch. 564 * @see #usearch_getText 565 * @stable ICU 2.4 566 */ 567 U_CAPI void U_EXPORT2 usearch_setText( UStringSearch *strsrch, 568 const UChar *text, 569 int32_t textlength, 570 UErrorCode *status); 571 572 /** 573 * Return the string text to be searched. 574 * @param strsrch search iterator data struct 575 * @param length returned string text length 576 * @return string text 577 * @see #usearch_setText 578 * @stable ICU 2.4 579 */ 580 U_CAPI const UChar * U_EXPORT2 usearch_getText(const UStringSearch *strsrch, 581 int32_t *length); 582 583 /** 584 * Gets the collator used for the language rules. 585 * <p> 586 * Deleting the returned <code>UCollator</code> before calling 587 * <code>usearch_close</code> would cause the string search to fail. 588 * <code>usearch_close</code> will delete the collator if this search owns it. 589 * @param strsrch search iterator data struct 590 * @return collator 591 * @stable ICU 2.4 592 */ 593 U_CAPI UCollator * U_EXPORT2 usearch_getCollator( 594 const UStringSearch *strsrch); 595 596 /** 597 * Sets the collator used for the language rules. User retains the ownership 598 * of this collator, thus the responsibility of deletion lies with the user. 599 * This method causes internal data such as the pattern collation elements 600 * and shift tables to be recalculated, but the iterator's position is unchanged. 601 * @param strsrch search iterator data struct 602 * @param collator to be used 603 * @param status for errors if it occurs 604 * @stable ICU 2.4 605 */ 606 U_CAPI void U_EXPORT2 usearch_setCollator( UStringSearch *strsrch, 607 const UCollator *collator, 608 UErrorCode *status); 609 610 /** 611 * Sets the pattern used for matching. 612 * Internal data like the pattern collation elements will be recalculated, but the 613 * iterator's position is unchanged. 614 * 615 * The UStringSearch retains a pointer to the pattern string. The caller must not 616 * modify or delete the string while using the UStringSearch. 617 * 618 * @param strsrch search iterator data struct 619 * @param pattern string 620 * @param patternlength pattern length, -1 for null-terminated string 621 * @param status for errors if it occurs. If text is NULL, or textlength is 0 622 * then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change 623 * done to strsrch. 624 * @stable ICU 2.4 625 */ 626 U_CAPI void U_EXPORT2 usearch_setPattern( UStringSearch *strsrch, 627 const UChar *pattern, 628 int32_t patternlength, 629 UErrorCode *status); 630 631 /** 632 * Gets the search pattern 633 * @param strsrch search iterator data struct 634 * @param length return length of the pattern, -1 indicates that the pattern 635 * is null-terminated 636 * @return pattern string 637 * @stable ICU 2.4 638 */ 639 U_CAPI const UChar * U_EXPORT2 usearch_getPattern( 640 const UStringSearch *strsrch, 641 int32_t *length); 642 643 /* methods ------------------------------------------------------------- */ 644 645 /** 646 * Returns the first index at which the string text matches the search 647 * pattern. 648 * The iterator is adjusted so that its current index (as returned by 649 * <code>usearch_getOffset</code>) is the match position if one was found. 650 * If a match is not found, <code>USEARCH_DONE</code> will be returned and 651 * the iterator will be adjusted to the index <code>USEARCH_DONE</code>. 652 * @param strsrch search iterator data struct 653 * @param status for errors if it occurs 654 * @return The character index of the first match, or 655 * <code>USEARCH_DONE</code> if there are no matches. 656 * @see #usearch_getOffset 657 * @see #USEARCH_DONE 658 * @stable ICU 2.4 659 */ 660 U_CAPI int32_t U_EXPORT2 usearch_first(UStringSearch *strsrch, 661 UErrorCode *status); 662 663 /** 664 * Returns the first index equal or greater than <code>position</code> at which 665 * the string text 666 * matches the search pattern. The iterator is adjusted so that its current 667 * index (as returned by <code>usearch_getOffset</code>) is the match position if 668 * one was found. 669 * If a match is not found, <code>USEARCH_DONE</code> will be returned and 670 * the iterator will be adjusted to the index <code>USEARCH_DONE</code> 671 * <p> 672 * Search positions that may render incorrect results are highlighted in the 673 * header comments. If position is less than or greater than the text range 674 * for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned 675 * @param strsrch search iterator data struct 676 * @param position to start the search at 677 * @param status for errors if it occurs 678 * @return The character index of the first match following <code>pos</code>, 679 * or <code>USEARCH_DONE</code> if there are no matches. 680 * @see #usearch_getOffset 681 * @see #USEARCH_DONE 682 * @stable ICU 2.4 683 */ 684 U_CAPI int32_t U_EXPORT2 usearch_following(UStringSearch *strsrch, 685 int32_t position, 686 UErrorCode *status); 687 688 /** 689 * Returns the last index in the target text at which it matches the search 690 * pattern. The iterator is adjusted so that its current 691 * index (as returned by <code>usearch_getOffset</code>) is the match position if 692 * one was found. 693 * If a match is not found, <code>USEARCH_DONE</code> will be returned and 694 * the iterator will be adjusted to the index <code>USEARCH_DONE</code>. 695 * @param strsrch search iterator data struct 696 * @param status for errors if it occurs 697 * @return The index of the first match, or <code>USEARCH_DONE</code> if there 698 * are no matches. 699 * @see #usearch_getOffset 700 * @see #USEARCH_DONE 701 * @stable ICU 2.4 702 */ 703 U_CAPI int32_t U_EXPORT2 usearch_last(UStringSearch *strsrch, 704 UErrorCode *status); 705 706 /** 707 * Returns the first index less than <code>position</code> at which the string text 708 * matches the search pattern. The iterator is adjusted so that its current 709 * index (as returned by <code>usearch_getOffset</code>) is the match position if 710 * one was found. 711 * If a match is not found, <code>USEARCH_DONE</code> will be returned and 712 * the iterator will be adjusted to the index <code>USEARCH_DONE</code> 713 * <p> 714 * Search positions that may render incorrect results are highlighted in the 715 * header comments. If position is less than or greater than the text range 716 * for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned. 717 * <p> 718 * When <code>USEARCH_OVERLAP</code> option is off, the last index of the 719 * result match is always less than <code>position</code>. 720 * When <code>USERARCH_OVERLAP</code> is on, the result match may span across 721 * <code>position</code>. 722 * @param strsrch search iterator data struct 723 * @param position index position the search is to begin at 724 * @param status for errors if it occurs 725 * @return The character index of the first match preceding <code>pos</code>, 726 * or <code>USEARCH_DONE</code> if there are no matches. 727 * @see #usearch_getOffset 728 * @see #USEARCH_DONE 729 * @stable ICU 2.4 730 */ 731 U_CAPI int32_t U_EXPORT2 usearch_preceding(UStringSearch *strsrch, 732 int32_t position, 733 UErrorCode *status); 734 735 /** 736 * Returns the index of the next point at which the string text matches the 737 * search pattern, starting from the current position. 738 * The iterator is adjusted so that its current 739 * index (as returned by <code>usearch_getOffset</code>) is the match position if 740 * one was found. 741 * If a match is not found, <code>USEARCH_DONE</code> will be returned and 742 * the iterator will be adjusted to the index <code>USEARCH_DONE</code> 743 * @param strsrch search iterator data struct 744 * @param status for errors if it occurs 745 * @return The index of the next match after the current position, or 746 * <code>USEARCH_DONE</code> if there are no more matches. 747 * @see #usearch_first 748 * @see #usearch_getOffset 749 * @see #USEARCH_DONE 750 * @stable ICU 2.4 751 */ 752 U_CAPI int32_t U_EXPORT2 usearch_next(UStringSearch *strsrch, 753 UErrorCode *status); 754 755 /** 756 * Returns the index of the previous point at which the string text matches 757 * the search pattern, starting at the current position. 758 * The iterator is adjusted so that its current 759 * index (as returned by <code>usearch_getOffset</code>) is the match position if 760 * one was found. 761 * If a match is not found, <code>USEARCH_DONE</code> will be returned and 762 * the iterator will be adjusted to the index <code>USEARCH_DONE</code> 763 * @param strsrch search iterator data struct 764 * @param status for errors if it occurs 765 * @return The index of the previous match before the current position, 766 * or <code>USEARCH_DONE</code> if there are no more matches. 767 * @see #usearch_last 768 * @see #usearch_getOffset 769 * @see #USEARCH_DONE 770 * @stable ICU 2.4 771 */ 772 U_CAPI int32_t U_EXPORT2 usearch_previous(UStringSearch *strsrch, 773 UErrorCode *status); 774 775 /** 776 * Reset the iteration. 777 * Search will begin at the start of the text string if a forward iteration 778 * is initiated before a backwards iteration. Otherwise if a backwards 779 * iteration is initiated before a forwards iteration, the search will begin 780 * at the end of the text string. 781 * @param strsrch search iterator data struct 782 * @see #usearch_first 783 * @stable ICU 2.4 784 */ 785 U_CAPI void U_EXPORT2 usearch_reset(UStringSearch *strsrch); 786 787 #ifndef U_HIDE_INTERNAL_API 788 /** 789 * Simple forward search for the pattern, starting at a specified index, 790 * and using a default set search options. 791 * 792 * This is an experimental function, and is not an official part of the 793 * ICU API. 794 * 795 * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored. 796 * 797 * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and 798 * any Break Iterator are ignored. 799 * 800 * Matches obey the following constraints: 801 * 802 * Characters at the start or end positions of a match that are ignorable 803 * for collation are not included as part of the match, unless they 804 * are part of a combining sequence, as described below. 805 * 806 * A match will not include a partial combining sequence. Combining 807 * character sequences are considered to be inseparable units, 808 * and either match the pattern completely, or are considered to not match 809 * at all. Thus, for example, an A followed a combining accent mark will 810 * not be found when searching for a plain (unaccented) A. (unless 811 * the collation strength has been set to ignore all accents). 812 * 813 * When beginning a search, the initial starting position, startIdx, 814 * is assumed to be an acceptable match boundary with respect to 815 * combining characters. A combining sequence that spans across the 816 * starting point will not suppress a match beginning at startIdx. 817 * 818 * Characters that expand to multiple collation elements 819 * (German sharp-S becoming 'ss', or the composed forms of accented 820 * characters, for example) also must match completely. 821 * Searching for a single 's' in a string containing only a sharp-s will 822 * find no match. 823 * 824 * 825 * @param strsrch the UStringSearch struct, which references both 826 * the text to be searched and the pattern being sought. 827 * @param startIdx The index into the text to begin the search. 828 * @param matchStart An out parameter, the starting index of the matched text. 829 * This parameter may be NULL. 830 * A value of -1 will be returned if no match was found. 831 * @param matchLimit Out parameter, the index of the first position following the matched text. 832 * The matchLimit will be at a suitable position for beginning a subsequent search 833 * in the input text. 834 * This parameter may be NULL. 835 * A value of -1 will be returned if no match was found. 836 * 837 * @param status Report any errors. Note that no match found is not an error. 838 * @return true if a match was found, false otherwise. 839 * 840 * @internal 841 */ 842 U_CAPI UBool U_EXPORT2 usearch_search(UStringSearch *strsrch, 843 int32_t startIdx, 844 int32_t *matchStart, 845 int32_t *matchLimit, 846 UErrorCode *status); 847 848 /** 849 * Simple backwards search for the pattern, starting at a specified index, 850 * and using using a default set search options. 851 * 852 * This is an experimental function, and is not an official part of the 853 * ICU API. 854 * 855 * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored. 856 * 857 * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and 858 * any Break Iterator are ignored. 859 * 860 * Matches obey the following constraints: 861 * 862 * Characters at the start or end positions of a match that are ignorable 863 * for collation are not included as part of the match, unless they 864 * are part of a combining sequence, as described below. 865 * 866 * A match will not include a partial combining sequence. Combining 867 * character sequences are considered to be inseparable units, 868 * and either match the pattern completely, or are considered to not match 869 * at all. Thus, for example, an A followed a combining accent mark will 870 * not be found when searching for a plain (unaccented) A. (unless 871 * the collation strength has been set to ignore all accents). 872 * 873 * When beginning a search, the initial starting position, startIdx, 874 * is assumed to be an acceptable match boundary with respect to 875 * combining characters. A combining sequence that spans across the 876 * starting point will not suppress a match beginning at startIdx. 877 * 878 * Characters that expand to multiple collation elements 879 * (German sharp-S becoming 'ss', or the composed forms of accented 880 * characters, for example) also must match completely. 881 * Searching for a single 's' in a string containing only a sharp-s will 882 * find no match. 883 * 884 * 885 * @param strsrch the UStringSearch struct, which references both 886 * the text to be searched and the pattern being sought. 887 * @param startIdx The index into the text to begin the search. 888 * @param matchStart An out parameter, the starting index of the matched text. 889 * This parameter may be NULL. 890 * A value of -1 will be returned if no match was found. 891 * @param matchLimit Out parameter, the index of the first position following the matched text. 892 * The matchLimit will be at a suitable position for beginning a subsequent search 893 * in the input text. 894 * This parameter may be NULL. 895 * A value of -1 will be returned if no match was found. 896 * 897 * @param status Report any errors. Note that no match found is not an error. 898 * @return true if a match was found, false otherwise. 899 * 900 * @internal 901 */ 902 U_CAPI UBool U_EXPORT2 usearch_searchBackwards(UStringSearch *strsrch, 903 int32_t startIdx, 904 int32_t *matchStart, 905 int32_t *matchLimit, 906 UErrorCode *status); 907 #endif /* U_HIDE_INTERNAL_API */ 908 909 #endif /* #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION */ 910 911 #endif