]> git.saurik.com Git - apple/icu.git/blob - icuSources/i18n/unicode/usearch.h
ICU-400.37.tar.gz
[apple/icu.git] / icuSources / i18n / unicode / usearch.h
1 /*
2 **********************************************************************
3 * Copyright (C) 2001-2008 IBM and others. All rights reserved.
4 **********************************************************************
5 * Date Name Description
6 * 06/28/2001 synwee Creation.
7 **********************************************************************
8 */
9 #ifndef USEARCH_H
10 #define USEARCH_H
11
12 #include "unicode/utypes.h"
13
14 #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION
15
16 #include "unicode/ucol.h"
17 #include "unicode/ucoleitr.h"
18 #include "unicode/ubrk.h"
19
20 /**
21 * \file
22 * \brief C API: StringSearch
23 *
24 * C Apis for an engine that provides language-sensitive text searching based
25 * on the comparison rules defined in a <tt>UCollator</tt> data struct,
26 * see <tt>ucol.h</tt>. This ensures that language eccentricity can be
27 * handled, e.g. for the German collator, characters &szlig; and SS will be matched
28 * if case is chosen to be ignored.
29 * See the <a href="http://source.icu-project.org/repos/icu/icuhtml/trunk/design/collation/ICU_collation_design.htm">
30 * "ICU Collation Design Document"</a> for more information.
31 * <p>
32 * The algorithm implemented is a modified form of the Boyer Moore's search.
33 * For more information see
34 * <a href="http://icu-project.org/docs/papers/efficient_text_searching_in_java.html">
35 * "Efficient Text Searching in Java"</a>, published in <i>Java Report</i>
36 * in February, 1999, for further information on the algorithm.
37 * <p>
38 * There are 2 match options for selection:<br>
39 * Let S' be the sub-string of a text string S between the offsets start and
40 * end <start, end>.
41 * <br>
42 * A pattern string P matches a text string S at the offsets <start, end>
43 * if
44 * <pre>
45 * option 1. Some canonical equivalent of P matches some canonical equivalent
46 * of S'
47 * option 2. P matches S' and if P starts or ends with a combining mark,
48 * there exists no non-ignorable combining mark before or after S'
49 * in S respectively.
50 * </pre>
51 * Option 2. will be the default.
52 * <p>
53 * This search has APIs similar to that of other text iteration mechanisms
54 * such as the break iterators in <tt>ubrk.h</tt>. Using these
55 * APIs, it is easy to scan through text looking for all occurances of
56 * a given pattern. This search iterator allows changing of direction by
57 * calling a <tt>reset</tt> followed by a <tt>next</tt> or <tt>previous</tt>.
58 * Though a direction change can occur without calling <tt>reset</tt> first,
59 * this operation comes with some speed penalty.
60 * Generally, match results in the forward direction will match the result
61 * matches in the backwards direction in the reverse order
62 * <p>
63 * <tt>usearch.h</tt> provides APIs to specify the starting position
64 * within the text string to be searched, e.g. <tt>usearch_setOffset</tt>,
65 * <tt>usearch_preceding</tt> and <tt>usearch_following</tt>. Since the
66 * starting position will be set as it is specified, please take note that
67 * there are some dangerous positions which the search may render incorrect
68 * results:
69 * <ul>
70 * <li> The midst of a substring that requires normalization.
71 * <li> If the following match is to be found, the position should not be the
72 * second character which requires to be swapped with the preceding
73 * character. Vice versa, if the preceding match is to be found,
74 * position to search from should not be the first character which
75 * requires to be swapped with the next character. E.g certain Thai and
76 * Lao characters require swapping.
77 * <li> If a following pattern match is to be found, any position within a
78 * contracting sequence except the first will fail. Vice versa if a
79 * preceding pattern match is to be found, a invalid starting point
80 * would be any character within a contracting sequence except the last.
81 * </ul>
82 * <p>
83 * A breakiterator can be used if only matches at logical breaks are desired.
84 * Using a breakiterator will only give you results that exactly matches the
85 * boundaries given by the breakiterator. For instance the pattern "e" will
86 * not be found in the string "\u00e9" if a character break iterator is used.
87 * <p>
88 * Options are provided to handle overlapping matches.
89 * E.g. In English, overlapping matches produces the result 0 and 2
90 * for the pattern "abab" in the text "ababab", where else mutually
91 * exclusive matches only produce the result of 0.
92 * <p>
93 * Though collator attributes will be taken into consideration while
94 * performing matches, there are no APIs here for setting and getting the
95 * attributes. These attributes can be set by getting the collator
96 * from <tt>usearch_getCollator</tt> and using the APIs in <tt>ucol.h</tt>.
97 * Lastly to update String Search to the new collator attributes,
98 * usearch_reset() has to be called.
99 * <p>
100 * Restriction: <br>
101 * Currently there are no composite characters that consists of a
102 * character with combining class > 0 before a character with combining
103 * class == 0. However, if such a character exists in the future, the
104 * search mechanism does not guarantee the results for option 1.
105 *
106 * <p>
107 * Example of use:<br>
108 * <pre><code>
109 * char *tgtstr = "The quick brown fox jumped over the lazy fox";
110 * char *patstr = "fox";
111 * UChar target[64];
112 * UChar pattern[16];
113 * UErrorCode status = U_ZERO_ERROR;
114 * u_uastrcpy(target, tgtstr);
115 * u_uastrcpy(pattern, patstr);
116 *
117 * UStringSearch *search = usearch_open(pattern, -1, target, -1, "en_US",
118 * NULL, &status);
119 * if (U_SUCCESS(status)) {
120 * for (int pos = usearch_first(search, &status);
121 * pos != USEARCH_DONE;
122 * pos = usearch_next(search, &status))
123 * {
124 * printf("Found match at %d pos, length is %d\n", pos,
125 * usearch_getMatchLength(search));
126 * }
127 * }
128 *
129 * usearch_close(search);
130 * </code></pre>
131 * @stable ICU 2.4
132 */
133
134 /**
135 * DONE is returned by previous() and next() after all valid matches have
136 * been returned, and by first() and last() if there are no matches at all.
137 * @stable ICU 2.4
138 */
139 #define USEARCH_DONE -1
140
141 /**
142 * Data structure for searching
143 * @stable ICU 2.4
144 */
145 struct UStringSearch;
146 /**
147 * Data structure for searching
148 * @stable ICU 2.4
149 */
150 typedef struct UStringSearch UStringSearch;
151
152 /**
153 * @stable ICU 2.4
154 */
155 typedef enum {
156 /** Option for overlapping matches */
157 USEARCH_OVERLAP,
158 /**
159 Option for canonical matches. option 1 in header documentation.
160 The default value will be USEARCH_OFF
161 */
162 USEARCH_CANONICAL_MATCH,
163 USEARCH_ATTRIBUTE_COUNT
164 } USearchAttribute;
165
166 /**
167 * @stable ICU 2.4
168 */
169 typedef enum {
170 /** default value for any USearchAttribute */
171 USEARCH_DEFAULT = -1,
172 /** value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH */
173 USEARCH_OFF,
174 /** value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH */
175 USEARCH_ON,
176 USEARCH_ATTRIBUTE_VALUE_COUNT
177 } USearchAttributeValue;
178
179 /* open and close ------------------------------------------------------ */
180
181 /**
182 * Creating a search iterator data struct using the argument locale language
183 * rule set. A collator will be created in the process, which will be owned by
184 * this search and will be deleted in <tt>usearch_close</tt>.
185 * @param pattern for matching
186 * @param patternlength length of the pattern, -1 for null-termination
187 * @param text text string
188 * @param textlength length of the text string, -1 for null-termination
189 * @param locale name of locale for the rules to be used
190 * @param breakiter A BreakIterator that will be used to restrict the points
191 * at which matches are detected. If a match is found, but
192 * the match's start or end index is not a boundary as
193 * determined by the <tt>BreakIterator</tt>, the match will
194 * be rejected and another will be searched for.
195 * If this parameter is <tt>NULL</tt>, no break detection is
196 * attempted.
197 * @param status for errors if it occurs. If pattern or text is NULL, or if
198 * patternlength or textlength is 0 then an
199 * U_ILLEGAL_ARGUMENT_ERROR is returned.
200 * @return search iterator data structure, or NULL if there is an error.
201 * @stable ICU 2.4
202 */
203 U_STABLE UStringSearch * U_EXPORT2 usearch_open(const UChar *pattern,
204 int32_t patternlength,
205 const UChar *text,
206 int32_t textlength,
207 const char *locale,
208 UBreakIterator *breakiter,
209 UErrorCode *status);
210
211 /**
212 * Creating a search iterator data struct using the argument collator language
213 * rule set. Note, user retains the ownership of this collator, thus the
214 * responsibility of deletion lies with the user.
215 * NOTE: string search cannot be instantiated from a collator that has
216 * collate digits as numbers (CODAN) turned on.
217 * @param pattern for matching
218 * @param patternlength length of the pattern, -1 for null-termination
219 * @param text text string
220 * @param textlength length of the text string, -1 for null-termination
221 * @param collator used for the language rules
222 * @param breakiter A BreakIterator that will be used to restrict the points
223 * at which matches are detected. If a match is found, but
224 * the match's start or end index is not a boundary as
225 * determined by the <tt>BreakIterator</tt>, the match will
226 * be rejected and another will be searched for.
227 * If this parameter is <tt>NULL</tt>, no break detection is
228 * attempted.
229 * @param status for errors if it occurs. If collator, pattern or text is NULL,
230 * or if patternlength or textlength is 0 then an
231 * U_ILLEGAL_ARGUMENT_ERROR is returned.
232 * @return search iterator data structure, or NULL if there is an error.
233 * @stable ICU 2.4
234 */
235 U_STABLE UStringSearch * U_EXPORT2 usearch_openFromCollator(
236 const UChar *pattern,
237 int32_t patternlength,
238 const UChar *text,
239 int32_t textlength,
240 const UCollator *collator,
241 UBreakIterator *breakiter,
242 UErrorCode *status);
243
244 /**
245 * Destroying and cleaning up the search iterator data struct.
246 * If a collator is created in <tt>usearch_open</tt>, it will be destroyed here.
247 * @param searchiter data struct to clean up
248 * @stable ICU 2.4
249 */
250 U_STABLE void U_EXPORT2 usearch_close(UStringSearch *searchiter);
251
252 /* get and set methods -------------------------------------------------- */
253
254 /**
255 * Sets the current position in the text string which the next search will
256 * start from. Clears previous states.
257 * This method takes the argument index and sets the position in the text
258 * string accordingly without checking if the index is pointing to a
259 * valid starting point to begin searching.
260 * Search positions that may render incorrect results are highlighted in the
261 * header comments
262 * @param strsrch search iterator data struct
263 * @param position position to start next search from. If position is less
264 * than or greater than the text range for searching,
265 * an U_INDEX_OUTOFBOUNDS_ERROR will be returned
266 * @param status error status if any.
267 * @stable ICU 2.4
268 */
269 U_STABLE void U_EXPORT2 usearch_setOffset(UStringSearch *strsrch,
270 int32_t position,
271 UErrorCode *status);
272
273 /**
274 * Return the current index in the string text being searched.
275 * If the iteration has gone past the end of the text (or past the beginning
276 * for a backwards search), <tt>USEARCH_DONE</tt> is returned.
277 * @param strsrch search iterator data struct
278 * @see #USEARCH_DONE
279 * @stable ICU 2.4
280 */
281 U_STABLE int32_t U_EXPORT2 usearch_getOffset(const UStringSearch *strsrch);
282
283 /**
284 * Sets the text searching attributes located in the enum USearchAttribute
285 * with values from the enum USearchAttributeValue.
286 * <tt>USEARCH_DEFAULT</tt> can be used for all attributes for resetting.
287 * @param strsrch search iterator data struct
288 * @param attribute text attribute to be set
289 * @param value text attribute value
290 * @param status for errors if it occurs
291 * @see #usearch_getAttribute
292 * @stable ICU 2.4
293 */
294 U_STABLE void U_EXPORT2 usearch_setAttribute(UStringSearch *strsrch,
295 USearchAttribute attribute,
296 USearchAttributeValue value,
297 UErrorCode *status);
298
299 /**
300 * Gets the text searching attributes.
301 * @param strsrch search iterator data struct
302 * @param attribute text attribute to be retrieve
303 * @return text attribute value
304 * @see #usearch_setAttribute
305 * @stable ICU 2.4
306 */
307 U_STABLE USearchAttributeValue U_EXPORT2 usearch_getAttribute(
308 const UStringSearch *strsrch,
309 USearchAttribute attribute);
310
311 /**
312 * Returns the index to the match in the text string that was searched.
313 * This call returns a valid result only after a successful call to
314 * <tt>usearch_first</tt>, <tt>usearch_next</tt>, <tt>usearch_previous</tt>,
315 * or <tt>usearch_last</tt>.
316 * Just after construction, or after a searching method returns
317 * <tt>USEARCH_DONE</tt>, this method will return <tt>USEARCH_DONE</tt>.
318 * <p>
319 * Use <tt>usearch_getMatchedLength</tt> to get the matched string length.
320 * @param strsrch search iterator data struct
321 * @return index to a substring within the text string that is being
322 * searched.
323 * @see #usearch_first
324 * @see #usearch_next
325 * @see #usearch_previous
326 * @see #usearch_last
327 * @see #USEARCH_DONE
328 * @stable ICU 2.4
329 */
330 U_STABLE int32_t U_EXPORT2 usearch_getMatchedStart(
331 const UStringSearch *strsrch);
332
333 /**
334 * Returns the length of text in the string which matches the search pattern.
335 * This call returns a valid result only after a successful call to
336 * <tt>usearch_first</tt>, <tt>usearch_next</tt>, <tt>usearch_previous</tt>,
337 * or <tt>usearch_last</tt>.
338 * Just after construction, or after a searching method returns
339 * <tt>USEARCH_DONE</tt>, this method will return 0.
340 * @param strsrch search iterator data struct
341 * @return The length of the match in the string text, or 0 if there is no
342 * match currently.
343 * @see #usearch_first
344 * @see #usearch_next
345 * @see #usearch_previous
346 * @see #usearch_last
347 * @see #USEARCH_DONE
348 * @stable ICU 2.4
349 */
350 U_STABLE int32_t U_EXPORT2 usearch_getMatchedLength(
351 const UStringSearch *strsrch);
352
353 /**
354 * Returns the text that was matched by the most recent call to
355 * <tt>usearch_first</tt>, <tt>usearch_next</tt>, <tt>usearch_previous</tt>,
356 * or <tt>usearch_last</tt>.
357 * If the iterator is not pointing at a valid match (e.g. just after
358 * construction or after <tt>USEARCH_DONE</tt> has been returned, returns
359 * an empty string. If result is not large enough to store the matched text,
360 * result will be filled with the partial text and an U_BUFFER_OVERFLOW_ERROR
361 * will be returned in status. result will be null-terminated whenever
362 * possible. If the buffer fits the matched text exactly, a null-termination
363 * is not possible, then a U_STRING_NOT_TERMINATED_ERROR set in status.
364 * Pre-flighting can be either done with length = 0 or the API
365 * <tt>usearch_getMatchLength</tt>.
366 * @param strsrch search iterator data struct
367 * @param result UChar buffer to store the matched string
368 * @param resultCapacity length of the result buffer
369 * @param status error returned if result is not large enough
370 * @return exact length of the matched text, not counting the null-termination
371 * @see #usearch_first
372 * @see #usearch_next
373 * @see #usearch_previous
374 * @see #usearch_last
375 * @see #USEARCH_DONE
376 * @stable ICU 2.4
377 */
378 U_STABLE int32_t U_EXPORT2 usearch_getMatchedText(const UStringSearch *strsrch,
379 UChar *result,
380 int32_t resultCapacity,
381 UErrorCode *status);
382
383 #if !UCONFIG_NO_BREAK_ITERATION
384
385 /**
386 * Set the BreakIterator that will be used to restrict the points at which
387 * matches are detected.
388 * @param strsrch search iterator data struct
389 * @param breakiter A BreakIterator that will be used to restrict the points
390 * at which matches are detected. If a match is found, but
391 * the match's start or end index is not a boundary as
392 * determined by the <tt>BreakIterator</tt>, the match will
393 * be rejected and another will be searched for.
394 * If this parameter is <tt>NULL</tt>, no break detection is
395 * attempted.
396 * @param status for errors if it occurs
397 * @see #usearch_getBreakIterator
398 * @stable ICU 2.4
399 */
400 U_STABLE void U_EXPORT2 usearch_setBreakIterator(UStringSearch *strsrch,
401 UBreakIterator *breakiter,
402 UErrorCode *status);
403
404 /**
405 * Returns the BreakIterator that is used to restrict the points at which
406 * matches are detected. This will be the same object that was passed to the
407 * constructor or to <tt>usearch_setBreakIterator</tt>. Note that
408 * <tt>NULL</tt>
409 * is a legal value; it means that break detection should not be attempted.
410 * @param strsrch search iterator data struct
411 * @return break iterator used
412 * @see #usearch_setBreakIterator
413 * @stable ICU 2.4
414 */
415 U_STABLE const UBreakIterator * U_EXPORT2 usearch_getBreakIterator(
416 const UStringSearch *strsrch);
417
418 #endif
419
420 /**
421 * Set the string text to be searched. Text iteration will hence begin at the
422 * start of the text string. This method is useful if you want to re-use an
423 * iterator to search for the same pattern within a different body of text.
424 * @param strsrch search iterator data struct
425 * @param text new string to look for match
426 * @param textlength length of the new string, -1 for null-termination
427 * @param status for errors if it occurs. If text is NULL, or textlength is 0
428 * then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change
429 * done to strsrch.
430 * @see #usearch_getText
431 * @stable ICU 2.4
432 */
433 U_STABLE void U_EXPORT2 usearch_setText( UStringSearch *strsrch,
434 const UChar *text,
435 int32_t textlength,
436 UErrorCode *status);
437
438 /**
439 * Return the string text to be searched.
440 * @param strsrch search iterator data struct
441 * @param length returned string text length
442 * @return string text
443 * @see #usearch_setText
444 * @stable ICU 2.4
445 */
446 U_STABLE const UChar * U_EXPORT2 usearch_getText(const UStringSearch *strsrch,
447 int32_t *length);
448
449 /**
450 * Gets the collator used for the language rules.
451 * <p>
452 * Deleting the returned <tt>UCollator</tt> before calling
453 * <tt>usearch_close</tt> would cause the string search to fail.
454 * <tt>usearch_close</tt> will delete the collator if this search owns it.
455 * @param strsrch search iterator data struct
456 * @return collator
457 * @stable ICU 2.4
458 */
459 U_STABLE UCollator * U_EXPORT2 usearch_getCollator(
460 const UStringSearch *strsrch);
461
462 /**
463 * Sets the collator used for the language rules. User retains the ownership
464 * of this collator, thus the responsibility of deletion lies with the user.
465 * This method causes internal data such as Boyer-Moore shift tables to
466 * be recalculated, but the iterator's position is unchanged.
467 * @param strsrch search iterator data struct
468 * @param collator to be used
469 * @param status for errors if it occurs
470 * @stable ICU 2.4
471 */
472 U_STABLE void U_EXPORT2 usearch_setCollator( UStringSearch *strsrch,
473 const UCollator *collator,
474 UErrorCode *status);
475
476 /**
477 * Sets the pattern used for matching.
478 * Internal data like the Boyer Moore table will be recalculated, but the
479 * iterator's position is unchanged.
480 * @param strsrch search iterator data struct
481 * @param pattern string
482 * @param patternlength pattern length, -1 for null-terminated string
483 * @param status for errors if it occurs. If text is NULL, or textlength is 0
484 * then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change
485 * done to strsrch.
486 * @stable ICU 2.4
487 */
488 U_STABLE void U_EXPORT2 usearch_setPattern( UStringSearch *strsrch,
489 const UChar *pattern,
490 int32_t patternlength,
491 UErrorCode *status);
492
493 /**
494 * Gets the search pattern
495 * @param strsrch search iterator data struct
496 * @param length return length of the pattern, -1 indicates that the pattern
497 * is null-terminated
498 * @return pattern string
499 * @stable ICU 2.4
500 */
501 U_STABLE const UChar * U_EXPORT2 usearch_getPattern(
502 const UStringSearch *strsrch,
503 int32_t *length);
504
505 /* methods ------------------------------------------------------------- */
506
507 /**
508 * Returns the first index at which the string text matches the search
509 * pattern.
510 * The iterator is adjusted so that its current index (as returned by
511 * <tt>usearch_getOffset</tt>) is the match position if one was found.
512 * If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
513 * the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>.
514 * @param strsrch search iterator data struct
515 * @param status for errors if it occurs
516 * @return The character index of the first match, or
517 * <tt>USEARCH_DONE</tt> if there are no matches.
518 * @see #usearch_getOffset
519 * @see #USEARCH_DONE
520 * @stable ICU 2.4
521 */
522 U_STABLE int32_t U_EXPORT2 usearch_first(UStringSearch *strsrch,
523 UErrorCode *status);
524
525 /**
526 * Returns the first index greater than <tt>position</tt> at which the string
527 * text
528 * matches the search pattern. The iterator is adjusted so that its current
529 * index (as returned by <tt>usearch_getOffset</tt>) is the match position if
530 * one was found.
531 * If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
532 * the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
533 * <p>
534 * Search positions that may render incorrect results are highlighted in the
535 * header comments. If position is less than or greater than the text range
536 * for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned
537 * @param strsrch search iterator data struct
538 * @param position to start the search at
539 * @param status for errors if it occurs
540 * @return The character index of the first match following <tt>pos</tt>,
541 * or <tt>USEARCH_DONE</tt> if there are no matches.
542 * @see #usearch_getOffset
543 * @see #USEARCH_DONE
544 * @stable ICU 2.4
545 */
546 U_STABLE int32_t U_EXPORT2 usearch_following(UStringSearch *strsrch,
547 int32_t position,
548 UErrorCode *status);
549
550 /**
551 * Returns the last index in the target text at which it matches the search
552 * pattern. The iterator is adjusted so that its current
553 * index (as returned by <tt>usearch_getOffset</tt>) is the match position if
554 * one was found.
555 * If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
556 * the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>.
557 * @param strsrch search iterator data struct
558 * @param status for errors if it occurs
559 * @return The index of the first match, or <tt>USEARCH_DONE</tt> if there
560 * are no matches.
561 * @see #usearch_getOffset
562 * @see #USEARCH_DONE
563 * @stable ICU 2.4
564 */
565 U_STABLE int32_t U_EXPORT2 usearch_last(UStringSearch *strsrch,
566 UErrorCode *status);
567
568 /**
569 * Returns the first index less than <tt>position</tt> at which the string text
570 * matches the search pattern. The iterator is adjusted so that its current
571 * index (as returned by <tt>usearch_getOffset</tt>) is the match position if
572 * one was found.
573 * If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
574 * the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
575 * <p>
576 * Search positions that may render incorrect results are highlighted in the
577 * header comments. If position is less than or greater than the text range
578 * for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned
579 * @param strsrch search iterator data struct
580 * @param position index position the search is to begin at
581 * @param status for errors if it occurs
582 * @return The character index of the first match preceding <tt>pos</tt>,
583 * or <tt>USEARCH_DONE</tt> if there are no matches.
584 * @see #usearch_getOffset
585 * @see #USEARCH_DONE
586 * @stable ICU 2.4
587 */
588 U_STABLE int32_t U_EXPORT2 usearch_preceding(UStringSearch *strsrch,
589 int32_t position,
590 UErrorCode *status);
591
592 /**
593 * Returns the index of the next point at which the string text matches the
594 * search pattern, starting from the current position.
595 * The iterator is adjusted so that its current
596 * index (as returned by <tt>usearch_getOffset</tt>) is the match position if
597 * one was found.
598 * If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
599 * the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
600 * @param strsrch search iterator data struct
601 * @param status for errors if it occurs
602 * @return The index of the next match after the current position, or
603 * <tt>USEARCH_DONE</tt> if there are no more matches.
604 * @see #usearch_first
605 * @see #usearch_getOffset
606 * @see #USEARCH_DONE
607 * @stable ICU 2.4
608 */
609 U_STABLE int32_t U_EXPORT2 usearch_next(UStringSearch *strsrch,
610 UErrorCode *status);
611
612 /**
613 * Returns the index of the previous point at which the string text matches
614 * the search pattern, starting at the current position.
615 * The iterator is adjusted so that its current
616 * index (as returned by <tt>usearch_getOffset</tt>) is the match position if
617 * one was found.
618 * If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
619 * the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
620 * @param strsrch search iterator data struct
621 * @param status for errors if it occurs
622 * @return The index of the previous match before the current position,
623 * or <tt>USEARCH_DONE</tt> if there are no more matches.
624 * @see #usearch_last
625 * @see #usearch_getOffset
626 * @see #USEARCH_DONE
627 * @stable ICU 2.4
628 */
629 U_STABLE int32_t U_EXPORT2 usearch_previous(UStringSearch *strsrch,
630 UErrorCode *status);
631
632 /**
633 * Reset the iteration.
634 * Search will begin at the start of the text string if a forward iteration
635 * is initiated before a backwards iteration. Otherwise if a backwards
636 * iteration is initiated before a forwards iteration, the search will begin
637 * at the end of the text string.
638 * @param strsrch search iterator data struct
639 * @see #usearch_first
640 * @stable ICU 2.4
641 */
642 U_STABLE void U_EXPORT2 usearch_reset(UStringSearch *strsrch);
643
644 /**
645 * Simple forward search for the pattern, starting at a specified index,
646 * and using using a default set search options.
647 *
648 * This is an experimental function, and is not an official part of the
649 * ICU API.
650 *
651 * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored.
652 *
653 * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and
654 * any Break Iterator are ignored.
655 *
656 * Matches obey the following constraints:
657 *
658 * Characters at the start or end positions of a match that are ignorable
659 * for collation are not included as part of the match, unless they
660 * are part of a combining sequence, as described below.
661 *
662 * A match will not include a partial combining sequence. Combining
663 * character sequences are considered to be inseperable units,
664 * and either match the pattern completely, or are considered to not match
665 * at all. Thus, for example, an A followed a combining accent mark will
666 * not be found when searching for a plain (unaccented) A. (unless
667 * the collation strength has been set to ignore all accents).
668 *
669 * When beginning a search, the initial starting position, startIdx,
670 * is assumed to be an acceptable match boundary with respect to
671 * combining characters. A combining sequence that spans across the
672 * starting point will not supress a match beginning at startIdx.
673 *
674 * Characters that expand to multiple collation elements
675 * (German sharp-S becoming 'ss', or the composed forms of accented
676 * characters, for example) also must match completely.
677 * Searching for a single 's' in a string containing only a sharp-s will
678 * find no match.
679 *
680 *
681 * @param strsrch the UStringSearch struct, which references both
682 * the text to be searched and the pattern being sought.
683 * @param startIdx The index into the text to begin the search.
684 * @param matchStart An out parameter, the starting index of the matched text.
685 * This parameter may be NULL.
686 * A value of -1 will be returned if no match was found.
687 * @param matchLimit Out parameter, the index of the first position following the matched text.
688 * The matchLimit will be at a suitable position for beginning a subsequent search
689 * in the input text.
690 * This parameter may be NULL.
691 * A value of -1 will be returned if no match was found.
692 *
693 * @param status Report any errors. Note that no match found is not an error.
694 * @return TRUE if a match was found, FALSE otherwise.
695 *
696 * @internal
697 */
698 U_INTERNAL UBool U_EXPORT2 usearch_search(UStringSearch *strsrch,
699 int32_t startIdx,
700 int32_t *matchStart,
701 int32_t *matchLimit,
702 UErrorCode *status);
703
704 /**
705 * Simple backwards search for the pattern, starting at a specified index,
706 * and using using a default set search options.
707 *
708 * This is an experimental function, and is not an official part of the
709 * ICU API.
710 *
711 * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored.
712 *
713 * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and
714 * any Break Iterator are ignored.
715 *
716 * Matches obey the following constraints:
717 *
718 * Characters at the start or end positions of a match that are ignorable
719 * for collation are not included as part of the match, unless they
720 * are part of a combining sequence, as described below.
721 *
722 * A match will not include a partial combining sequence. Combining
723 * character sequences are considered to be inseperable units,
724 * and either match the pattern completely, or are considered to not match
725 * at all. Thus, for example, an A followed a combining accent mark will
726 * not be found when searching for a plain (unaccented) A. (unless
727 * the collation strength has been set to ignore all accents).
728 *
729 * When beginning a search, the initial starting position, startIdx,
730 * is assumed to be an acceptable match boundary with respect to
731 * combining characters. A combining sequence that spans across the
732 * starting point will not supress a match beginning at startIdx.
733 *
734 * Characters that expand to multiple collation elements
735 * (German sharp-S becoming 'ss', or the composed forms of accented
736 * characters, for example) also must match completely.
737 * Searching for a single 's' in a string containing only a sharp-s will
738 * find no match.
739 *
740 *
741 * @param strsrch the UStringSearch struct, which references both
742 * the text to be searched and the pattern being sought.
743 * @param startIdx The index into the text to begin the search.
744 * @param matchStart An out parameter, the starting index of the matched text.
745 * This parameter may be NULL.
746 * A value of -1 will be returned if no match was found.
747 * @param matchLimit Out parameter, the index of the first position following the matched text.
748 * The matchLimit will be at a suitable position for beginning a subsequent search
749 * in the input text.
750 * This parameter may be NULL.
751 * A value of -1 will be returned if no match was found.
752 *
753 * @param status Report any errors. Note that no match found is not an error.
754 * @return TRUE if a match was found, FALSE otherwise.
755 *
756 * @internal
757 */
758 U_INTERNAL UBool U_EXPORT2 usearch_searchBackwards(UStringSearch *strsrch,
759 int32_t startIdx,
760 int32_t *matchStart,
761 int32_t *matchLimit,
762 UErrorCode *status);
763
764 #endif /* #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION */
765
766 #endif