1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/i18n/unicode/usearch.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,836 @@
1.4 +/*
1.5 +**********************************************************************
1.6 +* Copyright (C) 2001-2011 IBM and others. All rights reserved.
1.7 +**********************************************************************
1.8 +* Date Name Description
1.9 +* 06/28/2001 synwee Creation.
1.10 +**********************************************************************
1.11 +*/
1.12 +#ifndef USEARCH_H
1.13 +#define USEARCH_H
1.14 +
1.15 +#include "unicode/utypes.h"
1.16 +
1.17 +#if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION
1.18 +
1.19 +#include "unicode/localpointer.h"
1.20 +#include "unicode/ucol.h"
1.21 +#include "unicode/ucoleitr.h"
1.22 +#include "unicode/ubrk.h"
1.23 +
1.24 +/**
1.25 + * \file
1.26 + * \brief C API: StringSearch
1.27 + *
1.28 + * C Apis for an engine that provides language-sensitive text searching based
1.29 + * on the comparison rules defined in a <tt>UCollator</tt> data struct,
1.30 + * see <tt>ucol.h</tt>. This ensures that language eccentricity can be
1.31 + * handled, e.g. for the German collator, characters ß and SS will be matched
1.32 + * if case is chosen to be ignored.
1.33 + * See the <a href="http://source.icu-project.org/repos/icu/icuhtml/trunk/design/collation/ICU_collation_design.htm">
1.34 + * "ICU Collation Design Document"</a> for more information.
1.35 + * <p>
1.36 + * The algorithm implemented is a modified form of the Boyer Moore's search.
1.37 + * For more information see
1.38 + * <a href="http://icu-project.org/docs/papers/efficient_text_searching_in_java.html">
1.39 + * "Efficient Text Searching in Java"</a>, published in <i>Java Report</i>
1.40 + * in February, 1999, for further information on the algorithm.
1.41 + * <p>
1.42 + * There are 2 match options for selection:<br>
1.43 + * Let S' be the sub-string of a text string S between the offsets start and
1.44 + * end <start, end>.
1.45 + * <br>
1.46 + * A pattern string P matches a text string S at the offsets <start, end>
1.47 + * if
1.48 + * <pre>
1.49 + * option 1. Some canonical equivalent of P matches some canonical equivalent
1.50 + * of S'
1.51 + * option 2. P matches S' and if P starts or ends with a combining mark,
1.52 + * there exists no non-ignorable combining mark before or after S'
1.53 + * in S respectively.
1.54 + * </pre>
1.55 + * Option 2. will be the default.
1.56 + * <p>
1.57 + * This search has APIs similar to that of other text iteration mechanisms
1.58 + * such as the break iterators in <tt>ubrk.h</tt>. Using these
1.59 + * APIs, it is easy to scan through text looking for all occurances of
1.60 + * a given pattern. This search iterator allows changing of direction by
1.61 + * calling a <tt>reset</tt> followed by a <tt>next</tt> or <tt>previous</tt>.
1.62 + * Though a direction change can occur without calling <tt>reset</tt> first,
1.63 + * this operation comes with some speed penalty.
1.64 + * Generally, match results in the forward direction will match the result
1.65 + * matches in the backwards direction in the reverse order
1.66 + * <p>
1.67 + * <tt>usearch.h</tt> provides APIs to specify the starting position
1.68 + * within the text string to be searched, e.g. <tt>usearch_setOffset</tt>,
1.69 + * <tt>usearch_preceding</tt> and <tt>usearch_following</tt>. Since the
1.70 + * starting position will be set as it is specified, please take note that
1.71 + * there are some dangerous positions which the search may render incorrect
1.72 + * results:
1.73 + * <ul>
1.74 + * <li> The midst of a substring that requires normalization.
1.75 + * <li> If the following match is to be found, the position should not be the
1.76 + * second character which requires to be swapped with the preceding
1.77 + * character. Vice versa, if the preceding match is to be found,
1.78 + * position to search from should not be the first character which
1.79 + * requires to be swapped with the next character. E.g certain Thai and
1.80 + * Lao characters require swapping.
1.81 + * <li> If a following pattern match is to be found, any position within a
1.82 + * contracting sequence except the first will fail. Vice versa if a
1.83 + * preceding pattern match is to be found, a invalid starting point
1.84 + * would be any character within a contracting sequence except the last.
1.85 + * </ul>
1.86 + * <p>
1.87 + * A breakiterator can be used if only matches at logical breaks are desired.
1.88 + * Using a breakiterator will only give you results that exactly matches the
1.89 + * boundaries given by the breakiterator. For instance the pattern "e" will
1.90 + * not be found in the string "\u00e9" if a character break iterator is used.
1.91 + * <p>
1.92 + * Options are provided to handle overlapping matches.
1.93 + * E.g. In English, overlapping matches produces the result 0 and 2
1.94 + * for the pattern "abab" in the text "ababab", where else mutually
1.95 + * exclusive matches only produce the result of 0.
1.96 + * <p>
1.97 + * Though collator attributes will be taken into consideration while
1.98 + * performing matches, there are no APIs here for setting and getting the
1.99 + * attributes. These attributes can be set by getting the collator
1.100 + * from <tt>usearch_getCollator</tt> and using the APIs in <tt>ucol.h</tt>.
1.101 + * Lastly to update String Search to the new collator attributes,
1.102 + * usearch_reset() has to be called.
1.103 + * <p>
1.104 + * Restriction: <br>
1.105 + * Currently there are no composite characters that consists of a
1.106 + * character with combining class > 0 before a character with combining
1.107 + * class == 0. However, if such a character exists in the future, the
1.108 + * search mechanism does not guarantee the results for option 1.
1.109 + *
1.110 + * <p>
1.111 + * Example of use:<br>
1.112 + * <pre><code>
1.113 + * char *tgtstr = "The quick brown fox jumped over the lazy fox";
1.114 + * char *patstr = "fox";
1.115 + * UChar target[64];
1.116 + * UChar pattern[16];
1.117 + * UErrorCode status = U_ZERO_ERROR;
1.118 + * u_uastrcpy(target, tgtstr);
1.119 + * u_uastrcpy(pattern, patstr);
1.120 + *
1.121 + * UStringSearch *search = usearch_open(pattern, -1, target, -1, "en_US",
1.122 + * NULL, &status);
1.123 + * if (U_SUCCESS(status)) {
1.124 + * for (int pos = usearch_first(search, &status);
1.125 + * pos != USEARCH_DONE;
1.126 + * pos = usearch_next(search, &status))
1.127 + * {
1.128 + * printf("Found match at %d pos, length is %d\n", pos,
1.129 + * usearch_getMatchLength(search));
1.130 + * }
1.131 + * }
1.132 + *
1.133 + * usearch_close(search);
1.134 + * </code></pre>
1.135 + * @stable ICU 2.4
1.136 + */
1.137 +
1.138 +/**
1.139 +* DONE is returned by previous() and next() after all valid matches have
1.140 +* been returned, and by first() and last() if there are no matches at all.
1.141 +* @stable ICU 2.4
1.142 +*/
1.143 +#define USEARCH_DONE -1
1.144 +
1.145 +/**
1.146 +* Data structure for searching
1.147 +* @stable ICU 2.4
1.148 +*/
1.149 +struct UStringSearch;
1.150 +/**
1.151 +* Data structure for searching
1.152 +* @stable ICU 2.4
1.153 +*/
1.154 +typedef struct UStringSearch UStringSearch;
1.155 +
1.156 +/**
1.157 +* @stable ICU 2.4
1.158 +*/
1.159 +typedef enum {
1.160 + /** Option for overlapping matches */
1.161 + USEARCH_OVERLAP,
1.162 + /**
1.163 + * Option for canonical matches. option 1 in header documentation.
1.164 + * The default value will be USEARCH_OFF
1.165 + */
1.166 + USEARCH_CANONICAL_MATCH,
1.167 + /**
1.168 + * Option to control how collation elements are compared.
1.169 + * The default value will be USEARCH_STANDARD_ELEMENT_COMPARISON.
1.170 + * @stable ICU 4.4
1.171 + */
1.172 + USEARCH_ELEMENT_COMPARISON,
1.173 +
1.174 + USEARCH_ATTRIBUTE_COUNT
1.175 +} USearchAttribute;
1.176 +
1.177 +/**
1.178 +* @stable ICU 2.4
1.179 +*/
1.180 +typedef enum {
1.181 + /** Default value for any USearchAttribute */
1.182 + USEARCH_DEFAULT = -1,
1.183 + /** Value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH */
1.184 + USEARCH_OFF,
1.185 + /** Value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH */
1.186 + USEARCH_ON,
1.187 + /**
1.188 + * Value (default) for USEARCH_ELEMENT_COMPARISON;
1.189 + * standard collation element comparison at the specified collator
1.190 + * strength.
1.191 + * @stable ICU 4.4
1.192 + */
1.193 + USEARCH_STANDARD_ELEMENT_COMPARISON,
1.194 + /**
1.195 + * Value for USEARCH_ELEMENT_COMPARISON;
1.196 + * collation element comparison is modified to effectively provide
1.197 + * behavior between the specified strength and strength - 1. Collation
1.198 + * elements in the pattern that have the base weight for the specified
1.199 + * strength are treated as "wildcards" that match an element with any
1.200 + * other weight at that collation level in the searched text. For
1.201 + * example, with a secondary-strength English collator, a plain 'e' in
1.202 + * the pattern will match a plain e or an e with any diacritic in the
1.203 + * searched text, but an e with diacritic in the pattern will only
1.204 + * match an e with the same diacritic in the searched text.
1.205 + * @stable ICU 4.4
1.206 + */
1.207 + USEARCH_PATTERN_BASE_WEIGHT_IS_WILDCARD,
1.208 + /**
1.209 + * Value for USEARCH_ELEMENT_COMPARISON.
1.210 + * collation element comparison is modified to effectively provide
1.211 + * behavior between the specified strength and strength - 1. Collation
1.212 + * elements in either the pattern or the searched text that have the
1.213 + * base weight for the specified strength are treated as "wildcards"
1.214 + * that match an element with any other weight at that collation level.
1.215 + * For example, with a secondary-strength English collator, a plain 'e'
1.216 + * in the pattern will match a plain e or an e with any diacritic in the
1.217 + * searched text, but an e with diacritic in the pattern will only
1.218 + * match an e with the same diacritic or a plain e in the searched text.
1.219 + * @stable ICU 4.4
1.220 + */
1.221 + USEARCH_ANY_BASE_WEIGHT_IS_WILDCARD,
1.222 +
1.223 + USEARCH_ATTRIBUTE_VALUE_COUNT
1.224 +} USearchAttributeValue;
1.225 +
1.226 +/* open and close ------------------------------------------------------ */
1.227 +
1.228 +/**
1.229 +* Creating a search iterator data struct using the argument locale language
1.230 +* rule set. A collator will be created in the process, which will be owned by
1.231 +* this search and will be deleted in <tt>usearch_close</tt>.
1.232 +* @param pattern for matching
1.233 +* @param patternlength length of the pattern, -1 for null-termination
1.234 +* @param text text string
1.235 +* @param textlength length of the text string, -1 for null-termination
1.236 +* @param locale name of locale for the rules to be used
1.237 +* @param breakiter A BreakIterator that will be used to restrict the points
1.238 +* at which matches are detected. If a match is found, but
1.239 +* the match's start or end index is not a boundary as
1.240 +* determined by the <tt>BreakIterator</tt>, the match will
1.241 +* be rejected and another will be searched for.
1.242 +* If this parameter is <tt>NULL</tt>, no break detection is
1.243 +* attempted.
1.244 +* @param status for errors if it occurs. If pattern or text is NULL, or if
1.245 +* patternlength or textlength is 0 then an
1.246 +* U_ILLEGAL_ARGUMENT_ERROR is returned.
1.247 +* @return search iterator data structure, or NULL if there is an error.
1.248 +* @stable ICU 2.4
1.249 +*/
1.250 +U_STABLE UStringSearch * U_EXPORT2 usearch_open(const UChar *pattern,
1.251 + int32_t patternlength,
1.252 + const UChar *text,
1.253 + int32_t textlength,
1.254 + const char *locale,
1.255 + UBreakIterator *breakiter,
1.256 + UErrorCode *status);
1.257 +
1.258 +/**
1.259 +* Creating a search iterator data struct using the argument collator language
1.260 +* rule set. Note, user retains the ownership of this collator, thus the
1.261 +* responsibility of deletion lies with the user.
1.262 +* NOTE: string search cannot be instantiated from a collator that has
1.263 +* collate digits as numbers (CODAN) turned on.
1.264 +* @param pattern for matching
1.265 +* @param patternlength length of the pattern, -1 for null-termination
1.266 +* @param text text string
1.267 +* @param textlength length of the text string, -1 for null-termination
1.268 +* @param collator used for the language rules
1.269 +* @param breakiter A BreakIterator that will be used to restrict the points
1.270 +* at which matches are detected. If a match is found, but
1.271 +* the match's start or end index is not a boundary as
1.272 +* determined by the <tt>BreakIterator</tt>, the match will
1.273 +* be rejected and another will be searched for.
1.274 +* If this parameter is <tt>NULL</tt>, no break detection is
1.275 +* attempted.
1.276 +* @param status for errors if it occurs. If collator, pattern or text is NULL,
1.277 +* or if patternlength or textlength is 0 then an
1.278 +* U_ILLEGAL_ARGUMENT_ERROR is returned.
1.279 +* @return search iterator data structure, or NULL if there is an error.
1.280 +* @stable ICU 2.4
1.281 +*/
1.282 +U_STABLE UStringSearch * U_EXPORT2 usearch_openFromCollator(
1.283 + const UChar *pattern,
1.284 + int32_t patternlength,
1.285 + const UChar *text,
1.286 + int32_t textlength,
1.287 + const UCollator *collator,
1.288 + UBreakIterator *breakiter,
1.289 + UErrorCode *status);
1.290 +
1.291 +/**
1.292 +* Destroying and cleaning up the search iterator data struct.
1.293 +* If a collator is created in <tt>usearch_open</tt>, it will be destroyed here.
1.294 +* @param searchiter data struct to clean up
1.295 +* @stable ICU 2.4
1.296 +*/
1.297 +U_STABLE void U_EXPORT2 usearch_close(UStringSearch *searchiter);
1.298 +
1.299 +#if U_SHOW_CPLUSPLUS_API
1.300 +
1.301 +U_NAMESPACE_BEGIN
1.302 +
1.303 +/**
1.304 + * \class LocalUStringSearchPointer
1.305 + * "Smart pointer" class, closes a UStringSearch via usearch_close().
1.306 + * For most methods see the LocalPointerBase base class.
1.307 + *
1.308 + * @see LocalPointerBase
1.309 + * @see LocalPointer
1.310 + * @stable ICU 4.4
1.311 + */
1.312 +U_DEFINE_LOCAL_OPEN_POINTER(LocalUStringSearchPointer, UStringSearch, usearch_close);
1.313 +
1.314 +U_NAMESPACE_END
1.315 +
1.316 +#endif
1.317 +
1.318 +/* get and set methods -------------------------------------------------- */
1.319 +
1.320 +/**
1.321 +* Sets the current position in the text string which the next search will
1.322 +* start from. Clears previous states.
1.323 +* This method takes the argument index and sets the position in the text
1.324 +* string accordingly without checking if the index is pointing to a
1.325 +* valid starting point to begin searching.
1.326 +* Search positions that may render incorrect results are highlighted in the
1.327 +* header comments
1.328 +* @param strsrch search iterator data struct
1.329 +* @param position position to start next search from. If position is less
1.330 +* than or greater than the text range for searching,
1.331 +* an U_INDEX_OUTOFBOUNDS_ERROR will be returned
1.332 +* @param status error status if any.
1.333 +* @stable ICU 2.4
1.334 +*/
1.335 +U_STABLE void U_EXPORT2 usearch_setOffset(UStringSearch *strsrch,
1.336 + int32_t position,
1.337 + UErrorCode *status);
1.338 +
1.339 +/**
1.340 +* Return the current index in the string text being searched.
1.341 +* If the iteration has gone past the end of the text (or past the beginning
1.342 +* for a backwards search), <tt>USEARCH_DONE</tt> is returned.
1.343 +* @param strsrch search iterator data struct
1.344 +* @see #USEARCH_DONE
1.345 +* @stable ICU 2.4
1.346 +*/
1.347 +U_STABLE int32_t U_EXPORT2 usearch_getOffset(const UStringSearch *strsrch);
1.348 +
1.349 +/**
1.350 +* Sets the text searching attributes located in the enum USearchAttribute
1.351 +* with values from the enum USearchAttributeValue.
1.352 +* <tt>USEARCH_DEFAULT</tt> can be used for all attributes for resetting.
1.353 +* @param strsrch search iterator data struct
1.354 +* @param attribute text attribute to be set
1.355 +* @param value text attribute value
1.356 +* @param status for errors if it occurs
1.357 +* @see #usearch_getAttribute
1.358 +* @stable ICU 2.4
1.359 +*/
1.360 +U_STABLE void U_EXPORT2 usearch_setAttribute(UStringSearch *strsrch,
1.361 + USearchAttribute attribute,
1.362 + USearchAttributeValue value,
1.363 + UErrorCode *status);
1.364 +
1.365 +/**
1.366 +* Gets the text searching attributes.
1.367 +* @param strsrch search iterator data struct
1.368 +* @param attribute text attribute to be retrieve
1.369 +* @return text attribute value
1.370 +* @see #usearch_setAttribute
1.371 +* @stable ICU 2.4
1.372 +*/
1.373 +U_STABLE USearchAttributeValue U_EXPORT2 usearch_getAttribute(
1.374 + const UStringSearch *strsrch,
1.375 + USearchAttribute attribute);
1.376 +
1.377 +/**
1.378 +* Returns the index to the match in the text string that was searched.
1.379 +* This call returns a valid result only after a successful call to
1.380 +* <tt>usearch_first</tt>, <tt>usearch_next</tt>, <tt>usearch_previous</tt>,
1.381 +* or <tt>usearch_last</tt>.
1.382 +* Just after construction, or after a searching method returns
1.383 +* <tt>USEARCH_DONE</tt>, this method will return <tt>USEARCH_DONE</tt>.
1.384 +* <p>
1.385 +* Use <tt>usearch_getMatchedLength</tt> to get the matched string length.
1.386 +* @param strsrch search iterator data struct
1.387 +* @return index to a substring within the text string that is being
1.388 +* searched.
1.389 +* @see #usearch_first
1.390 +* @see #usearch_next
1.391 +* @see #usearch_previous
1.392 +* @see #usearch_last
1.393 +* @see #USEARCH_DONE
1.394 +* @stable ICU 2.4
1.395 +*/
1.396 +U_STABLE int32_t U_EXPORT2 usearch_getMatchedStart(
1.397 + const UStringSearch *strsrch);
1.398 +
1.399 +/**
1.400 +* Returns the length of text in the string which matches the search pattern.
1.401 +* This call returns a valid result only after a successful call to
1.402 +* <tt>usearch_first</tt>, <tt>usearch_next</tt>, <tt>usearch_previous</tt>,
1.403 +* or <tt>usearch_last</tt>.
1.404 +* Just after construction, or after a searching method returns
1.405 +* <tt>USEARCH_DONE</tt>, this method will return 0.
1.406 +* @param strsrch search iterator data struct
1.407 +* @return The length of the match in the string text, or 0 if there is no
1.408 +* match currently.
1.409 +* @see #usearch_first
1.410 +* @see #usearch_next
1.411 +* @see #usearch_previous
1.412 +* @see #usearch_last
1.413 +* @see #USEARCH_DONE
1.414 +* @stable ICU 2.4
1.415 +*/
1.416 +U_STABLE int32_t U_EXPORT2 usearch_getMatchedLength(
1.417 + const UStringSearch *strsrch);
1.418 +
1.419 +/**
1.420 +* Returns the text that was matched by the most recent call to
1.421 +* <tt>usearch_first</tt>, <tt>usearch_next</tt>, <tt>usearch_previous</tt>,
1.422 +* or <tt>usearch_last</tt>.
1.423 +* If the iterator is not pointing at a valid match (e.g. just after
1.424 +* construction or after <tt>USEARCH_DONE</tt> has been returned, returns
1.425 +* an empty string. If result is not large enough to store the matched text,
1.426 +* result will be filled with the partial text and an U_BUFFER_OVERFLOW_ERROR
1.427 +* will be returned in status. result will be null-terminated whenever
1.428 +* possible. If the buffer fits the matched text exactly, a null-termination
1.429 +* is not possible, then a U_STRING_NOT_TERMINATED_ERROR set in status.
1.430 +* Pre-flighting can be either done with length = 0 or the API
1.431 +* <tt>usearch_getMatchLength</tt>.
1.432 +* @param strsrch search iterator data struct
1.433 +* @param result UChar buffer to store the matched string
1.434 +* @param resultCapacity length of the result buffer
1.435 +* @param status error returned if result is not large enough
1.436 +* @return exact length of the matched text, not counting the null-termination
1.437 +* @see #usearch_first
1.438 +* @see #usearch_next
1.439 +* @see #usearch_previous
1.440 +* @see #usearch_last
1.441 +* @see #USEARCH_DONE
1.442 +* @stable ICU 2.4
1.443 +*/
1.444 +U_STABLE int32_t U_EXPORT2 usearch_getMatchedText(const UStringSearch *strsrch,
1.445 + UChar *result,
1.446 + int32_t resultCapacity,
1.447 + UErrorCode *status);
1.448 +
1.449 +#if !UCONFIG_NO_BREAK_ITERATION
1.450 +
1.451 +/**
1.452 +* Set the BreakIterator that will be used to restrict the points at which
1.453 +* matches are detected.
1.454 +* @param strsrch search iterator data struct
1.455 +* @param breakiter A BreakIterator that will be used to restrict the points
1.456 +* at which matches are detected. If a match is found, but
1.457 +* the match's start or end index is not a boundary as
1.458 +* determined by the <tt>BreakIterator</tt>, the match will
1.459 +* be rejected and another will be searched for.
1.460 +* If this parameter is <tt>NULL</tt>, no break detection is
1.461 +* attempted.
1.462 +* @param status for errors if it occurs
1.463 +* @see #usearch_getBreakIterator
1.464 +* @stable ICU 2.4
1.465 +*/
1.466 +U_STABLE void U_EXPORT2 usearch_setBreakIterator(UStringSearch *strsrch,
1.467 + UBreakIterator *breakiter,
1.468 + UErrorCode *status);
1.469 +
1.470 +/**
1.471 +* Returns the BreakIterator that is used to restrict the points at which
1.472 +* matches are detected. This will be the same object that was passed to the
1.473 +* constructor or to <tt>usearch_setBreakIterator</tt>. Note that
1.474 +* <tt>NULL</tt>
1.475 +* is a legal value; it means that break detection should not be attempted.
1.476 +* @param strsrch search iterator data struct
1.477 +* @return break iterator used
1.478 +* @see #usearch_setBreakIterator
1.479 +* @stable ICU 2.4
1.480 +*/
1.481 +U_STABLE const UBreakIterator * U_EXPORT2 usearch_getBreakIterator(
1.482 + const UStringSearch *strsrch);
1.483 +
1.484 +#endif
1.485 +
1.486 +/**
1.487 +* Set the string text to be searched. Text iteration will hence begin at the
1.488 +* start of the text string. This method is useful if you want to re-use an
1.489 +* iterator to search for the same pattern within a different body of text.
1.490 +* @param strsrch search iterator data struct
1.491 +* @param text new string to look for match
1.492 +* @param textlength length of the new string, -1 for null-termination
1.493 +* @param status for errors if it occurs. If text is NULL, or textlength is 0
1.494 +* then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change
1.495 +* done to strsrch.
1.496 +* @see #usearch_getText
1.497 +* @stable ICU 2.4
1.498 +*/
1.499 +U_STABLE void U_EXPORT2 usearch_setText( UStringSearch *strsrch,
1.500 + const UChar *text,
1.501 + int32_t textlength,
1.502 + UErrorCode *status);
1.503 +
1.504 +/**
1.505 +* Return the string text to be searched.
1.506 +* @param strsrch search iterator data struct
1.507 +* @param length returned string text length
1.508 +* @return string text
1.509 +* @see #usearch_setText
1.510 +* @stable ICU 2.4
1.511 +*/
1.512 +U_STABLE const UChar * U_EXPORT2 usearch_getText(const UStringSearch *strsrch,
1.513 + int32_t *length);
1.514 +
1.515 +/**
1.516 +* Gets the collator used for the language rules.
1.517 +* <p>
1.518 +* Deleting the returned <tt>UCollator</tt> before calling
1.519 +* <tt>usearch_close</tt> would cause the string search to fail.
1.520 +* <tt>usearch_close</tt> will delete the collator if this search owns it.
1.521 +* @param strsrch search iterator data struct
1.522 +* @return collator
1.523 +* @stable ICU 2.4
1.524 +*/
1.525 +U_STABLE UCollator * U_EXPORT2 usearch_getCollator(
1.526 + const UStringSearch *strsrch);
1.527 +
1.528 +/**
1.529 +* Sets the collator used for the language rules. User retains the ownership
1.530 +* of this collator, thus the responsibility of deletion lies with the user.
1.531 +* This method causes internal data such as Boyer-Moore shift tables to
1.532 +* be recalculated, but the iterator's position is unchanged.
1.533 +* @param strsrch search iterator data struct
1.534 +* @param collator to be used
1.535 +* @param status for errors if it occurs
1.536 +* @stable ICU 2.4
1.537 +*/
1.538 +U_STABLE void U_EXPORT2 usearch_setCollator( UStringSearch *strsrch,
1.539 + const UCollator *collator,
1.540 + UErrorCode *status);
1.541 +
1.542 +/**
1.543 +* Sets the pattern used for matching.
1.544 +* Internal data like the Boyer Moore table will be recalculated, but the
1.545 +* iterator's position is unchanged.
1.546 +* @param strsrch search iterator data struct
1.547 +* @param pattern string
1.548 +* @param patternlength pattern length, -1 for null-terminated string
1.549 +* @param status for errors if it occurs. If text is NULL, or textlength is 0
1.550 +* then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change
1.551 +* done to strsrch.
1.552 +* @stable ICU 2.4
1.553 +*/
1.554 +U_STABLE void U_EXPORT2 usearch_setPattern( UStringSearch *strsrch,
1.555 + const UChar *pattern,
1.556 + int32_t patternlength,
1.557 + UErrorCode *status);
1.558 +
1.559 +/**
1.560 +* Gets the search pattern
1.561 +* @param strsrch search iterator data struct
1.562 +* @param length return length of the pattern, -1 indicates that the pattern
1.563 +* is null-terminated
1.564 +* @return pattern string
1.565 +* @stable ICU 2.4
1.566 +*/
1.567 +U_STABLE const UChar * U_EXPORT2 usearch_getPattern(
1.568 + const UStringSearch *strsrch,
1.569 + int32_t *length);
1.570 +
1.571 +/* methods ------------------------------------------------------------- */
1.572 +
1.573 +/**
1.574 +* Returns the first index at which the string text matches the search
1.575 +* pattern.
1.576 +* The iterator is adjusted so that its current index (as returned by
1.577 +* <tt>usearch_getOffset</tt>) is the match position if one was found.
1.578 +* If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
1.579 +* the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>.
1.580 +* @param strsrch search iterator data struct
1.581 +* @param status for errors if it occurs
1.582 +* @return The character index of the first match, or
1.583 +* <tt>USEARCH_DONE</tt> if there are no matches.
1.584 +* @see #usearch_getOffset
1.585 +* @see #USEARCH_DONE
1.586 +* @stable ICU 2.4
1.587 +*/
1.588 +U_STABLE int32_t U_EXPORT2 usearch_first(UStringSearch *strsrch,
1.589 + UErrorCode *status);
1.590 +
1.591 +/**
1.592 +* Returns the first index equal or greater than <tt>position</tt> at which
1.593 +* the string text
1.594 +* matches the search pattern. The iterator is adjusted so that its current
1.595 +* index (as returned by <tt>usearch_getOffset</tt>) is the match position if
1.596 +* one was found.
1.597 +* If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
1.598 +* the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
1.599 +* <p>
1.600 +* Search positions that may render incorrect results are highlighted in the
1.601 +* header comments. If position is less than or greater than the text range
1.602 +* for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned
1.603 +* @param strsrch search iterator data struct
1.604 +* @param position to start the search at
1.605 +* @param status for errors if it occurs
1.606 +* @return The character index of the first match following <tt>pos</tt>,
1.607 +* or <tt>USEARCH_DONE</tt> if there are no matches.
1.608 +* @see #usearch_getOffset
1.609 +* @see #USEARCH_DONE
1.610 +* @stable ICU 2.4
1.611 +*/
1.612 +U_STABLE int32_t U_EXPORT2 usearch_following(UStringSearch *strsrch,
1.613 + int32_t position,
1.614 + UErrorCode *status);
1.615 +
1.616 +/**
1.617 +* Returns the last index in the target text at which it matches the search
1.618 +* pattern. The iterator is adjusted so that its current
1.619 +* index (as returned by <tt>usearch_getOffset</tt>) is the match position if
1.620 +* one was found.
1.621 +* If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
1.622 +* the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>.
1.623 +* @param strsrch search iterator data struct
1.624 +* @param status for errors if it occurs
1.625 +* @return The index of the first match, or <tt>USEARCH_DONE</tt> if there
1.626 +* are no matches.
1.627 +* @see #usearch_getOffset
1.628 +* @see #USEARCH_DONE
1.629 +* @stable ICU 2.4
1.630 +*/
1.631 +U_STABLE int32_t U_EXPORT2 usearch_last(UStringSearch *strsrch,
1.632 + UErrorCode *status);
1.633 +
1.634 +/**
1.635 +* Returns the first index less than <tt>position</tt> at which the string text
1.636 +* matches the search pattern. The iterator is adjusted so that its current
1.637 +* index (as returned by <tt>usearch_getOffset</tt>) is the match position if
1.638 +* one was found.
1.639 +* If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
1.640 +* the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
1.641 +* <p>
1.642 +* Search positions that may render incorrect results are highlighted in the
1.643 +* header comments. If position is less than or greater than the text range
1.644 +* for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned.
1.645 +* <p>
1.646 +* When <tt>USEARCH_OVERLAP</tt> option is off, the last index of the
1.647 +* result match is always less than <tt>position</tt>.
1.648 +* When <tt>USERARCH_OVERLAP</tt> is on, the result match may span across
1.649 +* <tt>position</tt>.
1.650 +* @param strsrch search iterator data struct
1.651 +* @param position index position the search is to begin at
1.652 +* @param status for errors if it occurs
1.653 +* @return The character index of the first match preceding <tt>pos</tt>,
1.654 +* or <tt>USEARCH_DONE</tt> if there are no matches.
1.655 +* @see #usearch_getOffset
1.656 +* @see #USEARCH_DONE
1.657 +* @stable ICU 2.4
1.658 +*/
1.659 +U_STABLE int32_t U_EXPORT2 usearch_preceding(UStringSearch *strsrch,
1.660 + int32_t position,
1.661 + UErrorCode *status);
1.662 +
1.663 +/**
1.664 +* Returns the index of the next point at which the string text matches the
1.665 +* search pattern, starting from the current position.
1.666 +* The iterator is adjusted so that its current
1.667 +* index (as returned by <tt>usearch_getOffset</tt>) is the match position if
1.668 +* one was found.
1.669 +* If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
1.670 +* the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
1.671 +* @param strsrch search iterator data struct
1.672 +* @param status for errors if it occurs
1.673 +* @return The index of the next match after the current position, or
1.674 +* <tt>USEARCH_DONE</tt> if there are no more matches.
1.675 +* @see #usearch_first
1.676 +* @see #usearch_getOffset
1.677 +* @see #USEARCH_DONE
1.678 +* @stable ICU 2.4
1.679 +*/
1.680 +U_STABLE int32_t U_EXPORT2 usearch_next(UStringSearch *strsrch,
1.681 + UErrorCode *status);
1.682 +
1.683 +/**
1.684 +* Returns the index of the previous point at which the string text matches
1.685 +* the search pattern, starting at the current position.
1.686 +* The iterator is adjusted so that its current
1.687 +* index (as returned by <tt>usearch_getOffset</tt>) is the match position if
1.688 +* one was found.
1.689 +* If a match is not found, <tt>USEARCH_DONE</tt> will be returned and
1.690 +* the iterator will be adjusted to the index <tt>USEARCH_DONE</tt>
1.691 +* @param strsrch search iterator data struct
1.692 +* @param status for errors if it occurs
1.693 +* @return The index of the previous match before the current position,
1.694 +* or <tt>USEARCH_DONE</tt> if there are no more matches.
1.695 +* @see #usearch_last
1.696 +* @see #usearch_getOffset
1.697 +* @see #USEARCH_DONE
1.698 +* @stable ICU 2.4
1.699 +*/
1.700 +U_STABLE int32_t U_EXPORT2 usearch_previous(UStringSearch *strsrch,
1.701 + UErrorCode *status);
1.702 +
1.703 +/**
1.704 +* Reset the iteration.
1.705 +* Search will begin at the start of the text string if a forward iteration
1.706 +* is initiated before a backwards iteration. Otherwise if a backwards
1.707 +* iteration is initiated before a forwards iteration, the search will begin
1.708 +* at the end of the text string.
1.709 +* @param strsrch search iterator data struct
1.710 +* @see #usearch_first
1.711 +* @stable ICU 2.4
1.712 +*/
1.713 +U_STABLE void U_EXPORT2 usearch_reset(UStringSearch *strsrch);
1.714 +
1.715 +#ifndef U_HIDE_INTERNAL_API
1.716 +/**
1.717 + * Simple forward search for the pattern, starting at a specified index,
1.718 + * and using using a default set search options.
1.719 + *
1.720 + * This is an experimental function, and is not an official part of the
1.721 + * ICU API.
1.722 + *
1.723 + * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored.
1.724 + *
1.725 + * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and
1.726 + * any Break Iterator are ignored.
1.727 + *
1.728 + * Matches obey the following constraints:
1.729 + *
1.730 + * Characters at the start or end positions of a match that are ignorable
1.731 + * for collation are not included as part of the match, unless they
1.732 + * are part of a combining sequence, as described below.
1.733 + *
1.734 + * A match will not include a partial combining sequence. Combining
1.735 + * character sequences are considered to be inseperable units,
1.736 + * and either match the pattern completely, or are considered to not match
1.737 + * at all. Thus, for example, an A followed a combining accent mark will
1.738 + * not be found when searching for a plain (unaccented) A. (unless
1.739 + * the collation strength has been set to ignore all accents).
1.740 + *
1.741 + * When beginning a search, the initial starting position, startIdx,
1.742 + * is assumed to be an acceptable match boundary with respect to
1.743 + * combining characters. A combining sequence that spans across the
1.744 + * starting point will not supress a match beginning at startIdx.
1.745 + *
1.746 + * Characters that expand to multiple collation elements
1.747 + * (German sharp-S becoming 'ss', or the composed forms of accented
1.748 + * characters, for example) also must match completely.
1.749 + * Searching for a single 's' in a string containing only a sharp-s will
1.750 + * find no match.
1.751 + *
1.752 + *
1.753 + * @param strsrch the UStringSearch struct, which references both
1.754 + * the text to be searched and the pattern being sought.
1.755 + * @param startIdx The index into the text to begin the search.
1.756 + * @param matchStart An out parameter, the starting index of the matched text.
1.757 + * This parameter may be NULL.
1.758 + * A value of -1 will be returned if no match was found.
1.759 + * @param matchLimit Out parameter, the index of the first position following the matched text.
1.760 + * The matchLimit will be at a suitable position for beginning a subsequent search
1.761 + * in the input text.
1.762 + * This parameter may be NULL.
1.763 + * A value of -1 will be returned if no match was found.
1.764 + *
1.765 + * @param status Report any errors. Note that no match found is not an error.
1.766 + * @return TRUE if a match was found, FALSE otherwise.
1.767 + *
1.768 + * @internal
1.769 + */
1.770 +U_INTERNAL UBool U_EXPORT2 usearch_search(UStringSearch *strsrch,
1.771 + int32_t startIdx,
1.772 + int32_t *matchStart,
1.773 + int32_t *matchLimit,
1.774 + UErrorCode *status);
1.775 +
1.776 +/**
1.777 + * Simple backwards search for the pattern, starting at a specified index,
1.778 + * and using using a default set search options.
1.779 + *
1.780 + * This is an experimental function, and is not an official part of the
1.781 + * ICU API.
1.782 + *
1.783 + * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored.
1.784 + *
1.785 + * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and
1.786 + * any Break Iterator are ignored.
1.787 + *
1.788 + * Matches obey the following constraints:
1.789 + *
1.790 + * Characters at the start or end positions of a match that are ignorable
1.791 + * for collation are not included as part of the match, unless they
1.792 + * are part of a combining sequence, as described below.
1.793 + *
1.794 + * A match will not include a partial combining sequence. Combining
1.795 + * character sequences are considered to be inseperable units,
1.796 + * and either match the pattern completely, or are considered to not match
1.797 + * at all. Thus, for example, an A followed a combining accent mark will
1.798 + * not be found when searching for a plain (unaccented) A. (unless
1.799 + * the collation strength has been set to ignore all accents).
1.800 + *
1.801 + * When beginning a search, the initial starting position, startIdx,
1.802 + * is assumed to be an acceptable match boundary with respect to
1.803 + * combining characters. A combining sequence that spans across the
1.804 + * starting point will not supress a match beginning at startIdx.
1.805 + *
1.806 + * Characters that expand to multiple collation elements
1.807 + * (German sharp-S becoming 'ss', or the composed forms of accented
1.808 + * characters, for example) also must match completely.
1.809 + * Searching for a single 's' in a string containing only a sharp-s will
1.810 + * find no match.
1.811 + *
1.812 + *
1.813 + * @param strsrch the UStringSearch struct, which references both
1.814 + * the text to be searched and the pattern being sought.
1.815 + * @param startIdx The index into the text to begin the search.
1.816 + * @param matchStart An out parameter, the starting index of the matched text.
1.817 + * This parameter may be NULL.
1.818 + * A value of -1 will be returned if no match was found.
1.819 + * @param matchLimit Out parameter, the index of the first position following the matched text.
1.820 + * The matchLimit will be at a suitable position for beginning a subsequent search
1.821 + * in the input text.
1.822 + * This parameter may be NULL.
1.823 + * A value of -1 will be returned if no match was found.
1.824 + *
1.825 + * @param status Report any errors. Note that no match found is not an error.
1.826 + * @return TRUE if a match was found, FALSE otherwise.
1.827 + *
1.828 + * @internal
1.829 + */
1.830 +U_INTERNAL UBool U_EXPORT2 usearch_searchBackwards(UStringSearch *strsrch,
1.831 + int32_t startIdx,
1.832 + int32_t *matchStart,
1.833 + int32_t *matchLimit,
1.834 + UErrorCode *status);
1.835 +#endif /* U_HIDE_INTERNAL_API */
1.836 +
1.837 +#endif /* #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION */
1.838 +
1.839 +#endif
