jpayne@69: * The implementation may use a linear search or a modified form of the Boyer-Moore jpayne@69: * search; for more information on the latter see jpayne@69: * jpayne@69: * "Efficient Text Searching in Java", published in Java Report jpayne@69: * in February, 1999. jpayne@69: *
jpayne@69: * There are 2 match options for selection:
jpayne@69: * This search has APIs similar to that of other text iteration mechanisms
jpayne@69: * such as the break iterators in ubrk.h. Using these
jpayne@69: * APIs, it is easy to scan through text looking for all occurrences of
jpayne@69: * a given pattern. This search iterator allows changing of direction by
jpayne@69: * calling a reset followed by a next or previous.
jpayne@69: * Though a direction change can occur without calling reset first,
jpayne@69: * this operation comes with some speed penalty.
jpayne@69: * Generally, match results in the forward direction will match the result
jpayne@69: * matches in the backwards direction in the reverse order
jpayne@69: *
jpayne@69: * usearch.h provides APIs to specify the starting position
jpayne@69: * within the text string to be searched, e.g. usearch_setOffset,
jpayne@69: * usearch_preceding and usearch_following. Since the
jpayne@69: * starting position will be set as it is specified, please take note that
jpayne@69: * there are some dangerous positions which the search may render incorrect
jpayne@69: * results:
jpayne@69: *
jpayne@69: * A breakiterator can be used if only matches at logical breaks are desired.
jpayne@69: * Using a breakiterator will only give you results that exactly matches the
jpayne@69: * boundaries given by the breakiterator. For instance the pattern "e" will
jpayne@69: * not be found in the string "\u00e9" if a character break iterator is used.
jpayne@69: *
jpayne@69: * Options are provided to handle overlapping matches.
jpayne@69: * E.g. In English, overlapping matches produces the result 0 and 2
jpayne@69: * for the pattern "abab" in the text "ababab", where else mutually
jpayne@69: * exclusive matches only produce the result of 0.
jpayne@69: *
jpayne@69: * Options are also provided to implement "asymmetric search" as described in
jpayne@69: *
jpayne@69: * UTS #10 Unicode Collation Algorithm, specifically the USearchAttribute
jpayne@69: * USEARCH_ELEMENT_COMPARISON and its values.
jpayne@69: *
jpayne@69: * Though collator attributes will be taken into consideration while
jpayne@69: * performing matches, there are no APIs here for setting and getting the
jpayne@69: * attributes. These attributes can be set by getting the collator
jpayne@69: * from usearch_getCollator and using the APIs in ucol.h.
jpayne@69: * Lastly to update String Search to the new collator attributes,
jpayne@69: * usearch_reset() has to be called.
jpayne@69: *
jpayne@69: * Restriction:
jpayne@69: * Example of use:
jpayne@69: * Use usearch_getMatchedLength to get the matched string length.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @return index to a substring within the text string that is being
jpayne@69: * searched.
jpayne@69: * @see #usearch_first
jpayne@69: * @see #usearch_next
jpayne@69: * @see #usearch_previous
jpayne@69: * @see #usearch_last
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_getMatchedStart(
jpayne@69: const UStringSearch *strsrch);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the length of text in the string which matches the search pattern.
jpayne@69: * This call returns a valid result only after a successful call to
jpayne@69: * usearch_first, usearch_next, usearch_previous,
jpayne@69: * or usearch_last.
jpayne@69: * Just after construction, or after a searching method returns
jpayne@69: * USEARCH_DONE, this method will return 0.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @return The length of the match in the string text, or 0 if there is no
jpayne@69: * match currently.
jpayne@69: * @see #usearch_first
jpayne@69: * @see #usearch_next
jpayne@69: * @see #usearch_previous
jpayne@69: * @see #usearch_last
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_getMatchedLength(
jpayne@69: const UStringSearch *strsrch);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the text that was matched by the most recent call to
jpayne@69: * usearch_first, usearch_next, usearch_previous,
jpayne@69: * or usearch_last.
jpayne@69: * If the iterator is not pointing at a valid match (e.g. just after
jpayne@69: * construction or after USEARCH_DONE has been returned, returns
jpayne@69: * an empty string. If result is not large enough to store the matched text,
jpayne@69: * result will be filled with the partial text and an U_BUFFER_OVERFLOW_ERROR
jpayne@69: * will be returned in status. result will be null-terminated whenever
jpayne@69: * possible. If the buffer fits the matched text exactly, a null-termination
jpayne@69: * is not possible, then a U_STRING_NOT_TERMINATED_ERROR set in status.
jpayne@69: * Pre-flighting can be either done with length = 0 or the API
jpayne@69: * usearch_getMatchedLength.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param result UChar buffer to store the matched string
jpayne@69: * @param resultCapacity length of the result buffer
jpayne@69: * @param status error returned if result is not large enough
jpayne@69: * @return exact length of the matched text, not counting the null-termination
jpayne@69: * @see #usearch_first
jpayne@69: * @see #usearch_next
jpayne@69: * @see #usearch_previous
jpayne@69: * @see #usearch_last
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_getMatchedText(const UStringSearch *strsrch,
jpayne@69: UChar *result,
jpayne@69: int32_t resultCapacity,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: #if !UCONFIG_NO_BREAK_ITERATION
jpayne@69:
jpayne@69: /**
jpayne@69: * Set the BreakIterator that will be used to restrict the points at which
jpayne@69: * matches are detected.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param breakiter A BreakIterator that will be used to restrict the points
jpayne@69: * at which matches are detected. If a match is found, but
jpayne@69: * the match's start or end index is not a boundary as
jpayne@69: * determined by the BreakIterator, the match will
jpayne@69: * be rejected and another will be searched for.
jpayne@69: * If this parameter is NULL, no break detection is
jpayne@69: * attempted.
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @see #usearch_getBreakIterator
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_setBreakIterator(UStringSearch *strsrch,
jpayne@69: UBreakIterator *breakiter,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the BreakIterator that is used to restrict the points at which
jpayne@69: * matches are detected. This will be the same object that was passed to the
jpayne@69: * constructor or to usearch_setBreakIterator. Note that
jpayne@69: * NULL
jpayne@69: * is a legal value; it means that break detection should not be attempted.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @return break iterator used
jpayne@69: * @see #usearch_setBreakIterator
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE const UBreakIterator * U_EXPORT2 usearch_getBreakIterator(
jpayne@69: const UStringSearch *strsrch);
jpayne@69:
jpayne@69: #endif
jpayne@69:
jpayne@69: /**
jpayne@69: * Set the string text to be searched. Text iteration will hence begin at the
jpayne@69: * start of the text string. This method is useful if you want to re-use an
jpayne@69: * iterator to search for the same pattern within a different body of text.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param text new string to look for match
jpayne@69: * @param textlength length of the new string, -1 for null-termination
jpayne@69: * @param status for errors if it occurs. If text is NULL, or textlength is 0
jpayne@69: * then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change
jpayne@69: * done to strsrch.
jpayne@69: * @see #usearch_getText
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_setText( UStringSearch *strsrch,
jpayne@69: const UChar *text,
jpayne@69: int32_t textlength,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Return the string text to be searched.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param length returned string text length
jpayne@69: * @return string text
jpayne@69: * @see #usearch_setText
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE const UChar * U_EXPORT2 usearch_getText(const UStringSearch *strsrch,
jpayne@69: int32_t *length);
jpayne@69:
jpayne@69: /**
jpayne@69: * Gets the collator used for the language rules.
jpayne@69: *
jpayne@69: * Deleting the returned UCollator before calling
jpayne@69: * usearch_close would cause the string search to fail.
jpayne@69: * usearch_close will delete the collator if this search owns it.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @return collator
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE UCollator * U_EXPORT2 usearch_getCollator(
jpayne@69: const UStringSearch *strsrch);
jpayne@69:
jpayne@69: /**
jpayne@69: * Sets the collator used for the language rules. User retains the ownership
jpayne@69: * of this collator, thus the responsibility of deletion lies with the user.
jpayne@69: * This method causes internal data such as Boyer-Moore shift tables to
jpayne@69: * be recalculated, but the iterator's position is unchanged.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param collator to be used
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_setCollator( UStringSearch *strsrch,
jpayne@69: const UCollator *collator,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Sets the pattern used for matching.
jpayne@69: * Internal data like the Boyer Moore table will be recalculated, but the
jpayne@69: * iterator's position is unchanged.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param pattern string
jpayne@69: * @param patternlength pattern length, -1 for null-terminated string
jpayne@69: * @param status for errors if it occurs. If text is NULL, or textlength is 0
jpayne@69: * then an U_ILLEGAL_ARGUMENT_ERROR is returned with no change
jpayne@69: * done to strsrch.
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_setPattern( UStringSearch *strsrch,
jpayne@69: const UChar *pattern,
jpayne@69: int32_t patternlength,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Gets the search pattern
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param length return length of the pattern, -1 indicates that the pattern
jpayne@69: * is null-terminated
jpayne@69: * @return pattern string
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE const UChar * U_EXPORT2 usearch_getPattern(
jpayne@69: const UStringSearch *strsrch,
jpayne@69: int32_t *length);
jpayne@69:
jpayne@69: /* methods ------------------------------------------------------------- */
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the first index at which the string text matches the search
jpayne@69: * pattern.
jpayne@69: * The iterator is adjusted so that its current index (as returned by
jpayne@69: * usearch_getOffset) is the match position if one was found.
jpayne@69: * If a match is not found, USEARCH_DONE will be returned and
jpayne@69: * the iterator will be adjusted to the index USEARCH_DONE.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @return The character index of the first match, or
jpayne@69: * USEARCH_DONE if there are no matches.
jpayne@69: * @see #usearch_getOffset
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_first(UStringSearch *strsrch,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the first index equal or greater than position at which
jpayne@69: * the string text
jpayne@69: * matches the search pattern. The iterator is adjusted so that its current
jpayne@69: * index (as returned by usearch_getOffset) is the match position if
jpayne@69: * one was found.
jpayne@69: * If a match is not found, USEARCH_DONE will be returned and
jpayne@69: * the iterator will be adjusted to the index USEARCH_DONE
jpayne@69: *
jpayne@69: * Search positions that may render incorrect results are highlighted in the
jpayne@69: * header comments. If position is less than or greater than the text range
jpayne@69: * for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param position to start the search at
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @return The character index of the first match following pos,
jpayne@69: * or USEARCH_DONE if there are no matches.
jpayne@69: * @see #usearch_getOffset
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_following(UStringSearch *strsrch,
jpayne@69: int32_t position,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the last index in the target text at which it matches the search
jpayne@69: * pattern. The iterator is adjusted so that its current
jpayne@69: * index (as returned by usearch_getOffset) is the match position if
jpayne@69: * one was found.
jpayne@69: * If a match is not found, USEARCH_DONE will be returned and
jpayne@69: * the iterator will be adjusted to the index USEARCH_DONE.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @return The index of the first match, or USEARCH_DONE if there
jpayne@69: * are no matches.
jpayne@69: * @see #usearch_getOffset
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_last(UStringSearch *strsrch,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the first index less than position at which the string text
jpayne@69: * matches the search pattern. The iterator is adjusted so that its current
jpayne@69: * index (as returned by usearch_getOffset) is the match position if
jpayne@69: * one was found.
jpayne@69: * If a match is not found, USEARCH_DONE will be returned and
jpayne@69: * the iterator will be adjusted to the index USEARCH_DONE
jpayne@69: *
jpayne@69: * Search positions that may render incorrect results are highlighted in the
jpayne@69: * header comments. If position is less than or greater than the text range
jpayne@69: * for searching, an U_INDEX_OUTOFBOUNDS_ERROR will be returned.
jpayne@69: *
jpayne@69: * When USEARCH_OVERLAP option is off, the last index of the
jpayne@69: * result match is always less than position.
jpayne@69: * When USERARCH_OVERLAP is on, the result match may span across
jpayne@69: * position.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param position index position the search is to begin at
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @return The character index of the first match preceding pos,
jpayne@69: * or USEARCH_DONE if there are no matches.
jpayne@69: * @see #usearch_getOffset
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_preceding(UStringSearch *strsrch,
jpayne@69: int32_t position,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the index of the next point at which the string text matches the
jpayne@69: * search pattern, starting from the current position.
jpayne@69: * The iterator is adjusted so that its current
jpayne@69: * index (as returned by usearch_getOffset) is the match position if
jpayne@69: * one was found.
jpayne@69: * If a match is not found, USEARCH_DONE will be returned and
jpayne@69: * the iterator will be adjusted to the index USEARCH_DONE
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @return The index of the next match after the current position, or
jpayne@69: * USEARCH_DONE if there are no more matches.
jpayne@69: * @see #usearch_first
jpayne@69: * @see #usearch_getOffset
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_next(UStringSearch *strsrch,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the index of the previous point at which the string text matches
jpayne@69: * the search pattern, starting at the current position.
jpayne@69: * The iterator is adjusted so that its current
jpayne@69: * index (as returned by usearch_getOffset) is the match position if
jpayne@69: * one was found.
jpayne@69: * If a match is not found, USEARCH_DONE will be returned and
jpayne@69: * the iterator will be adjusted to the index USEARCH_DONE
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @return The index of the previous match before the current position,
jpayne@69: * or USEARCH_DONE if there are no more matches.
jpayne@69: * @see #usearch_last
jpayne@69: * @see #usearch_getOffset
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_previous(UStringSearch *strsrch,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Reset the iteration.
jpayne@69: * Search will begin at the start of the text string if a forward iteration
jpayne@69: * is initiated before a backwards iteration. Otherwise if a backwards
jpayne@69: * iteration is initiated before a forwards iteration, the search will begin
jpayne@69: * at the end of the text string.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @see #usearch_first
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_reset(UStringSearch *strsrch);
jpayne@69:
jpayne@69: #ifndef U_HIDE_INTERNAL_API
jpayne@69: /**
jpayne@69: * Simple forward search for the pattern, starting at a specified index,
jpayne@69: * and using a default set search options.
jpayne@69: *
jpayne@69: * This is an experimental function, and is not an official part of the
jpayne@69: * ICU API.
jpayne@69: *
jpayne@69: * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored.
jpayne@69: *
jpayne@69: * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and
jpayne@69: * any Break Iterator are ignored.
jpayne@69: *
jpayne@69: * Matches obey the following constraints:
jpayne@69: *
jpayne@69: * Characters at the start or end positions of a match that are ignorable
jpayne@69: * for collation are not included as part of the match, unless they
jpayne@69: * are part of a combining sequence, as described below.
jpayne@69: *
jpayne@69: * A match will not include a partial combining sequence. Combining
jpayne@69: * character sequences are considered to be inseparable units,
jpayne@69: * and either match the pattern completely, or are considered to not match
jpayne@69: * at all. Thus, for example, an A followed a combining accent mark will
jpayne@69: * not be found when searching for a plain (unaccented) A. (unless
jpayne@69: * the collation strength has been set to ignore all accents).
jpayne@69: *
jpayne@69: * When beginning a search, the initial starting position, startIdx,
jpayne@69: * is assumed to be an acceptable match boundary with respect to
jpayne@69: * combining characters. A combining sequence that spans across the
jpayne@69: * starting point will not suppress a match beginning at startIdx.
jpayne@69: *
jpayne@69: * Characters that expand to multiple collation elements
jpayne@69: * (German sharp-S becoming 'ss', or the composed forms of accented
jpayne@69: * characters, for example) also must match completely.
jpayne@69: * Searching for a single 's' in a string containing only a sharp-s will
jpayne@69: * find no match.
jpayne@69: *
jpayne@69: *
jpayne@69: * @param strsrch the UStringSearch struct, which references both
jpayne@69: * the text to be searched and the pattern being sought.
jpayne@69: * @param startIdx The index into the text to begin the search.
jpayne@69: * @param matchStart An out parameter, the starting index of the matched text.
jpayne@69: * This parameter may be NULL.
jpayne@69: * A value of -1 will be returned if no match was found.
jpayne@69: * @param matchLimit Out parameter, the index of the first position following the matched text.
jpayne@69: * The matchLimit will be at a suitable position for beginning a subsequent search
jpayne@69: * in the input text.
jpayne@69: * This parameter may be NULL.
jpayne@69: * A value of -1 will be returned if no match was found.
jpayne@69: *
jpayne@69: * @param status Report any errors. Note that no match found is not an error.
jpayne@69: * @return TRUE if a match was found, FALSE otherwise.
jpayne@69: *
jpayne@69: * @internal
jpayne@69: */
jpayne@69: U_INTERNAL UBool U_EXPORT2 usearch_search(UStringSearch *strsrch,
jpayne@69: int32_t startIdx,
jpayne@69: int32_t *matchStart,
jpayne@69: int32_t *matchLimit,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Simple backwards search for the pattern, starting at a specified index,
jpayne@69: * and using using a default set search options.
jpayne@69: *
jpayne@69: * This is an experimental function, and is not an official part of the
jpayne@69: * ICU API.
jpayne@69: *
jpayne@69: * The collator options, such as UCOL_STRENGTH and UCOL_NORMALIZTION, are honored.
jpayne@69: *
jpayne@69: * The UStringSearch options USEARCH_CANONICAL_MATCH, USEARCH_OVERLAP and
jpayne@69: * any Break Iterator are ignored.
jpayne@69: *
jpayne@69: * Matches obey the following constraints:
jpayne@69: *
jpayne@69: * Characters at the start or end positions of a match that are ignorable
jpayne@69: * for collation are not included as part of the match, unless they
jpayne@69: * are part of a combining sequence, as described below.
jpayne@69: *
jpayne@69: * A match will not include a partial combining sequence. Combining
jpayne@69: * character sequences are considered to be inseparable units,
jpayne@69: * and either match the pattern completely, or are considered to not match
jpayne@69: * at all. Thus, for example, an A followed a combining accent mark will
jpayne@69: * not be found when searching for a plain (unaccented) A. (unless
jpayne@69: * the collation strength has been set to ignore all accents).
jpayne@69: *
jpayne@69: * When beginning a search, the initial starting position, startIdx,
jpayne@69: * is assumed to be an acceptable match boundary with respect to
jpayne@69: * combining characters. A combining sequence that spans across the
jpayne@69: * starting point will not suppress a match beginning at startIdx.
jpayne@69: *
jpayne@69: * Characters that expand to multiple collation elements
jpayne@69: * (German sharp-S becoming 'ss', or the composed forms of accented
jpayne@69: * characters, for example) also must match completely.
jpayne@69: * Searching for a single 's' in a string containing only a sharp-s will
jpayne@69: * find no match.
jpayne@69: *
jpayne@69: *
jpayne@69: * @param strsrch the UStringSearch struct, which references both
jpayne@69: * the text to be searched and the pattern being sought.
jpayne@69: * @param startIdx The index into the text to begin the search.
jpayne@69: * @param matchStart An out parameter, the starting index of the matched text.
jpayne@69: * This parameter may be NULL.
jpayne@69: * A value of -1 will be returned if no match was found.
jpayne@69: * @param matchLimit Out parameter, the index of the first position following the matched text.
jpayne@69: * The matchLimit will be at a suitable position for beginning a subsequent search
jpayne@69: * in the input text.
jpayne@69: * This parameter may be NULL.
jpayne@69: * A value of -1 will be returned if no match was found.
jpayne@69: *
jpayne@69: * @param status Report any errors. Note that no match found is not an error.
jpayne@69: * @return TRUE if a match was found, FALSE otherwise.
jpayne@69: *
jpayne@69: * @internal
jpayne@69: */
jpayne@69: U_INTERNAL UBool U_EXPORT2 usearch_searchBackwards(UStringSearch *strsrch,
jpayne@69: int32_t startIdx,
jpayne@69: int32_t *matchStart,
jpayne@69: int32_t *matchLimit,
jpayne@69: UErrorCode *status);
jpayne@69: #endif /* U_HIDE_INTERNAL_API */
jpayne@69:
jpayne@69: #endif /* #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION */
jpayne@69:
jpayne@69: #endif
jpayne@69: * Let S' be the sub-string of a text string S between the offsets start and
jpayne@69: * end
jpayne@69: * A pattern string P matches a text string S at the offsets
jpayne@69: * option 1. Some canonical equivalent of P matches some canonical equivalent
jpayne@69: * of S'
jpayne@69: * option 2. P matches S' and if P starts or ends with a combining mark,
jpayne@69: * there exists no non-ignorable combining mark before or after S'
jpayne@69: * in S respectively.
jpayne@69: *
jpayne@69: * Option 2. will be the default.
jpayne@69: *
jpayne@69: *
jpayne@69: *
jpayne@69: * Currently there are no composite characters that consists of a
jpayne@69: * character with combining class > 0 before a character with combining
jpayne@69: * class == 0. However, if such a character exists in the future, the
jpayne@69: * search mechanism does not guarantee the results for option 1.
jpayne@69: *
jpayne@69: *
jpayne@69: *
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69:
jpayne@69: /**
jpayne@69: * DONE is returned by previous() and next() after all valid matches have
jpayne@69: * been returned, and by first() and last() if there are no matches at all.
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: #define USEARCH_DONE -1
jpayne@69:
jpayne@69: /**
jpayne@69: * Data structure for searching
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: struct UStringSearch;
jpayne@69: /**
jpayne@69: * Data structure for searching
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: typedef struct UStringSearch UStringSearch;
jpayne@69:
jpayne@69: /**
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: typedef enum {
jpayne@69: /**
jpayne@69: * Option for overlapping matches
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: USEARCH_OVERLAP = 0,
jpayne@69: #ifndef U_HIDE_DEPRECATED_API
jpayne@69: /**
jpayne@69: * Option for canonical matches; option 1 in header documentation.
jpayne@69: * The default value will be USEARCH_OFF.
jpayne@69: * Note: Setting this option to USEARCH_ON currently has no effect on
jpayne@69: * search behavior, and this option is deprecated. Instead, to control
jpayne@69: * canonical match behavior, you must set UCOL_NORMALIZATION_MODE
jpayne@69: * appropriately (to UCOL_OFF or UCOL_ON) in the UCollator used by
jpayne@69: * the UStringSearch object.
jpayne@69: * @see usearch_openFromCollator
jpayne@69: * @see usearch_getCollator
jpayne@69: * @see usearch_setCollator
jpayne@69: * @see ucol_getAttribute
jpayne@69: * @deprecated ICU 53
jpayne@69: */
jpayne@69: USEARCH_CANONICAL_MATCH = 1,
jpayne@69: #endif /* U_HIDE_DEPRECATED_API */
jpayne@69: /**
jpayne@69: * Option to control how collation elements are compared.
jpayne@69: * The default value will be USEARCH_STANDARD_ELEMENT_COMPARISON.
jpayne@69: * @stable ICU 4.4
jpayne@69: */
jpayne@69: USEARCH_ELEMENT_COMPARISON = 2,
jpayne@69:
jpayne@69: #ifndef U_HIDE_DEPRECATED_API
jpayne@69: /**
jpayne@69: * One more than the highest normal USearchAttribute value.
jpayne@69: * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420.
jpayne@69: */
jpayne@69: USEARCH_ATTRIBUTE_COUNT = 3
jpayne@69: #endif /* U_HIDE_DEPRECATED_API */
jpayne@69: } USearchAttribute;
jpayne@69:
jpayne@69: /**
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: typedef enum {
jpayne@69: /**
jpayne@69: * Default value for any USearchAttribute
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: USEARCH_DEFAULT = -1,
jpayne@69: /**
jpayne@69: * Value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: USEARCH_OFF,
jpayne@69: /**
jpayne@69: * Value for USEARCH_OVERLAP and USEARCH_CANONICAL_MATCH
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: USEARCH_ON,
jpayne@69: /**
jpayne@69: * Value (default) for USEARCH_ELEMENT_COMPARISON;
jpayne@69: * standard collation element comparison at the specified collator
jpayne@69: * strength.
jpayne@69: * @stable ICU 4.4
jpayne@69: */
jpayne@69: USEARCH_STANDARD_ELEMENT_COMPARISON,
jpayne@69: /**
jpayne@69: * Value for USEARCH_ELEMENT_COMPARISON;
jpayne@69: * collation element comparison is modified to effectively provide
jpayne@69: * behavior between the specified strength and strength - 1. Collation
jpayne@69: * elements in the pattern that have the base weight for the specified
jpayne@69: * strength are treated as "wildcards" that match an element with any
jpayne@69: * other weight at that collation level in the searched text. For
jpayne@69: * example, with a secondary-strength English collator, a plain 'e' in
jpayne@69: * the pattern will match a plain e or an e with any diacritic in the
jpayne@69: * searched text, but an e with diacritic in the pattern will only
jpayne@69: * match an e with the same diacritic in the searched text.
jpayne@69: *
jpayne@69: * This supports "asymmetric search" as described in
jpayne@69: *
jpayne@69: * UTS #10 Unicode Collation Algorithm.
jpayne@69: *
jpayne@69: * @stable ICU 4.4
jpayne@69: */
jpayne@69: USEARCH_PATTERN_BASE_WEIGHT_IS_WILDCARD,
jpayne@69: /**
jpayne@69: * Value for USEARCH_ELEMENT_COMPARISON.
jpayne@69: * collation element comparison is modified to effectively provide
jpayne@69: * behavior between the specified strength and strength - 1. Collation
jpayne@69: * elements in either the pattern or the searched text that have the
jpayne@69: * base weight for the specified strength are treated as "wildcards"
jpayne@69: * that match an element with any other weight at that collation level.
jpayne@69: * For example, with a secondary-strength English collator, a plain 'e'
jpayne@69: * in the pattern will match a plain e or an e with any diacritic in the
jpayne@69: * searched text, but an e with diacritic in the pattern will only
jpayne@69: * match an e with the same diacritic or a plain e in the searched text.
jpayne@69: *
jpayne@69: * This option is similar to "asymmetric search" as described in
jpayne@69: * [UTS #10 Unicode Collation Algorithm](http://www.unicode.org/reports/tr10/#Asymmetric_Search),
jpayne@69: * but also allows unmarked characters in the searched text to match
jpayne@69: * marked or unmarked versions of that character in the pattern.
jpayne@69: *
jpayne@69: * @stable ICU 4.4
jpayne@69: */
jpayne@69: USEARCH_ANY_BASE_WEIGHT_IS_WILDCARD,
jpayne@69:
jpayne@69: #ifndef U_HIDE_DEPRECATED_API
jpayne@69: /**
jpayne@69: * One more than the highest normal USearchAttributeValue value.
jpayne@69: * @deprecated ICU 58 The numeric value may change over time, see ICU ticket #12420.
jpayne@69: */
jpayne@69: USEARCH_ATTRIBUTE_VALUE_COUNT
jpayne@69: #endif /* U_HIDE_DEPRECATED_API */
jpayne@69: } USearchAttributeValue;
jpayne@69:
jpayne@69: /* open and close ------------------------------------------------------ */
jpayne@69:
jpayne@69: /**
jpayne@69: * Creating a search iterator data struct using the argument locale language
jpayne@69: * rule set. A collator will be created in the process, which will be owned by
jpayne@69: * this search and will be deleted in usearch_close.
jpayne@69: * @param pattern for matching
jpayne@69: * @param patternlength length of the pattern, -1 for null-termination
jpayne@69: * @param text text string
jpayne@69: * @param textlength length of the text string, -1 for null-termination
jpayne@69: * @param locale name of locale for the rules to be used
jpayne@69: * @param breakiter A BreakIterator that will be used to restrict the points
jpayne@69: * at which matches are detected. If a match is found, but
jpayne@69: * the match's start or end index is not a boundary as
jpayne@69: * determined by the BreakIterator, the match will
jpayne@69: * be rejected and another will be searched for.
jpayne@69: * If this parameter is NULL, no break detection is
jpayne@69: * attempted.
jpayne@69: * @param status for errors if it occurs. If pattern or text is NULL, or if
jpayne@69: * patternlength or textlength is 0 then an
jpayne@69: * U_ILLEGAL_ARGUMENT_ERROR is returned.
jpayne@69: * @return search iterator data structure, or NULL if there is an error.
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE UStringSearch * U_EXPORT2 usearch_open(const UChar *pattern,
jpayne@69: int32_t patternlength,
jpayne@69: const UChar *text,
jpayne@69: int32_t textlength,
jpayne@69: const char *locale,
jpayne@69: UBreakIterator *breakiter,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Creating a search iterator data struct using the argument collator language
jpayne@69: * rule set. Note, user retains the ownership of this collator, thus the
jpayne@69: * responsibility of deletion lies with the user.
jpayne@69: * NOTE: string search cannot be instantiated from a collator that has
jpayne@69: * collate digits as numbers (CODAN) turned on.
jpayne@69: * @param pattern for matching
jpayne@69: * @param patternlength length of the pattern, -1 for null-termination
jpayne@69: * @param text text string
jpayne@69: * @param textlength length of the text string, -1 for null-termination
jpayne@69: * @param collator used for the language rules
jpayne@69: * @param breakiter A BreakIterator that will be used to restrict the points
jpayne@69: * at which matches are detected. If a match is found, but
jpayne@69: * the match's start or end index is not a boundary as
jpayne@69: * determined by the BreakIterator, the match will
jpayne@69: * be rejected and another will be searched for.
jpayne@69: * If this parameter is NULL, no break detection is
jpayne@69: * attempted.
jpayne@69: * @param status for errors if it occurs. If collator, pattern or text is NULL,
jpayne@69: * or if patternlength or textlength is 0 then an
jpayne@69: * U_ILLEGAL_ARGUMENT_ERROR is returned.
jpayne@69: * @return search iterator data structure, or NULL if there is an error.
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE UStringSearch * U_EXPORT2 usearch_openFromCollator(
jpayne@69: const UChar *pattern,
jpayne@69: int32_t patternlength,
jpayne@69: const UChar *text,
jpayne@69: int32_t textlength,
jpayne@69: const UCollator *collator,
jpayne@69: UBreakIterator *breakiter,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Destroying and cleaning up the search iterator data struct.
jpayne@69: * If a collator is created in usearch_open, it will be destroyed here.
jpayne@69: * @param searchiter data struct to clean up
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_close(UStringSearch *searchiter);
jpayne@69:
jpayne@69: #if U_SHOW_CPLUSPLUS_API
jpayne@69:
jpayne@69: U_NAMESPACE_BEGIN
jpayne@69:
jpayne@69: /**
jpayne@69: * \class LocalUStringSearchPointer
jpayne@69: * "Smart pointer" class, closes a UStringSearch via usearch_close().
jpayne@69: * For most methods see the LocalPointerBase base class.
jpayne@69: *
jpayne@69: * @see LocalPointerBase
jpayne@69: * @see LocalPointer
jpayne@69: * @stable ICU 4.4
jpayne@69: */
jpayne@69: U_DEFINE_LOCAL_OPEN_POINTER(LocalUStringSearchPointer, UStringSearch, usearch_close);
jpayne@69:
jpayne@69: U_NAMESPACE_END
jpayne@69:
jpayne@69: #endif
jpayne@69:
jpayne@69: /* get and set methods -------------------------------------------------- */
jpayne@69:
jpayne@69: /**
jpayne@69: * Sets the current position in the text string which the next search will
jpayne@69: * start from. Clears previous states.
jpayne@69: * This method takes the argument index and sets the position in the text
jpayne@69: * string accordingly without checking if the index is pointing to a
jpayne@69: * valid starting point to begin searching.
jpayne@69: * Search positions that may render incorrect results are highlighted in the
jpayne@69: * header comments
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param position position to start next search from. If position is less
jpayne@69: * than or greater than the text range for searching,
jpayne@69: * an U_INDEX_OUTOFBOUNDS_ERROR will be returned
jpayne@69: * @param status error status if any.
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_setOffset(UStringSearch *strsrch,
jpayne@69: int32_t position,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Return the current index in the string text being searched.
jpayne@69: * If the iteration has gone past the end of the text (or past the beginning
jpayne@69: * for a backwards search), USEARCH_DONE is returned.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @see #USEARCH_DONE
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE int32_t U_EXPORT2 usearch_getOffset(const UStringSearch *strsrch);
jpayne@69:
jpayne@69: /**
jpayne@69: * Sets the text searching attributes located in the enum USearchAttribute
jpayne@69: * with values from the enum USearchAttributeValue.
jpayne@69: * USEARCH_DEFAULT can be used for all attributes for resetting.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param attribute text attribute to be set
jpayne@69: * @param value text attribute value
jpayne@69: * @param status for errors if it occurs
jpayne@69: * @see #usearch_getAttribute
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE void U_EXPORT2 usearch_setAttribute(UStringSearch *strsrch,
jpayne@69: USearchAttribute attribute,
jpayne@69: USearchAttributeValue value,
jpayne@69: UErrorCode *status);
jpayne@69:
jpayne@69: /**
jpayne@69: * Gets the text searching attributes.
jpayne@69: * @param strsrch search iterator data struct
jpayne@69: * @param attribute text attribute to be retrieve
jpayne@69: * @return text attribute value
jpayne@69: * @see #usearch_setAttribute
jpayne@69: * @stable ICU 2.4
jpayne@69: */
jpayne@69: U_STABLE USearchAttributeValue U_EXPORT2 usearch_getAttribute(
jpayne@69: const UStringSearch *strsrch,
jpayne@69: USearchAttribute attribute);
jpayne@69:
jpayne@69: /**
jpayne@69: * Returns the index to the match in the text string that was searched.
jpayne@69: * This call returns a valid result only after a successful call to
jpayne@69: * usearch_first, usearch_next, usearch_previous,
jpayne@69: * or usearch_last.
jpayne@69: * Just after construction, or after a searching method returns
jpayne@69: * USEARCH_DONE, this method will return USEARCH_DONE.
jpayne@69: *
jpayne@69: * char *tgtstr = "The quick brown fox jumped over the lazy fox";
jpayne@69: * char *patstr = "fox";
jpayne@69: * UChar target[64];
jpayne@69: * UChar pattern[16];
jpayne@69: * UErrorCode status = U_ZERO_ERROR;
jpayne@69: * u_uastrcpy(target, tgtstr);
jpayne@69: * u_uastrcpy(pattern, patstr);
jpayne@69: *
jpayne@69: * UStringSearch *search = usearch_open(pattern, -1, target, -1, "en_US",
jpayne@69: * NULL, &status);
jpayne@69: * if (U_SUCCESS(status)) {
jpayne@69: * for (int pos = usearch_first(search, &status);
jpayne@69: * pos != USEARCH_DONE;
jpayne@69: * pos = usearch_next(search, &status))
jpayne@69: * {
jpayne@69: * printf("Found match at %d pos, length is %d\n", pos,
jpayne@69: * usearch_getMatchedLength(search));
jpayne@69: * }
jpayne@69: * }
jpayne@69: *
jpayne@69: * usearch_close(search);
jpayne@69: *