1 /* 2 ****************************************************************************** 3 * Copyright (C) 1996-2011, International Business Machines * 4 * Corporation and others. All Rights Reserved. * 5 ****************************************************************************** 6 */ 7 8 /** 9 * \file 10 * \brief C++ API: Boyer-Moore StringSearch technology preview 11 * \internal ICU 4.0.1 technology preview 12 */ 13 14 #ifndef B_M_SEARCH_H 15 #define B_M_SEARCH_H 16 17 #include "unicode/utypes.h" 18 19 #if !UCONFIG_NO_COLLATION && !UCONFIG_NO_BREAK_ITERATION 20 21 #include "unicode/uobject.h" 22 #include "unicode/ucol.h" 23 24 #include "unicode/colldata.h" 25 26 U_NAMESPACE_BEGIN 27 28 class BadCharacterTable; 29 class GoodSuffixTable; 30 class Target; 31 32 #ifndef U_HIDE_INTERNAL_API 33 /** 34 * BoyerMooreSearch 35 * 36 * This object holds the information needed to do a Collation sensitive Boyer-Moore search. It encapulates 37 * the pattern, the "bad character" and "good suffix" tables, the Collator-based data needed to compute them, 38 * and a reference to the text being searched. 39 * 40 * To do a search, you fist need to get a <code>CollData</code> object by calling <code>CollData::open</code>. 41 * Then you construct a <code>BoyerMooreSearch</code> object from the <code>CollData</code> object, the pattern 42 * string and the target string. Then you call the <code>search</code> method. Here's a code sample: 43 * 44 * <pre> 45 * void boyerMooreExample(UCollator *collator, UnicodeString *pattern, UnicodeString *target) 46 * { 47 * UErrorCode status = U_ZERO_ERROR; 48 * CollData *collData = CollData::open(collator, status); 49 * 50 * if (U_FAILURE(status)) { 51 * // could not create a CollData object 52 * return; 53 * } 54 * 55 * BoyerMooreSearch *search = new BoyerMooreSearch(collData, *patternString, target, status); 56 * 57 * if (U_FAILURE(status)) { 58 * // could not create a BoyerMooreSearch object 59 * CollData::close(collData); 60 * return; 61 * } 62 * 63 * int32_t offset = 0, start = -1, end = -1; 64 * 65 * // Find all matches 66 * while (search->search(offset, start, end)) { 67 * // process the match between start and end 68 * ... 69 * // advance past the match 70 * offset = end; 71 * } 72 * 73 * // at this point, if offset == 0, there were no matches 74 * if (offset == 0) { 75 * // handle the case of no matches 76 * } 77 * 78 * delete search; 79 * CollData::close(collData); 80 * 81 * // CollData objects are cached, so the call to 82 * // CollData::close doesn't delete the object. 83 * // Call this if you don't need the object any more. 84 * CollData::flushCollDataCache(); 85 * } 86 * </pre> 87 * 88 * NOTE: This is a technology preview. The final version of this API may not bear any resenblence to this API. 89 * 90 * Knows linitations: 91 * 1) Backwards searching has not been implemented. 92 * 93 * 2) For Han and Hangul characters, this code ignores any Collation tailorings. In general, 94 * this isn't a problem, but in Korean locals, at strength 1, Hangul characters are tailored 95 * to be equal to Han characters with the same pronounciation. Because this code ignroes 96 * tailorings, searching for a Hangul character will not find a Han character and visa-versa. 97 * 98 * 3) In some cases, searching for a pattern that needs to be normalized and ends 99 * in a discontiguous contraction may fail. The only known cases of this are with 100 * the Tibetan script. For example searching for the pattern 101 * "\u0F7F\u0F80\u0F81\u0F82\u0F83\u0F84\u0F85" will fail. (This case is artificial. We've 102 * been unable to find a pratical, real-world example of this failure.) 103 * 104 * @internal ICU 4.0.1 technology preview 105 * 106 * @see CollData 107 */ 108 class U_I18N_API BoyerMooreSearch : public UObject 109 { 110 public: 111 /** 112 * Construct a <code>BoyerMooreSearch</code> object. 113 * 114 * @param theData - A <code>CollData</code> object holding the Collator-sensitive data 115 * @param patternString - the string for which to search 116 * @param targetString - the string in which to search or <code>NULL</code> if youu will 117 * set it later by calling <code>setTargetString</code>. 118 * @param status - will be set if any errors occur. 119 * 120 * Note: if on return, status is set to an error code, 121 * the only safe thing to do with this object is to call 122 * the destructor. 123 * 124 * @internal ICU 4.0.1 technology preview 125 */ 126 BoyerMooreSearch(CollData *theData, const UnicodeString &patternString, const UnicodeString *targetString, UErrorCode &status); 127 128 /** 129 * The desstructor 130 * 131 * @internal ICU 4.0.1 technology preview 132 */ 133 ~BoyerMooreSearch(); 134 135 /** 136 * Test the pattern to see if it generates any CEs. 137 * 138 * @return <code>TRUE</code> if the pattern string did not generate any CEs 139 * 140 * @internal ICU 4.0.1 technology preview 141 */ 142 UBool empty(); 143 144 /** 145 * Search for the pattern string in the target string. 146 * 147 * @param offset - the offset in the target string at which to begin the search 148 * @param start - will be set to the starting offset of the match, or -1 if there's no match 149 * @param end - will be set to the ending offset of the match, or -1 if there's no match 150 * 151 * @return <code>TRUE</code> if the match succeeds, <code>FALSE</code> otherwise. 152 * 153 * @internal ICU 4.0.1 technology preview 154 */ 155 UBool search(int32_t offset, int32_t &start, int32_t &end); 156 157 /** 158 * Set the target string for the match. 159 * 160 * @param targetString - the new target string 161 * @param status - will be set if any errors occur. 162 * 163 * @internal ICU 4.0.1 technology preview 164 */ 165 void setTargetString(const UnicodeString *targetString, UErrorCode &status); 166 167 // **** no longer need these? **** 168 /** 169 * Return the <code>CollData</code> object used for searching 170 * 171 * @return the <code>CollData</code> object used for searching 172 * 173 * @internal ICU 4.0.1 technology preview 174 */ 175 CollData *getData(); 176 177 /** 178 * Return the CEs generated by the pattern string. 179 * 180 * @return a <code>CEList</code> object holding the CEs generated by the pattern string. 181 * 182 * @internal ICU 4.0.1 technology preview 183 */ 184 CEList *getPatternCEs(); 185 186 /** 187 * Return the <code>BadCharacterTable</code> object computed for the pattern string. 188 * 189 * @return the <code>BadCharacterTable</code> object. 190 * 191 * @internal ICU 4.0.1 technology preview 192 */ 193 BadCharacterTable *getBadCharacterTable(); 194 195 /** 196 * Return the <code>GoodSuffixTable</code> object computed for the pattern string. 197 * 198 * @return the <code>GoodSuffixTable</code> object computed for the pattern string. 199 * 200 * @internal ICU 4.0.1 technology preview 201 */ 202 GoodSuffixTable *getGoodSuffixTable(); 203 204 /** 205 * UObject glue... 206 * @internal ICU 4.0.1 technology preview 207 */ 208 virtual UClassID getDynamicClassID() const; 209 /** 210 * UObject glue... 211 * @internal ICU 4.0.1 technology preview 212 */ 213 static UClassID getStaticClassID(); 214 215 private: 216 CollData *data; 217 CEList *patCEs; 218 BadCharacterTable *badCharacterTable; 219 GoodSuffixTable *goodSuffixTable; 220 UnicodeString pattern; 221 Target *target; 222 }; 223 #endif /* U_HIDE_INTERNAL_API */ 224 225 U_NAMESPACE_END 226 227 #endif // #if !UCONFIG_NO_COLLATION 228 #endif // #ifndef B_M_SEARCH_H 229