1 /* 2 ****************************************************************************** 3 * Copyright (C) 1996-2010, 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 /** 33 * BoyerMooreSearch 34 * 35 * This object holds the information needed to do a Collation sensitive Boyer-Moore search. It encapulates 36 * the pattern, the "bad character" and "good suffix" tables, the Collator-based data needed to compute them, 37 * and a reference to the text being searched. 38 * 39 * To do a search, you fist need to get a <code>CollData</code> object by calling <code>CollData::open</code>. 40 * Then you construct a <code>BoyerMooreSearch</code> object from the <code>CollData</code> object, the pattern 41 * string and the target string. Then you call the <code>search</code> method. Here's a code sample: 42 * 43 * <pre> 44 * void boyerMooreExample(UCollator *collator, UnicodeString *pattern, UnicodeString *target) 45 * { 46 * UErrorCode status = U_ZERO_ERROR; 47 * CollData *collData = CollData::open(collator, status); 48 * 49 * if (U_FAILURE(status)) { 50 * // could not create a CollData object 51 * return; 52 * } 53 * 54 * BoyerMooreSearch *search = new BoyerMooreSearch(collData, *patternString, target, status); 55 * 56 * if (U_FAILURE(status)) { 57 * // could not create a BoyerMooreSearch object 58 * CollData::close(collData); 59 * return; 60 * } 61 * 62 * int32_t offset = 0, start = -1, end = -1; 63 * 64 * // Find all matches 65 * while (search->search(offset, start, end)) { 66 * // process the match between start and end 67 * ... 68 * // advance past the match 69 * offset = end; 70 * } 71 * 72 * // at this point, if offset == 0, there were no matches 73 * if (offset == 0) { 74 * // handle the case of no matches 75 * } 76 * 77 * delete search; 78 * CollData::close(collData); 79 * 80 * // CollData objects are cached, so the call to 81 * // CollData::close doesn't delete the object. 82 * // Call this if you don't need the object any more. 83 * CollData::flushCollDataCache(); 84 * } 85 * </pre> 86 * 87 * NOTE: This is a technology preview. The final version of this API may not bear any resenblence to this API. 88 * 89 * Knows linitations: 90 * 1) Backwards searching has not been implemented. 91 * 92 * 2) For Han and Hangul characters, this code ignores any Collation tailorings. In general, 93 * this isn't a problem, but in Korean locals, at strength 1, Hangul characters are tailored 94 * to be equal to Han characters with the same pronounciation. Because this code ignroes 95 * tailorings, searching for a Hangul character will not find a Han character and visa-versa. 96 * 97 * 3) In some cases, searching for a pattern that needs to be normalized and ends 98 * in a discontiguous contraction may fail. The only known cases of this are with 99 * the Tibetan script. For example searching for the pattern 100 * "\u0F7F\u0F80\u0F81\u0F82\u0F83\u0F84\u0F85" will fail. (This case is artificial. We've 101 * been unable to find a pratical, real-world example of this failure.) 102 * 103 * @internal ICU 4.0.1 technology preview 104 * 105 * @see CollData 106 */ 107 class U_I18N_API BoyerMooreSearch : public UObject 108 { 109 public: 110 /** 111 * Construct a <code>BoyerMooreSearch</code> object. 112 * 113 * @param theData - A <code>CollData</code> object holding the Collator-sensitive data 114 * @param patternString - the string for which to search 115 * @param targetString - the string in which to search or <code>NULL</code> if youu will 116 * set it later by calling <code>setTargetString</code>. 117 * @param status - will be set if any errors occur. 118 * 119 * Note: if on return, status is set to an error code, 120 * the only safe thing to do with this object is to call 121 * the destructor. 122 * 123 * @internal ICU 4.0.1 technology preview 124 */ 125 BoyerMooreSearch(CollData *theData, const UnicodeString &patternString, const UnicodeString *targetString, UErrorCode &status); 126 127 /** 128 * The desstructor 129 * 130 * @internal ICU 4.0.1 technology preview 131 */ 132 ~BoyerMooreSearch(); 133 134 /** 135 * Test the pattern to see if it generates any CEs. 136 * 137 * @return <code>TRUE</code> if the pattern string did not generate any CEs 138 * 139 * @internal ICU 4.0.1 technology preview 140 */ 141 UBool empty(); 142 143 /** 144 * Search for the pattern string in the target string. 145 * 146 * @param offset - the offset in the target string at which to begin the search 147 * @param start - will be set to the starting offset of the match, or -1 if there's no match 148 * @param end - will be set to the ending offset of the match, or -1 if there's no match 149 * 150 * @return <code>TRUE</code> if the match succeeds, <code>FALSE</code> otherwise. 151 * 152 * @internal ICU 4.0.1 technology preview 153 */ 154 UBool search(int32_t offset, int32_t &start, int32_t &end); 155 156 /** 157 * Set the target string for the match. 158 * 159 * @param targetString - the new target string 160 * @param status - will be set if any errors occur. 161 * 162 * @internal ICU 4.0.1 technology preview 163 */ 164 void setTargetString(const UnicodeString *targetString, UErrorCode &status); 165 166 // **** no longer need these? **** 167 /** 168 * Return the <code>CollData</code> object used for searching 169 * 170 * @return the <code>CollData</code> object used for searching 171 * 172 * @internal ICU 4.0.1 technology preview 173 */ 174 CollData *getData(); 175 176 /** 177 * Return the CEs generated by the pattern string. 178 * 179 * @return a <code>CEList</code> object holding the CEs generated by the pattern string. 180 * 181 * @internal ICU 4.0.1 technology preview 182 */ 183 CEList *getPatternCEs(); 184 185 /** 186 * Return the <code>BadCharacterTable</code> object computed for the pattern string. 187 * 188 * @return the <code>BadCharacterTable</code> object. 189 * 190 * @internal ICU 4.0.1 technology preview 191 */ 192 BadCharacterTable *getBadCharacterTable(); 193 194 /** 195 * Return the <code>GoodSuffixTable</code> object computed for the pattern string. 196 * 197 * @return the <code>GoodSuffixTable</code> object computed for the pattern string. 198 * 199 * @internal ICU 4.0.1 technology preview 200 */ 201 GoodSuffixTable *getGoodSuffixTable(); 202 203 /** 204 * UObject glue... 205 * @internal ICU 4.0.1 technology preview 206 */ 207 virtual UClassID getDynamicClassID() const; 208 /** 209 * UObject glue... 210 * @internal ICU 4.0.1 technology preview 211 */ 212 static UClassID getStaticClassID(); 213 214 private: 215 CollData *data; 216 CEList *patCEs; 217 BadCharacterTable *badCharacterTable; 218 GoodSuffixTable *goodSuffixTable; 219 UnicodeString pattern; 220 Target *target; 221 }; 222 223 U_NAMESPACE_END 224 225 #endif // #if !UCONFIG_NO_COLLATION 226 #endif // #ifndef B_M_SEARCH_H 227