1 // © 2017 and later: Unicode, Inc. and others. 2 // License & terms of use: http://www.unicode.org/copyright.html 3 4 // umutablecptrie.h (split out of ucptrie.h) 5 // created: 2018jan24 Markus W. Scherer 6 7 #ifndef __UMUTABLECPTRIE_H__ 8 #define __UMUTABLECPTRIE_H__ 9 10 #include "unicode/utypes.h" 11 12 #include "unicode/localpointer.h" 13 #include "unicode/ucpmap.h" 14 #include "unicode/ucptrie.h" 15 #include "unicode/utf8.h" 16 17 U_CDECL_BEGIN 18 19 /** 20 * \file 21 * 22 * This file defines a mutable Unicode code point trie. 23 * 24 * @see UCPTrie 25 * @see UMutableCPTrie 26 */ 27 28 /** 29 * Mutable Unicode code point trie. 30 * Fast map from Unicode code points (U+0000..U+10FFFF) to 32-bit integer values. 31 * For details see http://site.icu-project.org/design/struct/utrie 32 * 33 * Setting values (especially ranges) and lookup is fast. 34 * The mutable trie is only somewhat space-efficient. 35 * It builds a compacted, immutable UCPTrie. 36 * 37 * This trie can be modified while iterating over its contents. 38 * For example, it is possible to merge its values with those from another 39 * set of ranges (e.g., another mutable or immutable trie): 40 * Iterate over those source ranges; for each of them iterate over this trie; 41 * add the source value into the value of each trie range. 42 * 43 * @see UCPTrie 44 * @see umutablecptrie_buildImmutable 45 * @stable ICU 63 46 */ 47 typedef struct UMutableCPTrie UMutableCPTrie; 48 49 /** 50 * Creates a mutable trie that initially maps each Unicode code point to the same value. 51 * It uses 32-bit data values until umutablecptrie_buildImmutable() is called. 52 * umutablecptrie_buildImmutable() takes a valueWidth parameter which 53 * determines the number of bits in the data value in the resulting UCPTrie. 54 * You must umutablecptrie_close() the trie once you are done using it. 55 * 56 * @param initialValue the initial value that is set for all code points 57 * @param errorValue the value for out-of-range code points and ill-formed UTF-8/16 58 * @param pErrorCode an in/out ICU UErrorCode 59 * @return the trie 60 * @stable ICU 63 61 */ 62 U_CAPI UMutableCPTrie * U_EXPORT2 63 umutablecptrie_open(uint32_t initialValue, uint32_t errorValue, UErrorCode *pErrorCode); 64 65 /** 66 * Clones a mutable trie. 67 * You must umutablecptrie_close() the clone once you are done using it. 68 * 69 * @param other the trie to clone 70 * @param pErrorCode an in/out ICU UErrorCode 71 * @return the trie clone 72 * @stable ICU 63 73 */ 74 U_CAPI UMutableCPTrie * U_EXPORT2 75 umutablecptrie_clone(const UMutableCPTrie *other, UErrorCode *pErrorCode); 76 77 /** 78 * Closes a mutable trie and releases associated memory. 79 * 80 * @param trie the trie 81 * @stable ICU 63 82 */ 83 U_CAPI void U_EXPORT2 84 umutablecptrie_close(UMutableCPTrie *trie); 85 86 /** 87 * Creates a mutable trie with the same contents as the UCPMap. 88 * You must umutablecptrie_close() the mutable trie once you are done using it. 89 * 90 * @param map the source map 91 * @param pErrorCode an in/out ICU UErrorCode 92 * @return the mutable trie 93 * @stable ICU 63 94 */ 95 U_CAPI UMutableCPTrie * U_EXPORT2 96 umutablecptrie_fromUCPMap(const UCPMap *map, UErrorCode *pErrorCode); 97 98 /** 99 * Creates a mutable trie with the same contents as the immutable one. 100 * You must umutablecptrie_close() the mutable trie once you are done using it. 101 * 102 * @param trie the immutable trie 103 * @param pErrorCode an in/out ICU UErrorCode 104 * @return the mutable trie 105 * @stable ICU 63 106 */ 107 U_CAPI UMutableCPTrie * U_EXPORT2 108 umutablecptrie_fromUCPTrie(const UCPTrie *trie, UErrorCode *pErrorCode); 109 110 /** 111 * Returns the value for a code point as stored in the trie. 112 * 113 * @param trie the trie 114 * @param c the code point 115 * @return the value 116 * @stable ICU 63 117 */ 118 U_CAPI uint32_t U_EXPORT2 119 umutablecptrie_get(const UMutableCPTrie *trie, UChar32 c); 120 121 /** 122 * Returns the last code point such that all those from start to there have the same value. 123 * Can be used to efficiently iterate over all same-value ranges in a trie. 124 * (This is normally faster than iterating over code points and get()ting each value, 125 * but much slower than a data structure that stores ranges directly.) 126 * 127 * The trie can be modified between calls to this function. 128 * 129 * If the UCPMapValueFilter function pointer is not NULL, then 130 * the value to be delivered is passed through that function, and the return value is the end 131 * of the range where all values are modified to the same actual value. 132 * The value is unchanged if that function pointer is NULL. 133 * 134 * See the same-signature ucptrie_getRange() for a code sample. 135 * 136 * @param trie the trie 137 * @param start range start 138 * @param option defines whether surrogates are treated normally, 139 * or as having the surrogateValue; usually UCPMAP_RANGE_NORMAL 140 * @param surrogateValue value for surrogates; ignored if option==UCPMAP_RANGE_NORMAL 141 * @param filter a pointer to a function that may modify the trie data value, 142 * or NULL if the values from the trie are to be used unmodified 143 * @param context an opaque pointer that is passed on to the filter function 144 * @param pValue if not NULL, receives the value that every code point start..end has; 145 * may have been modified by filter(context, trie value) 146 * if that function pointer is not NULL 147 * @return the range end code point, or -1 if start is not a valid code point 148 * @stable ICU 63 149 */ 150 U_CAPI UChar32 U_EXPORT2 151 umutablecptrie_getRange(const UMutableCPTrie *trie, UChar32 start, 152 UCPMapRangeOption option, uint32_t surrogateValue, 153 UCPMapValueFilter *filter, const void *context, uint32_t *pValue); 154 155 /** 156 * Sets a value for a code point. 157 * 158 * @param trie the trie 159 * @param c the code point 160 * @param value the value 161 * @param pErrorCode an in/out ICU UErrorCode 162 * @stable ICU 63 163 */ 164 U_CAPI void U_EXPORT2 165 umutablecptrie_set(UMutableCPTrie *trie, UChar32 c, uint32_t value, UErrorCode *pErrorCode); 166 167 /** 168 * Sets a value for each code point [start..end]. 169 * Faster and more space-efficient than setting the value for each code point separately. 170 * 171 * @param trie the trie 172 * @param start the first code point to get the value 173 * @param end the last code point to get the value (inclusive) 174 * @param value the value 175 * @param pErrorCode an in/out ICU UErrorCode 176 * @stable ICU 63 177 */ 178 U_CAPI void U_EXPORT2 179 umutablecptrie_setRange(UMutableCPTrie *trie, 180 UChar32 start, UChar32 end, 181 uint32_t value, UErrorCode *pErrorCode); 182 183 /** 184 * Compacts the data and builds an immutable UCPTrie according to the parameters. 185 * After this, the mutable trie will be empty. 186 * 187 * The mutable trie stores 32-bit values until buildImmutable() is called. 188 * If values shorter than 32 bits are to be stored in the immutable trie, 189 * then the upper bits are discarded. 190 * For example, when the mutable trie contains values 0x81, -0x7f, and 0xa581, 191 * and the value width is 8 bits, then each of these is stored as 0x81 192 * and the immutable trie will return that as an unsigned value. 193 * (Some implementations may want to make productive temporary use of the upper bits 194 * until buildImmutable() discards them.) 195 * 196 * Not every possible set of mappings can be built into a UCPTrie, 197 * because of limitations resulting from speed and space optimizations. 198 * Every Unicode assigned character can be mapped to a unique value. 199 * Typical data yields data structures far smaller than the limitations. 200 * 201 * It is possible to construct extremely unusual mappings that exceed the data structure limits. 202 * In such a case this function will fail with a U_INDEX_OUTOFBOUNDS_ERROR. 203 * 204 * @param trie the trie trie 205 * @param type selects the trie type 206 * @param valueWidth selects the number of bits in a trie data value; if smaller than 32 bits, 207 * then the values stored in the trie will be truncated first 208 * @param pErrorCode an in/out ICU UErrorCode 209 * 210 * @see umutablecptrie_fromUCPTrie 211 * @stable ICU 63 212 */ 213 U_CAPI UCPTrie * U_EXPORT2 214 umutablecptrie_buildImmutable(UMutableCPTrie *trie, UCPTrieType type, UCPTrieValueWidth valueWidth, 215 UErrorCode *pErrorCode); 216 217 U_CDECL_END 218 219 #if U_SHOW_CPLUSPLUS_API 220 221 U_NAMESPACE_BEGIN 222 223 /** 224 * \class LocalUMutableCPTriePointer 225 * "Smart pointer" class, closes a UMutableCPTrie via umutablecptrie_close(). 226 * For most methods see the LocalPointerBase base class. 227 * 228 * @see LocalPointerBase 229 * @see LocalPointer 230 * @stable ICU 63 231 */ 232 U_DEFINE_LOCAL_OPEN_POINTER(LocalUMutableCPTriePointer, UMutableCPTrie, umutablecptrie_close); 233 234 U_NAMESPACE_END 235 236 #endif 237 238 #endif 239