1 // Copyright (C) 2016 and later: Unicode, Inc. and others. 2 // License & terms of use: http://www.unicode.org/copyright.html 3 /* 4 ******************************************************************************* 5 * 6 * Copyright (C) 2005-2012, International Business Machines 7 * Corporation and others. All Rights Reserved. 8 * 9 ******************************************************************************* 10 * file name: ucasemap.h 11 * encoding: US-ASCII 12 * tab size: 8 (not used) 13 * indentation:4 14 * 15 * created on: 2005may06 16 * created by: Markus W. Scherer 17 * 18 * Case mapping service object and functions using it. 19 */ 20 21 #ifndef __UCASEMAP_H__ 22 #define __UCASEMAP_H__ 23 24 #include "unicode/utypes.h" 25 #include "unicode/ustring.h" 26 #include "unicode/localpointer.h" 27 28 /** 29 * \file 30 * \brief C API: Unicode case mapping functions using a UCaseMap service object. 31 * 32 * The service object takes care of memory allocations, data loading, and setup 33 * for the attributes, as usual. 34 * 35 * Currently, the functionality provided here does not overlap with uchar.h 36 * and ustring.h, except for ucasemap_toTitle(). 37 * 38 * ucasemap_utf8XYZ() functions operate directly on UTF-8 strings. 39 */ 40 41 /** 42 * UCaseMap is an opaque service object for newer ICU case mapping functions. 43 * Older functions did not use a service object. 44 * @stable ICU 3.4 45 */ 46 struct UCaseMap; 47 typedef struct UCaseMap UCaseMap; /**< C typedef for struct UCaseMap. @stable ICU 3.4 */ 48 49 /** 50 * Open a UCaseMap service object for a locale and a set of options. 51 * The locale ID and options are preprocessed so that functions using the 52 * service object need not process them in each call. 53 * 54 * @param locale ICU locale ID, used for language-dependent 55 * upper-/lower-/title-casing according to the Unicode standard. 56 * Usual semantics: ""=root, NULL=default locale, etc. 57 * @param options Options bit set, used for case folding and string comparisons. 58 * Same flags as for u_foldCase(), u_strFoldCase(), 59 * u_strCaseCompare(), etc. 60 * Use 0 or U_FOLD_CASE_DEFAULT for default behavior. 61 * @param pErrorCode Must be a valid pointer to an error code value, 62 * which must not indicate a failure before the function call. 63 * @return Pointer to a UCaseMap service object, if successful. 64 * 65 * @see U_FOLD_CASE_DEFAULT 66 * @see U_FOLD_CASE_EXCLUDE_SPECIAL_I 67 * @see U_TITLECASE_NO_LOWERCASE 68 * @see U_TITLECASE_NO_BREAK_ADJUSTMENT 69 * @stable ICU 3.4 70 */ 71 U_STABLE UCaseMap * U_EXPORT2 72 ucasemap_open(const char *locale, uint32_t options, UErrorCode *pErrorCode); 73 74 /** 75 * Close a UCaseMap service object. 76 * @param csm Object to be closed. 77 * @stable ICU 3.4 78 */ 79 U_STABLE void U_EXPORT2 80 ucasemap_close(UCaseMap *csm); 81 82 #if U_SHOW_CPLUSPLUS_API 83 84 U_NAMESPACE_BEGIN 85 86 /** 87 * \class LocalUCaseMapPointer 88 * "Smart pointer" class, closes a UCaseMap via ucasemap_close(). 89 * For most methods see the LocalPointerBase base class. 90 * 91 * @see LocalPointerBase 92 * @see LocalPointer 93 * @stable ICU 4.4 94 */ 95 U_DEFINE_LOCAL_OPEN_POINTER(LocalUCaseMapPointer, UCaseMap, ucasemap_close); 96 97 U_NAMESPACE_END 98 99 #endif 100 101 /** 102 * Get the locale ID that is used for language-dependent case mappings. 103 * @param csm UCaseMap service object. 104 * @return locale ID 105 * @stable ICU 3.4 106 */ 107 U_STABLE const char * U_EXPORT2 108 ucasemap_getLocale(const UCaseMap *csm); 109 110 /** 111 * Get the options bit set that is used for case folding and string comparisons. 112 * @param csm UCaseMap service object. 113 * @return options bit set 114 * @stable ICU 3.4 115 */ 116 U_STABLE uint32_t U_EXPORT2 117 ucasemap_getOptions(const UCaseMap *csm); 118 119 /** 120 * Set the locale ID that is used for language-dependent case mappings. 121 * 122 * @param csm UCaseMap service object. 123 * @param locale Locale ID, see ucasemap_open(). 124 * @param pErrorCode Must be a valid pointer to an error code value, 125 * which must not indicate a failure before the function call. 126 * 127 * @see ucasemap_open 128 * @stable ICU 3.4 129 */ 130 U_STABLE void U_EXPORT2 131 ucasemap_setLocale(UCaseMap *csm, const char *locale, UErrorCode *pErrorCode); 132 133 /** 134 * Set the options bit set that is used for case folding and string comparisons. 135 * 136 * @param csm UCaseMap service object. 137 * @param options Options bit set, see ucasemap_open(). 138 * @param pErrorCode Must be a valid pointer to an error code value, 139 * which must not indicate a failure before the function call. 140 * 141 * @see ucasemap_open 142 * @stable ICU 3.4 143 */ 144 U_STABLE void U_EXPORT2 145 ucasemap_setOptions(UCaseMap *csm, uint32_t options, UErrorCode *pErrorCode); 146 147 /** 148 * Do not lowercase non-initial parts of words when titlecasing. 149 * Option bit for titlecasing APIs that take an options bit set. 150 * 151 * By default, titlecasing will titlecase the first cased character 152 * of a word and lowercase all other characters. 153 * With this option, the other characters will not be modified. 154 * 155 * @see ucasemap_setOptions 156 * @see ucasemap_toTitle 157 * @see ucasemap_utf8ToTitle 158 * @see UnicodeString::toTitle 159 * @stable ICU 3.8 160 */ 161 #define U_TITLECASE_NO_LOWERCASE 0x100 162 163 /** 164 * Do not adjust the titlecasing indexes from BreakIterator::next() indexes; 165 * titlecase exactly the characters at breaks from the iterator. 166 * Option bit for titlecasing APIs that take an options bit set. 167 * 168 * By default, titlecasing will take each break iterator index, 169 * adjust it by looking for the next cased character, and titlecase that one. 170 * Other characters are lowercased. 171 * 172 * This follows Unicode 4 & 5 section 3.13 Default Case Operations: 173 * 174 * R3 toTitlecase(X): Find the word boundaries based on Unicode Standard Annex 175 * #29, "Text Boundaries." Between each pair of word boundaries, find the first 176 * cased character F. If F exists, map F to default_title(F); then map each 177 * subsequent character C to default_lower(C). 178 * 179 * @see ucasemap_setOptions 180 * @see ucasemap_toTitle 181 * @see ucasemap_utf8ToTitle 182 * @see UnicodeString::toTitle 183 * @see U_TITLECASE_NO_LOWERCASE 184 * @stable ICU 3.8 185 */ 186 #define U_TITLECASE_NO_BREAK_ADJUSTMENT 0x200 187 188 #if !UCONFIG_NO_BREAK_ITERATION 189 190 /** 191 * Get the break iterator that is used for titlecasing. 192 * Do not modify the returned break iterator. 193 * @param csm UCaseMap service object. 194 * @return titlecasing break iterator 195 * @stable ICU 3.8 196 */ 197 U_STABLE const UBreakIterator * U_EXPORT2 198 ucasemap_getBreakIterator(const UCaseMap *csm); 199 200 /** 201 * Set the break iterator that is used for titlecasing. 202 * The UCaseMap service object releases a previously set break iterator 203 * and "adopts" this new one, taking ownership of it. 204 * It will be released in a subsequent call to ucasemap_setBreakIterator() 205 * or ucasemap_close(). 206 * 207 * Break iterator operations are not thread-safe. Therefore, titlecasing 208 * functions use non-const UCaseMap objects. It is not possible to titlecase 209 * strings concurrently using the same UCaseMap. 210 * 211 * @param csm UCaseMap service object. 212 * @param iterToAdopt Break iterator to be adopted for titlecasing. 213 * @param pErrorCode Must be a valid pointer to an error code value, 214 * which must not indicate a failure before the function call. 215 * 216 * @see ucasemap_toTitle 217 * @see ucasemap_utf8ToTitle 218 * @stable ICU 3.8 219 */ 220 U_STABLE void U_EXPORT2 221 ucasemap_setBreakIterator(UCaseMap *csm, UBreakIterator *iterToAdopt, UErrorCode *pErrorCode); 222 223 /** 224 * Titlecase a UTF-16 string. This function is almost a duplicate of u_strToTitle(), 225 * except that it takes ucasemap_setOptions() into account and has performance 226 * advantages from being able to use a UCaseMap object for multiple case mapping 227 * operations, saving setup time. 228 * 229 * Casing is locale-dependent and context-sensitive. 230 * Titlecasing uses a break iterator to find the first characters of words 231 * that are to be titlecased. It titlecases those characters and lowercases 232 * all others. (This can be modified with ucasemap_setOptions().) 233 * 234 * Note: This function takes a non-const UCaseMap pointer because it will 235 * open a default break iterator if no break iterator was set yet, 236 * and effectively call ucasemap_setBreakIterator(); 237 * also because the break iterator is stateful and will be modified during 238 * the iteration. 239 * 240 * The titlecase break iterator can be provided to customize for arbitrary 241 * styles, using rules and dictionaries beyond the standard iterators. 242 * The standard titlecase iterator for the root locale implements the 243 * algorithm of Unicode TR 21. 244 * 245 * This function uses only the setUText(), first(), next() and close() methods of the 246 * provided break iterator. 247 * 248 * The result may be longer or shorter than the original. 249 * The source string and the destination buffer must not overlap. 250 * 251 * @param csm UCaseMap service object. This pointer is non-const! 252 * See the note above for details. 253 * @param dest A buffer for the result string. The result will be NUL-terminated if 254 * the buffer is large enough. 255 * The contents is undefined in case of failure. 256 * @param destCapacity The size of the buffer (number of bytes). If it is 0, then 257 * dest may be NULL and the function will only return the length of the result 258 * without writing any of the result string. 259 * @param src The original string. 260 * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. 261 * @param pErrorCode Must be a valid pointer to an error code value, 262 * which must not indicate a failure before the function call. 263 * @return The length of the result string, if successful - or in case of a buffer overflow, 264 * in which case it will be greater than destCapacity. 265 * 266 * @see u_strToTitle 267 * @stable ICU 3.8 268 */ 269 U_STABLE int32_t U_EXPORT2 270 ucasemap_toTitle(UCaseMap *csm, 271 UChar *dest, int32_t destCapacity, 272 const UChar *src, int32_t srcLength, 273 UErrorCode *pErrorCode); 274 275 #endif 276 277 /** 278 * Lowercase the characters in a UTF-8 string. 279 * Casing is locale-dependent and context-sensitive. 280 * The result may be longer or shorter than the original. 281 * The source string and the destination buffer must not overlap. 282 * 283 * @param csm UCaseMap service object. 284 * @param dest A buffer for the result string. The result will be NUL-terminated if 285 * the buffer is large enough. 286 * The contents is undefined in case of failure. 287 * @param destCapacity The size of the buffer (number of bytes). If it is 0, then 288 * dest may be NULL and the function will only return the length of the result 289 * without writing any of the result string. 290 * @param src The original string. 291 * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. 292 * @param pErrorCode Must be a valid pointer to an error code value, 293 * which must not indicate a failure before the function call. 294 * @return The length of the result string, if successful - or in case of a buffer overflow, 295 * in which case it will be greater than destCapacity. 296 * 297 * @see u_strToLower 298 * @stable ICU 3.4 299 */ 300 U_STABLE int32_t U_EXPORT2 301 ucasemap_utf8ToLower(const UCaseMap *csm, 302 char *dest, int32_t destCapacity, 303 const char *src, int32_t srcLength, 304 UErrorCode *pErrorCode); 305 306 /** 307 * Uppercase the characters in a UTF-8 string. 308 * Casing is locale-dependent and context-sensitive. 309 * The result may be longer or shorter than the original. 310 * The source string and the destination buffer must not overlap. 311 * 312 * @param csm UCaseMap service object. 313 * @param dest A buffer for the result string. The result will be NUL-terminated if 314 * the buffer is large enough. 315 * The contents is undefined in case of failure. 316 * @param destCapacity The size of the buffer (number of bytes). If it is 0, then 317 * dest may be NULL and the function will only return the length of the result 318 * without writing any of the result string. 319 * @param src The original string. 320 * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. 321 * @param pErrorCode Must be a valid pointer to an error code value, 322 * which must not indicate a failure before the function call. 323 * @return The length of the result string, if successful - or in case of a buffer overflow, 324 * in which case it will be greater than destCapacity. 325 * 326 * @see u_strToUpper 327 * @stable ICU 3.4 328 */ 329 U_STABLE int32_t U_EXPORT2 330 ucasemap_utf8ToUpper(const UCaseMap *csm, 331 char *dest, int32_t destCapacity, 332 const char *src, int32_t srcLength, 333 UErrorCode *pErrorCode); 334 335 #if !UCONFIG_NO_BREAK_ITERATION 336 337 /** 338 * Titlecase a UTF-8 string. 339 * Casing is locale-dependent and context-sensitive. 340 * Titlecasing uses a break iterator to find the first characters of words 341 * that are to be titlecased. It titlecases those characters and lowercases 342 * all others. (This can be modified with ucasemap_setOptions().) 343 * 344 * Note: This function takes a non-const UCaseMap pointer because it will 345 * open a default break iterator if no break iterator was set yet, 346 * and effectively call ucasemap_setBreakIterator(); 347 * also because the break iterator is stateful and will be modified during 348 * the iteration. 349 * 350 * The titlecase break iterator can be provided to customize for arbitrary 351 * styles, using rules and dictionaries beyond the standard iterators. 352 * The standard titlecase iterator for the root locale implements the 353 * algorithm of Unicode TR 21. 354 * 355 * This function uses only the setUText(), first(), next() and close() methods of the 356 * provided break iterator. 357 * 358 * The result may be longer or shorter than the original. 359 * The source string and the destination buffer must not overlap. 360 * 361 * @param csm UCaseMap service object. This pointer is non-const! 362 * See the note above for details. 363 * @param dest A buffer for the result string. The result will be NUL-terminated if 364 * the buffer is large enough. 365 * The contents is undefined in case of failure. 366 * @param destCapacity The size of the buffer (number of bytes). If it is 0, then 367 * dest may be NULL and the function will only return the length of the result 368 * without writing any of the result string. 369 * @param src The original string. 370 * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. 371 * @param pErrorCode Must be a valid pointer to an error code value, 372 * which must not indicate a failure before the function call. 373 * @return The length of the result string, if successful - or in case of a buffer overflow, 374 * in which case it will be greater than destCapacity. 375 * 376 * @see u_strToTitle 377 * @see U_TITLECASE_NO_LOWERCASE 378 * @see U_TITLECASE_NO_BREAK_ADJUSTMENT 379 * @stable ICU 3.8 380 */ 381 U_STABLE int32_t U_EXPORT2 382 ucasemap_utf8ToTitle(UCaseMap *csm, 383 char *dest, int32_t destCapacity, 384 const char *src, int32_t srcLength, 385 UErrorCode *pErrorCode); 386 387 #endif 388 389 /** 390 * Case-folds the characters in a UTF-8 string. 391 * 392 * Case-folding is locale-independent and not context-sensitive, 393 * but there is an option for whether to include or exclude mappings for dotted I 394 * and dotless i that are marked with 'T' in CaseFolding.txt. 395 * 396 * The result may be longer or shorter than the original. 397 * The source string and the destination buffer must not overlap. 398 * 399 * @param csm UCaseMap service object. 400 * @param dest A buffer for the result string. The result will be NUL-terminated if 401 * the buffer is large enough. 402 * The contents is undefined in case of failure. 403 * @param destCapacity The size of the buffer (number of bytes). If it is 0, then 404 * dest may be NULL and the function will only return the length of the result 405 * without writing any of the result string. 406 * @param src The original string. 407 * @param srcLength The length of the original string. If -1, then src must be NUL-terminated. 408 * @param pErrorCode Must be a valid pointer to an error code value, 409 * which must not indicate a failure before the function call. 410 * @return The length of the result string, if successful - or in case of a buffer overflow, 411 * in which case it will be greater than destCapacity. 412 * 413 * @see u_strFoldCase 414 * @see ucasemap_setOptions 415 * @see U_FOLD_CASE_DEFAULT 416 * @see U_FOLD_CASE_EXCLUDE_SPECIAL_I 417 * @stable ICU 3.8 418 */ 419 U_STABLE int32_t U_EXPORT2 420 ucasemap_utf8FoldCase(const UCaseMap *csm, 421 char *dest, int32_t destCapacity, 422 const char *src, int32_t srcLength, 423 UErrorCode *pErrorCode); 424 425 #endif 426