1 /* 2 ******************************************************************************* 3 * Copyright (c) 1996-2015, International Business Machines Corporation and others. 4 * All Rights Reserved. 5 ******************************************************************************* 6 */ 7 8 #ifndef UCOL_H 9 #define UCOL_H 10 11 #include "unicode/utypes.h" 12 13 #if !UCONFIG_NO_COLLATION 14 15 #include "unicode/unorm.h" 16 #include "unicode/localpointer.h" 17 #include "unicode/parseerr.h" 18 #include "unicode/uloc.h" 19 #include "unicode/uset.h" 20 #include "unicode/uscript.h" 21 22 /** 23 * \file 24 * \brief C API: Collator 25 * 26 * <h2> Collator C API </h2> 27 * 28 * The C API for Collator performs locale-sensitive 29 * string comparison. You use this service to build 30 * searching and sorting routines for natural language text. 31 * <p> 32 * For more information about the collation service see 33 * <a href="http://userguide.icu-project.org/collation">the User Guide</a>. 34 * <p> 35 * Collation service provides correct sorting orders for most locales supported in ICU. 36 * If specific data for a locale is not available, the orders eventually falls back 37 * to the <a href="http://www.unicode.org/reports/tr35/tr35-collation.html#Root_Collation">CLDR root sort order</a>. 38 * <p> 39 * Sort ordering may be customized by providing your own set of rules. For more on 40 * this subject see the <a href="http://userguide.icu-project.org/collation/customization"> 41 * Collation Customization</a> section of the User Guide. 42 * <p> 43 * @see UCollationResult 44 * @see UNormalizationMode 45 * @see UCollationStrength 46 * @see UCollationElements 47 */ 48 49 /** A collator. 50 * For usage in C programs. 51 */ 52 struct UCollator; 53 /** structure representing a collator object instance 54 * @stable ICU 2.0 55 */ 56 typedef struct UCollator UCollator; 57 58 59 /** 60 * UCOL_LESS is returned if source string is compared to be less than target 61 * string in the ucol_strcoll() method. 62 * UCOL_EQUAL is returned if source string is compared to be equal to target 63 * string in the ucol_strcoll() method. 64 * UCOL_GREATER is returned if source string is compared to be greater than 65 * target string in the ucol_strcoll() method. 66 * @see ucol_strcoll() 67 * <p> 68 * Possible values for a comparison result 69 * @stable ICU 2.0 70 */ 71 typedef enum { 72 /** string a == string b */ 73 UCOL_EQUAL = 0, 74 /** string a > string b */ 75 UCOL_GREATER = 1, 76 /** string a < string b */ 77 UCOL_LESS = -1 78 } UCollationResult ; 79 80 81 /** Enum containing attribute values for controling collation behavior. 82 * Here are all the allowable values. Not every attribute can take every value. The only 83 * universal value is UCOL_DEFAULT, which resets the attribute value to the predefined 84 * value for that locale 85 * @stable ICU 2.0 86 */ 87 typedef enum { 88 /** accepted by most attributes */ 89 UCOL_DEFAULT = -1, 90 91 /** Primary collation strength */ 92 UCOL_PRIMARY = 0, 93 /** Secondary collation strength */ 94 UCOL_SECONDARY = 1, 95 /** Tertiary collation strength */ 96 UCOL_TERTIARY = 2, 97 /** Default collation strength */ 98 UCOL_DEFAULT_STRENGTH = UCOL_TERTIARY, 99 UCOL_CE_STRENGTH_LIMIT, 100 /** Quaternary collation strength */ 101 UCOL_QUATERNARY=3, 102 /** Identical collation strength */ 103 UCOL_IDENTICAL=15, 104 UCOL_STRENGTH_LIMIT, 105 106 /** Turn the feature off - works for UCOL_FRENCH_COLLATION, 107 UCOL_CASE_LEVEL, UCOL_HIRAGANA_QUATERNARY_MODE 108 & UCOL_DECOMPOSITION_MODE*/ 109 UCOL_OFF = 16, 110 /** Turn the feature on - works for UCOL_FRENCH_COLLATION, 111 UCOL_CASE_LEVEL, UCOL_HIRAGANA_QUATERNARY_MODE 112 & UCOL_DECOMPOSITION_MODE*/ 113 UCOL_ON = 17, 114 115 /** Valid for UCOL_ALTERNATE_HANDLING. Alternate handling will be shifted */ 116 UCOL_SHIFTED = 20, 117 /** Valid for UCOL_ALTERNATE_HANDLING. Alternate handling will be non ignorable */ 118 UCOL_NON_IGNORABLE = 21, 119 120 /** Valid for UCOL_CASE_FIRST - 121 lower case sorts before upper case */ 122 UCOL_LOWER_FIRST = 24, 123 /** upper case sorts before lower case */ 124 UCOL_UPPER_FIRST = 25, 125 126 UCOL_ATTRIBUTE_VALUE_COUNT 127 128 } UColAttributeValue; 129 130 /** 131 * Enum containing the codes for reordering segments of the collation table that are not script 132 * codes. These reordering codes are to be used in conjunction with the script codes. 133 * @see ucol_getReorderCodes 134 * @see ucol_setReorderCodes 135 * @see ucol_getEquivalentReorderCodes 136 * @see UScriptCode 137 * @stable ICU 4.8 138 */ 139 typedef enum { 140 /** 141 * A special reordering code that is used to specify the default 142 * reordering codes for a locale. 143 * @stable ICU 4.8 144 */ 145 UCOL_REORDER_CODE_DEFAULT = -1, 146 /** 147 * A special reordering code that is used to specify no reordering codes. 148 * @stable ICU 4.8 149 */ 150 UCOL_REORDER_CODE_NONE = USCRIPT_UNKNOWN, 151 /** 152 * A special reordering code that is used to specify all other codes used for 153 * reordering except for the codes lised as UColReorderCode values and those 154 * listed explicitly in a reordering. 155 * @stable ICU 4.8 156 */ 157 UCOL_REORDER_CODE_OTHERS = USCRIPT_UNKNOWN, 158 /** 159 * Characters with the space property. 160 * This is equivalent to the rule value "space". 161 * @stable ICU 4.8 162 */ 163 UCOL_REORDER_CODE_SPACE = 0x1000, 164 /** 165 * The first entry in the enumeration of reordering groups. This is intended for use in 166 * range checking and enumeration of the reorder codes. 167 * @stable ICU 4.8 168 */ 169 UCOL_REORDER_CODE_FIRST = UCOL_REORDER_CODE_SPACE, 170 /** 171 * Characters with the punctuation property. 172 * This is equivalent to the rule value "punct". 173 * @stable ICU 4.8 174 */ 175 UCOL_REORDER_CODE_PUNCTUATION = 0x1001, 176 /** 177 * Characters with the symbol property. 178 * This is equivalent to the rule value "symbol". 179 * @stable ICU 4.8 180 */ 181 UCOL_REORDER_CODE_SYMBOL = 0x1002, 182 /** 183 * Characters with the currency property. 184 * This is equivalent to the rule value "currency". 185 * @stable ICU 4.8 186 */ 187 UCOL_REORDER_CODE_CURRENCY = 0x1003, 188 /** 189 * Characters with the digit property. 190 * This is equivalent to the rule value "digit". 191 * @stable ICU 4.8 192 */ 193 UCOL_REORDER_CODE_DIGIT = 0x1004, 194 /** 195 * The limit of the reorder codes. This is intended for use in range checking 196 * and enumeration of the reorder codes. 197 * @stable ICU 4.8 198 */ 199 UCOL_REORDER_CODE_LIMIT = 0x1005 200 } UColReorderCode; 201 202 /** 203 * Base letter represents a primary difference. Set comparison 204 * level to UCOL_PRIMARY to ignore secondary and tertiary differences. 205 * Use this to set the strength of a Collator object. 206 * Example of primary difference, "abc" < "abd" 207 * 208 * Diacritical differences on the same base letter represent a secondary 209 * difference. Set comparison level to UCOL_SECONDARY to ignore tertiary 210 * differences. Use this to set the strength of a Collator object. 211 * Example of secondary difference, "ä" >> "a". 212 * 213 * Uppercase and lowercase versions of the same character represents a 214 * tertiary difference. Set comparison level to UCOL_TERTIARY to include 215 * all comparison differences. Use this to set the strength of a Collator 216 * object. 217 * Example of tertiary difference, "abc" <<< "ABC". 218 * 219 * Two characters are considered "identical" when they have the same 220 * unicode spellings. UCOL_IDENTICAL. 221 * For example, "ä" == "ä". 222 * 223 * UCollationStrength is also used to determine the strength of sort keys 224 * generated from UCollator objects 225 * These values can be now found in the UColAttributeValue enum. 226 * @stable ICU 2.0 227 **/ 228 typedef UColAttributeValue UCollationStrength; 229 230 /** Attributes that collation service understands. All the attributes can take UCOL_DEFAULT 231 * value, as well as the values specific to each one. 232 * @stable ICU 2.0 233 */ 234 typedef enum { 235 /** Attribute for direction of secondary weights - used in Canadian French. 236 * Acceptable values are UCOL_ON, which results in secondary weights 237 * being considered backwards and UCOL_OFF which treats secondary 238 * weights in the order they appear. 239 * @stable ICU 2.0 240 */ 241 UCOL_FRENCH_COLLATION, 242 /** Attribute for handling variable elements. 243 * Acceptable values are UCOL_NON_IGNORABLE (default) 244 * which treats all the codepoints with non-ignorable 245 * primary weights in the same way, 246 * and UCOL_SHIFTED which causes codepoints with primary 247 * weights that are equal or below the variable top value 248 * to be ignored on primary level and moved to the quaternary 249 * level. 250 * @stable ICU 2.0 251 */ 252 UCOL_ALTERNATE_HANDLING, 253 /** Controls the ordering of upper and lower case letters. 254 * Acceptable values are UCOL_OFF (default), which orders 255 * upper and lower case letters in accordance to their tertiary 256 * weights, UCOL_UPPER_FIRST which forces upper case letters to 257 * sort before lower case letters, and UCOL_LOWER_FIRST which does 258 * the opposite. 259 * @stable ICU 2.0 260 */ 261 UCOL_CASE_FIRST, 262 /** Controls whether an extra case level (positioned before the third 263 * level) is generated or not. Acceptable values are UCOL_OFF (default), 264 * when case level is not generated, and UCOL_ON which causes the case 265 * level to be generated. Contents of the case level are affected by 266 * the value of UCOL_CASE_FIRST attribute. A simple way to ignore 267 * accent differences in a string is to set the strength to UCOL_PRIMARY 268 * and enable case level. 269 * @stable ICU 2.0 270 */ 271 UCOL_CASE_LEVEL, 272 /** Controls whether the normalization check and necessary normalizations 273 * are performed. When set to UCOL_OFF (default) no normalization check 274 * is performed. The correctness of the result is guaranteed only if the 275 * input data is in so-called FCD form (see users manual for more info). 276 * When set to UCOL_ON, an incremental check is performed to see whether 277 * the input data is in the FCD form. If the data is not in the FCD form, 278 * incremental NFD normalization is performed. 279 * @stable ICU 2.0 280 */ 281 UCOL_NORMALIZATION_MODE, 282 /** An alias for UCOL_NORMALIZATION_MODE attribute. 283 * @stable ICU 2.0 284 */ 285 UCOL_DECOMPOSITION_MODE = UCOL_NORMALIZATION_MODE, 286 /** The strength attribute. Can be either UCOL_PRIMARY, UCOL_SECONDARY, 287 * UCOL_TERTIARY, UCOL_QUATERNARY or UCOL_IDENTICAL. The usual strength 288 * for most locales (except Japanese) is tertiary. 289 * 290 * Quaternary strength 291 * is useful when combined with shifted setting for alternate handling 292 * attribute and for JIS X 4061 collation, when it is used to distinguish 293 * between Katakana and Hiragana. 294 * Otherwise, quaternary level 295 * is affected only by the number of non-ignorable code points in 296 * the string. 297 * 298 * Identical strength is rarely useful, as it amounts 299 * to codepoints of the NFD form of the string. 300 * @stable ICU 2.0 301 */ 302 UCOL_STRENGTH, 303 #ifndef U_HIDE_DEPRECATED_API 304 /** When turned on, this attribute positions Hiragana before all 305 * non-ignorables on quaternary level This is a sneaky way to produce JIS 306 * sort order. 307 * 308 * This attribute was an implementation detail of the CLDR Japanese tailoring. 309 * Since ICU 50, this attribute is not settable any more via API functions. 310 * Since CLDR 25/ICU 53, explicit quaternary relations are used 311 * to achieve the same Japanese sort order. 312 * 313 * @deprecated ICU 50 Implementation detail, cannot be set via API, was removed from implementation. 314 */ 315 UCOL_HIRAGANA_QUATERNARY_MODE = UCOL_STRENGTH + 1, 316 #endif /* U_HIDE_DEPRECATED_API */ 317 /** 318 * When turned on, this attribute makes 319 * substrings of digits sort according to their numeric values. 320 * 321 * This is a way to get '100' to sort AFTER '2'. Note that the longest 322 * digit substring that can be treated as a single unit is 323 * 254 digits (not counting leading zeros). If a digit substring is 324 * longer than that, the digits beyond the limit will be treated as a 325 * separate digit substring. 326 * 327 * A "digit" in this sense is a code point with General_Category=Nd, 328 * which does not include circled numbers, roman numerals, etc. 329 * Only a contiguous digit substring is considered, that is, 330 * non-negative integers without separators. 331 * There is no support for plus/minus signs, decimals, exponents, etc. 332 * 333 * @stable ICU 2.8 334 */ 335 UCOL_NUMERIC_COLLATION = UCOL_STRENGTH + 2, 336 /** 337 * The number of UColAttribute constants. 338 * @stable ICU 2.0 339 */ 340 UCOL_ATTRIBUTE_COUNT 341 } UColAttribute; 342 343 /** Options for retrieving the rule string 344 * @stable ICU 2.0 345 */ 346 typedef enum { 347 /** 348 * Retrieves the tailoring rules only. 349 * Same as calling the version of getRules() without UColRuleOption. 350 * @stable ICU 2.0 351 */ 352 UCOL_TAILORING_ONLY, 353 /** 354 * Retrieves the "UCA rules" concatenated with the tailoring rules. 355 * The "UCA rules" are an <i>approximation</i> of the root collator's sort order. 356 * They are almost never used or useful at runtime and can be removed from the data. 357 * See http://userguide.icu-project.org/collation/customization#TOC-Building-on-Existing-Locales 358 * @stable ICU 2.0 359 */ 360 UCOL_FULL_RULES 361 } UColRuleOption ; 362 363 /** 364 * Open a UCollator for comparing strings. 365 * 366 * For some languages, multiple collation types are available; 367 * for example, "de@collation=phonebook". 368 * Starting with ICU 54, collation attributes can be specified via locale keywords as well, 369 * in the old locale extension syntax ("el@colCaseFirst=upper") 370 * or in language tag syntax ("el-u-kf-upper"). 371 * See <a href="http://userguide.icu-project.org/collation/api">User Guide: Collation API</a>. 372 * 373 * The UCollator pointer is used in all the calls to the Collation 374 * service. After finished, collator must be disposed of by calling 375 * {@link #ucol_close }. 376 * @param loc The locale containing the required collation rules. 377 * Special values for locales can be passed in - 378 * if NULL is passed for the locale, the default locale 379 * collation rules will be used. If empty string ("") or 380 * "root" are passed, the root collator will be returned. 381 * @param status A pointer to a UErrorCode to receive any errors 382 * @return A pointer to a UCollator, or 0 if an error occurred. 383 * @see ucol_openRules 384 * @see ucol_safeClone 385 * @see ucol_close 386 * @stable ICU 2.0 387 */ 388 U_STABLE UCollator* U_EXPORT2 389 ucol_open(const char *loc, UErrorCode *status); 390 391 /** 392 * Produce a UCollator instance according to the rules supplied. 393 * The rules are used to change the default ordering, defined in the 394 * UCA in a process called tailoring. The resulting UCollator pointer 395 * can be used in the same way as the one obtained by {@link #ucol_strcoll }. 396 * @param rules A string describing the collation rules. For the syntax 397 * of the rules please see users guide. 398 * @param rulesLength The length of rules, or -1 if null-terminated. 399 * @param normalizationMode The normalization mode: One of 400 * UCOL_OFF (expect the text to not need normalization), 401 * UCOL_ON (normalize), or 402 * UCOL_DEFAULT (set the mode according to the rules) 403 * @param strength The default collation strength; one of UCOL_PRIMARY, UCOL_SECONDARY, 404 * UCOL_TERTIARY, UCOL_IDENTICAL,UCOL_DEFAULT_STRENGTH - can be also set in the rules. 405 * @param parseError A pointer to UParseError to recieve information about errors 406 * occurred during parsing. This argument can currently be set 407 * to NULL, but at users own risk. Please provide a real structure. 408 * @param status A pointer to a UErrorCode to receive any errors 409 * @return A pointer to a UCollator. It is not guaranteed that NULL be returned in case 410 * of error - please use status argument to check for errors. 411 * @see ucol_open 412 * @see ucol_safeClone 413 * @see ucol_close 414 * @stable ICU 2.0 415 */ 416 U_STABLE UCollator* U_EXPORT2 417 ucol_openRules( const UChar *rules, 418 int32_t rulesLength, 419 UColAttributeValue normalizationMode, 420 UCollationStrength strength, 421 UParseError *parseError, 422 UErrorCode *status); 423 424 #ifndef U_HIDE_DEPRECATED_API 425 /** 426 * Open a collator defined by a short form string. 427 * The structure and the syntax of the string is defined in the "Naming collators" 428 * section of the users guide: 429 * http://userguide.icu-project.org/collation/concepts#TOC-Collator-naming-scheme 430 * Attributes are overriden by the subsequent attributes. So, for "S2_S3", final 431 * strength will be 3. 3066bis locale overrides individual locale parts. 432 * The call to this function is equivalent to a call to ucol_open, followed by a 433 * series of calls to ucol_setAttribute and ucol_setVariableTop. 434 * @param definition A short string containing a locale and a set of attributes. 435 * Attributes not explicitly mentioned are left at the default 436 * state for a locale. 437 * @param parseError if not NULL, structure that will get filled with error's pre 438 * and post context in case of error. 439 * @param forceDefaults if FALSE, the settings that are the same as the collator 440 * default settings will not be applied (for example, setting 441 * French secondary on a French collator would not be executed). 442 * If TRUE, all the settings will be applied regardless of the 443 * collator default value. If the definition 444 * strings are to be cached, should be set to FALSE. 445 * @param status Error code. Apart from regular error conditions connected to 446 * instantiating collators (like out of memory or similar), this 447 * API will return an error if an invalid attribute or attribute/value 448 * combination is specified. 449 * @return A pointer to a UCollator or 0 if an error occured (including an 450 * invalid attribute). 451 * @see ucol_open 452 * @see ucol_setAttribute 453 * @see ucol_setVariableTop 454 * @see ucol_getShortDefinitionString 455 * @see ucol_normalizeShortDefinitionString 456 * @deprecated ICU 54 Use ucol_open() with language tag collation keywords instead. 457 */ 458 U_DEPRECATED UCollator* U_EXPORT2 459 ucol_openFromShortString( const char *definition, 460 UBool forceDefaults, 461 UParseError *parseError, 462 UErrorCode *status); 463 #endif /* U_HIDE_DEPRECATED_API */ 464 465 #ifndef U_HIDE_DEPRECATED_API 466 /** 467 * Get a set containing the contractions defined by the collator. The set includes 468 * both the root collator's contractions and the contractions defined by the collator. This set 469 * will contain only strings. If a tailoring explicitly suppresses contractions from 470 * the root collator (like Russian), removed contractions will not be in the resulting set. 471 * @param coll collator 472 * @param conts the set to hold the result. It gets emptied before 473 * contractions are added. 474 * @param status to hold the error code 475 * @return the size of the contraction set 476 * 477 * @deprecated ICU 3.4, use ucol_getContractionsAndExpansions instead 478 */ 479 U_DEPRECATED int32_t U_EXPORT2 480 ucol_getContractions( const UCollator *coll, 481 USet *conts, 482 UErrorCode *status); 483 #endif /* U_HIDE_DEPRECATED_API */ 484 485 /** 486 * Get a set containing the expansions defined by the collator. The set includes 487 * both the root collator's expansions and the expansions defined by the tailoring 488 * @param coll collator 489 * @param contractions if not NULL, the set to hold the contractions 490 * @param expansions if not NULL, the set to hold the expansions 491 * @param addPrefixes add the prefix contextual elements to contractions 492 * @param status to hold the error code 493 * 494 * @stable ICU 3.4 495 */ 496 U_STABLE void U_EXPORT2 497 ucol_getContractionsAndExpansions( const UCollator *coll, 498 USet *contractions, USet *expansions, 499 UBool addPrefixes, UErrorCode *status); 500 501 /** 502 * Close a UCollator. 503 * Once closed, a UCollator should not be used. Every open collator should 504 * be closed. Otherwise, a memory leak will result. 505 * @param coll The UCollator to close. 506 * @see ucol_open 507 * @see ucol_openRules 508 * @see ucol_safeClone 509 * @stable ICU 2.0 510 */ 511 U_STABLE void U_EXPORT2 512 ucol_close(UCollator *coll); 513 514 #if U_SHOW_CPLUSPLUS_API 515 516 U_NAMESPACE_BEGIN 517 518 /** 519 * \class LocalUCollatorPointer 520 * "Smart pointer" class, closes a UCollator via ucol_close(). 521 * For most methods see the LocalPointerBase base class. 522 * 523 * @see LocalPointerBase 524 * @see LocalPointer 525 * @stable ICU 4.4 526 */ 527 U_DEFINE_LOCAL_OPEN_POINTER(LocalUCollatorPointer, UCollator, ucol_close); 528 529 U_NAMESPACE_END 530 531 #endif 532 533 /** 534 * Compare two strings. 535 * The strings will be compared using the options already specified. 536 * @param coll The UCollator containing the comparison rules. 537 * @param source The source string. 538 * @param sourceLength The length of source, or -1 if null-terminated. 539 * @param target The target string. 540 * @param targetLength The length of target, or -1 if null-terminated. 541 * @return The result of comparing the strings; one of UCOL_EQUAL, 542 * UCOL_GREATER, UCOL_LESS 543 * @see ucol_greater 544 * @see ucol_greaterOrEqual 545 * @see ucol_equal 546 * @stable ICU 2.0 547 */ 548 U_STABLE UCollationResult U_EXPORT2 549 ucol_strcoll( const UCollator *coll, 550 const UChar *source, 551 int32_t sourceLength, 552 const UChar *target, 553 int32_t targetLength); 554 555 /** 556 * Compare two strings in UTF-8. 557 * The strings will be compared using the options already specified. 558 * Note: When input string contains malformed a UTF-8 byte sequence, 559 * this function treats these bytes as REPLACEMENT CHARACTER (U+FFFD). 560 * @param coll The UCollator containing the comparison rules. 561 * @param source The source UTF-8 string. 562 * @param sourceLength The length of source, or -1 if null-terminated. 563 * @param target The target UTF-8 string. 564 * @param targetLength The length of target, or -1 if null-terminated. 565 * @param status A pointer to a UErrorCode to receive any errors 566 * @return The result of comparing the strings; one of UCOL_EQUAL, 567 * UCOL_GREATER, UCOL_LESS 568 * @see ucol_greater 569 * @see ucol_greaterOrEqual 570 * @see ucol_equal 571 * @stable ICU 50 572 */ 573 U_STABLE UCollationResult U_EXPORT2 574 ucol_strcollUTF8( 575 const UCollator *coll, 576 const char *source, 577 int32_t sourceLength, 578 const char *target, 579 int32_t targetLength, 580 UErrorCode *status); 581 582 /** 583 * Determine if one string is greater than another. 584 * This function is equivalent to {@link #ucol_strcoll } == UCOL_GREATER 585 * @param coll The UCollator containing the comparison rules. 586 * @param source The source string. 587 * @param sourceLength The length of source, or -1 if null-terminated. 588 * @param target The target string. 589 * @param targetLength The length of target, or -1 if null-terminated. 590 * @return TRUE if source is greater than target, FALSE otherwise. 591 * @see ucol_strcoll 592 * @see ucol_greaterOrEqual 593 * @see ucol_equal 594 * @stable ICU 2.0 595 */ 596 U_STABLE UBool U_EXPORT2 597 ucol_greater(const UCollator *coll, 598 const UChar *source, int32_t sourceLength, 599 const UChar *target, int32_t targetLength); 600 601 /** 602 * Determine if one string is greater than or equal to another. 603 * This function is equivalent to {@link #ucol_strcoll } != UCOL_LESS 604 * @param coll The UCollator containing the comparison rules. 605 * @param source The source string. 606 * @param sourceLength The length of source, or -1 if null-terminated. 607 * @param target The target string. 608 * @param targetLength The length of target, or -1 if null-terminated. 609 * @return TRUE if source is greater than or equal to target, FALSE otherwise. 610 * @see ucol_strcoll 611 * @see ucol_greater 612 * @see ucol_equal 613 * @stable ICU 2.0 614 */ 615 U_STABLE UBool U_EXPORT2 616 ucol_greaterOrEqual(const UCollator *coll, 617 const UChar *source, int32_t sourceLength, 618 const UChar *target, int32_t targetLength); 619 620 /** 621 * Compare two strings for equality. 622 * This function is equivalent to {@link #ucol_strcoll } == UCOL_EQUAL 623 * @param coll The UCollator containing the comparison rules. 624 * @param source The source string. 625 * @param sourceLength The length of source, or -1 if null-terminated. 626 * @param target The target string. 627 * @param targetLength The length of target, or -1 if null-terminated. 628 * @return TRUE if source is equal to target, FALSE otherwise 629 * @see ucol_strcoll 630 * @see ucol_greater 631 * @see ucol_greaterOrEqual 632 * @stable ICU 2.0 633 */ 634 U_STABLE UBool U_EXPORT2 635 ucol_equal(const UCollator *coll, 636 const UChar *source, int32_t sourceLength, 637 const UChar *target, int32_t targetLength); 638 639 /** 640 * Compare two UTF-8 encoded trings. 641 * The strings will be compared using the options already specified. 642 * @param coll The UCollator containing the comparison rules. 643 * @param sIter The source string iterator. 644 * @param tIter The target string iterator. 645 * @return The result of comparing the strings; one of UCOL_EQUAL, 646 * UCOL_GREATER, UCOL_LESS 647 * @param status A pointer to a UErrorCode to receive any errors 648 * @see ucol_strcoll 649 * @stable ICU 2.6 650 */ 651 U_STABLE UCollationResult U_EXPORT2 652 ucol_strcollIter( const UCollator *coll, 653 UCharIterator *sIter, 654 UCharIterator *tIter, 655 UErrorCode *status); 656 657 /** 658 * Get the collation strength used in a UCollator. 659 * The strength influences how strings are compared. 660 * @param coll The UCollator to query. 661 * @return The collation strength; one of UCOL_PRIMARY, UCOL_SECONDARY, 662 * UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL 663 * @see ucol_setStrength 664 * @stable ICU 2.0 665 */ 666 U_STABLE UCollationStrength U_EXPORT2 667 ucol_getStrength(const UCollator *coll); 668 669 /** 670 * Set the collation strength used in a UCollator. 671 * The strength influences how strings are compared. 672 * @param coll The UCollator to set. 673 * @param strength The desired collation strength; one of UCOL_PRIMARY, 674 * UCOL_SECONDARY, UCOL_TERTIARY, UCOL_QUATERNARY, UCOL_IDENTICAL, UCOL_DEFAULT 675 * @see ucol_getStrength 676 * @stable ICU 2.0 677 */ 678 U_STABLE void U_EXPORT2 679 ucol_setStrength(UCollator *coll, 680 UCollationStrength strength); 681 682 /** 683 * Retrieves the reordering codes for this collator. 684 * These reordering codes are a combination of UScript codes and UColReorderCode entries. 685 * @param coll The UCollator to query. 686 * @param dest The array to fill with the script ordering. 687 * @param destCapacity The length of dest. If it is 0, then dest may be NULL and the function 688 * will only return the length of the result without writing any codes (pre-flighting). 689 * @param pErrorCode Must be a valid pointer to an error code value, which must not indicate a 690 * failure before the function call. 691 * @return The number of reordering codes written to the dest array. 692 * @see ucol_setReorderCodes 693 * @see ucol_getEquivalentReorderCodes 694 * @see UScriptCode 695 * @see UColReorderCode 696 * @stable ICU 4.8 697 */ 698 U_STABLE int32_t U_EXPORT2 699 ucol_getReorderCodes(const UCollator* coll, 700 int32_t* dest, 701 int32_t destCapacity, 702 UErrorCode *pErrorCode); 703 /** 704 * Sets the reordering codes for this collator. 705 * Collation reordering allows scripts and some other groups of characters 706 * to be moved relative to each other. This reordering is done on top of 707 * the DUCET/CLDR standard collation order. Reordering can specify groups to be placed 708 * at the start and/or the end of the collation order. These groups are specified using 709 * UScript codes and UColReorderCode entries. 710 * 711 * <p>By default, reordering codes specified for the start of the order are placed in the 712 * order given after several special non-script blocks. These special groups of characters 713 * are space, punctuation, symbol, currency, and digit. These special groups are represented with 714 * UColReorderCode entries. Script groups can be intermingled with 715 * these special non-script groups if those special groups are explicitly specified in the reordering. 716 * 717 * <p>The special code OTHERS stands for any script that is not explicitly 718 * mentioned in the list of reordering codes given. Anything that is after OTHERS 719 * will go at the very end of the reordering in the order given. 720 * 721 * <p>The special reorder code DEFAULT will reset the reordering for this collator 722 * to the default for this collator. The default reordering may be the DUCET/CLDR order or may be a reordering that 723 * was specified when this collator was created from resource data or from rules. The 724 * DEFAULT code <b>must</b> be the sole code supplied when it is used. 725 * If not, then U_ILLEGAL_ARGUMENT_ERROR will be set. 726 * 727 * <p>The special reorder code NONE will remove any reordering for this collator. 728 * The result of setting no reordering will be to have the DUCET/CLDR ordering used. The 729 * NONE code <b>must</b> be the sole code supplied when it is used. 730 * 731 * @param coll The UCollator to set. 732 * @param reorderCodes An array of script codes in the new order. This can be NULL if the 733 * length is also set to 0. An empty array will clear any reordering codes on the collator. 734 * @param reorderCodesLength The length of reorderCodes. 735 * @param pErrorCode Must be a valid pointer to an error code value, which must not indicate a 736 * failure before the function call. 737 * @see ucol_getReorderCodes 738 * @see ucol_getEquivalentReorderCodes 739 * @see UScriptCode 740 * @see UColReorderCode 741 * @stable ICU 4.8 742 */ 743 U_STABLE void U_EXPORT2 744 ucol_setReorderCodes(UCollator* coll, 745 const int32_t* reorderCodes, 746 int32_t reorderCodesLength, 747 UErrorCode *pErrorCode); 748 749 /** 750 * Retrieves the reorder codes that are grouped with the given reorder code. Some reorder 751 * codes will be grouped and must reorder together. 752 * Beginning with ICU 55, scripts only reorder together if they are primary-equal, 753 * for example Hiragana and Katakana. 754 * 755 * @param reorderCode The reorder code to determine equivalence for. 756 * @param dest The array to fill with the script ordering. 757 * @param destCapacity The length of dest. If it is 0, then dest may be NULL and the function 758 * will only return the length of the result without writing any codes (pre-flighting). 759 * @param pErrorCode Must be a valid pointer to an error code value, which must not indicate 760 * a failure before the function call. 761 * @return The number of reordering codes written to the dest array. 762 * @see ucol_setReorderCodes 763 * @see ucol_getReorderCodes 764 * @see UScriptCode 765 * @see UColReorderCode 766 * @stable ICU 4.8 767 */ 768 U_STABLE int32_t U_EXPORT2 769 ucol_getEquivalentReorderCodes(int32_t reorderCode, 770 int32_t* dest, 771 int32_t destCapacity, 772 UErrorCode *pErrorCode); 773 774 /** 775 * Get the display name for a UCollator. 776 * The display name is suitable for presentation to a user. 777 * @param objLoc The locale of the collator in question. 778 * @param dispLoc The locale for display. 779 * @param result A pointer to a buffer to receive the attribute. 780 * @param resultLength The maximum size of result. 781 * @param status A pointer to a UErrorCode to receive any errors 782 * @return The total buffer size needed; if greater than resultLength, 783 * the output was truncated. 784 * @stable ICU 2.0 785 */ 786 U_STABLE int32_t U_EXPORT2 787 ucol_getDisplayName( const char *objLoc, 788 const char *dispLoc, 789 UChar *result, 790 int32_t resultLength, 791 UErrorCode *status); 792 793 /** 794 * Get a locale for which collation rules are available. 795 * A UCollator in a locale returned by this function will perform the correct 796 * collation for the locale. 797 * @param localeIndex The index of the desired locale. 798 * @return A locale for which collation rules are available, or 0 if none. 799 * @see ucol_countAvailable 800 * @stable ICU 2.0 801 */ 802 U_STABLE const char* U_EXPORT2 803 ucol_getAvailable(int32_t localeIndex); 804 805 /** 806 * Determine how many locales have collation rules available. 807 * This function is most useful as determining the loop ending condition for 808 * calls to {@link #ucol_getAvailable }. 809 * @return The number of locales for which collation rules are available. 810 * @see ucol_getAvailable 811 * @stable ICU 2.0 812 */ 813 U_STABLE int32_t U_EXPORT2 814 ucol_countAvailable(void); 815 816 #if !UCONFIG_NO_SERVICE 817 /** 818 * Create a string enumerator of all locales for which a valid 819 * collator may be opened. 820 * @param status input-output error code 821 * @return a string enumeration over locale strings. The caller is 822 * responsible for closing the result. 823 * @stable ICU 3.0 824 */ 825 U_STABLE UEnumeration* U_EXPORT2 826 ucol_openAvailableLocales(UErrorCode *status); 827 #endif 828 829 /** 830 * Create a string enumerator of all possible keywords that are relevant to 831 * collation. At this point, the only recognized keyword for this 832 * service is "collation". 833 * @param status input-output error code 834 * @return a string enumeration over locale strings. The caller is 835 * responsible for closing the result. 836 * @stable ICU 3.0 837 */ 838 U_STABLE UEnumeration* U_EXPORT2 839 ucol_getKeywords(UErrorCode *status); 840 841 /** 842 * Given a keyword, create a string enumeration of all values 843 * for that keyword that are currently in use. 844 * @param keyword a particular keyword as enumerated by 845 * ucol_getKeywords. If any other keyword is passed in, *status is set 846 * to U_ILLEGAL_ARGUMENT_ERROR. 847 * @param status input-output error code 848 * @return a string enumeration over collation keyword values, or NULL 849 * upon error. The caller is responsible for closing the result. 850 * @stable ICU 3.0 851 */ 852 U_STABLE UEnumeration* U_EXPORT2 853 ucol_getKeywordValues(const char *keyword, UErrorCode *status); 854 855 /** 856 * Given a key and a locale, returns an array of string values in a preferred 857 * order that would make a difference. These are all and only those values where 858 * the open (creation) of the service with the locale formed from the input locale 859 * plus input keyword and that value has different behavior than creation with the 860 * input locale alone. 861 * @param key one of the keys supported by this service. For now, only 862 * "collation" is supported. 863 * @param locale the locale 864 * @param commonlyUsed if set to true it will return only commonly used values 865 * with the given locale in preferred order. Otherwise, 866 * it will return all the available values for the locale. 867 * @param status error status 868 * @return a string enumeration over keyword values for the given key and the locale. 869 * @stable ICU 4.2 870 */ 871 U_STABLE UEnumeration* U_EXPORT2 872 ucol_getKeywordValuesForLocale(const char* key, 873 const char* locale, 874 UBool commonlyUsed, 875 UErrorCode* status); 876 877 /** 878 * Return the functionally equivalent locale for the specified 879 * input locale, with respect to given keyword, for the 880 * collation service. If two different input locale + keyword 881 * combinations produce the same result locale, then collators 882 * instantiated for these two different input locales will behave 883 * equivalently. The converse is not always true; two collators 884 * may in fact be equivalent, but return different results, due to 885 * internal details. The return result has no other meaning than 886 * that stated above, and implies nothing as to the relationship 887 * between the two locales. This is intended for use by 888 * applications who wish to cache collators, or otherwise reuse 889 * collators when possible. The functional equivalent may change 890 * over time. For more information, please see the <a 891 * href="http://userguide.icu-project.org/locale#TOC-Locales-and-Services"> 892 * Locales and Services</a> section of the ICU User Guide. 893 * @param result fillin for the functionally equivalent result locale 894 * @param resultCapacity capacity of the fillin buffer 895 * @param keyword a particular keyword as enumerated by 896 * ucol_getKeywords. 897 * @param locale the specified input locale 898 * @param isAvailable if non-NULL, pointer to a fillin parameter that 899 * on return indicates whether the specified input locale was 'available' 900 * to the collation service. A locale is defined as 'available' if it 901 * physically exists within the collation locale data. 902 * @param status pointer to input-output error code 903 * @return the actual buffer size needed for the locale. If greater 904 * than resultCapacity, the returned full name will be truncated and 905 * an error code will be returned. 906 * @stable ICU 3.0 907 */ 908 U_STABLE int32_t U_EXPORT2 909 ucol_getFunctionalEquivalent(char* result, int32_t resultCapacity, 910 const char* keyword, const char* locale, 911 UBool* isAvailable, UErrorCode* status); 912 913 /** 914 * Get the collation tailoring rules from a UCollator. 915 * The rules will follow the rule syntax. 916 * @param coll The UCollator to query. 917 * @param length 918 * @return The collation tailoring rules. 919 * @stable ICU 2.0 920 */ 921 U_STABLE const UChar* U_EXPORT2 922 ucol_getRules( const UCollator *coll, 923 int32_t *length); 924 925 #ifndef U_HIDE_DEPRECATED_API 926 /** Get the short definition string for a collator. This API harvests the collator's 927 * locale and the attribute set and produces a string that can be used for opening 928 * a collator with the same attributes using the ucol_openFromShortString API. 929 * This string will be normalized. 930 * The structure and the syntax of the string is defined in the "Naming collators" 931 * section of the users guide: 932 * http://userguide.icu-project.org/collation/concepts#TOC-Collator-naming-scheme 933 * This API supports preflighting. 934 * @param coll a collator 935 * @param locale a locale that will appear as a collators locale in the resulting 936 * short string definition. If NULL, the locale will be harvested 937 * from the collator. 938 * @param buffer space to hold the resulting string 939 * @param capacity capacity of the buffer 940 * @param status for returning errors. All the preflighting errors are featured 941 * @return length of the resulting string 942 * @see ucol_openFromShortString 943 * @see ucol_normalizeShortDefinitionString 944 * @deprecated ICU 54 945 */ 946 U_DEPRECATED int32_t U_EXPORT2 947 ucol_getShortDefinitionString(const UCollator *coll, 948 const char *locale, 949 char *buffer, 950 int32_t capacity, 951 UErrorCode *status); 952 953 /** Verifies and normalizes short definition string. 954 * Normalized short definition string has all the option sorted by the argument name, 955 * so that equivalent definition strings are the same. 956 * This API supports preflighting. 957 * @param source definition string 958 * @param destination space to hold the resulting string 959 * @param capacity capacity of the buffer 960 * @param parseError if not NULL, structure that will get filled with error's pre 961 * and post context in case of error. 962 * @param status Error code. This API will return an error if an invalid attribute 963 * or attribute/value combination is specified. All the preflighting 964 * errors are also featured 965 * @return length of the resulting normalized string. 966 * 967 * @see ucol_openFromShortString 968 * @see ucol_getShortDefinitionString 969 * 970 * @deprecated ICU 54 971 */ 972 973 U_DEPRECATED int32_t U_EXPORT2 974 ucol_normalizeShortDefinitionString(const char *source, 975 char *destination, 976 int32_t capacity, 977 UParseError *parseError, 978 UErrorCode *status); 979 #endif /* U_HIDE_DEPRECATED_API */ 980 981 982 /** 983 * Get a sort key for a string from a UCollator. 984 * Sort keys may be compared using <TT>strcmp</TT>. 985 * 986 * Note that sort keys are often less efficient than simply doing comparison. 987 * For more details, see the ICU User Guide. 988 * 989 * Like ICU functions that write to an output buffer, the buffer contents 990 * is undefined if the buffer capacity (resultLength parameter) is too small. 991 * Unlike ICU functions that write a string to an output buffer, 992 * the terminating zero byte is counted in the sort key length. 993 * @param coll The UCollator containing the collation rules. 994 * @param source The string to transform. 995 * @param sourceLength The length of source, or -1 if null-terminated. 996 * @param result A pointer to a buffer to receive the attribute. 997 * @param resultLength The maximum size of result. 998 * @return The size needed to fully store the sort key. 999 * If there was an internal error generating the sort key, 1000 * a zero value is returned. 1001 * @see ucol_keyHashCode 1002 * @stable ICU 2.0 1003 */ 1004 U_STABLE int32_t U_EXPORT2 1005 ucol_getSortKey(const UCollator *coll, 1006 const UChar *source, 1007 int32_t sourceLength, 1008 uint8_t *result, 1009 int32_t resultLength); 1010 1011 1012 /** Gets the next count bytes of a sort key. Caller needs 1013 * to preserve state array between calls and to provide 1014 * the same type of UCharIterator set with the same string. 1015 * The destination buffer provided must be big enough to store 1016 * the number of requested bytes. 1017 * 1018 * The generated sort key may or may not be compatible with 1019 * sort keys generated using ucol_getSortKey(). 1020 * @param coll The UCollator containing the collation rules. 1021 * @param iter UCharIterator containing the string we need 1022 * the sort key to be calculated for. 1023 * @param state Opaque state of sortkey iteration. 1024 * @param dest Buffer to hold the resulting sortkey part 1025 * @param count number of sort key bytes required. 1026 * @param status error code indicator. 1027 * @return the actual number of bytes of a sortkey. It can be 1028 * smaller than count if we have reached the end of 1029 * the sort key. 1030 * @stable ICU 2.6 1031 */ 1032 U_STABLE int32_t U_EXPORT2 1033 ucol_nextSortKeyPart(const UCollator *coll, 1034 UCharIterator *iter, 1035 uint32_t state[2], 1036 uint8_t *dest, int32_t count, 1037 UErrorCode *status); 1038 1039 /** enum that is taken by ucol_getBound API 1040 * See below for explanation 1041 * do not change the values assigned to the 1042 * members of this enum. Underlying code 1043 * depends on them having these numbers 1044 * @stable ICU 2.0 1045 */ 1046 typedef enum { 1047 /** lower bound */ 1048 UCOL_BOUND_LOWER = 0, 1049 /** upper bound that will match strings of exact size */ 1050 UCOL_BOUND_UPPER = 1, 1051 /** upper bound that will match all the strings that have the same initial substring as the given string */ 1052 UCOL_BOUND_UPPER_LONG = 2, 1053 UCOL_BOUND_VALUE_COUNT 1054 } UColBoundMode; 1055 1056 /** 1057 * Produce a bound for a given sortkey and a number of levels. 1058 * Return value is always the number of bytes needed, regardless of 1059 * whether the result buffer was big enough or even valid.<br> 1060 * Resulting bounds can be used to produce a range of strings that are 1061 * between upper and lower bounds. For example, if bounds are produced 1062 * for a sortkey of string "smith", strings between upper and lower 1063 * bounds with one level would include "Smith", "SMITH", "sMiTh".<br> 1064 * There are two upper bounds that can be produced. If UCOL_BOUND_UPPER 1065 * is produced, strings matched would be as above. However, if bound 1066 * produced using UCOL_BOUND_UPPER_LONG is used, the above example will 1067 * also match "Smithsonian" and similar.<br> 1068 * For more on usage, see example in cintltst/capitst.c in procedure 1069 * TestBounds. 1070 * Sort keys may be compared using <TT>strcmp</TT>. 1071 * @param source The source sortkey. 1072 * @param sourceLength The length of source, or -1 if null-terminated. 1073 * (If an unmodified sortkey is passed, it is always null 1074 * terminated). 1075 * @param boundType Type of bound required. It can be UCOL_BOUND_LOWER, which 1076 * produces a lower inclusive bound, UCOL_BOUND_UPPER, that 1077 * produces upper bound that matches strings of the same length 1078 * or UCOL_BOUND_UPPER_LONG that matches strings that have the 1079 * same starting substring as the source string. 1080 * @param noOfLevels Number of levels required in the resulting bound (for most 1081 * uses, the recommended value is 1). See users guide for 1082 * explanation on number of levels a sortkey can have. 1083 * @param result A pointer to a buffer to receive the resulting sortkey. 1084 * @param resultLength The maximum size of result. 1085 * @param status Used for returning error code if something went wrong. If the 1086 * number of levels requested is higher than the number of levels 1087 * in the source key, a warning (U_SORT_KEY_TOO_SHORT_WARNING) is 1088 * issued. 1089 * @return The size needed to fully store the bound. 1090 * @see ucol_keyHashCode 1091 * @stable ICU 2.1 1092 */ 1093 U_STABLE int32_t U_EXPORT2 1094 ucol_getBound(const uint8_t *source, 1095 int32_t sourceLength, 1096 UColBoundMode boundType, 1097 uint32_t noOfLevels, 1098 uint8_t *result, 1099 int32_t resultLength, 1100 UErrorCode *status); 1101 1102 /** 1103 * Gets the version information for a Collator. Version is currently 1104 * an opaque 32-bit number which depends, among other things, on major 1105 * versions of the collator tailoring and UCA. 1106 * @param coll The UCollator to query. 1107 * @param info the version # information, the result will be filled in 1108 * @stable ICU 2.0 1109 */ 1110 U_STABLE void U_EXPORT2 1111 ucol_getVersion(const UCollator* coll, UVersionInfo info); 1112 1113 /** 1114 * Gets the UCA version information for a Collator. Version is the 1115 * UCA version number (3.1.1, 4.0). 1116 * @param coll The UCollator to query. 1117 * @param info the version # information, the result will be filled in 1118 * @stable ICU 2.8 1119 */ 1120 U_STABLE void U_EXPORT2 1121 ucol_getUCAVersion(const UCollator* coll, UVersionInfo info); 1122 1123 /** 1124 * Merges two sort keys. The levels are merged with their corresponding counterparts 1125 * (primaries with primaries, secondaries with secondaries etc.). Between the values 1126 * from the same level a separator is inserted. 1127 * 1128 * This is useful, for example, for combining sort keys from first and last names 1129 * to sort such pairs. 1130 * See http://www.unicode.org/reports/tr10/#Merging_Sort_Keys 1131 * 1132 * The recommended way to achieve "merged" sorting is by 1133 * concatenating strings with U+FFFE between them. 1134 * The concatenation has the same sort order as the merged sort keys, 1135 * but merge(getSortKey(str1), getSortKey(str2)) may differ from getSortKey(str1 + '\uFFFE' + str2). 1136 * Using strings with U+FFFE may yield shorter sort keys. 1137 * 1138 * For details about Sort Key Features see 1139 * http://userguide.icu-project.org/collation/api#TOC-Sort-Key-Features 1140 * 1141 * It is possible to merge multiple sort keys by consecutively merging 1142 * another one with the intermediate result. 1143 * 1144 * The length of the merge result is the sum of the lengths of the input sort keys. 1145 * 1146 * Example (uncompressed): 1147 * <pre>191B1D 01 050505 01 910505 00 1148 * 1F2123 01 050505 01 910505 00</pre> 1149 * will be merged as 1150 * <pre>191B1D 02 1F2123 01 050505 02 050505 01 910505 02 910505 00</pre> 1151 * 1152 * If the destination buffer is not big enough, then its contents are undefined. 1153 * If any of source lengths are zero or any of the source pointers are NULL/undefined, 1154 * the result is of size zero. 1155 * 1156 * @param src1 the first sort key 1157 * @param src1Length the length of the first sort key, including the zero byte at the end; 1158 * can be -1 if the function is to find the length 1159 * @param src2 the second sort key 1160 * @param src2Length the length of the second sort key, including the zero byte at the end; 1161 * can be -1 if the function is to find the length 1162 * @param dest the buffer where the merged sort key is written, 1163 * can be NULL if destCapacity==0 1164 * @param destCapacity the number of bytes in the dest buffer 1165 * @return the length of the merged sort key, src1Length+src2Length; 1166 * can be larger than destCapacity, or 0 if an error occurs (only for illegal arguments), 1167 * in which cases the contents of dest is undefined 1168 * @stable ICU 2.0 1169 */ 1170 U_STABLE int32_t U_EXPORT2 1171 ucol_mergeSortkeys(const uint8_t *src1, int32_t src1Length, 1172 const uint8_t *src2, int32_t src2Length, 1173 uint8_t *dest, int32_t destCapacity); 1174 1175 /** 1176 * Universal attribute setter 1177 * @param coll collator which attributes are to be changed 1178 * @param attr attribute type 1179 * @param value attribute value 1180 * @param status to indicate whether the operation went on smoothly or there were errors 1181 * @see UColAttribute 1182 * @see UColAttributeValue 1183 * @see ucol_getAttribute 1184 * @stable ICU 2.0 1185 */ 1186 U_STABLE void U_EXPORT2 1187 ucol_setAttribute(UCollator *coll, UColAttribute attr, UColAttributeValue value, UErrorCode *status); 1188 1189 /** 1190 * Universal attribute getter 1191 * @param coll collator which attributes are to be changed 1192 * @param attr attribute type 1193 * @return attribute value 1194 * @param status to indicate whether the operation went on smoothly or there were errors 1195 * @see UColAttribute 1196 * @see UColAttributeValue 1197 * @see ucol_setAttribute 1198 * @stable ICU 2.0 1199 */ 1200 U_STABLE UColAttributeValue U_EXPORT2 1201 ucol_getAttribute(const UCollator *coll, UColAttribute attr, UErrorCode *status); 1202 1203 /** 1204 * Sets the variable top to the top of the specified reordering group. 1205 * The variable top determines the highest-sorting character 1206 * which is affected by UCOL_ALTERNATE_HANDLING. 1207 * If that attribute is set to UCOL_NON_IGNORABLE, then the variable top has no effect. 1208 * @param coll the collator 1209 * @param group one of UCOL_REORDER_CODE_SPACE, UCOL_REORDER_CODE_PUNCTUATION, 1210 * UCOL_REORDER_CODE_SYMBOL, UCOL_REORDER_CODE_CURRENCY; 1211 * or UCOL_REORDER_CODE_DEFAULT to restore the default max variable group 1212 * @param pErrorCode Standard ICU error code. Its input value must 1213 * pass the U_SUCCESS() test, or else the function returns 1214 * immediately. Check for U_FAILURE() on output or use with 1215 * function chaining. (See User Guide for details.) 1216 * @see ucol_getMaxVariable 1217 * @stable ICU 53 1218 */ 1219 U_STABLE void U_EXPORT2 1220 ucol_setMaxVariable(UCollator *coll, UColReorderCode group, UErrorCode *pErrorCode); 1221 1222 /** 1223 * Returns the maximum reordering group whose characters are affected by UCOL_ALTERNATE_HANDLING. 1224 * @param coll the collator 1225 * @return the maximum variable reordering group. 1226 * @see ucol_setMaxVariable 1227 * @stable ICU 53 1228 */ 1229 U_STABLE UColReorderCode U_EXPORT2 1230 ucol_getMaxVariable(const UCollator *coll); 1231 1232 #ifndef U_HIDE_DEPRECATED_API 1233 /** 1234 * Sets the variable top to the primary weight of the specified string. 1235 * 1236 * Beginning with ICU 53, the variable top is pinned to 1237 * the top of one of the supported reordering groups, 1238 * and it must not be beyond the last of those groups. 1239 * See ucol_setMaxVariable(). 1240 * @param coll the collator 1241 * @param varTop one or more (if contraction) UChars to which the variable top should be set 1242 * @param len length of variable top string. If -1 it is considered to be zero terminated. 1243 * @param status error code. If error code is set, the return value is undefined. 1244 * Errors set by this function are:<br> 1245 * U_CE_NOT_FOUND_ERROR if more than one character was passed and there is no such contraction<br> 1246 * U_ILLEGAL_ARGUMENT_ERROR if the variable top is beyond 1247 * the last reordering group supported by ucol_setMaxVariable() 1248 * @return variable top primary weight 1249 * @see ucol_getVariableTop 1250 * @see ucol_restoreVariableTop 1251 * @deprecated ICU 53 Call ucol_setMaxVariable() instead. 1252 */ 1253 U_DEPRECATED uint32_t U_EXPORT2 1254 ucol_setVariableTop(UCollator *coll, 1255 const UChar *varTop, int32_t len, 1256 UErrorCode *status); 1257 #endif /* U_HIDE_DEPRECATED_API */ 1258 1259 /** 1260 * Gets the variable top value of a Collator. 1261 * @param coll collator which variable top needs to be retrieved 1262 * @param status error code (not changed by function). If error code is set, 1263 * the return value is undefined. 1264 * @return the variable top primary weight 1265 * @see ucol_getMaxVariable 1266 * @see ucol_setVariableTop 1267 * @see ucol_restoreVariableTop 1268 * @stable ICU 2.0 1269 */ 1270 U_STABLE uint32_t U_EXPORT2 ucol_getVariableTop(const UCollator *coll, UErrorCode *status); 1271 1272 /** 1273 * Sets the variable top to the specified primary weight. 1274 * 1275 * Beginning with ICU 53, the variable top is pinned to 1276 * the top of one of the supported reordering groups, 1277 * and it must not be beyond the last of those groups. 1278 * See ucol_setMaxVariable(). 1279 * @param varTop primary weight, as returned by ucol_setVariableTop or ucol_getVariableTop 1280 * @param status error code 1281 * @see ucol_getVariableTop 1282 * @see ucol_setVariableTop 1283 * @deprecated ICU 53 Call ucol_setMaxVariable() instead. 1284 */ 1285 U_DEPRECATED void U_EXPORT2 1286 ucol_restoreVariableTop(UCollator *coll, const uint32_t varTop, UErrorCode *status); 1287 1288 /** 1289 * Thread safe cloning operation. The result is a clone of a given collator. 1290 * @param coll collator to be cloned 1291 * @param stackBuffer <em>Deprecated functionality as of ICU 52, use NULL.</em><br> 1292 * user allocated space for the new clone. 1293 * If NULL new memory will be allocated. 1294 * If buffer is not large enough, new memory will be allocated. 1295 * Clients can use the U_COL_SAFECLONE_BUFFERSIZE. 1296 * @param pBufferSize <em>Deprecated functionality as of ICU 52, use NULL or 1.</em><br> 1297 * pointer to size of allocated space. 1298 * If *pBufferSize == 0, a sufficient size for use in cloning will 1299 * be returned ('pre-flighting') 1300 * If *pBufferSize is not enough for a stack-based safe clone, 1301 * new memory will be allocated. 1302 * @param status to indicate whether the operation went on smoothly or there were errors 1303 * An informational status value, U_SAFECLONE_ALLOCATED_ERROR, is used if any 1304 * allocations were necessary. 1305 * @return pointer to the new clone 1306 * @see ucol_open 1307 * @see ucol_openRules 1308 * @see ucol_close 1309 * @stable ICU 2.0 1310 */ 1311 U_STABLE UCollator* U_EXPORT2 1312 ucol_safeClone(const UCollator *coll, 1313 void *stackBuffer, 1314 int32_t *pBufferSize, 1315 UErrorCode *status); 1316 1317 #ifndef U_HIDE_DEPRECATED_API 1318 1319 /** default memory size for the new clone. 1320 * @deprecated ICU 52. Do not rely on ucol_safeClone() cloning into any provided buffer. 1321 */ 1322 #define U_COL_SAFECLONE_BUFFERSIZE 1 1323 1324 #endif /* U_HIDE_DEPRECATED_API */ 1325 1326 /** 1327 * Returns current rules. Delta defines whether full rules are returned or just the tailoring. 1328 * Returns number of UChars needed to store rules. If buffer is NULL or bufferLen is not enough 1329 * to store rules, will store up to available space. 1330 * 1331 * ucol_getRules() should normally be used instead. 1332 * See http://userguide.icu-project.org/collation/customization#TOC-Building-on-Existing-Locales 1333 * @param coll collator to get the rules from 1334 * @param delta one of UCOL_TAILORING_ONLY, UCOL_FULL_RULES. 1335 * @param buffer buffer to store the result in. If NULL, you'll get no rules. 1336 * @param bufferLen length of buffer to store rules in. If less than needed you'll get only the part that fits in. 1337 * @return current rules 1338 * @stable ICU 2.0 1339 * @see UCOL_FULL_RULES 1340 */ 1341 U_STABLE int32_t U_EXPORT2 1342 ucol_getRulesEx(const UCollator *coll, UColRuleOption delta, UChar *buffer, int32_t bufferLen); 1343 1344 #ifndef U_HIDE_DEPRECATED_API 1345 /** 1346 * gets the locale name of the collator. If the collator 1347 * is instantiated from the rules, then this function returns 1348 * NULL. 1349 * @param coll The UCollator for which the locale is needed 1350 * @param type You can choose between requested, valid and actual 1351 * locale. For description see the definition of 1352 * ULocDataLocaleType in uloc.h 1353 * @param status error code of the operation 1354 * @return real locale name from which the collation data comes. 1355 * If the collator was instantiated from rules, returns 1356 * NULL. 1357 * @deprecated ICU 2.8 Use ucol_getLocaleByType instead 1358 */ 1359 U_DEPRECATED const char * U_EXPORT2 1360 ucol_getLocale(const UCollator *coll, ULocDataLocaleType type, UErrorCode *status); 1361 #endif /* U_HIDE_DEPRECATED_API */ 1362 1363 /** 1364 * gets the locale name of the collator. If the collator 1365 * is instantiated from the rules, then this function returns 1366 * NULL. 1367 * @param coll The UCollator for which the locale is needed 1368 * @param type You can choose between requested, valid and actual 1369 * locale. For description see the definition of 1370 * ULocDataLocaleType in uloc.h 1371 * @param status error code of the operation 1372 * @return real locale name from which the collation data comes. 1373 * If the collator was instantiated from rules, returns 1374 * NULL. 1375 * @stable ICU 2.8 1376 */ 1377 U_STABLE const char * U_EXPORT2 1378 ucol_getLocaleByType(const UCollator *coll, ULocDataLocaleType type, UErrorCode *status); 1379 1380 /** 1381 * Get a Unicode set that contains all the characters and sequences tailored in 1382 * this collator. The result must be disposed of by using uset_close. 1383 * @param coll The UCollator for which we want to get tailored chars 1384 * @param status error code of the operation 1385 * @return a pointer to newly created USet. Must be be disposed by using uset_close 1386 * @see ucol_openRules 1387 * @see uset_close 1388 * @stable ICU 2.4 1389 */ 1390 U_STABLE USet * U_EXPORT2 1391 ucol_getTailoredSet(const UCollator *coll, UErrorCode *status); 1392 1393 #ifndef U_HIDE_INTERNAL_API 1394 /** Calculates the set of unsafe code points, given a collator. 1395 * A character is unsafe if you could append any character and cause the ordering to alter significantly. 1396 * Collation sorts in normalized order, so anything that rearranges in normalization can cause this. 1397 * Thus if you have a character like a_umlaut, and you add a lower_dot to it, 1398 * then it normalizes to a_lower_dot + umlaut, and sorts differently. 1399 * @param coll Collator 1400 * @param unsafe a fill-in set to receive the unsafe points 1401 * @param status for catching errors 1402 * @return number of elements in the set 1403 * @internal ICU 3.0 1404 */ 1405 U_INTERNAL int32_t U_EXPORT2 1406 ucol_getUnsafeSet( const UCollator *coll, 1407 USet *unsafe, 1408 UErrorCode *status); 1409 1410 /** Touches all resources needed for instantiating a collator from a short string definition, 1411 * thus filling up the cache. 1412 * @param definition A short string containing a locale and a set of attributes. 1413 * Attributes not explicitly mentioned are left at the default 1414 * state for a locale. 1415 * @param parseError if not NULL, structure that will get filled with error's pre 1416 * and post context in case of error. 1417 * @param forceDefaults if FALSE, the settings that are the same as the collator 1418 * default settings will not be applied (for example, setting 1419 * French secondary on a French collator would not be executed). 1420 * If TRUE, all the settings will be applied regardless of the 1421 * collator default value. If the definition 1422 * strings are to be cached, should be set to FALSE. 1423 * @param status Error code. Apart from regular error conditions connected to 1424 * instantiating collators (like out of memory or similar), this 1425 * API will return an error if an invalid attribute or attribute/value 1426 * combination is specified. 1427 * @see ucol_openFromShortString 1428 * @internal ICU 3.2.1 1429 */ 1430 U_INTERNAL void U_EXPORT2 1431 ucol_prepareShortStringOpen( const char *definition, 1432 UBool forceDefaults, 1433 UParseError *parseError, 1434 UErrorCode *status); 1435 #endif /* U_HIDE_INTERNAL_API */ 1436 1437 /** Creates a binary image of a collator. This binary image can be stored and 1438 * later used to instantiate a collator using ucol_openBinary. 1439 * This API supports preflighting. 1440 * @param coll Collator 1441 * @param buffer a fill-in buffer to receive the binary image 1442 * @param capacity capacity of the destination buffer 1443 * @param status for catching errors 1444 * @return size of the image 1445 * @see ucol_openBinary 1446 * @stable ICU 3.2 1447 */ 1448 U_STABLE int32_t U_EXPORT2 1449 ucol_cloneBinary(const UCollator *coll, 1450 uint8_t *buffer, int32_t capacity, 1451 UErrorCode *status); 1452 1453 /** Opens a collator from a collator binary image created using 1454 * ucol_cloneBinary. Binary image used in instantiation of the 1455 * collator remains owned by the user and should stay around for 1456 * the lifetime of the collator. The API also takes a base collator 1457 * which must be the root collator. 1458 * @param bin binary image owned by the user and required through the 1459 * lifetime of the collator 1460 * @param length size of the image. If negative, the API will try to 1461 * figure out the length of the image 1462 * @param base Base collator, for lookup of untailored characters. 1463 * Must be the root collator, must not be NULL. 1464 * The base is required to be present through the lifetime of the collator. 1465 * @param status for catching errors 1466 * @return newly created collator 1467 * @see ucol_cloneBinary 1468 * @stable ICU 3.2 1469 */ 1470 U_STABLE UCollator* U_EXPORT2 1471 ucol_openBinary(const uint8_t *bin, int32_t length, 1472 const UCollator *base, 1473 UErrorCode *status); 1474 1475 1476 #endif /* #if !UCONFIG_NO_COLLATION */ 1477 1478 #endif 1479