|
|
|||
File indexing completed on 2026-01-07 10:23:56
0001 // © 2016 and later: Unicode, Inc. and others. 0002 // License & terms of use: http://www.unicode.org/copyright.html 0003 /* 0004 ****************************************************************************** 0005 * Copyright (C) 1996-2016, International Business Machines 0006 * Corporation and others. All Rights Reserved. 0007 ****************************************************************************** 0008 */ 0009 0010 /** 0011 * \file 0012 * \brief C++ API: Collation Service. 0013 */ 0014 0015 /** 0016 * File coll.h 0017 * 0018 * Created by: Helena Shih 0019 * 0020 * Modification History: 0021 * 0022 * Date Name Description 0023 * 02/5/97 aliu Modified createDefault to load collation data from 0024 * binary files when possible. Added related methods 0025 * createCollationFromFile, chopLocale, createPathName. 0026 * 02/11/97 aliu Added members addToCache, findInCache, and fgCache. 0027 * 02/12/97 aliu Modified to create objects from RuleBasedCollator cache. 0028 * Moved cache out of Collation class. 0029 * 02/13/97 aliu Moved several methods out of this class and into 0030 * RuleBasedCollator, with modifications. Modified 0031 * createDefault() to call new RuleBasedCollator(Locale&) 0032 * constructor. General clean up and documentation. 0033 * 02/20/97 helena Added clone, operator==, operator!=, operator=, copy 0034 * constructor and getDynamicClassID. 0035 * 03/25/97 helena Updated with platform independent data types. 0036 * 05/06/97 helena Added memory allocation error detection. 0037 * 06/20/97 helena Java class name change. 0038 * 09/03/97 helena Added createCollationKeyValues(). 0039 * 02/10/98 damiba Added compare() with length as parameter. 0040 * 04/23/99 stephen Removed EDecompositionMode, merged with 0041 * Normalizer::EMode. 0042 * 11/02/99 helena Collator performance enhancements. Eliminates the 0043 * UnicodeString construction and special case for NO_OP. 0044 * 11/23/99 srl More performance enhancements. Inlining of 0045 * critical accessors. 0046 * 05/15/00 helena Added version information API. 0047 * 01/29/01 synwee Modified into a C++ wrapper which calls C apis 0048 * (ucol.h). 0049 * 2012-2014 markus Rewritten in C++ again. 0050 */ 0051 0052 #ifndef COLL_H 0053 #define COLL_H 0054 0055 #include "unicode/utypes.h" 0056 0057 #if U_SHOW_CPLUSPLUS_API 0058 0059 #if !UCONFIG_NO_COLLATION 0060 0061 #include <functional> 0062 #include <string_view> 0063 #include <type_traits> 0064 0065 #include "unicode/char16ptr.h" 0066 #include "unicode/uobject.h" 0067 #include "unicode/ucol.h" 0068 #include "unicode/unorm.h" 0069 #include "unicode/locid.h" 0070 #include "unicode/uniset.h" 0071 #include "unicode/umisc.h" 0072 #include "unicode/unistr.h" 0073 #include "unicode/uiter.h" 0074 #include "unicode/stringpiece.h" 0075 0076 U_NAMESPACE_BEGIN 0077 0078 class StringEnumeration; 0079 0080 #if !UCONFIG_NO_SERVICE 0081 /** 0082 * @stable ICU 2.6 0083 */ 0084 class CollatorFactory; 0085 #endif 0086 0087 /** 0088 * @stable ICU 2.0 0089 */ 0090 class CollationKey; 0091 0092 /** 0093 * The <code>Collator</code> class performs locale-sensitive string 0094 * comparison.<br> 0095 * You use this class to build searching and sorting routines for natural 0096 * language text. 0097 * <p> 0098 * <code>Collator</code> is an abstract base class. Subclasses implement 0099 * specific collation strategies. One subclass, 0100 * <code>RuleBasedCollator</code>, is currently provided and is applicable 0101 * to a wide set of languages. Other subclasses may be created to handle more 0102 * specialized needs. 0103 * <p> 0104 * Like other locale-sensitive classes, you can use the static factory method, 0105 * <code>createInstance</code>, to obtain the appropriate 0106 * <code>Collator</code> object for a given locale. You will only need to 0107 * look at the subclasses of <code>Collator</code> if you need to 0108 * understand the details of a particular collation strategy or if you need to 0109 * modify that strategy. 0110 * <p> 0111 * The following example shows how to compare two strings using the 0112 * <code>Collator</code> for the default locale. 0113 * \htmlonly<blockquote>\endhtmlonly 0114 * <pre> 0115 * \code 0116 * // Compare two strings in the default locale 0117 * UErrorCode success = U_ZERO_ERROR; 0118 * Collator* myCollator = Collator::createInstance(success); 0119 * if (myCollator->compare("abc", "ABC") < 0) 0120 * cout << "abc is less than ABC" << endl; 0121 * else 0122 * cout << "abc is greater than or equal to ABC" << endl; 0123 * \endcode 0124 * </pre> 0125 * \htmlonly</blockquote>\endhtmlonly 0126 * <p> 0127 * You can set a <code>Collator</code>'s <em>strength</em> attribute to 0128 * determine the level of difference considered significant in comparisons. 0129 * Five strengths are provided: <code>PRIMARY</code>, <code>SECONDARY</code>, 0130 * <code>TERTIARY</code>, <code>QUATERNARY</code> and <code>IDENTICAL</code>. 0131 * The exact assignment of strengths to language features is locale dependent. 0132 * For example, in Czech, "e" and "f" are considered primary differences, 0133 * while "e" and "\u00EA" are secondary differences, "e" and "E" are tertiary 0134 * differences and "e" and "e" are identical. The following shows how both case 0135 * and accents could be ignored for US English. 0136 * \htmlonly<blockquote>\endhtmlonly 0137 * <pre> 0138 * \code 0139 * //Get the Collator for US English and set its strength to PRIMARY 0140 * UErrorCode success = U_ZERO_ERROR; 0141 * Collator* usCollator = Collator::createInstance(Locale::getUS(), success); 0142 * usCollator->setStrength(Collator::PRIMARY); 0143 * if (usCollator->compare("abc", "ABC") == 0) 0144 * cout << "'abc' and 'ABC' strings are equivalent with strength PRIMARY" << endl; 0145 * \endcode 0146 * </pre> 0147 * \htmlonly</blockquote>\endhtmlonly 0148 * 0149 * The <code>getSortKey</code> methods 0150 * convert a string to a series of bytes that can be compared bitwise against 0151 * other sort keys using <code>strcmp()</code>. Sort keys are written as 0152 * zero-terminated byte strings. 0153 * 0154 * Another set of APIs returns a <code>CollationKey</code> object that wraps 0155 * the sort key bytes instead of returning the bytes themselves. 0156 * </p> 0157 * <p> 0158 * <strong>Note:</strong> <code>Collator</code>s with different Locale, 0159 * and CollationStrength settings will return different sort 0160 * orders for the same set of strings. Locales have specific collation rules, 0161 * and the way in which secondary and tertiary differences are taken into 0162 * account, for example, will result in a different sorting order for same 0163 * strings. 0164 * </p> 0165 * @see RuleBasedCollator 0166 * @see CollationKey 0167 * @see CollationElementIterator 0168 * @see Locale 0169 * @see Normalizer2 0170 * @version 2.0 11/15/01 0171 */ 0172 0173 class U_I18N_API Collator : public UObject { 0174 public: 0175 0176 // Collator public enums ----------------------------------------------- 0177 0178 /** 0179 * Base letter represents a primary difference. Set comparison level to 0180 * PRIMARY to ignore secondary and tertiary differences.<br> 0181 * Use this to set the strength of a Collator object.<br> 0182 * Example of primary difference, "abc" < "abd" 0183 * 0184 * Diacritical differences on the same base letter represent a secondary 0185 * difference. Set comparison level to SECONDARY to ignore tertiary 0186 * differences. Use this to set the strength of a Collator object.<br> 0187 * Example of secondary difference, "ä" >> "a". 0188 * 0189 * Uppercase and lowercase versions of the same character represents a 0190 * tertiary difference. Set comparison level to TERTIARY to include all 0191 * comparison differences. Use this to set the strength of a Collator 0192 * object.<br> 0193 * Example of tertiary difference, "abc" <<< "ABC". 0194 * 0195 * Two characters are considered "identical" when they have the same unicode 0196 * spellings.<br> 0197 * For example, "ä" == "ä". 0198 * 0199 * UCollationStrength is also used to determine the strength of sort keys 0200 * generated from Collator objects. 0201 * @stable ICU 2.0 0202 */ 0203 enum ECollationStrength 0204 { 0205 PRIMARY = UCOL_PRIMARY, // 0 0206 SECONDARY = UCOL_SECONDARY, // 1 0207 TERTIARY = UCOL_TERTIARY, // 2 0208 QUATERNARY = UCOL_QUATERNARY, // 3 0209 IDENTICAL = UCOL_IDENTICAL // 15 0210 }; 0211 0212 0213 // Cannot use #ifndef U_HIDE_DEPRECATED_API for the following, it is 0214 // used by virtual methods that cannot have that conditional. 0215 #ifndef U_FORCE_HIDE_DEPRECATED_API 0216 /** 0217 * LESS is returned if source string is compared to be less than target 0218 * string in the compare() method. 0219 * EQUAL is returned if source string is compared to be equal to target 0220 * string in the compare() method. 0221 * GREATER is returned if source string is compared to be greater than 0222 * target string in the compare() method. 0223 * @see Collator#compare 0224 * @deprecated ICU 2.6. Use C enum UCollationResult defined in ucol.h 0225 */ 0226 enum EComparisonResult 0227 { 0228 LESS = UCOL_LESS, // -1 0229 EQUAL = UCOL_EQUAL, // 0 0230 GREATER = UCOL_GREATER // 1 0231 }; 0232 #endif // U_FORCE_HIDE_DEPRECATED_API 0233 0234 // Collator public destructor ----------------------------------------- 0235 0236 /** 0237 * Destructor 0238 * @stable ICU 2.0 0239 */ 0240 virtual ~Collator(); 0241 0242 // Collator public methods -------------------------------------------- 0243 0244 /** 0245 * Returns true if "other" is the same as "this". 0246 * 0247 * The base class implementation returns true if "other" has the same type/class as "this": 0248 * `typeid(*this) == typeid(other)`. 0249 * 0250 * Subclass implementations should do something like the following: 0251 * 0252 * if (this == &other) { return true; } 0253 * if (!Collator::operator==(other)) { return false; } // not the same class 0254 * 0255 * const MyCollator &o = (const MyCollator&)other; 0256 * (compare this vs. o's subclass fields) 0257 * 0258 * @param other Collator object to be compared 0259 * @return true if other is the same as this. 0260 * @stable ICU 2.0 0261 */ 0262 virtual bool operator==(const Collator& other) const; 0263 0264 /** 0265 * Returns true if "other" is not the same as "this". 0266 * Calls ! operator==(const Collator&) const which works for all subclasses. 0267 * @param other Collator object to be compared 0268 * @return true if other is not the same as this. 0269 * @stable ICU 2.0 0270 */ 0271 virtual bool operator!=(const Collator& other) const; 0272 0273 /** 0274 * Makes a copy of this object. 0275 * @return a copy of this object, owned by the caller 0276 * @stable ICU 2.0 0277 */ 0278 virtual Collator* clone() const = 0; 0279 0280 /** 0281 * Creates the Collator object for the current default locale. 0282 * The default locale is determined by Locale::getDefault. 0283 * The UErrorCode& err parameter is used to return status information to the user. 0284 * To check whether the construction succeeded or not, you should check the 0285 * value of U_SUCCESS(err). If you wish more detailed information, you can 0286 * check for informational error results which still indicate success. 0287 * U_USING_FALLBACK_ERROR indicates that a fall back locale was used. For 0288 * example, 'de_CH' was requested, but nothing was found there, so 'de' was 0289 * used. U_USING_DEFAULT_ERROR indicates that the default locale data was 0290 * used; neither the requested locale nor any of its fall back locales 0291 * could be found. 0292 * The caller owns the returned object and is responsible for deleting it. 0293 * 0294 * @param err the error code status. 0295 * @return the collation object of the default locale.(for example, en_US) 0296 * @see Locale#getDefault 0297 * @stable ICU 2.0 0298 */ 0299 static Collator* U_EXPORT2 createInstance(UErrorCode& err); 0300 0301 /** 0302 * Gets the collation object for the desired locale. The 0303 * resource of the desired locale will be loaded. 0304 * 0305 * Locale::getRoot() is the base collation table and all other languages are 0306 * built on top of it with additional language-specific modifications. 0307 * 0308 * For some languages, multiple collation types are available; 0309 * for example, "de@collation=phonebook". 0310 * Starting with ICU 54, collation attributes can be specified via locale keywords as well, 0311 * in the old locale extension syntax ("el@colCaseFirst=upper") 0312 * or in language tag syntax ("el-u-kf-upper"). 0313 * See <a href="https://unicode-org.github.io/icu/userguide/collation/api">User Guide: Collation API</a>. 0314 * 0315 * The UErrorCode& err parameter is used to return status information to the user. 0316 * To check whether the construction succeeded or not, you should check 0317 * the value of U_SUCCESS(err). If you wish more detailed information, you 0318 * can check for informational error results which still indicate success. 0319 * U_USING_FALLBACK_ERROR indicates that a fall back locale was used. For 0320 * example, 'de_CH' was requested, but nothing was found there, so 'de' was 0321 * used. U_USING_DEFAULT_ERROR indicates that the default locale data was 0322 * used; neither the requested locale nor any of its fall back locales 0323 * could be found. 0324 * 0325 * The caller owns the returned object and is responsible for deleting it. 0326 * @param loc The locale ID for which to open a collator. 0327 * @param err the error code status. 0328 * @return the created table-based collation object based on the desired 0329 * locale. 0330 * @see Locale 0331 * @see ResourceLoader 0332 * @stable ICU 2.2 0333 */ 0334 static Collator* U_EXPORT2 createInstance(const Locale& loc, UErrorCode& err); 0335 0336 #ifndef U_FORCE_HIDE_DEPRECATED_API 0337 /** 0338 * The comparison function compares the character data stored in two 0339 * different strings. Returns information about whether a string is less 0340 * than, greater than or equal to another string. 0341 * @param source the source string to be compared with. 0342 * @param target the string that is to be compared with the source string. 0343 * @return Returns a byte value. GREATER if source is greater 0344 * than target; EQUAL if source is equal to target; LESS if source is less 0345 * than target 0346 * @deprecated ICU 2.6 use the overload with UErrorCode & 0347 */ 0348 virtual EComparisonResult compare(const UnicodeString& source, 0349 const UnicodeString& target) const; 0350 #endif // U_FORCE_HIDE_DEPRECATED_API 0351 0352 /** 0353 * The comparison function compares the character data stored in two 0354 * different strings. Returns information about whether a string is less 0355 * than, greater than or equal to another string. 0356 * @param source the source string to be compared with. 0357 * @param target the string that is to be compared with the source string. 0358 * @param status possible error code 0359 * @return Returns an enum value. UCOL_GREATER if source is greater 0360 * than target; UCOL_EQUAL if source is equal to target; UCOL_LESS if source is less 0361 * than target 0362 * @stable ICU 2.6 0363 */ 0364 virtual UCollationResult compare(const UnicodeString& source, 0365 const UnicodeString& target, 0366 UErrorCode &status) const = 0; 0367 0368 #ifndef U_FORCE_HIDE_DEPRECATED_API 0369 /** 0370 * Does the same thing as compare but limits the comparison to a specified 0371 * length 0372 * @param source the source string to be compared with. 0373 * @param target the string that is to be compared with the source string. 0374 * @param length the length the comparison is limited to 0375 * @return Returns a byte value. GREATER if source (up to the specified 0376 * length) is greater than target; EQUAL if source (up to specified 0377 * length) is equal to target; LESS if source (up to the specified 0378 * length) is less than target. 0379 * @deprecated ICU 2.6 use the overload with UErrorCode & 0380 */ 0381 virtual EComparisonResult compare(const UnicodeString& source, 0382 const UnicodeString& target, 0383 int32_t length) const; 0384 #endif // U_FORCE_HIDE_DEPRECATED_API 0385 0386 /** 0387 * Does the same thing as compare but limits the comparison to a specified 0388 * length 0389 * @param source the source string to be compared with. 0390 * @param target the string that is to be compared with the source string. 0391 * @param length the length the comparison is limited to 0392 * @param status possible error code 0393 * @return Returns an enum value. UCOL_GREATER if source (up to the specified 0394 * length) is greater than target; UCOL_EQUAL if source (up to specified 0395 * length) is equal to target; UCOL_LESS if source (up to the specified 0396 * length) is less than target. 0397 * @stable ICU 2.6 0398 */ 0399 virtual UCollationResult compare(const UnicodeString& source, 0400 const UnicodeString& target, 0401 int32_t length, 0402 UErrorCode &status) const = 0; 0403 0404 #ifndef U_FORCE_HIDE_DEPRECATED_API 0405 /** 0406 * The comparison function compares the character data stored in two 0407 * different string arrays. Returns information about whether a string array 0408 * is less than, greater than or equal to another string array. 0409 * <p>Example of use: 0410 * <pre> 0411 * . char16_t ABC[] = {0x41, 0x42, 0x43, 0}; // = "ABC" 0412 * . char16_t abc[] = {0x61, 0x62, 0x63, 0}; // = "abc" 0413 * . UErrorCode status = U_ZERO_ERROR; 0414 * . Collator *myCollation = 0415 * . Collator::createInstance(Locale::getUS(), status); 0416 * . if (U_FAILURE(status)) return; 0417 * . myCollation->setStrength(Collator::PRIMARY); 0418 * . // result would be Collator::EQUAL ("abc" == "ABC") 0419 * . // (no primary difference between "abc" and "ABC") 0420 * . Collator::EComparisonResult result = 0421 * . myCollation->compare(abc, 3, ABC, 3); 0422 * . myCollation->setStrength(Collator::TERTIARY); 0423 * . // result would be Collator::LESS ("abc" <<< "ABC") 0424 * . // (with tertiary difference between "abc" and "ABC") 0425 * . result = myCollation->compare(abc, 3, ABC, 3); 0426 * </pre> 0427 * @param source the source string array to be compared with. 0428 * @param sourceLength the length of the source string array. If this value 0429 * is equal to -1, the string array is null-terminated. 0430 * @param target the string that is to be compared with the source string. 0431 * @param targetLength the length of the target string array. If this value 0432 * is equal to -1, the string array is null-terminated. 0433 * @return Returns a byte value. GREATER if source is greater than target; 0434 * EQUAL if source is equal to target; LESS if source is less than 0435 * target 0436 * @deprecated ICU 2.6 use the overload with UErrorCode & 0437 */ 0438 virtual EComparisonResult compare(const char16_t* source, int32_t sourceLength, 0439 const char16_t* target, int32_t targetLength) 0440 const; 0441 #endif // U_FORCE_HIDE_DEPRECATED_API 0442 0443 /** 0444 * The comparison function compares the character data stored in two 0445 * different string arrays. Returns information about whether a string array 0446 * is less than, greater than or equal to another string array. 0447 * @param source the source string array to be compared with. 0448 * @param sourceLength the length of the source string array. If this value 0449 * is equal to -1, the string array is null-terminated. 0450 * @param target the string that is to be compared with the source string. 0451 * @param targetLength the length of the target string array. If this value 0452 * is equal to -1, the string array is null-terminated. 0453 * @param status possible error code 0454 * @return Returns an enum value. UCOL_GREATER if source is greater 0455 * than target; UCOL_EQUAL if source is equal to target; UCOL_LESS if source is less 0456 * than target 0457 * @stable ICU 2.6 0458 */ 0459 virtual UCollationResult compare(const char16_t* source, int32_t sourceLength, 0460 const char16_t* target, int32_t targetLength, 0461 UErrorCode &status) const = 0; 0462 0463 /** 0464 * Compares two strings using the Collator. 0465 * Returns whether the first one compares less than/equal to/greater than 0466 * the second one. 0467 * This version takes UCharIterator input. 0468 * @param sIter the first ("source") string iterator 0469 * @param tIter the second ("target") string iterator 0470 * @param status ICU status 0471 * @return UCOL_LESS, UCOL_EQUAL or UCOL_GREATER 0472 * @stable ICU 4.2 0473 */ 0474 virtual UCollationResult compare(UCharIterator &sIter, 0475 UCharIterator &tIter, 0476 UErrorCode &status) const; 0477 0478 /** 0479 * Compares two UTF-8 strings using the Collator. 0480 * Returns whether the first one compares less than/equal to/greater than 0481 * the second one. 0482 * This version takes UTF-8 input. 0483 * Note that a StringPiece can be implicitly constructed 0484 * from a std::string or a NUL-terminated const char * string. 0485 * @param source the first UTF-8 string 0486 * @param target the second UTF-8 string 0487 * @param status ICU status 0488 * @return UCOL_LESS, UCOL_EQUAL or UCOL_GREATER 0489 * @stable ICU 4.2 0490 */ 0491 virtual UCollationResult compareUTF8(const StringPiece &source, 0492 const StringPiece &target, 0493 UErrorCode &status) const; 0494 0495 /** 0496 * Transforms the string into a series of characters that can be compared 0497 * with CollationKey::compareTo. It is not possible to restore the original 0498 * string from the chars in the sort key. 0499 * <p>Use CollationKey::equals or CollationKey::compare to compare the 0500 * generated sort keys. 0501 * If the source string is null, a null collation key will be returned. 0502 * 0503 * Note that sort keys are often less efficient than simply doing comparison. 0504 * For more details, see the ICU User Guide. 0505 * 0506 * @param source the source string to be transformed into a sort key. 0507 * @param key the collation key to be filled in 0508 * @param status the error code status. 0509 * @return the collation key of the string based on the collation rules. 0510 * @see CollationKey#compare 0511 * @stable ICU 2.0 0512 */ 0513 virtual CollationKey& getCollationKey(const UnicodeString& source, 0514 CollationKey& key, 0515 UErrorCode& status) const = 0; 0516 0517 /** 0518 * Transforms the string into a series of characters that can be compared 0519 * with CollationKey::compareTo. It is not possible to restore the original 0520 * string from the chars in the sort key. 0521 * <p>Use CollationKey::equals or CollationKey::compare to compare the 0522 * generated sort keys. 0523 * <p>If the source string is null, a null collation key will be returned. 0524 * 0525 * Note that sort keys are often less efficient than simply doing comparison. 0526 * For more details, see the ICU User Guide. 0527 * 0528 * @param source the source string to be transformed into a sort key. 0529 * @param sourceLength length of the collation key 0530 * @param key the collation key to be filled in 0531 * @param status the error code status. 0532 * @return the collation key of the string based on the collation rules. 0533 * @see CollationKey#compare 0534 * @stable ICU 2.0 0535 */ 0536 virtual CollationKey& getCollationKey(const char16_t*source, 0537 int32_t sourceLength, 0538 CollationKey& key, 0539 UErrorCode& status) const = 0; 0540 /** 0541 * Generates the hash code for the collation object 0542 * @stable ICU 2.0 0543 */ 0544 virtual int32_t hashCode() const = 0; 0545 0546 #ifndef U_FORCE_HIDE_DEPRECATED_API 0547 /** 0548 * Gets the locale of the Collator 0549 * 0550 * @param type can be either requested, valid or actual locale. For more 0551 * information see the definition of ULocDataLocaleType in 0552 * uloc.h 0553 * @param status the error code status. 0554 * @return locale where the collation data lives. If the collator 0555 * was instantiated from rules, locale is empty. 0556 * @deprecated ICU 2.8 This API is under consideration for revision 0557 * in ICU 3.0. 0558 */ 0559 virtual Locale getLocale(ULocDataLocaleType type, UErrorCode& status) const = 0; 0560 #endif // U_FORCE_HIDE_DEPRECATED_API 0561 0562 /** 0563 * Convenience method for comparing two strings based on the collation rules. 0564 * @param source the source string to be compared with. 0565 * @param target the target string to be compared with. 0566 * @return true if the first string is greater than the second one, 0567 * according to the collation rules. false, otherwise. 0568 * @see Collator#compare 0569 * @stable ICU 2.0 0570 */ 0571 UBool greater(const UnicodeString& source, const UnicodeString& target) 0572 const; 0573 0574 /** 0575 * Convenience method for comparing two strings based on the collation rules. 0576 * @param source the source string to be compared with. 0577 * @param target the target string to be compared with. 0578 * @return true if the first string is greater than or equal to the second 0579 * one, according to the collation rules. false, otherwise. 0580 * @see Collator#compare 0581 * @stable ICU 2.0 0582 */ 0583 UBool greaterOrEqual(const UnicodeString& source, 0584 const UnicodeString& target) const; 0585 0586 /** 0587 * Convenience method for comparing two strings based on the collation rules. 0588 * @param source the source string to be compared with. 0589 * @param target the target string to be compared with. 0590 * @return true if the strings are equal according to the collation rules. 0591 * false, otherwise. 0592 * @see Collator#compare 0593 * @stable ICU 2.0 0594 */ 0595 UBool equals(const UnicodeString& source, const UnicodeString& target) const; 0596 0597 #ifndef U_HIDE_DRAFT_API 0598 0599 /** 0600 * Creates a comparison function object that uses this collator. 0601 * Like <code>std::equal_to</code> but uses the collator instead of <code>operator==</code>. 0602 * @draft ICU 76 0603 */ 0604 inline auto equal_to() const { return Predicate<std::equal_to, UCOL_EQUAL>(*this); } 0605 0606 /** 0607 * Creates a comparison function object that uses this collator. 0608 * Like <code>std::greater</code> but uses the collator instead of <code>operator></code>. 0609 * @draft ICU 76 0610 */ 0611 inline auto greater() const { return Predicate<std::equal_to, UCOL_GREATER>(*this); } 0612 0613 /** 0614 * Creates a comparison function object that uses this collator. 0615 * Like <code>std::less</code> but uses the collator instead of <code>operator<</code>. 0616 * @draft ICU 76 0617 */ 0618 inline auto less() const { return Predicate<std::equal_to, UCOL_LESS>(*this); } 0619 0620 /** 0621 * Creates a comparison function object that uses this collator. 0622 * Like <code>std::not_equal_to</code> but uses the collator instead of <code>operator!=</code>. 0623 * @draft ICU 76 0624 */ 0625 inline auto not_equal_to() const { return Predicate<std::not_equal_to, UCOL_EQUAL>(*this); } 0626 0627 /** 0628 * Creates a comparison function object that uses this collator. 0629 * Like <code>std::greater_equal</code> but uses the collator instead of <code>operator>=</code>. 0630 * @draft ICU 76 0631 */ 0632 inline auto greater_equal() const { return Predicate<std::not_equal_to, UCOL_LESS>(*this); } 0633 0634 /** 0635 * Creates a comparison function object that uses this collator. 0636 * Like <code>std::less_equal</code> but uses the collator instead of <code>operator<=</code>. 0637 * @draft ICU 76 0638 */ 0639 inline auto less_equal() const { return Predicate<std::not_equal_to, UCOL_GREATER>(*this); } 0640 0641 #endif // U_HIDE_DRAFT_API 0642 0643 #ifndef U_FORCE_HIDE_DEPRECATED_API 0644 /** 0645 * Determines the minimum strength that will be used in comparison or 0646 * transformation. 0647 * <p>E.g. with strength == SECONDARY, the tertiary difference is ignored 0648 * <p>E.g. with strength == PRIMARY, the secondary and tertiary difference 0649 * are ignored. 0650 * @return the current comparison level. 0651 * @see Collator#setStrength 0652 * @deprecated ICU 2.6 Use getAttribute(UCOL_STRENGTH...) instead 0653 */ 0654 virtual ECollationStrength getStrength() const; 0655 0656 /** 0657 * Sets the minimum strength to be used in comparison or transformation. 0658 * <p>Example of use: 0659 * <pre> 0660 * \code 0661 * UErrorCode status = U_ZERO_ERROR; 0662 * Collator*myCollation = Collator::createInstance(Locale::getUS(), status); 0663 * if (U_FAILURE(status)) return; 0664 * myCollation->setStrength(Collator::PRIMARY); 0665 * // result will be "abc" == "ABC" 0666 * // tertiary differences will be ignored 0667 * Collator::ComparisonResult result = myCollation->compare("abc", "ABC"); 0668 * \endcode 0669 * </pre> 0670 * @see Collator#getStrength 0671 * @param newStrength the new comparison level. 0672 * @deprecated ICU 2.6 Use setAttribute(UCOL_STRENGTH...) instead 0673 */ 0674 virtual void setStrength(ECollationStrength newStrength); 0675 #endif // U_FORCE_HIDE_DEPRECATED_API 0676 0677 /** 0678 * Retrieves the reordering codes for this collator. 0679 * @param dest The array to fill with the script ordering. 0680 * @param destCapacity The length of dest. If it is 0, then dest may be nullptr and the function 0681 * will only return the length of the result without writing any codes (pre-flighting). 0682 * @param status A reference to an error code value, which must not indicate 0683 * a failure before the function call. 0684 * @return The length of the script ordering array. 0685 * @see ucol_setReorderCodes 0686 * @see Collator#getEquivalentReorderCodes 0687 * @see Collator#setReorderCodes 0688 * @see UScriptCode 0689 * @see UColReorderCode 0690 * @stable ICU 4.8 0691 */ 0692 virtual int32_t getReorderCodes(int32_t *dest, 0693 int32_t destCapacity, 0694 UErrorCode& status) const; 0695 0696 /** 0697 * Sets the ordering of scripts for this collator. 0698 * 0699 * <p>The reordering codes are a combination of script codes and reorder codes. 0700 * @param reorderCodes An array of script codes in the new order. This can be nullptr if the 0701 * length is also set to 0. An empty array will clear any reordering codes on the collator. 0702 * @param reorderCodesLength The length of reorderCodes. 0703 * @param status error code 0704 * @see ucol_setReorderCodes 0705 * @see Collator#getReorderCodes 0706 * @see Collator#getEquivalentReorderCodes 0707 * @see UScriptCode 0708 * @see UColReorderCode 0709 * @stable ICU 4.8 0710 */ 0711 virtual void setReorderCodes(const int32_t* reorderCodes, 0712 int32_t reorderCodesLength, 0713 UErrorCode& status) ; 0714 0715 /** 0716 * Retrieves the reorder codes that are grouped with the given reorder code. Some reorder 0717 * codes will be grouped and must reorder together. 0718 * Beginning with ICU 55, scripts only reorder together if they are primary-equal, 0719 * for example Hiragana and Katakana. 0720 * 0721 * @param reorderCode The reorder code to determine equivalence for. 0722 * @param dest The array to fill with the script equivalence reordering codes. 0723 * @param destCapacity The length of dest. If it is 0, then dest may be nullptr and the 0724 * function will only return the length of the result without writing any codes (pre-flighting). 0725 * @param status A reference to an error code value, which must not indicate 0726 * a failure before the function call. 0727 * @return The length of the of the reordering code equivalence array. 0728 * @see ucol_setReorderCodes 0729 * @see Collator#getReorderCodes 0730 * @see Collator#setReorderCodes 0731 * @see UScriptCode 0732 * @see UColReorderCode 0733 * @stable ICU 4.8 0734 */ 0735 static int32_t U_EXPORT2 getEquivalentReorderCodes(int32_t reorderCode, 0736 int32_t* dest, 0737 int32_t destCapacity, 0738 UErrorCode& status); 0739 0740 /** 0741 * Get name of the object for the desired Locale, in the desired language 0742 * @param objectLocale must be from getAvailableLocales 0743 * @param displayLocale specifies the desired locale for output 0744 * @param name the fill-in parameter of the return value 0745 * @return display-able name of the object for the object locale in the 0746 * desired language 0747 * @stable ICU 2.0 0748 */ 0749 static UnicodeString& U_EXPORT2 getDisplayName(const Locale& objectLocale, 0750 const Locale& displayLocale, 0751 UnicodeString& name); 0752 0753 /** 0754 * Get name of the object for the desired Locale, in the language of the 0755 * default locale. 0756 * @param objectLocale must be from getAvailableLocales 0757 * @param name the fill-in parameter of the return value 0758 * @return name of the object for the desired locale in the default language 0759 * @stable ICU 2.0 0760 */ 0761 static UnicodeString& U_EXPORT2 getDisplayName(const Locale& objectLocale, 0762 UnicodeString& name); 0763 0764 /** 0765 * Get the set of Locales for which Collations are installed. 0766 * 0767 * <p>Note this does not include locales supported by registered collators. 0768 * If collators might have been registered, use the overload of getAvailableLocales 0769 * that returns a StringEnumeration.</p> 0770 * 0771 * @param count the output parameter of number of elements in the locale list 0772 * @return the list of available locales for which collations are installed 0773 * @stable ICU 2.0 0774 */ 0775 static const Locale* U_EXPORT2 getAvailableLocales(int32_t& count); 0776 0777 /** 0778 * Return a StringEnumeration over the locales available at the time of the call, 0779 * including registered locales. If a severe error occurs (such as out of memory 0780 * condition) this will return null. If there is no locale data, an empty enumeration 0781 * will be returned. 0782 * @return a StringEnumeration over the locales available at the time of the call 0783 * @stable ICU 2.6 0784 */ 0785 static StringEnumeration* U_EXPORT2 getAvailableLocales(); 0786 0787 /** 0788 * Create a string enumerator of all possible keywords that are relevant to 0789 * collation. At this point, the only recognized keyword for this 0790 * service is "collation". 0791 * @param status input-output error code 0792 * @return a string enumeration over locale strings. The caller is 0793 * responsible for closing the result. 0794 * @stable ICU 3.0 0795 */ 0796 static StringEnumeration* U_EXPORT2 getKeywords(UErrorCode& status); 0797 0798 /** 0799 * Given a keyword, create a string enumeration of all values 0800 * for that keyword that are currently in use. 0801 * @param keyword a particular keyword as enumerated by 0802 * ucol_getKeywords. If any other keyword is passed in, status is set 0803 * to U_ILLEGAL_ARGUMENT_ERROR. 0804 * @param status input-output error code 0805 * @return a string enumeration over collation keyword values, or nullptr 0806 * upon error. The caller is responsible for deleting the result. 0807 * @stable ICU 3.0 0808 */ 0809 static StringEnumeration* U_EXPORT2 getKeywordValues(const char *keyword, UErrorCode& status); 0810 0811 /** 0812 * Given a key and a locale, returns an array of string values in a preferred 0813 * order that would make a difference. These are all and only those values where 0814 * the open (creation) of the service with the locale formed from the input locale 0815 * plus input keyword and that value has different behavior than creation with the 0816 * input locale alone. 0817 * @param keyword one of the keys supported by this service. For now, only 0818 * "collation" is supported. 0819 * @param locale the locale 0820 * @param commonlyUsed if set to true it will return only commonly used values 0821 * with the given locale in preferred order. Otherwise, 0822 * it will return all the available values for the locale. 0823 * @param status ICU status 0824 * @return a string enumeration over keyword values for the given key and the locale. 0825 * @stable ICU 4.2 0826 */ 0827 static StringEnumeration* U_EXPORT2 getKeywordValuesForLocale(const char* keyword, const Locale& locale, 0828 UBool commonlyUsed, UErrorCode& status); 0829 0830 /** 0831 * Return the functionally equivalent locale for the given 0832 * requested locale, with respect to given keyword, for the 0833 * collation service. If two locales return the same result, then 0834 * collators instantiated for these locales will behave 0835 * equivalently. The converse is not always true; two collators 0836 * may in fact be equivalent, but return different results, due to 0837 * internal details. The return result has no other meaning than 0838 * that stated above, and implies nothing as to the relationship 0839 * between the two locales. This is intended for use by 0840 * applications who wish to cache collators, or otherwise reuse 0841 * collators when possible. The functional equivalent may change 0842 * over time. For more information, please see the <a 0843 * href="https://unicode-org.github.io/icu/userguide/locale#locales-and-services"> 0844 * Locales and Services</a> section of the ICU User Guide. 0845 * @param keyword a particular keyword as enumerated by 0846 * ucol_getKeywords. 0847 * @param locale the requested locale 0848 * @param isAvailable reference to a fillin parameter that 0849 * indicates whether the requested locale was 'available' to the 0850 * collation service. A locale is defined as 'available' if it 0851 * physically exists within the collation locale data. 0852 * @param status reference to input-output error code 0853 * @return the functionally equivalent collation locale, or the root 0854 * locale upon error. 0855 * @stable ICU 3.0 0856 */ 0857 static Locale U_EXPORT2 getFunctionalEquivalent(const char* keyword, const Locale& locale, 0858 UBool& isAvailable, UErrorCode& status); 0859 0860 #if !UCONFIG_NO_SERVICE 0861 /** 0862 * Register a new Collator. The collator will be adopted. 0863 * Because ICU may choose to cache collators internally, this must be 0864 * called at application startup, prior to any calls to 0865 * Collator::createInstance to avoid undefined behavior. 0866 * @param toAdopt the Collator instance to be adopted 0867 * @param locale the locale with which the collator will be associated 0868 * @param status the in/out status code, no special meanings are assigned 0869 * @return a registry key that can be used to unregister this collator 0870 * @stable ICU 2.6 0871 */ 0872 static URegistryKey U_EXPORT2 registerInstance(Collator* toAdopt, const Locale& locale, UErrorCode& status); 0873 0874 /** 0875 * Register a new CollatorFactory. The factory will be adopted. 0876 * Because ICU may choose to cache collators internally, this must be 0877 * called at application startup, prior to any calls to 0878 * Collator::createInstance to avoid undefined behavior. 0879 * @param toAdopt the CollatorFactory instance to be adopted 0880 * @param status the in/out status code, no special meanings are assigned 0881 * @return a registry key that can be used to unregister this collator 0882 * @stable ICU 2.6 0883 */ 0884 static URegistryKey U_EXPORT2 registerFactory(CollatorFactory* toAdopt, UErrorCode& status); 0885 0886 /** 0887 * Unregister a previously-registered Collator or CollatorFactory 0888 * using the key returned from the register call. Key becomes 0889 * invalid after a successful call and should not be used again. 0890 * The object corresponding to the key will be deleted. 0891 * Because ICU may choose to cache collators internally, this should 0892 * be called during application shutdown, after all calls to 0893 * Collator::createInstance to avoid undefined behavior. 0894 * @param key the registry key returned by a previous call to registerInstance 0895 * @param status the in/out status code, no special meanings are assigned 0896 * @return true if the collator for the key was successfully unregistered 0897 * @stable ICU 2.6 0898 */ 0899 static UBool U_EXPORT2 unregister(URegistryKey key, UErrorCode& status); 0900 #endif /* UCONFIG_NO_SERVICE */ 0901 0902 /** 0903 * Gets the version information for a Collator. 0904 * @param info the version # information, the result will be filled in 0905 * @stable ICU 2.0 0906 */ 0907 virtual void getVersion(UVersionInfo info) const = 0; 0908 0909 /** 0910 * Returns a unique class ID POLYMORPHICALLY. Pure virtual method. 0911 * This method is to implement a simple version of RTTI, since not all C++ 0912 * compilers support genuine RTTI. Polymorphic operator==() and clone() 0913 * methods call this method. 0914 * @return The class ID for this object. All objects of a given class have 0915 * the same class ID. Objects of other classes have different class 0916 * IDs. 0917 * @stable ICU 2.0 0918 */ 0919 virtual UClassID getDynamicClassID() const override = 0; 0920 0921 /** 0922 * Universal attribute setter 0923 * @param attr attribute type 0924 * @param value attribute value 0925 * @param status to indicate whether the operation went on smoothly or 0926 * there were errors 0927 * @stable ICU 2.2 0928 */ 0929 virtual void setAttribute(UColAttribute attr, UColAttributeValue value, 0930 UErrorCode &status) = 0; 0931 0932 /** 0933 * Universal attribute getter 0934 * @param attr attribute type 0935 * @param status to indicate whether the operation went on smoothly or 0936 * there were errors 0937 * @return attribute value 0938 * @stable ICU 2.2 0939 */ 0940 virtual UColAttributeValue getAttribute(UColAttribute attr, 0941 UErrorCode &status) const = 0; 0942 0943 /** 0944 * Sets the variable top to the top of the specified reordering group. 0945 * The variable top determines the highest-sorting character 0946 * which is affected by UCOL_ALTERNATE_HANDLING. 0947 * If that attribute is set to UCOL_NON_IGNORABLE, then the variable top has no effect. 0948 * 0949 * The base class implementation sets U_UNSUPPORTED_ERROR. 0950 * @param group one of UCOL_REORDER_CODE_SPACE, UCOL_REORDER_CODE_PUNCTUATION, 0951 * UCOL_REORDER_CODE_SYMBOL, UCOL_REORDER_CODE_CURRENCY; 0952 * or UCOL_REORDER_CODE_DEFAULT to restore the default max variable group 0953 * @param errorCode Standard ICU error code. Its input value must 0954 * pass the U_SUCCESS() test, or else the function returns 0955 * immediately. Check for U_FAILURE() on output or use with 0956 * function chaining. (See User Guide for details.) 0957 * @return *this 0958 * @see getMaxVariable 0959 * @stable ICU 53 0960 */ 0961 virtual Collator &setMaxVariable(UColReorderCode group, UErrorCode &errorCode); 0962 0963 /** 0964 * Returns the maximum reordering group whose characters are affected by UCOL_ALTERNATE_HANDLING. 0965 * 0966 * The base class implementation returns UCOL_REORDER_CODE_PUNCTUATION. 0967 * @return the maximum variable reordering group. 0968 * @see setMaxVariable 0969 * @stable ICU 53 0970 */ 0971 virtual UColReorderCode getMaxVariable() const; 0972 0973 #ifndef U_FORCE_HIDE_DEPRECATED_API 0974 /** 0975 * Sets the variable top to the primary weight of the specified string. 0976 * 0977 * Beginning with ICU 53, the variable top is pinned to 0978 * the top of one of the supported reordering groups, 0979 * and it must not be beyond the last of those groups. 0980 * See setMaxVariable(). 0981 * @param varTop one or more (if contraction) char16_ts to which the variable top should be set 0982 * @param len length of variable top string. If -1 it is considered to be zero terminated. 0983 * @param status error code. If error code is set, the return value is undefined. Errors set by this function are: <br> 0984 * U_CE_NOT_FOUND_ERROR if more than one character was passed and there is no such contraction<br> 0985 * U_ILLEGAL_ARGUMENT_ERROR if the variable top is beyond 0986 * the last reordering group supported by setMaxVariable() 0987 * @return variable top primary weight 0988 * @deprecated ICU 53 Call setMaxVariable() instead. 0989 */ 0990 virtual uint32_t setVariableTop(const char16_t *varTop, int32_t len, UErrorCode &status) = 0; 0991 0992 /** 0993 * Sets the variable top to the primary weight of the specified string. 0994 * 0995 * Beginning with ICU 53, the variable top is pinned to 0996 * the top of one of the supported reordering groups, 0997 * and it must not be beyond the last of those groups. 0998 * See setMaxVariable(). 0999 * @param varTop a UnicodeString size 1 or more (if contraction) of char16_ts to which the variable top should be set 1000 * @param status error code. If error code is set, the return value is undefined. Errors set by this function are: <br> 1001 * U_CE_NOT_FOUND_ERROR if more than one character was passed and there is no such contraction<br> 1002 * U_ILLEGAL_ARGUMENT_ERROR if the variable top is beyond 1003 * the last reordering group supported by setMaxVariable() 1004 * @return variable top primary weight 1005 * @deprecated ICU 53 Call setMaxVariable() instead. 1006 */ 1007 virtual uint32_t setVariableTop(const UnicodeString &varTop, UErrorCode &status) = 0; 1008 1009 /** 1010 * Sets the variable top to the specified primary weight. 1011 * 1012 * Beginning with ICU 53, the variable top is pinned to 1013 * the top of one of the supported reordering groups, 1014 * and it must not be beyond the last of those groups. 1015 * See setMaxVariable(). 1016 * @param varTop primary weight, as returned by setVariableTop or ucol_getVariableTop 1017 * @param status error code 1018 * @deprecated ICU 53 Call setMaxVariable() instead. 1019 */ 1020 virtual void setVariableTop(uint32_t varTop, UErrorCode &status) = 0; 1021 #endif // U_FORCE_HIDE_DEPRECATED_API 1022 1023 /** 1024 * Gets the variable top value of a Collator. 1025 * @param status error code (not changed by function). If error code is set, the return value is undefined. 1026 * @return the variable top primary weight 1027 * @see getMaxVariable 1028 * @stable ICU 2.0 1029 */ 1030 virtual uint32_t getVariableTop(UErrorCode &status) const = 0; 1031 1032 /** 1033 * Get a UnicodeSet that contains all the characters and sequences 1034 * tailored in this collator. 1035 * @param status error code of the operation 1036 * @return a pointer to a UnicodeSet object containing all the 1037 * code points and sequences that may sort differently than 1038 * in the root collator. The object must be disposed of by using delete 1039 * @stable ICU 2.4 1040 */ 1041 virtual UnicodeSet *getTailoredSet(UErrorCode &status) const; 1042 1043 #ifndef U_FORCE_HIDE_DEPRECATED_API 1044 /** 1045 * Same as clone(). 1046 * The base class implementation simply calls clone(). 1047 * @return a copy of this object, owned by the caller 1048 * @see clone() 1049 * @deprecated ICU 50 no need to have two methods for cloning 1050 */ 1051 virtual Collator* safeClone() const; 1052 #endif // U_FORCE_HIDE_DEPRECATED_API 1053 1054 /** 1055 * Get the sort key as an array of bytes from a UnicodeString. 1056 * Sort key byte arrays are zero-terminated and can be compared using 1057 * strcmp(). 1058 * 1059 * Note that sort keys are often less efficient than simply doing comparison. 1060 * For more details, see the ICU User Guide. 1061 * 1062 * @param source string to be processed. 1063 * @param result buffer to store result in. If nullptr, number of bytes needed 1064 * will be returned. 1065 * @param resultLength length of the result buffer. If if not enough the 1066 * buffer will be filled to capacity. 1067 * @return Number of bytes needed for storing the sort key 1068 * @stable ICU 2.2 1069 */ 1070 virtual int32_t getSortKey(const UnicodeString& source, 1071 uint8_t* result, 1072 int32_t resultLength) const = 0; 1073 1074 /** 1075 * Get the sort key as an array of bytes from a char16_t buffer. 1076 * Sort key byte arrays are zero-terminated and can be compared using 1077 * strcmp(). 1078 * 1079 * Note that sort keys are often less efficient than simply doing comparison. 1080 * For more details, see the ICU User Guide. 1081 * 1082 * @param source string to be processed. 1083 * @param sourceLength length of string to be processed. 1084 * If -1, the string is 0 terminated and length will be decided by the 1085 * function. 1086 * @param result buffer to store result in. If nullptr, number of bytes needed 1087 * will be returned. 1088 * @param resultLength length of the result buffer. If if not enough the 1089 * buffer will be filled to capacity. 1090 * @return Number of bytes needed for storing the sort key 1091 * @stable ICU 2.2 1092 */ 1093 virtual int32_t getSortKey(const char16_t*source, int32_t sourceLength, 1094 uint8_t*result, int32_t resultLength) const = 0; 1095 1096 /** 1097 * Produce a bound for a given sortkey and a number of levels. 1098 * Return value is always the number of bytes needed, regardless of 1099 * whether the result buffer was big enough or even valid.<br> 1100 * Resulting bounds can be used to produce a range of strings that are 1101 * between upper and lower bounds. For example, if bounds are produced 1102 * for a sortkey of string "smith", strings between upper and lower 1103 * bounds with one level would include "Smith", "SMITH", "sMiTh".<br> 1104 * There are two upper bounds that can be produced. If UCOL_BOUND_UPPER 1105 * is produced, strings matched would be as above. However, if bound 1106 * produced using UCOL_BOUND_UPPER_LONG is used, the above example will 1107 * also match "Smithsonian" and similar.<br> 1108 * For more on usage, see example in cintltst/capitst.c in procedure 1109 * TestBounds. 1110 * Sort keys may be compared using <TT>strcmp</TT>. 1111 * @param source The source sortkey. 1112 * @param sourceLength The length of source, or -1 if null-terminated. 1113 * (If an unmodified sortkey is passed, it is always null 1114 * terminated). 1115 * @param boundType Type of bound required. It can be UCOL_BOUND_LOWER, which 1116 * produces a lower inclusive bound, UCOL_BOUND_UPPER, that 1117 * produces upper bound that matches strings of the same length 1118 * or UCOL_BOUND_UPPER_LONG that matches strings that have the 1119 * same starting substring as the source string. 1120 * @param noOfLevels Number of levels required in the resulting bound (for most 1121 * uses, the recommended value is 1). See users guide for 1122 * explanation on number of levels a sortkey can have. 1123 * @param result A pointer to a buffer to receive the resulting sortkey. 1124 * @param resultLength The maximum size of result. 1125 * @param status Used for returning error code if something went wrong. If the 1126 * number of levels requested is higher than the number of levels 1127 * in the source key, a warning (U_SORT_KEY_TOO_SHORT_WARNING) is 1128 * issued. 1129 * @return The size needed to fully store the bound. 1130 * @see ucol_keyHashCode 1131 * @stable ICU 2.1 1132 */ 1133 static int32_t U_EXPORT2 getBound(const uint8_t *source, 1134 int32_t sourceLength, 1135 UColBoundMode boundType, 1136 uint32_t noOfLevels, 1137 uint8_t *result, 1138 int32_t resultLength, 1139 UErrorCode &status); 1140 1141 1142 protected: 1143 1144 // Collator protected constructors ------------------------------------- 1145 1146 /** 1147 * Default constructor. 1148 * Constructor is different from the old default Collator constructor. 1149 * The task for determining the default collation strength and normalization 1150 * mode is left to the child class. 1151 * @stable ICU 2.0 1152 */ 1153 Collator(); 1154 1155 #ifndef U_HIDE_DEPRECATED_API 1156 /** 1157 * Constructor. 1158 * Empty constructor, does not handle the arguments. 1159 * This constructor is done for backward compatibility with 1.7 and 1.8. 1160 * The task for handling the argument collation strength and normalization 1161 * mode is left to the child class. 1162 * @param collationStrength collation strength 1163 * @param decompositionMode 1164 * @deprecated ICU 2.4. Subclasses should use the default constructor 1165 * instead and handle the strength and normalization mode themselves. 1166 */ 1167 Collator(UCollationStrength collationStrength, 1168 UNormalizationMode decompositionMode); 1169 #endif /* U_HIDE_DEPRECATED_API */ 1170 1171 /** 1172 * Copy constructor. 1173 * @param other Collator object to be copied from 1174 * @stable ICU 2.0 1175 */ 1176 Collator(const Collator& other); 1177 1178 public: 1179 /** 1180 * Used internally by registration to define the requested and valid locales. 1181 * @param requestedLocale the requested locale 1182 * @param validLocale the valid locale 1183 * @param actualLocale the actual locale 1184 * @internal 1185 */ 1186 virtual void setLocales(const Locale& requestedLocale, const Locale& validLocale, const Locale& actualLocale); 1187 1188 /** Get the short definition string for a collator. This internal API harvests the collator's 1189 * locale and the attribute set and produces a string that can be used for opening 1190 * a collator with the same attributes using the ucol_openFromShortString API. 1191 * This string will be normalized. 1192 * The structure and the syntax of the string is defined in the "Naming collators" 1193 * section of the users guide: 1194 * https://unicode-org.github.io/icu/userguide/collation/concepts#collator-naming-scheme 1195 * This function supports preflighting. 1196 * 1197 * This is internal, and intended to be used with delegate converters. 1198 * 1199 * @param locale a locale that will appear as a collators locale in the resulting 1200 * short string definition. If nullptr, the locale will be harvested 1201 * from the collator. 1202 * @param buffer space to hold the resulting string 1203 * @param capacity capacity of the buffer 1204 * @param status for returning errors. All the preflighting errors are featured 1205 * @return length of the resulting string 1206 * @see ucol_openFromShortString 1207 * @see ucol_normalizeShortDefinitionString 1208 * @see ucol_getShortDefinitionString 1209 * @internal 1210 */ 1211 virtual int32_t internalGetShortDefinitionString(const char *locale, 1212 char *buffer, 1213 int32_t capacity, 1214 UErrorCode &status) const; 1215 1216 /** 1217 * Implements ucol_strcollUTF8(). 1218 * @internal 1219 */ 1220 virtual UCollationResult internalCompareUTF8( 1221 const char *left, int32_t leftLength, 1222 const char *right, int32_t rightLength, 1223 UErrorCode &errorCode) const; 1224 1225 /** 1226 * Implements ucol_nextSortKeyPart(). 1227 * @internal 1228 */ 1229 virtual int32_t 1230 internalNextSortKeyPart( 1231 UCharIterator *iter, uint32_t state[2], 1232 uint8_t *dest, int32_t count, UErrorCode &errorCode) const; 1233 1234 #ifndef U_HIDE_INTERNAL_API 1235 /** @internal */ 1236 static inline Collator *fromUCollator(UCollator *uc) { 1237 return reinterpret_cast<Collator *>(uc); 1238 } 1239 /** @internal */ 1240 static inline const Collator *fromUCollator(const UCollator *uc) { 1241 return reinterpret_cast<const Collator *>(uc); 1242 } 1243 /** @internal */ 1244 inline UCollator *toUCollator() { 1245 return reinterpret_cast<UCollator *>(this); 1246 } 1247 /** @internal */ 1248 inline const UCollator *toUCollator() const { 1249 return reinterpret_cast<const UCollator *>(this); 1250 } 1251 #endif // U_HIDE_INTERNAL_API 1252 1253 private: 1254 /** 1255 * Assignment operator. Private for now. 1256 */ 1257 Collator& operator=(const Collator& other) = delete; 1258 1259 friend class CFactory; 1260 friend class SimpleCFactory; 1261 friend class ICUCollatorFactory; 1262 friend class ICUCollatorService; 1263 static Collator* makeInstance(const Locale& desiredLocale, 1264 UErrorCode& status); 1265 1266 #ifndef U_HIDE_DRAFT_API 1267 /** 1268 * Function object for performing comparisons using a Collator. 1269 * @internal 1270 */ 1271 template <template <typename...> typename Compare, UCollationResult result> 1272 class Predicate { 1273 public: 1274 explicit Predicate(const Collator& parent) : collator(parent) {} 1275 1276 template < 1277 typename T, typename U, 1278 typename = std::enable_if_t<ConvertibleToU16StringView<T> && ConvertibleToU16StringView<U>>> 1279 bool operator()(const T& lhs, const U& rhs) const { 1280 UErrorCode status = U_ZERO_ERROR; 1281 return compare( 1282 collator.compare( 1283 UnicodeString::readOnlyAlias(lhs), 1284 UnicodeString::readOnlyAlias(rhs), 1285 status), 1286 result); 1287 } 1288 1289 bool operator()(std::string_view lhs, std::string_view rhs) const { 1290 UErrorCode status = U_ZERO_ERROR; 1291 return compare(collator.compareUTF8(lhs, rhs, status), result); 1292 } 1293 1294 #if defined(__cpp_char8_t) 1295 bool operator()(std::u8string_view lhs, std::u8string_view rhs) const { 1296 UErrorCode status = U_ZERO_ERROR; 1297 return compare(collator.compareUTF8(lhs, rhs, status), result); 1298 } 1299 #endif 1300 1301 private: 1302 const Collator& collator; 1303 static constexpr Compare<UCollationResult> compare{}; 1304 }; 1305 #endif // U_HIDE_DRAFT_API 1306 }; 1307 1308 #if !UCONFIG_NO_SERVICE 1309 /** 1310 * A factory, used with registerFactory, the creates multiple collators and provides 1311 * display names for them. A factory supports some number of locales-- these are the 1312 * locales for which it can create collators. The factory can be visible, in which 1313 * case the supported locales will be enumerated by getAvailableLocales, or invisible, 1314 * in which they are not. Invisible locales are still supported, they are just not 1315 * listed by getAvailableLocales. 1316 * <p> 1317 * If standard locale display names are sufficient, Collator instances can 1318 * be registered using registerInstance instead.</p> 1319 * <p> 1320 * Note: if the collators are to be used from C APIs, they must be instances 1321 * of RuleBasedCollator.</p> 1322 * 1323 * @stable ICU 2.6 1324 */ 1325 class U_I18N_API CollatorFactory : public UObject { 1326 public: 1327 1328 /** 1329 * Destructor 1330 * @stable ICU 3.0 1331 */ 1332 virtual ~CollatorFactory(); 1333 1334 /** 1335 * Return true if this factory is visible. Default is true. 1336 * If not visible, the locales supported by this factory will not 1337 * be listed by getAvailableLocales. 1338 * @return true if the factory is visible. 1339 * @stable ICU 2.6 1340 */ 1341 virtual UBool visible() const; 1342 1343 /** 1344 * Return a collator for the provided locale. If the locale 1345 * is not supported, return nullptr. 1346 * @param loc the locale identifying the collator to be created. 1347 * @return a new collator if the locale is supported, otherwise nullptr. 1348 * @stable ICU 2.6 1349 */ 1350 virtual Collator* createCollator(const Locale& loc) = 0; 1351 1352 /** 1353 * Return the name of the collator for the objectLocale, localized for the displayLocale. 1354 * If objectLocale is not supported, or the factory is not visible, set the result string 1355 * to bogus. 1356 * @param objectLocale the locale identifying the collator 1357 * @param displayLocale the locale for which the display name of the collator should be localized 1358 * @param result an output parameter for the display name, set to bogus if not supported. 1359 * @return the display name 1360 * @stable ICU 2.6 1361 */ 1362 virtual UnicodeString& getDisplayName(const Locale& objectLocale, 1363 const Locale& displayLocale, 1364 UnicodeString& result); 1365 1366 /** 1367 * Return an array of all the locale names directly supported by this factory. 1368 * The number of names is returned in count. This array is owned by the factory. 1369 * Its contents must never change. 1370 * @param count output parameter for the number of locales supported by the factory 1371 * @param status the in/out error code 1372 * @return a pointer to an array of count UnicodeStrings. 1373 * @stable ICU 2.6 1374 */ 1375 virtual const UnicodeString * getSupportedIDs(int32_t &count, UErrorCode& status) = 0; 1376 }; 1377 #endif /* UCONFIG_NO_SERVICE */ 1378 1379 // Collator inline methods ----------------------------------------------- 1380 1381 U_NAMESPACE_END 1382 1383 #endif /* #if !UCONFIG_NO_COLLATION */ 1384 1385 #endif /* U_SHOW_CPLUSPLUS_API */ 1386 1387 #endif
| [ Source navigation ] | [ Diff markup ] | [ Identifier search ] | [ general search ] |
|
This page was automatically generated by the 2.3.7 LXR engine. The LXR team |
|