jpayne@69
|
1 // © 2016 and later: Unicode, Inc. and others.
|
jpayne@69
|
2 // License & terms of use: http://www.unicode.org/copyright.html
|
jpayne@69
|
3 /*
|
jpayne@69
|
4 ******************************************************************************
|
jpayne@69
|
5 * Copyright (C) 1996-2016, International Business Machines
|
jpayne@69
|
6 * Corporation and others. All Rights Reserved.
|
jpayne@69
|
7 ******************************************************************************
|
jpayne@69
|
8 */
|
jpayne@69
|
9
|
jpayne@69
|
10 /**
|
jpayne@69
|
11 * \file
|
jpayne@69
|
12 * \brief C++ API: Collation Service.
|
jpayne@69
|
13 */
|
jpayne@69
|
14
|
jpayne@69
|
15 /**
|
jpayne@69
|
16 * File coll.h
|
jpayne@69
|
17 *
|
jpayne@69
|
18 * Created by: Helena Shih
|
jpayne@69
|
19 *
|
jpayne@69
|
20 * Modification History:
|
jpayne@69
|
21 *
|
jpayne@69
|
22 * Date Name Description
|
jpayne@69
|
23 * 02/5/97 aliu Modified createDefault to load collation data from
|
jpayne@69
|
24 * binary files when possible. Added related methods
|
jpayne@69
|
25 * createCollationFromFile, chopLocale, createPathName.
|
jpayne@69
|
26 * 02/11/97 aliu Added members addToCache, findInCache, and fgCache.
|
jpayne@69
|
27 * 02/12/97 aliu Modified to create objects from RuleBasedCollator cache.
|
jpayne@69
|
28 * Moved cache out of Collation class.
|
jpayne@69
|
29 * 02/13/97 aliu Moved several methods out of this class and into
|
jpayne@69
|
30 * RuleBasedCollator, with modifications. Modified
|
jpayne@69
|
31 * createDefault() to call new RuleBasedCollator(Locale&)
|
jpayne@69
|
32 * constructor. General clean up and documentation.
|
jpayne@69
|
33 * 02/20/97 helena Added clone, operator==, operator!=, operator=, copy
|
jpayne@69
|
34 * constructor and getDynamicClassID.
|
jpayne@69
|
35 * 03/25/97 helena Updated with platform independent data types.
|
jpayne@69
|
36 * 05/06/97 helena Added memory allocation error detection.
|
jpayne@69
|
37 * 06/20/97 helena Java class name change.
|
jpayne@69
|
38 * 09/03/97 helena Added createCollationKeyValues().
|
jpayne@69
|
39 * 02/10/98 damiba Added compare() with length as parameter.
|
jpayne@69
|
40 * 04/23/99 stephen Removed EDecompositionMode, merged with
|
jpayne@69
|
41 * Normalizer::EMode.
|
jpayne@69
|
42 * 11/02/99 helena Collator performance enhancements. Eliminates the
|
jpayne@69
|
43 * UnicodeString construction and special case for NO_OP.
|
jpayne@69
|
44 * 11/23/99 srl More performance enhancements. Inlining of
|
jpayne@69
|
45 * critical accessors.
|
jpayne@69
|
46 * 05/15/00 helena Added version information API.
|
jpayne@69
|
47 * 01/29/01 synwee Modified into a C++ wrapper which calls C apis
|
jpayne@69
|
48 * (ucol.h).
|
jpayne@69
|
49 * 2012-2014 markus Rewritten in C++ again.
|
jpayne@69
|
50 */
|
jpayne@69
|
51
|
jpayne@69
|
52 #ifndef COLL_H
|
jpayne@69
|
53 #define COLL_H
|
jpayne@69
|
54
|
jpayne@69
|
55 #include "unicode/utypes.h"
|
jpayne@69
|
56
|
jpayne@69
|
57 #if U_SHOW_CPLUSPLUS_API
|
jpayne@69
|
58
|
jpayne@69
|
59 #if !UCONFIG_NO_COLLATION
|
jpayne@69
|
60
|
jpayne@69
|
61 #include "unicode/uobject.h"
|
jpayne@69
|
62 #include "unicode/ucol.h"
|
jpayne@69
|
63 #include "unicode/unorm.h"
|
jpayne@69
|
64 #include "unicode/locid.h"
|
jpayne@69
|
65 #include "unicode/uniset.h"
|
jpayne@69
|
66 #include "unicode/umisc.h"
|
jpayne@69
|
67 #include "unicode/uiter.h"
|
jpayne@69
|
68 #include "unicode/stringpiece.h"
|
jpayne@69
|
69
|
jpayne@69
|
70 U_NAMESPACE_BEGIN
|
jpayne@69
|
71
|
jpayne@69
|
72 class StringEnumeration;
|
jpayne@69
|
73
|
jpayne@69
|
74 #if !UCONFIG_NO_SERVICE
|
jpayne@69
|
75 /**
|
jpayne@69
|
76 * @stable ICU 2.6
|
jpayne@69
|
77 */
|
jpayne@69
|
78 class CollatorFactory;
|
jpayne@69
|
79 #endif
|
jpayne@69
|
80
|
jpayne@69
|
81 /**
|
jpayne@69
|
82 * @stable ICU 2.0
|
jpayne@69
|
83 */
|
jpayne@69
|
84 class CollationKey;
|
jpayne@69
|
85
|
jpayne@69
|
86 /**
|
jpayne@69
|
87 * The <code>Collator</code> class performs locale-sensitive string
|
jpayne@69
|
88 * comparison.<br>
|
jpayne@69
|
89 * You use this class to build searching and sorting routines for natural
|
jpayne@69
|
90 * language text.
|
jpayne@69
|
91 * <p>
|
jpayne@69
|
92 * <code>Collator</code> is an abstract base class. Subclasses implement
|
jpayne@69
|
93 * specific collation strategies. One subclass,
|
jpayne@69
|
94 * <code>RuleBasedCollator</code>, is currently provided and is applicable
|
jpayne@69
|
95 * to a wide set of languages. Other subclasses may be created to handle more
|
jpayne@69
|
96 * specialized needs.
|
jpayne@69
|
97 * <p>
|
jpayne@69
|
98 * Like other locale-sensitive classes, you can use the static factory method,
|
jpayne@69
|
99 * <code>createInstance</code>, to obtain the appropriate
|
jpayne@69
|
100 * <code>Collator</code> object for a given locale. You will only need to
|
jpayne@69
|
101 * look at the subclasses of <code>Collator</code> if you need to
|
jpayne@69
|
102 * understand the details of a particular collation strategy or if you need to
|
jpayne@69
|
103 * modify that strategy.
|
jpayne@69
|
104 * <p>
|
jpayne@69
|
105 * The following example shows how to compare two strings using the
|
jpayne@69
|
106 * <code>Collator</code> for the default locale.
|
jpayne@69
|
107 * \htmlonly<blockquote>\endhtmlonly
|
jpayne@69
|
108 * <pre>
|
jpayne@69
|
109 * \code
|
jpayne@69
|
110 * // Compare two strings in the default locale
|
jpayne@69
|
111 * UErrorCode success = U_ZERO_ERROR;
|
jpayne@69
|
112 * Collator* myCollator = Collator::createInstance(success);
|
jpayne@69
|
113 * if (myCollator->compare("abc", "ABC") < 0)
|
jpayne@69
|
114 * cout << "abc is less than ABC" << endl;
|
jpayne@69
|
115 * else
|
jpayne@69
|
116 * cout << "abc is greater than or equal to ABC" << endl;
|
jpayne@69
|
117 * \endcode
|
jpayne@69
|
118 * </pre>
|
jpayne@69
|
119 * \htmlonly</blockquote>\endhtmlonly
|
jpayne@69
|
120 * <p>
|
jpayne@69
|
121 * You can set a <code>Collator</code>'s <em>strength</em> attribute to
|
jpayne@69
|
122 * determine the level of difference considered significant in comparisons.
|
jpayne@69
|
123 * Five strengths are provided: <code>PRIMARY</code>, <code>SECONDARY</code>,
|
jpayne@69
|
124 * <code>TERTIARY</code>, <code>QUATERNARY</code> and <code>IDENTICAL</code>.
|
jpayne@69
|
125 * The exact assignment of strengths to language features is locale dependent.
|
jpayne@69
|
126 * For example, in Czech, "e" and "f" are considered primary differences,
|
jpayne@69
|
127 * while "e" and "\u00EA" are secondary differences, "e" and "E" are tertiary
|
jpayne@69
|
128 * differences and "e" and "e" are identical. The following shows how both case
|
jpayne@69
|
129 * and accents could be ignored for US English.
|
jpayne@69
|
130 * \htmlonly<blockquote>\endhtmlonly
|
jpayne@69
|
131 * <pre>
|
jpayne@69
|
132 * \code
|
jpayne@69
|
133 * //Get the Collator for US English and set its strength to PRIMARY
|
jpayne@69
|
134 * UErrorCode success = U_ZERO_ERROR;
|
jpayne@69
|
135 * Collator* usCollator = Collator::createInstance(Locale::getUS(), success);
|
jpayne@69
|
136 * usCollator->setStrength(Collator::PRIMARY);
|
jpayne@69
|
137 * if (usCollator->compare("abc", "ABC") == 0)
|
jpayne@69
|
138 * cout << "'abc' and 'ABC' strings are equivalent with strength PRIMARY" << endl;
|
jpayne@69
|
139 * \endcode
|
jpayne@69
|
140 * </pre>
|
jpayne@69
|
141 * \htmlonly</blockquote>\endhtmlonly
|
jpayne@69
|
142 *
|
jpayne@69
|
143 * The <code>getSortKey</code> methods
|
jpayne@69
|
144 * convert a string to a series of bytes that can be compared bitwise against
|
jpayne@69
|
145 * other sort keys using <code>strcmp()</code>. Sort keys are written as
|
jpayne@69
|
146 * zero-terminated byte strings.
|
jpayne@69
|
147 *
|
jpayne@69
|
148 * Another set of APIs returns a <code>CollationKey</code> object that wraps
|
jpayne@69
|
149 * the sort key bytes instead of returning the bytes themselves.
|
jpayne@69
|
150 * </p>
|
jpayne@69
|
151 * <p>
|
jpayne@69
|
152 * <strong>Note:</strong> <code>Collator</code>s with different Locale,
|
jpayne@69
|
153 * and CollationStrength settings will return different sort
|
jpayne@69
|
154 * orders for the same set of strings. Locales have specific collation rules,
|
jpayne@69
|
155 * and the way in which secondary and tertiary differences are taken into
|
jpayne@69
|
156 * account, for example, will result in a different sorting order for same
|
jpayne@69
|
157 * strings.
|
jpayne@69
|
158 * </p>
|
jpayne@69
|
159 * @see RuleBasedCollator
|
jpayne@69
|
160 * @see CollationKey
|
jpayne@69
|
161 * @see CollationElementIterator
|
jpayne@69
|
162 * @see Locale
|
jpayne@69
|
163 * @see Normalizer2
|
jpayne@69
|
164 * @version 2.0 11/15/01
|
jpayne@69
|
165 */
|
jpayne@69
|
166
|
jpayne@69
|
167 class U_I18N_API Collator : public UObject {
|
jpayne@69
|
168 public:
|
jpayne@69
|
169
|
jpayne@69
|
170 // Collator public enums -----------------------------------------------
|
jpayne@69
|
171
|
jpayne@69
|
172 /**
|
jpayne@69
|
173 * Base letter represents a primary difference. Set comparison level to
|
jpayne@69
|
174 * PRIMARY to ignore secondary and tertiary differences.<br>
|
jpayne@69
|
175 * Use this to set the strength of a Collator object.<br>
|
jpayne@69
|
176 * Example of primary difference, "abc" < "abd"
|
jpayne@69
|
177 *
|
jpayne@69
|
178 * Diacritical differences on the same base letter represent a secondary
|
jpayne@69
|
179 * difference. Set comparison level to SECONDARY to ignore tertiary
|
jpayne@69
|
180 * differences. Use this to set the strength of a Collator object.<br>
|
jpayne@69
|
181 * Example of secondary difference, "ä" >> "a".
|
jpayne@69
|
182 *
|
jpayne@69
|
183 * Uppercase and lowercase versions of the same character represents a
|
jpayne@69
|
184 * tertiary difference. Set comparison level to TERTIARY to include all
|
jpayne@69
|
185 * comparison differences. Use this to set the strength of a Collator
|
jpayne@69
|
186 * object.<br>
|
jpayne@69
|
187 * Example of tertiary difference, "abc" <<< "ABC".
|
jpayne@69
|
188 *
|
jpayne@69
|
189 * Two characters are considered "identical" when they have the same unicode
|
jpayne@69
|
190 * spellings.<br>
|
jpayne@69
|
191 * For example, "ä" == "ä".
|
jpayne@69
|
192 *
|
jpayne@69
|
193 * UCollationStrength is also used to determine the strength of sort keys
|
jpayne@69
|
194 * generated from Collator objects.
|
jpayne@69
|
195 * @stable ICU 2.0
|
jpayne@69
|
196 */
|
jpayne@69
|
197 enum ECollationStrength
|
jpayne@69
|
198 {
|
jpayne@69
|
199 PRIMARY = UCOL_PRIMARY, // 0
|
jpayne@69
|
200 SECONDARY = UCOL_SECONDARY, // 1
|
jpayne@69
|
201 TERTIARY = UCOL_TERTIARY, // 2
|
jpayne@69
|
202 QUATERNARY = UCOL_QUATERNARY, // 3
|
jpayne@69
|
203 IDENTICAL = UCOL_IDENTICAL // 15
|
jpayne@69
|
204 };
|
jpayne@69
|
205
|
jpayne@69
|
206
|
jpayne@69
|
207 // Cannot use #ifndef U_HIDE_DEPRECATED_API for the following, it is
|
jpayne@69
|
208 // used by virtual methods that cannot have that conditional.
|
jpayne@69
|
209 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
210 /**
|
jpayne@69
|
211 * LESS is returned if source string is compared to be less than target
|
jpayne@69
|
212 * string in the compare() method.
|
jpayne@69
|
213 * EQUAL is returned if source string is compared to be equal to target
|
jpayne@69
|
214 * string in the compare() method.
|
jpayne@69
|
215 * GREATER is returned if source string is compared to be greater than
|
jpayne@69
|
216 * target string in the compare() method.
|
jpayne@69
|
217 * @see Collator#compare
|
jpayne@69
|
218 * @deprecated ICU 2.6. Use C enum UCollationResult defined in ucol.h
|
jpayne@69
|
219 */
|
jpayne@69
|
220 enum EComparisonResult
|
jpayne@69
|
221 {
|
jpayne@69
|
222 LESS = UCOL_LESS, // -1
|
jpayne@69
|
223 EQUAL = UCOL_EQUAL, // 0
|
jpayne@69
|
224 GREATER = UCOL_GREATER // 1
|
jpayne@69
|
225 };
|
jpayne@69
|
226 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
227
|
jpayne@69
|
228 // Collator public destructor -----------------------------------------
|
jpayne@69
|
229
|
jpayne@69
|
230 /**
|
jpayne@69
|
231 * Destructor
|
jpayne@69
|
232 * @stable ICU 2.0
|
jpayne@69
|
233 */
|
jpayne@69
|
234 virtual ~Collator();
|
jpayne@69
|
235
|
jpayne@69
|
236 // Collator public methods --------------------------------------------
|
jpayne@69
|
237
|
jpayne@69
|
238 /**
|
jpayne@69
|
239 * Returns TRUE if "other" is the same as "this".
|
jpayne@69
|
240 *
|
jpayne@69
|
241 * The base class implementation returns TRUE if "other" has the same type/class as "this":
|
jpayne@69
|
242 * `typeid(*this) == typeid(other)`.
|
jpayne@69
|
243 *
|
jpayne@69
|
244 * Subclass implementations should do something like the following:
|
jpayne@69
|
245 *
|
jpayne@69
|
246 * if (this == &other) { return TRUE; }
|
jpayne@69
|
247 * if (!Collator::operator==(other)) { return FALSE; } // not the same class
|
jpayne@69
|
248 *
|
jpayne@69
|
249 * const MyCollator &o = (const MyCollator&)other;
|
jpayne@69
|
250 * (compare this vs. o's subclass fields)
|
jpayne@69
|
251 *
|
jpayne@69
|
252 * @param other Collator object to be compared
|
jpayne@69
|
253 * @return TRUE if other is the same as this.
|
jpayne@69
|
254 * @stable ICU 2.0
|
jpayne@69
|
255 */
|
jpayne@69
|
256 virtual UBool operator==(const Collator& other) const;
|
jpayne@69
|
257
|
jpayne@69
|
258 /**
|
jpayne@69
|
259 * Returns true if "other" is not the same as "this".
|
jpayne@69
|
260 * Calls ! operator==(const Collator&) const which works for all subclasses.
|
jpayne@69
|
261 * @param other Collator object to be compared
|
jpayne@69
|
262 * @return TRUE if other is not the same as this.
|
jpayne@69
|
263 * @stable ICU 2.0
|
jpayne@69
|
264 */
|
jpayne@69
|
265 virtual UBool operator!=(const Collator& other) const;
|
jpayne@69
|
266
|
jpayne@69
|
267 /**
|
jpayne@69
|
268 * Makes a copy of this object.
|
jpayne@69
|
269 * @return a copy of this object, owned by the caller
|
jpayne@69
|
270 * @stable ICU 2.0
|
jpayne@69
|
271 */
|
jpayne@69
|
272 virtual Collator* clone() const = 0;
|
jpayne@69
|
273
|
jpayne@69
|
274 /**
|
jpayne@69
|
275 * Creates the Collator object for the current default locale.
|
jpayne@69
|
276 * The default locale is determined by Locale::getDefault.
|
jpayne@69
|
277 * The UErrorCode& err parameter is used to return status information to the user.
|
jpayne@69
|
278 * To check whether the construction succeeded or not, you should check the
|
jpayne@69
|
279 * value of U_SUCCESS(err). If you wish more detailed information, you can
|
jpayne@69
|
280 * check for informational error results which still indicate success.
|
jpayne@69
|
281 * U_USING_FALLBACK_ERROR indicates that a fall back locale was used. For
|
jpayne@69
|
282 * example, 'de_CH' was requested, but nothing was found there, so 'de' was
|
jpayne@69
|
283 * used. U_USING_DEFAULT_ERROR indicates that the default locale data was
|
jpayne@69
|
284 * used; neither the requested locale nor any of its fall back locales
|
jpayne@69
|
285 * could be found.
|
jpayne@69
|
286 * The caller owns the returned object and is responsible for deleting it.
|
jpayne@69
|
287 *
|
jpayne@69
|
288 * @param err the error code status.
|
jpayne@69
|
289 * @return the collation object of the default locale.(for example, en_US)
|
jpayne@69
|
290 * @see Locale#getDefault
|
jpayne@69
|
291 * @stable ICU 2.0
|
jpayne@69
|
292 */
|
jpayne@69
|
293 static Collator* U_EXPORT2 createInstance(UErrorCode& err);
|
jpayne@69
|
294
|
jpayne@69
|
295 /**
|
jpayne@69
|
296 * Gets the collation object for the desired locale. The
|
jpayne@69
|
297 * resource of the desired locale will be loaded.
|
jpayne@69
|
298 *
|
jpayne@69
|
299 * Locale::getRoot() is the base collation table and all other languages are
|
jpayne@69
|
300 * built on top of it with additional language-specific modifications.
|
jpayne@69
|
301 *
|
jpayne@69
|
302 * For some languages, multiple collation types are available;
|
jpayne@69
|
303 * for example, "de@collation=phonebook".
|
jpayne@69
|
304 * Starting with ICU 54, collation attributes can be specified via locale keywords as well,
|
jpayne@69
|
305 * in the old locale extension syntax ("el@colCaseFirst=upper")
|
jpayne@69
|
306 * or in language tag syntax ("el-u-kf-upper").
|
jpayne@69
|
307 * See <a href="http://userguide.icu-project.org/collation/api">User Guide: Collation API</a>.
|
jpayne@69
|
308 *
|
jpayne@69
|
309 * The UErrorCode& err parameter is used to return status information to the user.
|
jpayne@69
|
310 * To check whether the construction succeeded or not, you should check
|
jpayne@69
|
311 * the value of U_SUCCESS(err). If you wish more detailed information, you
|
jpayne@69
|
312 * can check for informational error results which still indicate success.
|
jpayne@69
|
313 * U_USING_FALLBACK_ERROR indicates that a fall back locale was used. For
|
jpayne@69
|
314 * example, 'de_CH' was requested, but nothing was found there, so 'de' was
|
jpayne@69
|
315 * used. U_USING_DEFAULT_ERROR indicates that the default locale data was
|
jpayne@69
|
316 * used; neither the requested locale nor any of its fall back locales
|
jpayne@69
|
317 * could be found.
|
jpayne@69
|
318 *
|
jpayne@69
|
319 * The caller owns the returned object and is responsible for deleting it.
|
jpayne@69
|
320 * @param loc The locale ID for which to open a collator.
|
jpayne@69
|
321 * @param err the error code status.
|
jpayne@69
|
322 * @return the created table-based collation object based on the desired
|
jpayne@69
|
323 * locale.
|
jpayne@69
|
324 * @see Locale
|
jpayne@69
|
325 * @see ResourceLoader
|
jpayne@69
|
326 * @stable ICU 2.2
|
jpayne@69
|
327 */
|
jpayne@69
|
328 static Collator* U_EXPORT2 createInstance(const Locale& loc, UErrorCode& err);
|
jpayne@69
|
329
|
jpayne@69
|
330 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
331 /**
|
jpayne@69
|
332 * The comparison function compares the character data stored in two
|
jpayne@69
|
333 * different strings. Returns information about whether a string is less
|
jpayne@69
|
334 * than, greater than or equal to another string.
|
jpayne@69
|
335 * @param source the source string to be compared with.
|
jpayne@69
|
336 * @param target the string that is to be compared with the source string.
|
jpayne@69
|
337 * @return Returns a byte value. GREATER if source is greater
|
jpayne@69
|
338 * than target; EQUAL if source is equal to target; LESS if source is less
|
jpayne@69
|
339 * than target
|
jpayne@69
|
340 * @deprecated ICU 2.6 use the overload with UErrorCode &
|
jpayne@69
|
341 */
|
jpayne@69
|
342 virtual EComparisonResult compare(const UnicodeString& source,
|
jpayne@69
|
343 const UnicodeString& target) const;
|
jpayne@69
|
344 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
345
|
jpayne@69
|
346 /**
|
jpayne@69
|
347 * The comparison function compares the character data stored in two
|
jpayne@69
|
348 * different strings. Returns information about whether a string is less
|
jpayne@69
|
349 * than, greater than or equal to another string.
|
jpayne@69
|
350 * @param source the source string to be compared with.
|
jpayne@69
|
351 * @param target the string that is to be compared with the source string.
|
jpayne@69
|
352 * @param status possible error code
|
jpayne@69
|
353 * @return Returns an enum value. UCOL_GREATER if source is greater
|
jpayne@69
|
354 * than target; UCOL_EQUAL if source is equal to target; UCOL_LESS if source is less
|
jpayne@69
|
355 * than target
|
jpayne@69
|
356 * @stable ICU 2.6
|
jpayne@69
|
357 */
|
jpayne@69
|
358 virtual UCollationResult compare(const UnicodeString& source,
|
jpayne@69
|
359 const UnicodeString& target,
|
jpayne@69
|
360 UErrorCode &status) const = 0;
|
jpayne@69
|
361
|
jpayne@69
|
362 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
363 /**
|
jpayne@69
|
364 * Does the same thing as compare but limits the comparison to a specified
|
jpayne@69
|
365 * length
|
jpayne@69
|
366 * @param source the source string to be compared with.
|
jpayne@69
|
367 * @param target the string that is to be compared with the source string.
|
jpayne@69
|
368 * @param length the length the comparison is limited to
|
jpayne@69
|
369 * @return Returns a byte value. GREATER if source (up to the specified
|
jpayne@69
|
370 * length) is greater than target; EQUAL if source (up to specified
|
jpayne@69
|
371 * length) is equal to target; LESS if source (up to the specified
|
jpayne@69
|
372 * length) is less than target.
|
jpayne@69
|
373 * @deprecated ICU 2.6 use the overload with UErrorCode &
|
jpayne@69
|
374 */
|
jpayne@69
|
375 virtual EComparisonResult compare(const UnicodeString& source,
|
jpayne@69
|
376 const UnicodeString& target,
|
jpayne@69
|
377 int32_t length) const;
|
jpayne@69
|
378 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
379
|
jpayne@69
|
380 /**
|
jpayne@69
|
381 * Does the same thing as compare but limits the comparison to a specified
|
jpayne@69
|
382 * length
|
jpayne@69
|
383 * @param source the source string to be compared with.
|
jpayne@69
|
384 * @param target the string that is to be compared with the source string.
|
jpayne@69
|
385 * @param length the length the comparison is limited to
|
jpayne@69
|
386 * @param status possible error code
|
jpayne@69
|
387 * @return Returns an enum value. UCOL_GREATER if source (up to the specified
|
jpayne@69
|
388 * length) is greater than target; UCOL_EQUAL if source (up to specified
|
jpayne@69
|
389 * length) is equal to target; UCOL_LESS if source (up to the specified
|
jpayne@69
|
390 * length) is less than target.
|
jpayne@69
|
391 * @stable ICU 2.6
|
jpayne@69
|
392 */
|
jpayne@69
|
393 virtual UCollationResult compare(const UnicodeString& source,
|
jpayne@69
|
394 const UnicodeString& target,
|
jpayne@69
|
395 int32_t length,
|
jpayne@69
|
396 UErrorCode &status) const = 0;
|
jpayne@69
|
397
|
jpayne@69
|
398 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
399 /**
|
jpayne@69
|
400 * The comparison function compares the character data stored in two
|
jpayne@69
|
401 * different string arrays. Returns information about whether a string array
|
jpayne@69
|
402 * is less than, greater than or equal to another string array.
|
jpayne@69
|
403 * <p>Example of use:
|
jpayne@69
|
404 * <pre>
|
jpayne@69
|
405 * . char16_t ABC[] = {0x41, 0x42, 0x43, 0}; // = "ABC"
|
jpayne@69
|
406 * . char16_t abc[] = {0x61, 0x62, 0x63, 0}; // = "abc"
|
jpayne@69
|
407 * . UErrorCode status = U_ZERO_ERROR;
|
jpayne@69
|
408 * . Collator *myCollation =
|
jpayne@69
|
409 * . Collator::createInstance(Locale::getUS(), status);
|
jpayne@69
|
410 * . if (U_FAILURE(status)) return;
|
jpayne@69
|
411 * . myCollation->setStrength(Collator::PRIMARY);
|
jpayne@69
|
412 * . // result would be Collator::EQUAL ("abc" == "ABC")
|
jpayne@69
|
413 * . // (no primary difference between "abc" and "ABC")
|
jpayne@69
|
414 * . Collator::EComparisonResult result =
|
jpayne@69
|
415 * . myCollation->compare(abc, 3, ABC, 3);
|
jpayne@69
|
416 * . myCollation->setStrength(Collator::TERTIARY);
|
jpayne@69
|
417 * . // result would be Collator::LESS ("abc" <<< "ABC")
|
jpayne@69
|
418 * . // (with tertiary difference between "abc" and "ABC")
|
jpayne@69
|
419 * . result = myCollation->compare(abc, 3, ABC, 3);
|
jpayne@69
|
420 * </pre>
|
jpayne@69
|
421 * @param source the source string array to be compared with.
|
jpayne@69
|
422 * @param sourceLength the length of the source string array. If this value
|
jpayne@69
|
423 * is equal to -1, the string array is null-terminated.
|
jpayne@69
|
424 * @param target the string that is to be compared with the source string.
|
jpayne@69
|
425 * @param targetLength the length of the target string array. If this value
|
jpayne@69
|
426 * is equal to -1, the string array is null-terminated.
|
jpayne@69
|
427 * @return Returns a byte value. GREATER if source is greater than target;
|
jpayne@69
|
428 * EQUAL if source is equal to target; LESS if source is less than
|
jpayne@69
|
429 * target
|
jpayne@69
|
430 * @deprecated ICU 2.6 use the overload with UErrorCode &
|
jpayne@69
|
431 */
|
jpayne@69
|
432 virtual EComparisonResult compare(const char16_t* source, int32_t sourceLength,
|
jpayne@69
|
433 const char16_t* target, int32_t targetLength)
|
jpayne@69
|
434 const;
|
jpayne@69
|
435 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
436
|
jpayne@69
|
437 /**
|
jpayne@69
|
438 * The comparison function compares the character data stored in two
|
jpayne@69
|
439 * different string arrays. Returns information about whether a string array
|
jpayne@69
|
440 * is less than, greater than or equal to another string array.
|
jpayne@69
|
441 * @param source the source string array to be compared with.
|
jpayne@69
|
442 * @param sourceLength the length of the source string array. If this value
|
jpayne@69
|
443 * is equal to -1, the string array is null-terminated.
|
jpayne@69
|
444 * @param target the string that is to be compared with the source string.
|
jpayne@69
|
445 * @param targetLength the length of the target string array. If this value
|
jpayne@69
|
446 * is equal to -1, the string array is null-terminated.
|
jpayne@69
|
447 * @param status possible error code
|
jpayne@69
|
448 * @return Returns an enum value. UCOL_GREATER if source is greater
|
jpayne@69
|
449 * than target; UCOL_EQUAL if source is equal to target; UCOL_LESS if source is less
|
jpayne@69
|
450 * than target
|
jpayne@69
|
451 * @stable ICU 2.6
|
jpayne@69
|
452 */
|
jpayne@69
|
453 virtual UCollationResult compare(const char16_t* source, int32_t sourceLength,
|
jpayne@69
|
454 const char16_t* target, int32_t targetLength,
|
jpayne@69
|
455 UErrorCode &status) const = 0;
|
jpayne@69
|
456
|
jpayne@69
|
457 /**
|
jpayne@69
|
458 * Compares two strings using the Collator.
|
jpayne@69
|
459 * Returns whether the first one compares less than/equal to/greater than
|
jpayne@69
|
460 * the second one.
|
jpayne@69
|
461 * This version takes UCharIterator input.
|
jpayne@69
|
462 * @param sIter the first ("source") string iterator
|
jpayne@69
|
463 * @param tIter the second ("target") string iterator
|
jpayne@69
|
464 * @param status ICU status
|
jpayne@69
|
465 * @return UCOL_LESS, UCOL_EQUAL or UCOL_GREATER
|
jpayne@69
|
466 * @stable ICU 4.2
|
jpayne@69
|
467 */
|
jpayne@69
|
468 virtual UCollationResult compare(UCharIterator &sIter,
|
jpayne@69
|
469 UCharIterator &tIter,
|
jpayne@69
|
470 UErrorCode &status) const;
|
jpayne@69
|
471
|
jpayne@69
|
472 /**
|
jpayne@69
|
473 * Compares two UTF-8 strings using the Collator.
|
jpayne@69
|
474 * Returns whether the first one compares less than/equal to/greater than
|
jpayne@69
|
475 * the second one.
|
jpayne@69
|
476 * This version takes UTF-8 input.
|
jpayne@69
|
477 * Note that a StringPiece can be implicitly constructed
|
jpayne@69
|
478 * from a std::string or a NUL-terminated const char * string.
|
jpayne@69
|
479 * @param source the first UTF-8 string
|
jpayne@69
|
480 * @param target the second UTF-8 string
|
jpayne@69
|
481 * @param status ICU status
|
jpayne@69
|
482 * @return UCOL_LESS, UCOL_EQUAL or UCOL_GREATER
|
jpayne@69
|
483 * @stable ICU 4.2
|
jpayne@69
|
484 */
|
jpayne@69
|
485 virtual UCollationResult compareUTF8(const StringPiece &source,
|
jpayne@69
|
486 const StringPiece &target,
|
jpayne@69
|
487 UErrorCode &status) const;
|
jpayne@69
|
488
|
jpayne@69
|
489 /**
|
jpayne@69
|
490 * Transforms the string into a series of characters that can be compared
|
jpayne@69
|
491 * with CollationKey::compareTo. It is not possible to restore the original
|
jpayne@69
|
492 * string from the chars in the sort key.
|
jpayne@69
|
493 * <p>Use CollationKey::equals or CollationKey::compare to compare the
|
jpayne@69
|
494 * generated sort keys.
|
jpayne@69
|
495 * If the source string is null, a null collation key will be returned.
|
jpayne@69
|
496 *
|
jpayne@69
|
497 * Note that sort keys are often less efficient than simply doing comparison.
|
jpayne@69
|
498 * For more details, see the ICU User Guide.
|
jpayne@69
|
499 *
|
jpayne@69
|
500 * @param source the source string to be transformed into a sort key.
|
jpayne@69
|
501 * @param key the collation key to be filled in
|
jpayne@69
|
502 * @param status the error code status.
|
jpayne@69
|
503 * @return the collation key of the string based on the collation rules.
|
jpayne@69
|
504 * @see CollationKey#compare
|
jpayne@69
|
505 * @stable ICU 2.0
|
jpayne@69
|
506 */
|
jpayne@69
|
507 virtual CollationKey& getCollationKey(const UnicodeString& source,
|
jpayne@69
|
508 CollationKey& key,
|
jpayne@69
|
509 UErrorCode& status) const = 0;
|
jpayne@69
|
510
|
jpayne@69
|
511 /**
|
jpayne@69
|
512 * Transforms the string into a series of characters that can be compared
|
jpayne@69
|
513 * with CollationKey::compareTo. It is not possible to restore the original
|
jpayne@69
|
514 * string from the chars in the sort key.
|
jpayne@69
|
515 * <p>Use CollationKey::equals or CollationKey::compare to compare the
|
jpayne@69
|
516 * generated sort keys.
|
jpayne@69
|
517 * <p>If the source string is null, a null collation key will be returned.
|
jpayne@69
|
518 *
|
jpayne@69
|
519 * Note that sort keys are often less efficient than simply doing comparison.
|
jpayne@69
|
520 * For more details, see the ICU User Guide.
|
jpayne@69
|
521 *
|
jpayne@69
|
522 * @param source the source string to be transformed into a sort key.
|
jpayne@69
|
523 * @param sourceLength length of the collation key
|
jpayne@69
|
524 * @param key the collation key to be filled in
|
jpayne@69
|
525 * @param status the error code status.
|
jpayne@69
|
526 * @return the collation key of the string based on the collation rules.
|
jpayne@69
|
527 * @see CollationKey#compare
|
jpayne@69
|
528 * @stable ICU 2.0
|
jpayne@69
|
529 */
|
jpayne@69
|
530 virtual CollationKey& getCollationKey(const char16_t*source,
|
jpayne@69
|
531 int32_t sourceLength,
|
jpayne@69
|
532 CollationKey& key,
|
jpayne@69
|
533 UErrorCode& status) const = 0;
|
jpayne@69
|
534 /**
|
jpayne@69
|
535 * Generates the hash code for the collation object
|
jpayne@69
|
536 * @stable ICU 2.0
|
jpayne@69
|
537 */
|
jpayne@69
|
538 virtual int32_t hashCode(void) const = 0;
|
jpayne@69
|
539
|
jpayne@69
|
540 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
541 /**
|
jpayne@69
|
542 * Gets the locale of the Collator
|
jpayne@69
|
543 *
|
jpayne@69
|
544 * @param type can be either requested, valid or actual locale. For more
|
jpayne@69
|
545 * information see the definition of ULocDataLocaleType in
|
jpayne@69
|
546 * uloc.h
|
jpayne@69
|
547 * @param status the error code status.
|
jpayne@69
|
548 * @return locale where the collation data lives. If the collator
|
jpayne@69
|
549 * was instantiated from rules, locale is empty.
|
jpayne@69
|
550 * @deprecated ICU 2.8 This API is under consideration for revision
|
jpayne@69
|
551 * in ICU 3.0.
|
jpayne@69
|
552 */
|
jpayne@69
|
553 virtual Locale getLocale(ULocDataLocaleType type, UErrorCode& status) const = 0;
|
jpayne@69
|
554 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
555
|
jpayne@69
|
556 /**
|
jpayne@69
|
557 * Convenience method for comparing two strings based on the collation rules.
|
jpayne@69
|
558 * @param source the source string to be compared with.
|
jpayne@69
|
559 * @param target the target string to be compared with.
|
jpayne@69
|
560 * @return true if the first string is greater than the second one,
|
jpayne@69
|
561 * according to the collation rules. false, otherwise.
|
jpayne@69
|
562 * @see Collator#compare
|
jpayne@69
|
563 * @stable ICU 2.0
|
jpayne@69
|
564 */
|
jpayne@69
|
565 UBool greater(const UnicodeString& source, const UnicodeString& target)
|
jpayne@69
|
566 const;
|
jpayne@69
|
567
|
jpayne@69
|
568 /**
|
jpayne@69
|
569 * Convenience method for comparing two strings based on the collation rules.
|
jpayne@69
|
570 * @param source the source string to be compared with.
|
jpayne@69
|
571 * @param target the target string to be compared with.
|
jpayne@69
|
572 * @return true if the first string is greater than or equal to the second
|
jpayne@69
|
573 * one, according to the collation rules. false, otherwise.
|
jpayne@69
|
574 * @see Collator#compare
|
jpayne@69
|
575 * @stable ICU 2.0
|
jpayne@69
|
576 */
|
jpayne@69
|
577 UBool greaterOrEqual(const UnicodeString& source,
|
jpayne@69
|
578 const UnicodeString& target) const;
|
jpayne@69
|
579
|
jpayne@69
|
580 /**
|
jpayne@69
|
581 * Convenience method for comparing two strings based on the collation rules.
|
jpayne@69
|
582 * @param source the source string to be compared with.
|
jpayne@69
|
583 * @param target the target string to be compared with.
|
jpayne@69
|
584 * @return true if the strings are equal according to the collation rules.
|
jpayne@69
|
585 * false, otherwise.
|
jpayne@69
|
586 * @see Collator#compare
|
jpayne@69
|
587 * @stable ICU 2.0
|
jpayne@69
|
588 */
|
jpayne@69
|
589 UBool equals(const UnicodeString& source, const UnicodeString& target) const;
|
jpayne@69
|
590
|
jpayne@69
|
591 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
592 /**
|
jpayne@69
|
593 * Determines the minimum strength that will be used in comparison or
|
jpayne@69
|
594 * transformation.
|
jpayne@69
|
595 * <p>E.g. with strength == SECONDARY, the tertiary difference is ignored
|
jpayne@69
|
596 * <p>E.g. with strength == PRIMARY, the secondary and tertiary difference
|
jpayne@69
|
597 * are ignored.
|
jpayne@69
|
598 * @return the current comparison level.
|
jpayne@69
|
599 * @see Collator#setStrength
|
jpayne@69
|
600 * @deprecated ICU 2.6 Use getAttribute(UCOL_STRENGTH...) instead
|
jpayne@69
|
601 */
|
jpayne@69
|
602 virtual ECollationStrength getStrength(void) const;
|
jpayne@69
|
603
|
jpayne@69
|
604 /**
|
jpayne@69
|
605 * Sets the minimum strength to be used in comparison or transformation.
|
jpayne@69
|
606 * <p>Example of use:
|
jpayne@69
|
607 * <pre>
|
jpayne@69
|
608 * \code
|
jpayne@69
|
609 * UErrorCode status = U_ZERO_ERROR;
|
jpayne@69
|
610 * Collator*myCollation = Collator::createInstance(Locale::getUS(), status);
|
jpayne@69
|
611 * if (U_FAILURE(status)) return;
|
jpayne@69
|
612 * myCollation->setStrength(Collator::PRIMARY);
|
jpayne@69
|
613 * // result will be "abc" == "ABC"
|
jpayne@69
|
614 * // tertiary differences will be ignored
|
jpayne@69
|
615 * Collator::ComparisonResult result = myCollation->compare("abc", "ABC");
|
jpayne@69
|
616 * \endcode
|
jpayne@69
|
617 * </pre>
|
jpayne@69
|
618 * @see Collator#getStrength
|
jpayne@69
|
619 * @param newStrength the new comparison level.
|
jpayne@69
|
620 * @deprecated ICU 2.6 Use setAttribute(UCOL_STRENGTH...) instead
|
jpayne@69
|
621 */
|
jpayne@69
|
622 virtual void setStrength(ECollationStrength newStrength);
|
jpayne@69
|
623 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
624
|
jpayne@69
|
625 /**
|
jpayne@69
|
626 * Retrieves the reordering codes for this collator.
|
jpayne@69
|
627 * @param dest The array to fill with the script ordering.
|
jpayne@69
|
628 * @param destCapacity The length of dest. If it is 0, then dest may be NULL and the function
|
jpayne@69
|
629 * will only return the length of the result without writing any codes (pre-flighting).
|
jpayne@69
|
630 * @param status A reference to an error code value, which must not indicate
|
jpayne@69
|
631 * a failure before the function call.
|
jpayne@69
|
632 * @return The length of the script ordering array.
|
jpayne@69
|
633 * @see ucol_setReorderCodes
|
jpayne@69
|
634 * @see Collator#getEquivalentReorderCodes
|
jpayne@69
|
635 * @see Collator#setReorderCodes
|
jpayne@69
|
636 * @see UScriptCode
|
jpayne@69
|
637 * @see UColReorderCode
|
jpayne@69
|
638 * @stable ICU 4.8
|
jpayne@69
|
639 */
|
jpayne@69
|
640 virtual int32_t getReorderCodes(int32_t *dest,
|
jpayne@69
|
641 int32_t destCapacity,
|
jpayne@69
|
642 UErrorCode& status) const;
|
jpayne@69
|
643
|
jpayne@69
|
644 /**
|
jpayne@69
|
645 * Sets the ordering of scripts for this collator.
|
jpayne@69
|
646 *
|
jpayne@69
|
647 * <p>The reordering codes are a combination of script codes and reorder codes.
|
jpayne@69
|
648 * @param reorderCodes An array of script codes in the new order. This can be NULL if the
|
jpayne@69
|
649 * length is also set to 0. An empty array will clear any reordering codes on the collator.
|
jpayne@69
|
650 * @param reorderCodesLength The length of reorderCodes.
|
jpayne@69
|
651 * @param status error code
|
jpayne@69
|
652 * @see ucol_setReorderCodes
|
jpayne@69
|
653 * @see Collator#getReorderCodes
|
jpayne@69
|
654 * @see Collator#getEquivalentReorderCodes
|
jpayne@69
|
655 * @see UScriptCode
|
jpayne@69
|
656 * @see UColReorderCode
|
jpayne@69
|
657 * @stable ICU 4.8
|
jpayne@69
|
658 */
|
jpayne@69
|
659 virtual void setReorderCodes(const int32_t* reorderCodes,
|
jpayne@69
|
660 int32_t reorderCodesLength,
|
jpayne@69
|
661 UErrorCode& status) ;
|
jpayne@69
|
662
|
jpayne@69
|
663 /**
|
jpayne@69
|
664 * Retrieves the reorder codes that are grouped with the given reorder code. Some reorder
|
jpayne@69
|
665 * codes will be grouped and must reorder together.
|
jpayne@69
|
666 * Beginning with ICU 55, scripts only reorder together if they are primary-equal,
|
jpayne@69
|
667 * for example Hiragana and Katakana.
|
jpayne@69
|
668 *
|
jpayne@69
|
669 * @param reorderCode The reorder code to determine equivalence for.
|
jpayne@69
|
670 * @param dest The array to fill with the script equivalence reordering codes.
|
jpayne@69
|
671 * @param destCapacity The length of dest. If it is 0, then dest may be NULL and the
|
jpayne@69
|
672 * function will only return the length of the result without writing any codes (pre-flighting).
|
jpayne@69
|
673 * @param status A reference to an error code value, which must not indicate
|
jpayne@69
|
674 * a failure before the function call.
|
jpayne@69
|
675 * @return The length of the of the reordering code equivalence array.
|
jpayne@69
|
676 * @see ucol_setReorderCodes
|
jpayne@69
|
677 * @see Collator#getReorderCodes
|
jpayne@69
|
678 * @see Collator#setReorderCodes
|
jpayne@69
|
679 * @see UScriptCode
|
jpayne@69
|
680 * @see UColReorderCode
|
jpayne@69
|
681 * @stable ICU 4.8
|
jpayne@69
|
682 */
|
jpayne@69
|
683 static int32_t U_EXPORT2 getEquivalentReorderCodes(int32_t reorderCode,
|
jpayne@69
|
684 int32_t* dest,
|
jpayne@69
|
685 int32_t destCapacity,
|
jpayne@69
|
686 UErrorCode& status);
|
jpayne@69
|
687
|
jpayne@69
|
688 /**
|
jpayne@69
|
689 * Get name of the object for the desired Locale, in the desired language
|
jpayne@69
|
690 * @param objectLocale must be from getAvailableLocales
|
jpayne@69
|
691 * @param displayLocale specifies the desired locale for output
|
jpayne@69
|
692 * @param name the fill-in parameter of the return value
|
jpayne@69
|
693 * @return display-able name of the object for the object locale in the
|
jpayne@69
|
694 * desired language
|
jpayne@69
|
695 * @stable ICU 2.0
|
jpayne@69
|
696 */
|
jpayne@69
|
697 static UnicodeString& U_EXPORT2 getDisplayName(const Locale& objectLocale,
|
jpayne@69
|
698 const Locale& displayLocale,
|
jpayne@69
|
699 UnicodeString& name);
|
jpayne@69
|
700
|
jpayne@69
|
701 /**
|
jpayne@69
|
702 * Get name of the object for the desired Locale, in the language of the
|
jpayne@69
|
703 * default locale.
|
jpayne@69
|
704 * @param objectLocale must be from getAvailableLocales
|
jpayne@69
|
705 * @param name the fill-in parameter of the return value
|
jpayne@69
|
706 * @return name of the object for the desired locale in the default language
|
jpayne@69
|
707 * @stable ICU 2.0
|
jpayne@69
|
708 */
|
jpayne@69
|
709 static UnicodeString& U_EXPORT2 getDisplayName(const Locale& objectLocale,
|
jpayne@69
|
710 UnicodeString& name);
|
jpayne@69
|
711
|
jpayne@69
|
712 /**
|
jpayne@69
|
713 * Get the set of Locales for which Collations are installed.
|
jpayne@69
|
714 *
|
jpayne@69
|
715 * <p>Note this does not include locales supported by registered collators.
|
jpayne@69
|
716 * If collators might have been registered, use the overload of getAvailableLocales
|
jpayne@69
|
717 * that returns a StringEnumeration.</p>
|
jpayne@69
|
718 *
|
jpayne@69
|
719 * @param count the output parameter of number of elements in the locale list
|
jpayne@69
|
720 * @return the list of available locales for which collations are installed
|
jpayne@69
|
721 * @stable ICU 2.0
|
jpayne@69
|
722 */
|
jpayne@69
|
723 static const Locale* U_EXPORT2 getAvailableLocales(int32_t& count);
|
jpayne@69
|
724
|
jpayne@69
|
725 /**
|
jpayne@69
|
726 * Return a StringEnumeration over the locales available at the time of the call,
|
jpayne@69
|
727 * including registered locales. If a severe error occurs (such as out of memory
|
jpayne@69
|
728 * condition) this will return null. If there is no locale data, an empty enumeration
|
jpayne@69
|
729 * will be returned.
|
jpayne@69
|
730 * @return a StringEnumeration over the locales available at the time of the call
|
jpayne@69
|
731 * @stable ICU 2.6
|
jpayne@69
|
732 */
|
jpayne@69
|
733 static StringEnumeration* U_EXPORT2 getAvailableLocales(void);
|
jpayne@69
|
734
|
jpayne@69
|
735 /**
|
jpayne@69
|
736 * Create a string enumerator of all possible keywords that are relevant to
|
jpayne@69
|
737 * collation. At this point, the only recognized keyword for this
|
jpayne@69
|
738 * service is "collation".
|
jpayne@69
|
739 * @param status input-output error code
|
jpayne@69
|
740 * @return a string enumeration over locale strings. The caller is
|
jpayne@69
|
741 * responsible for closing the result.
|
jpayne@69
|
742 * @stable ICU 3.0
|
jpayne@69
|
743 */
|
jpayne@69
|
744 static StringEnumeration* U_EXPORT2 getKeywords(UErrorCode& status);
|
jpayne@69
|
745
|
jpayne@69
|
746 /**
|
jpayne@69
|
747 * Given a keyword, create a string enumeration of all values
|
jpayne@69
|
748 * for that keyword that are currently in use.
|
jpayne@69
|
749 * @param keyword a particular keyword as enumerated by
|
jpayne@69
|
750 * ucol_getKeywords. If any other keyword is passed in, status is set
|
jpayne@69
|
751 * to U_ILLEGAL_ARGUMENT_ERROR.
|
jpayne@69
|
752 * @param status input-output error code
|
jpayne@69
|
753 * @return a string enumeration over collation keyword values, or NULL
|
jpayne@69
|
754 * upon error. The caller is responsible for deleting the result.
|
jpayne@69
|
755 * @stable ICU 3.0
|
jpayne@69
|
756 */
|
jpayne@69
|
757 static StringEnumeration* U_EXPORT2 getKeywordValues(const char *keyword, UErrorCode& status);
|
jpayne@69
|
758
|
jpayne@69
|
759 /**
|
jpayne@69
|
760 * Given a key and a locale, returns an array of string values in a preferred
|
jpayne@69
|
761 * order that would make a difference. These are all and only those values where
|
jpayne@69
|
762 * the open (creation) of the service with the locale formed from the input locale
|
jpayne@69
|
763 * plus input keyword and that value has different behavior than creation with the
|
jpayne@69
|
764 * input locale alone.
|
jpayne@69
|
765 * @param keyword one of the keys supported by this service. For now, only
|
jpayne@69
|
766 * "collation" is supported.
|
jpayne@69
|
767 * @param locale the locale
|
jpayne@69
|
768 * @param commonlyUsed if set to true it will return only commonly used values
|
jpayne@69
|
769 * with the given locale in preferred order. Otherwise,
|
jpayne@69
|
770 * it will return all the available values for the locale.
|
jpayne@69
|
771 * @param status ICU status
|
jpayne@69
|
772 * @return a string enumeration over keyword values for the given key and the locale.
|
jpayne@69
|
773 * @stable ICU 4.2
|
jpayne@69
|
774 */
|
jpayne@69
|
775 static StringEnumeration* U_EXPORT2 getKeywordValuesForLocale(const char* keyword, const Locale& locale,
|
jpayne@69
|
776 UBool commonlyUsed, UErrorCode& status);
|
jpayne@69
|
777
|
jpayne@69
|
778 /**
|
jpayne@69
|
779 * Return the functionally equivalent locale for the given
|
jpayne@69
|
780 * requested locale, with respect to given keyword, for the
|
jpayne@69
|
781 * collation service. If two locales return the same result, then
|
jpayne@69
|
782 * collators instantiated for these locales will behave
|
jpayne@69
|
783 * equivalently. The converse is not always true; two collators
|
jpayne@69
|
784 * may in fact be equivalent, but return different results, due to
|
jpayne@69
|
785 * internal details. The return result has no other meaning than
|
jpayne@69
|
786 * that stated above, and implies nothing as to the relationship
|
jpayne@69
|
787 * between the two locales. This is intended for use by
|
jpayne@69
|
788 * applications who wish to cache collators, or otherwise reuse
|
jpayne@69
|
789 * collators when possible. The functional equivalent may change
|
jpayne@69
|
790 * over time. For more information, please see the <a
|
jpayne@69
|
791 * href="http://userguide.icu-project.org/locale#TOC-Locales-and-Services">
|
jpayne@69
|
792 * Locales and Services</a> section of the ICU User Guide.
|
jpayne@69
|
793 * @param keyword a particular keyword as enumerated by
|
jpayne@69
|
794 * ucol_getKeywords.
|
jpayne@69
|
795 * @param locale the requested locale
|
jpayne@69
|
796 * @param isAvailable reference to a fillin parameter that
|
jpayne@69
|
797 * indicates whether the requested locale was 'available' to the
|
jpayne@69
|
798 * collation service. A locale is defined as 'available' if it
|
jpayne@69
|
799 * physically exists within the collation locale data.
|
jpayne@69
|
800 * @param status reference to input-output error code
|
jpayne@69
|
801 * @return the functionally equivalent collation locale, or the root
|
jpayne@69
|
802 * locale upon error.
|
jpayne@69
|
803 * @stable ICU 3.0
|
jpayne@69
|
804 */
|
jpayne@69
|
805 static Locale U_EXPORT2 getFunctionalEquivalent(const char* keyword, const Locale& locale,
|
jpayne@69
|
806 UBool& isAvailable, UErrorCode& status);
|
jpayne@69
|
807
|
jpayne@69
|
808 #if !UCONFIG_NO_SERVICE
|
jpayne@69
|
809 /**
|
jpayne@69
|
810 * Register a new Collator. The collator will be adopted.
|
jpayne@69
|
811 * Because ICU may choose to cache collators internally, this must be
|
jpayne@69
|
812 * called at application startup, prior to any calls to
|
jpayne@69
|
813 * Collator::createInstance to avoid undefined behavior.
|
jpayne@69
|
814 * @param toAdopt the Collator instance to be adopted
|
jpayne@69
|
815 * @param locale the locale with which the collator will be associated
|
jpayne@69
|
816 * @param status the in/out status code, no special meanings are assigned
|
jpayne@69
|
817 * @return a registry key that can be used to unregister this collator
|
jpayne@69
|
818 * @stable ICU 2.6
|
jpayne@69
|
819 */
|
jpayne@69
|
820 static URegistryKey U_EXPORT2 registerInstance(Collator* toAdopt, const Locale& locale, UErrorCode& status);
|
jpayne@69
|
821
|
jpayne@69
|
822 /**
|
jpayne@69
|
823 * Register a new CollatorFactory. The factory will be adopted.
|
jpayne@69
|
824 * Because ICU may choose to cache collators internally, this must be
|
jpayne@69
|
825 * called at application startup, prior to any calls to
|
jpayne@69
|
826 * Collator::createInstance to avoid undefined behavior.
|
jpayne@69
|
827 * @param toAdopt the CollatorFactory instance to be adopted
|
jpayne@69
|
828 * @param status the in/out status code, no special meanings are assigned
|
jpayne@69
|
829 * @return a registry key that can be used to unregister this collator
|
jpayne@69
|
830 * @stable ICU 2.6
|
jpayne@69
|
831 */
|
jpayne@69
|
832 static URegistryKey U_EXPORT2 registerFactory(CollatorFactory* toAdopt, UErrorCode& status);
|
jpayne@69
|
833
|
jpayne@69
|
834 /**
|
jpayne@69
|
835 * Unregister a previously-registered Collator or CollatorFactory
|
jpayne@69
|
836 * using the key returned from the register call. Key becomes
|
jpayne@69
|
837 * invalid after a successful call and should not be used again.
|
jpayne@69
|
838 * The object corresponding to the key will be deleted.
|
jpayne@69
|
839 * Because ICU may choose to cache collators internally, this should
|
jpayne@69
|
840 * be called during application shutdown, after all calls to
|
jpayne@69
|
841 * Collator::createInstance to avoid undefined behavior.
|
jpayne@69
|
842 * @param key the registry key returned by a previous call to registerInstance
|
jpayne@69
|
843 * @param status the in/out status code, no special meanings are assigned
|
jpayne@69
|
844 * @return TRUE if the collator for the key was successfully unregistered
|
jpayne@69
|
845 * @stable ICU 2.6
|
jpayne@69
|
846 */
|
jpayne@69
|
847 static UBool U_EXPORT2 unregister(URegistryKey key, UErrorCode& status);
|
jpayne@69
|
848 #endif /* UCONFIG_NO_SERVICE */
|
jpayne@69
|
849
|
jpayne@69
|
850 /**
|
jpayne@69
|
851 * Gets the version information for a Collator.
|
jpayne@69
|
852 * @param info the version # information, the result will be filled in
|
jpayne@69
|
853 * @stable ICU 2.0
|
jpayne@69
|
854 */
|
jpayne@69
|
855 virtual void getVersion(UVersionInfo info) const = 0;
|
jpayne@69
|
856
|
jpayne@69
|
857 /**
|
jpayne@69
|
858 * Returns a unique class ID POLYMORPHICALLY. Pure virtual method.
|
jpayne@69
|
859 * This method is to implement a simple version of RTTI, since not all C++
|
jpayne@69
|
860 * compilers support genuine RTTI. Polymorphic operator==() and clone()
|
jpayne@69
|
861 * methods call this method.
|
jpayne@69
|
862 * @return The class ID for this object. All objects of a given class have
|
jpayne@69
|
863 * the same class ID. Objects of other classes have different class
|
jpayne@69
|
864 * IDs.
|
jpayne@69
|
865 * @stable ICU 2.0
|
jpayne@69
|
866 */
|
jpayne@69
|
867 virtual UClassID getDynamicClassID(void) const = 0;
|
jpayne@69
|
868
|
jpayne@69
|
869 /**
|
jpayne@69
|
870 * Universal attribute setter
|
jpayne@69
|
871 * @param attr attribute type
|
jpayne@69
|
872 * @param value attribute value
|
jpayne@69
|
873 * @param status to indicate whether the operation went on smoothly or
|
jpayne@69
|
874 * there were errors
|
jpayne@69
|
875 * @stable ICU 2.2
|
jpayne@69
|
876 */
|
jpayne@69
|
877 virtual void setAttribute(UColAttribute attr, UColAttributeValue value,
|
jpayne@69
|
878 UErrorCode &status) = 0;
|
jpayne@69
|
879
|
jpayne@69
|
880 /**
|
jpayne@69
|
881 * Universal attribute getter
|
jpayne@69
|
882 * @param attr attribute type
|
jpayne@69
|
883 * @param status to indicate whether the operation went on smoothly or
|
jpayne@69
|
884 * there were errors
|
jpayne@69
|
885 * @return attribute value
|
jpayne@69
|
886 * @stable ICU 2.2
|
jpayne@69
|
887 */
|
jpayne@69
|
888 virtual UColAttributeValue getAttribute(UColAttribute attr,
|
jpayne@69
|
889 UErrorCode &status) const = 0;
|
jpayne@69
|
890
|
jpayne@69
|
891 /**
|
jpayne@69
|
892 * Sets the variable top to the top of the specified reordering group.
|
jpayne@69
|
893 * The variable top determines the highest-sorting character
|
jpayne@69
|
894 * which is affected by UCOL_ALTERNATE_HANDLING.
|
jpayne@69
|
895 * If that attribute is set to UCOL_NON_IGNORABLE, then the variable top has no effect.
|
jpayne@69
|
896 *
|
jpayne@69
|
897 * The base class implementation sets U_UNSUPPORTED_ERROR.
|
jpayne@69
|
898 * @param group one of UCOL_REORDER_CODE_SPACE, UCOL_REORDER_CODE_PUNCTUATION,
|
jpayne@69
|
899 * UCOL_REORDER_CODE_SYMBOL, UCOL_REORDER_CODE_CURRENCY;
|
jpayne@69
|
900 * or UCOL_REORDER_CODE_DEFAULT to restore the default max variable group
|
jpayne@69
|
901 * @param errorCode Standard ICU error code. Its input value must
|
jpayne@69
|
902 * pass the U_SUCCESS() test, or else the function returns
|
jpayne@69
|
903 * immediately. Check for U_FAILURE() on output or use with
|
jpayne@69
|
904 * function chaining. (See User Guide for details.)
|
jpayne@69
|
905 * @return *this
|
jpayne@69
|
906 * @see getMaxVariable
|
jpayne@69
|
907 * @stable ICU 53
|
jpayne@69
|
908 */
|
jpayne@69
|
909 virtual Collator &setMaxVariable(UColReorderCode group, UErrorCode &errorCode);
|
jpayne@69
|
910
|
jpayne@69
|
911 /**
|
jpayne@69
|
912 * Returns the maximum reordering group whose characters are affected by UCOL_ALTERNATE_HANDLING.
|
jpayne@69
|
913 *
|
jpayne@69
|
914 * The base class implementation returns UCOL_REORDER_CODE_PUNCTUATION.
|
jpayne@69
|
915 * @return the maximum variable reordering group.
|
jpayne@69
|
916 * @see setMaxVariable
|
jpayne@69
|
917 * @stable ICU 53
|
jpayne@69
|
918 */
|
jpayne@69
|
919 virtual UColReorderCode getMaxVariable() const;
|
jpayne@69
|
920
|
jpayne@69
|
921 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
922 /**
|
jpayne@69
|
923 * Sets the variable top to the primary weight of the specified string.
|
jpayne@69
|
924 *
|
jpayne@69
|
925 * Beginning with ICU 53, the variable top is pinned to
|
jpayne@69
|
926 * the top of one of the supported reordering groups,
|
jpayne@69
|
927 * and it must not be beyond the last of those groups.
|
jpayne@69
|
928 * See setMaxVariable().
|
jpayne@69
|
929 * @param varTop one or more (if contraction) char16_ts to which the variable top should be set
|
jpayne@69
|
930 * @param len length of variable top string. If -1 it is considered to be zero terminated.
|
jpayne@69
|
931 * @param status error code. If error code is set, the return value is undefined. Errors set by this function are: <br>
|
jpayne@69
|
932 * U_CE_NOT_FOUND_ERROR if more than one character was passed and there is no such contraction<br>
|
jpayne@69
|
933 * U_ILLEGAL_ARGUMENT_ERROR if the variable top is beyond
|
jpayne@69
|
934 * the last reordering group supported by setMaxVariable()
|
jpayne@69
|
935 * @return variable top primary weight
|
jpayne@69
|
936 * @deprecated ICU 53 Call setMaxVariable() instead.
|
jpayne@69
|
937 */
|
jpayne@69
|
938 virtual uint32_t setVariableTop(const char16_t *varTop, int32_t len, UErrorCode &status) = 0;
|
jpayne@69
|
939
|
jpayne@69
|
940 /**
|
jpayne@69
|
941 * Sets the variable top to the primary weight of the specified string.
|
jpayne@69
|
942 *
|
jpayne@69
|
943 * Beginning with ICU 53, the variable top is pinned to
|
jpayne@69
|
944 * the top of one of the supported reordering groups,
|
jpayne@69
|
945 * and it must not be beyond the last of those groups.
|
jpayne@69
|
946 * See setMaxVariable().
|
jpayne@69
|
947 * @param varTop a UnicodeString size 1 or more (if contraction) of char16_ts to which the variable top should be set
|
jpayne@69
|
948 * @param status error code. If error code is set, the return value is undefined. Errors set by this function are: <br>
|
jpayne@69
|
949 * U_CE_NOT_FOUND_ERROR if more than one character was passed and there is no such contraction<br>
|
jpayne@69
|
950 * U_ILLEGAL_ARGUMENT_ERROR if the variable top is beyond
|
jpayne@69
|
951 * the last reordering group supported by setMaxVariable()
|
jpayne@69
|
952 * @return variable top primary weight
|
jpayne@69
|
953 * @deprecated ICU 53 Call setMaxVariable() instead.
|
jpayne@69
|
954 */
|
jpayne@69
|
955 virtual uint32_t setVariableTop(const UnicodeString &varTop, UErrorCode &status) = 0;
|
jpayne@69
|
956
|
jpayne@69
|
957 /**
|
jpayne@69
|
958 * Sets the variable top to the specified primary weight.
|
jpayne@69
|
959 *
|
jpayne@69
|
960 * Beginning with ICU 53, the variable top is pinned to
|
jpayne@69
|
961 * the top of one of the supported reordering groups,
|
jpayne@69
|
962 * and it must not be beyond the last of those groups.
|
jpayne@69
|
963 * See setMaxVariable().
|
jpayne@69
|
964 * @param varTop primary weight, as returned by setVariableTop or ucol_getVariableTop
|
jpayne@69
|
965 * @param status error code
|
jpayne@69
|
966 * @deprecated ICU 53 Call setMaxVariable() instead.
|
jpayne@69
|
967 */
|
jpayne@69
|
968 virtual void setVariableTop(uint32_t varTop, UErrorCode &status) = 0;
|
jpayne@69
|
969 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
970
|
jpayne@69
|
971 /**
|
jpayne@69
|
972 * Gets the variable top value of a Collator.
|
jpayne@69
|
973 * @param status error code (not changed by function). If error code is set, the return value is undefined.
|
jpayne@69
|
974 * @return the variable top primary weight
|
jpayne@69
|
975 * @see getMaxVariable
|
jpayne@69
|
976 * @stable ICU 2.0
|
jpayne@69
|
977 */
|
jpayne@69
|
978 virtual uint32_t getVariableTop(UErrorCode &status) const = 0;
|
jpayne@69
|
979
|
jpayne@69
|
980 /**
|
jpayne@69
|
981 * Get a UnicodeSet that contains all the characters and sequences
|
jpayne@69
|
982 * tailored in this collator.
|
jpayne@69
|
983 * @param status error code of the operation
|
jpayne@69
|
984 * @return a pointer to a UnicodeSet object containing all the
|
jpayne@69
|
985 * code points and sequences that may sort differently than
|
jpayne@69
|
986 * in the root collator. The object must be disposed of by using delete
|
jpayne@69
|
987 * @stable ICU 2.4
|
jpayne@69
|
988 */
|
jpayne@69
|
989 virtual UnicodeSet *getTailoredSet(UErrorCode &status) const;
|
jpayne@69
|
990
|
jpayne@69
|
991 #ifndef U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
992 /**
|
jpayne@69
|
993 * Same as clone().
|
jpayne@69
|
994 * The base class implementation simply calls clone().
|
jpayne@69
|
995 * @return a copy of this object, owned by the caller
|
jpayne@69
|
996 * @see clone()
|
jpayne@69
|
997 * @deprecated ICU 50 no need to have two methods for cloning
|
jpayne@69
|
998 */
|
jpayne@69
|
999 virtual Collator* safeClone() const;
|
jpayne@69
|
1000 #endif // U_FORCE_HIDE_DEPRECATED_API
|
jpayne@69
|
1001
|
jpayne@69
|
1002 /**
|
jpayne@69
|
1003 * Get the sort key as an array of bytes from a UnicodeString.
|
jpayne@69
|
1004 * Sort key byte arrays are zero-terminated and can be compared using
|
jpayne@69
|
1005 * strcmp().
|
jpayne@69
|
1006 *
|
jpayne@69
|
1007 * Note that sort keys are often less efficient than simply doing comparison.
|
jpayne@69
|
1008 * For more details, see the ICU User Guide.
|
jpayne@69
|
1009 *
|
jpayne@69
|
1010 * @param source string to be processed.
|
jpayne@69
|
1011 * @param result buffer to store result in. If NULL, number of bytes needed
|
jpayne@69
|
1012 * will be returned.
|
jpayne@69
|
1013 * @param resultLength length of the result buffer. If if not enough the
|
jpayne@69
|
1014 * buffer will be filled to capacity.
|
jpayne@69
|
1015 * @return Number of bytes needed for storing the sort key
|
jpayne@69
|
1016 * @stable ICU 2.2
|
jpayne@69
|
1017 */
|
jpayne@69
|
1018 virtual int32_t getSortKey(const UnicodeString& source,
|
jpayne@69
|
1019 uint8_t* result,
|
jpayne@69
|
1020 int32_t resultLength) const = 0;
|
jpayne@69
|
1021
|
jpayne@69
|
1022 /**
|
jpayne@69
|
1023 * Get the sort key as an array of bytes from a char16_t buffer.
|
jpayne@69
|
1024 * Sort key byte arrays are zero-terminated and can be compared using
|
jpayne@69
|
1025 * strcmp().
|
jpayne@69
|
1026 *
|
jpayne@69
|
1027 * Note that sort keys are often less efficient than simply doing comparison.
|
jpayne@69
|
1028 * For more details, see the ICU User Guide.
|
jpayne@69
|
1029 *
|
jpayne@69
|
1030 * @param source string to be processed.
|
jpayne@69
|
1031 * @param sourceLength length of string to be processed.
|
jpayne@69
|
1032 * If -1, the string is 0 terminated and length will be decided by the
|
jpayne@69
|
1033 * function.
|
jpayne@69
|
1034 * @param result buffer to store result in. If NULL, number of bytes needed
|
jpayne@69
|
1035 * will be returned.
|
jpayne@69
|
1036 * @param resultLength length of the result buffer. If if not enough the
|
jpayne@69
|
1037 * buffer will be filled to capacity.
|
jpayne@69
|
1038 * @return Number of bytes needed for storing the sort key
|
jpayne@69
|
1039 * @stable ICU 2.2
|
jpayne@69
|
1040 */
|
jpayne@69
|
1041 virtual int32_t getSortKey(const char16_t*source, int32_t sourceLength,
|
jpayne@69
|
1042 uint8_t*result, int32_t resultLength) const = 0;
|
jpayne@69
|
1043
|
jpayne@69
|
1044 /**
|
jpayne@69
|
1045 * Produce a bound for a given sortkey and a number of levels.
|
jpayne@69
|
1046 * Return value is always the number of bytes needed, regardless of
|
jpayne@69
|
1047 * whether the result buffer was big enough or even valid.<br>
|
jpayne@69
|
1048 * Resulting bounds can be used to produce a range of strings that are
|
jpayne@69
|
1049 * between upper and lower bounds. For example, if bounds are produced
|
jpayne@69
|
1050 * for a sortkey of string "smith", strings between upper and lower
|
jpayne@69
|
1051 * bounds with one level would include "Smith", "SMITH", "sMiTh".<br>
|
jpayne@69
|
1052 * There are two upper bounds that can be produced. If UCOL_BOUND_UPPER
|
jpayne@69
|
1053 * is produced, strings matched would be as above. However, if bound
|
jpayne@69
|
1054 * produced using UCOL_BOUND_UPPER_LONG is used, the above example will
|
jpayne@69
|
1055 * also match "Smithsonian" and similar.<br>
|
jpayne@69
|
1056 * For more on usage, see example in cintltst/capitst.c in procedure
|
jpayne@69
|
1057 * TestBounds.
|
jpayne@69
|
1058 * Sort keys may be compared using <TT>strcmp</TT>.
|
jpayne@69
|
1059 * @param source The source sortkey.
|
jpayne@69
|
1060 * @param sourceLength The length of source, or -1 if null-terminated.
|
jpayne@69
|
1061 * (If an unmodified sortkey is passed, it is always null
|
jpayne@69
|
1062 * terminated).
|
jpayne@69
|
1063 * @param boundType Type of bound required. It can be UCOL_BOUND_LOWER, which
|
jpayne@69
|
1064 * produces a lower inclusive bound, UCOL_BOUND_UPPER, that
|
jpayne@69
|
1065 * produces upper bound that matches strings of the same length
|
jpayne@69
|
1066 * or UCOL_BOUND_UPPER_LONG that matches strings that have the
|
jpayne@69
|
1067 * same starting substring as the source string.
|
jpayne@69
|
1068 * @param noOfLevels Number of levels required in the resulting bound (for most
|
jpayne@69
|
1069 * uses, the recommended value is 1). See users guide for
|
jpayne@69
|
1070 * explanation on number of levels a sortkey can have.
|
jpayne@69
|
1071 * @param result A pointer to a buffer to receive the resulting sortkey.
|
jpayne@69
|
1072 * @param resultLength The maximum size of result.
|
jpayne@69
|
1073 * @param status Used for returning error code if something went wrong. If the
|
jpayne@69
|
1074 * number of levels requested is higher than the number of levels
|
jpayne@69
|
1075 * in the source key, a warning (U_SORT_KEY_TOO_SHORT_WARNING) is
|
jpayne@69
|
1076 * issued.
|
jpayne@69
|
1077 * @return The size needed to fully store the bound.
|
jpayne@69
|
1078 * @see ucol_keyHashCode
|
jpayne@69
|
1079 * @stable ICU 2.1
|
jpayne@69
|
1080 */
|
jpayne@69
|
1081 static int32_t U_EXPORT2 getBound(const uint8_t *source,
|
jpayne@69
|
1082 int32_t sourceLength,
|
jpayne@69
|
1083 UColBoundMode boundType,
|
jpayne@69
|
1084 uint32_t noOfLevels,
|
jpayne@69
|
1085 uint8_t *result,
|
jpayne@69
|
1086 int32_t resultLength,
|
jpayne@69
|
1087 UErrorCode &status);
|
jpayne@69
|
1088
|
jpayne@69
|
1089
|
jpayne@69
|
1090 protected:
|
jpayne@69
|
1091
|
jpayne@69
|
1092 // Collator protected constructors -------------------------------------
|
jpayne@69
|
1093
|
jpayne@69
|
1094 /**
|
jpayne@69
|
1095 * Default constructor.
|
jpayne@69
|
1096 * Constructor is different from the old default Collator constructor.
|
jpayne@69
|
1097 * The task for determing the default collation strength and normalization
|
jpayne@69
|
1098 * mode is left to the child class.
|
jpayne@69
|
1099 * @stable ICU 2.0
|
jpayne@69
|
1100 */
|
jpayne@69
|
1101 Collator();
|
jpayne@69
|
1102
|
jpayne@69
|
1103 #ifndef U_HIDE_DEPRECATED_API
|
jpayne@69
|
1104 /**
|
jpayne@69
|
1105 * Constructor.
|
jpayne@69
|
1106 * Empty constructor, does not handle the arguments.
|
jpayne@69
|
1107 * This constructor is done for backward compatibility with 1.7 and 1.8.
|
jpayne@69
|
1108 * The task for handling the argument collation strength and normalization
|
jpayne@69
|
1109 * mode is left to the child class.
|
jpayne@69
|
1110 * @param collationStrength collation strength
|
jpayne@69
|
1111 * @param decompositionMode
|
jpayne@69
|
1112 * @deprecated ICU 2.4. Subclasses should use the default constructor
|
jpayne@69
|
1113 * instead and handle the strength and normalization mode themselves.
|
jpayne@69
|
1114 */
|
jpayne@69
|
1115 Collator(UCollationStrength collationStrength,
|
jpayne@69
|
1116 UNormalizationMode decompositionMode);
|
jpayne@69
|
1117 #endif /* U_HIDE_DEPRECATED_API */
|
jpayne@69
|
1118
|
jpayne@69
|
1119 /**
|
jpayne@69
|
1120 * Copy constructor.
|
jpayne@69
|
1121 * @param other Collator object to be copied from
|
jpayne@69
|
1122 * @stable ICU 2.0
|
jpayne@69
|
1123 */
|
jpayne@69
|
1124 Collator(const Collator& other);
|
jpayne@69
|
1125
|
jpayne@69
|
1126 public:
|
jpayne@69
|
1127 /**
|
jpayne@69
|
1128 * Used internally by registration to define the requested and valid locales.
|
jpayne@69
|
1129 * @param requestedLocale the requested locale
|
jpayne@69
|
1130 * @param validLocale the valid locale
|
jpayne@69
|
1131 * @param actualLocale the actual locale
|
jpayne@69
|
1132 * @internal
|
jpayne@69
|
1133 */
|
jpayne@69
|
1134 virtual void setLocales(const Locale& requestedLocale, const Locale& validLocale, const Locale& actualLocale);
|
jpayne@69
|
1135
|
jpayne@69
|
1136 /** Get the short definition string for a collator. This internal API harvests the collator's
|
jpayne@69
|
1137 * locale and the attribute set and produces a string that can be used for opening
|
jpayne@69
|
1138 * a collator with the same attributes using the ucol_openFromShortString API.
|
jpayne@69
|
1139 * This string will be normalized.
|
jpayne@69
|
1140 * The structure and the syntax of the string is defined in the "Naming collators"
|
jpayne@69
|
1141 * section of the users guide:
|
jpayne@69
|
1142 * http://userguide.icu-project.org/collation/concepts#TOC-Collator-naming-scheme
|
jpayne@69
|
1143 * This function supports preflighting.
|
jpayne@69
|
1144 *
|
jpayne@69
|
1145 * This is internal, and intended to be used with delegate converters.
|
jpayne@69
|
1146 *
|
jpayne@69
|
1147 * @param locale a locale that will appear as a collators locale in the resulting
|
jpayne@69
|
1148 * short string definition. If NULL, the locale will be harvested
|
jpayne@69
|
1149 * from the collator.
|
jpayne@69
|
1150 * @param buffer space to hold the resulting string
|
jpayne@69
|
1151 * @param capacity capacity of the buffer
|
jpayne@69
|
1152 * @param status for returning errors. All the preflighting errors are featured
|
jpayne@69
|
1153 * @return length of the resulting string
|
jpayne@69
|
1154 * @see ucol_openFromShortString
|
jpayne@69
|
1155 * @see ucol_normalizeShortDefinitionString
|
jpayne@69
|
1156 * @see ucol_getShortDefinitionString
|
jpayne@69
|
1157 * @internal
|
jpayne@69
|
1158 */
|
jpayne@69
|
1159 virtual int32_t internalGetShortDefinitionString(const char *locale,
|
jpayne@69
|
1160 char *buffer,
|
jpayne@69
|
1161 int32_t capacity,
|
jpayne@69
|
1162 UErrorCode &status) const;
|
jpayne@69
|
1163
|
jpayne@69
|
1164 /**
|
jpayne@69
|
1165 * Implements ucol_strcollUTF8().
|
jpayne@69
|
1166 * @internal
|
jpayne@69
|
1167 */
|
jpayne@69
|
1168 virtual UCollationResult internalCompareUTF8(
|
jpayne@69
|
1169 const char *left, int32_t leftLength,
|
jpayne@69
|
1170 const char *right, int32_t rightLength,
|
jpayne@69
|
1171 UErrorCode &errorCode) const;
|
jpayne@69
|
1172
|
jpayne@69
|
1173 /**
|
jpayne@69
|
1174 * Implements ucol_nextSortKeyPart().
|
jpayne@69
|
1175 * @internal
|
jpayne@69
|
1176 */
|
jpayne@69
|
1177 virtual int32_t
|
jpayne@69
|
1178 internalNextSortKeyPart(
|
jpayne@69
|
1179 UCharIterator *iter, uint32_t state[2],
|
jpayne@69
|
1180 uint8_t *dest, int32_t count, UErrorCode &errorCode) const;
|
jpayne@69
|
1181
|
jpayne@69
|
1182 #ifndef U_HIDE_INTERNAL_API
|
jpayne@69
|
1183 /** @internal */
|
jpayne@69
|
1184 static inline Collator *fromUCollator(UCollator *uc) {
|
jpayne@69
|
1185 return reinterpret_cast<Collator *>(uc);
|
jpayne@69
|
1186 }
|
jpayne@69
|
1187 /** @internal */
|
jpayne@69
|
1188 static inline const Collator *fromUCollator(const UCollator *uc) {
|
jpayne@69
|
1189 return reinterpret_cast<const Collator *>(uc);
|
jpayne@69
|
1190 }
|
jpayne@69
|
1191 /** @internal */
|
jpayne@69
|
1192 inline UCollator *toUCollator() {
|
jpayne@69
|
1193 return reinterpret_cast<UCollator *>(this);
|
jpayne@69
|
1194 }
|
jpayne@69
|
1195 /** @internal */
|
jpayne@69
|
1196 inline const UCollator *toUCollator() const {
|
jpayne@69
|
1197 return reinterpret_cast<const UCollator *>(this);
|
jpayne@69
|
1198 }
|
jpayne@69
|
1199 #endif // U_HIDE_INTERNAL_API
|
jpayne@69
|
1200
|
jpayne@69
|
1201 private:
|
jpayne@69
|
1202 /**
|
jpayne@69
|
1203 * Assignment operator. Private for now.
|
jpayne@69
|
1204 */
|
jpayne@69
|
1205 Collator& operator=(const Collator& other);
|
jpayne@69
|
1206
|
jpayne@69
|
1207 friend class CFactory;
|
jpayne@69
|
1208 friend class SimpleCFactory;
|
jpayne@69
|
1209 friend class ICUCollatorFactory;
|
jpayne@69
|
1210 friend class ICUCollatorService;
|
jpayne@69
|
1211 static Collator* makeInstance(const Locale& desiredLocale,
|
jpayne@69
|
1212 UErrorCode& status);
|
jpayne@69
|
1213 };
|
jpayne@69
|
1214
|
jpayne@69
|
1215 #if !UCONFIG_NO_SERVICE
|
jpayne@69
|
1216 /**
|
jpayne@69
|
1217 * A factory, used with registerFactory, the creates multiple collators and provides
|
jpayne@69
|
1218 * display names for them. A factory supports some number of locales-- these are the
|
jpayne@69
|
1219 * locales for which it can create collators. The factory can be visible, in which
|
jpayne@69
|
1220 * case the supported locales will be enumerated by getAvailableLocales, or invisible,
|
jpayne@69
|
1221 * in which they are not. Invisible locales are still supported, they are just not
|
jpayne@69
|
1222 * listed by getAvailableLocales.
|
jpayne@69
|
1223 * <p>
|
jpayne@69
|
1224 * If standard locale display names are sufficient, Collator instances can
|
jpayne@69
|
1225 * be registered using registerInstance instead.</p>
|
jpayne@69
|
1226 * <p>
|
jpayne@69
|
1227 * Note: if the collators are to be used from C APIs, they must be instances
|
jpayne@69
|
1228 * of RuleBasedCollator.</p>
|
jpayne@69
|
1229 *
|
jpayne@69
|
1230 * @stable ICU 2.6
|
jpayne@69
|
1231 */
|
jpayne@69
|
1232 class U_I18N_API CollatorFactory : public UObject {
|
jpayne@69
|
1233 public:
|
jpayne@69
|
1234
|
jpayne@69
|
1235 /**
|
jpayne@69
|
1236 * Destructor
|
jpayne@69
|
1237 * @stable ICU 3.0
|
jpayne@69
|
1238 */
|
jpayne@69
|
1239 virtual ~CollatorFactory();
|
jpayne@69
|
1240
|
jpayne@69
|
1241 /**
|
jpayne@69
|
1242 * Return true if this factory is visible. Default is true.
|
jpayne@69
|
1243 * If not visible, the locales supported by this factory will not
|
jpayne@69
|
1244 * be listed by getAvailableLocales.
|
jpayne@69
|
1245 * @return true if the factory is visible.
|
jpayne@69
|
1246 * @stable ICU 2.6
|
jpayne@69
|
1247 */
|
jpayne@69
|
1248 virtual UBool visible(void) const;
|
jpayne@69
|
1249
|
jpayne@69
|
1250 /**
|
jpayne@69
|
1251 * Return a collator for the provided locale. If the locale
|
jpayne@69
|
1252 * is not supported, return NULL.
|
jpayne@69
|
1253 * @param loc the locale identifying the collator to be created.
|
jpayne@69
|
1254 * @return a new collator if the locale is supported, otherwise NULL.
|
jpayne@69
|
1255 * @stable ICU 2.6
|
jpayne@69
|
1256 */
|
jpayne@69
|
1257 virtual Collator* createCollator(const Locale& loc) = 0;
|
jpayne@69
|
1258
|
jpayne@69
|
1259 /**
|
jpayne@69
|
1260 * Return the name of the collator for the objectLocale, localized for the displayLocale.
|
jpayne@69
|
1261 * If objectLocale is not supported, or the factory is not visible, set the result string
|
jpayne@69
|
1262 * to bogus.
|
jpayne@69
|
1263 * @param objectLocale the locale identifying the collator
|
jpayne@69
|
1264 * @param displayLocale the locale for which the display name of the collator should be localized
|
jpayne@69
|
1265 * @param result an output parameter for the display name, set to bogus if not supported.
|
jpayne@69
|
1266 * @return the display name
|
jpayne@69
|
1267 * @stable ICU 2.6
|
jpayne@69
|
1268 */
|
jpayne@69
|
1269 virtual UnicodeString& getDisplayName(const Locale& objectLocale,
|
jpayne@69
|
1270 const Locale& displayLocale,
|
jpayne@69
|
1271 UnicodeString& result);
|
jpayne@69
|
1272
|
jpayne@69
|
1273 /**
|
jpayne@69
|
1274 * Return an array of all the locale names directly supported by this factory.
|
jpayne@69
|
1275 * The number of names is returned in count. This array is owned by the factory.
|
jpayne@69
|
1276 * Its contents must never change.
|
jpayne@69
|
1277 * @param count output parameter for the number of locales supported by the factory
|
jpayne@69
|
1278 * @param status the in/out error code
|
jpayne@69
|
1279 * @return a pointer to an array of count UnicodeStrings.
|
jpayne@69
|
1280 * @stable ICU 2.6
|
jpayne@69
|
1281 */
|
jpayne@69
|
1282 virtual const UnicodeString * getSupportedIDs(int32_t &count, UErrorCode& status) = 0;
|
jpayne@69
|
1283 };
|
jpayne@69
|
1284 #endif /* UCONFIG_NO_SERVICE */
|
jpayne@69
|
1285
|
jpayne@69
|
1286 // Collator inline methods -----------------------------------------------
|
jpayne@69
|
1287
|
jpayne@69
|
1288 U_NAMESPACE_END
|
jpayne@69
|
1289
|
jpayne@69
|
1290 #endif /* #if !UCONFIG_NO_COLLATION */
|
jpayne@69
|
1291
|
jpayne@69
|
1292 #endif /* U_SHOW_CPLUSPLUS_API */
|
jpayne@69
|
1293
|
jpayne@69
|
1294 #endif
|