1 /*
2 * Copyright (c) 1996, 2019, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25
26 /*
27 * (C) Copyright Taligent, Inc. 1996, 1997 - All Rights Reserved
28 * (C) Copyright IBM Corp. 1996 - 1998 - All Rights Reserved
29 *
30 * The original version of this source code and documentation
31 * is copyrighted and owned by Taligent, Inc., a wholly-owned
32 * subsidiary of IBM. These materials are provided under terms
33 * of a License Agreement between Taligent and Sun. This technology
34 * is protected by multiple US and International patents.
35 *
36 * This notice and attribution to Taligent may not be removed.
37 * Taligent is a registered trademark of Taligent, Inc.
38 *
39 */
40
41 package java.util;
42
43 import java.io.IOException;
44 import java.io.ObjectInputStream;
45 import java.io.ObjectOutputStream;
46 import java.io.ObjectStreamField;
47 import java.io.Serializable;
48 import java.text.MessageFormat;
49 import java.util.concurrent.ConcurrentHashMap;
50 import java.util.spi.LocaleNameProvider;
51 import java.util.stream.Collectors;
52
53 import sun.security.action.GetPropertyAction;
54 import sun.util.locale.BaseLocale;
55 import sun.util.locale.InternalLocaleBuilder;
56 import sun.util.locale.LanguageTag;
57 import sun.util.locale.LocaleExtensions;
58 import sun.util.locale.LocaleMatcher;
59 import sun.util.locale.LocaleObjectCache;
60 import sun.util.locale.LocaleSyntaxException;
61 import sun.util.locale.LocaleUtils;
62 import sun.util.locale.ParseStatus;
63 import sun.util.locale.provider.LocaleProviderAdapter;
64 import sun.util.locale.provider.LocaleResources;
65 import sun.util.locale.provider.LocaleServiceProviderPool;
66 import sun.util.locale.provider.TimeZoneNameUtility;
67
68 /**
69 * A <code>Locale</code> object represents a specific geographical, political,
70 * or cultural region. An operation that requires a <code>Locale</code> to perform
71 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code>
72 * to tailor information for the user. For example, displaying a number
73 * is a locale-sensitive operation— the number should be formatted
74 * according to the customs and conventions of the user's native country,
75 * region, or culture.
76 *
77 * <p> The {@code Locale} class implements IETF BCP 47 which is composed of
78 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 "Matching of Language
79 * Tags"</a> and <a href="http://tools.ietf.org/html/rfc5646">RFC 5646 "Tags
80 * for Identifying Languages"</a> with support for the LDML (UTS#35, "Unicode
81 * Locale Data Markup Language") BCP 47-compatible extensions for locale data
82 * exchange.
83 *
84 * <p> A <code>Locale</code> object logically consists of the fields
85 * described below.
86 *
87 * <dl>
88 * <dt><a id="def_language"><b>language</b></a></dt>
89 *
90 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered
91 * language subtags up to 8 alpha letters (for future enhancements).
92 * When a language has both an alpha-2 code and an alpha-3 code, the
93 * alpha-2 code must be used. You can find a full list of valid
94 * language codes in the IANA Language Subtag Registry (search for
95 * "Type: language"). The language field is case insensitive, but
96 * <code>Locale</code> always canonicalizes to lower case.</dd>
97 *
98 * <dd>Well-formed language values have the form
99 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the full
100 * BCP47 language production, since it excludes extlang. They are
101 * not needed since modern three-letter language codes replace
102 * them.</dd>
103 *
104 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd>
105 *
106 * <dt><a id="def_script"><b>script</b></a></dt>
107 *
108 * <dd>ISO 15924 alpha-4 script code. You can find a full list of
109 * valid script codes in the IANA Language Subtag Registry (search
110 * for "Type: script"). The script field is case insensitive, but
111 * <code>Locale</code> always canonicalizes to title case (the first
112 * letter is upper case and the rest of the letters are lower
113 * case).</dd>
114 *
115 * <dd>Well-formed script values have the form
116 * <code>[a-zA-Z]{4}</code></dd>
117 *
118 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd>
119 *
120 * <dt><a id="def_region"><b>country (region)</b></a></dt>
121 *
122 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code.
123 * You can find a full list of valid country and region codes in the
124 * IANA Language Subtag Registry (search for "Type: region"). The
125 * country (region) field is case insensitive, but
126 * <code>Locale</code> always canonicalizes to upper case.</dd>
127 *
128 * <dd>Well-formed country/region values have
129 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd>
130 *
131 * <dd>Example: "US" (United States), "FR" (France), "029"
132 * (Caribbean)</dd>
133 *
134 * <dt><a id="def_variant"><b>variant</b></a></dt>
135 *
136 * <dd>Any arbitrary value used to indicate a variation of a
137 * <code>Locale</code>. Where there are two or more variant values
138 * each indicating its own semantics, these values should be ordered
139 * by importance, with most important first, separated by
140 * underscore('_'). The variant field is case sensitive.</dd>
141 *
142 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant
143 * subtags. Also BCP 47 subtags are strictly used to indicate
144 * additional variations that define a language or its dialects that
145 * are not covered by any combinations of language, script and
146 * region subtags. You can find a full list of valid variant codes
147 * in the IANA Language Subtag Registry (search for "Type: variant").
148 *
149 * <p>However, the variant field in <code>Locale</code> has
150 * historically been used for any kind of variation, not just
151 * language variations. For example, some supported variants
152 * available in Java SE Runtime Environments indicate alternative
153 * cultural behaviors such as calendar type or number script. In
154 * BCP 47 this kind of information, which does not identify the
155 * language, is supported by extension subtags or private use
156 * subtags.</dd>
157 *
158 * <dd>Well-formed variant values have the form <code>SUBTAG
159 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG =
160 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only
161 * uses hyphen ('-') as a delimiter, this is more lenient).</dd>
162 *
163 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd>
164 *
165 * <dt><a id="def_extensions"><b>extensions</b></a></dt>
166 *
167 * <dd>A map from single character keys to string values, indicating
168 * extensions apart from language identification. The extensions in
169 * <code>Locale</code> implement the semantics and syntax of BCP 47
170 * extension subtags and private use subtags. The extensions are
171 * case insensitive, but <code>Locale</code> canonicalizes all
172 * extension keys and values to lower case. Note that extensions
173 * cannot have empty values.</dd>
174 *
175 * <dd>Well-formed keys are single characters from the set
176 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form
177 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x'
178 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys
179 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows
180 * single-character subtags).</dd>
181 *
182 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar),
183 * key="x"/value="java-1-7"</dd>
184 * </dl>
185 *
186 * <b>Note:</b> Although BCP 47 requires field values to be registered
187 * in the IANA Language Subtag Registry, the <code>Locale</code> class
188 * does not provide any validation features. The <code>Builder</code>
189 * only checks if an individual field satisfies the syntactic
190 * requirement (is well-formed), but does not validate the value
191 * itself. See {@link Builder} for details.
192 *
193 * <h2><a id="def_locale_extension">Unicode locale/language extension</a></h2>
194 *
195 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional
196 * attributes and keywords to override or refine the default behavior
197 * associated with a locale. A keyword is represented by a pair of
198 * key and type. For example, "nu-thai" indicates that Thai local
199 * digits (value:"thai") should be used for formatting numbers
200 * (key:"nu").
201 *
202 * <p>The keywords are mapped to a BCP 47 extension value using the
203 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above
204 * example, "nu-thai", becomes the extension "u-nu-thai".
205 *
206 * <p>Thus, when a <code>Locale</code> object contains Unicode locale
207 * attributes and keywords,
208 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a
209 * String representing this information, for example, "nu-thai". The
210 * <code>Locale</code> class also provides {@link
211 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and
212 * {@link #getUnicodeLocaleType} which allow you to access Unicode
213 * locale attributes and key/type pairs directly. When represented as
214 * a string, the Unicode Locale Extension lists attributes
215 * alphabetically, followed by key/type sequences with keys listed
216 * alphabetically (the order of subtags comprising a key's type is
217 * fixed when the type is defined)
218 *
219 * <p>A well-formed locale key has the form
220 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the
221 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it
222 * can be empty, or a series of subtags 3-8 alphanums in length). A
223 * well-formed locale attribute has the form
224 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same
225 * form as a locale type subtag).
226 *
227 * <p>The Unicode locale extension specifies optional behavior in
228 * locale-sensitive services. Although the LDML specification defines
229 * various keys and values, actual locale-sensitive service
230 * implementations in a Java Runtime Environment might not support any
231 * particular Unicode locale attributes or key/type pairs.
232 *
233 * <h3>Creating a Locale</h3>
234 *
235 * <p>There are several different ways to create a <code>Locale</code>
236 * object.
237 *
238 * <h4>Builder</h4>
239 *
240 * <p>Using {@link Builder} you can construct a <code>Locale</code> object
241 * that conforms to BCP 47 syntax.
242 *
243 * <h4>Constructors</h4>
244 *
245 * <p>The <code>Locale</code> class provides three constructors:
246 * <blockquote>
247 * <pre>
248 * {@link #Locale(String language)}
249 * {@link #Locale(String language, String country)}
250 * {@link #Locale(String language, String country, String variant)}
251 * </pre>
252 * </blockquote>
253 * These constructors allow you to create a <code>Locale</code> object
254 * with language, country and variant, but you cannot specify
255 * script or extensions.
256 *
257 * <h4>Factory Methods</h4>
258 *
259 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code>
260 * object for a well-formed BCP 47 language tag.
261 *
262 * <h4>Locale Constants</h4>
263 *
264 * <p>The <code>Locale</code> class provides a number of convenient constants
265 * that you can use to create <code>Locale</code> objects for commonly used
266 * locales. For example, the following creates a <code>Locale</code> object
267 * for the United States:
268 * <blockquote>
269 * <pre>
270 * Locale.US
271 * </pre>
272 * </blockquote>
273 *
274 * <h3><a id="LocaleMatching">Locale Matching</a></h3>
275 *
276 * <p>If an application or a system is internationalized and provides localized
277 * resources for multiple locales, it sometimes needs to find one or more
278 * locales (or language tags) which meet each user's specific preferences. Note
279 * that a term "language tag" is used interchangeably with "locale" in this
280 * locale matching documentation.
281 *
282 * <p>In order to do matching a user's preferred locales to a set of language
283 * tags, <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
284 * Language Tags</a> defines two mechanisms: filtering and lookup.
285 * <em>Filtering</em> is used to get all matching locales, whereas
286 * <em>lookup</em> is to choose the best matching locale.
287 * Matching is done case-insensitively. These matching mechanisms are described
288 * in the following sections.
289 *
290 * <p>A user's preference is called a <em>Language Priority List</em> and is
291 * expressed as a list of language ranges. There are syntactically two types of
292 * language ranges: basic and extended. See
293 * {@link Locale.LanguageRange Locale.LanguageRange} for details.
294 *
295 * <h4>Filtering</h4>
296 *
297 * <p>The filtering operation returns all matching language tags. It is defined
298 * in RFC 4647 as follows:
299 * "In filtering, each language range represents the least specific language
300 * tag (that is, the language tag with fewest number of subtags) that is an
301 * acceptable match. All of the language tags in the matching set of tags will
302 * have an equal or greater number of subtags than the language range. Every
303 * non-wildcard subtag in the language range will appear in every one of the
304 * matching language tags."
305 *
306 * <p>There are two types of filtering: filtering for basic language ranges
307 * (called "basic filtering") and filtering for extended language ranges
308 * (called "extended filtering"). They may return different results by what
309 * kind of language ranges are included in the given Language Priority List.
310 * {@link Locale.FilteringMode} is a parameter to specify how filtering should
311 * be done.
312 *
313 * <h4>Lookup</h4>
314 *
315 * <p>The lookup operation returns the best matching language tags. It is
316 * defined in RFC 4647 as follows:
317 * "By contrast with filtering, each language range represents the most
318 * specific tag that is an acceptable match. The first matching tag found,
319 * according to the user's priority, is considered the closest match and is the
320 * item returned."
321 *
322 * <p>For example, if a Language Priority List consists of two language ranges,
323 * {@code "zh-Hant-TW"} and {@code "en-US"}, in prioritized order, lookup
324 * method progressively searches the language tags below in order to find the
325 * best matching language tag.
326 * <blockquote>
327 * <pre>
328 * 1. zh-Hant-TW
329 * 2. zh-Hant
330 * 3. zh
331 * 4. en-US
332 * 5. en
333 * </pre>
334 * </blockquote>
335 * If there is a language tag which matches completely to a language range
336 * above, the language tag is returned.
337 *
338 * <p>{@code "*"} is the special language range, and it is ignored in lookup.
339 *
340 * <p>If multiple language tags match as a result of the subtag {@code '*'}
341 * included in a language range, the first matching language tag returned by
342 * an {@link Iterator} over a {@link Collection} of language tags is treated as
343 * the best matching one.
344 *
345 * <h3>Use of Locale</h3>
346 *
347 * <p>Once you've created a <code>Locale</code> you can query it for information
348 * about itself. Use <code>getCountry</code> to get the country (or region)
349 * code and <code>getLanguage</code> to get the language code.
350 * You can use <code>getDisplayCountry</code> to get the
351 * name of the country suitable for displaying to the user. Similarly,
352 * you can use <code>getDisplayLanguage</code> to get the name of
353 * the language suitable for displaying to the user. Interestingly,
354 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive
355 * and have two versions: one that uses the default
356 * {@link Locale.Category#DISPLAY DISPLAY} locale and one
357 * that uses the locale specified as an argument.
358 *
359 * <p>The Java Platform provides a number of classes that perform locale-sensitive
360 * operations. For example, the <code>NumberFormat</code> class formats
361 * numbers, currency, and percentages in a locale-sensitive manner. Classes
362 * such as <code>NumberFormat</code> have several convenience methods
363 * for creating a default object of that type. For example, the
364 * <code>NumberFormat</code> class provides these three convenience methods
365 * for creating a default <code>NumberFormat</code> object:
366 * <blockquote>
367 * <pre>
368 * NumberFormat.getInstance()
369 * NumberFormat.getCurrencyInstance()
370 * NumberFormat.getPercentInstance()
371 * </pre>
372 * </blockquote>
373 * Each of these methods has two variants; one with an explicit locale
374 * and one without; the latter uses the default
375 * {@link Locale.Category#FORMAT FORMAT} locale:
376 * <blockquote>
377 * <pre>
378 * NumberFormat.getInstance(myLocale)
379 * NumberFormat.getCurrencyInstance(myLocale)
380 * NumberFormat.getPercentInstance(myLocale)
381 * </pre>
382 * </blockquote>
383 * A <code>Locale</code> is the mechanism for identifying the kind of object
384 * (<code>NumberFormat</code>) that you would like to get. The locale is
385 * <STRONG>just</STRONG> a mechanism for identifying objects,
386 * <STRONG>not</STRONG> a container for the objects themselves.
387 *
388 * <h3>Compatibility</h3>
389 *
390 * <p>In order to maintain compatibility with existing usage, Locale's
391 * constructors retain their behavior prior to the Java Runtime
392 * Environment version 1.7. The same is largely true for the
393 * <code>toString</code> method. Thus Locale objects can continue to
394 * be used as they were. In particular, clients who parse the output
395 * of toString into language, country, and variant fields can continue
396 * to do so (although this is strongly discouraged), although the
397 * variant field will have additional information in it if script or
398 * extensions are present.
399 *
400 * <p>In addition, BCP 47 imposes syntax restrictions that are not
401 * imposed by Locale's constructors. This means that conversions
402 * between some Locales and BCP 47 language tags cannot be made without
403 * losing information. Thus <code>toLanguageTag</code> cannot
404 * represent the state of locales whose language, country, or variant
405 * do not conform to BCP 47.
406 *
407 * <p>Because of these issues, it is recommended that clients migrate
408 * away from constructing non-conforming locales and use the
409 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead.
410 * Clients desiring a string representation of the complete locale can
411 * then always rely on <code>toLanguageTag</code> for this purpose.
412 *
413 * <h4><a id="special_cases_constructor">Special cases</a></h4>
414 *
415 * <p>For compatibility reasons, two
416 * non-conforming locales are treated as special cases. These are
417 * <b>{@code ja_JP_JP}</b> and <b>{@code th_TH_TH}</b>. These are ill-formed
418 * in BCP 47 since the variants are too short. To ease migration to BCP 47,
419 * these are treated specially during construction. These two cases (and only
420 * these) cause a constructor to generate an extension, all other values behave
421 * exactly as they did prior to Java 7.
422 *
423 * <p>Java has used {@code ja_JP_JP} to represent Japanese as used in
424 * Japan together with the Japanese Imperial calendar. This is now
425 * representable using a Unicode locale extension, by specifying the
426 * Unicode locale key {@code ca} (for "calendar") and type
427 * {@code japanese}. When the Locale constructor is called with the
428 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is
429 * automatically added.
430 *
431 * <p>Java has used {@code th_TH_TH} to represent Thai as used in
432 * Thailand together with Thai digits. This is also now representable using
433 * a Unicode locale extension, by specifying the Unicode locale key
434 * {@code nu} (for "number") and value {@code thai}. When the Locale
435 * constructor is called with the arguments "th", "TH", "TH", the
436 * extension "u-nu-thai" is automatically added.
437 *
438 * <h4>Serialization</h4>
439 *
440 * <p>During serialization, writeObject writes all fields to the output
441 * stream, including extensions.
442 *
443 * <p>During deserialization, readResolve adds extensions as described
444 * in <a href="#special_cases_constructor">Special Cases</a>, only
445 * for the two cases th_TH_TH and ja_JP_JP.
446 *
447 * <h4>Legacy language codes</h4>
448 *
449 * <p>Locale's constructor has always converted three language codes to
450 * their earlier, obsoleted forms: {@code he} maps to {@code iw},
451 * {@code yi} maps to {@code ji}, and {@code id} maps to
452 * {@code in}. This continues to be the case, in order to not break
453 * backwards compatibility.
454 *
455 * <p>The APIs added in 1.7 map between the old and new language codes,
456 * maintaining the old codes internal to Locale (so that
457 * <code>getLanguage</code> and <code>toString</code> reflect the old
458 * code), but using the new codes in the BCP 47 language tag APIs (so
459 * that <code>toLanguageTag</code> reflects the new one). This
460 * preserves the equivalence between Locales no matter which code or
461 * API is used to construct them. Java's default resource bundle
462 * lookup mechanism also implements this mapping, so that resources
463 * can be named using either convention, see {@link ResourceBundle.Control}.
464 *
465 * <h4>Three-letter language/country(region) codes</h4>
466 *
467 * <p>The Locale constructors have always specified that the language
468 * and the country param be two characters in length, although in
469 * practice they have accepted any length. The specification has now
470 * been relaxed to allow language codes of two to eight characters and
471 * country (region) codes of two to three characters, and in
472 * particular, three-letter language codes and three-digit region
473 * codes as specified in the IANA Language Subtag Registry. For
474 * compatibility, the implementation still does not impose a length
475 * constraint.
476 *
477 * @see Builder
478 * @see ResourceBundle
479 * @see java.text.Format
480 * @see java.text.NumberFormat
481 * @see java.text.Collator
482 * @author Mark Davis
483 * @since 1.1
484 */
485 public final class Locale implements Cloneable, Serializable {
486
487 /** Useful constant for language.
488 */
489 public static final Locale ENGLISH;
490
491 /** Useful constant for language.
492 */
493 public static final Locale FRENCH;
494
495 /** Useful constant for language.
496 */
497 public static final Locale GERMAN;
498
499 /** Useful constant for language.
500 */
501 public static final Locale ITALIAN;
502
503 /** Useful constant for language.
504 */
505 public static final Locale JAPANESE;
506
507 /** Useful constant for language.
508 */
509 public static final Locale KOREAN;
510
511 /** Useful constant for language.
512 */
513 public static final Locale CHINESE;
514
515 /** Useful constant for language.
516 */
517 public static final Locale SIMPLIFIED_CHINESE;
518
519 /** Useful constant for language.
520 */
521 public static final Locale TRADITIONAL_CHINESE;
522
523 /** Useful constant for country.
524 */
525 public static final Locale FRANCE;
526
527 /** Useful constant for country.
528 */
529 public static final Locale GERMANY;
530
531 /** Useful constant for country.
532 */
533 public static final Locale ITALY;
534
535 /** Useful constant for country.
536 */
537 public static final Locale JAPAN;
538
539 /** Useful constant for country.
540 */
541 public static final Locale KOREA;
542
543 /** Useful constant for country.
544 */
545 public static final Locale UK;
546
547 /** Useful constant for country.
548 */
549 public static final Locale US;
550
551 /** Useful constant for country.
552 */
553 public static final Locale CANADA;
554
555 /** Useful constant for country.
556 */
557 public static final Locale CANADA_FRENCH;
558
559 /**
560 * Useful constant for the root locale. The root locale is the locale whose
561 * language, country, and variant are empty ("") strings. This is regarded
562 * as the base locale of all locales, and is used as the language/country
563 * neutral locale for the locale sensitive operations.
564 *
565 * @since 1.6
566 */
567 public static final Locale ROOT;
568
569 private static final Map<BaseLocale, Locale> CONSTANT_LOCALES = new HashMap<>();
570
571 static {
572 ENGLISH = createConstant(BaseLocale.ENGLISH);
573 FRENCH = createConstant(BaseLocale.FRENCH);
574 GERMAN = createConstant(BaseLocale.GERMAN);
575 ITALIAN = createConstant(BaseLocale.ITALIAN);
576 JAPANESE = createConstant(BaseLocale.JAPANESE);
577 KOREAN = createConstant(BaseLocale.KOREAN);
578 CHINESE = createConstant(BaseLocale.CHINESE);
579 SIMPLIFIED_CHINESE = createConstant(BaseLocale.SIMPLIFIED_CHINESE);
580 TRADITIONAL_CHINESE = createConstant(BaseLocale.TRADITIONAL_CHINESE);
581 FRANCE = createConstant(BaseLocale.FRANCE);
582 GERMANY = createConstant(BaseLocale.GERMANY);
583 ITALY = createConstant(BaseLocale.ITALY);
584 JAPAN = createConstant(BaseLocale.JAPAN);
585 KOREA = createConstant(BaseLocale.KOREA);
586 UK = createConstant(BaseLocale.UK);
587 US = createConstant(BaseLocale.US);
588 CANADA = createConstant(BaseLocale.CANADA);
589 CANADA_FRENCH = createConstant(BaseLocale.CANADA_FRENCH);
590 ROOT = createConstant(BaseLocale.ROOT);
591 }
592
593 /** Useful constant for country.
594 */
595 public static final Locale CHINA = SIMPLIFIED_CHINESE;
596
597 /** Useful constant for country.
598 */
599 public static final Locale PRC = SIMPLIFIED_CHINESE;
600
601 /** Useful constant for country.
602 */
603 public static final Locale TAIWAN = TRADITIONAL_CHINESE;
604
605 /**
606 * This method must be called only for creating the Locale.*
607 * constants due to making shortcuts.
608 */
609 private static Locale createConstant(byte baseType) {
610 BaseLocale base = BaseLocale.constantBaseLocales[baseType];
611 Locale locale = new Locale(base, null);
612 CONSTANT_LOCALES.put(base, locale);
613 return locale;
614 }
615
616 /**
617 * The key for the private use extension ('x').
618 *
619 * @see #getExtension(char)
620 * @see Builder#setExtension(char, String)
621 * @since 1.7
622 */
623 public static final char PRIVATE_USE_EXTENSION = 'x';
624
625 /**
626 * The key for Unicode locale extension ('u').
627 *
628 * @see #getExtension(char)
629 * @see Builder#setExtension(char, String)
630 * @since 1.7
631 */
632 public static final char UNICODE_LOCALE_EXTENSION = 'u';
633
634 /** serialization ID
635 */
636 @java.io.Serial
637 static final long serialVersionUID = 9149081749638150636L;
638
639 /**
640 * Enum for specifying the type defined in ISO 3166. This enum is used to
641 * retrieve the two-letter ISO3166-1 alpha-2, three-letter ISO3166-1
642 * alpha-3, four-letter ISO3166-3 country codes.
643 *
644 * @see #getISOCountries(Locale.IsoCountryCode)
645 * @since 9
646 */
647 public static enum IsoCountryCode {
648 /**
649 * PART1_ALPHA2 is used to represent the ISO3166-1 alpha-2 two letter
650 * country codes.
651 */
652 PART1_ALPHA2 {
653 @Override
654 Set<String> createCountryCodeSet() {
655 return Set.of(Locale.getISOCountries());
656 }
657 },
658
659 /**
660 *
661 * PART1_ALPHA3 is used to represent the ISO3166-1 alpha-3 three letter
662 * country codes.
663 */
664 PART1_ALPHA3 {
665 @Override
666 Set<String> createCountryCodeSet() {
667 return LocaleISOData.computeISO3166_1Alpha3Countries();
668 }
669 },
670
671 /**
672 * PART3 is used to represent the ISO3166-3 four letter country codes.
673 */
674 PART3 {
675 @Override
676 Set<String> createCountryCodeSet() {
677 return Set.of(LocaleISOData.ISO3166_3);
678 }
679 };
680
681 /**
682 * Concrete implementation of this method attempts to compute value
683 * for iso3166CodesMap for each IsoCountryCode type key.
684 */
685 abstract Set<String> createCountryCodeSet();
686
687 /**
688 * Map to hold country codes for each ISO3166 part.
689 */
690 private static Map<IsoCountryCode, Set<String>> iso3166CodesMap = new ConcurrentHashMap<>();
691
692 /**
693 * This method is called from Locale class to retrieve country code set
694 * for getISOCountries(type)
695 */
696 static Set<String> retrieveISOCountryCodes(IsoCountryCode type) {
697 return iso3166CodesMap.computeIfAbsent(type, IsoCountryCode::createCountryCodeSet);
698 }
699 }
700
701 /**
702 * Display types for retrieving localized names from the name providers.
703 */
704 private static final int DISPLAY_LANGUAGE = 0;
705 private static final int DISPLAY_COUNTRY = 1;
706 private static final int DISPLAY_VARIANT = 2;
707 private static final int DISPLAY_SCRIPT = 3;
708 private static final int DISPLAY_UEXT_KEY = 4;
709 private static final int DISPLAY_UEXT_TYPE = 5;
710
711 /**
712 * Private constructor used by getInstance method
713 */
714 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) {
715 this.baseLocale = baseLocale;
716 this.localeExtensions = extensions;
717 }
718
719 /**
720 * Construct a locale from language, country and variant.
721 * This constructor normalizes the language value to lowercase and
722 * the country value to uppercase.
723 * <p>
724 * <b>Note:</b>
725 * <ul>
726 * <li>ISO 639 is not a stable standard; some of the language codes it defines
727 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
728 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
729 * API on Locale will return only the OLD codes.
730 * <li>For backward compatibility reasons, this constructor does not make
731 * any syntactic checks on the input.
732 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially,
733 * see <a href="#special_cases_constructor">Special Cases</a> for more information.
734 * </ul>
735 *
736 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
737 * up to 8 characters in length. See the <code>Locale</code> class description about
738 * valid language values.
739 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
740 * See the <code>Locale</code> class description about valid country values.
741 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>.
742 * See the <code>Locale</code> class description for the details.
743 * @exception NullPointerException thrown if any argument is null.
744 */
745 public Locale(String language, String country, String variant) {
746 if (language == null || country == null || variant == null) {
747 throw new NullPointerException();
748 }
749 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant);
750 localeExtensions = getCompatibilityExtensions(language, "", country, variant);
751 }
752
753 /**
754 * Construct a locale from language and country.
755 * This constructor normalizes the language value to lowercase and
756 * the country value to uppercase.
757 * <p>
758 * <b>Note:</b>
759 * <ul>
760 * <li>ISO 639 is not a stable standard; some of the language codes it defines
761 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
762 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
763 * API on Locale will return only the OLD codes.
764 * <li>For backward compatibility reasons, this constructor does not make
765 * any syntactic checks on the input.
766 * </ul>
767 *
768 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
769 * up to 8 characters in length. See the <code>Locale</code> class description about
770 * valid language values.
771 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
772 * See the <code>Locale</code> class description about valid country values.
773 * @exception NullPointerException thrown if either argument is null.
774 */
775 public Locale(String language, String country) {
776 this(language, country, "");
777 }
778
779 /**
780 * Construct a locale from a language code.
781 * This constructor normalizes the language value to lowercase.
782 * <p>
783 * <b>Note:</b>
784 * <ul>
785 * <li>ISO 639 is not a stable standard; some of the language codes it defines
786 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
787 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
788 * API on Locale will return only the OLD codes.
789 * <li>For backward compatibility reasons, this constructor does not make
790 * any syntactic checks on the input.
791 * </ul>
792 *
793 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
794 * up to 8 characters in length. See the <code>Locale</code> class description about
795 * valid language values.
796 * @exception NullPointerException thrown if argument is null.
797 * @since 1.4
798 */
799 public Locale(String language) {
800 this(language, "", "");
801 }
802
803 /**
804 * Returns a <code>Locale</code> constructed from the given
805 * <code>language</code>, <code>country</code> and
806 * <code>variant</code>. If the same <code>Locale</code> instance
807 * is available in the cache, then that instance is
808 * returned. Otherwise, a new <code>Locale</code> instance is
809 * created and cached.
810 *
811 * @param language lowercase 2 to 8 language code.
812 * @param country uppercase two-letter ISO-3166 code and numeric-3 UN M.49 area code.
813 * @param variant vendor and browser specific code. See class description.
814 * @return the <code>Locale</code> instance requested
815 * @exception NullPointerException if any argument is null.
816 */
817 static Locale getInstance(String language, String country, String variant) {
818 return getInstance(language, "", country, variant, null);
819 }
820
821 static Locale getInstance(String language, String script, String country,
822 String variant, LocaleExtensions extensions) {
823 if (language== null || script == null || country == null || variant == null) {
824 throw new NullPointerException();
825 }
826
827 if (extensions == null) {
828 extensions = getCompatibilityExtensions(language, script, country, variant);
829 }
830
831 BaseLocale baseloc = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant);
832 return getInstance(baseloc, extensions);
833 }
834
835 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) {
836 if (extensions == null) {
837 Locale locale = CONSTANT_LOCALES.get(baseloc);
838 if (locale != null) {
839 return locale;
840 }
841 return Cache.LOCALECACHE.get(baseloc);
842 } else {
843 LocaleKey key = new LocaleKey(baseloc, extensions);
844 return Cache.LOCALECACHE.get(key);
845 }
846 }
847
848 private static class Cache extends LocaleObjectCache<Object, Locale> {
849
850 private static final Cache LOCALECACHE = new Cache();
851
852 private Cache() {
853 }
854
855 @Override
856 protected Locale createObject(Object key) {
857 if (key instanceof BaseLocale) {
858 return new Locale((BaseLocale)key, null);
859 } else {
860 LocaleKey lk = (LocaleKey)key;
861 return new Locale(lk.base, lk.exts);
862 }
863 }
864 }
865
866 private static final class LocaleKey {
867 private final BaseLocale base;
868 private final LocaleExtensions exts;
869 private final int hash;
870
871 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) {
872 base = baseLocale;
873 exts = extensions;
874
875 // Calculate the hash value here because it's always used.
876 int h = base.hashCode();
877 if (exts != null) {
878 h ^= exts.hashCode();
879 }
880 hash = h;
881 }
882
883 @Override
884 public boolean equals(Object obj) {
885 if (this == obj) {
886 return true;
887 }
888 if (!(obj instanceof LocaleKey)) {
889 return false;
890 }
891 LocaleKey other = (LocaleKey)obj;
892 if (hash != other.hash || !base.equals(other.base)) {
893 return false;
894 }
895 if (exts == null) {
896 return other.exts == null;
897 }
898 return exts.equals(other.exts);
899 }
900
901 @Override
902 public int hashCode() {
903 return hash;
904 }
905 }
906
907 /**
908 * Gets the current value of the default locale for this instance
909 * of the Java Virtual Machine.
910 * <p>
911 * The Java Virtual Machine sets the default locale during startup
912 * based on the host environment. It is used by many locale-sensitive
913 * methods if no locale is explicitly specified.
914 * It can be changed using the
915 * {@link #setDefault(java.util.Locale) setDefault} method.
916 *
917 * @return the default locale for this instance of the Java Virtual Machine
918 */
919 public static Locale getDefault() {
920 // do not synchronize this method - see 4071298
921 return defaultLocale;
922 }
923
924 /**
925 * Gets the current value of the default locale for the specified Category
926 * for this instance of the Java Virtual Machine.
927 * <p>
928 * The Java Virtual Machine sets the default locale during startup based
929 * on the host environment. It is used by many locale-sensitive methods
930 * if no locale is explicitly specified. It can be changed using the
931 * setDefault(Locale.Category, Locale) method.
932 *
933 * @param category the specified category to get the default locale
934 * @throws NullPointerException if category is null
935 * @return the default locale for the specified Category for this instance
936 * of the Java Virtual Machine
937 * @see #setDefault(Locale.Category, Locale)
938 * @since 1.7
939 */
940 public static Locale getDefault(Locale.Category category) {
941 // do not synchronize this method - see 4071298
942 switch (category) {
943 case DISPLAY:
944 if (defaultDisplayLocale == null) {
945 synchronized(Locale.class) {
946 if (defaultDisplayLocale == null) {
947 defaultDisplayLocale = initDefault(category);
948 }
949 }
950 }
951 return defaultDisplayLocale;
952 case FORMAT:
953 if (defaultFormatLocale == null) {
954 synchronized(Locale.class) {
955 if (defaultFormatLocale == null) {
956 defaultFormatLocale = initDefault(category);
957 }
958 }
959 }
960 return defaultFormatLocale;
961 default:
962 assert false: "Unknown Category";
963 }
964 return getDefault();
965 }
966
967 private static Locale initDefault() {
968 String language, region, script, country, variant;
969 Properties props = GetPropertyAction.privilegedGetProperties();
970 language = props.getProperty("user.language", "en");
971 // for compatibility, check for old user.region property
972 region = props.getProperty("user.region");
973 if (region != null) {
974 // region can be of form country, country_variant, or _variant
975 int i = region.indexOf('_');
976 if (i >= 0) {
977 country = region.substring(0, i);
978 variant = region.substring(i + 1);
979 } else {
980 country = region;
981 variant = "";
982 }
983 script = "";
984 } else {
985 script = props.getProperty("user.script", "");
986 country = props.getProperty("user.country", "");
987 variant = props.getProperty("user.variant", "");
988 }
989
990 return getInstance(language, script, country, variant,
991 getDefaultExtensions(props.getProperty("user.extensions", ""))
992 .orElse(null));
993 }
994
995 private static Locale initDefault(Locale.Category category) {
996 Properties props = GetPropertyAction.privilegedGetProperties();
997
998 return getInstance(
999 props.getProperty(category.languageKey,
1000 defaultLocale.getLanguage()),
1001 props.getProperty(category.scriptKey,
1002 defaultLocale.getScript()),
1003 props.getProperty(category.countryKey,
1004 defaultLocale.getCountry()),
1005 props.getProperty(category.variantKey,
1006 defaultLocale.getVariant()),
1007 getDefaultExtensions(props.getProperty(category.extensionsKey, ""))
1008 .orElse(defaultLocale.getLocaleExtensions()));
1009 }
1010
1011 private static Optional<LocaleExtensions> getDefaultExtensions(String extensionsProp) {
1012 if (LocaleUtils.isEmpty(extensionsProp)) {
1013 return Optional.empty();
1014 }
1015
1016 LocaleExtensions exts = null;
1017 try {
1018 exts = new InternalLocaleBuilder()
1019 .setExtensions(extensionsProp)
1020 .getLocaleExtensions();
1021 } catch (LocaleSyntaxException e) {
1022 // just ignore this incorrect property
1023 }
1024
1025 return Optional.ofNullable(exts);
1026 }
1027
1028 /**
1029 * Sets the default locale for this instance of the Java Virtual Machine.
1030 * This does not affect the host locale.
1031 * <p>
1032 * If there is a security manager, its <code>checkPermission</code>
1033 * method is called with a <code>PropertyPermission("user.language", "write")</code>
1034 * permission before the default locale is changed.
1035 * <p>
1036 * The Java Virtual Machine sets the default locale during startup
1037 * based on the host environment. It is used by many locale-sensitive
1038 * methods if no locale is explicitly specified.
1039 * <p>
1040 * Since changing the default locale may affect many different areas
1041 * of functionality, this method should only be used if the caller
1042 * is prepared to reinitialize locale-sensitive code running
1043 * within the same Java Virtual Machine.
1044 * <p>
1045 * By setting the default locale with this method, all of the default
1046 * locales for each Category are also set to the specified default locale.
1047 *
1048 * @throws SecurityException
1049 * if a security manager exists and its
1050 * <code>checkPermission</code> method doesn't allow the operation.
1051 * @throws NullPointerException if <code>newLocale</code> is null
1052 * @param newLocale the new default locale
1053 * @see SecurityManager#checkPermission
1054 * @see java.util.PropertyPermission
1055 */
1056 public static synchronized void setDefault(Locale newLocale) {
1057 setDefault(Category.DISPLAY, newLocale);
1058 setDefault(Category.FORMAT, newLocale);
1059 defaultLocale = newLocale;
1060 }
1061
1062 /**
1063 * Sets the default locale for the specified Category for this instance
1064 * of the Java Virtual Machine. This does not affect the host locale.
1065 * <p>
1066 * If there is a security manager, its checkPermission method is called
1067 * with a PropertyPermission("user.language", "write") permission before
1068 * the default locale is changed.
1069 * <p>
1070 * The Java Virtual Machine sets the default locale during startup based
1071 * on the host environment. It is used by many locale-sensitive methods
1072 * if no locale is explicitly specified.
1073 * <p>
1074 * Since changing the default locale may affect many different areas of
1075 * functionality, this method should only be used if the caller is
1076 * prepared to reinitialize locale-sensitive code running within the
1077 * same Java Virtual Machine.
1078 *
1079 * @param category the specified category to set the default locale
1080 * @param newLocale the new default locale
1081 * @throws SecurityException if a security manager exists and its
1082 * checkPermission method doesn't allow the operation.
1083 * @throws NullPointerException if category and/or newLocale is null
1084 * @see SecurityManager#checkPermission(java.security.Permission)
1085 * @see PropertyPermission
1086 * @see #getDefault(Locale.Category)
1087 * @since 1.7
1088 */
1089 public static synchronized void setDefault(Locale.Category category,
1090 Locale newLocale) {
1091 if (category == null)
1092 throw new NullPointerException("Category cannot be NULL");
1093 if (newLocale == null)
1094 throw new NullPointerException("Can't set default locale to NULL");
1095
1096 SecurityManager sm = System.getSecurityManager();
1097 if (sm != null) sm.checkPermission(new PropertyPermission
1098 ("user.language", "write"));
1099 switch (category) {
1100 case DISPLAY:
1101 defaultDisplayLocale = newLocale;
1102 break;
1103 case FORMAT:
1104 defaultFormatLocale = newLocale;
1105 break;
1106 default:
1107 assert false: "Unknown Category";
1108 }
1109 }
1110
1111 /**
1112 * Returns an array of all installed locales.
1113 * The returned array represents the union of locales supported
1114 * by the Java runtime environment and by installed
1115 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider}
1116 * implementations. It must contain at least a <code>Locale</code>
1117 * instance equal to {@link java.util.Locale#US Locale.US}.
1118 *
1119 * @return An array of installed locales.
1120 */
1121 public static Locale[] getAvailableLocales() {
1122 return LocaleServiceProviderPool.getAllAvailableLocales();
1123 }
1124
1125 /**
1126 * Returns a list of all 2-letter country codes defined in ISO 3166.
1127 * Can be used to create Locales.
1128 * This method is equivalent to {@link #getISOCountries(Locale.IsoCountryCode type)}
1129 * with {@code type} {@link IsoCountryCode#PART1_ALPHA2}.
1130 * <p>
1131 * <b>Note:</b> The <code>Locale</code> class also supports other codes for
1132 * country (region), such as 3-letter numeric UN M.49 area codes.
1133 * Therefore, the list returned by this method does not contain ALL valid
1134 * codes that can be used to create Locales.
1135 * <p>
1136 * Note that this method does not return obsolete 2-letter country codes.
1137 * ISO3166-3 codes which designate country codes for those obsolete codes,
1138 * can be retrieved from {@link #getISOCountries(Locale.IsoCountryCode type)} with
1139 * {@code type} {@link IsoCountryCode#PART3}.
1140 * @return An array of ISO 3166 two-letter country codes.
1141 */
1142 public static String[] getISOCountries() {
1143 if (isoCountries == null) {
1144 isoCountries = getISO2Table(LocaleISOData.isoCountryTable);
1145 }
1146 String[] result = new String[isoCountries.length];
1147 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length);
1148 return result;
1149 }
1150
1151 /**
1152 * Returns a {@code Set} of ISO3166 country codes for the specified type.
1153 *
1154 * @param type {@link Locale.IsoCountryCode} specified ISO code type.
1155 * @see java.util.Locale.IsoCountryCode
1156 * @throws NullPointerException if type is null
1157 * @return a {@code Set} of ISO country codes for the specified type.
1158 * @since 9
1159 */
1160 public static Set<String> getISOCountries(IsoCountryCode type) {
1161 Objects.requireNonNull(type);
1162 return IsoCountryCode.retrieveISOCountryCodes(type);
1163 }
1164
1165 /**
1166 * Returns a list of all 2-letter language codes defined in ISO 639.
1167 * Can be used to create Locales.
1168 * <p>
1169 * <b>Note:</b>
1170 * <ul>
1171 * <li>ISO 639 is not a stable standard— some languages' codes have changed.
1172 * The list this function returns includes both the new and the old codes for the
1173 * languages whose codes have changed.
1174 * <li>The <code>Locale</code> class also supports language codes up to
1175 * 8 characters in length. Therefore, the list returned by this method does
1176 * not contain ALL valid codes that can be used to create Locales.
1177 * </ul>
1178 *
1179 * @return An array of ISO 639 two-letter language codes.
1180 */
1181 public static String[] getISOLanguages() {
1182 if (isoLanguages == null) {
1183 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable);
1184 }
1185 String[] result = new String[isoLanguages.length];
1186 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length);
1187 return result;
1188 }
1189
1190 private static String[] getISO2Table(String table) {
1191 int len = table.length() / 5;
1192 String[] isoTable = new String[len];
1193 for (int i = 0, j = 0; i < len; i++, j += 5) {
1194 isoTable[i] = table.substring(j, j + 2);
1195 }
1196 return isoTable;
1197 }
1198
1199 /**
1200 * Returns the language code of this Locale.
1201 *
1202 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed.
1203 * Locale's constructor recognizes both the new and the old codes for the languages
1204 * whose codes have changed, but this function always returns the old code. If you
1205 * want to check for a specific language whose code has changed, don't do
1206 * <pre>
1207 * if (locale.getLanguage().equals("he")) // BAD!
1208 * ...
1209 * </pre>
1210 * Instead, do
1211 * <pre>
1212 * if (locale.getLanguage().equals(new Locale("he").getLanguage()))
1213 * ...
1214 * </pre>
1215 * @return The language code, or the empty string if none is defined.
1216 * @see #getDisplayLanguage
1217 */
1218 public String getLanguage() {
1219 return baseLocale.getLanguage();
1220 }
1221
1222 /**
1223 * Returns the script for this locale, which should
1224 * either be the empty string or an ISO 15924 4-letter script
1225 * code. The first letter is uppercase and the rest are
1226 * lowercase, for example, 'Latn', 'Cyrl'.
1227 *
1228 * @return The script code, or the empty string if none is defined.
1229 * @see #getDisplayScript
1230 * @since 1.7
1231 */
1232 public String getScript() {
1233 return baseLocale.getScript();
1234 }
1235
1236 /**
1237 * Returns the country/region code for this locale, which should
1238 * either be the empty string, an uppercase ISO 3166 2-letter code,
1239 * or a UN M.49 3-digit code.
1240 *
1241 * @return The country/region code, or the empty string if none is defined.
1242 * @see #getDisplayCountry
1243 */
1244 public String getCountry() {
1245 return baseLocale.getRegion();
1246 }
1247
1248 /**
1249 * Returns the variant code for this locale.
1250 *
1251 * @return The variant code, or the empty string if none is defined.
1252 * @see #getDisplayVariant
1253 */
1254 public String getVariant() {
1255 return baseLocale.getVariant();
1256 }
1257
1258 /**
1259 * Returns {@code true} if this {@code Locale} has any <a href="#def_extensions">
1260 * extensions</a>.
1261 *
1262 * @return {@code true} if this {@code Locale} has any extensions
1263 * @since 1.8
1264 */
1265 public boolean hasExtensions() {
1266 return localeExtensions != null;
1267 }
1268
1269 /**
1270 * Returns a copy of this {@code Locale} with no <a href="#def_extensions">
1271 * extensions</a>. If this {@code Locale} has no extensions, this {@code Locale}
1272 * is returned.
1273 *
1274 * @return a copy of this {@code Locale} with no extensions, or {@code this}
1275 * if {@code this} has no extensions
1276 * @since 1.8
1277 */
1278 public Locale stripExtensions() {
1279 return hasExtensions() ? Locale.getInstance(baseLocale, null) : this;
1280 }
1281
1282 /**
1283 * Returns the extension (or private use) value associated with
1284 * the specified key, or null if there is no extension
1285 * associated with the key. To be well-formed, the key must be one
1286 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so
1287 * for example 'z' and 'Z' represent the same extension.
1288 *
1289 * @param key the extension key
1290 * @return The extension, or null if this locale defines no
1291 * extension for the specified key.
1292 * @throws IllegalArgumentException if key is not well-formed
1293 * @see #PRIVATE_USE_EXTENSION
1294 * @see #UNICODE_LOCALE_EXTENSION
1295 * @since 1.7
1296 */
1297 public String getExtension(char key) {
1298 if (!LocaleExtensions.isValidKey(key)) {
1299 throw new IllegalArgumentException("Ill-formed extension key: " + key);
1300 }
1301 return hasExtensions() ? localeExtensions.getExtensionValue(key) : null;
1302 }
1303
1304 /**
1305 * Returns the set of extension keys associated with this locale, or the
1306 * empty set if it has no extensions. The returned set is unmodifiable.
1307 * The keys will all be lower-case.
1308 *
1309 * @return The set of extension keys, or the empty set if this locale has
1310 * no extensions.
1311 * @since 1.7
1312 */
1313 public Set<Character> getExtensionKeys() {
1314 if (!hasExtensions()) {
1315 return Collections.emptySet();
1316 }
1317 return localeExtensions.getKeys();
1318 }
1319
1320 /**
1321 * Returns the set of unicode locale attributes associated with
1322 * this locale, or the empty set if it has no attributes. The
1323 * returned set is unmodifiable.
1324 *
1325 * @return The set of attributes.
1326 * @since 1.7
1327 */
1328 public Set<String> getUnicodeLocaleAttributes() {
1329 if (!hasExtensions()) {
1330 return Collections.emptySet();
1331 }
1332 return localeExtensions.getUnicodeLocaleAttributes();
1333 }
1334
1335 /**
1336 * Returns the Unicode locale type associated with the specified Unicode locale key
1337 * for this locale. Returns the empty string for keys that are defined with no type.
1338 * Returns null if the key is not defined. Keys are case-insensitive. The key must
1339 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is
1340 * thrown.
1341 *
1342 * @param key the Unicode locale key
1343 * @return The Unicode locale type associated with the key, or null if the
1344 * locale does not define the key.
1345 * @throws IllegalArgumentException if the key is not well-formed
1346 * @throws NullPointerException if <code>key</code> is null
1347 * @since 1.7
1348 */
1349 public String getUnicodeLocaleType(String key) {
1350 if (!isUnicodeExtensionKey(key)) {
1351 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key);
1352 }
1353 return hasExtensions() ? localeExtensions.getUnicodeLocaleType(key) : null;
1354 }
1355
1356 /**
1357 * Returns the set of Unicode locale keys defined by this locale, or the empty set if
1358 * this locale has none. The returned set is immutable. Keys are all lower case.
1359 *
1360 * @return The set of Unicode locale keys, or the empty set if this locale has
1361 * no Unicode locale keywords.
1362 * @since 1.7
1363 */
1364 public Set<String> getUnicodeLocaleKeys() {
1365 if (localeExtensions == null) {
1366 return Collections.emptySet();
1367 }
1368 return localeExtensions.getUnicodeLocaleKeys();
1369 }
1370
1371 /**
1372 * Package locale method returning the Locale's BaseLocale,
1373 * used by ResourceBundle
1374 * @return base locale of this Locale
1375 */
1376 BaseLocale getBaseLocale() {
1377 return baseLocale;
1378 }
1379
1380 /**
1381 * Package private method returning the Locale's LocaleExtensions,
1382 * used by ResourceBundle.
1383 * @return locale extensions of this Locale,
1384 * or {@code null} if no extensions are defined
1385 */
1386 LocaleExtensions getLocaleExtensions() {
1387 return localeExtensions;
1388 }
1389
1390 /**
1391 * Returns a string representation of this <code>Locale</code>
1392 * object, consisting of language, country, variant, script,
1393 * and extensions as below:
1394 * <blockquote>
1395 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "_" + extensions
1396 * </blockquote>
1397 *
1398 * Language is always lower case, country is always upper case, script is always title
1399 * case, and extensions are always lower case. Extensions and private use subtags
1400 * will be in canonical order as explained in {@link #toLanguageTag}.
1401 *
1402 * <p>When the locale has neither script nor extensions, the result is the same as in
1403 * Java 6 and prior.
1404 *
1405 * <p>If both the language and country fields are missing, this function will return
1406 * the empty string, even if the variant, script, or extensions field is present (you
1407 * can't have a locale with just a variant, the variant must accompany a well-formed
1408 * language or country code).
1409 *
1410 * <p>If script or extensions are present and variant is missing, no underscore is
1411 * added before the "#".
1412 *
1413 * <p>This behavior is designed to support debugging and to be compatible with
1414 * previous uses of <code>toString</code> that expected language, country, and variant
1415 * fields only. To represent a Locale as a String for interchange purposes, use
1416 * {@link #toLanguageTag}.
1417 *
1418 * <p>Examples: <ul>
1419 * <li>{@code en}</li>
1420 * <li>{@code de_DE}</li>
1421 * <li>{@code _GB}</li>
1422 * <li>{@code en_US_WIN}</li>
1423 * <li>{@code de__POSIX}</li>
1424 * <li>{@code zh_CN_#Hans}</li>
1425 * <li>{@code zh_TW_#Hant_x-java}</li>
1426 * <li>{@code th_TH_TH_#u-nu-thai}</li></ul>
1427 *
1428 * @return A string representation of the Locale, for debugging.
1429 * @see #getDisplayName
1430 * @see #toLanguageTag
1431 */
1432 @Override
1433 public final String toString() {
1434 boolean l = !baseLocale.getLanguage().isEmpty();
1435 boolean s = !baseLocale.getScript().isEmpty();
1436 boolean r = !baseLocale.getRegion().isEmpty();
1437 boolean v = !baseLocale.getVariant().isEmpty();
1438 boolean e = localeExtensions != null && !localeExtensions.getID().isEmpty();
1439
1440 StringBuilder result = new StringBuilder(baseLocale.getLanguage());
1441 if (r || (l && (v || s || e))) {
1442 result.append('_')
1443 .append(baseLocale.getRegion()); // This may just append '_'
1444 }
1445 if (v && (l || r)) {
1446 result.append('_')
1447 .append(baseLocale.getVariant());
1448 }
1449
1450 if (s && (l || r)) {
1451 result.append("_#")
1452 .append(baseLocale.getScript());
1453 }
1454
1455 if (e && (l || r)) {
1456 result.append('_');
1457 if (!s) {
1458 result.append('#');
1459 }
1460 result.append(localeExtensions.getID());
1461 }
1462
1463 return result.toString();
1464 }
1465
1466 /**
1467 * Returns a well-formed IETF BCP 47 language tag representing
1468 * this locale.
1469 *
1470 * <p>If this <code>Locale</code> has a language, country, or
1471 * variant that does not satisfy the IETF BCP 47 language tag
1472 * syntax requirements, this method handles these fields as
1473 * described below:
1474 *
1475 * <p><b>Language:</b> If language is empty, or not <a
1476 * href="#def_language" >well-formed</a> (for example "a" or
1477 * "e2"), it will be emitted as "und" (Undetermined).
1478 *
1479 * <p><b>Country:</b> If country is not <a
1480 * href="#def_region">well-formed</a> (for example "12" or "USA"),
1481 * it will be omitted.
1482 *
1483 * <p><b>Variant:</b> If variant <b>is</b> <a
1484 * href="#def_variant">well-formed</a>, each sub-segment
1485 * (delimited by '-' or '_') is emitted as a subtag. Otherwise:
1486 * <ul>
1487 *
1488 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code>
1489 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first
1490 * ill-formed sub-segment and all following will be appended to
1491 * the private use subtag. The first appended subtag will be
1492 * "lvariant", followed by the sub-segments in order, separated by
1493 * hyphen. For example, "x-lvariant-WIN",
1494 * "Oracle-x-lvariant-JDK-Standard-Edition".
1495 *
1496 * <li>if any sub-segment does not match
1497 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated
1498 * and the problematic sub-segment and all following sub-segments
1499 * will be omitted. If the remainder is non-empty, it will be
1500 * emitted as a private use subtag as above (even if the remainder
1501 * turns out to be well-formed). For example,
1502 * "Solaris_isjustthecoolestthing" is emitted as
1503 * "x-lvariant-Solaris", not as "solaris".</li></ul>
1504 *
1505 * <p><b>Special Conversions:</b> Java supports some old locale
1506 * representations, including deprecated ISO language codes,
1507 * for compatibility. This method performs the following
1508 * conversions:
1509 * <ul>
1510 *
1511 * <li>Deprecated ISO language codes "iw", "ji", and "in" are
1512 * converted to "he", "yi", and "id", respectively.
1513 *
1514 * <li>A locale with language "no", country "NO", and variant
1515 * "NY", representing Norwegian Nynorsk (Norway), is converted
1516 * to a language tag "nn-NO".</li></ul>
1517 *
1518 * <p><b>Note:</b> Although the language tag created by this
1519 * method is well-formed (satisfies the syntax requirements
1520 * defined by the IETF BCP 47 specification), it is not
1521 * necessarily a valid BCP 47 language tag. For example,
1522 * <pre>
1523 * new Locale("xx", "YY").toLanguageTag();</pre>
1524 *
1525 * will return "xx-YY", but the language subtag "xx" and the
1526 * region subtag "YY" are invalid because they are not registered
1527 * in the IANA Language Subtag Registry.
1528 *
1529 * @return a BCP47 language tag representing the locale
1530 * @see #forLanguageTag(String)
1531 * @since 1.7
1532 */
1533 public String toLanguageTag() {
1534 if (languageTag != null) {
1535 return languageTag;
1536 }
1537
1538 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions);
1539 StringBuilder buf = new StringBuilder();
1540
1541 String subtag = tag.getLanguage();
1542 if (!subtag.isEmpty()) {
1543 buf.append(LanguageTag.canonicalizeLanguage(subtag));
1544 }
1545
1546 subtag = tag.getScript();
1547 if (!subtag.isEmpty()) {
1548 buf.append(LanguageTag.SEP);
1549 buf.append(LanguageTag.canonicalizeScript(subtag));
1550 }
1551
1552 subtag = tag.getRegion();
1553 if (!subtag.isEmpty()) {
1554 buf.append(LanguageTag.SEP);
1555 buf.append(LanguageTag.canonicalizeRegion(subtag));
1556 }
1557
1558 List<String>subtags = tag.getVariants();
1559 for (String s : subtags) {
1560 buf.append(LanguageTag.SEP);
1561 // preserve casing
1562 buf.append(s);
1563 }
1564
1565 subtags = tag.getExtensions();
1566 for (String s : subtags) {
1567 buf.append(LanguageTag.SEP);
1568 buf.append(LanguageTag.canonicalizeExtension(s));
1569 }
1570
1571 subtag = tag.getPrivateuse();
1572 if (!subtag.isEmpty()) {
1573 if (buf.length() > 0) {
1574 buf.append(LanguageTag.SEP);
1575 }
1576 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP);
1577 // preserve casing
1578 buf.append(subtag);
1579 }
1580
1581 String langTag = buf.toString();
1582 synchronized (this) {
1583 if (languageTag == null) {
1584 languageTag = langTag;
1585 }
1586 }
1587 return languageTag;
1588 }
1589
1590 /**
1591 * Returns a locale for the specified IETF BCP 47 language tag string.
1592 *
1593 * <p>If the specified language tag contains any ill-formed subtags,
1594 * the first such subtag and all following subtags are ignored. Compare
1595 * to {@link Locale.Builder#setLanguageTag} which throws an exception
1596 * in this case.
1597 *
1598 * <p>The following <b>conversions</b> are performed:<ul>
1599 *
1600 * <li>The language code "und" is mapped to language "".
1601 *
1602 * <li>The language codes "he", "yi", and "id" are mapped to "iw",
1603 * "ji", and "in" respectively. (This is the same canonicalization
1604 * that's done in Locale's constructors.)
1605 *
1606 * <li>The portion of a private use subtag prefixed by "lvariant",
1607 * if any, is removed and appended to the variant field in the
1608 * result locale (without case normalization). If it is then
1609 * empty, the private use subtag is discarded:
1610 *
1611 * <pre>
1612 * Locale loc;
1613 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX");
1614 * loc.getVariant(); // returns "POSIX"
1615 * loc.getExtension('x'); // returns null
1616 *
1617 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def");
1618 * loc.getVariant(); // returns "POSIX_Abc_Def"
1619 * loc.getExtension('x'); // returns "urp"
1620 * </pre>
1621 *
1622 * <li>When the languageTag argument contains an extlang subtag,
1623 * the first such subtag is used as the language, and the primary
1624 * language subtag and other extlang subtags are ignored:
1625 *
1626 * <pre>
1627 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao"
1628 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US"
1629 * </pre>
1630 *
1631 * <li>Case is normalized except for variant tags, which are left
1632 * unchanged. Language is normalized to lower case, script to
1633 * title case, country to upper case, and extensions to lower
1634 * case.
1635 *
1636 * <li>If, after processing, the locale would exactly match either
1637 * ja_JP_JP or th_TH_TH with no extensions, the appropriate
1638 * extensions are added as though the constructor had been called:
1639 *
1640 * <pre>
1641 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag();
1642 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP"
1643 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag();
1644 * // returns "th-TH-u-nu-thai-x-lvariant-TH"
1645 * </pre></ul>
1646 *
1647 * <p>This implements the 'Language-Tag' production of BCP47, and
1648 * so supports grandfathered (regular and irregular) as well as
1649 * private use language tags. Stand alone private use tags are
1650 * represented as empty language and extension 'x-whatever',
1651 * and grandfathered tags are converted to their canonical replacements
1652 * where they exist.
1653 *
1654 * <p>Grandfathered tags with canonical replacements are as follows:
1655 *
1656 * <table class="striped">
1657 * <caption style="display:none">Grandfathered tags with canonical replacements</caption>
1658 * <thead style="text-align:center">
1659 * <tr><th scope="col" style="padding: 0 2px">grandfathered tag</th><th scope="col" style="padding: 0 2px">modern replacement</th></tr>
1660 * </thead>
1661 * <tbody style="text-align:center">
1662 * <tr><th scope="row">art-lojban</th><td>jbo</td></tr>
1663 * <tr><th scope="row">i-ami</th><td>ami</td></tr>
1664 * <tr><th scope="row">i-bnn</th><td>bnn</td></tr>
1665 * <tr><th scope="row">i-hak</th><td>hak</td></tr>
1666 * <tr><th scope="row">i-klingon</th><td>tlh</td></tr>
1667 * <tr><th scope="row">i-lux</th><td>lb</td></tr>
1668 * <tr><th scope="row">i-navajo</th><td>nv</td></tr>
1669 * <tr><th scope="row">i-pwn</th><td>pwn</td></tr>
1670 * <tr><th scope="row">i-tao</th><td>tao</td></tr>
1671 * <tr><th scope="row">i-tay</th><td>tay</td></tr>
1672 * <tr><th scope="row">i-tsu</th><td>tsu</td></tr>
1673 * <tr><th scope="row">no-bok</th><td>nb</td></tr>
1674 * <tr><th scope="row">no-nyn</th><td>nn</td></tr>
1675 * <tr><th scope="row">sgn-BE-FR</th><td>sfb</td></tr>
1676 * <tr><th scope="row">sgn-BE-NL</th><td>vgt</td></tr>
1677 * <tr><th scope="row">sgn-CH-DE</th><td>sgg</td></tr>
1678 * <tr><th scope="row">zh-guoyu</th><td>cmn</td></tr>
1679 * <tr><th scope="row">zh-hakka</th><td>hak</td></tr>
1680 * <tr><th scope="row">zh-min-nan</th><td>nan</td></tr>
1681 * <tr><th scope="row">zh-xiang</th><td>hsn</td></tr>
1682 * </tbody>
1683 * </table>
1684 *
1685 * <p>Grandfathered tags with no modern replacement will be
1686 * converted as follows:
1687 *
1688 * <table class="striped">
1689 * <caption style="display:none">Grandfathered tags with no modern replacement</caption>
1690 * <thead style="text-align:center">
1691 * <tr><th scope="col" style="padding: 0 2px">grandfathered tag</th><th scope="col" style="padding: 0 2px">converts to</th></tr>
1692 * </thead>
1693 * <tbody style="text-align:center">
1694 * <tr><th scope="row">cel-gaulish</th><td>xtg-x-cel-gaulish</td></tr>
1695 * <tr><th scope="row">en-GB-oed</th><td>en-GB-x-oed</td></tr>
1696 * <tr><th scope="row">i-default</th><td>en-x-i-default</td></tr>
1697 * <tr><th scope="row">i-enochian</th><td>und-x-i-enochian</td></tr>
1698 * <tr><th scope="row">i-mingo</th><td>see-x-i-mingo</td></tr>
1699 * <tr><th scope="row">zh-min</th><td>nan-x-zh-min</td></tr>
1700 * </tbody>
1701 * </table>
1702 *
1703 * <p>For a list of all grandfathered tags, see the
1704 * IANA Language Subtag Registry (search for "Type: grandfathered").
1705 *
1706 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code>
1707 * and <code>forLanguageTag</code> will round-trip.
1708 *
1709 * @param languageTag the language tag
1710 * @return The locale that best represents the language tag.
1711 * @throws NullPointerException if <code>languageTag</code> is <code>null</code>
1712 * @see #toLanguageTag()
1713 * @see java.util.Locale.Builder#setLanguageTag(String)
1714 * @since 1.7
1715 */
1716 public static Locale forLanguageTag(String languageTag) {
1717 LanguageTag tag = LanguageTag.parse(languageTag, null);
1718 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
1719 bldr.setLanguageTag(tag);
1720 BaseLocale base = bldr.getBaseLocale();
1721 LocaleExtensions exts = bldr.getLocaleExtensions();
1722 if (exts == null && !base.getVariant().isEmpty()) {
1723 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(),
1724 base.getRegion(), base.getVariant());
1725 }
1726 return getInstance(base, exts);
1727 }
1728
1729 /**
1730 * Returns a three-letter abbreviation of this locale's language.
1731 * If the language matches an ISO 639-1 two-letter code, the
1732 * corresponding ISO 639-2/T three-letter lowercase code is
1733 * returned. The ISO 639-2 language codes can be found on-line,
1734 * see "Codes for the Representation of Names of Languages Part 2:
1735 * Alpha-3 Code". If the locale specifies a three-letter
1736 * language, the language is returned as is. If the locale does
1737 * not specify a language the empty string is returned.
1738 *
1739 * @return A three-letter abbreviation of this locale's language.
1740 * @exception MissingResourceException Throws MissingResourceException if
1741 * three-letter language abbreviation is not available for this locale.
1742 */
1743 public String getISO3Language() throws MissingResourceException {
1744 String lang = baseLocale.getLanguage();
1745 if (lang.length() == 3) {
1746 return lang;
1747 }
1748
1749 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable);
1750 if (language3 == null) {
1751 throw new MissingResourceException("Couldn't find 3-letter language code for "
1752 + lang, "FormatData_" + toString(), "ShortLanguage");
1753 }
1754 return language3;
1755 }
1756
1757 /**
1758 * Returns a three-letter abbreviation for this locale's country.
1759 * If the country matches an ISO 3166-1 alpha-2 code, the
1760 * corresponding ISO 3166-1 alpha-3 uppercase code is returned.
1761 * If the locale doesn't specify a country, this will be the empty
1762 * string.
1763 *
1764 * <p>The ISO 3166-1 codes can be found on-line.
1765 *
1766 * @return A three-letter abbreviation of this locale's country.
1767 * @exception MissingResourceException Throws MissingResourceException if the
1768 * three-letter country abbreviation is not available for this locale.
1769 */
1770 public String getISO3Country() throws MissingResourceException {
1771 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable);
1772 if (country3 == null) {
1773 throw new MissingResourceException("Couldn't find 3-letter country code for "
1774 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry");
1775 }
1776 return country3;
1777 }
1778
1779 private static String getISO3Code(String iso2Code, String table) {
1780 int codeLength = iso2Code.length();
1781 if (codeLength == 0) {
1782 return "";
1783 }
1784
1785 int tableLength = table.length();
1786 int index = tableLength;
1787 if (codeLength == 2) {
1788 char c1 = iso2Code.charAt(0);
1789 char c2 = iso2Code.charAt(1);
1790 for (index = 0; index < tableLength; index += 5) {
1791 if (table.charAt(index) == c1
1792 && table.charAt(index + 1) == c2) {
1793 break;
1794 }
1795 }
1796 }
1797 return index < tableLength ? table.substring(index + 2, index + 5) : null;
1798 }
1799
1800 /**
1801 * Returns a name for the locale's language that is appropriate for display to the
1802 * user.
1803 * If possible, the name returned will be localized for the default
1804 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1805 * For example, if the locale is fr_FR and the default
1806 * {@link Locale.Category#DISPLAY DISPLAY} locale
1807 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1808 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1809 * getDisplayLanguage() will return "anglais".
1810 * If the name returned cannot be localized for the default
1811 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1812 * (say, we don't have a Japanese name for Croatian),
1813 * this function falls back on the English name, and uses the ISO code as a last-resort
1814 * value. If the locale doesn't specify a language, this function returns the empty string.
1815 *
1816 * @return The name of the display language.
1817 */
1818 public final String getDisplayLanguage() {
1819 return getDisplayLanguage(getDefault(Category.DISPLAY));
1820 }
1821
1822 /**
1823 * Returns a name for the locale's language that is appropriate for display to the
1824 * user.
1825 * If possible, the name returned will be localized according to inLocale.
1826 * For example, if the locale is fr_FR and inLocale
1827 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1828 * inLocale is fr_FR, getDisplayLanguage() will return "anglais".
1829 * If the name returned cannot be localized according to inLocale,
1830 * (say, we don't have a Japanese name for Croatian),
1831 * this function falls back on the English name, and finally
1832 * on the ISO code as a last-resort value. If the locale doesn't specify a language,
1833 * this function returns the empty string.
1834 *
1835 * @param inLocale The locale for which to retrieve the display language.
1836 * @return The name of the display language appropriate to the given locale.
1837 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1838 */
1839 public String getDisplayLanguage(Locale inLocale) {
1840 return getDisplayString(baseLocale.getLanguage(), null, inLocale, DISPLAY_LANGUAGE);
1841 }
1842
1843 /**
1844 * Returns a name for the locale's script that is appropriate for display to
1845 * the user. If possible, the name will be localized for the default
1846 * {@link Locale.Category#DISPLAY DISPLAY} locale. Returns
1847 * the empty string if this locale doesn't specify a script code.
1848 *
1849 * @return the display name of the script code for the current default
1850 * {@link Locale.Category#DISPLAY DISPLAY} locale
1851 * @since 1.7
1852 */
1853 public String getDisplayScript() {
1854 return getDisplayScript(getDefault(Category.DISPLAY));
1855 }
1856
1857 /**
1858 * Returns a name for the locale's script that is appropriate
1859 * for display to the user. If possible, the name will be
1860 * localized for the given locale. Returns the empty string if
1861 * this locale doesn't specify a script code.
1862 *
1863 * @param inLocale The locale for which to retrieve the display script.
1864 * @return the display name of the script code for the current default
1865 * {@link Locale.Category#DISPLAY DISPLAY} locale
1866 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1867 * @since 1.7
1868 */
1869 public String getDisplayScript(Locale inLocale) {
1870 return getDisplayString(baseLocale.getScript(), null, inLocale, DISPLAY_SCRIPT);
1871 }
1872
1873 /**
1874 * Returns a name for the locale's country that is appropriate for display to the
1875 * user.
1876 * If possible, the name returned will be localized for the default
1877 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1878 * For example, if the locale is fr_FR and the default
1879 * {@link Locale.Category#DISPLAY DISPLAY} locale
1880 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1881 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1882 * getDisplayCountry() will return "Etats-Unis".
1883 * If the name returned cannot be localized for the default
1884 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1885 * (say, we don't have a Japanese name for Croatia),
1886 * this function falls back on the English name, and uses the ISO code as a last-resort
1887 * value. If the locale doesn't specify a country, this function returns the empty string.
1888 *
1889 * @return The name of the country appropriate to the locale.
1890 */
1891 public final String getDisplayCountry() {
1892 return getDisplayCountry(getDefault(Category.DISPLAY));
1893 }
1894
1895 /**
1896 * Returns a name for the locale's country that is appropriate for display to the
1897 * user.
1898 * If possible, the name returned will be localized according to inLocale.
1899 * For example, if the locale is fr_FR and inLocale
1900 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1901 * inLocale is fr_FR, getDisplayCountry() will return "Etats-Unis".
1902 * If the name returned cannot be localized according to inLocale.
1903 * (say, we don't have a Japanese name for Croatia),
1904 * this function falls back on the English name, and finally
1905 * on the ISO code as a last-resort value. If the locale doesn't specify a country,
1906 * this function returns the empty string.
1907 *
1908 * @param inLocale The locale for which to retrieve the display country.
1909 * @return The name of the country appropriate to the given locale.
1910 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1911 */
1912 public String getDisplayCountry(Locale inLocale) {
1913 return getDisplayString(baseLocale.getRegion(), null, inLocale, DISPLAY_COUNTRY);
1914 }
1915
1916 private String getDisplayString(String code, String cat, Locale inLocale, int type) {
1917 Objects.requireNonNull(inLocale);
1918 Objects.requireNonNull(code);
1919
1920 if (code.isEmpty()) {
1921 return "";
1922 }
1923
1924 LocaleServiceProviderPool pool =
1925 LocaleServiceProviderPool.getPool(LocaleNameProvider.class);
1926 String rbKey = (type == DISPLAY_VARIANT ? "%%"+code : code);
1927 String result = pool.getLocalizedObject(
1928 LocaleNameGetter.INSTANCE,
1929 inLocale, rbKey, type, code, cat);
1930 return result != null ? result : code;
1931 }
1932
1933 /**
1934 * Returns a name for the locale's variant code that is appropriate for display to the
1935 * user. If possible, the name will be localized for the default
1936 * {@link Locale.Category#DISPLAY DISPLAY} locale. If the locale
1937 * doesn't specify a variant code, this function returns the empty string.
1938 *
1939 * @return The name of the display variant code appropriate to the locale.
1940 */
1941 public final String getDisplayVariant() {
1942 return getDisplayVariant(getDefault(Category.DISPLAY));
1943 }
1944
1945 /**
1946 * Returns a name for the locale's variant code that is appropriate for display to the
1947 * user. If possible, the name will be localized for inLocale. If the locale
1948 * doesn't specify a variant code, this function returns the empty string.
1949 *
1950 * @param inLocale The locale for which to retrieve the display variant code.
1951 * @return The name of the display variant code appropriate to the given locale.
1952 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1953 */
1954 public String getDisplayVariant(Locale inLocale) {
1955 if (baseLocale.getVariant().isEmpty())
1956 return "";
1957
1958 LocaleResources lr = LocaleProviderAdapter
1959 .getResourceBundleBased()
1960 .getLocaleResources(inLocale);
1961
1962 String names[] = getDisplayVariantArray(inLocale);
1963
1964 // Get the localized patterns for formatting a list, and use
1965 // them to format the list.
1966 return formatList(names,
1967 lr.getLocaleName("ListCompositionPattern"));
1968 }
1969
1970 /**
1971 * Returns a name for the locale that is appropriate for display to the
1972 * user. This will be the values returned by getDisplayLanguage(),
1973 * getDisplayScript(), getDisplayCountry(), getDisplayVariant() and
1974 * optional <a href="./Locale.html#def_locale_extension">Unicode extensions</a>
1975 * assembled into a single string. The non-empty values are used in order, with
1976 * the second and subsequent names in parentheses. For example:
1977 * <blockquote>
1978 * language (script, country, variant(, extension)*)<br>
1979 * language (country(, extension)*)<br>
1980 * language (variant(, extension)*)<br>
1981 * script (country(, extension)*)<br>
1982 * country (extension)*<br>
1983 * </blockquote>
1984 * depending on which fields are specified in the locale. The field
1985 * separator in the above parentheses, denoted as a comma character, may
1986 * be localized depending on the locale. If the language, script, country,
1987 * and variant fields are all empty, this function returns the empty string.
1988 *
1989 * @return The name of the locale appropriate to display.
1990 */
1991 public final String getDisplayName() {
1992 return getDisplayName(getDefault(Category.DISPLAY));
1993 }
1994
1995 /**
1996 * Returns a name for the locale that is appropriate for display
1997 * to the user. This will be the values returned by
1998 * getDisplayLanguage(), getDisplayScript(),getDisplayCountry()
1999 * getDisplayVariant(), and optional <a href="./Locale.html#def_locale_extension">
2000 * Unicode extensions</a> assembled into a single string. The non-empty
2001 * values are used in order, with the second and subsequent names in
2002 * parentheses. For example:
2003 * <blockquote>
2004 * language (script, country, variant(, extension)*)<br>
2005 * language (country(, extension)*)<br>
2006 * language (variant(, extension)*)<br>
2007 * script (country(, extension)*)<br>
2008 * country (extension)*<br>
2009 * </blockquote>
2010 * depending on which fields are specified in the locale. The field
2011 * separator in the above parentheses, denoted as a comma character, may
2012 * be localized depending on the locale. If the language, script, country,
2013 * and variant fields are all empty, this function returns the empty string.
2014 *
2015 * @param inLocale The locale for which to retrieve the display name.
2016 * @return The name of the locale appropriate to display.
2017 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
2018 */
2019 public String getDisplayName(Locale inLocale) {
2020 LocaleResources lr = LocaleProviderAdapter
2021 .getResourceBundleBased()
2022 .getLocaleResources(inLocale);
2023
2024 String languageName = getDisplayLanguage(inLocale);
2025 String scriptName = getDisplayScript(inLocale);
2026 String countryName = getDisplayCountry(inLocale);
2027 String[] variantNames = getDisplayVariantArray(inLocale);
2028
2029 // Get the localized patterns for formatting a display name.
2030 String displayNamePattern = lr.getLocaleName("DisplayNamePattern");
2031 String listCompositionPattern = lr.getLocaleName("ListCompositionPattern");
2032
2033 // The display name consists of a main name, followed by qualifiers.
2034 // Typically, the format is "MainName (Qualifier, Qualifier)" but this
2035 // depends on what pattern is stored in the display locale.
2036 String mainName;
2037 String[] qualifierNames;
2038
2039 // The main name is the language, or if there is no language, the script,
2040 // then if no script, the country. If there is no language/script/country
2041 // (an anomalous situation) then the display name is simply the variant's
2042 // display name.
2043 if (languageName.isEmpty() && scriptName.isEmpty() && countryName.isEmpty()) {
2044 if (variantNames.length == 0) {
2045 return "";
2046 } else {
2047 return formatList(variantNames, listCompositionPattern);
2048 }
2049 }
2050 ArrayList<String> names = new ArrayList<>(4);
2051 if (!languageName.isEmpty()) {
2052 names.add(languageName);
2053 }
2054 if (!scriptName.isEmpty()) {
2055 names.add(scriptName);
2056 }
2057 if (!countryName.isEmpty()) {
2058 names.add(countryName);
2059 }
2060 if (variantNames.length != 0) {
2061 names.addAll(Arrays.asList(variantNames));
2062 }
2063
2064 // add Unicode extensions
2065 if (localeExtensions != null) {
2066 localeExtensions.getUnicodeLocaleAttributes().stream()
2067 .map(key -> getDisplayString(key, null, inLocale, DISPLAY_UEXT_KEY))
2068 .forEach(names::add);
2069 localeExtensions.getUnicodeLocaleKeys().stream()
2070 .map(key -> getDisplayKeyTypeExtensionString(key, lr, inLocale))
2071 .forEach(names::add);
2072 }
2073
2074 // The first one in the main name
2075 mainName = names.get(0);
2076
2077 // Others are qualifiers
2078 int numNames = names.size();
2079 qualifierNames = (numNames > 1) ?
2080 names.subList(1, numNames).toArray(new String[numNames - 1]) : new String[0];
2081
2082 // Create an array whose first element is the number of remaining
2083 // elements. This serves as a selector into a ChoiceFormat pattern from
2084 // the resource. The second and third elements are the main name and
2085 // the qualifier; if there are no qualifiers, the third element is
2086 // unused by the format pattern.
2087 Object[] displayNames = {
2088 qualifierNames.length != 0 ? 2 : 1,
2089 mainName,
2090 // We could also just call formatList() and have it handle the empty
2091 // list case, but this is more efficient, and we want it to be
2092 // efficient since all the language-only locales will not have any
2093 // qualifiers.
2094 qualifierNames.length != 0 ? formatList(qualifierNames, listCompositionPattern) : null
2095 };
2096
2097 if (displayNamePattern != null) {
2098 return new MessageFormat(displayNamePattern).format(displayNames);
2099 }
2100 else {
2101 // If we cannot get the message format pattern, then we use a simple
2102 // hard-coded pattern. This should not occur in practice unless the
2103 // installation is missing some core files (FormatData etc.).
2104 StringBuilder result = new StringBuilder();
2105 result.append((String)displayNames[1]);
2106 if (displayNames.length > 2) {
2107 result.append(" (");
2108 result.append((String)displayNames[2]);
2109 result.append(')');
2110 }
2111 return result.toString();
2112 }
2113 }
2114
2115 /**
2116 * Overrides Cloneable.
2117 */
2118 @Override
2119 public Object clone()
2120 {
2121 try {
2122 Locale that = (Locale)super.clone();
2123 return that;
2124 } catch (CloneNotSupportedException e) {
2125 throw new InternalError(e);
2126 }
2127 }
2128
2129 /**
2130 * Override hashCode.
2131 * Since Locales are often used in hashtables, caches the value
2132 * for speed.
2133 */
2134 @Override
2135 public int hashCode() {
2136 int hc = hashCodeValue;
2137 if (hc == 0) {
2138 hc = baseLocale.hashCode();
2139 if (localeExtensions != null) {
2140 hc ^= localeExtensions.hashCode();
2141 }
2142 hashCodeValue = hc;
2143 }
2144 return hc;
2145 }
2146
2147 // Overrides
2148
2149 /**
2150 * Returns true if this Locale is equal to another object. A Locale is
2151 * deemed equal to another Locale with identical language, script, country,
2152 * variant and extensions, and unequal to all other objects.
2153 *
2154 * @return true if this Locale is equal to the specified object.
2155 */
2156 @Override
2157 public boolean equals(Object obj) {
2158 if (this == obj) // quick check
2159 return true;
2160 if (!(obj instanceof Locale))
2161 return false;
2162 BaseLocale otherBase = ((Locale)obj).baseLocale;
2163 if (!baseLocale.equals(otherBase)) {
2164 return false;
2165 }
2166 if (localeExtensions == null) {
2167 return ((Locale)obj).localeExtensions == null;
2168 }
2169 return localeExtensions.equals(((Locale)obj).localeExtensions);
2170 }
2171
2172 // ================= privates =====================================
2173
2174 private transient BaseLocale baseLocale;
2175 private transient LocaleExtensions localeExtensions;
2176
2177 /**
2178 * Calculated hashcode
2179 */
2180 private transient volatile int hashCodeValue;
2181
2182 private static volatile Locale defaultLocale = initDefault();
2183 private static volatile Locale defaultDisplayLocale;
2184 private static volatile Locale defaultFormatLocale;
2185
2186 private transient volatile String languageTag;
2187
2188 /**
2189 * Return an array of the display names of the variant.
2190 * @param bundle the ResourceBundle to use to get the display names
2191 * @return an array of display names, possible of zero length.
2192 */
2193 private String[] getDisplayVariantArray(Locale inLocale) {
2194 // Split the variant name into tokens separated by '_'.
2195 StringTokenizer tokenizer = new StringTokenizer(baseLocale.getVariant(), "_");
2196 String[] names = new String[tokenizer.countTokens()];
2197
2198 // For each variant token, lookup the display name. If
2199 // not found, use the variant name itself.
2200 for (int i=0; i<names.length; ++i) {
2201 names[i] = getDisplayString(tokenizer.nextToken(), null,
2202 inLocale, DISPLAY_VARIANT);
2203 }
2204
2205 return names;
2206 }
2207
2208 private String getDisplayKeyTypeExtensionString(String key, LocaleResources lr, Locale inLocale) {
2209 String type = localeExtensions.getUnicodeLocaleType(key);
2210 String ret = getDisplayString(type, key, inLocale, DISPLAY_UEXT_TYPE);
2211
2212 if (ret == null || ret.equals(type)) {
2213 // no localization for this type. try combining key/type separately
2214 String displayType = type;
2215 switch (key) {
2216 case "cu":
2217 displayType = lr.getCurrencyName(type.toLowerCase(Locale.ROOT));
2218 break;
2219 case "rg":
2220 if (type != null &&
2221 // UN M.49 code should not be allowed here
2222 type.matches("^[a-zA-Z]{2}[zZ]{4}$")) {
2223 displayType = lr.getLocaleName(type.substring(0, 2).toUpperCase(Locale.ROOT));
2224 }
2225 break;
2226 case "tz":
2227 displayType = TimeZoneNameUtility.convertLDMLShortID(type)
2228 .map(id -> TimeZoneNameUtility.retrieveGenericDisplayName(id, TimeZone.LONG, inLocale))
2229 .orElse(type);
2230 break;
2231 }
2232 ret = MessageFormat.format(lr.getLocaleName("ListKeyTypePattern"),
2233 getDisplayString(key, null, inLocale, DISPLAY_UEXT_KEY),
2234 Optional.ofNullable(displayType).orElse(type));
2235 }
2236
2237 return ret;
2238 }
2239
2240 /**
2241 * Format a list using given pattern strings.
2242 * If either of the patterns is null, then a the list is
2243 * formatted by concatenation with the delimiter ','.
2244 * @param stringList the list of strings to be formatted.
2245 * and formatting them into a list.
2246 * @param pattern should take 2 arguments for reduction
2247 * @return a string representing the list.
2248 */
2249 private static String formatList(String[] stringList, String pattern) {
2250 // If we have no list patterns, compose the list in a simple,
2251 // non-localized way.
2252 if (pattern == null) {
2253 return Arrays.stream(stringList).collect(Collectors.joining(","));
2254 }
2255
2256 switch (stringList.length) {
2257 case 0:
2258 return "";
2259 case 1:
2260 return stringList[0];
2261 default:
2262 return Arrays.stream(stringList).reduce("",
2263 (s1, s2) -> {
2264 if (s1.isEmpty()) {
2265 return s2;
2266 }
2267 if (s2.isEmpty()) {
2268 return s1;
2269 }
2270 return MessageFormat.format(pattern, s1, s2);
2271 });
2272 }
2273 }
2274
2275 // Duplicate of sun.util.locale.UnicodeLocaleExtension.isKey in order to
2276 // avoid its class loading.
2277 private static boolean isUnicodeExtensionKey(String s) {
2278 // 2alphanum
2279 return (s.length() == 2) && LocaleUtils.isAlphaNumericString(s);
2280 }
2281
2282 /**
2283 * @serialField language String
2284 * language subtag in lower case.
2285 * (See <a href="java.base/java/util/Locale.html#getLanguage()">getLanguage()</a>)
2286 * @serialField country String
2287 * country subtag in upper case.
2288 * (See <a href="java.base/java/util/Locale.html#getCountry()">getCountry()</a>)
2289 * @serialField variant String
2290 * variant subtags separated by LOWLINE characters.
2291 * (See <a href="java.base/java/util/Locale.html#getVariant()">getVariant()</a>)
2292 * @serialField hashcode int
2293 * deprecated, for forward compatibility only
2294 * @serialField script String
2295 * script subtag in title case
2296 * (See <a href="java.base/java/util/Locale.html#getScript()">getScript()</a>)
2297 * @serialField extensions String
2298 * canonical representation of extensions, that is,
2299 * BCP47 extensions in alphabetical order followed by
2300 * BCP47 private use subtags, all in lower case letters
2301 * separated by HYPHEN-MINUS characters.
2302 * (See <a href="java.base/java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>,
2303 * <a href="java.base/java/util/Locale.html#getExtension(char)">getExtension(char)</a>)
2304 */
2305 @java.io.Serial
2306 private static final ObjectStreamField[] serialPersistentFields = {
2307 new ObjectStreamField("language", String.class),
2308 new ObjectStreamField("country", String.class),
2309 new ObjectStreamField("variant", String.class),
2310 new ObjectStreamField("hashcode", int.class),
2311 new ObjectStreamField("script", String.class),
2312 new ObjectStreamField("extensions", String.class),
2313 };
2314
2315 /**
2316 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>.
2317 * @param out the <code>ObjectOutputStream</code> to write
2318 * @throws IOException
2319 * @since 1.7
2320 */
2321 @java.io.Serial
2322 private void writeObject(ObjectOutputStream out) throws IOException {
2323 ObjectOutputStream.PutField fields = out.putFields();
2324 fields.put("language", baseLocale.getLanguage());
2325 fields.put("script", baseLocale.getScript());
2326 fields.put("country", baseLocale.getRegion());
2327 fields.put("variant", baseLocale.getVariant());
2328 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID());
2329 fields.put("hashcode", -1); // place holder just for backward support
2330 out.writeFields();
2331 }
2332
2333 /**
2334 * Deserializes this <code>Locale</code>.
2335 * @param in the <code>ObjectInputStream</code> to read
2336 * @throws IOException
2337 * @throws ClassNotFoundException
2338 * @throws IllformedLocaleException
2339 * @since 1.7
2340 */
2341 @java.io.Serial
2342 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException {
2343 ObjectInputStream.GetField fields = in.readFields();
2344 String language = (String)fields.get("language", "");
2345 String script = (String)fields.get("script", "");
2346 String country = (String)fields.get("country", "");
2347 String variant = (String)fields.get("variant", "");
2348 String extStr = (String)fields.get("extensions", "");
2349
2350 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant);
2351 if (!extStr.isEmpty()) {
2352 try {
2353 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
2354 bldr.setExtensions(extStr);
2355 localeExtensions = bldr.getLocaleExtensions();
2356 } catch (LocaleSyntaxException e) {
2357 throw new IllformedLocaleException(e.getMessage());
2358 }
2359 } else {
2360 localeExtensions = null;
2361 }
2362 }
2363
2364 /**
2365 * Returns a cached <code>Locale</code> instance equivalent to
2366 * the deserialized <code>Locale</code>. When serialized
2367 * language, country and variant fields read from the object data stream
2368 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions
2369 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code>
2370 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script
2371 * type is "thai"). See <a href="Locale.html#special_cases_constructor">Special Cases</a>
2372 * for more information.
2373 *
2374 * @return an instance of <code>Locale</code> equivalent to
2375 * the deserialized <code>Locale</code>.
2376 * @throws java.io.ObjectStreamException
2377 */
2378 @java.io.Serial
2379 private Object readResolve() throws java.io.ObjectStreamException {
2380 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(),
2381 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions);
2382 }
2383
2384 private static volatile String[] isoLanguages;
2385
2386 private static volatile String[] isoCountries;
2387
2388 private static String convertOldISOCodes(String language) {
2389 // we accept both the old and the new ISO codes for the languages whose ISO
2390 // codes have changed, but we always store the OLD code, for backward compatibility
2391 language = LocaleUtils.toLowerString(language).intern();
2392 if (language == "he") {
2393 return "iw";
2394 } else if (language == "yi") {
2395 return "ji";
2396 } else if (language == "id") {
2397 return "in";
2398 } else {
2399 return language;
2400 }
2401 }
2402
2403 private static LocaleExtensions getCompatibilityExtensions(String language,
2404 String script,
2405 String country,
2406 String variant) {
2407 LocaleExtensions extensions = null;
2408 // Special cases for backward compatibility support
2409 if (LocaleUtils.caseIgnoreMatch(language, "ja")
2410 && script.isEmpty()
2411 && LocaleUtils.caseIgnoreMatch(country, "jp")
2412 && "JP".equals(variant)) {
2413 // ja_JP_JP -> u-ca-japanese (calendar = japanese)
2414 extensions = LocaleExtensions.CALENDAR_JAPANESE;
2415 } else if (LocaleUtils.caseIgnoreMatch(language, "th")
2416 && script.isEmpty()
2417 && LocaleUtils.caseIgnoreMatch(country, "th")
2418 && "TH".equals(variant)) {
2419 // th_TH_TH -> u-nu-thai (numbersystem = thai)
2420 extensions = LocaleExtensions.NUMBER_THAI;
2421 }
2422 return extensions;
2423 }
2424
2425 /**
2426 * Obtains a localized locale names from a LocaleNameProvider
2427 * implementation.
2428 */
2429 private static class LocaleNameGetter
2430 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> {
2431 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter();
2432
2433 @Override
2434 public String getObject(LocaleNameProvider localeNameProvider,
2435 Locale locale,
2436 String key,
2437 Object... params) {
2438 assert params.length == 3;
2439 int type = (Integer)params[0];
2440 String code = (String)params[1];
2441 String cat = (String)params[2];
2442
2443 switch(type) {
2444 case DISPLAY_LANGUAGE:
2445 return localeNameProvider.getDisplayLanguage(code, locale);
2446 case DISPLAY_COUNTRY:
2447 return localeNameProvider.getDisplayCountry(code, locale);
2448 case DISPLAY_VARIANT:
2449 return localeNameProvider.getDisplayVariant(code, locale);
2450 case DISPLAY_SCRIPT:
2451 return localeNameProvider.getDisplayScript(code, locale);
2452 case DISPLAY_UEXT_KEY:
2453 return localeNameProvider.getDisplayUnicodeExtensionKey(code, locale);
2454 case DISPLAY_UEXT_TYPE:
2455 return localeNameProvider.getDisplayUnicodeExtensionType(code, cat, locale);
2456 default:
2457 assert false; // shouldn't happen
2458 }
2459
2460 return null;
2461 }
2462 }
2463
2464 /**
2465 * Enum for locale categories. These locale categories are used to get/set
2466 * the default locale for the specific functionality represented by the
2467 * category.
2468 *
2469 * @see #getDefault(Locale.Category)
2470 * @see #setDefault(Locale.Category, Locale)
2471 * @since 1.7
2472 */
2473 public enum Category {
2474
2475 /**
2476 * Category used to represent the default locale for
2477 * displaying user interfaces.
2478 */
2479 DISPLAY("user.language.display",
2480 "user.script.display",
2481 "user.country.display",
2482 "user.variant.display",
2483 "user.extensions.display"),
2484
2485 /**
2486 * Category used to represent the default locale for
2487 * formatting dates, numbers, and/or currencies.
2488 */
2489 FORMAT("user.language.format",
2490 "user.script.format",
2491 "user.country.format",
2492 "user.variant.format",
2493 "user.extensions.format");
2494
2495 Category(String languageKey, String scriptKey, String countryKey,
2496 String variantKey, String extensionsKey) {
2497 this.languageKey = languageKey;
2498 this.scriptKey = scriptKey;
2499 this.countryKey = countryKey;
2500 this.variantKey = variantKey;
2501 this.extensionsKey = extensionsKey;
2502 }
2503
2504 final String languageKey;
2505 final String scriptKey;
2506 final String countryKey;
2507 final String variantKey;
2508 final String extensionsKey;
2509 }
2510
2511 /**
2512 * <code>Builder</code> is used to build instances of <code>Locale</code>
2513 * from values configured by the setters. Unlike the <code>Locale</code>
2514 * constructors, the <code>Builder</code> checks if a value configured by a
2515 * setter satisfies the syntax requirements defined by the <code>Locale</code>
2516 * class. A <code>Locale</code> object created by a <code>Builder</code> is
2517 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag
2518 * without losing information.
2519 *
2520 * <p><b>Note:</b> The <code>Locale</code> class does not provide any
2521 * syntactic restrictions on variant, while BCP 47 requires each variant
2522 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3
2523 * alphanumerics. The method <code>setVariant</code> throws
2524 * <code>IllformedLocaleException</code> for a variant that does not satisfy
2525 * this restriction. If it is necessary to support such a variant, use a
2526 * Locale constructor. However, keep in mind that a <code>Locale</code>
2527 * object created this way might lose the variant information when
2528 * transformed to a BCP 47 language tag.
2529 *
2530 * <p>The following example shows how to create a <code>Locale</code> object
2531 * with the <code>Builder</code>.
2532 * <blockquote>
2533 * <pre>
2534 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build();
2535 * </pre>
2536 * </blockquote>
2537 *
2538 * <p>Builders can be reused; <code>clear()</code> resets all
2539 * fields to their default values.
2540 *
2541 * @see Locale#forLanguageTag
2542 * @since 1.7
2543 */
2544 public static final class Builder {
2545 private final InternalLocaleBuilder localeBuilder;
2546
2547 /**
2548 * Constructs an empty Builder. The default value of all
2549 * fields, extensions, and private use information is the
2550 * empty string.
2551 */
2552 public Builder() {
2553 localeBuilder = new InternalLocaleBuilder();
2554 }
2555
2556 /**
2557 * Resets the <code>Builder</code> to match the provided
2558 * <code>locale</code>. Existing state is discarded.
2559 *
2560 * <p>All fields of the locale must be well-formed, see {@link Locale}.
2561 *
2562 * <p>Locales with any ill-formed fields cause
2563 * <code>IllformedLocaleException</code> to be thrown, except for the
2564 * following three cases which are accepted for compatibility
2565 * reasons:<ul>
2566 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese"
2567 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai"
2568 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul>
2569 *
2570 * @param locale the locale
2571 * @return This builder.
2572 * @throws IllformedLocaleException if <code>locale</code> has
2573 * any ill-formed fields.
2574 * @throws NullPointerException if <code>locale</code> is null.
2575 */
2576 public Builder setLocale(Locale locale) {
2577 try {
2578 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions);
2579 } catch (LocaleSyntaxException e) {
2580 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2581 }
2582 return this;
2583 }
2584
2585 /**
2586 * Resets the Builder to match the provided IETF BCP 47
2587 * language tag. Discards the existing state. Null and the
2588 * empty string cause the builder to be reset, like {@link
2589 * #clear}. Grandfathered tags (see {@link
2590 * Locale#forLanguageTag}) are converted to their canonical
2591 * form before being processed. Otherwise, the language tag
2592 * must be well-formed (see {@link Locale}) or an exception is
2593 * thrown (unlike <code>Locale.forLanguageTag</code>, which
2594 * just discards ill-formed and following portions of the
2595 * tag).
2596 *
2597 * @param languageTag the language tag
2598 * @return This builder.
2599 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed
2600 * @see Locale#forLanguageTag(String)
2601 */
2602 public Builder setLanguageTag(String languageTag) {
2603 ParseStatus sts = new ParseStatus();
2604 LanguageTag tag = LanguageTag.parse(languageTag, sts);
2605 if (sts.isError()) {
2606 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex());
2607 }
2608 localeBuilder.setLanguageTag(tag);
2609 return this;
2610 }
2611
2612 /**
2613 * Sets the language. If <code>language</code> is the empty string or
2614 * null, the language in this <code>Builder</code> is removed. Otherwise,
2615 * the language must be <a href="./Locale.html#def_language">well-formed</a>
2616 * or an exception is thrown.
2617 *
2618 * <p>The typical language value is a two or three-letter language
2619 * code as defined in ISO639.
2620 *
2621 * @param language the language
2622 * @return This builder.
2623 * @throws IllformedLocaleException if <code>language</code> is ill-formed
2624 */
2625 public Builder setLanguage(String language) {
2626 try {
2627 localeBuilder.setLanguage(language);
2628 } catch (LocaleSyntaxException e) {
2629 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2630 }
2631 return this;
2632 }
2633
2634 /**
2635 * Sets the script. If <code>script</code> is null or the empty string,
2636 * the script in this <code>Builder</code> is removed.
2637 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an
2638 * exception is thrown.
2639 *
2640 * <p>The typical script value is a four-letter script code as defined by ISO 15924.
2641 *
2642 * @param script the script
2643 * @return This builder.
2644 * @throws IllformedLocaleException if <code>script</code> is ill-formed
2645 */
2646 public Builder setScript(String script) {
2647 try {
2648 localeBuilder.setScript(script);
2649 } catch (LocaleSyntaxException e) {
2650 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2651 }
2652 return this;
2653 }
2654
2655 /**
2656 * Sets the region. If region is null or the empty string, the region
2657 * in this <code>Builder</code> is removed. Otherwise,
2658 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an
2659 * exception is thrown.
2660 *
2661 * <p>The typical region value is a two-letter ISO 3166 code or a
2662 * three-digit UN M.49 area code.
2663 *
2664 * <p>The country value in the <code>Locale</code> created by the
2665 * <code>Builder</code> is always normalized to upper case.
2666 *
2667 * @param region the region
2668 * @return This builder.
2669 * @throws IllformedLocaleException if <code>region</code> is ill-formed
2670 */
2671 public Builder setRegion(String region) {
2672 try {
2673 localeBuilder.setRegion(region);
2674 } catch (LocaleSyntaxException e) {
2675 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2676 }
2677 return this;
2678 }
2679
2680 /**
2681 * Sets the variant. If variant is null or the empty string, the
2682 * variant in this <code>Builder</code> is removed. Otherwise, it
2683 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a>
2684 * subtags, or an exception is thrown.
2685 *
2686 * <p><b>Note:</b> This method checks if <code>variant</code>
2687 * satisfies the IETF BCP 47 variant subtag's syntax requirements,
2688 * and normalizes the value to lowercase letters. However,
2689 * the <code>Locale</code> class does not impose any syntactic
2690 * restriction on variant, and the variant value in
2691 * <code>Locale</code> is case sensitive. To set such a variant,
2692 * use a Locale constructor.
2693 *
2694 * @param variant the variant
2695 * @return This builder.
2696 * @throws IllformedLocaleException if <code>variant</code> is ill-formed
2697 */
2698 public Builder setVariant(String variant) {
2699 try {
2700 localeBuilder.setVariant(variant);
2701 } catch (LocaleSyntaxException e) {
2702 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2703 }
2704 return this;
2705 }
2706
2707 /**
2708 * Sets the extension for the given key. If the value is null or the
2709 * empty string, the extension is removed. Otherwise, the extension
2710 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception
2711 * is thrown.
2712 *
2713 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION
2714 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension.
2715 * Setting a value for this key replaces any existing Unicode locale key/type
2716 * pairs with those defined in the extension.
2717 *
2718 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION
2719 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be
2720 * well-formed, the value for this key needs only to have subtags of one to
2721 * eight alphanumeric characters, not two to eight as in the general case.
2722 *
2723 * @param key the extension key
2724 * @param value the extension value
2725 * @return This builder.
2726 * @throws IllformedLocaleException if <code>key</code> is illegal
2727 * or <code>value</code> is ill-formed
2728 * @see #setUnicodeLocaleKeyword(String, String)
2729 */
2730 public Builder setExtension(char key, String value) {
2731 try {
2732 localeBuilder.setExtension(key, value);
2733 } catch (LocaleSyntaxException e) {
2734 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2735 }
2736 return this;
2737 }
2738
2739 /**
2740 * Sets the Unicode locale keyword type for the given key. If the type
2741 * is null, the Unicode keyword is removed. Otherwise, the key must be
2742 * non-null and both key and type must be <a
2743 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2744 * is thrown.
2745 *
2746 * <p>Keys and types are converted to lower case.
2747 *
2748 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension}
2749 * replaces all Unicode locale keywords with those defined in the
2750 * extension.
2751 *
2752 * @param key the Unicode locale key
2753 * @param type the Unicode locale type
2754 * @return This builder.
2755 * @throws IllformedLocaleException if <code>key</code> or <code>type</code>
2756 * is ill-formed
2757 * @throws NullPointerException if <code>key</code> is null
2758 * @see #setExtension(char, String)
2759 */
2760 public Builder setUnicodeLocaleKeyword(String key, String type) {
2761 try {
2762 localeBuilder.setUnicodeLocaleKeyword(key, type);
2763 } catch (LocaleSyntaxException e) {
2764 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2765 }
2766 return this;
2767 }
2768
2769 /**
2770 * Adds a unicode locale attribute, if not already present, otherwise
2771 * has no effect. The attribute must not be null and must be <a
2772 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2773 * is thrown.
2774 *
2775 * @param attribute the attribute
2776 * @return This builder.
2777 * @throws NullPointerException if <code>attribute</code> is null
2778 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2779 * @see #setExtension(char, String)
2780 */
2781 public Builder addUnicodeLocaleAttribute(String attribute) {
2782 try {
2783 localeBuilder.addUnicodeLocaleAttribute(attribute);
2784 } catch (LocaleSyntaxException e) {
2785 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2786 }
2787 return this;
2788 }
2789
2790 /**
2791 * Removes a unicode locale attribute, if present, otherwise has no
2792 * effect. The attribute must not be null and must be <a
2793 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2794 * is thrown.
2795 *
2796 * <p>Attribute comparison for removal is case-insensitive.
2797 *
2798 * @param attribute the attribute
2799 * @return This builder.
2800 * @throws NullPointerException if <code>attribute</code> is null
2801 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2802 * @see #setExtension(char, String)
2803 */
2804 public Builder removeUnicodeLocaleAttribute(String attribute) {
2805 Objects.requireNonNull(attribute);
2806 try {
2807 localeBuilder.removeUnicodeLocaleAttribute(attribute);
2808 } catch (LocaleSyntaxException e) {
2809 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2810 }
2811 return this;
2812 }
2813
2814 /**
2815 * Resets the builder to its initial, empty state.
2816 *
2817 * @return This builder.
2818 */
2819 public Builder clear() {
2820 localeBuilder.clear();
2821 return this;
2822 }
2823
2824 /**
2825 * Resets the extensions to their initial, empty state.
2826 * Language, script, region and variant are unchanged.
2827 *
2828 * @return This builder.
2829 * @see #setExtension(char, String)
2830 */
2831 public Builder clearExtensions() {
2832 localeBuilder.clearExtensions();
2833 return this;
2834 }
2835
2836 /**
2837 * Returns an instance of <code>Locale</code> created from the fields set
2838 * on this builder.
2839 *
2840 * <p>This applies the conversions listed in {@link Locale#forLanguageTag}
2841 * when constructing a Locale. (Grandfathered tags are handled in
2842 * {@link #setLanguageTag}.)
2843 *
2844 * @return A Locale.
2845 */
2846 public Locale build() {
2847 BaseLocale baseloc = localeBuilder.getBaseLocale();
2848 LocaleExtensions extensions = localeBuilder.getLocaleExtensions();
2849 if (extensions == null && !baseloc.getVariant().isEmpty()) {
2850 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(),
2851 baseloc.getRegion(), baseloc.getVariant());
2852 }
2853 return Locale.getInstance(baseloc, extensions);
2854 }
2855 }
2856
2857 /**
2858 * This enum provides constants to select a filtering mode for locale
2859 * matching. Refer to <a href="http://tools.ietf.org/html/rfc4647">RFC 4647
2860 * Matching of Language Tags</a> for details.
2861 *
2862 * <p>As an example, think of two Language Priority Lists each of which
2863 * includes only one language range and a set of following language tags:
2864 *
2865 * <pre>
2866 * de (German)
2867 * de-DE (German, Germany)
2868 * de-Deva (German, in Devanagari script)
2869 * de-Deva-DE (German, in Devanagari script, Germany)
2870 * de-DE-1996 (German, Germany, orthography of 1996)
2871 * de-Latn-DE (German, in Latin script, Germany)
2872 * de-Latn-DE-1996 (German, in Latin script, Germany, orthography of 1996)
2873 * </pre>
2874 *
2875 * The filtering method will behave as follows:
2876 *
2877 * <table class="striped">
2878 * <caption>Filtering method behavior</caption>
2879 * <thead>
2880 * <tr>
2881 * <th scope="col">Filtering Mode</th>
2882 * <th scope="col">Language Priority List: {@code "de-DE"}</th>
2883 * <th scope="col">Language Priority List: {@code "de-*-DE"}</th>
2884 * </tr>
2885 * </thead>
2886 * <tbody>
2887 * <tr>
2888 * <th scope="row" style="vertical-align:top">
2889 * {@link FilteringMode#AUTOSELECT_FILTERING AUTOSELECT_FILTERING}
2890 * </th>
2891 * <td style="vertical-align:top">
2892 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2893 * {@code "de-DE-1996"}.
2894 * </td>
2895 * <td style="vertical-align:top">
2896 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2897 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2898 * {@code "de-Latn-DE-1996"}.
2899 * </td>
2900 * </tr>
2901 * <tr>
2902 * <th scope="row" style="vertical-align:top">
2903 * {@link FilteringMode#EXTENDED_FILTERING EXTENDED_FILTERING}
2904 * </th>
2905 * <td style="vertical-align:top">
2906 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2907 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2908 * {@code "de-Latn-DE-1996"}.
2909 * </td>
2910 * <td style="vertical-align:top">Same as above.</td>
2911 * </tr>
2912 * <tr>
2913 * <th scope="row" style="vertical-align:top">
2914 * {@link FilteringMode#IGNORE_EXTENDED_RANGES IGNORE_EXTENDED_RANGES}
2915 * </th>
2916 * <td style="vertical-align:top">
2917 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2918 * {@code "de-DE-1996"}.
2919 * </td>
2920 * <td style="vertical-align:top">
2921 * Performs <em>basic</em> filtering and returns {@code null} because
2922 * nothing matches.
2923 * </td>
2924 * </tr>
2925 * <tr>
2926 * <th scope="row" style="vertical-align:top">
2927 * {@link FilteringMode#MAP_EXTENDED_RANGES MAP_EXTENDED_RANGES}
2928 * </th>
2929 * <td style="vertical-align:top">Same as above.</td>
2930 * <td style="vertical-align:top">
2931 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2932 * {@code "de-DE-1996"} because {@code "de-*-DE"} is mapped to
2933 * {@code "de-DE"}.
2934 * </td>
2935 * </tr>
2936 * <tr>
2937 * <th scope="row" style="vertical-align:top">
2938 * {@link FilteringMode#REJECT_EXTENDED_RANGES REJECT_EXTENDED_RANGES}
2939 * </th>
2940 * <td style="vertical-align:top">Same as above.</td>
2941 * <td style="vertical-align:top">
2942 * Throws {@link IllegalArgumentException} because {@code "de-*-DE"} is
2943 * not a valid basic language range.
2944 * </td>
2945 * </tr>
2946 * </tbody>
2947 * </table>
2948 *
2949 * @see #filter(List, Collection, FilteringMode)
2950 * @see #filterTags(List, Collection, FilteringMode)
2951 *
2952 * @since 1.8
2953 */
2954 public static enum FilteringMode {
2955 /**
2956 * Specifies automatic filtering mode based on the given Language
2957 * Priority List consisting of language ranges. If all of the ranges
2958 * are basic, basic filtering is selected. Otherwise, extended
2959 * filtering is selected.
2960 */
2961 AUTOSELECT_FILTERING,
2962
2963 /**
2964 * Specifies extended filtering.
2965 */
2966 EXTENDED_FILTERING,
2967
2968 /**
2969 * Specifies basic filtering: Note that any extended language ranges
2970 * included in the given Language Priority List are ignored.
2971 */
2972 IGNORE_EXTENDED_RANGES,
2973
2974 /**
2975 * Specifies basic filtering: If any extended language ranges are
2976 * included in the given Language Priority List, they are mapped to the
2977 * basic language range. Specifically, a language range starting with a
2978 * subtag {@code "*"} is treated as a language range {@code "*"}. For
2979 * example, {@code "*-US"} is treated as {@code "*"}. If {@code "*"} is
2980 * not the first subtag, {@code "*"} and extra {@code "-"} are removed.
2981 * For example, {@code "ja-*-JP"} is mapped to {@code "ja-JP"}.
2982 */
2983 MAP_EXTENDED_RANGES,
2984
2985 /**
2986 * Specifies basic filtering: If any extended language ranges are
2987 * included in the given Language Priority List, the list is rejected
2988 * and the filtering method throws {@link IllegalArgumentException}.
2989 */
2990 REJECT_EXTENDED_RANGES
2991 };
2992
2993 /**
2994 * This class expresses a <em>Language Range</em> defined in
2995 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
2996 * Language Tags</a>. A language range is an identifier which is used to
2997 * select language tag(s) meeting specific requirements by using the
2998 * mechanisms described in <a href="Locale.html#LocaleMatching">Locale
2999 * Matching</a>. A list which represents a user's preferences and consists
3000 * of language ranges is called a <em>Language Priority List</em>.
3001 *
3002 * <p>There are two types of language ranges: basic and extended. In RFC
3003 * 4647, the syntax of language ranges is expressed in
3004 * <a href="http://tools.ietf.org/html/rfc4234">ABNF</a> as follows:
3005 * <blockquote>
3006 * <pre>
3007 * basic-language-range = (1*8ALPHA *("-" 1*8alphanum)) / "*"
3008 * extended-language-range = (1*8ALPHA / "*")
3009 * *("-" (1*8alphanum / "*"))
3010 * alphanum = ALPHA / DIGIT
3011 * </pre>
3012 * </blockquote>
3013 * For example, {@code "en"} (English), {@code "ja-JP"} (Japanese, Japan),
3014 * {@code "*"} (special language range which matches any language tag) are
3015 * basic language ranges, whereas {@code "*-CH"} (any languages,
3016 * Switzerland), {@code "es-*"} (Spanish, any regions), and
3017 * {@code "zh-Hant-*"} (Traditional Chinese, any regions) are extended
3018 * language ranges.
3019 *
3020 * @see #filter
3021 * @see #filterTags
3022 * @see #lookup
3023 * @see #lookupTag
3024 *
3025 * @since 1.8
3026 */
3027 public static final class LanguageRange {
3028
3029 /**
3030 * A constant holding the maximum value of weight, 1.0, which indicates
3031 * that the language range is a good fit for the user.
3032 */
3033 public static final double MAX_WEIGHT = 1.0;
3034
3035 /**
3036 * A constant holding the minimum value of weight, 0.0, which indicates
3037 * that the language range is not a good fit for the user.
3038 */
3039 public static final double MIN_WEIGHT = 0.0;
3040
3041 private final String range;
3042 private final double weight;
3043
3044 private volatile int hash;
3045
3046 /**
3047 * Constructs a {@code LanguageRange} using the given {@code range}.
3048 * Note that no validation is done against the IANA Language Subtag
3049 * Registry at time of construction.
3050 *
3051 * <p>This is equivalent to {@code LanguageRange(range, MAX_WEIGHT)}.
3052 *
3053 * @param range a language range
3054 * @throws NullPointerException if the given {@code range} is
3055 * {@code null}
3056 * @throws IllegalArgumentException if the given {@code range} does not
3057 * comply with the syntax of the language range mentioned in RFC 4647
3058 */
3059 public LanguageRange(String range) {
3060 this(range, MAX_WEIGHT);
3061 }
3062
3063 /**
3064 * Constructs a {@code LanguageRange} using the given {@code range} and
3065 * {@code weight}. Note that no validation is done against the IANA
3066 * Language Subtag Registry at time of construction.
3067 *
3068 * @param range a language range
3069 * @param weight a weight value between {@code MIN_WEIGHT} and
3070 * {@code MAX_WEIGHT}
3071 * @throws NullPointerException if the given {@code range} is
3072 * {@code null}
3073 * @throws IllegalArgumentException if the given {@code range} does not
3074 * comply with the syntax of the language range mentioned in RFC 4647
3075 * or if the given {@code weight} is less than {@code MIN_WEIGHT}
3076 * or greater than {@code MAX_WEIGHT}
3077 */
3078 public LanguageRange(String range, double weight) {
3079 if (range == null) {
3080 throw new NullPointerException();
3081 }
3082 if (weight < MIN_WEIGHT || weight > MAX_WEIGHT) {
3083 throw new IllegalArgumentException("weight=" + weight);
3084 }
3085
3086 range = range.toLowerCase(Locale.ROOT);
3087
3088 // Do syntax check.
3089 boolean isIllFormed = false;
3090 String[] subtags = range.split("-");
3091 if (isSubtagIllFormed(subtags[0], true)
3092 || range.endsWith("-")) {
3093 isIllFormed = true;
3094 } else {
3095 for (int i = 1; i < subtags.length; i++) {
3096 if (isSubtagIllFormed(subtags[i], false)) {
3097 isIllFormed = true;
3098 break;
3099 }
3100 }
3101 }
3102 if (isIllFormed) {
3103 throw new IllegalArgumentException("range=" + range);
3104 }
3105
3106 this.range = range;
3107 this.weight = weight;
3108 }
3109
3110 private static boolean isSubtagIllFormed(String subtag,
3111 boolean isFirstSubtag) {
3112 if (subtag.isEmpty() || subtag.length() > 8) {
3113 return true;
3114 } else if (subtag.equals("*")) {
3115 return false;
3116 }
3117 char[] charArray = subtag.toCharArray();
3118 if (isFirstSubtag) { // ALPHA
3119 for (char c : charArray) {
3120 if (c < 'a' || c > 'z') {
3121 return true;
3122 }
3123 }
3124 } else { // ALPHA / DIGIT
3125 for (char c : charArray) {
3126 if (c < '0' || (c > '9' && c < 'a') || c > 'z') {
3127 return true;
3128 }
3129 }
3130 }
3131 return false;
3132 }
3133
3134 /**
3135 * Returns the language range of this {@code LanguageRange}.
3136 *
3137 * @return the language range.
3138 */
3139 public String getRange() {
3140 return range;
3141 }
3142
3143 /**
3144 * Returns the weight of this {@code LanguageRange}.
3145 *
3146 * @return the weight value.
3147 */
3148 public double getWeight() {
3149 return weight;
3150 }
3151
3152 /**
3153 * Parses the given {@code ranges} to generate a Language Priority List.
3154 *
3155 * <p>This method performs a syntactic check for each language range in
3156 * the given {@code ranges} but doesn't do validation using the IANA
3157 * Language Subtag Registry.
3158 *
3159 * <p>The {@code ranges} to be given can take one of the following
3160 * forms:
3161 *
3162 * <pre>
3163 * "Accept-Language: ja,en;q=0.4" (weighted list with Accept-Language prefix)
3164 * "ja,en;q=0.4" (weighted list)
3165 * "ja,en" (prioritized list)
3166 * </pre>
3167 *
3168 * In a weighted list, each language range is given a weight value.
3169 * The weight value is identical to the "quality value" in
3170 * <a href="http://tools.ietf.org/html/rfc2616">RFC 2616</a>, and it
3171 * expresses how much the user prefers the language. A weight value is
3172 * specified after a corresponding language range followed by
3173 * {@code ";q="}, and the default weight value is {@code MAX_WEIGHT}
3174 * when it is omitted.
3175 *
3176 * <p>Unlike a weighted list, language ranges in a prioritized list
3177 * are sorted in the descending order based on its priority. The first
3178 * language range has the highest priority and meets the user's
3179 * preference most.
3180 *
3181 * <p>In either case, language ranges are sorted in descending order in
3182 * the Language Priority List based on priority or weight. If a
3183 * language range appears in the given {@code ranges} more than once,
3184 * only the first one is included on the Language Priority List.
3185 *
3186 * <p>The returned list consists of language ranges from the given
3187 * {@code ranges} and their equivalents found in the IANA Language
3188 * Subtag Registry. For example, if the given {@code ranges} is
3189 * {@code "Accept-Language: iw,en-us;q=0.7,en;q=0.3"}, the elements in
3190 * the list to be returned are:
3191 *
3192 * <pre>
3193 * <b>Range</b> <b>Weight</b>
3194 * "iw" (older tag for Hebrew) 1.0
3195 * "he" (new preferred code for Hebrew) 1.0
3196 * "en-us" (English, United States) 0.7
3197 * "en" (English) 0.3
3198 * </pre>
3199 *
3200 * Two language ranges, {@code "iw"} and {@code "he"}, have the same
3201 * highest priority in the list. By adding {@code "he"} to the user's
3202 * Language Priority List, locale-matching method can find Hebrew as a
3203 * matching locale (or language tag) even if the application or system
3204 * offers only {@code "he"} as a supported locale (or language tag).
3205 *
3206 * @param ranges a list of comma-separated language ranges or a list of
3207 * language ranges in the form of the "Accept-Language" header
3208 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3209 * 2616</a>
3210 * @return a Language Priority List consisting of language ranges
3211 * included in the given {@code ranges} and their equivalent
3212 * language ranges if available. The list is modifiable.
3213 * @throws NullPointerException if {@code ranges} is null
3214 * @throws IllegalArgumentException if a language range or a weight
3215 * found in the given {@code ranges} is ill-formed
3216 */
3217 public static List<LanguageRange> parse(String ranges) {
3218 return LocaleMatcher.parse(ranges);
3219 }
3220
3221 /**
3222 * Parses the given {@code ranges} to generate a Language Priority
3223 * List, and then customizes the list using the given {@code map}.
3224 * This method is equivalent to
3225 * {@code mapEquivalents(parse(ranges), map)}.
3226 *
3227 * @param ranges a list of comma-separated language ranges or a list
3228 * of language ranges in the form of the "Accept-Language" header
3229 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3230 * 2616</a>
3231 * @param map a map containing information to customize language ranges
3232 * @return a Language Priority List with customization. The list is
3233 * modifiable.
3234 * @throws NullPointerException if {@code ranges} is null
3235 * @throws IllegalArgumentException if a language range or a weight
3236 * found in the given {@code ranges} is ill-formed
3237 * @see #parse(String)
3238 * @see #mapEquivalents
3239 */
3240 public static List<LanguageRange> parse(String ranges,
3241 Map<String, List<String>> map) {
3242 return mapEquivalents(parse(ranges), map);
3243 }
3244
3245 /**
3246 * Generates a new customized Language Priority List using the given
3247 * {@code priorityList} and {@code map}. If the given {@code map} is
3248 * empty, this method returns a copy of the given {@code priorityList}.
3249 *
3250 * <p>In the map, a key represents a language range whereas a value is
3251 * a list of equivalents of it. {@code '*'} cannot be used in the map.
3252 * Each equivalent language range has the same weight value as its
3253 * original language range.
3254 *
3255 * <pre>
3256 * An example of map:
3257 * <b>Key</b> <b>Value</b>
3258 * "zh" (Chinese) "zh",
3259 * "zh-Hans"(Simplified Chinese)
3260 * "zh-HK" (Chinese, Hong Kong) "zh-HK"
3261 * "zh-TW" (Chinese, Taiwan) "zh-TW"
3262 * </pre>
3263 *
3264 * The customization is performed after modification using the IANA
3265 * Language Subtag Registry.
3266 *
3267 * <p>For example, if a user's Language Priority List consists of five
3268 * language ranges ({@code "zh"}, {@code "zh-CN"}, {@code "en"},
3269 * {@code "zh-TW"}, and {@code "zh-HK"}), the newly generated Language
3270 * Priority List which is customized using the above map example will
3271 * consists of {@code "zh"}, {@code "zh-Hans"}, {@code "zh-CN"},
3272 * {@code "zh-Hans-CN"}, {@code "en"}, {@code "zh-TW"}, and
3273 * {@code "zh-HK"}.
3274 *
3275 * <p>{@code "zh-HK"} and {@code "zh-TW"} aren't converted to
3276 * {@code "zh-Hans-HK"} nor {@code "zh-Hans-TW"} even if they are
3277 * included in the Language Priority List. In this example, mapping
3278 * is used to clearly distinguish Simplified Chinese and Traditional
3279 * Chinese.
3280 *
3281 * <p>If the {@code "zh"}-to-{@code "zh"} mapping isn't included in the
3282 * map, a simple replacement will be performed and the customized list
3283 * won't include {@code "zh"} and {@code "zh-CN"}.
3284 *
3285 * @param priorityList user's Language Priority List
3286 * @param map a map containing information to customize language ranges
3287 * @return a new Language Priority List with customization. The list is
3288 * modifiable.
3289 * @throws NullPointerException if {@code priorityList} is {@code null}
3290 * @see #parse(String, Map)
3291 */
3292 public static List<LanguageRange> mapEquivalents(
3293 List<LanguageRange>priorityList,
3294 Map<String, List<String>> map) {
3295 return LocaleMatcher.mapEquivalents(priorityList, map);
3296 }
3297
3298 /**
3299 * Returns a hash code value for the object.
3300 *
3301 * @return a hash code value for this object.
3302 */
3303 @Override
3304 public int hashCode() {
3305 int h = hash;
3306 if (h == 0) {
3307 h = 17;
3308 h = 37*h + range.hashCode();
3309 long bitsWeight = Double.doubleToLongBits(weight);
3310 h = 37*h + (int)(bitsWeight ^ (bitsWeight >>> 32));
3311 if (h != 0) {
3312 hash = h;
3313 }
3314 }
3315 return h;
3316 }
3317
3318 /**
3319 * Compares this object to the specified object. The result is true if
3320 * and only if the argument is not {@code null} and is a
3321 * {@code LanguageRange} object that contains the same {@code range}
3322 * and {@code weight} values as this object.
3323 *
3324 * @param obj the object to compare with
3325 * @return {@code true} if this object's {@code range} and
3326 * {@code weight} are the same as the {@code obj}'s; {@code false}
3327 * otherwise.
3328 */
3329 @Override
3330 public boolean equals(Object obj) {
3331 if (this == obj) {
3332 return true;
3333 }
3334 if (!(obj instanceof LanguageRange)) {
3335 return false;
3336 }
3337 LanguageRange other = (LanguageRange)obj;
3338 return hash == other.hash
3339 && range.equals(other.range)
3340 && weight == other.weight;
3341 }
3342
3343 /**
3344 * Returns an informative string representation of this {@code LanguageRange}
3345 * object, consisting of language range and weight if the range is
3346 * weighted and the weight is less than the max weight.
3347 *
3348 * @return a string representation of this {@code LanguageRange} object.
3349 */
3350 @Override
3351 public String toString() {
3352 return (weight == MAX_WEIGHT) ? range : range + ";q=" + weight;
3353 }
3354 }
3355
3356 /**
3357 * Returns a list of matching {@code Locale} instances using the filtering
3358 * mechanism defined in RFC 4647.
3359 *
3360 * This filter operation on the given {@code locales} ensures that only
3361 * unique matching locale(s) are returned.
3362 *
3363 * @param priorityList user's Language Priority List in which each language
3364 * tag is sorted in descending order based on priority or weight
3365 * @param locales {@code Locale} instances used for matching
3366 * @param mode filtering mode
3367 * @return a list of {@code Locale} instances for matching language tags
3368 * sorted in descending order based on priority or weight, or an empty
3369 * list if nothing matches. The list is modifiable.
3370 * @throws NullPointerException if {@code priorityList} or {@code locales}
3371 * is {@code null}
3372 * @throws IllegalArgumentException if one or more extended language ranges
3373 * are included in the given list when
3374 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3375 *
3376 * @since 1.8
3377 */
3378 public static List<Locale> filter(List<LanguageRange> priorityList,
3379 Collection<Locale> locales,
3380 FilteringMode mode) {
3381 return LocaleMatcher.filter(priorityList, locales, mode);
3382 }
3383
3384 /**
3385 * Returns a list of matching {@code Locale} instances using the filtering
3386 * mechanism defined in RFC 4647. This is equivalent to
3387 * {@link #filter(List, Collection, FilteringMode)} when {@code mode} is
3388 * {@link FilteringMode#AUTOSELECT_FILTERING}.
3389 *
3390 * This filter operation on the given {@code locales} ensures that only
3391 * unique matching locale(s) are returned.
3392 *
3393 * @param priorityList user's Language Priority List in which each language
3394 * tag is sorted in descending order based on priority or weight
3395 * @param locales {@code Locale} instances used for matching
3396 * @return a list of {@code Locale} instances for matching language tags
3397 * sorted in descending order based on priority or weight, or an empty
3398 * list if nothing matches. The list is modifiable.
3399 * @throws NullPointerException if {@code priorityList} or {@code locales}
3400 * is {@code null}
3401 *
3402 * @since 1.8
3403 */
3404 public static List<Locale> filter(List<LanguageRange> priorityList,
3405 Collection<Locale> locales) {
3406 return filter(priorityList, locales, FilteringMode.AUTOSELECT_FILTERING);
3407 }
3408
3409 /**
3410 * Returns a list of matching languages tags using the basic filtering
3411 * mechanism defined in RFC 4647.
3412 *
3413 * This filter operation on the given {@code tags} ensures that only
3414 * unique matching tag(s) are returned with preserved case. In case of
3415 * duplicate matching tags with the case difference, the first matching
3416 * tag with preserved case is returned.
3417 * For example, "de-ch" is returned out of the duplicate matching tags
3418 * "de-ch" and "de-CH", if "de-ch" is checked first for matching in the
3419 * given {@code tags}. Note that if the given {@code tags} is an unordered
3420 * {@code Collection}, the returned matching tag out of duplicate tags is
3421 * subject to change, depending on the implementation of the
3422 * {@code Collection}.
3423 *
3424 * @param priorityList user's Language Priority List in which each language
3425 * tag is sorted in descending order based on priority or weight
3426 * @param tags language tags
3427 * @param mode filtering mode
3428 * @return a list of matching language tags sorted in descending order
3429 * based on priority or weight, or an empty list if nothing matches.
3430 * The list is modifiable.
3431 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3432 * {@code null}
3433 * @throws IllegalArgumentException if one or more extended language ranges
3434 * are included in the given list when
3435 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3436 *
3437 * @since 1.8
3438 */
3439 public static List<String> filterTags(List<LanguageRange> priorityList,
3440 Collection<String> tags,
3441 FilteringMode mode) {
3442 return LocaleMatcher.filterTags(priorityList, tags, mode);
3443 }
3444
3445 /**
3446 * Returns a list of matching languages tags using the basic filtering
3447 * mechanism defined in RFC 4647. This is equivalent to
3448 * {@link #filterTags(List, Collection, FilteringMode)} when {@code mode}
3449 * is {@link FilteringMode#AUTOSELECT_FILTERING}.
3450 *
3451 * This filter operation on the given {@code tags} ensures that only
3452 * unique matching tag(s) are returned with preserved case. In case of
3453 * duplicate matching tags with the case difference, the first matching
3454 * tag with preserved case is returned.
3455 * For example, "de-ch" is returned out of the duplicate matching tags
3456 * "de-ch" and "de-CH", if "de-ch" is checked first for matching in the
3457 * given {@code tags}. Note that if the given {@code tags} is an unordered
3458 * {@code Collection}, the returned matching tag out of duplicate tags is
3459 * subject to change, depending on the implementation of the
3460 * {@code Collection}.
3461 *
3462 * @param priorityList user's Language Priority List in which each language
3463 * tag is sorted in descending order based on priority or weight
3464 * @param tags language tags
3465 * @return a list of matching language tags sorted in descending order
3466 * based on priority or weight, or an empty list if nothing matches.
3467 * The list is modifiable.
3468 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3469 * {@code null}
3470 *
3471 * @since 1.8
3472 */
3473 public static List<String> filterTags(List<LanguageRange> priorityList,
3474 Collection<String> tags) {
3475 return filterTags(priorityList, tags, FilteringMode.AUTOSELECT_FILTERING);
3476 }
3477
3478 /**
3479 * Returns a {@code Locale} instance for the best-matching language
3480 * tag using the lookup mechanism defined in RFC 4647.
3481 *
3482 * @param priorityList user's Language Priority List in which each language
3483 * tag is sorted in descending order based on priority or weight
3484 * @param locales {@code Locale} instances used for matching
3485 * @return the best matching <code>Locale</code> instance chosen based on
3486 * priority or weight, or {@code null} if nothing matches.
3487 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3488 * {@code null}
3489 *
3490 * @since 1.8
3491 */
3492 public static Locale lookup(List<LanguageRange> priorityList,
3493 Collection<Locale> locales) {
3494 return LocaleMatcher.lookup(priorityList, locales);
3495 }
3496
3497 /**
3498 * Returns the best-matching language tag using the lookup mechanism
3499 * defined in RFC 4647.
3500 *
3501 * This lookup operation on the given {@code tags} ensures that the
3502 * first matching tag with preserved case is returned.
3503 *
3504 * @param priorityList user's Language Priority List in which each language
3505 * tag is sorted in descending order based on priority or weight
3506 * @param tags language tangs used for matching
3507 * @return the best matching language tag chosen based on priority or
3508 * weight, or {@code null} if nothing matches.
3509 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3510 * {@code null}
3511 *
3512 * @since 1.8
3513 */
3514 public static String lookupTag(List<LanguageRange> priorityList,
3515 Collection<String> tags) {
3516 return LocaleMatcher.lookupTag(priorityList, tags);
3517 }
3518
3519 }