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