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 var baseLocales = BaseLocale.constantBaseLocales;
573 boolean fromArchive = baseLocales != null;
574 if (!fromArchive) {
575 baseLocales = new HashMap<>();
576 }
577 ENGLISH = constant("en", "", baseLocales, fromArchive);
578 FRENCH = constant("fr", "", baseLocales, fromArchive);
579 GERMAN = constant("de", "", baseLocales, fromArchive);
580 ITALIAN = constant("it", "", baseLocales, fromArchive);
581 JAPANESE = constant("ja", "", baseLocales, fromArchive);
582 KOREAN = constant("ko", "", baseLocales, fromArchive);
583 CHINESE = constant("zh", "", baseLocales, fromArchive);
584 SIMPLIFIED_CHINESE = constant("zh", "CN", baseLocales, fromArchive);
585 TRADITIONAL_CHINESE = constant("zh", "TW", baseLocales, fromArchive);
586 FRANCE = constant("fr", "FR", baseLocales, fromArchive);
587 GERMANY = constant("de", "DE", baseLocales, fromArchive);
588 ITALY = constant("it", "IT", baseLocales, fromArchive);
589 JAPAN = constant("ja", "JP", baseLocales, fromArchive);
590 KOREA = constant("ko", "KR", baseLocales, fromArchive);
591 UK = constant("en", "GB", baseLocales, fromArchive);
592 US = constant("en", "US", baseLocales, fromArchive);
593 CANADA = constant("en", "CA", baseLocales, fromArchive);
594 CANADA_FRENCH = constant("fr", "CA", baseLocales, fromArchive);
595 ROOT = constant("", "", baseLocales, fromArchive);
596 if (!fromArchive) {
597 BaseLocale.constantBaseLocales = Map.copyOf(baseLocales);
598 }
599 }
600
601 /** Useful constant for country.
602 */
603 public static final Locale CHINA = SIMPLIFIED_CHINESE;
604
605 /** Useful constant for country.
606 */
607 public static final Locale PRC = SIMPLIFIED_CHINESE;
608
609 /** Useful constant for country.
610 */
611 public static final Locale TAIWAN = TRADITIONAL_CHINESE;
612
613 /**
614 * This method must be called only for creating the Locale.*
615 * constants due to making shortcuts.
616 */
617 private static Locale constant(String lang,
618 String country,
619 Map<String, Map<String, BaseLocale>> baseLocaleMap,
620 boolean fromArchive) {
621 BaseLocale base;
622 if (fromArchive) {
623 var countryMap = baseLocaleMap.get(lang);
624 base = countryMap.get(country);
625 } else {
626 base = BaseLocale.createInstance(lang, country);
627 var countryMap = baseLocaleMap.get(lang);
628 if (countryMap == null) {
629 countryMap = new HashMap<>();
630 baseLocaleMap.put(lang, countryMap);
631 }
632 countryMap.put(country, base);
633 }
634 Locale locale = new Locale(base, null);
635 CONSTANT_LOCALES.put(base, locale);
636 return locale;
637 }
638
639 /**
640 * The key for the private use extension ('x').
641 *
642 * @see #getExtension(char)
643 * @see Builder#setExtension(char, String)
644 * @since 1.7
645 */
646 public static final char PRIVATE_USE_EXTENSION = 'x';
647
648 /**
649 * The key for Unicode locale extension ('u').
650 *
651 * @see #getExtension(char)
652 * @see Builder#setExtension(char, String)
653 * @since 1.7
654 */
655 public static final char UNICODE_LOCALE_EXTENSION = 'u';
656
657 /** serialization ID
658 */
659 static final long serialVersionUID = 9149081749638150636L;
660
661 /**
662 * Enum for specifying the type defined in ISO 3166. This enum is used to
663 * retrieve the two-letter ISO3166-1 alpha-2, three-letter ISO3166-1
664 * alpha-3, four-letter ISO3166-3 country codes.
665 *
666 * @see #getISOCountries(Locale.IsoCountryCode)
667 * @since 9
668 */
669 public static enum IsoCountryCode {
670 /**
671 * PART1_ALPHA2 is used to represent the ISO3166-1 alpha-2 two letter
672 * country codes.
673 */
674 PART1_ALPHA2 {
675 @Override
676 Set<String> createCountryCodeSet() {
677 return Set.of(Locale.getISOCountries());
678 }
679 },
680
681 /**
682 *
683 * PART1_ALPHA3 is used to represent the ISO3166-1 alpha-3 three letter
684 * country codes.
685 */
686 PART1_ALPHA3 {
687 @Override
688 Set<String> createCountryCodeSet() {
689 return LocaleISOData.computeISO3166_1Alpha3Countries();
690 }
691 },
692
693 /**
694 * PART3 is used to represent the ISO3166-3 four letter country codes.
695 */
696 PART3 {
697 @Override
698 Set<String> createCountryCodeSet() {
699 return Set.of(LocaleISOData.ISO3166_3);
700 }
701 };
702
703 /**
704 * Concrete implementation of this method attempts to compute value
705 * for iso3166CodesMap for each IsoCountryCode type key.
706 */
707 abstract Set<String> createCountryCodeSet();
708
709 /**
710 * Map to hold country codes for each ISO3166 part.
711 */
712 private static Map<IsoCountryCode, Set<String>> iso3166CodesMap = new ConcurrentHashMap<>();
713
714 /**
715 * This method is called from Locale class to retrieve country code set
716 * for getISOCountries(type)
717 */
718 static Set<String> retrieveISOCountryCodes(IsoCountryCode type) {
719 return iso3166CodesMap.computeIfAbsent(type, IsoCountryCode::createCountryCodeSet);
720 }
721 }
722
723 /**
724 * Display types for retrieving localized names from the name providers.
725 */
726 private static final int DISPLAY_LANGUAGE = 0;
727 private static final int DISPLAY_COUNTRY = 1;
728 private static final int DISPLAY_VARIANT = 2;
729 private static final int DISPLAY_SCRIPT = 3;
730 private static final int DISPLAY_UEXT_KEY = 4;
731 private static final int DISPLAY_UEXT_TYPE = 5;
732
733 /**
734 * Private constructor used by getInstance method
735 */
736 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) {
737 this.baseLocale = baseLocale;
738 this.localeExtensions = extensions;
739 }
740
741 /**
742 * Construct a locale from language, country and variant.
743 * This constructor normalizes the language value to lowercase and
744 * the country value to uppercase.
745 * <p>
746 * <b>Note:</b>
747 * <ul>
748 * <li>ISO 639 is not a stable standard; some of the language codes it defines
749 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
750 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
751 * API on Locale will return only the OLD codes.
752 * <li>For backward compatibility reasons, this constructor does not make
753 * any syntactic checks on the input.
754 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially,
755 * see <a href="#special_cases_constructor">Special Cases</a> for more information.
756 * </ul>
757 *
758 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
759 * up to 8 characters in length. See the <code>Locale</code> class description about
760 * valid language values.
761 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
762 * See the <code>Locale</code> class description about valid country values.
763 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>.
764 * See the <code>Locale</code> class description for the details.
765 * @exception NullPointerException thrown if any argument is null.
766 */
767 public Locale(String language, String country, String variant) {
768 if (language == null || country == null || variant == null) {
769 throw new NullPointerException();
770 }
771 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant);
772 localeExtensions = getCompatibilityExtensions(language, "", country, variant);
773 }
774
775 /**
776 * Construct a locale from language and country.
777 * This constructor normalizes the language value to lowercase and
778 * the country value to uppercase.
779 * <p>
780 * <b>Note:</b>
781 * <ul>
782 * <li>ISO 639 is not a stable standard; some of the language codes it defines
783 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
784 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
785 * API on Locale will return only the OLD codes.
786 * <li>For backward compatibility reasons, this constructor does not make
787 * any syntactic checks on the input.
788 * </ul>
789 *
790 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
791 * up to 8 characters in length. See the <code>Locale</code> class description about
792 * valid language values.
793 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
794 * See the <code>Locale</code> class description about valid country values.
795 * @exception NullPointerException thrown if either argument is null.
796 */
797 public Locale(String language, String country) {
798 this(language, country, "");
799 }
800
801 /**
802 * Construct a locale from a language code.
803 * This constructor normalizes the language value to lowercase.
804 * <p>
805 * <b>Note:</b>
806 * <ul>
807 * <li>ISO 639 is not a stable standard; some of the language codes it defines
808 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
809 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
810 * API on Locale will return only the OLD codes.
811 * <li>For backward compatibility reasons, this constructor does not make
812 * any syntactic checks on the input.
813 * </ul>
814 *
815 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
816 * up to 8 characters in length. See the <code>Locale</code> class description about
817 * valid language values.
818 * @exception NullPointerException thrown if argument is null.
819 * @since 1.4
820 */
821 public Locale(String language) {
822 this(language, "", "");
823 }
824
825 /**
826 * Returns a <code>Locale</code> constructed from the given
827 * <code>language</code>, <code>country</code> and
828 * <code>variant</code>. If the same <code>Locale</code> instance
829 * is available in the cache, then that instance is
830 * returned. Otherwise, a new <code>Locale</code> instance is
831 * created and cached.
832 *
833 * @param language lowercase 2 to 8 language code.
834 * @param country uppercase two-letter ISO-3166 code and numeric-3 UN M.49 area code.
835 * @param variant vendor and browser specific code. See class description.
836 * @return the <code>Locale</code> instance requested
837 * @exception NullPointerException if any argument is null.
838 */
839 static Locale getInstance(String language, String country, String variant) {
840 return getInstance(language, "", country, variant, null);
841 }
842
843 static Locale getInstance(String language, String script, String country,
844 String variant, LocaleExtensions extensions) {
845 if (language== null || script == null || country == null || variant == null) {
846 throw new NullPointerException();
847 }
848
849 if (extensions == null) {
850 extensions = getCompatibilityExtensions(language, script, country, variant);
851 }
852
853 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant);
854 return getInstance(baseloc, extensions);
855 }
856
857 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) {
858 if (extensions == null) {
859 Locale locale = CONSTANT_LOCALES.get(baseloc);
860 if (locale != null) {
861 return locale;
862 }
863 return Cache.LOCALECACHE.get(baseloc);
864 } else {
865 LocaleKey key = new LocaleKey(baseloc, extensions);
866 return Cache.LOCALECACHE.get(key);
867 }
868 }
869
870 private static class Cache extends LocaleObjectCache<Object, Locale> {
871
872 private static final Cache LOCALECACHE = new Cache();
873
874 private Cache() {
875 }
876
877 @Override
878 protected Locale createObject(Object key) {
879 if (key instanceof BaseLocale) {
880 return new Locale((BaseLocale)key, null);
881 } else {
882 LocaleKey lk = (LocaleKey)key;
883 return new Locale(lk.base, lk.exts);
884 }
885 }
886 }
887
888 private static final class LocaleKey {
889 private final BaseLocale base;
890 private final LocaleExtensions exts;
891 private final int hash;
892
893 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) {
894 base = baseLocale;
895 exts = extensions;
896
897 // Calculate the hash value here because it's always used.
898 int h = base.hashCode();
899 if (exts != null) {
900 h ^= exts.hashCode();
901 }
902 hash = h;
903 }
904
905 @Override
906 public boolean equals(Object obj) {
907 if (this == obj) {
908 return true;
909 }
910 if (!(obj instanceof LocaleKey)) {
911 return false;
912 }
913 LocaleKey other = (LocaleKey)obj;
914 if (hash != other.hash || !base.equals(other.base)) {
915 return false;
916 }
917 if (exts == null) {
918 return other.exts == null;
919 }
920 return exts.equals(other.exts);
921 }
922
923 @Override
924 public int hashCode() {
925 return hash;
926 }
927 }
928
929 /**
930 * Gets the current value of the default locale for this instance
931 * of the Java Virtual Machine.
932 * <p>
933 * The Java Virtual Machine sets the default locale during startup
934 * based on the host environment. It is used by many locale-sensitive
935 * methods if no locale is explicitly specified.
936 * It can be changed using the
937 * {@link #setDefault(java.util.Locale) setDefault} method.
938 *
939 * @return the default locale for this instance of the Java Virtual Machine
940 */
941 public static Locale getDefault() {
942 // do not synchronize this method - see 4071298
943 return defaultLocale;
944 }
945
946 /**
947 * Gets the current value of the default locale for the specified Category
948 * for this instance of the Java Virtual Machine.
949 * <p>
950 * The Java Virtual Machine sets the default locale during startup based
951 * on the host environment. It is used by many locale-sensitive methods
952 * if no locale is explicitly specified. It can be changed using the
953 * setDefault(Locale.Category, Locale) method.
954 *
955 * @param category - the specified category to get the default locale
956 * @throws NullPointerException if category is null
957 * @return the default locale for the specified Category for this instance
958 * of the Java Virtual Machine
959 * @see #setDefault(Locale.Category, Locale)
960 * @since 1.7
961 */
962 public static Locale getDefault(Locale.Category category) {
963 // do not synchronize this method - see 4071298
964 switch (category) {
965 case DISPLAY:
966 if (defaultDisplayLocale == null) {
967 synchronized(Locale.class) {
968 if (defaultDisplayLocale == null) {
969 defaultDisplayLocale = initDefault(category);
970 }
971 }
972 }
973 return defaultDisplayLocale;
974 case FORMAT:
975 if (defaultFormatLocale == null) {
976 synchronized(Locale.class) {
977 if (defaultFormatLocale == null) {
978 defaultFormatLocale = initDefault(category);
979 }
980 }
981 }
982 return defaultFormatLocale;
983 default:
984 assert false: "Unknown Category";
985 }
986 return getDefault();
987 }
988
989 private static Locale initDefault() {
990 String language, region, script, country, variant;
991 Properties props = GetPropertyAction.privilegedGetProperties();
992 language = props.getProperty("user.language", "en");
993 // for compatibility, check for old user.region property
994 region = props.getProperty("user.region");
995 if (region != null) {
996 // region can be of form country, country_variant, or _variant
997 int i = region.indexOf('_');
998 if (i >= 0) {
999 country = region.substring(0, i);
1000 variant = region.substring(i + 1);
1001 } else {
1002 country = region;
1003 variant = "";
1004 }
1005 script = "";
1006 } else {
1007 script = props.getProperty("user.script", "");
1008 country = props.getProperty("user.country", "");
1009 variant = props.getProperty("user.variant", "");
1010 }
1011
1012 return getInstance(language, script, country, variant,
1013 getDefaultExtensions(props.getProperty("user.extensions", ""))
1014 .orElse(null));
1015 }
1016
1017 private static Locale initDefault(Locale.Category category) {
1018 Properties props = GetPropertyAction.privilegedGetProperties();
1019
1020 return getInstance(
1021 props.getProperty(category.languageKey,
1022 defaultLocale.getLanguage()),
1023 props.getProperty(category.scriptKey,
1024 defaultLocale.getScript()),
1025 props.getProperty(category.countryKey,
1026 defaultLocale.getCountry()),
1027 props.getProperty(category.variantKey,
1028 defaultLocale.getVariant()),
1029 getDefaultExtensions(props.getProperty(category.extensionsKey, ""))
1030 .orElse(defaultLocale.getLocaleExtensions()));
1031 }
1032
1033 private static Optional<LocaleExtensions> getDefaultExtensions(String extensionsProp) {
1034 if (LocaleUtils.isEmpty(extensionsProp)) {
1035 return Optional.empty();
1036 }
1037
1038 LocaleExtensions exts = null;
1039 try {
1040 exts = new InternalLocaleBuilder()
1041 .setExtensions(extensionsProp)
1042 .getLocaleExtensions();
1043 } catch (LocaleSyntaxException e) {
1044 // just ignore this incorrect property
1045 }
1046
1047 return Optional.ofNullable(exts);
1048 }
1049
1050 /**
1051 * Sets the default locale for this instance of the Java Virtual Machine.
1052 * This does not affect the host locale.
1053 * <p>
1054 * If there is a security manager, its <code>checkPermission</code>
1055 * method is called with a <code>PropertyPermission("user.language", "write")</code>
1056 * permission before the default locale is changed.
1057 * <p>
1058 * The Java Virtual Machine sets the default locale during startup
1059 * based on the host environment. It is used by many locale-sensitive
1060 * methods if no locale is explicitly specified.
1061 * <p>
1062 * Since changing the default locale may affect many different areas
1063 * of functionality, this method should only be used if the caller
1064 * is prepared to reinitialize locale-sensitive code running
1065 * within the same Java Virtual Machine.
1066 * <p>
1067 * By setting the default locale with this method, all of the default
1068 * locales for each Category are also set to the specified default locale.
1069 *
1070 * @throws SecurityException
1071 * if a security manager exists and its
1072 * <code>checkPermission</code> method doesn't allow the operation.
1073 * @throws NullPointerException if <code>newLocale</code> is null
1074 * @param newLocale the new default locale
1075 * @see SecurityManager#checkPermission
1076 * @see java.util.PropertyPermission
1077 */
1078 public static synchronized void setDefault(Locale newLocale) {
1079 setDefault(Category.DISPLAY, newLocale);
1080 setDefault(Category.FORMAT, newLocale);
1081 defaultLocale = newLocale;
1082 }
1083
1084 /**
1085 * Sets the default locale for the specified Category for this instance
1086 * of the Java Virtual Machine. This does not affect the host locale.
1087 * <p>
1088 * If there is a security manager, its checkPermission method is called
1089 * with a PropertyPermission("user.language", "write") permission before
1090 * the default locale is changed.
1091 * <p>
1092 * The Java Virtual Machine sets the default locale during startup based
1093 * on the host environment. It is used by many locale-sensitive methods
1094 * if no locale is explicitly specified.
1095 * <p>
1096 * Since changing the default locale may affect many different areas of
1097 * functionality, this method should only be used if the caller is
1098 * prepared to reinitialize locale-sensitive code running within the
1099 * same Java Virtual Machine.
1100 *
1101 * @param category - the specified category to set the default locale
1102 * @param newLocale - the new default locale
1103 * @throws SecurityException if a security manager exists and its
1104 * checkPermission method doesn't allow the operation.
1105 * @throws NullPointerException if category and/or newLocale is null
1106 * @see SecurityManager#checkPermission(java.security.Permission)
1107 * @see PropertyPermission
1108 * @see #getDefault(Locale.Category)
1109 * @since 1.7
1110 */
1111 public static synchronized void setDefault(Locale.Category category,
1112 Locale newLocale) {
1113 if (category == null)
1114 throw new NullPointerException("Category cannot be NULL");
1115 if (newLocale == null)
1116 throw new NullPointerException("Can't set default locale to NULL");
1117
1118 SecurityManager sm = System.getSecurityManager();
1119 if (sm != null) sm.checkPermission(new PropertyPermission
1120 ("user.language", "write"));
1121 switch (category) {
1122 case DISPLAY:
1123 defaultDisplayLocale = newLocale;
1124 break;
1125 case FORMAT:
1126 defaultFormatLocale = newLocale;
1127 break;
1128 default:
1129 assert false: "Unknown Category";
1130 }
1131 }
1132
1133 /**
1134 * Returns an array of all installed locales.
1135 * The returned array represents the union of locales supported
1136 * by the Java runtime environment and by installed
1137 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider}
1138 * implementations. It must contain at least a <code>Locale</code>
1139 * instance equal to {@link java.util.Locale#US Locale.US}.
1140 *
1141 * @return An array of installed locales.
1142 */
1143 public static Locale[] getAvailableLocales() {
1144 return LocaleServiceProviderPool.getAllAvailableLocales();
1145 }
1146
1147 /**
1148 * Returns a list of all 2-letter country codes defined in ISO 3166.
1149 * Can be used to create Locales.
1150 * This method is equivalent to {@link #getISOCountries(Locale.IsoCountryCode type)}
1151 * with {@code type} {@link IsoCountryCode#PART1_ALPHA2}.
1152 * <p>
1153 * <b>Note:</b> The <code>Locale</code> class also supports other codes for
1154 * country (region), such as 3-letter numeric UN M.49 area codes.
1155 * Therefore, the list returned by this method does not contain ALL valid
1156 * codes that can be used to create Locales.
1157 * <p>
1158 * Note that this method does not return obsolete 2-letter country codes.
1159 * ISO3166-3 codes which designate country codes for those obsolete codes,
1160 * can be retrieved from {@link #getISOCountries(Locale.IsoCountryCode type)} with
1161 * {@code type} {@link IsoCountryCode#PART3}.
1162 * @return An array of ISO 3166 two-letter country codes.
1163 */
1164 public static String[] getISOCountries() {
1165 if (isoCountries == null) {
1166 isoCountries = getISO2Table(LocaleISOData.isoCountryTable);
1167 }
1168 String[] result = new String[isoCountries.length];
1169 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length);
1170 return result;
1171 }
1172
1173 /**
1174 * Returns a {@code Set} of ISO3166 country codes for the specified type.
1175 *
1176 * @param type {@link Locale.IsoCountryCode} specified ISO code type.
1177 * @see java.util.Locale.IsoCountryCode
1178 * @throws NullPointerException if type is null
1179 * @return a {@code Set} of ISO country codes for the specified type.
1180 * @since 9
1181 */
1182 public static Set<String> getISOCountries(IsoCountryCode type) {
1183 Objects.requireNonNull(type);
1184 return IsoCountryCode.retrieveISOCountryCodes(type);
1185 }
1186
1187 /**
1188 * Returns a list of all 2-letter language codes defined in ISO 639.
1189 * Can be used to create Locales.
1190 * <p>
1191 * <b>Note:</b>
1192 * <ul>
1193 * <li>ISO 639 is not a stable standard— some languages' codes have changed.
1194 * The list this function returns includes both the new and the old codes for the
1195 * languages whose codes have changed.
1196 * <li>The <code>Locale</code> class also supports language codes up to
1197 * 8 characters in length. Therefore, the list returned by this method does
1198 * not contain ALL valid codes that can be used to create Locales.
1199 * </ul>
1200 *
1201 * @return An array of ISO 639 two-letter language codes.
1202 */
1203 public static String[] getISOLanguages() {
1204 if (isoLanguages == null) {
1205 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable);
1206 }
1207 String[] result = new String[isoLanguages.length];
1208 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length);
1209 return result;
1210 }
1211
1212 private static String[] getISO2Table(String table) {
1213 int len = table.length() / 5;
1214 String[] isoTable = new String[len];
1215 for (int i = 0, j = 0; i < len; i++, j += 5) {
1216 isoTable[i] = table.substring(j, j + 2);
1217 }
1218 return isoTable;
1219 }
1220
1221 /**
1222 * Returns the language code of this Locale.
1223 *
1224 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed.
1225 * Locale's constructor recognizes both the new and the old codes for the languages
1226 * whose codes have changed, but this function always returns the old code. If you
1227 * want to check for a specific language whose code has changed, don't do
1228 * <pre>
1229 * if (locale.getLanguage().equals("he")) // BAD!
1230 * ...
1231 * </pre>
1232 * Instead, do
1233 * <pre>
1234 * if (locale.getLanguage().equals(new Locale("he").getLanguage()))
1235 * ...
1236 * </pre>
1237 * @return The language code, or the empty string if none is defined.
1238 * @see #getDisplayLanguage
1239 */
1240 public String getLanguage() {
1241 return baseLocale.getLanguage();
1242 }
1243
1244 /**
1245 * Returns the script for this locale, which should
1246 * either be the empty string or an ISO 15924 4-letter script
1247 * code. The first letter is uppercase and the rest are
1248 * lowercase, for example, 'Latn', 'Cyrl'.
1249 *
1250 * @return The script code, or the empty string if none is defined.
1251 * @see #getDisplayScript
1252 * @since 1.7
1253 */
1254 public String getScript() {
1255 return baseLocale.getScript();
1256 }
1257
1258 /**
1259 * Returns the country/region code for this locale, which should
1260 * either be the empty string, an uppercase ISO 3166 2-letter code,
1261 * or a UN M.49 3-digit code.
1262 *
1263 * @return The country/region code, or the empty string if none is defined.
1264 * @see #getDisplayCountry
1265 */
1266 public String getCountry() {
1267 return baseLocale.getRegion();
1268 }
1269
1270 /**
1271 * Returns the variant code for this locale.
1272 *
1273 * @return The variant code, or the empty string if none is defined.
1274 * @see #getDisplayVariant
1275 */
1276 public String getVariant() {
1277 return baseLocale.getVariant();
1278 }
1279
1280 /**
1281 * Returns {@code true} if this {@code Locale} has any <a href="#def_extensions">
1282 * extensions</a>.
1283 *
1284 * @return {@code true} if this {@code Locale} has any extensions
1285 * @since 1.8
1286 */
1287 public boolean hasExtensions() {
1288 return localeExtensions != null;
1289 }
1290
1291 /**
1292 * Returns a copy of this {@code Locale} with no <a href="#def_extensions">
1293 * extensions</a>. If this {@code Locale} has no extensions, this {@code Locale}
1294 * is returned.
1295 *
1296 * @return a copy of this {@code Locale} with no extensions, or {@code this}
1297 * if {@code this} has no extensions
1298 * @since 1.8
1299 */
1300 public Locale stripExtensions() {
1301 return hasExtensions() ? Locale.getInstance(baseLocale, null) : this;
1302 }
1303
1304 /**
1305 * Returns the extension (or private use) value associated with
1306 * the specified key, or null if there is no extension
1307 * associated with the key. To be well-formed, the key must be one
1308 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so
1309 * for example 'z' and 'Z' represent the same extension.
1310 *
1311 * @param key the extension key
1312 * @return The extension, or null if this locale defines no
1313 * extension for the specified key.
1314 * @throws IllegalArgumentException if key is not well-formed
1315 * @see #PRIVATE_USE_EXTENSION
1316 * @see #UNICODE_LOCALE_EXTENSION
1317 * @since 1.7
1318 */
1319 public String getExtension(char key) {
1320 if (!LocaleExtensions.isValidKey(key)) {
1321 throw new IllegalArgumentException("Ill-formed extension key: " + key);
1322 }
1323 return hasExtensions() ? localeExtensions.getExtensionValue(key) : null;
1324 }
1325
1326 /**
1327 * Returns the set of extension keys associated with this locale, or the
1328 * empty set if it has no extensions. The returned set is unmodifiable.
1329 * The keys will all be lower-case.
1330 *
1331 * @return The set of extension keys, or the empty set if this locale has
1332 * no extensions.
1333 * @since 1.7
1334 */
1335 public Set<Character> getExtensionKeys() {
1336 if (!hasExtensions()) {
1337 return Collections.emptySet();
1338 }
1339 return localeExtensions.getKeys();
1340 }
1341
1342 /**
1343 * Returns the set of unicode locale attributes associated with
1344 * this locale, or the empty set if it has no attributes. The
1345 * returned set is unmodifiable.
1346 *
1347 * @return The set of attributes.
1348 * @since 1.7
1349 */
1350 public Set<String> getUnicodeLocaleAttributes() {
1351 if (!hasExtensions()) {
1352 return Collections.emptySet();
1353 }
1354 return localeExtensions.getUnicodeLocaleAttributes();
1355 }
1356
1357 /**
1358 * Returns the Unicode locale type associated with the specified Unicode locale key
1359 * for this locale. Returns the empty string for keys that are defined with no type.
1360 * Returns null if the key is not defined. Keys are case-insensitive. The key must
1361 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is
1362 * thrown.
1363 *
1364 * @param key the Unicode locale key
1365 * @return The Unicode locale type associated with the key, or null if the
1366 * locale does not define the key.
1367 * @throws IllegalArgumentException if the key is not well-formed
1368 * @throws NullPointerException if <code>key</code> is null
1369 * @since 1.7
1370 */
1371 public String getUnicodeLocaleType(String key) {
1372 if (!isUnicodeExtensionKey(key)) {
1373 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key);
1374 }
1375 return hasExtensions() ? localeExtensions.getUnicodeLocaleType(key) : null;
1376 }
1377
1378 /**
1379 * Returns the set of Unicode locale keys defined by this locale, or the empty set if
1380 * this locale has none. The returned set is immutable. Keys are all lower case.
1381 *
1382 * @return The set of Unicode locale keys, or the empty set if this locale has
1383 * no Unicode locale keywords.
1384 * @since 1.7
1385 */
1386 public Set<String> getUnicodeLocaleKeys() {
1387 if (localeExtensions == null) {
1388 return Collections.emptySet();
1389 }
1390 return localeExtensions.getUnicodeLocaleKeys();
1391 }
1392
1393 /**
1394 * Package locale method returning the Locale's BaseLocale,
1395 * used by ResourceBundle
1396 * @return base locale of this Locale
1397 */
1398 BaseLocale getBaseLocale() {
1399 return baseLocale;
1400 }
1401
1402 /**
1403 * Package private method returning the Locale's LocaleExtensions,
1404 * used by ResourceBundle.
1405 * @return locale extensions of this Locale,
1406 * or {@code null} if no extensions are defined
1407 */
1408 LocaleExtensions getLocaleExtensions() {
1409 return localeExtensions;
1410 }
1411
1412 /**
1413 * Returns a string representation of this <code>Locale</code>
1414 * object, consisting of language, country, variant, script,
1415 * and extensions as below:
1416 * <blockquote>
1417 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "_" + extensions
1418 * </blockquote>
1419 *
1420 * Language is always lower case, country is always upper case, script is always title
1421 * case, and extensions are always lower case. Extensions and private use subtags
1422 * will be in canonical order as explained in {@link #toLanguageTag}.
1423 *
1424 * <p>When the locale has neither script nor extensions, the result is the same as in
1425 * Java 6 and prior.
1426 *
1427 * <p>If both the language and country fields are missing, this function will return
1428 * the empty string, even if the variant, script, or extensions field is present (you
1429 * can't have a locale with just a variant, the variant must accompany a well-formed
1430 * language or country code).
1431 *
1432 * <p>If script or extensions are present and variant is missing, no underscore is
1433 * added before the "#".
1434 *
1435 * <p>This behavior is designed to support debugging and to be compatible with
1436 * previous uses of <code>toString</code> that expected language, country, and variant
1437 * fields only. To represent a Locale as a String for interchange purposes, use
1438 * {@link #toLanguageTag}.
1439 *
1440 * <p>Examples: <ul>
1441 * <li>{@code en}</li>
1442 * <li>{@code de_DE}</li>
1443 * <li>{@code _GB}</li>
1444 * <li>{@code en_US_WIN}</li>
1445 * <li>{@code de__POSIX}</li>
1446 * <li>{@code zh_CN_#Hans}</li>
1447 * <li>{@code zh_TW_#Hant_x-java}</li>
1448 * <li>{@code th_TH_TH_#u-nu-thai}</li></ul>
1449 *
1450 * @return A string representation of the Locale, for debugging.
1451 * @see #getDisplayName
1452 * @see #toLanguageTag
1453 */
1454 @Override
1455 public final String toString() {
1456 boolean l = !baseLocale.getLanguage().isEmpty();
1457 boolean s = !baseLocale.getScript().isEmpty();
1458 boolean r = !baseLocale.getRegion().isEmpty();
1459 boolean v = !baseLocale.getVariant().isEmpty();
1460 boolean e = localeExtensions != null && !localeExtensions.getID().isEmpty();
1461
1462 StringBuilder result = new StringBuilder(baseLocale.getLanguage());
1463 if (r || (l && (v || s || e))) {
1464 result.append('_')
1465 .append(baseLocale.getRegion()); // This may just append '_'
1466 }
1467 if (v && (l || r)) {
1468 result.append('_')
1469 .append(baseLocale.getVariant());
1470 }
1471
1472 if (s && (l || r)) {
1473 result.append("_#")
1474 .append(baseLocale.getScript());
1475 }
1476
1477 if (e && (l || r)) {
1478 result.append('_');
1479 if (!s) {
1480 result.append('#');
1481 }
1482 result.append(localeExtensions.getID());
1483 }
1484
1485 return result.toString();
1486 }
1487
1488 /**
1489 * Returns a well-formed IETF BCP 47 language tag representing
1490 * this locale.
1491 *
1492 * <p>If this <code>Locale</code> has a language, country, or
1493 * variant that does not satisfy the IETF BCP 47 language tag
1494 * syntax requirements, this method handles these fields as
1495 * described below:
1496 *
1497 * <p><b>Language:</b> If language is empty, or not <a
1498 * href="#def_language" >well-formed</a> (for example "a" or
1499 * "e2"), it will be emitted as "und" (Undetermined).
1500 *
1501 * <p><b>Country:</b> If country is not <a
1502 * href="#def_region">well-formed</a> (for example "12" or "USA"),
1503 * it will be omitted.
1504 *
1505 * <p><b>Variant:</b> If variant <b>is</b> <a
1506 * href="#def_variant">well-formed</a>, each sub-segment
1507 * (delimited by '-' or '_') is emitted as a subtag. Otherwise:
1508 * <ul>
1509 *
1510 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code>
1511 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first
1512 * ill-formed sub-segment and all following will be appended to
1513 * the private use subtag. The first appended subtag will be
1514 * "lvariant", followed by the sub-segments in order, separated by
1515 * hyphen. For example, "x-lvariant-WIN",
1516 * "Oracle-x-lvariant-JDK-Standard-Edition".
1517 *
1518 * <li>if any sub-segment does not match
1519 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated
1520 * and the problematic sub-segment and all following sub-segments
1521 * will be omitted. If the remainder is non-empty, it will be
1522 * emitted as a private use subtag as above (even if the remainder
1523 * turns out to be well-formed). For example,
1524 * "Solaris_isjustthecoolestthing" is emitted as
1525 * "x-lvariant-Solaris", not as "solaris".</li></ul>
1526 *
1527 * <p><b>Special Conversions:</b> Java supports some old locale
1528 * representations, including deprecated ISO language codes,
1529 * for compatibility. This method performs the following
1530 * conversions:
1531 * <ul>
1532 *
1533 * <li>Deprecated ISO language codes "iw", "ji", and "in" are
1534 * converted to "he", "yi", and "id", respectively.
1535 *
1536 * <li>A locale with language "no", country "NO", and variant
1537 * "NY", representing Norwegian Nynorsk (Norway), is converted
1538 * to a language tag "nn-NO".</li></ul>
1539 *
1540 * <p><b>Note:</b> Although the language tag created by this
1541 * method is well-formed (satisfies the syntax requirements
1542 * defined by the IETF BCP 47 specification), it is not
1543 * necessarily a valid BCP 47 language tag. For example,
1544 * <pre>
1545 * new Locale("xx", "YY").toLanguageTag();</pre>
1546 *
1547 * will return "xx-YY", but the language subtag "xx" and the
1548 * region subtag "YY" are invalid because they are not registered
1549 * in the IANA Language Subtag Registry.
1550 *
1551 * @return a BCP47 language tag representing the locale
1552 * @see #forLanguageTag(String)
1553 * @since 1.7
1554 */
1555 public String toLanguageTag() {
1556 if (languageTag != null) {
1557 return languageTag;
1558 }
1559
1560 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions);
1561 StringBuilder buf = new StringBuilder();
1562
1563 String subtag = tag.getLanguage();
1564 if (!subtag.isEmpty()) {
1565 buf.append(LanguageTag.canonicalizeLanguage(subtag));
1566 }
1567
1568 subtag = tag.getScript();
1569 if (!subtag.isEmpty()) {
1570 buf.append(LanguageTag.SEP);
1571 buf.append(LanguageTag.canonicalizeScript(subtag));
1572 }
1573
1574 subtag = tag.getRegion();
1575 if (!subtag.isEmpty()) {
1576 buf.append(LanguageTag.SEP);
1577 buf.append(LanguageTag.canonicalizeRegion(subtag));
1578 }
1579
1580 List<String>subtags = tag.getVariants();
1581 for (String s : subtags) {
1582 buf.append(LanguageTag.SEP);
1583 // preserve casing
1584 buf.append(s);
1585 }
1586
1587 subtags = tag.getExtensions();
1588 for (String s : subtags) {
1589 buf.append(LanguageTag.SEP);
1590 buf.append(LanguageTag.canonicalizeExtension(s));
1591 }
1592
1593 subtag = tag.getPrivateuse();
1594 if (!subtag.isEmpty()) {
1595 if (buf.length() > 0) {
1596 buf.append(LanguageTag.SEP);
1597 }
1598 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP);
1599 // preserve casing
1600 buf.append(subtag);
1601 }
1602
1603 String langTag = buf.toString();
1604 synchronized (this) {
1605 if (languageTag == null) {
1606 languageTag = langTag;
1607 }
1608 }
1609 return languageTag;
1610 }
1611
1612 /**
1613 * Returns a locale for the specified IETF BCP 47 language tag string.
1614 *
1615 * <p>If the specified language tag contains any ill-formed subtags,
1616 * the first such subtag and all following subtags are ignored. Compare
1617 * to {@link Locale.Builder#setLanguageTag} which throws an exception
1618 * in this case.
1619 *
1620 * <p>The following <b>conversions</b> are performed:<ul>
1621 *
1622 * <li>The language code "und" is mapped to language "".
1623 *
1624 * <li>The language codes "he", "yi", and "id" are mapped to "iw",
1625 * "ji", and "in" respectively. (This is the same canonicalization
1626 * that's done in Locale's constructors.)
1627 *
1628 * <li>The portion of a private use subtag prefixed by "lvariant",
1629 * if any, is removed and appended to the variant field in the
1630 * result locale (without case normalization). If it is then
1631 * empty, the private use subtag is discarded:
1632 *
1633 * <pre>
1634 * Locale loc;
1635 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX");
1636 * loc.getVariant(); // returns "POSIX"
1637 * loc.getExtension('x'); // returns null
1638 *
1639 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def");
1640 * loc.getVariant(); // returns "POSIX_Abc_Def"
1641 * loc.getExtension('x'); // returns "urp"
1642 * </pre>
1643 *
1644 * <li>When the languageTag argument contains an extlang subtag,
1645 * the first such subtag is used as the language, and the primary
1646 * language subtag and other extlang subtags are ignored:
1647 *
1648 * <pre>
1649 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao"
1650 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US"
1651 * </pre>
1652 *
1653 * <li>Case is normalized except for variant tags, which are left
1654 * unchanged. Language is normalized to lower case, script to
1655 * title case, country to upper case, and extensions to lower
1656 * case.
1657 *
1658 * <li>If, after processing, the locale would exactly match either
1659 * ja_JP_JP or th_TH_TH with no extensions, the appropriate
1660 * extensions are added as though the constructor had been called:
1661 *
1662 * <pre>
1663 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag();
1664 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP"
1665 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag();
1666 * // returns "th-TH-u-nu-thai-x-lvariant-TH"
1667 * </pre></ul>
1668 *
1669 * <p>This implements the 'Language-Tag' production of BCP47, and
1670 * so supports grandfathered (regular and irregular) as well as
1671 * private use language tags. Stand alone private use tags are
1672 * represented as empty language and extension 'x-whatever',
1673 * and grandfathered tags are converted to their canonical replacements
1674 * where they exist.
1675 *
1676 * <p>Grandfathered tags with canonical replacements are as follows:
1677 *
1678 * <table class="striped">
1679 * <caption style="display:none">Grandfathered tags with canonical replacements</caption>
1680 * <thead style="text-align:center">
1681 * <tr><th scope="col" style="padding: 0 2px">grandfathered tag</th><th scope="col" style="padding: 0 2px">modern replacement</th></tr>
1682 * </thead>
1683 * <tbody style="text-align:center">
1684 * <tr><th scope="row">art-lojban</th><td>jbo</td></tr>
1685 * <tr><th scope="row">i-ami</th><td>ami</td></tr>
1686 * <tr><th scope="row">i-bnn</th><td>bnn</td></tr>
1687 * <tr><th scope="row">i-hak</th><td>hak</td></tr>
1688 * <tr><th scope="row">i-klingon</th><td>tlh</td></tr>
1689 * <tr><th scope="row">i-lux</th><td>lb</td></tr>
1690 * <tr><th scope="row">i-navajo</th><td>nv</td></tr>
1691 * <tr><th scope="row">i-pwn</th><td>pwn</td></tr>
1692 * <tr><th scope="row">i-tao</th><td>tao</td></tr>
1693 * <tr><th scope="row">i-tay</th><td>tay</td></tr>
1694 * <tr><th scope="row">i-tsu</th><td>tsu</td></tr>
1695 * <tr><th scope="row">no-bok</th><td>nb</td></tr>
1696 * <tr><th scope="row">no-nyn</th><td>nn</td></tr>
1697 * <tr><th scope="row">sgn-BE-FR</th><td>sfb</td></tr>
1698 * <tr><th scope="row">sgn-BE-NL</th><td>vgt</td></tr>
1699 * <tr><th scope="row">sgn-CH-DE</th><td>sgg</td></tr>
1700 * <tr><th scope="row">zh-guoyu</th><td>cmn</td></tr>
1701 * <tr><th scope="row">zh-hakka</th><td>hak</td></tr>
1702 * <tr><th scope="row">zh-min-nan</th><td>nan</td></tr>
1703 * <tr><th scope="row">zh-xiang</th><td>hsn</td></tr>
1704 * </tbody>
1705 * </table>
1706 *
1707 * <p>Grandfathered tags with no modern replacement will be
1708 * converted as follows:
1709 *
1710 * <table class="striped">
1711 * <caption style="display:none">Grandfathered tags with no modern replacement</caption>
1712 * <thead style="text-align:center">
1713 * <tr><th scope="col" style="padding: 0 2px">grandfathered tag</th><th scope="col" style="padding: 0 2px">converts to</th></tr>
1714 * </thead>
1715 * <tbody style="text-align:center">
1716 * <tr><th scope="row">cel-gaulish</th><td>xtg-x-cel-gaulish</td></tr>
1717 * <tr><th scope="row">en-GB-oed</th><td>en-GB-x-oed</td></tr>
1718 * <tr><th scope="row">i-default</th><td>en-x-i-default</td></tr>
1719 * <tr><th scope="row">i-enochian</th><td>und-x-i-enochian</td></tr>
1720 * <tr><th scope="row">i-mingo</th><td>see-x-i-mingo</td></tr>
1721 * <tr><th scope="row">zh-min</th><td>nan-x-zh-min</td></tr>
1722 * </tbody>
1723 * </table>
1724 *
1725 * <p>For a list of all grandfathered tags, see the
1726 * IANA Language Subtag Registry (search for "Type: grandfathered").
1727 *
1728 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code>
1729 * and <code>forLanguageTag</code> will round-trip.
1730 *
1731 * @param languageTag the language tag
1732 * @return The locale that best represents the language tag.
1733 * @throws NullPointerException if <code>languageTag</code> is <code>null</code>
1734 * @see #toLanguageTag()
1735 * @see java.util.Locale.Builder#setLanguageTag(String)
1736 * @since 1.7
1737 */
1738 public static Locale forLanguageTag(String languageTag) {
1739 LanguageTag tag = LanguageTag.parse(languageTag, null);
1740 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
1741 bldr.setLanguageTag(tag);
1742 BaseLocale base = bldr.getBaseLocale();
1743 LocaleExtensions exts = bldr.getLocaleExtensions();
1744 if (exts == null && !base.getVariant().isEmpty()) {
1745 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(),
1746 base.getRegion(), base.getVariant());
1747 }
1748 return getInstance(base, exts);
1749 }
1750
1751 /**
1752 * Returns a three-letter abbreviation of this locale's language.
1753 * If the language matches an ISO 639-1 two-letter code, the
1754 * corresponding ISO 639-2/T three-letter lowercase code is
1755 * returned. The ISO 639-2 language codes can be found on-line,
1756 * see "Codes for the Representation of Names of Languages Part 2:
1757 * Alpha-3 Code". If the locale specifies a three-letter
1758 * language, the language is returned as is. If the locale does
1759 * not specify a language the empty string is returned.
1760 *
1761 * @return A three-letter abbreviation of this locale's language.
1762 * @exception MissingResourceException Throws MissingResourceException if
1763 * three-letter language abbreviation is not available for this locale.
1764 */
1765 public String getISO3Language() throws MissingResourceException {
1766 String lang = baseLocale.getLanguage();
1767 if (lang.length() == 3) {
1768 return lang;
1769 }
1770
1771 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable);
1772 if (language3 == null) {
1773 throw new MissingResourceException("Couldn't find 3-letter language code for "
1774 + lang, "FormatData_" + toString(), "ShortLanguage");
1775 }
1776 return language3;
1777 }
1778
1779 /**
1780 * Returns a three-letter abbreviation for this locale's country.
1781 * If the country matches an ISO 3166-1 alpha-2 code, the
1782 * corresponding ISO 3166-1 alpha-3 uppercase code is returned.
1783 * If the locale doesn't specify a country, this will be the empty
1784 * string.
1785 *
1786 * <p>The ISO 3166-1 codes can be found on-line.
1787 *
1788 * @return A three-letter abbreviation of this locale's country.
1789 * @exception MissingResourceException Throws MissingResourceException if the
1790 * three-letter country abbreviation is not available for this locale.
1791 */
1792 public String getISO3Country() throws MissingResourceException {
1793 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable);
1794 if (country3 == null) {
1795 throw new MissingResourceException("Couldn't find 3-letter country code for "
1796 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry");
1797 }
1798 return country3;
1799 }
1800
1801 private static String getISO3Code(String iso2Code, String table) {
1802 int codeLength = iso2Code.length();
1803 if (codeLength == 0) {
1804 return "";
1805 }
1806
1807 int tableLength = table.length();
1808 int index = tableLength;
1809 if (codeLength == 2) {
1810 char c1 = iso2Code.charAt(0);
1811 char c2 = iso2Code.charAt(1);
1812 for (index = 0; index < tableLength; index += 5) {
1813 if (table.charAt(index) == c1
1814 && table.charAt(index + 1) == c2) {
1815 break;
1816 }
1817 }
1818 }
1819 return index < tableLength ? table.substring(index + 2, index + 5) : null;
1820 }
1821
1822 /**
1823 * Returns a name for the locale's language that is appropriate for display to the
1824 * user.
1825 * If possible, the name returned will be localized for the default
1826 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1827 * For example, if the locale is fr_FR and the default
1828 * {@link Locale.Category#DISPLAY DISPLAY} locale
1829 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1830 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1831 * getDisplayLanguage() will return "anglais".
1832 * If the name returned cannot be localized for the default
1833 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1834 * (say, we don't have a Japanese name for Croatian),
1835 * this function falls back on the English name, and uses the ISO code as a last-resort
1836 * value. If the locale doesn't specify a language, this function returns the empty string.
1837 *
1838 * @return The name of the display language.
1839 */
1840 public final String getDisplayLanguage() {
1841 return getDisplayLanguage(getDefault(Category.DISPLAY));
1842 }
1843
1844 /**
1845 * Returns a name for the locale's language that is appropriate for display to the
1846 * user.
1847 * If possible, the name returned will be localized according to inLocale.
1848 * For example, if the locale is fr_FR and inLocale
1849 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1850 * inLocale is fr_FR, getDisplayLanguage() will return "anglais".
1851 * If the name returned cannot be localized according to inLocale,
1852 * (say, we don't have a Japanese name for Croatian),
1853 * this function falls back on the English name, and finally
1854 * on the ISO code as a last-resort value. If the locale doesn't specify a language,
1855 * this function returns the empty string.
1856 *
1857 * @param inLocale The locale for which to retrieve the display language.
1858 * @return The name of the display language appropriate to the given locale.
1859 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1860 */
1861 public String getDisplayLanguage(Locale inLocale) {
1862 return getDisplayString(baseLocale.getLanguage(), null, inLocale, DISPLAY_LANGUAGE);
1863 }
1864
1865 /**
1866 * Returns a name for the locale's script that is appropriate for display to
1867 * the user. If possible, the name will be localized for the default
1868 * {@link Locale.Category#DISPLAY DISPLAY} locale. Returns
1869 * the empty string if this locale doesn't specify a script code.
1870 *
1871 * @return the display name of the script code for the current default
1872 * {@link Locale.Category#DISPLAY DISPLAY} locale
1873 * @since 1.7
1874 */
1875 public String getDisplayScript() {
1876 return getDisplayScript(getDefault(Category.DISPLAY));
1877 }
1878
1879 /**
1880 * Returns a name for the locale's script that is appropriate
1881 * for display to the user. If possible, the name will be
1882 * localized for the given locale. Returns the empty string if
1883 * this locale doesn't specify a script code.
1884 *
1885 * @param inLocale The locale for which to retrieve the display script.
1886 * @return the display name of the script code for the current default
1887 * {@link Locale.Category#DISPLAY DISPLAY} locale
1888 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1889 * @since 1.7
1890 */
1891 public String getDisplayScript(Locale inLocale) {
1892 return getDisplayString(baseLocale.getScript(), null, inLocale, DISPLAY_SCRIPT);
1893 }
1894
1895 /**
1896 * Returns a name for the locale's country that is appropriate for display to the
1897 * user.
1898 * If possible, the name returned will be localized for the default
1899 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1900 * For example, if the locale is fr_FR and the default
1901 * {@link Locale.Category#DISPLAY DISPLAY} locale
1902 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1903 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1904 * getDisplayCountry() will return "Etats-Unis".
1905 * If the name returned cannot be localized for the default
1906 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1907 * (say, we don't have a Japanese name for Croatia),
1908 * this function falls back on the English name, and uses the ISO code as a last-resort
1909 * value. If the locale doesn't specify a country, this function returns the empty string.
1910 *
1911 * @return The name of the country appropriate to the locale.
1912 */
1913 public final String getDisplayCountry() {
1914 return getDisplayCountry(getDefault(Category.DISPLAY));
1915 }
1916
1917 /**
1918 * Returns a name for the locale's country that is appropriate for display to the
1919 * user.
1920 * If possible, the name returned will be localized according to inLocale.
1921 * For example, if the locale is fr_FR and inLocale
1922 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1923 * inLocale is fr_FR, getDisplayCountry() will return "Etats-Unis".
1924 * If the name returned cannot be localized according to inLocale.
1925 * (say, we don't have a Japanese name for Croatia),
1926 * this function falls back on the English name, and finally
1927 * on the ISO code as a last-resort value. If the locale doesn't specify a country,
1928 * this function returns the empty string.
1929 *
1930 * @param inLocale The locale for which to retrieve the display country.
1931 * @return The name of the country appropriate to the given locale.
1932 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1933 */
1934 public String getDisplayCountry(Locale inLocale) {
1935 return getDisplayString(baseLocale.getRegion(), null, inLocale, DISPLAY_COUNTRY);
1936 }
1937
1938 private String getDisplayString(String code, String cat, Locale inLocale, int type) {
1939 Objects.requireNonNull(inLocale);
1940 Objects.requireNonNull(code);
1941
1942 if (code.isEmpty()) {
1943 return "";
1944 }
1945
1946 LocaleServiceProviderPool pool =
1947 LocaleServiceProviderPool.getPool(LocaleNameProvider.class);
1948 String rbKey = (type == DISPLAY_VARIANT ? "%%"+code : code);
1949 String result = pool.getLocalizedObject(
1950 LocaleNameGetter.INSTANCE,
1951 inLocale, rbKey, type, code, cat);
1952 return result != null ? result : code;
1953 }
1954
1955 /**
1956 * Returns a name for the locale's variant code that is appropriate for display to the
1957 * user. If possible, the name will be localized for the default
1958 * {@link Locale.Category#DISPLAY DISPLAY} locale. If the locale
1959 * doesn't specify a variant code, this function returns the empty string.
1960 *
1961 * @return The name of the display variant code appropriate to the locale.
1962 */
1963 public final String getDisplayVariant() {
1964 return getDisplayVariant(getDefault(Category.DISPLAY));
1965 }
1966
1967 /**
1968 * Returns a name for the locale's variant code that is appropriate for display to the
1969 * user. If possible, the name will be localized for inLocale. If the locale
1970 * doesn't specify a variant code, this function returns the empty string.
1971 *
1972 * @param inLocale The locale for which to retrieve the display variant code.
1973 * @return The name of the display variant code appropriate to the given locale.
1974 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1975 */
1976 public String getDisplayVariant(Locale inLocale) {
1977 if (baseLocale.getVariant().isEmpty())
1978 return "";
1979
1980 LocaleResources lr = LocaleProviderAdapter
1981 .getResourceBundleBased()
1982 .getLocaleResources(inLocale);
1983
1984 String names[] = getDisplayVariantArray(inLocale);
1985
1986 // Get the localized patterns for formatting a list, and use
1987 // them to format the list.
1988 return formatList(names,
1989 lr.getLocaleName("ListCompositionPattern"));
1990 }
1991
1992 /**
1993 * Returns a name for the locale that is appropriate for display to the
1994 * user. This will be the values returned by getDisplayLanguage(),
1995 * getDisplayScript(), getDisplayCountry(), getDisplayVariant() and
1996 * optional <a href="./Locale.html#def_locale_extension">Unicode extensions</a>
1997 * assembled into a single string. The non-empty values are used in order, with
1998 * the second and subsequent names in parentheses. For example:
1999 * <blockquote>
2000 * language (script, country, variant(, extension)*)<br>
2001 * language (country(, extension)*)<br>
2002 * language (variant(, extension)*)<br>
2003 * script (country(, extension)*)<br>
2004 * country (extension)*<br>
2005 * </blockquote>
2006 * depending on which fields are specified in the locale. The field
2007 * separator in the above parentheses, denoted as a comma character, may
2008 * be localized depending on the locale. If the language, script, country,
2009 * and variant fields are all empty, this function returns the empty string.
2010 *
2011 * @return The name of the locale appropriate to display.
2012 */
2013 public final String getDisplayName() {
2014 return getDisplayName(getDefault(Category.DISPLAY));
2015 }
2016
2017 /**
2018 * Returns a name for the locale that is appropriate for display
2019 * to the user. This will be the values returned by
2020 * getDisplayLanguage(), getDisplayScript(),getDisplayCountry()
2021 * getDisplayVariant(), and optional <a href="./Locale.html#def_locale_extension">
2022 * Unicode extensions</a> assembled into a single string. The non-empty
2023 * values are used in order, with the second and subsequent names in
2024 * parentheses. For example:
2025 * <blockquote>
2026 * language (script, country, variant(, extension)*)<br>
2027 * language (country(, extension)*)<br>
2028 * language (variant(, extension)*)<br>
2029 * script (country(, extension)*)<br>
2030 * country (extension)*<br>
2031 * </blockquote>
2032 * depending on which fields are specified in the locale. The field
2033 * separator in the above parentheses, denoted as a comma character, may
2034 * be localized depending on the locale. If the language, script, country,
2035 * and variant fields are all empty, this function returns the empty string.
2036 *
2037 * @param inLocale The locale for which to retrieve the display name.
2038 * @return The name of the locale appropriate to display.
2039 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
2040 */
2041 public String getDisplayName(Locale inLocale) {
2042 LocaleResources lr = LocaleProviderAdapter
2043 .getResourceBundleBased()
2044 .getLocaleResources(inLocale);
2045
2046 String languageName = getDisplayLanguage(inLocale);
2047 String scriptName = getDisplayScript(inLocale);
2048 String countryName = getDisplayCountry(inLocale);
2049 String[] variantNames = getDisplayVariantArray(inLocale);
2050
2051 // Get the localized patterns for formatting a display name.
2052 String displayNamePattern = lr.getLocaleName("DisplayNamePattern");
2053 String listCompositionPattern = lr.getLocaleName("ListCompositionPattern");
2054
2055 // The display name consists of a main name, followed by qualifiers.
2056 // Typically, the format is "MainName (Qualifier, Qualifier)" but this
2057 // depends on what pattern is stored in the display locale.
2058 String mainName;
2059 String[] qualifierNames;
2060
2061 // The main name is the language, or if there is no language, the script,
2062 // then if no script, the country. If there is no language/script/country
2063 // (an anomalous situation) then the display name is simply the variant's
2064 // display name.
2065 if (languageName.isEmpty() && scriptName.isEmpty() && countryName.isEmpty()) {
2066 if (variantNames.length == 0) {
2067 return "";
2068 } else {
2069 return formatList(variantNames, listCompositionPattern);
2070 }
2071 }
2072 ArrayList<String> names = new ArrayList<>(4);
2073 if (!languageName.isEmpty()) {
2074 names.add(languageName);
2075 }
2076 if (!scriptName.isEmpty()) {
2077 names.add(scriptName);
2078 }
2079 if (!countryName.isEmpty()) {
2080 names.add(countryName);
2081 }
2082 if (variantNames.length != 0) {
2083 names.addAll(Arrays.asList(variantNames));
2084 }
2085
2086 // add Unicode extensions
2087 if (localeExtensions != null) {
2088 localeExtensions.getUnicodeLocaleAttributes().stream()
2089 .map(key -> getDisplayString(key, null, inLocale, DISPLAY_UEXT_KEY))
2090 .forEach(names::add);
2091 localeExtensions.getUnicodeLocaleKeys().stream()
2092 .map(key -> getDisplayKeyTypeExtensionString(key, lr, inLocale))
2093 .forEach(names::add);
2094 }
2095
2096 // The first one in the main name
2097 mainName = names.get(0);
2098
2099 // Others are qualifiers
2100 int numNames = names.size();
2101 qualifierNames = (numNames > 1) ?
2102 names.subList(1, numNames).toArray(new String[numNames - 1]) : new String[0];
2103
2104 // Create an array whose first element is the number of remaining
2105 // elements. This serves as a selector into a ChoiceFormat pattern from
2106 // the resource. The second and third elements are the main name and
2107 // the qualifier; if there are no qualifiers, the third element is
2108 // unused by the format pattern.
2109 Object[] displayNames = {
2110 qualifierNames.length != 0 ? 2 : 1,
2111 mainName,
2112 // We could also just call formatList() and have it handle the empty
2113 // list case, but this is more efficient, and we want it to be
2114 // efficient since all the language-only locales will not have any
2115 // qualifiers.
2116 qualifierNames.length != 0 ? formatList(qualifierNames, listCompositionPattern) : null
2117 };
2118
2119 if (displayNamePattern != null) {
2120 return new MessageFormat(displayNamePattern).format(displayNames);
2121 }
2122 else {
2123 // If we cannot get the message format pattern, then we use a simple
2124 // hard-coded pattern. This should not occur in practice unless the
2125 // installation is missing some core files (FormatData etc.).
2126 StringBuilder result = new StringBuilder();
2127 result.append((String)displayNames[1]);
2128 if (displayNames.length > 2) {
2129 result.append(" (");
2130 result.append((String)displayNames[2]);
2131 result.append(')');
2132 }
2133 return result.toString();
2134 }
2135 }
2136
2137 /**
2138 * Overrides Cloneable.
2139 */
2140 @Override
2141 public Object clone()
2142 {
2143 try {
2144 Locale that = (Locale)super.clone();
2145 return that;
2146 } catch (CloneNotSupportedException e) {
2147 throw new InternalError(e);
2148 }
2149 }
2150
2151 /**
2152 * Override hashCode.
2153 * Since Locales are often used in hashtables, caches the value
2154 * for speed.
2155 */
2156 @Override
2157 public int hashCode() {
2158 int hc = hashCodeValue;
2159 if (hc == 0) {
2160 hc = baseLocale.hashCode();
2161 if (localeExtensions != null) {
2162 hc ^= localeExtensions.hashCode();
2163 }
2164 hashCodeValue = hc;
2165 }
2166 return hc;
2167 }
2168
2169 // Overrides
2170
2171 /**
2172 * Returns true if this Locale is equal to another object. A Locale is
2173 * deemed equal to another Locale with identical language, script, country,
2174 * variant and extensions, and unequal to all other objects.
2175 *
2176 * @return true if this Locale is equal to the specified object.
2177 */
2178 @Override
2179 public boolean equals(Object obj) {
2180 if (this == obj) // quick check
2181 return true;
2182 if (!(obj instanceof Locale))
2183 return false;
2184 BaseLocale otherBase = ((Locale)obj).baseLocale;
2185 if (!baseLocale.equals(otherBase)) {
2186 return false;
2187 }
2188 if (localeExtensions == null) {
2189 return ((Locale)obj).localeExtensions == null;
2190 }
2191 return localeExtensions.equals(((Locale)obj).localeExtensions);
2192 }
2193
2194 // ================= privates =====================================
2195
2196 private transient BaseLocale baseLocale;
2197 private transient LocaleExtensions localeExtensions;
2198
2199 /**
2200 * Calculated hashcode
2201 */
2202 private transient volatile int hashCodeValue;
2203
2204 private static volatile Locale defaultLocale = initDefault();
2205 private static volatile Locale defaultDisplayLocale;
2206 private static volatile Locale defaultFormatLocale;
2207
2208 private transient volatile String languageTag;
2209
2210 /**
2211 * Return an array of the display names of the variant.
2212 * @param bundle the ResourceBundle to use to get the display names
2213 * @return an array of display names, possible of zero length.
2214 */
2215 private String[] getDisplayVariantArray(Locale inLocale) {
2216 // Split the variant name into tokens separated by '_'.
2217 StringTokenizer tokenizer = new StringTokenizer(baseLocale.getVariant(), "_");
2218 String[] names = new String[tokenizer.countTokens()];
2219
2220 // For each variant token, lookup the display name. If
2221 // not found, use the variant name itself.
2222 for (int i=0; i<names.length; ++i) {
2223 names[i] = getDisplayString(tokenizer.nextToken(), null,
2224 inLocale, DISPLAY_VARIANT);
2225 }
2226
2227 return names;
2228 }
2229
2230 private String getDisplayKeyTypeExtensionString(String key, LocaleResources lr, Locale inLocale) {
2231 String type = localeExtensions.getUnicodeLocaleType(key);
2232 String ret = getDisplayString(type, key, inLocale, DISPLAY_UEXT_TYPE);
2233
2234 if (ret == null || ret.equals(type)) {
2235 // no localization for this type. try combining key/type separately
2236 String displayType = type;
2237 switch (key) {
2238 case "cu":
2239 displayType = lr.getCurrencyName(type.toLowerCase(Locale.ROOT));
2240 break;
2241 case "rg":
2242 if (type != null &&
2243 // UN M.49 code should not be allowed here
2244 type.matches("^[a-zA-Z]{2}[zZ]{4}$")) {
2245 displayType = lr.getLocaleName(type.substring(0, 2).toUpperCase(Locale.ROOT));
2246 }
2247 break;
2248 case "tz":
2249 displayType = TimeZoneNameUtility.convertLDMLShortID(type)
2250 .map(id -> TimeZoneNameUtility.retrieveGenericDisplayName(id, TimeZone.LONG, inLocale))
2251 .orElse(type);
2252 break;
2253 }
2254 ret = MessageFormat.format(lr.getLocaleName("ListKeyTypePattern"),
2255 getDisplayString(key, null, inLocale, DISPLAY_UEXT_KEY),
2256 Optional.ofNullable(displayType).orElse(type));
2257 }
2258
2259 return ret;
2260 }
2261
2262 /**
2263 * Format a list using given pattern strings.
2264 * If either of the patterns is null, then a the list is
2265 * formatted by concatenation with the delimiter ','.
2266 * @param stringList the list of strings to be formatted.
2267 * and formatting them into a list.
2268 * @param pattern should take 2 arguments for reduction
2269 * @return a string representing the list.
2270 */
2271 private static String formatList(String[] stringList, String pattern) {
2272 // If we have no list patterns, compose the list in a simple,
2273 // non-localized way.
2274 if (pattern == null) {
2275 return Arrays.stream(stringList).collect(Collectors.joining(","));
2276 }
2277
2278 switch (stringList.length) {
2279 case 0:
2280 return "";
2281 case 1:
2282 return stringList[0];
2283 default:
2284 return Arrays.stream(stringList).reduce("",
2285 (s1, s2) -> {
2286 if (s1.isEmpty()) {
2287 return s2;
2288 }
2289 if (s2.isEmpty()) {
2290 return s1;
2291 }
2292 return MessageFormat.format(pattern, s1, s2);
2293 });
2294 }
2295 }
2296
2297 // Duplicate of sun.util.locale.UnicodeLocaleExtension.isKey in order to
2298 // avoid its class loading.
2299 private static boolean isUnicodeExtensionKey(String s) {
2300 // 2alphanum
2301 return (s.length() == 2) && LocaleUtils.isAlphaNumericString(s);
2302 }
2303
2304 /**
2305 * @serialField language String
2306 * language subtag in lower case.
2307 * (See <a href="java.base/java/util/Locale.html#getLanguage()">getLanguage()</a>)
2308 * @serialField country String
2309 * country subtag in upper case.
2310 * (See <a href="java.base/java/util/Locale.html#getCountry()">getCountry()</a>)
2311 * @serialField variant String
2312 * variant subtags separated by LOWLINE characters.
2313 * (See <a href="java.base/java/util/Locale.html#getVariant()">getVariant()</a>)
2314 * @serialField hashcode int
2315 * deprecated, for forward compatibility only
2316 * @serialField script String
2317 * script subtag in title case
2318 * (See <a href="java.base/java/util/Locale.html#getScript()">getScript()</a>)
2319 * @serialField extensions String
2320 * canonical representation of extensions, that is,
2321 * BCP47 extensions in alphabetical order followed by
2322 * BCP47 private use subtags, all in lower case letters
2323 * separated by HYPHEN-MINUS characters.
2324 * (See <a href="java.base/java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>,
2325 * <a href="java.base/java/util/Locale.html#getExtension(char)">getExtension(char)</a>)
2326 */
2327 private static final ObjectStreamField[] serialPersistentFields = {
2328 new ObjectStreamField("language", String.class),
2329 new ObjectStreamField("country", String.class),
2330 new ObjectStreamField("variant", String.class),
2331 new ObjectStreamField("hashcode", int.class),
2332 new ObjectStreamField("script", String.class),
2333 new ObjectStreamField("extensions", String.class),
2334 };
2335
2336 /**
2337 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>.
2338 * @param out the <code>ObjectOutputStream</code> to write
2339 * @throws IOException
2340 * @since 1.7
2341 */
2342 private void writeObject(ObjectOutputStream out) throws IOException {
2343 ObjectOutputStream.PutField fields = out.putFields();
2344 fields.put("language", baseLocale.getLanguage());
2345 fields.put("script", baseLocale.getScript());
2346 fields.put("country", baseLocale.getRegion());
2347 fields.put("variant", baseLocale.getVariant());
2348 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID());
2349 fields.put("hashcode", -1); // place holder just for backward support
2350 out.writeFields();
2351 }
2352
2353 /**
2354 * Deserializes this <code>Locale</code>.
2355 * @param in the <code>ObjectInputStream</code> to read
2356 * @throws IOException
2357 * @throws ClassNotFoundException
2358 * @throws IllformedLocaleException
2359 * @since 1.7
2360 */
2361 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException {
2362 ObjectInputStream.GetField fields = in.readFields();
2363 String language = (String)fields.get("language", "");
2364 String script = (String)fields.get("script", "");
2365 String country = (String)fields.get("country", "");
2366 String variant = (String)fields.get("variant", "");
2367 String extStr = (String)fields.get("extensions", "");
2368
2369 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant);
2370 if (!extStr.isEmpty()) {
2371 try {
2372 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
2373 bldr.setExtensions(extStr);
2374 localeExtensions = bldr.getLocaleExtensions();
2375 } catch (LocaleSyntaxException e) {
2376 throw new IllformedLocaleException(e.getMessage());
2377 }
2378 } else {
2379 localeExtensions = null;
2380 }
2381 }
2382
2383 /**
2384 * Returns a cached <code>Locale</code> instance equivalent to
2385 * the deserialized <code>Locale</code>. When serialized
2386 * language, country and variant fields read from the object data stream
2387 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions
2388 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code>
2389 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script
2390 * type is "thai"). See <a href="Locale.html#special_cases_constructor">Special Cases</a>
2391 * for more information.
2392 *
2393 * @return an instance of <code>Locale</code> equivalent to
2394 * the deserialized <code>Locale</code>.
2395 * @throws java.io.ObjectStreamException
2396 */
2397 private Object readResolve() throws java.io.ObjectStreamException {
2398 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(),
2399 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions);
2400 }
2401
2402 private static volatile String[] isoLanguages;
2403
2404 private static volatile String[] isoCountries;
2405
2406 private static String convertOldISOCodes(String language) {
2407 // we accept both the old and the new ISO codes for the languages whose ISO
2408 // codes have changed, but we always store the OLD code, for backward compatibility
2409 language = LocaleUtils.toLowerString(language).intern();
2410 if (language == "he") {
2411 return "iw";
2412 } else if (language == "yi") {
2413 return "ji";
2414 } else if (language == "id") {
2415 return "in";
2416 } else {
2417 return language;
2418 }
2419 }
2420
2421 private static LocaleExtensions getCompatibilityExtensions(String language,
2422 String script,
2423 String country,
2424 String variant) {
2425 LocaleExtensions extensions = null;
2426 // Special cases for backward compatibility support
2427 if (LocaleUtils.caseIgnoreMatch(language, "ja")
2428 && script.isEmpty()
2429 && LocaleUtils.caseIgnoreMatch(country, "jp")
2430 && "JP".equals(variant)) {
2431 // ja_JP_JP -> u-ca-japanese (calendar = japanese)
2432 extensions = LocaleExtensions.CALENDAR_JAPANESE;
2433 } else if (LocaleUtils.caseIgnoreMatch(language, "th")
2434 && script.isEmpty()
2435 && LocaleUtils.caseIgnoreMatch(country, "th")
2436 && "TH".equals(variant)) {
2437 // th_TH_TH -> u-nu-thai (numbersystem = thai)
2438 extensions = LocaleExtensions.NUMBER_THAI;
2439 }
2440 return extensions;
2441 }
2442
2443 /**
2444 * Obtains a localized locale names from a LocaleNameProvider
2445 * implementation.
2446 */
2447 private static class LocaleNameGetter
2448 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> {
2449 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter();
2450
2451 @Override
2452 public String getObject(LocaleNameProvider localeNameProvider,
2453 Locale locale,
2454 String key,
2455 Object... params) {
2456 assert params.length == 3;
2457 int type = (Integer)params[0];
2458 String code = (String)params[1];
2459 String cat = (String)params[2];
2460
2461 switch(type) {
2462 case DISPLAY_LANGUAGE:
2463 return localeNameProvider.getDisplayLanguage(code, locale);
2464 case DISPLAY_COUNTRY:
2465 return localeNameProvider.getDisplayCountry(code, locale);
2466 case DISPLAY_VARIANT:
2467 return localeNameProvider.getDisplayVariant(code, locale);
2468 case DISPLAY_SCRIPT:
2469 return localeNameProvider.getDisplayScript(code, locale);
2470 case DISPLAY_UEXT_KEY:
2471 return localeNameProvider.getDisplayUnicodeExtensionKey(code, locale);
2472 case DISPLAY_UEXT_TYPE:
2473 return localeNameProvider.getDisplayUnicodeExtensionType(code, cat, locale);
2474 default:
2475 assert false; // shouldn't happen
2476 }
2477
2478 return null;
2479 }
2480 }
2481
2482 /**
2483 * Enum for locale categories. These locale categories are used to get/set
2484 * the default locale for the specific functionality represented by the
2485 * category.
2486 *
2487 * @see #getDefault(Locale.Category)
2488 * @see #setDefault(Locale.Category, Locale)
2489 * @since 1.7
2490 */
2491 public enum Category {
2492
2493 /**
2494 * Category used to represent the default locale for
2495 * displaying user interfaces.
2496 */
2497 DISPLAY("user.language.display",
2498 "user.script.display",
2499 "user.country.display",
2500 "user.variant.display",
2501 "user.extensions.display"),
2502
2503 /**
2504 * Category used to represent the default locale for
2505 * formatting dates, numbers, and/or currencies.
2506 */
2507 FORMAT("user.language.format",
2508 "user.script.format",
2509 "user.country.format",
2510 "user.variant.format",
2511 "user.extensions.format");
2512
2513 Category(String languageKey, String scriptKey, String countryKey,
2514 String variantKey, String extensionsKey) {
2515 this.languageKey = languageKey;
2516 this.scriptKey = scriptKey;
2517 this.countryKey = countryKey;
2518 this.variantKey = variantKey;
2519 this.extensionsKey = extensionsKey;
2520 }
2521
2522 final String languageKey;
2523 final String scriptKey;
2524 final String countryKey;
2525 final String variantKey;
2526 final String extensionsKey;
2527 }
2528
2529 /**
2530 * <code>Builder</code> is used to build instances of <code>Locale</code>
2531 * from values configured by the setters. Unlike the <code>Locale</code>
2532 * constructors, the <code>Builder</code> checks if a value configured by a
2533 * setter satisfies the syntax requirements defined by the <code>Locale</code>
2534 * class. A <code>Locale</code> object created by a <code>Builder</code> is
2535 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag
2536 * without losing information.
2537 *
2538 * <p><b>Note:</b> The <code>Locale</code> class does not provide any
2539 * syntactic restrictions on variant, while BCP 47 requires each variant
2540 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3
2541 * alphanumerics. The method <code>setVariant</code> throws
2542 * <code>IllformedLocaleException</code> for a variant that does not satisfy
2543 * this restriction. If it is necessary to support such a variant, use a
2544 * Locale constructor. However, keep in mind that a <code>Locale</code>
2545 * object created this way might lose the variant information when
2546 * transformed to a BCP 47 language tag.
2547 *
2548 * <p>The following example shows how to create a <code>Locale</code> object
2549 * with the <code>Builder</code>.
2550 * <blockquote>
2551 * <pre>
2552 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build();
2553 * </pre>
2554 * </blockquote>
2555 *
2556 * <p>Builders can be reused; <code>clear()</code> resets all
2557 * fields to their default values.
2558 *
2559 * @see Locale#forLanguageTag
2560 * @since 1.7
2561 */
2562 public static final class Builder {
2563 private final InternalLocaleBuilder localeBuilder;
2564
2565 /**
2566 * Constructs an empty Builder. The default value of all
2567 * fields, extensions, and private use information is the
2568 * empty string.
2569 */
2570 public Builder() {
2571 localeBuilder = new InternalLocaleBuilder();
2572 }
2573
2574 /**
2575 * Resets the <code>Builder</code> to match the provided
2576 * <code>locale</code>. Existing state is discarded.
2577 *
2578 * <p>All fields of the locale must be well-formed, see {@link Locale}.
2579 *
2580 * <p>Locales with any ill-formed fields cause
2581 * <code>IllformedLocaleException</code> to be thrown, except for the
2582 * following three cases which are accepted for compatibility
2583 * reasons:<ul>
2584 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese"
2585 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai"
2586 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul>
2587 *
2588 * @param locale the locale
2589 * @return This builder.
2590 * @throws IllformedLocaleException if <code>locale</code> has
2591 * any ill-formed fields.
2592 * @throws NullPointerException if <code>locale</code> is null.
2593 */
2594 public Builder setLocale(Locale locale) {
2595 try {
2596 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions);
2597 } catch (LocaleSyntaxException e) {
2598 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2599 }
2600 return this;
2601 }
2602
2603 /**
2604 * Resets the Builder to match the provided IETF BCP 47
2605 * language tag. Discards the existing state. Null and the
2606 * empty string cause the builder to be reset, like {@link
2607 * #clear}. Grandfathered tags (see {@link
2608 * Locale#forLanguageTag}) are converted to their canonical
2609 * form before being processed. Otherwise, the language tag
2610 * must be well-formed (see {@link Locale}) or an exception is
2611 * thrown (unlike <code>Locale.forLanguageTag</code>, which
2612 * just discards ill-formed and following portions of the
2613 * tag).
2614 *
2615 * @param languageTag the language tag
2616 * @return This builder.
2617 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed
2618 * @see Locale#forLanguageTag(String)
2619 */
2620 public Builder setLanguageTag(String languageTag) {
2621 ParseStatus sts = new ParseStatus();
2622 LanguageTag tag = LanguageTag.parse(languageTag, sts);
2623 if (sts.isError()) {
2624 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex());
2625 }
2626 localeBuilder.setLanguageTag(tag);
2627 return this;
2628 }
2629
2630 /**
2631 * Sets the language. If <code>language</code> is the empty string or
2632 * null, the language in this <code>Builder</code> is removed. Otherwise,
2633 * the language must be <a href="./Locale.html#def_language">well-formed</a>
2634 * or an exception is thrown.
2635 *
2636 * <p>The typical language value is a two or three-letter language
2637 * code as defined in ISO639.
2638 *
2639 * @param language the language
2640 * @return This builder.
2641 * @throws IllformedLocaleException if <code>language</code> is ill-formed
2642 */
2643 public Builder setLanguage(String language) {
2644 try {
2645 localeBuilder.setLanguage(language);
2646 } catch (LocaleSyntaxException e) {
2647 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2648 }
2649 return this;
2650 }
2651
2652 /**
2653 * Sets the script. If <code>script</code> is null or the empty string,
2654 * the script in this <code>Builder</code> is removed.
2655 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an
2656 * exception is thrown.
2657 *
2658 * <p>The typical script value is a four-letter script code as defined by ISO 15924.
2659 *
2660 * @param script the script
2661 * @return This builder.
2662 * @throws IllformedLocaleException if <code>script</code> is ill-formed
2663 */
2664 public Builder setScript(String script) {
2665 try {
2666 localeBuilder.setScript(script);
2667 } catch (LocaleSyntaxException e) {
2668 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2669 }
2670 return this;
2671 }
2672
2673 /**
2674 * Sets the region. If region is null or the empty string, the region
2675 * in this <code>Builder</code> is removed. Otherwise,
2676 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an
2677 * exception is thrown.
2678 *
2679 * <p>The typical region value is a two-letter ISO 3166 code or a
2680 * three-digit UN M.49 area code.
2681 *
2682 * <p>The country value in the <code>Locale</code> created by the
2683 * <code>Builder</code> is always normalized to upper case.
2684 *
2685 * @param region the region
2686 * @return This builder.
2687 * @throws IllformedLocaleException if <code>region</code> is ill-formed
2688 */
2689 public Builder setRegion(String region) {
2690 try {
2691 localeBuilder.setRegion(region);
2692 } catch (LocaleSyntaxException e) {
2693 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2694 }
2695 return this;
2696 }
2697
2698 /**
2699 * Sets the variant. If variant is null or the empty string, the
2700 * variant in this <code>Builder</code> is removed. Otherwise, it
2701 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a>
2702 * subtags, or an exception is thrown.
2703 *
2704 * <p><b>Note:</b> This method checks if <code>variant</code>
2705 * satisfies the IETF BCP 47 variant subtag's syntax requirements,
2706 * and normalizes the value to lowercase letters. However,
2707 * the <code>Locale</code> class does not impose any syntactic
2708 * restriction on variant, and the variant value in
2709 * <code>Locale</code> is case sensitive. To set such a variant,
2710 * use a Locale constructor.
2711 *
2712 * @param variant the variant
2713 * @return This builder.
2714 * @throws IllformedLocaleException if <code>variant</code> is ill-formed
2715 */
2716 public Builder setVariant(String variant) {
2717 try {
2718 localeBuilder.setVariant(variant);
2719 } catch (LocaleSyntaxException e) {
2720 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2721 }
2722 return this;
2723 }
2724
2725 /**
2726 * Sets the extension for the given key. If the value is null or the
2727 * empty string, the extension is removed. Otherwise, the extension
2728 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception
2729 * is thrown.
2730 *
2731 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION
2732 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension.
2733 * Setting a value for this key replaces any existing Unicode locale key/type
2734 * pairs with those defined in the extension.
2735 *
2736 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION
2737 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be
2738 * well-formed, the value for this key needs only to have subtags of one to
2739 * eight alphanumeric characters, not two to eight as in the general case.
2740 *
2741 * @param key the extension key
2742 * @param value the extension value
2743 * @return This builder.
2744 * @throws IllformedLocaleException if <code>key</code> is illegal
2745 * or <code>value</code> is ill-formed
2746 * @see #setUnicodeLocaleKeyword(String, String)
2747 */
2748 public Builder setExtension(char key, String value) {
2749 try {
2750 localeBuilder.setExtension(key, value);
2751 } catch (LocaleSyntaxException e) {
2752 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2753 }
2754 return this;
2755 }
2756
2757 /**
2758 * Sets the Unicode locale keyword type for the given key. If the type
2759 * is null, the Unicode keyword is removed. Otherwise, the key must be
2760 * non-null and both key and type must be <a
2761 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2762 * is thrown.
2763 *
2764 * <p>Keys and types are converted to lower case.
2765 *
2766 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension}
2767 * replaces all Unicode locale keywords with those defined in the
2768 * extension.
2769 *
2770 * @param key the Unicode locale key
2771 * @param type the Unicode locale type
2772 * @return This builder.
2773 * @throws IllformedLocaleException if <code>key</code> or <code>type</code>
2774 * is ill-formed
2775 * @throws NullPointerException if <code>key</code> is null
2776 * @see #setExtension(char, String)
2777 */
2778 public Builder setUnicodeLocaleKeyword(String key, String type) {
2779 try {
2780 localeBuilder.setUnicodeLocaleKeyword(key, type);
2781 } catch (LocaleSyntaxException e) {
2782 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2783 }
2784 return this;
2785 }
2786
2787 /**
2788 * Adds a unicode locale attribute, if not already present, otherwise
2789 * has no effect. The attribute must not be null and must be <a
2790 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2791 * is thrown.
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 addUnicodeLocaleAttribute(String attribute) {
2800 try {
2801 localeBuilder.addUnicodeLocaleAttribute(attribute);
2802 } catch (LocaleSyntaxException e) {
2803 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2804 }
2805 return this;
2806 }
2807
2808 /**
2809 * Removes a unicode locale attribute, if present, otherwise has no
2810 * effect. The attribute must not be null and must be <a
2811 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2812 * is thrown.
2813 *
2814 * <p>Attribute comparison for removal is case-insensitive.
2815 *
2816 * @param attribute the attribute
2817 * @return This builder.
2818 * @throws NullPointerException if <code>attribute</code> is null
2819 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2820 * @see #setExtension(char, String)
2821 */
2822 public Builder removeUnicodeLocaleAttribute(String attribute) {
2823 Objects.requireNonNull(attribute);
2824 try {
2825 localeBuilder.removeUnicodeLocaleAttribute(attribute);
2826 } catch (LocaleSyntaxException e) {
2827 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2828 }
2829 return this;
2830 }
2831
2832 /**
2833 * Resets the builder to its initial, empty state.
2834 *
2835 * @return This builder.
2836 */
2837 public Builder clear() {
2838 localeBuilder.clear();
2839 return this;
2840 }
2841
2842 /**
2843 * Resets the extensions to their initial, empty state.
2844 * Language, script, region and variant are unchanged.
2845 *
2846 * @return This builder.
2847 * @see #setExtension(char, String)
2848 */
2849 public Builder clearExtensions() {
2850 localeBuilder.clearExtensions();
2851 return this;
2852 }
2853
2854 /**
2855 * Returns an instance of <code>Locale</code> created from the fields set
2856 * on this builder.
2857 *
2858 * <p>This applies the conversions listed in {@link Locale#forLanguageTag}
2859 * when constructing a Locale. (Grandfathered tags are handled in
2860 * {@link #setLanguageTag}.)
2861 *
2862 * @return A Locale.
2863 */
2864 public Locale build() {
2865 BaseLocale baseloc = localeBuilder.getBaseLocale();
2866 LocaleExtensions extensions = localeBuilder.getLocaleExtensions();
2867 if (extensions == null && !baseloc.getVariant().isEmpty()) {
2868 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(),
2869 baseloc.getRegion(), baseloc.getVariant());
2870 }
2871 return Locale.getInstance(baseloc, extensions);
2872 }
2873 }
2874
2875 /**
2876 * This enum provides constants to select a filtering mode for locale
2877 * matching. Refer to <a href="http://tools.ietf.org/html/rfc4647">RFC 4647
2878 * Matching of Language Tags</a> for details.
2879 *
2880 * <p>As an example, think of two Language Priority Lists each of which
2881 * includes only one language range and a set of following language tags:
2882 *
2883 * <pre>
2884 * de (German)
2885 * de-DE (German, Germany)
2886 * de-Deva (German, in Devanagari script)
2887 * de-Deva-DE (German, in Devanagari script, Germany)
2888 * de-DE-1996 (German, Germany, orthography of 1996)
2889 * de-Latn-DE (German, in Latin script, Germany)
2890 * de-Latn-DE-1996 (German, in Latin script, Germany, orthography of 1996)
2891 * </pre>
2892 *
2893 * The filtering method will behave as follows:
2894 *
2895 * <table class="striped">
2896 * <caption>Filtering method behavior</caption>
2897 * <thead>
2898 * <tr>
2899 * <th scope="col">Filtering Mode</th>
2900 * <th scope="col">Language Priority List: {@code "de-DE"}</th>
2901 * <th scope="col">Language Priority List: {@code "de-*-DE"}</th>
2902 * </tr>
2903 * </thead>
2904 * <tbody>
2905 * <tr>
2906 * <th scope="row" style="vertical-align:top">
2907 * {@link FilteringMode#AUTOSELECT_FILTERING AUTOSELECT_FILTERING}
2908 * </th>
2909 * <td style="vertical-align:top">
2910 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2911 * {@code "de-DE-1996"}.
2912 * </td>
2913 * <td style="vertical-align:top">
2914 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2915 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2916 * {@code "de-Latn-DE-1996"}.
2917 * </td>
2918 * </tr>
2919 * <tr>
2920 * <th scope="row" style="vertical-align:top">
2921 * {@link FilteringMode#EXTENDED_FILTERING EXTENDED_FILTERING}
2922 * </th>
2923 * <td style="vertical-align:top">
2924 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2925 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2926 * {@code "de-Latn-DE-1996"}.
2927 * </td>
2928 * <td style="vertical-align:top">Same as above.</td>
2929 * </tr>
2930 * <tr>
2931 * <th scope="row" style="vertical-align:top">
2932 * {@link FilteringMode#IGNORE_EXTENDED_RANGES IGNORE_EXTENDED_RANGES}
2933 * </th>
2934 * <td style="vertical-align:top">
2935 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2936 * {@code "de-DE-1996"}.
2937 * </td>
2938 * <td style="vertical-align:top">
2939 * Performs <em>basic</em> filtering and returns {@code null} because
2940 * nothing matches.
2941 * </td>
2942 * </tr>
2943 * <tr>
2944 * <th scope="row" style="vertical-align:top">
2945 * {@link FilteringMode#MAP_EXTENDED_RANGES MAP_EXTENDED_RANGES}
2946 * </th>
2947 * <td style="vertical-align:top">Same as above.</td>
2948 * <td style="vertical-align:top">
2949 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2950 * {@code "de-DE-1996"} because {@code "de-*-DE"} is mapped to
2951 * {@code "de-DE"}.
2952 * </td>
2953 * </tr>
2954 * <tr>
2955 * <th scope="row" style="vertical-align:top">
2956 * {@link FilteringMode#REJECT_EXTENDED_RANGES REJECT_EXTENDED_RANGES}
2957 * </th>
2958 * <td style="vertical-align:top">Same as above.</td>
2959 * <td style="vertical-align:top">
2960 * Throws {@link IllegalArgumentException} because {@code "de-*-DE"} is
2961 * not a valid basic language range.
2962 * </td>
2963 * </tr>
2964 * </tbody>
2965 * </table>
2966 *
2967 * @see #filter(List, Collection, FilteringMode)
2968 * @see #filterTags(List, Collection, FilteringMode)
2969 *
2970 * @since 1.8
2971 */
2972 public static enum FilteringMode {
2973 /**
2974 * Specifies automatic filtering mode based on the given Language
2975 * Priority List consisting of language ranges. If all of the ranges
2976 * are basic, basic filtering is selected. Otherwise, extended
2977 * filtering is selected.
2978 */
2979 AUTOSELECT_FILTERING,
2980
2981 /**
2982 * Specifies extended filtering.
2983 */
2984 EXTENDED_FILTERING,
2985
2986 /**
2987 * Specifies basic filtering: Note that any extended language ranges
2988 * included in the given Language Priority List are ignored.
2989 */
2990 IGNORE_EXTENDED_RANGES,
2991
2992 /**
2993 * Specifies basic filtering: If any extended language ranges are
2994 * included in the given Language Priority List, they are mapped to the
2995 * basic language range. Specifically, a language range starting with a
2996 * subtag {@code "*"} is treated as a language range {@code "*"}. For
2997 * example, {@code "*-US"} is treated as {@code "*"}. If {@code "*"} is
2998 * not the first subtag, {@code "*"} and extra {@code "-"} are removed.
2999 * For example, {@code "ja-*-JP"} is mapped to {@code "ja-JP"}.
3000 */
3001 MAP_EXTENDED_RANGES,
3002
3003 /**
3004 * Specifies basic filtering: If any extended language ranges are
3005 * included in the given Language Priority List, the list is rejected
3006 * and the filtering method throws {@link IllegalArgumentException}.
3007 */
3008 REJECT_EXTENDED_RANGES
3009 };
3010
3011 /**
3012 * This class expresses a <em>Language Range</em> defined in
3013 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
3014 * Language Tags</a>. A language range is an identifier which is used to
3015 * select language tag(s) meeting specific requirements by using the
3016 * mechanisms described in <a href="Locale.html#LocaleMatching">Locale
3017 * Matching</a>. A list which represents a user's preferences and consists
3018 * of language ranges is called a <em>Language Priority List</em>.
3019 *
3020 * <p>There are two types of language ranges: basic and extended. In RFC
3021 * 4647, the syntax of language ranges is expressed in
3022 * <a href="http://tools.ietf.org/html/rfc4234">ABNF</a> as follows:
3023 * <blockquote>
3024 * <pre>
3025 * basic-language-range = (1*8ALPHA *("-" 1*8alphanum)) / "*"
3026 * extended-language-range = (1*8ALPHA / "*")
3027 * *("-" (1*8alphanum / "*"))
3028 * alphanum = ALPHA / DIGIT
3029 * </pre>
3030 * </blockquote>
3031 * For example, {@code "en"} (English), {@code "ja-JP"} (Japanese, Japan),
3032 * {@code "*"} (special language range which matches any language tag) are
3033 * basic language ranges, whereas {@code "*-CH"} (any languages,
3034 * Switzerland), {@code "es-*"} (Spanish, any regions), and
3035 * {@code "zh-Hant-*"} (Traditional Chinese, any regions) are extended
3036 * language ranges.
3037 *
3038 * @see #filter
3039 * @see #filterTags
3040 * @see #lookup
3041 * @see #lookupTag
3042 *
3043 * @since 1.8
3044 */
3045 public static final class LanguageRange {
3046
3047 /**
3048 * A constant holding the maximum value of weight, 1.0, which indicates
3049 * that the language range is a good fit for the user.
3050 */
3051 public static final double MAX_WEIGHT = 1.0;
3052
3053 /**
3054 * A constant holding the minimum value of weight, 0.0, which indicates
3055 * that the language range is not a good fit for the user.
3056 */
3057 public static final double MIN_WEIGHT = 0.0;
3058
3059 private final String range;
3060 private final double weight;
3061
3062 private volatile int hash;
3063
3064 /**
3065 * Constructs a {@code LanguageRange} using the given {@code range}.
3066 * Note that no validation is done against the IANA Language Subtag
3067 * Registry at time of construction.
3068 *
3069 * <p>This is equivalent to {@code LanguageRange(range, MAX_WEIGHT)}.
3070 *
3071 * @param range a language range
3072 * @throws NullPointerException if the given {@code range} is
3073 * {@code null}
3074 * @throws IllegalArgumentException if the given {@code range} does not
3075 * comply with the syntax of the language range mentioned in RFC 4647
3076 */
3077 public LanguageRange(String range) {
3078 this(range, MAX_WEIGHT);
3079 }
3080
3081 /**
3082 * Constructs a {@code LanguageRange} using the given {@code range} and
3083 * {@code weight}. Note that no validation is done against the IANA
3084 * Language Subtag Registry at time of construction.
3085 *
3086 * @param range a language range
3087 * @param weight a weight value between {@code MIN_WEIGHT} and
3088 * {@code MAX_WEIGHT}
3089 * @throws NullPointerException if the given {@code range} is
3090 * {@code null}
3091 * @throws IllegalArgumentException if the given {@code range} does not
3092 * comply with the syntax of the language range mentioned in RFC 4647
3093 * or if the given {@code weight} is less than {@code MIN_WEIGHT}
3094 * or greater than {@code MAX_WEIGHT}
3095 */
3096 public LanguageRange(String range, double weight) {
3097 if (range == null) {
3098 throw new NullPointerException();
3099 }
3100 if (weight < MIN_WEIGHT || weight > MAX_WEIGHT) {
3101 throw new IllegalArgumentException("weight=" + weight);
3102 }
3103
3104 range = range.toLowerCase(Locale.ROOT);
3105
3106 // Do syntax check.
3107 boolean isIllFormed = false;
3108 String[] subtags = range.split("-");
3109 if (isSubtagIllFormed(subtags[0], true)
3110 || range.endsWith("-")) {
3111 isIllFormed = true;
3112 } else {
3113 for (int i = 1; i < subtags.length; i++) {
3114 if (isSubtagIllFormed(subtags[i], false)) {
3115 isIllFormed = true;
3116 break;
3117 }
3118 }
3119 }
3120 if (isIllFormed) {
3121 throw new IllegalArgumentException("range=" + range);
3122 }
3123
3124 this.range = range;
3125 this.weight = weight;
3126 }
3127
3128 private static boolean isSubtagIllFormed(String subtag,
3129 boolean isFirstSubtag) {
3130 if (subtag.isEmpty() || subtag.length() > 8) {
3131 return true;
3132 } else if (subtag.equals("*")) {
3133 return false;
3134 }
3135 char[] charArray = subtag.toCharArray();
3136 if (isFirstSubtag) { // ALPHA
3137 for (char c : charArray) {
3138 if (c < 'a' || c > 'z') {
3139 return true;
3140 }
3141 }
3142 } else { // ALPHA / DIGIT
3143 for (char c : charArray) {
3144 if (c < '0' || (c > '9' && c < 'a') || c > 'z') {
3145 return true;
3146 }
3147 }
3148 }
3149 return false;
3150 }
3151
3152 /**
3153 * Returns the language range of this {@code LanguageRange}.
3154 *
3155 * @return the language range.
3156 */
3157 public String getRange() {
3158 return range;
3159 }
3160
3161 /**
3162 * Returns the weight of this {@code LanguageRange}.
3163 *
3164 * @return the weight value.
3165 */
3166 public double getWeight() {
3167 return weight;
3168 }
3169
3170 /**
3171 * Parses the given {@code ranges} to generate a Language Priority List.
3172 *
3173 * <p>This method performs a syntactic check for each language range in
3174 * the given {@code ranges} but doesn't do validation using the IANA
3175 * Language Subtag Registry.
3176 *
3177 * <p>The {@code ranges} to be given can take one of the following
3178 * forms:
3179 *
3180 * <pre>
3181 * "Accept-Language: ja,en;q=0.4" (weighted list with Accept-Language prefix)
3182 * "ja,en;q=0.4" (weighted list)
3183 * "ja,en" (prioritized list)
3184 * </pre>
3185 *
3186 * In a weighted list, each language range is given a weight value.
3187 * The weight value is identical to the "quality value" in
3188 * <a href="http://tools.ietf.org/html/rfc2616">RFC 2616</a>, and it
3189 * expresses how much the user prefers the language. A weight value is
3190 * specified after a corresponding language range followed by
3191 * {@code ";q="}, and the default weight value is {@code MAX_WEIGHT}
3192 * when it is omitted.
3193 *
3194 * <p>Unlike a weighted list, language ranges in a prioritized list
3195 * are sorted in the descending order based on its priority. The first
3196 * language range has the highest priority and meets the user's
3197 * preference most.
3198 *
3199 * <p>In either case, language ranges are sorted in descending order in
3200 * the Language Priority List based on priority or weight. If a
3201 * language range appears in the given {@code ranges} more than once,
3202 * only the first one is included on the Language Priority List.
3203 *
3204 * <p>The returned list consists of language ranges from the given
3205 * {@code ranges} and their equivalents found in the IANA Language
3206 * Subtag Registry. For example, if the given {@code ranges} is
3207 * {@code "Accept-Language: iw,en-us;q=0.7,en;q=0.3"}, the elements in
3208 * the list to be returned are:
3209 *
3210 * <pre>
3211 * <b>Range</b> <b>Weight</b>
3212 * "iw" (older tag for Hebrew) 1.0
3213 * "he" (new preferred code for Hebrew) 1.0
3214 * "en-us" (English, United States) 0.7
3215 * "en" (English) 0.3
3216 * </pre>
3217 *
3218 * Two language ranges, {@code "iw"} and {@code "he"}, have the same
3219 * highest priority in the list. By adding {@code "he"} to the user's
3220 * Language Priority List, locale-matching method can find Hebrew as a
3221 * matching locale (or language tag) even if the application or system
3222 * offers only {@code "he"} as a supported locale (or language tag).
3223 *
3224 * @param ranges a list of comma-separated language ranges or a list of
3225 * language ranges in the form of the "Accept-Language" header
3226 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3227 * 2616</a>
3228 * @return a Language Priority List consisting of language ranges
3229 * included in the given {@code ranges} and their equivalent
3230 * language ranges if available. The list is modifiable.
3231 * @throws NullPointerException if {@code ranges} is null
3232 * @throws IllegalArgumentException if a language range or a weight
3233 * found in the given {@code ranges} is ill-formed
3234 */
3235 public static List<LanguageRange> parse(String ranges) {
3236 return LocaleMatcher.parse(ranges);
3237 }
3238
3239 /**
3240 * Parses the given {@code ranges} to generate a Language Priority
3241 * List, and then customizes the list using the given {@code map}.
3242 * This method is equivalent to
3243 * {@code mapEquivalents(parse(ranges), map)}.
3244 *
3245 * @param ranges a list of comma-separated language ranges or a list
3246 * of language ranges in the form of the "Accept-Language" header
3247 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3248 * 2616</a>
3249 * @param map a map containing information to customize language ranges
3250 * @return a Language Priority List with customization. The list is
3251 * modifiable.
3252 * @throws NullPointerException if {@code ranges} is null
3253 * @throws IllegalArgumentException if a language range or a weight
3254 * found in the given {@code ranges} is ill-formed
3255 * @see #parse(String)
3256 * @see #mapEquivalents
3257 */
3258 public static List<LanguageRange> parse(String ranges,
3259 Map<String, List<String>> map) {
3260 return mapEquivalents(parse(ranges), map);
3261 }
3262
3263 /**
3264 * Generates a new customized Language Priority List using the given
3265 * {@code priorityList} and {@code map}. If the given {@code map} is
3266 * empty, this method returns a copy of the given {@code priorityList}.
3267 *
3268 * <p>In the map, a key represents a language range whereas a value is
3269 * a list of equivalents of it. {@code '*'} cannot be used in the map.
3270 * Each equivalent language range has the same weight value as its
3271 * original language range.
3272 *
3273 * <pre>
3274 * An example of map:
3275 * <b>Key</b> <b>Value</b>
3276 * "zh" (Chinese) "zh",
3277 * "zh-Hans"(Simplified Chinese)
3278 * "zh-HK" (Chinese, Hong Kong) "zh-HK"
3279 * "zh-TW" (Chinese, Taiwan) "zh-TW"
3280 * </pre>
3281 *
3282 * The customization is performed after modification using the IANA
3283 * Language Subtag Registry.
3284 *
3285 * <p>For example, if a user's Language Priority List consists of five
3286 * language ranges ({@code "zh"}, {@code "zh-CN"}, {@code "en"},
3287 * {@code "zh-TW"}, and {@code "zh-HK"}), the newly generated Language
3288 * Priority List which is customized using the above map example will
3289 * consists of {@code "zh"}, {@code "zh-Hans"}, {@code "zh-CN"},
3290 * {@code "zh-Hans-CN"}, {@code "en"}, {@code "zh-TW"}, and
3291 * {@code "zh-HK"}.
3292 *
3293 * <p>{@code "zh-HK"} and {@code "zh-TW"} aren't converted to
3294 * {@code "zh-Hans-HK"} nor {@code "zh-Hans-TW"} even if they are
3295 * included in the Language Priority List. In this example, mapping
3296 * is used to clearly distinguish Simplified Chinese and Traditional
3297 * Chinese.
3298 *
3299 * <p>If the {@code "zh"}-to-{@code "zh"} mapping isn't included in the
3300 * map, a simple replacement will be performed and the customized list
3301 * won't include {@code "zh"} and {@code "zh-CN"}.
3302 *
3303 * @param priorityList user's Language Priority List
3304 * @param map a map containing information to customize language ranges
3305 * @return a new Language Priority List with customization. The list is
3306 * modifiable.
3307 * @throws NullPointerException if {@code priorityList} is {@code null}
3308 * @see #parse(String, Map)
3309 */
3310 public static List<LanguageRange> mapEquivalents(
3311 List<LanguageRange>priorityList,
3312 Map<String, List<String>> map) {
3313 return LocaleMatcher.mapEquivalents(priorityList, map);
3314 }
3315
3316 /**
3317 * Returns a hash code value for the object.
3318 *
3319 * @return a hash code value for this object.
3320 */
3321 @Override
3322 public int hashCode() {
3323 int h = hash;
3324 if (h == 0) {
3325 h = 17;
3326 h = 37*h + range.hashCode();
3327 long bitsWeight = Double.doubleToLongBits(weight);
3328 h = 37*h + (int)(bitsWeight ^ (bitsWeight >>> 32));
3329 if (h != 0) {
3330 hash = h;
3331 }
3332 }
3333 return h;
3334 }
3335
3336 /**
3337 * Compares this object to the specified object. The result is true if
3338 * and only if the argument is not {@code null} and is a
3339 * {@code LanguageRange} object that contains the same {@code range}
3340 * and {@code weight} values as this object.
3341 *
3342 * @param obj the object to compare with
3343 * @return {@code true} if this object's {@code range} and
3344 * {@code weight} are the same as the {@code obj}'s; {@code false}
3345 * otherwise.
3346 */
3347 @Override
3348 public boolean equals(Object obj) {
3349 if (this == obj) {
3350 return true;
3351 }
3352 if (!(obj instanceof LanguageRange)) {
3353 return false;
3354 }
3355 LanguageRange other = (LanguageRange)obj;
3356 return hash == other.hash
3357 && range.equals(other.range)
3358 && weight == other.weight;
3359 }
3360
3361 /**
3362 * Returns an informative string representation of this {@code LanguageRange}
3363 * object, consisting of language range and weight if the range is
3364 * weighted and the weight is less than the max weight.
3365 *
3366 * @return a string representation of this {@code LanguageRange} object.
3367 */
3368 @Override
3369 public String toString() {
3370 return (weight == MAX_WEIGHT) ? range : range + ";q=" + weight;
3371 }
3372 }
3373
3374 /**
3375 * Returns a list of matching {@code Locale} instances using the filtering
3376 * mechanism defined in RFC 4647.
3377 *
3378 * This filter operation on the given {@code locales} ensures that only
3379 * unique matching locale(s) are returned.
3380 *
3381 * @param priorityList user's Language Priority List in which each language
3382 * tag is sorted in descending order based on priority or weight
3383 * @param locales {@code Locale} instances used for matching
3384 * @param mode filtering mode
3385 * @return a list of {@code Locale} instances for matching language tags
3386 * sorted in descending order based on priority or weight, or an empty
3387 * list if nothing matches. The list is modifiable.
3388 * @throws NullPointerException if {@code priorityList} or {@code locales}
3389 * is {@code null}
3390 * @throws IllegalArgumentException if one or more extended language ranges
3391 * are included in the given list when
3392 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3393 *
3394 * @since 1.8
3395 */
3396 public static List<Locale> filter(List<LanguageRange> priorityList,
3397 Collection<Locale> locales,
3398 FilteringMode mode) {
3399 return LocaleMatcher.filter(priorityList, locales, mode);
3400 }
3401
3402 /**
3403 * Returns a list of matching {@code Locale} instances using the filtering
3404 * mechanism defined in RFC 4647. This is equivalent to
3405 * {@link #filter(List, Collection, FilteringMode)} when {@code mode} is
3406 * {@link FilteringMode#AUTOSELECT_FILTERING}.
3407 *
3408 * This filter operation on the given {@code locales} ensures that only
3409 * unique matching locale(s) are returned.
3410 *
3411 * @param priorityList user's Language Priority List in which each language
3412 * tag is sorted in descending order based on priority or weight
3413 * @param locales {@code Locale} instances used for matching
3414 * @return a list of {@code Locale} instances for matching language tags
3415 * sorted in descending order based on priority or weight, or an empty
3416 * list if nothing matches. The list is modifiable.
3417 * @throws NullPointerException if {@code priorityList} or {@code locales}
3418 * is {@code null}
3419 *
3420 * @since 1.8
3421 */
3422 public static List<Locale> filter(List<LanguageRange> priorityList,
3423 Collection<Locale> locales) {
3424 return filter(priorityList, locales, FilteringMode.AUTOSELECT_FILTERING);
3425 }
3426
3427 /**
3428 * Returns a list of matching languages tags using the basic filtering
3429 * mechanism defined in RFC 4647.
3430 *
3431 * This filter operation on the given {@code tags} ensures that only
3432 * unique matching tag(s) are returned with preserved case. In case of
3433 * duplicate matching tags with the case difference, the first matching
3434 * tag with preserved case is returned.
3435 * For example, "de-ch" is returned out of the duplicate matching tags
3436 * "de-ch" and "de-CH", if "de-ch" is checked first for matching in the
3437 * given {@code tags}. Note that if the given {@code tags} is an unordered
3438 * {@code Collection}, the returned matching tag out of duplicate tags is
3439 * subject to change, depending on the implementation of the
3440 * {@code Collection}.
3441 *
3442 * @param priorityList user's Language Priority List in which each language
3443 * tag is sorted in descending order based on priority or weight
3444 * @param tags language tags
3445 * @param mode filtering mode
3446 * @return a list of matching language tags sorted in descending order
3447 * based on priority or weight, or an empty list if nothing matches.
3448 * The list is modifiable.
3449 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3450 * {@code null}
3451 * @throws IllegalArgumentException if one or more extended language ranges
3452 * are included in the given list when
3453 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3454 *
3455 * @since 1.8
3456 */
3457 public static List<String> filterTags(List<LanguageRange> priorityList,
3458 Collection<String> tags,
3459 FilteringMode mode) {
3460 return LocaleMatcher.filterTags(priorityList, tags, mode);
3461 }
3462
3463 /**
3464 * Returns a list of matching languages tags using the basic filtering
3465 * mechanism defined in RFC 4647. This is equivalent to
3466 * {@link #filterTags(List, Collection, FilteringMode)} when {@code mode}
3467 * is {@link FilteringMode#AUTOSELECT_FILTERING}.
3468 *
3469 * This filter operation on the given {@code tags} ensures that only
3470 * unique matching tag(s) are returned with preserved case. In case of
3471 * duplicate matching tags with the case difference, the first matching
3472 * tag with preserved case is returned.
3473 * For example, "de-ch" is returned out of the duplicate matching tags
3474 * "de-ch" and "de-CH", if "de-ch" is checked first for matching in the
3475 * given {@code tags}. Note that if the given {@code tags} is an unordered
3476 * {@code Collection}, the returned matching tag out of duplicate tags is
3477 * subject to change, depending on the implementation of the
3478 * {@code Collection}.
3479 *
3480 * @param priorityList user's Language Priority List in which each language
3481 * tag is sorted in descending order based on priority or weight
3482 * @param tags language tags
3483 * @return a list of matching language tags sorted in descending order
3484 * based on priority or weight, or an empty list if nothing matches.
3485 * The list is modifiable.
3486 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3487 * {@code null}
3488 *
3489 * @since 1.8
3490 */
3491 public static List<String> filterTags(List<LanguageRange> priorityList,
3492 Collection<String> tags) {
3493 return filterTags(priorityList, tags, FilteringMode.AUTOSELECT_FILTERING);
3494 }
3495
3496 /**
3497 * Returns a {@code Locale} instance for the best-matching language
3498 * tag using the lookup mechanism defined in RFC 4647.
3499 *
3500 * @param priorityList user's Language Priority List in which each language
3501 * tag is sorted in descending order based on priority or weight
3502 * @param locales {@code Locale} instances used for matching
3503 * @return the best matching <code>Locale</code> instance chosen based on
3504 * priority or weight, or {@code null} if nothing matches.
3505 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3506 * {@code null}
3507 *
3508 * @since 1.8
3509 */
3510 public static Locale lookup(List<LanguageRange> priorityList,
3511 Collection<Locale> locales) {
3512 return LocaleMatcher.lookup(priorityList, locales);
3513 }
3514
3515 /**
3516 * Returns the best-matching language tag using the lookup mechanism
3517 * defined in RFC 4647.
3518 *
3519 * This lookup operation on the given {@code tags} ensures that the
3520 * first matching tag with preserved case is returned.
3521 *
3522 * @param priorityList user's Language Priority List in which each language
3523 * tag is sorted in descending order based on priority or weight
3524 * @param tags language tangs used for matching
3525 * @return the best matching language tag chosen based on priority or
3526 * weight, or {@code null} if nothing matches.
3527 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3528 * {@code null}
3529 *
3530 * @since 1.8
3531 */
3532 public static String lookupTag(List<LanguageRange> priorityList,
3533 Collection<String> tags) {
3534 return LocaleMatcher.lookupTag(priorityList, tags);
3535 }
3536
3537 }