1 /*
2 * Copyright (c) 1996, 2016, 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
52 import sun.security.action.GetPropertyAction;
53 import sun.util.locale.BaseLocale;
54 import sun.util.locale.InternalLocaleBuilder;
55 import sun.util.locale.LanguageTag;
56 import sun.util.locale.LocaleExtensions;
57 import sun.util.locale.LocaleMatcher;
58 import sun.util.locale.LocaleObjectCache;
59 import sun.util.locale.LocaleSyntaxException;
60 import sun.util.locale.LocaleUtils;
61 import sun.util.locale.ParseStatus;
62 import sun.util.locale.provider.LocaleProviderAdapter;
63 import sun.util.locale.provider.LocaleResources;
64 import sun.util.locale.provider.LocaleServiceProviderPool;
65
66 /**
67 * A <code>Locale</code> object represents a specific geographical, political,
68 * or cultural region. An operation that requires a <code>Locale</code> to perform
69 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code>
70 * to tailor information for the user. For example, displaying a number
71 * is a locale-sensitive operation— the number should be formatted
72 * according to the customs and conventions of the user's native country,
73 * region, or culture.
74 *
75 * <p> The {@code Locale} class implements IETF BCP 47 which is composed of
76 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 "Matching of Language
77 * Tags"</a> and <a href="http://tools.ietf.org/html/rfc5646">RFC 5646 "Tags
78 * for Identifying Languages"</a> with support for the LDML (UTS#35, "Unicode
79 * Locale Data Markup Language") BCP 47-compatible extensions for locale data
80 * exchange.
81 *
82 * <p> A <code>Locale</code> object logically consists of the fields
83 * described below.
84 *
85 * <dl>
86 * <dt><a name="def_language"><b>language</b></a></dt>
87 *
88 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered
89 * language subtags up to 8 alpha letters (for future enhancements).
90 * When a language has both an alpha-2 code and an alpha-3 code, the
91 * alpha-2 code must be used. You can find a full list of valid
92 * language codes in the IANA Language Subtag Registry (search for
93 * "Type: language"). The language field is case insensitive, but
94 * <code>Locale</code> always canonicalizes to lower case.</dd>
95 *
96 * <dd>Well-formed language values have the form
97 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the full
98 * BCP47 language production, since it excludes extlang. They are
99 * not needed since modern three-letter language codes replace
100 * them.</dd>
101 *
102 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd>
103 *
104 * <dt><a name="def_script"><b>script</b></a></dt>
105 *
106 * <dd>ISO 15924 alpha-4 script code. You can find a full list of
107 * valid script codes in the IANA Language Subtag Registry (search
108 * for "Type: script"). The script field is case insensitive, but
109 * <code>Locale</code> always canonicalizes to title case (the first
110 * letter is upper case and the rest of the letters are lower
111 * case).</dd>
112 *
113 * <dd>Well-formed script values have the form
114 * <code>[a-zA-Z]{4}</code></dd>
115 *
116 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd>
117 *
118 * <dt><a name="def_region"><b>country (region)</b></a></dt>
119 *
120 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code.
121 * You can find a full list of valid country and region codes in the
122 * IANA Language Subtag Registry (search for "Type: region"). The
123 * country (region) field is case insensitive, but
124 * <code>Locale</code> always canonicalizes to upper case.</dd>
125 *
126 * <dd>Well-formed country/region values have
127 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd>
128 *
129 * <dd>Example: "US" (United States), "FR" (France), "029"
130 * (Caribbean)</dd>
131 *
132 * <dt><a name="def_variant"><b>variant</b></a></dt>
133 *
134 * <dd>Any arbitrary value used to indicate a variation of a
135 * <code>Locale</code>. Where there are two or more variant values
136 * each indicating its own semantics, these values should be ordered
137 * by importance, with most important first, separated by
138 * underscore('_'). The variant field is case sensitive.</dd>
139 *
140 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant
141 * subtags. Also BCP 47 subtags are strictly used to indicate
142 * additional variations that define a language or its dialects that
143 * are not covered by any combinations of language, script and
144 * region subtags. You can find a full list of valid variant codes
145 * in the IANA Language Subtag Registry (search for "Type: variant").
146 *
147 * <p>However, the variant field in <code>Locale</code> has
148 * historically been used for any kind of variation, not just
149 * language variations. For example, some supported variants
150 * available in Java SE Runtime Environments indicate alternative
151 * cultural behaviors such as calendar type or number script. In
152 * BCP 47 this kind of information, which does not identify the
153 * language, is supported by extension subtags or private use
154 * subtags.</dd>
155 *
156 * <dd>Well-formed variant values have the form <code>SUBTAG
157 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG =
158 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only
159 * uses hyphen ('-') as a delimiter, this is more lenient).</dd>
160 *
161 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd>
162 *
163 * <dt><a name="def_extensions"><b>extensions</b></a></dt>
164 *
165 * <dd>A map from single character keys to string values, indicating
166 * extensions apart from language identification. The extensions in
167 * <code>Locale</code> implement the semantics and syntax of BCP 47
168 * extension subtags and private use subtags. The extensions are
169 * case insensitive, but <code>Locale</code> canonicalizes all
170 * extension keys and values to lower case. Note that extensions
171 * cannot have empty values.</dd>
172 *
173 * <dd>Well-formed keys are single characters from the set
174 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form
175 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x'
176 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys
177 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows
178 * single-character subtags).</dd>
179 *
180 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar),
181 * key="x"/value="java-1-7"</dd>
182 * </dl>
183 *
184 * <b>Note:</b> Although BCP 47 requires field values to be registered
185 * in the IANA Language Subtag Registry, the <code>Locale</code> class
186 * does not provide any validation features. The <code>Builder</code>
187 * only checks if an individual field satisfies the syntactic
188 * requirement (is well-formed), but does not validate the value
189 * itself. See {@link Builder} for details.
190 *
191 * <h3><a name="def_locale_extension">Unicode locale/language extension</a></h3>
192 *
193 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional
194 * attributes and keywords to override or refine the default behavior
195 * associated with a locale. A keyword is represented by a pair of
196 * key and type. For example, "nu-thai" indicates that Thai local
197 * digits (value:"thai") should be used for formatting numbers
198 * (key:"nu").
199 *
200 * <p>The keywords are mapped to a BCP 47 extension value using the
201 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above
202 * example, "nu-thai", becomes the extension "u-nu-thai".
203 *
204 * <p>Thus, when a <code>Locale</code> object contains Unicode locale
205 * attributes and keywords,
206 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a
207 * String representing this information, for example, "nu-thai". The
208 * <code>Locale</code> class also provides {@link
209 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and
210 * {@link #getUnicodeLocaleType} which allow you to access Unicode
211 * locale attributes and key/type pairs directly. When represented as
212 * a string, the Unicode Locale Extension lists attributes
213 * alphabetically, followed by key/type sequences with keys listed
214 * alphabetically (the order of subtags comprising a key's type is
215 * fixed when the type is defined)
216 *
217 * <p>A well-formed locale key has the form
218 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the
219 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it
220 * can be empty, or a series of subtags 3-8 alphanums in length). A
221 * well-formed locale attribute has the form
222 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same
223 * form as a locale type subtag).
224 *
225 * <p>The Unicode locale extension specifies optional behavior in
226 * locale-sensitive services. Although the LDML specification defines
227 * various keys and values, actual locale-sensitive service
228 * implementations in a Java Runtime Environment might not support any
229 * particular Unicode locale attributes or key/type pairs.
230 *
231 * <h4>Creating a Locale</h4>
232 *
233 * <p>There are several different ways to create a <code>Locale</code>
234 * object.
235 *
236 * <h5>Builder</h5>
237 *
238 * <p>Using {@link Builder} you can construct a <code>Locale</code> object
239 * that conforms to BCP 47 syntax.
240 *
241 * <h5>Constructors</h5>
242 *
243 * <p>The <code>Locale</code> class provides three constructors:
244 * <blockquote>
245 * <pre>
246 * {@link #Locale(String language)}
247 * {@link #Locale(String language, String country)}
248 * {@link #Locale(String language, String country, String variant)}
249 * </pre>
250 * </blockquote>
251 * These constructors allow you to create a <code>Locale</code> object
252 * with language, country and variant, but you cannot specify
253 * script or extensions.
254 *
255 * <h5>Factory Methods</h5>
256 *
257 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code>
258 * object for a well-formed BCP 47 language tag.
259 *
260 * <h5>Locale Constants</h5>
261 *
262 * <p>The <code>Locale</code> class provides a number of convenient constants
263 * that you can use to create <code>Locale</code> objects for commonly used
264 * locales. For example, the following creates a <code>Locale</code> object
265 * for the United States:
266 * <blockquote>
267 * <pre>
268 * Locale.US
269 * </pre>
270 * </blockquote>
271 *
272 * <h4><a name="LocaleMatching">Locale Matching</a></h4>
273 *
274 * <p>If an application or a system is internationalized and provides localized
275 * resources for multiple locales, it sometimes needs to find one or more
276 * locales (or language tags) which meet each user's specific preferences. Note
277 * that a term "language tag" is used interchangeably with "locale" in this
278 * locale matching documentation.
279 *
280 * <p>In order to do matching a user's preferred locales to a set of language
281 * tags, <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
282 * Language Tags</a> defines two mechanisms: filtering and lookup.
283 * <em>Filtering</em> is used to get all matching locales, whereas
284 * <em>lookup</em> is to choose the best matching locale.
285 * Matching is done case-insensitively. These matching mechanisms are described
286 * in the following sections.
287 *
288 * <p>A user's preference is called a <em>Language Priority List</em> and is
289 * expressed as a list of language ranges. There are syntactically two types of
290 * language ranges: basic and extended. See
291 * {@link Locale.LanguageRange Locale.LanguageRange} for details.
292 *
293 * <h5>Filtering</h5>
294 *
295 * <p>The filtering operation returns all matching language tags. It is defined
296 * in RFC 4647 as follows:
297 * "In filtering, each language range represents the least specific language
298 * tag (that is, the language tag with fewest number of subtags) that is an
299 * acceptable match. All of the language tags in the matching set of tags will
300 * have an equal or greater number of subtags than the language range. Every
301 * non-wildcard subtag in the language range will appear in every one of the
302 * matching language tags."
303 *
304 * <p>There are two types of filtering: filtering for basic language ranges
305 * (called "basic filtering") and filtering for extended language ranges
306 * (called "extended filtering"). They may return different results by what
307 * kind of language ranges are included in the given Language Priority List.
308 * {@link Locale.FilteringMode} is a parameter to specify how filtering should
309 * be done.
310 *
311 * <h5>Lookup</h5>
312 *
313 * <p>The lookup operation returns the best matching language tags. It is
314 * defined in RFC 4647 as follows:
315 * "By contrast with filtering, each language range represents the most
316 * specific tag that is an acceptable match. The first matching tag found,
317 * according to the user's priority, is considered the closest match and is the
318 * item returned."
319 *
320 * <p>For example, if a Language Priority List consists of two language ranges,
321 * {@code "zh-Hant-TW"} and {@code "en-US"}, in prioritized order, lookup
322 * method progressively searches the language tags below in order to find the
323 * best matching language tag.
324 * <blockquote>
325 * <pre>
326 * 1. zh-Hant-TW
327 * 2. zh-Hant
328 * 3. zh
329 * 4. en-US
330 * 5. en
331 * </pre>
332 * </blockquote>
333 * If there is a language tag which matches completely to a language range
334 * above, the language tag is returned.
335 *
336 * <p>{@code "*"} is the special language range, and it is ignored in lookup.
337 *
338 * <p>If multiple language tags match as a result of the subtag {@code '*'}
339 * included in a language range, the first matching language tag returned by
340 * an {@link Iterator} over a {@link Collection} of language tags is treated as
341 * the best matching one.
342 *
343 * <h4>Use of Locale</h4>
344 *
345 * <p>Once you've created a <code>Locale</code> you can query it for information
346 * about itself. Use <code>getCountry</code> to get the country (or region)
347 * code and <code>getLanguage</code> to get the language code.
348 * You can use <code>getDisplayCountry</code> to get the
349 * name of the country suitable for displaying to the user. Similarly,
350 * you can use <code>getDisplayLanguage</code> to get the name of
351 * the language suitable for displaying to the user. Interestingly,
352 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive
353 * and have two versions: one that uses the default
354 * {@link Locale.Category#DISPLAY DISPLAY} locale and one
355 * that uses the locale specified as an argument.
356 *
357 * <p>The Java Platform provides a number of classes that perform locale-sensitive
358 * operations. For example, the <code>NumberFormat</code> class formats
359 * numbers, currency, and percentages in a locale-sensitive manner. Classes
360 * such as <code>NumberFormat</code> have several convenience methods
361 * for creating a default object of that type. For example, the
362 * <code>NumberFormat</code> class provides these three convenience methods
363 * for creating a default <code>NumberFormat</code> object:
364 * <blockquote>
365 * <pre>
366 * NumberFormat.getInstance()
367 * NumberFormat.getCurrencyInstance()
368 * NumberFormat.getPercentInstance()
369 * </pre>
370 * </blockquote>
371 * Each of these methods has two variants; one with an explicit locale
372 * and one without; the latter uses the default
373 * {@link Locale.Category#FORMAT FORMAT} locale:
374 * <blockquote>
375 * <pre>
376 * NumberFormat.getInstance(myLocale)
377 * NumberFormat.getCurrencyInstance(myLocale)
378 * NumberFormat.getPercentInstance(myLocale)
379 * </pre>
380 * </blockquote>
381 * A <code>Locale</code> is the mechanism for identifying the kind of object
382 * (<code>NumberFormat</code>) that you would like to get. The locale is
383 * <STRONG>just</STRONG> a mechanism for identifying objects,
384 * <STRONG>not</STRONG> a container for the objects themselves.
385 *
386 * <h4>Compatibility</h4>
387 *
388 * <p>In order to maintain compatibility with existing usage, Locale's
389 * constructors retain their behavior prior to the Java Runtime
390 * Environment version 1.7. The same is largely true for the
391 * <code>toString</code> method. Thus Locale objects can continue to
392 * be used as they were. In particular, clients who parse the output
393 * of toString into language, country, and variant fields can continue
394 * to do so (although this is strongly discouraged), although the
395 * variant field will have additional information in it if script or
396 * extensions are present.
397 *
398 * <p>In addition, BCP 47 imposes syntax restrictions that are not
399 * imposed by Locale's constructors. This means that conversions
400 * between some Locales and BCP 47 language tags cannot be made without
401 * losing information. Thus <code>toLanguageTag</code> cannot
402 * represent the state of locales whose language, country, or variant
403 * do not conform to BCP 47.
404 *
405 * <p>Because of these issues, it is recommended that clients migrate
406 * away from constructing non-conforming locales and use the
407 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead.
408 * Clients desiring a string representation of the complete locale can
409 * then always rely on <code>toLanguageTag</code> for this purpose.
410 *
411 * <h5><a name="special_cases_constructor">Special cases</a></h5>
412 *
413 * <p>For compatibility reasons, two
414 * non-conforming locales are treated as special cases. These are
415 * <b>{@code ja_JP_JP}</b> and <b>{@code th_TH_TH}</b>. These are ill-formed
416 * in BCP 47 since the variants are too short. To ease migration to BCP 47,
417 * these are treated specially during construction. These two cases (and only
418 * these) cause a constructor to generate an extension, all other values behave
419 * exactly as they did prior to Java 7.
420 *
421 * <p>Java has used {@code ja_JP_JP} to represent Japanese as used in
422 * Japan together with the Japanese Imperial calendar. This is now
423 * representable using a Unicode locale extension, by specifying the
424 * Unicode locale key {@code ca} (for "calendar") and type
425 * {@code japanese}. When the Locale constructor is called with the
426 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is
427 * automatically added.
428 *
429 * <p>Java has used {@code th_TH_TH} to represent Thai as used in
430 * Thailand together with Thai digits. This is also now representable using
431 * a Unicode locale extension, by specifying the Unicode locale key
432 * {@code nu} (for "number") and value {@code thai}. When the Locale
433 * constructor is called with the arguments "th", "TH", "TH", the
434 * extension "u-nu-thai" is automatically added.
435 *
436 * <h5>Serialization</h5>
437 *
438 * <p>During serialization, writeObject writes all fields to the output
439 * stream, including extensions.
440 *
441 * <p>During deserialization, readResolve adds extensions as described
442 * in <a href="#special_cases_constructor">Special Cases</a>, only
443 * for the two cases th_TH_TH and ja_JP_JP.
444 *
445 * <h5>Legacy language codes</h5>
446 *
447 * <p>Locale's constructor has always converted three language codes to
448 * their earlier, obsoleted forms: {@code he} maps to {@code iw},
449 * {@code yi} maps to {@code ji}, and {@code id} maps to
450 * {@code in}. This continues to be the case, in order to not break
451 * backwards compatibility.
452 *
453 * <p>The APIs added in 1.7 map between the old and new language codes,
454 * maintaining the old codes internal to Locale (so that
455 * <code>getLanguage</code> and <code>toString</code> reflect the old
456 * code), but using the new codes in the BCP 47 language tag APIs (so
457 * that <code>toLanguageTag</code> reflects the new one). This
458 * preserves the equivalence between Locales no matter which code or
459 * API is used to construct them. Java's default resource bundle
460 * lookup mechanism also implements this mapping, so that resources
461 * can be named using either convention, see {@link ResourceBundle.Control}.
462 *
463 * <h5>Three-letter language/country(region) codes</h5>
464 *
465 * <p>The Locale constructors have always specified that the language
466 * and the country param be two characters in length, although in
467 * practice they have accepted any length. The specification has now
468 * been relaxed to allow language codes of two to eight characters and
469 * country (region) codes of two to three characters, and in
470 * particular, three-letter language codes and three-digit region
471 * codes as specified in the IANA Language Subtag Registry. For
472 * compatibility, the implementation still does not impose a length
473 * constraint.
474 *
475 * @see Builder
476 * @see ResourceBundle
477 * @see java.text.Format
478 * @see java.text.NumberFormat
479 * @see java.text.Collator
480 * @author Mark Davis
481 * @since 1.1
482 */
483 public final class Locale implements Cloneable, Serializable {
484
485 private static final Cache LOCALECACHE = new Cache();
486
487 /** Useful constant for language.
488 */
489 public static final Locale ENGLISH = createConstant("en", "");
490
491 /** Useful constant for language.
492 */
493 public static final Locale FRENCH = createConstant("fr", "");
494
495 /** Useful constant for language.
496 */
497 public static final Locale GERMAN = createConstant("de", "");
498
499 /** Useful constant for language.
500 */
501 public static final Locale ITALIAN = createConstant("it", "");
502
503 /** Useful constant for language.
504 */
505 public static final Locale JAPANESE = createConstant("ja", "");
506
507 /** Useful constant for language.
508 */
509 public static final Locale KOREAN = createConstant("ko", "");
510
511 /** Useful constant for language.
512 */
513 public static final Locale CHINESE = createConstant("zh", "");
514
515 /** Useful constant for language.
516 */
517 public static final Locale SIMPLIFIED_CHINESE = createConstant("zh", "CN");
518
519 /** Useful constant for language.
520 */
521 public static final Locale TRADITIONAL_CHINESE = createConstant("zh", "TW");
522
523 /** Useful constant for country.
524 */
525 public static final Locale FRANCE = createConstant("fr", "FR");
526
527 /** Useful constant for country.
528 */
529 public static final Locale GERMANY = createConstant("de", "DE");
530
531 /** Useful constant for country.
532 */
533 public static final Locale ITALY = createConstant("it", "IT");
534
535 /** Useful constant for country.
536 */
537 public static final Locale JAPAN = createConstant("ja", "JP");
538
539 /** Useful constant for country.
540 */
541 public static final Locale KOREA = createConstant("ko", "KR");
542
543 /** Useful constant for country.
544 */
545 public static final Locale CHINA = SIMPLIFIED_CHINESE;
546
547 /** Useful constant for country.
548 */
549 public static final Locale PRC = SIMPLIFIED_CHINESE;
550
551 /** Useful constant for country.
552 */
553 public static final Locale TAIWAN = TRADITIONAL_CHINESE;
554
555 /** Useful constant for country.
556 */
557 public static final Locale UK = createConstant("en", "GB");
558
559 /** Useful constant for country.
560 */
561 public static final Locale US = createConstant("en", "US");
562
563 /** Useful constant for country.
564 */
565 public static final Locale CANADA = createConstant("en", "CA");
566
567 /** Useful constant for country.
568 */
569 public static final Locale CANADA_FRENCH = createConstant("fr", "CA");
570
571 /**
572 * Useful constant for the root locale. The root locale is the locale whose
573 * language, country, and variant are empty ("") strings. This is regarded
574 * as the base locale of all locales, and is used as the language/country
575 * neutral locale for the locale sensitive operations.
576 *
577 * @since 1.6
578 */
579 public static final Locale ROOT = createConstant("", "");
580
581 /**
582 * The key for the private use extension ('x').
583 *
584 * @see #getExtension(char)
585 * @see Builder#setExtension(char, String)
586 * @since 1.7
587 */
588 public static final char PRIVATE_USE_EXTENSION = 'x';
589
590 /**
591 * The key for Unicode locale extension ('u').
592 *
593 * @see #getExtension(char)
594 * @see Builder#setExtension(char, String)
595 * @since 1.7
596 */
597 public static final char UNICODE_LOCALE_EXTENSION = 'u';
598
599 /** serialization ID
600 */
601 static final long serialVersionUID = 9149081749638150636L;
602
603 /**
604 * Enum for specifying the type defined in ISO 3166. This enum is used to
605 * retrieve the two-letter ISO3166-1 alpha-2, three-letter ISO3166-1
606 * alpha-3, four-letter ISO3166-3 country codes.
607 *
608 * @see #getISOCountries(Locale.IsoCountryCode)
609 * @since 9
610 */
611 public static enum IsoCountryCode {
612 /**
613 * PART1_ALPHA2 is used to represent the ISO3166-1 alpha-2 two letter
614 * country codes.
615 */
616 PART1_ALPHA2 {
617 @Override
618 Set<String> createCountryCodeSet() {
619 return Set.of(Locale.getISOCountries());
620 }
621 },
622
623 /**
624 *
625 * PART1_ALPHA3 is used to represent the ISO3166-1 alpha-3 three letter
626 * country codes.
627 */
628 PART1_ALPHA3 {
629 @Override
630 Set<String> createCountryCodeSet() {
631 return LocaleISOData.computeISO3166_1Alpha3Countries();
632 }
633 },
634
635 /**
636 * PART3 is used to represent the ISO3166-3 four letter country codes.
637 */
638 PART3 {
639 @Override
640 Set<String> createCountryCodeSet() {
641 return Set.of(LocaleISOData.ISO3166_3);
642 }
643 };
644
645 /**
646 * Concrete implementation of this method attempts to compute value
647 * for iso3166CodesMap for each IsoCountryCode type key.
648 */
649 abstract Set<String> createCountryCodeSet();
650
651 /**
652 * Map to hold country codes for each ISO3166 part.
653 */
654 private static Map<IsoCountryCode, Set<String>> iso3166CodesMap = new ConcurrentHashMap<>();
655
656 /**
657 * This method is called from Locale class to retrieve country code set
658 * for getISOCountries(type)
659 */
660 static Set<String> retrieveISOCountryCodes(IsoCountryCode type) {
661 return iso3166CodesMap.computeIfAbsent(type, IsoCountryCode::createCountryCodeSet);
662 }
663 }
664
665 /**
666 * Display types for retrieving localized names from the name providers.
667 */
668 private static final int DISPLAY_LANGUAGE = 0;
669 private static final int DISPLAY_COUNTRY = 1;
670 private static final int DISPLAY_VARIANT = 2;
671 private static final int DISPLAY_SCRIPT = 3;
672
673 /**
674 * Private constructor used by getInstance method
675 */
676 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) {
677 this.baseLocale = baseLocale;
678 this.localeExtensions = extensions;
679 }
680
681 /**
682 * Construct a locale from language, country and variant.
683 * This constructor normalizes the language value to lowercase and
684 * the country value to uppercase.
685 * <p>
686 * <b>Note:</b>
687 * <ul>
688 * <li>ISO 639 is not a stable standard; some of the language codes it defines
689 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
690 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
691 * API on Locale will return only the OLD codes.
692 * <li>For backward compatibility reasons, this constructor does not make
693 * any syntactic checks on the input.
694 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially,
695 * see <a href="#special_cases_constructor">Special Cases</a> for more information.
696 * </ul>
697 *
698 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
699 * up to 8 characters in length. See the <code>Locale</code> class description about
700 * valid language values.
701 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
702 * See the <code>Locale</code> class description about valid country values.
703 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>.
704 * See the <code>Locale</code> class description for the details.
705 * @exception NullPointerException thrown if any argument is null.
706 */
707 public Locale(String language, String country, String variant) {
708 if (language== null || country == null || variant == null) {
709 throw new NullPointerException();
710 }
711 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant);
712 localeExtensions = getCompatibilityExtensions(language, "", country, variant);
713 }
714
715 /**
716 * Construct a locale from language and country.
717 * This constructor normalizes the language value to lowercase and
718 * the country value to uppercase.
719 * <p>
720 * <b>Note:</b>
721 * <ul>
722 * <li>ISO 639 is not a stable standard; some of the language codes it defines
723 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
724 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
725 * API on Locale will return only the OLD codes.
726 * <li>For backward compatibility reasons, this constructor does not make
727 * any syntactic checks on the input.
728 * </ul>
729 *
730 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
731 * up to 8 characters in length. See the <code>Locale</code> class description about
732 * valid language values.
733 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
734 * See the <code>Locale</code> class description about valid country values.
735 * @exception NullPointerException thrown if either argument is null.
736 */
737 public Locale(String language, String country) {
738 this(language, country, "");
739 }
740
741 /**
742 * Construct a locale from a language code.
743 * This constructor normalizes the language value to lowercase.
744 * <p>
745 * <b>Note:</b>
746 * <ul>
747 * <li>ISO 639 is not a stable standard; some of the language codes it defines
748 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
749 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
750 * API on Locale will return only the OLD codes.
751 * <li>For backward compatibility reasons, this constructor does not make
752 * any syntactic checks on the input.
753 * </ul>
754 *
755 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
756 * up to 8 characters in length. See the <code>Locale</code> class description about
757 * valid language values.
758 * @exception NullPointerException thrown if argument is null.
759 * @since 1.4
760 */
761 public Locale(String language) {
762 this(language, "", "");
763 }
764
765 /**
766 * This method must be called only for creating the Locale.*
767 * constants due to making shortcuts.
768 */
769 private static Locale createConstant(String lang, String country) {
770 BaseLocale base = BaseLocale.createInstance(lang, country);
771 return getInstance(base, null);
772 }
773
774 /**
775 * Returns a <code>Locale</code> constructed from the given
776 * <code>language</code>, <code>country</code> and
777 * <code>variant</code>. If the same <code>Locale</code> instance
778 * is available in the cache, then that instance is
779 * returned. Otherwise, a new <code>Locale</code> instance is
780 * created and cached.
781 *
782 * @param language lowercase 2 to 8 language code.
783 * @param country uppercase two-letter ISO-3166 code and numeric-3 UN M.49 area code.
784 * @param variant vendor and browser specific code. See class description.
785 * @return the <code>Locale</code> instance requested
786 * @exception NullPointerException if any argument is null.
787 */
788 static Locale getInstance(String language, String country, String variant) {
789 return getInstance(language, "", country, variant, null);
790 }
791
792 static Locale getInstance(String language, String script, String country,
793 String variant, LocaleExtensions extensions) {
794 if (language== null || script == null || country == null || variant == null) {
795 throw new NullPointerException();
796 }
797
798 if (extensions == null) {
799 extensions = getCompatibilityExtensions(language, script, country, variant);
800 }
801
802 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant);
803 return getInstance(baseloc, extensions);
804 }
805
806 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) {
807 LocaleKey key = new LocaleKey(baseloc, extensions);
808 return LOCALECACHE.get(key);
809 }
810
811 private static class Cache extends LocaleObjectCache<LocaleKey, Locale> {
812 private Cache() {
813 }
814
815 @Override
816 protected Locale createObject(LocaleKey key) {
817 return new Locale(key.base, key.exts);
818 }
819 }
820
821 private static final class LocaleKey {
822 private final BaseLocale base;
823 private final LocaleExtensions exts;
824 private final int hash;
825
826 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) {
827 base = baseLocale;
828 exts = extensions;
829
830 // Calculate the hash value here because it's always used.
831 int h = base.hashCode();
832 if (exts != null) {
833 h ^= exts.hashCode();
834 }
835 hash = h;
836 }
837
838 @Override
839 public boolean equals(Object obj) {
840 if (this == obj) {
841 return true;
842 }
843 if (!(obj instanceof LocaleKey)) {
844 return false;
845 }
846 LocaleKey other = (LocaleKey)obj;
847 if (hash != other.hash || !base.equals(other.base)) {
848 return false;
849 }
850 if (exts == null) {
851 return other.exts == null;
852 }
853 return exts.equals(other.exts);
854 }
855
856 @Override
857 public int hashCode() {
858 return hash;
859 }
860 }
861
862 /**
863 * Gets the current value of the default locale for this instance
864 * of the Java Virtual Machine.
865 * <p>
866 * The Java Virtual Machine sets the default locale during startup
867 * based on the host environment. It is used by many locale-sensitive
868 * methods if no locale is explicitly specified.
869 * It can be changed using the
870 * {@link #setDefault(java.util.Locale) setDefault} method.
871 *
872 * @return the default locale for this instance of the Java Virtual Machine
873 */
874 public static Locale getDefault() {
875 // do not synchronize this method - see 4071298
876 return defaultLocale;
877 }
878
879 /**
880 * Gets the current value of the default locale for the specified Category
881 * for this instance of the Java Virtual Machine.
882 * <p>
883 * The Java Virtual Machine sets the default locale during startup based
884 * on the host environment. It is used by many locale-sensitive methods
885 * if no locale is explicitly specified. It can be changed using the
886 * setDefault(Locale.Category, Locale) method.
887 *
888 * @param category - the specified category to get the default locale
889 * @throws NullPointerException if category is null
890 * @return the default locale for the specified Category for this instance
891 * of the Java Virtual Machine
892 * @see #setDefault(Locale.Category, Locale)
893 * @since 1.7
894 */
895 public static Locale getDefault(Locale.Category category) {
896 // do not synchronize this method - see 4071298
897 switch (category) {
898 case DISPLAY:
899 if (defaultDisplayLocale == null) {
900 synchronized(Locale.class) {
901 if (defaultDisplayLocale == null) {
902 defaultDisplayLocale = initDefault(category);
903 }
904 }
905 }
906 return defaultDisplayLocale;
907 case FORMAT:
908 if (defaultFormatLocale == null) {
909 synchronized(Locale.class) {
910 if (defaultFormatLocale == null) {
911 defaultFormatLocale = initDefault(category);
912 }
913 }
914 }
915 return defaultFormatLocale;
916 default:
917 assert false: "Unknown Category";
918 }
919 return getDefault();
920 }
921
922 private static Locale initDefault() {
923 String language, region, script, country, variant;
924 Properties props = GetPropertyAction.privilegedGetProperties();
925 language = props.getProperty("user.language", "en");
926 // for compatibility, check for old user.region property
927 region = props.getProperty("user.region");
928 if (region != null) {
929 // region can be of form country, country_variant, or _variant
930 int i = region.indexOf('_');
931 if (i >= 0) {
932 country = region.substring(0, i);
933 variant = region.substring(i + 1);
934 } else {
935 country = region;
936 variant = "";
937 }
938 script = "";
939 } else {
940 script = props.getProperty("user.script", "");
941 country = props.getProperty("user.country", "");
942 variant = props.getProperty("user.variant", "");
943 }
944
945 return getInstance(language, script, country, variant, null);
946 }
947
948 private static Locale initDefault(Locale.Category category) {
949 Properties props = GetPropertyAction.privilegedGetProperties();
950 return getInstance(
951 props.getProperty(category.languageKey,
952 defaultLocale.getLanguage()),
953 props.getProperty(category.scriptKey,
954 defaultLocale.getScript()),
955 props.getProperty(category.countryKey,
956 defaultLocale.getCountry()),
957 props.getProperty(category.variantKey,
958 defaultLocale.getVariant()),
959 null);
960 }
961
962 /**
963 * Sets the default locale for this instance of the Java Virtual Machine.
964 * This does not affect the host locale.
965 * <p>
966 * If there is a security manager, its <code>checkPermission</code>
967 * method is called with a <code>PropertyPermission("user.language", "write")</code>
968 * permission before the default locale is changed.
969 * <p>
970 * The Java Virtual Machine sets the default locale during startup
971 * based on the host environment. It is used by many locale-sensitive
972 * methods if no locale is explicitly specified.
973 * <p>
974 * Since changing the default locale may affect many different areas
975 * of functionality, this method should only be used if the caller
976 * is prepared to reinitialize locale-sensitive code running
977 * within the same Java Virtual Machine.
978 * <p>
979 * By setting the default locale with this method, all of the default
980 * locales for each Category are also set to the specified default locale.
981 *
982 * @throws SecurityException
983 * if a security manager exists and its
984 * <code>checkPermission</code> method doesn't allow the operation.
985 * @throws NullPointerException if <code>newLocale</code> is null
986 * @param newLocale the new default locale
987 * @see SecurityManager#checkPermission
988 * @see java.util.PropertyPermission
989 */
990 public static synchronized void setDefault(Locale newLocale) {
991 setDefault(Category.DISPLAY, newLocale);
992 setDefault(Category.FORMAT, newLocale);
993 defaultLocale = newLocale;
994 }
995
996 /**
997 * Sets the default locale for the specified Category for this instance
998 * of the Java Virtual Machine. This does not affect the host locale.
999 * <p>
1000 * If there is a security manager, its checkPermission method is called
1001 * with a PropertyPermission("user.language", "write") permission before
1002 * the default locale is changed.
1003 * <p>
1004 * The Java Virtual Machine sets the default locale during startup based
1005 * on the host environment. It is used by many locale-sensitive methods
1006 * if no locale is explicitly specified.
1007 * <p>
1008 * Since changing the default locale may affect many different areas of
1009 * functionality, this method should only be used if the caller is
1010 * prepared to reinitialize locale-sensitive code running within the
1011 * same Java Virtual Machine.
1012 *
1013 * @param category - the specified category to set the default locale
1014 * @param newLocale - the new default locale
1015 * @throws SecurityException if a security manager exists and its
1016 * checkPermission method doesn't allow the operation.
1017 * @throws NullPointerException if category and/or newLocale is null
1018 * @see SecurityManager#checkPermission(java.security.Permission)
1019 * @see PropertyPermission
1020 * @see #getDefault(Locale.Category)
1021 * @since 1.7
1022 */
1023 public static synchronized void setDefault(Locale.Category category,
1024 Locale newLocale) {
1025 if (category == null)
1026 throw new NullPointerException("Category cannot be NULL");
1027 if (newLocale == null)
1028 throw new NullPointerException("Can't set default locale to NULL");
1029
1030 SecurityManager sm = System.getSecurityManager();
1031 if (sm != null) sm.checkPermission(new PropertyPermission
1032 ("user.language", "write"));
1033 switch (category) {
1034 case DISPLAY:
1035 defaultDisplayLocale = newLocale;
1036 break;
1037 case FORMAT:
1038 defaultFormatLocale = newLocale;
1039 break;
1040 default:
1041 assert false: "Unknown Category";
1042 }
1043 }
1044
1045 /**
1046 * Returns an array of all installed locales.
1047 * The returned array represents the union of locales supported
1048 * by the Java runtime environment and by installed
1049 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider}
1050 * implementations. It must contain at least a <code>Locale</code>
1051 * instance equal to {@link java.util.Locale#US Locale.US}.
1052 *
1053 * @return An array of installed locales.
1054 */
1055 public static Locale[] getAvailableLocales() {
1056 return LocaleServiceProviderPool.getAllAvailableLocales();
1057 }
1058
1059 /**
1060 * Returns a list of all 2-letter country codes defined in ISO 3166.
1061 * Can be used to create Locales.
1062 * This method is equivalent to {@link #getISOCountries(Locale.IsoCountryCode type)}
1063 * with {@code type} {@link IsoCountryCode#PART1_ALPHA2}.
1064 * <p>
1065 * <b>Note:</b> The <code>Locale</code> class also supports other codes for
1066 * country (region), such as 3-letter numeric UN M.49 area codes.
1067 * Therefore, the list returned by this method does not contain ALL valid
1068 * codes that can be used to create Locales.
1069 * <p>
1070 * Note that this method does not return obsolete 2-letter country codes.
1071 * ISO3166-3 codes which designate country codes for those obsolete codes,
1072 * can be retrieved from {@link #getISOCountries(Locale.IsoCountryCode type)} with
1073 * {@code type} {@link IsoCountryCode#PART3}.
1074 * @return An array of ISO 3166 two-letter country codes.
1075 */
1076 public static String[] getISOCountries() {
1077 if (isoCountries == null) {
1078 isoCountries = getISO2Table(LocaleISOData.isoCountryTable);
1079 }
1080 String[] result = new String[isoCountries.length];
1081 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length);
1082 return result;
1083 }
1084
1085 /**
1086 * Returns a {@code Set} of ISO3166 country codes for the specified type.
1087 *
1088 * @param type {@link Locale.IsoCountryCode} specified ISO code type.
1089 * @see java.util.Locale.IsoCountryCode
1090 * @throws NullPointerException if type is null
1091 * @return a {@code Set} of ISO country codes for the specified type.
1092 * @since 9
1093 */
1094 public static Set<String> getISOCountries(IsoCountryCode type) {
1095 Objects.requireNonNull(type);
1096 return IsoCountryCode.retrieveISOCountryCodes(type);
1097 }
1098
1099 /**
1100 * Returns a list of all 2-letter language codes defined in ISO 639.
1101 * Can be used to create Locales.
1102 * <p>
1103 * <b>Note:</b>
1104 * <ul>
1105 * <li>ISO 639 is not a stable standard— some languages' codes have changed.
1106 * The list this function returns includes both the new and the old codes for the
1107 * languages whose codes have changed.
1108 * <li>The <code>Locale</code> class also supports language codes up to
1109 * 8 characters in length. Therefore, the list returned by this method does
1110 * not contain ALL valid codes that can be used to create Locales.
1111 * </ul>
1112 *
1113 * @return An array of ISO 639 two-letter language codes.
1114 */
1115 public static String[] getISOLanguages() {
1116 if (isoLanguages == null) {
1117 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable);
1118 }
1119 String[] result = new String[isoLanguages.length];
1120 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length);
1121 return result;
1122 }
1123
1124 private static String[] getISO2Table(String table) {
1125 int len = table.length() / 5;
1126 String[] isoTable = new String[len];
1127 for (int i = 0, j = 0; i < len; i++, j += 5) {
1128 isoTable[i] = table.substring(j, j + 2);
1129 }
1130 return isoTable;
1131 }
1132
1133 /**
1134 * Returns the language code of this Locale.
1135 *
1136 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed.
1137 * Locale's constructor recognizes both the new and the old codes for the languages
1138 * whose codes have changed, but this function always returns the old code. If you
1139 * want to check for a specific language whose code has changed, don't do
1140 * <pre>
1141 * if (locale.getLanguage().equals("he")) // BAD!
1142 * ...
1143 * </pre>
1144 * Instead, do
1145 * <pre>
1146 * if (locale.getLanguage().equals(new Locale("he").getLanguage()))
1147 * ...
1148 * </pre>
1149 * @return The language code, or the empty string if none is defined.
1150 * @see #getDisplayLanguage
1151 */
1152 public String getLanguage() {
1153 return baseLocale.getLanguage();
1154 }
1155
1156 /**
1157 * Returns the script for this locale, which should
1158 * either be the empty string or an ISO 15924 4-letter script
1159 * code. The first letter is uppercase and the rest are
1160 * lowercase, for example, 'Latn', 'Cyrl'.
1161 *
1162 * @return The script code, or the empty string if none is defined.
1163 * @see #getDisplayScript
1164 * @since 1.7
1165 */
1166 public String getScript() {
1167 return baseLocale.getScript();
1168 }
1169
1170 /**
1171 * Returns the country/region code for this locale, which should
1172 * either be the empty string, an uppercase ISO 3166 2-letter code,
1173 * or a UN M.49 3-digit code.
1174 *
1175 * @return The country/region code, or the empty string if none is defined.
1176 * @see #getDisplayCountry
1177 */
1178 public String getCountry() {
1179 return baseLocale.getRegion();
1180 }
1181
1182 /**
1183 * Returns the variant code for this locale.
1184 *
1185 * @return The variant code, or the empty string if none is defined.
1186 * @see #getDisplayVariant
1187 */
1188 public String getVariant() {
1189 return baseLocale.getVariant();
1190 }
1191
1192 /**
1193 * Returns {@code true} if this {@code Locale} has any <a href="#def_extensions">
1194 * extensions</a>.
1195 *
1196 * @return {@code true} if this {@code Locale} has any extensions
1197 * @since 1.8
1198 */
1199 public boolean hasExtensions() {
1200 return localeExtensions != null;
1201 }
1202
1203 /**
1204 * Returns a copy of this {@code Locale} with no <a href="#def_extensions">
1205 * extensions</a>. If this {@code Locale} has no extensions, this {@code Locale}
1206 * is returned.
1207 *
1208 * @return a copy of this {@code Locale} with no extensions, or {@code this}
1209 * if {@code this} has no extensions
1210 * @since 1.8
1211 */
1212 public Locale stripExtensions() {
1213 return hasExtensions() ? Locale.getInstance(baseLocale, null) : this;
1214 }
1215
1216 /**
1217 * Returns the extension (or private use) value associated with
1218 * the specified key, or null if there is no extension
1219 * associated with the key. To be well-formed, the key must be one
1220 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so
1221 * for example 'z' and 'Z' represent the same extension.
1222 *
1223 * @param key the extension key
1224 * @return The extension, or null if this locale defines no
1225 * extension for the specified key.
1226 * @throws IllegalArgumentException if key is not well-formed
1227 * @see #PRIVATE_USE_EXTENSION
1228 * @see #UNICODE_LOCALE_EXTENSION
1229 * @since 1.7
1230 */
1231 public String getExtension(char key) {
1232 if (!LocaleExtensions.isValidKey(key)) {
1233 throw new IllegalArgumentException("Ill-formed extension key: " + key);
1234 }
1235 return hasExtensions() ? localeExtensions.getExtensionValue(key) : null;
1236 }
1237
1238 /**
1239 * Returns the set of extension keys associated with this locale, or the
1240 * empty set if it has no extensions. The returned set is unmodifiable.
1241 * The keys will all be lower-case.
1242 *
1243 * @return The set of extension keys, or the empty set if this locale has
1244 * no extensions.
1245 * @since 1.7
1246 */
1247 public Set<Character> getExtensionKeys() {
1248 if (!hasExtensions()) {
1249 return Collections.emptySet();
1250 }
1251 return localeExtensions.getKeys();
1252 }
1253
1254 /**
1255 * Returns the set of unicode locale attributes associated with
1256 * this locale, or the empty set if it has no attributes. The
1257 * returned set is unmodifiable.
1258 *
1259 * @return The set of attributes.
1260 * @since 1.7
1261 */
1262 public Set<String> getUnicodeLocaleAttributes() {
1263 if (!hasExtensions()) {
1264 return Collections.emptySet();
1265 }
1266 return localeExtensions.getUnicodeLocaleAttributes();
1267 }
1268
1269 /**
1270 * Returns the Unicode locale type associated with the specified Unicode locale key
1271 * for this locale. Returns the empty string for keys that are defined with no type.
1272 * Returns null if the key is not defined. Keys are case-insensitive. The key must
1273 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is
1274 * thrown.
1275 *
1276 * @param key the Unicode locale key
1277 * @return The Unicode locale type associated with the key, or null if the
1278 * locale does not define the key.
1279 * @throws IllegalArgumentException if the key is not well-formed
1280 * @throws NullPointerException if <code>key</code> is null
1281 * @since 1.7
1282 */
1283 public String getUnicodeLocaleType(String key) {
1284 if (!isUnicodeExtensionKey(key)) {
1285 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key);
1286 }
1287 return hasExtensions() ? localeExtensions.getUnicodeLocaleType(key) : null;
1288 }
1289
1290 /**
1291 * Returns the set of Unicode locale keys defined by this locale, or the empty set if
1292 * this locale has none. The returned set is immutable. Keys are all lower case.
1293 *
1294 * @return The set of Unicode locale keys, or the empty set if this locale has
1295 * no Unicode locale keywords.
1296 * @since 1.7
1297 */
1298 public Set<String> getUnicodeLocaleKeys() {
1299 if (localeExtensions == null) {
1300 return Collections.emptySet();
1301 }
1302 return localeExtensions.getUnicodeLocaleKeys();
1303 }
1304
1305 /**
1306 * Package locale method returning the Locale's BaseLocale,
1307 * used by ResourceBundle
1308 * @return base locale of this Locale
1309 */
1310 BaseLocale getBaseLocale() {
1311 return baseLocale;
1312 }
1313
1314 /**
1315 * Package private method returning the Locale's LocaleExtensions,
1316 * used by ResourceBundle.
1317 * @return locale extensions of this Locale,
1318 * or {@code null} if no extensions are defined
1319 */
1320 LocaleExtensions getLocaleExtensions() {
1321 return localeExtensions;
1322 }
1323
1324 /**
1325 * Returns a string representation of this <code>Locale</code>
1326 * object, consisting of language, country, variant, script,
1327 * and extensions as below:
1328 * <blockquote>
1329 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "_" + extensions
1330 * </blockquote>
1331 *
1332 * Language is always lower case, country is always upper case, script is always title
1333 * case, and extensions are always lower case. Extensions and private use subtags
1334 * will be in canonical order as explained in {@link #toLanguageTag}.
1335 *
1336 * <p>When the locale has neither script nor extensions, the result is the same as in
1337 * Java 6 and prior.
1338 *
1339 * <p>If both the language and country fields are missing, this function will return
1340 * the empty string, even if the variant, script, or extensions field is present (you
1341 * can't have a locale with just a variant, the variant must accompany a well-formed
1342 * language or country code).
1343 *
1344 * <p>If script or extensions are present and variant is missing, no underscore is
1345 * added before the "#".
1346 *
1347 * <p>This behavior is designed to support debugging and to be compatible with
1348 * previous uses of <code>toString</code> that expected language, country, and variant
1349 * fields only. To represent a Locale as a String for interchange purposes, use
1350 * {@link #toLanguageTag}.
1351 *
1352 * <p>Examples: <ul>
1353 * <li>{@code en}</li>
1354 * <li>{@code de_DE}</li>
1355 * <li>{@code _GB}</li>
1356 * <li>{@code en_US_WIN}</li>
1357 * <li>{@code de__POSIX}</li>
1358 * <li>{@code zh_CN_#Hans}</li>
1359 * <li>{@code zh_TW_#Hant_x-java}</li>
1360 * <li>{@code th_TH_TH_#u-nu-thai}</li></ul>
1361 *
1362 * @return A string representation of the Locale, for debugging.
1363 * @see #getDisplayName
1364 * @see #toLanguageTag
1365 */
1366 @Override
1367 public final String toString() {
1368 boolean l = (baseLocale.getLanguage().length() != 0);
1369 boolean s = (baseLocale.getScript().length() != 0);
1370 boolean r = (baseLocale.getRegion().length() != 0);
1371 boolean v = (baseLocale.getVariant().length() != 0);
1372 boolean e = (localeExtensions != null && localeExtensions.getID().length() != 0);
1373
1374 StringBuilder result = new StringBuilder(baseLocale.getLanguage());
1375 if (r || (l && (v || s || e))) {
1376 result.append('_')
1377 .append(baseLocale.getRegion()); // This may just append '_'
1378 }
1379 if (v && (l || r)) {
1380 result.append('_')
1381 .append(baseLocale.getVariant());
1382 }
1383
1384 if (s && (l || r)) {
1385 result.append("_#")
1386 .append(baseLocale.getScript());
1387 }
1388
1389 if (e && (l || r)) {
1390 result.append('_');
1391 if (!s) {
1392 result.append('#');
1393 }
1394 result.append(localeExtensions.getID());
1395 }
1396
1397 return result.toString();
1398 }
1399
1400 /**
1401 * Returns a well-formed IETF BCP 47 language tag representing
1402 * this locale.
1403 *
1404 * <p>If this <code>Locale</code> has a language, country, or
1405 * variant that does not satisfy the IETF BCP 47 language tag
1406 * syntax requirements, this method handles these fields as
1407 * described below:
1408 *
1409 * <p><b>Language:</b> If language is empty, or not <a
1410 * href="#def_language" >well-formed</a> (for example "a" or
1411 * "e2"), it will be emitted as "und" (Undetermined).
1412 *
1413 * <p><b>Country:</b> If country is not <a
1414 * href="#def_region">well-formed</a> (for example "12" or "USA"),
1415 * it will be omitted.
1416 *
1417 * <p><b>Variant:</b> If variant <b>is</b> <a
1418 * href="#def_variant">well-formed</a>, each sub-segment
1419 * (delimited by '-' or '_') is emitted as a subtag. Otherwise:
1420 * <ul>
1421 *
1422 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code>
1423 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first
1424 * ill-formed sub-segment and all following will be appended to
1425 * the private use subtag. The first appended subtag will be
1426 * "lvariant", followed by the sub-segments in order, separated by
1427 * hyphen. For example, "x-lvariant-WIN",
1428 * "Oracle-x-lvariant-JDK-Standard-Edition".
1429 *
1430 * <li>if any sub-segment does not match
1431 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated
1432 * and the problematic sub-segment and all following sub-segments
1433 * will be omitted. If the remainder is non-empty, it will be
1434 * emitted as a private use subtag as above (even if the remainder
1435 * turns out to be well-formed). For example,
1436 * "Solaris_isjustthecoolestthing" is emitted as
1437 * "x-lvariant-Solaris", not as "solaris".</li></ul>
1438 *
1439 * <p><b>Special Conversions:</b> Java supports some old locale
1440 * representations, including deprecated ISO language codes,
1441 * for compatibility. This method performs the following
1442 * conversions:
1443 * <ul>
1444 *
1445 * <li>Deprecated ISO language codes "iw", "ji", and "in" are
1446 * converted to "he", "yi", and "id", respectively.
1447 *
1448 * <li>A locale with language "no", country "NO", and variant
1449 * "NY", representing Norwegian Nynorsk (Norway), is converted
1450 * to a language tag "nn-NO".</li></ul>
1451 *
1452 * <p><b>Note:</b> Although the language tag created by this
1453 * method is well-formed (satisfies the syntax requirements
1454 * defined by the IETF BCP 47 specification), it is not
1455 * necessarily a valid BCP 47 language tag. For example,
1456 * <pre>
1457 * new Locale("xx", "YY").toLanguageTag();</pre>
1458 *
1459 * will return "xx-YY", but the language subtag "xx" and the
1460 * region subtag "YY" are invalid because they are not registered
1461 * in the IANA Language Subtag Registry.
1462 *
1463 * @return a BCP47 language tag representing the locale
1464 * @see #forLanguageTag(String)
1465 * @since 1.7
1466 */
1467 public String toLanguageTag() {
1468 if (languageTag != null) {
1469 return languageTag;
1470 }
1471
1472 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions);
1473 StringBuilder buf = new StringBuilder();
1474
1475 String subtag = tag.getLanguage();
1476 if (subtag.length() > 0) {
1477 buf.append(LanguageTag.canonicalizeLanguage(subtag));
1478 }
1479
1480 subtag = tag.getScript();
1481 if (subtag.length() > 0) {
1482 buf.append(LanguageTag.SEP);
1483 buf.append(LanguageTag.canonicalizeScript(subtag));
1484 }
1485
1486 subtag = tag.getRegion();
1487 if (subtag.length() > 0) {
1488 buf.append(LanguageTag.SEP);
1489 buf.append(LanguageTag.canonicalizeRegion(subtag));
1490 }
1491
1492 List<String>subtags = tag.getVariants();
1493 for (String s : subtags) {
1494 buf.append(LanguageTag.SEP);
1495 // preserve casing
1496 buf.append(s);
1497 }
1498
1499 subtags = tag.getExtensions();
1500 for (String s : subtags) {
1501 buf.append(LanguageTag.SEP);
1502 buf.append(LanguageTag.canonicalizeExtension(s));
1503 }
1504
1505 subtag = tag.getPrivateuse();
1506 if (subtag.length() > 0) {
1507 if (buf.length() > 0) {
1508 buf.append(LanguageTag.SEP);
1509 }
1510 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP);
1511 // preserve casing
1512 buf.append(subtag);
1513 }
1514
1515 String langTag = buf.toString();
1516 synchronized (this) {
1517 if (languageTag == null) {
1518 languageTag = langTag;
1519 }
1520 }
1521 return languageTag;
1522 }
1523
1524 /**
1525 * Returns a locale for the specified IETF BCP 47 language tag string.
1526 *
1527 * <p>If the specified language tag contains any ill-formed subtags,
1528 * the first such subtag and all following subtags are ignored. Compare
1529 * to {@link Locale.Builder#setLanguageTag} which throws an exception
1530 * in this case.
1531 *
1532 * <p>The following <b>conversions</b> are performed:<ul>
1533 *
1534 * <li>The language code "und" is mapped to language "".
1535 *
1536 * <li>The language codes "he", "yi", and "id" are mapped to "iw",
1537 * "ji", and "in" respectively. (This is the same canonicalization
1538 * that's done in Locale's constructors.)
1539 *
1540 * <li>The portion of a private use subtag prefixed by "lvariant",
1541 * if any, is removed and appended to the variant field in the
1542 * result locale (without case normalization). If it is then
1543 * empty, the private use subtag is discarded:
1544 *
1545 * <pre>
1546 * Locale loc;
1547 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX");
1548 * loc.getVariant(); // returns "POSIX"
1549 * loc.getExtension('x'); // returns null
1550 *
1551 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def");
1552 * loc.getVariant(); // returns "POSIX_Abc_Def"
1553 * loc.getExtension('x'); // returns "urp"
1554 * </pre>
1555 *
1556 * <li>When the languageTag argument contains an extlang subtag,
1557 * the first such subtag is used as the language, and the primary
1558 * language subtag and other extlang subtags are ignored:
1559 *
1560 * <pre>
1561 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao"
1562 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US"
1563 * </pre>
1564 *
1565 * <li>Case is normalized except for variant tags, which are left
1566 * unchanged. Language is normalized to lower case, script to
1567 * title case, country to upper case, and extensions to lower
1568 * case.
1569 *
1570 * <li>If, after processing, the locale would exactly match either
1571 * ja_JP_JP or th_TH_TH with no extensions, the appropriate
1572 * extensions are added as though the constructor had been called:
1573 *
1574 * <pre>
1575 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag();
1576 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP"
1577 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag();
1578 * // returns "th-TH-u-nu-thai-x-lvariant-TH"
1579 * </pre></ul>
1580 *
1581 * <p>This implements the 'Language-Tag' production of BCP47, and
1582 * so supports grandfathered (regular and irregular) as well as
1583 * private use language tags. Stand alone private use tags are
1584 * represented as empty language and extension 'x-whatever',
1585 * and grandfathered tags are converted to their canonical replacements
1586 * where they exist.
1587 *
1588 * <p>Grandfathered tags with canonical replacements are as follows:
1589 *
1590 * <table summary="Grandfathered tags with canonical replacements">
1591 * <tbody align="center">
1592 * <tr><th>grandfathered tag</th><th> </th><th>modern replacement</th></tr>
1593 * <tr><td>art-lojban</td><td> </td><td>jbo</td></tr>
1594 * <tr><td>i-ami</td><td> </td><td>ami</td></tr>
1595 * <tr><td>i-bnn</td><td> </td><td>bnn</td></tr>
1596 * <tr><td>i-hak</td><td> </td><td>hak</td></tr>
1597 * <tr><td>i-klingon</td><td> </td><td>tlh</td></tr>
1598 * <tr><td>i-lux</td><td> </td><td>lb</td></tr>
1599 * <tr><td>i-navajo</td><td> </td><td>nv</td></tr>
1600 * <tr><td>i-pwn</td><td> </td><td>pwn</td></tr>
1601 * <tr><td>i-tao</td><td> </td><td>tao</td></tr>
1602 * <tr><td>i-tay</td><td> </td><td>tay</td></tr>
1603 * <tr><td>i-tsu</td><td> </td><td>tsu</td></tr>
1604 * <tr><td>no-bok</td><td> </td><td>nb</td></tr>
1605 * <tr><td>no-nyn</td><td> </td><td>nn</td></tr>
1606 * <tr><td>sgn-BE-FR</td><td> </td><td>sfb</td></tr>
1607 * <tr><td>sgn-BE-NL</td><td> </td><td>vgt</td></tr>
1608 * <tr><td>sgn-CH-DE</td><td> </td><td>sgg</td></tr>
1609 * <tr><td>zh-guoyu</td><td> </td><td>cmn</td></tr>
1610 * <tr><td>zh-hakka</td><td> </td><td>hak</td></tr>
1611 * <tr><td>zh-min-nan</td><td> </td><td>nan</td></tr>
1612 * <tr><td>zh-xiang</td><td> </td><td>hsn</td></tr>
1613 * </tbody>
1614 * </table>
1615 *
1616 * <p>Grandfathered tags with no modern replacement will be
1617 * converted as follows:
1618 *
1619 * <table summary="Grandfathered tags with no modern replacement">
1620 * <tbody align="center">
1621 * <tr><th>grandfathered tag</th><th> </th><th>converts to</th></tr>
1622 * <tr><td>cel-gaulish</td><td> </td><td>xtg-x-cel-gaulish</td></tr>
1623 * <tr><td>en-GB-oed</td><td> </td><td>en-GB-x-oed</td></tr>
1624 * <tr><td>i-default</td><td> </td><td>en-x-i-default</td></tr>
1625 * <tr><td>i-enochian</td><td> </td><td>und-x-i-enochian</td></tr>
1626 * <tr><td>i-mingo</td><td> </td><td>see-x-i-mingo</td></tr>
1627 * <tr><td>zh-min</td><td> </td><td>nan-x-zh-min</td></tr>
1628 * </tbody>
1629 * </table>
1630 *
1631 * <p>For a list of all grandfathered tags, see the
1632 * IANA Language Subtag Registry (search for "Type: grandfathered").
1633 *
1634 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code>
1635 * and <code>forLanguageTag</code> will round-trip.
1636 *
1637 * @param languageTag the language tag
1638 * @return The locale that best represents the language tag.
1639 * @throws NullPointerException if <code>languageTag</code> is <code>null</code>
1640 * @see #toLanguageTag()
1641 * @see java.util.Locale.Builder#setLanguageTag(String)
1642 * @since 1.7
1643 */
1644 public static Locale forLanguageTag(String languageTag) {
1645 LanguageTag tag = LanguageTag.parse(languageTag, null);
1646 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
1647 bldr.setLanguageTag(tag);
1648 BaseLocale base = bldr.getBaseLocale();
1649 LocaleExtensions exts = bldr.getLocaleExtensions();
1650 if (exts == null && base.getVariant().length() > 0) {
1651 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(),
1652 base.getRegion(), base.getVariant());
1653 }
1654 return getInstance(base, exts);
1655 }
1656
1657 /**
1658 * Returns a three-letter abbreviation of this locale's language.
1659 * If the language matches an ISO 639-1 two-letter code, the
1660 * corresponding ISO 639-2/T three-letter lowercase code is
1661 * returned. The ISO 639-2 language codes can be found on-line,
1662 * see "Codes for the Representation of Names of Languages Part 2:
1663 * Alpha-3 Code". If the locale specifies a three-letter
1664 * language, the language is returned as is. If the locale does
1665 * not specify a language the empty string is returned.
1666 *
1667 * @return A three-letter abbreviation of this locale's language.
1668 * @exception MissingResourceException Throws MissingResourceException if
1669 * three-letter language abbreviation is not available for this locale.
1670 */
1671 public String getISO3Language() throws MissingResourceException {
1672 String lang = baseLocale.getLanguage();
1673 if (lang.length() == 3) {
1674 return lang;
1675 }
1676
1677 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable);
1678 if (language3 == null) {
1679 throw new MissingResourceException("Couldn't find 3-letter language code for "
1680 + lang, "FormatData_" + toString(), "ShortLanguage");
1681 }
1682 return language3;
1683 }
1684
1685 /**
1686 * Returns a three-letter abbreviation for this locale's country.
1687 * If the country matches an ISO 3166-1 alpha-2 code, the
1688 * corresponding ISO 3166-1 alpha-3 uppercase code is returned.
1689 * If the locale doesn't specify a country, this will be the empty
1690 * string.
1691 *
1692 * <p>The ISO 3166-1 codes can be found on-line.
1693 *
1694 * @return A three-letter abbreviation of this locale's country.
1695 * @exception MissingResourceException Throws MissingResourceException if the
1696 * three-letter country abbreviation is not available for this locale.
1697 */
1698 public String getISO3Country() throws MissingResourceException {
1699 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable);
1700 if (country3 == null) {
1701 throw new MissingResourceException("Couldn't find 3-letter country code for "
1702 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry");
1703 }
1704 return country3;
1705 }
1706
1707 private static String getISO3Code(String iso2Code, String table) {
1708 int codeLength = iso2Code.length();
1709 if (codeLength == 0) {
1710 return "";
1711 }
1712
1713 int tableLength = table.length();
1714 int index = tableLength;
1715 if (codeLength == 2) {
1716 char c1 = iso2Code.charAt(0);
1717 char c2 = iso2Code.charAt(1);
1718 for (index = 0; index < tableLength; index += 5) {
1719 if (table.charAt(index) == c1
1720 && table.charAt(index + 1) == c2) {
1721 break;
1722 }
1723 }
1724 }
1725 return index < tableLength ? table.substring(index + 2, index + 5) : null;
1726 }
1727
1728 /**
1729 * Returns a name for the locale's language that is appropriate for display to the
1730 * user.
1731 * If possible, the name returned will be localized for the default
1732 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1733 * For example, if the locale is fr_FR and the default
1734 * {@link Locale.Category#DISPLAY DISPLAY} locale
1735 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1736 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1737 * getDisplayLanguage() will return "anglais".
1738 * If the name returned cannot be localized for the default
1739 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1740 * (say, we don't have a Japanese name for Croatian),
1741 * this function falls back on the English name, and uses the ISO code as a last-resort
1742 * value. If the locale doesn't specify a language, this function returns the empty string.
1743 *
1744 * @return The name of the display language.
1745 */
1746 public final String getDisplayLanguage() {
1747 return getDisplayLanguage(getDefault(Category.DISPLAY));
1748 }
1749
1750 /**
1751 * Returns a name for the locale's language that is appropriate for display to the
1752 * user.
1753 * If possible, the name returned will be localized according to inLocale.
1754 * For example, if the locale is fr_FR and inLocale
1755 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1756 * inLocale is fr_FR, getDisplayLanguage() will return "anglais".
1757 * If the name returned cannot be localized according to inLocale,
1758 * (say, we don't have a Japanese name for Croatian),
1759 * this function falls back on the English name, and finally
1760 * on the ISO code as a last-resort value. If the locale doesn't specify a language,
1761 * this function returns the empty string.
1762 *
1763 * @param inLocale The locale for which to retrieve the display language.
1764 * @return The name of the display language appropriate to the given locale.
1765 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1766 */
1767 public String getDisplayLanguage(Locale inLocale) {
1768 return getDisplayString(baseLocale.getLanguage(), inLocale, DISPLAY_LANGUAGE);
1769 }
1770
1771 /**
1772 * Returns a name for the locale's script that is appropriate for display to
1773 * the user. If possible, the name will be localized for the default
1774 * {@link Locale.Category#DISPLAY DISPLAY} locale. Returns
1775 * the empty string if this locale doesn't specify a script code.
1776 *
1777 * @return the display name of the script code for the current default
1778 * {@link Locale.Category#DISPLAY DISPLAY} locale
1779 * @since 1.7
1780 */
1781 public String getDisplayScript() {
1782 return getDisplayScript(getDefault(Category.DISPLAY));
1783 }
1784
1785 /**
1786 * Returns a name for the locale's script that is appropriate
1787 * for display to the user. If possible, the name will be
1788 * localized for the given locale. Returns the empty string if
1789 * this locale doesn't specify a script code.
1790 *
1791 * @param inLocale The locale for which to retrieve the display script.
1792 * @return the display name of the script code for the current default
1793 * {@link Locale.Category#DISPLAY DISPLAY} locale
1794 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1795 * @since 1.7
1796 */
1797 public String getDisplayScript(Locale inLocale) {
1798 return getDisplayString(baseLocale.getScript(), inLocale, DISPLAY_SCRIPT);
1799 }
1800
1801 /**
1802 * Returns a name for the locale's country that is appropriate for display to the
1803 * user.
1804 * If possible, the name returned will be localized for the default
1805 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1806 * For example, if the locale is fr_FR and the default
1807 * {@link Locale.Category#DISPLAY DISPLAY} locale
1808 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1809 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1810 * getDisplayCountry() will return "Etats-Unis".
1811 * If the name returned cannot be localized for the default
1812 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1813 * (say, we don't have a Japanese name for Croatia),
1814 * this function falls back on the English name, and uses the ISO code as a last-resort
1815 * value. If the locale doesn't specify a country, this function returns the empty string.
1816 *
1817 * @return The name of the country appropriate to the locale.
1818 */
1819 public final String getDisplayCountry() {
1820 return getDisplayCountry(getDefault(Category.DISPLAY));
1821 }
1822
1823 /**
1824 * Returns a name for the locale's country that is appropriate for display to the
1825 * user.
1826 * If possible, the name returned will be localized according to inLocale.
1827 * For example, if the locale is fr_FR and inLocale
1828 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1829 * inLocale is fr_FR, getDisplayCountry() will return "Etats-Unis".
1830 * If the name returned cannot be localized according to inLocale.
1831 * (say, we don't have a Japanese name for Croatia),
1832 * this function falls back on the English name, and finally
1833 * on the ISO code as a last-resort value. If the locale doesn't specify a country,
1834 * this function returns the empty string.
1835 *
1836 * @param inLocale The locale for which to retrieve the display country.
1837 * @return The name of the country appropriate to the given locale.
1838 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1839 */
1840 public String getDisplayCountry(Locale inLocale) {
1841 return getDisplayString(baseLocale.getRegion(), inLocale, DISPLAY_COUNTRY);
1842 }
1843
1844 private String getDisplayString(String code, Locale inLocale, int type) {
1845 if (code.length() == 0) {
1846 return "";
1847 }
1848
1849 if (inLocale == null) {
1850 throw new NullPointerException();
1851 }
1852
1853 LocaleServiceProviderPool pool =
1854 LocaleServiceProviderPool.getPool(LocaleNameProvider.class);
1855 String key = (type == DISPLAY_VARIANT ? "%%"+code : code);
1856 String result = pool.getLocalizedObject(
1857 LocaleNameGetter.INSTANCE,
1858 inLocale, key, type, code);
1859 if (result != null) {
1860 return result;
1861 }
1862
1863 return code;
1864 }
1865
1866 /**
1867 * Returns a name for the locale's variant code that is appropriate for display to the
1868 * user. If possible, the name will be localized for the default
1869 * {@link Locale.Category#DISPLAY DISPLAY} locale. If the locale
1870 * doesn't specify a variant code, this function returns the empty string.
1871 *
1872 * @return The name of the display variant code appropriate to the locale.
1873 */
1874 public final String getDisplayVariant() {
1875 return getDisplayVariant(getDefault(Category.DISPLAY));
1876 }
1877
1878 /**
1879 * Returns a name for the locale's variant code that is appropriate for display to the
1880 * user. If possible, the name will be localized for inLocale. If the locale
1881 * doesn't specify a variant code, this function returns the empty string.
1882 *
1883 * @param inLocale The locale for which to retrieve the display variant code.
1884 * @return The name of the display variant code appropriate to the given locale.
1885 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1886 */
1887 public String getDisplayVariant(Locale inLocale) {
1888 if (baseLocale.getVariant().length() == 0)
1889 return "";
1890
1891 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale);
1892
1893 String names[] = getDisplayVariantArray(inLocale);
1894
1895 // Get the localized patterns for formatting a list, and use
1896 // them to format the list.
1897 return formatList(names,
1898 lr.getLocaleName("ListPattern"),
1899 lr.getLocaleName("ListCompositionPattern"));
1900 }
1901
1902 /**
1903 * Returns a name for the locale that is appropriate for display to the
1904 * user. This will be the values returned by getDisplayLanguage(),
1905 * getDisplayScript(), getDisplayCountry(), and getDisplayVariant() assembled
1906 * into a single string. The the non-empty values are used in order,
1907 * with the second and subsequent names in parentheses. For example:
1908 * <blockquote>
1909 * language (script, country, variant)<br>
1910 * language (country)<br>
1911 * language (variant)<br>
1912 * script (country)<br>
1913 * country<br>
1914 * </blockquote>
1915 * depending on which fields are specified in the locale. If the
1916 * language, script, country, and variant fields are all empty,
1917 * this function returns the empty string.
1918 *
1919 * @return The name of the locale appropriate to display.
1920 */
1921 public final String getDisplayName() {
1922 return getDisplayName(getDefault(Category.DISPLAY));
1923 }
1924
1925 /**
1926 * Returns a name for the locale that is appropriate for display
1927 * to the user. This will be the values returned by
1928 * getDisplayLanguage(), getDisplayScript(),getDisplayCountry(),
1929 * and getDisplayVariant() assembled into a single string.
1930 * The non-empty values are used in order,
1931 * with the second and subsequent names in parentheses. For example:
1932 * <blockquote>
1933 * language (script, country, variant)<br>
1934 * language (country)<br>
1935 * language (variant)<br>
1936 * script (country)<br>
1937 * country<br>
1938 * </blockquote>
1939 * depending on which fields are specified in the locale. If the
1940 * language, script, country, and variant fields are all empty,
1941 * this function returns the empty string.
1942 *
1943 * @param inLocale The locale for which to retrieve the display name.
1944 * @return The name of the locale appropriate to display.
1945 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1946 */
1947 public String getDisplayName(Locale inLocale) {
1948 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale);
1949
1950 String languageName = getDisplayLanguage(inLocale);
1951 String scriptName = getDisplayScript(inLocale);
1952 String countryName = getDisplayCountry(inLocale);
1953 String[] variantNames = getDisplayVariantArray(inLocale);
1954
1955 // Get the localized patterns for formatting a display name.
1956 String displayNamePattern = lr.getLocaleName("DisplayNamePattern");
1957 String listPattern = lr.getLocaleName("ListPattern");
1958 String listCompositionPattern = lr.getLocaleName("ListCompositionPattern");
1959
1960 // The display name consists of a main name, followed by qualifiers.
1961 // Typically, the format is "MainName (Qualifier, Qualifier)" but this
1962 // depends on what pattern is stored in the display locale.
1963 String mainName = null;
1964 String[] qualifierNames = null;
1965
1966 // The main name is the language, or if there is no language, the script,
1967 // then if no script, the country. If there is no language/script/country
1968 // (an anomalous situation) then the display name is simply the variant's
1969 // display name.
1970 if (languageName.length() == 0 && scriptName.length() == 0 && countryName.length() == 0) {
1971 if (variantNames.length == 0) {
1972 return "";
1973 } else {
1974 return formatList(variantNames, listPattern, listCompositionPattern);
1975 }
1976 }
1977 ArrayList<String> names = new ArrayList<>(4);
1978 if (languageName.length() != 0) {
1979 names.add(languageName);
1980 }
1981 if (scriptName.length() != 0) {
1982 names.add(scriptName);
1983 }
1984 if (countryName.length() != 0) {
1985 names.add(countryName);
1986 }
1987 if (variantNames.length != 0) {
1988 names.addAll(Arrays.asList(variantNames));
1989 }
1990
1991 // The first one in the main name
1992 mainName = names.get(0);
1993
1994 // Others are qualifiers
1995 int numNames = names.size();
1996 qualifierNames = (numNames > 1) ?
1997 names.subList(1, numNames).toArray(new String[numNames - 1]) : new String[0];
1998
1999 // Create an array whose first element is the number of remaining
2000 // elements. This serves as a selector into a ChoiceFormat pattern from
2001 // the resource. The second and third elements are the main name and
2002 // the qualifier; if there are no qualifiers, the third element is
2003 // unused by the format pattern.
2004 Object[] displayNames = {
2005 qualifierNames.length != 0 ? 2 : 1,
2006 mainName,
2007 // We could also just call formatList() and have it handle the empty
2008 // list case, but this is more efficient, and we want it to be
2009 // efficient since all the language-only locales will not have any
2010 // qualifiers.
2011 qualifierNames.length != 0 ? formatList(qualifierNames, listPattern, listCompositionPattern) : null
2012 };
2013
2014 if (displayNamePattern != null) {
2015 return new MessageFormat(displayNamePattern).format(displayNames);
2016 }
2017 else {
2018 // If we cannot get the message format pattern, then we use a simple
2019 // hard-coded pattern. This should not occur in practice unless the
2020 // installation is missing some core files (FormatData etc.).
2021 StringBuilder result = new StringBuilder();
2022 result.append((String)displayNames[1]);
2023 if (displayNames.length > 2) {
2024 result.append(" (");
2025 result.append((String)displayNames[2]);
2026 result.append(')');
2027 }
2028 return result.toString();
2029 }
2030 }
2031
2032 /**
2033 * Overrides Cloneable.
2034 */
2035 @Override
2036 public Object clone()
2037 {
2038 try {
2039 Locale that = (Locale)super.clone();
2040 return that;
2041 } catch (CloneNotSupportedException e) {
2042 throw new InternalError(e);
2043 }
2044 }
2045
2046 /**
2047 * Override hashCode.
2048 * Since Locales are often used in hashtables, caches the value
2049 * for speed.
2050 */
2051 @Override
2052 public int hashCode() {
2053 int hc = hashCodeValue;
2054 if (hc == 0) {
2055 hc = baseLocale.hashCode();
2056 if (localeExtensions != null) {
2057 hc ^= localeExtensions.hashCode();
2058 }
2059 hashCodeValue = hc;
2060 }
2061 return hc;
2062 }
2063
2064 // Overrides
2065
2066 /**
2067 * Returns true if this Locale is equal to another object. A Locale is
2068 * deemed equal to another Locale with identical language, script, country,
2069 * variant and extensions, and unequal to all other objects.
2070 *
2071 * @return true if this Locale is equal to the specified object.
2072 */
2073 @Override
2074 public boolean equals(Object obj) {
2075 if (this == obj) // quick check
2076 return true;
2077 if (!(obj instanceof Locale))
2078 return false;
2079 BaseLocale otherBase = ((Locale)obj).baseLocale;
2080 if (!baseLocale.equals(otherBase)) {
2081 return false;
2082 }
2083 if (localeExtensions == null) {
2084 return ((Locale)obj).localeExtensions == null;
2085 }
2086 return localeExtensions.equals(((Locale)obj).localeExtensions);
2087 }
2088
2089 // ================= privates =====================================
2090
2091 private transient BaseLocale baseLocale;
2092 private transient LocaleExtensions localeExtensions;
2093
2094 /**
2095 * Calculated hashcode
2096 */
2097 private transient volatile int hashCodeValue;
2098
2099 private static volatile Locale defaultLocale = initDefault();
2100 private static volatile Locale defaultDisplayLocale;
2101 private static volatile Locale defaultFormatLocale;
2102
2103 private transient volatile String languageTag;
2104
2105 /**
2106 * Return an array of the display names of the variant.
2107 * @param bundle the ResourceBundle to use to get the display names
2108 * @return an array of display names, possible of zero length.
2109 */
2110 private String[] getDisplayVariantArray(Locale inLocale) {
2111 // Split the variant name into tokens separated by '_'.
2112 StringTokenizer tokenizer = new StringTokenizer(baseLocale.getVariant(), "_");
2113 String[] names = new String[tokenizer.countTokens()];
2114
2115 // For each variant token, lookup the display name. If
2116 // not found, use the variant name itself.
2117 for (int i=0; i<names.length; ++i) {
2118 names[i] = getDisplayString(tokenizer.nextToken(),
2119 inLocale, DISPLAY_VARIANT);
2120 }
2121
2122 return names;
2123 }
2124
2125 /**
2126 * Format a list using given pattern strings.
2127 * If either of the patterns is null, then a the list is
2128 * formatted by concatenation with the delimiter ','.
2129 * @param stringList the list of strings to be formatted.
2130 * @param listPattern should create a MessageFormat taking 0-3 arguments
2131 * and formatting them into a list.
2132 * @param listCompositionPattern should take 2 arguments
2133 * and is used by composeList.
2134 * @return a string representing the list.
2135 */
2136 private static String formatList(String[] stringList, String listPattern, String listCompositionPattern) {
2137 // If we have no list patterns, compose the list in a simple,
2138 // non-localized way.
2139 if (listPattern == null || listCompositionPattern == null) {
2140 StringJoiner sj = new StringJoiner(",");
2141 for (int i = 0; i < stringList.length; ++i) {
2142 sj.add(stringList[i]);
2143 }
2144 return sj.toString();
2145 }
2146
2147 // Compose the list down to three elements if necessary
2148 if (stringList.length > 3) {
2149 MessageFormat format = new MessageFormat(listCompositionPattern);
2150 stringList = composeList(format, stringList);
2151 }
2152
2153 // Rebuild the argument list with the list length as the first element
2154 Object[] args = new Object[stringList.length + 1];
2155 System.arraycopy(stringList, 0, args, 1, stringList.length);
2156 args[0] = stringList.length;
2157
2158 // Format it using the pattern in the resource
2159 MessageFormat format = new MessageFormat(listPattern);
2160 return format.format(args);
2161 }
2162
2163 /**
2164 * Given a list of strings, return a list shortened to three elements.
2165 * Shorten it by applying the given format to the first two elements
2166 * recursively.
2167 * @param format a format which takes two arguments
2168 * @param list a list of strings
2169 * @return if the list is three elements or shorter, the same list;
2170 * otherwise, a new list of three elements.
2171 */
2172 private static String[] composeList(MessageFormat format, String[] list) {
2173 if (list.length <= 3) return list;
2174
2175 // Use the given format to compose the first two elements into one
2176 String[] listItems = { list[0], list[1] };
2177 String newItem = format.format(listItems);
2178
2179 // Form a new list one element shorter
2180 String[] newList = new String[list.length-1];
2181 System.arraycopy(list, 2, newList, 1, newList.length-1);
2182 newList[0] = newItem;
2183
2184 // Recurse
2185 return composeList(format, newList);
2186 }
2187
2188 // Duplicate of sun.util.locale.UnicodeLocaleExtension.isKey in order to
2189 // avoid its class loading.
2190 private static boolean isUnicodeExtensionKey(String s) {
2191 // 2alphanum
2192 return (s.length() == 2) && LocaleUtils.isAlphaNumericString(s);
2193 }
2194
2195 /**
2196 * @serialField language String
2197 * language subtag in lower case. (See <a href="java/util/Locale.html#getLanguage()">getLanguage()</a>)
2198 * @serialField country String
2199 * country subtag in upper case. (See <a href="java/util/Locale.html#getCountry()">getCountry()</a>)
2200 * @serialField variant String
2201 * variant subtags separated by LOWLINE characters. (See <a href="java/util/Locale.html#getVariant()">getVariant()</a>)
2202 * @serialField hashcode int
2203 * deprecated, for forward compatibility only
2204 * @serialField script String
2205 * script subtag in title case (See <a href="java/util/Locale.html#getScript()">getScript()</a>)
2206 * @serialField extensions String
2207 * canonical representation of extensions, that is,
2208 * BCP47 extensions in alphabetical order followed by
2209 * BCP47 private use subtags, all in lower case letters
2210 * separated by HYPHEN-MINUS characters.
2211 * (See <a href="java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>,
2212 * <a href="java/util/Locale.html#getExtension(char)">getExtension(char)</a>)
2213 */
2214 private static final ObjectStreamField[] serialPersistentFields = {
2215 new ObjectStreamField("language", String.class),
2216 new ObjectStreamField("country", String.class),
2217 new ObjectStreamField("variant", String.class),
2218 new ObjectStreamField("hashcode", int.class),
2219 new ObjectStreamField("script", String.class),
2220 new ObjectStreamField("extensions", String.class),
2221 };
2222
2223 /**
2224 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>.
2225 * @param out the <code>ObjectOutputStream</code> to write
2226 * @throws IOException
2227 * @since 1.7
2228 */
2229 private void writeObject(ObjectOutputStream out) throws IOException {
2230 ObjectOutputStream.PutField fields = out.putFields();
2231 fields.put("language", baseLocale.getLanguage());
2232 fields.put("script", baseLocale.getScript());
2233 fields.put("country", baseLocale.getRegion());
2234 fields.put("variant", baseLocale.getVariant());
2235 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID());
2236 fields.put("hashcode", -1); // place holder just for backward support
2237 out.writeFields();
2238 }
2239
2240 /**
2241 * Deserializes this <code>Locale</code>.
2242 * @param in the <code>ObjectInputStream</code> to read
2243 * @throws IOException
2244 * @throws ClassNotFoundException
2245 * @throws IllformedLocaleException
2246 * @since 1.7
2247 */
2248 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException {
2249 ObjectInputStream.GetField fields = in.readFields();
2250 String language = (String)fields.get("language", "");
2251 String script = (String)fields.get("script", "");
2252 String country = (String)fields.get("country", "");
2253 String variant = (String)fields.get("variant", "");
2254 String extStr = (String)fields.get("extensions", "");
2255 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant);
2256 if (extStr.length() > 0) {
2257 try {
2258 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
2259 bldr.setExtensions(extStr);
2260 localeExtensions = bldr.getLocaleExtensions();
2261 } catch (LocaleSyntaxException e) {
2262 throw new IllformedLocaleException(e.getMessage());
2263 }
2264 } else {
2265 localeExtensions = null;
2266 }
2267 }
2268
2269 /**
2270 * Returns a cached <code>Locale</code> instance equivalent to
2271 * the deserialized <code>Locale</code>. When serialized
2272 * language, country and variant fields read from the object data stream
2273 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions
2274 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code>
2275 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script
2276 * type is "thai"). See <a href="Locale.html#special_cases_constructor">Special Cases</a>
2277 * for more information.
2278 *
2279 * @return an instance of <code>Locale</code> equivalent to
2280 * the deserialized <code>Locale</code>.
2281 * @throws java.io.ObjectStreamException
2282 */
2283 private Object readResolve() throws java.io.ObjectStreamException {
2284 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(),
2285 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions);
2286 }
2287
2288 private static volatile String[] isoLanguages;
2289
2290 private static volatile String[] isoCountries;
2291
2292 private static String convertOldISOCodes(String language) {
2293 // we accept both the old and the new ISO codes for the languages whose ISO
2294 // codes have changed, but we always store the OLD code, for backward compatibility
2295 language = LocaleUtils.toLowerString(language).intern();
2296 if (language == "he") {
2297 return "iw";
2298 } else if (language == "yi") {
2299 return "ji";
2300 } else if (language == "id") {
2301 return "in";
2302 } else {
2303 return language;
2304 }
2305 }
2306
2307 private static LocaleExtensions getCompatibilityExtensions(String language,
2308 String script,
2309 String country,
2310 String variant) {
2311 LocaleExtensions extensions = null;
2312 // Special cases for backward compatibility support
2313 if (LocaleUtils.caseIgnoreMatch(language, "ja")
2314 && script.length() == 0
2315 && LocaleUtils.caseIgnoreMatch(country, "jp")
2316 && "JP".equals(variant)) {
2317 // ja_JP_JP -> u-ca-japanese (calendar = japanese)
2318 extensions = LocaleExtensions.CALENDAR_JAPANESE;
2319 } else if (LocaleUtils.caseIgnoreMatch(language, "th")
2320 && script.length() == 0
2321 && LocaleUtils.caseIgnoreMatch(country, "th")
2322 && "TH".equals(variant)) {
2323 // th_TH_TH -> u-nu-thai (numbersystem = thai)
2324 extensions = LocaleExtensions.NUMBER_THAI;
2325 }
2326 return extensions;
2327 }
2328
2329 /**
2330 * Obtains a localized locale names from a LocaleNameProvider
2331 * implementation.
2332 */
2333 private static class LocaleNameGetter
2334 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> {
2335 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter();
2336
2337 @Override
2338 public String getObject(LocaleNameProvider localeNameProvider,
2339 Locale locale,
2340 String key,
2341 Object... params) {
2342 assert params.length == 2;
2343 int type = (Integer)params[0];
2344 String code = (String)params[1];
2345
2346 switch(type) {
2347 case DISPLAY_LANGUAGE:
2348 return localeNameProvider.getDisplayLanguage(code, locale);
2349 case DISPLAY_COUNTRY:
2350 return localeNameProvider.getDisplayCountry(code, locale);
2351 case DISPLAY_VARIANT:
2352 return localeNameProvider.getDisplayVariant(code, locale);
2353 case DISPLAY_SCRIPT:
2354 return localeNameProvider.getDisplayScript(code, locale);
2355 default:
2356 assert false; // shouldn't happen
2357 }
2358
2359 return null;
2360 }
2361 }
2362
2363 /**
2364 * Enum for locale categories. These locale categories are used to get/set
2365 * the default locale for the specific functionality represented by the
2366 * category.
2367 *
2368 * @see #getDefault(Locale.Category)
2369 * @see #setDefault(Locale.Category, Locale)
2370 * @since 1.7
2371 */
2372 public enum Category {
2373
2374 /**
2375 * Category used to represent the default locale for
2376 * displaying user interfaces.
2377 */
2378 DISPLAY("user.language.display",
2379 "user.script.display",
2380 "user.country.display",
2381 "user.variant.display"),
2382
2383 /**
2384 * Category used to represent the default locale for
2385 * formatting dates, numbers, and/or currencies.
2386 */
2387 FORMAT("user.language.format",
2388 "user.script.format",
2389 "user.country.format",
2390 "user.variant.format");
2391
2392 Category(String languageKey, String scriptKey, String countryKey, String variantKey) {
2393 this.languageKey = languageKey;
2394 this.scriptKey = scriptKey;
2395 this.countryKey = countryKey;
2396 this.variantKey = variantKey;
2397 }
2398
2399 final String languageKey;
2400 final String scriptKey;
2401 final String countryKey;
2402 final String variantKey;
2403 }
2404
2405 /**
2406 * <code>Builder</code> is used to build instances of <code>Locale</code>
2407 * from values configured by the setters. Unlike the <code>Locale</code>
2408 * constructors, the <code>Builder</code> checks if a value configured by a
2409 * setter satisfies the syntax requirements defined by the <code>Locale</code>
2410 * class. A <code>Locale</code> object created by a <code>Builder</code> is
2411 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag
2412 * without losing information.
2413 *
2414 * <p><b>Note:</b> The <code>Locale</code> class does not provide any
2415 * syntactic restrictions on variant, while BCP 47 requires each variant
2416 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3
2417 * alphanumerics. The method <code>setVariant</code> throws
2418 * <code>IllformedLocaleException</code> for a variant that does not satisfy
2419 * this restriction. If it is necessary to support such a variant, use a
2420 * Locale constructor. However, keep in mind that a <code>Locale</code>
2421 * object created this way might lose the variant information when
2422 * transformed to a BCP 47 language tag.
2423 *
2424 * <p>The following example shows how to create a <code>Locale</code> object
2425 * with the <code>Builder</code>.
2426 * <blockquote>
2427 * <pre>
2428 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build();
2429 * </pre>
2430 * </blockquote>
2431 *
2432 * <p>Builders can be reused; <code>clear()</code> resets all
2433 * fields to their default values.
2434 *
2435 * @see Locale#forLanguageTag
2436 * @since 1.7
2437 */
2438 public static final class Builder {
2439 private final InternalLocaleBuilder localeBuilder;
2440
2441 /**
2442 * Constructs an empty Builder. The default value of all
2443 * fields, extensions, and private use information is the
2444 * empty string.
2445 */
2446 public Builder() {
2447 localeBuilder = new InternalLocaleBuilder();
2448 }
2449
2450 /**
2451 * Resets the <code>Builder</code> to match the provided
2452 * <code>locale</code>. Existing state is discarded.
2453 *
2454 * <p>All fields of the locale must be well-formed, see {@link Locale}.
2455 *
2456 * <p>Locales with any ill-formed fields cause
2457 * <code>IllformedLocaleException</code> to be thrown, except for the
2458 * following three cases which are accepted for compatibility
2459 * reasons:<ul>
2460 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese"
2461 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai"
2462 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul>
2463 *
2464 * @param locale the locale
2465 * @return This builder.
2466 * @throws IllformedLocaleException if <code>locale</code> has
2467 * any ill-formed fields.
2468 * @throws NullPointerException if <code>locale</code> is null.
2469 */
2470 public Builder setLocale(Locale locale) {
2471 try {
2472 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions);
2473 } catch (LocaleSyntaxException e) {
2474 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2475 }
2476 return this;
2477 }
2478
2479 /**
2480 * Resets the Builder to match the provided IETF BCP 47
2481 * language tag. Discards the existing state. Null and the
2482 * empty string cause the builder to be reset, like {@link
2483 * #clear}. Grandfathered tags (see {@link
2484 * Locale#forLanguageTag}) are converted to their canonical
2485 * form before being processed. Otherwise, the language tag
2486 * must be well-formed (see {@link Locale}) or an exception is
2487 * thrown (unlike <code>Locale.forLanguageTag</code>, which
2488 * just discards ill-formed and following portions of the
2489 * tag).
2490 *
2491 * @param languageTag the language tag
2492 * @return This builder.
2493 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed
2494 * @see Locale#forLanguageTag(String)
2495 */
2496 public Builder setLanguageTag(String languageTag) {
2497 ParseStatus sts = new ParseStatus();
2498 LanguageTag tag = LanguageTag.parse(languageTag, sts);
2499 if (sts.isError()) {
2500 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex());
2501 }
2502 localeBuilder.setLanguageTag(tag);
2503 return this;
2504 }
2505
2506 /**
2507 * Sets the language. If <code>language</code> is the empty string or
2508 * null, the language in this <code>Builder</code> is removed. Otherwise,
2509 * the language must be <a href="./Locale.html#def_language">well-formed</a>
2510 * or an exception is thrown.
2511 *
2512 * <p>The typical language value is a two or three-letter language
2513 * code as defined in ISO639.
2514 *
2515 * @param language the language
2516 * @return This builder.
2517 * @throws IllformedLocaleException if <code>language</code> is ill-formed
2518 */
2519 public Builder setLanguage(String language) {
2520 try {
2521 localeBuilder.setLanguage(language);
2522 } catch (LocaleSyntaxException e) {
2523 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2524 }
2525 return this;
2526 }
2527
2528 /**
2529 * Sets the script. If <code>script</code> is null or the empty string,
2530 * the script in this <code>Builder</code> is removed.
2531 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an
2532 * exception is thrown.
2533 *
2534 * <p>The typical script value is a four-letter script code as defined by ISO 15924.
2535 *
2536 * @param script the script
2537 * @return This builder.
2538 * @throws IllformedLocaleException if <code>script</code> is ill-formed
2539 */
2540 public Builder setScript(String script) {
2541 try {
2542 localeBuilder.setScript(script);
2543 } catch (LocaleSyntaxException e) {
2544 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2545 }
2546 return this;
2547 }
2548
2549 /**
2550 * Sets the region. If region is null or the empty string, the region
2551 * in this <code>Builder</code> is removed. Otherwise,
2552 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an
2553 * exception is thrown.
2554 *
2555 * <p>The typical region value is a two-letter ISO 3166 code or a
2556 * three-digit UN M.49 area code.
2557 *
2558 * <p>The country value in the <code>Locale</code> created by the
2559 * <code>Builder</code> is always normalized to upper case.
2560 *
2561 * @param region the region
2562 * @return This builder.
2563 * @throws IllformedLocaleException if <code>region</code> is ill-formed
2564 */
2565 public Builder setRegion(String region) {
2566 try {
2567 localeBuilder.setRegion(region);
2568 } catch (LocaleSyntaxException e) {
2569 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2570 }
2571 return this;
2572 }
2573
2574 /**
2575 * Sets the variant. If variant is null or the empty string, the
2576 * variant in this <code>Builder</code> is removed. Otherwise, it
2577 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a>
2578 * subtags, or an exception is thrown.
2579 *
2580 * <p><b>Note:</b> This method checks if <code>variant</code>
2581 * satisfies the IETF BCP 47 variant subtag's syntax requirements,
2582 * and normalizes the value to lowercase letters. However,
2583 * the <code>Locale</code> class does not impose any syntactic
2584 * restriction on variant, and the variant value in
2585 * <code>Locale</code> is case sensitive. To set such a variant,
2586 * use a Locale constructor.
2587 *
2588 * @param variant the variant
2589 * @return This builder.
2590 * @throws IllformedLocaleException if <code>variant</code> is ill-formed
2591 */
2592 public Builder setVariant(String variant) {
2593 try {
2594 localeBuilder.setVariant(variant);
2595 } catch (LocaleSyntaxException e) {
2596 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2597 }
2598 return this;
2599 }
2600
2601 /**
2602 * Sets the extension for the given key. If the value is null or the
2603 * empty string, the extension is removed. Otherwise, the extension
2604 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception
2605 * is thrown.
2606 *
2607 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION
2608 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension.
2609 * Setting a value for this key replaces any existing Unicode locale key/type
2610 * pairs with those defined in the extension.
2611 *
2612 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION
2613 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be
2614 * well-formed, the value for this key needs only to have subtags of one to
2615 * eight alphanumeric characters, not two to eight as in the general case.
2616 *
2617 * @param key the extension key
2618 * @param value the extension value
2619 * @return This builder.
2620 * @throws IllformedLocaleException if <code>key</code> is illegal
2621 * or <code>value</code> is ill-formed
2622 * @see #setUnicodeLocaleKeyword(String, String)
2623 */
2624 public Builder setExtension(char key, String value) {
2625 try {
2626 localeBuilder.setExtension(key, value);
2627 } catch (LocaleSyntaxException e) {
2628 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2629 }
2630 return this;
2631 }
2632
2633 /**
2634 * Sets the Unicode locale keyword type for the given key. If the type
2635 * is null, the Unicode keyword is removed. Otherwise, the key must be
2636 * non-null and both key and type must be <a
2637 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2638 * is thrown.
2639 *
2640 * <p>Keys and types are converted to lower case.
2641 *
2642 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension}
2643 * replaces all Unicode locale keywords with those defined in the
2644 * extension.
2645 *
2646 * @param key the Unicode locale key
2647 * @param type the Unicode locale type
2648 * @return This builder.
2649 * @throws IllformedLocaleException if <code>key</code> or <code>type</code>
2650 * is ill-formed
2651 * @throws NullPointerException if <code>key</code> is null
2652 * @see #setExtension(char, String)
2653 */
2654 public Builder setUnicodeLocaleKeyword(String key, String type) {
2655 try {
2656 localeBuilder.setUnicodeLocaleKeyword(key, type);
2657 } catch (LocaleSyntaxException e) {
2658 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2659 }
2660 return this;
2661 }
2662
2663 /**
2664 * Adds a unicode locale attribute, if not already present, otherwise
2665 * has no effect. The attribute must not be null and must be <a
2666 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2667 * is thrown.
2668 *
2669 * @param attribute the attribute
2670 * @return This builder.
2671 * @throws NullPointerException if <code>attribute</code> is null
2672 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2673 * @see #setExtension(char, String)
2674 */
2675 public Builder addUnicodeLocaleAttribute(String attribute) {
2676 try {
2677 localeBuilder.addUnicodeLocaleAttribute(attribute);
2678 } catch (LocaleSyntaxException e) {
2679 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2680 }
2681 return this;
2682 }
2683
2684 /**
2685 * Removes a unicode locale attribute, if present, otherwise has no
2686 * effect. The attribute must not be null and must be <a
2687 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2688 * is thrown.
2689 *
2690 * <p>Attribute comparison for removal is case-insensitive.
2691 *
2692 * @param attribute the attribute
2693 * @return This builder.
2694 * @throws NullPointerException if <code>attribute</code> is null
2695 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2696 * @see #setExtension(char, String)
2697 */
2698 public Builder removeUnicodeLocaleAttribute(String attribute) {
2699 try {
2700 localeBuilder.removeUnicodeLocaleAttribute(attribute);
2701 } catch (LocaleSyntaxException e) {
2702 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2703 }
2704 return this;
2705 }
2706
2707 /**
2708 * Resets the builder to its initial, empty state.
2709 *
2710 * @return This builder.
2711 */
2712 public Builder clear() {
2713 localeBuilder.clear();
2714 return this;
2715 }
2716
2717 /**
2718 * Resets the extensions to their initial, empty state.
2719 * Language, script, region and variant are unchanged.
2720 *
2721 * @return This builder.
2722 * @see #setExtension(char, String)
2723 */
2724 public Builder clearExtensions() {
2725 localeBuilder.clearExtensions();
2726 return this;
2727 }
2728
2729 /**
2730 * Returns an instance of <code>Locale</code> created from the fields set
2731 * on this builder.
2732 *
2733 * <p>This applies the conversions listed in {@link Locale#forLanguageTag}
2734 * when constructing a Locale. (Grandfathered tags are handled in
2735 * {@link #setLanguageTag}.)
2736 *
2737 * @return A Locale.
2738 */
2739 public Locale build() {
2740 BaseLocale baseloc = localeBuilder.getBaseLocale();
2741 LocaleExtensions extensions = localeBuilder.getLocaleExtensions();
2742 if (extensions == null && baseloc.getVariant().length() > 0) {
2743 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(),
2744 baseloc.getRegion(), baseloc.getVariant());
2745 }
2746 return Locale.getInstance(baseloc, extensions);
2747 }
2748 }
2749
2750 /**
2751 * This enum provides constants to select a filtering mode for locale
2752 * matching. Refer to <a href="http://tools.ietf.org/html/rfc4647">RFC 4647
2753 * Matching of Language Tags</a> for details.
2754 *
2755 * <p>As an example, think of two Language Priority Lists each of which
2756 * includes only one language range and a set of following language tags:
2757 *
2758 * <pre>
2759 * de (German)
2760 * de-DE (German, Germany)
2761 * de-Deva (German, in Devanagari script)
2762 * de-Deva-DE (German, in Devanagari script, Germany)
2763 * de-DE-1996 (German, Germany, orthography of 1996)
2764 * de-Latn-DE (German, in Latin script, Germany)
2765 * de-Latn-DE-1996 (German, in Latin script, Germany, orthography of 1996)
2766 * </pre>
2767 *
2768 * The filtering method will behave as follows:
2769 *
2770 * <table cellpadding=2 summary="Filtering method behavior">
2771 * <tr>
2772 * <th>Filtering Mode</th>
2773 * <th>Language Priority List: {@code "de-DE"}</th>
2774 * <th>Language Priority List: {@code "de-*-DE"}</th>
2775 * </tr>
2776 * <tr>
2777 * <td valign=top>
2778 * {@link FilteringMode#AUTOSELECT_FILTERING AUTOSELECT_FILTERING}
2779 * </td>
2780 * <td valign=top>
2781 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2782 * {@code "de-DE-1996"}.
2783 * </td>
2784 * <td valign=top>
2785 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2786 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2787 * {@code "de-Latn-DE-1996"}.
2788 * </td>
2789 * </tr>
2790 * <tr>
2791 * <td valign=top>
2792 * {@link FilteringMode#EXTENDED_FILTERING EXTENDED_FILTERING}
2793 * </td>
2794 * <td valign=top>
2795 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2796 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2797 * {@code "de-Latn-DE-1996"}.
2798 * </td>
2799 * <td valign=top>Same as above.</td>
2800 * </tr>
2801 * <tr>
2802 * <td valign=top>
2803 * {@link FilteringMode#IGNORE_EXTENDED_RANGES IGNORE_EXTENDED_RANGES}
2804 * </td>
2805 * <td valign=top>
2806 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2807 * {@code "de-DE-1996"}.
2808 * </td>
2809 * <td valign=top>
2810 * Performs <em>basic</em> filtering and returns {@code null} because
2811 * nothing matches.
2812 * </td>
2813 * </tr>
2814 * <tr>
2815 * <td valign=top>
2816 * {@link FilteringMode#MAP_EXTENDED_RANGES MAP_EXTENDED_RANGES}
2817 * </td>
2818 * <td valign=top>Same as above.</td>
2819 * <td valign=top>
2820 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2821 * {@code "de-DE-1996"} because {@code "de-*-DE"} is mapped to
2822 * {@code "de-DE"}.
2823 * </td>
2824 * </tr>
2825 * <tr>
2826 * <td valign=top>
2827 * {@link FilteringMode#REJECT_EXTENDED_RANGES REJECT_EXTENDED_RANGES}
2828 * </td>
2829 * <td valign=top>Same as above.</td>
2830 * <td valign=top>
2831 * Throws {@link IllegalArgumentException} because {@code "de-*-DE"} is
2832 * not a valid basic language range.
2833 * </td>
2834 * </tr>
2835 * </table>
2836 *
2837 * @see #filter(List, Collection, FilteringMode)
2838 * @see #filterTags(List, Collection, FilteringMode)
2839 *
2840 * @since 1.8
2841 */
2842 public static enum FilteringMode {
2843 /**
2844 * Specifies automatic filtering mode based on the given Language
2845 * Priority List consisting of language ranges. If all of the ranges
2846 * are basic, basic filtering is selected. Otherwise, extended
2847 * filtering is selected.
2848 */
2849 AUTOSELECT_FILTERING,
2850
2851 /**
2852 * Specifies extended filtering.
2853 */
2854 EXTENDED_FILTERING,
2855
2856 /**
2857 * Specifies basic filtering: Note that any extended language ranges
2858 * included in the given Language Priority List are ignored.
2859 */
2860 IGNORE_EXTENDED_RANGES,
2861
2862 /**
2863 * Specifies basic filtering: If any extended language ranges are
2864 * included in the given Language Priority List, they are mapped to the
2865 * basic language range. Specifically, a language range starting with a
2866 * subtag {@code "*"} is treated as a language range {@code "*"}. For
2867 * example, {@code "*-US"} is treated as {@code "*"}. If {@code "*"} is
2868 * not the first subtag, {@code "*"} and extra {@code "-"} are removed.
2869 * For example, {@code "ja-*-JP"} is mapped to {@code "ja-JP"}.
2870 */
2871 MAP_EXTENDED_RANGES,
2872
2873 /**
2874 * Specifies basic filtering: If any extended language ranges are
2875 * included in the given Language Priority List, the list is rejected
2876 * and the filtering method throws {@link IllegalArgumentException}.
2877 */
2878 REJECT_EXTENDED_RANGES
2879 };
2880
2881 /**
2882 * This class expresses a <em>Language Range</em> defined in
2883 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
2884 * Language Tags</a>. A language range is an identifier which is used to
2885 * select language tag(s) meeting specific requirements by using the
2886 * mechanisms described in <a href="Locale.html#LocaleMatching">Locale
2887 * Matching</a>. A list which represents a user's preferences and consists
2888 * of language ranges is called a <em>Language Priority List</em>.
2889 *
2890 * <p>There are two types of language ranges: basic and extended. In RFC
2891 * 4647, the syntax of language ranges is expressed in
2892 * <a href="http://tools.ietf.org/html/rfc4234">ABNF</a> as follows:
2893 * <blockquote>
2894 * <pre>
2895 * basic-language-range = (1*8ALPHA *("-" 1*8alphanum)) / "*"
2896 * extended-language-range = (1*8ALPHA / "*")
2897 * *("-" (1*8alphanum / "*"))
2898 * alphanum = ALPHA / DIGIT
2899 * </pre>
2900 * </blockquote>
2901 * For example, {@code "en"} (English), {@code "ja-JP"} (Japanese, Japan),
2902 * {@code "*"} (special language range which matches any language tag) are
2903 * basic language ranges, whereas {@code "*-CH"} (any languages,
2904 * Switzerland), {@code "es-*"} (Spanish, any regions), and
2905 * {@code "zh-Hant-*"} (Traditional Chinese, any regions) are extended
2906 * language ranges.
2907 *
2908 * @see #filter
2909 * @see #filterTags
2910 * @see #lookup
2911 * @see #lookupTag
2912 *
2913 * @since 1.8
2914 */
2915 public static final class LanguageRange {
2916
2917 /**
2918 * A constant holding the maximum value of weight, 1.0, which indicates
2919 * that the language range is a good fit for the user.
2920 */
2921 public static final double MAX_WEIGHT = 1.0;
2922
2923 /**
2924 * A constant holding the minimum value of weight, 0.0, which indicates
2925 * that the language range is not a good fit for the user.
2926 */
2927 public static final double MIN_WEIGHT = 0.0;
2928
2929 private final String range;
2930 private final double weight;
2931
2932 private volatile int hash;
2933
2934 /**
2935 * Constructs a {@code LanguageRange} using the given {@code range}.
2936 * Note that no validation is done against the IANA Language Subtag
2937 * Registry at time of construction.
2938 *
2939 * <p>This is equivalent to {@code LanguageRange(range, MAX_WEIGHT)}.
2940 *
2941 * @param range a language range
2942 * @throws NullPointerException if the given {@code range} is
2943 * {@code null}
2944 * @throws IllegalArgumentException if the given {@code range} does not
2945 * comply with the syntax of the language range mentioned in RFC 4647
2946 */
2947 public LanguageRange(String range) {
2948 this(range, MAX_WEIGHT);
2949 }
2950
2951 /**
2952 * Constructs a {@code LanguageRange} using the given {@code range} and
2953 * {@code weight}. Note that no validation is done against the IANA
2954 * Language Subtag Registry at time of construction.
2955 *
2956 * @param range a language range
2957 * @param weight a weight value between {@code MIN_WEIGHT} and
2958 * {@code MAX_WEIGHT}
2959 * @throws NullPointerException if the given {@code range} is
2960 * {@code null}
2961 * @throws IllegalArgumentException if the given {@code range} does not
2962 * comply with the syntax of the language range mentioned in RFC 4647
2963 * or if the given {@code weight} is less than {@code MIN_WEIGHT}
2964 * or greater than {@code MAX_WEIGHT}
2965 */
2966 public LanguageRange(String range, double weight) {
2967 if (range == null) {
2968 throw new NullPointerException();
2969 }
2970 if (weight < MIN_WEIGHT || weight > MAX_WEIGHT) {
2971 throw new IllegalArgumentException("weight=" + weight);
2972 }
2973
2974 range = range.toLowerCase(Locale.ROOT);
2975
2976 // Do syntax check.
2977 boolean isIllFormed = false;
2978 String[] subtags = range.split("-");
2979 if (isSubtagIllFormed(subtags[0], true)
2980 || range.endsWith("-")) {
2981 isIllFormed = true;
2982 } else {
2983 for (int i = 1; i < subtags.length; i++) {
2984 if (isSubtagIllFormed(subtags[i], false)) {
2985 isIllFormed = true;
2986 break;
2987 }
2988 }
2989 }
2990 if (isIllFormed) {
2991 throw new IllegalArgumentException("range=" + range);
2992 }
2993
2994 this.range = range;
2995 this.weight = weight;
2996 }
2997
2998 private static boolean isSubtagIllFormed(String subtag,
2999 boolean isFirstSubtag) {
3000 if (subtag.equals("") || subtag.length() > 8) {
3001 return true;
3002 } else if (subtag.equals("*")) {
3003 return false;
3004 }
3005 char[] charArray = subtag.toCharArray();
3006 if (isFirstSubtag) { // ALPHA
3007 for (char c : charArray) {
3008 if (c < 'a' || c > 'z') {
3009 return true;
3010 }
3011 }
3012 } else { // ALPHA / DIGIT
3013 for (char c : charArray) {
3014 if (c < '0' || (c > '9' && c < 'a') || c > 'z') {
3015 return true;
3016 }
3017 }
3018 }
3019 return false;
3020 }
3021
3022 /**
3023 * Returns the language range of this {@code LanguageRange}.
3024 *
3025 * @return the language range.
3026 */
3027 public String getRange() {
3028 return range;
3029 }
3030
3031 /**
3032 * Returns the weight of this {@code LanguageRange}.
3033 *
3034 * @return the weight value.
3035 */
3036 public double getWeight() {
3037 return weight;
3038 }
3039
3040 /**
3041 * Parses the given {@code ranges} to generate a Language Priority List.
3042 *
3043 * <p>This method performs a syntactic check for each language range in
3044 * the given {@code ranges} but doesn't do validation using the IANA
3045 * Language Subtag Registry.
3046 *
3047 * <p>The {@code ranges} to be given can take one of the following
3048 * forms:
3049 *
3050 * <pre>
3051 * "Accept-Language: ja,en;q=0.4" (weighted list with Accept-Language prefix)
3052 * "ja,en;q=0.4" (weighted list)
3053 * "ja,en" (prioritized list)
3054 * </pre>
3055 *
3056 * In a weighted list, each language range is given a weight value.
3057 * The weight value is identical to the "quality value" in
3058 * <a href="http://tools.ietf.org/html/rfc2616">RFC 2616</a>, and it
3059 * expresses how much the user prefers the language. A weight value is
3060 * specified after a corresponding language range followed by
3061 * {@code ";q="}, and the default weight value is {@code MAX_WEIGHT}
3062 * when it is omitted.
3063 *
3064 * <p>Unlike a weighted list, language ranges in a prioritized list
3065 * are sorted in the descending order based on its priority. The first
3066 * language range has the highest priority and meets the user's
3067 * preference most.
3068 *
3069 * <p>In either case, language ranges are sorted in descending order in
3070 * the Language Priority List based on priority or weight. If a
3071 * language range appears in the given {@code ranges} more than once,
3072 * only the first one is included on the Language Priority List.
3073 *
3074 * <p>The returned list consists of language ranges from the given
3075 * {@code ranges} and their equivalents found in the IANA Language
3076 * Subtag Registry. For example, if the given {@code ranges} is
3077 * {@code "Accept-Language: iw,en-us;q=0.7,en;q=0.3"}, the elements in
3078 * the list to be returned are:
3079 *
3080 * <pre>
3081 * <b>Range</b> <b>Weight</b>
3082 * "iw" (older tag for Hebrew) 1.0
3083 * "he" (new preferred code for Hebrew) 1.0
3084 * "en-us" (English, United States) 0.7
3085 * "en" (English) 0.3
3086 * </pre>
3087 *
3088 * Two language ranges, {@code "iw"} and {@code "he"}, have the same
3089 * highest priority in the list. By adding {@code "he"} to the user's
3090 * Language Priority List, locale-matching method can find Hebrew as a
3091 * matching locale (or language tag) even if the application or system
3092 * offers only {@code "he"} as a supported locale (or language tag).
3093 *
3094 * @param ranges a list of comma-separated language ranges or a list of
3095 * language ranges in the form of the "Accept-Language" header
3096 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3097 * 2616</a>
3098 * @return a Language Priority List consisting of language ranges
3099 * included in the given {@code ranges} and their equivalent
3100 * language ranges if available. The list is modifiable.
3101 * @throws NullPointerException if {@code ranges} is null
3102 * @throws IllegalArgumentException if a language range or a weight
3103 * found in the given {@code ranges} is ill-formed
3104 */
3105 public static List<LanguageRange> parse(String ranges) {
3106 return LocaleMatcher.parse(ranges);
3107 }
3108
3109 /**
3110 * Parses the given {@code ranges} to generate a Language Priority
3111 * List, and then customizes the list using the given {@code map}.
3112 * This method is equivalent to
3113 * {@code mapEquivalents(parse(ranges), map)}.
3114 *
3115 * @param ranges a list of comma-separated language ranges or a list
3116 * of language ranges in the form of the "Accept-Language" header
3117 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3118 * 2616</a>
3119 * @param map a map containing information to customize language ranges
3120 * @return a Language Priority List with customization. The list is
3121 * modifiable.
3122 * @throws NullPointerException if {@code ranges} is null
3123 * @throws IllegalArgumentException if a language range or a weight
3124 * found in the given {@code ranges} is ill-formed
3125 * @see #parse(String)
3126 * @see #mapEquivalents
3127 */
3128 public static List<LanguageRange> parse(String ranges,
3129 Map<String, List<String>> map) {
3130 return mapEquivalents(parse(ranges), map);
3131 }
3132
3133 /**
3134 * Generates a new customized Language Priority List using the given
3135 * {@code priorityList} and {@code map}. If the given {@code map} is
3136 * empty, this method returns a copy of the given {@code priorityList}.
3137 *
3138 * <p>In the map, a key represents a language range whereas a value is
3139 * a list of equivalents of it. {@code '*'} cannot be used in the map.
3140 * Each equivalent language range has the same weight value as its
3141 * original language range.
3142 *
3143 * <pre>
3144 * An example of map:
3145 * <b>Key</b> <b>Value</b>
3146 * "zh" (Chinese) "zh",
3147 * "zh-Hans"(Simplified Chinese)
3148 * "zh-HK" (Chinese, Hong Kong) "zh-HK"
3149 * "zh-TW" (Chinese, Taiwan) "zh-TW"
3150 * </pre>
3151 *
3152 * The customization is performed after modification using the IANA
3153 * Language Subtag Registry.
3154 *
3155 * <p>For example, if a user's Language Priority List consists of five
3156 * language ranges ({@code "zh"}, {@code "zh-CN"}, {@code "en"},
3157 * {@code "zh-TW"}, and {@code "zh-HK"}), the newly generated Language
3158 * Priority List which is customized using the above map example will
3159 * consists of {@code "zh"}, {@code "zh-Hans"}, {@code "zh-CN"},
3160 * {@code "zh-Hans-CN"}, {@code "en"}, {@code "zh-TW"}, and
3161 * {@code "zh-HK"}.
3162 *
3163 * <p>{@code "zh-HK"} and {@code "zh-TW"} aren't converted to
3164 * {@code "zh-Hans-HK"} nor {@code "zh-Hans-TW"} even if they are
3165 * included in the Language Priority List. In this example, mapping
3166 * is used to clearly distinguish Simplified Chinese and Traditional
3167 * Chinese.
3168 *
3169 * <p>If the {@code "zh"}-to-{@code "zh"} mapping isn't included in the
3170 * map, a simple replacement will be performed and the customized list
3171 * won't include {@code "zh"} and {@code "zh-CN"}.
3172 *
3173 * @param priorityList user's Language Priority List
3174 * @param map a map containing information to customize language ranges
3175 * @return a new Language Priority List with customization. The list is
3176 * modifiable.
3177 * @throws NullPointerException if {@code priorityList} is {@code null}
3178 * @see #parse(String, Map)
3179 */
3180 public static List<LanguageRange> mapEquivalents(
3181 List<LanguageRange>priorityList,
3182 Map<String, List<String>> map) {
3183 return LocaleMatcher.mapEquivalents(priorityList, map);
3184 }
3185
3186 /**
3187 * Returns a hash code value for the object.
3188 *
3189 * @return a hash code value for this object.
3190 */
3191 @Override
3192 public int hashCode() {
3193 int h = hash;
3194 if (h == 0) {
3195 h = 17;
3196 h = 37*h + range.hashCode();
3197 long bitsWeight = Double.doubleToLongBits(weight);
3198 h = 37*h + (int)(bitsWeight ^ (bitsWeight >>> 32));
3199 if (h != 0) {
3200 hash = h;
3201 }
3202 }
3203 return h;
3204 }
3205
3206 /**
3207 * Compares this object to the specified object. The result is true if
3208 * and only if the argument is not {@code null} and is a
3209 * {@code LanguageRange} object that contains the same {@code range}
3210 * and {@code weight} values as this object.
3211 *
3212 * @param obj the object to compare with
3213 * @return {@code true} if this object's {@code range} and
3214 * {@code weight} are the same as the {@code obj}'s; {@code false}
3215 * otherwise.
3216 */
3217 @Override
3218 public boolean equals(Object obj) {
3219 if (this == obj) {
3220 return true;
3221 }
3222 if (!(obj instanceof LanguageRange)) {
3223 return false;
3224 }
3225 LanguageRange other = (LanguageRange)obj;
3226 return hash == other.hash
3227 && range.equals(other.range)
3228 && weight == other.weight;
3229 }
3230
3231 /**
3232 * Returns an informative string representation of this {@code LanguageRange}
3233 * object, consisting of language range and weight if the range is
3234 * weighted and the weight is less than the max weight.
3235 *
3236 * @return a string representation of this {@code LanguageRange} object.
3237 */
3238 @Override
3239 public String toString() {
3240 return (weight == MAX_WEIGHT) ? range : range + ";q=" + weight;
3241 }
3242 }
3243
3244 /**
3245 * Returns a list of matching {@code Locale} instances using the filtering
3246 * mechanism defined in RFC 4647.
3247 *
3248 * @param priorityList user's Language Priority List in which each language
3249 * tag is sorted in descending order based on priority or weight
3250 * @param locales {@code Locale} instances used for matching
3251 * @param mode filtering mode
3252 * @return a list of {@code Locale} instances for matching language tags
3253 * sorted in descending order based on priority or weight, or an empty
3254 * list if nothing matches. The list is modifiable.
3255 * @throws NullPointerException if {@code priorityList} or {@code locales}
3256 * is {@code null}
3257 * @throws IllegalArgumentException if one or more extended language ranges
3258 * are included in the given list when
3259 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3260 *
3261 * @since 1.8
3262 */
3263 public static List<Locale> filter(List<LanguageRange> priorityList,
3264 Collection<Locale> locales,
3265 FilteringMode mode) {
3266 return LocaleMatcher.filter(priorityList, locales, mode);
3267 }
3268
3269 /**
3270 * Returns a list of matching {@code Locale} instances using the filtering
3271 * mechanism defined in RFC 4647. This is equivalent to
3272 * {@link #filter(List, Collection, FilteringMode)} when {@code mode} is
3273 * {@link FilteringMode#AUTOSELECT_FILTERING}.
3274 *
3275 * @param priorityList user's Language Priority List in which each language
3276 * tag is sorted in descending order based on priority or weight
3277 * @param locales {@code Locale} instances used for matching
3278 * @return a list of {@code Locale} instances for matching language tags
3279 * sorted in descending order based on priority or weight, or an empty
3280 * list if nothing matches. The list is modifiable.
3281 * @throws NullPointerException if {@code priorityList} or {@code locales}
3282 * is {@code null}
3283 *
3284 * @since 1.8
3285 */
3286 public static List<Locale> filter(List<LanguageRange> priorityList,
3287 Collection<Locale> locales) {
3288 return filter(priorityList, locales, FilteringMode.AUTOSELECT_FILTERING);
3289 }
3290
3291 /**
3292 * Returns a list of matching languages tags using the basic filtering
3293 * mechanism defined in RFC 4647.
3294 *
3295 * @param priorityList user's Language Priority List in which each language
3296 * tag is sorted in descending order based on priority or weight
3297 * @param tags language tags
3298 * @param mode filtering mode
3299 * @return a list of matching language tags sorted in descending order
3300 * based on priority or weight, or an empty list if nothing matches.
3301 * The list is modifiable.
3302 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3303 * {@code null}
3304 * @throws IllegalArgumentException if one or more extended language ranges
3305 * are included in the given list when
3306 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3307 *
3308 * @since 1.8
3309 */
3310 public static List<String> filterTags(List<LanguageRange> priorityList,
3311 Collection<String> tags,
3312 FilteringMode mode) {
3313 return LocaleMatcher.filterTags(priorityList, tags, mode);
3314 }
3315
3316 /**
3317 * Returns a list of matching languages tags using the basic filtering
3318 * mechanism defined in RFC 4647. This is equivalent to
3319 * {@link #filterTags(List, Collection, FilteringMode)} when {@code mode}
3320 * is {@link FilteringMode#AUTOSELECT_FILTERING}.
3321 *
3322 * @param priorityList user's Language Priority List in which each language
3323 * tag is sorted in descending order based on priority or weight
3324 * @param tags language tags
3325 * @return a list of matching language tags sorted in descending order
3326 * based on priority or weight, or an empty list if nothing matches.
3327 * The list is modifiable.
3328 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3329 * {@code null}
3330 *
3331 * @since 1.8
3332 */
3333 public static List<String> filterTags(List<LanguageRange> priorityList,
3334 Collection<String> tags) {
3335 return filterTags(priorityList, tags, FilteringMode.AUTOSELECT_FILTERING);
3336 }
3337
3338 /**
3339 * Returns a {@code Locale} instance for the best-matching language
3340 * tag using the lookup mechanism defined in RFC 4647.
3341 *
3342 * @param priorityList user's Language Priority List in which each language
3343 * tag is sorted in descending order based on priority or weight
3344 * @param locales {@code Locale} instances used for matching
3345 * @return the best matching <code>Locale</code> instance chosen based on
3346 * priority or weight, or {@code null} if nothing matches.
3347 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3348 * {@code null}
3349 *
3350 * @since 1.8
3351 */
3352 public static Locale lookup(List<LanguageRange> priorityList,
3353 Collection<Locale> locales) {
3354 return LocaleMatcher.lookup(priorityList, locales);
3355 }
3356
3357 /**
3358 * Returns the best-matching language tag using the lookup mechanism
3359 * defined in RFC 4647.
3360 *
3361 * @param priorityList user's Language Priority List in which each language
3362 * tag is sorted in descending order based on priority or weight
3363 * @param tags language tangs used for matching
3364 * @return the best matching language tag chosen based on priority or
3365 * weight, or {@code null} if nothing matches.
3366 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3367 * {@code null}
3368 *
3369 * @since 1.8
3370 */
3371 public static String lookupTag(List<LanguageRange> priorityList,
3372 Collection<String> tags) {
3373 return LocaleMatcher.lookupTag(priorityList, tags);
3374 }
3375
3376 }