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.spi.LocaleNameProvider;
50
51 import sun.security.action.GetPropertyAction;
52 import sun.util.locale.BaseLocale;
53 import sun.util.locale.InternalLocaleBuilder;
54 import sun.util.locale.LanguageTag;
55 import sun.util.locale.LocaleExtensions;
56 import sun.util.locale.LocaleMatcher;
57 import sun.util.locale.LocaleObjectCache;
58 import sun.util.locale.LocaleSyntaxException;
59 import sun.util.locale.LocaleUtils;
60 import sun.util.locale.ParseStatus;
61 import sun.util.locale.provider.LocaleProviderAdapter;
62 import sun.util.locale.provider.LocaleResources;
63 import sun.util.locale.provider.LocaleServiceProviderPool;
64
65 /**
66 * A <code>Locale</code> object represents a specific geographical, political,
67 * or cultural region. An operation that requires a <code>Locale</code> to perform
68 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code>
69 * to tailor information for the user. For example, displaying a number
70 * is a locale-sensitive operation— the number should be formatted
71 * according to the customs and conventions of the user's native country,
72 * region, or culture.
73 *
74 * <p> The {@code Locale} class implements IETF BCP 47 which is composed of
75 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 "Matching of Language
76 * Tags"</a> and <a href="http://tools.ietf.org/html/rfc5646">RFC 5646 "Tags
77 * for Identifying Languages"</a> with support for the LDML (UTS#35, "Unicode
78 * Locale Data Markup Language") BCP 47-compatible extensions for locale data
79 * exchange.
80 *
81 * <p> A <code>Locale</code> object logically consists of the fields
82 * described below.
83 *
84 * <dl>
85 * <dt><a name="def_language"><b>language</b></a></dt>
86 *
87 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered
88 * language subtags up to 8 alpha letters (for future enhancements).
89 * When a language has both an alpha-2 code and an alpha-3 code, the
90 * alpha-2 code must be used. You can find a full list of valid
91 * language codes in the IANA Language Subtag Registry (search for
92 * "Type: language"). The language field is case insensitive, but
93 * <code>Locale</code> always canonicalizes to lower case.</dd>
94 *
95 * <dd>Well-formed language values have the form
96 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the full
97 * BCP47 language production, since it excludes extlang. They are
98 * not needed since modern three-letter language codes replace
99 * them.</dd>
100 *
101 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd>
102 *
103 * <dt><a name="def_script"><b>script</b></a></dt>
104 *
105 * <dd>ISO 15924 alpha-4 script code. You can find a full list of
106 * valid script codes in the IANA Language Subtag Registry (search
107 * for "Type: script"). The script field is case insensitive, but
108 * <code>Locale</code> always canonicalizes to title case (the first
109 * letter is upper case and the rest of the letters are lower
110 * case).</dd>
111 *
112 * <dd>Well-formed script values have the form
113 * <code>[a-zA-Z]{4}</code></dd>
114 *
115 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd>
116 *
117 * <dt><a name="def_region"><b>country (region)</b></a></dt>
118 *
119 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code.
120 * You can find a full list of valid country and region codes in the
121 * IANA Language Subtag Registry (search for "Type: region"). The
122 * country (region) field is case insensitive, but
123 * <code>Locale</code> always canonicalizes to upper case.</dd>
124 *
125 * <dd>Well-formed country/region values have
126 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd>
127 *
128 * <dd>Example: "US" (United States), "FR" (France), "029"
129 * (Caribbean)</dd>
130 *
131 * <dt><a name="def_variant"><b>variant</b></a></dt>
132 *
133 * <dd>Any arbitrary value used to indicate a variation of a
134 * <code>Locale</code>. Where there are two or more variant values
135 * each indicating its own semantics, these values should be ordered
136 * by importance, with most important first, separated by
137 * underscore('_'). The variant field is case sensitive.</dd>
138 *
139 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant
140 * subtags. Also BCP 47 subtags are strictly used to indicate
141 * additional variations that define a language or its dialects that
142 * are not covered by any combinations of language, script and
143 * region subtags. You can find a full list of valid variant codes
144 * in the IANA Language Subtag Registry (search for "Type: variant").
145 *
146 * <p>However, the variant field in <code>Locale</code> has
147 * historically been used for any kind of variation, not just
148 * language variations. For example, some supported variants
149 * available in Java SE Runtime Environments indicate alternative
150 * cultural behaviors such as calendar type or number script. In
151 * BCP 47 this kind of information, which does not identify the
152 * language, is supported by extension subtags or private use
153 * subtags.</dd>
154 *
155 * <dd>Well-formed variant values have the form <code>SUBTAG
156 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG =
157 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only
158 * uses hyphen ('-') as a delimiter, this is more lenient).</dd>
159 *
160 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd>
161 *
162 * <dt><a name="def_extensions"><b>extensions</b></a></dt>
163 *
164 * <dd>A map from single character keys to string values, indicating
165 * extensions apart from language identification. The extensions in
166 * <code>Locale</code> implement the semantics and syntax of BCP 47
167 * extension subtags and private use subtags. The extensions are
168 * case insensitive, but <code>Locale</code> canonicalizes all
169 * extension keys and values to lower case. Note that extensions
170 * cannot have empty values.</dd>
171 *
172 * <dd>Well-formed keys are single characters from the set
173 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form
174 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x'
175 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys
176 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows
177 * single-character subtags).</dd>
178 *
179 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar),
180 * key="x"/value="java-1-7"</dd>
181 * </dl>
182 *
183 * <b>Note:</b> Although BCP 47 requires field values to be registered
184 * in the IANA Language Subtag Registry, the <code>Locale</code> class
185 * does not provide any validation features. The <code>Builder</code>
186 * only checks if an individual field satisfies the syntactic
187 * requirement (is well-formed), but does not validate the value
188 * itself. See {@link Builder} for details.
189 *
190 * <h3><a name="def_locale_extension">Unicode locale/language extension</a></h3>
191 *
192 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional
193 * attributes and keywords to override or refine the default behavior
194 * associated with a locale. A keyword is represented by a pair of
195 * key and type. For example, "nu-thai" indicates that Thai local
196 * digits (value:"thai") should be used for formatting numbers
197 * (key:"nu").
198 *
199 * <p>The keywords are mapped to a BCP 47 extension value using the
200 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above
201 * example, "nu-thai", becomes the extension "u-nu-thai".
202 *
203 * <p>Thus, when a <code>Locale</code> object contains Unicode locale
204 * attributes and keywords,
205 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a
206 * String representing this information, for example, "nu-thai". The
207 * <code>Locale</code> class also provides {@link
208 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and
209 * {@link #getUnicodeLocaleType} which allow you to access Unicode
210 * locale attributes and key/type pairs directly. When represented as
211 * a string, the Unicode Locale Extension lists attributes
212 * alphabetically, followed by key/type sequences with keys listed
213 * alphabetically (the order of subtags comprising a key's type is
214 * fixed when the type is defined)
215 *
216 * <p>A well-formed locale key has the form
217 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the
218 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it
219 * can be empty, or a series of subtags 3-8 alphanums in length). A
220 * well-formed locale attribute has the form
221 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same
222 * form as a locale type subtag).
223 *
224 * <p>The Unicode locale extension specifies optional behavior in
225 * locale-sensitive services. Although the LDML specification defines
226 * various keys and values, actual locale-sensitive service
227 * implementations in a Java Runtime Environment might not support any
228 * particular Unicode locale attributes or key/type pairs.
229 *
230 * <h4>Creating a Locale</h4>
231 *
232 * <p>There are several different ways to create a <code>Locale</code>
233 * object.
234 *
235 * <h5>Builder</h5>
236 *
237 * <p>Using {@link Builder} you can construct a <code>Locale</code> object
238 * that conforms to BCP 47 syntax.
239 *
240 * <h5>Constructors</h5>
241 *
242 * <p>The <code>Locale</code> class provides three constructors:
243 * <blockquote>
244 * <pre>
245 * {@link #Locale(String language)}
246 * {@link #Locale(String language, String country)}
247 * {@link #Locale(String language, String country, String variant)}
248 * </pre>
249 * </blockquote>
250 * These constructors allow you to create a <code>Locale</code> object
251 * with language, country and variant, but you cannot specify
252 * script or extensions.
253 *
254 * <h5>Factory Methods</h5>
255 *
256 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code>
257 * object for a well-formed BCP 47 language tag.
258 *
259 * <h5>Locale Constants</h5>
260 *
261 * <p>The <code>Locale</code> class provides a number of convenient constants
262 * that you can use to create <code>Locale</code> objects for commonly used
263 * locales. For example, the following creates a <code>Locale</code> object
264 * for the United States:
265 * <blockquote>
266 * <pre>
267 * Locale.US
268 * </pre>
269 * </blockquote>
270 *
271 * <h4><a name="LocaleMatching">Locale Matching</a></h4>
272 *
273 * <p>If an application or a system is internationalized and provides localized
274 * resources for multiple locales, it sometimes needs to find one or more
275 * locales (or language tags) which meet each user's specific preferences. Note
276 * that a term "language tag" is used interchangeably with "locale" in this
277 * locale matching documentation.
278 *
279 * <p>In order to do matching a user's preferred locales to a set of language
280 * tags, <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
281 * Language Tags</a> defines two mechanisms: filtering and lookup.
282 * <em>Filtering</em> is used to get all matching locales, whereas
283 * <em>lookup</em> is to choose the best matching locale.
284 * Matching is done case-insensitively. These matching mechanisms are described
285 * in the following sections.
286 *
287 * <p>A user's preference is called a <em>Language Priority List</em> and is
288 * expressed as a list of language ranges. There are syntactically two types of
289 * language ranges: basic and extended. See
290 * {@link Locale.LanguageRange Locale.LanguageRange} for details.
291 *
292 * <h5>Filtering</h5>
293 *
294 * <p>The filtering operation returns all matching language tags. It is defined
295 * in RFC 4647 as follows:
296 * "In filtering, each language range represents the least specific language
297 * tag (that is, the language tag with fewest number of subtags) that is an
298 * acceptable match. All of the language tags in the matching set of tags will
299 * have an equal or greater number of subtags than the language range. Every
300 * non-wildcard subtag in the language range will appear in every one of the
301 * matching language tags."
302 *
303 * <p>There are two types of filtering: filtering for basic language ranges
304 * (called "basic filtering") and filtering for extended language ranges
305 * (called "extended filtering"). They may return different results by what
306 * kind of language ranges are included in the given Language Priority List.
307 * {@link Locale.FilteringMode} is a parameter to specify how filtering should
308 * be done.
309 *
310 * <h5>Lookup</h5>
311 *
312 * <p>The lookup operation returns the best matching language tags. It is
313 * defined in RFC 4647 as follows:
314 * "By contrast with filtering, each language range represents the most
315 * specific tag that is an acceptable match. The first matching tag found,
316 * according to the user's priority, is considered the closest match and is the
317 * item returned."
318 *
319 * <p>For example, if a Language Priority List consists of two language ranges,
320 * {@code "zh-Hant-TW"} and {@code "en-US"}, in prioritized order, lookup
321 * method progressively searches the language tags below in order to find the
322 * best matching language tag.
323 * <blockquote>
324 * <pre>
325 * 1. zh-Hant-TW
326 * 2. zh-Hant
327 * 3. zh
328 * 4. en-US
329 * 5. en
330 * </pre>
331 * </blockquote>
332 * If there is a language tag which matches completely to a language range
333 * above, the language tag is returned.
334 *
335 * <p>{@code "*"} is the special language range, and it is ignored in lookup.
336 *
337 * <p>If multiple language tags match as a result of the subtag {@code '*'}
338 * included in a language range, the first matching language tag returned by
339 * an {@link Iterator} over a {@link Collection} of language tags is treated as
340 * the best matching one.
341 *
342 * <h4>Use of Locale</h4>
343 *
344 * <p>Once you've created a <code>Locale</code> you can query it for information
345 * about itself. Use <code>getCountry</code> to get the country (or region)
346 * code and <code>getLanguage</code> to get the language code.
347 * You can use <code>getDisplayCountry</code> to get the
348 * name of the country suitable for displaying to the user. Similarly,
349 * you can use <code>getDisplayLanguage</code> to get the name of
350 * the language suitable for displaying to the user. Interestingly,
351 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive
352 * and have two versions: one that uses the default
353 * {@link Locale.Category#DISPLAY DISPLAY} locale and one
354 * that uses the locale specified as an argument.
355 *
356 * <p>The Java Platform provides a number of classes that perform locale-sensitive
357 * operations. For example, the <code>NumberFormat</code> class formats
358 * numbers, currency, and percentages in a locale-sensitive manner. Classes
359 * such as <code>NumberFormat</code> have several convenience methods
360 * for creating a default object of that type. For example, the
361 * <code>NumberFormat</code> class provides these three convenience methods
362 * for creating a default <code>NumberFormat</code> object:
363 * <blockquote>
364 * <pre>
365 * NumberFormat.getInstance()
366 * NumberFormat.getCurrencyInstance()
367 * NumberFormat.getPercentInstance()
368 * </pre>
369 * </blockquote>
370 * Each of these methods has two variants; one with an explicit locale
371 * and one without; the latter uses the default
372 * {@link Locale.Category#FORMAT FORMAT} locale:
373 * <blockquote>
374 * <pre>
375 * NumberFormat.getInstance(myLocale)
376 * NumberFormat.getCurrencyInstance(myLocale)
377 * NumberFormat.getPercentInstance(myLocale)
378 * </pre>
379 * </blockquote>
380 * A <code>Locale</code> is the mechanism for identifying the kind of object
381 * (<code>NumberFormat</code>) that you would like to get. The locale is
382 * <STRONG>just</STRONG> a mechanism for identifying objects,
383 * <STRONG>not</STRONG> a container for the objects themselves.
384 *
385 * <h4>Compatibility</h4>
386 *
387 * <p>In order to maintain compatibility with existing usage, Locale's
388 * constructors retain their behavior prior to the Java Runtime
389 * Environment version 1.7. The same is largely true for the
390 * <code>toString</code> method. Thus Locale objects can continue to
391 * be used as they were. In particular, clients who parse the output
392 * of toString into language, country, and variant fields can continue
393 * to do so (although this is strongly discouraged), although the
394 * variant field will have additional information in it if script or
395 * extensions are present.
396 *
397 * <p>In addition, BCP 47 imposes syntax restrictions that are not
398 * imposed by Locale's constructors. This means that conversions
399 * between some Locales and BCP 47 language tags cannot be made without
400 * losing information. Thus <code>toLanguageTag</code> cannot
401 * represent the state of locales whose language, country, or variant
402 * do not conform to BCP 47.
403 *
404 * <p>Because of these issues, it is recommended that clients migrate
405 * away from constructing non-conforming locales and use the
406 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead.
407 * Clients desiring a string representation of the complete locale can
408 * then always rely on <code>toLanguageTag</code> for this purpose.
409 *
410 * <h5><a name="special_cases_constructor">Special cases</a></h5>
411 *
412 * <p>For compatibility reasons, two
413 * non-conforming locales are treated as special cases. These are
414 * <b>{@code ja_JP_JP}</b> and <b>{@code th_TH_TH}</b>. These are ill-formed
415 * in BCP 47 since the variants are too short. To ease migration to BCP 47,
416 * these are treated specially during construction. These two cases (and only
417 * these) cause a constructor to generate an extension, all other values behave
418 * exactly as they did prior to Java 7.
419 *
420 * <p>Java has used {@code ja_JP_JP} to represent Japanese as used in
421 * Japan together with the Japanese Imperial calendar. This is now
422 * representable using a Unicode locale extension, by specifying the
423 * Unicode locale key {@code ca} (for "calendar") and type
424 * {@code japanese}. When the Locale constructor is called with the
425 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is
426 * automatically added.
427 *
428 * <p>Java has used {@code th_TH_TH} to represent Thai as used in
429 * Thailand together with Thai digits. This is also now representable using
430 * a Unicode locale extension, by specifying the Unicode locale key
431 * {@code nu} (for "number") and value {@code thai}. When the Locale
432 * constructor is called with the arguments "th", "TH", "TH", the
433 * extension "u-nu-thai" is automatically added.
434 *
435 * <h5>Serialization</h5>
436 *
437 * <p>During serialization, writeObject writes all fields to the output
438 * stream, including extensions.
439 *
440 * <p>During deserialization, readResolve adds extensions as described
441 * in <a href="#special_cases_constructor">Special Cases</a>, only
442 * for the two cases th_TH_TH and ja_JP_JP.
443 *
444 * <h5>Legacy language codes</h5>
445 *
446 * <p>Locale's constructor has always converted three language codes to
447 * their earlier, obsoleted forms: {@code he} maps to {@code iw},
448 * {@code yi} maps to {@code ji}, and {@code id} maps to
449 * {@code in}. This continues to be the case, in order to not break
450 * backwards compatibility.
451 *
452 * <p>The APIs added in 1.7 map between the old and new language codes,
453 * maintaining the old codes internal to Locale (so that
454 * <code>getLanguage</code> and <code>toString</code> reflect the old
455 * code), but using the new codes in the BCP 47 language tag APIs (so
456 * that <code>toLanguageTag</code> reflects the new one). This
457 * preserves the equivalence between Locales no matter which code or
458 * API is used to construct them. Java's default resource bundle
459 * lookup mechanism also implements this mapping, so that resources
460 * can be named using either convention, see {@link ResourceBundle.Control}.
461 *
462 * <h5>Three-letter language/country(region) codes</h5>
463 *
464 * <p>The Locale constructors have always specified that the language
465 * and the country param be two characters in length, although in
466 * practice they have accepted any length. The specification has now
467 * been relaxed to allow language codes of two to eight characters and
468 * country (region) codes of two to three characters, and in
469 * particular, three-letter language codes and three-digit region
470 * codes as specified in the IANA Language Subtag Registry. For
471 * compatibility, the implementation still does not impose a length
472 * constraint.
473 *
474 * @see Builder
475 * @see ResourceBundle
476 * @see java.text.Format
477 * @see java.text.NumberFormat
478 * @see java.text.Collator
479 * @author Mark Davis
480 * @since 1.1
481 */
482 public final class Locale implements Cloneable, Serializable {
483
484 private static final Cache LOCALECACHE = new Cache();
485
486 /** Useful constant for language.
487 */
488 public static final Locale ENGLISH = createConstant("en", "");
489
490 /** Useful constant for language.
491 */
492 public static final Locale FRENCH = createConstant("fr", "");
493
494 /** Useful constant for language.
495 */
496 public static final Locale GERMAN = createConstant("de", "");
497
498 /** Useful constant for language.
499 */
500 public static final Locale ITALIAN = createConstant("it", "");
501
502 /** Useful constant for language.
503 */
504 public static final Locale JAPANESE = createConstant("ja", "");
505
506 /** Useful constant for language.
507 */
508 public static final Locale KOREAN = createConstant("ko", "");
509
510 /** Useful constant for language.
511 */
512 public static final Locale CHINESE = createConstant("zh", "");
513
514 /** Useful constant for language.
515 */
516 public static final Locale SIMPLIFIED_CHINESE = createConstant("zh", "CN");
517
518 /** Useful constant for language.
519 */
520 public static final Locale TRADITIONAL_CHINESE = createConstant("zh", "TW");
521
522 /** Useful constant for country.
523 */
524 public static final Locale FRANCE = createConstant("fr", "FR");
525
526 /** Useful constant for country.
527 */
528 public static final Locale GERMANY = createConstant("de", "DE");
529
530 /** Useful constant for country.
531 */
532 public static final Locale ITALY = createConstant("it", "IT");
533
534 /** Useful constant for country.
535 */
536 public static final Locale JAPAN = createConstant("ja", "JP");
537
538 /** Useful constant for country.
539 */
540 public static final Locale KOREA = createConstant("ko", "KR");
541
542 /** Useful constant for country.
543 */
544 public static final Locale CHINA = SIMPLIFIED_CHINESE;
545
546 /** Useful constant for country.
547 */
548 public static final Locale PRC = SIMPLIFIED_CHINESE;
549
550 /** Useful constant for country.
551 */
552 public static final Locale TAIWAN = TRADITIONAL_CHINESE;
553
554 /** Useful constant for country.
555 */
556 public static final Locale UK = createConstant("en", "GB");
557
558 /** Useful constant for country.
559 */
560 public static final Locale US = createConstant("en", "US");
561
562 /** Useful constant for country.
563 */
564 public static final Locale CANADA = createConstant("en", "CA");
565
566 /** Useful constant for country.
567 */
568 public static final Locale CANADA_FRENCH = createConstant("fr", "CA");
569
570 /**
571 * Useful constant for the root locale. The root locale is the locale whose
572 * language, country, and variant are empty ("") strings. This is regarded
573 * as the base locale of all locales, and is used as the language/country
574 * neutral locale for the locale sensitive operations.
575 *
576 * @since 1.6
577 */
578 public static final Locale ROOT = createConstant("", "");
579
580 /**
581 * The key for the private use extension ('x').
582 *
583 * @see #getExtension(char)
584 * @see Builder#setExtension(char, String)
585 * @since 1.7
586 */
587 public static final char PRIVATE_USE_EXTENSION = 'x';
588
589 /**
590 * The key for Unicode locale extension ('u').
591 *
592 * @see #getExtension(char)
593 * @see Builder#setExtension(char, String)
594 * @since 1.7
595 */
596 public static final char UNICODE_LOCALE_EXTENSION = 'u';
597
598 /** serialization ID
599 */
600 static final long serialVersionUID = 9149081749638150636L;
601
602 /**
603 * Display types for retrieving localized names from the name providers.
604 */
605 private static final int DISPLAY_LANGUAGE = 0;
606 private static final int DISPLAY_COUNTRY = 1;
607 private static final int DISPLAY_VARIANT = 2;
608 private static final int DISPLAY_SCRIPT = 3;
609
610 /**
611 * Private constructor used by getInstance method
612 */
613 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) {
614 this.baseLocale = baseLocale;
615 this.localeExtensions = extensions;
616 }
617
618 /**
619 * Construct a locale from language, country and variant.
620 * This constructor normalizes the language value to lowercase and
621 * the country value to uppercase.
622 * <p>
623 * <b>Note:</b>
624 * <ul>
625 * <li>ISO 639 is not a stable standard; some of the language codes it defines
626 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
627 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
628 * API on Locale will return only the OLD codes.
629 * <li>For backward compatibility reasons, this constructor does not make
630 * any syntactic checks on the input.
631 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially,
632 * see <a href="#special_cases_constructor">Special Cases</a> for more information.
633 * </ul>
634 *
635 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
636 * up to 8 characters in length. See the <code>Locale</code> class description about
637 * valid language values.
638 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
639 * See the <code>Locale</code> class description about valid country values.
640 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>.
641 * See the <code>Locale</code> class description for the details.
642 * @exception NullPointerException thrown if any argument is null.
643 */
644 public Locale(String language, String country, String variant) {
645 if (language== null || country == null || variant == null) {
646 throw new NullPointerException();
647 }
648 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant);
649 localeExtensions = getCompatibilityExtensions(language, "", country, variant);
650 }
651
652 /**
653 * Construct a locale from language and country.
654 * This constructor normalizes the language value to lowercase and
655 * the country value to uppercase.
656 * <p>
657 * <b>Note:</b>
658 * <ul>
659 * <li>ISO 639 is not a stable standard; some of the language codes it defines
660 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
661 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
662 * API on Locale will return only the OLD codes.
663 * <li>For backward compatibility reasons, this constructor does not make
664 * any syntactic checks on the input.
665 * </ul>
666 *
667 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
668 * up to 8 characters in length. See the <code>Locale</code> class description about
669 * valid language values.
670 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
671 * See the <code>Locale</code> class description about valid country values.
672 * @exception NullPointerException thrown if either argument is null.
673 */
674 public Locale(String language, String country) {
675 this(language, country, "");
676 }
677
678 /**
679 * Construct a locale from a language code.
680 * This constructor normalizes the language value to lowercase.
681 * <p>
682 * <b>Note:</b>
683 * <ul>
684 * <li>ISO 639 is not a stable standard; some of the language codes it defines
685 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
686 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
687 * API on Locale will return only the OLD codes.
688 * <li>For backward compatibility reasons, this constructor does not make
689 * any syntactic checks on the input.
690 * </ul>
691 *
692 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
693 * up to 8 characters in length. See the <code>Locale</code> class description about
694 * valid language values.
695 * @exception NullPointerException thrown if argument is null.
696 * @since 1.4
697 */
698 public Locale(String language) {
699 this(language, "", "");
700 }
701
702 /**
703 * This method must be called only for creating the Locale.*
704 * constants due to making shortcuts.
705 */
706 private static Locale createConstant(String lang, String country) {
707 BaseLocale base = BaseLocale.createInstance(lang, country);
708 return getInstance(base, null);
709 }
710
711 /**
712 * Returns a <code>Locale</code> constructed from the given
713 * <code>language</code>, <code>country</code> and
714 * <code>variant</code>. If the same <code>Locale</code> instance
715 * is available in the cache, then that instance is
716 * returned. Otherwise, a new <code>Locale</code> instance is
717 * created and cached.
718 *
719 * @param language lowercase 2 to 8 language code.
720 * @param country uppercase two-letter ISO-3166 code and numeric-3 UN M.49 area code.
721 * @param variant vendor and browser specific code. See class description.
722 * @return the <code>Locale</code> instance requested
723 * @exception NullPointerException if any argument is null.
724 */
725 static Locale getInstance(String language, String country, String variant) {
726 return getInstance(language, "", country, variant, null);
727 }
728
729 static Locale getInstance(String language, String script, String country,
730 String variant, LocaleExtensions extensions) {
731 if (language== null || script == null || country == null || variant == null) {
732 throw new NullPointerException();
733 }
734
735 if (extensions == null) {
736 extensions = getCompatibilityExtensions(language, script, country, variant);
737 }
738
739 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant);
740 return getInstance(baseloc, extensions);
741 }
742
743 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) {
744 LocaleKey key = new LocaleKey(baseloc, extensions);
745 return LOCALECACHE.get(key);
746 }
747
748 private static class Cache extends LocaleObjectCache<LocaleKey, Locale> {
749 private Cache() {
750 }
751
752 @Override
753 protected Locale createObject(LocaleKey key) {
754 return new Locale(key.base, key.exts);
755 }
756 }
757
758 private static final class LocaleKey {
759 private final BaseLocale base;
760 private final LocaleExtensions exts;
761 private final int hash;
762
763 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) {
764 base = baseLocale;
765 exts = extensions;
766
767 // Calculate the hash value here because it's always used.
768 int h = base.hashCode();
769 if (exts != null) {
770 h ^= exts.hashCode();
771 }
772 hash = h;
773 }
774
775 @Override
776 public boolean equals(Object obj) {
777 if (this == obj) {
778 return true;
779 }
780 if (!(obj instanceof LocaleKey)) {
781 return false;
782 }
783 LocaleKey other = (LocaleKey)obj;
784 if (hash != other.hash || !base.equals(other.base)) {
785 return false;
786 }
787 if (exts == null) {
788 return other.exts == null;
789 }
790 return exts.equals(other.exts);
791 }
792
793 @Override
794 public int hashCode() {
795 return hash;
796 }
797 }
798
799 /**
800 * Gets the current value of the default locale for this instance
801 * of the Java Virtual Machine.
802 * <p>
803 * The Java Virtual Machine sets the default locale during startup
804 * based on the host environment. It is used by many locale-sensitive
805 * methods if no locale is explicitly specified.
806 * It can be changed using the
807 * {@link #setDefault(java.util.Locale) setDefault} method.
808 *
809 * @return the default locale for this instance of the Java Virtual Machine
810 */
811 public static Locale getDefault() {
812 // do not synchronize this method - see 4071298
813 return defaultLocale;
814 }
815
816 /**
817 * Gets the current value of the default locale for the specified Category
818 * for this instance of the Java Virtual Machine.
819 * <p>
820 * The Java Virtual Machine sets the default locale during startup based
821 * on the host environment. It is used by many locale-sensitive methods
822 * if no locale is explicitly specified. It can be changed using the
823 * setDefault(Locale.Category, Locale) method.
824 *
825 * @param category - the specified category to get the default locale
826 * @throws NullPointerException if category is null
827 * @return the default locale for the specified Category for this instance
828 * of the Java Virtual Machine
829 * @see #setDefault(Locale.Category, Locale)
830 * @since 1.7
831 */
832 public static Locale getDefault(Locale.Category category) {
833 // do not synchronize this method - see 4071298
834 switch (category) {
835 case DISPLAY:
836 if (defaultDisplayLocale == null) {
837 synchronized(Locale.class) {
838 if (defaultDisplayLocale == null) {
839 defaultDisplayLocale = initDefault(category);
840 }
841 }
842 }
843 return defaultDisplayLocale;
844 case FORMAT:
845 if (defaultFormatLocale == null) {
846 synchronized(Locale.class) {
847 if (defaultFormatLocale == null) {
848 defaultFormatLocale = initDefault(category);
849 }
850 }
851 }
852 return defaultFormatLocale;
853 default:
854 assert false: "Unknown Category";
855 }
856 return getDefault();
857 }
858
859 private static Locale initDefault() {
860 String language, region, script, country, variant;
861 Properties props = GetPropertyAction.privilegedGetProperties();
862 language = props.getProperty("user.language", "en");
863 // for compatibility, check for old user.region property
864 region = props.getProperty("user.region");
865 if (region != null) {
866 // region can be of form country, country_variant, or _variant
867 int i = region.indexOf('_');
868 if (i >= 0) {
869 country = region.substring(0, i);
870 variant = region.substring(i + 1);
871 } else {
872 country = region;
873 variant = "";
874 }
875 script = "";
876 } else {
877 script = props.getProperty("user.script", "");
878 country = props.getProperty("user.country", "");
879 variant = props.getProperty("user.variant", "");
880 }
881
882 return getInstance(language, script, country, variant, null);
883 }
884
885 private static Locale initDefault(Locale.Category category) {
886 Properties props = GetPropertyAction.privilegedGetProperties();
887 return getInstance(
888 props.getProperty(category.languageKey,
889 defaultLocale.getLanguage()),
890 props.getProperty(category.scriptKey,
891 defaultLocale.getScript()),
892 props.getProperty(category.countryKey,
893 defaultLocale.getCountry()),
894 props.getProperty(category.variantKey,
895 defaultLocale.getVariant()),
896 null);
897 }
898
899 /**
900 * Sets the default locale for this instance of the Java Virtual Machine.
901 * This does not affect the host locale.
902 * <p>
903 * If there is a security manager, its <code>checkPermission</code>
904 * method is called with a <code>PropertyPermission("user.language", "write")</code>
905 * permission before the default locale is changed.
906 * <p>
907 * The Java Virtual Machine sets the default locale during startup
908 * based on the host environment. It is used by many locale-sensitive
909 * methods if no locale is explicitly specified.
910 * <p>
911 * Since changing the default locale may affect many different areas
912 * of functionality, this method should only be used if the caller
913 * is prepared to reinitialize locale-sensitive code running
914 * within the same Java Virtual Machine.
915 * <p>
916 * By setting the default locale with this method, all of the default
917 * locales for each Category are also set to the specified default locale.
918 *
919 * @throws SecurityException
920 * if a security manager exists and its
921 * <code>checkPermission</code> method doesn't allow the operation.
922 * @throws NullPointerException if <code>newLocale</code> is null
923 * @param newLocale the new default locale
924 * @see SecurityManager#checkPermission
925 * @see java.util.PropertyPermission
926 */
927 public static synchronized void setDefault(Locale newLocale) {
928 setDefault(Category.DISPLAY, newLocale);
929 setDefault(Category.FORMAT, newLocale);
930 defaultLocale = newLocale;
931 }
932
933 /**
934 * Sets the default locale for the specified Category for this instance
935 * of the Java Virtual Machine. This does not affect the host locale.
936 * <p>
937 * If there is a security manager, its checkPermission method is called
938 * with a PropertyPermission("user.language", "write") permission before
939 * the default locale is changed.
940 * <p>
941 * The Java Virtual Machine sets the default locale during startup based
942 * on the host environment. It is used by many locale-sensitive methods
943 * if no locale is explicitly specified.
944 * <p>
945 * Since changing the default locale may affect many different areas of
946 * functionality, this method should only be used if the caller is
947 * prepared to reinitialize locale-sensitive code running within the
948 * same Java Virtual Machine.
949 *
950 * @param category - the specified category to set the default locale
951 * @param newLocale - the new default locale
952 * @throws SecurityException if a security manager exists and its
953 * checkPermission method doesn't allow the operation.
954 * @throws NullPointerException if category and/or newLocale is null
955 * @see SecurityManager#checkPermission(java.security.Permission)
956 * @see PropertyPermission
957 * @see #getDefault(Locale.Category)
958 * @since 1.7
959 */
960 public static synchronized void setDefault(Locale.Category category,
961 Locale newLocale) {
962 if (category == null)
963 throw new NullPointerException("Category cannot be NULL");
964 if (newLocale == null)
965 throw new NullPointerException("Can't set default locale to NULL");
966
967 SecurityManager sm = System.getSecurityManager();
968 if (sm != null) sm.checkPermission(new PropertyPermission
969 ("user.language", "write"));
970 switch (category) {
971 case DISPLAY:
972 defaultDisplayLocale = newLocale;
973 break;
974 case FORMAT:
975 defaultFormatLocale = newLocale;
976 break;
977 default:
978 assert false: "Unknown Category";
979 }
980 }
981
982 /**
983 * Returns an array of all installed locales.
984 * The returned array represents the union of locales supported
985 * by the Java runtime environment and by installed
986 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider}
987 * implementations. It must contain at least a <code>Locale</code>
988 * instance equal to {@link java.util.Locale#US Locale.US}.
989 *
990 * @return An array of installed locales.
991 */
992 public static Locale[] getAvailableLocales() {
993 return LocaleServiceProviderPool.getAllAvailableLocales();
994 }
995
996 /**
997 * Returns a list of all 2-letter country codes defined in ISO 3166.
998 * Can be used to create Locales.
999 * <p>
1000 * <b>Note:</b> The <code>Locale</code> class also supports other codes for
1001 * country (region), such as 3-letter numeric UN M.49 area codes.
1002 * Therefore, the list returned by this method does not contain ALL valid
1003 * codes that can be used to create Locales.
1004 *
1005 * @return An array of ISO 3166 two-letter country codes.
1006 */
1007 public static String[] getISOCountries() {
1008 if (isoCountries == null) {
1009 isoCountries = getISO2Table(LocaleISOData.isoCountryTable);
1010 }
1011 String[] result = new String[isoCountries.length];
1012 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length);
1013 return result;
1014 }
1015
1016 /**
1017 * Returns a list of all 2-letter language codes defined in ISO 639.
1018 * Can be used to create Locales.
1019 * <p>
1020 * <b>Note:</b>
1021 * <ul>
1022 * <li>ISO 639 is not a stable standard— some languages' codes have changed.
1023 * The list this function returns includes both the new and the old codes for the
1024 * languages whose codes have changed.
1025 * <li>The <code>Locale</code> class also supports language codes up to
1026 * 8 characters in length. Therefore, the list returned by this method does
1027 * not contain ALL valid codes that can be used to create Locales.
1028 * </ul>
1029 *
1030 * @return An array of ISO 639 two-letter language codes.
1031 */
1032 public static String[] getISOLanguages() {
1033 if (isoLanguages == null) {
1034 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable);
1035 }
1036 String[] result = new String[isoLanguages.length];
1037 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length);
1038 return result;
1039 }
1040
1041 private static String[] getISO2Table(String table) {
1042 int len = table.length() / 5;
1043 String[] isoTable = new String[len];
1044 for (int i = 0, j = 0; i < len; i++, j += 5) {
1045 isoTable[i] = table.substring(j, j + 2);
1046 }
1047 return isoTable;
1048 }
1049
1050 /**
1051 * Returns the language code of this Locale.
1052 *
1053 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed.
1054 * Locale's constructor recognizes both the new and the old codes for the languages
1055 * whose codes have changed, but this function always returns the old code. If you
1056 * want to check for a specific language whose code has changed, don't do
1057 * <pre>
1058 * if (locale.getLanguage().equals("he")) // BAD!
1059 * ...
1060 * </pre>
1061 * Instead, do
1062 * <pre>
1063 * if (locale.getLanguage().equals(new Locale("he").getLanguage()))
1064 * ...
1065 * </pre>
1066 * @return The language code, or the empty string if none is defined.
1067 * @see #getDisplayLanguage
1068 */
1069 public String getLanguage() {
1070 return baseLocale.getLanguage();
1071 }
1072
1073 /**
1074 * Returns the script for this locale, which should
1075 * either be the empty string or an ISO 15924 4-letter script
1076 * code. The first letter is uppercase and the rest are
1077 * lowercase, for example, 'Latn', 'Cyrl'.
1078 *
1079 * @return The script code, or the empty string if none is defined.
1080 * @see #getDisplayScript
1081 * @since 1.7
1082 */
1083 public String getScript() {
1084 return baseLocale.getScript();
1085 }
1086
1087 /**
1088 * Returns the country/region code for this locale, which should
1089 * either be the empty string, an uppercase ISO 3166 2-letter code,
1090 * or a UN M.49 3-digit code.
1091 *
1092 * @return The country/region code, or the empty string if none is defined.
1093 * @see #getDisplayCountry
1094 */
1095 public String getCountry() {
1096 return baseLocale.getRegion();
1097 }
1098
1099 /**
1100 * Returns the variant code for this locale.
1101 *
1102 * @return The variant code, or the empty string if none is defined.
1103 * @see #getDisplayVariant
1104 */
1105 public String getVariant() {
1106 return baseLocale.getVariant();
1107 }
1108
1109 /**
1110 * Returns {@code true} if this {@code Locale} has any <a href="#def_extensions">
1111 * extensions</a>.
1112 *
1113 * @return {@code true} if this {@code Locale} has any extensions
1114 * @since 1.8
1115 */
1116 public boolean hasExtensions() {
1117 return localeExtensions != null;
1118 }
1119
1120 /**
1121 * Returns a copy of this {@code Locale} with no <a href="#def_extensions">
1122 * extensions</a>. If this {@code Locale} has no extensions, this {@code Locale}
1123 * is returned.
1124 *
1125 * @return a copy of this {@code Locale} with no extensions, or {@code this}
1126 * if {@code this} has no extensions
1127 * @since 1.8
1128 */
1129 public Locale stripExtensions() {
1130 return hasExtensions() ? Locale.getInstance(baseLocale, null) : this;
1131 }
1132
1133 /**
1134 * Returns the extension (or private use) value associated with
1135 * the specified key, or null if there is no extension
1136 * associated with the key. To be well-formed, the key must be one
1137 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so
1138 * for example 'z' and 'Z' represent the same extension.
1139 *
1140 * @param key the extension key
1141 * @return The extension, or null if this locale defines no
1142 * extension for the specified key.
1143 * @throws IllegalArgumentException if key is not well-formed
1144 * @see #PRIVATE_USE_EXTENSION
1145 * @see #UNICODE_LOCALE_EXTENSION
1146 * @since 1.7
1147 */
1148 public String getExtension(char key) {
1149 if (!LocaleExtensions.isValidKey(key)) {
1150 throw new IllegalArgumentException("Ill-formed extension key: " + key);
1151 }
1152 return hasExtensions() ? localeExtensions.getExtensionValue(key) : null;
1153 }
1154
1155 /**
1156 * Returns the set of extension keys associated with this locale, or the
1157 * empty set if it has no extensions. The returned set is unmodifiable.
1158 * The keys will all be lower-case.
1159 *
1160 * @return The set of extension keys, or the empty set if this locale has
1161 * no extensions.
1162 * @since 1.7
1163 */
1164 public Set<Character> getExtensionKeys() {
1165 if (!hasExtensions()) {
1166 return Collections.emptySet();
1167 }
1168 return localeExtensions.getKeys();
1169 }
1170
1171 /**
1172 * Returns the set of unicode locale attributes associated with
1173 * this locale, or the empty set if it has no attributes. The
1174 * returned set is unmodifiable.
1175 *
1176 * @return The set of attributes.
1177 * @since 1.7
1178 */
1179 public Set<String> getUnicodeLocaleAttributes() {
1180 if (!hasExtensions()) {
1181 return Collections.emptySet();
1182 }
1183 return localeExtensions.getUnicodeLocaleAttributes();
1184 }
1185
1186 /**
1187 * Returns the Unicode locale type associated with the specified Unicode locale key
1188 * for this locale. Returns the empty string for keys that are defined with no type.
1189 * Returns null if the key is not defined. Keys are case-insensitive. The key must
1190 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is
1191 * thrown.
1192 *
1193 * @param key the Unicode locale key
1194 * @return The Unicode locale type associated with the key, or null if the
1195 * locale does not define the key.
1196 * @throws IllegalArgumentException if the key is not well-formed
1197 * @throws NullPointerException if <code>key</code> is null
1198 * @since 1.7
1199 */
1200 public String getUnicodeLocaleType(String key) {
1201 if (!isUnicodeExtensionKey(key)) {
1202 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key);
1203 }
1204 return hasExtensions() ? localeExtensions.getUnicodeLocaleType(key) : null;
1205 }
1206
1207 /**
1208 * Returns the set of Unicode locale keys defined by this locale, or the empty set if
1209 * this locale has none. The returned set is immutable. Keys are all lower case.
1210 *
1211 * @return The set of Unicode locale keys, or the empty set if this locale has
1212 * no Unicode locale keywords.
1213 * @since 1.7
1214 */
1215 public Set<String> getUnicodeLocaleKeys() {
1216 if (localeExtensions == null) {
1217 return Collections.emptySet();
1218 }
1219 return localeExtensions.getUnicodeLocaleKeys();
1220 }
1221
1222 /**
1223 * Package locale method returning the Locale's BaseLocale,
1224 * used by ResourceBundle
1225 * @return base locale of this Locale
1226 */
1227 BaseLocale getBaseLocale() {
1228 return baseLocale;
1229 }
1230
1231 /**
1232 * Package private method returning the Locale's LocaleExtensions,
1233 * used by ResourceBundle.
1234 * @return locale extensions of this Locale,
1235 * or {@code null} if no extensions are defined
1236 */
1237 LocaleExtensions getLocaleExtensions() {
1238 return localeExtensions;
1239 }
1240
1241 /**
1242 * Returns a string representation of this <code>Locale</code>
1243 * object, consisting of language, country, variant, script,
1244 * and extensions as below:
1245 * <blockquote>
1246 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "_" + extensions
1247 * </blockquote>
1248 *
1249 * Language is always lower case, country is always upper case, script is always title
1250 * case, and extensions are always lower case. Extensions and private use subtags
1251 * will be in canonical order as explained in {@link #toLanguageTag}.
1252 *
1253 * <p>When the locale has neither script nor extensions, the result is the same as in
1254 * Java 6 and prior.
1255 *
1256 * <p>If both the language and country fields are missing, this function will return
1257 * the empty string, even if the variant, script, or extensions field is present (you
1258 * can't have a locale with just a variant, the variant must accompany a well-formed
1259 * language or country code).
1260 *
1261 * <p>If script or extensions are present and variant is missing, no underscore is
1262 * added before the "#".
1263 *
1264 * <p>This behavior is designed to support debugging and to be compatible with
1265 * previous uses of <code>toString</code> that expected language, country, and variant
1266 * fields only. To represent a Locale as a String for interchange purposes, use
1267 * {@link #toLanguageTag}.
1268 *
1269 * <p>Examples: <ul>
1270 * <li>{@code en}</li>
1271 * <li>{@code de_DE}</li>
1272 * <li>{@code _GB}</li>
1273 * <li>{@code en_US_WIN}</li>
1274 * <li>{@code de__POSIX}</li>
1275 * <li>{@code zh_CN_#Hans}</li>
1276 * <li>{@code zh_TW_#Hant_x-java}</li>
1277 * <li>{@code th_TH_TH_#u-nu-thai}</li></ul>
1278 *
1279 * @return A string representation of the Locale, for debugging.
1280 * @see #getDisplayName
1281 * @see #toLanguageTag
1282 */
1283 @Override
1284 public final String toString() {
1285 boolean l = (baseLocale.getLanguage().length() != 0);
1286 boolean s = (baseLocale.getScript().length() != 0);
1287 boolean r = (baseLocale.getRegion().length() != 0);
1288 boolean v = (baseLocale.getVariant().length() != 0);
1289 boolean e = (localeExtensions != null && localeExtensions.getID().length() != 0);
1290
1291 StringBuilder result = new StringBuilder(baseLocale.getLanguage());
1292 if (r || (l && (v || s || e))) {
1293 result.append('_')
1294 .append(baseLocale.getRegion()); // This may just append '_'
1295 }
1296 if (v && (l || r)) {
1297 result.append('_')
1298 .append(baseLocale.getVariant());
1299 }
1300
1301 if (s && (l || r)) {
1302 result.append("_#")
1303 .append(baseLocale.getScript());
1304 }
1305
1306 if (e && (l || r)) {
1307 result.append('_');
1308 if (!s) {
1309 result.append('#');
1310 }
1311 result.append(localeExtensions.getID());
1312 }
1313
1314 return result.toString();
1315 }
1316
1317 /**
1318 * Returns a well-formed IETF BCP 47 language tag representing
1319 * this locale.
1320 *
1321 * <p>If this <code>Locale</code> has a language, country, or
1322 * variant that does not satisfy the IETF BCP 47 language tag
1323 * syntax requirements, this method handles these fields as
1324 * described below:
1325 *
1326 * <p><b>Language:</b> If language is empty, or not <a
1327 * href="#def_language" >well-formed</a> (for example "a" or
1328 * "e2"), it will be emitted as "und" (Undetermined).
1329 *
1330 * <p><b>Country:</b> If country is not <a
1331 * href="#def_region">well-formed</a> (for example "12" or "USA"),
1332 * it will be omitted.
1333 *
1334 * <p><b>Variant:</b> If variant <b>is</b> <a
1335 * href="#def_variant">well-formed</a>, each sub-segment
1336 * (delimited by '-' or '_') is emitted as a subtag. Otherwise:
1337 * <ul>
1338 *
1339 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code>
1340 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first
1341 * ill-formed sub-segment and all following will be appended to
1342 * the private use subtag. The first appended subtag will be
1343 * "lvariant", followed by the sub-segments in order, separated by
1344 * hyphen. For example, "x-lvariant-WIN",
1345 * "Oracle-x-lvariant-JDK-Standard-Edition".
1346 *
1347 * <li>if any sub-segment does not match
1348 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated
1349 * and the problematic sub-segment and all following sub-segments
1350 * will be omitted. If the remainder is non-empty, it will be
1351 * emitted as a private use subtag as above (even if the remainder
1352 * turns out to be well-formed). For example,
1353 * "Solaris_isjustthecoolestthing" is emitted as
1354 * "x-lvariant-Solaris", not as "solaris".</li></ul>
1355 *
1356 * <p><b>Special Conversions:</b> Java supports some old locale
1357 * representations, including deprecated ISO language codes,
1358 * for compatibility. This method performs the following
1359 * conversions:
1360 * <ul>
1361 *
1362 * <li>Deprecated ISO language codes "iw", "ji", and "in" are
1363 * converted to "he", "yi", and "id", respectively.
1364 *
1365 * <li>A locale with language "no", country "NO", and variant
1366 * "NY", representing Norwegian Nynorsk (Norway), is converted
1367 * to a language tag "nn-NO".</li></ul>
1368 *
1369 * <p><b>Note:</b> Although the language tag created by this
1370 * method is well-formed (satisfies the syntax requirements
1371 * defined by the IETF BCP 47 specification), it is not
1372 * necessarily a valid BCP 47 language tag. For example,
1373 * <pre>
1374 * new Locale("xx", "YY").toLanguageTag();</pre>
1375 *
1376 * will return "xx-YY", but the language subtag "xx" and the
1377 * region subtag "YY" are invalid because they are not registered
1378 * in the IANA Language Subtag Registry.
1379 *
1380 * @return a BCP47 language tag representing the locale
1381 * @see #forLanguageTag(String)
1382 * @since 1.7
1383 */
1384 public String toLanguageTag() {
1385 if (languageTag != null) {
1386 return languageTag;
1387 }
1388
1389 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions);
1390 StringBuilder buf = new StringBuilder();
1391
1392 String subtag = tag.getLanguage();
1393 if (subtag.length() > 0) {
1394 buf.append(LanguageTag.canonicalizeLanguage(subtag));
1395 }
1396
1397 subtag = tag.getScript();
1398 if (subtag.length() > 0) {
1399 buf.append(LanguageTag.SEP);
1400 buf.append(LanguageTag.canonicalizeScript(subtag));
1401 }
1402
1403 subtag = tag.getRegion();
1404 if (subtag.length() > 0) {
1405 buf.append(LanguageTag.SEP);
1406 buf.append(LanguageTag.canonicalizeRegion(subtag));
1407 }
1408
1409 List<String>subtags = tag.getVariants();
1410 for (String s : subtags) {
1411 buf.append(LanguageTag.SEP);
1412 // preserve casing
1413 buf.append(s);
1414 }
1415
1416 subtags = tag.getExtensions();
1417 for (String s : subtags) {
1418 buf.append(LanguageTag.SEP);
1419 buf.append(LanguageTag.canonicalizeExtension(s));
1420 }
1421
1422 subtag = tag.getPrivateuse();
1423 if (subtag.length() > 0) {
1424 if (buf.length() > 0) {
1425 buf.append(LanguageTag.SEP);
1426 }
1427 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP);
1428 // preserve casing
1429 buf.append(subtag);
1430 }
1431
1432 String langTag = buf.toString();
1433 synchronized (this) {
1434 if (languageTag == null) {
1435 languageTag = langTag;
1436 }
1437 }
1438 return languageTag;
1439 }
1440
1441 /**
1442 * Returns a locale for the specified IETF BCP 47 language tag string.
1443 *
1444 * <p>If the specified language tag contains any ill-formed subtags,
1445 * the first such subtag and all following subtags are ignored. Compare
1446 * to {@link Locale.Builder#setLanguageTag} which throws an exception
1447 * in this case.
1448 *
1449 * <p>The following <b>conversions</b> are performed:<ul>
1450 *
1451 * <li>The language code "und" is mapped to language "".
1452 *
1453 * <li>The language codes "he", "yi", and "id" are mapped to "iw",
1454 * "ji", and "in" respectively. (This is the same canonicalization
1455 * that's done in Locale's constructors.)
1456 *
1457 * <li>The portion of a private use subtag prefixed by "lvariant",
1458 * if any, is removed and appended to the variant field in the
1459 * result locale (without case normalization). If it is then
1460 * empty, the private use subtag is discarded:
1461 *
1462 * <pre>
1463 * Locale loc;
1464 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX");
1465 * loc.getVariant(); // returns "POSIX"
1466 * loc.getExtension('x'); // returns null
1467 *
1468 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def");
1469 * loc.getVariant(); // returns "POSIX_Abc_Def"
1470 * loc.getExtension('x'); // returns "urp"
1471 * </pre>
1472 *
1473 * <li>When the languageTag argument contains an extlang subtag,
1474 * the first such subtag is used as the language, and the primary
1475 * language subtag and other extlang subtags are ignored:
1476 *
1477 * <pre>
1478 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao"
1479 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US"
1480 * </pre>
1481 *
1482 * <li>Case is normalized except for variant tags, which are left
1483 * unchanged. Language is normalized to lower case, script to
1484 * title case, country to upper case, and extensions to lower
1485 * case.
1486 *
1487 * <li>If, after processing, the locale would exactly match either
1488 * ja_JP_JP or th_TH_TH with no extensions, the appropriate
1489 * extensions are added as though the constructor had been called:
1490 *
1491 * <pre>
1492 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag();
1493 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP"
1494 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag();
1495 * // returns "th-TH-u-nu-thai-x-lvariant-TH"
1496 * </pre></ul>
1497 *
1498 * <p>This implements the 'Language-Tag' production of BCP47, and
1499 * so supports grandfathered (regular and irregular) as well as
1500 * private use language tags. Stand alone private use tags are
1501 * represented as empty language and extension 'x-whatever',
1502 * and grandfathered tags are converted to their canonical replacements
1503 * where they exist.
1504 *
1505 * <p>Grandfathered tags with canonical replacements are as follows:
1506 *
1507 * <table summary="Grandfathered tags with canonical replacements">
1508 * <tbody align="center">
1509 * <tr><th>grandfathered tag</th><th> </th><th>modern replacement</th></tr>
1510 * <tr><td>art-lojban</td><td> </td><td>jbo</td></tr>
1511 * <tr><td>i-ami</td><td> </td><td>ami</td></tr>
1512 * <tr><td>i-bnn</td><td> </td><td>bnn</td></tr>
1513 * <tr><td>i-hak</td><td> </td><td>hak</td></tr>
1514 * <tr><td>i-klingon</td><td> </td><td>tlh</td></tr>
1515 * <tr><td>i-lux</td><td> </td><td>lb</td></tr>
1516 * <tr><td>i-navajo</td><td> </td><td>nv</td></tr>
1517 * <tr><td>i-pwn</td><td> </td><td>pwn</td></tr>
1518 * <tr><td>i-tao</td><td> </td><td>tao</td></tr>
1519 * <tr><td>i-tay</td><td> </td><td>tay</td></tr>
1520 * <tr><td>i-tsu</td><td> </td><td>tsu</td></tr>
1521 * <tr><td>no-bok</td><td> </td><td>nb</td></tr>
1522 * <tr><td>no-nyn</td><td> </td><td>nn</td></tr>
1523 * <tr><td>sgn-BE-FR</td><td> </td><td>sfb</td></tr>
1524 * <tr><td>sgn-BE-NL</td><td> </td><td>vgt</td></tr>
1525 * <tr><td>sgn-CH-DE</td><td> </td><td>sgg</td></tr>
1526 * <tr><td>zh-guoyu</td><td> </td><td>cmn</td></tr>
1527 * <tr><td>zh-hakka</td><td> </td><td>hak</td></tr>
1528 * <tr><td>zh-min-nan</td><td> </td><td>nan</td></tr>
1529 * <tr><td>zh-xiang</td><td> </td><td>hsn</td></tr>
1530 * </tbody>
1531 * </table>
1532 *
1533 * <p>Grandfathered tags with no modern replacement will be
1534 * converted as follows:
1535 *
1536 * <table summary="Grandfathered tags with no modern replacement">
1537 * <tbody align="center">
1538 * <tr><th>grandfathered tag</th><th> </th><th>converts to</th></tr>
1539 * <tr><td>cel-gaulish</td><td> </td><td>xtg-x-cel-gaulish</td></tr>
1540 * <tr><td>en-GB-oed</td><td> </td><td>en-GB-x-oed</td></tr>
1541 * <tr><td>i-default</td><td> </td><td>en-x-i-default</td></tr>
1542 * <tr><td>i-enochian</td><td> </td><td>und-x-i-enochian</td></tr>
1543 * <tr><td>i-mingo</td><td> </td><td>see-x-i-mingo</td></tr>
1544 * <tr><td>zh-min</td><td> </td><td>nan-x-zh-min</td></tr>
1545 * </tbody>
1546 * </table>
1547 *
1548 * <p>For a list of all grandfathered tags, see the
1549 * IANA Language Subtag Registry (search for "Type: grandfathered").
1550 *
1551 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code>
1552 * and <code>forLanguageTag</code> will round-trip.
1553 *
1554 * @param languageTag the language tag
1555 * @return The locale that best represents the language tag.
1556 * @throws NullPointerException if <code>languageTag</code> is <code>null</code>
1557 * @see #toLanguageTag()
1558 * @see java.util.Locale.Builder#setLanguageTag(String)
1559 * @since 1.7
1560 */
1561 public static Locale forLanguageTag(String languageTag) {
1562 LanguageTag tag = LanguageTag.parse(languageTag, null);
1563 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
1564 bldr.setLanguageTag(tag);
1565 BaseLocale base = bldr.getBaseLocale();
1566 LocaleExtensions exts = bldr.getLocaleExtensions();
1567 if (exts == null && base.getVariant().length() > 0) {
1568 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(),
1569 base.getRegion(), base.getVariant());
1570 }
1571 return getInstance(base, exts);
1572 }
1573
1574 /**
1575 * Returns a three-letter abbreviation of this locale's language.
1576 * If the language matches an ISO 639-1 two-letter code, the
1577 * corresponding ISO 639-2/T three-letter lowercase code is
1578 * returned. The ISO 639-2 language codes can be found on-line,
1579 * see "Codes for the Representation of Names of Languages Part 2:
1580 * Alpha-3 Code". If the locale specifies a three-letter
1581 * language, the language is returned as is. If the locale does
1582 * not specify a language the empty string is returned.
1583 *
1584 * @return A three-letter abbreviation of this locale's language.
1585 * @exception MissingResourceException Throws MissingResourceException if
1586 * three-letter language abbreviation is not available for this locale.
1587 */
1588 public String getISO3Language() throws MissingResourceException {
1589 String lang = baseLocale.getLanguage();
1590 if (lang.length() == 3) {
1591 return lang;
1592 }
1593
1594 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable);
1595 if (language3 == null) {
1596 throw new MissingResourceException("Couldn't find 3-letter language code for "
1597 + lang, "FormatData_" + toString(), "ShortLanguage");
1598 }
1599 return language3;
1600 }
1601
1602 /**
1603 * Returns a three-letter abbreviation for this locale's country.
1604 * If the country matches an ISO 3166-1 alpha-2 code, the
1605 * corresponding ISO 3166-1 alpha-3 uppercase code is returned.
1606 * If the locale doesn't specify a country, this will be the empty
1607 * string.
1608 *
1609 * <p>The ISO 3166-1 codes can be found on-line.
1610 *
1611 * @return A three-letter abbreviation of this locale's country.
1612 * @exception MissingResourceException Throws MissingResourceException if the
1613 * three-letter country abbreviation is not available for this locale.
1614 */
1615 public String getISO3Country() throws MissingResourceException {
1616 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable);
1617 if (country3 == null) {
1618 throw new MissingResourceException("Couldn't find 3-letter country code for "
1619 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry");
1620 }
1621 return country3;
1622 }
1623
1624 private static String getISO3Code(String iso2Code, String table) {
1625 int codeLength = iso2Code.length();
1626 if (codeLength == 0) {
1627 return "";
1628 }
1629
1630 int tableLength = table.length();
1631 int index = tableLength;
1632 if (codeLength == 2) {
1633 char c1 = iso2Code.charAt(0);
1634 char c2 = iso2Code.charAt(1);
1635 for (index = 0; index < tableLength; index += 5) {
1636 if (table.charAt(index) == c1
1637 && table.charAt(index + 1) == c2) {
1638 break;
1639 }
1640 }
1641 }
1642 return index < tableLength ? table.substring(index + 2, index + 5) : null;
1643 }
1644
1645 /**
1646 * Returns a name for the locale's language that is appropriate for display to the
1647 * user.
1648 * If possible, the name returned will be localized for the default
1649 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1650 * For example, if the locale is fr_FR and the default
1651 * {@link Locale.Category#DISPLAY DISPLAY} locale
1652 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1653 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1654 * getDisplayLanguage() will return "anglais".
1655 * If the name returned cannot be localized for the default
1656 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1657 * (say, we don't have a Japanese name for Croatian),
1658 * this function falls back on the English name, and uses the ISO code as a last-resort
1659 * value. If the locale doesn't specify a language, this function returns the empty string.
1660 *
1661 * @return The name of the display language.
1662 */
1663 public final String getDisplayLanguage() {
1664 return getDisplayLanguage(getDefault(Category.DISPLAY));
1665 }
1666
1667 /**
1668 * Returns a name for the locale's language that is appropriate for display to the
1669 * user.
1670 * If possible, the name returned will be localized according to inLocale.
1671 * For example, if the locale is fr_FR and inLocale
1672 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1673 * inLocale is fr_FR, getDisplayLanguage() will return "anglais".
1674 * If the name returned cannot be localized according to inLocale,
1675 * (say, we don't have a Japanese name for Croatian),
1676 * this function falls back on the English name, and finally
1677 * on the ISO code as a last-resort value. If the locale doesn't specify a language,
1678 * this function returns the empty string.
1679 *
1680 * @param inLocale The locale for which to retrieve the display language.
1681 * @return The name of the display language appropriate to the given locale.
1682 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1683 */
1684 public String getDisplayLanguage(Locale inLocale) {
1685 return getDisplayString(baseLocale.getLanguage(), inLocale, DISPLAY_LANGUAGE);
1686 }
1687
1688 /**
1689 * Returns a name for the locale's script that is appropriate for display to
1690 * the user. If possible, the name will be localized for the default
1691 * {@link Locale.Category#DISPLAY DISPLAY} locale. Returns
1692 * the empty string if this locale doesn't specify a script code.
1693 *
1694 * @return the display name of the script code for the current default
1695 * {@link Locale.Category#DISPLAY DISPLAY} locale
1696 * @since 1.7
1697 */
1698 public String getDisplayScript() {
1699 return getDisplayScript(getDefault(Category.DISPLAY));
1700 }
1701
1702 /**
1703 * Returns a name for the locale's script that is appropriate
1704 * for display to the user. If possible, the name will be
1705 * localized for the given locale. Returns the empty string if
1706 * this locale doesn't specify a script code.
1707 *
1708 * @param inLocale The locale for which to retrieve the display script.
1709 * @return the display name of the script code for the current default
1710 * {@link Locale.Category#DISPLAY DISPLAY} locale
1711 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1712 * @since 1.7
1713 */
1714 public String getDisplayScript(Locale inLocale) {
1715 return getDisplayString(baseLocale.getScript(), inLocale, DISPLAY_SCRIPT);
1716 }
1717
1718 /**
1719 * Returns a name for the locale's country that is appropriate for display to the
1720 * user.
1721 * If possible, the name returned will be localized for the default
1722 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1723 * For example, if the locale is fr_FR and the default
1724 * {@link Locale.Category#DISPLAY DISPLAY} locale
1725 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1726 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1727 * getDisplayCountry() will return "Etats-Unis".
1728 * If the name returned cannot be localized for the default
1729 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1730 * (say, we don't have a Japanese name for Croatia),
1731 * this function falls back on the English name, and uses the ISO code as a last-resort
1732 * value. If the locale doesn't specify a country, this function returns the empty string.
1733 *
1734 * @return The name of the country appropriate to the locale.
1735 */
1736 public final String getDisplayCountry() {
1737 return getDisplayCountry(getDefault(Category.DISPLAY));
1738 }
1739
1740 /**
1741 * Returns a name for the locale's country that is appropriate for display to the
1742 * user.
1743 * If possible, the name returned will be localized according to inLocale.
1744 * For example, if the locale is fr_FR and inLocale
1745 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1746 * inLocale is fr_FR, getDisplayCountry() will return "Etats-Unis".
1747 * If the name returned cannot be localized according to inLocale.
1748 * (say, we don't have a Japanese name for Croatia),
1749 * this function falls back on the English name, and finally
1750 * on the ISO code as a last-resort value. If the locale doesn't specify a country,
1751 * this function returns the empty string.
1752 *
1753 * @param inLocale The locale for which to retrieve the display country.
1754 * @return The name of the country appropriate to the given locale.
1755 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1756 */
1757 public String getDisplayCountry(Locale inLocale) {
1758 return getDisplayString(baseLocale.getRegion(), inLocale, DISPLAY_COUNTRY);
1759 }
1760
1761 private String getDisplayString(String code, Locale inLocale, int type) {
1762 if (code.length() == 0) {
1763 return "";
1764 }
1765
1766 if (inLocale == null) {
1767 throw new NullPointerException();
1768 }
1769
1770 LocaleServiceProviderPool pool =
1771 LocaleServiceProviderPool.getPool(LocaleNameProvider.class);
1772 String key = (type == DISPLAY_VARIANT ? "%%"+code : code);
1773 String result = pool.getLocalizedObject(
1774 LocaleNameGetter.INSTANCE,
1775 inLocale, key, type, code);
1776 if (result != null) {
1777 return result;
1778 }
1779
1780 return code;
1781 }
1782
1783 /**
1784 * Returns a name for the locale's variant code that is appropriate for display to the
1785 * user. If possible, the name will be localized for the default
1786 * {@link Locale.Category#DISPLAY DISPLAY} locale. If the locale
1787 * doesn't specify a variant code, this function returns the empty string.
1788 *
1789 * @return The name of the display variant code appropriate to the locale.
1790 */
1791 public final String getDisplayVariant() {
1792 return getDisplayVariant(getDefault(Category.DISPLAY));
1793 }
1794
1795 /**
1796 * Returns a name for the locale's variant code that is appropriate for display to the
1797 * user. If possible, the name will be localized for inLocale. If the locale
1798 * doesn't specify a variant code, this function returns the empty string.
1799 *
1800 * @param inLocale The locale for which to retrieve the display variant code.
1801 * @return The name of the display variant code appropriate to the given locale.
1802 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1803 */
1804 public String getDisplayVariant(Locale inLocale) {
1805 if (baseLocale.getVariant().length() == 0)
1806 return "";
1807
1808 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale);
1809
1810 String names[] = getDisplayVariantArray(inLocale);
1811
1812 // Get the localized patterns for formatting a list, and use
1813 // them to format the list.
1814 return formatList(names,
1815 lr.getLocaleName("ListPattern"),
1816 lr.getLocaleName("ListCompositionPattern"));
1817 }
1818
1819 /**
1820 * Returns a name for the locale that is appropriate for display to the
1821 * user. This will be the values returned by getDisplayLanguage(),
1822 * getDisplayScript(), getDisplayCountry(), and getDisplayVariant() assembled
1823 * into a single string. The the non-empty values are used in order,
1824 * with the second and subsequent names in parentheses. For example:
1825 * <blockquote>
1826 * language (script, country, variant)<br>
1827 * language (country)<br>
1828 * language (variant)<br>
1829 * script (country)<br>
1830 * country<br>
1831 * </blockquote>
1832 * depending on which fields are specified in the locale. If the
1833 * language, script, country, and variant fields are all empty,
1834 * this function returns the empty string.
1835 *
1836 * @return The name of the locale appropriate to display.
1837 */
1838 public final String getDisplayName() {
1839 return getDisplayName(getDefault(Category.DISPLAY));
1840 }
1841
1842 /**
1843 * Returns a name for the locale that is appropriate for display
1844 * to the user. This will be the values returned by
1845 * getDisplayLanguage(), getDisplayScript(),getDisplayCountry(),
1846 * and getDisplayVariant() assembled into a single string.
1847 * The non-empty values are used in order,
1848 * with the second and subsequent names in parentheses. For example:
1849 * <blockquote>
1850 * language (script, country, variant)<br>
1851 * language (country)<br>
1852 * language (variant)<br>
1853 * script (country)<br>
1854 * country<br>
1855 * </blockquote>
1856 * depending on which fields are specified in the locale. If the
1857 * language, script, country, and variant fields are all empty,
1858 * this function returns the empty string.
1859 *
1860 * @param inLocale The locale for which to retrieve the display name.
1861 * @return The name of the locale appropriate to display.
1862 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1863 */
1864 public String getDisplayName(Locale inLocale) {
1865 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale);
1866
1867 String languageName = getDisplayLanguage(inLocale);
1868 String scriptName = getDisplayScript(inLocale);
1869 String countryName = getDisplayCountry(inLocale);
1870 String[] variantNames = getDisplayVariantArray(inLocale);
1871
1872 // Get the localized patterns for formatting a display name.
1873 String displayNamePattern = lr.getLocaleName("DisplayNamePattern");
1874 String listPattern = lr.getLocaleName("ListPattern");
1875 String listCompositionPattern = lr.getLocaleName("ListCompositionPattern");
1876
1877 // The display name consists of a main name, followed by qualifiers.
1878 // Typically, the format is "MainName (Qualifier, Qualifier)" but this
1879 // depends on what pattern is stored in the display locale.
1880 String mainName = null;
1881 String[] qualifierNames = null;
1882
1883 // The main name is the language, or if there is no language, the script,
1884 // then if no script, the country. If there is no language/script/country
1885 // (an anomalous situation) then the display name is simply the variant's
1886 // display name.
1887 if (languageName.length() == 0 && scriptName.length() == 0 && countryName.length() == 0) {
1888 if (variantNames.length == 0) {
1889 return "";
1890 } else {
1891 return formatList(variantNames, listPattern, listCompositionPattern);
1892 }
1893 }
1894 ArrayList<String> names = new ArrayList<>(4);
1895 if (languageName.length() != 0) {
1896 names.add(languageName);
1897 }
1898 if (scriptName.length() != 0) {
1899 names.add(scriptName);
1900 }
1901 if (countryName.length() != 0) {
1902 names.add(countryName);
1903 }
1904 if (variantNames.length != 0) {
1905 names.addAll(Arrays.asList(variantNames));
1906 }
1907
1908 // The first one in the main name
1909 mainName = names.get(0);
1910
1911 // Others are qualifiers
1912 int numNames = names.size();
1913 qualifierNames = (numNames > 1) ?
1914 names.subList(1, numNames).toArray(new String[numNames - 1]) : new String[0];
1915
1916 // Create an array whose first element is the number of remaining
1917 // elements. This serves as a selector into a ChoiceFormat pattern from
1918 // the resource. The second and third elements are the main name and
1919 // the qualifier; if there are no qualifiers, the third element is
1920 // unused by the format pattern.
1921 Object[] displayNames = {
1922 qualifierNames.length != 0 ? 2 : 1,
1923 mainName,
1924 // We could also just call formatList() and have it handle the empty
1925 // list case, but this is more efficient, and we want it to be
1926 // efficient since all the language-only locales will not have any
1927 // qualifiers.
1928 qualifierNames.length != 0 ? formatList(qualifierNames, listPattern, listCompositionPattern) : null
1929 };
1930
1931 if (displayNamePattern != null) {
1932 return new MessageFormat(displayNamePattern).format(displayNames);
1933 }
1934 else {
1935 // If we cannot get the message format pattern, then we use a simple
1936 // hard-coded pattern. This should not occur in practice unless the
1937 // installation is missing some core files (FormatData etc.).
1938 StringBuilder result = new StringBuilder();
1939 result.append((String)displayNames[1]);
1940 if (displayNames.length > 2) {
1941 result.append(" (");
1942 result.append((String)displayNames[2]);
1943 result.append(')');
1944 }
1945 return result.toString();
1946 }
1947 }
1948
1949 /**
1950 * Overrides Cloneable.
1951 */
1952 @Override
1953 public Object clone()
1954 {
1955 try {
1956 Locale that = (Locale)super.clone();
1957 return that;
1958 } catch (CloneNotSupportedException e) {
1959 throw new InternalError(e);
1960 }
1961 }
1962
1963 /**
1964 * Override hashCode.
1965 * Since Locales are often used in hashtables, caches the value
1966 * for speed.
1967 */
1968 @Override
1969 public int hashCode() {
1970 int hc = hashCodeValue;
1971 if (hc == 0) {
1972 hc = baseLocale.hashCode();
1973 if (localeExtensions != null) {
1974 hc ^= localeExtensions.hashCode();
1975 }
1976 hashCodeValue = hc;
1977 }
1978 return hc;
1979 }
1980
1981 // Overrides
1982
1983 /**
1984 * Returns true if this Locale is equal to another object. A Locale is
1985 * deemed equal to another Locale with identical language, script, country,
1986 * variant and extensions, and unequal to all other objects.
1987 *
1988 * @return true if this Locale is equal to the specified object.
1989 */
1990 @Override
1991 public boolean equals(Object obj) {
1992 if (this == obj) // quick check
1993 return true;
1994 if (!(obj instanceof Locale))
1995 return false;
1996 BaseLocale otherBase = ((Locale)obj).baseLocale;
1997 if (!baseLocale.equals(otherBase)) {
1998 return false;
1999 }
2000 if (localeExtensions == null) {
2001 return ((Locale)obj).localeExtensions == null;
2002 }
2003 return localeExtensions.equals(((Locale)obj).localeExtensions);
2004 }
2005
2006 // ================= privates =====================================
2007
2008 private transient BaseLocale baseLocale;
2009 private transient LocaleExtensions localeExtensions;
2010
2011 /**
2012 * Calculated hashcode
2013 */
2014 private transient volatile int hashCodeValue;
2015
2016 private static volatile Locale defaultLocale = initDefault();
2017 private static volatile Locale defaultDisplayLocale;
2018 private static volatile Locale defaultFormatLocale;
2019
2020 private transient volatile String languageTag;
2021
2022 /**
2023 * Return an array of the display names of the variant.
2024 * @param bundle the ResourceBundle to use to get the display names
2025 * @return an array of display names, possible of zero length.
2026 */
2027 private String[] getDisplayVariantArray(Locale inLocale) {
2028 // Split the variant name into tokens separated by '_'.
2029 StringTokenizer tokenizer = new StringTokenizer(baseLocale.getVariant(), "_");
2030 String[] names = new String[tokenizer.countTokens()];
2031
2032 // For each variant token, lookup the display name. If
2033 // not found, use the variant name itself.
2034 for (int i=0; i<names.length; ++i) {
2035 names[i] = getDisplayString(tokenizer.nextToken(),
2036 inLocale, DISPLAY_VARIANT);
2037 }
2038
2039 return names;
2040 }
2041
2042 /**
2043 * Format a list using given pattern strings.
2044 * If either of the patterns is null, then a the list is
2045 * formatted by concatenation with the delimiter ','.
2046 * @param stringList the list of strings to be formatted.
2047 * @param listPattern should create a MessageFormat taking 0-3 arguments
2048 * and formatting them into a list.
2049 * @param listCompositionPattern should take 2 arguments
2050 * and is used by composeList.
2051 * @return a string representing the list.
2052 */
2053 private static String formatList(String[] stringList, String listPattern, String listCompositionPattern) {
2054 // If we have no list patterns, compose the list in a simple,
2055 // non-localized way.
2056 if (listPattern == null || listCompositionPattern == null) {
2057 StringJoiner sj = new StringJoiner(",");
2058 for (int i = 0; i < stringList.length; ++i) {
2059 sj.add(stringList[i]);
2060 }
2061 return sj.toString();
2062 }
2063
2064 // Compose the list down to three elements if necessary
2065 if (stringList.length > 3) {
2066 MessageFormat format = new MessageFormat(listCompositionPattern);
2067 stringList = composeList(format, stringList);
2068 }
2069
2070 // Rebuild the argument list with the list length as the first element
2071 Object[] args = new Object[stringList.length + 1];
2072 System.arraycopy(stringList, 0, args, 1, stringList.length);
2073 args[0] = stringList.length;
2074
2075 // Format it using the pattern in the resource
2076 MessageFormat format = new MessageFormat(listPattern);
2077 return format.format(args);
2078 }
2079
2080 /**
2081 * Given a list of strings, return a list shortened to three elements.
2082 * Shorten it by applying the given format to the first two elements
2083 * recursively.
2084 * @param format a format which takes two arguments
2085 * @param list a list of strings
2086 * @return if the list is three elements or shorter, the same list;
2087 * otherwise, a new list of three elements.
2088 */
2089 private static String[] composeList(MessageFormat format, String[] list) {
2090 if (list.length <= 3) return list;
2091
2092 // Use the given format to compose the first two elements into one
2093 String[] listItems = { list[0], list[1] };
2094 String newItem = format.format(listItems);
2095
2096 // Form a new list one element shorter
2097 String[] newList = new String[list.length-1];
2098 System.arraycopy(list, 2, newList, 1, newList.length-1);
2099 newList[0] = newItem;
2100
2101 // Recurse
2102 return composeList(format, newList);
2103 }
2104
2105 // Duplicate of sun.util.locale.UnicodeLocaleExtension.isKey in order to
2106 // avoid its class loading.
2107 private static boolean isUnicodeExtensionKey(String s) {
2108 // 2alphanum
2109 return (s.length() == 2) && LocaleUtils.isAlphaNumericString(s);
2110 }
2111
2112 /**
2113 * @serialField language String
2114 * language subtag in lower case. (See <a href="java/util/Locale.html#getLanguage()">getLanguage()</a>)
2115 * @serialField country String
2116 * country subtag in upper case. (See <a href="java/util/Locale.html#getCountry()">getCountry()</a>)
2117 * @serialField variant String
2118 * variant subtags separated by LOWLINE characters. (See <a href="java/util/Locale.html#getVariant()">getVariant()</a>)
2119 * @serialField hashcode int
2120 * deprecated, for forward compatibility only
2121 * @serialField script String
2122 * script subtag in title case (See <a href="java/util/Locale.html#getScript()">getScript()</a>)
2123 * @serialField extensions String
2124 * canonical representation of extensions, that is,
2125 * BCP47 extensions in alphabetical order followed by
2126 * BCP47 private use subtags, all in lower case letters
2127 * separated by HYPHEN-MINUS characters.
2128 * (See <a href="java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>,
2129 * <a href="java/util/Locale.html#getExtension(char)">getExtension(char)</a>)
2130 */
2131 private static final ObjectStreamField[] serialPersistentFields = {
2132 new ObjectStreamField("language", String.class),
2133 new ObjectStreamField("country", String.class),
2134 new ObjectStreamField("variant", String.class),
2135 new ObjectStreamField("hashcode", int.class),
2136 new ObjectStreamField("script", String.class),
2137 new ObjectStreamField("extensions", String.class),
2138 };
2139
2140 /**
2141 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>.
2142 * @param out the <code>ObjectOutputStream</code> to write
2143 * @throws IOException
2144 * @since 1.7
2145 */
2146 private void writeObject(ObjectOutputStream out) throws IOException {
2147 ObjectOutputStream.PutField fields = out.putFields();
2148 fields.put("language", baseLocale.getLanguage());
2149 fields.put("script", baseLocale.getScript());
2150 fields.put("country", baseLocale.getRegion());
2151 fields.put("variant", baseLocale.getVariant());
2152 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID());
2153 fields.put("hashcode", -1); // place holder just for backward support
2154 out.writeFields();
2155 }
2156
2157 /**
2158 * Deserializes this <code>Locale</code>.
2159 * @param in the <code>ObjectInputStream</code> to read
2160 * @throws IOException
2161 * @throws ClassNotFoundException
2162 * @throws IllformedLocaleException
2163 * @since 1.7
2164 */
2165 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException {
2166 ObjectInputStream.GetField fields = in.readFields();
2167 String language = (String)fields.get("language", "");
2168 String script = (String)fields.get("script", "");
2169 String country = (String)fields.get("country", "");
2170 String variant = (String)fields.get("variant", "");
2171 String extStr = (String)fields.get("extensions", "");
2172 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant);
2173 if (extStr.length() > 0) {
2174 try {
2175 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
2176 bldr.setExtensions(extStr);
2177 localeExtensions = bldr.getLocaleExtensions();
2178 } catch (LocaleSyntaxException e) {
2179 throw new IllformedLocaleException(e.getMessage());
2180 }
2181 } else {
2182 localeExtensions = null;
2183 }
2184 }
2185
2186 /**
2187 * Returns a cached <code>Locale</code> instance equivalent to
2188 * the deserialized <code>Locale</code>. When serialized
2189 * language, country and variant fields read from the object data stream
2190 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions
2191 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code>
2192 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script
2193 * type is "thai"). See <a href="Locale.html#special_cases_constructor">Special Cases</a>
2194 * for more information.
2195 *
2196 * @return an instance of <code>Locale</code> equivalent to
2197 * the deserialized <code>Locale</code>.
2198 * @throws java.io.ObjectStreamException
2199 */
2200 private Object readResolve() throws java.io.ObjectStreamException {
2201 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(),
2202 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions);
2203 }
2204
2205 private static volatile String[] isoLanguages;
2206
2207 private static volatile String[] isoCountries;
2208
2209 private static String convertOldISOCodes(String language) {
2210 // we accept both the old and the new ISO codes for the languages whose ISO
2211 // codes have changed, but we always store the OLD code, for backward compatibility
2212 language = LocaleUtils.toLowerString(language).intern();
2213 if (language == "he") {
2214 return "iw";
2215 } else if (language == "yi") {
2216 return "ji";
2217 } else if (language == "id") {
2218 return "in";
2219 } else {
2220 return language;
2221 }
2222 }
2223
2224 private static LocaleExtensions getCompatibilityExtensions(String language,
2225 String script,
2226 String country,
2227 String variant) {
2228 LocaleExtensions extensions = null;
2229 // Special cases for backward compatibility support
2230 if (LocaleUtils.caseIgnoreMatch(language, "ja")
2231 && script.length() == 0
2232 && LocaleUtils.caseIgnoreMatch(country, "jp")
2233 && "JP".equals(variant)) {
2234 // ja_JP_JP -> u-ca-japanese (calendar = japanese)
2235 extensions = LocaleExtensions.CALENDAR_JAPANESE;
2236 } else if (LocaleUtils.caseIgnoreMatch(language, "th")
2237 && script.length() == 0
2238 && LocaleUtils.caseIgnoreMatch(country, "th")
2239 && "TH".equals(variant)) {
2240 // th_TH_TH -> u-nu-thai (numbersystem = thai)
2241 extensions = LocaleExtensions.NUMBER_THAI;
2242 }
2243 return extensions;
2244 }
2245
2246 /**
2247 * Obtains a localized locale names from a LocaleNameProvider
2248 * implementation.
2249 */
2250 private static class LocaleNameGetter
2251 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> {
2252 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter();
2253
2254 @Override
2255 public String getObject(LocaleNameProvider localeNameProvider,
2256 Locale locale,
2257 String key,
2258 Object... params) {
2259 assert params.length == 2;
2260 int type = (Integer)params[0];
2261 String code = (String)params[1];
2262
2263 switch(type) {
2264 case DISPLAY_LANGUAGE:
2265 return localeNameProvider.getDisplayLanguage(code, locale);
2266 case DISPLAY_COUNTRY:
2267 return localeNameProvider.getDisplayCountry(code, locale);
2268 case DISPLAY_VARIANT:
2269 return localeNameProvider.getDisplayVariant(code, locale);
2270 case DISPLAY_SCRIPT:
2271 return localeNameProvider.getDisplayScript(code, locale);
2272 default:
2273 assert false; // shouldn't happen
2274 }
2275
2276 return null;
2277 }
2278 }
2279
2280 /**
2281 * Enum for locale categories. These locale categories are used to get/set
2282 * the default locale for the specific functionality represented by the
2283 * category.
2284 *
2285 * @see #getDefault(Locale.Category)
2286 * @see #setDefault(Locale.Category, Locale)
2287 * @since 1.7
2288 */
2289 public enum Category {
2290
2291 /**
2292 * Category used to represent the default locale for
2293 * displaying user interfaces.
2294 */
2295 DISPLAY("user.language.display",
2296 "user.script.display",
2297 "user.country.display",
2298 "user.variant.display"),
2299
2300 /**
2301 * Category used to represent the default locale for
2302 * formatting dates, numbers, and/or currencies.
2303 */
2304 FORMAT("user.language.format",
2305 "user.script.format",
2306 "user.country.format",
2307 "user.variant.format");
2308
2309 Category(String languageKey, String scriptKey, String countryKey, String variantKey) {
2310 this.languageKey = languageKey;
2311 this.scriptKey = scriptKey;
2312 this.countryKey = countryKey;
2313 this.variantKey = variantKey;
2314 }
2315
2316 final String languageKey;
2317 final String scriptKey;
2318 final String countryKey;
2319 final String variantKey;
2320 }
2321
2322 /**
2323 * <code>Builder</code> is used to build instances of <code>Locale</code>
2324 * from values configured by the setters. Unlike the <code>Locale</code>
2325 * constructors, the <code>Builder</code> checks if a value configured by a
2326 * setter satisfies the syntax requirements defined by the <code>Locale</code>
2327 * class. A <code>Locale</code> object created by a <code>Builder</code> is
2328 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag
2329 * without losing information.
2330 *
2331 * <p><b>Note:</b> The <code>Locale</code> class does not provide any
2332 * syntactic restrictions on variant, while BCP 47 requires each variant
2333 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3
2334 * alphanumerics. The method <code>setVariant</code> throws
2335 * <code>IllformedLocaleException</code> for a variant that does not satisfy
2336 * this restriction. If it is necessary to support such a variant, use a
2337 * Locale constructor. However, keep in mind that a <code>Locale</code>
2338 * object created this way might lose the variant information when
2339 * transformed to a BCP 47 language tag.
2340 *
2341 * <p>The following example shows how to create a <code>Locale</code> object
2342 * with the <code>Builder</code>.
2343 * <blockquote>
2344 * <pre>
2345 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build();
2346 * </pre>
2347 * </blockquote>
2348 *
2349 * <p>Builders can be reused; <code>clear()</code> resets all
2350 * fields to their default values.
2351 *
2352 * @see Locale#forLanguageTag
2353 * @since 1.7
2354 */
2355 public static final class Builder {
2356 private final InternalLocaleBuilder localeBuilder;
2357
2358 /**
2359 * Constructs an empty Builder. The default value of all
2360 * fields, extensions, and private use information is the
2361 * empty string.
2362 */
2363 public Builder() {
2364 localeBuilder = new InternalLocaleBuilder();
2365 }
2366
2367 /**
2368 * Resets the <code>Builder</code> to match the provided
2369 * <code>locale</code>. Existing state is discarded.
2370 *
2371 * <p>All fields of the locale must be well-formed, see {@link Locale}.
2372 *
2373 * <p>Locales with any ill-formed fields cause
2374 * <code>IllformedLocaleException</code> to be thrown, except for the
2375 * following three cases which are accepted for compatibility
2376 * reasons:<ul>
2377 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese"
2378 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai"
2379 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul>
2380 *
2381 * @param locale the locale
2382 * @return This builder.
2383 * @throws IllformedLocaleException if <code>locale</code> has
2384 * any ill-formed fields.
2385 * @throws NullPointerException if <code>locale</code> is null.
2386 */
2387 public Builder setLocale(Locale locale) {
2388 try {
2389 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions);
2390 } catch (LocaleSyntaxException e) {
2391 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2392 }
2393 return this;
2394 }
2395
2396 /**
2397 * Resets the Builder to match the provided IETF BCP 47
2398 * language tag. Discards the existing state. Null and the
2399 * empty string cause the builder to be reset, like {@link
2400 * #clear}. Grandfathered tags (see {@link
2401 * Locale#forLanguageTag}) are converted to their canonical
2402 * form before being processed. Otherwise, the language tag
2403 * must be well-formed (see {@link Locale}) or an exception is
2404 * thrown (unlike <code>Locale.forLanguageTag</code>, which
2405 * just discards ill-formed and following portions of the
2406 * tag).
2407 *
2408 * @param languageTag the language tag
2409 * @return This builder.
2410 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed
2411 * @see Locale#forLanguageTag(String)
2412 */
2413 public Builder setLanguageTag(String languageTag) {
2414 ParseStatus sts = new ParseStatus();
2415 LanguageTag tag = LanguageTag.parse(languageTag, sts);
2416 if (sts.isError()) {
2417 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex());
2418 }
2419 localeBuilder.setLanguageTag(tag);
2420 return this;
2421 }
2422
2423 /**
2424 * Sets the language. If <code>language</code> is the empty string or
2425 * null, the language in this <code>Builder</code> is removed. Otherwise,
2426 * the language must be <a href="./Locale.html#def_language">well-formed</a>
2427 * or an exception is thrown.
2428 *
2429 * <p>The typical language value is a two or three-letter language
2430 * code as defined in ISO639.
2431 *
2432 * @param language the language
2433 * @return This builder.
2434 * @throws IllformedLocaleException if <code>language</code> is ill-formed
2435 */
2436 public Builder setLanguage(String language) {
2437 try {
2438 localeBuilder.setLanguage(language);
2439 } catch (LocaleSyntaxException e) {
2440 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2441 }
2442 return this;
2443 }
2444
2445 /**
2446 * Sets the script. If <code>script</code> is null or the empty string,
2447 * the script in this <code>Builder</code> is removed.
2448 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an
2449 * exception is thrown.
2450 *
2451 * <p>The typical script value is a four-letter script code as defined by ISO 15924.
2452 *
2453 * @param script the script
2454 * @return This builder.
2455 * @throws IllformedLocaleException if <code>script</code> is ill-formed
2456 */
2457 public Builder setScript(String script) {
2458 try {
2459 localeBuilder.setScript(script);
2460 } catch (LocaleSyntaxException e) {
2461 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2462 }
2463 return this;
2464 }
2465
2466 /**
2467 * Sets the region. If region is null or the empty string, the region
2468 * in this <code>Builder</code> is removed. Otherwise,
2469 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an
2470 * exception is thrown.
2471 *
2472 * <p>The typical region value is a two-letter ISO 3166 code or a
2473 * three-digit UN M.49 area code.
2474 *
2475 * <p>The country value in the <code>Locale</code> created by the
2476 * <code>Builder</code> is always normalized to upper case.
2477 *
2478 * @param region the region
2479 * @return This builder.
2480 * @throws IllformedLocaleException if <code>region</code> is ill-formed
2481 */
2482 public Builder setRegion(String region) {
2483 try {
2484 localeBuilder.setRegion(region);
2485 } catch (LocaleSyntaxException e) {
2486 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2487 }
2488 return this;
2489 }
2490
2491 /**
2492 * Sets the variant. If variant is null or the empty string, the
2493 * variant in this <code>Builder</code> is removed. Otherwise, it
2494 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a>
2495 * subtags, or an exception is thrown.
2496 *
2497 * <p><b>Note:</b> This method checks if <code>variant</code>
2498 * satisfies the IETF BCP 47 variant subtag's syntax requirements,
2499 * and normalizes the value to lowercase letters. However,
2500 * the <code>Locale</code> class does not impose any syntactic
2501 * restriction on variant, and the variant value in
2502 * <code>Locale</code> is case sensitive. To set such a variant,
2503 * use a Locale constructor.
2504 *
2505 * @param variant the variant
2506 * @return This builder.
2507 * @throws IllformedLocaleException if <code>variant</code> is ill-formed
2508 */
2509 public Builder setVariant(String variant) {
2510 try {
2511 localeBuilder.setVariant(variant);
2512 } catch (LocaleSyntaxException e) {
2513 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2514 }
2515 return this;
2516 }
2517
2518 /**
2519 * Sets the extension for the given key. If the value is null or the
2520 * empty string, the extension is removed. Otherwise, the extension
2521 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception
2522 * is thrown.
2523 *
2524 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION
2525 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension.
2526 * Setting a value for this key replaces any existing Unicode locale key/type
2527 * pairs with those defined in the extension.
2528 *
2529 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION
2530 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be
2531 * well-formed, the value for this key needs only to have subtags of one to
2532 * eight alphanumeric characters, not two to eight as in the general case.
2533 *
2534 * @param key the extension key
2535 * @param value the extension value
2536 * @return This builder.
2537 * @throws IllformedLocaleException if <code>key</code> is illegal
2538 * or <code>value</code> is ill-formed
2539 * @see #setUnicodeLocaleKeyword(String, String)
2540 */
2541 public Builder setExtension(char key, String value) {
2542 try {
2543 localeBuilder.setExtension(key, value);
2544 } catch (LocaleSyntaxException e) {
2545 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2546 }
2547 return this;
2548 }
2549
2550 /**
2551 * Sets the Unicode locale keyword type for the given key. If the type
2552 * is null, the Unicode keyword is removed. Otherwise, the key must be
2553 * non-null and both key and type must be <a
2554 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2555 * is thrown.
2556 *
2557 * <p>Keys and types are converted to lower case.
2558 *
2559 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension}
2560 * replaces all Unicode locale keywords with those defined in the
2561 * extension.
2562 *
2563 * @param key the Unicode locale key
2564 * @param type the Unicode locale type
2565 * @return This builder.
2566 * @throws IllformedLocaleException if <code>key</code> or <code>type</code>
2567 * is ill-formed
2568 * @throws NullPointerException if <code>key</code> is null
2569 * @see #setExtension(char, String)
2570 */
2571 public Builder setUnicodeLocaleKeyword(String key, String type) {
2572 try {
2573 localeBuilder.setUnicodeLocaleKeyword(key, type);
2574 } catch (LocaleSyntaxException e) {
2575 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2576 }
2577 return this;
2578 }
2579
2580 /**
2581 * Adds a unicode locale attribute, if not already present, otherwise
2582 * has no effect. The attribute must not be null and must be <a
2583 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2584 * is thrown.
2585 *
2586 * @param attribute the attribute
2587 * @return This builder.
2588 * @throws NullPointerException if <code>attribute</code> is null
2589 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2590 * @see #setExtension(char, String)
2591 */
2592 public Builder addUnicodeLocaleAttribute(String attribute) {
2593 try {
2594 localeBuilder.addUnicodeLocaleAttribute(attribute);
2595 } catch (LocaleSyntaxException e) {
2596 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2597 }
2598 return this;
2599 }
2600
2601 /**
2602 * Removes a unicode locale attribute, if present, otherwise has no
2603 * effect. The attribute must not be null and must be <a
2604 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2605 * is thrown.
2606 *
2607 * <p>Attribute comparison for removal is case-insensitive.
2608 *
2609 * @param attribute the attribute
2610 * @return This builder.
2611 * @throws NullPointerException if <code>attribute</code> is null
2612 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2613 * @see #setExtension(char, String)
2614 */
2615 public Builder removeUnicodeLocaleAttribute(String attribute) {
2616 try {
2617 localeBuilder.removeUnicodeLocaleAttribute(attribute);
2618 } catch (LocaleSyntaxException e) {
2619 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2620 }
2621 return this;
2622 }
2623
2624 /**
2625 * Resets the builder to its initial, empty state.
2626 *
2627 * @return This builder.
2628 */
2629 public Builder clear() {
2630 localeBuilder.clear();
2631 return this;
2632 }
2633
2634 /**
2635 * Resets the extensions to their initial, empty state.
2636 * Language, script, region and variant are unchanged.
2637 *
2638 * @return This builder.
2639 * @see #setExtension(char, String)
2640 */
2641 public Builder clearExtensions() {
2642 localeBuilder.clearExtensions();
2643 return this;
2644 }
2645
2646 /**
2647 * Returns an instance of <code>Locale</code> created from the fields set
2648 * on this builder.
2649 *
2650 * <p>This applies the conversions listed in {@link Locale#forLanguageTag}
2651 * when constructing a Locale. (Grandfathered tags are handled in
2652 * {@link #setLanguageTag}.)
2653 *
2654 * @return A Locale.
2655 */
2656 public Locale build() {
2657 BaseLocale baseloc = localeBuilder.getBaseLocale();
2658 LocaleExtensions extensions = localeBuilder.getLocaleExtensions();
2659 if (extensions == null && baseloc.getVariant().length() > 0) {
2660 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(),
2661 baseloc.getRegion(), baseloc.getVariant());
2662 }
2663 return Locale.getInstance(baseloc, extensions);
2664 }
2665 }
2666
2667 /**
2668 * This enum provides constants to select a filtering mode for locale
2669 * matching. Refer to <a href="http://tools.ietf.org/html/rfc4647">RFC 4647
2670 * Matching of Language Tags</a> for details.
2671 *
2672 * <p>As an example, think of two Language Priority Lists each of which
2673 * includes only one language range and a set of following language tags:
2674 *
2675 * <pre>
2676 * de (German)
2677 * de-DE (German, Germany)
2678 * de-Deva (German, in Devanagari script)
2679 * de-Deva-DE (German, in Devanagari script, Germany)
2680 * de-DE-1996 (German, Germany, orthography of 1996)
2681 * de-Latn-DE (German, in Latin script, Germany)
2682 * de-Latn-DE-1996 (German, in Latin script, Germany, orthography of 1996)
2683 * </pre>
2684 *
2685 * The filtering method will behave as follows:
2686 *
2687 * <table cellpadding=2 summary="Filtering method behavior">
2688 * <tr>
2689 * <th>Filtering Mode</th>
2690 * <th>Language Priority List: {@code "de-DE"}</th>
2691 * <th>Language Priority List: {@code "de-*-DE"}</th>
2692 * </tr>
2693 * <tr>
2694 * <td valign=top>
2695 * {@link FilteringMode#AUTOSELECT_FILTERING AUTOSELECT_FILTERING}
2696 * </td>
2697 * <td valign=top>
2698 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2699 * {@code "de-DE-1996"}.
2700 * </td>
2701 * <td valign=top>
2702 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2703 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2704 * {@code "de-Latn-DE-1996"}.
2705 * </td>
2706 * </tr>
2707 * <tr>
2708 * <td valign=top>
2709 * {@link FilteringMode#EXTENDED_FILTERING EXTENDED_FILTERING}
2710 * </td>
2711 * <td valign=top>
2712 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2713 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2714 * {@code "de-Latn-DE-1996"}.
2715 * </td>
2716 * <td valign=top>Same as above.</td>
2717 * </tr>
2718 * <tr>
2719 * <td valign=top>
2720 * {@link FilteringMode#IGNORE_EXTENDED_RANGES IGNORE_EXTENDED_RANGES}
2721 * </td>
2722 * <td valign=top>
2723 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2724 * {@code "de-DE-1996"}.
2725 * </td>
2726 * <td valign=top>
2727 * Performs <em>basic</em> filtering and returns {@code null} because
2728 * nothing matches.
2729 * </td>
2730 * </tr>
2731 * <tr>
2732 * <td valign=top>
2733 * {@link FilteringMode#MAP_EXTENDED_RANGES MAP_EXTENDED_RANGES}
2734 * </td>
2735 * <td valign=top>Same as above.</td>
2736 * <td valign=top>
2737 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2738 * {@code "de-DE-1996"} because {@code "de-*-DE"} is mapped to
2739 * {@code "de-DE"}.
2740 * </td>
2741 * </tr>
2742 * <tr>
2743 * <td valign=top>
2744 * {@link FilteringMode#REJECT_EXTENDED_RANGES REJECT_EXTENDED_RANGES}
2745 * </td>
2746 * <td valign=top>Same as above.</td>
2747 * <td valign=top>
2748 * Throws {@link IllegalArgumentException} because {@code "de-*-DE"} is
2749 * not a valid basic language range.
2750 * </td>
2751 * </tr>
2752 * </table>
2753 *
2754 * @see #filter(List, Collection, FilteringMode)
2755 * @see #filterTags(List, Collection, FilteringMode)
2756 *
2757 * @since 1.8
2758 */
2759 public static enum FilteringMode {
2760 /**
2761 * Specifies automatic filtering mode based on the given Language
2762 * Priority List consisting of language ranges. If all of the ranges
2763 * are basic, basic filtering is selected. Otherwise, extended
2764 * filtering is selected.
2765 */
2766 AUTOSELECT_FILTERING,
2767
2768 /**
2769 * Specifies extended filtering.
2770 */
2771 EXTENDED_FILTERING,
2772
2773 /**
2774 * Specifies basic filtering: Note that any extended language ranges
2775 * included in the given Language Priority List are ignored.
2776 */
2777 IGNORE_EXTENDED_RANGES,
2778
2779 /**
2780 * Specifies basic filtering: If any extended language ranges are
2781 * included in the given Language Priority List, they are mapped to the
2782 * basic language range. Specifically, a language range starting with a
2783 * subtag {@code "*"} is treated as a language range {@code "*"}. For
2784 * example, {@code "*-US"} is treated as {@code "*"}. If {@code "*"} is
2785 * not the first subtag, {@code "*"} and extra {@code "-"} are removed.
2786 * For example, {@code "ja-*-JP"} is mapped to {@code "ja-JP"}.
2787 */
2788 MAP_EXTENDED_RANGES,
2789
2790 /**
2791 * Specifies basic filtering: If any extended language ranges are
2792 * included in the given Language Priority List, the list is rejected
2793 * and the filtering method throws {@link IllegalArgumentException}.
2794 */
2795 REJECT_EXTENDED_RANGES
2796 };
2797
2798 /**
2799 * This class expresses a <em>Language Range</em> defined in
2800 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
2801 * Language Tags</a>. A language range is an identifier which is used to
2802 * select language tag(s) meeting specific requirements by using the
2803 * mechanisms described in <a href="Locale.html#LocaleMatching">Locale
2804 * Matching</a>. A list which represents a user's preferences and consists
2805 * of language ranges is called a <em>Language Priority List</em>.
2806 *
2807 * <p>There are two types of language ranges: basic and extended. In RFC
2808 * 4647, the syntax of language ranges is expressed in
2809 * <a href="http://tools.ietf.org/html/rfc4234">ABNF</a> as follows:
2810 * <blockquote>
2811 * <pre>
2812 * basic-language-range = (1*8ALPHA *("-" 1*8alphanum)) / "*"
2813 * extended-language-range = (1*8ALPHA / "*")
2814 * *("-" (1*8alphanum / "*"))
2815 * alphanum = ALPHA / DIGIT
2816 * </pre>
2817 * </blockquote>
2818 * For example, {@code "en"} (English), {@code "ja-JP"} (Japanese, Japan),
2819 * {@code "*"} (special language range which matches any language tag) are
2820 * basic language ranges, whereas {@code "*-CH"} (any languages,
2821 * Switzerland), {@code "es-*"} (Spanish, any regions), and
2822 * {@code "zh-Hant-*"} (Traditional Chinese, any regions) are extended
2823 * language ranges.
2824 *
2825 * @see #filter
2826 * @see #filterTags
2827 * @see #lookup
2828 * @see #lookupTag
2829 *
2830 * @since 1.8
2831 */
2832 public static final class LanguageRange {
2833
2834 /**
2835 * A constant holding the maximum value of weight, 1.0, which indicates
2836 * that the language range is a good fit for the user.
2837 */
2838 public static final double MAX_WEIGHT = 1.0;
2839
2840 /**
2841 * A constant holding the minimum value of weight, 0.0, which indicates
2842 * that the language range is not a good fit for the user.
2843 */
2844 public static final double MIN_WEIGHT = 0.0;
2845
2846 private final String range;
2847 private final double weight;
2848
2849 private volatile int hash;
2850
2851 /**
2852 * Constructs a {@code LanguageRange} using the given {@code range}.
2853 * Note that no validation is done against the IANA Language Subtag
2854 * Registry at time of construction.
2855 *
2856 * <p>This is equivalent to {@code LanguageRange(range, MAX_WEIGHT)}.
2857 *
2858 * @param range a language range
2859 * @throws NullPointerException if the given {@code range} is
2860 * {@code null}
2861 * @throws IllegalArgumentException if the given {@code range} does not
2862 * comply with the syntax of the language range mentioned in RFC 4647
2863 */
2864 public LanguageRange(String range) {
2865 this(range, MAX_WEIGHT);
2866 }
2867
2868 /**
2869 * Constructs a {@code LanguageRange} using the given {@code range} and
2870 * {@code weight}. Note that no validation is done against the IANA
2871 * Language Subtag Registry at time of construction.
2872 *
2873 * @param range a language range
2874 * @param weight a weight value between {@code MIN_WEIGHT} and
2875 * {@code MAX_WEIGHT}
2876 * @throws NullPointerException if the given {@code range} is
2877 * {@code null}
2878 * @throws IllegalArgumentException if the given {@code range} does not
2879 * comply with the syntax of the language range mentioned in RFC 4647
2880 * or if the given {@code weight} is less than {@code MIN_WEIGHT}
2881 * or greater than {@code MAX_WEIGHT}
2882 */
2883 public LanguageRange(String range, double weight) {
2884 if (range == null) {
2885 throw new NullPointerException();
2886 }
2887 if (weight < MIN_WEIGHT || weight > MAX_WEIGHT) {
2888 throw new IllegalArgumentException("weight=" + weight);
2889 }
2890
2891 range = range.toLowerCase(Locale.ROOT);
2892
2893 // Do syntax check.
2894 boolean isIllFormed = false;
2895 String[] subtags = range.split("-");
2896 if (isSubtagIllFormed(subtags[0], true)
2897 || range.endsWith("-")) {
2898 isIllFormed = true;
2899 } else {
2900 for (int i = 1; i < subtags.length; i++) {
2901 if (isSubtagIllFormed(subtags[i], false)) {
2902 isIllFormed = true;
2903 break;
2904 }
2905 }
2906 }
2907 if (isIllFormed) {
2908 throw new IllegalArgumentException("range=" + range);
2909 }
2910
2911 this.range = range;
2912 this.weight = weight;
2913 }
2914
2915 private static boolean isSubtagIllFormed(String subtag,
2916 boolean isFirstSubtag) {
2917 if (subtag.equals("") || subtag.length() > 8) {
2918 return true;
2919 } else if (subtag.equals("*")) {
2920 return false;
2921 }
2922 char[] charArray = subtag.toCharArray();
2923 if (isFirstSubtag) { // ALPHA
2924 for (char c : charArray) {
2925 if (c < 'a' || c > 'z') {
2926 return true;
2927 }
2928 }
2929 } else { // ALPHA / DIGIT
2930 for (char c : charArray) {
2931 if (c < '0' || (c > '9' && c < 'a') || c > 'z') {
2932 return true;
2933 }
2934 }
2935 }
2936 return false;
2937 }
2938
2939 /**
2940 * Returns the language range of this {@code LanguageRange}.
2941 *
2942 * @return the language range.
2943 */
2944 public String getRange() {
2945 return range;
2946 }
2947
2948 /**
2949 * Returns the weight of this {@code LanguageRange}.
2950 *
2951 * @return the weight value.
2952 */
2953 public double getWeight() {
2954 return weight;
2955 }
2956
2957 /**
2958 * Parses the given {@code ranges} to generate a Language Priority List.
2959 *
2960 * <p>This method performs a syntactic check for each language range in
2961 * the given {@code ranges} but doesn't do validation using the IANA
2962 * Language Subtag Registry.
2963 *
2964 * <p>The {@code ranges} to be given can take one of the following
2965 * forms:
2966 *
2967 * <pre>
2968 * "Accept-Language: ja,en;q=0.4" (weighted list with Accept-Language prefix)
2969 * "ja,en;q=0.4" (weighted list)
2970 * "ja,en" (prioritized list)
2971 * </pre>
2972 *
2973 * In a weighted list, each language range is given a weight value.
2974 * The weight value is identical to the "quality value" in
2975 * <a href="http://tools.ietf.org/html/rfc2616">RFC 2616</a>, and it
2976 * expresses how much the user prefers the language. A weight value is
2977 * specified after a corresponding language range followed by
2978 * {@code ";q="}, and the default weight value is {@code MAX_WEIGHT}
2979 * when it is omitted.
2980 *
2981 * <p>Unlike a weighted list, language ranges in a prioritized list
2982 * are sorted in the descending order based on its priority. The first
2983 * language range has the highest priority and meets the user's
2984 * preference most.
2985 *
2986 * <p>In either case, language ranges are sorted in descending order in
2987 * the Language Priority List based on priority or weight. If a
2988 * language range appears in the given {@code ranges} more than once,
2989 * only the first one is included on the Language Priority List.
2990 *
2991 * <p>The returned list consists of language ranges from the given
2992 * {@code ranges} and their equivalents found in the IANA Language
2993 * Subtag Registry. For example, if the given {@code ranges} is
2994 * {@code "Accept-Language: iw,en-us;q=0.7,en;q=0.3"}, the elements in
2995 * the list to be returned are:
2996 *
2997 * <pre>
2998 * <b>Range</b> <b>Weight</b>
2999 * "iw" (older tag for Hebrew) 1.0
3000 * "he" (new preferred code for Hebrew) 1.0
3001 * "en-us" (English, United States) 0.7
3002 * "en" (English) 0.3
3003 * </pre>
3004 *
3005 * Two language ranges, {@code "iw"} and {@code "he"}, have the same
3006 * highest priority in the list. By adding {@code "he"} to the user's
3007 * Language Priority List, locale-matching method can find Hebrew as a
3008 * matching locale (or language tag) even if the application or system
3009 * offers only {@code "he"} as a supported locale (or language tag).
3010 *
3011 * @param ranges a list of comma-separated language ranges or a list of
3012 * language ranges in the form of the "Accept-Language" header
3013 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3014 * 2616</a>
3015 * @return a Language Priority List consisting of language ranges
3016 * included in the given {@code ranges} and their equivalent
3017 * language ranges if available. The list is modifiable.
3018 * @throws NullPointerException if {@code ranges} is null
3019 * @throws IllegalArgumentException if a language range or a weight
3020 * found in the given {@code ranges} is ill-formed
3021 */
3022 public static List<LanguageRange> parse(String ranges) {
3023 return LocaleMatcher.parse(ranges);
3024 }
3025
3026 /**
3027 * Parses the given {@code ranges} to generate a Language Priority
3028 * List, and then customizes the list using the given {@code map}.
3029 * This method is equivalent to
3030 * {@code mapEquivalents(parse(ranges), map)}.
3031 *
3032 * @param ranges a list of comma-separated language ranges or a list
3033 * of language ranges in the form of the "Accept-Language" header
3034 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3035 * 2616</a>
3036 * @param map a map containing information to customize language ranges
3037 * @return a Language Priority List with customization. The list is
3038 * modifiable.
3039 * @throws NullPointerException if {@code ranges} is null
3040 * @throws IllegalArgumentException if a language range or a weight
3041 * found in the given {@code ranges} is ill-formed
3042 * @see #parse(String)
3043 * @see #mapEquivalents
3044 */
3045 public static List<LanguageRange> parse(String ranges,
3046 Map<String, List<String>> map) {
3047 return mapEquivalents(parse(ranges), map);
3048 }
3049
3050 /**
3051 * Generates a new customized Language Priority List using the given
3052 * {@code priorityList} and {@code map}. If the given {@code map} is
3053 * empty, this method returns a copy of the given {@code priorityList}.
3054 *
3055 * <p>In the map, a key represents a language range whereas a value is
3056 * a list of equivalents of it. {@code '*'} cannot be used in the map.
3057 * Each equivalent language range has the same weight value as its
3058 * original language range.
3059 *
3060 * <pre>
3061 * An example of map:
3062 * <b>Key</b> <b>Value</b>
3063 * "zh" (Chinese) "zh",
3064 * "zh-Hans"(Simplified Chinese)
3065 * "zh-HK" (Chinese, Hong Kong) "zh-HK"
3066 * "zh-TW" (Chinese, Taiwan) "zh-TW"
3067 * </pre>
3068 *
3069 * The customization is performed after modification using the IANA
3070 * Language Subtag Registry.
3071 *
3072 * <p>For example, if a user's Language Priority List consists of five
3073 * language ranges ({@code "zh"}, {@code "zh-CN"}, {@code "en"},
3074 * {@code "zh-TW"}, and {@code "zh-HK"}), the newly generated Language
3075 * Priority List which is customized using the above map example will
3076 * consists of {@code "zh"}, {@code "zh-Hans"}, {@code "zh-CN"},
3077 * {@code "zh-Hans-CN"}, {@code "en"}, {@code "zh-TW"}, and
3078 * {@code "zh-HK"}.
3079 *
3080 * <p>{@code "zh-HK"} and {@code "zh-TW"} aren't converted to
3081 * {@code "zh-Hans-HK"} nor {@code "zh-Hans-TW"} even if they are
3082 * included in the Language Priority List. In this example, mapping
3083 * is used to clearly distinguish Simplified Chinese and Traditional
3084 * Chinese.
3085 *
3086 * <p>If the {@code "zh"}-to-{@code "zh"} mapping isn't included in the
3087 * map, a simple replacement will be performed and the customized list
3088 * won't include {@code "zh"} and {@code "zh-CN"}.
3089 *
3090 * @param priorityList user's Language Priority List
3091 * @param map a map containing information to customize language ranges
3092 * @return a new Language Priority List with customization. The list is
3093 * modifiable.
3094 * @throws NullPointerException if {@code priorityList} is {@code null}
3095 * @see #parse(String, Map)
3096 */
3097 public static List<LanguageRange> mapEquivalents(
3098 List<LanguageRange>priorityList,
3099 Map<String, List<String>> map) {
3100 return LocaleMatcher.mapEquivalents(priorityList, map);
3101 }
3102
3103 /**
3104 * Returns a hash code value for the object.
3105 *
3106 * @return a hash code value for this object.
3107 */
3108 @Override
3109 public int hashCode() {
3110 int h = hash;
3111 if (h == 0) {
3112 h = 17;
3113 h = 37*h + range.hashCode();
3114 long bitsWeight = Double.doubleToLongBits(weight);
3115 h = 37*h + (int)(bitsWeight ^ (bitsWeight >>> 32));
3116 if (h != 0) {
3117 hash = h;
3118 }
3119 }
3120 return h;
3121 }
3122
3123 /**
3124 * Compares this object to the specified object. The result is true if
3125 * and only if the argument is not {@code null} and is a
3126 * {@code LanguageRange} object that contains the same {@code range}
3127 * and {@code weight} values as this object.
3128 *
3129 * @param obj the object to compare with
3130 * @return {@code true} if this object's {@code range} and
3131 * {@code weight} are the same as the {@code obj}'s; {@code false}
3132 * otherwise.
3133 */
3134 @Override
3135 public boolean equals(Object obj) {
3136 if (this == obj) {
3137 return true;
3138 }
3139 if (!(obj instanceof LanguageRange)) {
3140 return false;
3141 }
3142 LanguageRange other = (LanguageRange)obj;
3143 return hash == other.hash
3144 && range.equals(other.range)
3145 && weight == other.weight;
3146 }
3147
3148 /**
3149 * Returns an informative string representation of this {@code LanguageRange}
3150 * object, consisting of language range and weight if the range is
3151 * weighted and the weight is less than the max weight.
3152 *
3153 * @return a string representation of this {@code LanguageRange} object.
3154 */
3155 @Override
3156 public String toString() {
3157 return (weight == MAX_WEIGHT) ? range : range + ";q=" + weight;
3158 }
3159 }
3160
3161 /**
3162 * Returns a list of matching {@code Locale} instances using the filtering
3163 * mechanism defined in RFC 4647.
3164 *
3165 * @param priorityList user's Language Priority List in which each language
3166 * tag is sorted in descending order based on priority or weight
3167 * @param locales {@code Locale} instances used for matching
3168 * @param mode filtering mode
3169 * @return a list of {@code Locale} instances for matching language tags
3170 * sorted in descending order based on priority or weight, or an empty
3171 * list if nothing matches. The list is modifiable.
3172 * @throws NullPointerException if {@code priorityList} or {@code locales}
3173 * is {@code null}
3174 * @throws IllegalArgumentException if one or more extended language ranges
3175 * are included in the given list when
3176 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3177 *
3178 * @since 1.8
3179 */
3180 public static List<Locale> filter(List<LanguageRange> priorityList,
3181 Collection<Locale> locales,
3182 FilteringMode mode) {
3183 return LocaleMatcher.filter(priorityList, locales, mode);
3184 }
3185
3186 /**
3187 * Returns a list of matching {@code Locale} instances using the filtering
3188 * mechanism defined in RFC 4647. This is equivalent to
3189 * {@link #filter(List, Collection, FilteringMode)} when {@code mode} is
3190 * {@link FilteringMode#AUTOSELECT_FILTERING}.
3191 *
3192 * @param priorityList user's Language Priority List in which each language
3193 * tag is sorted in descending order based on priority or weight
3194 * @param locales {@code Locale} instances used for matching
3195 * @return a list of {@code Locale} instances for matching language tags
3196 * sorted in descending order based on priority or weight, or an empty
3197 * list if nothing matches. The list is modifiable.
3198 * @throws NullPointerException if {@code priorityList} or {@code locales}
3199 * is {@code null}
3200 *
3201 * @since 1.8
3202 */
3203 public static List<Locale> filter(List<LanguageRange> priorityList,
3204 Collection<Locale> locales) {
3205 return filter(priorityList, locales, FilteringMode.AUTOSELECT_FILTERING);
3206 }
3207
3208 /**
3209 * Returns a list of matching languages tags using the basic filtering
3210 * mechanism defined in RFC 4647.
3211 *
3212 * @param priorityList user's Language Priority List in which each language
3213 * tag is sorted in descending order based on priority or weight
3214 * @param tags language tags
3215 * @param mode filtering mode
3216 * @return a list of matching language tags sorted in descending order
3217 * based on priority or weight, or an empty list if nothing matches.
3218 * The list is modifiable.
3219 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3220 * {@code null}
3221 * @throws IllegalArgumentException if one or more extended language ranges
3222 * are included in the given list when
3223 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3224 *
3225 * @since 1.8
3226 */
3227 public static List<String> filterTags(List<LanguageRange> priorityList,
3228 Collection<String> tags,
3229 FilteringMode mode) {
3230 return LocaleMatcher.filterTags(priorityList, tags, mode);
3231 }
3232
3233 /**
3234 * Returns a list of matching languages tags using the basic filtering
3235 * mechanism defined in RFC 4647. This is equivalent to
3236 * {@link #filterTags(List, Collection, FilteringMode)} when {@code mode}
3237 * is {@link FilteringMode#AUTOSELECT_FILTERING}.
3238 *
3239 * @param priorityList user's Language Priority List in which each language
3240 * tag is sorted in descending order based on priority or weight
3241 * @param tags language tags
3242 * @return a list of matching language tags sorted in descending order
3243 * based on priority or weight, or an empty list if nothing matches.
3244 * The list is modifiable.
3245 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3246 * {@code null}
3247 *
3248 * @since 1.8
3249 */
3250 public static List<String> filterTags(List<LanguageRange> priorityList,
3251 Collection<String> tags) {
3252 return filterTags(priorityList, tags, FilteringMode.AUTOSELECT_FILTERING);
3253 }
3254
3255 /**
3256 * Returns a {@code Locale} instance for the best-matching language
3257 * tag using the lookup mechanism defined in RFC 4647.
3258 *
3259 * @param priorityList user's Language Priority List in which each language
3260 * tag is sorted in descending order based on priority or weight
3261 * @param locales {@code Locale} instances used for matching
3262 * @return the best matching <code>Locale</code> instance chosen based on
3263 * priority or weight, or {@code null} if nothing matches.
3264 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3265 * {@code null}
3266 *
3267 * @since 1.8
3268 */
3269 public static Locale lookup(List<LanguageRange> priorityList,
3270 Collection<Locale> locales) {
3271 return LocaleMatcher.lookup(priorityList, locales);
3272 }
3273
3274 /**
3275 * Returns the best-matching language tag using the lookup mechanism
3276 * defined in RFC 4647.
3277 *
3278 * @param priorityList user's Language Priority List in which each language
3279 * tag is sorted in descending order based on priority or weight
3280 * @param tags language tangs used for matching
3281 * @return the best matching language tag chosen based on priority or
3282 * weight, or {@code null} if nothing matches.
3283 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3284 * {@code null}
3285 *
3286 * @since 1.8
3287 */
3288 public static String lookupTag(List<LanguageRange> priorityList,
3289 Collection<String> tags) {
3290 return LocaleMatcher.lookupTag(priorityList, tags);
3291 }
3292
3293 }