1 /*
2 * Copyright (c) 1996, 2016, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25
26 /*
27 * (C) Copyright Taligent, Inc. 1996, 1997 - All Rights Reserved
28 * (C) Copyright IBM Corp. 1996 - 1998 - All Rights Reserved
29 *
30 * The original version of this source code and documentation
31 * is copyrighted and owned by Taligent, Inc., a wholly-owned
32 * subsidiary of IBM. These materials are provided under terms
33 * of a License Agreement between Taligent and Sun. This technology
34 * is protected by multiple US and International patents.
35 *
36 * This notice and attribution to Taligent may not be removed.
37 * Taligent is a registered trademark of Taligent, Inc.
38 *
39 */
40
41 package java.util;
42
43 import java.io.IOException;
44 import java.io.ObjectInputStream;
45 import java.io.ObjectOutputStream;
46 import java.io.ObjectStreamField;
47 import java.io.Serializable;
48 import java.text.MessageFormat;
49 import java.util.concurrent.ConcurrentHashMap;
50 import java.util.spi.LocaleNameProvider;
51
52 import sun.security.action.GetPropertyAction;
53 import sun.util.locale.BaseLocale;
54 import sun.util.locale.InternalLocaleBuilder;
55 import sun.util.locale.LanguageTag;
56 import sun.util.locale.LocaleExtensions;
57 import sun.util.locale.LocaleMatcher;
58 import sun.util.locale.LocaleObjectCache;
59 import sun.util.locale.LocaleSyntaxException;
60 import sun.util.locale.LocaleUtils;
61 import sun.util.locale.ParseStatus;
62 import sun.util.locale.provider.LocaleProviderAdapter;
63 import sun.util.locale.provider.LocaleResources;
64 import sun.util.locale.provider.LocaleServiceProviderPool;
65
66 /**
67 * A <code>Locale</code> object represents a specific geographical, political,
68 * or cultural region. An operation that requires a <code>Locale</code> to perform
69 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code>
70 * to tailor information for the user. For example, displaying a number
71 * is a locale-sensitive operation— the number should be formatted
72 * according to the customs and conventions of the user's native country,
73 * region, or culture.
74 *
75 * <p> The {@code Locale} class implements IETF BCP 47 which is composed of
76 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 "Matching of Language
77 * Tags"</a> and <a href="http://tools.ietf.org/html/rfc5646">RFC 5646 "Tags
78 * for Identifying Languages"</a> with support for the LDML (UTS#35, "Unicode
79 * Locale Data Markup Language") BCP 47-compatible extensions for locale data
80 * exchange.
81 *
82 * <p> A <code>Locale</code> object logically consists of the fields
83 * described below.
84 *
85 * <dl>
86 * <dt><a name="def_language"><b>language</b></a></dt>
87 *
88 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered
89 * language subtags up to 8 alpha letters (for future enhancements).
90 * When a language has both an alpha-2 code and an alpha-3 code, the
91 * alpha-2 code must be used. You can find a full list of valid
92 * language codes in the IANA Language Subtag Registry (search for
93 * "Type: language"). The language field is case insensitive, but
94 * <code>Locale</code> always canonicalizes to lower case.</dd>
95 *
96 * <dd>Well-formed language values have the form
97 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the full
98 * BCP47 language production, since it excludes extlang. They are
99 * not needed since modern three-letter language codes replace
100 * them.</dd>
101 *
102 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd>
103 *
104 * <dt><a name="def_script"><b>script</b></a></dt>
105 *
106 * <dd>ISO 15924 alpha-4 script code. You can find a full list of
107 * valid script codes in the IANA Language Subtag Registry (search
108 * for "Type: script"). The script field is case insensitive, but
109 * <code>Locale</code> always canonicalizes to title case (the first
110 * letter is upper case and the rest of the letters are lower
111 * case).</dd>
112 *
113 * <dd>Well-formed script values have the form
114 * <code>[a-zA-Z]{4}</code></dd>
115 *
116 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd>
117 *
118 * <dt><a name="def_region"><b>country (region)</b></a></dt>
119 *
120 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code.
121 * You can find a full list of valid country and region codes in the
122 * IANA Language Subtag Registry (search for "Type: region"). The
123 * country (region) field is case insensitive, but
124 * <code>Locale</code> always canonicalizes to upper case.</dd>
125 *
126 * <dd>Well-formed country/region values have
127 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd>
128 *
129 * <dd>Example: "US" (United States), "FR" (France), "029"
130 * (Caribbean)</dd>
131 *
132 * <dt><a name="def_variant"><b>variant</b></a></dt>
133 *
134 * <dd>Any arbitrary value used to indicate a variation of a
135 * <code>Locale</code>. Where there are two or more variant values
136 * each indicating its own semantics, these values should be ordered
137 * by importance, with most important first, separated by
138 * underscore('_'). The variant field is case sensitive.</dd>
139 *
140 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant
141 * subtags. Also BCP 47 subtags are strictly used to indicate
142 * additional variations that define a language or its dialects that
143 * are not covered by any combinations of language, script and
144 * region subtags. You can find a full list of valid variant codes
145 * in the IANA Language Subtag Registry (search for "Type: variant").
146 *
147 * <p>However, the variant field in <code>Locale</code> has
148 * historically been used for any kind of variation, not just
149 * language variations. For example, some supported variants
150 * available in Java SE Runtime Environments indicate alternative
151 * cultural behaviors such as calendar type or number script. In
152 * BCP 47 this kind of information, which does not identify the
153 * language, is supported by extension subtags or private use
154 * subtags.</dd>
155 *
156 * <dd>Well-formed variant values have the form <code>SUBTAG
157 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG =
158 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only
159 * uses hyphen ('-') as a delimiter, this is more lenient).</dd>
160 *
161 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd>
162 *
163 * <dt><a name="def_extensions"><b>extensions</b></a></dt>
164 *
165 * <dd>A map from single character keys to string values, indicating
166 * extensions apart from language identification. The extensions in
167 * <code>Locale</code> implement the semantics and syntax of BCP 47
168 * extension subtags and private use subtags. The extensions are
169 * case insensitive, but <code>Locale</code> canonicalizes all
170 * extension keys and values to lower case. Note that extensions
171 * cannot have empty values.</dd>
172 *
173 * <dd>Well-formed keys are single characters from the set
174 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form
175 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x'
176 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys
177 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows
178 * single-character subtags).</dd>
179 *
180 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar),
181 * key="x"/value="java-1-7"</dd>
182 * </dl>
183 *
184 * <b>Note:</b> Although BCP 47 requires field values to be registered
185 * in the IANA Language Subtag Registry, the <code>Locale</code> class
186 * does not provide any validation features. The <code>Builder</code>
187 * only checks if an individual field satisfies the syntactic
188 * requirement (is well-formed), but does not validate the value
189 * itself. See {@link Builder} for details.
190 *
191 * <h3><a name="def_locale_extension">Unicode locale/language extension</a></h3>
192 *
193 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional
194 * attributes and keywords to override or refine the default behavior
195 * associated with a locale. A keyword is represented by a pair of
196 * key and type. For example, "nu-thai" indicates that Thai local
197 * digits (value:"thai") should be used for formatting numbers
198 * (key:"nu").
199 *
200 * <p>The keywords are mapped to a BCP 47 extension value using the
201 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above
202 * example, "nu-thai", becomes the extension "u-nu-thai".
203 *
204 * <p>Thus, when a <code>Locale</code> object contains Unicode locale
205 * attributes and keywords,
206 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a
207 * String representing this information, for example, "nu-thai". The
208 * <code>Locale</code> class also provides {@link
209 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and
210 * {@link #getUnicodeLocaleType} which allow you to access Unicode
211 * locale attributes and key/type pairs directly. When represented as
212 * a string, the Unicode Locale Extension lists attributes
213 * alphabetically, followed by key/type sequences with keys listed
214 * alphabetically (the order of subtags comprising a key's type is
215 * fixed when the type is defined)
216 *
217 * <p>A well-formed locale key has the form
218 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the
219 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it
220 * can be empty, or a series of subtags 3-8 alphanums in length). A
221 * well-formed locale attribute has the form
222 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same
223 * form as a locale type subtag).
224 *
225 * <p>The Unicode locale extension specifies optional behavior in
226 * locale-sensitive services. Although the LDML specification defines
227 * various keys and values, actual locale-sensitive service
228 * implementations in a Java Runtime Environment might not support any
229 * particular Unicode locale attributes or key/type pairs.
230 *
231 * <h4>Creating a Locale</h4>
232 *
233 * <p>There are several different ways to create a <code>Locale</code>
234 * object.
235 *
236 * <h5>Builder</h5>
237 *
238 * <p>Using {@link Builder} you can construct a <code>Locale</code> object
239 * that conforms to BCP 47 syntax.
240 *
241 * <h5>Constructors</h5>
242 *
243 * <p>The <code>Locale</code> class provides three constructors:
244 * <blockquote>
245 * <pre>
246 * {@link #Locale(String language)}
247 * {@link #Locale(String language, String country)}
248 * {@link #Locale(String language, String country, String variant)}
249 * </pre>
250 * </blockquote>
251 * These constructors allow you to create a <code>Locale</code> object
252 * with language, country and variant, but you cannot specify
253 * script or extensions.
254 *
255 * <h5>Factory Methods</h5>
256 *
257 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code>
258 * object for a well-formed BCP 47 language tag.
259 *
260 * <h5>Locale Constants</h5>
261 *
262 * <p>The <code>Locale</code> class provides a number of convenient constants
263 * that you can use to create <code>Locale</code> objects for commonly used
264 * locales. For example, the following creates a <code>Locale</code> object
265 * for the United States:
266 * <blockquote>
267 * <pre>
268 * Locale.US
269 * </pre>
270 * </blockquote>
271 *
272 * <h4><a name="LocaleMatching">Locale Matching</a></h4>
273 *
274 * <p>If an application or a system is internationalized and provides localized
275 * resources for multiple locales, it sometimes needs to find one or more
276 * locales (or language tags) which meet each user's specific preferences. Note
277 * that a term "language tag" is used interchangeably with "locale" in this
278 * locale matching documentation.
279 *
280 * <p>In order to do matching a user's preferred locales to a set of language
281 * tags, <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
282 * Language Tags</a> defines two mechanisms: filtering and lookup.
283 * <em>Filtering</em> is used to get all matching locales, whereas
284 * <em>lookup</em> is to choose the best matching locale.
285 * Matching is done case-insensitively. These matching mechanisms are described
286 * in the following sections.
287 *
288 * <p>A user's preference is called a <em>Language Priority List</em> and is
289 * expressed as a list of language ranges. There are syntactically two types of
290 * language ranges: basic and extended. See
291 * {@link Locale.LanguageRange Locale.LanguageRange} for details.
292 *
293 * <h5>Filtering</h5>
294 *
295 * <p>The filtering operation returns all matching language tags. It is defined
296 * in RFC 4647 as follows:
297 * "In filtering, each language range represents the least specific language
298 * tag (that is, the language tag with fewest number of subtags) that is an
299 * acceptable match. All of the language tags in the matching set of tags will
300 * have an equal or greater number of subtags than the language range. Every
301 * non-wildcard subtag in the language range will appear in every one of the
302 * matching language tags."
303 *
304 * <p>There are two types of filtering: filtering for basic language ranges
305 * (called "basic filtering") and filtering for extended language ranges
306 * (called "extended filtering"). They may return different results by what
307 * kind of language ranges are included in the given Language Priority List.
308 * {@link Locale.FilteringMode} is a parameter to specify how filtering should
309 * be done.
310 *
311 * <h5>Lookup</h5>
312 *
313 * <p>The lookup operation returns the best matching language tags. It is
314 * defined in RFC 4647 as follows:
315 * "By contrast with filtering, each language range represents the most
316 * specific tag that is an acceptable match. The first matching tag found,
317 * according to the user's priority, is considered the closest match and is the
318 * item returned."
319 *
320 * <p>For example, if a Language Priority List consists of two language ranges,
321 * {@code "zh-Hant-TW"} and {@code "en-US"}, in prioritized order, lookup
322 * method progressively searches the language tags below in order to find the
323 * best matching language tag.
324 * <blockquote>
325 * <pre>
326 * 1. zh-Hant-TW
327 * 2. zh-Hant
328 * 3. zh
329 * 4. en-US
330 * 5. en
331 * </pre>
332 * </blockquote>
333 * If there is a language tag which matches completely to a language range
334 * above, the language tag is returned.
335 *
336 * <p>{@code "*"} is the special language range, and it is ignored in lookup.
337 *
338 * <p>If multiple language tags match as a result of the subtag {@code '*'}
339 * included in a language range, the first matching language tag returned by
340 * an {@link Iterator} over a {@link Collection} of language tags is treated as
341 * the best matching one.
342 *
343 * <h4>Use of Locale</h4>
344 *
345 * <p>Once you've created a <code>Locale</code> you can query it for information
346 * about itself. Use <code>getCountry</code> to get the country (or region)
347 * code and <code>getLanguage</code> to get the language code.
348 * You can use <code>getDisplayCountry</code> to get the
349 * name of the country suitable for displaying to the user. Similarly,
350 * you can use <code>getDisplayLanguage</code> to get the name of
351 * the language suitable for displaying to the user. Interestingly,
352 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive
353 * and have two versions: one that uses the default
354 * {@link Locale.Category#DISPLAY DISPLAY} locale and one
355 * that uses the locale specified as an argument.
356 *
357 * <p>The Java Platform provides a number of classes that perform locale-sensitive
358 * operations. For example, the <code>NumberFormat</code> class formats
359 * numbers, currency, and percentages in a locale-sensitive manner. Classes
360 * such as <code>NumberFormat</code> have several convenience methods
361 * for creating a default object of that type. For example, the
362 * <code>NumberFormat</code> class provides these three convenience methods
363 * for creating a default <code>NumberFormat</code> object:
364 * <blockquote>
365 * <pre>
366 * NumberFormat.getInstance()
367 * NumberFormat.getCurrencyInstance()
368 * NumberFormat.getPercentInstance()
369 * </pre>
370 * </blockquote>
371 * Each of these methods has two variants; one with an explicit locale
372 * and one without; the latter uses the default
373 * {@link Locale.Category#FORMAT FORMAT} locale:
374 * <blockquote>
375 * <pre>
376 * NumberFormat.getInstance(myLocale)
377 * NumberFormat.getCurrencyInstance(myLocale)
378 * NumberFormat.getPercentInstance(myLocale)
379 * </pre>
380 * </blockquote>
381 * A <code>Locale</code> is the mechanism for identifying the kind of object
382 * (<code>NumberFormat</code>) that you would like to get. The locale is
383 * <STRONG>just</STRONG> a mechanism for identifying objects,
384 * <STRONG>not</STRONG> a container for the objects themselves.
385 *
386 * <h4>Compatibility</h4>
387 *
388 * <p>In order to maintain compatibility with existing usage, Locale's
389 * constructors retain their behavior prior to the Java Runtime
390 * Environment version 1.7. The same is largely true for the
391 * <code>toString</code> method. Thus Locale objects can continue to
392 * be used as they were. In particular, clients who parse the output
393 * of toString into language, country, and variant fields can continue
394 * to do so (although this is strongly discouraged), although the
395 * variant field will have additional information in it if script or
396 * extensions are present.
397 *
398 * <p>In addition, BCP 47 imposes syntax restrictions that are not
399 * imposed by Locale's constructors. This means that conversions
400 * between some Locales and BCP 47 language tags cannot be made without
401 * losing information. Thus <code>toLanguageTag</code> cannot
402 * represent the state of locales whose language, country, or variant
403 * do not conform to BCP 47.
404 *
405 * <p>Because of these issues, it is recommended that clients migrate
406 * away from constructing non-conforming locales and use the
407 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead.
408 * Clients desiring a string representation of the complete locale can
409 * then always rely on <code>toLanguageTag</code> for this purpose.
410 *
411 * <h5><a name="special_cases_constructor">Special cases</a></h5>
412 *
413 * <p>For compatibility reasons, two
414 * non-conforming locales are treated as special cases. These are
415 * <b>{@code ja_JP_JP}</b> and <b>{@code th_TH_TH}</b>. These are ill-formed
416 * in BCP 47 since the variants are too short. To ease migration to BCP 47,
417 * these are treated specially during construction. These two cases (and only
418 * these) cause a constructor to generate an extension, all other values behave
419 * exactly as they did prior to Java 7.
420 *
421 * <p>Java has used {@code ja_JP_JP} to represent Japanese as used in
422 * Japan together with the Japanese Imperial calendar. This is now
423 * representable using a Unicode locale extension, by specifying the
424 * Unicode locale key {@code ca} (for "calendar") and type
425 * {@code japanese}. When the Locale constructor is called with the
426 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is
427 * automatically added.
428 *
429 * <p>Java has used {@code th_TH_TH} to represent Thai as used in
430 * Thailand together with Thai digits. This is also now representable using
431 * a Unicode locale extension, by specifying the Unicode locale key
432 * {@code nu} (for "number") and value {@code thai}. When the Locale
433 * constructor is called with the arguments "th", "TH", "TH", the
434 * extension "u-nu-thai" is automatically added.
435 *
436 * <h5>Serialization</h5>
437 *
438 * <p>During serialization, writeObject writes all fields to the output
439 * stream, including extensions.
440 *
441 * <p>During deserialization, readResolve adds extensions as described
442 * in <a href="#special_cases_constructor">Special Cases</a>, only
443 * for the two cases th_TH_TH and ja_JP_JP.
444 *
445 * <h5>Legacy language codes</h5>
446 *
447 * <p>Locale's constructor has always converted three language codes to
448 * their earlier, obsoleted forms: {@code he} maps to {@code iw},
449 * {@code yi} maps to {@code ji}, and {@code id} maps to
450 * {@code in}. This continues to be the case, in order to not break
451 * backwards compatibility.
452 *
453 * <p>The APIs added in 1.7 map between the old and new language codes,
454 * maintaining the old codes internal to Locale (so that
455 * <code>getLanguage</code> and <code>toString</code> reflect the old
456 * code), but using the new codes in the BCP 47 language tag APIs (so
457 * that <code>toLanguageTag</code> reflects the new one). This
458 * preserves the equivalence between Locales no matter which code or
459 * API is used to construct them. Java's default resource bundle
460 * lookup mechanism also implements this mapping, so that resources
461 * can be named using either convention, see {@link ResourceBundle.Control}.
462 *
463 * <h5>Three-letter language/country(region) codes</h5>
464 *
465 * <p>The Locale constructors have always specified that the language
466 * and the country param be two characters in length, although in
467 * practice they have accepted any length. The specification has now
468 * been relaxed to allow language codes of two to eight characters and
469 * country (region) codes of two to three characters, and in
470 * particular, three-letter language codes and three-digit region
471 * codes as specified in the IANA Language Subtag Registry. For
472 * compatibility, the implementation still does not impose a length
473 * constraint.
474 *
475 * @see Builder
476 * @see ResourceBundle
477 * @see java.text.Format
478 * @see java.text.NumberFormat
479 * @see java.text.Collator
480 * @author Mark Davis
481 * @since 1.1
482 */
483 public final class Locale implements Cloneable, Serializable {
484
485 private static final Cache LOCALECACHE = new Cache();
486
487 /** Useful constant for language.
488 */
489 public static final Locale ENGLISH = createConstant("en", "");
490
491 /** Useful constant for language.
492 */
493 public static final Locale FRENCH = createConstant("fr", "");
494
495 /** Useful constant for language.
496 */
497 public static final Locale GERMAN = createConstant("de", "");
498
499 /** Useful constant for language.
500 */
501 public static final Locale ITALIAN = createConstant("it", "");
502
503 /** Useful constant for language.
504 */
505 public static final Locale JAPANESE = createConstant("ja", "");
506
507 /** Useful constant for language.
508 */
509 public static final Locale KOREAN = createConstant("ko", "");
510
511 /** Useful constant for language.
512 */
513 public static final Locale CHINESE = createConstant("zh", "");
514
515 /** Useful constant for language.
516 */
517 public static final Locale SIMPLIFIED_CHINESE = createConstant("zh", "CN");
518
519 /** Useful constant for language.
520 */
521 public static final Locale TRADITIONAL_CHINESE = createConstant("zh", "TW");
522
523 /** Useful constant for country.
524 */
525 public static final Locale FRANCE = createConstant("fr", "FR");
526
527 /** Useful constant for country.
528 */
529 public static final Locale GERMANY = createConstant("de", "DE");
530
531 /** Useful constant for country.
532 */
533 public static final Locale ITALY = createConstant("it", "IT");
534
535 /** Useful constant for country.
536 */
537 public static final Locale JAPAN = createConstant("ja", "JP");
538
539 /** Useful constant for country.
540 */
541 public static final Locale KOREA = createConstant("ko", "KR");
542
543 /** Useful constant for country.
544 */
545 public static final Locale CHINA = SIMPLIFIED_CHINESE;
546
547 /** Useful constant for country.
548 */
549 public static final Locale PRC = SIMPLIFIED_CHINESE;
550
551 /** Useful constant for country.
552 */
553 public static final Locale TAIWAN = TRADITIONAL_CHINESE;
554
555 /** Useful constant for country.
556 */
557 public static final Locale UK = createConstant("en", "GB");
558
559 /** Useful constant for country.
560 */
561 public static final Locale US = createConstant("en", "US");
562
563 /** Useful constant for country.
564 */
565 public static final Locale CANADA = createConstant("en", "CA");
566
567 /** Useful constant for country.
568 */
569 public static final Locale CANADA_FRENCH = createConstant("fr", "CA");
570
571 /**
572 * Useful constant for the root locale. The root locale is the locale whose
573 * language, country, and variant are empty ("") strings. This is regarded
574 * as the base locale of all locales, and is used as the language/country
575 * neutral locale for the locale sensitive operations.
576 *
577 * @since 1.6
578 */
579 public static final Locale ROOT = createConstant("", "");
580
581 /**
582 * The key for the private use extension ('x').
583 *
584 * @see #getExtension(char)
585 * @see Builder#setExtension(char, String)
586 * @since 1.7
587 */
588 public static final char PRIVATE_USE_EXTENSION = 'x';
589
590 /**
591 * The key for Unicode locale extension ('u').
592 *
593 * @see #getExtension(char)
594 * @see Builder#setExtension(char, String)
595 * @since 1.7
596 */
597 public static final char UNICODE_LOCALE_EXTENSION = 'u';
598
599 /** serialization ID
600 */
601 static final long serialVersionUID = 9149081749638150636L;
602
603 /**
604 * Enum for specifying the type defined in ISO 3166. This enum is used to
605 * retrieve the two-letter ISO3166-1 alpha-2, three-letter ISO3166-1
606 * alpha-3, four-letter ISO3166-3 country codes.
607 *
608 * @see #getISOCountries(Locale.IsoCountryCode)
609 * @since 9
610 */
611 public static enum IsoCountryCode {
612 /**
613 * PART1_ALPHA2 is used to represent the ISO3166-1 alpha-2 two letter
614 * country codes.
615 */
616 PART1_ALPHA2 {
617 @Override
618 Set<String> getISOCountriesImpl() {
619 return Set.of(Locale.getISOCountries());
620 }
621 },
622
623 /**
624 *
625 * PART1_ALPHA3 is used to represent the ISO3166-1 alpha-3 three letter
626 * country codes.
627 */
628
629 PART1_ALPHA3 {
630 @Override
631 Set<String> getISOCountriesImpl() {
632 return LocaleISOData.computeISO3166_1Alpha3Countries();
633 }
634 },
635
636 /**
637 * PART3 is used to represent the ISO3166-3 four letter country codes.
638 */
639
640 PART3 {
641 @Override
642 Set<String> getISOCountriesImpl() {
643 return Set.of(LocaleISOData.ISO3166_3);
644 }
645 };
646
647 /**
648 * Concrete implementation of this mapping function attempts to compute value
649 * for iso3166CodesMap for each IsoCountryCode type key.
650 */
651 abstract Set<String> getISOCountriesImpl();
652
653 /**
654 * Map to hold country codes for each ISO3166 part.
655 */
656 private static Map<IsoCountryCode, Set<String>> iso3166CodesMap = new ConcurrentHashMap<>();
657
658 /**
659 * This method is called from Locale class to retrieve country code set
660 * for getISOCountries(type)
661 */
662 static Set<String> retrieveISOCountryCodes(IsoCountryCode type) {
663 return iso3166CodesMap.computeIfAbsent(type, IsoCountryCode::getISOCountriesImpl);
664 }
665 }
666
667 /**
668 * Display types for retrieving localized names from the name providers.
669 */
670 private static final int DISPLAY_LANGUAGE = 0;
671 private static final int DISPLAY_COUNTRY = 1;
672 private static final int DISPLAY_VARIANT = 2;
673 private static final int DISPLAY_SCRIPT = 3;
674
675 /**
676 * Private constructor used by getInstance method
677 */
678 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) {
679 this.baseLocale = baseLocale;
680 this.localeExtensions = extensions;
681 }
682
683 /**
684 * Construct a locale from language, country and variant.
685 * This constructor normalizes the language value to lowercase and
686 * the country value to uppercase.
687 * <p>
688 * <b>Note:</b>
689 * <ul>
690 * <li>ISO 639 is not a stable standard; some of the language codes it defines
691 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
692 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
693 * API on Locale will return only the OLD codes.
694 * <li>For backward compatibility reasons, this constructor does not make
695 * any syntactic checks on the input.
696 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially,
697 * see <a href="#special_cases_constructor">Special Cases</a> for more information.
698 * </ul>
699 *
700 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
701 * up to 8 characters in length. See the <code>Locale</code> class description about
702 * valid language values.
703 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
704 * See the <code>Locale</code> class description about valid country values.
705 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>.
706 * See the <code>Locale</code> class description for the details.
707 * @exception NullPointerException thrown if any argument is null.
708 */
709 public Locale(String language, String country, String variant) {
710 if (language== null || country == null || variant == null) {
711 throw new NullPointerException();
712 }
713 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant);
714 localeExtensions = getCompatibilityExtensions(language, "", country, variant);
715 }
716
717 /**
718 * Construct a locale from language and country.
719 * This constructor normalizes the language value to lowercase and
720 * the country value to uppercase.
721 * <p>
722 * <b>Note:</b>
723 * <ul>
724 * <li>ISO 639 is not a stable standard; some of the language codes it defines
725 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
726 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
727 * API on Locale will return only the OLD codes.
728 * <li>For backward compatibility reasons, this constructor does not make
729 * any syntactic checks on the input.
730 * </ul>
731 *
732 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
733 * up to 8 characters in length. See the <code>Locale</code> class description about
734 * valid language values.
735 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
736 * See the <code>Locale</code> class description about valid country values.
737 * @exception NullPointerException thrown if either argument is null.
738 */
739 public Locale(String language, String country) {
740 this(language, country, "");
741 }
742
743 /**
744 * Construct a locale from a language code.
745 * This constructor normalizes the language value to lowercase.
746 * <p>
747 * <b>Note:</b>
748 * <ul>
749 * <li>ISO 639 is not a stable standard; some of the language codes it defines
750 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
751 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
752 * API on Locale will return only the OLD codes.
753 * <li>For backward compatibility reasons, this constructor does not make
754 * any syntactic checks on the input.
755 * </ul>
756 *
757 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
758 * up to 8 characters in length. See the <code>Locale</code> class description about
759 * valid language values.
760 * @exception NullPointerException thrown if argument is null.
761 * @since 1.4
762 */
763 public Locale(String language) {
764 this(language, "", "");
765 }
766
767 /**
768 * This method must be called only for creating the Locale.*
769 * constants due to making shortcuts.
770 */
771 private static Locale createConstant(String lang, String country) {
772 BaseLocale base = BaseLocale.createInstance(lang, country);
773 return getInstance(base, null);
774 }
775
776 /**
777 * Returns a <code>Locale</code> constructed from the given
778 * <code>language</code>, <code>country</code> and
779 * <code>variant</code>. If the same <code>Locale</code> instance
780 * is available in the cache, then that instance is
781 * returned. Otherwise, a new <code>Locale</code> instance is
782 * created and cached.
783 *
784 * @param language lowercase 2 to 8 language code.
785 * @param country uppercase two-letter ISO-3166 code and numeric-3 UN M.49 area code.
786 * @param variant vendor and browser specific code. See class description.
787 * @return the <code>Locale</code> instance requested
788 * @exception NullPointerException if any argument is null.
789 */
790 static Locale getInstance(String language, String country, String variant) {
791 return getInstance(language, "", country, variant, null);
792 }
793
794 static Locale getInstance(String language, String script, String country,
795 String variant, LocaleExtensions extensions) {
796 if (language== null || script == null || country == null || variant == null) {
797 throw new NullPointerException();
798 }
799
800 if (extensions == null) {
801 extensions = getCompatibilityExtensions(language, script, country, variant);
802 }
803
804 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant);
805 return getInstance(baseloc, extensions);
806 }
807
808 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) {
809 LocaleKey key = new LocaleKey(baseloc, extensions);
810 return LOCALECACHE.get(key);
811 }
812
813 private static class Cache extends LocaleObjectCache<LocaleKey, Locale> {
814 private Cache() {
815 }
816
817 @Override
818 protected Locale createObject(LocaleKey key) {
819 return new Locale(key.base, key.exts);
820 }
821 }
822
823 private static final class LocaleKey {
824 private final BaseLocale base;
825 private final LocaleExtensions exts;
826 private final int hash;
827
828 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) {
829 base = baseLocale;
830 exts = extensions;
831
832 // Calculate the hash value here because it's always used.
833 int h = base.hashCode();
834 if (exts != null) {
835 h ^= exts.hashCode();
836 }
837 hash = h;
838 }
839
840 @Override
841 public boolean equals(Object obj) {
842 if (this == obj) {
843 return true;
844 }
845 if (!(obj instanceof LocaleKey)) {
846 return false;
847 }
848 LocaleKey other = (LocaleKey)obj;
849 if (hash != other.hash || !base.equals(other.base)) {
850 return false;
851 }
852 if (exts == null) {
853 return other.exts == null;
854 }
855 return exts.equals(other.exts);
856 }
857
858 @Override
859 public int hashCode() {
860 return hash;
861 }
862 }
863
864 /**
865 * Gets the current value of the default locale for this instance
866 * of the Java Virtual Machine.
867 * <p>
868 * The Java Virtual Machine sets the default locale during startup
869 * based on the host environment. It is used by many locale-sensitive
870 * methods if no locale is explicitly specified.
871 * It can be changed using the
872 * {@link #setDefault(java.util.Locale) setDefault} method.
873 *
874 * @return the default locale for this instance of the Java Virtual Machine
875 */
876 public static Locale getDefault() {
877 // do not synchronize this method - see 4071298
878 return defaultLocale;
879 }
880
881 /**
882 * Gets the current value of the default locale for the specified Category
883 * for this instance of the Java Virtual Machine.
884 * <p>
885 * The Java Virtual Machine sets the default locale during startup based
886 * on the host environment. It is used by many locale-sensitive methods
887 * if no locale is explicitly specified. It can be changed using the
888 * setDefault(Locale.Category, Locale) method.
889 *
890 * @param category - the specified category to get the default locale
891 * @throws NullPointerException if category is null
892 * @return the default locale for the specified Category for this instance
893 * of the Java Virtual Machine
894 * @see #setDefault(Locale.Category, Locale)
895 * @since 1.7
896 */
897 public static Locale getDefault(Locale.Category category) {
898 // do not synchronize this method - see 4071298
899 switch (category) {
900 case DISPLAY:
901 if (defaultDisplayLocale == null) {
902 synchronized(Locale.class) {
903 if (defaultDisplayLocale == null) {
904 defaultDisplayLocale = initDefault(category);
905 }
906 }
907 }
908 return defaultDisplayLocale;
909 case FORMAT:
910 if (defaultFormatLocale == null) {
911 synchronized(Locale.class) {
912 if (defaultFormatLocale == null) {
913 defaultFormatLocale = initDefault(category);
914 }
915 }
916 }
917 return defaultFormatLocale;
918 default:
919 assert false: "Unknown Category";
920 }
921 return getDefault();
922 }
923
924 private static Locale initDefault() {
925 String language, region, script, country, variant;
926 Properties props = GetPropertyAction.privilegedGetProperties();
927 language = props.getProperty("user.language", "en");
928 // for compatibility, check for old user.region property
929 region = props.getProperty("user.region");
930 if (region != null) {
931 // region can be of form country, country_variant, or _variant
932 int i = region.indexOf('_');
933 if (i >= 0) {
934 country = region.substring(0, i);
935 variant = region.substring(i + 1);
936 } else {
937 country = region;
938 variant = "";
939 }
940 script = "";
941 } else {
942 script = props.getProperty("user.script", "");
943 country = props.getProperty("user.country", "");
944 variant = props.getProperty("user.variant", "");
945 }
946
947 return getInstance(language, script, country, variant, null);
948 }
949
950 private static Locale initDefault(Locale.Category category) {
951 Properties props = GetPropertyAction.privilegedGetProperties();
952 return getInstance(
953 props.getProperty(category.languageKey,
954 defaultLocale.getLanguage()),
955 props.getProperty(category.scriptKey,
956 defaultLocale.getScript()),
957 props.getProperty(category.countryKey,
958 defaultLocale.getCountry()),
959 props.getProperty(category.variantKey,
960 defaultLocale.getVariant()),
961 null);
962 }
963
964 /**
965 * Sets the default locale for this instance of the Java Virtual Machine.
966 * This does not affect the host locale.
967 * <p>
968 * If there is a security manager, its <code>checkPermission</code>
969 * method is called with a <code>PropertyPermission("user.language", "write")</code>
970 * permission before the default locale is changed.
971 * <p>
972 * The Java Virtual Machine sets the default locale during startup
973 * based on the host environment. It is used by many locale-sensitive
974 * methods if no locale is explicitly specified.
975 * <p>
976 * Since changing the default locale may affect many different areas
977 * of functionality, this method should only be used if the caller
978 * is prepared to reinitialize locale-sensitive code running
979 * within the same Java Virtual Machine.
980 * <p>
981 * By setting the default locale with this method, all of the default
982 * locales for each Category are also set to the specified default locale.
983 *
984 * @throws SecurityException
985 * if a security manager exists and its
986 * <code>checkPermission</code> method doesn't allow the operation.
987 * @throws NullPointerException if <code>newLocale</code> is null
988 * @param newLocale the new default locale
989 * @see SecurityManager#checkPermission
990 * @see java.util.PropertyPermission
991 */
992 public static synchronized void setDefault(Locale newLocale) {
993 setDefault(Category.DISPLAY, newLocale);
994 setDefault(Category.FORMAT, newLocale);
995 defaultLocale = newLocale;
996 }
997
998 /**
999 * Sets the default locale for the specified Category for this instance
1000 * of the Java Virtual Machine. This does not affect the host locale.
1001 * <p>
1002 * If there is a security manager, its checkPermission method is called
1003 * with a PropertyPermission("user.language", "write") permission before
1004 * the default locale is changed.
1005 * <p>
1006 * The Java Virtual Machine sets the default locale during startup based
1007 * on the host environment. It is used by many locale-sensitive methods
1008 * if no locale is explicitly specified.
1009 * <p>
1010 * Since changing the default locale may affect many different areas of
1011 * functionality, this method should only be used if the caller is
1012 * prepared to reinitialize locale-sensitive code running within the
1013 * same Java Virtual Machine.
1014 *
1015 * @param category - the specified category to set the default locale
1016 * @param newLocale - the new default locale
1017 * @throws SecurityException if a security manager exists and its
1018 * checkPermission method doesn't allow the operation.
1019 * @throws NullPointerException if category and/or newLocale is null
1020 * @see SecurityManager#checkPermission(java.security.Permission)
1021 * @see PropertyPermission
1022 * @see #getDefault(Locale.Category)
1023 * @since 1.7
1024 */
1025 public static synchronized void setDefault(Locale.Category category,
1026 Locale newLocale) {
1027 if (category == null)
1028 throw new NullPointerException("Category cannot be NULL");
1029 if (newLocale == null)
1030 throw new NullPointerException("Can't set default locale to NULL");
1031
1032 SecurityManager sm = System.getSecurityManager();
1033 if (sm != null) sm.checkPermission(new PropertyPermission
1034 ("user.language", "write"));
1035 switch (category) {
1036 case DISPLAY:
1037 defaultDisplayLocale = newLocale;
1038 break;
1039 case FORMAT:
1040 defaultFormatLocale = newLocale;
1041 break;
1042 default:
1043 assert false: "Unknown Category";
1044 }
1045 }
1046
1047 /**
1048 * Returns an array of all installed locales.
1049 * The returned array represents the union of locales supported
1050 * by the Java runtime environment and by installed
1051 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider}
1052 * implementations. It must contain at least a <code>Locale</code>
1053 * instance equal to {@link java.util.Locale#US Locale.US}.
1054 *
1055 * @return An array of installed locales.
1056 */
1057 public static Locale[] getAvailableLocales() {
1058 return LocaleServiceProviderPool.getAllAvailableLocales();
1059 }
1060
1061 /**
1062 * Returns a list of all 2-letter country codes defined in ISO 3166.
1063 * Can be used to create Locales.
1064 * This method is equivalent to {@link #getISOCountries(Locale.IsoCountryCode type)}
1065 * with {@code type} {@link IsoCountryCode#PART1_ALPHA2}.
1066 * <p>
1067 * <b>Note:</b> The <code>Locale</code> class also supports other codes for
1068 * country (region), such as 3-letter numeric UN M.49 area codes.
1069 * Therefore, the list returned by this method does not contain ALL valid
1070 * codes that can be used to create Locales.
1071 * <p>
1072 * Note that this method does not return obsolete 2-letter country codes.
1073 * ISO3166-3 codes which designate country codes for those obsolete codes,
1074 * can be retrieved from {@link #getISOCountries(Locale.IsoCountryCode type)} with
1075 * {@code type} {@link IsoCountryCode#PART3}.
1076 * @return An array of ISO 3166 two-letter country codes.
1077 */
1078 public static String[] getISOCountries() {
1079 if (isoCountries == null) {
1080 isoCountries = getISO2Table(LocaleISOData.isoCountryTable);
1081 }
1082 String[] result = new String[isoCountries.length];
1083 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length);
1084 return result;
1085 }
1086
1087 /**
1088 * Returns a {@code Set} of ISO3166 country codes for the specified type.
1089 *
1090 * @param type {@link Locale.IsoCountryCode} specified ISO code type.
1091 * @see java.util.Locale.IsoCountryCode
1092 * @throws NullPointerException if type is null
1093 * @return a {@code Set} of ISO country codes for the specified type.
1094 * @since 9
1095 */
1096 public static Set<String> getISOCountries(IsoCountryCode type) {
1097 Objects.requireNonNull(type);
1098 return IsoCountryCode.retrieveISOCountryCodes(type);
1099 }
1100
1101 /**
1102 * Returns a list of all 2-letter language codes defined in ISO 639.
1103 * Can be used to create Locales.
1104 * <p>
1105 * <b>Note:</b>
1106 * <ul>
1107 * <li>ISO 639 is not a stable standard— some languages' codes have changed.
1108 * The list this function returns includes both the new and the old codes for the
1109 * languages whose codes have changed.
1110 * <li>The <code>Locale</code> class also supports language codes up to
1111 * 8 characters in length. Therefore, the list returned by this method does
1112 * not contain ALL valid codes that can be used to create Locales.
1113 * </ul>
1114 *
1115 * @return An array of ISO 639 two-letter language codes.
1116 */
1117 public static String[] getISOLanguages() {
1118 if (isoLanguages == null) {
1119 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable);
1120 }
1121 String[] result = new String[isoLanguages.length];
1122 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length);
1123 return result;
1124 }
1125
1126 private static String[] getISO2Table(String table) {
1127 int len = table.length() / 5;
1128 String[] isoTable = new String[len];
1129 for (int i = 0, j = 0; i < len; i++, j += 5) {
1130 isoTable[i] = table.substring(j, j + 2);
1131 }
1132 return isoTable;
1133 }
1134
1135 /**
1136 * Returns the language code of this Locale.
1137 *
1138 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed.
1139 * Locale's constructor recognizes both the new and the old codes for the languages
1140 * whose codes have changed, but this function always returns the old code. If you
1141 * want to check for a specific language whose code has changed, don't do
1142 * <pre>
1143 * if (locale.getLanguage().equals("he")) // BAD!
1144 * ...
1145 * </pre>
1146 * Instead, do
1147 * <pre>
1148 * if (locale.getLanguage().equals(new Locale("he").getLanguage()))
1149 * ...
1150 * </pre>
1151 * @return The language code, or the empty string if none is defined.
1152 * @see #getDisplayLanguage
1153 */
1154 public String getLanguage() {
1155 return baseLocale.getLanguage();
1156 }
1157
1158 /**
1159 * Returns the script for this locale, which should
1160 * either be the empty string or an ISO 15924 4-letter script
1161 * code. The first letter is uppercase and the rest are
1162 * lowercase, for example, 'Latn', 'Cyrl'.
1163 *
1164 * @return The script code, or the empty string if none is defined.
1165 * @see #getDisplayScript
1166 * @since 1.7
1167 */
1168 public String getScript() {
1169 return baseLocale.getScript();
1170 }
1171
1172 /**
1173 * Returns the country/region code for this locale, which should
1174 * either be the empty string, an uppercase ISO 3166 2-letter code,
1175 * or a UN M.49 3-digit code.
1176 *
1177 * @return The country/region code, or the empty string if none is defined.
1178 * @see #getDisplayCountry
1179 */
1180 public String getCountry() {
1181 return baseLocale.getRegion();
1182 }
1183
1184 /**
1185 * Returns the variant code for this locale.
1186 *
1187 * @return The variant code, or the empty string if none is defined.
1188 * @see #getDisplayVariant
1189 */
1190 public String getVariant() {
1191 return baseLocale.getVariant();
1192 }
1193
1194 /**
1195 * Returns {@code true} if this {@code Locale} has any <a href="#def_extensions">
1196 * extensions</a>.
1197 *
1198 * @return {@code true} if this {@code Locale} has any extensions
1199 * @since 1.8
1200 */
1201 public boolean hasExtensions() {
1202 return localeExtensions != null;
1203 }
1204
1205 /**
1206 * Returns a copy of this {@code Locale} with no <a href="#def_extensions">
1207 * extensions</a>. If this {@code Locale} has no extensions, this {@code Locale}
1208 * is returned.
1209 *
1210 * @return a copy of this {@code Locale} with no extensions, or {@code this}
1211 * if {@code this} has no extensions
1212 * @since 1.8
1213 */
1214 public Locale stripExtensions() {
1215 return hasExtensions() ? Locale.getInstance(baseLocale, null) : this;
1216 }
1217
1218 /**
1219 * Returns the extension (or private use) value associated with
1220 * the specified key, or null if there is no extension
1221 * associated with the key. To be well-formed, the key must be one
1222 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so
1223 * for example 'z' and 'Z' represent the same extension.
1224 *
1225 * @param key the extension key
1226 * @return The extension, or null if this locale defines no
1227 * extension for the specified key.
1228 * @throws IllegalArgumentException if key is not well-formed
1229 * @see #PRIVATE_USE_EXTENSION
1230 * @see #UNICODE_LOCALE_EXTENSION
1231 * @since 1.7
1232 */
1233 public String getExtension(char key) {
1234 if (!LocaleExtensions.isValidKey(key)) {
1235 throw new IllegalArgumentException("Ill-formed extension key: " + key);
1236 }
1237 return hasExtensions() ? localeExtensions.getExtensionValue(key) : null;
1238 }
1239
1240 /**
1241 * Returns the set of extension keys associated with this locale, or the
1242 * empty set if it has no extensions. The returned set is unmodifiable.
1243 * The keys will all be lower-case.
1244 *
1245 * @return The set of extension keys, or the empty set if this locale has
1246 * no extensions.
1247 * @since 1.7
1248 */
1249 public Set<Character> getExtensionKeys() {
1250 if (!hasExtensions()) {
1251 return Collections.emptySet();
1252 }
1253 return localeExtensions.getKeys();
1254 }
1255
1256 /**
1257 * Returns the set of unicode locale attributes associated with
1258 * this locale, or the empty set if it has no attributes. The
1259 * returned set is unmodifiable.
1260 *
1261 * @return The set of attributes.
1262 * @since 1.7
1263 */
1264 public Set<String> getUnicodeLocaleAttributes() {
1265 if (!hasExtensions()) {
1266 return Collections.emptySet();
1267 }
1268 return localeExtensions.getUnicodeLocaleAttributes();
1269 }
1270
1271 /**
1272 * Returns the Unicode locale type associated with the specified Unicode locale key
1273 * for this locale. Returns the empty string for keys that are defined with no type.
1274 * Returns null if the key is not defined. Keys are case-insensitive. The key must
1275 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is
1276 * thrown.
1277 *
1278 * @param key the Unicode locale key
1279 * @return The Unicode locale type associated with the key, or null if the
1280 * locale does not define the key.
1281 * @throws IllegalArgumentException if the key is not well-formed
1282 * @throws NullPointerException if <code>key</code> is null
1283 * @since 1.7
1284 */
1285 public String getUnicodeLocaleType(String key) {
1286 if (!isUnicodeExtensionKey(key)) {
1287 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key);
1288 }
1289 return hasExtensions() ? localeExtensions.getUnicodeLocaleType(key) : null;
1290 }
1291
1292 /**
1293 * Returns the set of Unicode locale keys defined by this locale, or the empty set if
1294 * this locale has none. The returned set is immutable. Keys are all lower case.
1295 *
1296 * @return The set of Unicode locale keys, or the empty set if this locale has
1297 * no Unicode locale keywords.
1298 * @since 1.7
1299 */
1300 public Set<String> getUnicodeLocaleKeys() {
1301 if (localeExtensions == null) {
1302 return Collections.emptySet();
1303 }
1304 return localeExtensions.getUnicodeLocaleKeys();
1305 }
1306
1307 /**
1308 * Package locale method returning the Locale's BaseLocale,
1309 * used by ResourceBundle
1310 * @return base locale of this Locale
1311 */
1312 BaseLocale getBaseLocale() {
1313 return baseLocale;
1314 }
1315
1316 /**
1317 * Package private method returning the Locale's LocaleExtensions,
1318 * used by ResourceBundle.
1319 * @return locale extensions of this Locale,
1320 * or {@code null} if no extensions are defined
1321 */
1322 LocaleExtensions getLocaleExtensions() {
1323 return localeExtensions;
1324 }
1325
1326 /**
1327 * Returns a string representation of this <code>Locale</code>
1328 * object, consisting of language, country, variant, script,
1329 * and extensions as below:
1330 * <blockquote>
1331 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "_" + extensions
1332 * </blockquote>
1333 *
1334 * Language is always lower case, country is always upper case, script is always title
1335 * case, and extensions are always lower case. Extensions and private use subtags
1336 * will be in canonical order as explained in {@link #toLanguageTag}.
1337 *
1338 * <p>When the locale has neither script nor extensions, the result is the same as in
1339 * Java 6 and prior.
1340 *
1341 * <p>If both the language and country fields are missing, this function will return
1342 * the empty string, even if the variant, script, or extensions field is present (you
1343 * can't have a locale with just a variant, the variant must accompany a well-formed
1344 * language or country code).
1345 *
1346 * <p>If script or extensions are present and variant is missing, no underscore is
1347 * added before the "#".
1348 *
1349 * <p>This behavior is designed to support debugging and to be compatible with
1350 * previous uses of <code>toString</code> that expected language, country, and variant
1351 * fields only. To represent a Locale as a String for interchange purposes, use
1352 * {@link #toLanguageTag}.
1353 *
1354 * <p>Examples: <ul>
1355 * <li>{@code en}</li>
1356 * <li>{@code de_DE}</li>
1357 * <li>{@code _GB}</li>
1358 * <li>{@code en_US_WIN}</li>
1359 * <li>{@code de__POSIX}</li>
1360 * <li>{@code zh_CN_#Hans}</li>
1361 * <li>{@code zh_TW_#Hant_x-java}</li>
1362 * <li>{@code th_TH_TH_#u-nu-thai}</li></ul>
1363 *
1364 * @return A string representation of the Locale, for debugging.
1365 * @see #getDisplayName
1366 * @see #toLanguageTag
1367 */
1368 @Override
1369 public final String toString() {
1370 boolean l = (baseLocale.getLanguage().length() != 0);
1371 boolean s = (baseLocale.getScript().length() != 0);
1372 boolean r = (baseLocale.getRegion().length() != 0);
1373 boolean v = (baseLocale.getVariant().length() != 0);
1374 boolean e = (localeExtensions != null && localeExtensions.getID().length() != 0);
1375
1376 StringBuilder result = new StringBuilder(baseLocale.getLanguage());
1377 if (r || (l && (v || s || e))) {
1378 result.append('_')
1379 .append(baseLocale.getRegion()); // This may just append '_'
1380 }
1381 if (v && (l || r)) {
1382 result.append('_')
1383 .append(baseLocale.getVariant());
1384 }
1385
1386 if (s && (l || r)) {
1387 result.append("_#")
1388 .append(baseLocale.getScript());
1389 }
1390
1391 if (e && (l || r)) {
1392 result.append('_');
1393 if (!s) {
1394 result.append('#');
1395 }
1396 result.append(localeExtensions.getID());
1397 }
1398
1399 return result.toString();
1400 }
1401
1402 /**
1403 * Returns a well-formed IETF BCP 47 language tag representing
1404 * this locale.
1405 *
1406 * <p>If this <code>Locale</code> has a language, country, or
1407 * variant that does not satisfy the IETF BCP 47 language tag
1408 * syntax requirements, this method handles these fields as
1409 * described below:
1410 *
1411 * <p><b>Language:</b> If language is empty, or not <a
1412 * href="#def_language" >well-formed</a> (for example "a" or
1413 * "e2"), it will be emitted as "und" (Undetermined).
1414 *
1415 * <p><b>Country:</b> If country is not <a
1416 * href="#def_region">well-formed</a> (for example "12" or "USA"),
1417 * it will be omitted.
1418 *
1419 * <p><b>Variant:</b> If variant <b>is</b> <a
1420 * href="#def_variant">well-formed</a>, each sub-segment
1421 * (delimited by '-' or '_') is emitted as a subtag. Otherwise:
1422 * <ul>
1423 *
1424 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code>
1425 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first
1426 * ill-formed sub-segment and all following will be appended to
1427 * the private use subtag. The first appended subtag will be
1428 * "lvariant", followed by the sub-segments in order, separated by
1429 * hyphen. For example, "x-lvariant-WIN",
1430 * "Oracle-x-lvariant-JDK-Standard-Edition".
1431 *
1432 * <li>if any sub-segment does not match
1433 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated
1434 * and the problematic sub-segment and all following sub-segments
1435 * will be omitted. If the remainder is non-empty, it will be
1436 * emitted as a private use subtag as above (even if the remainder
1437 * turns out to be well-formed). For example,
1438 * "Solaris_isjustthecoolestthing" is emitted as
1439 * "x-lvariant-Solaris", not as "solaris".</li></ul>
1440 *
1441 * <p><b>Special Conversions:</b> Java supports some old locale
1442 * representations, including deprecated ISO language codes,
1443 * for compatibility. This method performs the following
1444 * conversions:
1445 * <ul>
1446 *
1447 * <li>Deprecated ISO language codes "iw", "ji", and "in" are
1448 * converted to "he", "yi", and "id", respectively.
1449 *
1450 * <li>A locale with language "no", country "NO", and variant
1451 * "NY", representing Norwegian Nynorsk (Norway), is converted
1452 * to a language tag "nn-NO".</li></ul>
1453 *
1454 * <p><b>Note:</b> Although the language tag created by this
1455 * method is well-formed (satisfies the syntax requirements
1456 * defined by the IETF BCP 47 specification), it is not
1457 * necessarily a valid BCP 47 language tag. For example,
1458 * <pre>
1459 * new Locale("xx", "YY").toLanguageTag();</pre>
1460 *
1461 * will return "xx-YY", but the language subtag "xx" and the
1462 * region subtag "YY" are invalid because they are not registered
1463 * in the IANA Language Subtag Registry.
1464 *
1465 * @return a BCP47 language tag representing the locale
1466 * @see #forLanguageTag(String)
1467 * @since 1.7
1468 */
1469 public String toLanguageTag() {
1470 if (languageTag != null) {
1471 return languageTag;
1472 }
1473
1474 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions);
1475 StringBuilder buf = new StringBuilder();
1476
1477 String subtag = tag.getLanguage();
1478 if (subtag.length() > 0) {
1479 buf.append(LanguageTag.canonicalizeLanguage(subtag));
1480 }
1481
1482 subtag = tag.getScript();
1483 if (subtag.length() > 0) {
1484 buf.append(LanguageTag.SEP);
1485 buf.append(LanguageTag.canonicalizeScript(subtag));
1486 }
1487
1488 subtag = tag.getRegion();
1489 if (subtag.length() > 0) {
1490 buf.append(LanguageTag.SEP);
1491 buf.append(LanguageTag.canonicalizeRegion(subtag));
1492 }
1493
1494 List<String>subtags = tag.getVariants();
1495 for (String s : subtags) {
1496 buf.append(LanguageTag.SEP);
1497 // preserve casing
1498 buf.append(s);
1499 }
1500
1501 subtags = tag.getExtensions();
1502 for (String s : subtags) {
1503 buf.append(LanguageTag.SEP);
1504 buf.append(LanguageTag.canonicalizeExtension(s));
1505 }
1506
1507 subtag = tag.getPrivateuse();
1508 if (subtag.length() > 0) {
1509 if (buf.length() > 0) {
1510 buf.append(LanguageTag.SEP);
1511 }
1512 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP);
1513 // preserve casing
1514 buf.append(subtag);
1515 }
1516
1517 String langTag = buf.toString();
1518 synchronized (this) {
1519 if (languageTag == null) {
1520 languageTag = langTag;
1521 }
1522 }
1523 return languageTag;
1524 }
1525
1526 /**
1527 * Returns a locale for the specified IETF BCP 47 language tag string.
1528 *
1529 * <p>If the specified language tag contains any ill-formed subtags,
1530 * the first such subtag and all following subtags are ignored. Compare
1531 * to {@link Locale.Builder#setLanguageTag} which throws an exception
1532 * in this case.
1533 *
1534 * <p>The following <b>conversions</b> are performed:<ul>
1535 *
1536 * <li>The language code "und" is mapped to language "".
1537 *
1538 * <li>The language codes "he", "yi", and "id" are mapped to "iw",
1539 * "ji", and "in" respectively. (This is the same canonicalization
1540 * that's done in Locale's constructors.)
1541 *
1542 * <li>The portion of a private use subtag prefixed by "lvariant",
1543 * if any, is removed and appended to the variant field in the
1544 * result locale (without case normalization). If it is then
1545 * empty, the private use subtag is discarded:
1546 *
1547 * <pre>
1548 * Locale loc;
1549 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX");
1550 * loc.getVariant(); // returns "POSIX"
1551 * loc.getExtension('x'); // returns null
1552 *
1553 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def");
1554 * loc.getVariant(); // returns "POSIX_Abc_Def"
1555 * loc.getExtension('x'); // returns "urp"
1556 * </pre>
1557 *
1558 * <li>When the languageTag argument contains an extlang subtag,
1559 * the first such subtag is used as the language, and the primary
1560 * language subtag and other extlang subtags are ignored:
1561 *
1562 * <pre>
1563 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao"
1564 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US"
1565 * </pre>
1566 *
1567 * <li>Case is normalized except for variant tags, which are left
1568 * unchanged. Language is normalized to lower case, script to
1569 * title case, country to upper case, and extensions to lower
1570 * case.
1571 *
1572 * <li>If, after processing, the locale would exactly match either
1573 * ja_JP_JP or th_TH_TH with no extensions, the appropriate
1574 * extensions are added as though the constructor had been called:
1575 *
1576 * <pre>
1577 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag();
1578 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP"
1579 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag();
1580 * // returns "th-TH-u-nu-thai-x-lvariant-TH"
1581 * </pre></ul>
1582 *
1583 * <p>This implements the 'Language-Tag' production of BCP47, and
1584 * so supports grandfathered (regular and irregular) as well as
1585 * private use language tags. Stand alone private use tags are
1586 * represented as empty language and extension 'x-whatever',
1587 * and grandfathered tags are converted to their canonical replacements
1588 * where they exist.
1589 *
1590 * <p>Grandfathered tags with canonical replacements are as follows:
1591 *
1592 * <table summary="Grandfathered tags with canonical replacements">
1593 * <tbody align="center">
1594 * <tr><th>grandfathered tag</th><th> </th><th>modern replacement</th></tr>
1595 * <tr><td>art-lojban</td><td> </td><td>jbo</td></tr>
1596 * <tr><td>i-ami</td><td> </td><td>ami</td></tr>
1597 * <tr><td>i-bnn</td><td> </td><td>bnn</td></tr>
1598 * <tr><td>i-hak</td><td> </td><td>hak</td></tr>
1599 * <tr><td>i-klingon</td><td> </td><td>tlh</td></tr>
1600 * <tr><td>i-lux</td><td> </td><td>lb</td></tr>
1601 * <tr><td>i-navajo</td><td> </td><td>nv</td></tr>
1602 * <tr><td>i-pwn</td><td> </td><td>pwn</td></tr>
1603 * <tr><td>i-tao</td><td> </td><td>tao</td></tr>
1604 * <tr><td>i-tay</td><td> </td><td>tay</td></tr>
1605 * <tr><td>i-tsu</td><td> </td><td>tsu</td></tr>
1606 * <tr><td>no-bok</td><td> </td><td>nb</td></tr>
1607 * <tr><td>no-nyn</td><td> </td><td>nn</td></tr>
1608 * <tr><td>sgn-BE-FR</td><td> </td><td>sfb</td></tr>
1609 * <tr><td>sgn-BE-NL</td><td> </td><td>vgt</td></tr>
1610 * <tr><td>sgn-CH-DE</td><td> </td><td>sgg</td></tr>
1611 * <tr><td>zh-guoyu</td><td> </td><td>cmn</td></tr>
1612 * <tr><td>zh-hakka</td><td> </td><td>hak</td></tr>
1613 * <tr><td>zh-min-nan</td><td> </td><td>nan</td></tr>
1614 * <tr><td>zh-xiang</td><td> </td><td>hsn</td></tr>
1615 * </tbody>
1616 * </table>
1617 *
1618 * <p>Grandfathered tags with no modern replacement will be
1619 * converted as follows:
1620 *
1621 * <table summary="Grandfathered tags with no modern replacement">
1622 * <tbody align="center">
1623 * <tr><th>grandfathered tag</th><th> </th><th>converts to</th></tr>
1624 * <tr><td>cel-gaulish</td><td> </td><td>xtg-x-cel-gaulish</td></tr>
1625 * <tr><td>en-GB-oed</td><td> </td><td>en-GB-x-oed</td></tr>
1626 * <tr><td>i-default</td><td> </td><td>en-x-i-default</td></tr>
1627 * <tr><td>i-enochian</td><td> </td><td>und-x-i-enochian</td></tr>
1628 * <tr><td>i-mingo</td><td> </td><td>see-x-i-mingo</td></tr>
1629 * <tr><td>zh-min</td><td> </td><td>nan-x-zh-min</td></tr>
1630 * </tbody>
1631 * </table>
1632 *
1633 * <p>For a list of all grandfathered tags, see the
1634 * IANA Language Subtag Registry (search for "Type: grandfathered").
1635 *
1636 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code>
1637 * and <code>forLanguageTag</code> will round-trip.
1638 *
1639 * @param languageTag the language tag
1640 * @return The locale that best represents the language tag.
1641 * @throws NullPointerException if <code>languageTag</code> is <code>null</code>
1642 * @see #toLanguageTag()
1643 * @see java.util.Locale.Builder#setLanguageTag(String)
1644 * @since 1.7
1645 */
1646 public static Locale forLanguageTag(String languageTag) {
1647 LanguageTag tag = LanguageTag.parse(languageTag, null);
1648 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
1649 bldr.setLanguageTag(tag);
1650 BaseLocale base = bldr.getBaseLocale();
1651 LocaleExtensions exts = bldr.getLocaleExtensions();
1652 if (exts == null && base.getVariant().length() > 0) {
1653 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(),
1654 base.getRegion(), base.getVariant());
1655 }
1656 return getInstance(base, exts);
1657 }
1658
1659 /**
1660 * Returns a three-letter abbreviation of this locale's language.
1661 * If the language matches an ISO 639-1 two-letter code, the
1662 * corresponding ISO 639-2/T three-letter lowercase code is
1663 * returned. The ISO 639-2 language codes can be found on-line,
1664 * see "Codes for the Representation of Names of Languages Part 2:
1665 * Alpha-3 Code". If the locale specifies a three-letter
1666 * language, the language is returned as is. If the locale does
1667 * not specify a language the empty string is returned.
1668 *
1669 * @return A three-letter abbreviation of this locale's language.
1670 * @exception MissingResourceException Throws MissingResourceException if
1671 * three-letter language abbreviation is not available for this locale.
1672 */
1673 public String getISO3Language() throws MissingResourceException {
1674 String lang = baseLocale.getLanguage();
1675 if (lang.length() == 3) {
1676 return lang;
1677 }
1678
1679 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable);
1680 if (language3 == null) {
1681 throw new MissingResourceException("Couldn't find 3-letter language code for "
1682 + lang, "FormatData_" + toString(), "ShortLanguage");
1683 }
1684 return language3;
1685 }
1686
1687 /**
1688 * Returns a three-letter abbreviation for this locale's country.
1689 * If the country matches an ISO 3166-1 alpha-2 code, the
1690 * corresponding ISO 3166-1 alpha-3 uppercase code is returned.
1691 * If the locale doesn't specify a country, this will be the empty
1692 * string.
1693 *
1694 * <p>The ISO 3166-1 codes can be found on-line.
1695 *
1696 * @return A three-letter abbreviation of this locale's country.
1697 * @exception MissingResourceException Throws MissingResourceException if the
1698 * three-letter country abbreviation is not available for this locale.
1699 */
1700 public String getISO3Country() throws MissingResourceException {
1701 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable);
1702 if (country3 == null) {
1703 throw new MissingResourceException("Couldn't find 3-letter country code for "
1704 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry");
1705 }
1706 return country3;
1707 }
1708
1709 private static String getISO3Code(String iso2Code, String table) {
1710 int codeLength = iso2Code.length();
1711 if (codeLength == 0) {
1712 return "";
1713 }
1714
1715 int tableLength = table.length();
1716 int index = tableLength;
1717 if (codeLength == 2) {
1718 char c1 = iso2Code.charAt(0);
1719 char c2 = iso2Code.charAt(1);
1720 for (index = 0; index < tableLength; index += 5) {
1721 if (table.charAt(index) == c1
1722 && table.charAt(index + 1) == c2) {
1723 break;
1724 }
1725 }
1726 }
1727 return index < tableLength ? table.substring(index + 2, index + 5) : null;
1728 }
1729
1730 /**
1731 * Returns a name for the locale's language that is appropriate for display to the
1732 * user.
1733 * If possible, the name returned will be localized for the default
1734 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1735 * For example, if the locale is fr_FR and the default
1736 * {@link Locale.Category#DISPLAY DISPLAY} locale
1737 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1738 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1739 * getDisplayLanguage() will return "anglais".
1740 * If the name returned cannot be localized for the default
1741 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1742 * (say, we don't have a Japanese name for Croatian),
1743 * this function falls back on the English name, and uses the ISO code as a last-resort
1744 * value. If the locale doesn't specify a language, this function returns the empty string.
1745 *
1746 * @return The name of the display language.
1747 */
1748 public final String getDisplayLanguage() {
1749 return getDisplayLanguage(getDefault(Category.DISPLAY));
1750 }
1751
1752 /**
1753 * Returns a name for the locale's language that is appropriate for display to the
1754 * user.
1755 * If possible, the name returned will be localized according to inLocale.
1756 * For example, if the locale is fr_FR and inLocale
1757 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1758 * inLocale is fr_FR, getDisplayLanguage() will return "anglais".
1759 * If the name returned cannot be localized according to inLocale,
1760 * (say, we don't have a Japanese name for Croatian),
1761 * this function falls back on the English name, and finally
1762 * on the ISO code as a last-resort value. If the locale doesn't specify a language,
1763 * this function returns the empty string.
1764 *
1765 * @param inLocale The locale for which to retrieve the display language.
1766 * @return The name of the display language appropriate to the given locale.
1767 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1768 */
1769 public String getDisplayLanguage(Locale inLocale) {
1770 return getDisplayString(baseLocale.getLanguage(), inLocale, DISPLAY_LANGUAGE);
1771 }
1772
1773 /**
1774 * Returns a name for the locale's script that is appropriate for display to
1775 * the user. If possible, the name will be localized for the default
1776 * {@link Locale.Category#DISPLAY DISPLAY} locale. Returns
1777 * the empty string if this locale doesn't specify a script code.
1778 *
1779 * @return the display name of the script code for the current default
1780 * {@link Locale.Category#DISPLAY DISPLAY} locale
1781 * @since 1.7
1782 */
1783 public String getDisplayScript() {
1784 return getDisplayScript(getDefault(Category.DISPLAY));
1785 }
1786
1787 /**
1788 * Returns a name for the locale's script that is appropriate
1789 * for display to the user. If possible, the name will be
1790 * localized for the given locale. Returns the empty string if
1791 * this locale doesn't specify a script code.
1792 *
1793 * @param inLocale The locale for which to retrieve the display script.
1794 * @return the display name of the script code for the current default
1795 * {@link Locale.Category#DISPLAY DISPLAY} locale
1796 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1797 * @since 1.7
1798 */
1799 public String getDisplayScript(Locale inLocale) {
1800 return getDisplayString(baseLocale.getScript(), inLocale, DISPLAY_SCRIPT);
1801 }
1802
1803 /**
1804 * Returns a name for the locale's country that is appropriate for display to the
1805 * user.
1806 * If possible, the name returned will be localized for the default
1807 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1808 * For example, if the locale is fr_FR and the default
1809 * {@link Locale.Category#DISPLAY DISPLAY} locale
1810 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1811 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1812 * getDisplayCountry() will return "Etats-Unis".
1813 * If the name returned cannot be localized for the default
1814 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1815 * (say, we don't have a Japanese name for Croatia),
1816 * this function falls back on the English name, and uses the ISO code as a last-resort
1817 * value. If the locale doesn't specify a country, this function returns the empty string.
1818 *
1819 * @return The name of the country appropriate to the locale.
1820 */
1821 public final String getDisplayCountry() {
1822 return getDisplayCountry(getDefault(Category.DISPLAY));
1823 }
1824
1825 /**
1826 * Returns a name for the locale's country that is appropriate for display to the
1827 * user.
1828 * If possible, the name returned will be localized according to inLocale.
1829 * For example, if the locale is fr_FR and inLocale
1830 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1831 * inLocale is fr_FR, getDisplayCountry() will return "Etats-Unis".
1832 * If the name returned cannot be localized according to inLocale.
1833 * (say, we don't have a Japanese name for Croatia),
1834 * this function falls back on the English name, and finally
1835 * on the ISO code as a last-resort value. If the locale doesn't specify a country,
1836 * this function returns the empty string.
1837 *
1838 * @param inLocale The locale for which to retrieve the display country.
1839 * @return The name of the country appropriate to the given locale.
1840 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1841 */
1842 public String getDisplayCountry(Locale inLocale) {
1843 return getDisplayString(baseLocale.getRegion(), inLocale, DISPLAY_COUNTRY);
1844 }
1845
1846 private String getDisplayString(String code, Locale inLocale, int type) {
1847 if (code.length() == 0) {
1848 return "";
1849 }
1850
1851 if (inLocale == null) {
1852 throw new NullPointerException();
1853 }
1854
1855 LocaleServiceProviderPool pool =
1856 LocaleServiceProviderPool.getPool(LocaleNameProvider.class);
1857 String key = (type == DISPLAY_VARIANT ? "%%"+code : code);
1858 String result = pool.getLocalizedObject(
1859 LocaleNameGetter.INSTANCE,
1860 inLocale, key, type, code);
1861 if (result != null) {
1862 return result;
1863 }
1864
1865 return code;
1866 }
1867
1868 /**
1869 * Returns a name for the locale's variant code that is appropriate for display to the
1870 * user. If possible, the name will be localized for the default
1871 * {@link Locale.Category#DISPLAY DISPLAY} locale. If the locale
1872 * doesn't specify a variant code, this function returns the empty string.
1873 *
1874 * @return The name of the display variant code appropriate to the locale.
1875 */
1876 public final String getDisplayVariant() {
1877 return getDisplayVariant(getDefault(Category.DISPLAY));
1878 }
1879
1880 /**
1881 * Returns a name for the locale's variant code that is appropriate for display to the
1882 * user. If possible, the name will be localized for inLocale. If the locale
1883 * doesn't specify a variant code, this function returns the empty string.
1884 *
1885 * @param inLocale The locale for which to retrieve the display variant code.
1886 * @return The name of the display variant code appropriate to the given locale.
1887 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1888 */
1889 public String getDisplayVariant(Locale inLocale) {
1890 if (baseLocale.getVariant().length() == 0)
1891 return "";
1892
1893 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale);
1894
1895 String names[] = getDisplayVariantArray(inLocale);
1896
1897 // Get the localized patterns for formatting a list, and use
1898 // them to format the list.
1899 return formatList(names,
1900 lr.getLocaleName("ListPattern"),
1901 lr.getLocaleName("ListCompositionPattern"));
1902 }
1903
1904 /**
1905 * Returns a name for the locale that is appropriate for display to the
1906 * user. This will be the values returned by getDisplayLanguage(),
1907 * getDisplayScript(), getDisplayCountry(), and getDisplayVariant() assembled
1908 * into a single string. The the non-empty values are used in order,
1909 * with the second and subsequent names in parentheses. For example:
1910 * <blockquote>
1911 * language (script, country, variant)<br>
1912 * language (country)<br>
1913 * language (variant)<br>
1914 * script (country)<br>
1915 * country<br>
1916 * </blockquote>
1917 * depending on which fields are specified in the locale. If the
1918 * language, script, country, and variant fields are all empty,
1919 * this function returns the empty string.
1920 *
1921 * @return The name of the locale appropriate to display.
1922 */
1923 public final String getDisplayName() {
1924 return getDisplayName(getDefault(Category.DISPLAY));
1925 }
1926
1927 /**
1928 * Returns a name for the locale that is appropriate for display
1929 * to the user. This will be the values returned by
1930 * getDisplayLanguage(), getDisplayScript(),getDisplayCountry(),
1931 * and getDisplayVariant() assembled into a single string.
1932 * The non-empty values are used in order,
1933 * with the second and subsequent names in parentheses. For example:
1934 * <blockquote>
1935 * language (script, country, variant)<br>
1936 * language (country)<br>
1937 * language (variant)<br>
1938 * script (country)<br>
1939 * country<br>
1940 * </blockquote>
1941 * depending on which fields are specified in the locale. If the
1942 * language, script, country, and variant fields are all empty,
1943 * this function returns the empty string.
1944 *
1945 * @param inLocale The locale for which to retrieve the display name.
1946 * @return The name of the locale appropriate to display.
1947 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1948 */
1949 public String getDisplayName(Locale inLocale) {
1950 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale);
1951
1952 String languageName = getDisplayLanguage(inLocale);
1953 String scriptName = getDisplayScript(inLocale);
1954 String countryName = getDisplayCountry(inLocale);
1955 String[] variantNames = getDisplayVariantArray(inLocale);
1956
1957 // Get the localized patterns for formatting a display name.
1958 String displayNamePattern = lr.getLocaleName("DisplayNamePattern");
1959 String listPattern = lr.getLocaleName("ListPattern");
1960 String listCompositionPattern = lr.getLocaleName("ListCompositionPattern");
1961
1962 // The display name consists of a main name, followed by qualifiers.
1963 // Typically, the format is "MainName (Qualifier, Qualifier)" but this
1964 // depends on what pattern is stored in the display locale.
1965 String mainName = null;
1966 String[] qualifierNames = null;
1967
1968 // The main name is the language, or if there is no language, the script,
1969 // then if no script, the country. If there is no language/script/country
1970 // (an anomalous situation) then the display name is simply the variant's
1971 // display name.
1972 if (languageName.length() == 0 && scriptName.length() == 0 && countryName.length() == 0) {
1973 if (variantNames.length == 0) {
1974 return "";
1975 } else {
1976 return formatList(variantNames, listPattern, listCompositionPattern);
1977 }
1978 }
1979 ArrayList<String> names = new ArrayList<>(4);
1980 if (languageName.length() != 0) {
1981 names.add(languageName);
1982 }
1983 if (scriptName.length() != 0) {
1984 names.add(scriptName);
1985 }
1986 if (countryName.length() != 0) {
1987 names.add(countryName);
1988 }
1989 if (variantNames.length != 0) {
1990 names.addAll(Arrays.asList(variantNames));
1991 }
1992
1993 // The first one in the main name
1994 mainName = names.get(0);
1995
1996 // Others are qualifiers
1997 int numNames = names.size();
1998 qualifierNames = (numNames > 1) ?
1999 names.subList(1, numNames).toArray(new String[numNames - 1]) : new String[0];
2000
2001 // Create an array whose first element is the number of remaining
2002 // elements. This serves as a selector into a ChoiceFormat pattern from
2003 // the resource. The second and third elements are the main name and
2004 // the qualifier; if there are no qualifiers, the third element is
2005 // unused by the format pattern.
2006 Object[] displayNames = {
2007 qualifierNames.length != 0 ? 2 : 1,
2008 mainName,
2009 // We could also just call formatList() and have it handle the empty
2010 // list case, but this is more efficient, and we want it to be
2011 // efficient since all the language-only locales will not have any
2012 // qualifiers.
2013 qualifierNames.length != 0 ? formatList(qualifierNames, listPattern, listCompositionPattern) : null
2014 };
2015
2016 if (displayNamePattern != null) {
2017 return new MessageFormat(displayNamePattern).format(displayNames);
2018 }
2019 else {
2020 // If we cannot get the message format pattern, then we use a simple
2021 // hard-coded pattern. This should not occur in practice unless the
2022 // installation is missing some core files (FormatData etc.).
2023 StringBuilder result = new StringBuilder();
2024 result.append((String)displayNames[1]);
2025 if (displayNames.length > 2) {
2026 result.append(" (");
2027 result.append((String)displayNames[2]);
2028 result.append(')');
2029 }
2030 return result.toString();
2031 }
2032 }
2033
2034 /**
2035 * Overrides Cloneable.
2036 */
2037 @Override
2038 public Object clone()
2039 {
2040 try {
2041 Locale that = (Locale)super.clone();
2042 return that;
2043 } catch (CloneNotSupportedException e) {
2044 throw new InternalError(e);
2045 }
2046 }
2047
2048 /**
2049 * Override hashCode.
2050 * Since Locales are often used in hashtables, caches the value
2051 * for speed.
2052 */
2053 @Override
2054 public int hashCode() {
2055 int hc = hashCodeValue;
2056 if (hc == 0) {
2057 hc = baseLocale.hashCode();
2058 if (localeExtensions != null) {
2059 hc ^= localeExtensions.hashCode();
2060 }
2061 hashCodeValue = hc;
2062 }
2063 return hc;
2064 }
2065
2066 // Overrides
2067
2068 /**
2069 * Returns true if this Locale is equal to another object. A Locale is
2070 * deemed equal to another Locale with identical language, script, country,
2071 * variant and extensions, and unequal to all other objects.
2072 *
2073 * @return true if this Locale is equal to the specified object.
2074 */
2075 @Override
2076 public boolean equals(Object obj) {
2077 if (this == obj) // quick check
2078 return true;
2079 if (!(obj instanceof Locale))
2080 return false;
2081 BaseLocale otherBase = ((Locale)obj).baseLocale;
2082 if (!baseLocale.equals(otherBase)) {
2083 return false;
2084 }
2085 if (localeExtensions == null) {
2086 return ((Locale)obj).localeExtensions == null;
2087 }
2088 return localeExtensions.equals(((Locale)obj).localeExtensions);
2089 }
2090
2091 // ================= privates =====================================
2092
2093 private transient BaseLocale baseLocale;
2094 private transient LocaleExtensions localeExtensions;
2095
2096 /**
2097 * Calculated hashcode
2098 */
2099 private transient volatile int hashCodeValue;
2100
2101 private static volatile Locale defaultLocale = initDefault();
2102 private static volatile Locale defaultDisplayLocale;
2103 private static volatile Locale defaultFormatLocale;
2104
2105 private transient volatile String languageTag;
2106
2107 /**
2108 * Return an array of the display names of the variant.
2109 * @param bundle the ResourceBundle to use to get the display names
2110 * @return an array of display names, possible of zero length.
2111 */
2112 private String[] getDisplayVariantArray(Locale inLocale) {
2113 // Split the variant name into tokens separated by '_'.
2114 StringTokenizer tokenizer = new StringTokenizer(baseLocale.getVariant(), "_");
2115 String[] names = new String[tokenizer.countTokens()];
2116
2117 // For each variant token, lookup the display name. If
2118 // not found, use the variant name itself.
2119 for (int i=0; i<names.length; ++i) {
2120 names[i] = getDisplayString(tokenizer.nextToken(),
2121 inLocale, DISPLAY_VARIANT);
2122 }
2123
2124 return names;
2125 }
2126
2127 /**
2128 * Format a list using given pattern strings.
2129 * If either of the patterns is null, then a the list is
2130 * formatted by concatenation with the delimiter ','.
2131 * @param stringList the list of strings to be formatted.
2132 * @param listPattern should create a MessageFormat taking 0-3 arguments
2133 * and formatting them into a list.
2134 * @param listCompositionPattern should take 2 arguments
2135 * and is used by composeList.
2136 * @return a string representing the list.
2137 */
2138 private static String formatList(String[] stringList, String listPattern, String listCompositionPattern) {
2139 // If we have no list patterns, compose the list in a simple,
2140 // non-localized way.
2141 if (listPattern == null || listCompositionPattern == null) {
2142 StringJoiner sj = new StringJoiner(",");
2143 for (int i = 0; i < stringList.length; ++i) {
2144 sj.add(stringList[i]);
2145 }
2146 return sj.toString();
2147 }
2148
2149 // Compose the list down to three elements if necessary
2150 if (stringList.length > 3) {
2151 MessageFormat format = new MessageFormat(listCompositionPattern);
2152 stringList = composeList(format, stringList);
2153 }
2154
2155 // Rebuild the argument list with the list length as the first element
2156 Object[] args = new Object[stringList.length + 1];
2157 System.arraycopy(stringList, 0, args, 1, stringList.length);
2158 args[0] = stringList.length;
2159
2160 // Format it using the pattern in the resource
2161 MessageFormat format = new MessageFormat(listPattern);
2162 return format.format(args);
2163 }
2164
2165 /**
2166 * Given a list of strings, return a list shortened to three elements.
2167 * Shorten it by applying the given format to the first two elements
2168 * recursively.
2169 * @param format a format which takes two arguments
2170 * @param list a list of strings
2171 * @return if the list is three elements or shorter, the same list;
2172 * otherwise, a new list of three elements.
2173 */
2174 private static String[] composeList(MessageFormat format, String[] list) {
2175 if (list.length <= 3) return list;
2176
2177 // Use the given format to compose the first two elements into one
2178 String[] listItems = { list[0], list[1] };
2179 String newItem = format.format(listItems);
2180
2181 // Form a new list one element shorter
2182 String[] newList = new String[list.length-1];
2183 System.arraycopy(list, 2, newList, 1, newList.length-1);
2184 newList[0] = newItem;
2185
2186 // Recurse
2187 return composeList(format, newList);
2188 }
2189
2190 // Duplicate of sun.util.locale.UnicodeLocaleExtension.isKey in order to
2191 // avoid its class loading.
2192 private static boolean isUnicodeExtensionKey(String s) {
2193 // 2alphanum
2194 return (s.length() == 2) && LocaleUtils.isAlphaNumericString(s);
2195 }
2196
2197 /**
2198 * @serialField language String
2199 * language subtag in lower case. (See <a href="java/util/Locale.html#getLanguage()">getLanguage()</a>)
2200 * @serialField country String
2201 * country subtag in upper case. (See <a href="java/util/Locale.html#getCountry()">getCountry()</a>)
2202 * @serialField variant String
2203 * variant subtags separated by LOWLINE characters. (See <a href="java/util/Locale.html#getVariant()">getVariant()</a>)
2204 * @serialField hashcode int
2205 * deprecated, for forward compatibility only
2206 * @serialField script String
2207 * script subtag in title case (See <a href="java/util/Locale.html#getScript()">getScript()</a>)
2208 * @serialField extensions String
2209 * canonical representation of extensions, that is,
2210 * BCP47 extensions in alphabetical order followed by
2211 * BCP47 private use subtags, all in lower case letters
2212 * separated by HYPHEN-MINUS characters.
2213 * (See <a href="java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>,
2214 * <a href="java/util/Locale.html#getExtension(char)">getExtension(char)</a>)
2215 */
2216 private static final ObjectStreamField[] serialPersistentFields = {
2217 new ObjectStreamField("language", String.class),
2218 new ObjectStreamField("country", String.class),
2219 new ObjectStreamField("variant", String.class),
2220 new ObjectStreamField("hashcode", int.class),
2221 new ObjectStreamField("script", String.class),
2222 new ObjectStreamField("extensions", String.class),
2223 };
2224
2225 /**
2226 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>.
2227 * @param out the <code>ObjectOutputStream</code> to write
2228 * @throws IOException
2229 * @since 1.7
2230 */
2231 private void writeObject(ObjectOutputStream out) throws IOException {
2232 ObjectOutputStream.PutField fields = out.putFields();
2233 fields.put("language", baseLocale.getLanguage());
2234 fields.put("script", baseLocale.getScript());
2235 fields.put("country", baseLocale.getRegion());
2236 fields.put("variant", baseLocale.getVariant());
2237 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID());
2238 fields.put("hashcode", -1); // place holder just for backward support
2239 out.writeFields();
2240 }
2241
2242 /**
2243 * Deserializes this <code>Locale</code>.
2244 * @param in the <code>ObjectInputStream</code> to read
2245 * @throws IOException
2246 * @throws ClassNotFoundException
2247 * @throws IllformedLocaleException
2248 * @since 1.7
2249 */
2250 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException {
2251 ObjectInputStream.GetField fields = in.readFields();
2252 String language = (String)fields.get("language", "");
2253 String script = (String)fields.get("script", "");
2254 String country = (String)fields.get("country", "");
2255 String variant = (String)fields.get("variant", "");
2256 String extStr = (String)fields.get("extensions", "");
2257 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant);
2258 if (extStr.length() > 0) {
2259 try {
2260 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
2261 bldr.setExtensions(extStr);
2262 localeExtensions = bldr.getLocaleExtensions();
2263 } catch (LocaleSyntaxException e) {
2264 throw new IllformedLocaleException(e.getMessage());
2265 }
2266 } else {
2267 localeExtensions = null;
2268 }
2269 }
2270
2271 /**
2272 * Returns a cached <code>Locale</code> instance equivalent to
2273 * the deserialized <code>Locale</code>. When serialized
2274 * language, country and variant fields read from the object data stream
2275 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions
2276 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code>
2277 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script
2278 * type is "thai"). See <a href="Locale.html#special_cases_constructor">Special Cases</a>
2279 * for more information.
2280 *
2281 * @return an instance of <code>Locale</code> equivalent to
2282 * the deserialized <code>Locale</code>.
2283 * @throws java.io.ObjectStreamException
2284 */
2285 private Object readResolve() throws java.io.ObjectStreamException {
2286 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(),
2287 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions);
2288 }
2289
2290 private static volatile String[] isoLanguages;
2291
2292 private static volatile String[] isoCountries;
2293
2294 private static String convertOldISOCodes(String language) {
2295 // we accept both the old and the new ISO codes for the languages whose ISO
2296 // codes have changed, but we always store the OLD code, for backward compatibility
2297 language = LocaleUtils.toLowerString(language).intern();
2298 if (language == "he") {
2299 return "iw";
2300 } else if (language == "yi") {
2301 return "ji";
2302 } else if (language == "id") {
2303 return "in";
2304 } else {
2305 return language;
2306 }
2307 }
2308
2309 private static LocaleExtensions getCompatibilityExtensions(String language,
2310 String script,
2311 String country,
2312 String variant) {
2313 LocaleExtensions extensions = null;
2314 // Special cases for backward compatibility support
2315 if (LocaleUtils.caseIgnoreMatch(language, "ja")
2316 && script.length() == 0
2317 && LocaleUtils.caseIgnoreMatch(country, "jp")
2318 && "JP".equals(variant)) {
2319 // ja_JP_JP -> u-ca-japanese (calendar = japanese)
2320 extensions = LocaleExtensions.CALENDAR_JAPANESE;
2321 } else if (LocaleUtils.caseIgnoreMatch(language, "th")
2322 && script.length() == 0
2323 && LocaleUtils.caseIgnoreMatch(country, "th")
2324 && "TH".equals(variant)) {
2325 // th_TH_TH -> u-nu-thai (numbersystem = thai)
2326 extensions = LocaleExtensions.NUMBER_THAI;
2327 }
2328 return extensions;
2329 }
2330
2331 /**
2332 * Obtains a localized locale names from a LocaleNameProvider
2333 * implementation.
2334 */
2335 private static class LocaleNameGetter
2336 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> {
2337 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter();
2338
2339 @Override
2340 public String getObject(LocaleNameProvider localeNameProvider,
2341 Locale locale,
2342 String key,
2343 Object... params) {
2344 assert params.length == 2;
2345 int type = (Integer)params[0];
2346 String code = (String)params[1];
2347
2348 switch(type) {
2349 case DISPLAY_LANGUAGE:
2350 return localeNameProvider.getDisplayLanguage(code, locale);
2351 case DISPLAY_COUNTRY:
2352 return localeNameProvider.getDisplayCountry(code, locale);
2353 case DISPLAY_VARIANT:
2354 return localeNameProvider.getDisplayVariant(code, locale);
2355 case DISPLAY_SCRIPT:
2356 return localeNameProvider.getDisplayScript(code, locale);
2357 default:
2358 assert false; // shouldn't happen
2359 }
2360
2361 return null;
2362 }
2363 }
2364
2365 /**
2366 * Enum for locale categories. These locale categories are used to get/set
2367 * the default locale for the specific functionality represented by the
2368 * category.
2369 *
2370 * @see #getDefault(Locale.Category)
2371 * @see #setDefault(Locale.Category, Locale)
2372 * @since 1.7
2373 */
2374 public enum Category {
2375
2376 /**
2377 * Category used to represent the default locale for
2378 * displaying user interfaces.
2379 */
2380 DISPLAY("user.language.display",
2381 "user.script.display",
2382 "user.country.display",
2383 "user.variant.display"),
2384
2385 /**
2386 * Category used to represent the default locale for
2387 * formatting dates, numbers, and/or currencies.
2388 */
2389 FORMAT("user.language.format",
2390 "user.script.format",
2391 "user.country.format",
2392 "user.variant.format");
2393
2394 Category(String languageKey, String scriptKey, String countryKey, String variantKey) {
2395 this.languageKey = languageKey;
2396 this.scriptKey = scriptKey;
2397 this.countryKey = countryKey;
2398 this.variantKey = variantKey;
2399 }
2400
2401 final String languageKey;
2402 final String scriptKey;
2403 final String countryKey;
2404 final String variantKey;
2405 }
2406
2407 /**
2408 * <code>Builder</code> is used to build instances of <code>Locale</code>
2409 * from values configured by the setters. Unlike the <code>Locale</code>
2410 * constructors, the <code>Builder</code> checks if a value configured by a
2411 * setter satisfies the syntax requirements defined by the <code>Locale</code>
2412 * class. A <code>Locale</code> object created by a <code>Builder</code> is
2413 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag
2414 * without losing information.
2415 *
2416 * <p><b>Note:</b> The <code>Locale</code> class does not provide any
2417 * syntactic restrictions on variant, while BCP 47 requires each variant
2418 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3
2419 * alphanumerics. The method <code>setVariant</code> throws
2420 * <code>IllformedLocaleException</code> for a variant that does not satisfy
2421 * this restriction. If it is necessary to support such a variant, use a
2422 * Locale constructor. However, keep in mind that a <code>Locale</code>
2423 * object created this way might lose the variant information when
2424 * transformed to a BCP 47 language tag.
2425 *
2426 * <p>The following example shows how to create a <code>Locale</code> object
2427 * with the <code>Builder</code>.
2428 * <blockquote>
2429 * <pre>
2430 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build();
2431 * </pre>
2432 * </blockquote>
2433 *
2434 * <p>Builders can be reused; <code>clear()</code> resets all
2435 * fields to their default values.
2436 *
2437 * @see Locale#forLanguageTag
2438 * @since 1.7
2439 */
2440 public static final class Builder {
2441 private final InternalLocaleBuilder localeBuilder;
2442
2443 /**
2444 * Constructs an empty Builder. The default value of all
2445 * fields, extensions, and private use information is the
2446 * empty string.
2447 */
2448 public Builder() {
2449 localeBuilder = new InternalLocaleBuilder();
2450 }
2451
2452 /**
2453 * Resets the <code>Builder</code> to match the provided
2454 * <code>locale</code>. Existing state is discarded.
2455 *
2456 * <p>All fields of the locale must be well-formed, see {@link Locale}.
2457 *
2458 * <p>Locales with any ill-formed fields cause
2459 * <code>IllformedLocaleException</code> to be thrown, except for the
2460 * following three cases which are accepted for compatibility
2461 * reasons:<ul>
2462 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese"
2463 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai"
2464 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul>
2465 *
2466 * @param locale the locale
2467 * @return This builder.
2468 * @throws IllformedLocaleException if <code>locale</code> has
2469 * any ill-formed fields.
2470 * @throws NullPointerException if <code>locale</code> is null.
2471 */
2472 public Builder setLocale(Locale locale) {
2473 try {
2474 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions);
2475 } catch (LocaleSyntaxException e) {
2476 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2477 }
2478 return this;
2479 }
2480
2481 /**
2482 * Resets the Builder to match the provided IETF BCP 47
2483 * language tag. Discards the existing state. Null and the
2484 * empty string cause the builder to be reset, like {@link
2485 * #clear}. Grandfathered tags (see {@link
2486 * Locale#forLanguageTag}) are converted to their canonical
2487 * form before being processed. Otherwise, the language tag
2488 * must be well-formed (see {@link Locale}) or an exception is
2489 * thrown (unlike <code>Locale.forLanguageTag</code>, which
2490 * just discards ill-formed and following portions of the
2491 * tag).
2492 *
2493 * @param languageTag the language tag
2494 * @return This builder.
2495 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed
2496 * @see Locale#forLanguageTag(String)
2497 */
2498 public Builder setLanguageTag(String languageTag) {
2499 ParseStatus sts = new ParseStatus();
2500 LanguageTag tag = LanguageTag.parse(languageTag, sts);
2501 if (sts.isError()) {
2502 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex());
2503 }
2504 localeBuilder.setLanguageTag(tag);
2505 return this;
2506 }
2507
2508 /**
2509 * Sets the language. If <code>language</code> is the empty string or
2510 * null, the language in this <code>Builder</code> is removed. Otherwise,
2511 * the language must be <a href="./Locale.html#def_language">well-formed</a>
2512 * or an exception is thrown.
2513 *
2514 * <p>The typical language value is a two or three-letter language
2515 * code as defined in ISO639.
2516 *
2517 * @param language the language
2518 * @return This builder.
2519 * @throws IllformedLocaleException if <code>language</code> is ill-formed
2520 */
2521 public Builder setLanguage(String language) {
2522 try {
2523 localeBuilder.setLanguage(language);
2524 } catch (LocaleSyntaxException e) {
2525 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2526 }
2527 return this;
2528 }
2529
2530 /**
2531 * Sets the script. If <code>script</code> is null or the empty string,
2532 * the script in this <code>Builder</code> is removed.
2533 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an
2534 * exception is thrown.
2535 *
2536 * <p>The typical script value is a four-letter script code as defined by ISO 15924.
2537 *
2538 * @param script the script
2539 * @return This builder.
2540 * @throws IllformedLocaleException if <code>script</code> is ill-formed
2541 */
2542 public Builder setScript(String script) {
2543 try {
2544 localeBuilder.setScript(script);
2545 } catch (LocaleSyntaxException e) {
2546 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2547 }
2548 return this;
2549 }
2550
2551 /**
2552 * Sets the region. If region is null or the empty string, the region
2553 * in this <code>Builder</code> is removed. Otherwise,
2554 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an
2555 * exception is thrown.
2556 *
2557 * <p>The typical region value is a two-letter ISO 3166 code or a
2558 * three-digit UN M.49 area code.
2559 *
2560 * <p>The country value in the <code>Locale</code> created by the
2561 * <code>Builder</code> is always normalized to upper case.
2562 *
2563 * @param region the region
2564 * @return This builder.
2565 * @throws IllformedLocaleException if <code>region</code> is ill-formed
2566 */
2567 public Builder setRegion(String region) {
2568 try {
2569 localeBuilder.setRegion(region);
2570 } catch (LocaleSyntaxException e) {
2571 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2572 }
2573 return this;
2574 }
2575
2576 /**
2577 * Sets the variant. If variant is null or the empty string, the
2578 * variant in this <code>Builder</code> is removed. Otherwise, it
2579 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a>
2580 * subtags, or an exception is thrown.
2581 *
2582 * <p><b>Note:</b> This method checks if <code>variant</code>
2583 * satisfies the IETF BCP 47 variant subtag's syntax requirements,
2584 * and normalizes the value to lowercase letters. However,
2585 * the <code>Locale</code> class does not impose any syntactic
2586 * restriction on variant, and the variant value in
2587 * <code>Locale</code> is case sensitive. To set such a variant,
2588 * use a Locale constructor.
2589 *
2590 * @param variant the variant
2591 * @return This builder.
2592 * @throws IllformedLocaleException if <code>variant</code> is ill-formed
2593 */
2594 public Builder setVariant(String variant) {
2595 try {
2596 localeBuilder.setVariant(variant);
2597 } catch (LocaleSyntaxException e) {
2598 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2599 }
2600 return this;
2601 }
2602
2603 /**
2604 * Sets the extension for the given key. If the value is null or the
2605 * empty string, the extension is removed. Otherwise, the extension
2606 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception
2607 * is thrown.
2608 *
2609 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION
2610 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension.
2611 * Setting a value for this key replaces any existing Unicode locale key/type
2612 * pairs with those defined in the extension.
2613 *
2614 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION
2615 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be
2616 * well-formed, the value for this key needs only to have subtags of one to
2617 * eight alphanumeric characters, not two to eight as in the general case.
2618 *
2619 * @param key the extension key
2620 * @param value the extension value
2621 * @return This builder.
2622 * @throws IllformedLocaleException if <code>key</code> is illegal
2623 * or <code>value</code> is ill-formed
2624 * @see #setUnicodeLocaleKeyword(String, String)
2625 */
2626 public Builder setExtension(char key, String value) {
2627 try {
2628 localeBuilder.setExtension(key, value);
2629 } catch (LocaleSyntaxException e) {
2630 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2631 }
2632 return this;
2633 }
2634
2635 /**
2636 * Sets the Unicode locale keyword type for the given key. If the type
2637 * is null, the Unicode keyword is removed. Otherwise, the key must be
2638 * non-null and both key and type must be <a
2639 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2640 * is thrown.
2641 *
2642 * <p>Keys and types are converted to lower case.
2643 *
2644 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension}
2645 * replaces all Unicode locale keywords with those defined in the
2646 * extension.
2647 *
2648 * @param key the Unicode locale key
2649 * @param type the Unicode locale type
2650 * @return This builder.
2651 * @throws IllformedLocaleException if <code>key</code> or <code>type</code>
2652 * is ill-formed
2653 * @throws NullPointerException if <code>key</code> is null
2654 * @see #setExtension(char, String)
2655 */
2656 public Builder setUnicodeLocaleKeyword(String key, String type) {
2657 try {
2658 localeBuilder.setUnicodeLocaleKeyword(key, type);
2659 } catch (LocaleSyntaxException e) {
2660 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2661 }
2662 return this;
2663 }
2664
2665 /**
2666 * Adds a unicode locale attribute, if not already present, otherwise
2667 * has no effect. The attribute must not be null and must be <a
2668 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2669 * is thrown.
2670 *
2671 * @param attribute the attribute
2672 * @return This builder.
2673 * @throws NullPointerException if <code>attribute</code> is null
2674 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2675 * @see #setExtension(char, String)
2676 */
2677 public Builder addUnicodeLocaleAttribute(String attribute) {
2678 try {
2679 localeBuilder.addUnicodeLocaleAttribute(attribute);
2680 } catch (LocaleSyntaxException e) {
2681 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2682 }
2683 return this;
2684 }
2685
2686 /**
2687 * Removes a unicode locale attribute, if present, otherwise has no
2688 * effect. The attribute must not be null and must be <a
2689 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2690 * is thrown.
2691 *
2692 * <p>Attribute comparison for removal is case-insensitive.
2693 *
2694 * @param attribute the attribute
2695 * @return This builder.
2696 * @throws NullPointerException if <code>attribute</code> is null
2697 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2698 * @see #setExtension(char, String)
2699 */
2700 public Builder removeUnicodeLocaleAttribute(String attribute) {
2701 try {
2702 localeBuilder.removeUnicodeLocaleAttribute(attribute);
2703 } catch (LocaleSyntaxException e) {
2704 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2705 }
2706 return this;
2707 }
2708
2709 /**
2710 * Resets the builder to its initial, empty state.
2711 *
2712 * @return This builder.
2713 */
2714 public Builder clear() {
2715 localeBuilder.clear();
2716 return this;
2717 }
2718
2719 /**
2720 * Resets the extensions to their initial, empty state.
2721 * Language, script, region and variant are unchanged.
2722 *
2723 * @return This builder.
2724 * @see #setExtension(char, String)
2725 */
2726 public Builder clearExtensions() {
2727 localeBuilder.clearExtensions();
2728 return this;
2729 }
2730
2731 /**
2732 * Returns an instance of <code>Locale</code> created from the fields set
2733 * on this builder.
2734 *
2735 * <p>This applies the conversions listed in {@link Locale#forLanguageTag}
2736 * when constructing a Locale. (Grandfathered tags are handled in
2737 * {@link #setLanguageTag}.)
2738 *
2739 * @return A Locale.
2740 */
2741 public Locale build() {
2742 BaseLocale baseloc = localeBuilder.getBaseLocale();
2743 LocaleExtensions extensions = localeBuilder.getLocaleExtensions();
2744 if (extensions == null && baseloc.getVariant().length() > 0) {
2745 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(),
2746 baseloc.getRegion(), baseloc.getVariant());
2747 }
2748 return Locale.getInstance(baseloc, extensions);
2749 }
2750 }
2751
2752 /**
2753 * This enum provides constants to select a filtering mode for locale
2754 * matching. Refer to <a href="http://tools.ietf.org/html/rfc4647">RFC 4647
2755 * Matching of Language Tags</a> for details.
2756 *
2757 * <p>As an example, think of two Language Priority Lists each of which
2758 * includes only one language range and a set of following language tags:
2759 *
2760 * <pre>
2761 * de (German)
2762 * de-DE (German, Germany)
2763 * de-Deva (German, in Devanagari script)
2764 * de-Deva-DE (German, in Devanagari script, Germany)
2765 * de-DE-1996 (German, Germany, orthography of 1996)
2766 * de-Latn-DE (German, in Latin script, Germany)
2767 * de-Latn-DE-1996 (German, in Latin script, Germany, orthography of 1996)
2768 * </pre>
2769 *
2770 * The filtering method will behave as follows:
2771 *
2772 * <table cellpadding=2 summary="Filtering method behavior">
2773 * <tr>
2774 * <th>Filtering Mode</th>
2775 * <th>Language Priority List: {@code "de-DE"}</th>
2776 * <th>Language Priority List: {@code "de-*-DE"}</th>
2777 * </tr>
2778 * <tr>
2779 * <td valign=top>
2780 * {@link FilteringMode#AUTOSELECT_FILTERING AUTOSELECT_FILTERING}
2781 * </td>
2782 * <td valign=top>
2783 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2784 * {@code "de-DE-1996"}.
2785 * </td>
2786 * <td valign=top>
2787 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2788 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2789 * {@code "de-Latn-DE-1996"}.
2790 * </td>
2791 * </tr>
2792 * <tr>
2793 * <td valign=top>
2794 * {@link FilteringMode#EXTENDED_FILTERING EXTENDED_FILTERING}
2795 * </td>
2796 * <td valign=top>
2797 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2798 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2799 * {@code "de-Latn-DE-1996"}.
2800 * </td>
2801 * <td valign=top>Same as above.</td>
2802 * </tr>
2803 * <tr>
2804 * <td valign=top>
2805 * {@link FilteringMode#IGNORE_EXTENDED_RANGES IGNORE_EXTENDED_RANGES}
2806 * </td>
2807 * <td valign=top>
2808 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2809 * {@code "de-DE-1996"}.
2810 * </td>
2811 * <td valign=top>
2812 * Performs <em>basic</em> filtering and returns {@code null} because
2813 * nothing matches.
2814 * </td>
2815 * </tr>
2816 * <tr>
2817 * <td valign=top>
2818 * {@link FilteringMode#MAP_EXTENDED_RANGES MAP_EXTENDED_RANGES}
2819 * </td>
2820 * <td valign=top>Same as above.</td>
2821 * <td valign=top>
2822 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2823 * {@code "de-DE-1996"} because {@code "de-*-DE"} is mapped to
2824 * {@code "de-DE"}.
2825 * </td>
2826 * </tr>
2827 * <tr>
2828 * <td valign=top>
2829 * {@link FilteringMode#REJECT_EXTENDED_RANGES REJECT_EXTENDED_RANGES}
2830 * </td>
2831 * <td valign=top>Same as above.</td>
2832 * <td valign=top>
2833 * Throws {@link IllegalArgumentException} because {@code "de-*-DE"} is
2834 * not a valid basic language range.
2835 * </td>
2836 * </tr>
2837 * </table>
2838 *
2839 * @see #filter(List, Collection, FilteringMode)
2840 * @see #filterTags(List, Collection, FilteringMode)
2841 *
2842 * @since 1.8
2843 */
2844 public static enum FilteringMode {
2845 /**
2846 * Specifies automatic filtering mode based on the given Language
2847 * Priority List consisting of language ranges. If all of the ranges
2848 * are basic, basic filtering is selected. Otherwise, extended
2849 * filtering is selected.
2850 */
2851 AUTOSELECT_FILTERING,
2852
2853 /**
2854 * Specifies extended filtering.
2855 */
2856 EXTENDED_FILTERING,
2857
2858 /**
2859 * Specifies basic filtering: Note that any extended language ranges
2860 * included in the given Language Priority List are ignored.
2861 */
2862 IGNORE_EXTENDED_RANGES,
2863
2864 /**
2865 * Specifies basic filtering: If any extended language ranges are
2866 * included in the given Language Priority List, they are mapped to the
2867 * basic language range. Specifically, a language range starting with a
2868 * subtag {@code "*"} is treated as a language range {@code "*"}. For
2869 * example, {@code "*-US"} is treated as {@code "*"}. If {@code "*"} is
2870 * not the first subtag, {@code "*"} and extra {@code "-"} are removed.
2871 * For example, {@code "ja-*-JP"} is mapped to {@code "ja-JP"}.
2872 */
2873 MAP_EXTENDED_RANGES,
2874
2875 /**
2876 * Specifies basic filtering: If any extended language ranges are
2877 * included in the given Language Priority List, the list is rejected
2878 * and the filtering method throws {@link IllegalArgumentException}.
2879 */
2880 REJECT_EXTENDED_RANGES
2881 };
2882
2883 /**
2884 * This class expresses a <em>Language Range</em> defined in
2885 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
2886 * Language Tags</a>. A language range is an identifier which is used to
2887 * select language tag(s) meeting specific requirements by using the
2888 * mechanisms described in <a href="Locale.html#LocaleMatching">Locale
2889 * Matching</a>. A list which represents a user's preferences and consists
2890 * of language ranges is called a <em>Language Priority List</em>.
2891 *
2892 * <p>There are two types of language ranges: basic and extended. In RFC
2893 * 4647, the syntax of language ranges is expressed in
2894 * <a href="http://tools.ietf.org/html/rfc4234">ABNF</a> as follows:
2895 * <blockquote>
2896 * <pre>
2897 * basic-language-range = (1*8ALPHA *("-" 1*8alphanum)) / "*"
2898 * extended-language-range = (1*8ALPHA / "*")
2899 * *("-" (1*8alphanum / "*"))
2900 * alphanum = ALPHA / DIGIT
2901 * </pre>
2902 * </blockquote>
2903 * For example, {@code "en"} (English), {@code "ja-JP"} (Japanese, Japan),
2904 * {@code "*"} (special language range which matches any language tag) are
2905 * basic language ranges, whereas {@code "*-CH"} (any languages,
2906 * Switzerland), {@code "es-*"} (Spanish, any regions), and
2907 * {@code "zh-Hant-*"} (Traditional Chinese, any regions) are extended
2908 * language ranges.
2909 *
2910 * @see #filter
2911 * @see #filterTags
2912 * @see #lookup
2913 * @see #lookupTag
2914 *
2915 * @since 1.8
2916 */
2917 public static final class LanguageRange {
2918
2919 /**
2920 * A constant holding the maximum value of weight, 1.0, which indicates
2921 * that the language range is a good fit for the user.
2922 */
2923 public static final double MAX_WEIGHT = 1.0;
2924
2925 /**
2926 * A constant holding the minimum value of weight, 0.0, which indicates
2927 * that the language range is not a good fit for the user.
2928 */
2929 public static final double MIN_WEIGHT = 0.0;
2930
2931 private final String range;
2932 private final double weight;
2933
2934 private volatile int hash;
2935
2936 /**
2937 * Constructs a {@code LanguageRange} using the given {@code range}.
2938 * Note that no validation is done against the IANA Language Subtag
2939 * Registry at time of construction.
2940 *
2941 * <p>This is equivalent to {@code LanguageRange(range, MAX_WEIGHT)}.
2942 *
2943 * @param range a language range
2944 * @throws NullPointerException if the given {@code range} is
2945 * {@code null}
2946 * @throws IllegalArgumentException if the given {@code range} does not
2947 * comply with the syntax of the language range mentioned in RFC 4647
2948 */
2949 public LanguageRange(String range) {
2950 this(range, MAX_WEIGHT);
2951 }
2952
2953 /**
2954 * Constructs a {@code LanguageRange} using the given {@code range} and
2955 * {@code weight}. Note that no validation is done against the IANA
2956 * Language Subtag Registry at time of construction.
2957 *
2958 * @param range a language range
2959 * @param weight a weight value between {@code MIN_WEIGHT} and
2960 * {@code MAX_WEIGHT}
2961 * @throws NullPointerException if the given {@code range} is
2962 * {@code null}
2963 * @throws IllegalArgumentException if the given {@code range} does not
2964 * comply with the syntax of the language range mentioned in RFC 4647
2965 * or if the given {@code weight} is less than {@code MIN_WEIGHT}
2966 * or greater than {@code MAX_WEIGHT}
2967 */
2968 public LanguageRange(String range, double weight) {
2969 if (range == null) {
2970 throw new NullPointerException();
2971 }
2972 if (weight < MIN_WEIGHT || weight > MAX_WEIGHT) {
2973 throw new IllegalArgumentException("weight=" + weight);
2974 }
2975
2976 range = range.toLowerCase(Locale.ROOT);
2977
2978 // Do syntax check.
2979 boolean isIllFormed = false;
2980 String[] subtags = range.split("-");
2981 if (isSubtagIllFormed(subtags[0], true)
2982 || range.endsWith("-")) {
2983 isIllFormed = true;
2984 } else {
2985 for (int i = 1; i < subtags.length; i++) {
2986 if (isSubtagIllFormed(subtags[i], false)) {
2987 isIllFormed = true;
2988 break;
2989 }
2990 }
2991 }
2992 if (isIllFormed) {
2993 throw new IllegalArgumentException("range=" + range);
2994 }
2995
2996 this.range = range;
2997 this.weight = weight;
2998 }
2999
3000 private static boolean isSubtagIllFormed(String subtag,
3001 boolean isFirstSubtag) {
3002 if (subtag.equals("") || subtag.length() > 8) {
3003 return true;
3004 } else if (subtag.equals("*")) {
3005 return false;
3006 }
3007 char[] charArray = subtag.toCharArray();
3008 if (isFirstSubtag) { // ALPHA
3009 for (char c : charArray) {
3010 if (c < 'a' || c > 'z') {
3011 return true;
3012 }
3013 }
3014 } else { // ALPHA / DIGIT
3015 for (char c : charArray) {
3016 if (c < '0' || (c > '9' && c < 'a') || c > 'z') {
3017 return true;
3018 }
3019 }
3020 }
3021 return false;
3022 }
3023
3024 /**
3025 * Returns the language range of this {@code LanguageRange}.
3026 *
3027 * @return the language range.
3028 */
3029 public String getRange() {
3030 return range;
3031 }
3032
3033 /**
3034 * Returns the weight of this {@code LanguageRange}.
3035 *
3036 * @return the weight value.
3037 */
3038 public double getWeight() {
3039 return weight;
3040 }
3041
3042 /**
3043 * Parses the given {@code ranges} to generate a Language Priority List.
3044 *
3045 * <p>This method performs a syntactic check for each language range in
3046 * the given {@code ranges} but doesn't do validation using the IANA
3047 * Language Subtag Registry.
3048 *
3049 * <p>The {@code ranges} to be given can take one of the following
3050 * forms:
3051 *
3052 * <pre>
3053 * "Accept-Language: ja,en;q=0.4" (weighted list with Accept-Language prefix)
3054 * "ja,en;q=0.4" (weighted list)
3055 * "ja,en" (prioritized list)
3056 * </pre>
3057 *
3058 * In a weighted list, each language range is given a weight value.
3059 * The weight value is identical to the "quality value" in
3060 * <a href="http://tools.ietf.org/html/rfc2616">RFC 2616</a>, and it
3061 * expresses how much the user prefers the language. A weight value is
3062 * specified after a corresponding language range followed by
3063 * {@code ";q="}, and the default weight value is {@code MAX_WEIGHT}
3064 * when it is omitted.
3065 *
3066 * <p>Unlike a weighted list, language ranges in a prioritized list
3067 * are sorted in the descending order based on its priority. The first
3068 * language range has the highest priority and meets the user's
3069 * preference most.
3070 *
3071 * <p>In either case, language ranges are sorted in descending order in
3072 * the Language Priority List based on priority or weight. If a
3073 * language range appears in the given {@code ranges} more than once,
3074 * only the first one is included on the Language Priority List.
3075 *
3076 * <p>The returned list consists of language ranges from the given
3077 * {@code ranges} and their equivalents found in the IANA Language
3078 * Subtag Registry. For example, if the given {@code ranges} is
3079 * {@code "Accept-Language: iw,en-us;q=0.7,en;q=0.3"}, the elements in
3080 * the list to be returned are:
3081 *
3082 * <pre>
3083 * <b>Range</b> <b>Weight</b>
3084 * "iw" (older tag for Hebrew) 1.0
3085 * "he" (new preferred code for Hebrew) 1.0
3086 * "en-us" (English, United States) 0.7
3087 * "en" (English) 0.3
3088 * </pre>
3089 *
3090 * Two language ranges, {@code "iw"} and {@code "he"}, have the same
3091 * highest priority in the list. By adding {@code "he"} to the user's
3092 * Language Priority List, locale-matching method can find Hebrew as a
3093 * matching locale (or language tag) even if the application or system
3094 * offers only {@code "he"} as a supported locale (or language tag).
3095 *
3096 * @param ranges a list of comma-separated language ranges or a list of
3097 * language ranges in the form of the "Accept-Language" header
3098 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3099 * 2616</a>
3100 * @return a Language Priority List consisting of language ranges
3101 * included in the given {@code ranges} and their equivalent
3102 * language ranges if available. The list is modifiable.
3103 * @throws NullPointerException if {@code ranges} is null
3104 * @throws IllegalArgumentException if a language range or a weight
3105 * found in the given {@code ranges} is ill-formed
3106 */
3107 public static List<LanguageRange> parse(String ranges) {
3108 return LocaleMatcher.parse(ranges);
3109 }
3110
3111 /**
3112 * Parses the given {@code ranges} to generate a Language Priority
3113 * List, and then customizes the list using the given {@code map}.
3114 * This method is equivalent to
3115 * {@code mapEquivalents(parse(ranges), map)}.
3116 *
3117 * @param ranges a list of comma-separated language ranges or a list
3118 * of language ranges in the form of the "Accept-Language" header
3119 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3120 * 2616</a>
3121 * @param map a map containing information to customize language ranges
3122 * @return a Language Priority List with customization. The list is
3123 * modifiable.
3124 * @throws NullPointerException if {@code ranges} is null
3125 * @throws IllegalArgumentException if a language range or a weight
3126 * found in the given {@code ranges} is ill-formed
3127 * @see #parse(String)
3128 * @see #mapEquivalents
3129 */
3130 public static List<LanguageRange> parse(String ranges,
3131 Map<String, List<String>> map) {
3132 return mapEquivalents(parse(ranges), map);
3133 }
3134
3135 /**
3136 * Generates a new customized Language Priority List using the given
3137 * {@code priorityList} and {@code map}. If the given {@code map} is
3138 * empty, this method returns a copy of the given {@code priorityList}.
3139 *
3140 * <p>In the map, a key represents a language range whereas a value is
3141 * a list of equivalents of it. {@code '*'} cannot be used in the map.
3142 * Each equivalent language range has the same weight value as its
3143 * original language range.
3144 *
3145 * <pre>
3146 * An example of map:
3147 * <b>Key</b> <b>Value</b>
3148 * "zh" (Chinese) "zh",
3149 * "zh-Hans"(Simplified Chinese)
3150 * "zh-HK" (Chinese, Hong Kong) "zh-HK"
3151 * "zh-TW" (Chinese, Taiwan) "zh-TW"
3152 * </pre>
3153 *
3154 * The customization is performed after modification using the IANA
3155 * Language Subtag Registry.
3156 *
3157 * <p>For example, if a user's Language Priority List consists of five
3158 * language ranges ({@code "zh"}, {@code "zh-CN"}, {@code "en"},
3159 * {@code "zh-TW"}, and {@code "zh-HK"}), the newly generated Language
3160 * Priority List which is customized using the above map example will
3161 * consists of {@code "zh"}, {@code "zh-Hans"}, {@code "zh-CN"},
3162 * {@code "zh-Hans-CN"}, {@code "en"}, {@code "zh-TW"}, and
3163 * {@code "zh-HK"}.
3164 *
3165 * <p>{@code "zh-HK"} and {@code "zh-TW"} aren't converted to
3166 * {@code "zh-Hans-HK"} nor {@code "zh-Hans-TW"} even if they are
3167 * included in the Language Priority List. In this example, mapping
3168 * is used to clearly distinguish Simplified Chinese and Traditional
3169 * Chinese.
3170 *
3171 * <p>If the {@code "zh"}-to-{@code "zh"} mapping isn't included in the
3172 * map, a simple replacement will be performed and the customized list
3173 * won't include {@code "zh"} and {@code "zh-CN"}.
3174 *
3175 * @param priorityList user's Language Priority List
3176 * @param map a map containing information to customize language ranges
3177 * @return a new Language Priority List with customization. The list is
3178 * modifiable.
3179 * @throws NullPointerException if {@code priorityList} is {@code null}
3180 * @see #parse(String, Map)
3181 */
3182 public static List<LanguageRange> mapEquivalents(
3183 List<LanguageRange>priorityList,
3184 Map<String, List<String>> map) {
3185 return LocaleMatcher.mapEquivalents(priorityList, map);
3186 }
3187
3188 /**
3189 * Returns a hash code value for the object.
3190 *
3191 * @return a hash code value for this object.
3192 */
3193 @Override
3194 public int hashCode() {
3195 int h = hash;
3196 if (h == 0) {
3197 h = 17;
3198 h = 37*h + range.hashCode();
3199 long bitsWeight = Double.doubleToLongBits(weight);
3200 h = 37*h + (int)(bitsWeight ^ (bitsWeight >>> 32));
3201 if (h != 0) {
3202 hash = h;
3203 }
3204 }
3205 return h;
3206 }
3207
3208 /**
3209 * Compares this object to the specified object. The result is true if
3210 * and only if the argument is not {@code null} and is a
3211 * {@code LanguageRange} object that contains the same {@code range}
3212 * and {@code weight} values as this object.
3213 *
3214 * @param obj the object to compare with
3215 * @return {@code true} if this object's {@code range} and
3216 * {@code weight} are the same as the {@code obj}'s; {@code false}
3217 * otherwise.
3218 */
3219 @Override
3220 public boolean equals(Object obj) {
3221 if (this == obj) {
3222 return true;
3223 }
3224 if (!(obj instanceof LanguageRange)) {
3225 return false;
3226 }
3227 LanguageRange other = (LanguageRange)obj;
3228 return hash == other.hash
3229 && range.equals(other.range)
3230 && weight == other.weight;
3231 }
3232
3233 /**
3234 * Returns an informative string representation of this {@code LanguageRange}
3235 * object, consisting of language range and weight if the range is
3236 * weighted and the weight is less than the max weight.
3237 *
3238 * @return a string representation of this {@code LanguageRange} object.
3239 */
3240 @Override
3241 public String toString() {
3242 return (weight == MAX_WEIGHT) ? range : range + ";q=" + weight;
3243 }
3244 }
3245
3246 /**
3247 * Returns a list of matching {@code Locale} instances using the filtering
3248 * mechanism defined in RFC 4647.
3249 *
3250 * @param priorityList user's Language Priority List in which each language
3251 * tag is sorted in descending order based on priority or weight
3252 * @param locales {@code Locale} instances used for matching
3253 * @param mode filtering mode
3254 * @return a list of {@code Locale} instances for matching language tags
3255 * sorted in descending order based on priority or weight, or an empty
3256 * list if nothing matches. The list is modifiable.
3257 * @throws NullPointerException if {@code priorityList} or {@code locales}
3258 * is {@code null}
3259 * @throws IllegalArgumentException if one or more extended language ranges
3260 * are included in the given list when
3261 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3262 *
3263 * @since 1.8
3264 */
3265 public static List<Locale> filter(List<LanguageRange> priorityList,
3266 Collection<Locale> locales,
3267 FilteringMode mode) {
3268 return LocaleMatcher.filter(priorityList, locales, mode);
3269 }
3270
3271 /**
3272 * Returns a list of matching {@code Locale} instances using the filtering
3273 * mechanism defined in RFC 4647. This is equivalent to
3274 * {@link #filter(List, Collection, FilteringMode)} when {@code mode} is
3275 * {@link FilteringMode#AUTOSELECT_FILTERING}.
3276 *
3277 * @param priorityList user's Language Priority List in which each language
3278 * tag is sorted in descending order based on priority or weight
3279 * @param locales {@code Locale} instances used for matching
3280 * @return a list of {@code Locale} instances for matching language tags
3281 * sorted in descending order based on priority or weight, or an empty
3282 * list if nothing matches. The list is modifiable.
3283 * @throws NullPointerException if {@code priorityList} or {@code locales}
3284 * is {@code null}
3285 *
3286 * @since 1.8
3287 */
3288 public static List<Locale> filter(List<LanguageRange> priorityList,
3289 Collection<Locale> locales) {
3290 return filter(priorityList, locales, FilteringMode.AUTOSELECT_FILTERING);
3291 }
3292
3293 /**
3294 * Returns a list of matching languages tags using the basic filtering
3295 * mechanism defined in RFC 4647.
3296 *
3297 * @param priorityList user's Language Priority List in which each language
3298 * tag is sorted in descending order based on priority or weight
3299 * @param tags language tags
3300 * @param mode filtering mode
3301 * @return a list of matching language tags sorted in descending order
3302 * based on priority or weight, or an empty list if nothing matches.
3303 * The list is modifiable.
3304 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3305 * {@code null}
3306 * @throws IllegalArgumentException if one or more extended language ranges
3307 * are included in the given list when
3308 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3309 *
3310 * @since 1.8
3311 */
3312 public static List<String> filterTags(List<LanguageRange> priorityList,
3313 Collection<String> tags,
3314 FilteringMode mode) {
3315 return LocaleMatcher.filterTags(priorityList, tags, mode);
3316 }
3317
3318 /**
3319 * Returns a list of matching languages tags using the basic filtering
3320 * mechanism defined in RFC 4647. This is equivalent to
3321 * {@link #filterTags(List, Collection, FilteringMode)} when {@code mode}
3322 * is {@link FilteringMode#AUTOSELECT_FILTERING}.
3323 *
3324 * @param priorityList user's Language Priority List in which each language
3325 * tag is sorted in descending order based on priority or weight
3326 * @param tags language tags
3327 * @return a list of matching language tags sorted in descending order
3328 * based on priority or weight, or an empty list if nothing matches.
3329 * The list is modifiable.
3330 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3331 * {@code null}
3332 *
3333 * @since 1.8
3334 */
3335 public static List<String> filterTags(List<LanguageRange> priorityList,
3336 Collection<String> tags) {
3337 return filterTags(priorityList, tags, FilteringMode.AUTOSELECT_FILTERING);
3338 }
3339
3340 /**
3341 * Returns a {@code Locale} instance for the best-matching language
3342 * tag using the lookup mechanism defined in RFC 4647.
3343 *
3344 * @param priorityList user's Language Priority List in which each language
3345 * tag is sorted in descending order based on priority or weight
3346 * @param locales {@code Locale} instances used for matching
3347 * @return the best matching <code>Locale</code> instance chosen based on
3348 * priority or weight, or {@code null} if nothing matches.
3349 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3350 * {@code null}
3351 *
3352 * @since 1.8
3353 */
3354 public static Locale lookup(List<LanguageRange> priorityList,
3355 Collection<Locale> locales) {
3356 return LocaleMatcher.lookup(priorityList, locales);
3357 }
3358
3359 /**
3360 * Returns the best-matching language tag using the lookup mechanism
3361 * defined in RFC 4647.
3362 *
3363 * @param priorityList user's Language Priority List in which each language
3364 * tag is sorted in descending order based on priority or weight
3365 * @param tags language tangs used for matching
3366 * @return the best matching language tag chosen based on priority or
3367 * weight, or {@code null} if nothing matches.
3368 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3369 * {@code null}
3370 *
3371 * @since 1.8
3372 */
3373 public static String lookupTag(List<LanguageRange> priorityList,
3374 Collection<String> tags) {
3375 return LocaleMatcher.lookupTag(priorityList, tags);
3376 }
3377
3378 }