1 /*
2 * Copyright (c) 1996, 2017, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
10 *
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
16 *
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
20 *
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
23 * questions.
24 */
25
26 /*
27 * (C) Copyright Taligent, Inc. 1996, 1997 - All Rights Reserved
28 * (C) Copyright IBM Corp. 1996 - 1998 - All Rights Reserved
29 *
30 * The original version of this source code and documentation
31 * is copyrighted and owned by Taligent, Inc., a wholly-owned
32 * subsidiary of IBM. These materials are provided under terms
33 * of a License Agreement between Taligent and Sun. This technology
34 * is protected by multiple US and International patents.
35 *
36 * This notice and attribution to Taligent may not be removed.
37 * Taligent is a registered trademark of Taligent, Inc.
38 *
39 */
40
41 package java.util;
42
43 import java.io.IOException;
44 import java.io.ObjectInputStream;
45 import java.io.ObjectOutputStream;
46 import java.io.ObjectStreamField;
47 import java.io.Serializable;
48 import java.text.MessageFormat;
49 import java.util.concurrent.ConcurrentHashMap;
50 import java.util.spi.LocaleNameProvider;
51 import java.util.stream.Collectors;
52
53 import sun.security.action.GetPropertyAction;
54 import sun.util.locale.BaseLocale;
55 import sun.util.locale.InternalLocaleBuilder;
56 import sun.util.locale.LanguageTag;
57 import sun.util.locale.LocaleExtensions;
58 import sun.util.locale.LocaleMatcher;
59 import sun.util.locale.LocaleObjectCache;
60 import sun.util.locale.LocaleSyntaxException;
61 import sun.util.locale.LocaleUtils;
62 import sun.util.locale.ParseStatus;
63 import sun.util.locale.provider.LocaleProviderAdapter;
64 import sun.util.locale.provider.LocaleResources;
65 import sun.util.locale.provider.LocaleServiceProviderPool;
66 import sun.util.locale.provider.TimeZoneNameUtility;
67
68 /**
69 * A <code>Locale</code> object represents a specific geographical, political,
70 * or cultural region. An operation that requires a <code>Locale</code> to perform
71 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code>
72 * to tailor information for the user. For example, displaying a number
73 * is a locale-sensitive operation— the number should be formatted
74 * according to the customs and conventions of the user's native country,
75 * region, or culture.
76 *
77 * <p> The {@code Locale} class implements IETF BCP 47 which is composed of
78 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 "Matching of Language
79 * Tags"</a> and <a href="http://tools.ietf.org/html/rfc5646">RFC 5646 "Tags
80 * for Identifying Languages"</a> with support for the LDML (UTS#35, "Unicode
81 * Locale Data Markup Language") BCP 47-compatible extensions for locale data
82 * exchange.
83 *
84 * <p> A <code>Locale</code> object logically consists of the fields
85 * described below.
86 *
87 * <dl>
88 * <dt><a id="def_language"><b>language</b></a></dt>
89 *
90 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered
91 * language subtags up to 8 alpha letters (for future enhancements).
92 * When a language has both an alpha-2 code and an alpha-3 code, the
93 * alpha-2 code must be used. You can find a full list of valid
94 * language codes in the IANA Language Subtag Registry (search for
95 * "Type: language"). The language field is case insensitive, but
96 * <code>Locale</code> always canonicalizes to lower case.</dd>
97 *
98 * <dd>Well-formed language values have the form
99 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the full
100 * BCP47 language production, since it excludes extlang. They are
101 * not needed since modern three-letter language codes replace
102 * them.</dd>
103 *
104 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd>
105 *
106 * <dt><a id="def_script"><b>script</b></a></dt>
107 *
108 * <dd>ISO 15924 alpha-4 script code. You can find a full list of
109 * valid script codes in the IANA Language Subtag Registry (search
110 * for "Type: script"). The script field is case insensitive, but
111 * <code>Locale</code> always canonicalizes to title case (the first
112 * letter is upper case and the rest of the letters are lower
113 * case).</dd>
114 *
115 * <dd>Well-formed script values have the form
116 * <code>[a-zA-Z]{4}</code></dd>
117 *
118 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd>
119 *
120 * <dt><a id="def_region"><b>country (region)</b></a></dt>
121 *
122 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code.
123 * You can find a full list of valid country and region codes in the
124 * IANA Language Subtag Registry (search for "Type: region"). The
125 * country (region) field is case insensitive, but
126 * <code>Locale</code> always canonicalizes to upper case.</dd>
127 *
128 * <dd>Well-formed country/region values have
129 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd>
130 *
131 * <dd>Example: "US" (United States), "FR" (France), "029"
132 * (Caribbean)</dd>
133 *
134 * <dt><a id="def_variant"><b>variant</b></a></dt>
135 *
136 * <dd>Any arbitrary value used to indicate a variation of a
137 * <code>Locale</code>. Where there are two or more variant values
138 * each indicating its own semantics, these values should be ordered
139 * by importance, with most important first, separated by
140 * underscore('_'). The variant field is case sensitive.</dd>
141 *
142 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant
143 * subtags. Also BCP 47 subtags are strictly used to indicate
144 * additional variations that define a language or its dialects that
145 * are not covered by any combinations of language, script and
146 * region subtags. You can find a full list of valid variant codes
147 * in the IANA Language Subtag Registry (search for "Type: variant").
148 *
149 * <p>However, the variant field in <code>Locale</code> has
150 * historically been used for any kind of variation, not just
151 * language variations. For example, some supported variants
152 * available in Java SE Runtime Environments indicate alternative
153 * cultural behaviors such as calendar type or number script. In
154 * BCP 47 this kind of information, which does not identify the
155 * language, is supported by extension subtags or private use
156 * subtags.</dd>
157 *
158 * <dd>Well-formed variant values have the form <code>SUBTAG
159 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG =
160 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only
161 * uses hyphen ('-') as a delimiter, this is more lenient).</dd>
162 *
163 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd>
164 *
165 * <dt><a id="def_extensions"><b>extensions</b></a></dt>
166 *
167 * <dd>A map from single character keys to string values, indicating
168 * extensions apart from language identification. The extensions in
169 * <code>Locale</code> implement the semantics and syntax of BCP 47
170 * extension subtags and private use subtags. The extensions are
171 * case insensitive, but <code>Locale</code> canonicalizes all
172 * extension keys and values to lower case. Note that extensions
173 * cannot have empty values.</dd>
174 *
175 * <dd>Well-formed keys are single characters from the set
176 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form
177 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x'
178 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys
179 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows
180 * single-character subtags).</dd>
181 *
182 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar),
183 * key="x"/value="java-1-7"</dd>
184 * </dl>
185 *
186 * <b>Note:</b> Although BCP 47 requires field values to be registered
187 * in the IANA Language Subtag Registry, the <code>Locale</code> class
188 * does not provide any validation features. The <code>Builder</code>
189 * only checks if an individual field satisfies the syntactic
190 * requirement (is well-formed), but does not validate the value
191 * itself. See {@link Builder} for details.
192 *
193 * <h3><a id="def_locale_extension">Unicode locale/language extension</a></h3>
194 *
195 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional
196 * attributes and keywords to override or refine the default behavior
197 * associated with a locale. A keyword is represented by a pair of
198 * key and type. For example, "nu-thai" indicates that Thai local
199 * digits (value:"thai") should be used for formatting numbers
200 * (key:"nu").
201 *
202 * <p>The keywords are mapped to a BCP 47 extension value using the
203 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above
204 * example, "nu-thai", becomes the extension "u-nu-thai".
205 *
206 * <p>Thus, when a <code>Locale</code> object contains Unicode locale
207 * attributes and keywords,
208 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a
209 * String representing this information, for example, "nu-thai". The
210 * <code>Locale</code> class also provides {@link
211 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and
212 * {@link #getUnicodeLocaleType} which allow you to access Unicode
213 * locale attributes and key/type pairs directly. When represented as
214 * a string, the Unicode Locale Extension lists attributes
215 * alphabetically, followed by key/type sequences with keys listed
216 * alphabetically (the order of subtags comprising a key's type is
217 * fixed when the type is defined)
218 *
219 * <p>A well-formed locale key has the form
220 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the
221 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it
222 * can be empty, or a series of subtags 3-8 alphanums in length). A
223 * well-formed locale attribute has the form
224 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same
225 * form as a locale type subtag).
226 *
227 * <p>The Unicode locale extension specifies optional behavior in
228 * locale-sensitive services. Although the LDML specification defines
229 * various keys and values, actual locale-sensitive service
230 * implementations in a Java Runtime Environment might not support any
231 * particular Unicode locale attributes or key/type pairs.
232 *
233 * <h4>Creating a Locale</h4>
234 *
235 * <p>There are several different ways to create a <code>Locale</code>
236 * object.
237 *
238 * <h5>Builder</h5>
239 *
240 * <p>Using {@link Builder} you can construct a <code>Locale</code> object
241 * that conforms to BCP 47 syntax.
242 *
243 * <h5>Constructors</h5>
244 *
245 * <p>The <code>Locale</code> class provides three constructors:
246 * <blockquote>
247 * <pre>
248 * {@link #Locale(String language)}
249 * {@link #Locale(String language, String country)}
250 * {@link #Locale(String language, String country, String variant)}
251 * </pre>
252 * </blockquote>
253 * These constructors allow you to create a <code>Locale</code> object
254 * with language, country and variant, but you cannot specify
255 * script or extensions.
256 *
257 * <h5>Factory Methods</h5>
258 *
259 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code>
260 * object for a well-formed BCP 47 language tag.
261 *
262 * <h5>Locale Constants</h5>
263 *
264 * <p>The <code>Locale</code> class provides a number of convenient constants
265 * that you can use to create <code>Locale</code> objects for commonly used
266 * locales. For example, the following creates a <code>Locale</code> object
267 * for the United States:
268 * <blockquote>
269 * <pre>
270 * Locale.US
271 * </pre>
272 * </blockquote>
273 *
274 * <h4><a id="LocaleMatching">Locale Matching</a></h4>
275 *
276 * <p>If an application or a system is internationalized and provides localized
277 * resources for multiple locales, it sometimes needs to find one or more
278 * locales (or language tags) which meet each user's specific preferences. Note
279 * that a term "language tag" is used interchangeably with "locale" in this
280 * locale matching documentation.
281 *
282 * <p>In order to do matching a user's preferred locales to a set of language
283 * tags, <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
284 * Language Tags</a> defines two mechanisms: filtering and lookup.
285 * <em>Filtering</em> is used to get all matching locales, whereas
286 * <em>lookup</em> is to choose the best matching locale.
287 * Matching is done case-insensitively. These matching mechanisms are described
288 * in the following sections.
289 *
290 * <p>A user's preference is called a <em>Language Priority List</em> and is
291 * expressed as a list of language ranges. There are syntactically two types of
292 * language ranges: basic and extended. See
293 * {@link Locale.LanguageRange Locale.LanguageRange} for details.
294 *
295 * <h5>Filtering</h5>
296 *
297 * <p>The filtering operation returns all matching language tags. It is defined
298 * in RFC 4647 as follows:
299 * "In filtering, each language range represents the least specific language
300 * tag (that is, the language tag with fewest number of subtags) that is an
301 * acceptable match. All of the language tags in the matching set of tags will
302 * have an equal or greater number of subtags than the language range. Every
303 * non-wildcard subtag in the language range will appear in every one of the
304 * matching language tags."
305 *
306 * <p>There are two types of filtering: filtering for basic language ranges
307 * (called "basic filtering") and filtering for extended language ranges
308 * (called "extended filtering"). They may return different results by what
309 * kind of language ranges are included in the given Language Priority List.
310 * {@link Locale.FilteringMode} is a parameter to specify how filtering should
311 * be done.
312 *
313 * <h5>Lookup</h5>
314 *
315 * <p>The lookup operation returns the best matching language tags. It is
316 * defined in RFC 4647 as follows:
317 * "By contrast with filtering, each language range represents the most
318 * specific tag that is an acceptable match. The first matching tag found,
319 * according to the user's priority, is considered the closest match and is the
320 * item returned."
321 *
322 * <p>For example, if a Language Priority List consists of two language ranges,
323 * {@code "zh-Hant-TW"} and {@code "en-US"}, in prioritized order, lookup
324 * method progressively searches the language tags below in order to find the
325 * best matching language tag.
326 * <blockquote>
327 * <pre>
328 * 1. zh-Hant-TW
329 * 2. zh-Hant
330 * 3. zh
331 * 4. en-US
332 * 5. en
333 * </pre>
334 * </blockquote>
335 * If there is a language tag which matches completely to a language range
336 * above, the language tag is returned.
337 *
338 * <p>{@code "*"} is the special language range, and it is ignored in lookup.
339 *
340 * <p>If multiple language tags match as a result of the subtag {@code '*'}
341 * included in a language range, the first matching language tag returned by
342 * an {@link Iterator} over a {@link Collection} of language tags is treated as
343 * the best matching one.
344 *
345 * <h4>Use of Locale</h4>
346 *
347 * <p>Once you've created a <code>Locale</code> you can query it for information
348 * about itself. Use <code>getCountry</code> to get the country (or region)
349 * code and <code>getLanguage</code> to get the language code.
350 * You can use <code>getDisplayCountry</code> to get the
351 * name of the country suitable for displaying to the user. Similarly,
352 * you can use <code>getDisplayLanguage</code> to get the name of
353 * the language suitable for displaying to the user. Interestingly,
354 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive
355 * and have two versions: one that uses the default
356 * {@link Locale.Category#DISPLAY DISPLAY} locale and one
357 * that uses the locale specified as an argument.
358 *
359 * <p>The Java Platform provides a number of classes that perform locale-sensitive
360 * operations. For example, the <code>NumberFormat</code> class formats
361 * numbers, currency, and percentages in a locale-sensitive manner. Classes
362 * such as <code>NumberFormat</code> have several convenience methods
363 * for creating a default object of that type. For example, the
364 * <code>NumberFormat</code> class provides these three convenience methods
365 * for creating a default <code>NumberFormat</code> object:
366 * <blockquote>
367 * <pre>
368 * NumberFormat.getInstance()
369 * NumberFormat.getCurrencyInstance()
370 * NumberFormat.getPercentInstance()
371 * </pre>
372 * </blockquote>
373 * Each of these methods has two variants; one with an explicit locale
374 * and one without; the latter uses the default
375 * {@link Locale.Category#FORMAT FORMAT} locale:
376 * <blockquote>
377 * <pre>
378 * NumberFormat.getInstance(myLocale)
379 * NumberFormat.getCurrencyInstance(myLocale)
380 * NumberFormat.getPercentInstance(myLocale)
381 * </pre>
382 * </blockquote>
383 * A <code>Locale</code> is the mechanism for identifying the kind of object
384 * (<code>NumberFormat</code>) that you would like to get. The locale is
385 * <STRONG>just</STRONG> a mechanism for identifying objects,
386 * <STRONG>not</STRONG> a container for the objects themselves.
387 *
388 * <h4>Compatibility</h4>
389 *
390 * <p>In order to maintain compatibility with existing usage, Locale's
391 * constructors retain their behavior prior to the Java Runtime
392 * Environment version 1.7. The same is largely true for the
393 * <code>toString</code> method. Thus Locale objects can continue to
394 * be used as they were. In particular, clients who parse the output
395 * of toString into language, country, and variant fields can continue
396 * to do so (although this is strongly discouraged), although the
397 * variant field will have additional information in it if script or
398 * extensions are present.
399 *
400 * <p>In addition, BCP 47 imposes syntax restrictions that are not
401 * imposed by Locale's constructors. This means that conversions
402 * between some Locales and BCP 47 language tags cannot be made without
403 * losing information. Thus <code>toLanguageTag</code> cannot
404 * represent the state of locales whose language, country, or variant
405 * do not conform to BCP 47.
406 *
407 * <p>Because of these issues, it is recommended that clients migrate
408 * away from constructing non-conforming locales and use the
409 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead.
410 * Clients desiring a string representation of the complete locale can
411 * then always rely on <code>toLanguageTag</code> for this purpose.
412 *
413 * <h5><a id="special_cases_constructor">Special cases</a></h5>
414 *
415 * <p>For compatibility reasons, two
416 * non-conforming locales are treated as special cases. These are
417 * <b>{@code ja_JP_JP}</b> and <b>{@code th_TH_TH}</b>. These are ill-formed
418 * in BCP 47 since the variants are too short. To ease migration to BCP 47,
419 * these are treated specially during construction. These two cases (and only
420 * these) cause a constructor to generate an extension, all other values behave
421 * exactly as they did prior to Java 7.
422 *
423 * <p>Java has used {@code ja_JP_JP} to represent Japanese as used in
424 * Japan together with the Japanese Imperial calendar. This is now
425 * representable using a Unicode locale extension, by specifying the
426 * Unicode locale key {@code ca} (for "calendar") and type
427 * {@code japanese}. When the Locale constructor is called with the
428 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is
429 * automatically added.
430 *
431 * <p>Java has used {@code th_TH_TH} to represent Thai as used in
432 * Thailand together with Thai digits. This is also now representable using
433 * a Unicode locale extension, by specifying the Unicode locale key
434 * {@code nu} (for "number") and value {@code thai}. When the Locale
435 * constructor is called with the arguments "th", "TH", "TH", the
436 * extension "u-nu-thai" is automatically added.
437 *
438 * <h5>Serialization</h5>
439 *
440 * <p>During serialization, writeObject writes all fields to the output
441 * stream, including extensions.
442 *
443 * <p>During deserialization, readResolve adds extensions as described
444 * in <a href="#special_cases_constructor">Special Cases</a>, only
445 * for the two cases th_TH_TH and ja_JP_JP.
446 *
447 * <h5>Legacy language codes</h5>
448 *
449 * <p>Locale's constructor has always converted three language codes to
450 * their earlier, obsoleted forms: {@code he} maps to {@code iw},
451 * {@code yi} maps to {@code ji}, and {@code id} maps to
452 * {@code in}. This continues to be the case, in order to not break
453 * backwards compatibility.
454 *
455 * <p>The APIs added in 1.7 map between the old and new language codes,
456 * maintaining the old codes internal to Locale (so that
457 * <code>getLanguage</code> and <code>toString</code> reflect the old
458 * code), but using the new codes in the BCP 47 language tag APIs (so
459 * that <code>toLanguageTag</code> reflects the new one). This
460 * preserves the equivalence between Locales no matter which code or
461 * API is used to construct them. Java's default resource bundle
462 * lookup mechanism also implements this mapping, so that resources
463 * can be named using either convention, see {@link ResourceBundle.Control}.
464 *
465 * <h5>Three-letter language/country(region) codes</h5>
466 *
467 * <p>The Locale constructors have always specified that the language
468 * and the country param be two characters in length, although in
469 * practice they have accepted any length. The specification has now
470 * been relaxed to allow language codes of two to eight characters and
471 * country (region) codes of two to three characters, and in
472 * particular, three-letter language codes and three-digit region
473 * codes as specified in the IANA Language Subtag Registry. For
474 * compatibility, the implementation still does not impose a length
475 * constraint.
476 *
477 * @see Builder
478 * @see ResourceBundle
479 * @see java.text.Format
480 * @see java.text.NumberFormat
481 * @see java.text.Collator
482 * @author Mark Davis
483 * @since 1.1
484 */
485 public final class Locale implements Cloneable, Serializable {
486
487 private static final Cache LOCALECACHE = new Cache();
488
489 /** Useful constant for language.
490 */
491 public static final Locale ENGLISH = createConstant("en", "");
492
493 /** Useful constant for language.
494 */
495 public static final Locale FRENCH = createConstant("fr", "");
496
497 /** Useful constant for language.
498 */
499 public static final Locale GERMAN = createConstant("de", "");
500
501 /** Useful constant for language.
502 */
503 public static final Locale ITALIAN = createConstant("it", "");
504
505 /** Useful constant for language.
506 */
507 public static final Locale JAPANESE = createConstant("ja", "");
508
509 /** Useful constant for language.
510 */
511 public static final Locale KOREAN = createConstant("ko", "");
512
513 /** Useful constant for language.
514 */
515 public static final Locale CHINESE = createConstant("zh", "");
516
517 /** Useful constant for language.
518 */
519 public static final Locale SIMPLIFIED_CHINESE = createConstant("zh", "CN");
520
521 /** Useful constant for language.
522 */
523 public static final Locale TRADITIONAL_CHINESE = createConstant("zh", "TW");
524
525 /** Useful constant for country.
526 */
527 public static final Locale FRANCE = createConstant("fr", "FR");
528
529 /** Useful constant for country.
530 */
531 public static final Locale GERMANY = createConstant("de", "DE");
532
533 /** Useful constant for country.
534 */
535 public static final Locale ITALY = createConstant("it", "IT");
536
537 /** Useful constant for country.
538 */
539 public static final Locale JAPAN = createConstant("ja", "JP");
540
541 /** Useful constant for country.
542 */
543 public static final Locale KOREA = createConstant("ko", "KR");
544
545 /** Useful constant for country.
546 */
547 public static final Locale CHINA = SIMPLIFIED_CHINESE;
548
549 /** Useful constant for country.
550 */
551 public static final Locale PRC = SIMPLIFIED_CHINESE;
552
553 /** Useful constant for country.
554 */
555 public static final Locale TAIWAN = TRADITIONAL_CHINESE;
556
557 /** Useful constant for country.
558 */
559 public static final Locale UK = createConstant("en", "GB");
560
561 /** Useful constant for country.
562 */
563 public static final Locale US = createConstant("en", "US");
564
565 /** Useful constant for country.
566 */
567 public static final Locale CANADA = createConstant("en", "CA");
568
569 /** Useful constant for country.
570 */
571 public static final Locale CANADA_FRENCH = createConstant("fr", "CA");
572
573 /**
574 * Useful constant for the root locale. The root locale is the locale whose
575 * language, country, and variant are empty ("") strings. This is regarded
576 * as the base locale of all locales, and is used as the language/country
577 * neutral locale for the locale sensitive operations.
578 *
579 * @since 1.6
580 */
581 public static final Locale ROOT = createConstant("", "");
582
583 /**
584 * The key for the private use extension ('x').
585 *
586 * @see #getExtension(char)
587 * @see Builder#setExtension(char, String)
588 * @since 1.7
589 */
590 public static final char PRIVATE_USE_EXTENSION = 'x';
591
592 /**
593 * The key for Unicode locale extension ('u').
594 *
595 * @see #getExtension(char)
596 * @see Builder#setExtension(char, String)
597 * @since 1.7
598 */
599 public static final char UNICODE_LOCALE_EXTENSION = 'u';
600
601 /** serialization ID
602 */
603 static final long serialVersionUID = 9149081749638150636L;
604
605 /**
606 * Enum for specifying the type defined in ISO 3166. This enum is used to
607 * retrieve the two-letter ISO3166-1 alpha-2, three-letter ISO3166-1
608 * alpha-3, four-letter ISO3166-3 country codes.
609 *
610 * @see #getISOCountries(Locale.IsoCountryCode)
611 * @since 9
612 */
613 public static enum IsoCountryCode {
614 /**
615 * PART1_ALPHA2 is used to represent the ISO3166-1 alpha-2 two letter
616 * country codes.
617 */
618 PART1_ALPHA2 {
619 @Override
620 Set<String> createCountryCodeSet() {
621 return Set.of(Locale.getISOCountries());
622 }
623 },
624
625 /**
626 *
627 * PART1_ALPHA3 is used to represent the ISO3166-1 alpha-3 three letter
628 * country codes.
629 */
630 PART1_ALPHA3 {
631 @Override
632 Set<String> createCountryCodeSet() {
633 return LocaleISOData.computeISO3166_1Alpha3Countries();
634 }
635 },
636
637 /**
638 * PART3 is used to represent the ISO3166-3 four letter country codes.
639 */
640 PART3 {
641 @Override
642 Set<String> createCountryCodeSet() {
643 return Set.of(LocaleISOData.ISO3166_3);
644 }
645 };
646
647 /**
648 * Concrete implementation of this method attempts to compute value
649 * for iso3166CodesMap for each IsoCountryCode type key.
650 */
651 abstract Set<String> createCountryCodeSet();
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::createCountryCodeSet);
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 private static final int DISPLAY_UEXT_KEY = 4;
675 private static final int DISPLAY_UEXT_TYPE = 5;
676
677 /**
678 * Private constructor used by getInstance method
679 */
680 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) {
681 this.baseLocale = baseLocale;
682 this.localeExtensions = extensions;
683 }
684
685 /**
686 * Construct a locale from language, country and variant.
687 * This constructor normalizes the language value to lowercase and
688 * the country value to uppercase.
689 * <p>
690 * <b>Note:</b>
691 * <ul>
692 * <li>ISO 639 is not a stable standard; some of the language codes it defines
693 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
694 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
695 * API on Locale will return only the OLD codes.
696 * <li>For backward compatibility reasons, this constructor does not make
697 * any syntactic checks on the input.
698 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially,
699 * see <a href="#special_cases_constructor">Special Cases</a> for more information.
700 * </ul>
701 *
702 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
703 * up to 8 characters in length. See the <code>Locale</code> class description about
704 * valid language values.
705 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
706 * See the <code>Locale</code> class description about valid country values.
707 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>.
708 * See the <code>Locale</code> class description for the details.
709 * @exception NullPointerException thrown if any argument is null.
710 */
711 public Locale(String language, String country, String variant) {
712 if (language== null || country == null || variant == null) {
713 throw new NullPointerException();
714 }
715 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant);
716 localeExtensions = getCompatibilityExtensions(language, "", country, variant);
717 }
718
719 /**
720 * Construct a locale from language and country.
721 * This constructor normalizes the language value to lowercase and
722 * the country value to uppercase.
723 * <p>
724 * <b>Note:</b>
725 * <ul>
726 * <li>ISO 639 is not a stable standard; some of the language codes it defines
727 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
728 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
729 * API on Locale will return only the OLD codes.
730 * <li>For backward compatibility reasons, this constructor does not make
731 * any syntactic checks on the input.
732 * </ul>
733 *
734 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
735 * up to 8 characters in length. See the <code>Locale</code> class description about
736 * valid language values.
737 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code.
738 * See the <code>Locale</code> class description about valid country values.
739 * @exception NullPointerException thrown if either argument is null.
740 */
741 public Locale(String language, String country) {
742 this(language, country, "");
743 }
744
745 /**
746 * Construct a locale from a language code.
747 * This constructor normalizes the language value to lowercase.
748 * <p>
749 * <b>Note:</b>
750 * <ul>
751 * <li>ISO 639 is not a stable standard; some of the language codes it defines
752 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the
753 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other
754 * API on Locale will return only the OLD codes.
755 * <li>For backward compatibility reasons, this constructor does not make
756 * any syntactic checks on the input.
757 * </ul>
758 *
759 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag
760 * up to 8 characters in length. See the <code>Locale</code> class description about
761 * valid language values.
762 * @exception NullPointerException thrown if argument is null.
763 * @since 1.4
764 */
765 public Locale(String language) {
766 this(language, "", "");
767 }
768
769 /**
770 * This method must be called only for creating the Locale.*
771 * constants due to making shortcuts.
772 */
773 private static Locale createConstant(String lang, String country) {
774 BaseLocale base = BaseLocale.createInstance(lang, country);
775 return getInstance(base, null);
776 }
777
778 /**
779 * Returns a <code>Locale</code> constructed from the given
780 * <code>language</code>, <code>country</code> and
781 * <code>variant</code>. If the same <code>Locale</code> instance
782 * is available in the cache, then that instance is
783 * returned. Otherwise, a new <code>Locale</code> instance is
784 * created and cached.
785 *
786 * @param language lowercase 2 to 8 language code.
787 * @param country uppercase two-letter ISO-3166 code and numeric-3 UN M.49 area code.
788 * @param variant vendor and browser specific code. See class description.
789 * @return the <code>Locale</code> instance requested
790 * @exception NullPointerException if any argument is null.
791 */
792 static Locale getInstance(String language, String country, String variant) {
793 return getInstance(language, "", country, variant, null);
794 }
795
796 static Locale getInstance(String language, String script, String country,
797 String variant, LocaleExtensions extensions) {
798 if (language== null || script == null || country == null || variant == null) {
799 throw new NullPointerException();
800 }
801
802 if (extensions == null) {
803 extensions = getCompatibilityExtensions(language, script, country, variant);
804 }
805
806 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant);
807 return getInstance(baseloc, extensions);
808 }
809
810 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) {
811 LocaleKey key = new LocaleKey(baseloc, extensions);
812 return LOCALECACHE.get(key);
813 }
814
815 private static class Cache extends LocaleObjectCache<LocaleKey, Locale> {
816 private Cache() {
817 }
818
819 @Override
820 protected Locale createObject(LocaleKey key) {
821 return new Locale(key.base, key.exts);
822 }
823 }
824
825 private static final class LocaleKey {
826 private final BaseLocale base;
827 private final LocaleExtensions exts;
828 private final int hash;
829
830 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) {
831 base = baseLocale;
832 exts = extensions;
833
834 // Calculate the hash value here because it's always used.
835 int h = base.hashCode();
836 if (exts != null) {
837 h ^= exts.hashCode();
838 }
839 hash = h;
840 }
841
842 @Override
843 public boolean equals(Object obj) {
844 if (this == obj) {
845 return true;
846 }
847 if (!(obj instanceof LocaleKey)) {
848 return false;
849 }
850 LocaleKey other = (LocaleKey)obj;
851 if (hash != other.hash || !base.equals(other.base)) {
852 return false;
853 }
854 if (exts == null) {
855 return other.exts == null;
856 }
857 return exts.equals(other.exts);
858 }
859
860 @Override
861 public int hashCode() {
862 return hash;
863 }
864 }
865
866 /**
867 * Gets the current value of the default locale for this instance
868 * of the Java Virtual Machine.
869 * <p>
870 * The Java Virtual Machine sets the default locale during startup
871 * based on the host environment. It is used by many locale-sensitive
872 * methods if no locale is explicitly specified.
873 * It can be changed using the
874 * {@link #setDefault(java.util.Locale) setDefault} method.
875 *
876 * @return the default locale for this instance of the Java Virtual Machine
877 */
878 public static Locale getDefault() {
879 // do not synchronize this method - see 4071298
880 return defaultLocale;
881 }
882
883 /**
884 * Gets the current value of the default locale for the specified Category
885 * for this instance of the Java Virtual Machine.
886 * <p>
887 * The Java Virtual Machine sets the default locale during startup based
888 * on the host environment. It is used by many locale-sensitive methods
889 * if no locale is explicitly specified. It can be changed using the
890 * setDefault(Locale.Category, Locale) method.
891 *
892 * @param category - the specified category to get the default locale
893 * @throws NullPointerException if category is null
894 * @return the default locale for the specified Category for this instance
895 * of the Java Virtual Machine
896 * @see #setDefault(Locale.Category, Locale)
897 * @since 1.7
898 */
899 public static Locale getDefault(Locale.Category category) {
900 // do not synchronize this method - see 4071298
901 switch (category) {
902 case DISPLAY:
903 if (defaultDisplayLocale == null) {
904 synchronized(Locale.class) {
905 if (defaultDisplayLocale == null) {
906 defaultDisplayLocale = initDefault(category);
907 }
908 }
909 }
910 return defaultDisplayLocale;
911 case FORMAT:
912 if (defaultFormatLocale == null) {
913 synchronized(Locale.class) {
914 if (defaultFormatLocale == null) {
915 defaultFormatLocale = initDefault(category);
916 }
917 }
918 }
919 return defaultFormatLocale;
920 default:
921 assert false: "Unknown Category";
922 }
923 return getDefault();
924 }
925
926 private static Locale initDefault() {
927 String language, region, script, country, variant;
928 Properties props = GetPropertyAction.privilegedGetProperties();
929 language = props.getProperty("user.language", "en");
930 // for compatibility, check for old user.region property
931 region = props.getProperty("user.region");
932 if (region != null) {
933 // region can be of form country, country_variant, or _variant
934 int i = region.indexOf('_');
935 if (i >= 0) {
936 country = region.substring(0, i);
937 variant = region.substring(i + 1);
938 } else {
939 country = region;
940 variant = "";
941 }
942 script = "";
943 } else {
944 script = props.getProperty("user.script", "");
945 country = props.getProperty("user.country", "");
946 variant = props.getProperty("user.variant", "");
947 }
948
949 return getInstance(language, script, country, variant,
950 getDefaultExtensions(props.getProperty("user.extensions", ""))
951 .orElse(null));
952 }
953
954 private static Locale initDefault(Locale.Category category) {
955 Properties props = GetPropertyAction.privilegedGetProperties();
956
957 return getInstance(
958 props.getProperty(category.languageKey,
959 defaultLocale.getLanguage()),
960 props.getProperty(category.scriptKey,
961 defaultLocale.getScript()),
962 props.getProperty(category.countryKey,
963 defaultLocale.getCountry()),
964 props.getProperty(category.variantKey,
965 defaultLocale.getVariant()),
966 getDefaultExtensions(props.getProperty(category.extensionsKey, ""))
967 .orElse(defaultLocale.getLocaleExtensions()));
968 }
969
970 private static Optional<LocaleExtensions> getDefaultExtensions(String extensionsProp) {
971 LocaleExtensions exts = null;
972
973 try {
974 exts = new InternalLocaleBuilder()
975 .setExtensions(extensionsProp)
976 .getLocaleExtensions();
977 } catch (LocaleSyntaxException e) {
978 // just ignore this incorrect property
979 }
980
981 return Optional.ofNullable(exts);
982 }
983
984 /**
985 * Sets the default locale for this instance of the Java Virtual Machine.
986 * This does not affect the host locale.
987 * <p>
988 * If there is a security manager, its <code>checkPermission</code>
989 * method is called with a <code>PropertyPermission("user.language", "write")</code>
990 * permission before the default locale is changed.
991 * <p>
992 * The Java Virtual Machine sets the default locale during startup
993 * based on the host environment. It is used by many locale-sensitive
994 * methods if no locale is explicitly specified.
995 * <p>
996 * Since changing the default locale may affect many different areas
997 * of functionality, this method should only be used if the caller
998 * is prepared to reinitialize locale-sensitive code running
999 * within the same Java Virtual Machine.
1000 * <p>
1001 * By setting the default locale with this method, all of the default
1002 * locales for each Category are also set to the specified default locale.
1003 *
1004 * @throws SecurityException
1005 * if a security manager exists and its
1006 * <code>checkPermission</code> method doesn't allow the operation.
1007 * @throws NullPointerException if <code>newLocale</code> is null
1008 * @param newLocale the new default locale
1009 * @see SecurityManager#checkPermission
1010 * @see java.util.PropertyPermission
1011 */
1012 public static synchronized void setDefault(Locale newLocale) {
1013 setDefault(Category.DISPLAY, newLocale);
1014 setDefault(Category.FORMAT, newLocale);
1015 defaultLocale = newLocale;
1016 }
1017
1018 /**
1019 * Sets the default locale for the specified Category for this instance
1020 * of the Java Virtual Machine. This does not affect the host locale.
1021 * <p>
1022 * If there is a security manager, its checkPermission method is called
1023 * with a PropertyPermission("user.language", "write") permission before
1024 * the default locale is changed.
1025 * <p>
1026 * The Java Virtual Machine sets the default locale during startup based
1027 * on the host environment. It is used by many locale-sensitive methods
1028 * if no locale is explicitly specified.
1029 * <p>
1030 * Since changing the default locale may affect many different areas of
1031 * functionality, this method should only be used if the caller is
1032 * prepared to reinitialize locale-sensitive code running within the
1033 * same Java Virtual Machine.
1034 *
1035 * @param category - the specified category to set the default locale
1036 * @param newLocale - the new default locale
1037 * @throws SecurityException if a security manager exists and its
1038 * checkPermission method doesn't allow the operation.
1039 * @throws NullPointerException if category and/or newLocale is null
1040 * @see SecurityManager#checkPermission(java.security.Permission)
1041 * @see PropertyPermission
1042 * @see #getDefault(Locale.Category)
1043 * @since 1.7
1044 */
1045 public static synchronized void setDefault(Locale.Category category,
1046 Locale newLocale) {
1047 if (category == null)
1048 throw new NullPointerException("Category cannot be NULL");
1049 if (newLocale == null)
1050 throw new NullPointerException("Can't set default locale to NULL");
1051
1052 SecurityManager sm = System.getSecurityManager();
1053 if (sm != null) sm.checkPermission(new PropertyPermission
1054 ("user.language", "write"));
1055 switch (category) {
1056 case DISPLAY:
1057 defaultDisplayLocale = newLocale;
1058 break;
1059 case FORMAT:
1060 defaultFormatLocale = newLocale;
1061 break;
1062 default:
1063 assert false: "Unknown Category";
1064 }
1065 }
1066
1067 /**
1068 * Returns an array of all installed locales.
1069 * The returned array represents the union of locales supported
1070 * by the Java runtime environment and by installed
1071 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider}
1072 * implementations. It must contain at least a <code>Locale</code>
1073 * instance equal to {@link java.util.Locale#US Locale.US}.
1074 *
1075 * @return An array of installed locales.
1076 */
1077 public static Locale[] getAvailableLocales() {
1078 return LocaleServiceProviderPool.getAllAvailableLocales();
1079 }
1080
1081 /**
1082 * Returns a list of all 2-letter country codes defined in ISO 3166.
1083 * Can be used to create Locales.
1084 * This method is equivalent to {@link #getISOCountries(Locale.IsoCountryCode type)}
1085 * with {@code type} {@link IsoCountryCode#PART1_ALPHA2}.
1086 * <p>
1087 * <b>Note:</b> The <code>Locale</code> class also supports other codes for
1088 * country (region), such as 3-letter numeric UN M.49 area codes.
1089 * Therefore, the list returned by this method does not contain ALL valid
1090 * codes that can be used to create Locales.
1091 * <p>
1092 * Note that this method does not return obsolete 2-letter country codes.
1093 * ISO3166-3 codes which designate country codes for those obsolete codes,
1094 * can be retrieved from {@link #getISOCountries(Locale.IsoCountryCode type)} with
1095 * {@code type} {@link IsoCountryCode#PART3}.
1096 * @return An array of ISO 3166 two-letter country codes.
1097 */
1098 public static String[] getISOCountries() {
1099 if (isoCountries == null) {
1100 isoCountries = getISO2Table(LocaleISOData.isoCountryTable);
1101 }
1102 String[] result = new String[isoCountries.length];
1103 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length);
1104 return result;
1105 }
1106
1107 /**
1108 * Returns a {@code Set} of ISO3166 country codes for the specified type.
1109 *
1110 * @param type {@link Locale.IsoCountryCode} specified ISO code type.
1111 * @see java.util.Locale.IsoCountryCode
1112 * @throws NullPointerException if type is null
1113 * @return a {@code Set} of ISO country codes for the specified type.
1114 * @since 9
1115 */
1116 public static Set<String> getISOCountries(IsoCountryCode type) {
1117 Objects.requireNonNull(type);
1118 return IsoCountryCode.retrieveISOCountryCodes(type);
1119 }
1120
1121 /**
1122 * Returns a list of all 2-letter language codes defined in ISO 639.
1123 * Can be used to create Locales.
1124 * <p>
1125 * <b>Note:</b>
1126 * <ul>
1127 * <li>ISO 639 is not a stable standard— some languages' codes have changed.
1128 * The list this function returns includes both the new and the old codes for the
1129 * languages whose codes have changed.
1130 * <li>The <code>Locale</code> class also supports language codes up to
1131 * 8 characters in length. Therefore, the list returned by this method does
1132 * not contain ALL valid codes that can be used to create Locales.
1133 * </ul>
1134 *
1135 * @return An array of ISO 639 two-letter language codes.
1136 */
1137 public static String[] getISOLanguages() {
1138 if (isoLanguages == null) {
1139 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable);
1140 }
1141 String[] result = new String[isoLanguages.length];
1142 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length);
1143 return result;
1144 }
1145
1146 private static String[] getISO2Table(String table) {
1147 int len = table.length() / 5;
1148 String[] isoTable = new String[len];
1149 for (int i = 0, j = 0; i < len; i++, j += 5) {
1150 isoTable[i] = table.substring(j, j + 2);
1151 }
1152 return isoTable;
1153 }
1154
1155 /**
1156 * Returns the language code of this Locale.
1157 *
1158 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed.
1159 * Locale's constructor recognizes both the new and the old codes for the languages
1160 * whose codes have changed, but this function always returns the old code. If you
1161 * want to check for a specific language whose code has changed, don't do
1162 * <pre>
1163 * if (locale.getLanguage().equals("he")) // BAD!
1164 * ...
1165 * </pre>
1166 * Instead, do
1167 * <pre>
1168 * if (locale.getLanguage().equals(new Locale("he").getLanguage()))
1169 * ...
1170 * </pre>
1171 * @return The language code, or the empty string if none is defined.
1172 * @see #getDisplayLanguage
1173 */
1174 public String getLanguage() {
1175 return baseLocale.getLanguage();
1176 }
1177
1178 /**
1179 * Returns the script for this locale, which should
1180 * either be the empty string or an ISO 15924 4-letter script
1181 * code. The first letter is uppercase and the rest are
1182 * lowercase, for example, 'Latn', 'Cyrl'.
1183 *
1184 * @return The script code, or the empty string if none is defined.
1185 * @see #getDisplayScript
1186 * @since 1.7
1187 */
1188 public String getScript() {
1189 return baseLocale.getScript();
1190 }
1191
1192 /**
1193 * Returns the country/region code for this locale, which should
1194 * either be the empty string, an uppercase ISO 3166 2-letter code,
1195 * or a UN M.49 3-digit code.
1196 *
1197 * @return The country/region code, or the empty string if none is defined.
1198 * @see #getDisplayCountry
1199 */
1200 public String getCountry() {
1201 return baseLocale.getRegion();
1202 }
1203
1204 /**
1205 * Returns the variant code for this locale.
1206 *
1207 * @return The variant code, or the empty string if none is defined.
1208 * @see #getDisplayVariant
1209 */
1210 public String getVariant() {
1211 return baseLocale.getVariant();
1212 }
1213
1214 /**
1215 * Returns {@code true} if this {@code Locale} has any <a href="#def_extensions">
1216 * extensions</a>.
1217 *
1218 * @return {@code true} if this {@code Locale} has any extensions
1219 * @since 1.8
1220 */
1221 public boolean hasExtensions() {
1222 return localeExtensions != null;
1223 }
1224
1225 /**
1226 * Returns a copy of this {@code Locale} with no <a href="#def_extensions">
1227 * extensions</a>. If this {@code Locale} has no extensions, this {@code Locale}
1228 * is returned.
1229 *
1230 * @return a copy of this {@code Locale} with no extensions, or {@code this}
1231 * if {@code this} has no extensions
1232 * @since 1.8
1233 */
1234 public Locale stripExtensions() {
1235 return hasExtensions() ? Locale.getInstance(baseLocale, null) : this;
1236 }
1237
1238 /**
1239 * Returns the extension (or private use) value associated with
1240 * the specified key, or null if there is no extension
1241 * associated with the key. To be well-formed, the key must be one
1242 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so
1243 * for example 'z' and 'Z' represent the same extension.
1244 *
1245 * @param key the extension key
1246 * @return The extension, or null if this locale defines no
1247 * extension for the specified key.
1248 * @throws IllegalArgumentException if key is not well-formed
1249 * @see #PRIVATE_USE_EXTENSION
1250 * @see #UNICODE_LOCALE_EXTENSION
1251 * @since 1.7
1252 */
1253 public String getExtension(char key) {
1254 if (!LocaleExtensions.isValidKey(key)) {
1255 throw new IllegalArgumentException("Ill-formed extension key: " + key);
1256 }
1257 return hasExtensions() ? localeExtensions.getExtensionValue(key) : null;
1258 }
1259
1260 /**
1261 * Returns the set of extension keys associated with this locale, or the
1262 * empty set if it has no extensions. The returned set is unmodifiable.
1263 * The keys will all be lower-case.
1264 *
1265 * @return The set of extension keys, or the empty set if this locale has
1266 * no extensions.
1267 * @since 1.7
1268 */
1269 public Set<Character> getExtensionKeys() {
1270 if (!hasExtensions()) {
1271 return Collections.emptySet();
1272 }
1273 return localeExtensions.getKeys();
1274 }
1275
1276 /**
1277 * Returns the set of unicode locale attributes associated with
1278 * this locale, or the empty set if it has no attributes. The
1279 * returned set is unmodifiable.
1280 *
1281 * @return The set of attributes.
1282 * @since 1.7
1283 */
1284 public Set<String> getUnicodeLocaleAttributes() {
1285 if (!hasExtensions()) {
1286 return Collections.emptySet();
1287 }
1288 return localeExtensions.getUnicodeLocaleAttributes();
1289 }
1290
1291 /**
1292 * Returns the Unicode locale type associated with the specified Unicode locale key
1293 * for this locale. Returns the empty string for keys that are defined with no type.
1294 * Returns null if the key is not defined. Keys are case-insensitive. The key must
1295 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is
1296 * thrown.
1297 *
1298 * @param key the Unicode locale key
1299 * @return The Unicode locale type associated with the key, or null if the
1300 * locale does not define the key.
1301 * @throws IllegalArgumentException if the key is not well-formed
1302 * @throws NullPointerException if <code>key</code> is null
1303 * @since 1.7
1304 */
1305 public String getUnicodeLocaleType(String key) {
1306 if (!isUnicodeExtensionKey(key)) {
1307 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key);
1308 }
1309 return hasExtensions() ? localeExtensions.getUnicodeLocaleType(key) : null;
1310 }
1311
1312 /**
1313 * Returns the set of Unicode locale keys defined by this locale, or the empty set if
1314 * this locale has none. The returned set is immutable. Keys are all lower case.
1315 *
1316 * @return The set of Unicode locale keys, or the empty set if this locale has
1317 * no Unicode locale keywords.
1318 * @since 1.7
1319 */
1320 public Set<String> getUnicodeLocaleKeys() {
1321 if (localeExtensions == null) {
1322 return Collections.emptySet();
1323 }
1324 return localeExtensions.getUnicodeLocaleKeys();
1325 }
1326
1327 /**
1328 * Package locale method returning the Locale's BaseLocale,
1329 * used by ResourceBundle
1330 * @return base locale of this Locale
1331 */
1332 BaseLocale getBaseLocale() {
1333 return baseLocale;
1334 }
1335
1336 /**
1337 * Package private method returning the Locale's LocaleExtensions,
1338 * used by ResourceBundle.
1339 * @return locale extensions of this Locale,
1340 * or {@code null} if no extensions are defined
1341 */
1342 LocaleExtensions getLocaleExtensions() {
1343 return localeExtensions;
1344 }
1345
1346 /**
1347 * Returns a string representation of this <code>Locale</code>
1348 * object, consisting of language, country, variant, script,
1349 * and extensions as below:
1350 * <blockquote>
1351 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "_" + extensions
1352 * </blockquote>
1353 *
1354 * Language is always lower case, country is always upper case, script is always title
1355 * case, and extensions are always lower case. Extensions and private use subtags
1356 * will be in canonical order as explained in {@link #toLanguageTag}.
1357 *
1358 * <p>When the locale has neither script nor extensions, the result is the same as in
1359 * Java 6 and prior.
1360 *
1361 * <p>If both the language and country fields are missing, this function will return
1362 * the empty string, even if the variant, script, or extensions field is present (you
1363 * can't have a locale with just a variant, the variant must accompany a well-formed
1364 * language or country code).
1365 *
1366 * <p>If script or extensions are present and variant is missing, no underscore is
1367 * added before the "#".
1368 *
1369 * <p>This behavior is designed to support debugging and to be compatible with
1370 * previous uses of <code>toString</code> that expected language, country, and variant
1371 * fields only. To represent a Locale as a String for interchange purposes, use
1372 * {@link #toLanguageTag}.
1373 *
1374 * <p>Examples: <ul>
1375 * <li>{@code en}</li>
1376 * <li>{@code de_DE}</li>
1377 * <li>{@code _GB}</li>
1378 * <li>{@code en_US_WIN}</li>
1379 * <li>{@code de__POSIX}</li>
1380 * <li>{@code zh_CN_#Hans}</li>
1381 * <li>{@code zh_TW_#Hant_x-java}</li>
1382 * <li>{@code th_TH_TH_#u-nu-thai}</li></ul>
1383 *
1384 * @return A string representation of the Locale, for debugging.
1385 * @see #getDisplayName
1386 * @see #toLanguageTag
1387 */
1388 @Override
1389 public final String toString() {
1390 boolean l = (baseLocale.getLanguage().length() != 0);
1391 boolean s = (baseLocale.getScript().length() != 0);
1392 boolean r = (baseLocale.getRegion().length() != 0);
1393 boolean v = (baseLocale.getVariant().length() != 0);
1394 boolean e = (localeExtensions != null && localeExtensions.getID().length() != 0);
1395
1396 StringBuilder result = new StringBuilder(baseLocale.getLanguage());
1397 if (r || (l && (v || s || e))) {
1398 result.append('_')
1399 .append(baseLocale.getRegion()); // This may just append '_'
1400 }
1401 if (v && (l || r)) {
1402 result.append('_')
1403 .append(baseLocale.getVariant());
1404 }
1405
1406 if (s && (l || r)) {
1407 result.append("_#")
1408 .append(baseLocale.getScript());
1409 }
1410
1411 if (e && (l || r)) {
1412 result.append('_');
1413 if (!s) {
1414 result.append('#');
1415 }
1416 result.append(localeExtensions.getID());
1417 }
1418
1419 return result.toString();
1420 }
1421
1422 /**
1423 * Returns a well-formed IETF BCP 47 language tag representing
1424 * this locale.
1425 *
1426 * <p>If this <code>Locale</code> has a language, country, or
1427 * variant that does not satisfy the IETF BCP 47 language tag
1428 * syntax requirements, this method handles these fields as
1429 * described below:
1430 *
1431 * <p><b>Language:</b> If language is empty, or not <a
1432 * href="#def_language" >well-formed</a> (for example "a" or
1433 * "e2"), it will be emitted as "und" (Undetermined).
1434 *
1435 * <p><b>Country:</b> If country is not <a
1436 * href="#def_region">well-formed</a> (for example "12" or "USA"),
1437 * it will be omitted.
1438 *
1439 * <p><b>Variant:</b> If variant <b>is</b> <a
1440 * href="#def_variant">well-formed</a>, each sub-segment
1441 * (delimited by '-' or '_') is emitted as a subtag. Otherwise:
1442 * <ul>
1443 *
1444 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code>
1445 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first
1446 * ill-formed sub-segment and all following will be appended to
1447 * the private use subtag. The first appended subtag will be
1448 * "lvariant", followed by the sub-segments in order, separated by
1449 * hyphen. For example, "x-lvariant-WIN",
1450 * "Oracle-x-lvariant-JDK-Standard-Edition".
1451 *
1452 * <li>if any sub-segment does not match
1453 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated
1454 * and the problematic sub-segment and all following sub-segments
1455 * will be omitted. If the remainder is non-empty, it will be
1456 * emitted as a private use subtag as above (even if the remainder
1457 * turns out to be well-formed). For example,
1458 * "Solaris_isjustthecoolestthing" is emitted as
1459 * "x-lvariant-Solaris", not as "solaris".</li></ul>
1460 *
1461 * <p><b>Special Conversions:</b> Java supports some old locale
1462 * representations, including deprecated ISO language codes,
1463 * for compatibility. This method performs the following
1464 * conversions:
1465 * <ul>
1466 *
1467 * <li>Deprecated ISO language codes "iw", "ji", and "in" are
1468 * converted to "he", "yi", and "id", respectively.
1469 *
1470 * <li>A locale with language "no", country "NO", and variant
1471 * "NY", representing Norwegian Nynorsk (Norway), is converted
1472 * to a language tag "nn-NO".</li></ul>
1473 *
1474 * <p><b>Note:</b> Although the language tag created by this
1475 * method is well-formed (satisfies the syntax requirements
1476 * defined by the IETF BCP 47 specification), it is not
1477 * necessarily a valid BCP 47 language tag. For example,
1478 * <pre>
1479 * new Locale("xx", "YY").toLanguageTag();</pre>
1480 *
1481 * will return "xx-YY", but the language subtag "xx" and the
1482 * region subtag "YY" are invalid because they are not registered
1483 * in the IANA Language Subtag Registry.
1484 *
1485 * @return a BCP47 language tag representing the locale
1486 * @see #forLanguageTag(String)
1487 * @since 1.7
1488 */
1489 public String toLanguageTag() {
1490 if (languageTag != null) {
1491 return languageTag;
1492 }
1493
1494 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions);
1495 StringBuilder buf = new StringBuilder();
1496
1497 String subtag = tag.getLanguage();
1498 if (subtag.length() > 0) {
1499 buf.append(LanguageTag.canonicalizeLanguage(subtag));
1500 }
1501
1502 subtag = tag.getScript();
1503 if (subtag.length() > 0) {
1504 buf.append(LanguageTag.SEP);
1505 buf.append(LanguageTag.canonicalizeScript(subtag));
1506 }
1507
1508 subtag = tag.getRegion();
1509 if (subtag.length() > 0) {
1510 buf.append(LanguageTag.SEP);
1511 buf.append(LanguageTag.canonicalizeRegion(subtag));
1512 }
1513
1514 List<String>subtags = tag.getVariants();
1515 for (String s : subtags) {
1516 buf.append(LanguageTag.SEP);
1517 // preserve casing
1518 buf.append(s);
1519 }
1520
1521 subtags = tag.getExtensions();
1522 for (String s : subtags) {
1523 buf.append(LanguageTag.SEP);
1524 buf.append(LanguageTag.canonicalizeExtension(s));
1525 }
1526
1527 subtag = tag.getPrivateuse();
1528 if (subtag.length() > 0) {
1529 if (buf.length() > 0) {
1530 buf.append(LanguageTag.SEP);
1531 }
1532 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP);
1533 // preserve casing
1534 buf.append(subtag);
1535 }
1536
1537 String langTag = buf.toString();
1538 synchronized (this) {
1539 if (languageTag == null) {
1540 languageTag = langTag;
1541 }
1542 }
1543 return languageTag;
1544 }
1545
1546 /**
1547 * Returns a locale for the specified IETF BCP 47 language tag string.
1548 *
1549 * <p>If the specified language tag contains any ill-formed subtags,
1550 * the first such subtag and all following subtags are ignored. Compare
1551 * to {@link Locale.Builder#setLanguageTag} which throws an exception
1552 * in this case.
1553 *
1554 * <p>The following <b>conversions</b> are performed:<ul>
1555 *
1556 * <li>The language code "und" is mapped to language "".
1557 *
1558 * <li>The language codes "he", "yi", and "id" are mapped to "iw",
1559 * "ji", and "in" respectively. (This is the same canonicalization
1560 * that's done in Locale's constructors.)
1561 *
1562 * <li>The portion of a private use subtag prefixed by "lvariant",
1563 * if any, is removed and appended to the variant field in the
1564 * result locale (without case normalization). If it is then
1565 * empty, the private use subtag is discarded:
1566 *
1567 * <pre>
1568 * Locale loc;
1569 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX");
1570 * loc.getVariant(); // returns "POSIX"
1571 * loc.getExtension('x'); // returns null
1572 *
1573 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def");
1574 * loc.getVariant(); // returns "POSIX_Abc_Def"
1575 * loc.getExtension('x'); // returns "urp"
1576 * </pre>
1577 *
1578 * <li>When the languageTag argument contains an extlang subtag,
1579 * the first such subtag is used as the language, and the primary
1580 * language subtag and other extlang subtags are ignored:
1581 *
1582 * <pre>
1583 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao"
1584 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US"
1585 * </pre>
1586 *
1587 * <li>Case is normalized except for variant tags, which are left
1588 * unchanged. Language is normalized to lower case, script to
1589 * title case, country to upper case, and extensions to lower
1590 * case.
1591 *
1592 * <li>If, after processing, the locale would exactly match either
1593 * ja_JP_JP or th_TH_TH with no extensions, the appropriate
1594 * extensions are added as though the constructor had been called:
1595 *
1596 * <pre>
1597 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag();
1598 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP"
1599 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag();
1600 * // returns "th-TH-u-nu-thai-x-lvariant-TH"
1601 * </pre></ul>
1602 *
1603 * <p>This implements the 'Language-Tag' production of BCP47, and
1604 * so supports grandfathered (regular and irregular) as well as
1605 * private use language tags. Stand alone private use tags are
1606 * represented as empty language and extension 'x-whatever',
1607 * and grandfathered tags are converted to their canonical replacements
1608 * where they exist.
1609 *
1610 * <p>Grandfathered tags with canonical replacements are as follows:
1611 *
1612 * <table class="striped">
1613 * <caption style="display:none">Grandfathered tags with canonical replacements</caption>
1614 * <thead style="text-align:center">
1615 * <tr><th scope="col" style="padding: 0 2px">grandfathered tag</th><th scope="col" style="padding: 0 2px">modern replacement</th></tr>
1616 * </thead>
1617 * <tbody style="text-align:center">
1618 * <tr><th scope="row">art-lojban</th><td>jbo</td></tr>
1619 * <tr><th scope="row">i-ami</th><td>ami</td></tr>
1620 * <tr><th scope="row">i-bnn</th><td>bnn</td></tr>
1621 * <tr><th scope="row">i-hak</th><td>hak</td></tr>
1622 * <tr><th scope="row">i-klingon</th><td>tlh</td></tr>
1623 * <tr><th scope="row">i-lux</th><td>lb</td></tr>
1624 * <tr><th scope="row">i-navajo</th><td>nv</td></tr>
1625 * <tr><th scope="row">i-pwn</th><td>pwn</td></tr>
1626 * <tr><th scope="row">i-tao</th><td>tao</td></tr>
1627 * <tr><th scope="row">i-tay</th><td>tay</td></tr>
1628 * <tr><th scope="row">i-tsu</th><td>tsu</td></tr>
1629 * <tr><th scope="row">no-bok</th><td>nb</td></tr>
1630 * <tr><th scope="row">no-nyn</th><td>nn</td></tr>
1631 * <tr><th scope="row">sgn-BE-FR</th><td>sfb</td></tr>
1632 * <tr><th scope="row">sgn-BE-NL</th><td>vgt</td></tr>
1633 * <tr><th scope="row">sgn-CH-DE</th><td>sgg</td></tr>
1634 * <tr><th scope="row">zh-guoyu</th><td>cmn</td></tr>
1635 * <tr><th scope="row">zh-hakka</th><td>hak</td></tr>
1636 * <tr><th scope="row">zh-min-nan</th><td>nan</td></tr>
1637 * <tr><th scope="row">zh-xiang</th><td>hsn</td></tr>
1638 * </tbody>
1639 * </table>
1640 *
1641 * <p>Grandfathered tags with no modern replacement will be
1642 * converted as follows:
1643 *
1644 * <table class="striped">
1645 * <caption style="display:none">Grandfathered tags with no modern replacement</caption>
1646 * <thead style="text-align:center">
1647 * <tr><th scope="col" style="padding: 0 2px">grandfathered tag</th><th scope="col" style="padding: 0 2px">converts to</th></tr>
1648 * </thead>
1649 * <tbody style="text-align:center">
1650 * <tr><th scope="row">cel-gaulish</th><td>xtg-x-cel-gaulish</td></tr>
1651 * <tr><th scope="row">en-GB-oed</th><td>en-GB-x-oed</td></tr>
1652 * <tr><th scope="row">i-default</th><td>en-x-i-default</td></tr>
1653 * <tr><th scope="row">i-enochian</th><td>und-x-i-enochian</td></tr>
1654 * <tr><th scope="row">i-mingo</th><td>see-x-i-mingo</td></tr>
1655 * <tr><th scope="row">zh-min</th><td>nan-x-zh-min</td></tr>
1656 * </tbody>
1657 * </table>
1658 *
1659 * <p>For a list of all grandfathered tags, see the
1660 * IANA Language Subtag Registry (search for "Type: grandfathered").
1661 *
1662 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code>
1663 * and <code>forLanguageTag</code> will round-trip.
1664 *
1665 * @param languageTag the language tag
1666 * @return The locale that best represents the language tag.
1667 * @throws NullPointerException if <code>languageTag</code> is <code>null</code>
1668 * @see #toLanguageTag()
1669 * @see java.util.Locale.Builder#setLanguageTag(String)
1670 * @since 1.7
1671 */
1672 public static Locale forLanguageTag(String languageTag) {
1673 LanguageTag tag = LanguageTag.parse(languageTag, null);
1674 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
1675 bldr.setLanguageTag(tag);
1676 BaseLocale base = bldr.getBaseLocale();
1677 LocaleExtensions exts = bldr.getLocaleExtensions();
1678 if (exts == null && base.getVariant().length() > 0) {
1679 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(),
1680 base.getRegion(), base.getVariant());
1681 }
1682 return getInstance(base, exts);
1683 }
1684
1685 /**
1686 * Returns a three-letter abbreviation of this locale's language.
1687 * If the language matches an ISO 639-1 two-letter code, the
1688 * corresponding ISO 639-2/T three-letter lowercase code is
1689 * returned. The ISO 639-2 language codes can be found on-line,
1690 * see "Codes for the Representation of Names of Languages Part 2:
1691 * Alpha-3 Code". If the locale specifies a three-letter
1692 * language, the language is returned as is. If the locale does
1693 * not specify a language the empty string is returned.
1694 *
1695 * @return A three-letter abbreviation of this locale's language.
1696 * @exception MissingResourceException Throws MissingResourceException if
1697 * three-letter language abbreviation is not available for this locale.
1698 */
1699 public String getISO3Language() throws MissingResourceException {
1700 String lang = baseLocale.getLanguage();
1701 if (lang.length() == 3) {
1702 return lang;
1703 }
1704
1705 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable);
1706 if (language3 == null) {
1707 throw new MissingResourceException("Couldn't find 3-letter language code for "
1708 + lang, "FormatData_" + toString(), "ShortLanguage");
1709 }
1710 return language3;
1711 }
1712
1713 /**
1714 * Returns a three-letter abbreviation for this locale's country.
1715 * If the country matches an ISO 3166-1 alpha-2 code, the
1716 * corresponding ISO 3166-1 alpha-3 uppercase code is returned.
1717 * If the locale doesn't specify a country, this will be the empty
1718 * string.
1719 *
1720 * <p>The ISO 3166-1 codes can be found on-line.
1721 *
1722 * @return A three-letter abbreviation of this locale's country.
1723 * @exception MissingResourceException Throws MissingResourceException if the
1724 * three-letter country abbreviation is not available for this locale.
1725 */
1726 public String getISO3Country() throws MissingResourceException {
1727 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable);
1728 if (country3 == null) {
1729 throw new MissingResourceException("Couldn't find 3-letter country code for "
1730 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry");
1731 }
1732 return country3;
1733 }
1734
1735 private static String getISO3Code(String iso2Code, String table) {
1736 int codeLength = iso2Code.length();
1737 if (codeLength == 0) {
1738 return "";
1739 }
1740
1741 int tableLength = table.length();
1742 int index = tableLength;
1743 if (codeLength == 2) {
1744 char c1 = iso2Code.charAt(0);
1745 char c2 = iso2Code.charAt(1);
1746 for (index = 0; index < tableLength; index += 5) {
1747 if (table.charAt(index) == c1
1748 && table.charAt(index + 1) == c2) {
1749 break;
1750 }
1751 }
1752 }
1753 return index < tableLength ? table.substring(index + 2, index + 5) : null;
1754 }
1755
1756 /**
1757 * Returns a name for the locale's language that is appropriate for display to the
1758 * user.
1759 * If possible, the name returned will be localized for the default
1760 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1761 * For example, if the locale is fr_FR and the default
1762 * {@link Locale.Category#DISPLAY DISPLAY} locale
1763 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1764 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1765 * getDisplayLanguage() will return "anglais".
1766 * If the name returned cannot be localized for the default
1767 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1768 * (say, we don't have a Japanese name for Croatian),
1769 * this function falls back on the English name, and uses the ISO code as a last-resort
1770 * value. If the locale doesn't specify a language, this function returns the empty string.
1771 *
1772 * @return The name of the display language.
1773 */
1774 public final String getDisplayLanguage() {
1775 return getDisplayLanguage(getDefault(Category.DISPLAY));
1776 }
1777
1778 /**
1779 * Returns a name for the locale's language that is appropriate for display to the
1780 * user.
1781 * If possible, the name returned will be localized according to inLocale.
1782 * For example, if the locale is fr_FR and inLocale
1783 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and
1784 * inLocale is fr_FR, getDisplayLanguage() will return "anglais".
1785 * If the name returned cannot be localized according to inLocale,
1786 * (say, we don't have a Japanese name for Croatian),
1787 * this function falls back on the English name, and finally
1788 * on the ISO code as a last-resort value. If the locale doesn't specify a language,
1789 * this function returns the empty string.
1790 *
1791 * @param inLocale The locale for which to retrieve the display language.
1792 * @return The name of the display language appropriate to the given locale.
1793 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1794 */
1795 public String getDisplayLanguage(Locale inLocale) {
1796 return getDisplayString(baseLocale.getLanguage(), null, inLocale, DISPLAY_LANGUAGE);
1797 }
1798
1799 /**
1800 * Returns a name for the locale's script that is appropriate for display to
1801 * the user. If possible, the name will be localized for the default
1802 * {@link Locale.Category#DISPLAY DISPLAY} locale. Returns
1803 * the empty string if this locale doesn't specify a script code.
1804 *
1805 * @return the display name of the script code for the current default
1806 * {@link Locale.Category#DISPLAY DISPLAY} locale
1807 * @since 1.7
1808 */
1809 public String getDisplayScript() {
1810 return getDisplayScript(getDefault(Category.DISPLAY));
1811 }
1812
1813 /**
1814 * Returns a name for the locale's script that is appropriate
1815 * for display to the user. If possible, the name will be
1816 * localized for the given locale. Returns the empty string if
1817 * this locale doesn't specify a script code.
1818 *
1819 * @param inLocale The locale for which to retrieve the display script.
1820 * @return the display name of the script code for the current default
1821 * {@link Locale.Category#DISPLAY DISPLAY} locale
1822 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1823 * @since 1.7
1824 */
1825 public String getDisplayScript(Locale inLocale) {
1826 return getDisplayString(baseLocale.getScript(), null, inLocale, DISPLAY_SCRIPT);
1827 }
1828
1829 /**
1830 * Returns a name for the locale's country that is appropriate for display to the
1831 * user.
1832 * If possible, the name returned will be localized for the default
1833 * {@link Locale.Category#DISPLAY DISPLAY} locale.
1834 * For example, if the locale is fr_FR and the default
1835 * {@link Locale.Category#DISPLAY DISPLAY} locale
1836 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1837 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR,
1838 * getDisplayCountry() will return "Etats-Unis".
1839 * If the name returned cannot be localized for the default
1840 * {@link Locale.Category#DISPLAY DISPLAY} locale,
1841 * (say, we don't have a Japanese name for Croatia),
1842 * this function falls back on the English name, and uses the ISO code as a last-resort
1843 * value. If the locale doesn't specify a country, this function returns the empty string.
1844 *
1845 * @return The name of the country appropriate to the locale.
1846 */
1847 public final String getDisplayCountry() {
1848 return getDisplayCountry(getDefault(Category.DISPLAY));
1849 }
1850
1851 /**
1852 * Returns a name for the locale's country that is appropriate for display to the
1853 * user.
1854 * If possible, the name returned will be localized according to inLocale.
1855 * For example, if the locale is fr_FR and inLocale
1856 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and
1857 * inLocale is fr_FR, getDisplayCountry() will return "Etats-Unis".
1858 * If the name returned cannot be localized according to inLocale.
1859 * (say, we don't have a Japanese name for Croatia),
1860 * this function falls back on the English name, and finally
1861 * on the ISO code as a last-resort value. If the locale doesn't specify a country,
1862 * this function returns the empty string.
1863 *
1864 * @param inLocale The locale for which to retrieve the display country.
1865 * @return The name of the country appropriate to the given locale.
1866 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1867 */
1868 public String getDisplayCountry(Locale inLocale) {
1869 return getDisplayString(baseLocale.getRegion(), null, inLocale, DISPLAY_COUNTRY);
1870 }
1871
1872 private String getDisplayString(String code, String cat, Locale inLocale, int type) {
1873 Objects.requireNonNull(inLocale);
1874 Objects.requireNonNull(code);
1875
1876 if (code.isEmpty()) {
1877 return "";
1878 }
1879
1880 LocaleServiceProviderPool pool =
1881 LocaleServiceProviderPool.getPool(LocaleNameProvider.class);
1882 String rbKey = (type == DISPLAY_VARIANT ? "%%"+code : code);
1883 String result = pool.getLocalizedObject(
1884 LocaleNameGetter.INSTANCE,
1885 inLocale, rbKey, type, code, cat);
1886 return result != null ? result : code;
1887 }
1888
1889 /**
1890 * Returns a name for the locale's variant code that is appropriate for display to the
1891 * user. If possible, the name will be localized for the default
1892 * {@link Locale.Category#DISPLAY DISPLAY} locale. If the locale
1893 * doesn't specify a variant code, this function returns the empty string.
1894 *
1895 * @return The name of the display variant code appropriate to the locale.
1896 */
1897 public final String getDisplayVariant() {
1898 return getDisplayVariant(getDefault(Category.DISPLAY));
1899 }
1900
1901 /**
1902 * Returns a name for the locale's variant code that is appropriate for display to the
1903 * user. If possible, the name will be localized for inLocale. If the locale
1904 * doesn't specify a variant code, this function returns the empty string.
1905 *
1906 * @param inLocale The locale for which to retrieve the display variant code.
1907 * @return The name of the display variant code appropriate to the given locale.
1908 * @exception NullPointerException if <code>inLocale</code> is <code>null</code>
1909 */
1910 public String getDisplayVariant(Locale inLocale) {
1911 if (baseLocale.getVariant().length() == 0)
1912 return "";
1913
1914 LocaleResources lr = LocaleProviderAdapter
1915 .getResourceBundleBased()
1916 .getLocaleResources(inLocale);
1917
1918 String names[] = getDisplayVariantArray(inLocale);
1919
1920 // Get the localized patterns for formatting a list, and use
1921 // them to format the list.
1922 return formatList(names,
1923 lr.getLocaleName("ListCompositionPattern"));
1924 }
1925
1926 /**
1927 * Returns a name for the locale that is appropriate for display to the
1928 * user. This will be the values returned by getDisplayLanguage(),
1929 * getDisplayScript(), getDisplayCountry(), getDisplayVariant() and
1930 * optional <a href="./Locale.html#def_locale_extension">Unicode extensions</a>
1931 * assembled into a single string. The non-empty values are used in order, with
1932 * the second and subsequent names in parentheses. For example:
1933 * <blockquote>
1934 * language (script, country, variant(, extension)*)<br>
1935 * language (country(, extension)*)<br>
1936 * language (variant(, extension)*)<br>
1937 * script (country(, extension)*)<br>
1938 * country (extension)*<br>
1939 * </blockquote>
1940 * depending on which fields are specified in the locale. If the
1941 * language, script, country, and variant fields are all empty,
1942 * this function returns the empty string.
1943 *
1944 * @return The name of the locale appropriate to display.
1945 */
1946 public final String getDisplayName() {
1947 return getDisplayName(getDefault(Category.DISPLAY));
1948 }
1949
1950 /**
1951 * Returns a name for the locale that is appropriate for display
1952 * to the user. This will be the values returned by
1953 * getDisplayLanguage(), getDisplayScript(),getDisplayCountry()
1954 * getDisplayVariant(), and optional <a href="./Locale.html#def_locale_extension">
1955 * Unicode extensions</a> assembled into a single string. The non-empty
1956 * values are used in order, with the second and subsequent names in
1957 * parentheses. For example:
1958 * <blockquote>
1959 * language (script, country, variant(, extension)*)<br>
1960 * language (country(, extension)*)<br>
1961 * language (variant(, extension)*)<br>
1962 * script (country(, extension)*)<br>
1963 * country (extension)*<br>
1964 * </blockquote>
1965 * depending on which fields are specified in the locale. If the
1966 * language, script, country, and variant fields are all empty,
1967 * this function returns the empty string.
1968 *
1969 * @param inLocale The locale for which to retrieve the display name.
1970 * @return The name of the locale appropriate to display.
1971 * @throws NullPointerException if <code>inLocale</code> is <code>null</code>
1972 */
1973 public String getDisplayName(Locale inLocale) {
1974 LocaleResources lr = LocaleProviderAdapter
1975 .getResourceBundleBased()
1976 .getLocaleResources(inLocale);
1977
1978 String languageName = getDisplayLanguage(inLocale);
1979 String scriptName = getDisplayScript(inLocale);
1980 String countryName = getDisplayCountry(inLocale);
1981 String[] variantNames = getDisplayVariantArray(inLocale);
1982
1983 // Get the localized patterns for formatting a display name.
1984 String displayNamePattern = lr.getLocaleName("DisplayNamePattern");
1985 String listCompositionPattern = lr.getLocaleName("ListCompositionPattern");
1986
1987 // The display name consists of a main name, followed by qualifiers.
1988 // Typically, the format is "MainName (Qualifier, Qualifier)" but this
1989 // depends on what pattern is stored in the display locale.
1990 String mainName = null;
1991 String[] qualifierNames = null;
1992
1993 // The main name is the language, or if there is no language, the script,
1994 // then if no script, the country. If there is no language/script/country
1995 // (an anomalous situation) then the display name is simply the variant's
1996 // display name.
1997 if (languageName.length() == 0 && scriptName.length() == 0 && countryName.length() == 0) {
1998 if (variantNames.length == 0) {
1999 return "";
2000 } else {
2001 return formatList(variantNames, listCompositionPattern);
2002 }
2003 }
2004 ArrayList<String> names = new ArrayList<>(4);
2005 if (languageName.length() != 0) {
2006 names.add(languageName);
2007 }
2008 if (scriptName.length() != 0) {
2009 names.add(scriptName);
2010 }
2011 if (countryName.length() != 0) {
2012 names.add(countryName);
2013 }
2014 if (variantNames.length != 0) {
2015 names.addAll(Arrays.asList(variantNames));
2016 }
2017
2018 // add Unicode extensions
2019 if (localeExtensions != null) {
2020 localeExtensions.getUnicodeLocaleAttributes().stream()
2021 .map(key -> getDisplayString(key, null, inLocale, DISPLAY_UEXT_KEY))
2022 .forEach(names::add);
2023 localeExtensions.getUnicodeLocaleKeys().stream()
2024 .map(key -> getDisplayKeyTypeExtensionString(key, lr, inLocale))
2025 .forEach(names::add);
2026 }
2027
2028 // The first one in the main name
2029 mainName = names.get(0);
2030
2031 // Others are qualifiers
2032 int numNames = names.size();
2033 qualifierNames = (numNames > 1) ?
2034 names.subList(1, numNames).toArray(new String[numNames - 1]) : new String[0];
2035
2036 // Create an array whose first element is the number of remaining
2037 // elements. This serves as a selector into a ChoiceFormat pattern from
2038 // the resource. The second and third elements are the main name and
2039 // the qualifier; if there are no qualifiers, the third element is
2040 // unused by the format pattern.
2041 Object[] displayNames = {
2042 qualifierNames.length != 0 ? 2 : 1,
2043 mainName,
2044 // We could also just call formatList() and have it handle the empty
2045 // list case, but this is more efficient, and we want it to be
2046 // efficient since all the language-only locales will not have any
2047 // qualifiers.
2048 qualifierNames.length != 0 ? formatList(qualifierNames, listCompositionPattern) : null
2049 };
2050
2051 if (displayNamePattern != null) {
2052 return new MessageFormat(displayNamePattern).format(displayNames);
2053 }
2054 else {
2055 // If we cannot get the message format pattern, then we use a simple
2056 // hard-coded pattern. This should not occur in practice unless the
2057 // installation is missing some core files (FormatData etc.).
2058 StringBuilder result = new StringBuilder();
2059 result.append((String)displayNames[1]);
2060 if (displayNames.length > 2) {
2061 result.append(" (");
2062 result.append((String)displayNames[2]);
2063 result.append(')');
2064 }
2065 return result.toString();
2066 }
2067 }
2068
2069 /**
2070 * Overrides Cloneable.
2071 */
2072 @Override
2073 public Object clone()
2074 {
2075 try {
2076 Locale that = (Locale)super.clone();
2077 return that;
2078 } catch (CloneNotSupportedException e) {
2079 throw new InternalError(e);
2080 }
2081 }
2082
2083 /**
2084 * Override hashCode.
2085 * Since Locales are often used in hashtables, caches the value
2086 * for speed.
2087 */
2088 @Override
2089 public int hashCode() {
2090 int hc = hashCodeValue;
2091 if (hc == 0) {
2092 hc = baseLocale.hashCode();
2093 if (localeExtensions != null) {
2094 hc ^= localeExtensions.hashCode();
2095 }
2096 hashCodeValue = hc;
2097 }
2098 return hc;
2099 }
2100
2101 // Overrides
2102
2103 /**
2104 * Returns true if this Locale is equal to another object. A Locale is
2105 * deemed equal to another Locale with identical language, script, country,
2106 * variant and extensions, and unequal to all other objects.
2107 *
2108 * @return true if this Locale is equal to the specified object.
2109 */
2110 @Override
2111 public boolean equals(Object obj) {
2112 if (this == obj) // quick check
2113 return true;
2114 if (!(obj instanceof Locale))
2115 return false;
2116 BaseLocale otherBase = ((Locale)obj).baseLocale;
2117 if (!baseLocale.equals(otherBase)) {
2118 return false;
2119 }
2120 if (localeExtensions == null) {
2121 return ((Locale)obj).localeExtensions == null;
2122 }
2123 return localeExtensions.equals(((Locale)obj).localeExtensions);
2124 }
2125
2126 // ================= privates =====================================
2127
2128 private transient BaseLocale baseLocale;
2129 private transient LocaleExtensions localeExtensions;
2130
2131 /**
2132 * Calculated hashcode
2133 */
2134 private transient volatile int hashCodeValue;
2135
2136 private static volatile Locale defaultLocale = initDefault();
2137 private static volatile Locale defaultDisplayLocale;
2138 private static volatile Locale defaultFormatLocale;
2139
2140 private transient volatile String languageTag;
2141
2142 /**
2143 * Return an array of the display names of the variant.
2144 * @param bundle the ResourceBundle to use to get the display names
2145 * @return an array of display names, possible of zero length.
2146 */
2147 private String[] getDisplayVariantArray(Locale inLocale) {
2148 // Split the variant name into tokens separated by '_'.
2149 StringTokenizer tokenizer = new StringTokenizer(baseLocale.getVariant(), "_");
2150 String[] names = new String[tokenizer.countTokens()];
2151
2152 // For each variant token, lookup the display name. If
2153 // not found, use the variant name itself.
2154 for (int i=0; i<names.length; ++i) {
2155 names[i] = getDisplayString(tokenizer.nextToken(), null,
2156 inLocale, DISPLAY_VARIANT);
2157 }
2158
2159 return names;
2160 }
2161
2162 private String getDisplayKeyTypeExtensionString(String key, LocaleResources lr, Locale inLocale) {
2163 String type = localeExtensions.getUnicodeLocaleType(key);
2164 String ret = getDisplayString(type, key, inLocale, DISPLAY_UEXT_TYPE);
2165
2166 if (ret == null || ret.equals(type)) {
2167 // no localization for this type. try combining key/type separately
2168 String displayType = type;
2169 switch (key) {
2170 case "cu":
2171 displayType = lr.getCurrencyName(type.toLowerCase(Locale.ROOT));
2172 break;
2173 case "rg":
2174 if (type != null &&
2175 // UN M.49 code should not be allowed here
2176 type.matches("^[a-zA-Z]{2}[zZ]{4}$")) {
2177 displayType = lr.getLocaleName(type.substring(0, 2).toUpperCase(Locale.ROOT));
2178 }
2179 break;
2180 case "tz":
2181 displayType = TimeZoneNameUtility.retrieveGenericDisplayName(
2182 TimeZoneNameUtility.convertLDMLShortID(type).orElse(type),
2183 TimeZone.LONG, inLocale);
2184 break;
2185 }
2186 ret = MessageFormat.format(lr.getLocaleName("ListKeyTypePattern"),
2187 getDisplayString(key, null, inLocale, DISPLAY_UEXT_KEY), displayType);
2188 }
2189
2190 return ret;
2191 }
2192
2193 /**
2194 * Format a list using given pattern strings.
2195 * If either of the patterns is null, then a the list is
2196 * formatted by concatenation with the delimiter ','.
2197 * @param stringList the list of strings to be formatted.
2198 * and formatting them into a list.
2199 * @param pattern should take 2 arguments for reduction
2200 * @return a string representing the list.
2201 */
2202 private static String formatList(String[] stringList, String pattern) {
2203 // If we have no list patterns, compose the list in a simple,
2204 // non-localized way.
2205 if (pattern == null) {
2206 return Arrays.stream(stringList).collect(Collectors.joining(","));
2207 }
2208
2209 switch (stringList.length) {
2210 case 0:
2211 return "";
2212 case 1:
2213 return stringList[0];
2214 default:
2215 return Arrays.stream(stringList).reduce("",
2216 (s1, s2) -> {
2217 if (s1.equals("")) {
2218 return s2;
2219 }
2220 if (s2.equals("")) {
2221 return s1;
2222 }
2223 return MessageFormat.format(pattern, s1, s2);
2224 });
2225 }
2226 }
2227
2228 // Duplicate of sun.util.locale.UnicodeLocaleExtension.isKey in order to
2229 // avoid its class loading.
2230 private static boolean isUnicodeExtensionKey(String s) {
2231 // 2alphanum
2232 return (s.length() == 2) && LocaleUtils.isAlphaNumericString(s);
2233 }
2234
2235 /**
2236 * @serialField language String
2237 * language subtag in lower case. (See <a href="java/util/Locale.html#getLanguage()">getLanguage()</a>)
2238 * @serialField country String
2239 * country subtag in upper case. (See <a href="java/util/Locale.html#getCountry()">getCountry()</a>)
2240 * @serialField variant String
2241 * variant subtags separated by LOWLINE characters. (See <a href="java/util/Locale.html#getVariant()">getVariant()</a>)
2242 * @serialField hashcode int
2243 * deprecated, for forward compatibility only
2244 * @serialField script String
2245 * script subtag in title case (See <a href="java/util/Locale.html#getScript()">getScript()</a>)
2246 * @serialField extensions String
2247 * canonical representation of extensions, that is,
2248 * BCP47 extensions in alphabetical order followed by
2249 * BCP47 private use subtags, all in lower case letters
2250 * separated by HYPHEN-MINUS characters.
2251 * (See <a href="java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>,
2252 * <a href="java/util/Locale.html#getExtension(char)">getExtension(char)</a>)
2253 */
2254 private static final ObjectStreamField[] serialPersistentFields = {
2255 new ObjectStreamField("language", String.class),
2256 new ObjectStreamField("country", String.class),
2257 new ObjectStreamField("variant", String.class),
2258 new ObjectStreamField("hashcode", int.class),
2259 new ObjectStreamField("script", String.class),
2260 new ObjectStreamField("extensions", String.class),
2261 };
2262
2263 /**
2264 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>.
2265 * @param out the <code>ObjectOutputStream</code> to write
2266 * @throws IOException
2267 * @since 1.7
2268 */
2269 private void writeObject(ObjectOutputStream out) throws IOException {
2270 ObjectOutputStream.PutField fields = out.putFields();
2271 fields.put("language", baseLocale.getLanguage());
2272 fields.put("script", baseLocale.getScript());
2273 fields.put("country", baseLocale.getRegion());
2274 fields.put("variant", baseLocale.getVariant());
2275 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID());
2276 fields.put("hashcode", -1); // place holder just for backward support
2277 out.writeFields();
2278 }
2279
2280 /**
2281 * Deserializes this <code>Locale</code>.
2282 * @param in the <code>ObjectInputStream</code> to read
2283 * @throws IOException
2284 * @throws ClassNotFoundException
2285 * @throws IllformedLocaleException
2286 * @since 1.7
2287 */
2288 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException {
2289 ObjectInputStream.GetField fields = in.readFields();
2290 String language = (String)fields.get("language", "");
2291 String script = (String)fields.get("script", "");
2292 String country = (String)fields.get("country", "");
2293 String variant = (String)fields.get("variant", "");
2294 String extStr = (String)fields.get("extensions", "");
2295 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant);
2296 if (extStr.length() > 0) {
2297 try {
2298 InternalLocaleBuilder bldr = new InternalLocaleBuilder();
2299 bldr.setExtensions(extStr);
2300 localeExtensions = bldr.getLocaleExtensions();
2301 } catch (LocaleSyntaxException e) {
2302 throw new IllformedLocaleException(e.getMessage());
2303 }
2304 } else {
2305 localeExtensions = null;
2306 }
2307 }
2308
2309 /**
2310 * Returns a cached <code>Locale</code> instance equivalent to
2311 * the deserialized <code>Locale</code>. When serialized
2312 * language, country and variant fields read from the object data stream
2313 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions
2314 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code>
2315 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script
2316 * type is "thai"). See <a href="Locale.html#special_cases_constructor">Special Cases</a>
2317 * for more information.
2318 *
2319 * @return an instance of <code>Locale</code> equivalent to
2320 * the deserialized <code>Locale</code>.
2321 * @throws java.io.ObjectStreamException
2322 */
2323 private Object readResolve() throws java.io.ObjectStreamException {
2324 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(),
2325 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions);
2326 }
2327
2328 private static volatile String[] isoLanguages;
2329
2330 private static volatile String[] isoCountries;
2331
2332 private static String convertOldISOCodes(String language) {
2333 // we accept both the old and the new ISO codes for the languages whose ISO
2334 // codes have changed, but we always store the OLD code, for backward compatibility
2335 language = LocaleUtils.toLowerString(language).intern();
2336 if (language == "he") {
2337 return "iw";
2338 } else if (language == "yi") {
2339 return "ji";
2340 } else if (language == "id") {
2341 return "in";
2342 } else {
2343 return language;
2344 }
2345 }
2346
2347 private static LocaleExtensions getCompatibilityExtensions(String language,
2348 String script,
2349 String country,
2350 String variant) {
2351 LocaleExtensions extensions = null;
2352 // Special cases for backward compatibility support
2353 if (LocaleUtils.caseIgnoreMatch(language, "ja")
2354 && script.length() == 0
2355 && LocaleUtils.caseIgnoreMatch(country, "jp")
2356 && "JP".equals(variant)) {
2357 // ja_JP_JP -> u-ca-japanese (calendar = japanese)
2358 extensions = LocaleExtensions.CALENDAR_JAPANESE;
2359 } else if (LocaleUtils.caseIgnoreMatch(language, "th")
2360 && script.length() == 0
2361 && LocaleUtils.caseIgnoreMatch(country, "th")
2362 && "TH".equals(variant)) {
2363 // th_TH_TH -> u-nu-thai (numbersystem = thai)
2364 extensions = LocaleExtensions.NUMBER_THAI;
2365 }
2366 return extensions;
2367 }
2368
2369 /**
2370 * Obtains a localized locale names from a LocaleNameProvider
2371 * implementation.
2372 */
2373 private static class LocaleNameGetter
2374 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> {
2375 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter();
2376
2377 @Override
2378 public String getObject(LocaleNameProvider localeNameProvider,
2379 Locale locale,
2380 String key,
2381 Object... params) {
2382 assert params.length == 3;
2383 int type = (Integer)params[0];
2384 String code = (String)params[1];
2385 String cat = (String)params[2];
2386
2387 switch(type) {
2388 case DISPLAY_LANGUAGE:
2389 return localeNameProvider.getDisplayLanguage(code, locale);
2390 case DISPLAY_COUNTRY:
2391 return localeNameProvider.getDisplayCountry(code, locale);
2392 case DISPLAY_VARIANT:
2393 return localeNameProvider.getDisplayVariant(code, locale);
2394 case DISPLAY_SCRIPT:
2395 return localeNameProvider.getDisplayScript(code, locale);
2396 case DISPLAY_UEXT_KEY:
2397 return localeNameProvider.getDisplayUnicodeExtensionKey(code, locale);
2398 case DISPLAY_UEXT_TYPE:
2399 return localeNameProvider.getDisplayUnicodeExtensionType(code, cat, locale);
2400 default:
2401 assert false; // shouldn't happen
2402 }
2403
2404 return null;
2405 }
2406 }
2407
2408 /**
2409 * Enum for locale categories. These locale categories are used to get/set
2410 * the default locale for the specific functionality represented by the
2411 * category.
2412 *
2413 * @see #getDefault(Locale.Category)
2414 * @see #setDefault(Locale.Category, Locale)
2415 * @since 1.7
2416 */
2417 public enum Category {
2418
2419 /**
2420 * Category used to represent the default locale for
2421 * displaying user interfaces.
2422 */
2423 DISPLAY("user.language.display",
2424 "user.script.display",
2425 "user.country.display",
2426 "user.variant.display",
2427 "user.extensions.display"),
2428
2429 /**
2430 * Category used to represent the default locale for
2431 * formatting dates, numbers, and/or currencies.
2432 */
2433 FORMAT("user.language.format",
2434 "user.script.format",
2435 "user.country.format",
2436 "user.variant.format",
2437 "user.extensions.format");
2438
2439 Category(String languageKey, String scriptKey, String countryKey,
2440 String variantKey, String extensionsKey) {
2441 this.languageKey = languageKey;
2442 this.scriptKey = scriptKey;
2443 this.countryKey = countryKey;
2444 this.variantKey = variantKey;
2445 this.extensionsKey = extensionsKey;
2446 }
2447
2448 final String languageKey;
2449 final String scriptKey;
2450 final String countryKey;
2451 final String variantKey;
2452 final String extensionsKey;
2453 }
2454
2455 /**
2456 * <code>Builder</code> is used to build instances of <code>Locale</code>
2457 * from values configured by the setters. Unlike the <code>Locale</code>
2458 * constructors, the <code>Builder</code> checks if a value configured by a
2459 * setter satisfies the syntax requirements defined by the <code>Locale</code>
2460 * class. A <code>Locale</code> object created by a <code>Builder</code> is
2461 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag
2462 * without losing information.
2463 *
2464 * <p><b>Note:</b> The <code>Locale</code> class does not provide any
2465 * syntactic restrictions on variant, while BCP 47 requires each variant
2466 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3
2467 * alphanumerics. The method <code>setVariant</code> throws
2468 * <code>IllformedLocaleException</code> for a variant that does not satisfy
2469 * this restriction. If it is necessary to support such a variant, use a
2470 * Locale constructor. However, keep in mind that a <code>Locale</code>
2471 * object created this way might lose the variant information when
2472 * transformed to a BCP 47 language tag.
2473 *
2474 * <p>The following example shows how to create a <code>Locale</code> object
2475 * with the <code>Builder</code>.
2476 * <blockquote>
2477 * <pre>
2478 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build();
2479 * </pre>
2480 * </blockquote>
2481 *
2482 * <p>Builders can be reused; <code>clear()</code> resets all
2483 * fields to their default values.
2484 *
2485 * @see Locale#forLanguageTag
2486 * @since 1.7
2487 */
2488 public static final class Builder {
2489 private final InternalLocaleBuilder localeBuilder;
2490
2491 /**
2492 * Constructs an empty Builder. The default value of all
2493 * fields, extensions, and private use information is the
2494 * empty string.
2495 */
2496 public Builder() {
2497 localeBuilder = new InternalLocaleBuilder();
2498 }
2499
2500 /**
2501 * Resets the <code>Builder</code> to match the provided
2502 * <code>locale</code>. Existing state is discarded.
2503 *
2504 * <p>All fields of the locale must be well-formed, see {@link Locale}.
2505 *
2506 * <p>Locales with any ill-formed fields cause
2507 * <code>IllformedLocaleException</code> to be thrown, except for the
2508 * following three cases which are accepted for compatibility
2509 * reasons:<ul>
2510 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese"
2511 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai"
2512 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul>
2513 *
2514 * @param locale the locale
2515 * @return This builder.
2516 * @throws IllformedLocaleException if <code>locale</code> has
2517 * any ill-formed fields.
2518 * @throws NullPointerException if <code>locale</code> is null.
2519 */
2520 public Builder setLocale(Locale locale) {
2521 try {
2522 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions);
2523 } catch (LocaleSyntaxException e) {
2524 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2525 }
2526 return this;
2527 }
2528
2529 /**
2530 * Resets the Builder to match the provided IETF BCP 47
2531 * language tag. Discards the existing state. Null and the
2532 * empty string cause the builder to be reset, like {@link
2533 * #clear}. Grandfathered tags (see {@link
2534 * Locale#forLanguageTag}) are converted to their canonical
2535 * form before being processed. Otherwise, the language tag
2536 * must be well-formed (see {@link Locale}) or an exception is
2537 * thrown (unlike <code>Locale.forLanguageTag</code>, which
2538 * just discards ill-formed and following portions of the
2539 * tag).
2540 *
2541 * @param languageTag the language tag
2542 * @return This builder.
2543 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed
2544 * @see Locale#forLanguageTag(String)
2545 */
2546 public Builder setLanguageTag(String languageTag) {
2547 ParseStatus sts = new ParseStatus();
2548 LanguageTag tag = LanguageTag.parse(languageTag, sts);
2549 if (sts.isError()) {
2550 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex());
2551 }
2552 localeBuilder.setLanguageTag(tag);
2553 return this;
2554 }
2555
2556 /**
2557 * Sets the language. If <code>language</code> is the empty string or
2558 * null, the language in this <code>Builder</code> is removed. Otherwise,
2559 * the language must be <a href="./Locale.html#def_language">well-formed</a>
2560 * or an exception is thrown.
2561 *
2562 * <p>The typical language value is a two or three-letter language
2563 * code as defined in ISO639.
2564 *
2565 * @param language the language
2566 * @return This builder.
2567 * @throws IllformedLocaleException if <code>language</code> is ill-formed
2568 */
2569 public Builder setLanguage(String language) {
2570 try {
2571 localeBuilder.setLanguage(language);
2572 } catch (LocaleSyntaxException e) {
2573 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2574 }
2575 return this;
2576 }
2577
2578 /**
2579 * Sets the script. If <code>script</code> is null or the empty string,
2580 * the script in this <code>Builder</code> is removed.
2581 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an
2582 * exception is thrown.
2583 *
2584 * <p>The typical script value is a four-letter script code as defined by ISO 15924.
2585 *
2586 * @param script the script
2587 * @return This builder.
2588 * @throws IllformedLocaleException if <code>script</code> is ill-formed
2589 */
2590 public Builder setScript(String script) {
2591 try {
2592 localeBuilder.setScript(script);
2593 } catch (LocaleSyntaxException e) {
2594 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2595 }
2596 return this;
2597 }
2598
2599 /**
2600 * Sets the region. If region is null or the empty string, the region
2601 * in this <code>Builder</code> is removed. Otherwise,
2602 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an
2603 * exception is thrown.
2604 *
2605 * <p>The typical region value is a two-letter ISO 3166 code or a
2606 * three-digit UN M.49 area code.
2607 *
2608 * <p>The country value in the <code>Locale</code> created by the
2609 * <code>Builder</code> is always normalized to upper case.
2610 *
2611 * @param region the region
2612 * @return This builder.
2613 * @throws IllformedLocaleException if <code>region</code> is ill-formed
2614 */
2615 public Builder setRegion(String region) {
2616 try {
2617 localeBuilder.setRegion(region);
2618 } catch (LocaleSyntaxException e) {
2619 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2620 }
2621 return this;
2622 }
2623
2624 /**
2625 * Sets the variant. If variant is null or the empty string, the
2626 * variant in this <code>Builder</code> is removed. Otherwise, it
2627 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a>
2628 * subtags, or an exception is thrown.
2629 *
2630 * <p><b>Note:</b> This method checks if <code>variant</code>
2631 * satisfies the IETF BCP 47 variant subtag's syntax requirements,
2632 * and normalizes the value to lowercase letters. However,
2633 * the <code>Locale</code> class does not impose any syntactic
2634 * restriction on variant, and the variant value in
2635 * <code>Locale</code> is case sensitive. To set such a variant,
2636 * use a Locale constructor.
2637 *
2638 * @param variant the variant
2639 * @return This builder.
2640 * @throws IllformedLocaleException if <code>variant</code> is ill-formed
2641 */
2642 public Builder setVariant(String variant) {
2643 try {
2644 localeBuilder.setVariant(variant);
2645 } catch (LocaleSyntaxException e) {
2646 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2647 }
2648 return this;
2649 }
2650
2651 /**
2652 * Sets the extension for the given key. If the value is null or the
2653 * empty string, the extension is removed. Otherwise, the extension
2654 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception
2655 * is thrown.
2656 *
2657 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION
2658 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension.
2659 * Setting a value for this key replaces any existing Unicode locale key/type
2660 * pairs with those defined in the extension.
2661 *
2662 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION
2663 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be
2664 * well-formed, the value for this key needs only to have subtags of one to
2665 * eight alphanumeric characters, not two to eight as in the general case.
2666 *
2667 * @param key the extension key
2668 * @param value the extension value
2669 * @return This builder.
2670 * @throws IllformedLocaleException if <code>key</code> is illegal
2671 * or <code>value</code> is ill-formed
2672 * @see #setUnicodeLocaleKeyword(String, String)
2673 */
2674 public Builder setExtension(char key, String value) {
2675 try {
2676 localeBuilder.setExtension(key, value);
2677 } catch (LocaleSyntaxException e) {
2678 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2679 }
2680 return this;
2681 }
2682
2683 /**
2684 * Sets the Unicode locale keyword type for the given key. If the type
2685 * is null, the Unicode keyword is removed. Otherwise, the key must be
2686 * non-null and both key and type must be <a
2687 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2688 * is thrown.
2689 *
2690 * <p>Keys and types are converted to lower case.
2691 *
2692 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension}
2693 * replaces all Unicode locale keywords with those defined in the
2694 * extension.
2695 *
2696 * @param key the Unicode locale key
2697 * @param type the Unicode locale type
2698 * @return This builder.
2699 * @throws IllformedLocaleException if <code>key</code> or <code>type</code>
2700 * is ill-formed
2701 * @throws NullPointerException if <code>key</code> is null
2702 * @see #setExtension(char, String)
2703 */
2704 public Builder setUnicodeLocaleKeyword(String key, String type) {
2705 try {
2706 localeBuilder.setUnicodeLocaleKeyword(key, type);
2707 } catch (LocaleSyntaxException e) {
2708 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2709 }
2710 return this;
2711 }
2712
2713 /**
2714 * Adds a unicode locale attribute, if not already present, otherwise
2715 * has no effect. The attribute must not be null and must be <a
2716 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2717 * is thrown.
2718 *
2719 * @param attribute the attribute
2720 * @return This builder.
2721 * @throws NullPointerException if <code>attribute</code> is null
2722 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2723 * @see #setExtension(char, String)
2724 */
2725 public Builder addUnicodeLocaleAttribute(String attribute) {
2726 try {
2727 localeBuilder.addUnicodeLocaleAttribute(attribute);
2728 } catch (LocaleSyntaxException e) {
2729 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2730 }
2731 return this;
2732 }
2733
2734 /**
2735 * Removes a unicode locale attribute, if present, otherwise has no
2736 * effect. The attribute must not be null and must be <a
2737 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception
2738 * is thrown.
2739 *
2740 * <p>Attribute comparison for removal is case-insensitive.
2741 *
2742 * @param attribute the attribute
2743 * @return This builder.
2744 * @throws NullPointerException if <code>attribute</code> is null
2745 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed
2746 * @see #setExtension(char, String)
2747 */
2748 public Builder removeUnicodeLocaleAttribute(String attribute) {
2749 Objects.requireNonNull(attribute);
2750 try {
2751 localeBuilder.removeUnicodeLocaleAttribute(attribute);
2752 } catch (LocaleSyntaxException e) {
2753 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex());
2754 }
2755 return this;
2756 }
2757
2758 /**
2759 * Resets the builder to its initial, empty state.
2760 *
2761 * @return This builder.
2762 */
2763 public Builder clear() {
2764 localeBuilder.clear();
2765 return this;
2766 }
2767
2768 /**
2769 * Resets the extensions to their initial, empty state.
2770 * Language, script, region and variant are unchanged.
2771 *
2772 * @return This builder.
2773 * @see #setExtension(char, String)
2774 */
2775 public Builder clearExtensions() {
2776 localeBuilder.clearExtensions();
2777 return this;
2778 }
2779
2780 /**
2781 * Returns an instance of <code>Locale</code> created from the fields set
2782 * on this builder.
2783 *
2784 * <p>This applies the conversions listed in {@link Locale#forLanguageTag}
2785 * when constructing a Locale. (Grandfathered tags are handled in
2786 * {@link #setLanguageTag}.)
2787 *
2788 * @return A Locale.
2789 */
2790 public Locale build() {
2791 BaseLocale baseloc = localeBuilder.getBaseLocale();
2792 LocaleExtensions extensions = localeBuilder.getLocaleExtensions();
2793 if (extensions == null && baseloc.getVariant().length() > 0) {
2794 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(),
2795 baseloc.getRegion(), baseloc.getVariant());
2796 }
2797 return Locale.getInstance(baseloc, extensions);
2798 }
2799 }
2800
2801 /**
2802 * This enum provides constants to select a filtering mode for locale
2803 * matching. Refer to <a href="http://tools.ietf.org/html/rfc4647">RFC 4647
2804 * Matching of Language Tags</a> for details.
2805 *
2806 * <p>As an example, think of two Language Priority Lists each of which
2807 * includes only one language range and a set of following language tags:
2808 *
2809 * <pre>
2810 * de (German)
2811 * de-DE (German, Germany)
2812 * de-Deva (German, in Devanagari script)
2813 * de-Deva-DE (German, in Devanagari script, Germany)
2814 * de-DE-1996 (German, Germany, orthography of 1996)
2815 * de-Latn-DE (German, in Latin script, Germany)
2816 * de-Latn-DE-1996 (German, in Latin script, Germany, orthography of 1996)
2817 * </pre>
2818 *
2819 * The filtering method will behave as follows:
2820 *
2821 * <table class="striped">
2822 * <caption>Filtering method behavior</caption>
2823 * <thead>
2824 * <tr>
2825 * <th scope="col">Filtering Mode</th>
2826 * <th scope="col">Language Priority List: {@code "de-DE"}</th>
2827 * <th scope="col">Language Priority List: {@code "de-*-DE"}</th>
2828 * </tr>
2829 * </thead>
2830 * <tbody>
2831 * <tr>
2832 * <th scope="row" style="vertical-align:top">
2833 * {@link FilteringMode#AUTOSELECT_FILTERING AUTOSELECT_FILTERING}
2834 * </th>
2835 * <td style="vertical-align:top">
2836 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2837 * {@code "de-DE-1996"}.
2838 * </td>
2839 * <td style="vertical-align:top">
2840 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2841 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2842 * {@code "de-Latn-DE-1996"}.
2843 * </td>
2844 * </tr>
2845 * <tr>
2846 * <th scope="row" style="vertical-align:top">
2847 * {@link FilteringMode#EXTENDED_FILTERING EXTENDED_FILTERING}
2848 * </th>
2849 * <td style="vertical-align:top">
2850 * Performs <em>extended</em> filtering and returns {@code "de-DE"},
2851 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and
2852 * {@code "de-Latn-DE-1996"}.
2853 * </td>
2854 * <td style="vertical-align:top">Same as above.</td>
2855 * </tr>
2856 * <tr>
2857 * <th scope="row" style="vertical-align:top">
2858 * {@link FilteringMode#IGNORE_EXTENDED_RANGES IGNORE_EXTENDED_RANGES}
2859 * </th>
2860 * <td style="vertical-align:top">
2861 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2862 * {@code "de-DE-1996"}.
2863 * </td>
2864 * <td style="vertical-align:top">
2865 * Performs <em>basic</em> filtering and returns {@code null} because
2866 * nothing matches.
2867 * </td>
2868 * </tr>
2869 * <tr>
2870 * <th scope="row" style="vertical-align:top">
2871 * {@link FilteringMode#MAP_EXTENDED_RANGES MAP_EXTENDED_RANGES}
2872 * </th>
2873 * <td style="vertical-align:top">Same as above.</td>
2874 * <td style="vertical-align:top">
2875 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and
2876 * {@code "de-DE-1996"} because {@code "de-*-DE"} is mapped to
2877 * {@code "de-DE"}.
2878 * </td>
2879 * </tr>
2880 * <tr>
2881 * <th scope="row" style="vertical-align:top">
2882 * {@link FilteringMode#REJECT_EXTENDED_RANGES REJECT_EXTENDED_RANGES}
2883 * </th>
2884 * <td style="vertical-align:top">Same as above.</td>
2885 * <td style="vertical-align:top">
2886 * Throws {@link IllegalArgumentException} because {@code "de-*-DE"} is
2887 * not a valid basic language range.
2888 * </td>
2889 * </tr>
2890 * </tbody>
2891 * </table>
2892 *
2893 * @see #filter(List, Collection, FilteringMode)
2894 * @see #filterTags(List, Collection, FilteringMode)
2895 *
2896 * @since 1.8
2897 */
2898 public static enum FilteringMode {
2899 /**
2900 * Specifies automatic filtering mode based on the given Language
2901 * Priority List consisting of language ranges. If all of the ranges
2902 * are basic, basic filtering is selected. Otherwise, extended
2903 * filtering is selected.
2904 */
2905 AUTOSELECT_FILTERING,
2906
2907 /**
2908 * Specifies extended filtering.
2909 */
2910 EXTENDED_FILTERING,
2911
2912 /**
2913 * Specifies basic filtering: Note that any extended language ranges
2914 * included in the given Language Priority List are ignored.
2915 */
2916 IGNORE_EXTENDED_RANGES,
2917
2918 /**
2919 * Specifies basic filtering: If any extended language ranges are
2920 * included in the given Language Priority List, they are mapped to the
2921 * basic language range. Specifically, a language range starting with a
2922 * subtag {@code "*"} is treated as a language range {@code "*"}. For
2923 * example, {@code "*-US"} is treated as {@code "*"}. If {@code "*"} is
2924 * not the first subtag, {@code "*"} and extra {@code "-"} are removed.
2925 * For example, {@code "ja-*-JP"} is mapped to {@code "ja-JP"}.
2926 */
2927 MAP_EXTENDED_RANGES,
2928
2929 /**
2930 * Specifies basic filtering: If any extended language ranges are
2931 * included in the given Language Priority List, the list is rejected
2932 * and the filtering method throws {@link IllegalArgumentException}.
2933 */
2934 REJECT_EXTENDED_RANGES
2935 };
2936
2937 /**
2938 * This class expresses a <em>Language Range</em> defined in
2939 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of
2940 * Language Tags</a>. A language range is an identifier which is used to
2941 * select language tag(s) meeting specific requirements by using the
2942 * mechanisms described in <a href="Locale.html#LocaleMatching">Locale
2943 * Matching</a>. A list which represents a user's preferences and consists
2944 * of language ranges is called a <em>Language Priority List</em>.
2945 *
2946 * <p>There are two types of language ranges: basic and extended. In RFC
2947 * 4647, the syntax of language ranges is expressed in
2948 * <a href="http://tools.ietf.org/html/rfc4234">ABNF</a> as follows:
2949 * <blockquote>
2950 * <pre>
2951 * basic-language-range = (1*8ALPHA *("-" 1*8alphanum)) / "*"
2952 * extended-language-range = (1*8ALPHA / "*")
2953 * *("-" (1*8alphanum / "*"))
2954 * alphanum = ALPHA / DIGIT
2955 * </pre>
2956 * </blockquote>
2957 * For example, {@code "en"} (English), {@code "ja-JP"} (Japanese, Japan),
2958 * {@code "*"} (special language range which matches any language tag) are
2959 * basic language ranges, whereas {@code "*-CH"} (any languages,
2960 * Switzerland), {@code "es-*"} (Spanish, any regions), and
2961 * {@code "zh-Hant-*"} (Traditional Chinese, any regions) are extended
2962 * language ranges.
2963 *
2964 * @see #filter
2965 * @see #filterTags
2966 * @see #lookup
2967 * @see #lookupTag
2968 *
2969 * @since 1.8
2970 */
2971 public static final class LanguageRange {
2972
2973 /**
2974 * A constant holding the maximum value of weight, 1.0, which indicates
2975 * that the language range is a good fit for the user.
2976 */
2977 public static final double MAX_WEIGHT = 1.0;
2978
2979 /**
2980 * A constant holding the minimum value of weight, 0.0, which indicates
2981 * that the language range is not a good fit for the user.
2982 */
2983 public static final double MIN_WEIGHT = 0.0;
2984
2985 private final String range;
2986 private final double weight;
2987
2988 private volatile int hash;
2989
2990 /**
2991 * Constructs a {@code LanguageRange} using the given {@code range}.
2992 * Note that no validation is done against the IANA Language Subtag
2993 * Registry at time of construction.
2994 *
2995 * <p>This is equivalent to {@code LanguageRange(range, MAX_WEIGHT)}.
2996 *
2997 * @param range a language range
2998 * @throws NullPointerException if the given {@code range} is
2999 * {@code null}
3000 * @throws IllegalArgumentException if the given {@code range} does not
3001 * comply with the syntax of the language range mentioned in RFC 4647
3002 */
3003 public LanguageRange(String range) {
3004 this(range, MAX_WEIGHT);
3005 }
3006
3007 /**
3008 * Constructs a {@code LanguageRange} using the given {@code range} and
3009 * {@code weight}. Note that no validation is done against the IANA
3010 * Language Subtag Registry at time of construction.
3011 *
3012 * @param range a language range
3013 * @param weight a weight value between {@code MIN_WEIGHT} and
3014 * {@code MAX_WEIGHT}
3015 * @throws NullPointerException if the given {@code range} is
3016 * {@code null}
3017 * @throws IllegalArgumentException if the given {@code range} does not
3018 * comply with the syntax of the language range mentioned in RFC 4647
3019 * or if the given {@code weight} is less than {@code MIN_WEIGHT}
3020 * or greater than {@code MAX_WEIGHT}
3021 */
3022 public LanguageRange(String range, double weight) {
3023 if (range == null) {
3024 throw new NullPointerException();
3025 }
3026 if (weight < MIN_WEIGHT || weight > MAX_WEIGHT) {
3027 throw new IllegalArgumentException("weight=" + weight);
3028 }
3029
3030 range = range.toLowerCase(Locale.ROOT);
3031
3032 // Do syntax check.
3033 boolean isIllFormed = false;
3034 String[] subtags = range.split("-");
3035 if (isSubtagIllFormed(subtags[0], true)
3036 || range.endsWith("-")) {
3037 isIllFormed = true;
3038 } else {
3039 for (int i = 1; i < subtags.length; i++) {
3040 if (isSubtagIllFormed(subtags[i], false)) {
3041 isIllFormed = true;
3042 break;
3043 }
3044 }
3045 }
3046 if (isIllFormed) {
3047 throw new IllegalArgumentException("range=" + range);
3048 }
3049
3050 this.range = range;
3051 this.weight = weight;
3052 }
3053
3054 private static boolean isSubtagIllFormed(String subtag,
3055 boolean isFirstSubtag) {
3056 if (subtag.equals("") || subtag.length() > 8) {
3057 return true;
3058 } else if (subtag.equals("*")) {
3059 return false;
3060 }
3061 char[] charArray = subtag.toCharArray();
3062 if (isFirstSubtag) { // ALPHA
3063 for (char c : charArray) {
3064 if (c < 'a' || c > 'z') {
3065 return true;
3066 }
3067 }
3068 } else { // ALPHA / DIGIT
3069 for (char c : charArray) {
3070 if (c < '0' || (c > '9' && c < 'a') || c > 'z') {
3071 return true;
3072 }
3073 }
3074 }
3075 return false;
3076 }
3077
3078 /**
3079 * Returns the language range of this {@code LanguageRange}.
3080 *
3081 * @return the language range.
3082 */
3083 public String getRange() {
3084 return range;
3085 }
3086
3087 /**
3088 * Returns the weight of this {@code LanguageRange}.
3089 *
3090 * @return the weight value.
3091 */
3092 public double getWeight() {
3093 return weight;
3094 }
3095
3096 /**
3097 * Parses the given {@code ranges} to generate a Language Priority List.
3098 *
3099 * <p>This method performs a syntactic check for each language range in
3100 * the given {@code ranges} but doesn't do validation using the IANA
3101 * Language Subtag Registry.
3102 *
3103 * <p>The {@code ranges} to be given can take one of the following
3104 * forms:
3105 *
3106 * <pre>
3107 * "Accept-Language: ja,en;q=0.4" (weighted list with Accept-Language prefix)
3108 * "ja,en;q=0.4" (weighted list)
3109 * "ja,en" (prioritized list)
3110 * </pre>
3111 *
3112 * In a weighted list, each language range is given a weight value.
3113 * The weight value is identical to the "quality value" in
3114 * <a href="http://tools.ietf.org/html/rfc2616">RFC 2616</a>, and it
3115 * expresses how much the user prefers the language. A weight value is
3116 * specified after a corresponding language range followed by
3117 * {@code ";q="}, and the default weight value is {@code MAX_WEIGHT}
3118 * when it is omitted.
3119 *
3120 * <p>Unlike a weighted list, language ranges in a prioritized list
3121 * are sorted in the descending order based on its priority. The first
3122 * language range has the highest priority and meets the user's
3123 * preference most.
3124 *
3125 * <p>In either case, language ranges are sorted in descending order in
3126 * the Language Priority List based on priority or weight. If a
3127 * language range appears in the given {@code ranges} more than once,
3128 * only the first one is included on the Language Priority List.
3129 *
3130 * <p>The returned list consists of language ranges from the given
3131 * {@code ranges} and their equivalents found in the IANA Language
3132 * Subtag Registry. For example, if the given {@code ranges} is
3133 * {@code "Accept-Language: iw,en-us;q=0.7,en;q=0.3"}, the elements in
3134 * the list to be returned are:
3135 *
3136 * <pre>
3137 * <b>Range</b> <b>Weight</b>
3138 * "iw" (older tag for Hebrew) 1.0
3139 * "he" (new preferred code for Hebrew) 1.0
3140 * "en-us" (English, United States) 0.7
3141 * "en" (English) 0.3
3142 * </pre>
3143 *
3144 * Two language ranges, {@code "iw"} and {@code "he"}, have the same
3145 * highest priority in the list. By adding {@code "he"} to the user's
3146 * Language Priority List, locale-matching method can find Hebrew as a
3147 * matching locale (or language tag) even if the application or system
3148 * offers only {@code "he"} as a supported locale (or language tag).
3149 *
3150 * @param ranges a list of comma-separated language ranges or a list of
3151 * language ranges in the form of the "Accept-Language" header
3152 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3153 * 2616</a>
3154 * @return a Language Priority List consisting of language ranges
3155 * included in the given {@code ranges} and their equivalent
3156 * language ranges if available. The list is modifiable.
3157 * @throws NullPointerException if {@code ranges} is null
3158 * @throws IllegalArgumentException if a language range or a weight
3159 * found in the given {@code ranges} is ill-formed
3160 */
3161 public static List<LanguageRange> parse(String ranges) {
3162 return LocaleMatcher.parse(ranges);
3163 }
3164
3165 /**
3166 * Parses the given {@code ranges} to generate a Language Priority
3167 * List, and then customizes the list using the given {@code map}.
3168 * This method is equivalent to
3169 * {@code mapEquivalents(parse(ranges), map)}.
3170 *
3171 * @param ranges a list of comma-separated language ranges or a list
3172 * of language ranges in the form of the "Accept-Language" header
3173 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC
3174 * 2616</a>
3175 * @param map a map containing information to customize language ranges
3176 * @return a Language Priority List with customization. The list is
3177 * modifiable.
3178 * @throws NullPointerException if {@code ranges} is null
3179 * @throws IllegalArgumentException if a language range or a weight
3180 * found in the given {@code ranges} is ill-formed
3181 * @see #parse(String)
3182 * @see #mapEquivalents
3183 */
3184 public static List<LanguageRange> parse(String ranges,
3185 Map<String, List<String>> map) {
3186 return mapEquivalents(parse(ranges), map);
3187 }
3188
3189 /**
3190 * Generates a new customized Language Priority List using the given
3191 * {@code priorityList} and {@code map}. If the given {@code map} is
3192 * empty, this method returns a copy of the given {@code priorityList}.
3193 *
3194 * <p>In the map, a key represents a language range whereas a value is
3195 * a list of equivalents of it. {@code '*'} cannot be used in the map.
3196 * Each equivalent language range has the same weight value as its
3197 * original language range.
3198 *
3199 * <pre>
3200 * An example of map:
3201 * <b>Key</b> <b>Value</b>
3202 * "zh" (Chinese) "zh",
3203 * "zh-Hans"(Simplified Chinese)
3204 * "zh-HK" (Chinese, Hong Kong) "zh-HK"
3205 * "zh-TW" (Chinese, Taiwan) "zh-TW"
3206 * </pre>
3207 *
3208 * The customization is performed after modification using the IANA
3209 * Language Subtag Registry.
3210 *
3211 * <p>For example, if a user's Language Priority List consists of five
3212 * language ranges ({@code "zh"}, {@code "zh-CN"}, {@code "en"},
3213 * {@code "zh-TW"}, and {@code "zh-HK"}), the newly generated Language
3214 * Priority List which is customized using the above map example will
3215 * consists of {@code "zh"}, {@code "zh-Hans"}, {@code "zh-CN"},
3216 * {@code "zh-Hans-CN"}, {@code "en"}, {@code "zh-TW"}, and
3217 * {@code "zh-HK"}.
3218 *
3219 * <p>{@code "zh-HK"} and {@code "zh-TW"} aren't converted to
3220 * {@code "zh-Hans-HK"} nor {@code "zh-Hans-TW"} even if they are
3221 * included in the Language Priority List. In this example, mapping
3222 * is used to clearly distinguish Simplified Chinese and Traditional
3223 * Chinese.
3224 *
3225 * <p>If the {@code "zh"}-to-{@code "zh"} mapping isn't included in the
3226 * map, a simple replacement will be performed and the customized list
3227 * won't include {@code "zh"} and {@code "zh-CN"}.
3228 *
3229 * @param priorityList user's Language Priority List
3230 * @param map a map containing information to customize language ranges
3231 * @return a new Language Priority List with customization. The list is
3232 * modifiable.
3233 * @throws NullPointerException if {@code priorityList} is {@code null}
3234 * @see #parse(String, Map)
3235 */
3236 public static List<LanguageRange> mapEquivalents(
3237 List<LanguageRange>priorityList,
3238 Map<String, List<String>> map) {
3239 return LocaleMatcher.mapEquivalents(priorityList, map);
3240 }
3241
3242 /**
3243 * Returns a hash code value for the object.
3244 *
3245 * @return a hash code value for this object.
3246 */
3247 @Override
3248 public int hashCode() {
3249 int h = hash;
3250 if (h == 0) {
3251 h = 17;
3252 h = 37*h + range.hashCode();
3253 long bitsWeight = Double.doubleToLongBits(weight);
3254 h = 37*h + (int)(bitsWeight ^ (bitsWeight >>> 32));
3255 if (h != 0) {
3256 hash = h;
3257 }
3258 }
3259 return h;
3260 }
3261
3262 /**
3263 * Compares this object to the specified object. The result is true if
3264 * and only if the argument is not {@code null} and is a
3265 * {@code LanguageRange} object that contains the same {@code range}
3266 * and {@code weight} values as this object.
3267 *
3268 * @param obj the object to compare with
3269 * @return {@code true} if this object's {@code range} and
3270 * {@code weight} are the same as the {@code obj}'s; {@code false}
3271 * otherwise.
3272 */
3273 @Override
3274 public boolean equals(Object obj) {
3275 if (this == obj) {
3276 return true;
3277 }
3278 if (!(obj instanceof LanguageRange)) {
3279 return false;
3280 }
3281 LanguageRange other = (LanguageRange)obj;
3282 return hash == other.hash
3283 && range.equals(other.range)
3284 && weight == other.weight;
3285 }
3286
3287 /**
3288 * Returns an informative string representation of this {@code LanguageRange}
3289 * object, consisting of language range and weight if the range is
3290 * weighted and the weight is less than the max weight.
3291 *
3292 * @return a string representation of this {@code LanguageRange} object.
3293 */
3294 @Override
3295 public String toString() {
3296 return (weight == MAX_WEIGHT) ? range : range + ";q=" + weight;
3297 }
3298 }
3299
3300 /**
3301 * Returns a list of matching {@code Locale} instances using the filtering
3302 * mechanism defined in RFC 4647.
3303 *
3304 * This filter operation on the given {@code locales} ensures that only
3305 * unique matching locale(s) are returned.
3306 *
3307 * @param priorityList user's Language Priority List in which each language
3308 * tag is sorted in descending order based on priority or weight
3309 * @param locales {@code Locale} instances used for matching
3310 * @param mode filtering mode
3311 * @return a list of {@code Locale} instances for matching language tags
3312 * sorted in descending order based on priority or weight, or an empty
3313 * list if nothing matches. The list is modifiable.
3314 * @throws NullPointerException if {@code priorityList} or {@code locales}
3315 * is {@code null}
3316 * @throws IllegalArgumentException if one or more extended language ranges
3317 * are included in the given list when
3318 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3319 *
3320 * @since 1.8
3321 */
3322 public static List<Locale> filter(List<LanguageRange> priorityList,
3323 Collection<Locale> locales,
3324 FilteringMode mode) {
3325 return LocaleMatcher.filter(priorityList, locales, mode);
3326 }
3327
3328 /**
3329 * Returns a list of matching {@code Locale} instances using the filtering
3330 * mechanism defined in RFC 4647. This is equivalent to
3331 * {@link #filter(List, Collection, FilteringMode)} when {@code mode} is
3332 * {@link FilteringMode#AUTOSELECT_FILTERING}.
3333 *
3334 * This filter operation on the given {@code locales} ensures that only
3335 * unique matching locale(s) are returned.
3336 *
3337 * @param priorityList user's Language Priority List in which each language
3338 * tag is sorted in descending order based on priority or weight
3339 * @param locales {@code Locale} instances used for matching
3340 * @return a list of {@code Locale} instances for matching language tags
3341 * sorted in descending order based on priority or weight, or an empty
3342 * list if nothing matches. The list is modifiable.
3343 * @throws NullPointerException if {@code priorityList} or {@code locales}
3344 * is {@code null}
3345 *
3346 * @since 1.8
3347 */
3348 public static List<Locale> filter(List<LanguageRange> priorityList,
3349 Collection<Locale> locales) {
3350 return filter(priorityList, locales, FilteringMode.AUTOSELECT_FILTERING);
3351 }
3352
3353 /**
3354 * Returns a list of matching languages tags using the basic filtering
3355 * mechanism defined in RFC 4647.
3356 *
3357 * This filter operation on the given {@code tags} ensures that only
3358 * unique matching tag(s) are returned with preserved case. In case of
3359 * duplicate matching tags with the case difference, the first matching
3360 * tag with preserved case is returned.
3361 * For example, "de-ch" is returned out of the duplicate matching tags
3362 * "de-ch" and "de-CH", if "de-ch" is checked first for matching in the
3363 * given {@code tags}. Note that if the given {@code tags} is an unordered
3364 * {@code Collection}, the returned matching tag out of duplicate tags is
3365 * subject to change, depending on the implementation of the
3366 * {@code Collection}.
3367 *
3368 * @param priorityList user's Language Priority List in which each language
3369 * tag is sorted in descending order based on priority or weight
3370 * @param tags language tags
3371 * @param mode filtering mode
3372 * @return a list of matching language tags sorted in descending order
3373 * based on priority or weight, or an empty list if nothing matches.
3374 * The list is modifiable.
3375 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3376 * {@code null}
3377 * @throws IllegalArgumentException if one or more extended language ranges
3378 * are included in the given list when
3379 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified
3380 *
3381 * @since 1.8
3382 */
3383 public static List<String> filterTags(List<LanguageRange> priorityList,
3384 Collection<String> tags,
3385 FilteringMode mode) {
3386 return LocaleMatcher.filterTags(priorityList, tags, mode);
3387 }
3388
3389 /**
3390 * Returns a list of matching languages tags using the basic filtering
3391 * mechanism defined in RFC 4647. This is equivalent to
3392 * {@link #filterTags(List, Collection, FilteringMode)} when {@code mode}
3393 * is {@link FilteringMode#AUTOSELECT_FILTERING}.
3394 *
3395 * This filter operation on the given {@code tags} ensures that only
3396 * unique matching tag(s) are returned with preserved case. In case of
3397 * duplicate matching tags with the case difference, the first matching
3398 * tag with preserved case is returned.
3399 * For example, "de-ch" is returned out of the duplicate matching tags
3400 * "de-ch" and "de-CH", if "de-ch" is checked first for matching in the
3401 * given {@code tags}. Note that if the given {@code tags} is an unordered
3402 * {@code Collection}, the returned matching tag out of duplicate tags is
3403 * subject to change, depending on the implementation of the
3404 * {@code Collection}.
3405 *
3406 * @param priorityList user's Language Priority List in which each language
3407 * tag is sorted in descending order based on priority or weight
3408 * @param tags language tags
3409 * @return a list of matching language tags sorted in descending order
3410 * based on priority or weight, or an empty list if nothing matches.
3411 * The list is modifiable.
3412 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3413 * {@code null}
3414 *
3415 * @since 1.8
3416 */
3417 public static List<String> filterTags(List<LanguageRange> priorityList,
3418 Collection<String> tags) {
3419 return filterTags(priorityList, tags, FilteringMode.AUTOSELECT_FILTERING);
3420 }
3421
3422 /**
3423 * Returns a {@code Locale} instance for the best-matching language
3424 * tag using the lookup mechanism defined in RFC 4647.
3425 *
3426 * @param priorityList user's Language Priority List in which each language
3427 * tag is sorted in descending order based on priority or weight
3428 * @param locales {@code Locale} instances used for matching
3429 * @return the best matching <code>Locale</code> instance chosen based on
3430 * priority or weight, or {@code null} if nothing matches.
3431 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3432 * {@code null}
3433 *
3434 * @since 1.8
3435 */
3436 public static Locale lookup(List<LanguageRange> priorityList,
3437 Collection<Locale> locales) {
3438 return LocaleMatcher.lookup(priorityList, locales);
3439 }
3440
3441 /**
3442 * Returns the best-matching language tag using the lookup mechanism
3443 * defined in RFC 4647.
3444 *
3445 * This lookup operation on the given {@code tags} ensures that the
3446 * first matching tag with preserved case is returned.
3447 *
3448 * @param priorityList user's Language Priority List in which each language
3449 * tag is sorted in descending order based on priority or weight
3450 * @param tags language tangs used for matching
3451 * @return the best matching language tag chosen based on priority or
3452 * weight, or {@code null} if nothing matches.
3453 * @throws NullPointerException if {@code priorityList} or {@code tags} is
3454 * {@code null}
3455 *
3456 * @since 1.8
3457 */
3458 public static String lookupTag(List<LanguageRange> priorityList,
3459 Collection<String> tags) {
3460 return LocaleMatcher.lookupTag(priorityList, tags);
3461 }
3462
3463 }