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