1 /* 2 * Copyright (c) 1996, 2016, Oracle and/or its affiliates. All rights reserved. 3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 4 * 5 * This code is free software; you can redistribute it and/or modify it 6 * under the terms of the GNU General Public License version 2 only, as 7 * published by the Free Software Foundation. Oracle designates this 8 * particular file as subject to the "Classpath" exception as provided 9 * by Oracle in the LICENSE file that accompanied this code. 10 * 11 * This code is distributed in the hope that it will be useful, but WITHOUT 12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 14 * version 2 for more details (a copy is included in the LICENSE file that 15 * accompanied this code). 16 * 17 * You should have received a copy of the GNU General Public License version 18 * 2 along with this work; if not, write to the Free Software Foundation, 19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 20 * 21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 22 * or visit www.oracle.com if you need additional information or have any 23 * questions. 24 */ 25 26 /* 27 * (C) Copyright Taligent, Inc. 1996, 1997 - All Rights Reserved 28 * (C) Copyright IBM Corp. 1996 - 1998 - All Rights Reserved 29 * 30 * The original version of this source code and documentation 31 * is copyrighted and owned by Taligent, Inc., a wholly-owned 32 * subsidiary of IBM. These materials are provided under terms 33 * of a License Agreement between Taligent and Sun. This technology 34 * is protected by multiple US and International patents. 35 * 36 * This notice and attribution to Taligent may not be removed. 37 * Taligent is a registered trademark of Taligent, Inc. 38 * 39 */ 40 41 package java.util; 42 43 import java.io.IOException; 44 import java.io.ObjectInputStream; 45 import java.io.ObjectOutputStream; 46 import java.io.ObjectStreamField; 47 import java.io.Serializable; 48 import java.text.MessageFormat; 49 import java.util.concurrent.ConcurrentHashMap; 50 import java.util.spi.LocaleNameProvider; 51 52 import sun.security.action.GetPropertyAction; 53 import sun.util.locale.BaseLocale; 54 import sun.util.locale.InternalLocaleBuilder; 55 import sun.util.locale.LanguageTag; 56 import sun.util.locale.LocaleExtensions; 57 import sun.util.locale.LocaleMatcher; 58 import sun.util.locale.LocaleObjectCache; 59 import sun.util.locale.LocaleSyntaxException; 60 import sun.util.locale.LocaleUtils; 61 import sun.util.locale.ParseStatus; 62 import sun.util.locale.provider.LocaleProviderAdapter; 63 import sun.util.locale.provider.LocaleResources; 64 import sun.util.locale.provider.LocaleServiceProviderPool; 65 66 /** 67 * A <code>Locale</code> object represents a specific geographical, political, 68 * or cultural region. An operation that requires a <code>Locale</code> to perform 69 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code> 70 * to tailor information for the user. For example, displaying a number 71 * is a locale-sensitive operation— the number should be formatted 72 * according to the customs and conventions of the user's native country, 73 * region, or culture. 74 * 75 * <p> The {@code Locale} class implements IETF BCP 47 which is composed of 76 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 "Matching of Language 77 * Tags"</a> and <a href="http://tools.ietf.org/html/rfc5646">RFC 5646 "Tags 78 * for Identifying Languages"</a> with support for the LDML (UTS#35, "Unicode 79 * Locale Data Markup Language") BCP 47-compatible extensions for locale data 80 * exchange. 81 * 82 * <p> A <code>Locale</code> object logically consists of the fields 83 * described below. 84 * 85 * <dl> 86 * <dt><a id="def_language"><b>language</b></a></dt> 87 * 88 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered 89 * language subtags up to 8 alpha letters (for future enhancements). 90 * When a language has both an alpha-2 code and an alpha-3 code, the 91 * alpha-2 code must be used. You can find a full list of valid 92 * language codes in the IANA Language Subtag Registry (search for 93 * "Type: language"). The language field is case insensitive, but 94 * <code>Locale</code> always canonicalizes to lower case.</dd> 95 * 96 * <dd>Well-formed language values have the form 97 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the full 98 * BCP47 language production, since it excludes extlang. They are 99 * not needed since modern three-letter language codes replace 100 * them.</dd> 101 * 102 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd> 103 * 104 * <dt><a id="def_script"><b>script</b></a></dt> 105 * 106 * <dd>ISO 15924 alpha-4 script code. You can find a full list of 107 * valid script codes in the IANA Language Subtag Registry (search 108 * for "Type: script"). The script field is case insensitive, but 109 * <code>Locale</code> always canonicalizes to title case (the first 110 * letter is upper case and the rest of the letters are lower 111 * case).</dd> 112 * 113 * <dd>Well-formed script values have the form 114 * <code>[a-zA-Z]{4}</code></dd> 115 * 116 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd> 117 * 118 * <dt><a id="def_region"><b>country (region)</b></a></dt> 119 * 120 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code. 121 * You can find a full list of valid country and region codes in the 122 * IANA Language Subtag Registry (search for "Type: region"). The 123 * country (region) field is case insensitive, but 124 * <code>Locale</code> always canonicalizes to upper case.</dd> 125 * 126 * <dd>Well-formed country/region values have 127 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd> 128 * 129 * <dd>Example: "US" (United States), "FR" (France), "029" 130 * (Caribbean)</dd> 131 * 132 * <dt><a id="def_variant"><b>variant</b></a></dt> 133 * 134 * <dd>Any arbitrary value used to indicate a variation of a 135 * <code>Locale</code>. Where there are two or more variant values 136 * each indicating its own semantics, these values should be ordered 137 * by importance, with most important first, separated by 138 * underscore('_'). The variant field is case sensitive.</dd> 139 * 140 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant 141 * subtags. Also BCP 47 subtags are strictly used to indicate 142 * additional variations that define a language or its dialects that 143 * are not covered by any combinations of language, script and 144 * region subtags. You can find a full list of valid variant codes 145 * in the IANA Language Subtag Registry (search for "Type: variant"). 146 * 147 * <p>However, the variant field in <code>Locale</code> has 148 * historically been used for any kind of variation, not just 149 * language variations. For example, some supported variants 150 * available in Java SE Runtime Environments indicate alternative 151 * cultural behaviors such as calendar type or number script. In 152 * BCP 47 this kind of information, which does not identify the 153 * language, is supported by extension subtags or private use 154 * subtags.</dd> 155 * 156 * <dd>Well-formed variant values have the form <code>SUBTAG 157 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG = 158 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only 159 * uses hyphen ('-') as a delimiter, this is more lenient).</dd> 160 * 161 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd> 162 * 163 * <dt><a id="def_extensions"><b>extensions</b></a></dt> 164 * 165 * <dd>A map from single character keys to string values, indicating 166 * extensions apart from language identification. The extensions in 167 * <code>Locale</code> implement the semantics and syntax of BCP 47 168 * extension subtags and private use subtags. The extensions are 169 * case insensitive, but <code>Locale</code> canonicalizes all 170 * extension keys and values to lower case. Note that extensions 171 * cannot have empty values.</dd> 172 * 173 * <dd>Well-formed keys are single characters from the set 174 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form 175 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x' 176 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys 177 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows 178 * single-character subtags).</dd> 179 * 180 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar), 181 * key="x"/value="java-1-7"</dd> 182 * </dl> 183 * 184 * <b>Note:</b> Although BCP 47 requires field values to be registered 185 * in the IANA Language Subtag Registry, the <code>Locale</code> class 186 * does not provide any validation features. The <code>Builder</code> 187 * only checks if an individual field satisfies the syntactic 188 * requirement (is well-formed), but does not validate the value 189 * itself. See {@link Builder} for details. 190 * 191 * <h3><a id="def_locale_extension">Unicode locale/language extension</a></h3> 192 * 193 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional 194 * attributes and keywords to override or refine the default behavior 195 * associated with a locale. A keyword is represented by a pair of 196 * key and type. For example, "nu-thai" indicates that Thai local 197 * digits (value:"thai") should be used for formatting numbers 198 * (key:"nu"). 199 * 200 * <p>The keywords are mapped to a BCP 47 extension value using the 201 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above 202 * example, "nu-thai", becomes the extension "u-nu-thai". 203 * 204 * <p>Thus, when a <code>Locale</code> object contains Unicode locale 205 * attributes and keywords, 206 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a 207 * String representing this information, for example, "nu-thai". The 208 * <code>Locale</code> class also provides {@link 209 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and 210 * {@link #getUnicodeLocaleType} which allow you to access Unicode 211 * locale attributes and key/type pairs directly. When represented as 212 * a string, the Unicode Locale Extension lists attributes 213 * alphabetically, followed by key/type sequences with keys listed 214 * alphabetically (the order of subtags comprising a key's type is 215 * fixed when the type is defined) 216 * 217 * <p>A well-formed locale key has the form 218 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the 219 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it 220 * can be empty, or a series of subtags 3-8 alphanums in length). A 221 * well-formed locale attribute has the form 222 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same 223 * form as a locale type subtag). 224 * 225 * <p>The Unicode locale extension specifies optional behavior in 226 * locale-sensitive services. Although the LDML specification defines 227 * various keys and values, actual locale-sensitive service 228 * implementations in a Java Runtime Environment might not support any 229 * particular Unicode locale attributes or key/type pairs. 230 * 231 * <h4>Creating a Locale</h4> 232 * 233 * <p>There are several different ways to create a <code>Locale</code> 234 * object. 235 * 236 * <h5>Builder</h5> 237 * 238 * <p>Using {@link Builder} you can construct a <code>Locale</code> object 239 * that conforms to BCP 47 syntax. 240 * 241 * <h5>Constructors</h5> 242 * 243 * <p>The <code>Locale</code> class provides three constructors: 244 * <blockquote> 245 * <pre> 246 * {@link #Locale(String language)} 247 * {@link #Locale(String language, String country)} 248 * {@link #Locale(String language, String country, String variant)} 249 * </pre> 250 * </blockquote> 251 * These constructors allow you to create a <code>Locale</code> object 252 * with language, country and variant, but you cannot specify 253 * script or extensions. 254 * 255 * <h5>Factory Methods</h5> 256 * 257 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code> 258 * object for a well-formed BCP 47 language tag. 259 * 260 * <h5>Locale Constants</h5> 261 * 262 * <p>The <code>Locale</code> class provides a number of convenient constants 263 * that you can use to create <code>Locale</code> objects for commonly used 264 * locales. For example, the following creates a <code>Locale</code> object 265 * for the United States: 266 * <blockquote> 267 * <pre> 268 * Locale.US 269 * </pre> 270 * </blockquote> 271 * 272 * <h4><a id="LocaleMatching">Locale Matching</a></h4> 273 * 274 * <p>If an application or a system is internationalized and provides localized 275 * resources for multiple locales, it sometimes needs to find one or more 276 * locales (or language tags) which meet each user's specific preferences. Note 277 * that a term "language tag" is used interchangeably with "locale" in this 278 * locale matching documentation. 279 * 280 * <p>In order to do matching a user's preferred locales to a set of language 281 * tags, <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of 282 * Language Tags</a> defines two mechanisms: filtering and lookup. 283 * <em>Filtering</em> is used to get all matching locales, whereas 284 * <em>lookup</em> is to choose the best matching locale. 285 * Matching is done case-insensitively. These matching mechanisms are described 286 * in the following sections. 287 * 288 * <p>A user's preference is called a <em>Language Priority List</em> and is 289 * expressed as a list of language ranges. There are syntactically two types of 290 * language ranges: basic and extended. See 291 * {@link Locale.LanguageRange Locale.LanguageRange} for details. 292 * 293 * <h5>Filtering</h5> 294 * 295 * <p>The filtering operation returns all matching language tags. It is defined 296 * in RFC 4647 as follows: 297 * "In filtering, each language range represents the least specific language 298 * tag (that is, the language tag with fewest number of subtags) that is an 299 * acceptable match. All of the language tags in the matching set of tags will 300 * have an equal or greater number of subtags than the language range. Every 301 * non-wildcard subtag in the language range will appear in every one of the 302 * matching language tags." 303 * 304 * <p>There are two types of filtering: filtering for basic language ranges 305 * (called "basic filtering") and filtering for extended language ranges 306 * (called "extended filtering"). They may return different results by what 307 * kind of language ranges are included in the given Language Priority List. 308 * {@link Locale.FilteringMode} is a parameter to specify how filtering should 309 * be done. 310 * 311 * <h5>Lookup</h5> 312 * 313 * <p>The lookup operation returns the best matching language tags. It is 314 * defined in RFC 4647 as follows: 315 * "By contrast with filtering, each language range represents the most 316 * specific tag that is an acceptable match. The first matching tag found, 317 * according to the user's priority, is considered the closest match and is the 318 * item returned." 319 * 320 * <p>For example, if a Language Priority List consists of two language ranges, 321 * {@code "zh-Hant-TW"} and {@code "en-US"}, in prioritized order, lookup 322 * method progressively searches the language tags below in order to find the 323 * best matching language tag. 324 * <blockquote> 325 * <pre> 326 * 1. zh-Hant-TW 327 * 2. zh-Hant 328 * 3. zh 329 * 4. en-US 330 * 5. en 331 * </pre> 332 * </blockquote> 333 * If there is a language tag which matches completely to a language range 334 * above, the language tag is returned. 335 * 336 * <p>{@code "*"} is the special language range, and it is ignored in lookup. 337 * 338 * <p>If multiple language tags match as a result of the subtag {@code '*'} 339 * included in a language range, the first matching language tag returned by 340 * an {@link Iterator} over a {@link Collection} of language tags is treated as 341 * the best matching one. 342 * 343 * <h4>Use of Locale</h4> 344 * 345 * <p>Once you've created a <code>Locale</code> you can query it for information 346 * about itself. Use <code>getCountry</code> to get the country (or region) 347 * code and <code>getLanguage</code> to get the language code. 348 * You can use <code>getDisplayCountry</code> to get the 349 * name of the country suitable for displaying to the user. Similarly, 350 * you can use <code>getDisplayLanguage</code> to get the name of 351 * the language suitable for displaying to the user. Interestingly, 352 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive 353 * and have two versions: one that uses the default 354 * {@link Locale.Category#DISPLAY DISPLAY} locale and one 355 * that uses the locale specified as an argument. 356 * 357 * <p>The Java Platform provides a number of classes that perform locale-sensitive 358 * operations. For example, the <code>NumberFormat</code> class formats 359 * numbers, currency, and percentages in a locale-sensitive manner. Classes 360 * such as <code>NumberFormat</code> have several convenience methods 361 * for creating a default object of that type. For example, the 362 * <code>NumberFormat</code> class provides these three convenience methods 363 * for creating a default <code>NumberFormat</code> object: 364 * <blockquote> 365 * <pre> 366 * NumberFormat.getInstance() 367 * NumberFormat.getCurrencyInstance() 368 * NumberFormat.getPercentInstance() 369 * </pre> 370 * </blockquote> 371 * Each of these methods has two variants; one with an explicit locale 372 * and one without; the latter uses the default 373 * {@link Locale.Category#FORMAT FORMAT} locale: 374 * <blockquote> 375 * <pre> 376 * NumberFormat.getInstance(myLocale) 377 * NumberFormat.getCurrencyInstance(myLocale) 378 * NumberFormat.getPercentInstance(myLocale) 379 * </pre> 380 * </blockquote> 381 * A <code>Locale</code> is the mechanism for identifying the kind of object 382 * (<code>NumberFormat</code>) that you would like to get. The locale is 383 * <STRONG>just</STRONG> a mechanism for identifying objects, 384 * <STRONG>not</STRONG> a container for the objects themselves. 385 * 386 * <h4>Compatibility</h4> 387 * 388 * <p>In order to maintain compatibility with existing usage, Locale's 389 * constructors retain their behavior prior to the Java Runtime 390 * Environment version 1.7. The same is largely true for the 391 * <code>toString</code> method. Thus Locale objects can continue to 392 * be used as they were. In particular, clients who parse the output 393 * of toString into language, country, and variant fields can continue 394 * to do so (although this is strongly discouraged), although the 395 * variant field will have additional information in it if script or 396 * extensions are present. 397 * 398 * <p>In addition, BCP 47 imposes syntax restrictions that are not 399 * imposed by Locale's constructors. This means that conversions 400 * between some Locales and BCP 47 language tags cannot be made without 401 * losing information. Thus <code>toLanguageTag</code> cannot 402 * represent the state of locales whose language, country, or variant 403 * do not conform to BCP 47. 404 * 405 * <p>Because of these issues, it is recommended that clients migrate 406 * away from constructing non-conforming locales and use the 407 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead. 408 * Clients desiring a string representation of the complete locale can 409 * then always rely on <code>toLanguageTag</code> for this purpose. 410 * 411 * <h5><a id="special_cases_constructor">Special cases</a></h5> 412 * 413 * <p>For compatibility reasons, two 414 * non-conforming locales are treated as special cases. These are 415 * <b>{@code ja_JP_JP}</b> and <b>{@code th_TH_TH}</b>. These are ill-formed 416 * in BCP 47 since the variants are too short. To ease migration to BCP 47, 417 * these are treated specially during construction. These two cases (and only 418 * these) cause a constructor to generate an extension, all other values behave 419 * exactly as they did prior to Java 7. 420 * 421 * <p>Java has used {@code ja_JP_JP} to represent Japanese as used in 422 * Japan together with the Japanese Imperial calendar. This is now 423 * representable using a Unicode locale extension, by specifying the 424 * Unicode locale key {@code ca} (for "calendar") and type 425 * {@code japanese}. When the Locale constructor is called with the 426 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is 427 * automatically added. 428 * 429 * <p>Java has used {@code th_TH_TH} to represent Thai as used in 430 * Thailand together with Thai digits. This is also now representable using 431 * a Unicode locale extension, by specifying the Unicode locale key 432 * {@code nu} (for "number") and value {@code thai}. When the Locale 433 * constructor is called with the arguments "th", "TH", "TH", the 434 * extension "u-nu-thai" is automatically added. 435 * 436 * <h5>Serialization</h5> 437 * 438 * <p>During serialization, writeObject writes all fields to the output 439 * stream, including extensions. 440 * 441 * <p>During deserialization, readResolve adds extensions as described 442 * in <a href="#special_cases_constructor">Special Cases</a>, only 443 * for the two cases th_TH_TH and ja_JP_JP. 444 * 445 * <h5>Legacy language codes</h5> 446 * 447 * <p>Locale's constructor has always converted three language codes to 448 * their earlier, obsoleted forms: {@code he} maps to {@code iw}, 449 * {@code yi} maps to {@code ji}, and {@code id} maps to 450 * {@code in}. This continues to be the case, in order to not break 451 * backwards compatibility. 452 * 453 * <p>The APIs added in 1.7 map between the old and new language codes, 454 * maintaining the old codes internal to Locale (so that 455 * <code>getLanguage</code> and <code>toString</code> reflect the old 456 * code), but using the new codes in the BCP 47 language tag APIs (so 457 * that <code>toLanguageTag</code> reflects the new one). This 458 * preserves the equivalence between Locales no matter which code or 459 * API is used to construct them. Java's default resource bundle 460 * lookup mechanism also implements this mapping, so that resources 461 * can be named using either convention, see {@link ResourceBundle.Control}. 462 * 463 * <h5>Three-letter language/country(region) codes</h5> 464 * 465 * <p>The Locale constructors have always specified that the language 466 * and the country param be two characters in length, although in 467 * practice they have accepted any length. The specification has now 468 * been relaxed to allow language codes of two to eight characters and 469 * country (region) codes of two to three characters, and in 470 * particular, three-letter language codes and three-digit region 471 * codes as specified in the IANA Language Subtag Registry. For 472 * compatibility, the implementation still does not impose a length 473 * constraint. 474 * 475 * @see Builder 476 * @see ResourceBundle 477 * @see java.text.Format 478 * @see java.text.NumberFormat 479 * @see java.text.Collator 480 * @author Mark Davis 481 * @since 1.1 482 */ 483 public final class Locale implements Cloneable, Serializable { 484 485 private static final Cache LOCALECACHE = new Cache(); 486 487 /** Useful constant for language. 488 */ 489 public static final Locale ENGLISH = createConstant("en", ""); 490 491 /** Useful constant for language. 492 */ 493 public static final Locale FRENCH = createConstant("fr", ""); 494 495 /** Useful constant for language. 496 */ 497 public static final Locale GERMAN = createConstant("de", ""); 498 499 /** Useful constant for language. 500 */ 501 public static final Locale ITALIAN = createConstant("it", ""); 502 503 /** Useful constant for language. 504 */ 505 public static final Locale JAPANESE = createConstant("ja", ""); 506 507 /** Useful constant for language. 508 */ 509 public static final Locale KOREAN = createConstant("ko", ""); 510 511 /** Useful constant for language. 512 */ 513 public static final Locale CHINESE = createConstant("zh", ""); 514 515 /** Useful constant for language. 516 */ 517 public static final Locale SIMPLIFIED_CHINESE = createConstant("zh", "CN"); 518 519 /** Useful constant for language. 520 */ 521 public static final Locale TRADITIONAL_CHINESE = createConstant("zh", "TW"); 522 523 /** Useful constant for country. 524 */ 525 public static final Locale FRANCE = createConstant("fr", "FR"); 526 527 /** Useful constant for country. 528 */ 529 public static final Locale GERMANY = createConstant("de", "DE"); 530 531 /** Useful constant for country. 532 */ 533 public static final Locale ITALY = createConstant("it", "IT"); 534 535 /** Useful constant for country. 536 */ 537 public static final Locale JAPAN = createConstant("ja", "JP"); 538 539 /** Useful constant for country. 540 */ 541 public static final Locale KOREA = createConstant("ko", "KR"); 542 543 /** Useful constant for country. 544 */ 545 public static final Locale CHINA = SIMPLIFIED_CHINESE; 546 547 /** Useful constant for country. 548 */ 549 public static final Locale PRC = SIMPLIFIED_CHINESE; 550 551 /** Useful constant for country. 552 */ 553 public static final Locale TAIWAN = TRADITIONAL_CHINESE; 554 555 /** Useful constant for country. 556 */ 557 public static final Locale UK = createConstant("en", "GB"); 558 559 /** Useful constant for country. 560 */ 561 public static final Locale US = createConstant("en", "US"); 562 563 /** Useful constant for country. 564 */ 565 public static final Locale CANADA = createConstant("en", "CA"); 566 567 /** Useful constant for country. 568 */ 569 public static final Locale CANADA_FRENCH = createConstant("fr", "CA"); 570 571 /** 572 * Useful constant for the root locale. The root locale is the locale whose 573 * language, country, and variant are empty ("") strings. This is regarded 574 * as the base locale of all locales, and is used as the language/country 575 * neutral locale for the locale sensitive operations. 576 * 577 * @since 1.6 578 */ 579 public static final Locale ROOT = createConstant("", ""); 580 581 /** 582 * The key for the private use extension ('x'). 583 * 584 * @see #getExtension(char) 585 * @see Builder#setExtension(char, String) 586 * @since 1.7 587 */ 588 public static final char PRIVATE_USE_EXTENSION = 'x'; 589 590 /** 591 * The key for Unicode locale extension ('u'). 592 * 593 * @see #getExtension(char) 594 * @see Builder#setExtension(char, String) 595 * @since 1.7 596 */ 597 public static final char UNICODE_LOCALE_EXTENSION = 'u'; 598 599 /** serialization ID 600 */ 601 static final long serialVersionUID = 9149081749638150636L; 602 603 /** 604 * Enum for specifying the type defined in ISO 3166. This enum is used to 605 * retrieve the two-letter ISO3166-1 alpha-2, three-letter ISO3166-1 606 * alpha-3, four-letter ISO3166-3 country codes. 607 * 608 * @see #getISOCountries(Locale.IsoCountryCode) 609 * @since 9 610 */ 611 public static enum IsoCountryCode { 612 /** 613 * PART1_ALPHA2 is used to represent the ISO3166-1 alpha-2 two letter 614 * country codes. 615 */ 616 PART1_ALPHA2 { 617 @Override 618 Set<String> createCountryCodeSet() { 619 return Set.of(Locale.getISOCountries()); 620 } 621 }, 622 623 /** 624 * 625 * PART1_ALPHA3 is used to represent the ISO3166-1 alpha-3 three letter 626 * country codes. 627 */ 628 PART1_ALPHA3 { 629 @Override 630 Set<String> createCountryCodeSet() { 631 return LocaleISOData.computeISO3166_1Alpha3Countries(); 632 } 633 }, 634 635 /** 636 * PART3 is used to represent the ISO3166-3 four letter country codes. 637 */ 638 PART3 { 639 @Override 640 Set<String> createCountryCodeSet() { 641 return Set.of(LocaleISOData.ISO3166_3); 642 } 643 }; 644 645 /** 646 * Concrete implementation of this method attempts to compute value 647 * for iso3166CodesMap for each IsoCountryCode type key. 648 */ 649 abstract Set<String> createCountryCodeSet(); 650 651 /** 652 * Map to hold country codes for each ISO3166 part. 653 */ 654 private static Map<IsoCountryCode, Set<String>> iso3166CodesMap = new ConcurrentHashMap<>(); 655 656 /** 657 * This method is called from Locale class to retrieve country code set 658 * for getISOCountries(type) 659 */ 660 static Set<String> retrieveISOCountryCodes(IsoCountryCode type) { 661 return iso3166CodesMap.computeIfAbsent(type, IsoCountryCode::createCountryCodeSet); 662 } 663 } 664 665 /** 666 * Display types for retrieving localized names from the name providers. 667 */ 668 private static final int DISPLAY_LANGUAGE = 0; 669 private static final int DISPLAY_COUNTRY = 1; 670 private static final int DISPLAY_VARIANT = 2; 671 private static final int DISPLAY_SCRIPT = 3; 672 673 /** 674 * Private constructor used by getInstance method 675 */ 676 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) { 677 this.baseLocale = baseLocale; 678 this.localeExtensions = extensions; 679 } 680 681 /** 682 * Construct a locale from language, country and variant. 683 * This constructor normalizes the language value to lowercase and 684 * the country value to uppercase. 685 * <p> 686 * <b>Note:</b> 687 * <ul> 688 * <li>ISO 639 is not a stable standard; some of the language codes it defines 689 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 690 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 691 * API on Locale will return only the OLD codes. 692 * <li>For backward compatibility reasons, this constructor does not make 693 * any syntactic checks on the input. 694 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially, 695 * see <a href="#special_cases_constructor">Special Cases</a> for more information. 696 * </ul> 697 * 698 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 699 * up to 8 characters in length. See the <code>Locale</code> class description about 700 * valid language values. 701 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code. 702 * See the <code>Locale</code> class description about valid country values. 703 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>. 704 * See the <code>Locale</code> class description for the details. 705 * @exception NullPointerException thrown if any argument is null. 706 */ 707 public Locale(String language, String country, String variant) { 708 if (language== null || country == null || variant == null) { 709 throw new NullPointerException(); 710 } 711 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant); 712 localeExtensions = getCompatibilityExtensions(language, "", country, variant); 713 } 714 715 /** 716 * Construct a locale from language and country. 717 * This constructor normalizes the language value to lowercase and 718 * the country value to uppercase. 719 * <p> 720 * <b>Note:</b> 721 * <ul> 722 * <li>ISO 639 is not a stable standard; some of the language codes it defines 723 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 724 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 725 * API on Locale will return only the OLD codes. 726 * <li>For backward compatibility reasons, this constructor does not make 727 * any syntactic checks on the input. 728 * </ul> 729 * 730 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 731 * up to 8 characters in length. See the <code>Locale</code> class description about 732 * valid language values. 733 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code. 734 * See the <code>Locale</code> class description about valid country values. 735 * @exception NullPointerException thrown if either argument is null. 736 */ 737 public Locale(String language, String country) { 738 this(language, country, ""); 739 } 740 741 /** 742 * Construct a locale from a language code. 743 * This constructor normalizes the language value to lowercase. 744 * <p> 745 * <b>Note:</b> 746 * <ul> 747 * <li>ISO 639 is not a stable standard; some of the language codes it defines 748 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 749 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 750 * API on Locale will return only the OLD codes. 751 * <li>For backward compatibility reasons, this constructor does not make 752 * any syntactic checks on the input. 753 * </ul> 754 * 755 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 756 * up to 8 characters in length. See the <code>Locale</code> class description about 757 * valid language values. 758 * @exception NullPointerException thrown if argument is null. 759 * @since 1.4 760 */ 761 public Locale(String language) { 762 this(language, "", ""); 763 } 764 765 /** 766 * This method must be called only for creating the Locale.* 767 * constants due to making shortcuts. 768 */ 769 private static Locale createConstant(String lang, String country) { 770 BaseLocale base = BaseLocale.createInstance(lang, country); 771 return getInstance(base, null); 772 } 773 774 /** 775 * Returns a <code>Locale</code> constructed from the given 776 * <code>language</code>, <code>country</code> and 777 * <code>variant</code>. If the same <code>Locale</code> instance 778 * is available in the cache, then that instance is 779 * returned. Otherwise, a new <code>Locale</code> instance is 780 * created and cached. 781 * 782 * @param language lowercase 2 to 8 language code. 783 * @param country uppercase two-letter ISO-3166 code and numeric-3 UN M.49 area code. 784 * @param variant vendor and browser specific code. See class description. 785 * @return the <code>Locale</code> instance requested 786 * @exception NullPointerException if any argument is null. 787 */ 788 static Locale getInstance(String language, String country, String variant) { 789 return getInstance(language, "", country, variant, null); 790 } 791 792 static Locale getInstance(String language, String script, String country, 793 String variant, LocaleExtensions extensions) { 794 if (language== null || script == null || country == null || variant == null) { 795 throw new NullPointerException(); 796 } 797 798 if (extensions == null) { 799 extensions = getCompatibilityExtensions(language, script, country, variant); 800 } 801 802 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant); 803 return getInstance(baseloc, extensions); 804 } 805 806 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) { 807 LocaleKey key = new LocaleKey(baseloc, extensions); 808 return LOCALECACHE.get(key); 809 } 810 811 private static class Cache extends LocaleObjectCache<LocaleKey, Locale> { 812 private Cache() { 813 } 814 815 @Override 816 protected Locale createObject(LocaleKey key) { 817 return new Locale(key.base, key.exts); 818 } 819 } 820 821 private static final class LocaleKey { 822 private final BaseLocale base; 823 private final LocaleExtensions exts; 824 private final int hash; 825 826 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) { 827 base = baseLocale; 828 exts = extensions; 829 830 // Calculate the hash value here because it's always used. 831 int h = base.hashCode(); 832 if (exts != null) { 833 h ^= exts.hashCode(); 834 } 835 hash = h; 836 } 837 838 @Override 839 public boolean equals(Object obj) { 840 if (this == obj) { 841 return true; 842 } 843 if (!(obj instanceof LocaleKey)) { 844 return false; 845 } 846 LocaleKey other = (LocaleKey)obj; 847 if (hash != other.hash || !base.equals(other.base)) { 848 return false; 849 } 850 if (exts == null) { 851 return other.exts == null; 852 } 853 return exts.equals(other.exts); 854 } 855 856 @Override 857 public int hashCode() { 858 return hash; 859 } 860 } 861 862 /** 863 * Gets the current value of the default locale for this instance 864 * of the Java Virtual Machine. 865 * <p> 866 * The Java Virtual Machine sets the default locale during startup 867 * based on the host environment. It is used by many locale-sensitive 868 * methods if no locale is explicitly specified. 869 * It can be changed using the 870 * {@link #setDefault(java.util.Locale) setDefault} method. 871 * 872 * @return the default locale for this instance of the Java Virtual Machine 873 */ 874 public static Locale getDefault() { 875 // do not synchronize this method - see 4071298 876 return defaultLocale; 877 } 878 879 /** 880 * Gets the current value of the default locale for the specified Category 881 * for this instance of the Java Virtual Machine. 882 * <p> 883 * The Java Virtual Machine sets the default locale during startup based 884 * on the host environment. It is used by many locale-sensitive methods 885 * if no locale is explicitly specified. It can be changed using the 886 * setDefault(Locale.Category, Locale) method. 887 * 888 * @param category - the specified category to get the default locale 889 * @throws NullPointerException if category is null 890 * @return the default locale for the specified Category for this instance 891 * of the Java Virtual Machine 892 * @see #setDefault(Locale.Category, Locale) 893 * @since 1.7 894 */ 895 public static Locale getDefault(Locale.Category category) { 896 // do not synchronize this method - see 4071298 897 switch (category) { 898 case DISPLAY: 899 if (defaultDisplayLocale == null) { 900 synchronized(Locale.class) { 901 if (defaultDisplayLocale == null) { 902 defaultDisplayLocale = initDefault(category); 903 } 904 } 905 } 906 return defaultDisplayLocale; 907 case FORMAT: 908 if (defaultFormatLocale == null) { 909 synchronized(Locale.class) { 910 if (defaultFormatLocale == null) { 911 defaultFormatLocale = initDefault(category); 912 } 913 } 914 } 915 return defaultFormatLocale; 916 default: 917 assert false: "Unknown Category"; 918 } 919 return getDefault(); 920 } 921 922 private static Locale initDefault() { 923 String language, region, script, country, variant; 924 Properties props = GetPropertyAction.privilegedGetProperties(); 925 language = props.getProperty("user.language", "en"); 926 // for compatibility, check for old user.region property 927 region = props.getProperty("user.region"); 928 if (region != null) { 929 // region can be of form country, country_variant, or _variant 930 int i = region.indexOf('_'); 931 if (i >= 0) { 932 country = region.substring(0, i); 933 variant = region.substring(i + 1); 934 } else { 935 country = region; 936 variant = ""; 937 } 938 script = ""; 939 } else { 940 script = props.getProperty("user.script", ""); 941 country = props.getProperty("user.country", ""); 942 variant = props.getProperty("user.variant", ""); 943 } 944 945 return getInstance(language, script, country, variant, null); 946 } 947 948 private static Locale initDefault(Locale.Category category) { 949 Properties props = GetPropertyAction.privilegedGetProperties(); 950 return getInstance( 951 props.getProperty(category.languageKey, 952 defaultLocale.getLanguage()), 953 props.getProperty(category.scriptKey, 954 defaultLocale.getScript()), 955 props.getProperty(category.countryKey, 956 defaultLocale.getCountry()), 957 props.getProperty(category.variantKey, 958 defaultLocale.getVariant()), 959 null); 960 } 961 962 /** 963 * Sets the default locale for this instance of the Java Virtual Machine. 964 * This does not affect the host locale. 965 * <p> 966 * If there is a security manager, its <code>checkPermission</code> 967 * method is called with a <code>PropertyPermission("user.language", "write")</code> 968 * permission before the default locale is changed. 969 * <p> 970 * The Java Virtual Machine sets the default locale during startup 971 * based on the host environment. It is used by many locale-sensitive 972 * methods if no locale is explicitly specified. 973 * <p> 974 * Since changing the default locale may affect many different areas 975 * of functionality, this method should only be used if the caller 976 * is prepared to reinitialize locale-sensitive code running 977 * within the same Java Virtual Machine. 978 * <p> 979 * By setting the default locale with this method, all of the default 980 * locales for each Category are also set to the specified default locale. 981 * 982 * @throws SecurityException 983 * if a security manager exists and its 984 * <code>checkPermission</code> method doesn't allow the operation. 985 * @throws NullPointerException if <code>newLocale</code> is null 986 * @param newLocale the new default locale 987 * @see SecurityManager#checkPermission 988 * @see java.util.PropertyPermission 989 */ 990 public static synchronized void setDefault(Locale newLocale) { 991 setDefault(Category.DISPLAY, newLocale); 992 setDefault(Category.FORMAT, newLocale); 993 defaultLocale = newLocale; 994 } 995 996 /** 997 * Sets the default locale for the specified Category for this instance 998 * of the Java Virtual Machine. This does not affect the host locale. 999 * <p> 1000 * If there is a security manager, its checkPermission method is called 1001 * with a PropertyPermission("user.language", "write") permission before 1002 * the default locale is changed. 1003 * <p> 1004 * The Java Virtual Machine sets the default locale during startup based 1005 * on the host environment. It is used by many locale-sensitive methods 1006 * if no locale is explicitly specified. 1007 * <p> 1008 * Since changing the default locale may affect many different areas of 1009 * functionality, this method should only be used if the caller is 1010 * prepared to reinitialize locale-sensitive code running within the 1011 * same Java Virtual Machine. 1012 * 1013 * @param category - the specified category to set the default locale 1014 * @param newLocale - the new default locale 1015 * @throws SecurityException if a security manager exists and its 1016 * checkPermission method doesn't allow the operation. 1017 * @throws NullPointerException if category and/or newLocale is null 1018 * @see SecurityManager#checkPermission(java.security.Permission) 1019 * @see PropertyPermission 1020 * @see #getDefault(Locale.Category) 1021 * @since 1.7 1022 */ 1023 public static synchronized void setDefault(Locale.Category category, 1024 Locale newLocale) { 1025 if (category == null) 1026 throw new NullPointerException("Category cannot be NULL"); 1027 if (newLocale == null) 1028 throw new NullPointerException("Can't set default locale to NULL"); 1029 1030 SecurityManager sm = System.getSecurityManager(); 1031 if (sm != null) sm.checkPermission(new PropertyPermission 1032 ("user.language", "write")); 1033 switch (category) { 1034 case DISPLAY: 1035 defaultDisplayLocale = newLocale; 1036 break; 1037 case FORMAT: 1038 defaultFormatLocale = newLocale; 1039 break; 1040 default: 1041 assert false: "Unknown Category"; 1042 } 1043 } 1044 1045 /** 1046 * Returns an array of all installed locales. 1047 * The returned array represents the union of locales supported 1048 * by the Java runtime environment and by installed 1049 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider} 1050 * implementations. It must contain at least a <code>Locale</code> 1051 * instance equal to {@link java.util.Locale#US Locale.US}. 1052 * 1053 * @return An array of installed locales. 1054 */ 1055 public static Locale[] getAvailableLocales() { 1056 return LocaleServiceProviderPool.getAllAvailableLocales(); 1057 } 1058 1059 /** 1060 * Returns a list of all 2-letter country codes defined in ISO 3166. 1061 * Can be used to create Locales. 1062 * This method is equivalent to {@link #getISOCountries(Locale.IsoCountryCode type)} 1063 * with {@code type} {@link IsoCountryCode#PART1_ALPHA2}. 1064 * <p> 1065 * <b>Note:</b> The <code>Locale</code> class also supports other codes for 1066 * country (region), such as 3-letter numeric UN M.49 area codes. 1067 * Therefore, the list returned by this method does not contain ALL valid 1068 * codes that can be used to create Locales. 1069 * <p> 1070 * Note that this method does not return obsolete 2-letter country codes. 1071 * ISO3166-3 codes which designate country codes for those obsolete codes, 1072 * can be retrieved from {@link #getISOCountries(Locale.IsoCountryCode type)} with 1073 * {@code type} {@link IsoCountryCode#PART3}. 1074 * @return An array of ISO 3166 two-letter country codes. 1075 */ 1076 public static String[] getISOCountries() { 1077 if (isoCountries == null) { 1078 isoCountries = getISO2Table(LocaleISOData.isoCountryTable); 1079 } 1080 String[] result = new String[isoCountries.length]; 1081 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length); 1082 return result; 1083 } 1084 1085 /** 1086 * Returns a {@code Set} of ISO3166 country codes for the specified type. 1087 * 1088 * @param type {@link Locale.IsoCountryCode} specified ISO code type. 1089 * @see java.util.Locale.IsoCountryCode 1090 * @throws NullPointerException if type is null 1091 * @return a {@code Set} of ISO country codes for the specified type. 1092 * @since 9 1093 */ 1094 public static Set<String> getISOCountries(IsoCountryCode type) { 1095 Objects.requireNonNull(type); 1096 return IsoCountryCode.retrieveISOCountryCodes(type); 1097 } 1098 1099 /** 1100 * Returns a list of all 2-letter language codes defined in ISO 639. 1101 * Can be used to create Locales. 1102 * <p> 1103 * <b>Note:</b> 1104 * <ul> 1105 * <li>ISO 639 is not a stable standard— some languages' codes have changed. 1106 * The list this function returns includes both the new and the old codes for the 1107 * languages whose codes have changed. 1108 * <li>The <code>Locale</code> class also supports language codes up to 1109 * 8 characters in length. Therefore, the list returned by this method does 1110 * not contain ALL valid codes that can be used to create Locales. 1111 * </ul> 1112 * 1113 * @return An array of ISO 639 two-letter language codes. 1114 */ 1115 public static String[] getISOLanguages() { 1116 if (isoLanguages == null) { 1117 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable); 1118 } 1119 String[] result = new String[isoLanguages.length]; 1120 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length); 1121 return result; 1122 } 1123 1124 private static String[] getISO2Table(String table) { 1125 int len = table.length() / 5; 1126 String[] isoTable = new String[len]; 1127 for (int i = 0, j = 0; i < len; i++, j += 5) { 1128 isoTable[i] = table.substring(j, j + 2); 1129 } 1130 return isoTable; 1131 } 1132 1133 /** 1134 * Returns the language code of this Locale. 1135 * 1136 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed. 1137 * Locale's constructor recognizes both the new and the old codes for the languages 1138 * whose codes have changed, but this function always returns the old code. If you 1139 * want to check for a specific language whose code has changed, don't do 1140 * <pre> 1141 * if (locale.getLanguage().equals("he")) // BAD! 1142 * ... 1143 * </pre> 1144 * Instead, do 1145 * <pre> 1146 * if (locale.getLanguage().equals(new Locale("he").getLanguage())) 1147 * ... 1148 * </pre> 1149 * @return The language code, or the empty string if none is defined. 1150 * @see #getDisplayLanguage 1151 */ 1152 public String getLanguage() { 1153 return baseLocale.getLanguage(); 1154 } 1155 1156 /** 1157 * Returns the script for this locale, which should 1158 * either be the empty string or an ISO 15924 4-letter script 1159 * code. The first letter is uppercase and the rest are 1160 * lowercase, for example, 'Latn', 'Cyrl'. 1161 * 1162 * @return The script code, or the empty string if none is defined. 1163 * @see #getDisplayScript 1164 * @since 1.7 1165 */ 1166 public String getScript() { 1167 return baseLocale.getScript(); 1168 } 1169 1170 /** 1171 * Returns the country/region code for this locale, which should 1172 * either be the empty string, an uppercase ISO 3166 2-letter code, 1173 * or a UN M.49 3-digit code. 1174 * 1175 * @return The country/region code, or the empty string if none is defined. 1176 * @see #getDisplayCountry 1177 */ 1178 public String getCountry() { 1179 return baseLocale.getRegion(); 1180 } 1181 1182 /** 1183 * Returns the variant code for this locale. 1184 * 1185 * @return The variant code, or the empty string if none is defined. 1186 * @see #getDisplayVariant 1187 */ 1188 public String getVariant() { 1189 return baseLocale.getVariant(); 1190 } 1191 1192 /** 1193 * Returns {@code true} if this {@code Locale} has any <a href="#def_extensions"> 1194 * extensions</a>. 1195 * 1196 * @return {@code true} if this {@code Locale} has any extensions 1197 * @since 1.8 1198 */ 1199 public boolean hasExtensions() { 1200 return localeExtensions != null; 1201 } 1202 1203 /** 1204 * Returns a copy of this {@code Locale} with no <a href="#def_extensions"> 1205 * extensions</a>. If this {@code Locale} has no extensions, this {@code Locale} 1206 * is returned. 1207 * 1208 * @return a copy of this {@code Locale} with no extensions, or {@code this} 1209 * if {@code this} has no extensions 1210 * @since 1.8 1211 */ 1212 public Locale stripExtensions() { 1213 return hasExtensions() ? Locale.getInstance(baseLocale, null) : this; 1214 } 1215 1216 /** 1217 * Returns the extension (or private use) value associated with 1218 * the specified key, or null if there is no extension 1219 * associated with the key. To be well-formed, the key must be one 1220 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so 1221 * for example 'z' and 'Z' represent the same extension. 1222 * 1223 * @param key the extension key 1224 * @return The extension, or null if this locale defines no 1225 * extension for the specified key. 1226 * @throws IllegalArgumentException if key is not well-formed 1227 * @see #PRIVATE_USE_EXTENSION 1228 * @see #UNICODE_LOCALE_EXTENSION 1229 * @since 1.7 1230 */ 1231 public String getExtension(char key) { 1232 if (!LocaleExtensions.isValidKey(key)) { 1233 throw new IllegalArgumentException("Ill-formed extension key: " + key); 1234 } 1235 return hasExtensions() ? localeExtensions.getExtensionValue(key) : null; 1236 } 1237 1238 /** 1239 * Returns the set of extension keys associated with this locale, or the 1240 * empty set if it has no extensions. The returned set is unmodifiable. 1241 * The keys will all be lower-case. 1242 * 1243 * @return The set of extension keys, or the empty set if this locale has 1244 * no extensions. 1245 * @since 1.7 1246 */ 1247 public Set<Character> getExtensionKeys() { 1248 if (!hasExtensions()) { 1249 return Collections.emptySet(); 1250 } 1251 return localeExtensions.getKeys(); 1252 } 1253 1254 /** 1255 * Returns the set of unicode locale attributes associated with 1256 * this locale, or the empty set if it has no attributes. The 1257 * returned set is unmodifiable. 1258 * 1259 * @return The set of attributes. 1260 * @since 1.7 1261 */ 1262 public Set<String> getUnicodeLocaleAttributes() { 1263 if (!hasExtensions()) { 1264 return Collections.emptySet(); 1265 } 1266 return localeExtensions.getUnicodeLocaleAttributes(); 1267 } 1268 1269 /** 1270 * Returns the Unicode locale type associated with the specified Unicode locale key 1271 * for this locale. Returns the empty string for keys that are defined with no type. 1272 * Returns null if the key is not defined. Keys are case-insensitive. The key must 1273 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is 1274 * thrown. 1275 * 1276 * @param key the Unicode locale key 1277 * @return The Unicode locale type associated with the key, or null if the 1278 * locale does not define the key. 1279 * @throws IllegalArgumentException if the key is not well-formed 1280 * @throws NullPointerException if <code>key</code> is null 1281 * @since 1.7 1282 */ 1283 public String getUnicodeLocaleType(String key) { 1284 if (!isUnicodeExtensionKey(key)) { 1285 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key); 1286 } 1287 return hasExtensions() ? localeExtensions.getUnicodeLocaleType(key) : null; 1288 } 1289 1290 /** 1291 * Returns the set of Unicode locale keys defined by this locale, or the empty set if 1292 * this locale has none. The returned set is immutable. Keys are all lower case. 1293 * 1294 * @return The set of Unicode locale keys, or the empty set if this locale has 1295 * no Unicode locale keywords. 1296 * @since 1.7 1297 */ 1298 public Set<String> getUnicodeLocaleKeys() { 1299 if (localeExtensions == null) { 1300 return Collections.emptySet(); 1301 } 1302 return localeExtensions.getUnicodeLocaleKeys(); 1303 } 1304 1305 /** 1306 * Package locale method returning the Locale's BaseLocale, 1307 * used by ResourceBundle 1308 * @return base locale of this Locale 1309 */ 1310 BaseLocale getBaseLocale() { 1311 return baseLocale; 1312 } 1313 1314 /** 1315 * Package private method returning the Locale's LocaleExtensions, 1316 * used by ResourceBundle. 1317 * @return locale extensions of this Locale, 1318 * or {@code null} if no extensions are defined 1319 */ 1320 LocaleExtensions getLocaleExtensions() { 1321 return localeExtensions; 1322 } 1323 1324 /** 1325 * Returns a string representation of this <code>Locale</code> 1326 * object, consisting of language, country, variant, script, 1327 * and extensions as below: 1328 * <blockquote> 1329 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "_" + extensions 1330 * </blockquote> 1331 * 1332 * Language is always lower case, country is always upper case, script is always title 1333 * case, and extensions are always lower case. Extensions and private use subtags 1334 * will be in canonical order as explained in {@link #toLanguageTag}. 1335 * 1336 * <p>When the locale has neither script nor extensions, the result is the same as in 1337 * Java 6 and prior. 1338 * 1339 * <p>If both the language and country fields are missing, this function will return 1340 * the empty string, even if the variant, script, or extensions field is present (you 1341 * can't have a locale with just a variant, the variant must accompany a well-formed 1342 * language or country code). 1343 * 1344 * <p>If script or extensions are present and variant is missing, no underscore is 1345 * added before the "#". 1346 * 1347 * <p>This behavior is designed to support debugging and to be compatible with 1348 * previous uses of <code>toString</code> that expected language, country, and variant 1349 * fields only. To represent a Locale as a String for interchange purposes, use 1350 * {@link #toLanguageTag}. 1351 * 1352 * <p>Examples: <ul> 1353 * <li>{@code en}</li> 1354 * <li>{@code de_DE}</li> 1355 * <li>{@code _GB}</li> 1356 * <li>{@code en_US_WIN}</li> 1357 * <li>{@code de__POSIX}</li> 1358 * <li>{@code zh_CN_#Hans}</li> 1359 * <li>{@code zh_TW_#Hant_x-java}</li> 1360 * <li>{@code th_TH_TH_#u-nu-thai}</li></ul> 1361 * 1362 * @return A string representation of the Locale, for debugging. 1363 * @see #getDisplayName 1364 * @see #toLanguageTag 1365 */ 1366 @Override 1367 public final String toString() { 1368 boolean l = (baseLocale.getLanguage().length() != 0); 1369 boolean s = (baseLocale.getScript().length() != 0); 1370 boolean r = (baseLocale.getRegion().length() != 0); 1371 boolean v = (baseLocale.getVariant().length() != 0); 1372 boolean e = (localeExtensions != null && localeExtensions.getID().length() != 0); 1373 1374 StringBuilder result = new StringBuilder(baseLocale.getLanguage()); 1375 if (r || (l && (v || s || e))) { 1376 result.append('_') 1377 .append(baseLocale.getRegion()); // This may just append '_' 1378 } 1379 if (v && (l || r)) { 1380 result.append('_') 1381 .append(baseLocale.getVariant()); 1382 } 1383 1384 if (s && (l || r)) { 1385 result.append("_#") 1386 .append(baseLocale.getScript()); 1387 } 1388 1389 if (e && (l || r)) { 1390 result.append('_'); 1391 if (!s) { 1392 result.append('#'); 1393 } 1394 result.append(localeExtensions.getID()); 1395 } 1396 1397 return result.toString(); 1398 } 1399 1400 /** 1401 * Returns a well-formed IETF BCP 47 language tag representing 1402 * this locale. 1403 * 1404 * <p>If this <code>Locale</code> has a language, country, or 1405 * variant that does not satisfy the IETF BCP 47 language tag 1406 * syntax requirements, this method handles these fields as 1407 * described below: 1408 * 1409 * <p><b>Language:</b> If language is empty, or not <a 1410 * href="#def_language" >well-formed</a> (for example "a" or 1411 * "e2"), it will be emitted as "und" (Undetermined). 1412 * 1413 * <p><b>Country:</b> If country is not <a 1414 * href="#def_region">well-formed</a> (for example "12" or "USA"), 1415 * it will be omitted. 1416 * 1417 * <p><b>Variant:</b> If variant <b>is</b> <a 1418 * href="#def_variant">well-formed</a>, each sub-segment 1419 * (delimited by '-' or '_') is emitted as a subtag. Otherwise: 1420 * <ul> 1421 * 1422 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code> 1423 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first 1424 * ill-formed sub-segment and all following will be appended to 1425 * the private use subtag. The first appended subtag will be 1426 * "lvariant", followed by the sub-segments in order, separated by 1427 * hyphen. For example, "x-lvariant-WIN", 1428 * "Oracle-x-lvariant-JDK-Standard-Edition". 1429 * 1430 * <li>if any sub-segment does not match 1431 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated 1432 * and the problematic sub-segment and all following sub-segments 1433 * will be omitted. If the remainder is non-empty, it will be 1434 * emitted as a private use subtag as above (even if the remainder 1435 * turns out to be well-formed). For example, 1436 * "Solaris_isjustthecoolestthing" is emitted as 1437 * "x-lvariant-Solaris", not as "solaris".</li></ul> 1438 * 1439 * <p><b>Special Conversions:</b> Java supports some old locale 1440 * representations, including deprecated ISO language codes, 1441 * for compatibility. This method performs the following 1442 * conversions: 1443 * <ul> 1444 * 1445 * <li>Deprecated ISO language codes "iw", "ji", and "in" are 1446 * converted to "he", "yi", and "id", respectively. 1447 * 1448 * <li>A locale with language "no", country "NO", and variant 1449 * "NY", representing Norwegian Nynorsk (Norway), is converted 1450 * to a language tag "nn-NO".</li></ul> 1451 * 1452 * <p><b>Note:</b> Although the language tag created by this 1453 * method is well-formed (satisfies the syntax requirements 1454 * defined by the IETF BCP 47 specification), it is not 1455 * necessarily a valid BCP 47 language tag. For example, 1456 * <pre> 1457 * new Locale("xx", "YY").toLanguageTag();</pre> 1458 * 1459 * will return "xx-YY", but the language subtag "xx" and the 1460 * region subtag "YY" are invalid because they are not registered 1461 * in the IANA Language Subtag Registry. 1462 * 1463 * @return a BCP47 language tag representing the locale 1464 * @see #forLanguageTag(String) 1465 * @since 1.7 1466 */ 1467 public String toLanguageTag() { 1468 if (languageTag != null) { 1469 return languageTag; 1470 } 1471 1472 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions); 1473 StringBuilder buf = new StringBuilder(); 1474 1475 String subtag = tag.getLanguage(); 1476 if (subtag.length() > 0) { 1477 buf.append(LanguageTag.canonicalizeLanguage(subtag)); 1478 } 1479 1480 subtag = tag.getScript(); 1481 if (subtag.length() > 0) { 1482 buf.append(LanguageTag.SEP); 1483 buf.append(LanguageTag.canonicalizeScript(subtag)); 1484 } 1485 1486 subtag = tag.getRegion(); 1487 if (subtag.length() > 0) { 1488 buf.append(LanguageTag.SEP); 1489 buf.append(LanguageTag.canonicalizeRegion(subtag)); 1490 } 1491 1492 List<String>subtags = tag.getVariants(); 1493 for (String s : subtags) { 1494 buf.append(LanguageTag.SEP); 1495 // preserve casing 1496 buf.append(s); 1497 } 1498 1499 subtags = tag.getExtensions(); 1500 for (String s : subtags) { 1501 buf.append(LanguageTag.SEP); 1502 buf.append(LanguageTag.canonicalizeExtension(s)); 1503 } 1504 1505 subtag = tag.getPrivateuse(); 1506 if (subtag.length() > 0) { 1507 if (buf.length() > 0) { 1508 buf.append(LanguageTag.SEP); 1509 } 1510 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP); 1511 // preserve casing 1512 buf.append(subtag); 1513 } 1514 1515 String langTag = buf.toString(); 1516 synchronized (this) { 1517 if (languageTag == null) { 1518 languageTag = langTag; 1519 } 1520 } 1521 return languageTag; 1522 } 1523 1524 /** 1525 * Returns a locale for the specified IETF BCP 47 language tag string. 1526 * 1527 * <p>If the specified language tag contains any ill-formed subtags, 1528 * the first such subtag and all following subtags are ignored. Compare 1529 * to {@link Locale.Builder#setLanguageTag} which throws an exception 1530 * in this case. 1531 * 1532 * <p>The following <b>conversions</b> are performed:<ul> 1533 * 1534 * <li>The language code "und" is mapped to language "". 1535 * 1536 * <li>The language codes "he", "yi", and "id" are mapped to "iw", 1537 * "ji", and "in" respectively. (This is the same canonicalization 1538 * that's done in Locale's constructors.) 1539 * 1540 * <li>The portion of a private use subtag prefixed by "lvariant", 1541 * if any, is removed and appended to the variant field in the 1542 * result locale (without case normalization). If it is then 1543 * empty, the private use subtag is discarded: 1544 * 1545 * <pre> 1546 * Locale loc; 1547 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX"); 1548 * loc.getVariant(); // returns "POSIX" 1549 * loc.getExtension('x'); // returns null 1550 * 1551 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def"); 1552 * loc.getVariant(); // returns "POSIX_Abc_Def" 1553 * loc.getExtension('x'); // returns "urp" 1554 * </pre> 1555 * 1556 * <li>When the languageTag argument contains an extlang subtag, 1557 * the first such subtag is used as the language, and the primary 1558 * language subtag and other extlang subtags are ignored: 1559 * 1560 * <pre> 1561 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao" 1562 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US" 1563 * </pre> 1564 * 1565 * <li>Case is normalized except for variant tags, which are left 1566 * unchanged. Language is normalized to lower case, script to 1567 * title case, country to upper case, and extensions to lower 1568 * case. 1569 * 1570 * <li>If, after processing, the locale would exactly match either 1571 * ja_JP_JP or th_TH_TH with no extensions, the appropriate 1572 * extensions are added as though the constructor had been called: 1573 * 1574 * <pre> 1575 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag(); 1576 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP" 1577 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag(); 1578 * // returns "th-TH-u-nu-thai-x-lvariant-TH" 1579 * </pre></ul> 1580 * 1581 * <p>This implements the 'Language-Tag' production of BCP47, and 1582 * so supports grandfathered (regular and irregular) as well as 1583 * private use language tags. Stand alone private use tags are 1584 * represented as empty language and extension 'x-whatever', 1585 * and grandfathered tags are converted to their canonical replacements 1586 * where they exist. 1587 * 1588 * <p>Grandfathered tags with canonical replacements are as follows: 1589 * 1590 * <table class="altrows"> 1591 * <caption style="display:none">Grandfathered tags with canonical replacements</caption> 1592 * <thead style="text-align:center"> 1593 * <tr><th style="padding: 0 2px">grandfathered tag</th><th style="padding: 0 2px">modern replacement</th></tr> 1594 * </thead> 1595 * <tbody style="text-align:center"> 1596 * <tr><td>art-lojban</td><td>jbo</td></tr> 1597 * <tr><td>i-ami</td><td>ami</td></tr> 1598 * <tr><td>i-bnn</td><td>bnn</td></tr> 1599 * <tr><td>i-hak</td><td>hak</td></tr> 1600 * <tr><td>i-klingon</td><td>tlh</td></tr> 1601 * <tr><td>i-lux</td><td>lb</td></tr> 1602 * <tr><td>i-navajo</td><td>nv</td></tr> 1603 * <tr><td>i-pwn</td><td>pwn</td></tr> 1604 * <tr><td>i-tao</td><td>tao</td></tr> 1605 * <tr><td>i-tay</td><td>tay</td></tr> 1606 * <tr><td>i-tsu</td><td>tsu</td></tr> 1607 * <tr><td>no-bok</td><td>nb</td></tr> 1608 * <tr><td>no-nyn</td><td>nn</td></tr> 1609 * <tr><td>sgn-BE-FR</td><td>sfb</td></tr> 1610 * <tr><td>sgn-BE-NL</td><td>vgt</td></tr> 1611 * <tr><td>sgn-CH-DE</td><td>sgg</td></tr> 1612 * <tr><td>zh-guoyu</td><td>cmn</td></tr> 1613 * <tr><td>zh-hakka</td><td>hak</td></tr> 1614 * <tr><td>zh-min-nan</td><td>nan</td></tr> 1615 * <tr><td>zh-xiang</td><td>hsn</td></tr> 1616 * </tbody> 1617 * </table> 1618 * 1619 * <p>Grandfathered tags with no modern replacement will be 1620 * converted as follows: 1621 * 1622 * <table class="altrows"> 1623 * <caption style="display:none">Grandfathered tags with no modern replacement</caption> 1624 * <thead style="text-align:center"> 1625 * <tr><th style="padding: 0 2px">grandfathered tag</th><th style="padding: 0 2px">converts to</th></tr> 1626 * </thead> 1627 * <tbody style="text-align:center"> 1628 * <tr><td>cel-gaulish</td><td>xtg-x-cel-gaulish</td></tr> 1629 * <tr><td>en-GB-oed</td><td>en-GB-x-oed</td></tr> 1630 * <tr><td>i-default</td><td>en-x-i-default</td></tr> 1631 * <tr><td>i-enochian</td><td>und-x-i-enochian</td></tr> 1632 * <tr><td>i-mingo</td><td>see-x-i-mingo</td></tr> 1633 * <tr><td>zh-min</td><td>nan-x-zh-min</td></tr> 1634 * </tbody> 1635 * </table> 1636 * 1637 * <p>For a list of all grandfathered tags, see the 1638 * IANA Language Subtag Registry (search for "Type: grandfathered"). 1639 * 1640 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code> 1641 * and <code>forLanguageTag</code> will round-trip. 1642 * 1643 * @param languageTag the language tag 1644 * @return The locale that best represents the language tag. 1645 * @throws NullPointerException if <code>languageTag</code> is <code>null</code> 1646 * @see #toLanguageTag() 1647 * @see java.util.Locale.Builder#setLanguageTag(String) 1648 * @since 1.7 1649 */ 1650 public static Locale forLanguageTag(String languageTag) { 1651 LanguageTag tag = LanguageTag.parse(languageTag, null); 1652 InternalLocaleBuilder bldr = new InternalLocaleBuilder(); 1653 bldr.setLanguageTag(tag); 1654 BaseLocale base = bldr.getBaseLocale(); 1655 LocaleExtensions exts = bldr.getLocaleExtensions(); 1656 if (exts == null && base.getVariant().length() > 0) { 1657 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(), 1658 base.getRegion(), base.getVariant()); 1659 } 1660 return getInstance(base, exts); 1661 } 1662 1663 /** 1664 * Returns a three-letter abbreviation of this locale's language. 1665 * If the language matches an ISO 639-1 two-letter code, the 1666 * corresponding ISO 639-2/T three-letter lowercase code is 1667 * returned. The ISO 639-2 language codes can be found on-line, 1668 * see "Codes for the Representation of Names of Languages Part 2: 1669 * Alpha-3 Code". If the locale specifies a three-letter 1670 * language, the language is returned as is. If the locale does 1671 * not specify a language the empty string is returned. 1672 * 1673 * @return A three-letter abbreviation of this locale's language. 1674 * @exception MissingResourceException Throws MissingResourceException if 1675 * three-letter language abbreviation is not available for this locale. 1676 */ 1677 public String getISO3Language() throws MissingResourceException { 1678 String lang = baseLocale.getLanguage(); 1679 if (lang.length() == 3) { 1680 return lang; 1681 } 1682 1683 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable); 1684 if (language3 == null) { 1685 throw new MissingResourceException("Couldn't find 3-letter language code for " 1686 + lang, "FormatData_" + toString(), "ShortLanguage"); 1687 } 1688 return language3; 1689 } 1690 1691 /** 1692 * Returns a three-letter abbreviation for this locale's country. 1693 * If the country matches an ISO 3166-1 alpha-2 code, the 1694 * corresponding ISO 3166-1 alpha-3 uppercase code is returned. 1695 * If the locale doesn't specify a country, this will be the empty 1696 * string. 1697 * 1698 * <p>The ISO 3166-1 codes can be found on-line. 1699 * 1700 * @return A three-letter abbreviation of this locale's country. 1701 * @exception MissingResourceException Throws MissingResourceException if the 1702 * three-letter country abbreviation is not available for this locale. 1703 */ 1704 public String getISO3Country() throws MissingResourceException { 1705 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable); 1706 if (country3 == null) { 1707 throw new MissingResourceException("Couldn't find 3-letter country code for " 1708 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry"); 1709 } 1710 return country3; 1711 } 1712 1713 private static String getISO3Code(String iso2Code, String table) { 1714 int codeLength = iso2Code.length(); 1715 if (codeLength == 0) { 1716 return ""; 1717 } 1718 1719 int tableLength = table.length(); 1720 int index = tableLength; 1721 if (codeLength == 2) { 1722 char c1 = iso2Code.charAt(0); 1723 char c2 = iso2Code.charAt(1); 1724 for (index = 0; index < tableLength; index += 5) { 1725 if (table.charAt(index) == c1 1726 && table.charAt(index + 1) == c2) { 1727 break; 1728 } 1729 } 1730 } 1731 return index < tableLength ? table.substring(index + 2, index + 5) : null; 1732 } 1733 1734 /** 1735 * Returns a name for the locale's language that is appropriate for display to the 1736 * user. 1737 * If possible, the name returned will be localized for the default 1738 * {@link Locale.Category#DISPLAY DISPLAY} locale. 1739 * For example, if the locale is fr_FR and the default 1740 * {@link Locale.Category#DISPLAY DISPLAY} locale 1741 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and 1742 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR, 1743 * getDisplayLanguage() will return "anglais". 1744 * If the name returned cannot be localized for the default 1745 * {@link Locale.Category#DISPLAY DISPLAY} locale, 1746 * (say, we don't have a Japanese name for Croatian), 1747 * this function falls back on the English name, and uses the ISO code as a last-resort 1748 * value. If the locale doesn't specify a language, this function returns the empty string. 1749 * 1750 * @return The name of the display language. 1751 */ 1752 public final String getDisplayLanguage() { 1753 return getDisplayLanguage(getDefault(Category.DISPLAY)); 1754 } 1755 1756 /** 1757 * Returns a name for the locale's language that is appropriate for display to the 1758 * user. 1759 * If possible, the name returned will be localized according to inLocale. 1760 * For example, if the locale is fr_FR and inLocale 1761 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and 1762 * inLocale is fr_FR, getDisplayLanguage() will return "anglais". 1763 * If the name returned cannot be localized according to inLocale, 1764 * (say, we don't have a Japanese name for Croatian), 1765 * this function falls back on the English name, and finally 1766 * on the ISO code as a last-resort value. If the locale doesn't specify a language, 1767 * this function returns the empty string. 1768 * 1769 * @param inLocale The locale for which to retrieve the display language. 1770 * @return The name of the display language appropriate to the given locale. 1771 * @exception NullPointerException if <code>inLocale</code> is <code>null</code> 1772 */ 1773 public String getDisplayLanguage(Locale inLocale) { 1774 return getDisplayString(baseLocale.getLanguage(), inLocale, DISPLAY_LANGUAGE); 1775 } 1776 1777 /** 1778 * Returns a name for the locale's script that is appropriate for display to 1779 * the user. If possible, the name will be localized for the default 1780 * {@link Locale.Category#DISPLAY DISPLAY} locale. Returns 1781 * the empty string if this locale doesn't specify a script code. 1782 * 1783 * @return the display name of the script code for the current default 1784 * {@link Locale.Category#DISPLAY DISPLAY} locale 1785 * @since 1.7 1786 */ 1787 public String getDisplayScript() { 1788 return getDisplayScript(getDefault(Category.DISPLAY)); 1789 } 1790 1791 /** 1792 * Returns a name for the locale's script that is appropriate 1793 * for display to the user. If possible, the name will be 1794 * localized for the given locale. Returns the empty string if 1795 * this locale doesn't specify a script code. 1796 * 1797 * @param inLocale The locale for which to retrieve the display script. 1798 * @return the display name of the script code for the current default 1799 * {@link Locale.Category#DISPLAY DISPLAY} locale 1800 * @throws NullPointerException if <code>inLocale</code> is <code>null</code> 1801 * @since 1.7 1802 */ 1803 public String getDisplayScript(Locale inLocale) { 1804 return getDisplayString(baseLocale.getScript(), inLocale, DISPLAY_SCRIPT); 1805 } 1806 1807 /** 1808 * Returns a name for the locale's country that is appropriate for display to the 1809 * user. 1810 * If possible, the name returned will be localized for the default 1811 * {@link Locale.Category#DISPLAY DISPLAY} locale. 1812 * For example, if the locale is fr_FR and the default 1813 * {@link Locale.Category#DISPLAY DISPLAY} locale 1814 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and 1815 * the default {@link Locale.Category#DISPLAY DISPLAY} locale is fr_FR, 1816 * getDisplayCountry() will return "Etats-Unis". 1817 * If the name returned cannot be localized for the default 1818 * {@link Locale.Category#DISPLAY DISPLAY} locale, 1819 * (say, we don't have a Japanese name for Croatia), 1820 * this function falls back on the English name, and uses the ISO code as a last-resort 1821 * value. If the locale doesn't specify a country, this function returns the empty string. 1822 * 1823 * @return The name of the country appropriate to the locale. 1824 */ 1825 public final String getDisplayCountry() { 1826 return getDisplayCountry(getDefault(Category.DISPLAY)); 1827 } 1828 1829 /** 1830 * Returns a name for the locale's country that is appropriate for display to the 1831 * user. 1832 * If possible, the name returned will be localized according to inLocale. 1833 * For example, if the locale is fr_FR and inLocale 1834 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and 1835 * inLocale is fr_FR, getDisplayCountry() will return "Etats-Unis". 1836 * If the name returned cannot be localized according to inLocale. 1837 * (say, we don't have a Japanese name for Croatia), 1838 * this function falls back on the English name, and finally 1839 * on the ISO code as a last-resort value. If the locale doesn't specify a country, 1840 * this function returns the empty string. 1841 * 1842 * @param inLocale The locale for which to retrieve the display country. 1843 * @return The name of the country appropriate to the given locale. 1844 * @exception NullPointerException if <code>inLocale</code> is <code>null</code> 1845 */ 1846 public String getDisplayCountry(Locale inLocale) { 1847 return getDisplayString(baseLocale.getRegion(), inLocale, DISPLAY_COUNTRY); 1848 } 1849 1850 private String getDisplayString(String code, Locale inLocale, int type) { 1851 if (code.length() == 0) { 1852 return ""; 1853 } 1854 1855 if (inLocale == null) { 1856 throw new NullPointerException(); 1857 } 1858 1859 LocaleServiceProviderPool pool = 1860 LocaleServiceProviderPool.getPool(LocaleNameProvider.class); 1861 String key = (type == DISPLAY_VARIANT ? "%%"+code : code); 1862 String result = pool.getLocalizedObject( 1863 LocaleNameGetter.INSTANCE, 1864 inLocale, key, type, code); 1865 if (result != null) { 1866 return result; 1867 } 1868 1869 return code; 1870 } 1871 1872 /** 1873 * Returns a name for the locale's variant code that is appropriate for display to the 1874 * user. If possible, the name will be localized for the default 1875 * {@link Locale.Category#DISPLAY DISPLAY} locale. If the locale 1876 * doesn't specify a variant code, this function returns the empty string. 1877 * 1878 * @return The name of the display variant code appropriate to the locale. 1879 */ 1880 public final String getDisplayVariant() { 1881 return getDisplayVariant(getDefault(Category.DISPLAY)); 1882 } 1883 1884 /** 1885 * Returns a name for the locale's variant code that is appropriate for display to the 1886 * user. If possible, the name will be localized for inLocale. If the locale 1887 * doesn't specify a variant code, this function returns the empty string. 1888 * 1889 * @param inLocale The locale for which to retrieve the display variant code. 1890 * @return The name of the display variant code appropriate to the given locale. 1891 * @exception NullPointerException if <code>inLocale</code> is <code>null</code> 1892 */ 1893 public String getDisplayVariant(Locale inLocale) { 1894 if (baseLocale.getVariant().length() == 0) 1895 return ""; 1896 1897 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale); 1898 1899 String names[] = getDisplayVariantArray(inLocale); 1900 1901 // Get the localized patterns for formatting a list, and use 1902 // them to format the list. 1903 return formatList(names, 1904 lr.getLocaleName("ListPattern"), 1905 lr.getLocaleName("ListCompositionPattern")); 1906 } 1907 1908 /** 1909 * Returns a name for the locale that is appropriate for display to the 1910 * user. This will be the values returned by getDisplayLanguage(), 1911 * getDisplayScript(), getDisplayCountry(), and getDisplayVariant() assembled 1912 * into a single string. The the non-empty values are used in order, 1913 * with the second and subsequent names in parentheses. For example: 1914 * <blockquote> 1915 * language (script, country, variant)<br> 1916 * language (country)<br> 1917 * language (variant)<br> 1918 * script (country)<br> 1919 * country<br> 1920 * </blockquote> 1921 * depending on which fields are specified in the locale. If the 1922 * language, script, country, and variant fields are all empty, 1923 * this function returns the empty string. 1924 * 1925 * @return The name of the locale appropriate to display. 1926 */ 1927 public final String getDisplayName() { 1928 return getDisplayName(getDefault(Category.DISPLAY)); 1929 } 1930 1931 /** 1932 * Returns a name for the locale that is appropriate for display 1933 * to the user. This will be the values returned by 1934 * getDisplayLanguage(), getDisplayScript(),getDisplayCountry(), 1935 * and getDisplayVariant() assembled into a single string. 1936 * The non-empty values are used in order, 1937 * with the second and subsequent names in parentheses. For example: 1938 * <blockquote> 1939 * language (script, country, variant)<br> 1940 * language (country)<br> 1941 * language (variant)<br> 1942 * script (country)<br> 1943 * country<br> 1944 * </blockquote> 1945 * depending on which fields are specified in the locale. If the 1946 * language, script, country, and variant fields are all empty, 1947 * this function returns the empty string. 1948 * 1949 * @param inLocale The locale for which to retrieve the display name. 1950 * @return The name of the locale appropriate to display. 1951 * @throws NullPointerException if <code>inLocale</code> is <code>null</code> 1952 */ 1953 public String getDisplayName(Locale inLocale) { 1954 LocaleResources lr = LocaleProviderAdapter.forJRE().getLocaleResources(inLocale); 1955 1956 String languageName = getDisplayLanguage(inLocale); 1957 String scriptName = getDisplayScript(inLocale); 1958 String countryName = getDisplayCountry(inLocale); 1959 String[] variantNames = getDisplayVariantArray(inLocale); 1960 1961 // Get the localized patterns for formatting a display name. 1962 String displayNamePattern = lr.getLocaleName("DisplayNamePattern"); 1963 String listPattern = lr.getLocaleName("ListPattern"); 1964 String listCompositionPattern = lr.getLocaleName("ListCompositionPattern"); 1965 1966 // The display name consists of a main name, followed by qualifiers. 1967 // Typically, the format is "MainName (Qualifier, Qualifier)" but this 1968 // depends on what pattern is stored in the display locale. 1969 String mainName = null; 1970 String[] qualifierNames = null; 1971 1972 // The main name is the language, or if there is no language, the script, 1973 // then if no script, the country. If there is no language/script/country 1974 // (an anomalous situation) then the display name is simply the variant's 1975 // display name. 1976 if (languageName.length() == 0 && scriptName.length() == 0 && countryName.length() == 0) { 1977 if (variantNames.length == 0) { 1978 return ""; 1979 } else { 1980 return formatList(variantNames, listPattern, listCompositionPattern); 1981 } 1982 } 1983 ArrayList<String> names = new ArrayList<>(4); 1984 if (languageName.length() != 0) { 1985 names.add(languageName); 1986 } 1987 if (scriptName.length() != 0) { 1988 names.add(scriptName); 1989 } 1990 if (countryName.length() != 0) { 1991 names.add(countryName); 1992 } 1993 if (variantNames.length != 0) { 1994 names.addAll(Arrays.asList(variantNames)); 1995 } 1996 1997 // The first one in the main name 1998 mainName = names.get(0); 1999 2000 // Others are qualifiers 2001 int numNames = names.size(); 2002 qualifierNames = (numNames > 1) ? 2003 names.subList(1, numNames).toArray(new String[numNames - 1]) : new String[0]; 2004 2005 // Create an array whose first element is the number of remaining 2006 // elements. This serves as a selector into a ChoiceFormat pattern from 2007 // the resource. The second and third elements are the main name and 2008 // the qualifier; if there are no qualifiers, the third element is 2009 // unused by the format pattern. 2010 Object[] displayNames = { 2011 qualifierNames.length != 0 ? 2 : 1, 2012 mainName, 2013 // We could also just call formatList() and have it handle the empty 2014 // list case, but this is more efficient, and we want it to be 2015 // efficient since all the language-only locales will not have any 2016 // qualifiers. 2017 qualifierNames.length != 0 ? formatList(qualifierNames, listPattern, listCompositionPattern) : null 2018 }; 2019 2020 if (displayNamePattern != null) { 2021 return new MessageFormat(displayNamePattern).format(displayNames); 2022 } 2023 else { 2024 // If we cannot get the message format pattern, then we use a simple 2025 // hard-coded pattern. This should not occur in practice unless the 2026 // installation is missing some core files (FormatData etc.). 2027 StringBuilder result = new StringBuilder(); 2028 result.append((String)displayNames[1]); 2029 if (displayNames.length > 2) { 2030 result.append(" ("); 2031 result.append((String)displayNames[2]); 2032 result.append(')'); 2033 } 2034 return result.toString(); 2035 } 2036 } 2037 2038 /** 2039 * Overrides Cloneable. 2040 */ 2041 @Override 2042 public Object clone() 2043 { 2044 try { 2045 Locale that = (Locale)super.clone(); 2046 return that; 2047 } catch (CloneNotSupportedException e) { 2048 throw new InternalError(e); 2049 } 2050 } 2051 2052 /** 2053 * Override hashCode. 2054 * Since Locales are often used in hashtables, caches the value 2055 * for speed. 2056 */ 2057 @Override 2058 public int hashCode() { 2059 int hc = hashCodeValue; 2060 if (hc == 0) { 2061 hc = baseLocale.hashCode(); 2062 if (localeExtensions != null) { 2063 hc ^= localeExtensions.hashCode(); 2064 } 2065 hashCodeValue = hc; 2066 } 2067 return hc; 2068 } 2069 2070 // Overrides 2071 2072 /** 2073 * Returns true if this Locale is equal to another object. A Locale is 2074 * deemed equal to another Locale with identical language, script, country, 2075 * variant and extensions, and unequal to all other objects. 2076 * 2077 * @return true if this Locale is equal to the specified object. 2078 */ 2079 @Override 2080 public boolean equals(Object obj) { 2081 if (this == obj) // quick check 2082 return true; 2083 if (!(obj instanceof Locale)) 2084 return false; 2085 BaseLocale otherBase = ((Locale)obj).baseLocale; 2086 if (!baseLocale.equals(otherBase)) { 2087 return false; 2088 } 2089 if (localeExtensions == null) { 2090 return ((Locale)obj).localeExtensions == null; 2091 } 2092 return localeExtensions.equals(((Locale)obj).localeExtensions); 2093 } 2094 2095 // ================= privates ===================================== 2096 2097 private transient BaseLocale baseLocale; 2098 private transient LocaleExtensions localeExtensions; 2099 2100 /** 2101 * Calculated hashcode 2102 */ 2103 private transient volatile int hashCodeValue; 2104 2105 private static volatile Locale defaultLocale = initDefault(); 2106 private static volatile Locale defaultDisplayLocale; 2107 private static volatile Locale defaultFormatLocale; 2108 2109 private transient volatile String languageTag; 2110 2111 /** 2112 * Return an array of the display names of the variant. 2113 * @param bundle the ResourceBundle to use to get the display names 2114 * @return an array of display names, possible of zero length. 2115 */ 2116 private String[] getDisplayVariantArray(Locale inLocale) { 2117 // Split the variant name into tokens separated by '_'. 2118 StringTokenizer tokenizer = new StringTokenizer(baseLocale.getVariant(), "_"); 2119 String[] names = new String[tokenizer.countTokens()]; 2120 2121 // For each variant token, lookup the display name. If 2122 // not found, use the variant name itself. 2123 for (int i=0; i<names.length; ++i) { 2124 names[i] = getDisplayString(tokenizer.nextToken(), 2125 inLocale, DISPLAY_VARIANT); 2126 } 2127 2128 return names; 2129 } 2130 2131 /** 2132 * Format a list using given pattern strings. 2133 * If either of the patterns is null, then a the list is 2134 * formatted by concatenation with the delimiter ','. 2135 * @param stringList the list of strings to be formatted. 2136 * @param listPattern should create a MessageFormat taking 0-3 arguments 2137 * and formatting them into a list. 2138 * @param listCompositionPattern should take 2 arguments 2139 * and is used by composeList. 2140 * @return a string representing the list. 2141 */ 2142 private static String formatList(String[] stringList, String listPattern, String listCompositionPattern) { 2143 // If we have no list patterns, compose the list in a simple, 2144 // non-localized way. 2145 if (listPattern == null || listCompositionPattern == null) { 2146 StringJoiner sj = new StringJoiner(","); 2147 for (int i = 0; i < stringList.length; ++i) { 2148 sj.add(stringList[i]); 2149 } 2150 return sj.toString(); 2151 } 2152 2153 // Compose the list down to three elements if necessary 2154 if (stringList.length > 3) { 2155 MessageFormat format = new MessageFormat(listCompositionPattern); 2156 stringList = composeList(format, stringList); 2157 } 2158 2159 // Rebuild the argument list with the list length as the first element 2160 Object[] args = new Object[stringList.length + 1]; 2161 System.arraycopy(stringList, 0, args, 1, stringList.length); 2162 args[0] = stringList.length; 2163 2164 // Format it using the pattern in the resource 2165 MessageFormat format = new MessageFormat(listPattern); 2166 return format.format(args); 2167 } 2168 2169 /** 2170 * Given a list of strings, return a list shortened to three elements. 2171 * Shorten it by applying the given format to the first two elements 2172 * recursively. 2173 * @param format a format which takes two arguments 2174 * @param list a list of strings 2175 * @return if the list is three elements or shorter, the same list; 2176 * otherwise, a new list of three elements. 2177 */ 2178 private static String[] composeList(MessageFormat format, String[] list) { 2179 if (list.length <= 3) return list; 2180 2181 // Use the given format to compose the first two elements into one 2182 String[] listItems = { list[0], list[1] }; 2183 String newItem = format.format(listItems); 2184 2185 // Form a new list one element shorter 2186 String[] newList = new String[list.length-1]; 2187 System.arraycopy(list, 2, newList, 1, newList.length-1); 2188 newList[0] = newItem; 2189 2190 // Recurse 2191 return composeList(format, newList); 2192 } 2193 2194 // Duplicate of sun.util.locale.UnicodeLocaleExtension.isKey in order to 2195 // avoid its class loading. 2196 private static boolean isUnicodeExtensionKey(String s) { 2197 // 2alphanum 2198 return (s.length() == 2) && LocaleUtils.isAlphaNumericString(s); 2199 } 2200 2201 /** 2202 * @serialField language String 2203 * language subtag in lower case. (See <a href="java/util/Locale.html#getLanguage()">getLanguage()</a>) 2204 * @serialField country String 2205 * country subtag in upper case. (See <a href="java/util/Locale.html#getCountry()">getCountry()</a>) 2206 * @serialField variant String 2207 * variant subtags separated by LOWLINE characters. (See <a href="java/util/Locale.html#getVariant()">getVariant()</a>) 2208 * @serialField hashcode int 2209 * deprecated, for forward compatibility only 2210 * @serialField script String 2211 * script subtag in title case (See <a href="java/util/Locale.html#getScript()">getScript()</a>) 2212 * @serialField extensions String 2213 * canonical representation of extensions, that is, 2214 * BCP47 extensions in alphabetical order followed by 2215 * BCP47 private use subtags, all in lower case letters 2216 * separated by HYPHEN-MINUS characters. 2217 * (See <a href="java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>, 2218 * <a href="java/util/Locale.html#getExtension(char)">getExtension(char)</a>) 2219 */ 2220 private static final ObjectStreamField[] serialPersistentFields = { 2221 new ObjectStreamField("language", String.class), 2222 new ObjectStreamField("country", String.class), 2223 new ObjectStreamField("variant", String.class), 2224 new ObjectStreamField("hashcode", int.class), 2225 new ObjectStreamField("script", String.class), 2226 new ObjectStreamField("extensions", String.class), 2227 }; 2228 2229 /** 2230 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>. 2231 * @param out the <code>ObjectOutputStream</code> to write 2232 * @throws IOException 2233 * @since 1.7 2234 */ 2235 private void writeObject(ObjectOutputStream out) throws IOException { 2236 ObjectOutputStream.PutField fields = out.putFields(); 2237 fields.put("language", baseLocale.getLanguage()); 2238 fields.put("script", baseLocale.getScript()); 2239 fields.put("country", baseLocale.getRegion()); 2240 fields.put("variant", baseLocale.getVariant()); 2241 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID()); 2242 fields.put("hashcode", -1); // place holder just for backward support 2243 out.writeFields(); 2244 } 2245 2246 /** 2247 * Deserializes this <code>Locale</code>. 2248 * @param in the <code>ObjectInputStream</code> to read 2249 * @throws IOException 2250 * @throws ClassNotFoundException 2251 * @throws IllformedLocaleException 2252 * @since 1.7 2253 */ 2254 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException { 2255 ObjectInputStream.GetField fields = in.readFields(); 2256 String language = (String)fields.get("language", ""); 2257 String script = (String)fields.get("script", ""); 2258 String country = (String)fields.get("country", ""); 2259 String variant = (String)fields.get("variant", ""); 2260 String extStr = (String)fields.get("extensions", ""); 2261 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant); 2262 if (extStr.length() > 0) { 2263 try { 2264 InternalLocaleBuilder bldr = new InternalLocaleBuilder(); 2265 bldr.setExtensions(extStr); 2266 localeExtensions = bldr.getLocaleExtensions(); 2267 } catch (LocaleSyntaxException e) { 2268 throw new IllformedLocaleException(e.getMessage()); 2269 } 2270 } else { 2271 localeExtensions = null; 2272 } 2273 } 2274 2275 /** 2276 * Returns a cached <code>Locale</code> instance equivalent to 2277 * the deserialized <code>Locale</code>. When serialized 2278 * language, country and variant fields read from the object data stream 2279 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions 2280 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code> 2281 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script 2282 * type is "thai"). See <a href="Locale.html#special_cases_constructor">Special Cases</a> 2283 * for more information. 2284 * 2285 * @return an instance of <code>Locale</code> equivalent to 2286 * the deserialized <code>Locale</code>. 2287 * @throws java.io.ObjectStreamException 2288 */ 2289 private Object readResolve() throws java.io.ObjectStreamException { 2290 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(), 2291 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions); 2292 } 2293 2294 private static volatile String[] isoLanguages; 2295 2296 private static volatile String[] isoCountries; 2297 2298 private static String convertOldISOCodes(String language) { 2299 // we accept both the old and the new ISO codes for the languages whose ISO 2300 // codes have changed, but we always store the OLD code, for backward compatibility 2301 language = LocaleUtils.toLowerString(language).intern(); 2302 if (language == "he") { 2303 return "iw"; 2304 } else if (language == "yi") { 2305 return "ji"; 2306 } else if (language == "id") { 2307 return "in"; 2308 } else { 2309 return language; 2310 } 2311 } 2312 2313 private static LocaleExtensions getCompatibilityExtensions(String language, 2314 String script, 2315 String country, 2316 String variant) { 2317 LocaleExtensions extensions = null; 2318 // Special cases for backward compatibility support 2319 if (LocaleUtils.caseIgnoreMatch(language, "ja") 2320 && script.length() == 0 2321 && LocaleUtils.caseIgnoreMatch(country, "jp") 2322 && "JP".equals(variant)) { 2323 // ja_JP_JP -> u-ca-japanese (calendar = japanese) 2324 extensions = LocaleExtensions.CALENDAR_JAPANESE; 2325 } else if (LocaleUtils.caseIgnoreMatch(language, "th") 2326 && script.length() == 0 2327 && LocaleUtils.caseIgnoreMatch(country, "th") 2328 && "TH".equals(variant)) { 2329 // th_TH_TH -> u-nu-thai (numbersystem = thai) 2330 extensions = LocaleExtensions.NUMBER_THAI; 2331 } 2332 return extensions; 2333 } 2334 2335 /** 2336 * Obtains a localized locale names from a LocaleNameProvider 2337 * implementation. 2338 */ 2339 private static class LocaleNameGetter 2340 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> { 2341 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter(); 2342 2343 @Override 2344 public String getObject(LocaleNameProvider localeNameProvider, 2345 Locale locale, 2346 String key, 2347 Object... params) { 2348 assert params.length == 2; 2349 int type = (Integer)params[0]; 2350 String code = (String)params[1]; 2351 2352 switch(type) { 2353 case DISPLAY_LANGUAGE: 2354 return localeNameProvider.getDisplayLanguage(code, locale); 2355 case DISPLAY_COUNTRY: 2356 return localeNameProvider.getDisplayCountry(code, locale); 2357 case DISPLAY_VARIANT: 2358 return localeNameProvider.getDisplayVariant(code, locale); 2359 case DISPLAY_SCRIPT: 2360 return localeNameProvider.getDisplayScript(code, locale); 2361 default: 2362 assert false; // shouldn't happen 2363 } 2364 2365 return null; 2366 } 2367 } 2368 2369 /** 2370 * Enum for locale categories. These locale categories are used to get/set 2371 * the default locale for the specific functionality represented by the 2372 * category. 2373 * 2374 * @see #getDefault(Locale.Category) 2375 * @see #setDefault(Locale.Category, Locale) 2376 * @since 1.7 2377 */ 2378 public enum Category { 2379 2380 /** 2381 * Category used to represent the default locale for 2382 * displaying user interfaces. 2383 */ 2384 DISPLAY("user.language.display", 2385 "user.script.display", 2386 "user.country.display", 2387 "user.variant.display"), 2388 2389 /** 2390 * Category used to represent the default locale for 2391 * formatting dates, numbers, and/or currencies. 2392 */ 2393 FORMAT("user.language.format", 2394 "user.script.format", 2395 "user.country.format", 2396 "user.variant.format"); 2397 2398 Category(String languageKey, String scriptKey, String countryKey, String variantKey) { 2399 this.languageKey = languageKey; 2400 this.scriptKey = scriptKey; 2401 this.countryKey = countryKey; 2402 this.variantKey = variantKey; 2403 } 2404 2405 final String languageKey; 2406 final String scriptKey; 2407 final String countryKey; 2408 final String variantKey; 2409 } 2410 2411 /** 2412 * <code>Builder</code> is used to build instances of <code>Locale</code> 2413 * from values configured by the setters. Unlike the <code>Locale</code> 2414 * constructors, the <code>Builder</code> checks if a value configured by a 2415 * setter satisfies the syntax requirements defined by the <code>Locale</code> 2416 * class. A <code>Locale</code> object created by a <code>Builder</code> is 2417 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag 2418 * without losing information. 2419 * 2420 * <p><b>Note:</b> The <code>Locale</code> class does not provide any 2421 * syntactic restrictions on variant, while BCP 47 requires each variant 2422 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3 2423 * alphanumerics. The method <code>setVariant</code> throws 2424 * <code>IllformedLocaleException</code> for a variant that does not satisfy 2425 * this restriction. If it is necessary to support such a variant, use a 2426 * Locale constructor. However, keep in mind that a <code>Locale</code> 2427 * object created this way might lose the variant information when 2428 * transformed to a BCP 47 language tag. 2429 * 2430 * <p>The following example shows how to create a <code>Locale</code> object 2431 * with the <code>Builder</code>. 2432 * <blockquote> 2433 * <pre> 2434 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build(); 2435 * </pre> 2436 * </blockquote> 2437 * 2438 * <p>Builders can be reused; <code>clear()</code> resets all 2439 * fields to their default values. 2440 * 2441 * @see Locale#forLanguageTag 2442 * @since 1.7 2443 */ 2444 public static final class Builder { 2445 private final InternalLocaleBuilder localeBuilder; 2446 2447 /** 2448 * Constructs an empty Builder. The default value of all 2449 * fields, extensions, and private use information is the 2450 * empty string. 2451 */ 2452 public Builder() { 2453 localeBuilder = new InternalLocaleBuilder(); 2454 } 2455 2456 /** 2457 * Resets the <code>Builder</code> to match the provided 2458 * <code>locale</code>. Existing state is discarded. 2459 * 2460 * <p>All fields of the locale must be well-formed, see {@link Locale}. 2461 * 2462 * <p>Locales with any ill-formed fields cause 2463 * <code>IllformedLocaleException</code> to be thrown, except for the 2464 * following three cases which are accepted for compatibility 2465 * reasons:<ul> 2466 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese" 2467 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai" 2468 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul> 2469 * 2470 * @param locale the locale 2471 * @return This builder. 2472 * @throws IllformedLocaleException if <code>locale</code> has 2473 * any ill-formed fields. 2474 * @throws NullPointerException if <code>locale</code> is null. 2475 */ 2476 public Builder setLocale(Locale locale) { 2477 try { 2478 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions); 2479 } catch (LocaleSyntaxException e) { 2480 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2481 } 2482 return this; 2483 } 2484 2485 /** 2486 * Resets the Builder to match the provided IETF BCP 47 2487 * language tag. Discards the existing state. Null and the 2488 * empty string cause the builder to be reset, like {@link 2489 * #clear}. Grandfathered tags (see {@link 2490 * Locale#forLanguageTag}) are converted to their canonical 2491 * form before being processed. Otherwise, the language tag 2492 * must be well-formed (see {@link Locale}) or an exception is 2493 * thrown (unlike <code>Locale.forLanguageTag</code>, which 2494 * just discards ill-formed and following portions of the 2495 * tag). 2496 * 2497 * @param languageTag the language tag 2498 * @return This builder. 2499 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed 2500 * @see Locale#forLanguageTag(String) 2501 */ 2502 public Builder setLanguageTag(String languageTag) { 2503 ParseStatus sts = new ParseStatus(); 2504 LanguageTag tag = LanguageTag.parse(languageTag, sts); 2505 if (sts.isError()) { 2506 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex()); 2507 } 2508 localeBuilder.setLanguageTag(tag); 2509 return this; 2510 } 2511 2512 /** 2513 * Sets the language. If <code>language</code> is the empty string or 2514 * null, the language in this <code>Builder</code> is removed. Otherwise, 2515 * the language must be <a href="./Locale.html#def_language">well-formed</a> 2516 * or an exception is thrown. 2517 * 2518 * <p>The typical language value is a two or three-letter language 2519 * code as defined in ISO639. 2520 * 2521 * @param language the language 2522 * @return This builder. 2523 * @throws IllformedLocaleException if <code>language</code> is ill-formed 2524 */ 2525 public Builder setLanguage(String language) { 2526 try { 2527 localeBuilder.setLanguage(language); 2528 } catch (LocaleSyntaxException e) { 2529 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2530 } 2531 return this; 2532 } 2533 2534 /** 2535 * Sets the script. If <code>script</code> is null or the empty string, 2536 * the script in this <code>Builder</code> is removed. 2537 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an 2538 * exception is thrown. 2539 * 2540 * <p>The typical script value is a four-letter script code as defined by ISO 15924. 2541 * 2542 * @param script the script 2543 * @return This builder. 2544 * @throws IllformedLocaleException if <code>script</code> is ill-formed 2545 */ 2546 public Builder setScript(String script) { 2547 try { 2548 localeBuilder.setScript(script); 2549 } catch (LocaleSyntaxException e) { 2550 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2551 } 2552 return this; 2553 } 2554 2555 /** 2556 * Sets the region. If region is null or the empty string, the region 2557 * in this <code>Builder</code> is removed. Otherwise, 2558 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an 2559 * exception is thrown. 2560 * 2561 * <p>The typical region value is a two-letter ISO 3166 code or a 2562 * three-digit UN M.49 area code. 2563 * 2564 * <p>The country value in the <code>Locale</code> created by the 2565 * <code>Builder</code> is always normalized to upper case. 2566 * 2567 * @param region the region 2568 * @return This builder. 2569 * @throws IllformedLocaleException if <code>region</code> is ill-formed 2570 */ 2571 public Builder setRegion(String region) { 2572 try { 2573 localeBuilder.setRegion(region); 2574 } catch (LocaleSyntaxException e) { 2575 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2576 } 2577 return this; 2578 } 2579 2580 /** 2581 * Sets the variant. If variant is null or the empty string, the 2582 * variant in this <code>Builder</code> is removed. Otherwise, it 2583 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a> 2584 * subtags, or an exception is thrown. 2585 * 2586 * <p><b>Note:</b> This method checks if <code>variant</code> 2587 * satisfies the IETF BCP 47 variant subtag's syntax requirements, 2588 * and normalizes the value to lowercase letters. However, 2589 * the <code>Locale</code> class does not impose any syntactic 2590 * restriction on variant, and the variant value in 2591 * <code>Locale</code> is case sensitive. To set such a variant, 2592 * use a Locale constructor. 2593 * 2594 * @param variant the variant 2595 * @return This builder. 2596 * @throws IllformedLocaleException if <code>variant</code> is ill-formed 2597 */ 2598 public Builder setVariant(String variant) { 2599 try { 2600 localeBuilder.setVariant(variant); 2601 } catch (LocaleSyntaxException e) { 2602 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2603 } 2604 return this; 2605 } 2606 2607 /** 2608 * Sets the extension for the given key. If the value is null or the 2609 * empty string, the extension is removed. Otherwise, the extension 2610 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception 2611 * is thrown. 2612 * 2613 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION 2614 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension. 2615 * Setting a value for this key replaces any existing Unicode locale key/type 2616 * pairs with those defined in the extension. 2617 * 2618 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION 2619 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be 2620 * well-formed, the value for this key needs only to have subtags of one to 2621 * eight alphanumeric characters, not two to eight as in the general case. 2622 * 2623 * @param key the extension key 2624 * @param value the extension value 2625 * @return This builder. 2626 * @throws IllformedLocaleException if <code>key</code> is illegal 2627 * or <code>value</code> is ill-formed 2628 * @see #setUnicodeLocaleKeyword(String, String) 2629 */ 2630 public Builder setExtension(char key, String value) { 2631 try { 2632 localeBuilder.setExtension(key, value); 2633 } catch (LocaleSyntaxException e) { 2634 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2635 } 2636 return this; 2637 } 2638 2639 /** 2640 * Sets the Unicode locale keyword type for the given key. If the type 2641 * is null, the Unicode keyword is removed. Otherwise, the key must be 2642 * non-null and both key and type must be <a 2643 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2644 * is thrown. 2645 * 2646 * <p>Keys and types are converted to lower case. 2647 * 2648 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension} 2649 * replaces all Unicode locale keywords with those defined in the 2650 * extension. 2651 * 2652 * @param key the Unicode locale key 2653 * @param type the Unicode locale type 2654 * @return This builder. 2655 * @throws IllformedLocaleException if <code>key</code> or <code>type</code> 2656 * is ill-formed 2657 * @throws NullPointerException if <code>key</code> is null 2658 * @see #setExtension(char, String) 2659 */ 2660 public Builder setUnicodeLocaleKeyword(String key, String type) { 2661 try { 2662 localeBuilder.setUnicodeLocaleKeyword(key, type); 2663 } catch (LocaleSyntaxException e) { 2664 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2665 } 2666 return this; 2667 } 2668 2669 /** 2670 * Adds a unicode locale attribute, if not already present, otherwise 2671 * has no effect. The attribute must not be null and must be <a 2672 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2673 * is thrown. 2674 * 2675 * @param attribute the attribute 2676 * @return This builder. 2677 * @throws NullPointerException if <code>attribute</code> is null 2678 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed 2679 * @see #setExtension(char, String) 2680 */ 2681 public Builder addUnicodeLocaleAttribute(String attribute) { 2682 try { 2683 localeBuilder.addUnicodeLocaleAttribute(attribute); 2684 } catch (LocaleSyntaxException e) { 2685 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2686 } 2687 return this; 2688 } 2689 2690 /** 2691 * Removes a unicode locale attribute, if present, otherwise has no 2692 * effect. The attribute must not be null and must be <a 2693 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2694 * is thrown. 2695 * 2696 * <p>Attribute comparison for removal is case-insensitive. 2697 * 2698 * @param attribute the attribute 2699 * @return This builder. 2700 * @throws NullPointerException if <code>attribute</code> is null 2701 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed 2702 * @see #setExtension(char, String) 2703 */ 2704 public Builder removeUnicodeLocaleAttribute(String attribute) { 2705 try { 2706 localeBuilder.removeUnicodeLocaleAttribute(attribute); 2707 } catch (LocaleSyntaxException e) { 2708 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2709 } 2710 return this; 2711 } 2712 2713 /** 2714 * Resets the builder to its initial, empty state. 2715 * 2716 * @return This builder. 2717 */ 2718 public Builder clear() { 2719 localeBuilder.clear(); 2720 return this; 2721 } 2722 2723 /** 2724 * Resets the extensions to their initial, empty state. 2725 * Language, script, region and variant are unchanged. 2726 * 2727 * @return This builder. 2728 * @see #setExtension(char, String) 2729 */ 2730 public Builder clearExtensions() { 2731 localeBuilder.clearExtensions(); 2732 return this; 2733 } 2734 2735 /** 2736 * Returns an instance of <code>Locale</code> created from the fields set 2737 * on this builder. 2738 * 2739 * <p>This applies the conversions listed in {@link Locale#forLanguageTag} 2740 * when constructing a Locale. (Grandfathered tags are handled in 2741 * {@link #setLanguageTag}.) 2742 * 2743 * @return A Locale. 2744 */ 2745 public Locale build() { 2746 BaseLocale baseloc = localeBuilder.getBaseLocale(); 2747 LocaleExtensions extensions = localeBuilder.getLocaleExtensions(); 2748 if (extensions == null && baseloc.getVariant().length() > 0) { 2749 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(), 2750 baseloc.getRegion(), baseloc.getVariant()); 2751 } 2752 return Locale.getInstance(baseloc, extensions); 2753 } 2754 } 2755 2756 /** 2757 * This enum provides constants to select a filtering mode for locale 2758 * matching. Refer to <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 2759 * Matching of Language Tags</a> for details. 2760 * 2761 * <p>As an example, think of two Language Priority Lists each of which 2762 * includes only one language range and a set of following language tags: 2763 * 2764 * <pre> 2765 * de (German) 2766 * de-DE (German, Germany) 2767 * de-Deva (German, in Devanagari script) 2768 * de-Deva-DE (German, in Devanagari script, Germany) 2769 * de-DE-1996 (German, Germany, orthography of 1996) 2770 * de-Latn-DE (German, in Latin script, Germany) 2771 * de-Latn-DE-1996 (German, in Latin script, Germany, orthography of 1996) 2772 * </pre> 2773 * 2774 * The filtering method will behave as follows: 2775 * 2776 * <table class="altrows"> 2777 * <caption>Filtering method behavior</caption> 2778 * <thead> 2779 * <tr> 2780 * <th>Filtering Mode</th> 2781 * <th>Language Priority List: {@code "de-DE"}</th> 2782 * <th>Language Priority List: {@code "de-*-DE"}</th> 2783 * </tr> 2784 * </thead> 2785 * <tbody> 2786 * <tr> 2787 * <td style="vertical-align:top"> 2788 * {@link FilteringMode#AUTOSELECT_FILTERING AUTOSELECT_FILTERING} 2789 * </td> 2790 * <td style="vertical-align:top"> 2791 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and 2792 * {@code "de-DE-1996"}. 2793 * </td> 2794 * <td style="vertical-align:top"> 2795 * Performs <em>extended</em> filtering and returns {@code "de-DE"}, 2796 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and 2797 * {@code "de-Latn-DE-1996"}. 2798 * </td> 2799 * </tr> 2800 * <tr> 2801 * <td style="vertical-align:top"> 2802 * {@link FilteringMode#EXTENDED_FILTERING EXTENDED_FILTERING} 2803 * </td> 2804 * <td style="vertical-align:top"> 2805 * Performs <em>extended</em> filtering and returns {@code "de-DE"}, 2806 * {@code "de-Deva-DE"}, {@code "de-DE-1996"}, {@code "de-Latn-DE"}, and 2807 * {@code "de-Latn-DE-1996"}. 2808 * </td> 2809 * <td style="vertical-align:top">Same as above.</td> 2810 * </tr> 2811 * <tr> 2812 * <td style="vertical-align:top"> 2813 * {@link FilteringMode#IGNORE_EXTENDED_RANGES IGNORE_EXTENDED_RANGES} 2814 * </td> 2815 * <td style="vertical-align:top"> 2816 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and 2817 * {@code "de-DE-1996"}. 2818 * </td> 2819 * <td style="vertical-align:top"> 2820 * Performs <em>basic</em> filtering and returns {@code null} because 2821 * nothing matches. 2822 * </td> 2823 * </tr> 2824 * <tr> 2825 * <td style="vertical-align:top"> 2826 * {@link FilteringMode#MAP_EXTENDED_RANGES MAP_EXTENDED_RANGES} 2827 * </td> 2828 * <td style="vertical-align:top">Same as above.</td> 2829 * <td style="vertical-align:top"> 2830 * Performs <em>basic</em> filtering and returns {@code "de-DE"} and 2831 * {@code "de-DE-1996"} because {@code "de-*-DE"} is mapped to 2832 * {@code "de-DE"}. 2833 * </td> 2834 * </tr> 2835 * <tr> 2836 * <td style="vertical-align:top"> 2837 * {@link FilteringMode#REJECT_EXTENDED_RANGES REJECT_EXTENDED_RANGES} 2838 * </td> 2839 * <td style="vertical-align:top">Same as above.</td> 2840 * <td style="vertical-align:top"> 2841 * Throws {@link IllegalArgumentException} because {@code "de-*-DE"} is 2842 * not a valid basic language range. 2843 * </td> 2844 * </tr> 2845 * </tbody> 2846 * </table> 2847 * 2848 * @see #filter(List, Collection, FilteringMode) 2849 * @see #filterTags(List, Collection, FilteringMode) 2850 * 2851 * @since 1.8 2852 */ 2853 public static enum FilteringMode { 2854 /** 2855 * Specifies automatic filtering mode based on the given Language 2856 * Priority List consisting of language ranges. If all of the ranges 2857 * are basic, basic filtering is selected. Otherwise, extended 2858 * filtering is selected. 2859 */ 2860 AUTOSELECT_FILTERING, 2861 2862 /** 2863 * Specifies extended filtering. 2864 */ 2865 EXTENDED_FILTERING, 2866 2867 /** 2868 * Specifies basic filtering: Note that any extended language ranges 2869 * included in the given Language Priority List are ignored. 2870 */ 2871 IGNORE_EXTENDED_RANGES, 2872 2873 /** 2874 * Specifies basic filtering: If any extended language ranges are 2875 * included in the given Language Priority List, they are mapped to the 2876 * basic language range. Specifically, a language range starting with a 2877 * subtag {@code "*"} is treated as a language range {@code "*"}. For 2878 * example, {@code "*-US"} is treated as {@code "*"}. If {@code "*"} is 2879 * not the first subtag, {@code "*"} and extra {@code "-"} are removed. 2880 * For example, {@code "ja-*-JP"} is mapped to {@code "ja-JP"}. 2881 */ 2882 MAP_EXTENDED_RANGES, 2883 2884 /** 2885 * Specifies basic filtering: If any extended language ranges are 2886 * included in the given Language Priority List, the list is rejected 2887 * and the filtering method throws {@link IllegalArgumentException}. 2888 */ 2889 REJECT_EXTENDED_RANGES 2890 }; 2891 2892 /** 2893 * This class expresses a <em>Language Range</em> defined in 2894 * <a href="http://tools.ietf.org/html/rfc4647">RFC 4647 Matching of 2895 * Language Tags</a>. A language range is an identifier which is used to 2896 * select language tag(s) meeting specific requirements by using the 2897 * mechanisms described in <a href="Locale.html#LocaleMatching">Locale 2898 * Matching</a>. A list which represents a user's preferences and consists 2899 * of language ranges is called a <em>Language Priority List</em>. 2900 * 2901 * <p>There are two types of language ranges: basic and extended. In RFC 2902 * 4647, the syntax of language ranges is expressed in 2903 * <a href="http://tools.ietf.org/html/rfc4234">ABNF</a> as follows: 2904 * <blockquote> 2905 * <pre> 2906 * basic-language-range = (1*8ALPHA *("-" 1*8alphanum)) / "*" 2907 * extended-language-range = (1*8ALPHA / "*") 2908 * *("-" (1*8alphanum / "*")) 2909 * alphanum = ALPHA / DIGIT 2910 * </pre> 2911 * </blockquote> 2912 * For example, {@code "en"} (English), {@code "ja-JP"} (Japanese, Japan), 2913 * {@code "*"} (special language range which matches any language tag) are 2914 * basic language ranges, whereas {@code "*-CH"} (any languages, 2915 * Switzerland), {@code "es-*"} (Spanish, any regions), and 2916 * {@code "zh-Hant-*"} (Traditional Chinese, any regions) are extended 2917 * language ranges. 2918 * 2919 * @see #filter 2920 * @see #filterTags 2921 * @see #lookup 2922 * @see #lookupTag 2923 * 2924 * @since 1.8 2925 */ 2926 public static final class LanguageRange { 2927 2928 /** 2929 * A constant holding the maximum value of weight, 1.0, which indicates 2930 * that the language range is a good fit for the user. 2931 */ 2932 public static final double MAX_WEIGHT = 1.0; 2933 2934 /** 2935 * A constant holding the minimum value of weight, 0.0, which indicates 2936 * that the language range is not a good fit for the user. 2937 */ 2938 public static final double MIN_WEIGHT = 0.0; 2939 2940 private final String range; 2941 private final double weight; 2942 2943 private volatile int hash; 2944 2945 /** 2946 * Constructs a {@code LanguageRange} using the given {@code range}. 2947 * Note that no validation is done against the IANA Language Subtag 2948 * Registry at time of construction. 2949 * 2950 * <p>This is equivalent to {@code LanguageRange(range, MAX_WEIGHT)}. 2951 * 2952 * @param range a language range 2953 * @throws NullPointerException if the given {@code range} is 2954 * {@code null} 2955 * @throws IllegalArgumentException if the given {@code range} does not 2956 * comply with the syntax of the language range mentioned in RFC 4647 2957 */ 2958 public LanguageRange(String range) { 2959 this(range, MAX_WEIGHT); 2960 } 2961 2962 /** 2963 * Constructs a {@code LanguageRange} using the given {@code range} and 2964 * {@code weight}. Note that no validation is done against the IANA 2965 * Language Subtag Registry at time of construction. 2966 * 2967 * @param range a language range 2968 * @param weight a weight value between {@code MIN_WEIGHT} and 2969 * {@code MAX_WEIGHT} 2970 * @throws NullPointerException if the given {@code range} is 2971 * {@code null} 2972 * @throws IllegalArgumentException if the given {@code range} does not 2973 * comply with the syntax of the language range mentioned in RFC 4647 2974 * or if the given {@code weight} is less than {@code MIN_WEIGHT} 2975 * or greater than {@code MAX_WEIGHT} 2976 */ 2977 public LanguageRange(String range, double weight) { 2978 if (range == null) { 2979 throw new NullPointerException(); 2980 } 2981 if (weight < MIN_WEIGHT || weight > MAX_WEIGHT) { 2982 throw new IllegalArgumentException("weight=" + weight); 2983 } 2984 2985 range = range.toLowerCase(Locale.ROOT); 2986 2987 // Do syntax check. 2988 boolean isIllFormed = false; 2989 String[] subtags = range.split("-"); 2990 if (isSubtagIllFormed(subtags[0], true) 2991 || range.endsWith("-")) { 2992 isIllFormed = true; 2993 } else { 2994 for (int i = 1; i < subtags.length; i++) { 2995 if (isSubtagIllFormed(subtags[i], false)) { 2996 isIllFormed = true; 2997 break; 2998 } 2999 } 3000 } 3001 if (isIllFormed) { 3002 throw new IllegalArgumentException("range=" + range); 3003 } 3004 3005 this.range = range; 3006 this.weight = weight; 3007 } 3008 3009 private static boolean isSubtagIllFormed(String subtag, 3010 boolean isFirstSubtag) { 3011 if (subtag.equals("") || subtag.length() > 8) { 3012 return true; 3013 } else if (subtag.equals("*")) { 3014 return false; 3015 } 3016 char[] charArray = subtag.toCharArray(); 3017 if (isFirstSubtag) { // ALPHA 3018 for (char c : charArray) { 3019 if (c < 'a' || c > 'z') { 3020 return true; 3021 } 3022 } 3023 } else { // ALPHA / DIGIT 3024 for (char c : charArray) { 3025 if (c < '0' || (c > '9' && c < 'a') || c > 'z') { 3026 return true; 3027 } 3028 } 3029 } 3030 return false; 3031 } 3032 3033 /** 3034 * Returns the language range of this {@code LanguageRange}. 3035 * 3036 * @return the language range. 3037 */ 3038 public String getRange() { 3039 return range; 3040 } 3041 3042 /** 3043 * Returns the weight of this {@code LanguageRange}. 3044 * 3045 * @return the weight value. 3046 */ 3047 public double getWeight() { 3048 return weight; 3049 } 3050 3051 /** 3052 * Parses the given {@code ranges} to generate a Language Priority List. 3053 * 3054 * <p>This method performs a syntactic check for each language range in 3055 * the given {@code ranges} but doesn't do validation using the IANA 3056 * Language Subtag Registry. 3057 * 3058 * <p>The {@code ranges} to be given can take one of the following 3059 * forms: 3060 * 3061 * <pre> 3062 * "Accept-Language: ja,en;q=0.4" (weighted list with Accept-Language prefix) 3063 * "ja,en;q=0.4" (weighted list) 3064 * "ja,en" (prioritized list) 3065 * </pre> 3066 * 3067 * In a weighted list, each language range is given a weight value. 3068 * The weight value is identical to the "quality value" in 3069 * <a href="http://tools.ietf.org/html/rfc2616">RFC 2616</a>, and it 3070 * expresses how much the user prefers the language. A weight value is 3071 * specified after a corresponding language range followed by 3072 * {@code ";q="}, and the default weight value is {@code MAX_WEIGHT} 3073 * when it is omitted. 3074 * 3075 * <p>Unlike a weighted list, language ranges in a prioritized list 3076 * are sorted in the descending order based on its priority. The first 3077 * language range has the highest priority and meets the user's 3078 * preference most. 3079 * 3080 * <p>In either case, language ranges are sorted in descending order in 3081 * the Language Priority List based on priority or weight. If a 3082 * language range appears in the given {@code ranges} more than once, 3083 * only the first one is included on the Language Priority List. 3084 * 3085 * <p>The returned list consists of language ranges from the given 3086 * {@code ranges} and their equivalents found in the IANA Language 3087 * Subtag Registry. For example, if the given {@code ranges} is 3088 * {@code "Accept-Language: iw,en-us;q=0.7,en;q=0.3"}, the elements in 3089 * the list to be returned are: 3090 * 3091 * <pre> 3092 * <b>Range</b> <b>Weight</b> 3093 * "iw" (older tag for Hebrew) 1.0 3094 * "he" (new preferred code for Hebrew) 1.0 3095 * "en-us" (English, United States) 0.7 3096 * "en" (English) 0.3 3097 * </pre> 3098 * 3099 * Two language ranges, {@code "iw"} and {@code "he"}, have the same 3100 * highest priority in the list. By adding {@code "he"} to the user's 3101 * Language Priority List, locale-matching method can find Hebrew as a 3102 * matching locale (or language tag) even if the application or system 3103 * offers only {@code "he"} as a supported locale (or language tag). 3104 * 3105 * @param ranges a list of comma-separated language ranges or a list of 3106 * language ranges in the form of the "Accept-Language" header 3107 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC 3108 * 2616</a> 3109 * @return a Language Priority List consisting of language ranges 3110 * included in the given {@code ranges} and their equivalent 3111 * language ranges if available. The list is modifiable. 3112 * @throws NullPointerException if {@code ranges} is null 3113 * @throws IllegalArgumentException if a language range or a weight 3114 * found in the given {@code ranges} is ill-formed 3115 */ 3116 public static List<LanguageRange> parse(String ranges) { 3117 return LocaleMatcher.parse(ranges); 3118 } 3119 3120 /** 3121 * Parses the given {@code ranges} to generate a Language Priority 3122 * List, and then customizes the list using the given {@code map}. 3123 * This method is equivalent to 3124 * {@code mapEquivalents(parse(ranges), map)}. 3125 * 3126 * @param ranges a list of comma-separated language ranges or a list 3127 * of language ranges in the form of the "Accept-Language" header 3128 * defined in <a href="http://tools.ietf.org/html/rfc2616">RFC 3129 * 2616</a> 3130 * @param map a map containing information to customize language ranges 3131 * @return a Language Priority List with customization. The list is 3132 * modifiable. 3133 * @throws NullPointerException if {@code ranges} is null 3134 * @throws IllegalArgumentException if a language range or a weight 3135 * found in the given {@code ranges} is ill-formed 3136 * @see #parse(String) 3137 * @see #mapEquivalents 3138 */ 3139 public static List<LanguageRange> parse(String ranges, 3140 Map<String, List<String>> map) { 3141 return mapEquivalents(parse(ranges), map); 3142 } 3143 3144 /** 3145 * Generates a new customized Language Priority List using the given 3146 * {@code priorityList} and {@code map}. If the given {@code map} is 3147 * empty, this method returns a copy of the given {@code priorityList}. 3148 * 3149 * <p>In the map, a key represents a language range whereas a value is 3150 * a list of equivalents of it. {@code '*'} cannot be used in the map. 3151 * Each equivalent language range has the same weight value as its 3152 * original language range. 3153 * 3154 * <pre> 3155 * An example of map: 3156 * <b>Key</b> <b>Value</b> 3157 * "zh" (Chinese) "zh", 3158 * "zh-Hans"(Simplified Chinese) 3159 * "zh-HK" (Chinese, Hong Kong) "zh-HK" 3160 * "zh-TW" (Chinese, Taiwan) "zh-TW" 3161 * </pre> 3162 * 3163 * The customization is performed after modification using the IANA 3164 * Language Subtag Registry. 3165 * 3166 * <p>For example, if a user's Language Priority List consists of five 3167 * language ranges ({@code "zh"}, {@code "zh-CN"}, {@code "en"}, 3168 * {@code "zh-TW"}, and {@code "zh-HK"}), the newly generated Language 3169 * Priority List which is customized using the above map example will 3170 * consists of {@code "zh"}, {@code "zh-Hans"}, {@code "zh-CN"}, 3171 * {@code "zh-Hans-CN"}, {@code "en"}, {@code "zh-TW"}, and 3172 * {@code "zh-HK"}. 3173 * 3174 * <p>{@code "zh-HK"} and {@code "zh-TW"} aren't converted to 3175 * {@code "zh-Hans-HK"} nor {@code "zh-Hans-TW"} even if they are 3176 * included in the Language Priority List. In this example, mapping 3177 * is used to clearly distinguish Simplified Chinese and Traditional 3178 * Chinese. 3179 * 3180 * <p>If the {@code "zh"}-to-{@code "zh"} mapping isn't included in the 3181 * map, a simple replacement will be performed and the customized list 3182 * won't include {@code "zh"} and {@code "zh-CN"}. 3183 * 3184 * @param priorityList user's Language Priority List 3185 * @param map a map containing information to customize language ranges 3186 * @return a new Language Priority List with customization. The list is 3187 * modifiable. 3188 * @throws NullPointerException if {@code priorityList} is {@code null} 3189 * @see #parse(String, Map) 3190 */ 3191 public static List<LanguageRange> mapEquivalents( 3192 List<LanguageRange>priorityList, 3193 Map<String, List<String>> map) { 3194 return LocaleMatcher.mapEquivalents(priorityList, map); 3195 } 3196 3197 /** 3198 * Returns a hash code value for the object. 3199 * 3200 * @return a hash code value for this object. 3201 */ 3202 @Override 3203 public int hashCode() { 3204 int h = hash; 3205 if (h == 0) { 3206 h = 17; 3207 h = 37*h + range.hashCode(); 3208 long bitsWeight = Double.doubleToLongBits(weight); 3209 h = 37*h + (int)(bitsWeight ^ (bitsWeight >>> 32)); 3210 if (h != 0) { 3211 hash = h; 3212 } 3213 } 3214 return h; 3215 } 3216 3217 /** 3218 * Compares this object to the specified object. The result is true if 3219 * and only if the argument is not {@code null} and is a 3220 * {@code LanguageRange} object that contains the same {@code range} 3221 * and {@code weight} values as this object. 3222 * 3223 * @param obj the object to compare with 3224 * @return {@code true} if this object's {@code range} and 3225 * {@code weight} are the same as the {@code obj}'s; {@code false} 3226 * otherwise. 3227 */ 3228 @Override 3229 public boolean equals(Object obj) { 3230 if (this == obj) { 3231 return true; 3232 } 3233 if (!(obj instanceof LanguageRange)) { 3234 return false; 3235 } 3236 LanguageRange other = (LanguageRange)obj; 3237 return hash == other.hash 3238 && range.equals(other.range) 3239 && weight == other.weight; 3240 } 3241 3242 /** 3243 * Returns an informative string representation of this {@code LanguageRange} 3244 * object, consisting of language range and weight if the range is 3245 * weighted and the weight is less than the max weight. 3246 * 3247 * @return a string representation of this {@code LanguageRange} object. 3248 */ 3249 @Override 3250 public String toString() { 3251 return (weight == MAX_WEIGHT) ? range : range + ";q=" + weight; 3252 } 3253 } 3254 3255 /** 3256 * Returns a list of matching {@code Locale} instances using the filtering 3257 * mechanism defined in RFC 4647. 3258 * 3259 * @param priorityList user's Language Priority List in which each language 3260 * tag is sorted in descending order based on priority or weight 3261 * @param locales {@code Locale} instances used for matching 3262 * @param mode filtering mode 3263 * @return a list of {@code Locale} instances for matching language tags 3264 * sorted in descending order based on priority or weight, or an empty 3265 * list if nothing matches. The list is modifiable. 3266 * @throws NullPointerException if {@code priorityList} or {@code locales} 3267 * is {@code null} 3268 * @throws IllegalArgumentException if one or more extended language ranges 3269 * are included in the given list when 3270 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified 3271 * 3272 * @since 1.8 3273 */ 3274 public static List<Locale> filter(List<LanguageRange> priorityList, 3275 Collection<Locale> locales, 3276 FilteringMode mode) { 3277 return LocaleMatcher.filter(priorityList, locales, mode); 3278 } 3279 3280 /** 3281 * Returns a list of matching {@code Locale} instances using the filtering 3282 * mechanism defined in RFC 4647. This is equivalent to 3283 * {@link #filter(List, Collection, FilteringMode)} when {@code mode} is 3284 * {@link FilteringMode#AUTOSELECT_FILTERING}. 3285 * 3286 * @param priorityList user's Language Priority List in which each language 3287 * tag is sorted in descending order based on priority or weight 3288 * @param locales {@code Locale} instances used for matching 3289 * @return a list of {@code Locale} instances for matching language tags 3290 * sorted in descending order based on priority or weight, or an empty 3291 * list if nothing matches. The list is modifiable. 3292 * @throws NullPointerException if {@code priorityList} or {@code locales} 3293 * is {@code null} 3294 * 3295 * @since 1.8 3296 */ 3297 public static List<Locale> filter(List<LanguageRange> priorityList, 3298 Collection<Locale> locales) { 3299 return filter(priorityList, locales, FilteringMode.AUTOSELECT_FILTERING); 3300 } 3301 3302 /** 3303 * Returns a list of matching languages tags using the basic filtering 3304 * mechanism defined in RFC 4647. 3305 * 3306 * @param priorityList user's Language Priority List in which each language 3307 * tag is sorted in descending order based on priority or weight 3308 * @param tags language tags 3309 * @param mode filtering mode 3310 * @return a list of matching language tags sorted in descending order 3311 * based on priority or weight, or an empty list if nothing matches. 3312 * The list is modifiable. 3313 * @throws NullPointerException if {@code priorityList} or {@code tags} is 3314 * {@code null} 3315 * @throws IllegalArgumentException if one or more extended language ranges 3316 * are included in the given list when 3317 * {@link FilteringMode#REJECT_EXTENDED_RANGES} is specified 3318 * 3319 * @since 1.8 3320 */ 3321 public static List<String> filterTags(List<LanguageRange> priorityList, 3322 Collection<String> tags, 3323 FilteringMode mode) { 3324 return LocaleMatcher.filterTags(priorityList, tags, mode); 3325 } 3326 3327 /** 3328 * Returns a list of matching languages tags using the basic filtering 3329 * mechanism defined in RFC 4647. This is equivalent to 3330 * {@link #filterTags(List, Collection, FilteringMode)} when {@code mode} 3331 * is {@link FilteringMode#AUTOSELECT_FILTERING}. 3332 * 3333 * @param priorityList user's Language Priority List in which each language 3334 * tag is sorted in descending order based on priority or weight 3335 * @param tags language tags 3336 * @return a list of matching language tags sorted in descending order 3337 * based on priority or weight, or an empty list if nothing matches. 3338 * The list is modifiable. 3339 * @throws NullPointerException if {@code priorityList} or {@code tags} is 3340 * {@code null} 3341 * 3342 * @since 1.8 3343 */ 3344 public static List<String> filterTags(List<LanguageRange> priorityList, 3345 Collection<String> tags) { 3346 return filterTags(priorityList, tags, FilteringMode.AUTOSELECT_FILTERING); 3347 } 3348 3349 /** 3350 * Returns a {@code Locale} instance for the best-matching language 3351 * tag using the lookup mechanism defined in RFC 4647. 3352 * 3353 * @param priorityList user's Language Priority List in which each language 3354 * tag is sorted in descending order based on priority or weight 3355 * @param locales {@code Locale} instances used for matching 3356 * @return the best matching <code>Locale</code> instance chosen based on 3357 * priority or weight, or {@code null} if nothing matches. 3358 * @throws NullPointerException if {@code priorityList} or {@code tags} is 3359 * {@code null} 3360 * 3361 * @since 1.8 3362 */ 3363 public static Locale lookup(List<LanguageRange> priorityList, 3364 Collection<Locale> locales) { 3365 return LocaleMatcher.lookup(priorityList, locales); 3366 } 3367 3368 /** 3369 * Returns the best-matching language tag using the lookup mechanism 3370 * defined in RFC 4647. 3371 * 3372 * @param priorityList user's Language Priority List in which each language 3373 * tag is sorted in descending order based on priority or weight 3374 * @param tags language tangs used for matching 3375 * @return the best matching language tag chosen based on priority or 3376 * weight, or {@code null} if nothing matches. 3377 * @throws NullPointerException if {@code priorityList} or {@code tags} is 3378 * {@code null} 3379 * 3380 * @since 1.8 3381 */ 3382 public static String lookupTag(List<LanguageRange> priorityList, 3383 Collection<String> tags) { 3384 return LocaleMatcher.lookupTag(priorityList, tags); 3385 } 3386 3387 }