Locale.java revision 983b2c6ff9ea6d35adf7ab6398dccf870b7e180a
1/* 2 * Copyright (c) 1996, 2011, 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 41package java.util; 42 43import java.io.IOException; 44import java.io.ObjectInputStream; 45import java.io.ObjectOutputStream; 46import java.io.ObjectStreamField; 47import java.io.Serializable; 48import java.security.AccessController; 49import java.text.MessageFormat; 50import java.util.spi.LocaleNameProvider; 51import libcore.icu.ICU; 52 53import sun.security.action.GetPropertyAction; 54import sun.util.LocaleServiceProviderPool; 55import sun.util.locale.BaseLocale; 56import sun.util.locale.InternalLocaleBuilder; 57import sun.util.locale.LanguageTag; 58import sun.util.locale.LocaleExtensions; 59import sun.util.locale.LocaleObjectCache; 60import sun.util.locale.LocaleSyntaxException; 61import sun.util.locale.LocaleUtils; 62import sun.util.locale.ParseStatus; 63import sun.util.locale.UnicodeLocaleExtension; 64 65/** 66 * A <code>Locale</code> object represents a specific geographical, political, 67 * or cultural region. An operation that requires a <code>Locale</code> to perform 68 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code> 69 * to tailor information for the user. For example, displaying a number 70 * is a locale-sensitive operation— the number should be formatted 71 * according to the customs and conventions of the user's native country, 72 * region, or culture. 73 * 74 * <p> The <code>Locale</code> class implements identifiers 75 * interchangeable with BCP 47 (IETF BCP 47, "Tags for Identifying 76 * Languages"), with support for the LDML (UTS#35, "Unicode Locale 77 * Data Markup Language") BCP 47-compatible extensions for locale data 78 * exchange. 79 * 80 * <p> A <code>Locale</code> object logically consists of the fields 81 * described below. 82 * 83 * <dl> 84 * <dt><a name="def_language"/><b>language</b></dt> 85 * 86 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered 87 * language subtags up to 8 alpha letters (for future enhancements). 88 * When a language has both an alpha-2 code and an alpha-3 code, the 89 * alpha-2 code must be used. You can find a full list of valid 90 * language codes in the IANA Language Subtag Registry (search for 91 * "Type: language"). The language field is case insensitive, but 92 * <code>Locale</code> always canonicalizes to lower case.</dd><br> 93 * 94 * <dd>Well-formed language values have the form 95 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the the full 96 * BCP47 language production, since it excludes extlang. They are 97 * not needed since modern three-letter language codes replace 98 * them.</dd><br> 99 * 100 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd><br> 101 * 102 * <dt><a name="def_script"/><b>script</b></dt> 103 * 104 * <dd>ISO 15924 alpha-4 script code. You can find a full list of 105 * valid script codes in the IANA Language Subtag Registry (search 106 * for "Type: script"). The script field is case insensitive, but 107 * <code>Locale</code> always canonicalizes to title case (the first 108 * letter is upper case and the rest of the letters are lower 109 * case).</dd><br> 110 * 111 * <dd>Well-formed script values have the form 112 * <code>[a-zA-Z]{4}</code></dd><br> 113 * 114 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd><br> 115 * 116 * <dt><a name="def_region"/><b>country (region)</b></dt> 117 * 118 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code. 119 * You can find a full list of valid country and region codes in the 120 * IANA Language Subtag Registry (search for "Type: region"). The 121 * country (region) field is case insensitive, but 122 * <code>Locale</code> always canonicalizes to upper case.</dd><br> 123 * 124 * <dd>Well-formed country/region values have 125 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd><br> 126 * 127 * <dd>Example: "US" (United States), "FR" (France), "029" 128 * (Caribbean)</dd><br> 129 * 130 * <dt><a name="def_variant"/><b>variant</b></dt> 131 * 132 * <dd>Any arbitrary value used to indicate a variation of a 133 * <code>Locale</code>. Where there are two or more variant values 134 * each indicating its own semantics, these values should be ordered 135 * by importance, with most important first, separated by 136 * underscore('_'). The variant field is case sensitive.</dd><br> 137 * 138 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant 139 * subtags. Also BCP 47 subtags are strictly used to indicate 140 * additional variations that define a language or its dialects that 141 * are not covered by any combinations of language, script and 142 * region subtags. You can find a full list of valid variant codes 143 * in the IANA Language Subtag Registry (search for "Type: variant"). 144 * 145 * <p>However, the variant field in <code>Locale</code> has 146 * historically been used for any kind of variation, not just 147 * language variations. For example, some supported variants 148 * available in Java SE Runtime Environments indicate alternative 149 * cultural behaviors such as calendar type or number script. In 150 * BCP 47 this kind of information, which does not identify the 151 * language, is supported by extension subtags or private use 152 * subtags.</dd><br> 153 * 154 * <dd>Well-formed variant values have the form <code>SUBTAG 155 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG = 156 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only 157 * uses hyphen ('-') as a delimiter, this is more lenient).</dd><br> 158 * 159 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd><br> 160 * 161 * <dt><a name="def_extensions"/><b>extensions</b></dt> 162 * 163 * <dd>A map from single character keys to string values, indicating 164 * extensions apart from language identification. The extensions in 165 * <code>Locale</code> implement the semantics and syntax of BCP 47 166 * extension subtags and private use subtags. The extensions are 167 * case insensitive, but <code>Locale</code> canonicalizes all 168 * extension keys and values to lower case. Note that extensions 169 * cannot have empty values.</dd><br> 170 * 171 * <dd>Well-formed keys are single characters from the set 172 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form 173 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x' 174 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys 175 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows 176 * single-character subtags).</dd><br> 177 * 178 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar), 179 * key="x"/value="java-1-7"</dd> 180 * </dl> 181 * 182 * <b>Note:</b> Although BCP 47 requires field values to be registered 183 * in the IANA Language Subtag Registry, the <code>Locale</code> class 184 * does not provide any validation features. The <code>Builder</code> 185 * only checks if an individual field satisfies the syntactic 186 * requirement (is well-formed), but does not validate the value 187 * itself. See {@link Builder} for details. 188 * 189 * <h4><a name="def_locale_extension">Unicode locale/language extension</h4> 190 * 191 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional 192 * attributes and keywords to override or refine the default behavior 193 * associated with a locale. A keyword is represented by a pair of 194 * key and type. For example, "nu-thai" indicates that Thai local 195 * digits (value:"thai") should be used for formatting numbers 196 * (key:"nu"). 197 * 198 * <p>The keywords are mapped to a BCP 47 extension value using the 199 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above 200 * example, "nu-thai", becomes the extension "u-nu-thai".code 201 * 202 * <p>Thus, when a <code>Locale</code> object contains Unicode locale 203 * attributes and keywords, 204 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a 205 * String representing this information, for example, "nu-thai". The 206 * <code>Locale</code> class also provides {@link 207 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and 208 * {@link #getUnicodeLocaleType} which allow you to access Unicode 209 * locale attributes and key/type pairs directly. When represented as 210 * a string, the Unicode Locale Extension lists attributes 211 * alphabetically, followed by key/type sequences with keys listed 212 * alphabetically (the order of subtags comprising a key's type is 213 * fixed when the type is defined) 214 * 215 * <p>A well-formed locale key has the form 216 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the 217 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it 218 * can be empty, or a series of subtags 3-8 alphanums in length). A 219 * well-formed locale attribute has the form 220 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same 221 * form as a locale type subtag). 222 * 223 * <p>The Unicode locale extension specifies optional behavior in 224 * locale-sensitive services. Although the LDML specification defines 225 * various keys and values, actual locale-sensitive service 226 * implementations in a Java Runtime Environment might not support any 227 * particular Unicode locale attributes or key/type pairs. 228 * 229 * <h4>Creating a Locale</h4> 230 * 231 * <p>There are several different ways to create a <code>Locale</code> 232 * object. 233 * 234 * <h5>Builder</h5> 235 * 236 * <p>Using {@link Builder} you can construct a <code>Locale</code> object 237 * that conforms to BCP 47 syntax. 238 * 239 * <h5>Constructors</h5> 240 * 241 * <p>The <code>Locale</code> class provides three constructors: 242 * <blockquote> 243 * <pre> 244 * {@link #Locale(String language)} 245 * {@link #Locale(String language, String country)} 246 * {@link #Locale(String language, String country, String variant)} 247 * </pre> 248 * </blockquote> 249 * These constructors allow you to create a <code>Locale</code> object 250 * with language, country and variant, but you cannot specify 251 * script or extensions. 252 * 253 * <h5>Factory Methods</h5> 254 * 255 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code> 256 * object for a well-formed BCP 47 language tag. 257 * 258 * <h5>Locale Constants</h5> 259 * 260 * <p>The <code>Locale</code> class provides a number of convenient constants 261 * that you can use to create <code>Locale</code> objects for commonly used 262 * locales. For example, the following creates a <code>Locale</code> object 263 * for the United States: 264 * <blockquote> 265 * <pre> 266 * Locale.US 267 * </pre> 268 * </blockquote> 269 * 270 * <h4>Use of Locale</h4> 271 * 272 * <p>Once you've created a <code>Locale</code> you can query it for information 273 * about itself. Use <code>getCountry</code> to get the country (or region) 274 * code and <code>getLanguage</code> to get the language code. 275 * You can use <code>getDisplayCountry</code> to get the 276 * name of the country suitable for displaying to the user. Similarly, 277 * you can use <code>getDisplayLanguage</code> to get the name of 278 * the language suitable for displaying to the user. Interestingly, 279 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive 280 * and have two versions: one that uses the default locale and one 281 * that uses the locale specified as an argument. 282 * 283 * <p>The Java Platform provides a number of classes that perform locale-sensitive 284 * operations. For example, the <code>NumberFormat</code> class formats 285 * numbers, currency, and percentages in a locale-sensitive manner. Classes 286 * such as <code>NumberFormat</code> have several convenience methods 287 * for creating a default object of that type. For example, the 288 * <code>NumberFormat</code> class provides these three convenience methods 289 * for creating a default <code>NumberFormat</code> object: 290 * <blockquote> 291 * <pre> 292 * NumberFormat.getInstance() 293 * NumberFormat.getCurrencyInstance() 294 * NumberFormat.getPercentInstance() 295 * </pre> 296 * </blockquote> 297 * Each of these methods has two variants; one with an explicit locale 298 * and one without; the latter uses the default locale: 299 * <blockquote> 300 * <pre> 301 * NumberFormat.getInstance(myLocale) 302 * NumberFormat.getCurrencyInstance(myLocale) 303 * NumberFormat.getPercentInstance(myLocale) 304 * </pre> 305 * </blockquote> 306 * A <code>Locale</code> is the mechanism for identifying the kind of object 307 * (<code>NumberFormat</code>) that you would like to get. The locale is 308 * <STRONG>just</STRONG> a mechanism for identifying objects, 309 * <STRONG>not</STRONG> a container for the objects themselves. 310 * 311 * <h4>Compatibility</h4> 312 * 313 * <p>In order to maintain compatibility with existing usage, Locale's 314 * constructors retain their behavior prior to the Java Runtime 315 * Environment version 1.7. The same is largely true for the 316 * <code>toString</code> method. Thus Locale objects can continue to 317 * be used as they were. In particular, clients who parse the output 318 * of toString into language, country, and variant fields can continue 319 * to do so (although this is strongly discouraged), although the 320 * variant field will have additional information in it if script or 321 * extensions are present. 322 * 323 * <p>In addition, BCP 47 imposes syntax restrictions that are not 324 * imposed by Locale's constructors. This means that conversions 325 * between some Locales and BCP 47 language tags cannot be made without 326 * losing information. Thus <code>toLanguageTag</code> cannot 327 * represent the state of locales whose language, country, or variant 328 * do not conform to BCP 47. 329 * 330 * <p>Because of these issues, it is recommended that clients migrate 331 * away from constructing non-conforming locales and use the 332 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead. 333 * Clients desiring a string representation of the complete locale can 334 * then always rely on <code>toLanguageTag</code> for this purpose. 335 * 336 * <h5><a name="special_cases_constructor"/>Special cases</h5> 337 * 338 * <p>For compatibility reasons, two 339 * non-conforming locales are treated as special cases. These are 340 * <b><tt>ja_JP_JP</tt></b> and <b><tt>th_TH_TH</tt></b>. These are ill-formed 341 * in BCP 47 since the variants are too short. To ease migration to BCP 47, 342 * these are treated specially during construction. These two cases (and only 343 * these) cause a constructor to generate an extension, all other values behave 344 * exactly as they did prior to Java 7. 345 * 346 * <p>Java has used <tt>ja_JP_JP</tt> to represent Japanese as used in 347 * Japan together with the Japanese Imperial calendar. This is now 348 * representable using a Unicode locale extension, by specifying the 349 * Unicode locale key <tt>ca</tt> (for "calendar") and type 350 * <tt>japanese</tt>. When the Locale constructor is called with the 351 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is 352 * automatically added. 353 * 354 * <p>Java has used <tt>th_TH_TH</tt> to represent Thai as used in 355 * Thailand together with Thai digits. This is also now representable using 356 * a Unicode locale extension, by specifying the Unicode locale key 357 * <tt>nu</tt> (for "number") and value <tt>thai</tt>. When the Locale 358 * constructor is called with the arguments "th", "TH", "TH", the 359 * extension "u-nu-thai" is automatically added. 360 * 361 * <h5>Serialization</h5> 362 * 363 * <p>During serialization, writeObject writes all fields to the output 364 * stream, including extensions. 365 * 366 * <p>During deserialization, readResolve adds extensions as described 367 * in <a href="#special_cases_constructor">Special Cases</a>, only 368 * for the two cases th_TH_TH and ja_JP_JP. 369 * 370 * <h5>Legacy language codes</h5> 371 * 372 * <p>Locale's constructor has always converted three language codes to 373 * their earlier, obsoleted forms: <tt>he</tt> maps to <tt>iw</tt>, 374 * <tt>yi</tt> maps to <tt>ji</tt>, and <tt>id</tt> maps to 375 * <tt>in</tt>. This continues to be the case, in order to not break 376 * backwards compatibility. 377 * 378 * <p>The APIs added in 1.7 map between the old and new language codes, 379 * maintaining the old codes internal to Locale (so that 380 * <code>getLanguage</code> and <code>toString</code> reflect the old 381 * code), but using the new codes in the BCP 47 language tag APIs (so 382 * that <code>toLanguageTag</code> reflects the new one). This 383 * preserves the equivalence between Locales no matter which code or 384 * API is used to construct them. Java's default resource bundle 385 * lookup mechanism also implements this mapping, so that resources 386 * can be named using either convention, see {@link ResourceBundle.Control}. 387 * 388 * <h5>Three-letter language/country(region) codes</h5> 389 * 390 * <p>The Locale constructors have always specified that the language 391 * and the country param be two characters in length, although in 392 * practice they have accepted any length. The specification has now 393 * been relaxed to allow language codes of two to eight characters and 394 * country (region) codes of two to three characters, and in 395 * particular, three-letter language codes and three-digit region 396 * codes as specified in the IANA Language Subtag Registry. For 397 * compatibility, the implementation still does not impose a length 398 * constraint. 399 * 400 * @see Builder 401 * @see ResourceBundle 402 * @see java.text.Format 403 * @see java.text.NumberFormat 404 * @see java.text.Collator 405 * @author Mark Davis 406 * @since 1.1 407 */ 408public final class Locale implements Cloneable, Serializable { 409 410 static private final Cache LOCALECACHE = new Cache(); 411 412 /** Useful constant for language. 413 */ 414 static public final Locale ENGLISH = createConstant("en", ""); 415 416 /** Useful constant for language. 417 */ 418 static public final Locale FRENCH = createConstant("fr", ""); 419 420 /** Useful constant for language. 421 */ 422 static public final Locale GERMAN = createConstant("de", ""); 423 424 /** Useful constant for language. 425 */ 426 static public final Locale ITALIAN = createConstant("it", ""); 427 428 /** Useful constant for language. 429 */ 430 static public final Locale JAPANESE = createConstant("ja", ""); 431 432 /** Useful constant for language. 433 */ 434 static public final Locale KOREAN = createConstant("ko", ""); 435 436 /** Useful constant for language. 437 */ 438 static public final Locale CHINESE = createConstant("zh", ""); 439 440 /** Useful constant for language. 441 */ 442 static public final Locale SIMPLIFIED_CHINESE = createConstant("zh", "CN"); 443 444 /** Useful constant for language. 445 */ 446 static public final Locale TRADITIONAL_CHINESE = createConstant("zh", "TW"); 447 448 /** Useful constant for country. 449 */ 450 static public final Locale FRANCE = createConstant("fr", "FR"); 451 452 /** Useful constant for country. 453 */ 454 static public final Locale GERMANY = createConstant("de", "DE"); 455 456 /** Useful constant for country. 457 */ 458 static public final Locale ITALY = createConstant("it", "IT"); 459 460 /** Useful constant for country. 461 */ 462 static public final Locale JAPAN = createConstant("ja", "JP"); 463 464 /** Useful constant for country. 465 */ 466 static public final Locale KOREA = createConstant("ko", "KR"); 467 468 /** Useful constant for country. 469 */ 470 static public final Locale CHINA = SIMPLIFIED_CHINESE; 471 472 /** Useful constant for country. 473 */ 474 static public final Locale PRC = SIMPLIFIED_CHINESE; 475 476 /** Useful constant for country. 477 */ 478 static public final Locale TAIWAN = TRADITIONAL_CHINESE; 479 480 /** Useful constant for country. 481 */ 482 static public final Locale UK = createConstant("en", "GB"); 483 484 /** Useful constant for country. 485 */ 486 static public final Locale US = createConstant("en", "US"); 487 488 /** Useful constant for country. 489 */ 490 static public final Locale CANADA = createConstant("en", "CA"); 491 492 /** Useful constant for country. 493 */ 494 static public final Locale CANADA_FRENCH = createConstant("fr", "CA"); 495 496 /** 497 * ISO 639-3 generic code for undetermined languages. 498 */ 499 private static final String UNDETERMINED_LANGUAGE = "und"; 500 501 /** 502 * Useful constant for the root locale. The root locale is the locale whose 503 * language, country, and variant are empty ("") strings. This is regarded 504 * as the base locale of all locales, and is used as the language/country 505 * neutral locale for the locale sensitive operations. 506 * 507 * @since 1.6 508 */ 509 static public final Locale ROOT = createConstant("", ""); 510 511 /** 512 * The key for the private use extension ('x'). 513 * 514 * @see #getExtension(char) 515 * @see Builder#setExtension(char, String) 516 * @since 1.7 517 */ 518 static public final char PRIVATE_USE_EXTENSION = 'x'; 519 520 /** 521 * The key for Unicode locale extension ('u'). 522 * 523 * @see #getExtension(char) 524 * @see Builder#setExtension(char, String) 525 * @since 1.7 526 */ 527 static public final char UNICODE_LOCALE_EXTENSION = 'u'; 528 529 /** serialization ID 530 */ 531 static final long serialVersionUID = 9149081749638150636L; 532 533 /** 534 * Display types for retrieving localized names from the name providers. 535 */ 536 private static final int DISPLAY_LANGUAGE = 0; 537 private static final int DISPLAY_COUNTRY = 1; 538 private static final int DISPLAY_VARIANT = 2; 539 private static final int DISPLAY_SCRIPT = 3; 540 541 /** 542 * Private constructor used by getInstance method 543 */ 544 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) { 545 this.baseLocale = baseLocale; 546 this.localeExtensions = extensions; 547 } 548 549 /** 550 * Construct a locale from language, country and variant. 551 * This constructor normalizes the language value to lowercase and 552 * the country value to uppercase. 553 * <p> 554 * <b>Note:</b> 555 * <ul> 556 * <li>ISO 639 is not a stable standard; some of the language codes it defines 557 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 558 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 559 * API on Locale will return only the OLD codes. 560 * <li>For backward compatibility reasons, this constructor does not make 561 * any syntactic checks on the input. 562 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially, 563 * see <a href="#special_cases_constructor">Special Cases</a> for more information. 564 * </ul> 565 * 566 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 567 * up to 8 characters in length. See the <code>Locale</code> class description about 568 * valid language values. 569 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code. 570 * See the <code>Locale</code> class description about valid country values. 571 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>. 572 * See the <code>Locale</code> class description for the details. 573 * @exception NullPointerException thrown if any argument is null. 574 */ 575 public Locale(String language, String country, String variant) { 576 if (language== null || country == null || variant == null) { 577 throw new NullPointerException(); 578 } 579 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant); 580 localeExtensions = getCompatibilityExtensions(language, "", country, variant); 581 } 582 583 /** 584 * Construct a locale from language and country. 585 * This constructor normalizes the language value to lowercase and 586 * the country value to uppercase. 587 * <p> 588 * <b>Note:</b> 589 * <ul> 590 * <li>ISO 639 is not a stable standard; some of the language codes it defines 591 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 592 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 593 * API on Locale will return only the OLD codes. 594 * <li>For backward compatibility reasons, this constructor does not make 595 * any syntactic checks on the input. 596 * </ul> 597 * 598 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 599 * up to 8 characters in length. See the <code>Locale</code> class description about 600 * valid language values. 601 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code. 602 * See the <code>Locale</code> class description about valid country values. 603 * @exception NullPointerException thrown if either argument is null. 604 */ 605 public Locale(String language, String country) { 606 this(language, country, ""); 607 } 608 609 /** 610 * Construct a locale from a language code. 611 * This constructor normalizes the language value to lowercase. 612 * <p> 613 * <b>Note:</b> 614 * <ul> 615 * <li>ISO 639 is not a stable standard; some of the language codes it defines 616 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 617 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 618 * API on Locale will return only the OLD codes. 619 * <li>For backward compatibility reasons, this constructor does not make 620 * any syntactic checks on the input. 621 * </ul> 622 * 623 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 624 * up to 8 characters in length. See the <code>Locale</code> class description about 625 * valid language values. 626 * @exception NullPointerException thrown if argument is null. 627 * @since 1.4 628 */ 629 public Locale(String language) { 630 this(language, "", ""); 631 } 632 633 /** 634 * This method must be called only for creating the Locale.* 635 * constants due to making shortcuts. 636 */ 637 private static Locale createConstant(String lang, String country) { 638 BaseLocale base = BaseLocale.createInstance(lang, country); 639 return getInstance(base, null); 640 } 641 642 /** 643 * Returns a <code>Locale</code> constructed from the given 644 * <code>language</code>, <code>country</code> and 645 * <code>variant</code>. If the same <code>Locale</code> instance 646 * is available in the cache, then that instance is 647 * returned. Otherwise, a new <code>Locale</code> instance is 648 * created and cached. 649 * 650 * @param language lowercase 2 to 8 language code. 651 * @param country uppercase two-letter ISO-3166 code and numric-3 UN M.49 area code. 652 * @param variant vendor and browser specific code. See class description. 653 * @return the <code>Locale</code> instance requested 654 * @exception NullPointerException if any argument is null. 655 */ 656 static Locale getInstance(String language, String country, String variant) { 657 return getInstance(language, "", country, variant, null); 658 } 659 660 static Locale getInstance(String language, String script, String country, 661 String variant, LocaleExtensions extensions) { 662 if (language== null || script == null || country == null || variant == null) { 663 throw new NullPointerException(); 664 } 665 666 if (extensions == null) { 667 extensions = getCompatibilityExtensions(language, script, country, variant); 668 } 669 670 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant); 671 return getInstance(baseloc, extensions); 672 } 673 674 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) { 675 LocaleKey key = new LocaleKey(baseloc, extensions); 676 return LOCALECACHE.get(key); 677 } 678 679 private static class Cache extends LocaleObjectCache<LocaleKey, Locale> { 680 private Cache() { 681 } 682 683 @Override 684 protected Locale createObject(LocaleKey key) { 685 return new Locale(key.base, key.exts); 686 } 687 } 688 689 private static final class LocaleKey { 690 private final BaseLocale base; 691 private final LocaleExtensions exts; 692 private final int hash; 693 694 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) { 695 base = baseLocale; 696 exts = extensions; 697 698 // Calculate the hash value here because it's always used. 699 int h = base.hashCode(); 700 if (exts != null) { 701 h ^= exts.hashCode(); 702 } 703 hash = h; 704 } 705 706 @Override 707 public boolean equals(Object obj) { 708 if (this == obj) { 709 return true; 710 } 711 if (!(obj instanceof LocaleKey)) { 712 return false; 713 } 714 LocaleKey other = (LocaleKey)obj; 715 if (hash != other.hash || !base.equals(other.base)) { 716 return false; 717 } 718 if (exts == null) { 719 return other.exts == null; 720 } 721 return exts.equals(other.exts); 722 } 723 724 @Override 725 public int hashCode() { 726 return hash; 727 } 728 } 729 730 /** 731 * Gets the current value of the default locale for this instance 732 * of the Java Virtual Machine. 733 * <p> 734 * The Java Virtual Machine sets the default locale during startup 735 * based on the host environment. It is used by many locale-sensitive 736 * methods if no locale is explicitly specified. 737 * It can be changed using the 738 * {@link #setDefault(java.util.Locale) setDefault} method. 739 * 740 * @return the default locale for this instance of the Java Virtual Machine 741 */ 742 public static Locale getDefault() { 743 // do not synchronize this method - see 4071298 744 // it's OK if more than one default locale happens to be created 745 if (defaultLocale == null) { 746 initDefault(); 747 } 748 return defaultLocale; 749 } 750 751 /** 752 * Gets the current value of the default locale for the specified Category 753 * for this instance of the Java Virtual Machine. 754 * <p> 755 * The Java Virtual Machine sets the default locale during startup based 756 * on the host environment. It is used by many locale-sensitive methods 757 * if no locale is explicitly specified. It can be changed using the 758 * setDefault(Locale.Category, Locale) method. 759 * 760 * @param category - the specified category to get the default locale 761 * @throws NullPointerException - if category is null 762 * @return the default locale for the specified Category for this instance 763 * of the Java Virtual Machine 764 * @see #setDefault(Locale.Category, Locale) 765 * @since 1.7 766 */ 767 public static Locale getDefault(Locale.Category category) { 768 // do not synchronize this method - see 4071298 769 // it's OK if more than one default locale happens to be created 770 switch (category) { 771 case DISPLAY: 772 if (defaultDisplayLocale == null) { 773 initDefault(category); 774 } 775 return defaultDisplayLocale; 776 case FORMAT: 777 if (defaultFormatLocale == null) { 778 initDefault(category); 779 } 780 return defaultFormatLocale; 781 default: 782 assert false: "Unknown Category"; 783 } 784 return getDefault(); 785 } 786 787 private static void initDefault() { 788 String language, region, script, country, variant; 789 language = AccessController.doPrivileged( 790 new GetPropertyAction("user.language", "en")); 791 // for compatibility, check for old user.region property 792 region = AccessController.doPrivileged( 793 new GetPropertyAction("user.region")); 794 if (region != null) { 795 // region can be of form country, country_variant, or _variant 796 int i = region.indexOf('_'); 797 if (i >= 0) { 798 country = region.substring(0, i); 799 variant = region.substring(i + 1); 800 } else { 801 country = region; 802 variant = ""; 803 } 804 script = ""; 805 } else { 806 script = AccessController.doPrivileged( 807 new GetPropertyAction("user.script", "")); 808 country = AccessController.doPrivileged( 809 new GetPropertyAction("user.country", "")); 810 variant = AccessController.doPrivileged( 811 new GetPropertyAction("user.variant", "")); 812 } 813 defaultLocale = getInstance(language, script, country, variant, null); 814 } 815 816 private static void initDefault(Locale.Category category) { 817 // make sure defaultLocale is initialized 818 if (defaultLocale == null) { 819 initDefault(); 820 } 821 822 Locale defaultCategoryLocale = getInstance( 823 AccessController.doPrivileged( 824 new GetPropertyAction(category.languageKey, defaultLocale.getLanguage())), 825 AccessController.doPrivileged( 826 new GetPropertyAction(category.scriptKey, defaultLocale.getScript())), 827 AccessController.doPrivileged( 828 new GetPropertyAction(category.countryKey, defaultLocale.getCountry())), 829 AccessController.doPrivileged( 830 new GetPropertyAction(category.variantKey, defaultLocale.getVariant())), 831 null); 832 833 switch (category) { 834 case DISPLAY: 835 defaultDisplayLocale = defaultCategoryLocale; 836 break; 837 case FORMAT: 838 defaultFormatLocale = defaultCategoryLocale; 839 break; 840 } 841 } 842 843 /** 844 * Sets the default locale for this instance of the Java Virtual Machine. 845 * This does not affect the host locale. 846 * <p> 847 * If there is a security manager, its <code>checkPermission</code> 848 * method is called with a <code>PropertyPermission("user.language", "write")</code> 849 * permission before the default locale is changed. 850 * <p> 851 * The Java Virtual Machine sets the default locale during startup 852 * based on the host environment. It is used by many locale-sensitive 853 * methods if no locale is explicitly specified. 854 * <p> 855 * Since changing the default locale may affect many different areas 856 * of functionality, this method should only be used if the caller 857 * is prepared to reinitialize locale-sensitive code running 858 * within the same Java Virtual Machine. 859 * <p> 860 * By setting the default locale with this method, all of the default 861 * locales for each Category are also set to the specified default locale. 862 * 863 * @throws SecurityException 864 * if a security manager exists and its 865 * <code>checkPermission</code> method doesn't allow the operation. 866 * @throws NullPointerException if <code>newLocale</code> is null 867 * @param newLocale the new default locale 868 * @see SecurityManager#checkPermission 869 * @see java.util.PropertyPermission 870 */ 871 public static synchronized void setDefault(Locale newLocale) { 872 setDefault(Category.DISPLAY, newLocale); 873 setDefault(Category.FORMAT, newLocale); 874 defaultLocale = newLocale; 875 } 876 877 /** 878 * Sets the default locale for the specified Category for this instance 879 * of the Java Virtual Machine. This does not affect the host locale. 880 * <p> 881 * If there is a security manager, its checkPermission method is called 882 * with a PropertyPermission("user.language", "write") permission before 883 * the default locale is changed. 884 * <p> 885 * The Java Virtual Machine sets the default locale during startup based 886 * on the host environment. It is used by many locale-sensitive methods 887 * if no locale is explicitly specified. 888 * <p> 889 * Since changing the default locale may affect many different areas of 890 * functionality, this method should only be used if the caller is 891 * prepared to reinitialize locale-sensitive code running within the 892 * same Java Virtual Machine. 893 * <p> 894 * 895 * @param category - the specified category to set the default locale 896 * @param newLocale - the new default locale 897 * @throws SecurityException - if a security manager exists and its 898 * checkPermission method doesn't allow the operation. 899 * @throws NullPointerException - if category and/or newLocale is null 900 * @see SecurityManager#checkPermission(java.security.Permission) 901 * @see PropertyPermission 902 * @see #getDefault(Locale.Category) 903 * @since 1.7 904 */ 905 public static synchronized void setDefault(Locale.Category category, 906 Locale newLocale) { 907 if (category == null) 908 throw new NullPointerException("Category cannot be NULL"); 909 if (newLocale == null) 910 throw new NullPointerException("Can't set default locale to NULL"); 911 912 SecurityManager sm = System.getSecurityManager(); 913 if (sm != null) sm.checkPermission(new PropertyPermission 914 ("user.language", "write")); 915 switch (category) { 916 case DISPLAY: 917 defaultDisplayLocale = newLocale; 918 break; 919 case FORMAT: 920 defaultFormatLocale = newLocale; 921 break; 922 default: 923 assert false: "Unknown Category"; 924 } 925 } 926 927 /** 928 * Returns an array of all installed locales. 929 * The returned array represents the union of locales supported 930 * by the Java runtime environment and by installed 931 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider} 932 * implementations. It must contain at least a <code>Locale</code> 933 * instance equal to {@link java.util.Locale#US Locale.US}. 934 * 935 * @return An array of installed locales. 936 */ 937 public static Locale[] getAvailableLocales() { 938 return LocaleServiceProviderPool.getAllAvailableLocales(); 939 } 940 941 /** 942 * Returns a list of all 2-letter country codes defined in ISO 3166. 943 * Can be used to create Locales. 944 * <p> 945 * <b>Note:</b> The <code>Locale</code> class also supports other codes for 946 * country (region), such as 3-letter numeric UN M.49 area codes. 947 * Therefore, the list returned by this method does not contain ALL valid 948 * codes that can be used to create Locales. 949 */ 950 public static String[] getISOCountries() { 951 if (isoCountries == null) { 952 isoCountries = getISO2Table(LocaleISOData.isoCountryTable); 953 } 954 String[] result = new String[isoCountries.length]; 955 System.arraycopy(isoCountries, 0, result, 0, isoCountries.length); 956 return result; 957 } 958 959 /** 960 * Returns a list of all 2-letter language codes defined in ISO 639. 961 * Can be used to create Locales. 962 * <p> 963 * <b>Note:</b> 964 * <ul> 965 * <li>ISO 639 is not a stable standard— some languages' codes have changed. 966 * The list this function returns includes both the new and the old codes for the 967 * languages whose codes have changed. 968 * <li>The <code>Locale</code> class also supports language codes up to 969 * 8 characters in length. Therefore, the list returned by this method does 970 * not contain ALL valid codes that can be used to create Locales. 971 * </ul> 972 */ 973 public static String[] getISOLanguages() { 974 if (isoLanguages == null) { 975 isoLanguages = getISO2Table(LocaleISOData.isoLanguageTable); 976 } 977 String[] result = new String[isoLanguages.length]; 978 System.arraycopy(isoLanguages, 0, result, 0, isoLanguages.length); 979 return result; 980 } 981 982 private static final String[] getISO2Table(String table) { 983 int len = table.length() / 5; 984 String[] isoTable = new String[len]; 985 for (int i = 0, j = 0; i < len; i++, j += 5) { 986 isoTable[i] = table.substring(j, j + 2); 987 } 988 return isoTable; 989 } 990 991 /** 992 * Returns the language code of this Locale. 993 * 994 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed. 995 * Locale's constructor recognizes both the new and the old codes for the languages 996 * whose codes have changed, but this function always returns the old code. If you 997 * want to check for a specific language whose code has changed, don't do 998 * <pre> 999 * if (locale.getLanguage().equals("he")) // BAD! 1000 * ... 1001 * </pre> 1002 * Instead, do 1003 * <pre> 1004 * if (locale.getLanguage().equals(new Locale("he").getLanguage())) 1005 * ... 1006 * </pre> 1007 * @return The language code, or the empty string if none is defined. 1008 * @see #getDisplayLanguage 1009 */ 1010 public String getLanguage() { 1011 return baseLocale.getLanguage(); 1012 } 1013 1014 /** 1015 * Returns the script for this locale, which should 1016 * either be the empty string or an ISO 15924 4-letter script 1017 * code. The first letter is uppercase and the rest are 1018 * lowercase, for example, 'Latn', 'Cyrl'. 1019 * 1020 * @return The script code, or the empty string if none is defined. 1021 * @see #getDisplayScript 1022 * @since 1.7 1023 */ 1024 public String getScript() { 1025 return baseLocale.getScript(); 1026 } 1027 1028 /** 1029 * Returns the country/region code for this locale, which should 1030 * either be the empty string, an uppercase ISO 3166 2-letter code, 1031 * or a UN M.49 3-digit code. 1032 * 1033 * @return The country/region code, or the empty string if none is defined. 1034 * @see #getDisplayCountry 1035 */ 1036 public String getCountry() { 1037 return baseLocale.getRegion(); 1038 } 1039 1040 /** 1041 * Returns the variant code for this locale. 1042 * 1043 * @return The variant code, or the empty string if none is defined. 1044 * @see #getDisplayVariant 1045 */ 1046 public String getVariant() { 1047 return baseLocale.getVariant(); 1048 } 1049 1050 /** 1051 * Returns the extension (or private use) value associated with 1052 * the specified key, or null if there is no extension 1053 * associated with the key. To be well-formed, the key must be one 1054 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so 1055 * for example 'z' and 'Z' represent the same extension. 1056 * 1057 * @param key the extension key 1058 * @return The extension, or null if this locale defines no 1059 * extension for the specified key. 1060 * @throws IllegalArgumentException if key is not well-formed 1061 * @see #PRIVATE_USE_EXTENSION 1062 * @see #UNICODE_LOCALE_EXTENSION 1063 * @since 1.7 1064 */ 1065 public String getExtension(char key) { 1066 if (!LocaleExtensions.isValidKey(key)) { 1067 throw new IllegalArgumentException("Ill-formed extension key: " + key); 1068 } 1069 return (localeExtensions == null) ? null : localeExtensions.getExtensionValue(key); 1070 } 1071 1072 /** 1073 * Returns the set of extension keys associated with this locale, or the 1074 * empty set if it has no extensions. The returned set is unmodifiable. 1075 * The keys will all be lower-case. 1076 * 1077 * @return The set of extension keys, or the empty set if this locale has 1078 * no extensions. 1079 * @since 1.7 1080 */ 1081 public Set<Character> getExtensionKeys() { 1082 if (localeExtensions == null) { 1083 return Collections.emptySet(); 1084 } 1085 return localeExtensions.getKeys(); 1086 } 1087 1088 /** 1089 * Returns the set of unicode locale attributes associated with 1090 * this locale, or the empty set if it has no attributes. The 1091 * returned set is unmodifiable. 1092 * 1093 * @return The set of attributes. 1094 * @since 1.7 1095 */ 1096 public Set<String> getUnicodeLocaleAttributes() { 1097 if (localeExtensions == null) { 1098 return Collections.emptySet(); 1099 } 1100 return localeExtensions.getUnicodeLocaleAttributes(); 1101 } 1102 1103 /** 1104 * Returns the Unicode locale type associated with the specified Unicode locale key 1105 * for this locale. Returns the empty string for keys that are defined with no type. 1106 * Returns null if the key is not defined. Keys are case-insensitive. The key must 1107 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is 1108 * thrown. 1109 * 1110 * @param key the Unicode locale key 1111 * @return The Unicode locale type associated with the key, or null if the 1112 * locale does not define the key. 1113 * @throws IllegalArgumentException if the key is not well-formed 1114 * @throws NullPointerException if <code>key</code> is null 1115 * @since 1.7 1116 */ 1117 public String getUnicodeLocaleType(String key) { 1118 if (!UnicodeLocaleExtension.isKey(key)) { 1119 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key); 1120 } 1121 return (localeExtensions == null) ? null : localeExtensions.getUnicodeLocaleType(key); 1122 } 1123 1124 /** 1125 * Returns the set of Unicode locale keys defined by this locale, or the empty set if 1126 * this locale has none. The returned set is immutable. Keys are all lower case. 1127 * 1128 * @return The set of Unicode locale keys, or the empty set if this locale has 1129 * no Unicode locale keywords. 1130 * @since 1.7 1131 */ 1132 public Set<String> getUnicodeLocaleKeys() { 1133 if (localeExtensions == null) { 1134 return Collections.emptySet(); 1135 } 1136 return localeExtensions.getUnicodeLocaleKeys(); 1137 } 1138 1139 /** 1140 * Package locale method returning the Locale's BaseLocale, 1141 * used by ResourceBundle 1142 * @return base locale of this Locale 1143 */ 1144 BaseLocale getBaseLocale() { 1145 return baseLocale; 1146 } 1147 1148 /** 1149 * Package private method returning the Locale's LocaleExtensions, 1150 * used by ResourceBundle. 1151 * @return locale exnteions of this Locale, 1152 * or {@code null} if no extensions are defined 1153 */ 1154 LocaleExtensions getLocaleExtensions() { 1155 return localeExtensions; 1156 } 1157 1158 /** 1159 * Returns a string representation of this <code>Locale</code> 1160 * object, consisting of language, country, variant, script, 1161 * and extensions as below: 1162 * <p><blockquote> 1163 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "-" + extensions 1164 * </blockquote> 1165 * 1166 * Language is always lower case, country is always upper case, script is always title 1167 * case, and extensions are always lower case. Extensions and private use subtags 1168 * will be in canonical order as explained in {@link #toLanguageTag}. 1169 * 1170 * <p>When the locale has neither script nor extensions, the result is the same as in 1171 * Java 6 and prior. 1172 * 1173 * <p>If both the language and country fields are missing, this function will return 1174 * the empty string, even if the variant, script, or extensions field is present (you 1175 * can't have a locale with just a variant, the variant must accompany a well-formed 1176 * language or country code). 1177 * 1178 * <p>If script or extensions are present and variant is missing, no underscore is 1179 * added before the "#". 1180 * 1181 * <p>This behavior is designed to support debugging and to be compatible with 1182 * previous uses of <code>toString</code> that expected language, country, and variant 1183 * fields only. To represent a Locale as a String for interchange purposes, use 1184 * {@link #toLanguageTag}. 1185 * 1186 * <p>Examples: <ul><tt> 1187 * <li>en 1188 * <li>de_DE 1189 * <li>_GB 1190 * <li>en_US_WIN 1191 * <li>de__POSIX 1192 * <li>zh_CN_#Hans 1193 * <li>zh_TW_#Hant-x-java 1194 * <li>th_TH_TH_#u-nu-thai</tt></ul> 1195 * 1196 * @return A string representation of the Locale, for debugging. 1197 * @see #getDisplayName 1198 * @see #toLanguageTag 1199 */ 1200 @Override 1201 public final String toString() { 1202 boolean l = (baseLocale.getLanguage().length() != 0); 1203 boolean s = (baseLocale.getScript().length() != 0); 1204 boolean r = (baseLocale.getRegion().length() != 0); 1205 boolean v = (baseLocale.getVariant().length() != 0); 1206 boolean e = (localeExtensions != null && localeExtensions.getID().length() != 0); 1207 1208 StringBuilder result = new StringBuilder(baseLocale.getLanguage()); 1209 if (r || (l && (v || s || e))) { 1210 result.append('_') 1211 .append(baseLocale.getRegion()); // This may just append '_' 1212 } 1213 if (v && (l || r)) { 1214 result.append('_') 1215 .append(baseLocale.getVariant()); 1216 } 1217 1218 if (s && (l || r)) { 1219 result.append("_#") 1220 .append(baseLocale.getScript()); 1221 } 1222 1223 if (e && (l || r)) { 1224 result.append('_'); 1225 if (!s) { 1226 result.append('#'); 1227 } 1228 result.append(localeExtensions.getID()); 1229 } 1230 1231 return result.toString(); 1232 } 1233 1234 /** 1235 * Returns a well-formed IETF BCP 47 language tag representing 1236 * this locale. 1237 * 1238 * <p>If this <code>Locale</code> has a language, country, or 1239 * variant that does not satisfy the IETF BCP 47 language tag 1240 * syntax requirements, this method handles these fields as 1241 * described below: 1242 * 1243 * <p><b>Language:</b> If language is empty, or not <a 1244 * href="#def_language" >well-formed</a> (for example "a" or 1245 * "e2"), it will be emitted as "und" (Undetermined). 1246 * 1247 * <p><b>Country:</b> If country is not <a 1248 * href="#def_region">well-formed</a> (for example "12" or "USA"), 1249 * it will be omitted. 1250 * 1251 * <p><b>Variant:</b> If variant <b>is</b> <a 1252 * href="#def_variant">well-formed</a>, each sub-segment 1253 * (delimited by '-' or '_') is emitted as a subtag. Otherwise: 1254 * <ul> 1255 * 1256 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code> 1257 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first 1258 * ill-formed sub-segment and all following will be appended to 1259 * the private use subtag. The first appended subtag will be 1260 * "lvariant", followed by the sub-segments in order, separated by 1261 * hyphen. For example, "x-lvariant-WIN", 1262 * "Oracle-x-lvariant-JDK-Standard-Edition". 1263 * 1264 * <li>if any sub-segment does not match 1265 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated 1266 * and the problematic sub-segment and all following sub-segments 1267 * will be omitted. If the remainder is non-empty, it will be 1268 * emitted as a private use subtag as above (even if the remainder 1269 * turns out to be well-formed). For example, 1270 * "Solaris_isjustthecoolestthing" is emitted as 1271 * "x-lvariant-Solaris", not as "solaris".</li></ul> 1272 * 1273 * <p><b>Special Conversions:</b> Java supports some old locale 1274 * representations, including deprecated ISO language codes, 1275 * for compatibility. This method performs the following 1276 * conversions: 1277 * <ul> 1278 * 1279 * <li>Deprecated ISO language codes "iw", "ji", and "in" are 1280 * converted to "he", "yi", and "id", respectively. 1281 * 1282 * <li>A locale with language "no", country "NO", and variant 1283 * "NY", representing Norwegian Nynorsk (Norway), is converted 1284 * to a language tag "nn-NO".</li></ul> 1285 * 1286 * <p><b>Note:</b> Although the language tag created by this 1287 * method is well-formed (satisfies the syntax requirements 1288 * defined by the IETF BCP 47 specification), it is not 1289 * necessarily a valid BCP 47 language tag. For example, 1290 * <pre> 1291 * new Locale("xx", "YY").toLanguageTag();</pre> 1292 * 1293 * will return "xx-YY", but the language subtag "xx" and the 1294 * region subtag "YY" are invalid because they are not registered 1295 * in the IANA Language Subtag Registry. 1296 * 1297 * @return a BCP47 language tag representing the locale 1298 * @see #forLanguageTag(String) 1299 * @since 1.7 1300 */ 1301 public String toLanguageTag() { 1302 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions); 1303 StringBuilder buf = new StringBuilder(); 1304 1305 String subtag = tag.getLanguage(); 1306 if (subtag.length() > 0) { 1307 buf.append(LanguageTag.canonicalizeLanguage(subtag)); 1308 } 1309 1310 subtag = tag.getScript(); 1311 if (subtag.length() > 0) { 1312 buf.append(LanguageTag.SEP); 1313 buf.append(LanguageTag.canonicalizeScript(subtag)); 1314 } 1315 1316 subtag = tag.getRegion(); 1317 if (subtag.length() > 0) { 1318 buf.append(LanguageTag.SEP); 1319 buf.append(LanguageTag.canonicalizeRegion(subtag)); 1320 } 1321 1322 List<String>subtags = tag.getVariants(); 1323 for (String s : subtags) { 1324 buf.append(LanguageTag.SEP); 1325 // preserve casing 1326 buf.append(s); 1327 } 1328 1329 subtags = tag.getExtensions(); 1330 for (String s : subtags) { 1331 buf.append(LanguageTag.SEP); 1332 buf.append(LanguageTag.canonicalizeExtension(s)); 1333 } 1334 1335 subtag = tag.getPrivateuse(); 1336 if (subtag.length() > 0) { 1337 if (buf.length() > 0) { 1338 buf.append(LanguageTag.SEP); 1339 } 1340 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP); 1341 // preserve casing 1342 buf.append(subtag); 1343 } 1344 1345 return buf.toString(); 1346 } 1347 1348 /** 1349 * Returns a locale for the specified IETF BCP 47 language tag string. 1350 * 1351 * <p>If the specified language tag contains any ill-formed subtags, 1352 * the first such subtag and all following subtags are ignored. Compare 1353 * to {@link Locale.Builder#setLanguageTag} which throws an exception 1354 * in this case. 1355 * 1356 * <p>The following <b>conversions</b> are performed:<ul> 1357 * 1358 * <li>The language code "und" is mapped to language "". 1359 * 1360 * <li>The language codes "he", "yi", and "id" are mapped to "iw", 1361 * "ji", and "in" respectively. (This is the same canonicalization 1362 * that's done in Locale's constructors.) 1363 * 1364 * <li>The portion of a private use subtag prefixed by "lvariant", 1365 * if any, is removed and appended to the variant field in the 1366 * result locale (without case normalization). If it is then 1367 * empty, the private use subtag is discarded: 1368 * 1369 * <pre> 1370 * Locale loc; 1371 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX"); 1372 * loc.getVariant(); // returns "POSIX" 1373 * loc.getExtension('x'); // returns null 1374 * 1375 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def"); 1376 * loc.getVariant(); // returns "POSIX_Abc_Def" 1377 * loc.getExtension('x'); // returns "urp" 1378 * </pre> 1379 * 1380 * <li>When the languageTag argument contains an extlang subtag, 1381 * the first such subtag is used as the language, and the primary 1382 * language subtag and other extlang subtags are ignored: 1383 * 1384 * <pre> 1385 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao" 1386 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US" 1387 * </pre> 1388 * 1389 * <li>Case is normalized except for variant tags, which are left 1390 * unchanged. Language is normalized to lower case, script to 1391 * title case, country to upper case, and extensions to lower 1392 * case. 1393 * 1394 * <li>If, after processing, the locale would exactly match either 1395 * ja_JP_JP or th_TH_TH with no extensions, the appropriate 1396 * extensions are added as though the constructor had been called: 1397 * 1398 * <pre> 1399 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag(); 1400 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP" 1401 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag(); 1402 * // returns "th-TH-u-nu-thai-x-lvariant-TH" 1403 * <pre></ul> 1404 * 1405 * <p>This implements the 'Language-Tag' production of BCP47, and 1406 * so supports grandfathered (regular and irregular) as well as 1407 * private use language tags. Stand alone private use tags are 1408 * represented as empty language and extension 'x-whatever', 1409 * and grandfathered tags are converted to their canonical replacements 1410 * where they exist. 1411 * 1412 * <p>Grandfathered tags with canonical replacements are as follows: 1413 * 1414 * <table> 1415 * <tbody align="center"> 1416 * <tr><th>grandfathered tag</th><th> </th><th>modern replacement</th></tr> 1417 * <tr><td>art-lojban</td><td> </td><td>jbo</td></tr> 1418 * <tr><td>i-ami</td><td> </td><td>ami</td></tr> 1419 * <tr><td>i-bnn</td><td> </td><td>bnn</td></tr> 1420 * <tr><td>i-hak</td><td> </td><td>hak</td></tr> 1421 * <tr><td>i-klingon</td><td> </td><td>tlh</td></tr> 1422 * <tr><td>i-lux</td><td> </td><td>lb</td></tr> 1423 * <tr><td>i-navajo</td><td> </td><td>nv</td></tr> 1424 * <tr><td>i-pwn</td><td> </td><td>pwn</td></tr> 1425 * <tr><td>i-tao</td><td> </td><td>tao</td></tr> 1426 * <tr><td>i-tay</td><td> </td><td>tay</td></tr> 1427 * <tr><td>i-tsu</td><td> </td><td>tsu</td></tr> 1428 * <tr><td>no-bok</td><td> </td><td>nb</td></tr> 1429 * <tr><td>no-nyn</td><td> </td><td>nn</td></tr> 1430 * <tr><td>sgn-BE-FR</td><td> </td><td>sfb</td></tr> 1431 * <tr><td>sgn-BE-NL</td><td> </td><td>vgt</td></tr> 1432 * <tr><td>sgn-CH-DE</td><td> </td><td>sgg</td></tr> 1433 * <tr><td>zh-guoyu</td><td> </td><td>cmn</td></tr> 1434 * <tr><td>zh-hakka</td><td> </td><td>hak</td></tr> 1435 * <tr><td>zh-min-nan</td><td> </td><td>nan</td></tr> 1436 * <tr><td>zh-xiang</td><td> </td><td>hsn</td></tr> 1437 * </tbody> 1438 * </table> 1439 * 1440 * <p>Grandfathered tags with no modern replacement will be 1441 * converted as follows: 1442 * 1443 * <table> 1444 * <tbody align="center"> 1445 * <tr><th>grandfathered tag</th><th> </th><th>converts to</th></tr> 1446 * <tr><td>cel-gaulish</td><td> </td><td>xtg-x-cel-gaulish</td></tr> 1447 * <tr><td>en-GB-oed</td><td> </td><td>en-GB-x-oed</td></tr> 1448 * <tr><td>i-default</td><td> </td><td>en-x-i-default</td></tr> 1449 * <tr><td>i-enochian</td><td> </td><td>und-x-i-enochian</td></tr> 1450 * <tr><td>i-mingo</td><td> </td><td>see-x-i-mingo</td></tr> 1451 * <tr><td>zh-min</td><td> </td><td>nan-x-zh-min</td></tr> 1452 * </tbody> 1453 * </table> 1454 * 1455 * <p>For a list of all grandfathered tags, see the 1456 * IANA Language Subtag Registry (search for "Type: grandfathered"). 1457 * 1458 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code> 1459 * and <code>forLanguageTag</code> will round-trip. 1460 * 1461 * @param languageTag the language tag 1462 * @return The locale that best represents the language tag. 1463 * @throws NullPointerException if <code>languageTag</code> is <code>null</code> 1464 * @see #toLanguageTag() 1465 * @see java.util.Locale.Builder#setLanguageTag(String) 1466 * @since 1.7 1467 */ 1468 public static Locale forLanguageTag(String languageTag) { 1469 LanguageTag tag = LanguageTag.parse(languageTag, null); 1470 InternalLocaleBuilder bldr = new InternalLocaleBuilder(); 1471 bldr.setLanguageTag(tag); 1472 BaseLocale base = bldr.getBaseLocale(); 1473 LocaleExtensions exts = bldr.getLocaleExtensions(); 1474 if (exts == null && base.getVariant().length() > 0) { 1475 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(), 1476 base.getRegion(), base.getVariant()); 1477 } 1478 return getInstance(base, exts); 1479 } 1480 1481 /** 1482 * Returns a three-letter abbreviation of this locale's language. 1483 * If the language matches an ISO 639-1 two-letter code, the 1484 * corresponding ISO 639-2/T three-letter lowercase code is 1485 * returned. The ISO 639-2 language codes can be found on-line, 1486 * see "Codes for the Representation of Names of Languages Part 2: 1487 * Alpha-3 Code". If the locale specifies a three-letter 1488 * language, the language is returned as is. If the locale does 1489 * not specify a language the empty string is returned. 1490 * 1491 * @return A three-letter abbreviation of this locale's language. 1492 * @exception MissingResourceException Throws MissingResourceException if 1493 * three-letter language abbreviation is not available for this locale. 1494 */ 1495 public String getISO3Language() throws MissingResourceException { 1496 String lang = baseLocale.getLanguage(); 1497 if (lang.length() == 3) { 1498 return lang; 1499 } 1500 1501 String language3 = getISO3Code(lang, LocaleISOData.isoLanguageTable); 1502 if (language3 == null) { 1503 throw new MissingResourceException("Couldn't find 3-letter language code for " 1504 + lang, "FormatData_" + toString(), "ShortLanguage"); 1505 } 1506 return language3; 1507 } 1508 1509 /** 1510 * Returns a three-letter abbreviation for this locale's country. 1511 * If the country matches an ISO 3166-1 alpha-2 code, the 1512 * corresponding ISO 3166-1 alpha-3 uppercase code is returned. 1513 * If the locale doesn't specify a country, this will be the empty 1514 * string. 1515 * 1516 * <p>The ISO 3166-1 codes can be found on-line. 1517 * 1518 * @return A three-letter abbreviation of this locale's country. 1519 * @exception MissingResourceException Throws MissingResourceException if the 1520 * three-letter country abbreviation is not available for this locale. 1521 */ 1522 public String getISO3Country() throws MissingResourceException { 1523 // Android-changed : Hack for UN-M49 location codes. 1524 if (baseLocale.getRegion().length() == 3) { 1525 return baseLocale.getRegion(); 1526 } 1527 1528 String country3 = getISO3Code(baseLocale.getRegion(), LocaleISOData.isoCountryTable); 1529 if (country3 == null) { 1530 throw new MissingResourceException("Couldn't find 3-letter country code for " 1531 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry"); 1532 } 1533 return country3; 1534 } 1535 1536 private static final String getISO3Code(String iso2Code, String table) { 1537 int codeLength = iso2Code.length(); 1538 if (codeLength == 0) { 1539 return ""; 1540 } 1541 1542 int tableLength = table.length(); 1543 int index = tableLength; 1544 if (codeLength == 2) { 1545 char c1 = iso2Code.charAt(0); 1546 char c2 = iso2Code.charAt(1); 1547 for (index = 0; index < tableLength; index += 5) { 1548 if (table.charAt(index) == c1 1549 && table.charAt(index + 1) == c2) { 1550 break; 1551 } 1552 } 1553 } 1554 return index < tableLength ? table.substring(index + 2, index + 5) : null; 1555 } 1556 1557 /** 1558 * Returns a name for the locale's language that is appropriate for display to the 1559 * user. 1560 * If possible, the name returned will be localized for the default locale. 1561 * For example, if the locale is fr_FR and the default locale 1562 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and 1563 * the default locale is fr_FR, getDisplayLanguage() will return "anglais". 1564 * If the name returned cannot be localized for the default locale, 1565 * (say, we don't have a Japanese name for Croatian), 1566 * this function falls back on the English name, and uses the ISO code as a last-resort 1567 * value. If the locale doesn't specify a language, this function returns the empty string. 1568 */ 1569 public final String getDisplayLanguage() { 1570 return getDisplayLanguage(getDefault(Category.DISPLAY)); 1571 } 1572 1573 /** 1574 * Returns the name of this locale's language, localized to {@code locale}. 1575 * If the language name is unknown, the language code is returned. 1576 */ 1577 public String getDisplayLanguage(Locale locale) { 1578 String languageCode = baseLocale.getLanguage(); 1579 if (languageCode.isEmpty()) { 1580 return ""; 1581 } 1582 1583 // Hacks for backward compatibility. 1584 // 1585 // Our language tag will contain "und" if the languageCode is invalid 1586 // or missing. ICU will then return "langue indéterminée" or the equivalent 1587 // display language for the indeterminate language code. 1588 // 1589 // Sigh... ugh... and what not. 1590 final String normalizedLanguage = normalizeAndValidateLanguage( 1591 languageCode, false /* strict */); 1592 if (UNDETERMINED_LANGUAGE.equals(normalizedLanguage)) { 1593 return languageCode; 1594 } 1595 1596 // TODO: We need a new hack or a complete fix for http://b/8049507 --- We would 1597 // cover the frameworks' tracks when they were using "tl" instead of "fil". 1598 String result = ICU.getDisplayLanguage(this, locale); 1599 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1600 result = ICU.getDisplayLanguage(this, Locale.getDefault()); 1601 } 1602 return result; 1603 } 1604 1605 private static String normalizeAndValidateLanguage(String language, boolean strict) { 1606 if (language == null || language.isEmpty()) { 1607 return ""; 1608 } 1609 1610 final String lowercaseLanguage = language.toLowerCase(Locale.ROOT); 1611 if (!isValidBcp47Alpha(lowercaseLanguage, 2, 3)) { 1612 if (strict) { 1613 throw new IllformedLocaleException("Invalid language: " + language); 1614 } else { 1615 return UNDETERMINED_LANGUAGE; 1616 } 1617 } 1618 1619 return lowercaseLanguage; 1620 } 1621 1622 /* 1623 * Checks whether a given string is an ASCII alphanumeric string. 1624 */ 1625 private static boolean isAsciiAlphaNum(String string) { 1626 for (int i = 0; i < string.length(); i++) { 1627 final char character = string.charAt(i); 1628 if (!(character >= 'a' && character <= 'z' || 1629 character >= 'A' && character <= 'Z' || 1630 character >= '0' && character <= '9')) { 1631 return false; 1632 } 1633 } 1634 1635 return true; 1636 } 1637 1638 /** 1639 * Returns a name for the the locale's script that is appropriate for display to 1640 * the user. If possible, the name will be localized for the default locale. Returns 1641 * the empty string if this locale doesn't specify a script code. 1642 * 1643 * @return the display name of the script code for the current default locale 1644 * @since 1.7 1645 */ 1646 public String getDisplayScript() { 1647 return getDisplayScript(getDefault()); 1648 } 1649 1650 /** 1651 * Returns the name of this locale's script code, localized to {@link Locale}. If the 1652 * script code is unknown, the return value of this method is the same as that of 1653 * {@link #getScript()}. 1654 * 1655 * @since 1.7 1656 */ 1657 public String getDisplayScript(Locale locale) { 1658 String scriptCode = baseLocale.getScript(); 1659 if (scriptCode.isEmpty()) { 1660 return ""; 1661 } 1662 1663 String result = ICU.getDisplayScript(this, locale); 1664 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1665 result = ICU.getDisplayScript(this, Locale.getDefault()); 1666 } 1667 1668 return result; 1669 1670 } 1671 1672 /** 1673 * Returns a name for the locale's country that is appropriate for display to the 1674 * user. 1675 * If possible, the name returned will be localized for the default locale. 1676 * For example, if the locale is fr_FR and the default locale 1677 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and 1678 * the default locale is fr_FR, getDisplayCountry() will return "Etats-Unis". 1679 * If the name returned cannot be localized for the default locale, 1680 * (say, we don't have a Japanese name for Croatia), 1681 * this function falls back on the English name, and uses the ISO code as a last-resort 1682 * value. If the locale doesn't specify a country, this function returns the empty string. 1683 */ 1684 public final String getDisplayCountry() { 1685 return getDisplayCountry(getDefault(Category.DISPLAY)); 1686 } 1687 /** 1688 * Returns the name of this locale's country, localized to {@code locale}. 1689 * Returns the empty string if this locale does not correspond to a specific 1690 * country. 1691 */ 1692 public String getDisplayCountry(Locale locale) { 1693 String countryCode = baseLocale.getRegion(); 1694 if (countryCode.isEmpty()) { 1695 return ""; 1696 } 1697 1698 final String normalizedRegion = normalizeAndValidateRegion( 1699 countryCode, false /* strict */); 1700 if (normalizedRegion.isEmpty()) { 1701 return countryCode; 1702 } 1703 1704 String result = ICU.getDisplayCountry(this, locale); 1705 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1706 result = ICU.getDisplayCountry(this, Locale.getDefault()); 1707 } 1708 return result; 1709 } 1710 1711 private static String normalizeAndValidateRegion(String region, boolean strict) { 1712 if (region == null || region.isEmpty()) { 1713 return ""; 1714 } 1715 1716 final String uppercaseRegion = region.toUpperCase(Locale.ROOT); 1717 if (!isValidBcp47Alpha(uppercaseRegion, 2, 2) && 1718 !isUnM49AreaCode(uppercaseRegion)) { 1719 if (strict) { 1720 throw new IllformedLocaleException("Invalid region: " + region); 1721 } else { 1722 return ""; 1723 } 1724 } 1725 1726 return uppercaseRegion; 1727 } 1728 1729 private static boolean isValidBcp47Alpha(String string, int lowerBound, int upperBound) { 1730 final int length = string.length(); 1731 if (length < lowerBound || length > upperBound) { 1732 return false; 1733 } 1734 1735 for (int i = 0; i < length; ++i) { 1736 final char character = string.charAt(i); 1737 if (!(character >= 'a' && character <= 'z' || 1738 character >= 'A' && character <= 'Z')) { 1739 return false; 1740 } 1741 } 1742 1743 return true; 1744 } 1745 1746 /** 1747 * A UN M.49 is a 3 digit numeric code. 1748 */ 1749 private static boolean isUnM49AreaCode(String code) { 1750 if (code.length() != 3) { 1751 return false; 1752 } 1753 1754 for (int i = 0; i < 3; ++i) { 1755 final char character = code.charAt(i); 1756 if (!(character >= '0' && character <= '9')) { 1757 return false; 1758 } 1759 } 1760 1761 return true; 1762 } 1763 1764 /** 1765 * Returns a name for the locale's variant code that is appropriate for display to the 1766 * user. If possible, the name will be localized for the default locale. If the locale 1767 * doesn't specify a variant code, this function returns the empty string. 1768 */ 1769 public final String getDisplayVariant() { 1770 return getDisplayVariant(getDefault(Category.DISPLAY)); 1771 } 1772 1773 /** 1774 * Returns the full variant name in the specified {@code Locale} for the variant code 1775 * of this {@code Locale}. If there is no matching variant name, the variant code is 1776 * returned. 1777 * 1778 * @since 1.7 1779 */ 1780 public String getDisplayVariant(Locale locale) { 1781 String variantCode = baseLocale.getVariant(); 1782 if (variantCode.isEmpty()) { 1783 return ""; 1784 } 1785 1786 try { 1787 normalizeAndValidateVariant(variantCode); 1788 } catch (IllformedLocaleException ilfe) { 1789 return variantCode; 1790 } 1791 1792 String result = ICU.getDisplayVariant(this, locale); 1793 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1794 result = ICU.getDisplayVariant(this, Locale.getDefault()); 1795 } 1796 1797 // The "old style" locale constructors allow us to pass in variants that aren't 1798 // valid BCP-47 variant subtags. When that happens, toLanguageTag will not emit 1799 // them. Note that we know variantCode.length() > 0 due to the isEmpty check at 1800 // the beginning of this function. 1801 if (result.isEmpty()) { 1802 return variantCode; 1803 } 1804 return result; 1805 } 1806 1807 private static String normalizeAndValidateVariant(String variant) { 1808 if (variant == null || variant.isEmpty()) { 1809 return ""; 1810 } 1811 1812 // Note that unlike extensions, we canonicalize to lower case alphabets 1813 // and underscores instead of hyphens. 1814 final String normalizedVariant = variant.replace('-', '_'); 1815 String[] subTags = normalizedVariant.split("_"); 1816 1817 for (String subTag : subTags) { 1818 if (!isValidVariantSubtag(subTag)) { 1819 throw new IllformedLocaleException("Invalid variant: " + variant); 1820 } 1821 } 1822 1823 return normalizedVariant; 1824 } 1825 1826 private static boolean isValidVariantSubtag(String subTag) { 1827 // The BCP-47 spec states that : 1828 // - Subtags can be between [5, 8] alphanumeric chars in length. 1829 // - Subtags that start with a number are allowed to be 4 chars in length. 1830 if (subTag.length() >= 5 && subTag.length() <= 8) { 1831 if (isAsciiAlphaNum(subTag)) { 1832 return true; 1833 } 1834 } else if (subTag.length() == 4) { 1835 final char firstChar = subTag.charAt(0); 1836 if ((firstChar >= '0' && firstChar <= '9') && isAsciiAlphaNum(subTag)) { 1837 return true; 1838 } 1839 } 1840 1841 return false; 1842 } 1843 1844 /** 1845 * Returns a name for the locale that is appropriate for display to the 1846 * user. This will be the values returned by getDisplayLanguage(), 1847 * getDisplayScript(), getDisplayCountry(), and getDisplayVariant() assembled 1848 * into a single string. The the non-empty values are used in order, 1849 * with the second and subsequent names in parentheses. For example: 1850 * <blockquote> 1851 * language (script, country, variant)<br> 1852 * language (country)<br> 1853 * language (variant)<br> 1854 * script (country)<br> 1855 * country<br> 1856 * </blockquote> 1857 * depending on which fields are specified in the locale. If the 1858 * language, sacript, country, and variant fields are all empty, 1859 * this function returns the empty string. 1860 */ 1861 public final String getDisplayName() { 1862 return getDisplayName(getDefault(Category.DISPLAY)); 1863 } 1864 1865 /** 1866 * Returns this locale's language name, country name, and variant, localized 1867 * to {@code locale}. The exact output form depends on whether this locale 1868 * corresponds to a specific language, script, country and variant. 1869 * 1870 * <p>For example: 1871 * <ul> 1872 * <li>{@code new Locale("en").getDisplayName(Locale.US)} -> {@code English} 1873 * <li>{@code new Locale("en", "US").getDisplayName(Locale.US)} -> {@code English (United States)} 1874 * <li>{@code new Locale("en", "US", "POSIX").getDisplayName(Locale.US)} -> {@code English (United States,Computer)} 1875 * <li>{@code Locale.fromLanguageTag("zh-Hant-CN").getDisplayName(Locale.US)} -> {@code Chinese (Traditional Han,China)} 1876 * <li>{@code new Locale("en").getDisplayName(Locale.FRANCE)} -> {@code anglais} 1877 * <li>{@code new Locale("en", "US").getDisplayName(Locale.FRANCE)} -> {@code anglais (États-Unis)} 1878 * <li>{@code new Locale("en", "US", "POSIX").getDisplayName(Locale.FRANCE)} -> {@code anglais (États-Unis,informatique)}. 1879 * </ul> 1880 */ 1881 public String getDisplayName(Locale locale) { 1882 int count = 0; 1883 StringBuilder buffer = new StringBuilder(); 1884 String languageCode = baseLocale.getLanguage(); 1885 if (!languageCode.isEmpty()) { 1886 String displayLanguage = getDisplayLanguage(locale); 1887 buffer.append(displayLanguage.isEmpty() ? languageCode : displayLanguage); 1888 ++count; 1889 } 1890 String scriptCode = baseLocale.getScript(); 1891 if (!scriptCode.isEmpty()) { 1892 if (count == 1) { 1893 buffer.append(" ("); 1894 } 1895 String displayScript = getDisplayScript(locale); 1896 buffer.append(displayScript.isEmpty() ? scriptCode : displayScript); 1897 ++count; 1898 } 1899 String countryCode = baseLocale.getRegion(); 1900 if (!countryCode.isEmpty()) { 1901 if (count == 1) { 1902 buffer.append(" ("); 1903 } else if (count == 2) { 1904 buffer.append(","); 1905 } 1906 String displayCountry = getDisplayCountry(locale); 1907 buffer.append(displayCountry.isEmpty() ? countryCode : displayCountry); 1908 ++count; 1909 } 1910 String variantCode = baseLocale.getVariant(); 1911 if (!variantCode.isEmpty()) { 1912 if (count == 1) { 1913 buffer.append(" ("); 1914 } else if (count == 2 || count == 3) { 1915 buffer.append(","); 1916 } 1917 String displayVariant = getDisplayVariant(locale); 1918 buffer.append(displayVariant.isEmpty() ? variantCode : displayVariant); 1919 ++count; 1920 } 1921 if (count > 1) { 1922 buffer.append(")"); 1923 } 1924 return buffer.toString(); 1925 } 1926 1927 /** 1928 * Overrides Cloneable. 1929 */ 1930 public Object clone() 1931 { 1932 try { 1933 Locale that = (Locale)super.clone(); 1934 return that; 1935 } catch (CloneNotSupportedException e) { 1936 throw new InternalError(); 1937 } 1938 } 1939 1940 /** 1941 * Override hashCode. 1942 * Since Locales are often used in hashtables, caches the value 1943 * for speed. 1944 */ 1945 @Override 1946 public int hashCode() { 1947 int hc = hashCodeValue; 1948 if (hc == 0) { 1949 hc = baseLocale.hashCode(); 1950 if (localeExtensions != null) { 1951 hc ^= localeExtensions.hashCode(); 1952 } 1953 hashCodeValue = hc; 1954 } 1955 return hc; 1956 } 1957 1958 // Overrides 1959 1960 /** 1961 * Returns true if this Locale is equal to another object. A Locale is 1962 * deemed equal to another Locale with identical language, script, country, 1963 * variant and extensions, and unequal to all other objects. 1964 * 1965 * @return true if this Locale is equal to the specified object. 1966 */ 1967 @Override 1968 public boolean equals(Object obj) { 1969 if (this == obj) // quick check 1970 return true; 1971 if (!(obj instanceof Locale)) 1972 return false; 1973 BaseLocale otherBase = ((Locale)obj).baseLocale; 1974 if (!baseLocale.equals(otherBase)) { 1975 return false; 1976 } 1977 if (localeExtensions == null) { 1978 return ((Locale)obj).localeExtensions == null; 1979 } 1980 return localeExtensions.equals(((Locale)obj).localeExtensions); 1981 } 1982 1983 // ================= privates ===================================== 1984 1985 private transient BaseLocale baseLocale; 1986 private transient LocaleExtensions localeExtensions; 1987 1988 /** 1989 * Calculated hashcode 1990 */ 1991 private transient volatile int hashCodeValue = 0; 1992 1993 private static Locale defaultLocale = null; 1994 private static Locale defaultDisplayLocale = null; 1995 private static Locale defaultFormatLocale = null; 1996 1997 /** 1998 * Format a list using given pattern strings. 1999 * If either of the patterns is null, then a the list is 2000 * formatted by concatenation with the delimiter ','. 2001 * @param stringList the list of strings to be formatted. 2002 * @param listPattern should create a MessageFormat taking 0-3 arguments 2003 * and formatting them into a list. 2004 * @param listCompositionPattern should take 2 arguments 2005 * and is used by composeList. 2006 * @return a string representing the list. 2007 */ 2008 private static String formatList(String[] stringList, String listPattern, String listCompositionPattern) { 2009 // If we have no list patterns, compose the list in a simple, 2010 // non-localized way. 2011 if (listPattern == null || listCompositionPattern == null) { 2012 StringBuffer result = new StringBuffer(); 2013 for (int i=0; i<stringList.length; ++i) { 2014 if (i>0) result.append(','); 2015 result.append(stringList[i]); 2016 } 2017 return result.toString(); 2018 } 2019 2020 // Compose the list down to three elements if necessary 2021 if (stringList.length > 3) { 2022 MessageFormat format = new MessageFormat(listCompositionPattern); 2023 stringList = composeList(format, stringList); 2024 } 2025 2026 // Rebuild the argument list with the list length as the first element 2027 Object[] args = new Object[stringList.length + 1]; 2028 System.arraycopy(stringList, 0, args, 1, stringList.length); 2029 args[0] = new Integer(stringList.length); 2030 2031 // Format it using the pattern in the resource 2032 MessageFormat format = new MessageFormat(listPattern); 2033 return format.format(args); 2034 } 2035 2036 /** 2037 * Given a list of strings, return a list shortened to three elements. 2038 * Shorten it by applying the given format to the first two elements 2039 * recursively. 2040 * @param format a format which takes two arguments 2041 * @param list a list of strings 2042 * @return if the list is three elements or shorter, the same list; 2043 * otherwise, a new list of three elements. 2044 */ 2045 private static String[] composeList(MessageFormat format, String[] list) { 2046 if (list.length <= 3) return list; 2047 2048 // Use the given format to compose the first two elements into one 2049 String[] listItems = { list[0], list[1] }; 2050 String newItem = format.format(listItems); 2051 2052 // Form a new list one element shorter 2053 String[] newList = new String[list.length-1]; 2054 System.arraycopy(list, 2, newList, 1, newList.length-1); 2055 newList[0] = newItem; 2056 2057 // Recurse 2058 return composeList(format, newList); 2059 } 2060 2061 /** 2062 * @serialField language String 2063 * language subtag in lower case. (See <a href="java/util/Locale.html#getLanguage()">getLanguage()</a>) 2064 * @serialField country String 2065 * country subtag in upper case. (See <a href="java/util/Locale.html#getCountry()">getCountry()</a>) 2066 * @serialField variant String 2067 * variant subtags separated by LOWLINE characters. (See <a href="java/util/Locale.html#getVariant()">getVariant()</a>) 2068 * @serialField hashcode int 2069 * deprecated, for forward compatibility only 2070 * @serialField script String 2071 * script subtag in title case (See <a href="java/util/Locale.html#getScript()">getScript()</a>) 2072 * @serialField extensions String 2073 * canonical representation of extensions, that is, 2074 * BCP47 extensions in alphabetical order followed by 2075 * BCP47 private use subtags, all in lower case letters 2076 * separated by HYPHEN-MINUS characters. 2077 * (See <a href="java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>, 2078 * <a href="java/util/Locale.html#getExtension(char)">getExtension(char)</a>) 2079 */ 2080 private static final ObjectStreamField[] serialPersistentFields = { 2081 new ObjectStreamField("language", String.class), 2082 new ObjectStreamField("country", String.class), 2083 new ObjectStreamField("variant", String.class), 2084 new ObjectStreamField("hashcode", int.class), 2085 new ObjectStreamField("script", String.class), 2086 new ObjectStreamField("extensions", String.class), 2087 }; 2088 2089 /** 2090 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>. 2091 * @param out the <code>ObjectOutputStream</code> to write 2092 * @throws IOException 2093 * @since 1.7 2094 */ 2095 private void writeObject(ObjectOutputStream out) throws IOException { 2096 ObjectOutputStream.PutField fields = out.putFields(); 2097 fields.put("language", baseLocale.getLanguage()); 2098 fields.put("script", baseLocale.getScript()); 2099 fields.put("country", baseLocale.getRegion()); 2100 fields.put("variant", baseLocale.getVariant()); 2101 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID()); 2102 fields.put("hashcode", -1); // place holder just for backward support 2103 out.writeFields(); 2104 } 2105 2106 /** 2107 * Deserializes this <code>Locale</code>. 2108 * @param in the <code>ObjectInputStream</code> to read 2109 * @throws IOException 2110 * @throws ClassNotFoundException 2111 * @throws IllformdLocaleException 2112 * @since 1.7 2113 */ 2114 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException { 2115 ObjectInputStream.GetField fields = in.readFields(); 2116 String language = (String)fields.get("language", ""); 2117 String script = (String)fields.get("script", ""); 2118 String country = (String)fields.get("country", ""); 2119 String variant = (String)fields.get("variant", ""); 2120 String extStr = (String)fields.get("extensions", ""); 2121 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant); 2122 if (extStr.length() > 0) { 2123 try { 2124 InternalLocaleBuilder bldr = new InternalLocaleBuilder(); 2125 bldr.setExtensions(extStr); 2126 localeExtensions = bldr.getLocaleExtensions(); 2127 } catch (LocaleSyntaxException e) { 2128 throw new IllformedLocaleException(e.getMessage()); 2129 } 2130 } else { 2131 localeExtensions = null; 2132 } 2133 } 2134 2135 /** 2136 * Returns a cached <code>Locale</code> instance equivalent to 2137 * the deserialized <code>Locale</code>. When serialized 2138 * language, country and variant fields read from the object data stream 2139 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions 2140 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code> 2141 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script 2142 * type is "thai"). See <a href="Locale.html#special_cases_constructor"/>Special Cases</a> 2143 * for more information. 2144 * 2145 * @return an instance of <code>Locale</code> equivalent to 2146 * the deserialized <code>Locale</code>. 2147 * @throws java.io.ObjectStreamException 2148 */ 2149 private Object readResolve() throws java.io.ObjectStreamException { 2150 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(), 2151 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions); 2152 } 2153 2154 private static volatile String[] isoLanguages = null; 2155 2156 private static volatile String[] isoCountries = null; 2157 2158 private static String convertOldISOCodes(String language) { 2159 // we accept both the old and the new ISO codes for the languages whose ISO 2160 // codes have changed, but we always store the OLD code, for backward compatibility 2161 language = LocaleUtils.toLowerString(language).intern(); 2162 if (language == "he") { 2163 return "iw"; 2164 } else if (language == "yi") { 2165 return "ji"; 2166 } else if (language == "id") { 2167 return "in"; 2168 } else { 2169 return language; 2170 } 2171 } 2172 2173 private static LocaleExtensions getCompatibilityExtensions(String language, 2174 String script, 2175 String country, 2176 String variant) { 2177 LocaleExtensions extensions = null; 2178 // Special cases for backward compatibility support 2179 if (LocaleUtils.caseIgnoreMatch(language, "ja") 2180 && script.length() == 0 2181 && LocaleUtils.caseIgnoreMatch(country, "jp") 2182 && "JP".equals(variant)) { 2183 // ja_JP_JP -> u-ca-japanese (calendar = japanese) 2184 extensions = LocaleExtensions.CALENDAR_JAPANESE; 2185 } else if (LocaleUtils.caseIgnoreMatch(language, "th") 2186 && script.length() == 0 2187 && LocaleUtils.caseIgnoreMatch(country, "th") 2188 && "TH".equals(variant)) { 2189 // th_TH_TH -> u-nu-thai (numbersystem = thai) 2190 extensions = LocaleExtensions.NUMBER_THAI; 2191 } 2192 return extensions; 2193 } 2194 2195 /** 2196 * @hide for internal use only. 2197 */ 2198 public static String adjustLanguageCode(String languageCode) { 2199 String adjusted = languageCode.toLowerCase(Locale.US); 2200 // Map new language codes to the obsolete language 2201 // codes so the correct resource bundles will be used. 2202 if (languageCode.equals("he")) { 2203 adjusted = "iw"; 2204 } else if (languageCode.equals("id")) { 2205 adjusted = "in"; 2206 } else if (languageCode.equals("yi")) { 2207 adjusted = "ji"; 2208 } 2209 2210 return adjusted; 2211 } 2212 2213 /** 2214 * Obtains a localized locale names from a LocaleNameProvider 2215 * implementation. 2216 */ 2217 private static class LocaleNameGetter 2218 implements LocaleServiceProviderPool.LocalizedObjectGetter<LocaleNameProvider, String> { 2219 private static final LocaleNameGetter INSTANCE = new LocaleNameGetter(); 2220 2221 public String getObject(LocaleNameProvider localeNameProvider, 2222 Locale locale, 2223 String key, 2224 Object... params) { 2225 assert params.length == 2; 2226 int type = (Integer)params[0]; 2227 String code = (String)params[1]; 2228 2229 switch(type) { 2230 case DISPLAY_LANGUAGE: 2231 return localeNameProvider.getDisplayLanguage(code, locale); 2232 case DISPLAY_COUNTRY: 2233 return localeNameProvider.getDisplayCountry(code, locale); 2234 case DISPLAY_VARIANT: 2235 return localeNameProvider.getDisplayVariant(code, locale); 2236 case DISPLAY_SCRIPT: 2237 return localeNameProvider.getDisplayScript(code, locale); 2238 default: 2239 assert false; // shouldn't happen 2240 } 2241 2242 return null; 2243 } 2244 } 2245 2246 /** 2247 * Enum for locale categories. These locale categories are used to get/set 2248 * the default locale for the specific functionality represented by the 2249 * category. 2250 * 2251 * @see #getDefault(Locale.Category) 2252 * @see #setDefault(Locale.Category, Locale) 2253 * @since 1.7 2254 */ 2255 public enum Category { 2256 2257 /** 2258 * Category used to represent the default locale for 2259 * displaying user interfaces. 2260 */ 2261 DISPLAY("user.language.display", 2262 "user.script.display", 2263 "user.country.display", 2264 "user.variant.display"), 2265 2266 /** 2267 * Category used to represent the default locale for 2268 * formatting dates, numbers, and/or currencies. 2269 */ 2270 FORMAT("user.language.format", 2271 "user.script.format", 2272 "user.country.format", 2273 "user.variant.format"); 2274 2275 Category(String languageKey, String scriptKey, String countryKey, String variantKey) { 2276 this.languageKey = languageKey; 2277 this.scriptKey = scriptKey; 2278 this.countryKey = countryKey; 2279 this.variantKey = variantKey; 2280 } 2281 2282 final String languageKey; 2283 final String scriptKey; 2284 final String countryKey; 2285 final String variantKey; 2286 } 2287 2288 /** 2289 * <code>Builder</code> is used to build instances of <code>Locale</code> 2290 * from values configured by the setters. Unlike the <code>Locale</code> 2291 * constructors, the <code>Builder</code> checks if a value configured by a 2292 * setter satisfies the syntax requirements defined by the <code>Locale</code> 2293 * class. A <code>Locale</code> object created by a <code>Builder</code> is 2294 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag 2295 * without losing information. 2296 * 2297 * <p><b>Note:</b> The <code>Locale</code> class does not provide any 2298 * syntactic restrictions on variant, while BCP 47 requires each variant 2299 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3 2300 * alphanumerics. The method <code>setVariant</code> throws 2301 * <code>IllformedLocaleException</code> for a variant that does not satisfy 2302 * this restriction. If it is necessary to support such a variant, use a 2303 * Locale constructor. However, keep in mind that a <code>Locale</code> 2304 * object created this way might lose the variant information when 2305 * transformed to a BCP 47 language tag. 2306 * 2307 * <p>The following example shows how to create a <code>Locale</code> object 2308 * with the <code>Builder</code>. 2309 * <blockquote> 2310 * <pre> 2311 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build(); 2312 * </pre> 2313 * </blockquote> 2314 * 2315 * <p>Builders can be reused; <code>clear()</code> resets all 2316 * fields to their default values. 2317 * 2318 * @see Locale#forLanguageTag 2319 * @since 1.7 2320 */ 2321 public static final class Builder { 2322 private final InternalLocaleBuilder localeBuilder; 2323 2324 /** 2325 * Constructs an empty Builder. The default value of all 2326 * fields, extensions, and private use information is the 2327 * empty string. 2328 */ 2329 public Builder() { 2330 localeBuilder = new InternalLocaleBuilder(); 2331 } 2332 2333 /** 2334 * Resets the <code>Builder</code> to match the provided 2335 * <code>locale</code>. Existing state is discarded. 2336 * 2337 * <p>All fields of the locale must be well-formed, see {@link Locale}. 2338 * 2339 * <p>Locales with any ill-formed fields cause 2340 * <code>IllformedLocaleException</code> to be thrown, except for the 2341 * following three cases which are accepted for compatibility 2342 * reasons:<ul> 2343 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese" 2344 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai" 2345 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul> 2346 * 2347 * @param locale the locale 2348 * @return This builder. 2349 * @throws IllformedLocaleException if <code>locale</code> has 2350 * any ill-formed fields. 2351 * @throws NullPointerException if <code>locale</code> is null. 2352 */ 2353 public Builder setLocale(Locale locale) { 2354 try { 2355 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions); 2356 } catch (LocaleSyntaxException e) { 2357 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2358 } 2359 return this; 2360 } 2361 2362 /** 2363 * Resets the Builder to match the provided IETF BCP 47 2364 * language tag. Discards the existing state. Null and the 2365 * empty string cause the builder to be reset, like {@link 2366 * #clear}. Grandfathered tags (see {@link 2367 * Locale#forLanguageTag}) are converted to their canonical 2368 * form before being processed. Otherwise, the language tag 2369 * must be well-formed (see {@link Locale}) or an exception is 2370 * thrown (unlike <code>Locale.forLanguageTag</code>, which 2371 * just discards ill-formed and following portions of the 2372 * tag). 2373 * 2374 * @param languageTag the language tag 2375 * @return This builder. 2376 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed 2377 * @see Locale#forLanguageTag(String) 2378 */ 2379 public Builder setLanguageTag(String languageTag) { 2380 ParseStatus sts = new ParseStatus(); 2381 LanguageTag tag = LanguageTag.parse(languageTag, sts); 2382 if (sts.isError()) { 2383 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex()); 2384 } 2385 localeBuilder.setLanguageTag(tag); 2386 return this; 2387 } 2388 2389 /** 2390 * Sets the language. If <code>language</code> is the empty string or 2391 * null, the language in this <code>Builder</code> is removed. Otherwise, 2392 * the language must be <a href="./Locale.html#def_language">well-formed</a> 2393 * or an exception is thrown. 2394 * 2395 * <p>The typical language value is a two or three-letter language 2396 * code as defined in ISO639. 2397 * 2398 * @param language the language 2399 * @return This builder. 2400 * @throws IllformedLocaleException if <code>language</code> is ill-formed 2401 */ 2402 public Builder setLanguage(String language) { 2403 try { 2404 localeBuilder.setLanguage(language); 2405 } catch (LocaleSyntaxException e) { 2406 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2407 } 2408 return this; 2409 } 2410 2411 /** 2412 * Sets the script. If <code>script</code> is null or the empty string, 2413 * the script in this <code>Builder</code> is removed. 2414 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an 2415 * exception is thrown. 2416 * 2417 * <p>The typical script value is a four-letter script code as defined by ISO 15924. 2418 * 2419 * @param script the script 2420 * @return This builder. 2421 * @throws IllformedLocaleException if <code>script</code> is ill-formed 2422 */ 2423 public Builder setScript(String script) { 2424 try { 2425 localeBuilder.setScript(script); 2426 } catch (LocaleSyntaxException e) { 2427 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2428 } 2429 return this; 2430 } 2431 2432 /** 2433 * Sets the region. If region is null or the empty string, the region 2434 * in this <code>Builder</code> is removed. Otherwise, 2435 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an 2436 * exception is thrown. 2437 * 2438 * <p>The typical region value is a two-letter ISO 3166 code or a 2439 * three-digit UN M.49 area code. 2440 * 2441 * <p>The country value in the <code>Locale</code> created by the 2442 * <code>Builder</code> is always normalized to upper case. 2443 * 2444 * @param region the region 2445 * @return This builder. 2446 * @throws IllformedLocaleException if <code>region</code> is ill-formed 2447 */ 2448 public Builder setRegion(String region) { 2449 try { 2450 localeBuilder.setRegion(region); 2451 } catch (LocaleSyntaxException e) { 2452 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2453 } 2454 return this; 2455 } 2456 2457 /** 2458 * Sets the variant. If variant is null or the empty string, the 2459 * variant in this <code>Builder</code> is removed. Otherwise, it 2460 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a> 2461 * subtags, or an exception is thrown. 2462 * 2463 * <p><b>Note:</b> This method checks if <code>variant</code> 2464 * satisfies the IETF BCP 47 variant subtag's syntax requirements, 2465 * and normalizes the value to lowercase letters. However, 2466 * the <code>Locale</code> class does not impose any syntactic 2467 * restriction on variant, and the variant value in 2468 * <code>Locale</code> is case sensitive. To set such a variant, 2469 * use a Locale constructor. 2470 * 2471 * @param variant the variant 2472 * @return This builder. 2473 * @throws IllformedLocaleException if <code>variant</code> is ill-formed 2474 */ 2475 public Builder setVariant(String variant) { 2476 try { 2477 localeBuilder.setVariant(variant); 2478 } catch (LocaleSyntaxException e) { 2479 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2480 } 2481 return this; 2482 } 2483 2484 /** 2485 * Sets the extension for the given key. If the value is null or the 2486 * empty string, the extension is removed. Otherwise, the extension 2487 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception 2488 * is thrown. 2489 * 2490 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION 2491 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension. 2492 * Setting a value for this key replaces any existing Unicode locale key/type 2493 * pairs with those defined in the extension. 2494 * 2495 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION 2496 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be 2497 * well-formed, the value for this key needs only to have subtags of one to 2498 * eight alphanumeric characters, not two to eight as in the general case. 2499 * 2500 * @param key the extension key 2501 * @param value the extension value 2502 * @return This builder. 2503 * @throws IllformedLocaleException if <code>key</code> is illegal 2504 * or <code>value</code> is ill-formed 2505 * @see #setUnicodeLocaleKeyword(String, String) 2506 */ 2507 public Builder setExtension(char key, String value) { 2508 try { 2509 localeBuilder.setExtension(key, value); 2510 } catch (LocaleSyntaxException e) { 2511 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2512 } 2513 return this; 2514 } 2515 2516 /** 2517 * Sets the Unicode locale keyword type for the given key. If the type 2518 * is null, the Unicode keyword is removed. Otherwise, the key must be 2519 * non-null and both key and type must be <a 2520 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2521 * is thrown. 2522 * 2523 * <p>Keys and types are converted to lower case. 2524 * 2525 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension} 2526 * replaces all Unicode locale keywords with those defined in the 2527 * extension. 2528 * 2529 * @param key the Unicode locale key 2530 * @param type the Unicode locale type 2531 * @return This builder. 2532 * @throws IllformedLocaleException if <code>key</code> or <code>type</code> 2533 * is ill-formed 2534 * @throws NullPointerException if <code>key</code> is null 2535 * @see #setExtension(char, String) 2536 */ 2537 public Builder setUnicodeLocaleKeyword(String key, String type) { 2538 try { 2539 localeBuilder.setUnicodeLocaleKeyword(key, type); 2540 } catch (LocaleSyntaxException e) { 2541 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2542 } 2543 return this; 2544 } 2545 2546 /** 2547 * Adds a unicode locale attribute, if not already present, otherwise 2548 * has no effect. The attribute must not be null and must be <a 2549 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2550 * is thrown. 2551 * 2552 * @param attribute the attribute 2553 * @return This builder. 2554 * @throws NullPointerException if <code>attribute</code> is null 2555 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed 2556 * @see #setExtension(char, String) 2557 */ 2558 public Builder addUnicodeLocaleAttribute(String attribute) { 2559 try { 2560 localeBuilder.addUnicodeLocaleAttribute(attribute); 2561 } catch (LocaleSyntaxException e) { 2562 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2563 } 2564 return this; 2565 } 2566 2567 /** 2568 * Removes a unicode locale attribute, if present, otherwise has no 2569 * effect. The attribute must not be null and must be <a 2570 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2571 * is thrown. 2572 * 2573 * <p>Attribute comparision for removal is case-insensitive. 2574 * 2575 * @param attribute the attribute 2576 * @return This builder. 2577 * @throws NullPointerException if <code>attribute</code> is null 2578 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed 2579 * @see #setExtension(char, String) 2580 */ 2581 public Builder removeUnicodeLocaleAttribute(String attribute) { 2582 try { 2583 localeBuilder.removeUnicodeLocaleAttribute(attribute); 2584 } catch (LocaleSyntaxException e) { 2585 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2586 } 2587 return this; 2588 } 2589 2590 /** 2591 * Resets the builder to its initial, empty state. 2592 * 2593 * @return This builder. 2594 */ 2595 public Builder clear() { 2596 localeBuilder.clear(); 2597 return this; 2598 } 2599 2600 /** 2601 * Resets the extensions to their initial, empty state. 2602 * Language, script, region and variant are unchanged. 2603 * 2604 * @return This builder. 2605 * @see #setExtension(char, String) 2606 */ 2607 public Builder clearExtensions() { 2608 localeBuilder.clearExtensions(); 2609 return this; 2610 } 2611 2612 /** 2613 * Returns an instance of <code>Locale</code> created from the fields set 2614 * on this builder. 2615 * 2616 * <p>This applies the conversions listed in {@link Locale#forLanguageTag} 2617 * when constructing a Locale. (Grandfathered tags are handled in 2618 * {@link #setLanguageTag}.) 2619 * 2620 * @return A Locale. 2621 */ 2622 public Locale build() { 2623 BaseLocale baseloc = localeBuilder.getBaseLocale(); 2624 LocaleExtensions extensions = localeBuilder.getLocaleExtensions(); 2625 if (extensions == null && baseloc.getVariant().length() > 0) { 2626 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(), 2627 baseloc.getRegion(), baseloc.getVariant()); 2628 } 2629 return Locale.getInstance(baseloc, extensions); 2630 } 2631 } 2632} 2633