Locale.java revision 5a65597bdfe80ca0a0940b0f5c429dece7328b38
1/* 2 * Copyright (C) 2014 The Android Open Source Project 3 * Copyright (c) 1996, 2011, Oracle and/or its affiliates. All rights reserved. 4 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 5 * 6 * This code is free software; you can redistribute it and/or modify it 7 * under the terms of the GNU General Public License version 2 only, as 8 * published by the Free Software Foundation. Oracle designates this 9 * particular file as subject to the "Classpath" exception as provided 10 * by Oracle in the LICENSE file that accompanied this code. 11 * 12 * This code is distributed in the hope that it will be useful, but WITHOUT 13 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 14 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 15 * version 2 for more details (a copy is included in the LICENSE file that 16 * accompanied this code). 17 * 18 * You should have received a copy of the GNU General Public License version 19 * 2 along with this work; if not, write to the Free Software Foundation, 20 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 21 * 22 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 23 * or visit www.oracle.com if you need additional information or have any 24 * questions. 25 */ 26 27/* 28 * (C) Copyright Taligent, Inc. 1996, 1997 - All Rights Reserved 29 * (C) Copyright IBM Corp. 1996 - 1998 - All Rights Reserved 30 * 31 * The original version of this source code and documentation 32 * is copyrighted and owned by Taligent, Inc., a wholly-owned 33 * subsidiary of IBM. These materials are provided under terms 34 * of a License Agreement between Taligent and Sun. This technology 35 * is protected by multiple US and International patents. 36 * 37 * This notice and attribution to Taligent may not be removed. 38 * Taligent is a registered trademark of Taligent, Inc. 39 * 40 */ 41 42package java.util; 43 44import java.io.IOException; 45import java.io.ObjectInputStream; 46import java.io.ObjectOutputStream; 47import java.io.ObjectStreamField; 48import java.io.Serializable; 49import java.security.AccessController; 50import java.text.MessageFormat; 51import java.util.spi.LocaleNameProvider; 52import libcore.icu.ICU; 53 54import sun.security.action.GetPropertyAction; 55import sun.util.LocaleServiceProviderPool; 56import sun.util.locale.BaseLocale; 57import sun.util.locale.InternalLocaleBuilder; 58import sun.util.locale.LanguageTag; 59import sun.util.locale.LocaleExtensions; 60import sun.util.locale.LocaleObjectCache; 61import sun.util.locale.LocaleSyntaxException; 62import sun.util.locale.LocaleUtils; 63import sun.util.locale.ParseStatus; 64import sun.util.locale.UnicodeLocaleExtension; 65 66/** 67 * A <code>Locale</code> object represents a specific geographical, political, 68 * or cultural region. An operation that requires a <code>Locale</code> to perform 69 * its task is called <em>locale-sensitive</em> and uses the <code>Locale</code> 70 * to tailor information for the user. For example, displaying a number 71 * is a locale-sensitive operation— the number should be formatted 72 * according to the customs and conventions of the user's native country, 73 * region, or culture. 74 * 75 * <p> The <code>Locale</code> class implements identifiers 76 * interchangeable with BCP 47 (IETF BCP 47, "Tags for Identifying 77 * Languages"), with support for the LDML (UTS#35, "Unicode Locale 78 * Data Markup Language") BCP 47-compatible extensions for locale data 79 * exchange. 80 * 81 * <p> A <code>Locale</code> object logically consists of the fields 82 * described below. 83 * 84 * <dl> 85 * <dt><a name="def_language"/><b>language</b></dt> 86 * 87 * <dd>ISO 639 alpha-2 or alpha-3 language code, or registered 88 * language subtags up to 8 alpha letters (for future enhancements). 89 * When a language has both an alpha-2 code and an alpha-3 code, the 90 * alpha-2 code must be used. You can find a full list of valid 91 * language codes in the IANA Language Subtag Registry (search for 92 * "Type: language"). The language field is case insensitive, but 93 * <code>Locale</code> always canonicalizes to lower case.</dd><br> 94 * 95 * <dd>Well-formed language values have the form 96 * <code>[a-zA-Z]{2,8}</code>. Note that this is not the the full 97 * BCP47 language production, since it excludes extlang. They are 98 * not needed since modern three-letter language codes replace 99 * them.</dd><br> 100 * 101 * <dd>Example: "en" (English), "ja" (Japanese), "kok" (Konkani)</dd><br> 102 * 103 * <dt><a name="def_script"/><b>script</b></dt> 104 * 105 * <dd>ISO 15924 alpha-4 script code. You can find a full list of 106 * valid script codes in the IANA Language Subtag Registry (search 107 * for "Type: script"). The script field is case insensitive, but 108 * <code>Locale</code> always canonicalizes to title case (the first 109 * letter is upper case and the rest of the letters are lower 110 * case).</dd><br> 111 * 112 * <dd>Well-formed script values have the form 113 * <code>[a-zA-Z]{4}</code></dd><br> 114 * 115 * <dd>Example: "Latn" (Latin), "Cyrl" (Cyrillic)</dd><br> 116 * 117 * <dt><a name="def_region"/><b>country (region)</b></dt> 118 * 119 * <dd>ISO 3166 alpha-2 country code or UN M.49 numeric-3 area code. 120 * You can find a full list of valid country and region codes in the 121 * IANA Language Subtag Registry (search for "Type: region"). The 122 * country (region) field is case insensitive, but 123 * <code>Locale</code> always canonicalizes to upper case.</dd><br> 124 * 125 * <dd>Well-formed country/region values have 126 * the form <code>[a-zA-Z]{2} | [0-9]{3}</code></dd><br> 127 * 128 * <dd>Example: "US" (United States), "FR" (France), "029" 129 * (Caribbean)</dd><br> 130 * 131 * <dt><a name="def_variant"/><b>variant</b></dt> 132 * 133 * <dd>Any arbitrary value used to indicate a variation of a 134 * <code>Locale</code>. Where there are two or more variant values 135 * each indicating its own semantics, these values should be ordered 136 * by importance, with most important first, separated by 137 * underscore('_'). The variant field is case sensitive.</dd><br> 138 * 139 * <dd>Note: IETF BCP 47 places syntactic restrictions on variant 140 * subtags. Also BCP 47 subtags are strictly used to indicate 141 * additional variations that define a language or its dialects that 142 * are not covered by any combinations of language, script and 143 * region subtags. You can find a full list of valid variant codes 144 * in the IANA Language Subtag Registry (search for "Type: variant"). 145 * 146 * <p>However, the variant field in <code>Locale</code> has 147 * historically been used for any kind of variation, not just 148 * language variations. For example, some supported variants 149 * available in Java SE Runtime Environments indicate alternative 150 * cultural behaviors such as calendar type or number script. In 151 * BCP 47 this kind of information, which does not identify the 152 * language, is supported by extension subtags or private use 153 * subtags.</dd><br> 154 * 155 * <dd>Well-formed variant values have the form <code>SUBTAG 156 * (('_'|'-') SUBTAG)*</code> where <code>SUBTAG = 157 * [0-9][0-9a-zA-Z]{3} | [0-9a-zA-Z]{5,8}</code>. (Note: BCP 47 only 158 * uses hyphen ('-') as a delimiter, this is more lenient).</dd><br> 159 * 160 * <dd>Example: "polyton" (Polytonic Greek), "POSIX"</dd><br> 161 * 162 * <dt><a name="def_extensions"/><b>extensions</b></dt> 163 * 164 * <dd>A map from single character keys to string values, indicating 165 * extensions apart from language identification. The extensions in 166 * <code>Locale</code> implement the semantics and syntax of BCP 47 167 * extension subtags and private use subtags. The extensions are 168 * case insensitive, but <code>Locale</code> canonicalizes all 169 * extension keys and values to lower case. Note that extensions 170 * cannot have empty values.</dd><br> 171 * 172 * <dd>Well-formed keys are single characters from the set 173 * <code>[0-9a-zA-Z]</code>. Well-formed values have the form 174 * <code>SUBTAG ('-' SUBTAG)*</code> where for the key 'x' 175 * <code>SUBTAG = [0-9a-zA-Z]{1,8}</code> and for other keys 176 * <code>SUBTAG = [0-9a-zA-Z]{2,8}</code> (that is, 'x' allows 177 * single-character subtags).</dd><br> 178 * 179 * <dd>Example: key="u"/value="ca-japanese" (Japanese Calendar), 180 * key="x"/value="java-1-7"</dd> 181 * </dl> 182 * 183 * <b>Note:</b> Although BCP 47 requires field values to be registered 184 * in the IANA Language Subtag Registry, the <code>Locale</code> class 185 * does not provide any validation features. The <code>Builder</code> 186 * only checks if an individual field satisfies the syntactic 187 * requirement (is well-formed), but does not validate the value 188 * itself. See {@link Builder} for details. 189 * 190 * <h4><a name="def_locale_extension">Unicode locale/language extension</h4> 191 * 192 * <p>UTS#35, "Unicode Locale Data Markup Language" defines optional 193 * attributes and keywords to override or refine the default behavior 194 * associated with a locale. A keyword is represented by a pair of 195 * key and type. For example, "nu-thai" indicates that Thai local 196 * digits (value:"thai") should be used for formatting numbers 197 * (key:"nu"). 198 * 199 * <p>The keywords are mapped to a BCP 47 extension value using the 200 * extension key 'u' ({@link #UNICODE_LOCALE_EXTENSION}). The above 201 * example, "nu-thai", becomes the extension "u-nu-thai".code 202 * 203 * <p>Thus, when a <code>Locale</code> object contains Unicode locale 204 * attributes and keywords, 205 * <code>getExtension(UNICODE_LOCALE_EXTENSION)</code> will return a 206 * String representing this information, for example, "nu-thai". The 207 * <code>Locale</code> class also provides {@link 208 * #getUnicodeLocaleAttributes}, {@link #getUnicodeLocaleKeys}, and 209 * {@link #getUnicodeLocaleType} which allow you to access Unicode 210 * locale attributes and key/type pairs directly. When represented as 211 * a string, the Unicode Locale Extension lists attributes 212 * alphabetically, followed by key/type sequences with keys listed 213 * alphabetically (the order of subtags comprising a key's type is 214 * fixed when the type is defined) 215 * 216 * <p>A well-formed locale key has the form 217 * <code>[0-9a-zA-Z]{2}</code>. A well-formed locale type has the 218 * form <code>"" | [0-9a-zA-Z]{3,8} ('-' [0-9a-zA-Z]{3,8})*</code> (it 219 * can be empty, or a series of subtags 3-8 alphanums in length). A 220 * well-formed locale attribute has the form 221 * <code>[0-9a-zA-Z]{3,8}</code> (it is a single subtag with the same 222 * form as a locale type subtag). 223 * 224 * <p>The Unicode locale extension specifies optional behavior in 225 * locale-sensitive services. Although the LDML specification defines 226 * various keys and values, actual locale-sensitive service 227 * implementations in a Java Runtime Environment might not support any 228 * particular Unicode locale attributes or key/type pairs. 229 * 230 * <h4>Creating a Locale</h4> 231 * 232 * <p>There are several different ways to create a <code>Locale</code> 233 * object. 234 * 235 * <h5>Builder</h5> 236 * 237 * <p>Using {@link Builder} you can construct a <code>Locale</code> object 238 * that conforms to BCP 47 syntax. 239 * 240 * <h5>Constructors</h5> 241 * 242 * <p>The <code>Locale</code> class provides three constructors: 243 * <blockquote> 244 * <pre> 245 * {@link #Locale(String language)} 246 * {@link #Locale(String language, String country)} 247 * {@link #Locale(String language, String country, String variant)} 248 * </pre> 249 * </blockquote> 250 * These constructors allow you to create a <code>Locale</code> object 251 * with language, country and variant, but you cannot specify 252 * script or extensions. 253 * 254 * <h5>Factory Methods</h5> 255 * 256 * <p>The method {@link #forLanguageTag} creates a <code>Locale</code> 257 * object for a well-formed BCP 47 language tag. 258 * 259 * <h5>Locale Constants</h5> 260 * 261 * <p>The <code>Locale</code> class provides a number of convenient constants 262 * that you can use to create <code>Locale</code> objects for commonly used 263 * locales. For example, the following creates a <code>Locale</code> object 264 * for the United States: 265 * <blockquote> 266 * <pre> 267 * Locale.US 268 * </pre> 269 * </blockquote> 270 * 271 * <h4>Use of Locale</h4> 272 * 273 * <p>Once you've created a <code>Locale</code> you can query it for information 274 * about itself. Use <code>getCountry</code> to get the country (or region) 275 * code and <code>getLanguage</code> to get the language code. 276 * You can use <code>getDisplayCountry</code> to get the 277 * name of the country suitable for displaying to the user. Similarly, 278 * you can use <code>getDisplayLanguage</code> to get the name of 279 * the language suitable for displaying to the user. Interestingly, 280 * the <code>getDisplayXXX</code> methods are themselves locale-sensitive 281 * and have two versions: one that uses the default locale and one 282 * that uses the locale specified as an argument. 283 * 284 * <p>The Java Platform provides a number of classes that perform locale-sensitive 285 * operations. For example, the <code>NumberFormat</code> class formats 286 * numbers, currency, and percentages in a locale-sensitive manner. Classes 287 * such as <code>NumberFormat</code> have several convenience methods 288 * for creating a default object of that type. For example, the 289 * <code>NumberFormat</code> class provides these three convenience methods 290 * for creating a default <code>NumberFormat</code> object: 291 * <blockquote> 292 * <pre> 293 * NumberFormat.getInstance() 294 * NumberFormat.getCurrencyInstance() 295 * NumberFormat.getPercentInstance() 296 * </pre> 297 * </blockquote> 298 * Each of these methods has two variants; one with an explicit locale 299 * and one without; the latter uses the default locale: 300 * <blockquote> 301 * <pre> 302 * NumberFormat.getInstance(myLocale) 303 * NumberFormat.getCurrencyInstance(myLocale) 304 * NumberFormat.getPercentInstance(myLocale) 305 * </pre> 306 * </blockquote> 307 * A <code>Locale</code> is the mechanism for identifying the kind of object 308 * (<code>NumberFormat</code>) that you would like to get. The locale is 309 * <STRONG>just</STRONG> a mechanism for identifying objects, 310 * <STRONG>not</STRONG> a container for the objects themselves. 311 * 312 * <h4>Compatibility</h4> 313 * 314 * <p>In order to maintain compatibility with existing usage, Locale's 315 * constructors retain their behavior prior to the Java Runtime 316 * Environment version 1.7. The same is largely true for the 317 * <code>toString</code> method. Thus Locale objects can continue to 318 * be used as they were. In particular, clients who parse the output 319 * of toString into language, country, and variant fields can continue 320 * to do so (although this is strongly discouraged), although the 321 * variant field will have additional information in it if script or 322 * extensions are present. 323 * 324 * <p>In addition, BCP 47 imposes syntax restrictions that are not 325 * imposed by Locale's constructors. This means that conversions 326 * between some Locales and BCP 47 language tags cannot be made without 327 * losing information. Thus <code>toLanguageTag</code> cannot 328 * represent the state of locales whose language, country, or variant 329 * do not conform to BCP 47. 330 * 331 * <p>Because of these issues, it is recommended that clients migrate 332 * away from constructing non-conforming locales and use the 333 * <code>forLanguageTag</code> and <code>Locale.Builder</code> APIs instead. 334 * Clients desiring a string representation of the complete locale can 335 * then always rely on <code>toLanguageTag</code> for this purpose. 336 * 337 * <h5><a name="special_cases_constructor"/>Special cases</h5> 338 * 339 * <p>For compatibility reasons, two 340 * non-conforming locales are treated as special cases. These are 341 * <b><tt>ja_JP_JP</tt></b> and <b><tt>th_TH_TH</tt></b>. These are ill-formed 342 * in BCP 47 since the variants are too short. To ease migration to BCP 47, 343 * these are treated specially during construction. These two cases (and only 344 * these) cause a constructor to generate an extension, all other values behave 345 * exactly as they did prior to Java 7. 346 * 347 * <p>Java has used <tt>ja_JP_JP</tt> to represent Japanese as used in 348 * Japan together with the Japanese Imperial calendar. This is now 349 * representable using a Unicode locale extension, by specifying the 350 * Unicode locale key <tt>ca</tt> (for "calendar") and type 351 * <tt>japanese</tt>. When the Locale constructor is called with the 352 * arguments "ja", "JP", "JP", the extension "u-ca-japanese" is 353 * automatically added. 354 * 355 * <p>Java has used <tt>th_TH_TH</tt> to represent Thai as used in 356 * Thailand together with Thai digits. This is also now representable using 357 * a Unicode locale extension, by specifying the Unicode locale key 358 * <tt>nu</tt> (for "number") and value <tt>thai</tt>. When the Locale 359 * constructor is called with the arguments "th", "TH", "TH", the 360 * extension "u-nu-thai" is automatically added. 361 * 362 * <h5>Serialization</h5> 363 * 364 * <p>During serialization, writeObject writes all fields to the output 365 * stream, including extensions. 366 * 367 * <p>During deserialization, readResolve adds extensions as described 368 * in <a href="#special_cases_constructor">Special Cases</a>, only 369 * for the two cases th_TH_TH and ja_JP_JP. 370 * 371 * <h5>Legacy language codes</h5> 372 * 373 * <p>Locale's constructor has always converted three language codes to 374 * their earlier, obsoleted forms: <tt>he</tt> maps to <tt>iw</tt>, 375 * <tt>yi</tt> maps to <tt>ji</tt>, and <tt>id</tt> maps to 376 * <tt>in</tt>. This continues to be the case, in order to not break 377 * backwards compatibility. 378 * 379 * <p>The APIs added in 1.7 map between the old and new language codes, 380 * maintaining the old codes internal to Locale (so that 381 * <code>getLanguage</code> and <code>toString</code> reflect the old 382 * code), but using the new codes in the BCP 47 language tag APIs (so 383 * that <code>toLanguageTag</code> reflects the new one). This 384 * preserves the equivalence between Locales no matter which code or 385 * API is used to construct them. Java's default resource bundle 386 * lookup mechanism also implements this mapping, so that resources 387 * can be named using either convention, see {@link ResourceBundle.Control}. 388 * 389 * <h5>Three-letter language/country(region) codes</h5> 390 * 391 * <p>The Locale constructors have always specified that the language 392 * and the country param be two characters in length, although in 393 * practice they have accepted any length. The specification has now 394 * been relaxed to allow language codes of two to eight characters and 395 * country (region) codes of two to three characters, and in 396 * particular, three-letter language codes and three-digit region 397 * codes as specified in the IANA Language Subtag Registry. For 398 * compatibility, the implementation still does not impose a length 399 * constraint. 400 * 401 * @see Builder 402 * @see ResourceBundle 403 * @see java.text.Format 404 * @see java.text.NumberFormat 405 * @see java.text.Collator 406 * @author Mark Davis 407 * @since 1.1 408 */ 409public final class Locale implements Cloneable, Serializable { 410 411 static private final Cache LOCALECACHE = new Cache(); 412 413 /** Useful constant for language. 414 */ 415 static public final Locale ENGLISH = createConstant("en", ""); 416 417 /** Useful constant for language. 418 */ 419 static public final Locale FRENCH = createConstant("fr", ""); 420 421 /** Useful constant for language. 422 */ 423 static public final Locale GERMAN = createConstant("de", ""); 424 425 /** Useful constant for language. 426 */ 427 static public final Locale ITALIAN = createConstant("it", ""); 428 429 /** Useful constant for language. 430 */ 431 static public final Locale JAPANESE = createConstant("ja", ""); 432 433 /** Useful constant for language. 434 */ 435 static public final Locale KOREAN = createConstant("ko", ""); 436 437 /** Useful constant for language. 438 */ 439 static public final Locale CHINESE = createConstant("zh", ""); 440 441 /** Useful constant for language. 442 */ 443 static public final Locale SIMPLIFIED_CHINESE = createConstant("zh", "CN"); 444 445 /** Useful constant for language. 446 */ 447 static public final Locale TRADITIONAL_CHINESE = createConstant("zh", "TW"); 448 449 /** Useful constant for country. 450 */ 451 static public final Locale FRANCE = createConstant("fr", "FR"); 452 453 /** Useful constant for country. 454 */ 455 static public final Locale GERMANY = createConstant("de", "DE"); 456 457 /** Useful constant for country. 458 */ 459 static public final Locale ITALY = createConstant("it", "IT"); 460 461 /** Useful constant for country. 462 */ 463 static public final Locale JAPAN = createConstant("ja", "JP"); 464 465 /** Useful constant for country. 466 */ 467 static public final Locale KOREA = createConstant("ko", "KR"); 468 469 /** Useful constant for country. 470 */ 471 static public final Locale CHINA = SIMPLIFIED_CHINESE; 472 473 /** Useful constant for country. 474 */ 475 static public final Locale PRC = SIMPLIFIED_CHINESE; 476 477 /** Useful constant for country. 478 */ 479 static public final Locale TAIWAN = TRADITIONAL_CHINESE; 480 481 /** Useful constant for country. 482 */ 483 static public final Locale UK = createConstant("en", "GB"); 484 485 /** Useful constant for country. 486 */ 487 static public final Locale US = createConstant("en", "US"); 488 489 /** Useful constant for country. 490 */ 491 static public final Locale CANADA = createConstant("en", "CA"); 492 493 /** Useful constant for country. 494 */ 495 static public final Locale CANADA_FRENCH = createConstant("fr", "CA"); 496 497 /** 498 * ISO 639-3 generic code for undetermined languages. 499 */ 500 private static final String UNDETERMINED_LANGUAGE = "und"; 501 502 /** 503 * Useful constant for the root locale. The root locale is the locale whose 504 * language, country, and variant are empty ("") strings. This is regarded 505 * as the base locale of all locales, and is used as the language/country 506 * neutral locale for the locale sensitive operations. 507 * 508 * @since 1.6 509 */ 510 static public final Locale ROOT = createConstant("", ""); 511 512 /** 513 * The key for the private use extension ('x'). 514 * 515 * @see #getExtension(char) 516 * @see Builder#setExtension(char, String) 517 * @since 1.7 518 */ 519 static public final char PRIVATE_USE_EXTENSION = 'x'; 520 521 /** 522 * The key for Unicode locale extension ('u'). 523 * 524 * @see #getExtension(char) 525 * @see Builder#setExtension(char, String) 526 * @since 1.7 527 */ 528 static public final char UNICODE_LOCALE_EXTENSION = 'u'; 529 530 /** serialization ID 531 */ 532 static final long serialVersionUID = 9149081749638150636L; 533 534 /** 535 * Display types for retrieving localized names from the name providers. 536 */ 537 private static final int DISPLAY_LANGUAGE = 0; 538 private static final int DISPLAY_COUNTRY = 1; 539 private static final int DISPLAY_VARIANT = 2; 540 private static final int DISPLAY_SCRIPT = 3; 541 542 /** 543 * Private constructor used by getInstance method 544 */ 545 private Locale(BaseLocale baseLocale, LocaleExtensions extensions) { 546 this.baseLocale = baseLocale; 547 this.localeExtensions = extensions; 548 } 549 550 /** 551 * Construct a locale from language, country and variant. 552 * This constructor normalizes the language value to lowercase and 553 * the country value to uppercase. 554 * <p> 555 * <b>Note:</b> 556 * <ul> 557 * <li>ISO 639 is not a stable standard; some of the language codes it defines 558 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 559 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 560 * API on Locale will return only the OLD codes. 561 * <li>For backward compatibility reasons, this constructor does not make 562 * any syntactic checks on the input. 563 * <li>The two cases ("ja", "JP", "JP") and ("th", "TH", "TH") are handled specially, 564 * see <a href="#special_cases_constructor">Special Cases</a> for more information. 565 * </ul> 566 * 567 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 568 * up to 8 characters in length. See the <code>Locale</code> class description about 569 * valid language values. 570 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code. 571 * See the <code>Locale</code> class description about valid country values. 572 * @param variant Any arbitrary value used to indicate a variation of a <code>Locale</code>. 573 * See the <code>Locale</code> class description for the details. 574 * @exception NullPointerException thrown if any argument is null. 575 */ 576 public Locale(String language, String country, String variant) { 577 if (language== null || country == null || variant == null) { 578 throw new NullPointerException(); 579 } 580 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), "", country, variant); 581 localeExtensions = getCompatibilityExtensions(language, "", country, variant); 582 } 583 584 /** 585 * Construct a locale from language and country. 586 * This constructor normalizes the language value to lowercase and 587 * the country value to uppercase. 588 * <p> 589 * <b>Note:</b> 590 * <ul> 591 * <li>ISO 639 is not a stable standard; some of the language codes it defines 592 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 593 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 594 * API on Locale will return only the OLD codes. 595 * <li>For backward compatibility reasons, this constructor does not make 596 * any syntactic checks on the input. 597 * </ul> 598 * 599 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 600 * up to 8 characters in length. See the <code>Locale</code> class description about 601 * valid language values. 602 * @param country An ISO 3166 alpha-2 country code or a UN M.49 numeric-3 area code. 603 * See the <code>Locale</code> class description about valid country values. 604 * @exception NullPointerException thrown if either argument is null. 605 */ 606 public Locale(String language, String country) { 607 this(language, country, ""); 608 } 609 610 /** 611 * Construct a locale from a language code. 612 * This constructor normalizes the language value to lowercase. 613 * <p> 614 * <b>Note:</b> 615 * <ul> 616 * <li>ISO 639 is not a stable standard; some of the language codes it defines 617 * (specifically "iw", "ji", and "in") have changed. This constructor accepts both the 618 * old codes ("iw", "ji", and "in") and the new codes ("he", "yi", and "id"), but all other 619 * API on Locale will return only the OLD codes. 620 * <li>For backward compatibility reasons, this constructor does not make 621 * any syntactic checks on the input. 622 * </ul> 623 * 624 * @param language An ISO 639 alpha-2 or alpha-3 language code, or a language subtag 625 * up to 8 characters in length. See the <code>Locale</code> class description about 626 * valid language values. 627 * @exception NullPointerException thrown if argument is null. 628 * @since 1.4 629 */ 630 public Locale(String language) { 631 this(language, "", ""); 632 } 633 634 /** 635 * This method must be called only for creating the Locale.* 636 * constants due to making shortcuts. 637 */ 638 private static Locale createConstant(String lang, String country) { 639 BaseLocale base = BaseLocale.createInstance(lang, country); 640 return getInstance(base, null); 641 } 642 643 /** 644 * Returns a <code>Locale</code> constructed from the given 645 * <code>language</code>, <code>country</code> and 646 * <code>variant</code>. If the same <code>Locale</code> instance 647 * is available in the cache, then that instance is 648 * returned. Otherwise, a new <code>Locale</code> instance is 649 * created and cached. 650 * 651 * @param language lowercase 2 to 8 language code. 652 * @param country uppercase two-letter ISO-3166 code and numric-3 UN M.49 area code. 653 * @param variant vendor and browser specific code. See class description. 654 * @return the <code>Locale</code> instance requested 655 * @exception NullPointerException if any argument is null. 656 */ 657 static Locale getInstance(String language, String country, String variant) { 658 return getInstance(language, "", country, variant, null); 659 } 660 661 static Locale getInstance(String language, String script, String country, 662 String variant, LocaleExtensions extensions) { 663 if (language== null || script == null || country == null || variant == null) { 664 throw new NullPointerException(); 665 } 666 667 if (extensions == null) { 668 extensions = getCompatibilityExtensions(language, script, country, variant); 669 } 670 671 BaseLocale baseloc = BaseLocale.getInstance(language, script, country, variant); 672 return getInstance(baseloc, extensions); 673 } 674 675 static Locale getInstance(BaseLocale baseloc, LocaleExtensions extensions) { 676 LocaleKey key = new LocaleKey(baseloc, extensions); 677 return LOCALECACHE.get(key); 678 } 679 680 private static class Cache extends LocaleObjectCache<LocaleKey, Locale> { 681 private Cache() { 682 } 683 684 @Override 685 protected Locale createObject(LocaleKey key) { 686 return new Locale(key.base, key.exts); 687 } 688 } 689 690 private static final class LocaleKey { 691 private final BaseLocale base; 692 private final LocaleExtensions exts; 693 private final int hash; 694 695 private LocaleKey(BaseLocale baseLocale, LocaleExtensions extensions) { 696 base = baseLocale; 697 exts = extensions; 698 699 // Calculate the hash value here because it's always used. 700 int h = base.hashCode(); 701 if (exts != null) { 702 h ^= exts.hashCode(); 703 } 704 hash = h; 705 } 706 707 @Override 708 public boolean equals(Object obj) { 709 if (this == obj) { 710 return true; 711 } 712 if (!(obj instanceof LocaleKey)) { 713 return false; 714 } 715 LocaleKey other = (LocaleKey)obj; 716 if (hash != other.hash || !base.equals(other.base)) { 717 return false; 718 } 719 if (exts == null) { 720 return other.exts == null; 721 } 722 return exts.equals(other.exts); 723 } 724 725 @Override 726 public int hashCode() { 727 return hash; 728 } 729 } 730 731 /** 732 * Gets the current value of the default locale for this instance 733 * of the Java Virtual Machine. 734 * <p> 735 * The Java Virtual Machine sets the default locale during startup 736 * based on the host environment. It is used by many locale-sensitive 737 * methods if no locale is explicitly specified. 738 * It can be changed using the 739 * {@link #setDefault(java.util.Locale) setDefault} method. 740 * 741 * @return the default locale for this instance of the Java Virtual Machine 742 */ 743 public static Locale getDefault() { 744 // do not synchronize this method - see 4071298 745 // it's OK if more than one default locale happens to be created 746 if (defaultLocale == null) { 747 initDefault(); 748 } 749 return defaultLocale; 750 } 751 752 /** 753 * Gets the current value of the default locale for the specified Category 754 * for this instance of the Java Virtual Machine. 755 * <p> 756 * The Java Virtual Machine sets the default locale during startup based 757 * on the host environment. It is used by many locale-sensitive methods 758 * if no locale is explicitly specified. It can be changed using the 759 * setDefault(Locale.Category, Locale) method. 760 * 761 * @param category - the specified category to get the default locale 762 * @throws NullPointerException - if category is null 763 * @return the default locale for the specified Category for this instance 764 * of the Java Virtual Machine 765 * @see #setDefault(Locale.Category, Locale) 766 * @since 1.7 767 */ 768 public static Locale getDefault(Locale.Category category) { 769 // do not synchronize this method - see 4071298 770 // it's OK if more than one default locale happens to be created 771 switch (category) { 772 case DISPLAY: 773 if (defaultDisplayLocale == null) { 774 initDefault(category); 775 } 776 return defaultDisplayLocale; 777 case FORMAT: 778 if (defaultFormatLocale == null) { 779 initDefault(category); 780 } 781 return defaultFormatLocale; 782 default: 783 assert false: "Unknown Category"; 784 } 785 return getDefault(); 786 } 787 788 private static void initDefault() { 789 String language, region, script, country, variant; 790 language = AccessController.doPrivileged( 791 new GetPropertyAction("user.language", "en")); 792 // for compatibility, check for old user.region property 793 region = AccessController.doPrivileged( 794 new GetPropertyAction("user.region")); 795 if (region != null) { 796 // region can be of form country, country_variant, or _variant 797 int i = region.indexOf('_'); 798 if (i >= 0) { 799 country = region.substring(0, i); 800 variant = region.substring(i + 1); 801 } else { 802 country = region; 803 variant = ""; 804 } 805 script = ""; 806 } else { 807 script = AccessController.doPrivileged( 808 new GetPropertyAction("user.script", "")); 809 country = AccessController.doPrivileged( 810 new GetPropertyAction("user.country", "")); 811 variant = AccessController.doPrivileged( 812 new GetPropertyAction("user.variant", "")); 813 } 814 defaultLocale = getInstance(language, script, country, variant, null); 815 } 816 817 private static void initDefault(Locale.Category category) { 818 // make sure defaultLocale is initialized 819 if (defaultLocale == null) { 820 initDefault(); 821 } 822 823 Locale defaultCategoryLocale = getInstance( 824 AccessController.doPrivileged( 825 new GetPropertyAction(category.languageKey, defaultLocale.getLanguage())), 826 AccessController.doPrivileged( 827 new GetPropertyAction(category.scriptKey, defaultLocale.getScript())), 828 AccessController.doPrivileged( 829 new GetPropertyAction(category.countryKey, defaultLocale.getCountry())), 830 AccessController.doPrivileged( 831 new GetPropertyAction(category.variantKey, defaultLocale.getVariant())), 832 null); 833 834 switch (category) { 835 case DISPLAY: 836 defaultDisplayLocale = defaultCategoryLocale; 837 break; 838 case FORMAT: 839 defaultFormatLocale = defaultCategoryLocale; 840 break; 841 } 842 } 843 844 /** 845 * Sets the default locale for this instance of the Java Virtual Machine. 846 * This does not affect the host locale. 847 * <p> 848 * If there is a security manager, its <code>checkPermission</code> 849 * method is called with a <code>PropertyPermission("user.language", "write")</code> 850 * permission before the default locale is changed. 851 * <p> 852 * The Java Virtual Machine sets the default locale during startup 853 * based on the host environment. It is used by many locale-sensitive 854 * methods if no locale is explicitly specified. 855 * <p> 856 * Since changing the default locale may affect many different areas 857 * of functionality, this method should only be used if the caller 858 * is prepared to reinitialize locale-sensitive code running 859 * within the same Java Virtual Machine. 860 * <p> 861 * By setting the default locale with this method, all of the default 862 * locales for each Category are also set to the specified default locale. 863 * 864 * @throws SecurityException 865 * if a security manager exists and its 866 * <code>checkPermission</code> method doesn't allow the operation. 867 * @throws NullPointerException if <code>newLocale</code> is null 868 * @param newLocale the new default locale 869 * @see SecurityManager#checkPermission 870 * @see java.util.PropertyPermission 871 */ 872 public static synchronized void setDefault(Locale newLocale) { 873 setDefault(Category.DISPLAY, newLocale); 874 setDefault(Category.FORMAT, newLocale); 875 defaultLocale = newLocale; 876 ICU.setDefaultLocale(newLocale.toLanguageTag()); 877 } 878 879 /** 880 * Sets the default locale for the specified Category for this instance 881 * of the Java Virtual Machine. This does not affect the host locale. 882 * <p> 883 * If there is a security manager, its checkPermission method is called 884 * with a PropertyPermission("user.language", "write") permission before 885 * the default locale is changed. 886 * <p> 887 * The Java Virtual Machine sets the default locale during startup based 888 * on the host environment. It is used by many locale-sensitive methods 889 * if no locale is explicitly specified. 890 * <p> 891 * Since changing the default locale may affect many different areas of 892 * functionality, this method should only be used if the caller is 893 * prepared to reinitialize locale-sensitive code running within the 894 * same Java Virtual Machine. 895 * <p> 896 * 897 * @param category - the specified category to set the default locale 898 * @param newLocale - the new default locale 899 * @throws SecurityException - if a security manager exists and its 900 * checkPermission method doesn't allow the operation. 901 * @throws NullPointerException - if category and/or newLocale is null 902 * @see SecurityManager#checkPermission(java.security.Permission) 903 * @see PropertyPermission 904 * @see #getDefault(Locale.Category) 905 * @since 1.7 906 */ 907 public static synchronized void setDefault(Locale.Category category, 908 Locale newLocale) { 909 if (category == null) 910 throw new NullPointerException("Category cannot be NULL"); 911 if (newLocale == null) 912 throw new NullPointerException("Can't set default locale to NULL"); 913 914 SecurityManager sm = System.getSecurityManager(); 915 if (sm != null) sm.checkPermission(new PropertyPermission 916 ("user.language", "write")); 917 switch (category) { 918 case DISPLAY: 919 defaultDisplayLocale = newLocale; 920 break; 921 case FORMAT: 922 defaultFormatLocale = newLocale; 923 break; 924 default: 925 assert false: "Unknown Category"; 926 } 927 } 928 929 /** 930 * Returns an array of all installed locales. 931 * The returned array represents the union of locales supported 932 * by the Java runtime environment and by installed 933 * {@link java.util.spi.LocaleServiceProvider LocaleServiceProvider} 934 * implementations. It must contain at least a <code>Locale</code> 935 * instance equal to {@link java.util.Locale#US Locale.US}. 936 * 937 * @return An array of installed locales. 938 */ 939 public static Locale[] getAvailableLocales() { 940 return LocaleServiceProviderPool.getAllAvailableLocales(); 941 } 942 943 /** 944 * Returns a list of all 2-letter country codes defined in ISO 3166. 945 * Can be used to create Locales. 946 * <p> 947 * <b>Note:</b> The <code>Locale</code> class also supports other codes for 948 * country (region), such as 3-letter numeric UN M.49 area codes. 949 * Therefore, the list returned by this method does not contain ALL valid 950 * codes that can be used to create Locales. 951 */ 952 public static String[] getISOCountries() { 953 // Android-changed: Use ICU. 954 return ICU.getISOCountries(); 955 } 956 957 /** 958 * Returns a list of all 2-letter language codes defined in ISO 639. 959 * Can be used to create Locales. 960 * <p> 961 * <b>Note:</b> 962 * <ul> 963 * <li>ISO 639 is not a stable standard— some languages' codes have changed. 964 * The list this function returns includes both the new and the old codes for the 965 * languages whose codes have changed. 966 * <li>The <code>Locale</code> class also supports language codes up to 967 * 8 characters in length. Therefore, the list returned by this method does 968 * not contain ALL valid codes that can be used to create Locales. 969 * </ul> 970 */ 971 public static String[] getISOLanguages() { 972 // Android-changed: Use ICU. 973 return ICU.getISOLanguages(); 974 } 975 976 /** 977 * Returns the language code of this Locale. 978 * 979 * <p><b>Note:</b> ISO 639 is not a stable standard— some languages' codes have changed. 980 * Locale's constructor recognizes both the new and the old codes for the languages 981 * whose codes have changed, but this function always returns the old code. If you 982 * want to check for a specific language whose code has changed, don't do 983 * <pre> 984 * if (locale.getLanguage().equals("he")) // BAD! 985 * ... 986 * </pre> 987 * Instead, do 988 * <pre> 989 * if (locale.getLanguage().equals(new Locale("he").getLanguage())) 990 * ... 991 * </pre> 992 * @return The language code, or the empty string if none is defined. 993 * @see #getDisplayLanguage 994 */ 995 public String getLanguage() { 996 return baseLocale.getLanguage(); 997 } 998 999 /** 1000 * Returns the script for this locale, which should 1001 * either be the empty string or an ISO 15924 4-letter script 1002 * code. The first letter is uppercase and the rest are 1003 * lowercase, for example, 'Latn', 'Cyrl'. 1004 * 1005 * @return The script code, or the empty string if none is defined. 1006 * @see #getDisplayScript 1007 * @since 1.7 1008 */ 1009 public String getScript() { 1010 return baseLocale.getScript(); 1011 } 1012 1013 /** 1014 * Returns the country/region code for this locale, which should 1015 * either be the empty string, an uppercase ISO 3166 2-letter code, 1016 * or a UN M.49 3-digit code. 1017 * 1018 * @return The country/region code, or the empty string if none is defined. 1019 * @see #getDisplayCountry 1020 */ 1021 public String getCountry() { 1022 return baseLocale.getRegion(); 1023 } 1024 1025 /** 1026 * Returns the variant code for this locale. 1027 * 1028 * @return The variant code, or the empty string if none is defined. 1029 * @see #getDisplayVariant 1030 */ 1031 public String getVariant() { 1032 return baseLocale.getVariant(); 1033 } 1034 1035 /** 1036 * Returns the extension (or private use) value associated with 1037 * the specified key, or null if there is no extension 1038 * associated with the key. To be well-formed, the key must be one 1039 * of <code>[0-9A-Za-z]</code>. Keys are case-insensitive, so 1040 * for example 'z' and 'Z' represent the same extension. 1041 * 1042 * @param key the extension key 1043 * @return The extension, or null if this locale defines no 1044 * extension for the specified key. 1045 * @throws IllegalArgumentException if key is not well-formed 1046 * @see #PRIVATE_USE_EXTENSION 1047 * @see #UNICODE_LOCALE_EXTENSION 1048 * @since 1.7 1049 */ 1050 public String getExtension(char key) { 1051 if (!LocaleExtensions.isValidKey(key)) { 1052 throw new IllegalArgumentException("Ill-formed extension key: " + key); 1053 } 1054 return (localeExtensions == null) ? null : localeExtensions.getExtensionValue(key); 1055 } 1056 1057 /** 1058 * Returns the set of extension keys associated with this locale, or the 1059 * empty set if it has no extensions. The returned set is unmodifiable. 1060 * The keys will all be lower-case. 1061 * 1062 * @return The set of extension keys, or the empty set if this locale has 1063 * no extensions. 1064 * @since 1.7 1065 */ 1066 public Set<Character> getExtensionKeys() { 1067 if (localeExtensions == null) { 1068 return Collections.emptySet(); 1069 } 1070 return localeExtensions.getKeys(); 1071 } 1072 1073 /** 1074 * Returns the set of unicode locale attributes associated with 1075 * this locale, or the empty set if it has no attributes. The 1076 * returned set is unmodifiable. 1077 * 1078 * @return The set of attributes. 1079 * @since 1.7 1080 */ 1081 public Set<String> getUnicodeLocaleAttributes() { 1082 if (localeExtensions == null) { 1083 return Collections.emptySet(); 1084 } 1085 return localeExtensions.getUnicodeLocaleAttributes(); 1086 } 1087 1088 /** 1089 * Returns the Unicode locale type associated with the specified Unicode locale key 1090 * for this locale. Returns the empty string for keys that are defined with no type. 1091 * Returns null if the key is not defined. Keys are case-insensitive. The key must 1092 * be two alphanumeric characters ([0-9a-zA-Z]), or an IllegalArgumentException is 1093 * thrown. 1094 * 1095 * @param key the Unicode locale key 1096 * @return The Unicode locale type associated with the key, or null if the 1097 * locale does not define the key. 1098 * @throws IllegalArgumentException if the key is not well-formed 1099 * @throws NullPointerException if <code>key</code> is null 1100 * @since 1.7 1101 */ 1102 public String getUnicodeLocaleType(String key) { 1103 if (!UnicodeLocaleExtension.isKey(key)) { 1104 throw new IllegalArgumentException("Ill-formed Unicode locale key: " + key); 1105 } 1106 return (localeExtensions == null) ? null : localeExtensions.getUnicodeLocaleType(key); 1107 } 1108 1109 /** 1110 * Returns the set of Unicode locale keys defined by this locale, or the empty set if 1111 * this locale has none. The returned set is immutable. Keys are all lower case. 1112 * 1113 * @return The set of Unicode locale keys, or the empty set if this locale has 1114 * no Unicode locale keywords. 1115 * @since 1.7 1116 */ 1117 public Set<String> getUnicodeLocaleKeys() { 1118 if (localeExtensions == null) { 1119 return Collections.emptySet(); 1120 } 1121 return localeExtensions.getUnicodeLocaleKeys(); 1122 } 1123 1124 /** 1125 * Package locale method returning the Locale's BaseLocale, 1126 * used by ResourceBundle 1127 * @return base locale of this Locale 1128 */ 1129 BaseLocale getBaseLocale() { 1130 return baseLocale; 1131 } 1132 1133 /** 1134 * Package private method returning the Locale's LocaleExtensions, 1135 * used by ResourceBundle. 1136 * @return locale exnteions of this Locale, 1137 * or {@code null} if no extensions are defined 1138 */ 1139 LocaleExtensions getLocaleExtensions() { 1140 return localeExtensions; 1141 } 1142 1143 /** 1144 * Returns a string representation of this <code>Locale</code> 1145 * object, consisting of language, country, variant, script, 1146 * and extensions as below: 1147 * <p><blockquote> 1148 * language + "_" + country + "_" + (variant + "_#" | "#") + script + "-" + extensions 1149 * </blockquote> 1150 * 1151 * Language is always lower case, country is always upper case, script is always title 1152 * case, and extensions are always lower case. Extensions and private use subtags 1153 * will be in canonical order as explained in {@link #toLanguageTag}. 1154 * 1155 * <p>When the locale has neither script nor extensions, the result is the same as in 1156 * Java 6 and prior. 1157 * 1158 * <p>If both the language and country fields are missing, this function will return 1159 * the empty string, even if the variant, script, or extensions field is present (you 1160 * can't have a locale with just a variant, the variant must accompany a well-formed 1161 * language or country code). 1162 * 1163 * <p>If script or extensions are present and variant is missing, no underscore is 1164 * added before the "#". 1165 * 1166 * <p>This behavior is designed to support debugging and to be compatible with 1167 * previous uses of <code>toString</code> that expected language, country, and variant 1168 * fields only. To represent a Locale as a String for interchange purposes, use 1169 * {@link #toLanguageTag}. 1170 * 1171 * <p>Examples: <ul><tt> 1172 * <li>en 1173 * <li>de_DE 1174 * <li>_GB 1175 * <li>en_US_WIN 1176 * <li>de__POSIX 1177 * <li>zh_CN_#Hans 1178 * <li>zh_TW_#Hant-x-java 1179 * <li>th_TH_TH_#u-nu-thai</tt></ul> 1180 * 1181 * @return A string representation of the Locale, for debugging. 1182 * @see #getDisplayName 1183 * @see #toLanguageTag 1184 */ 1185 @Override 1186 public final String toString() { 1187 boolean l = (baseLocale.getLanguage().length() != 0); 1188 boolean s = (baseLocale.getScript().length() != 0); 1189 boolean r = (baseLocale.getRegion().length() != 0); 1190 boolean v = (baseLocale.getVariant().length() != 0); 1191 boolean e = (localeExtensions != null && localeExtensions.getID().length() != 0); 1192 1193 StringBuilder result = new StringBuilder(baseLocale.getLanguage()); 1194 if (r || (l && (v || s || e))) { 1195 result.append('_') 1196 .append(baseLocale.getRegion()); // This may just append '_' 1197 } 1198 if (v && (l || r)) { 1199 result.append('_') 1200 .append(baseLocale.getVariant()); 1201 } 1202 1203 if (s && (l || r)) { 1204 result.append("_#") 1205 .append(baseLocale.getScript()); 1206 } 1207 1208 if (e && (l || r)) { 1209 result.append('_'); 1210 if (!s) { 1211 result.append('#'); 1212 } 1213 result.append(localeExtensions.getID()); 1214 } 1215 1216 return result.toString(); 1217 } 1218 1219 /** 1220 * Returns a well-formed IETF BCP 47 language tag representing 1221 * this locale. 1222 * 1223 * <p>If this <code>Locale</code> has a language, country, or 1224 * variant that does not satisfy the IETF BCP 47 language tag 1225 * syntax requirements, this method handles these fields as 1226 * described below: 1227 * 1228 * <p><b>Language:</b> If language is empty, or not <a 1229 * href="#def_language" >well-formed</a> (for example "a" or 1230 * "e2"), it will be emitted as "und" (Undetermined). 1231 * 1232 * <p><b>Country:</b> If country is not <a 1233 * href="#def_region">well-formed</a> (for example "12" or "USA"), 1234 * it will be omitted. 1235 * 1236 * <p><b>Variant:</b> If variant <b>is</b> <a 1237 * href="#def_variant">well-formed</a>, each sub-segment 1238 * (delimited by '-' or '_') is emitted as a subtag. Otherwise: 1239 * <ul> 1240 * 1241 * <li>if all sub-segments match <code>[0-9a-zA-Z]{1,8}</code> 1242 * (for example "WIN" or "Oracle_JDK_Standard_Edition"), the first 1243 * ill-formed sub-segment and all following will be appended to 1244 * the private use subtag. The first appended subtag will be 1245 * "lvariant", followed by the sub-segments in order, separated by 1246 * hyphen. For example, "x-lvariant-WIN", 1247 * "Oracle-x-lvariant-JDK-Standard-Edition". 1248 * 1249 * <li>if any sub-segment does not match 1250 * <code>[0-9a-zA-Z]{1,8}</code>, the variant will be truncated 1251 * and the problematic sub-segment and all following sub-segments 1252 * will be omitted. If the remainder is non-empty, it will be 1253 * emitted as a private use subtag as above (even if the remainder 1254 * turns out to be well-formed). For example, 1255 * "Solaris_isjustthecoolestthing" is emitted as 1256 * "x-lvariant-Solaris", not as "solaris".</li></ul> 1257 * 1258 * <p><b>Special Conversions:</b> Java supports some old locale 1259 * representations, including deprecated ISO language codes, 1260 * for compatibility. This method performs the following 1261 * conversions: 1262 * <ul> 1263 * 1264 * <li>Deprecated ISO language codes "iw", "ji", and "in" are 1265 * converted to "he", "yi", and "id", respectively. 1266 * 1267 * <li>A locale with language "no", country "NO", and variant 1268 * "NY", representing Norwegian Nynorsk (Norway), is converted 1269 * to a language tag "nn-NO".</li></ul> 1270 * 1271 * <p><b>Note:</b> Although the language tag created by this 1272 * method is well-formed (satisfies the syntax requirements 1273 * defined by the IETF BCP 47 specification), it is not 1274 * necessarily a valid BCP 47 language tag. For example, 1275 * <pre> 1276 * new Locale("xx", "YY").toLanguageTag();</pre> 1277 * 1278 * will return "xx-YY", but the language subtag "xx" and the 1279 * region subtag "YY" are invalid because they are not registered 1280 * in the IANA Language Subtag Registry. 1281 * 1282 * @return a BCP47 language tag representing the locale 1283 * @see #forLanguageTag(String) 1284 * @since 1.7 1285 */ 1286 public String toLanguageTag() { 1287 LanguageTag tag = LanguageTag.parseLocale(baseLocale, localeExtensions); 1288 StringBuilder buf = new StringBuilder(); 1289 1290 String subtag = tag.getLanguage(); 1291 if (subtag.length() > 0) { 1292 buf.append(LanguageTag.canonicalizeLanguage(subtag)); 1293 } 1294 1295 subtag = tag.getScript(); 1296 if (subtag.length() > 0) { 1297 buf.append(LanguageTag.SEP); 1298 buf.append(LanguageTag.canonicalizeScript(subtag)); 1299 } 1300 1301 subtag = tag.getRegion(); 1302 if (subtag.length() > 0) { 1303 buf.append(LanguageTag.SEP); 1304 buf.append(LanguageTag.canonicalizeRegion(subtag)); 1305 } 1306 1307 List<String>subtags = tag.getVariants(); 1308 for (String s : subtags) { 1309 buf.append(LanguageTag.SEP); 1310 // preserve casing 1311 buf.append(s); 1312 } 1313 1314 subtags = tag.getExtensions(); 1315 for (String s : subtags) { 1316 buf.append(LanguageTag.SEP); 1317 buf.append(LanguageTag.canonicalizeExtension(s)); 1318 } 1319 1320 subtag = tag.getPrivateuse(); 1321 if (subtag.length() > 0) { 1322 if (buf.length() > 0) { 1323 buf.append(LanguageTag.SEP); 1324 } 1325 buf.append(LanguageTag.PRIVATEUSE).append(LanguageTag.SEP); 1326 // preserve casing 1327 buf.append(subtag); 1328 } 1329 1330 return buf.toString(); 1331 } 1332 1333 /** 1334 * Returns a locale for the specified IETF BCP 47 language tag string. 1335 * 1336 * <p>If the specified language tag contains any ill-formed subtags, 1337 * the first such subtag and all following subtags are ignored. Compare 1338 * to {@link Locale.Builder#setLanguageTag} which throws an exception 1339 * in this case. 1340 * 1341 * <p>The following <b>conversions</b> are performed:<ul> 1342 * 1343 * <li>The language code "und" is mapped to language "". 1344 * 1345 * <li>The language codes "he", "yi", and "id" are mapped to "iw", 1346 * "ji", and "in" respectively. (This is the same canonicalization 1347 * that's done in Locale's constructors.) 1348 * 1349 * <li>The portion of a private use subtag prefixed by "lvariant", 1350 * if any, is removed and appended to the variant field in the 1351 * result locale (without case normalization). If it is then 1352 * empty, the private use subtag is discarded: 1353 * 1354 * <pre> 1355 * Locale loc; 1356 * loc = Locale.forLanguageTag("en-US-x-lvariant-POSIX"); 1357 * loc.getVariant(); // returns "POSIX" 1358 * loc.getExtension('x'); // returns null 1359 * 1360 * loc = Locale.forLanguageTag("de-POSIX-x-URP-lvariant-Abc-Def"); 1361 * loc.getVariant(); // returns "POSIX_Abc_Def" 1362 * loc.getExtension('x'); // returns "urp" 1363 * </pre> 1364 * 1365 * <li>When the languageTag argument contains an extlang subtag, 1366 * the first such subtag is used as the language, and the primary 1367 * language subtag and other extlang subtags are ignored: 1368 * 1369 * <pre> 1370 * Locale.forLanguageTag("ar-aao").getLanguage(); // returns "aao" 1371 * Locale.forLanguageTag("en-abc-def-us").toString(); // returns "abc_US" 1372 * </pre> 1373 * 1374 * <li>Case is normalized except for variant tags, which are left 1375 * unchanged. Language is normalized to lower case, script to 1376 * title case, country to upper case, and extensions to lower 1377 * case. 1378 * 1379 * <li>If, after processing, the locale would exactly match either 1380 * ja_JP_JP or th_TH_TH with no extensions, the appropriate 1381 * extensions are added as though the constructor had been called: 1382 * 1383 * <pre> 1384 * Locale.forLanguageTag("ja-JP-x-lvariant-JP").toLanguageTag(); 1385 * // returns "ja-JP-u-ca-japanese-x-lvariant-JP" 1386 * Locale.forLanguageTag("th-TH-x-lvariant-TH").toLanguageTag(); 1387 * // returns "th-TH-u-nu-thai-x-lvariant-TH" 1388 * <pre></ul> 1389 * 1390 * <p>This implements the 'Language-Tag' production of BCP47, and 1391 * so supports grandfathered (regular and irregular) as well as 1392 * private use language tags. Stand alone private use tags are 1393 * represented as empty language and extension 'x-whatever', 1394 * and grandfathered tags are converted to their canonical replacements 1395 * where they exist. 1396 * 1397 * <p>Grandfathered tags with canonical replacements are as follows: 1398 * 1399 * <table> 1400 * <tbody align="center"> 1401 * <tr><th>grandfathered tag</th><th> </th><th>modern replacement</th></tr> 1402 * <tr><td>art-lojban</td><td> </td><td>jbo</td></tr> 1403 * <tr><td>i-ami</td><td> </td><td>ami</td></tr> 1404 * <tr><td>i-bnn</td><td> </td><td>bnn</td></tr> 1405 * <tr><td>i-hak</td><td> </td><td>hak</td></tr> 1406 * <tr><td>i-klingon</td><td> </td><td>tlh</td></tr> 1407 * <tr><td>i-lux</td><td> </td><td>lb</td></tr> 1408 * <tr><td>i-navajo</td><td> </td><td>nv</td></tr> 1409 * <tr><td>i-pwn</td><td> </td><td>pwn</td></tr> 1410 * <tr><td>i-tao</td><td> </td><td>tao</td></tr> 1411 * <tr><td>i-tay</td><td> </td><td>tay</td></tr> 1412 * <tr><td>i-tsu</td><td> </td><td>tsu</td></tr> 1413 * <tr><td>no-bok</td><td> </td><td>nb</td></tr> 1414 * <tr><td>no-nyn</td><td> </td><td>nn</td></tr> 1415 * <tr><td>sgn-BE-FR</td><td> </td><td>sfb</td></tr> 1416 * <tr><td>sgn-BE-NL</td><td> </td><td>vgt</td></tr> 1417 * <tr><td>sgn-CH-DE</td><td> </td><td>sgg</td></tr> 1418 * <tr><td>zh-guoyu</td><td> </td><td>cmn</td></tr> 1419 * <tr><td>zh-hakka</td><td> </td><td>hak</td></tr> 1420 * <tr><td>zh-min-nan</td><td> </td><td>nan</td></tr> 1421 * <tr><td>zh-xiang</td><td> </td><td>hsn</td></tr> 1422 * </tbody> 1423 * </table> 1424 * 1425 * <p>Grandfathered tags with no modern replacement will be 1426 * converted as follows: 1427 * 1428 * <table> 1429 * <tbody align="center"> 1430 * <tr><th>grandfathered tag</th><th> </th><th>converts to</th></tr> 1431 * <tr><td>cel-gaulish</td><td> </td><td>xtg-x-cel-gaulish</td></tr> 1432 * <tr><td>en-GB-oed</td><td> </td><td>en-GB-x-oed</td></tr> 1433 * <tr><td>i-default</td><td> </td><td>en-x-i-default</td></tr> 1434 * <tr><td>i-enochian</td><td> </td><td>und-x-i-enochian</td></tr> 1435 * <tr><td>i-mingo</td><td> </td><td>see-x-i-mingo</td></tr> 1436 * <tr><td>zh-min</td><td> </td><td>nan-x-zh-min</td></tr> 1437 * </tbody> 1438 * </table> 1439 * 1440 * <p>For a list of all grandfathered tags, see the 1441 * IANA Language Subtag Registry (search for "Type: grandfathered"). 1442 * 1443 * <p><b>Note</b>: there is no guarantee that <code>toLanguageTag</code> 1444 * and <code>forLanguageTag</code> will round-trip. 1445 * 1446 * @param languageTag the language tag 1447 * @return The locale that best represents the language tag. 1448 * @throws NullPointerException if <code>languageTag</code> is <code>null</code> 1449 * @see #toLanguageTag() 1450 * @see java.util.Locale.Builder#setLanguageTag(String) 1451 * @since 1.7 1452 */ 1453 public static Locale forLanguageTag(String languageTag) { 1454 LanguageTag tag = LanguageTag.parse(languageTag, null); 1455 InternalLocaleBuilder bldr = new InternalLocaleBuilder(); 1456 bldr.setLanguageTag(tag); 1457 BaseLocale base = bldr.getBaseLocale(); 1458 LocaleExtensions exts = bldr.getLocaleExtensions(); 1459 if (exts == null && base.getVariant().length() > 0) { 1460 exts = getCompatibilityExtensions(base.getLanguage(), base.getScript(), 1461 base.getRegion(), base.getVariant()); 1462 } 1463 return getInstance(base, exts); 1464 } 1465 1466 /** 1467 * Returns a three-letter abbreviation of this locale's language. 1468 * If the language matches an ISO 639-1 two-letter code, the 1469 * corresponding ISO 639-2/T three-letter lowercase code is 1470 * returned. The ISO 639-2 language codes can be found on-line, 1471 * see "Codes for the Representation of Names of Languages Part 2: 1472 * Alpha-3 Code". If the locale specifies a three-letter 1473 * language, the language is returned as is. If the locale does 1474 * not specify a language the empty string is returned. 1475 * 1476 * @return A three-letter abbreviation of this locale's language. 1477 * @exception MissingResourceException Throws MissingResourceException if 1478 * three-letter language abbreviation is not available for this locale. 1479 */ 1480 public String getISO3Language() throws MissingResourceException { 1481 // Android-changed: Use ICU.getIso3Language. Also return "" for empty languages 1482 // for the sake of backwards compatibility. 1483 String lang = baseLocale.getLanguage(); 1484 if (lang.length() == 3) { 1485 return lang; 1486 } else if (lang.isEmpty()) { 1487 return ""; 1488 } 1489 1490 String language3 = ICU.getISO3Language(lang); 1491 if (language3.isEmpty()) { 1492 throw new MissingResourceException("Couldn't find 3-letter language code for " 1493 + lang, "FormatData_" + toString(), "ShortLanguage"); 1494 } 1495 1496 return language3; 1497 } 1498 1499 /** 1500 * Returns a three-letter abbreviation for this locale's country. 1501 * If the country matches an ISO 3166-1 alpha-2 code, the 1502 * corresponding ISO 3166-1 alpha-3 uppercase code is returned. 1503 * If the locale doesn't specify a country, this will be the empty 1504 * string. 1505 * 1506 * <p>The ISO 3166-1 codes can be found on-line. 1507 * 1508 * @return A three-letter abbreviation of this locale's country. 1509 * @exception MissingResourceException Throws MissingResourceException if the 1510 * three-letter country abbreviation is not available for this locale. 1511 */ 1512 public String getISO3Country() throws MissingResourceException { 1513 // Android changed: Use.getIso3Country. Also return "" for missing regions. 1514 final String region = baseLocale.getRegion(); 1515 // Note that this will return an UN.M49 region code 1516 if (region.length() == 3) { 1517 return baseLocale.getRegion(); 1518 } else if (region.isEmpty()) { 1519 return ""; 1520 } 1521 1522 // Prefix "en-" because ICU doesn't really care about what the language is. 1523 String country3 = ICU.getISO3Country("en-" + region); 1524 if (country3 == null) { 1525 throw new MissingResourceException("Couldn't find 3-letter country code for " 1526 + baseLocale.getRegion(), "FormatData_" + toString(), "ShortCountry"); 1527 } 1528 return country3; 1529 } 1530 1531 /** 1532 * Returns a name for the locale's language that is appropriate for display to the 1533 * user. 1534 * If possible, the name returned will be localized for the default locale. 1535 * For example, if the locale is fr_FR and the default locale 1536 * is en_US, getDisplayLanguage() will return "French"; if the locale is en_US and 1537 * the default locale is fr_FR, getDisplayLanguage() will return "anglais". 1538 * If the name returned cannot be localized for the default locale, 1539 * (say, we don't have a Japanese name for Croatian), 1540 * this function falls back on the English name, and uses the ISO code as a last-resort 1541 * value. If the locale doesn't specify a language, this function returns the empty string. 1542 */ 1543 public final String getDisplayLanguage() { 1544 return getDisplayLanguage(getDefault(Category.DISPLAY)); 1545 } 1546 1547 /** 1548 * Returns the name of this locale's language, localized to {@code locale}. 1549 * If the language name is unknown, the language code is returned. 1550 */ 1551 public String getDisplayLanguage(Locale locale) { 1552 String languageCode = baseLocale.getLanguage(); 1553 if (languageCode.isEmpty()) { 1554 return ""; 1555 } 1556 1557 // Hacks for backward compatibility. 1558 // 1559 // Our language tag will contain "und" if the languageCode is invalid 1560 // or missing. ICU will then return "langue indéterminée" or the equivalent 1561 // display language for the indeterminate language code. 1562 // 1563 // Sigh... ugh... and what not. 1564 final String normalizedLanguage = normalizeAndValidateLanguage( 1565 languageCode, false /* strict */); 1566 if (UNDETERMINED_LANGUAGE.equals(normalizedLanguage)) { 1567 return languageCode; 1568 } 1569 1570 // TODO: We need a new hack or a complete fix for http://b/8049507 --- We would 1571 // cover the frameworks' tracks when they were using "tl" instead of "fil". 1572 String result = ICU.getDisplayLanguage(this, locale); 1573 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1574 result = ICU.getDisplayLanguage(this, Locale.getDefault()); 1575 } 1576 return result; 1577 } 1578 1579 private static String normalizeAndValidateLanguage(String language, boolean strict) { 1580 if (language == null || language.isEmpty()) { 1581 return ""; 1582 } 1583 1584 final String lowercaseLanguage = language.toLowerCase(Locale.ROOT); 1585 if (!isValidBcp47Alpha(lowercaseLanguage, 2, 3)) { 1586 if (strict) { 1587 throw new IllformedLocaleException("Invalid language: " + language); 1588 } else { 1589 return UNDETERMINED_LANGUAGE; 1590 } 1591 } 1592 1593 return lowercaseLanguage; 1594 } 1595 1596 /* 1597 * Checks whether a given string is an ASCII alphanumeric string. 1598 */ 1599 private static boolean isAsciiAlphaNum(String string) { 1600 for (int i = 0; i < string.length(); i++) { 1601 final char character = string.charAt(i); 1602 if (!(character >= 'a' && character <= 'z' || 1603 character >= 'A' && character <= 'Z' || 1604 character >= '0' && character <= '9')) { 1605 return false; 1606 } 1607 } 1608 1609 return true; 1610 } 1611 1612 /** 1613 * Returns a name for the the locale's script that is appropriate for display to 1614 * the user. If possible, the name will be localized for the default locale. Returns 1615 * the empty string if this locale doesn't specify a script code. 1616 * 1617 * @return the display name of the script code for the current default locale 1618 * @since 1.7 1619 */ 1620 public String getDisplayScript() { 1621 return getDisplayScript(getDefault()); 1622 } 1623 1624 /** 1625 * Returns the name of this locale's script code, localized to {@link Locale}. If the 1626 * script code is unknown, the return value of this method is the same as that of 1627 * {@link #getScript()}. 1628 * 1629 * @since 1.7 1630 */ 1631 public String getDisplayScript(Locale locale) { 1632 String scriptCode = baseLocale.getScript(); 1633 if (scriptCode.isEmpty()) { 1634 return ""; 1635 } 1636 1637 String result = ICU.getDisplayScript(this, locale); 1638 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1639 result = ICU.getDisplayScript(this, Locale.getDefault()); 1640 } 1641 1642 return result; 1643 1644 } 1645 1646 /** 1647 * Returns a name for the locale's country that is appropriate for display to the 1648 * user. 1649 * If possible, the name returned will be localized for the default locale. 1650 * For example, if the locale is fr_FR and the default locale 1651 * is en_US, getDisplayCountry() will return "France"; if the locale is en_US and 1652 * the default locale is fr_FR, getDisplayCountry() will return "Etats-Unis". 1653 * If the name returned cannot be localized for the default locale, 1654 * (say, we don't have a Japanese name for Croatia), 1655 * this function falls back on the English name, and uses the ISO code as a last-resort 1656 * value. If the locale doesn't specify a country, this function returns the empty string. 1657 */ 1658 public final String getDisplayCountry() { 1659 return getDisplayCountry(getDefault(Category.DISPLAY)); 1660 } 1661 /** 1662 * Returns the name of this locale's country, localized to {@code locale}. 1663 * Returns the empty string if this locale does not correspond to a specific 1664 * country. 1665 */ 1666 public String getDisplayCountry(Locale locale) { 1667 String countryCode = baseLocale.getRegion(); 1668 if (countryCode.isEmpty()) { 1669 return ""; 1670 } 1671 1672 final String normalizedRegion = normalizeAndValidateRegion( 1673 countryCode, false /* strict */); 1674 if (normalizedRegion.isEmpty()) { 1675 return countryCode; 1676 } 1677 1678 String result = ICU.getDisplayCountry(this, locale); 1679 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1680 result = ICU.getDisplayCountry(this, Locale.getDefault()); 1681 } 1682 return result; 1683 } 1684 1685 private static String normalizeAndValidateRegion(String region, boolean strict) { 1686 if (region == null || region.isEmpty()) { 1687 return ""; 1688 } 1689 1690 final String uppercaseRegion = region.toUpperCase(Locale.ROOT); 1691 if (!isValidBcp47Alpha(uppercaseRegion, 2, 2) && 1692 !isUnM49AreaCode(uppercaseRegion)) { 1693 if (strict) { 1694 throw new IllformedLocaleException("Invalid region: " + region); 1695 } else { 1696 return ""; 1697 } 1698 } 1699 1700 return uppercaseRegion; 1701 } 1702 1703 private static boolean isValidBcp47Alpha(String string, int lowerBound, int upperBound) { 1704 final int length = string.length(); 1705 if (length < lowerBound || length > upperBound) { 1706 return false; 1707 } 1708 1709 for (int i = 0; i < length; ++i) { 1710 final char character = string.charAt(i); 1711 if (!(character >= 'a' && character <= 'z' || 1712 character >= 'A' && character <= 'Z')) { 1713 return false; 1714 } 1715 } 1716 1717 return true; 1718 } 1719 1720 /** 1721 * A UN M.49 is a 3 digit numeric code. 1722 */ 1723 private static boolean isUnM49AreaCode(String code) { 1724 if (code.length() != 3) { 1725 return false; 1726 } 1727 1728 for (int i = 0; i < 3; ++i) { 1729 final char character = code.charAt(i); 1730 if (!(character >= '0' && character <= '9')) { 1731 return false; 1732 } 1733 } 1734 1735 return true; 1736 } 1737 1738 /** 1739 * Returns a name for the locale's variant code that is appropriate for display to the 1740 * user. If possible, the name will be localized for the default locale. If the locale 1741 * doesn't specify a variant code, this function returns the empty string. 1742 */ 1743 public final String getDisplayVariant() { 1744 return getDisplayVariant(getDefault(Category.DISPLAY)); 1745 } 1746 1747 /** 1748 * Returns the full variant name in the specified {@code Locale} for the variant code 1749 * of this {@code Locale}. If there is no matching variant name, the variant code is 1750 * returned. 1751 * 1752 * @since 1.7 1753 */ 1754 public String getDisplayVariant(Locale locale) { 1755 String variantCode = baseLocale.getVariant(); 1756 if (variantCode.isEmpty()) { 1757 return ""; 1758 } 1759 1760 try { 1761 normalizeAndValidateVariant(variantCode); 1762 } catch (IllformedLocaleException ilfe) { 1763 return variantCode; 1764 } 1765 1766 String result = ICU.getDisplayVariant(this, locale); 1767 if (result == null) { // TODO: do we need to do this, or does ICU do it for us? 1768 result = ICU.getDisplayVariant(this, Locale.getDefault()); 1769 } 1770 1771 // The "old style" locale constructors allow us to pass in variants that aren't 1772 // valid BCP-47 variant subtags. When that happens, toLanguageTag will not emit 1773 // them. Note that we know variantCode.length() > 0 due to the isEmpty check at 1774 // the beginning of this function. 1775 if (result.isEmpty()) { 1776 return variantCode; 1777 } 1778 return result; 1779 } 1780 1781 private static String normalizeAndValidateVariant(String variant) { 1782 if (variant == null || variant.isEmpty()) { 1783 return ""; 1784 } 1785 1786 // Note that unlike extensions, we canonicalize to lower case alphabets 1787 // and underscores instead of hyphens. 1788 final String normalizedVariant = variant.replace('-', '_'); 1789 String[] subTags = normalizedVariant.split("_"); 1790 1791 for (String subTag : subTags) { 1792 if (!isValidVariantSubtag(subTag)) { 1793 throw new IllformedLocaleException("Invalid variant: " + variant); 1794 } 1795 } 1796 1797 return normalizedVariant; 1798 } 1799 1800 private static boolean isValidVariantSubtag(String subTag) { 1801 // The BCP-47 spec states that : 1802 // - Subtags can be between [5, 8] alphanumeric chars in length. 1803 // - Subtags that start with a number are allowed to be 4 chars in length. 1804 if (subTag.length() >= 5 && subTag.length() <= 8) { 1805 if (isAsciiAlphaNum(subTag)) { 1806 return true; 1807 } 1808 } else if (subTag.length() == 4) { 1809 final char firstChar = subTag.charAt(0); 1810 if ((firstChar >= '0' && firstChar <= '9') && isAsciiAlphaNum(subTag)) { 1811 return true; 1812 } 1813 } 1814 1815 return false; 1816 } 1817 1818 /** 1819 * Returns a name for the locale that is appropriate for display to the 1820 * user. This will be the values returned by getDisplayLanguage(), 1821 * getDisplayScript(), getDisplayCountry(), and getDisplayVariant() assembled 1822 * into a single string. The the non-empty values are used in order, 1823 * with the second and subsequent names in parentheses. For example: 1824 * <blockquote> 1825 * language (script, country, variant)<br> 1826 * language (country)<br> 1827 * language (variant)<br> 1828 * script (country)<br> 1829 * country<br> 1830 * </blockquote> 1831 * depending on which fields are specified in the locale. If the 1832 * language, sacript, country, and variant fields are all empty, 1833 * this function returns the empty string. 1834 */ 1835 public final String getDisplayName() { 1836 return getDisplayName(getDefault(Category.DISPLAY)); 1837 } 1838 1839 /** 1840 * Returns this locale's language name, country name, and variant, localized 1841 * to {@code locale}. The exact output form depends on whether this locale 1842 * corresponds to a specific language, script, country and variant. 1843 * 1844 * <p>For example: 1845 * <ul> 1846 * <li>{@code new Locale("en").getDisplayName(Locale.US)} -> {@code English} 1847 * <li>{@code new Locale("en", "US").getDisplayName(Locale.US)} -> {@code English (United States)} 1848 * <li>{@code new Locale("en", "US", "POSIX").getDisplayName(Locale.US)} -> {@code English (United States,Computer)} 1849 * <li>{@code Locale.fromLanguageTag("zh-Hant-CN").getDisplayName(Locale.US)} -> {@code Chinese (Traditional Han,China)} 1850 * <li>{@code new Locale("en").getDisplayName(Locale.FRANCE)} -> {@code anglais} 1851 * <li>{@code new Locale("en", "US").getDisplayName(Locale.FRANCE)} -> {@code anglais (États-Unis)} 1852 * <li>{@code new Locale("en", "US", "POSIX").getDisplayName(Locale.FRANCE)} -> {@code anglais (États-Unis,informatique)}. 1853 * </ul> 1854 */ 1855 public String getDisplayName(Locale locale) { 1856 int count = 0; 1857 StringBuilder buffer = new StringBuilder(); 1858 String languageCode = baseLocale.getLanguage(); 1859 if (!languageCode.isEmpty()) { 1860 String displayLanguage = getDisplayLanguage(locale); 1861 buffer.append(displayLanguage.isEmpty() ? languageCode : displayLanguage); 1862 ++count; 1863 } 1864 String scriptCode = baseLocale.getScript(); 1865 if (!scriptCode.isEmpty()) { 1866 if (count == 1) { 1867 buffer.append(" ("); 1868 } 1869 String displayScript = getDisplayScript(locale); 1870 buffer.append(displayScript.isEmpty() ? scriptCode : displayScript); 1871 ++count; 1872 } 1873 String countryCode = baseLocale.getRegion(); 1874 if (!countryCode.isEmpty()) { 1875 if (count == 1) { 1876 buffer.append(" ("); 1877 } else if (count == 2) { 1878 buffer.append(","); 1879 } 1880 String displayCountry = getDisplayCountry(locale); 1881 buffer.append(displayCountry.isEmpty() ? countryCode : displayCountry); 1882 ++count; 1883 } 1884 String variantCode = baseLocale.getVariant(); 1885 if (!variantCode.isEmpty()) { 1886 if (count == 1) { 1887 buffer.append(" ("); 1888 } else if (count == 2 || count == 3) { 1889 buffer.append(","); 1890 } 1891 String displayVariant = getDisplayVariant(locale); 1892 buffer.append(displayVariant.isEmpty() ? variantCode : displayVariant); 1893 ++count; 1894 } 1895 if (count > 1) { 1896 buffer.append(")"); 1897 } 1898 return buffer.toString(); 1899 } 1900 1901 /** 1902 * Overrides Cloneable. 1903 */ 1904 public Object clone() 1905 { 1906 try { 1907 Locale that = (Locale)super.clone(); 1908 return that; 1909 } catch (CloneNotSupportedException e) { 1910 throw new InternalError(); 1911 } 1912 } 1913 1914 /** 1915 * Override hashCode. 1916 * Since Locales are often used in hashtables, caches the value 1917 * for speed. 1918 */ 1919 @Override 1920 public int hashCode() { 1921 int hc = hashCodeValue; 1922 if (hc == 0) { 1923 hc = baseLocale.hashCode(); 1924 if (localeExtensions != null) { 1925 hc ^= localeExtensions.hashCode(); 1926 } 1927 hashCodeValue = hc; 1928 } 1929 return hc; 1930 } 1931 1932 // Overrides 1933 1934 /** 1935 * Returns true if this Locale is equal to another object. A Locale is 1936 * deemed equal to another Locale with identical language, script, country, 1937 * variant and extensions, and unequal to all other objects. 1938 * 1939 * @return true if this Locale is equal to the specified object. 1940 */ 1941 @Override 1942 public boolean equals(Object obj) { 1943 if (this == obj) // quick check 1944 return true; 1945 if (!(obj instanceof Locale)) 1946 return false; 1947 BaseLocale otherBase = ((Locale)obj).baseLocale; 1948 if (!baseLocale.equals(otherBase)) { 1949 return false; 1950 } 1951 if (localeExtensions == null) { 1952 return ((Locale)obj).localeExtensions == null; 1953 } 1954 return localeExtensions.equals(((Locale)obj).localeExtensions); 1955 } 1956 1957 // ================= privates ===================================== 1958 1959 private transient BaseLocale baseLocale; 1960 private transient LocaleExtensions localeExtensions; 1961 1962 /** 1963 * Calculated hashcode 1964 */ 1965 private transient volatile int hashCodeValue = 0; 1966 1967 private static Locale defaultLocale = null; 1968 private static Locale defaultDisplayLocale = null; 1969 private static Locale defaultFormatLocale = null; 1970 1971 /** 1972 * Format a list using given pattern strings. 1973 * If either of the patterns is null, then a the list is 1974 * formatted by concatenation with the delimiter ','. 1975 * @param stringList the list of strings to be formatted. 1976 * @param listPattern should create a MessageFormat taking 0-3 arguments 1977 * and formatting them into a list. 1978 * @param listCompositionPattern should take 2 arguments 1979 * and is used by composeList. 1980 * @return a string representing the list. 1981 */ 1982 private static String formatList(String[] stringList, String listPattern, String listCompositionPattern) { 1983 // If we have no list patterns, compose the list in a simple, 1984 // non-localized way. 1985 if (listPattern == null || listCompositionPattern == null) { 1986 StringBuffer result = new StringBuffer(); 1987 for (int i=0; i<stringList.length; ++i) { 1988 if (i>0) result.append(','); 1989 result.append(stringList[i]); 1990 } 1991 return result.toString(); 1992 } 1993 1994 // Compose the list down to three elements if necessary 1995 if (stringList.length > 3) { 1996 MessageFormat format = new MessageFormat(listCompositionPattern); 1997 stringList = composeList(format, stringList); 1998 } 1999 2000 // Rebuild the argument list with the list length as the first element 2001 Object[] args = new Object[stringList.length + 1]; 2002 System.arraycopy(stringList, 0, args, 1, stringList.length); 2003 args[0] = new Integer(stringList.length); 2004 2005 // Format it using the pattern in the resource 2006 MessageFormat format = new MessageFormat(listPattern); 2007 return format.format(args); 2008 } 2009 2010 /** 2011 * Given a list of strings, return a list shortened to three elements. 2012 * Shorten it by applying the given format to the first two elements 2013 * recursively. 2014 * @param format a format which takes two arguments 2015 * @param list a list of strings 2016 * @return if the list is three elements or shorter, the same list; 2017 * otherwise, a new list of three elements. 2018 */ 2019 private static String[] composeList(MessageFormat format, String[] list) { 2020 if (list.length <= 3) return list; 2021 2022 // Use the given format to compose the first two elements into one 2023 String[] listItems = { list[0], list[1] }; 2024 String newItem = format.format(listItems); 2025 2026 // Form a new list one element shorter 2027 String[] newList = new String[list.length-1]; 2028 System.arraycopy(list, 2, newList, 1, newList.length-1); 2029 newList[0] = newItem; 2030 2031 // Recurse 2032 return composeList(format, newList); 2033 } 2034 2035 /** 2036 * @serialField language String 2037 * language subtag in lower case. (See <a href="java/util/Locale.html#getLanguage()">getLanguage()</a>) 2038 * @serialField country String 2039 * country subtag in upper case. (See <a href="java/util/Locale.html#getCountry()">getCountry()</a>) 2040 * @serialField variant String 2041 * variant subtags separated by LOWLINE characters. (See <a href="java/util/Locale.html#getVariant()">getVariant()</a>) 2042 * @serialField hashcode int 2043 * deprecated, for forward compatibility only 2044 * @serialField script String 2045 * script subtag in title case (See <a href="java/util/Locale.html#getScript()">getScript()</a>) 2046 * @serialField extensions String 2047 * canonical representation of extensions, that is, 2048 * BCP47 extensions in alphabetical order followed by 2049 * BCP47 private use subtags, all in lower case letters 2050 * separated by HYPHEN-MINUS characters. 2051 * (See <a href="java/util/Locale.html#getExtensionKeys()">getExtensionKeys()</a>, 2052 * <a href="java/util/Locale.html#getExtension(char)">getExtension(char)</a>) 2053 */ 2054 private static final ObjectStreamField[] serialPersistentFields = { 2055 new ObjectStreamField("language", String.class), 2056 new ObjectStreamField("country", String.class), 2057 new ObjectStreamField("variant", String.class), 2058 new ObjectStreamField("hashcode", int.class), 2059 new ObjectStreamField("script", String.class), 2060 new ObjectStreamField("extensions", String.class), 2061 }; 2062 2063 /** 2064 * Serializes this <code>Locale</code> to the specified <code>ObjectOutputStream</code>. 2065 * @param out the <code>ObjectOutputStream</code> to write 2066 * @throws IOException 2067 * @since 1.7 2068 */ 2069 private void writeObject(ObjectOutputStream out) throws IOException { 2070 ObjectOutputStream.PutField fields = out.putFields(); 2071 fields.put("language", baseLocale.getLanguage()); 2072 fields.put("script", baseLocale.getScript()); 2073 fields.put("country", baseLocale.getRegion()); 2074 fields.put("variant", baseLocale.getVariant()); 2075 fields.put("extensions", localeExtensions == null ? "" : localeExtensions.getID()); 2076 fields.put("hashcode", -1); // place holder just for backward support 2077 out.writeFields(); 2078 } 2079 2080 /** 2081 * Deserializes this <code>Locale</code>. 2082 * @param in the <code>ObjectInputStream</code> to read 2083 * @throws IOException 2084 * @throws ClassNotFoundException 2085 * @throws IllformedLocaleException 2086 * @since 1.7 2087 */ 2088 private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException { 2089 ObjectInputStream.GetField fields = in.readFields(); 2090 String language = (String)fields.get("language", ""); 2091 String script = (String)fields.get("script", ""); 2092 String country = (String)fields.get("country", ""); 2093 String variant = (String)fields.get("variant", ""); 2094 String extStr = (String)fields.get("extensions", ""); 2095 baseLocale = BaseLocale.getInstance(convertOldISOCodes(language), script, country, variant); 2096 if (extStr.length() > 0) { 2097 try { 2098 InternalLocaleBuilder bldr = new InternalLocaleBuilder(); 2099 bldr.setExtensions(extStr); 2100 localeExtensions = bldr.getLocaleExtensions(); 2101 } catch (LocaleSyntaxException e) { 2102 throw new IllformedLocaleException(e.getMessage()); 2103 } 2104 } else { 2105 localeExtensions = null; 2106 } 2107 } 2108 2109 /** 2110 * Returns a cached <code>Locale</code> instance equivalent to 2111 * the deserialized <code>Locale</code>. When serialized 2112 * language, country and variant fields read from the object data stream 2113 * are exactly "ja", "JP", "JP" or "th", "TH", "TH" and script/extensions 2114 * fields are empty, this method supplies <code>UNICODE_LOCALE_EXTENSION</code> 2115 * "ca"/"japanese" (calendar type is "japanese") or "nu"/"thai" (number script 2116 * type is "thai"). See <a href="Locale.html#special_cases_constructor"/>Special Cases</a> 2117 * for more information. 2118 * 2119 * @return an instance of <code>Locale</code> equivalent to 2120 * the deserialized <code>Locale</code>. 2121 * @throws java.io.ObjectStreamException 2122 */ 2123 private Object readResolve() throws java.io.ObjectStreamException { 2124 return getInstance(baseLocale.getLanguage(), baseLocale.getScript(), 2125 baseLocale.getRegion(), baseLocale.getVariant(), localeExtensions); 2126 } 2127 2128 private static volatile String[] isoLanguages = null; 2129 2130 private static volatile String[] isoCountries = null; 2131 2132 private static String convertOldISOCodes(String language) { 2133 // we accept both the old and the new ISO codes for the languages whose ISO 2134 // codes have changed, but we always store the OLD code, for backward compatibility 2135 language = LocaleUtils.toLowerString(language).intern(); 2136 if (language == "he") { 2137 return "iw"; 2138 } else if (language == "yi") { 2139 return "ji"; 2140 } else if (language == "id") { 2141 return "in"; 2142 } else { 2143 return language; 2144 } 2145 } 2146 2147 private static LocaleExtensions getCompatibilityExtensions(String language, 2148 String script, 2149 String country, 2150 String variant) { 2151 LocaleExtensions extensions = null; 2152 // Special cases for backward compatibility support 2153 if (LocaleUtils.caseIgnoreMatch(language, "ja") 2154 && script.length() == 0 2155 && LocaleUtils.caseIgnoreMatch(country, "jp") 2156 && "JP".equals(variant)) { 2157 // ja_JP_JP -> u-ca-japanese (calendar = japanese) 2158 extensions = LocaleExtensions.CALENDAR_JAPANESE; 2159 } else if (LocaleUtils.caseIgnoreMatch(language, "th") 2160 && script.length() == 0 2161 && LocaleUtils.caseIgnoreMatch(country, "th") 2162 && "TH".equals(variant)) { 2163 // th_TH_TH -> u-nu-thai (numbersystem = thai) 2164 extensions = LocaleExtensions.NUMBER_THAI; 2165 } 2166 return extensions; 2167 } 2168 2169 /** 2170 * @hide for internal use only. 2171 */ 2172 public static String adjustLanguageCode(String languageCode) { 2173 String adjusted = languageCode.toLowerCase(Locale.US); 2174 // Map new language codes to the obsolete language 2175 // codes so the correct resource bundles will be used. 2176 if (languageCode.equals("he")) { 2177 adjusted = "iw"; 2178 } else if (languageCode.equals("id")) { 2179 adjusted = "in"; 2180 } else if (languageCode.equals("yi")) { 2181 adjusted = "ji"; 2182 } 2183 2184 return adjusted; 2185 } 2186 2187 /** 2188 * Enum for locale categories. These locale categories are used to get/set 2189 * the default locale for the specific functionality represented by the 2190 * category. 2191 * 2192 * @see #getDefault(Locale.Category) 2193 * @see #setDefault(Locale.Category, Locale) 2194 * @since 1.7 2195 */ 2196 public enum Category { 2197 2198 /** 2199 * Category used to represent the default locale for 2200 * displaying user interfaces. 2201 */ 2202 DISPLAY("user.language.display", 2203 "user.script.display", 2204 "user.country.display", 2205 "user.variant.display"), 2206 2207 /** 2208 * Category used to represent the default locale for 2209 * formatting dates, numbers, and/or currencies. 2210 */ 2211 FORMAT("user.language.format", 2212 "user.script.format", 2213 "user.country.format", 2214 "user.variant.format"); 2215 2216 Category(String languageKey, String scriptKey, String countryKey, String variantKey) { 2217 this.languageKey = languageKey; 2218 this.scriptKey = scriptKey; 2219 this.countryKey = countryKey; 2220 this.variantKey = variantKey; 2221 } 2222 2223 final String languageKey; 2224 final String scriptKey; 2225 final String countryKey; 2226 final String variantKey; 2227 } 2228 2229 /** 2230 * <code>Builder</code> is used to build instances of <code>Locale</code> 2231 * from values configured by the setters. Unlike the <code>Locale</code> 2232 * constructors, the <code>Builder</code> checks if a value configured by a 2233 * setter satisfies the syntax requirements defined by the <code>Locale</code> 2234 * class. A <code>Locale</code> object created by a <code>Builder</code> is 2235 * well-formed and can be transformed to a well-formed IETF BCP 47 language tag 2236 * without losing information. 2237 * 2238 * <p><b>Note:</b> The <code>Locale</code> class does not provide any 2239 * syntactic restrictions on variant, while BCP 47 requires each variant 2240 * subtag to be 5 to 8 alphanumerics or a single numeric followed by 3 2241 * alphanumerics. The method <code>setVariant</code> throws 2242 * <code>IllformedLocaleException</code> for a variant that does not satisfy 2243 * this restriction. If it is necessary to support such a variant, use a 2244 * Locale constructor. However, keep in mind that a <code>Locale</code> 2245 * object created this way might lose the variant information when 2246 * transformed to a BCP 47 language tag. 2247 * 2248 * <p>The following example shows how to create a <code>Locale</code> object 2249 * with the <code>Builder</code>. 2250 * <blockquote> 2251 * <pre> 2252 * Locale aLocale = new Builder().setLanguage("sr").setScript("Latn").setRegion("RS").build(); 2253 * </pre> 2254 * </blockquote> 2255 * 2256 * <p>Builders can be reused; <code>clear()</code> resets all 2257 * fields to their default values. 2258 * 2259 * @see Locale#forLanguageTag 2260 * @since 1.7 2261 */ 2262 public static final class Builder { 2263 private final InternalLocaleBuilder localeBuilder; 2264 2265 /** 2266 * Constructs an empty Builder. The default value of all 2267 * fields, extensions, and private use information is the 2268 * empty string. 2269 */ 2270 public Builder() { 2271 localeBuilder = new InternalLocaleBuilder(); 2272 } 2273 2274 /** 2275 * Resets the <code>Builder</code> to match the provided 2276 * <code>locale</code>. Existing state is discarded. 2277 * 2278 * <p>All fields of the locale must be well-formed, see {@link Locale}. 2279 * 2280 * <p>Locales with any ill-formed fields cause 2281 * <code>IllformedLocaleException</code> to be thrown, except for the 2282 * following three cases which are accepted for compatibility 2283 * reasons:<ul> 2284 * <li>Locale("ja", "JP", "JP") is treated as "ja-JP-u-ca-japanese" 2285 * <li>Locale("th", "TH", "TH") is treated as "th-TH-u-nu-thai" 2286 * <li>Locale("no", "NO", "NY") is treated as "nn-NO"</ul> 2287 * 2288 * @param locale the locale 2289 * @return This builder. 2290 * @throws IllformedLocaleException if <code>locale</code> has 2291 * any ill-formed fields. 2292 * @throws NullPointerException if <code>locale</code> is null. 2293 */ 2294 public Builder setLocale(Locale locale) { 2295 try { 2296 localeBuilder.setLocale(locale.baseLocale, locale.localeExtensions); 2297 } catch (LocaleSyntaxException e) { 2298 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2299 } 2300 return this; 2301 } 2302 2303 /** 2304 * Resets the Builder to match the provided IETF BCP 47 2305 * language tag. Discards the existing state. Null and the 2306 * empty string cause the builder to be reset, like {@link 2307 * #clear}. Grandfathered tags (see {@link 2308 * Locale#forLanguageTag}) are converted to their canonical 2309 * form before being processed. Otherwise, the language tag 2310 * must be well-formed (see {@link Locale}) or an exception is 2311 * thrown (unlike <code>Locale.forLanguageTag</code>, which 2312 * just discards ill-formed and following portions of the 2313 * tag). 2314 * 2315 * @param languageTag the language tag 2316 * @return This builder. 2317 * @throws IllformedLocaleException if <code>languageTag</code> is ill-formed 2318 * @see Locale#forLanguageTag(String) 2319 */ 2320 public Builder setLanguageTag(String languageTag) { 2321 ParseStatus sts = new ParseStatus(); 2322 LanguageTag tag = LanguageTag.parse(languageTag, sts); 2323 if (sts.isError()) { 2324 throw new IllformedLocaleException(sts.getErrorMessage(), sts.getErrorIndex()); 2325 } 2326 localeBuilder.setLanguageTag(tag); 2327 return this; 2328 } 2329 2330 /** 2331 * Sets the language. If <code>language</code> is the empty string or 2332 * null, the language in this <code>Builder</code> is removed. Otherwise, 2333 * the language must be <a href="./Locale.html#def_language">well-formed</a> 2334 * or an exception is thrown. 2335 * 2336 * <p>The typical language value is a two or three-letter language 2337 * code as defined in ISO639. 2338 * 2339 * @param language the language 2340 * @return This builder. 2341 * @throws IllformedLocaleException if <code>language</code> is ill-formed 2342 */ 2343 public Builder setLanguage(String language) { 2344 try { 2345 localeBuilder.setLanguage(language); 2346 } catch (LocaleSyntaxException e) { 2347 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2348 } 2349 return this; 2350 } 2351 2352 /** 2353 * Sets the script. If <code>script</code> is null or the empty string, 2354 * the script in this <code>Builder</code> is removed. 2355 * Otherwise, the script must be <a href="./Locale.html#def_script">well-formed</a> or an 2356 * exception is thrown. 2357 * 2358 * <p>The typical script value is a four-letter script code as defined by ISO 15924. 2359 * 2360 * @param script the script 2361 * @return This builder. 2362 * @throws IllformedLocaleException if <code>script</code> is ill-formed 2363 */ 2364 public Builder setScript(String script) { 2365 try { 2366 localeBuilder.setScript(script); 2367 } catch (LocaleSyntaxException e) { 2368 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2369 } 2370 return this; 2371 } 2372 2373 /** 2374 * Sets the region. If region is null or the empty string, the region 2375 * in this <code>Builder</code> is removed. Otherwise, 2376 * the region must be <a href="./Locale.html#def_region">well-formed</a> or an 2377 * exception is thrown. 2378 * 2379 * <p>The typical region value is a two-letter ISO 3166 code or a 2380 * three-digit UN M.49 area code. 2381 * 2382 * <p>The country value in the <code>Locale</code> created by the 2383 * <code>Builder</code> is always normalized to upper case. 2384 * 2385 * @param region the region 2386 * @return This builder. 2387 * @throws IllformedLocaleException if <code>region</code> is ill-formed 2388 */ 2389 public Builder setRegion(String region) { 2390 try { 2391 localeBuilder.setRegion(region); 2392 } catch (LocaleSyntaxException e) { 2393 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2394 } 2395 return this; 2396 } 2397 2398 /** 2399 * Sets the variant. If variant is null or the empty string, the 2400 * variant in this <code>Builder</code> is removed. Otherwise, it 2401 * must consist of one or more <a href="./Locale.html#def_variant">well-formed</a> 2402 * subtags, or an exception is thrown. 2403 * 2404 * <p><b>Note:</b> This method checks if <code>variant</code> 2405 * satisfies the IETF BCP 47 variant subtag's syntax requirements, 2406 * and normalizes the value to lowercase letters. However, 2407 * the <code>Locale</code> class does not impose any syntactic 2408 * restriction on variant, and the variant value in 2409 * <code>Locale</code> is case sensitive. To set such a variant, 2410 * use a Locale constructor. 2411 * 2412 * @param variant the variant 2413 * @return This builder. 2414 * @throws IllformedLocaleException if <code>variant</code> is ill-formed 2415 */ 2416 public Builder setVariant(String variant) { 2417 try { 2418 localeBuilder.setVariant(variant); 2419 } catch (LocaleSyntaxException e) { 2420 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2421 } 2422 return this; 2423 } 2424 2425 /** 2426 * Sets the extension for the given key. If the value is null or the 2427 * empty string, the extension is removed. Otherwise, the extension 2428 * must be <a href="./Locale.html#def_extensions">well-formed</a> or an exception 2429 * is thrown. 2430 * 2431 * <p><b>Note:</b> The key {@link Locale#UNICODE_LOCALE_EXTENSION 2432 * UNICODE_LOCALE_EXTENSION} ('u') is used for the Unicode locale extension. 2433 * Setting a value for this key replaces any existing Unicode locale key/type 2434 * pairs with those defined in the extension. 2435 * 2436 * <p><b>Note:</b> The key {@link Locale#PRIVATE_USE_EXTENSION 2437 * PRIVATE_USE_EXTENSION} ('x') is used for the private use code. To be 2438 * well-formed, the value for this key needs only to have subtags of one to 2439 * eight alphanumeric characters, not two to eight as in the general case. 2440 * 2441 * @param key the extension key 2442 * @param value the extension value 2443 * @return This builder. 2444 * @throws IllformedLocaleException if <code>key</code> is illegal 2445 * or <code>value</code> is ill-formed 2446 * @see #setUnicodeLocaleKeyword(String, String) 2447 */ 2448 public Builder setExtension(char key, String value) { 2449 try { 2450 localeBuilder.setExtension(key, value); 2451 } catch (LocaleSyntaxException e) { 2452 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2453 } 2454 return this; 2455 } 2456 2457 /** 2458 * Sets the Unicode locale keyword type for the given key. If the type 2459 * is null, the Unicode keyword is removed. Otherwise, the key must be 2460 * non-null and both key and type must be <a 2461 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2462 * is thrown. 2463 * 2464 * <p>Keys and types are converted to lower case. 2465 * 2466 * <p><b>Note</b>:Setting the 'u' extension via {@link #setExtension} 2467 * replaces all Unicode locale keywords with those defined in the 2468 * extension. 2469 * 2470 * @param key the Unicode locale key 2471 * @param type the Unicode locale type 2472 * @return This builder. 2473 * @throws IllformedLocaleException if <code>key</code> or <code>type</code> 2474 * is ill-formed 2475 * @throws NullPointerException if <code>key</code> is null 2476 * @see #setExtension(char, String) 2477 */ 2478 public Builder setUnicodeLocaleKeyword(String key, String type) { 2479 try { 2480 localeBuilder.setUnicodeLocaleKeyword(key, type); 2481 } catch (LocaleSyntaxException e) { 2482 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2483 } 2484 return this; 2485 } 2486 2487 /** 2488 * Adds a unicode locale attribute, if not already present, otherwise 2489 * has no effect. The attribute must not be null and must be <a 2490 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2491 * is thrown. 2492 * 2493 * @param attribute the attribute 2494 * @return This builder. 2495 * @throws NullPointerException if <code>attribute</code> is null 2496 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed 2497 * @see #setExtension(char, String) 2498 */ 2499 public Builder addUnicodeLocaleAttribute(String attribute) { 2500 try { 2501 localeBuilder.addUnicodeLocaleAttribute(attribute); 2502 } catch (LocaleSyntaxException e) { 2503 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2504 } 2505 return this; 2506 } 2507 2508 /** 2509 * Removes a unicode locale attribute, if present, otherwise has no 2510 * effect. The attribute must not be null and must be <a 2511 * href="./Locale.html#def_locale_extension">well-formed</a> or an exception 2512 * is thrown. 2513 * 2514 * <p>Attribute comparision for removal is case-insensitive. 2515 * 2516 * @param attribute the attribute 2517 * @return This builder. 2518 * @throws NullPointerException if <code>attribute</code> is null 2519 * @throws IllformedLocaleException if <code>attribute</code> is ill-formed 2520 * @see #setExtension(char, String) 2521 */ 2522 public Builder removeUnicodeLocaleAttribute(String attribute) { 2523 try { 2524 localeBuilder.removeUnicodeLocaleAttribute(attribute); 2525 } catch (LocaleSyntaxException e) { 2526 throw new IllformedLocaleException(e.getMessage(), e.getErrorIndex()); 2527 } 2528 return this; 2529 } 2530 2531 /** 2532 * Resets the builder to its initial, empty state. 2533 * 2534 * @return This builder. 2535 */ 2536 public Builder clear() { 2537 localeBuilder.clear(); 2538 return this; 2539 } 2540 2541 /** 2542 * Resets the extensions to their initial, empty state. 2543 * Language, script, region and variant are unchanged. 2544 * 2545 * @return This builder. 2546 * @see #setExtension(char, String) 2547 */ 2548 public Builder clearExtensions() { 2549 localeBuilder.clearExtensions(); 2550 return this; 2551 } 2552 2553 /** 2554 * Returns an instance of <code>Locale</code> created from the fields set 2555 * on this builder. 2556 * 2557 * <p>This applies the conversions listed in {@link Locale#forLanguageTag} 2558 * when constructing a Locale. (Grandfathered tags are handled in 2559 * {@link #setLanguageTag}.) 2560 * 2561 * @return A Locale. 2562 */ 2563 public Locale build() { 2564 BaseLocale baseloc = localeBuilder.getBaseLocale(); 2565 LocaleExtensions extensions = localeBuilder.getLocaleExtensions(); 2566 if (extensions == null && baseloc.getVariant().length() > 0) { 2567 extensions = getCompatibilityExtensions(baseloc.getLanguage(), baseloc.getScript(), 2568 baseloc.getRegion(), baseloc.getVariant()); 2569 } 2570 return Locale.getInstance(baseloc, extensions); 2571 } 2572 } 2573} 2574