1 /* 2 * Copyright (c) 2000, 2017, Oracle and/or its affiliates. All rights reserved. 3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 4 * 5 * This code is free software; you can redistribute it and/or modify it 6 * under the terms of the GNU General Public License version 2 only, as 7 * published by the Free Software Foundation. Oracle designates this 8 * particular file as subject to the "Classpath" exception as provided 9 * by Oracle in the LICENSE file that accompanied this code. 10 * 11 * This code is distributed in the hope that it will be useful, but WITHOUT 12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 14 * version 2 for more details (a copy is included in the LICENSE file that 15 * accompanied this code). 16 * 17 * You should have received a copy of the GNU General Public License version 18 * 2 along with this work; if not, write to the Free Software Foundation, 19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 20 * 21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 22 * or visit www.oracle.com if you need additional information or have any 23 * questions. 24 */ 25 26 package java.nio.charset; 27 28 import jdk.internal.misc.VM; 29 import sun.nio.cs.ThreadLocalCoders; 30 import sun.security.action.GetPropertyAction; 31 32 import java.nio.ByteBuffer; 33 import java.nio.CharBuffer; 34 import java.nio.charset.spi.CharsetProvider; 35 import java.security.AccessController; 36 import java.security.PrivilegedAction; 37 import java.util.Arrays; 38 import java.util.Collections; 39 import java.util.HashSet; 40 import java.util.Iterator; 41 import java.util.Locale; 42 import java.util.Map; 43 import java.util.NoSuchElementException; 44 import java.util.Objects; 45 import java.util.ServiceConfigurationError; 46 import java.util.ServiceLoader; 47 import java.util.Set; 48 import java.util.SortedMap; 49 import java.util.TreeMap; 50 51 52 /** 53 * A named mapping between sequences of sixteen-bit Unicode <a 54 * href="../../lang/Character.html#unicode">code units</a> and sequences of 55 * bytes. This class defines methods for creating decoders and encoders and 56 * for retrieving the various names associated with a charset. Instances of 57 * this class are immutable. 58 * 59 * <p> This class also defines static methods for testing whether a particular 60 * charset is supported, for locating charset instances by name, and for 61 * constructing a map that contains every charset for which support is 62 * available in the current Java virtual machine. Support for new charsets can 63 * be added via the service-provider interface defined in the {@link 64 * java.nio.charset.spi.CharsetProvider} class. 65 * 66 * <p> All of the methods defined in this class are safe for use by multiple 67 * concurrent threads. 68 * 69 * 70 * <a id="names"></a><a id="charenc"></a> 71 * <h2>Charset names</h2> 72 * 73 * <p> Charsets are named by strings composed of the following characters: 74 * 75 * <ul> 76 * 77 * <li> The uppercase letters {@code 'A'} through {@code 'Z'} 78 * (<code>'\u0041'</code> through <code>'\u005a'</code>), 79 * 80 * <li> The lowercase letters {@code 'a'} through {@code 'z'} 81 * (<code>'\u0061'</code> through <code>'\u007a'</code>), 82 * 83 * <li> The digits {@code '0'} through {@code '9'} 84 * (<code>'\u0030'</code> through <code>'\u0039'</code>), 85 * 86 * <li> The dash character {@code '-'} 87 * (<code>'\u002d'</code>, <small>HYPHEN-MINUS</small>), 88 * 89 * <li> The plus character {@code '+'} 90 * (<code>'\u002b'</code>, <small>PLUS SIGN</small>), 91 * 92 * <li> The period character {@code '.'} 93 * (<code>'\u002e'</code>, <small>FULL STOP</small>), 94 * 95 * <li> The colon character {@code ':'} 96 * (<code>'\u003a'</code>, <small>COLON</small>), and 97 * 98 * <li> The underscore character {@code '_'} 99 * (<code>'\u005f'</code>, <small>LOW LINE</small>). 100 * 101 * </ul> 102 * 103 * A charset name must begin with either a letter or a digit. The empty string 104 * is not a legal charset name. Charset names are not case-sensitive; that is, 105 * case is always ignored when comparing charset names. Charset names 106 * generally follow the conventions documented in <a 107 * href="http://www.ietf.org/rfc/rfc2278.txt"><i>RFC 2278: IANA Charset 108 * Registration Procedures</i></a>. 109 * 110 * <p> Every charset has a <i>canonical name</i> and may also have one or more 111 * <i>aliases</i>. The canonical name is returned by the {@link #name() name} method 112 * of this class. Canonical names are, by convention, usually in upper case. 113 * The aliases of a charset are returned by the {@link #aliases() aliases} 114 * method. 115 * 116 * <p><a id="hn">Some charsets have an <i>historical name</i> that is defined for 117 * compatibility with previous versions of the Java platform.</a> A charset's 118 * historical name is either its canonical name or one of its aliases. The 119 * historical name is returned by the {@code getEncoding()} methods of the 120 * {@link java.io.InputStreamReader#getEncoding InputStreamReader} and {@link 121 * java.io.OutputStreamWriter#getEncoding OutputStreamWriter} classes. 122 * 123 * <p><a id="iana"> </a>If a charset listed in the <a 124 * href="http://www.iana.org/assignments/character-sets"><i>IANA Charset 125 * Registry</i></a> is supported by an implementation of the Java platform then 126 * its canonical name must be the name listed in the registry. Many charsets 127 * are given more than one name in the registry, in which case the registry 128 * identifies one of the names as <i>MIME-preferred</i>. If a charset has more 129 * than one registry name then its canonical name must be the MIME-preferred 130 * name and the other names in the registry must be valid aliases. If a 131 * supported charset is not listed in the IANA registry then its canonical name 132 * must begin with one of the strings {@code "X-"} or {@code "x-"}. 133 * 134 * <p> The IANA charset registry does change over time, and so the canonical 135 * name and the aliases of a particular charset may also change over time. To 136 * ensure compatibility it is recommended that no alias ever be removed from a 137 * charset, and that if the canonical name of a charset is changed then its 138 * previous canonical name be made into an alias. 139 * 140 * 141 * <h2>Standard charsets</h2> 142 * 143 * 144 * 145 * <p><a id="standard">Every implementation of the Java platform is required to support the 146 * following standard charsets.</a> Consult the release documentation for your 147 * implementation to see if any other charsets are supported. The behavior 148 * of such optional charsets may differ between implementations. 149 * 150 * <blockquote><table class="striped" style="width:80%"> 151 * <caption style="display:none">Description of standard charsets</caption> 152 * <thead> 153 * <tr><th scope="col" style="text-align:left">Charset</th><th scope="col" style="text-align:left">Description</th></tr> 154 * </thead> 155 * <tbody> 156 * <tr><th scope="row" style="vertical-align:top">{@code US-ASCII}</th> 157 * <td>Seven-bit ASCII, a.k.a. {@code ISO646-US}, 158 * a.k.a. the Basic Latin block of the Unicode character set</td></tr> 159 * <tr><th scope="row" style="vertical-align:top"><code>ISO-8859-1 </code></th> 160 * <td>ISO Latin Alphabet No. 1, a.k.a. {@code ISO-LATIN-1}</td></tr> 161 * <tr><th scope="row" style="vertical-align:top">{@code UTF-8}</th> 162 * <td>Eight-bit UCS Transformation Format</td></tr> 163 * <tr><th scope="row" style="vertical-align:top">{@code UTF-16BE}</th> 164 * <td>Sixteen-bit UCS Transformation Format, 165 * big-endian byte order</td></tr> 166 * <tr><th scope="row" style="vertical-align:top">{@code UTF-16LE}</th> 167 * <td>Sixteen-bit UCS Transformation Format, 168 * little-endian byte order</td></tr> 169 * <tr><th scope="row" style="vertical-align:top">{@code UTF-16}</th> 170 * <td>Sixteen-bit UCS Transformation Format, 171 * byte order identified by an optional byte-order mark</td></tr> 172 * </tbody> 173 * </table></blockquote> 174 * 175 * <p> The {@code UTF-8} charset is specified by <a 176 * href="http://www.ietf.org/rfc/rfc2279.txt"><i>RFC 2279</i></a>; the 177 * transformation format upon which it is based is specified in 178 * Amendment 2 of ISO 10646-1 and is also described in the <a 179 * href="http://www.unicode.org/unicode/standard/standard.html"><i>Unicode 180 * Standard</i></a>. 181 * 182 * <p> The {@code UTF-16} charsets are specified by <a 183 * href="http://www.ietf.org/rfc/rfc2781.txt"><i>RFC 2781</i></a>; the 184 * transformation formats upon which they are based are specified in 185 * Amendment 1 of ISO 10646-1 and are also described in the <a 186 * href="http://www.unicode.org/unicode/standard/standard.html"><i>Unicode 187 * Standard</i></a>. 188 * 189 * <p> The {@code UTF-16} charsets use sixteen-bit quantities and are 190 * therefore sensitive to byte order. In these encodings the byte order of a 191 * stream may be indicated by an initial <i>byte-order mark</i> represented by 192 * the Unicode character <code>'\uFEFF'</code>. Byte-order marks are handled 193 * as follows: 194 * 195 * <ul> 196 * 197 * <li><p> When decoding, the {@code UTF-16BE} and {@code UTF-16LE} 198 * charsets interpret the initial byte-order marks as a <small>ZERO-WIDTH 199 * NON-BREAKING SPACE</small>; when encoding, they do not write 200 * byte-order marks. </p></li> 201 202 * 203 * <li><p> When decoding, the {@code UTF-16} charset interprets the 204 * byte-order mark at the beginning of the input stream to indicate the 205 * byte-order of the stream but defaults to big-endian if there is no 206 * byte-order mark; when encoding, it uses big-endian byte order and writes 207 * a big-endian byte-order mark. </p></li> 208 * 209 * </ul> 210 * 211 * In any case, byte order marks occurring after the first element of an 212 * input sequence are not omitted since the same code is used to represent 213 * <small>ZERO-WIDTH NON-BREAKING SPACE</small>. 214 * 215 * <p> Every instance of the Java virtual machine has a default charset, which 216 * may or may not be one of the standard charsets. The default charset is 217 * determined during virtual-machine startup and typically depends upon the 218 * locale and charset being used by the underlying operating system. </p> 219 * 220 * <p>The {@link StandardCharsets} class defines constants for each of the 221 * standard charsets. 222 * 223 * <h2>Terminology</h2> 224 * 225 * <p> The name of this class is taken from the terms used in 226 * <a href="http://www.ietf.org/rfc/rfc2278.txt"><i>RFC 2278</i></a>. 227 * In that document a <i>charset</i> is defined as the combination of 228 * one or more coded character sets and a character-encoding scheme. 229 * (This definition is confusing; some other software systems define 230 * <i>charset</i> as a synonym for <i>coded character set</i>.) 231 * 232 * <p> A <i>coded character set</i> is a mapping between a set of abstract 233 * characters and a set of integers. US-ASCII, ISO 8859-1, 234 * JIS X 0201, and Unicode are examples of coded character sets. 235 * 236 * <p> Some standards have defined a <i>character set</i> to be simply a 237 * set of abstract characters without an associated assigned numbering. 238 * An alphabet is an example of such a character set. However, the subtle 239 * distinction between <i>character set</i> and <i>coded character set</i> 240 * is rarely used in practice; the former has become a short form for the 241 * latter, including in the Java API specification. 242 * 243 * <p> A <i>character-encoding scheme</i> is a mapping between one or more 244 * coded character sets and a set of octet (eight-bit byte) sequences. 245 * UTF-8, UTF-16, ISO 2022, and EUC are examples of 246 * character-encoding schemes. Encoding schemes are often associated with 247 * a particular coded character set; UTF-8, for example, is used only to 248 * encode Unicode. Some schemes, however, are associated with multiple 249 * coded character sets; EUC, for example, can be used to encode 250 * characters in a variety of Asian coded character sets. 251 * 252 * <p> When a coded character set is used exclusively with a single 253 * character-encoding scheme then the corresponding charset is usually 254 * named for the coded character set; otherwise a charset is usually named 255 * for the encoding scheme and, possibly, the locale of the coded 256 * character sets that it supports. Hence {@code US-ASCII} is both the 257 * name of a coded character set and of the charset that encodes it, while 258 * {@code EUC-JP} is the name of the charset that encodes the 259 * JIS X 0201, JIS X 0208, and JIS X 0212 260 * coded character sets for the Japanese language. 261 * 262 * <p> The native character encoding of the Java programming language is 263 * UTF-16. A charset in the Java platform therefore defines a mapping 264 * between sequences of sixteen-bit UTF-16 code units (that is, sequences 265 * of chars) and sequences of bytes. </p> 266 * 267 * 268 * @author Mark Reinhold 269 * @author JSR-51 Expert Group 270 * @since 1.4 271 * 272 * @see CharsetDecoder 273 * @see CharsetEncoder 274 * @see java.nio.charset.spi.CharsetProvider 275 * @see java.lang.Character 276 */ 277 278 public abstract class Charset 279 implements Comparable<Charset> 280 { 281 282 /* -- Static methods -- */ 283 284 /** 285 * Checks that the given string is a legal charset name. </p> 286 * 287 * @param s 288 * A purported charset name 289 * 290 * @throws IllegalCharsetNameException 291 * If the given name is not a legal charset name 292 */ 293 private static void checkName(String s) { 294 int n = s.length(); 295 if (n == 0) { 296 throw new IllegalCharsetNameException(s); 297 } 298 for (int i = 0; i < n; i++) { 299 char c = s.charAt(i); 300 if (c >= 'A' && c <= 'Z') continue; 301 if (c >= 'a' && c <= 'z') continue; 302 if (c >= '0' && c <= '9') continue; 303 if (c == '-' && i != 0) continue; 304 if (c == '+' && i != 0) continue; 305 if (c == ':' && i != 0) continue; 306 if (c == '_' && i != 0) continue; 307 if (c == '.' && i != 0) continue; 308 throw new IllegalCharsetNameException(s); 309 } 310 } 311 312 /* The standard set of charsets */ 313 private static final CharsetProvider standardProvider 314 = new sun.nio.cs.StandardCharsets(); 315 316 private static final String[] zeroAliases = new String[0]; 317 318 // Cache of the most-recently-returned charsets, 319 // along with the names that were used to find them 320 // 321 private static volatile Object[] cache1; // "Level 1" cache 322 private static volatile Object[] cache2; // "Level 2" cache 323 324 private static void cache(String charsetName, Charset cs) { 325 cache2 = cache1; 326 cache1 = new Object[] { charsetName, cs }; 327 } 328 329 // Creates an iterator that walks over the available providers, ignoring 330 // those whose lookup or instantiation causes a security exception to be 331 // thrown. Should be invoked with full privileges. 332 // 333 private static Iterator<CharsetProvider> providers() { 334 return new Iterator<>() { 335 ClassLoader cl = ClassLoader.getSystemClassLoader(); 336 ServiceLoader<CharsetProvider> sl = 337 ServiceLoader.load(CharsetProvider.class, cl); 338 Iterator<CharsetProvider> i = sl.iterator(); 339 CharsetProvider next = null; 340 341 private boolean getNext() { 342 while (next == null) { 343 try { 344 if (!i.hasNext()) 345 return false; 346 next = i.next(); 347 } catch (ServiceConfigurationError sce) { 348 if (sce.getCause() instanceof SecurityException) { 349 // Ignore security exceptions 350 continue; 351 } 352 throw sce; 353 } 354 } 355 return true; 356 } 357 358 public boolean hasNext() { 359 return getNext(); 360 } 361 362 public CharsetProvider next() { 363 if (!getNext()) 364 throw new NoSuchElementException(); 365 CharsetProvider n = next; 366 next = null; 367 return n; 368 } 369 370 public void remove() { 371 throw new UnsupportedOperationException(); 372 } 373 374 }; 375 } 376 377 // Thread-local gate to prevent recursive provider lookups 378 private static ThreadLocal<ThreadLocal<?>> gate = 379 new ThreadLocal<ThreadLocal<?>>(); 380 381 private static Charset lookupViaProviders(final String charsetName) { 382 383 // The runtime startup sequence looks up standard charsets as a 384 // consequence of the VM's invocation of System.initializeSystemClass 385 // in order to, e.g., set system properties and encode filenames. At 386 // that point the application class loader has not been initialized, 387 // however, so we can't look for providers because doing so will cause 388 // that loader to be prematurely initialized with incomplete 389 // information. 390 // 391 if (!VM.isBooted()) 392 return null; 393 394 if (gate.get() != null) 395 // Avoid recursive provider lookups 396 return null; 397 try { 398 gate.set(gate); 399 400 return AccessController.doPrivileged( 401 new PrivilegedAction<>() { 402 public Charset run() { 403 for (Iterator<CharsetProvider> i = providers(); 404 i.hasNext();) { 405 CharsetProvider cp = i.next(); 406 Charset cs = cp.charsetForName(charsetName); 407 if (cs != null) 408 return cs; 409 } 410 return null; 411 } 412 }); 413 414 } finally { 415 gate.set(null); 416 } 417 } 418 419 /* The extended set of charsets */ 420 private static class ExtendedProviderHolder { 421 static final CharsetProvider[] extendedProviders = extendedProviders(); 422 // returns ExtendedProvider, if installed 423 private static CharsetProvider[] extendedProviders() { 424 return AccessController.doPrivileged(new PrivilegedAction<>() { 425 public CharsetProvider[] run() { 426 CharsetProvider[] cps = new CharsetProvider[1]; 427 int n = 0; 428 ServiceLoader<CharsetProvider> sl = 429 ServiceLoader.loadInstalled(CharsetProvider.class); 430 for (CharsetProvider cp : sl) { 431 if (n + 1 > cps.length) { 432 cps = Arrays.copyOf(cps, cps.length << 1); 433 } 434 cps[n++] = cp; 435 } 436 return n == cps.length ? cps : Arrays.copyOf(cps, n); 437 }}); 438 } 439 } 440 441 private static Charset lookupExtendedCharset(String charsetName) { 442 if (!VM.isBooted()) // see lookupViaProviders() 443 return null; 444 CharsetProvider[] ecps = ExtendedProviderHolder.extendedProviders; 445 for (CharsetProvider cp : ecps) { 446 Charset cs = cp.charsetForName(charsetName); 447 if (cs != null) 448 return cs; 449 } 450 return null; 451 } 452 453 private static Charset lookup(String charsetName) { 454 if (charsetName == null) 455 throw new IllegalArgumentException("Null charset name"); 456 Object[] a; 457 if ((a = cache1) != null && charsetName.equals(a[0])) 458 return (Charset)a[1]; 459 // We expect most programs to use one Charset repeatedly. 460 // We convey a hint to this effect to the VM by putting the 461 // level 1 cache miss code in a separate method. 462 return lookup2(charsetName); 463 } 464 465 private static Charset lookup2(String charsetName) { 466 Object[] a; 467 if ((a = cache2) != null && charsetName.equals(a[0])) { 468 cache2 = cache1; 469 cache1 = a; 470 return (Charset)a[1]; 471 } 472 Charset cs; 473 if ((cs = standardProvider.charsetForName(charsetName)) != null || 474 (cs = lookupExtendedCharset(charsetName)) != null || 475 (cs = lookupViaProviders(charsetName)) != null) 476 { 477 cache(charsetName, cs); 478 return cs; 479 } 480 481 /* Only need to check the name if we didn't find a charset for it */ 482 checkName(charsetName); 483 return null; 484 } 485 486 /** 487 * Tells whether the named charset is supported. 488 * 489 * @param charsetName 490 * The name of the requested charset; may be either 491 * a canonical name or an alias 492 * 493 * @return {@code true} if, and only if, support for the named charset 494 * is available in the current Java virtual machine 495 * 496 * @throws IllegalCharsetNameException 497 * If the given charset name is illegal 498 * 499 * @throws IllegalArgumentException 500 * If the given {@code charsetName} is null 501 */ 502 public static boolean isSupported(String charsetName) { 503 return (lookup(charsetName) != null); 504 } 505 506 /** 507 * Returns a charset object for the named charset. 508 * 509 * @param charsetName 510 * The name of the requested charset; may be either 511 * a canonical name or an alias 512 * 513 * @return A charset object for the named charset 514 * 515 * @throws IllegalCharsetNameException 516 * If the given charset name is illegal 517 * 518 * @throws IllegalArgumentException 519 * If the given {@code charsetName} is null 520 * 521 * @throws UnsupportedCharsetException 522 * If no support for the named charset is available 523 * in this instance of the Java virtual machine 524 */ 525 public static Charset forName(String charsetName) { 526 Charset cs = lookup(charsetName); 527 if (cs != null) 528 return cs; 529 throw new UnsupportedCharsetException(charsetName); 530 } 531 532 // Fold charsets from the given iterator into the given map, ignoring 533 // charsets whose names already have entries in the map. 534 // 535 private static void put(Iterator<Charset> i, Map<String,Charset> m) { 536 while (i.hasNext()) { 537 Charset cs = i.next(); 538 if (!m.containsKey(cs.name())) 539 m.put(cs.name(), cs); 540 } 541 } 542 543 /** 544 * Constructs a sorted map from canonical charset names to charset objects. 545 * 546 * <p> The map returned by this method will have one entry for each charset 547 * for which support is available in the current Java virtual machine. If 548 * two or more supported charsets have the same canonical name then the 549 * resulting map will contain just one of them; which one it will contain 550 * is not specified. </p> 551 * 552 * <p> The invocation of this method, and the subsequent use of the 553 * resulting map, may cause time-consuming disk or network I/O operations 554 * to occur. This method is provided for applications that need to 555 * enumerate all of the available charsets, for example to allow user 556 * charset selection. This method is not used by the {@link #forName 557 * forName} method, which instead employs an efficient incremental lookup 558 * algorithm. 559 * 560 * <p> This method may return different results at different times if new 561 * charset providers are dynamically made available to the current Java 562 * virtual machine. In the absence of such changes, the charsets returned 563 * by this method are exactly those that can be retrieved via the {@link 564 * #forName forName} method. </p> 565 * 566 * @return An immutable, case-insensitive map from canonical charset names 567 * to charset objects 568 */ 569 public static SortedMap<String,Charset> availableCharsets() { 570 return AccessController.doPrivileged( 571 new PrivilegedAction<>() { 572 public SortedMap<String,Charset> run() { 573 TreeMap<String,Charset> m = 574 new TreeMap<>( 575 String.CASE_INSENSITIVE_ORDER); 576 put(standardProvider.charsets(), m); 577 CharsetProvider[] ecps = ExtendedProviderHolder.extendedProviders; 578 for (CharsetProvider ecp :ecps) { 579 put(ecp.charsets(), m); 580 } 581 for (Iterator<CharsetProvider> i = providers(); i.hasNext();) { 582 CharsetProvider cp = i.next(); 583 put(cp.charsets(), m); 584 } 585 return Collections.unmodifiableSortedMap(m); 586 } 587 }); 588 } 589 590 private static volatile Charset defaultCharset; 591 592 /** 593 * Returns the default charset of this Java virtual machine. 594 * 595 * <p> The default charset is determined during virtual-machine startup and 596 * typically depends upon the locale and charset of the underlying 597 * operating system. 598 * 599 * @return A charset object for the default charset 600 * 601 * @since 1.5 602 */ 603 public static Charset defaultCharset() { 604 if (defaultCharset == null) { 605 synchronized (Charset.class) { 606 String csn = GetPropertyAction 607 .privilegedGetProperty("file.encoding"); 608 Charset cs = lookup(csn); 609 if (cs != null) 610 defaultCharset = cs; 611 else 612 defaultCharset = StandardCharsets.UTF_8; 613 } 614 } 615 return defaultCharset; 616 } 617 618 619 /* -- Instance fields and methods -- */ 620 621 private final String name; // tickles a bug in oldjavac 622 private final String[] aliases; // tickles a bug in oldjavac 623 private Set<String> aliasSet = null; 624 625 /** 626 * Initializes a new charset with the given canonical name and alias 627 * set. 628 * 629 * @param canonicalName 630 * The canonical name of this charset 631 * 632 * @param aliases 633 * An array of this charset's aliases, or null if it has no aliases 634 * 635 * @throws IllegalCharsetNameException 636 * If the canonical name or any of the aliases are illegal 637 */ 638 protected Charset(String canonicalName, String[] aliases) { 639 String[] as = Objects.requireNonNullElse(aliases, zeroAliases); 640 641 // Skip checks for the standard, built-in Charsets we always load 642 // during initialization. 643 if (canonicalName != "ISO-8859-1" 644 && canonicalName != "US-ASCII" 645 && canonicalName != "UTF-8") { 646 checkName(canonicalName); 647 for (int i = 0; i < as.length; i++) { 648 checkName(as[i]); 649 } 650 } 651 this.name = canonicalName; 652 this.aliases = as; 653 } 654 655 /** 656 * Returns this charset's canonical name. 657 * 658 * @return The canonical name of this charset 659 */ 660 public final String name() { 661 return name; 662 } 663 664 /** 665 * Returns a set containing this charset's aliases. 666 * 667 * @return An immutable set of this charset's aliases 668 */ 669 public final Set<String> aliases() { 670 if (aliasSet != null) 671 return aliasSet; 672 int n = aliases.length; 673 HashSet<String> hs = new HashSet<>(n); 674 for (int i = 0; i < n; i++) 675 hs.add(aliases[i]); 676 aliasSet = Collections.unmodifiableSet(hs); 677 return aliasSet; 678 } 679 680 /** 681 * Returns this charset's human-readable name for the default locale. 682 * 683 * <p> The default implementation of this method simply returns this 684 * charset's canonical name. Concrete subclasses of this class may 685 * override this method in order to provide a localized display name. </p> 686 * 687 * @return The display name of this charset in the default locale 688 */ 689 public String displayName() { 690 return name; 691 } 692 693 /** 694 * Tells whether or not this charset is registered in the <a 695 * href="http://www.iana.org/assignments/character-sets">IANA Charset 696 * Registry</a>. 697 * 698 * @return {@code true} if, and only if, this charset is known by its 699 * implementor to be registered with the IANA 700 */ 701 public final boolean isRegistered() { 702 return !name.startsWith("X-") && !name.startsWith("x-"); 703 } 704 705 /** 706 * Returns this charset's human-readable name for the given locale. 707 * 708 * <p> The default implementation of this method simply returns this 709 * charset's canonical name. Concrete subclasses of this class may 710 * override this method in order to provide a localized display name. </p> 711 * 712 * @param locale 713 * The locale for which the display name is to be retrieved 714 * 715 * @return The display name of this charset in the given locale 716 */ 717 public String displayName(Locale locale) { 718 return name; 719 } 720 721 /** 722 * Tells whether or not this charset contains the given charset. 723 * 724 * <p> A charset <i>C</i> is said to <i>contain</i> a charset <i>D</i> if, 725 * and only if, every character representable in <i>D</i> is also 726 * representable in <i>C</i>. If this relationship holds then it is 727 * guaranteed that every string that can be encoded in <i>D</i> can also be 728 * encoded in <i>C</i> without performing any replacements. 729 * 730 * <p> That <i>C</i> contains <i>D</i> does not imply that each character 731 * representable in <i>C</i> by a particular byte sequence is represented 732 * in <i>D</i> by the same byte sequence, although sometimes this is the 733 * case. 734 * 735 * <p> Every charset contains itself. 736 * 737 * <p> This method computes an approximation of the containment relation: 738 * If it returns {@code true} then the given charset is known to be 739 * contained by this charset; if it returns {@code false}, however, then 740 * it is not necessarily the case that the given charset is not contained 741 * in this charset. 742 * 743 * @param cs 744 * The given charset 745 * 746 * @return {@code true} if the given charset is contained in this charset 747 */ 748 public abstract boolean contains(Charset cs); 749 750 /** 751 * Constructs a new decoder for this charset. 752 * 753 * @return A new decoder for this charset 754 */ 755 public abstract CharsetDecoder newDecoder(); 756 757 /** 758 * Constructs a new encoder for this charset. 759 * 760 * @return A new encoder for this charset 761 * 762 * @throws UnsupportedOperationException 763 * If this charset does not support encoding 764 */ 765 public abstract CharsetEncoder newEncoder(); 766 767 /** 768 * Tells whether or not this charset supports encoding. 769 * 770 * <p> Nearly all charsets support encoding. The primary exceptions are 771 * special-purpose <i>auto-detect</i> charsets whose decoders can determine 772 * which of several possible encoding schemes is in use by examining the 773 * input byte sequence. Such charsets do not support encoding because 774 * there is no way to determine which encoding should be used on output. 775 * Implementations of such charsets should override this method to return 776 * {@code false}. </p> 777 * 778 * @return {@code true} if, and only if, this charset supports encoding 779 */ 780 public boolean canEncode() { 781 return true; 782 } 783 784 /** 785 * Convenience method that decodes bytes in this charset into Unicode 786 * characters. 787 * 788 * <p> An invocation of this method upon a charset {@code cs} returns the 789 * same result as the expression 790 * 791 * <pre> 792 * cs.newDecoder() 793 * .onMalformedInput(CodingErrorAction.REPLACE) 794 * .onUnmappableCharacter(CodingErrorAction.REPLACE) 795 * .decode(bb); </pre> 796 * 797 * except that it is potentially more efficient because it can cache 798 * decoders between successive invocations. 799 * 800 * <p> This method always replaces malformed-input and unmappable-character 801 * sequences with this charset's default replacement byte array. In order 802 * to detect such sequences, use the {@link 803 * CharsetDecoder#decode(java.nio.ByteBuffer)} method directly. </p> 804 * 805 * @param bb The byte buffer to be decoded 806 * 807 * @return A char buffer containing the decoded characters 808 */ 809 public final CharBuffer decode(ByteBuffer bb) { 810 try { 811 return ThreadLocalCoders.decoderFor(this) 812 .onMalformedInput(CodingErrorAction.REPLACE) 813 .onUnmappableCharacter(CodingErrorAction.REPLACE) 814 .decode(bb); 815 } catch (CharacterCodingException x) { 816 throw new Error(x); // Can't happen 817 } 818 } 819 820 /** 821 * Convenience method that encodes Unicode characters into bytes in this 822 * charset. 823 * 824 * <p> An invocation of this method upon a charset {@code cs} returns the 825 * same result as the expression 826 * 827 * <pre> 828 * cs.newEncoder() 829 * .onMalformedInput(CodingErrorAction.REPLACE) 830 * .onUnmappableCharacter(CodingErrorAction.REPLACE) 831 * .encode(bb); </pre> 832 * 833 * except that it is potentially more efficient because it can cache 834 * encoders between successive invocations. 835 * 836 * <p> This method always replaces malformed-input and unmappable-character 837 * sequences with this charset's default replacement string. In order to 838 * detect such sequences, use the {@link 839 * CharsetEncoder#encode(java.nio.CharBuffer)} method directly. </p> 840 * 841 * @param cb The char buffer to be encoded 842 * 843 * @return A byte buffer containing the encoded characters 844 */ 845 public final ByteBuffer encode(CharBuffer cb) { 846 try { 847 return ThreadLocalCoders.encoderFor(this) 848 .onMalformedInput(CodingErrorAction.REPLACE) 849 .onUnmappableCharacter(CodingErrorAction.REPLACE) 850 .encode(cb); 851 } catch (CharacterCodingException x) { 852 throw new Error(x); // Can't happen 853 } 854 } 855 856 /** 857 * Convenience method that encodes a string into bytes in this charset. 858 * 859 * <p> An invocation of this method upon a charset {@code cs} returns the 860 * same result as the expression 861 * 862 * <pre> 863 * cs.encode(CharBuffer.wrap(s)); </pre> 864 * 865 * @param str The string to be encoded 866 * 867 * @return A byte buffer containing the encoded characters 868 */ 869 public final ByteBuffer encode(String str) { 870 return encode(CharBuffer.wrap(str)); 871 } 872 873 /** 874 * Compares this charset to another. 875 * 876 * <p> Charsets are ordered by their canonical names, without regard to 877 * case. </p> 878 * 879 * @param that 880 * The charset to which this charset is to be compared 881 * 882 * @return A negative integer, zero, or a positive integer as this charset 883 * is less than, equal to, or greater than the specified charset 884 */ 885 public final int compareTo(Charset that) { 886 return (name().compareToIgnoreCase(that.name())); 887 } 888 889 /** 890 * Computes a hashcode for this charset. 891 * 892 * @return An integer hashcode 893 */ 894 public final int hashCode() { 895 return name().hashCode(); 896 } 897 898 /** 899 * Tells whether or not this object is equal to another. 900 * 901 * <p> Two charsets are equal if, and only if, they have the same canonical 902 * names. A charset is never equal to any other type of object. </p> 903 * 904 * @return {@code true} if, and only if, this charset is equal to the 905 * given object 906 */ 907 public final boolean equals(Object ob) { 908 if (!(ob instanceof Charset)) 909 return false; 910 if (this == ob) 911 return true; 912 return name.equals(((Charset)ob).name()); 913 } 914 915 /** 916 * Returns a string describing this charset. 917 * 918 * @return A string describing this charset 919 */ 920 public final String toString() { 921 return name(); 922 } 923 924 }