1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/rt/emul/compact/src/main/java/java/nio/charset/Charset.java Thu Oct 03 15:40:35 2013 +0200
1.3 @@ -0,0 +1,923 @@
1.4 +/*
1.5 + * Copyright (c) 2000, 2010, Oracle and/or its affiliates. All rights reserved.
1.6 + * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
1.7 + *
1.8 + * This code is free software; you can redistribute it and/or modify it
1.9 + * under the terms of the GNU General Public License version 2 only, as
1.10 + * published by the Free Software Foundation. Oracle designates this
1.11 + * particular file as subject to the "Classpath" exception as provided
1.12 + * by Oracle in the LICENSE file that accompanied this code.
1.13 + *
1.14 + * This code is distributed in the hope that it will be useful, but WITHOUT
1.15 + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
1.16 + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
1.17 + * version 2 for more details (a copy is included in the LICENSE file that
1.18 + * accompanied this code).
1.19 + *
1.20 + * You should have received a copy of the GNU General Public License version
1.21 + * 2 along with this work; if not, write to the Free Software Foundation,
1.22 + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
1.23 + *
1.24 + * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
1.25 + * or visit www.oracle.com if you need additional information or have any
1.26 + * questions.
1.27 + */
1.28 +
1.29 +package java.nio.charset;
1.30 +
1.31 +import java.nio.ByteBuffer;
1.32 +import java.nio.CharBuffer;
1.33 +import java.nio.charset.spi.CharsetProvider;
1.34 +import java.security.AccessController;
1.35 +import java.security.AccessControlException;
1.36 +import java.security.PrivilegedAction;
1.37 +import java.util.Collections;
1.38 +import java.util.HashSet;
1.39 +import java.util.Iterator;
1.40 +import java.util.Locale;
1.41 +import java.util.Map;
1.42 +import java.util.NoSuchElementException;
1.43 +import java.util.Set;
1.44 +import java.util.ServiceLoader;
1.45 +import java.util.ServiceConfigurationError;
1.46 +import java.util.SortedMap;
1.47 +import java.util.TreeMap;
1.48 +import sun.misc.ASCIICaseInsensitiveComparator;
1.49 +import sun.nio.cs.StandardCharsets;
1.50 +import sun.nio.cs.ThreadLocalCoders;
1.51 +import sun.security.action.GetPropertyAction;
1.52 +
1.53 +
1.54 +/**
1.55 + * A named mapping between sequences of sixteen-bit Unicode <a
1.56 + * href="../../lang/Character.html#unicode">code units</a> and sequences of
1.57 + * bytes. This class defines methods for creating decoders and encoders and
1.58 + * for retrieving the various names associated with a charset. Instances of
1.59 + * this class are immutable.
1.60 + *
1.61 + * <p> This class also defines static methods for testing whether a particular
1.62 + * charset is supported, for locating charset instances by name, and for
1.63 + * constructing a map that contains every charset for which support is
1.64 + * available in the current Java virtual machine. Support for new charsets can
1.65 + * be added via the service-provider interface defined in the {@link
1.66 + * java.nio.charset.spi.CharsetProvider} class.
1.67 + *
1.68 + * <p> All of the methods defined in this class are safe for use by multiple
1.69 + * concurrent threads.
1.70 + *
1.71 + *
1.72 + * <a name="names"><a name="charenc">
1.73 + * <h4>Charset names</h4>
1.74 + *
1.75 + * <p> Charsets are named by strings composed of the following characters:
1.76 + *
1.77 + * <ul>
1.78 + *
1.79 + * <li> The uppercase letters <tt>'A'</tt> through <tt>'Z'</tt>
1.80 + * (<tt>'\u0041'</tt> through <tt>'\u005a'</tt>),
1.81 + *
1.82 + * <li> The lowercase letters <tt>'a'</tt> through <tt>'z'</tt>
1.83 + * (<tt>'\u0061'</tt> through <tt>'\u007a'</tt>),
1.84 + *
1.85 + * <li> The digits <tt>'0'</tt> through <tt>'9'</tt>
1.86 + * (<tt>'\u0030'</tt> through <tt>'\u0039'</tt>),
1.87 + *
1.88 + * <li> The dash character <tt>'-'</tt>
1.89 + * (<tt>'\u002d'</tt>, <small>HYPHEN-MINUS</small>),
1.90 + *
1.91 + * <li> The plus character <tt>'+'</tt>
1.92 + * (<tt>'\u002b'</tt>, <small>PLUS SIGN</small>),
1.93 + *
1.94 + * <li> The period character <tt>'.'</tt>
1.95 + * (<tt>'\u002e'</tt>, <small>FULL STOP</small>),
1.96 + *
1.97 + * <li> The colon character <tt>':'</tt>
1.98 + * (<tt>'\u003a'</tt>, <small>COLON</small>), and
1.99 + *
1.100 + * <li> The underscore character <tt>'_'</tt>
1.101 + * (<tt>'\u005f'</tt>, <small>LOW LINE</small>).
1.102 + *
1.103 + * </ul>
1.104 + *
1.105 + * A charset name must begin with either a letter or a digit. The empty string
1.106 + * is not a legal charset name. Charset names are not case-sensitive; that is,
1.107 + * case is always ignored when comparing charset names. Charset names
1.108 + * generally follow the conventions documented in <a
1.109 + * href="http://www.ietf.org/rfc/rfc2278.txt"><i>RFC 2278: IANA Charset
1.110 + * Registration Procedures</i></a>.
1.111 + *
1.112 + * <p> Every charset has a <i>canonical name</i> and may also have one or more
1.113 + * <i>aliases</i>. The canonical name is returned by the {@link #name() name} method
1.114 + * of this class. Canonical names are, by convention, usually in upper case.
1.115 + * The aliases of a charset are returned by the {@link #aliases() aliases}
1.116 + * method.
1.117 + *
1.118 + * <a name="hn">
1.119 + *
1.120 + * <p> Some charsets have an <i>historical name</i> that is defined for
1.121 + * compatibility with previous versions of the Java platform. A charset's
1.122 + * historical name is either its canonical name or one of its aliases. The
1.123 + * historical name is returned by the <tt>getEncoding()</tt> methods of the
1.124 + * {@link java.io.InputStreamReader#getEncoding InputStreamReader} and {@link
1.125 + * java.io.OutputStreamWriter#getEncoding OutputStreamWriter} classes.
1.126 + *
1.127 + * <a name="iana">
1.128 + *
1.129 + * <p> If a charset listed in the <a
1.130 + * href="http://www.iana.org/assignments/character-sets"><i>IANA Charset
1.131 + * Registry</i></a> is supported by an implementation of the Java platform then
1.132 + * its canonical name must be the name listed in the registry. Many charsets
1.133 + * are given more than one name in the registry, in which case the registry
1.134 + * identifies one of the names as <i>MIME-preferred</i>. If a charset has more
1.135 + * than one registry name then its canonical name must be the MIME-preferred
1.136 + * name and the other names in the registry must be valid aliases. If a
1.137 + * supported charset is not listed in the IANA registry then its canonical name
1.138 + * must begin with one of the strings <tt>"X-"</tt> or <tt>"x-"</tt>.
1.139 + *
1.140 + * <p> The IANA charset registry does change over time, and so the canonical
1.141 + * name and the aliases of a particular charset may also change over time. To
1.142 + * ensure compatibility it is recommended that no alias ever be removed from a
1.143 + * charset, and that if the canonical name of a charset is changed then its
1.144 + * previous canonical name be made into an alias.
1.145 + *
1.146 + *
1.147 + * <h4>Standard charsets</h4>
1.148 + *
1.149 + * <a name="standard">
1.150 + *
1.151 + * <p> Every implementation of the Java platform is required to support the
1.152 + * following standard charsets. Consult the release documentation for your
1.153 + * implementation to see if any other charsets are supported. The behavior
1.154 + * of such optional charsets may differ between implementations.
1.155 + *
1.156 + * <blockquote><table width="80%" summary="Description of standard charsets">
1.157 + * <tr><th><p align="left">Charset</p></th><th><p align="left">Description</p></th></tr>
1.158 + * <tr><td valign=top><tt>US-ASCII</tt></td>
1.159 + * <td>Seven-bit ASCII, a.k.a. <tt>ISO646-US</tt>,
1.160 + * a.k.a. the Basic Latin block of the Unicode character set</td></tr>
1.161 + * <tr><td valign=top><tt>ISO-8859-1 </tt></td>
1.162 + * <td>ISO Latin Alphabet No. 1, a.k.a. <tt>ISO-LATIN-1</tt></td></tr>
1.163 + * <tr><td valign=top><tt>UTF-8</tt></td>
1.164 + * <td>Eight-bit UCS Transformation Format</td></tr>
1.165 + * <tr><td valign=top><tt>UTF-16BE</tt></td>
1.166 + * <td>Sixteen-bit UCS Transformation Format,
1.167 + * big-endian byte order</td></tr>
1.168 + * <tr><td valign=top><tt>UTF-16LE</tt></td>
1.169 + * <td>Sixteen-bit UCS Transformation Format,
1.170 + * little-endian byte order</td></tr>
1.171 + * <tr><td valign=top><tt>UTF-16</tt></td>
1.172 + * <td>Sixteen-bit UCS Transformation Format,
1.173 + * byte order identified by an optional byte-order mark</td></tr>
1.174 + * </table></blockquote>
1.175 + *
1.176 + * <p> The <tt>UTF-8</tt> charset is specified by <a
1.177 + * href="http://www.ietf.org/rfc/rfc2279.txt"><i>RFC 2279</i></a>; the
1.178 + * transformation format upon which it is based is specified in
1.179 + * Amendment 2 of ISO 10646-1 and is also described in the <a
1.180 + * href="http://www.unicode.org/unicode/standard/standard.html"><i>Unicode
1.181 + * Standard</i></a>.
1.182 + *
1.183 + * <p> The <tt>UTF-16</tt> charsets are specified by <a
1.184 + * href="http://www.ietf.org/rfc/rfc2781.txt"><i>RFC 2781</i></a>; the
1.185 + * transformation formats upon which they are based are specified in
1.186 + * Amendment 1 of ISO 10646-1 and are also described in the <a
1.187 + * href="http://www.unicode.org/unicode/standard/standard.html"><i>Unicode
1.188 + * Standard</i></a>.
1.189 + *
1.190 + * <p> The <tt>UTF-16</tt> charsets use sixteen-bit quantities and are
1.191 + * therefore sensitive to byte order. In these encodings the byte order of a
1.192 + * stream may be indicated by an initial <i>byte-order mark</i> represented by
1.193 + * the Unicode character <tt>'\uFEFF'</tt>. Byte-order marks are handled
1.194 + * as follows:
1.195 + *
1.196 + * <ul>
1.197 + *
1.198 + * <li><p> When decoding, the <tt>UTF-16BE</tt> and <tt>UTF-16LE</tt>
1.199 + * charsets interpret the initial byte-order marks as a <small>ZERO-WIDTH
1.200 + * NON-BREAKING SPACE</small>; when encoding, they do not write
1.201 + * byte-order marks. </p></li>
1.202 +
1.203 + *
1.204 + * <li><p> When decoding, the <tt>UTF-16</tt> charset interprets the
1.205 + * byte-order mark at the beginning of the input stream to indicate the
1.206 + * byte-order of the stream but defaults to big-endian if there is no
1.207 + * byte-order mark; when encoding, it uses big-endian byte order and writes
1.208 + * a big-endian byte-order mark. </p></li>
1.209 + *
1.210 + * </ul>
1.211 + *
1.212 + * In any case, byte order marks occuring after the first element of an
1.213 + * input sequence are not omitted since the same code is used to represent
1.214 + * <small>ZERO-WIDTH NON-BREAKING SPACE</small>.
1.215 + *
1.216 + * <p> Every instance of the Java virtual machine has a default charset, which
1.217 + * may or may not be one of the standard charsets. The default charset is
1.218 + * determined during virtual-machine startup and typically depends upon the
1.219 + * locale and charset being used by the underlying operating system. </p>
1.220 + *
1.221 + * <p>The {@link StandardCharsets} class defines constants for each of the
1.222 + * standard charsets.
1.223 + *
1.224 + * <h4>Terminology</h4>
1.225 + *
1.226 + * <p> The name of this class is taken from the terms used in
1.227 + * <a href="http://www.ietf.org/rfc/rfc2278.txt"><i>RFC 2278</i></a>.
1.228 + * In that document a <i>charset</i> is defined as the combination of
1.229 + * one or more coded character sets and a character-encoding scheme.
1.230 + * (This definition is confusing; some other software systems define
1.231 + * <i>charset</i> as a synonym for <i>coded character set</i>.)
1.232 + *
1.233 + * <p> A <i>coded character set</i> is a mapping between a set of abstract
1.234 + * characters and a set of integers. US-ASCII, ISO 8859-1,
1.235 + * JIS X 0201, and Unicode are examples of coded character sets.
1.236 + *
1.237 + * <p> Some standards have defined a <i>character set</i> to be simply a
1.238 + * set of abstract characters without an associated assigned numbering.
1.239 + * An alphabet is an example of such a character set. However, the subtle
1.240 + * distinction between <i>character set</i> and <i>coded character set</i>
1.241 + * is rarely used in practice; the former has become a short form for the
1.242 + * latter, including in the Java API specification.
1.243 + *
1.244 + * <p> A <i>character-encoding scheme</i> is a mapping between one or more
1.245 + * coded character sets and a set of octet (eight-bit byte) sequences.
1.246 + * UTF-8, UTF-16, ISO 2022, and EUC are examples of
1.247 + * character-encoding schemes. Encoding schemes are often associated with
1.248 + * a particular coded character set; UTF-8, for example, is used only to
1.249 + * encode Unicode. Some schemes, however, are associated with multiple
1.250 + * coded character sets; EUC, for example, can be used to encode
1.251 + * characters in a variety of Asian coded character sets.
1.252 + *
1.253 + * <p> When a coded character set is used exclusively with a single
1.254 + * character-encoding scheme then the corresponding charset is usually
1.255 + * named for the coded character set; otherwise a charset is usually named
1.256 + * for the encoding scheme and, possibly, the locale of the coded
1.257 + * character sets that it supports. Hence <tt>US-ASCII</tt> is both the
1.258 + * name of a coded character set and of the charset that encodes it, while
1.259 + * <tt>EUC-JP</tt> is the name of the charset that encodes the
1.260 + * JIS X 0201, JIS X 0208, and JIS X 0212
1.261 + * coded character sets for the Japanese language.
1.262 + *
1.263 + * <p> The native character encoding of the Java programming language is
1.264 + * UTF-16. A charset in the Java platform therefore defines a mapping
1.265 + * between sequences of sixteen-bit UTF-16 code units (that is, sequences
1.266 + * of chars) and sequences of bytes. </p>
1.267 + *
1.268 + *
1.269 + * @author Mark Reinhold
1.270 + * @author JSR-51 Expert Group
1.271 + * @since 1.4
1.272 + *
1.273 + * @see CharsetDecoder
1.274 + * @see CharsetEncoder
1.275 + * @see java.nio.charset.spi.CharsetProvider
1.276 + * @see java.lang.Character
1.277 + */
1.278 +
1.279 +public abstract class Charset
1.280 + implements Comparable<Charset>
1.281 +{
1.282 +
1.283 + /* -- Static methods -- */
1.284 +
1.285 + private static volatile String bugLevel = null;
1.286 +
1.287 + static boolean atBugLevel(String bl) { // package-private
1.288 + String level = bugLevel;
1.289 + if (level == null) {
1.290 + if (!sun.misc.VM.isBooted())
1.291 + return false;
1.292 + bugLevel = level = AccessController.doPrivileged(
1.293 + new GetPropertyAction("sun.nio.cs.bugLevel", ""));
1.294 + }
1.295 + return level.equals(bl);
1.296 + }
1.297 +
1.298 + /**
1.299 + * Checks that the given string is a legal charset name. </p>
1.300 + *
1.301 + * @param s
1.302 + * A purported charset name
1.303 + *
1.304 + * @throws IllegalCharsetNameException
1.305 + * If the given name is not a legal charset name
1.306 + */
1.307 + private static void checkName(String s) {
1.308 + int n = s.length();
1.309 + if (!atBugLevel("1.4")) {
1.310 + if (n == 0)
1.311 + throw new IllegalCharsetNameException(s);
1.312 + }
1.313 + for (int i = 0; i < n; i++) {
1.314 + char c = s.charAt(i);
1.315 + if (c >= 'A' && c <= 'Z') continue;
1.316 + if (c >= 'a' && c <= 'z') continue;
1.317 + if (c >= '0' && c <= '9') continue;
1.318 + if (c == '-' && i != 0) continue;
1.319 + if (c == '+' && i != 0) continue;
1.320 + if (c == ':' && i != 0) continue;
1.321 + if (c == '_' && i != 0) continue;
1.322 + if (c == '.' && i != 0) continue;
1.323 + throw new IllegalCharsetNameException(s);
1.324 + }
1.325 + }
1.326 +
1.327 + /* The standard set of charsets */
1.328 + private static CharsetProvider standardProvider = new StandardCharsets();
1.329 +
1.330 + // Cache of the most-recently-returned charsets,
1.331 + // along with the names that were used to find them
1.332 + //
1.333 + private static volatile Object[] cache1 = null; // "Level 1" cache
1.334 + private static volatile Object[] cache2 = null; // "Level 2" cache
1.335 +
1.336 + private static void cache(String charsetName, Charset cs) {
1.337 + cache2 = cache1;
1.338 + cache1 = new Object[] { charsetName, cs };
1.339 + }
1.340 +
1.341 + // Creates an iterator that walks over the available providers, ignoring
1.342 + // those whose lookup or instantiation causes a security exception to be
1.343 + // thrown. Should be invoked with full privileges.
1.344 + //
1.345 + private static Iterator providers() {
1.346 + return new Iterator() {
1.347 +
1.348 + ClassLoader cl = ClassLoader.getSystemClassLoader();
1.349 + ServiceLoader<CharsetProvider> sl =
1.350 + ServiceLoader.load(CharsetProvider.class, cl);
1.351 + Iterator<CharsetProvider> i = sl.iterator();
1.352 +
1.353 + Object next = null;
1.354 +
1.355 + private boolean getNext() {
1.356 + while (next == null) {
1.357 + try {
1.358 + if (!i.hasNext())
1.359 + return false;
1.360 + next = i.next();
1.361 + } catch (ServiceConfigurationError sce) {
1.362 + if (sce.getCause() instanceof SecurityException) {
1.363 + // Ignore security exceptions
1.364 + continue;
1.365 + }
1.366 + throw sce;
1.367 + }
1.368 + }
1.369 + return true;
1.370 + }
1.371 +
1.372 + public boolean hasNext() {
1.373 + return getNext();
1.374 + }
1.375 +
1.376 + public Object next() {
1.377 + if (!getNext())
1.378 + throw new NoSuchElementException();
1.379 + Object n = next;
1.380 + next = null;
1.381 + return n;
1.382 + }
1.383 +
1.384 + public void remove() {
1.385 + throw new UnsupportedOperationException();
1.386 + }
1.387 +
1.388 + };
1.389 + }
1.390 +
1.391 + // Thread-local gate to prevent recursive provider lookups
1.392 + private static ThreadLocal<ThreadLocal> gate = new ThreadLocal<ThreadLocal>();
1.393 +
1.394 + private static Charset lookupViaProviders(final String charsetName) {
1.395 +
1.396 + // The runtime startup sequence looks up standard charsets as a
1.397 + // consequence of the VM's invocation of System.initializeSystemClass
1.398 + // in order to, e.g., set system properties and encode filenames. At
1.399 + // that point the application class loader has not been initialized,
1.400 + // however, so we can't look for providers because doing so will cause
1.401 + // that loader to be prematurely initialized with incomplete
1.402 + // information.
1.403 + //
1.404 + if (!sun.misc.VM.isBooted())
1.405 + return null;
1.406 +
1.407 + if (gate.get() != null)
1.408 + // Avoid recursive provider lookups
1.409 + return null;
1.410 + try {
1.411 + gate.set(gate);
1.412 +
1.413 + return AccessController.doPrivileged(
1.414 + new PrivilegedAction<Charset>() {
1.415 + public Charset run() {
1.416 + for (Iterator i = providers(); i.hasNext();) {
1.417 + CharsetProvider cp = (CharsetProvider)i.next();
1.418 + Charset cs = cp.charsetForName(charsetName);
1.419 + if (cs != null)
1.420 + return cs;
1.421 + }
1.422 + return null;
1.423 + }
1.424 + });
1.425 +
1.426 + } finally {
1.427 + gate.set(null);
1.428 + }
1.429 + }
1.430 +
1.431 + /* The extended set of charsets */
1.432 + private static Object extendedProviderLock = new Object();
1.433 + private static boolean extendedProviderProbed = false;
1.434 + private static CharsetProvider extendedProvider = null;
1.435 +
1.436 + private static void probeExtendedProvider() {
1.437 + AccessController.doPrivileged(new PrivilegedAction<Object>() {
1.438 + public Object run() {
1.439 + try {
1.440 + Class epc
1.441 + = Class.forName("sun.nio.cs.ext.ExtendedCharsets");
1.442 + extendedProvider = (CharsetProvider)epc.newInstance();
1.443 + } catch (ClassNotFoundException x) {
1.444 + // Extended charsets not available
1.445 + // (charsets.jar not present)
1.446 + } catch (InstantiationException x) {
1.447 + throw new Error(x);
1.448 + } catch (IllegalAccessException x) {
1.449 + throw new Error(x);
1.450 + }
1.451 + return null;
1.452 + }
1.453 + });
1.454 + }
1.455 +
1.456 + private static Charset lookupExtendedCharset(String charsetName) {
1.457 + CharsetProvider ecp = null;
1.458 + synchronized (extendedProviderLock) {
1.459 + if (!extendedProviderProbed) {
1.460 + probeExtendedProvider();
1.461 + extendedProviderProbed = true;
1.462 + }
1.463 + ecp = extendedProvider;
1.464 + }
1.465 + return (ecp != null) ? ecp.charsetForName(charsetName) : null;
1.466 + }
1.467 +
1.468 + private static Charset lookup(String charsetName) {
1.469 + if (charsetName == null)
1.470 + throw new IllegalArgumentException("Null charset name");
1.471 +
1.472 + Object[] a;
1.473 + if ((a = cache1) != null && charsetName.equals(a[0]))
1.474 + return (Charset)a[1];
1.475 + // We expect most programs to use one Charset repeatedly.
1.476 + // We convey a hint to this effect to the VM by putting the
1.477 + // level 1 cache miss code in a separate method.
1.478 + return lookup2(charsetName);
1.479 + }
1.480 +
1.481 + private static Charset lookup2(String charsetName) {
1.482 + Object[] a;
1.483 + if ((a = cache2) != null && charsetName.equals(a[0])) {
1.484 + cache2 = cache1;
1.485 + cache1 = a;
1.486 + return (Charset)a[1];
1.487 + }
1.488 +
1.489 + Charset cs;
1.490 + if ((cs = standardProvider.charsetForName(charsetName)) != null ||
1.491 + (cs = lookupExtendedCharset(charsetName)) != null ||
1.492 + (cs = lookupViaProviders(charsetName)) != null)
1.493 + {
1.494 + cache(charsetName, cs);
1.495 + return cs;
1.496 + }
1.497 +
1.498 + /* Only need to check the name if we didn't find a charset for it */
1.499 + checkName(charsetName);
1.500 + return null;
1.501 + }
1.502 +
1.503 + /**
1.504 + * Tells whether the named charset is supported. </p>
1.505 + *
1.506 + * @param charsetName
1.507 + * The name of the requested charset; may be either
1.508 + * a canonical name or an alias
1.509 + *
1.510 + * @return <tt>true</tt> if, and only if, support for the named charset
1.511 + * is available in the current Java virtual machine
1.512 + *
1.513 + * @throws IllegalCharsetNameException
1.514 + * If the given charset name is illegal
1.515 + *
1.516 + * @throws IllegalArgumentException
1.517 + * If the given <tt>charsetName</tt> is null
1.518 + */
1.519 + public static boolean isSupported(String charsetName) {
1.520 + return (lookup(charsetName) != null);
1.521 + }
1.522 +
1.523 + /**
1.524 + * Returns a charset object for the named charset. </p>
1.525 + *
1.526 + * @param charsetName
1.527 + * The name of the requested charset; may be either
1.528 + * a canonical name or an alias
1.529 + *
1.530 + * @return A charset object for the named charset
1.531 + *
1.532 + * @throws IllegalCharsetNameException
1.533 + * If the given charset name is illegal
1.534 + *
1.535 + * @throws IllegalArgumentException
1.536 + * If the given <tt>charsetName</tt> is null
1.537 + *
1.538 + * @throws UnsupportedCharsetException
1.539 + * If no support for the named charset is available
1.540 + * in this instance of the Java virtual machine
1.541 + */
1.542 + public static Charset forName(String charsetName) {
1.543 + Charset cs = lookup(charsetName);
1.544 + if (cs != null)
1.545 + return cs;
1.546 + throw new UnsupportedCharsetException(charsetName);
1.547 + }
1.548 +
1.549 + // Fold charsets from the given iterator into the given map, ignoring
1.550 + // charsets whose names already have entries in the map.
1.551 + //
1.552 + private static void put(Iterator<Charset> i, Map<String,Charset> m) {
1.553 + while (i.hasNext()) {
1.554 + Charset cs = i.next();
1.555 + if (!m.containsKey(cs.name()))
1.556 + m.put(cs.name(), cs);
1.557 + }
1.558 + }
1.559 +
1.560 + /**
1.561 + * Constructs a sorted map from canonical charset names to charset objects.
1.562 + *
1.563 + * <p> The map returned by this method will have one entry for each charset
1.564 + * for which support is available in the current Java virtual machine. If
1.565 + * two or more supported charsets have the same canonical name then the
1.566 + * resulting map will contain just one of them; which one it will contain
1.567 + * is not specified. </p>
1.568 + *
1.569 + * <p> The invocation of this method, and the subsequent use of the
1.570 + * resulting map, may cause time-consuming disk or network I/O operations
1.571 + * to occur. This method is provided for applications that need to
1.572 + * enumerate all of the available charsets, for example to allow user
1.573 + * charset selection. This method is not used by the {@link #forName
1.574 + * forName} method, which instead employs an efficient incremental lookup
1.575 + * algorithm.
1.576 + *
1.577 + * <p> This method may return different results at different times if new
1.578 + * charset providers are dynamically made available to the current Java
1.579 + * virtual machine. In the absence of such changes, the charsets returned
1.580 + * by this method are exactly those that can be retrieved via the {@link
1.581 + * #forName forName} method. </p>
1.582 + *
1.583 + * @return An immutable, case-insensitive map from canonical charset names
1.584 + * to charset objects
1.585 + */
1.586 + public static SortedMap<String,Charset> availableCharsets() {
1.587 + return AccessController.doPrivileged(
1.588 + new PrivilegedAction<SortedMap<String,Charset>>() {
1.589 + public SortedMap<String,Charset> run() {
1.590 + TreeMap<String,Charset> m =
1.591 + new TreeMap<String,Charset>(
1.592 + ASCIICaseInsensitiveComparator.CASE_INSENSITIVE_ORDER);
1.593 + put(standardProvider.charsets(), m);
1.594 + for (Iterator i = providers(); i.hasNext();) {
1.595 + CharsetProvider cp = (CharsetProvider)i.next();
1.596 + put(cp.charsets(), m);
1.597 + }
1.598 + return Collections.unmodifiableSortedMap(m);
1.599 + }
1.600 + });
1.601 + }
1.602 +
1.603 + private static volatile Charset defaultCharset;
1.604 +
1.605 + /**
1.606 + * Returns the default charset of this Java virtual machine.
1.607 + *
1.608 + * <p> The default charset is determined during virtual-machine startup and
1.609 + * typically depends upon the locale and charset of the underlying
1.610 + * operating system.
1.611 + *
1.612 + * @return A charset object for the default charset
1.613 + *
1.614 + * @since 1.5
1.615 + */
1.616 + public static Charset defaultCharset() {
1.617 + if (defaultCharset == null) {
1.618 + synchronized (Charset.class) {
1.619 + String csn = AccessController.doPrivileged(
1.620 + new GetPropertyAction("file.encoding"));
1.621 + Charset cs = lookup(csn);
1.622 + if (cs != null)
1.623 + defaultCharset = cs;
1.624 + else
1.625 + defaultCharset = forName("UTF-8");
1.626 + }
1.627 + }
1.628 + return defaultCharset;
1.629 + }
1.630 +
1.631 +
1.632 + /* -- Instance fields and methods -- */
1.633 +
1.634 + private final String name; // tickles a bug in oldjavac
1.635 + private final String[] aliases; // tickles a bug in oldjavac
1.636 + private Set<String> aliasSet = null;
1.637 +
1.638 + /**
1.639 + * Initializes a new charset with the given canonical name and alias
1.640 + * set. </p>
1.641 + *
1.642 + * @param canonicalName
1.643 + * The canonical name of this charset
1.644 + *
1.645 + * @param aliases
1.646 + * An array of this charset's aliases, or null if it has no aliases
1.647 + *
1.648 + * @throws IllegalCharsetNameException
1.649 + * If the canonical name or any of the aliases are illegal
1.650 + */
1.651 + protected Charset(String canonicalName, String[] aliases) {
1.652 + checkName(canonicalName);
1.653 + String[] as = (aliases == null) ? new String[0] : aliases;
1.654 + for (int i = 0; i < as.length; i++)
1.655 + checkName(as[i]);
1.656 + this.name = canonicalName;
1.657 + this.aliases = as;
1.658 + }
1.659 +
1.660 + /**
1.661 + * Returns this charset's canonical name. </p>
1.662 + *
1.663 + * @return The canonical name of this charset
1.664 + */
1.665 + public final String name() {
1.666 + return name;
1.667 + }
1.668 +
1.669 + /**
1.670 + * Returns a set containing this charset's aliases. </p>
1.671 + *
1.672 + * @return An immutable set of this charset's aliases
1.673 + */
1.674 + public final Set<String> aliases() {
1.675 + if (aliasSet != null)
1.676 + return aliasSet;
1.677 + int n = aliases.length;
1.678 + HashSet<String> hs = new HashSet<String>(n);
1.679 + for (int i = 0; i < n; i++)
1.680 + hs.add(aliases[i]);
1.681 + aliasSet = Collections.unmodifiableSet(hs);
1.682 + return aliasSet;
1.683 + }
1.684 +
1.685 + /**
1.686 + * Returns this charset's human-readable name for the default locale.
1.687 + *
1.688 + * <p> The default implementation of this method simply returns this
1.689 + * charset's canonical name. Concrete subclasses of this class may
1.690 + * override this method in order to provide a localized display name. </p>
1.691 + *
1.692 + * @return The display name of this charset in the default locale
1.693 + */
1.694 + public String displayName() {
1.695 + return name;
1.696 + }
1.697 +
1.698 + /**
1.699 + * Tells whether or not this charset is registered in the <a
1.700 + * href="http://www.iana.org/assignments/character-sets">IANA Charset
1.701 + * Registry</a>. </p>
1.702 + *
1.703 + * @return <tt>true</tt> if, and only if, this charset is known by its
1.704 + * implementor to be registered with the IANA
1.705 + */
1.706 + public final boolean isRegistered() {
1.707 + return !name.startsWith("X-") && !name.startsWith("x-");
1.708 + }
1.709 +
1.710 + /**
1.711 + * Returns this charset's human-readable name for the given locale.
1.712 + *
1.713 + * <p> The default implementation of this method simply returns this
1.714 + * charset's canonical name. Concrete subclasses of this class may
1.715 + * override this method in order to provide a localized display name. </p>
1.716 + *
1.717 + * @param locale
1.718 + * The locale for which the display name is to be retrieved
1.719 + *
1.720 + * @return The display name of this charset in the given locale
1.721 + */
1.722 + public String displayName(Locale locale) {
1.723 + return name;
1.724 + }
1.725 +
1.726 + /**
1.727 + * Tells whether or not this charset contains the given charset.
1.728 + *
1.729 + * <p> A charset <i>C</i> is said to <i>contain</i> a charset <i>D</i> if,
1.730 + * and only if, every character representable in <i>D</i> is also
1.731 + * representable in <i>C</i>. If this relationship holds then it is
1.732 + * guaranteed that every string that can be encoded in <i>D</i> can also be
1.733 + * encoded in <i>C</i> without performing any replacements.
1.734 + *
1.735 + * <p> That <i>C</i> contains <i>D</i> does not imply that each character
1.736 + * representable in <i>C</i> by a particular byte sequence is represented
1.737 + * in <i>D</i> by the same byte sequence, although sometimes this is the
1.738 + * case.
1.739 + *
1.740 + * <p> Every charset contains itself.
1.741 + *
1.742 + * <p> This method computes an approximation of the containment relation:
1.743 + * If it returns <tt>true</tt> then the given charset is known to be
1.744 + * contained by this charset; if it returns <tt>false</tt>, however, then
1.745 + * it is not necessarily the case that the given charset is not contained
1.746 + * in this charset.
1.747 + *
1.748 + * @return <tt>true</tt> if the given charset is contained in this charset
1.749 + */
1.750 + public abstract boolean contains(Charset cs);
1.751 +
1.752 + /**
1.753 + * Constructs a new decoder for this charset. </p>
1.754 + *
1.755 + * @return A new decoder for this charset
1.756 + */
1.757 + public abstract CharsetDecoder newDecoder();
1.758 +
1.759 + /**
1.760 + * Constructs a new encoder for this charset. </p>
1.761 + *
1.762 + * @return A new encoder for this charset
1.763 + *
1.764 + * @throws UnsupportedOperationException
1.765 + * If this charset does not support encoding
1.766 + */
1.767 + public abstract CharsetEncoder newEncoder();
1.768 +
1.769 + /**
1.770 + * Tells whether or not this charset supports encoding.
1.771 + *
1.772 + * <p> Nearly all charsets support encoding. The primary exceptions are
1.773 + * special-purpose <i>auto-detect</i> charsets whose decoders can determine
1.774 + * which of several possible encoding schemes is in use by examining the
1.775 + * input byte sequence. Such charsets do not support encoding because
1.776 + * there is no way to determine which encoding should be used on output.
1.777 + * Implementations of such charsets should override this method to return
1.778 + * <tt>false</tt>. </p>
1.779 + *
1.780 + * @return <tt>true</tt> if, and only if, this charset supports encoding
1.781 + */
1.782 + public boolean canEncode() {
1.783 + return true;
1.784 + }
1.785 +
1.786 + /**
1.787 + * Convenience method that decodes bytes in this charset into Unicode
1.788 + * characters.
1.789 + *
1.790 + * <p> An invocation of this method upon a charset <tt>cs</tt> returns the
1.791 + * same result as the expression
1.792 + *
1.793 + * <pre>
1.794 + * cs.newDecoder()
1.795 + * .onMalformedInput(CodingErrorAction.REPLACE)
1.796 + * .onUnmappableCharacter(CodingErrorAction.REPLACE)
1.797 + * .decode(bb); </pre>
1.798 + *
1.799 + * except that it is potentially more efficient because it can cache
1.800 + * decoders between successive invocations.
1.801 + *
1.802 + * <p> This method always replaces malformed-input and unmappable-character
1.803 + * sequences with this charset's default replacement byte array. In order
1.804 + * to detect such sequences, use the {@link
1.805 + * CharsetDecoder#decode(java.nio.ByteBuffer)} method directly. </p>
1.806 + *
1.807 + * @param bb The byte buffer to be decoded
1.808 + *
1.809 + * @return A char buffer containing the decoded characters
1.810 + */
1.811 + public final CharBuffer decode(ByteBuffer bb) {
1.812 + try {
1.813 + return ThreadLocalCoders.decoderFor(this)
1.814 + .onMalformedInput(CodingErrorAction.REPLACE)
1.815 + .onUnmappableCharacter(CodingErrorAction.REPLACE)
1.816 + .decode(bb);
1.817 + } catch (CharacterCodingException x) {
1.818 + throw new Error(x); // Can't happen
1.819 + }
1.820 + }
1.821 +
1.822 + /**
1.823 + * Convenience method that encodes Unicode characters into bytes in this
1.824 + * charset.
1.825 + *
1.826 + * <p> An invocation of this method upon a charset <tt>cs</tt> returns the
1.827 + * same result as the expression
1.828 + *
1.829 + * <pre>
1.830 + * cs.newEncoder()
1.831 + * .onMalformedInput(CodingErrorAction.REPLACE)
1.832 + * .onUnmappableCharacter(CodingErrorAction.REPLACE)
1.833 + * .encode(bb); </pre>
1.834 + *
1.835 + * except that it is potentially more efficient because it can cache
1.836 + * encoders between successive invocations.
1.837 + *
1.838 + * <p> This method always replaces malformed-input and unmappable-character
1.839 + * sequences with this charset's default replacement string. In order to
1.840 + * detect such sequences, use the {@link
1.841 + * CharsetEncoder#encode(java.nio.CharBuffer)} method directly. </p>
1.842 + *
1.843 + * @param cb The char buffer to be encoded
1.844 + *
1.845 + * @return A byte buffer containing the encoded characters
1.846 + */
1.847 + public final ByteBuffer encode(CharBuffer cb) {
1.848 + try {
1.849 + return ThreadLocalCoders.encoderFor(this)
1.850 + .onMalformedInput(CodingErrorAction.REPLACE)
1.851 + .onUnmappableCharacter(CodingErrorAction.REPLACE)
1.852 + .encode(cb);
1.853 + } catch (CharacterCodingException x) {
1.854 + throw new Error(x); // Can't happen
1.855 + }
1.856 + }
1.857 +
1.858 + /**
1.859 + * Convenience method that encodes a string into bytes in this charset.
1.860 + *
1.861 + * <p> An invocation of this method upon a charset <tt>cs</tt> returns the
1.862 + * same result as the expression
1.863 + *
1.864 + * <pre>
1.865 + * cs.encode(CharBuffer.wrap(s)); </pre>
1.866 + *
1.867 + * @param str The string to be encoded
1.868 + *
1.869 + * @return A byte buffer containing the encoded characters
1.870 + */
1.871 + public final ByteBuffer encode(String str) {
1.872 + return encode(CharBuffer.wrap(str));
1.873 + }
1.874 +
1.875 + /**
1.876 + * Compares this charset to another.
1.877 + *
1.878 + * <p> Charsets are ordered by their canonical names, without regard to
1.879 + * case. </p>
1.880 + *
1.881 + * @param that
1.882 + * The charset to which this charset is to be compared
1.883 + *
1.884 + * @return A negative integer, zero, or a positive integer as this charset
1.885 + * is less than, equal to, or greater than the specified charset
1.886 + */
1.887 + public final int compareTo(Charset that) {
1.888 + return (name().compareToIgnoreCase(that.name()));
1.889 + }
1.890 +
1.891 + /**
1.892 + * Computes a hashcode for this charset. </p>
1.893 + *
1.894 + * @return An integer hashcode
1.895 + */
1.896 + public final int hashCode() {
1.897 + return name().hashCode();
1.898 + }
1.899 +
1.900 + /**
1.901 + * Tells whether or not this object is equal to another.
1.902 + *
1.903 + * <p> Two charsets are equal if, and only if, they have the same canonical
1.904 + * names. A charset is never equal to any other type of object. </p>
1.905 + *
1.906 + * @return <tt>true</tt> if, and only if, this charset is equal to the
1.907 + * given object
1.908 + */
1.909 + public final boolean equals(Object ob) {
1.910 + if (!(ob instanceof Charset))
1.911 + return false;
1.912 + if (this == ob)
1.913 + return true;
1.914 + return name.equals(((Charset)ob).name());
1.915 + }
1.916 +
1.917 + /**
1.918 + * Returns a string describing this charset. </p>
1.919 + *
1.920 + * @return A string describing this charset
1.921 + */
1.922 + public final String toString() {
1.923 + return name();
1.924 + }
1.925 +
1.926 +}