Source Home >> Java Source 1.6.0 >> java.nio.charset.Charset V 0.09

001/*
002 * @(#)Charset.java	1.53 06/06/28
003 *
004 * Copyright 2006 Sun Microsystems, Inc. All rights reserved.
005 * SUN PROPRIETARY/CONFIDENTIAL. Use is subject to license terms.
006 */
007
008package java.nio.charset;
009
010import java.nio.ByteBuffer;
011import java.nio.CharBuffer;
012import java.nio.charset.spi.CharsetProvider;
013import java.security.AccessController;
014import java.security.AccessControlException;
015import java.security.PrivilegedAction;
016import java.util.Collections;
017import java.util.HashSet;
018import java.util.Iterator;
019import java.util.Locale;
020import java.util.Map;
021import java.util.NoSuchElementException;
022import java.util.Set;
023import java.util.SortedMap;
024import java.util.TreeMap;
025import sun.misc.ASCIICaseInsensitiveComparator;
026import sun.misc.Service;
027import sun.misc.ServiceConfigurationError;
028import sun.nio.cs.StandardCharsets;
029import sun.nio.cs.ThreadLocalCoders;
030import sun.security.action.GetPropertyAction;
031
032
033/**
034 * A named mapping between sequences of sixteen-bit Unicode <a
035 * href="../../lang/Character.html#unicode">code units</a> and sequences of
036 * bytes.  This class defines methods for creating decoders and encoders and
037 * for retrieving the various names associated with a charset.  Instances of
038 * this class are immutable.
039 *
040 * <p> This class also defines static methods for testing whether a particular
041 * charset is supported, for locating charset instances by name, and for
042 * constructing a map that contains every charset for which support is
043 * available in the current Java virtual machine.  Support for new charsets can
044 * be added via the service-provider interface defined in the {@link
045 * java.nio.charset.spi.CharsetProvider} class.
046 *
047 * <p> All of the methods defined in this class are safe for use by multiple
048 * concurrent threads.
049 *
050 *
051 * <a name="names"><a name="charenc">
052 * <h4>Charset names</h4>
053 *
054 * <p> Charsets are named by strings composed of the following characters:
055 *
056 * <ul>
057 *
058 *   <li> The uppercase letters <tt>'A'</tt> through <tt>'Z'</tt>
059 *        (<tt>'\u0041'</tt> through <tt>'\u005a'</tt>),
060 * 
061 *   <li> The lowercase letters <tt>'a'</tt> through <tt>'z'</tt>
062 *        (<tt>'\u0061'</tt> through <tt>'\u007a'</tt>),
063 * 
064 *   <li> The digits <tt>'0'</tt> through <tt>'9'</tt>
065 *        (<tt>'\u0030'</tt> through <tt>'\u0039'</tt>),
066 * 
067 *   <li> The dash character <tt>'-'</tt>
068 *        (<tt>'\u002d'</tt>, <small>HYPHEN-MINUS</small>),
069 *
070 *   <li> The period character <tt>'.'</tt>
071 *        (<tt>'\u002e'</tt>, <small>FULL STOP</small>),
072 * 
073 *   <li> The colon character <tt>':'</tt>
074 *        (<tt>'\u003a'</tt>, <small>COLON</small>), and
075 * 
076 *   <li> The underscore character <tt>'_'</tt>
077 *        (<tt>'\u005f'</tt>, <small>LOW LINE</small>).
078 * 
079 * </ul>
080 *
081 * A charset name must begin with either a letter or a digit.  The empty string
082 * is not a legal charset name.  Charset names are not case-sensitive; that is,
083 * case is always ignored when comparing charset names.  Charset names
084 * generally follow the conventions documented in <a
085 * href="http://ietf.org/rfc/rfc2278.txt"><i>RFC 2278: IANA Charset
086 * Registration Procedures</i></a>.
087 *
088 * <p> Every charset has a <i>canonical name</i> and may also have one or more
089 * <i>aliases</i>.  The canonical name is returned by the {@link #name() name} method
090 * of this class.  Canonical names are, by convention, usually in upper case.
091 * The aliases of a charset are returned by the {@link #aliases() aliases}
092 * method.
093 *
094 * <a name="hn">
095 *
096 * <p> Some charsets have an <i>historical name</i> that is defined for
097 * compatibility with previous versions of the Java platform.  A charset's
098 * historical name is either its canonical name or one of its aliases.  The
099 * historical name is returned by the <tt>getEncoding()</tt> methods of the
100 * {@link java.io.InputStreamReader#getEncoding InputStreamReader} and {@link
101 * java.io.OutputStreamWriter#getEncoding OutputStreamWriter} classes.
102 *
103 * <a name="iana">
104 *
105 * <p> If a charset listed in the <a
106 * href="http://www.iana.org/assignments/character-sets"><i>IANA Charset
107 * Registry</i></a> is supported by an implementation of the Java platform then
108 * its canonical name must be the name listed in the registry.  Many charsets
109 * are given more than one name in the registry, in which case the registry
110 * identifies one of the names as <i>MIME-preferred</i>.  If a charset has more
111 * than one registry name then its canonical name must be the MIME-preferred
112 * name and the other names in the registry must be valid aliases.  If a
113 * supported charset is not listed in the IANA registry then its canonical name
114 * must begin with one of the strings <tt>"X-"</tt> or <tt>"x-"</tt>.
115 *
116 * <p> The IANA charset registry does change over time, and so the canonical
117 * name and the aliases of a particular charset may also change over time.  To
118 * ensure compatibility it is recommended that no alias ever be removed from a
119 * charset, and that if the canonical name of a charset is changed then its
120 * previous canonical name be made into an alias.
121 *
122 *
123 * <h4>Standard charsets</h4>
124 *
125 * <p> Every implementation of the Java platform is required to support the
126 * following standard charsets.  Consult the release documentation for your
127 * implementation to see if any other charsets are supported.  The behavior
128 * of such optional charsets may differ between implementations.
129 * 
130 * <blockquote><table width="80%" summary="Description of standard charsets">
131 * <tr><th><p align="left">Charset</p></th><th><p align="left">Description</p></th></tr>
132 * <tr><td valign=top><tt>US-ASCII</tt></td>
133 *     <td>Seven-bit ASCII, a.k.a. <tt>ISO646-US</tt>,
134 *         a.k.a. the Basic Latin block of the Unicode character set</td></tr>
135 * <tr><td valign=top><tt>ISO-8859-1  </tt></td>
136 *     <td>ISO Latin Alphabet No. 1, a.k.a. <tt>ISO-LATIN-1</tt></td></tr>
137 * <tr><td valign=top><tt>UTF-8</tt></td>
138 *     <td>Eight-bit UCS Transformation Format</td></tr>
139 * <tr><td valign=top><tt>UTF-16BE</tt></td>
140 *     <td>Sixteen-bit UCS Transformation Format,
141 *         big-endian byte order</td></tr>
142 * <tr><td valign=top><tt>UTF-16LE</tt></td>
143 *     <td>Sixteen-bit UCS Transformation Format,
144 *         little-endian byte order</td></tr>
145 * <tr><td valign=top><tt>UTF-16</tt></td>
146 *     <td>Sixteen-bit UCS Transformation Format,
147 *         byte order identified by an optional byte-order mark</td></tr>
148 * </table></blockquote>
149 * 
150 * <p> The <tt>UTF-8</tt> charset is specified by <a
151 * href="http://ietf.org/rfc/rfc2279.txt"><i>RFC 2279</i></a>; the
152 * transformation format upon which it is based is specified in
153 * Amendment 2 of ISO 10646-1 and is also described in the <a
154 * href="http://www.unicode.org/unicode/standard/standard.html"><i>Unicode
155 * Standard</i></a>.
156 *
157 * <p> The <tt>UTF-16</tt> charsets are specified by <a
158 * href="http://ietf.org/rfc/rfc2781.txt"><i>RFC 2781</i></a>; the
159 * transformation formats upon which they are based are specified in
160 * Amendment 1 of ISO 10646-1 and are also described in the <a
161 * href="http://www.unicode.org/unicode/standard/standard.html"><i>Unicode
162 * Standard</i></a>.
163 *
164 * <p> The <tt>UTF-16</tt> charsets use sixteen-bit quantities and are
165 * therefore sensitive to byte order.  In these encodings the byte order of a
166 * stream may be indicated by an initial <i>byte-order mark</i> represented by
167 * the Unicode character <tt>'\uFEFF'</tt>.  Byte-order marks are handled
168 * as follows:
169 *
170 * <ul>
171 *
172 *   <li><p> When decoding, the <tt>UTF-16BE</tt> and <tt>UTF-16LE</tt>
173 *   charsets ignore byte-order marks; when encoding, they do not write
174 *   byte-order marks. </p></li>
175 *
176 *   <li><p> When decoding, the <tt>UTF-16</tt> charset interprets a byte-order
177 *   mark to indicate the byte order of the stream but defaults to big-endian
178 *   if there is no byte-order mark; when encoding, it uses big-endian byte
179 *   order and writes a big-endian byte-order mark. </p></li>
180 *
181 * </ul>
182 *
183 * In any case, when a byte-order mark is read at the beginning of a decoding
184 * operation it is omitted from the resulting sequence of characters.  Byte
185 * order marks occuring after the first element of an input sequence are not
186 * omitted since the same code is used to represent <small>ZERO-WIDTH
187 * NON-BREAKING SPACE</small>.
188 *
189 * <p> Every instance of the Java virtual machine has a default charset, which
190 * may or may not be one of the standard charsets.  The default charset is
191 * determined during virtual-machine startup and typically depends upon the
192 * locale and charset being used by the underlying operating system. </p>
193 *
194 *
195 * <h4>Terminology</h4>
196 *
197 * <p> The name of this class is taken from the terms used in <a
198 * href="http://ietf.org/rfc/rfc2278.txt""><i>RFC 2278</i></a>.  In that
199 * document a <i>charset</i> is defined as the combination of a coded character
200 * set and a character-encoding scheme.
201 *
202 * <p> A <i>coded character set</i> is a mapping between a set of abstract
203 * characters and a set of integers.  US-ASCII, ISO 8859-1,
204 * JIS X 0201, and full Unicode, which is the same as
205 * ISO 10646-1, are examples of coded character sets.
206 *
207 * <p> A <i>character-encoding scheme</i> is a mapping between a coded
208 * character set and a set of octet (eight-bit byte) sequences.  UTF-8, UCS-2,
209 * UTF-16, ISO 2022, and EUC are examples of character-encoding schemes.
210 * Encoding schemes are often associated with a particular coded character set;
211 * UTF-8, for example, is used only to encode Unicode.  Some schemes, however,
212 * are associated with multiple character sets; EUC, for example, can be used
213 * to encode characters in a variety of Asian character sets.
214 *
215 * <p> When a coded character set is used exclusively with a single
216 * character-encoding scheme then the corresponding charset is usually named
217 * for the character set; otherwise a charset is usually named for the encoding
218 * scheme and, possibly, the locale of the character sets that it supports.
219 * Hence <tt>US-ASCII</tt> is the name of the charset for US-ASCII while
220 * <tt>EUC-JP</tt> is the name of the charset that encodes the
221 * JIS X 0201, JIS X 0208, and JIS X 0212
222 * character sets.
223 *
224 * <p> The native character encoding of the Java programming language is
225 * UTF-16.  A charset in the Java platform therefore defines a mapping between
226 * sequences of sixteen-bit UTF-16 code units and sequences of bytes. </p>
227 *
228 *
229 * @author Mark Reinhold
230 * @author JSR-51 Expert Group
231 * @version 1.53, 06/06/28
232 * @since 1.4
233 *
234 * @see CharsetDecoder
235 * @see CharsetEncoder
236 * @see java.nio.charset.spi.CharsetProvider
237 * @see java.lang.Character
238 */
239
240public abstract class Charset
241    implements Comparable<Charset>
242{
243
244    /* -- Static methods -- */
245
246    private static String bugLevel = null;
247
248    static boolean atBugLevel(String bl) {		// package-private
249	if (bugLevel == null) {
250	    if (!sun.misc.VM.isBooted())
251		return false;
252	    java.security.PrivilegedAction pa =
253		new GetPropertyAction("sun.nio.cs.bugLevel");
254	    bugLevel = (String)AccessController.doPrivileged(pa);
255	    if (bugLevel == null)
256		bugLevel = "";
257	}
258	return (bugLevel != null) && bugLevel.equals(bl);
259    }
260
261    /**
262     * Checks that the given string is a legal charset name. </p>
263     *
264     * @param  s
265     *         A purported charset name
266     *
267     * @throws  IllegalCharsetNameException
268     *          If the given name is not a legal charset name
269     */
270    private static void checkName(String s) {
271	int n = s.length();
272	if (!atBugLevel("1.4")) {
273	    if (n == 0)
274		throw new IllegalCharsetNameException(s);
275	}
276	for (int i = 0; i < n; i++) {
277	    char c = s.charAt(i);
278	    if (c >= 'A' && c <= 'Z') continue;
279	    if (c >= 'a' && c <= 'z') continue;
280	    if (c >= '0' && c <= '9') continue;
281	    if (c == '-' && i != 0) continue;
282	    if (c == ':' && i != 0) continue;
283	    if (c == '_' && i != 0) continue;
284	    if (c == '.' && i != 0) continue;
285	    throw new IllegalCharsetNameException(s);
286	}
287    }
288
289    /* The standard set of charsets */
290    private static CharsetProvider standardProvider = new StandardCharsets();
291
292    // Cache of the most-recently-returned charsets,
293    // along with the names that were used to find them
294    //
295    private static volatile Object[] cache1 = null; // "Level 1" cache
296    private static volatile Object[] cache2 = null; // "Level 2" cache
297
298    private static void cache(String charsetName, Charset cs) {
299	cache2 = cache1;
300	cache1 = new Object[] { charsetName, cs };
301    }
302
303    // Creates an iterator that walks over the available providers, ignoring
304    // those whose lookup or instantiation causes a security exception to be
305    // thrown.  Should be invoked with full privileges.
306    //
307    private static Iterator providers() {
308	return new Iterator() {
309
310		Class c = java.nio.charset.spi.CharsetProvider.class;
311		ClassLoader cl = ClassLoader.getSystemClassLoader();
312		Iterator i = Service.providers(c, cl);
313		Object next = null;
314
315		private boolean getNext() {
316		    while (next == null) {
317			try {
318			    if (!i.hasNext())
319				return false;
320			    next = i.next();
321			} catch (ServiceConfigurationError sce) {
322			    if (sce.getCause() instanceof SecurityException) {
323				// Ignore security exceptions
324				continue;
325			    }
326			    throw sce;
327			}
328		    }
329		    return true;
330		}
331
332		public boolean hasNext() {
333		    return getNext();
334		}
335
336		public Object next() {
337		    if (!getNext())
338			throw new NoSuchElementException();
339		    Object n = next;
340		    next = null;
341		    return n;
342		}
343
344		public void remove() {
345		    throw new UnsupportedOperationException();
346		}
347
348	    };
349    }
350
351    // Thread-local gate to prevent recursive provider lookups
352    private static ThreadLocal gate = new ThreadLocal();
353
354    private static Charset lookupViaProviders(final String charsetName) {
355
356	// The runtime startup sequence looks up standard charsets as a
357	// consequence of the VM's invocation of System.initializeSystemClass
358	// in order to, e.g., set system properties and encode filenames.  At
359	// that point the application class loader has not been initialized,
360	// however, so we can't look for providers because doing so will cause
361	// that loader to be prematurely initialized with incomplete
362	// information.
363	//
364	if (!sun.misc.VM.isBooted())
365	    return null;
366
367	if (gate.get() != null)
368	    // Avoid recursive provider lookups
369	    return null;
370	try {
371	    gate.set(gate);
372
373	    return (Charset)AccessController
374		.doPrivileged(new PrivilegedAction() {
375		    public Object run() {
376			for (Iterator i = providers(); i.hasNext();) {
377			    CharsetProvider cp = (CharsetProvider)i.next();
378			    Charset cs = cp.charsetForName(charsetName);
379			    if (cs != null)
380				return cs;
381			}
382			return null;
383		    }
384		});
385
386	} finally {
387	    gate.set(null);
388	}
389    }
390
391    /* The extended set of charsets */
392    private static Object extendedProviderLock = new Object();
393    private static boolean extendedProviderProbed = false;
394    private static CharsetProvider extendedProvider = null;
395
396    private static void probeExtendedProvider() {
397	AccessController.doPrivileged(new PrivilegedAction() {
398		public Object run() {
399		    try {
400			Class epc
401			    = Class.forName("sun.nio.cs.ext.ExtendedCharsets");
402			extendedProvider = (CharsetProvider)epc.newInstance();
403		    } catch (ClassNotFoundException x) {
404			// Extended charsets not available
405			// (charsets.jar not present)
406		    } catch (InstantiationException x) {
407			throw new Error(x);
408		    } catch (IllegalAccessException x) {
409			throw new Error(x);
410		    }
411		    return null;
412		}
413	    });
414    }
415
416    private static Charset lookupExtendedCharset(String charsetName) {
417	CharsetProvider ecp = null;
418	synchronized (extendedProviderLock) {
419	    if (!extendedProviderProbed) {
420		probeExtendedProvider();
421		extendedProviderProbed = true;
422	    }
423	    ecp = extendedProvider;
424	}
425	return (ecp != null) ? ecp.charsetForName(charsetName) : null;
426    }
427
428    private static Charset lookup(String charsetName) {
429	if (charsetName == null)
430	    throw new IllegalArgumentException("Null charset name");
431
432	Object[] a;
433	if ((a = cache1) != null && charsetName.equals(a[0]))
434	    return (Charset)a[1];
435	// We expect most programs to use one Charset repeatedly.
436	// We convey a hint to this effect to the VM by putting the
437	// level 1 cache miss code in a separate method.
438	return lookup2(charsetName);
439    }
440
441    private static Charset lookup2(String charsetName) {
442	Object[] a;
443	if ((a = cache2) != null && charsetName.equals(a[0])) {
444	    cache2 = cache1;
445	    cache1 = a;
446	    return (Charset)a[1];
447	}
448
449	Charset cs;
450	if ((cs = standardProvider.charsetForName(charsetName)) != null ||
451	    (cs = lookupExtendedCharset(charsetName))           != null ||
452	    (cs = lookupViaProviders(charsetName))              != null)
453	{
454	    cache(charsetName, cs);
455	    return cs;
456	}
457
458	/* Only need to check the name if we didn't find a charset for it */
459	checkName(charsetName);
460	return null;
461    }
462
463    /**
464     * Tells whether the named charset is supported. </p>
465     *
466     * @param  charsetName
467     *         The name of the requested charset; may be either
468     *         a canonical name or an alias
469     *
470     * @return  <tt>true</tt> if, and only if, support for the named charset
471     *          is available in the current Java virtual machine
472     *
473     * @throws IllegalCharsetNameException
474     *         If the given charset name is illegal
475     *
476     * @throws  IllegalArgumentException
477     *          If the given <tt>charsetName</tt> is null
478     */
479    public static boolean isSupported(String charsetName) {
480	return (lookup(charsetName) != null);
481    }
482
483    /**
484     * Returns a charset object for the named charset. </p>
485     *
486     * @param  charsetName
487     *         The name of the requested charset; may be either
488     *         a canonical name or an alias
489     *
490     * @return  A charset object for the named charset
491     *
492     * @throws  IllegalCharsetNameException
493     *          If the given charset name is illegal
494     *
495     * @throws  IllegalArgumentException
496     *          If the given <tt>charsetName</tt> is null
497     *
498     * @throws  UnsupportedCharsetException
499     *          If no support for the named charset is available
500     *          in this instance of the Java virtual machine
501     */
502    public static Charset forName(String charsetName) {
503	Charset cs = lookup(charsetName);
504	if (cs != null)
505	    return cs;
506	throw new UnsupportedCharsetException(charsetName);
507    }
508
509    // Fold charsets from the given iterator into the given map, ignoring
510    // charsets whose names already have entries in the map.
511    //
512    private static void put(Iterator i, Map m) {
513	while (i.hasNext()) {
514	    Charset cs = (Charset)i.next();
515	    if (!m.containsKey(cs.name()))
516		m.put(cs.name(), cs);
517	}
518    }
519
520    /**
521     * Constructs a sorted map from canonical charset names to charset objects.
522     *
523     * <p> The map returned by this method will have one entry for each charset
524     * for which support is available in the current Java virtual machine.  If
525     * two or more supported charsets have the same canonical name then the
526     * resulting map will contain just one of them; which one it will contain
527     * is not specified. </p>
528     *
529     * <p> The invocation of this method, and the subsequent use of the
530     * resulting map, may cause time-consuming disk or network I/O operations
531     * to occur.  This method is provided for applications that need to
532     * enumerate all of the available charsets, for example to allow user
533     * charset selection.  This method is not used by the {@link #forName
534     * forName} method, which instead employs an efficient incremental lookup
535     * algorithm.
536     *
537     * <p> This method may return different results at different times if new
538     * charset providers are dynamically made available to the current Java
539     * virtual machine.  In the absence of such changes, the charsets returned
540     * by this method are exactly those that can be retrieved via the {@link
541     * #forName forName} method.  </p>
542     *
543     * @return An immutable, case-insensitive map from canonical charset names
544     *         to charset objects
545     */
546    public static SortedMap<String,Charset> availableCharsets() {
547	return (SortedMap)AccessController
548	    .doPrivileged(new PrivilegedAction() {
549		public Object run() {
550		    TreeMap m = new TreeMap(ASCIICaseInsensitiveComparator.CASE_INSENSITIVE_ORDER);
551		    put(standardProvider.charsets(), m);
552		    for (Iterator i = providers(); i.hasNext();) {
553			CharsetProvider cp = (CharsetProvider)i.next();
554			put(cp.charsets(), m);
555		    }
556		    return Collections.unmodifiableSortedMap(m);
557		}
558	    });
559    }
560
561    private static volatile Charset defaultCharset;
562
563    /**
564     * Returns the default charset of this Java virtual machine.
565     *
566     * <p> The default charset is determined during virtual-machine startup and
567     * typically depends upon the locale and charset of the underlying
568     * operating system.
569     *
570     * @return  A charset object for the default charset
571     *
572     * @since 1.5
573     */
574    public static Charset defaultCharset() {
575        if (defaultCharset == null) {
576	    synchronized (Charset.class) {
577		java.security.PrivilegedAction pa =
578		    new GetPropertyAction("file.encoding");
579		String csn = (String)AccessController.doPrivileged(pa);
580		Charset cs = lookup(csn);
581		if (cs != null)
582		    defaultCharset = cs;
583                else 
584		    defaultCharset = forName("UTF-8");
585            }
586	}
587	return defaultCharset;
588    }
589
590
591    /* -- Instance fields and methods -- */
592
593    private final String name;		// tickles a bug in oldjavac
594    private final String[] aliases;	// tickles a bug in oldjavac
595    private Set aliasSet = null;
596
597    /**
598     * Initializes a new charset with the given canonical name and alias
599     * set. </p>
600     *
601     * @param  canonicalName
602     *         The canonical name of this charset
603     *
604     * @param  aliases
605     *         An array of this charset's aliases, or null if it has no aliases
606     *
607     * @throws IllegalCharsetNameException
608     *         If the canonical name or any of the aliases are illegal
609     */
610    protected Charset(String canonicalName, String[] aliases) {
611	checkName(canonicalName);
612	String[] as = (aliases == null) ? new String[0] : aliases;
613	for (int i = 0; i < as.length; i++)
614	    checkName(as[i]);
615	this.name = canonicalName;
616	this.aliases = as;
617    }
618
619    /**
620     * Returns this charset's canonical name. </p>
621     *
622     * @return  The canonical name of this charset
623     */
624    public final String name() {
625	return name;
626    }
627
628    /**
629     * Returns a set containing this charset's aliases. </p>
630     *
631     * @return  An immutable set of this charset's aliases
632     */
633    public final Set<String> aliases() {
634	if (aliasSet != null)
635	    return aliasSet;
636	int n = aliases.length;
637	HashSet hs = new HashSet(n);
638	for (int i = 0; i < n; i++)
639	    hs.add(aliases[i]);
640	aliasSet = Collections.unmodifiableSet(hs);
641	return aliasSet;
642    }
643
644    /**
645     * Returns this charset's human-readable name for the default locale.
646     *
647     * <p> The default implementation of this method simply returns this
648     * charset's canonical name.  Concrete subclasses of this class may
649     * override this method in order to provide a localized display name. </p>
650     *
651     * @return  The display name of this charset in the default locale
652     */
653    public String displayName() {
654	return name;
655    }
656
657    /**
658     * Tells whether or not this charset is registered in the <a
659     * href="http://www.iana.org/assignments/character-sets">IANA Charset
660     * Registry</a>.  </p>
661     *
662     * @return  <tt>true</tt> if, and only if, this charset is known by its
663     *          implementor to be registered with the IANA
664     */
665    public final boolean isRegistered() {
666	return !name.startsWith("X-") && !name.startsWith("x-");
667    }
668
669    /**
670     * Returns this charset's human-readable name for the given locale.
671     *
672     * <p> The default implementation of this method simply returns this
673     * charset's canonical name.  Concrete subclasses of this class may
674     * override this method in order to provide a localized display name. </p>
675     *
676     * @param  locale
677     *         The locale for which the display name is to be retrieved
678     *
679     * @return  The display name of this charset in the given locale
680     */
681    public String displayName(Locale locale) {
682	return name;
683    }
684
685    /**
686     * Tells whether or not this charset contains the given charset.
687     *
688     * <p> A charset <i>C</i> is said to <i>contain</i> a charset <i>D</i> if,
689     * and only if, every character representable in <i>D</i> is also
690     * representable in <i>C</i>.  If this relationship holds then it is
691     * guaranteed that every string that can be encoded in <i>D</i> can also be
692     * encoded in <i>C</i> without performing any replacements.
693     *
694     * <p> That <i>C</i> contains <i>D</i> does not imply that each character
695     * representable in <i>C</i> by a particular byte sequence is represented
696     * in <i>D</i> by the same byte sequence, although sometimes this is the
697     * case.
698     *
699     * <p> Every charset contains itself.
700     *
701     * <p> This method computes an approximation of the containment relation:
702     * If it returns <tt>true</tt> then the given charset is known to be
703     * contained by this charset; if it returns <tt>false</tt>, however, then
704     * it is not necessarily the case that the given charset is not contained
705     * in this charset.
706     *
707     * @return  <tt>true</tt> if the given charset is contained in this charset
708     */
709    public abstract boolean contains(Charset cs);
710
711    /**
712     * Constructs a new decoder for this charset. </p>
713     *
714     * @return  A new decoder for this charset
715     */
716    public abstract CharsetDecoder newDecoder();
717
718    /**
719     * Constructs a new encoder for this charset. </p>
720     *
721     * @return  A new encoder for this charset
722     *
723     * @throws  UnsupportedOperationException
724     *          If this charset does not support encoding
725     */
726    public abstract CharsetEncoder newEncoder();
727
728    /**
729     * Tells whether or not this charset supports encoding.
730     *
731     * <p> Nearly all charsets support encoding.  The primary exceptions are
732     * special-purpose <i>auto-detect</i> charsets whose decoders can determine
733     * which of several possible encoding schemes is in use by examining the
734     * input byte sequence.  Such charsets do not support encoding because
735     * there is no way to determine which encoding should be used on output.
736     * Implementations of such charsets should override this method to return
737     * <tt>false</tt>. </p>
738     *
739     * @return  <tt>true</tt> if, and only if, this charset supports encoding
740     */
741    public boolean canEncode() {
742	return true;
743    }
744
745    /**
746     * Convenience method that decodes bytes in this charset into Unicode
747     * characters.
748     *
749     * <p> An invocation of this method upon a charset <tt>cs</tt> returns the
750     * same result as the expression
751     *
752     * <pre>
753     *     cs.newDecoder()
754     *       .onMalformedInput(CodingErrorAction.REPLACE)
755     *       .onUnmappableCharacter(CodingErrorAction.REPLACE)
756     *       .decode(bb); </pre>
757     *
758     * except that it is potentially more efficient because it can cache
759     * decoders between successive invocations.
760     *
761     * <p> This method always replaces malformed-input and unmappable-character
762     * sequences with this charset's default replacement byte array.  In order
763     * to detect such sequences, use the {@link
764     * CharsetDecoder#decode(java.nio.ByteBuffer)} method directly.  </p>
765     *
766     * @param  bb  The byte buffer to be decoded
767     *
768     * @return  A char buffer containing the decoded characters
769     */
770    public final CharBuffer decode(ByteBuffer bb) {
771	try {
772	    return ThreadLocalCoders.decoderFor(this)
773		.onMalformedInput(CodingErrorAction.REPLACE)
774		.onUnmappableCharacter(CodingErrorAction.REPLACE)
775		.decode(bb);
776	} catch (CharacterCodingException x) {
777	    throw new Error(x);		// Can't happen
778	}
779    }
780
781    /**
782     * Convenience method that encodes Unicode characters into bytes in this
783     * charset.
784     *
785     * <p> An invocation of this method upon a charset <tt>cs</tt> returns the
786     * same result as the expression
787     *
788     * <pre>
789     *     cs.newEncoder()
790     *       .onMalformedInput(CodingErrorAction.REPLACE)
791     *       .onUnmappableCharacter(CodingErrorAction.REPLACE)
792     *       .encode(bb); </pre>
793     *
794     * except that it is potentially more efficient because it can cache
795     * encoders between successive invocations.
796     *
797     * <p> This method always replaces malformed-input and unmappable-character
798     * sequences with this charset's default replacement string.  In order to
799     * detect such sequences, use the {@link
800     * CharsetEncoder#encode(java.nio.CharBuffer)} method directly.  </p>
801     *
802     * @param  cb  The char buffer to be encoded
803     *
804     * @return  A byte buffer containing the encoded characters
805     */
806    public final ByteBuffer encode(CharBuffer cb) {
807	try {
808	    return ThreadLocalCoders.encoderFor(this)
809		.onMalformedInput(CodingErrorAction.REPLACE)
810		.onUnmappableCharacter(CodingErrorAction.REPLACE)
811		.encode(cb);
812	} catch (CharacterCodingException x) {
813	    throw new Error(x);		// Can't happen
814	}
815    }
816
817    /**
818     * Convenience method that encodes a string into bytes in this charset.
819     *
820     * <p> An invocation of this method upon a charset <tt>cs</tt> returns the
821     * same result as the expression
822     *
823     * <pre>
824     *     cs.encode(CharBuffer.wrap(s)); </pre>
825     *
826     * @param  str  The string to be encoded
827     *
828     * @return  A byte buffer containing the encoded characters
829     */
830    public final ByteBuffer encode(String str) {
831	return encode(CharBuffer.wrap(str));
832    }
833
834    /**
835     * Compares this charset to another.
836     *
837     * <p> Charsets are ordered by their canonical names, without regard to
838     * case. </p>
839     *
840     * @param  that
841     *         The charset to which this charset is to be compared
842     *
843     * @return A negative integer, zero, or a positive integer as this charset
844     *         is less than, equal to, or greater than the specified charset
845     */
846    public final int compareTo(Charset that) {
847	return (name().compareToIgnoreCase(that.name()));
848    }
849
850    /**
851     * Computes a hashcode for this charset. </p>
852     *
853     * @return  An integer hashcode
854     */
855    public final int hashCode() {
856	return name().hashCode();
857    }
858
859    /**
860     * Tells whether or not this object is equal to another.
861     *
862     * <p> Two charsets are equal if, and only if, they have the same canonical
863     * names.  A charset is never equal to any other type of object.  </p>
864     *
865     * @return  <tt>true</tt> if, and only if, this charset is equal to the
866     *          given object
867     */
868    public final boolean equals(Object ob) {
869	if (!(ob instanceof Charset))
870	    return false;
871	if (this == ob)
872	    return true;
873	return name.equals(((Charset)ob).name());
874    }
875
876    /**
877     * Returns a string describing this charset. </p>
878     *
879     * @return  A string describing this charset
880     */
881    public final String toString() {
882	return name();
883    }
884
885}