1 /*
  2  * Copyright (c) 2000, 2021, 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.Gate;
 29 import jdk.internal.misc.VM;
 30 import sun.nio.cs.ThreadLocalCoders;
 31 import sun.security.action.GetPropertyAction;
 32 
 33 import java.nio.ByteBuffer;
 34 import java.nio.CharBuffer;
 35 import java.nio.charset.spi.CharsetProvider;
 36 import java.security.AccessController;
 37 import java.security.PrivilegedAction;
 38 import java.util.Arrays;
 39 import java.util.Collections;
 40 import java.util.HashSet;
 41 import java.util.Iterator;
 42 import java.util.Locale;
 43 import java.util.Map;
 44 import java.util.NoSuchElementException;
 45 import java.util.Objects;
 46 import java.util.ServiceConfigurationError;
 47 import java.util.ServiceLoader;
 48 import java.util.Set;
 49 import java.util.SortedMap;
 50 import java.util.TreeMap;
 51 
 52 
 53 /**
 54  * A named mapping between sequences of sixteen-bit Unicode <a
 55  * href="../../lang/Character.html#unicode">code units</a> and sequences of
 56  * bytes.  This class defines methods for creating decoders and encoders and
 57  * for retrieving the various names associated with a charset.  Instances of
 58  * this class are immutable.
 59  *
 60  * <p> This class also defines static methods for testing whether a particular
 61  * charset is supported, for locating charset instances by name, and for
 62  * constructing a map that contains every charset for which support is
 63  * available in the current Java virtual machine.  Support for new charsets can
 64  * be added via the service-provider interface defined in the {@link
 65  * java.nio.charset.spi.CharsetProvider} class.
 66  *
 67  * <p> All of the methods defined in this class are safe for use by multiple
 68  * concurrent threads.
 69  *
 70  *
 71  * <h2><a id="names">Charset names</a></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>'&#92;u0041'</code>&nbsp;through&nbsp;<code>'&#92;u005a'</code>),
 79  *
 80  *   <li> The lowercase letters {@code 'a'} through {@code 'z'}
 81  *        (<code>'&#92;u0061'</code>&nbsp;through&nbsp;<code>'&#92;u007a'</code>),
 82  *
 83  *   <li> The digits {@code '0'} through {@code '9'}
 84  *        (<code>'&#92;u0030'</code>&nbsp;through&nbsp;<code>'&#92;u0039'</code>),
 85  *
 86  *   <li> The dash character {@code '-'}
 87  *        (<code>'&#92;u002d'</code>,&nbsp;<small>HYPHEN-MINUS</small>),
 88  *
 89  *   <li> The plus character {@code '+'}
 90  *        (<code>'&#92;u002b'</code>,&nbsp;<small>PLUS SIGN</small>),
 91  *
 92  *   <li> The period character {@code '.'}
 93  *        (<code>'&#92;u002e'</code>,&nbsp;<small>FULL STOP</small>),
 94  *
 95  *   <li> The colon character {@code ':'}
 96  *        (<code>'&#92;u003a'</code>,&nbsp;<small>COLON</small>), and
 97  *
 98  *   <li> The underscore character {@code '_'}
 99  *        (<code>'&#92;u005f'</code>,&nbsp;<small>LOW&nbsp;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&nbsp;2278:&nbsp;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><a id="standard">Standard charsets</a></h2>
142  *
143  *
144  * <p> Every implementation of the Java platform is required to support the
145  * following standard charsets.  Consult the release documentation for your
146  * implementation to see if any other charsets are supported.  The behavior
147  * of such optional charsets may differ between implementations.
148  *
149  * <blockquote><table class="striped" style="width:80%">
150  * <caption style="display:none">Description of standard charsets</caption>
151  * <thead>
152  * <tr><th scope="col" style="text-align:left">Charset</th><th scope="col" style="text-align:left">Description</th></tr>
153  * </thead>
154  * <tbody>
155  * <tr><th scope="row" style="vertical-align:top">{@code US-ASCII}</th>
156  *     <td>Seven-bit ASCII, a.k.a. {@code ISO646-US},
157  *         a.k.a. the Basic Latin block of the Unicode character set</td></tr>
158  * <tr><th scope="row" style="vertical-align:top"><code>ISO-8859-1&nbsp;&nbsp;</code></th>
159  *     <td>ISO Latin Alphabet No. 1, a.k.a. {@code ISO-LATIN-1}</td></tr>
160  * <tr><th scope="row" style="vertical-align:top">{@code UTF-8}</th>
161  *     <td>Eight-bit UCS Transformation Format</td></tr>
162  * <tr><th scope="row" style="vertical-align:top">{@code UTF-16BE}</th>
163  *     <td>Sixteen-bit UCS Transformation Format,
164  *         big-endian byte&nbsp;order</td></tr>
165  * <tr><th scope="row" style="vertical-align:top">{@code UTF-16LE}</th>
166  *     <td>Sixteen-bit UCS Transformation Format,
167  *         little-endian byte&nbsp;order</td></tr>
168  * <tr><th scope="row" style="vertical-align:top">{@code UTF-16}</th>
169  *     <td>Sixteen-bit UCS Transformation Format,
170  *         byte&nbsp;order identified by an optional byte-order mark</td></tr>
171  * </tbody>
172  * </table></blockquote>
173  *
174  * <p> The {@code UTF-8} charset is specified by <a
175  * href="http://www.ietf.org/rfc/rfc2279.txt"><i>RFC&nbsp;2279</i></a>; the
176  * transformation format upon which it is based is specified in
177  * Amendment&nbsp;2 of ISO&nbsp;10646-1 and is also described in the <a
178  * href="http://www.unicode.org/standard/standard.html"><i>Unicode
179  * Standard</i></a>.
180  *
181  * <p> The {@code UTF-16} charsets are specified by <a
182  * href="http://www.ietf.org/rfc/rfc2781.txt"><i>RFC&nbsp;2781</i></a>; the
183  * transformation formats upon which they are based are specified in
184  * Amendment&nbsp;1 of ISO&nbsp;10646-1 and are also described in the <a
185  * href="http://www.unicode.org/standard/standard.html"><i>Unicode
186  * Standard</i></a>.
187  *
188  * <p> The {@code UTF-16} charsets use sixteen-bit quantities and are
189  * therefore sensitive to byte order.  In these encodings the byte order of a
190  * stream may be indicated by an initial <i>byte-order mark</i> represented by
191  * the Unicode character <code>'&#92;uFEFF'</code>.  Byte-order marks are handled
192  * as follows:
193  *
194  * <ul>
195  *
196  *   <li><p> When decoding, the {@code UTF-16BE} and {@code UTF-16LE}
197  *   charsets interpret the initial byte-order marks as a <small>ZERO-WIDTH
198  *   NON-BREAKING SPACE</small>; when encoding, they do not write
199  *   byte-order marks. </p></li>
200  *
201  *   <li><p> When decoding, the {@code UTF-16} charset interprets the
202  *   byte-order mark at the beginning of the input stream to indicate the
203  *   byte-order of the stream but defaults to big-endian if there is no
204  *   byte-order mark; when encoding, it uses big-endian byte order and writes
205  *   a big-endian byte-order mark. </p></li>
206  *
207  * </ul>
208  *
209  * In any case, byte order marks occurring after the first element of an
210  * input sequence are not omitted since the same code is used to represent
211  * <small>ZERO-WIDTH NON-BREAKING SPACE</small>.
212  *
213  * <p> Every instance of the Java virtual machine has a default charset, which
214  * is {@code UTF-8} unless changed in an implementation specific manner. Refer to
215  * {@link #defaultCharset()} for more detail.
216  *
217  * <p> The {@link StandardCharsets} class defines constants for each of the
218  * standard charsets.
219  *
220  * <h2>Terminology</h2>
221  *
222  * <p> The name of this class is taken from the terms used in
223  * <a href="http://www.ietf.org/rfc/rfc2278.txt"><i>RFC&nbsp;2278</i></a>.
224  * In that document a <i>charset</i> is defined as the combination of
225  * one or more coded character sets and a character-encoding scheme.
226  * (This definition is confusing; some other software systems define
227  * <i>charset</i> as a synonym for <i>coded character set</i>.)
228  *
229  * <p> A <i>coded character set</i> is a mapping between a set of abstract
230  * characters and a set of integers.  US-ASCII, ISO&nbsp;8859-1,
231  * JIS&nbsp;X&nbsp;0201, and Unicode are examples of coded character sets.
232  *
233  * <p> Some standards have defined a <i>character set</i> to be simply a
234  * set of abstract characters without an associated assigned numbering.
235  * An alphabet is an example of such a character set.  However, the subtle
236  * distinction between <i>character set</i> and <i>coded character set</i>
237  * is rarely used in practice; the former has become a short form for the
238  * latter, including in the Java API specification.
239  *
240  * <p> A <i>character-encoding scheme</i> is a mapping between one or more
241  * coded character sets and a set of octet (eight-bit byte) sequences.
242  * UTF-8, UTF-16, ISO&nbsp;2022, and EUC are examples of
243  * character-encoding schemes.  Encoding schemes are often associated with
244  * a particular coded character set; UTF-8, for example, is used only to
245  * encode Unicode.  Some schemes, however, are associated with multiple
246  * coded character sets; EUC, for example, can be used to encode
247  * characters in a variety of Asian coded character sets.
248  *
249  * <p> When a coded character set is used exclusively with a single
250  * character-encoding scheme then the corresponding charset is usually
251  * named for the coded character set; otherwise a charset is usually named
252  * for the encoding scheme and, possibly, the locale of the coded
253  * character sets that it supports.  Hence {@code US-ASCII} is both the
254  * name of a coded character set and of the charset that encodes it, while
255  * {@code EUC-JP} is the name of the charset that encodes the
256  * JIS&nbsp;X&nbsp;0201, JIS&nbsp;X&nbsp;0208, and JIS&nbsp;X&nbsp;0212
257  * coded character sets for the Japanese language.
258  *
259  * <p> The native character encoding of the Java programming language is
260  * UTF-16.  A charset in the Java platform therefore defines a mapping
261  * between sequences of sixteen-bit UTF-16 code units (that is, sequences
262  * of chars) and sequences of bytes. </p>
263  *
264  *
265  * @author Mark Reinhold
266  * @author JSR-51 Expert Group
267  * @since 1.4
268  *
269  * @see CharsetDecoder
270  * @see CharsetEncoder
271  * @see java.nio.charset.spi.CharsetProvider
272  * @see java.lang.Character
273  */
274 
275 public abstract class Charset
276     implements Comparable<Charset>
277 {
278 
279     /* -- Static methods -- */
280 
281     /**
282      * Checks that the given string is a legal charset name. </p>
283      *
284      * @param  s
285      *         A purported charset name
286      *
287      * @throws  IllegalCharsetNameException
288      *          If the given name is not a legal charset name
289      */
290     private static void checkName(String s) {
291         int n = s.length();
292         if (n == 0) {
293             throw new IllegalCharsetNameException(s);
294         }
295         for (int i = 0; i < n; i++) {
296             char c = s.charAt(i);
297             if (c >= 'A' && c <= 'Z') continue;
298             if (c >= 'a' && c <= 'z') continue;
299             if (c >= '0' && c <= '9') continue;
300             if (c == '-' && i != 0) continue;
301             if (c == '+' && i != 0) continue;
302             if (c == ':' && i != 0) continue;
303             if (c == '_' && i != 0) continue;
304             if (c == '.' && i != 0) continue;
305             throw new IllegalCharsetNameException(s);
306         }
307     }
308 
309     /* The standard set of charsets */
310     private static final CharsetProvider standardProvider
311         = new sun.nio.cs.StandardCharsets();
312 
313     private static final String[] zeroAliases = new String[0];
314 
315     // Cache of the most-recently-returned charsets,
316     // along with the names that were used to find them
317     //
318     private static volatile Object[] cache1; // "Level 1" cache
319     private static volatile Object[] cache2; // "Level 2" cache
320 
321     private static void cache(String charsetName, Charset cs) {
322         cache2 = cache1;
323         cache1 = new Object[] { charsetName, cs };
324     }
325 
326     // Creates an iterator that walks over the available providers, ignoring
327     // those whose lookup or instantiation causes a security exception to be
328     // thrown.  Should be invoked with full privileges.
329     //
330     private static Iterator<CharsetProvider> providers() {
331         return new Iterator<>() {
332                 ClassLoader cl = ClassLoader.getSystemClassLoader();
333                 ServiceLoader<CharsetProvider> sl =
334                     ServiceLoader.load(CharsetProvider.class, cl);
335                 Iterator<CharsetProvider> i = sl.iterator();
336                 CharsetProvider next = null;
337 
338                 private boolean getNext() {
339                     while (next == null) {
340                         try {
341                             if (!i.hasNext())
342                                 return false;
343                             next = i.next();
344                         } catch (ServiceConfigurationError sce) {
345                             if (sce.getCause() instanceof SecurityException) {
346                                 // Ignore security exceptions
347                                 continue;
348                             }
349                             throw sce;
350                         }
351                     }
352                     return true;
353                 }
354 
355                 public boolean hasNext() {
356                     return getNext();
357                 }
358 
359                 public CharsetProvider next() {
360                     if (!getNext())
361                         throw new NoSuchElementException();
362                     CharsetProvider n = next;
363                     next = null;
364                     return n;
365                 }
366 
367                 public void remove() {
368                     throw new UnsupportedOperationException();
369                 }
370 
371             };
372     }
373 
374     // gate to prevent recursive provider lookups
375     private static final Gate LOOKUP_GATE = Gate.create();
376 
377     @SuppressWarnings("removal")
378     private static Charset lookupViaProviders(final String charsetName) {
379 
380         // The runtime startup sequence looks up standard charsets as a
381         // consequence of the VM's invocation of System.initializeSystemClass
382         // in order to, e.g., set system properties and encode filenames.  At
383         // that point the application class loader has not been initialized,
384         // however, so we can't look for providers because doing so will cause
385         // that loader to be prematurely initialized with incomplete
386         // information.
387         //
388         if (!VM.isBooted())
389             return null;
390 
391         if (!LOOKUP_GATE.tryEnter())
392             // Avoid recursive provider lookups
393             return null;
394         try {
395             return AccessController.doPrivileged(
396                 new PrivilegedAction<>() {
397                     public Charset run() {
398                         for (Iterator<CharsetProvider> i = providers();
399                              i.hasNext();) {
400                             CharsetProvider cp = i.next();
401                             Charset cs = cp.charsetForName(charsetName);
402                             if (cs != null)
403                                 return cs;
404                         }
405                         return null;
406                     }
407                 });
408 
409         } finally {
410             LOOKUP_GATE.exit();
411         }
412     }
413 
414     /* The extended set of charsets */
415     private static class ExtendedProviderHolder {
416         static final CharsetProvider[] extendedProviders = extendedProviders();
417         // returns ExtendedProvider, if installed
418         @SuppressWarnings("removal")
419         private static CharsetProvider[] extendedProviders() {
420             return AccessController.doPrivileged(new PrivilegedAction<>() {
421                     public CharsetProvider[] run() {
422                         CharsetProvider[] cps = new CharsetProvider[1];
423                         int n = 0;
424                         ServiceLoader<CharsetProvider> sl =
425                             ServiceLoader.loadInstalled(CharsetProvider.class);
426                         for (CharsetProvider cp : sl) {
427                             if (n + 1 > cps.length) {
428                                 cps = Arrays.copyOf(cps, cps.length << 1);
429                             }
430                             cps[n++] = cp;
431                         }
432                         return n == cps.length ? cps : Arrays.copyOf(cps, n);
433                     }});
434         }
435     }
436 
437     private static Charset lookupExtendedCharset(String charsetName) {
438         if (!VM.isBooted())  // see lookupViaProviders()
439             return null;
440         CharsetProvider[] ecps = ExtendedProviderHolder.extendedProviders;
441         for (CharsetProvider cp : ecps) {
442             Charset cs = cp.charsetForName(charsetName);
443             if (cs != null)
444                 return cs;
445         }
446         return null;
447     }
448 
449     private static Charset lookup(String charsetName) {
450         if (charsetName == null)
451             throw new IllegalArgumentException("Null charset name");
452         Object[] a;
453         if ((a = cache1) != null && charsetName.equals(a[0]))
454             return (Charset)a[1];
455         // We expect most programs to use one Charset repeatedly.
456         // We convey a hint to this effect to the VM by putting the
457         // level 1 cache miss code in a separate method.
458         return lookup2(charsetName);
459     }
460 
461     private static Charset lookup2(String charsetName) {
462         Object[] a;
463         if ((a = cache2) != null && charsetName.equals(a[0])) {
464             cache2 = cache1;
465             cache1 = a;
466             return (Charset)a[1];
467         }
468         Charset cs;
469         if ((cs = standardProvider.charsetForName(charsetName)) != null ||
470             (cs = lookupExtendedCharset(charsetName))           != null ||
471             (cs = lookupViaProviders(charsetName))              != null)
472         {
473             cache(charsetName, cs);
474             return cs;
475         }
476 
477         /* Only need to check the name if we didn't find a charset for it */
478         checkName(charsetName);
479         return null;
480     }
481 
482     /**
483      * Tells whether the named charset is supported.
484      *
485      * @param  charsetName
486      *         The name of the requested charset; may be either
487      *         a canonical name or an alias
488      *
489      * @return  {@code true} if, and only if, support for the named charset
490      *          is available in the current Java virtual machine
491      *
492      * @throws IllegalCharsetNameException
493      *         If the given charset name is illegal
494      *
495      * @throws  IllegalArgumentException
496      *          If the given {@code charsetName} is null
497      */
498     public static boolean isSupported(String charsetName) {
499         return (lookup(charsetName) != null);
500     }
501 
502     /**
503      * Returns a charset object for the named charset.
504      *
505      * @param  charsetName
506      *         The name of the requested charset; may be either
507      *         a canonical name or an alias
508      *
509      * @return  A charset object for the named charset
510      *
511      * @throws  IllegalCharsetNameException
512      *          If the given charset name is illegal
513      *
514      * @throws  IllegalArgumentException
515      *          If the given {@code charsetName} is null
516      *
517      * @throws  UnsupportedCharsetException
518      *          If no support for the named charset is available
519      *          in this instance of the Java virtual machine
520      */
521     public static Charset forName(String charsetName) {
522         Charset cs = lookup(charsetName);
523         if (cs != null)
524             return cs;
525         throw new UnsupportedCharsetException(charsetName);
526     }
527 
528     // Fold charsets from the given iterator into the given map, ignoring
529     // charsets whose names already have entries in the map.
530     //
531     private static void put(Iterator<Charset> i, Map<String,Charset> m) {
532         while (i.hasNext()) {
533             Charset cs = i.next();
534             if (!m.containsKey(cs.name()))
535                 m.put(cs.name(), cs);
536         }
537     }
538 
539     /**
540      * Constructs a sorted map from canonical charset names to charset objects.
541      *
542      * <p> The map returned by this method will have one entry for each charset
543      * for which support is available in the current Java virtual machine.  If
544      * two or more supported charsets have the same canonical name then the
545      * resulting map will contain just one of them; which one it will contain
546      * is not specified. </p>
547      *
548      * <p> The invocation of this method, and the subsequent use of the
549      * resulting map, may cause time-consuming disk or network I/O operations
550      * to occur.  This method is provided for applications that need to
551      * enumerate all of the available charsets, for example to allow user
552      * charset selection.  This method is not used by the {@link #forName
553      * forName} method, which instead employs an efficient incremental lookup
554      * algorithm.
555      *
556      * <p> This method may return different results at different times if new
557      * charset providers are dynamically made available to the current Java
558      * virtual machine.  In the absence of such changes, the charsets returned
559      * by this method are exactly those that can be retrieved via the {@link
560      * #forName forName} method.  </p>
561      *
562      * @return An immutable, case-insensitive map from canonical charset names
563      *         to charset objects
564      */
565     @SuppressWarnings("removal")
566     public static SortedMap<String,Charset> availableCharsets() {
567         return AccessController.doPrivileged(
568             new PrivilegedAction<>() {
569                 public SortedMap<String,Charset> run() {
570                     TreeMap<String,Charset> m =
571                         new TreeMap<>(
572                             String.CASE_INSENSITIVE_ORDER);
573                     put(standardProvider.charsets(), m);
574                     CharsetProvider[] ecps = ExtendedProviderHolder.extendedProviders;
575                     for (CharsetProvider ecp :ecps) {
576                         put(ecp.charsets(), m);
577                     }
578                     for (Iterator<CharsetProvider> i = providers(); i.hasNext();) {
579                         CharsetProvider cp = i.next();
580                         put(cp.charsets(), m);
581                     }
582                     return Collections.unmodifiableSortedMap(m);
583                 }
584             });
585     }
586 
587     private static volatile Charset defaultCharset;
588 
589     /**
590      * Returns the default charset of this Java virtual machine.
591      *
592      * <p> The default charset is {@code UTF-8}, unless changed in an
593      * implementation specific manner.
594      *
595      * @implNote An implementation may override the default charset with
596      * the system property {@code file.encoding} on the command line. If the
597      * value is {@code COMPAT}, the default charset is derived from
598      * the {@code native.encoding} system property, which typically depends
599      * upon the locale and charset of the underlying operating system.
600      *
601      * @return  A charset object for the default charset
602      * @see <a href="../../lang/System.html#file.encoding">file.encoding</a>
603      * @see <a href="../../lang/System.html#native.encoding">native.encoding</a>
604      *
605      * @since 1.5
606      */
607     public static Charset defaultCharset() {
608         if (defaultCharset == null) {
609             synchronized (Charset.class) {
610                 String csn = GetPropertyAction
611                         .privilegedGetProperty("file.encoding");
612                 Charset cs = lookup(csn);
613                 if (cs != null)
614                     defaultCharset = cs;
615                 else
616                     defaultCharset = sun.nio.cs.UTF_8.INSTANCE;
617             }
618         }
619         return defaultCharset;
620     }
621 
622 
623     /* -- Instance fields and methods -- */
624 
625     private final String name;          // tickles a bug in oldjavac
626     private final String[] aliases;     // tickles a bug in oldjavac
627     private Set<String> aliasSet = null;
628 
629     /**
630      * Initializes a new charset with the given canonical name and alias
631      * set.
632      *
633      * @param  canonicalName
634      *         The canonical name of this charset
635      *
636      * @param  aliases
637      *         An array of this charset's aliases, or null if it has no aliases
638      *
639      * @throws IllegalCharsetNameException
640      *         If the canonical name or any of the aliases are illegal
641      */
642     protected Charset(String canonicalName, String[] aliases) {
643         String[] as = Objects.requireNonNullElse(aliases, zeroAliases);
644 
645         // Skip checks for the standard, built-in Charsets we always load
646         // during initialization.
647         if (canonicalName != "ISO-8859-1"
648                 && canonicalName != "US-ASCII"
649                 && canonicalName != "UTF-8") {
650             checkName(canonicalName);
651             for (int i = 0; i < as.length; i++) {
652                 checkName(as[i]);
653             }
654         }
655         this.name = canonicalName;
656         this.aliases = as;
657     }
658 
659     /**
660      * Returns this charset's canonical name.
661      *
662      * @return  The canonical name of this charset
663      */
664     public final String name() {
665         return name;
666     }
667 
668     /**
669      * Returns a set containing this charset's aliases.
670      *
671      * @return  An immutable set of this charset's aliases
672      */
673     public final Set<String> aliases() {
674         if (aliasSet != null)
675             return aliasSet;
676         int n = aliases.length;
677         HashSet<String> hs = new HashSet<>(n);
678         for (int i = 0; i < n; i++)
679             hs.add(aliases[i]);
680         aliasSet = Collections.unmodifiableSet(hs);
681         return aliasSet;
682     }
683 
684     /**
685      * Returns this charset's human-readable name for the default locale.
686      *
687      * <p> The default implementation of this method simply returns this
688      * charset's canonical name.  Concrete subclasses of this class may
689      * override this method in order to provide a localized display name. </p>
690      *
691      * @return  The display name of this charset in the default locale
692      */
693     public String displayName() {
694         return name;
695     }
696 
697     /**
698      * Tells whether or not this charset is registered in the <a
699      * href="http://www.iana.org/assignments/character-sets">IANA Charset
700      * Registry</a>.
701      *
702      * @return  {@code true} if, and only if, this charset is known by its
703      *          implementor to be registered with the IANA
704      */
705     public final boolean isRegistered() {
706         return !name.startsWith("X-") && !name.startsWith("x-");
707     }
708 
709     /**
710      * Returns this charset's human-readable name for the given locale.
711      *
712      * <p> The default implementation of this method simply returns this
713      * charset's canonical name.  Concrete subclasses of this class may
714      * override this method in order to provide a localized display name. </p>
715      *
716      * @param  locale
717      *         The locale for which the display name is to be retrieved
718      *
719      * @return  The display name of this charset in the given locale
720      */
721     public String displayName(Locale locale) {
722         return name;
723     }
724 
725     /**
726      * Tells whether or not this charset contains the given charset.
727      *
728      * <p> A charset <i>C</i> is said to <i>contain</i> a charset <i>D</i> if,
729      * and only if, every character representable in <i>D</i> is also
730      * representable in <i>C</i>.  If this relationship holds then it is
731      * guaranteed that every string that can be encoded in <i>D</i> can also be
732      * encoded in <i>C</i> without performing any replacements.
733      *
734      * <p> That <i>C</i> contains <i>D</i> does not imply that each character
735      * representable in <i>C</i> by a particular byte sequence is represented
736      * in <i>D</i> by the same byte sequence, although sometimes this is the
737      * case.
738      *
739      * <p> Every charset contains itself.
740      *
741      * <p> This method computes an approximation of the containment relation:
742      * If it returns {@code true} then the given charset is known to be
743      * contained by this charset; if it returns {@code false}, however, then
744      * it is not necessarily the case that the given charset is not contained
745      * in this charset.
746      *
747      * @param   cs
748      *          The given charset
749      *
750      * @return  {@code true} if the given charset is contained in this charset
751      */
752     public abstract boolean contains(Charset cs);
753 
754     /**
755      * Constructs a new decoder for this charset.
756      *
757      * @return  A new decoder for this charset
758      */
759     public abstract CharsetDecoder newDecoder();
760 
761     /**
762      * Constructs a new encoder for this charset.
763      *
764      * @return  A new encoder for this charset
765      *
766      * @throws  UnsupportedOperationException
767      *          If this charset does not support encoding
768      */
769     public abstract CharsetEncoder newEncoder();
770 
771     /**
772      * Tells whether or not this charset supports encoding.
773      *
774      * <p> Nearly all charsets support encoding.  The primary exceptions are
775      * special-purpose <i>auto-detect</i> charsets whose decoders can determine
776      * which of several possible encoding schemes is in use by examining the
777      * input byte sequence.  Such charsets do not support encoding because
778      * there is no way to determine which encoding should be used on output.
779      * Implementations of such charsets should override this method to return
780      * {@code false}. </p>
781      *
782      * @return  {@code true} if, and only if, this charset supports encoding
783      */
784     public boolean canEncode() {
785         return true;
786     }
787 
788     /**
789      * Convenience method that decodes bytes in this charset into Unicode
790      * characters.
791      *
792      * <p> An invocation of this method upon a charset {@code cs} returns the
793      * same result as the expression
794      *
795      * <pre>
796      *     cs.newDecoder()
797      *       .onMalformedInput(CodingErrorAction.REPLACE)
798      *       .onUnmappableCharacter(CodingErrorAction.REPLACE)
799      *       .decode(bb); </pre>
800      *
801      * except that it is potentially more efficient because it can cache
802      * decoders between successive invocations.
803      *
804      * <p> This method always replaces malformed-input and unmappable-character
805      * sequences with this charset's default replacement byte array.  In order
806      * to detect such sequences, use the {@link
807      * CharsetDecoder#decode(java.nio.ByteBuffer)} method directly.  </p>
808      *
809      * @param  bb  The byte buffer to be decoded
810      *
811      * @return  A char buffer containing the decoded characters
812      */
813     public final CharBuffer decode(ByteBuffer bb) {
814         CharsetDecoder decoder;
815         if (Thread.currentThread().isVirtual()) {
816             decoder = newDecoder();
817         } else {
818             decoder = ThreadLocalCoders.decoderFor(this);
819         }
820         try {
821             return decoder.onMalformedInput(CodingErrorAction.REPLACE)
822                 .onUnmappableCharacter(CodingErrorAction.REPLACE)
823                 .decode(bb);
824         } catch (CharacterCodingException x) {
825             throw new Error(x);         // Can't happen
826         }
827     }
828 
829     /**
830      * Convenience method that encodes Unicode characters into bytes in this
831      * charset.
832      *
833      * <p> An invocation of this method upon a charset {@code cs} returns the
834      * same result as the expression
835      *
836      * <pre>
837      *     cs.newEncoder()
838      *       .onMalformedInput(CodingErrorAction.REPLACE)
839      *       .onUnmappableCharacter(CodingErrorAction.REPLACE)
840      *       .encode(bb); </pre>
841      *
842      * except that it is potentially more efficient because it can cache
843      * encoders between successive invocations.
844      *
845      * <p> This method always replaces malformed-input and unmappable-character
846      * sequences with this charset's default replacement string.  In order to
847      * detect such sequences, use the {@link
848      * CharsetEncoder#encode(java.nio.CharBuffer)} method directly.  </p>
849      *
850      * @param  cb  The char buffer to be encoded
851      *
852      * @return  A byte buffer containing the encoded characters
853      */
854     public final ByteBuffer encode(CharBuffer cb) {
855         CharsetEncoder encoder;
856         if (Thread.currentThread().isVirtual()) {
857             encoder = newEncoder();
858         } else {
859             encoder = ThreadLocalCoders.encoderFor(this);
860         }
861         try {
862             return encoder.onMalformedInput(CodingErrorAction.REPLACE)
863                 .onUnmappableCharacter(CodingErrorAction.REPLACE)
864                 .encode(cb);
865         } catch (CharacterCodingException x) {
866             throw new Error(x);         // Can't happen
867         }
868     }
869 
870     /**
871      * Convenience method that encodes a string into bytes in this charset.
872      *
873      * <p> An invocation of this method upon a charset {@code cs} returns the
874      * same result as the expression
875      *
876      * <pre>
877      *     cs.encode(CharBuffer.wrap(s)); </pre>
878      *
879      * @param  str  The string to be encoded
880      *
881      * @return  A byte buffer containing the encoded characters
882      */
883     public final ByteBuffer encode(String str) {
884         return encode(CharBuffer.wrap(str));
885     }
886 
887     /**
888      * Compares this charset to another.
889      *
890      * <p> Charsets are ordered by their canonical names, without regard to
891      * case. </p>
892      *
893      * @param  that
894      *         The charset to which this charset is to be compared
895      *
896      * @return A negative integer, zero, or a positive integer as this charset
897      *         is less than, equal to, or greater than the specified charset
898      */
899     public final int compareTo(Charset that) {
900         return (name().compareToIgnoreCase(that.name()));
901     }
902 
903     /**
904      * Computes a hashcode for this charset.
905      *
906      * @return  An integer hashcode
907      */
908     public final int hashCode() {
909         return name().hashCode();
910     }
911 
912     /**
913      * Tells whether or not this object is equal to another.
914      *
915      * <p> Two charsets are equal if, and only if, they have the same canonical
916      * names.  A charset is never equal to any other type of object.  </p>
917      *
918      * @return  {@code true} if, and only if, this charset is equal to the
919      *          given object
920      */
921     public final boolean equals(Object ob) {
922         if (!(ob instanceof Charset))
923             return false;
924         if (this == ob)
925             return true;
926         return name.equals(((Charset)ob).name());
927     }
928 
929     /**
930      * Returns a string describing this charset.
931      *
932      * @return  A string describing this charset
933      */
934     public final String toString() {
935         return name();
936     }
937 
938 }