1 /*
   2  * Copyright (c) 1994, 2016, 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.lang;
  27 
  28 import java.io.ObjectStreamField;
  29 import java.io.UnsupportedEncodingException;
  30 import java.lang.invoke.Constable;
  31 import java.lang.invoke.MethodHandles;
  32 import java.nio.charset.Charset;
  33 import java.util.ArrayList;
  34 import java.util.Arrays;
  35 import java.util.Comparator;
  36 import java.util.Formatter;
  37 import java.util.Locale;
  38 import java.util.Objects;
  39 import java.util.Spliterator;
  40 import java.util.StringJoiner;
  41 import java.util.regex.Pattern;
  42 import java.util.regex.PatternSyntaxException;
  43 import java.util.stream.IntStream;
  44 import java.util.stream.StreamSupport;
  45 import jdk.internal.HotSpotIntrinsicCandidate;
  46 import jdk.internal.vm.annotation.Stable;
  47 
  48 /**
  49  * The {@code String} class represents character strings. All
  50  * string literals in Java programs, such as {@code "abc"}, are
  51  * implemented as instances of this class.
  52  * <p>
  53  * Strings are constant; their values cannot be changed after they
  54  * are created. String buffers support mutable strings.
  55  * Because String objects are immutable they can be shared. For example:
  56  * <blockquote><pre>
  57  *     String str = "abc";
  58  * </pre></blockquote><p>
  59  * is equivalent to:
  60  * <blockquote><pre>
  61  *     char data[] = {'a', 'b', 'c'};
  62  *     String str = new String(data);
  63  * </pre></blockquote><p>
  64  * Here are some more examples of how strings can be used:
  65  * <blockquote><pre>
  66  *     System.out.println("abc");
  67  *     String cde = "cde";
  68  *     System.out.println("abc" + cde);
  69  *     String c = "abc".substring(2,3);
  70  *     String d = cde.substring(1, 2);
  71  * </pre></blockquote>
  72  * <p>
  73  * The class {@code String} includes methods for examining
  74  * individual characters of the sequence, for comparing strings, for
  75  * searching strings, for extracting substrings, and for creating a
  76  * copy of a string with all characters translated to uppercase or to
  77  * lowercase. Case mapping is based on the Unicode Standard version
  78  * specified by the {@link java.lang.Character Character} class.
  79  * <p>
  80  * The Java language provides special support for the string
  81  * concatenation operator (&nbsp;+&nbsp;), and for conversion of
  82  * other objects to strings. For additional information on string
  83  * concatenation and conversion, see <i>The Java&trade; Language Specification</i>.
  84  *
  85  * <p> Unless otherwise noted, passing a {@code null} argument to a constructor
  86  * or method in this class will cause a {@link NullPointerException} to be
  87  * thrown.
  88  *
  89  * <p>A {@code String} represents a string in the UTF-16 format
  90  * in which <em>supplementary characters</em> are represented by <em>surrogate
  91  * pairs</em> (see the section <a href="Character.html#unicode">Unicode
  92  * Character Representations</a> in the {@code Character} class for
  93  * more information).
  94  * Index values refer to {@code char} code units, so a supplementary
  95  * character uses two positions in a {@code String}.
  96  * <p>The {@code String} class provides methods for dealing with
  97  * Unicode code points (i.e., characters), in addition to those for
  98  * dealing with Unicode code units (i.e., {@code char} values).
  99  *
 100  * <p>Unless otherwise noted, methods for comparing Strings do not take locale
 101  * into account.  The {@link java.text.Collator} class provides methods for
 102  * finer-grain, locale-sensitive String comparison.
 103  *
 104  * @implNote The implementation of the string concatenation operator is left to
 105  * the discretion of a Java compiler, as long as the compiler ultimately conforms
 106  * to <i>The Java&trade; Language Specification</i>. For example, the {@code javac} compiler
 107  * may implement the operator with {@code StringBuffer}, {@code StringBuilder},
 108  * or {@code java.lang.invoke.StringConcatFactory} depending on the JDK version. The
 109  * implementation of string conversion is typically through the method {@code toString},
 110  * defined by {@code Object} and inherited by all classes in Java.
 111  *
 112  * @author  Lee Boynton
 113  * @author  Arthur van Hoff
 114  * @author  Martin Buchholz
 115  * @author  Ulf Zibis
 116  * @see     java.lang.Object#toString()
 117  * @see     java.lang.StringBuffer
 118  * @see     java.lang.StringBuilder
 119  * @see     java.nio.charset.Charset
 120  * @since   1.0
 121  * @jls     15.18.1 String Concatenation Operator +
 122  */
 123 
 124 public final class String
 125     implements java.io.Serializable, Comparable<String>, CharSequence, Constable<String> {
 126 
 127     /**
 128      * The value is used for character storage.
 129      *
 130      * @implNote This field is trusted by the VM, and is a subject to
 131      * constant folding if String instance is constant. Overwriting this
 132      * field after construction will cause problems.
 133      *
 134      * Additionally, it is marked with {@link Stable} to trust the contents
 135      * of the array. No other facility in JDK provides this functionality (yet).
 136      * {@link Stable} is safe here, because value is never null.
 137      */
 138     @Stable
 139     private final byte[] value;
 140 
 141     /**
 142      * The identifier of the encoding used to encode the bytes in
 143      * {@code value}. The supported values in this implementation are
 144      *
 145      * LATIN1
 146      * UTF16
 147      *
 148      * @implNote This field is trusted by the VM, and is a subject to
 149      * constant folding if String instance is constant. Overwriting this
 150      * field after construction will cause problems.
 151      */
 152     private final byte coder;
 153 
 154     /** Cache the hash code for the string */
 155     private int hash; // Default to 0
 156 
 157     /** use serialVersionUID from JDK 1.0.2 for interoperability */
 158     private static final long serialVersionUID = -6849794470754667710L;
 159 
 160     /**
 161      * If String compaction is disabled, the bytes in {@code value} are
 162      * always encoded in UTF16.
 163      *
 164      * For methods with several possible implementation paths, when String
 165      * compaction is disabled, only one code path is taken.
 166      *
 167      * The instance field value is generally opaque to optimizing JIT
 168      * compilers. Therefore, in performance-sensitive place, an explicit
 169      * check of the static boolean {@code COMPACT_STRINGS} is done first
 170      * before checking the {@code coder} field since the static boolean
 171      * {@code COMPACT_STRINGS} would be constant folded away by an
 172      * optimizing JIT compiler. The idioms for these cases are as follows.
 173      *
 174      * For code such as:
 175      *
 176      *    if (coder == LATIN1) { ... }
 177      *
 178      * can be written more optimally as
 179      *
 180      *    if (coder() == LATIN1) { ... }
 181      *
 182      * or:
 183      *
 184      *    if (COMPACT_STRINGS && coder == LATIN1) { ... }
 185      *
 186      * An optimizing JIT compiler can fold the above conditional as:
 187      *
 188      *    COMPACT_STRINGS == true  => if (coder == LATIN1) { ... }
 189      *    COMPACT_STRINGS == false => if (false)           { ... }
 190      *
 191      * @implNote
 192      * The actual value for this field is injected by JVM. The static
 193      * initialization block is used to set the value here to communicate
 194      * that this static final field is not statically foldable, and to
 195      * avoid any possible circular dependency during vm initialization.
 196      */
 197     static final boolean COMPACT_STRINGS;
 198 
 199     static {
 200         COMPACT_STRINGS = true;
 201     }
 202 
 203     /**
 204      * Class String is special cased within the Serialization Stream Protocol.
 205      *
 206      * A String instance is written into an ObjectOutputStream according to
 207      * <a href="{@docRoot}/../platform/serialization/spec/output.html">
 208      * Object Serialization Specification, Section 6.2, "Stream Elements"</a>
 209      */
 210     private static final ObjectStreamField[] serialPersistentFields =
 211         new ObjectStreamField[0];
 212 
 213     /**
 214      * Initializes a newly created {@code String} object so that it represents
 215      * an empty character sequence.  Note that use of this constructor is
 216      * unnecessary since Strings are immutable.
 217      */
 218     public String() {
 219         this.value = "".value;
 220         this.coder = "".coder;
 221     }
 222 
 223     /**
 224      * Initializes a newly created {@code String} object so that it represents
 225      * the same sequence of characters as the argument; in other words, the
 226      * newly created string is a copy of the argument string. Unless an
 227      * explicit copy of {@code original} is needed, use of this constructor is
 228      * unnecessary since Strings are immutable.
 229      *
 230      * @param  original
 231      *         A {@code String}
 232      */
 233     @HotSpotIntrinsicCandidate
 234     public String(String original) {
 235         this.value = original.value;
 236         this.coder = original.coder;
 237         this.hash = original.hash;
 238     }
 239 
 240     /**
 241      * Allocates a new {@code String} so that it represents the sequence of
 242      * characters currently contained in the character array argument. The
 243      * contents of the character array are copied; subsequent modification of
 244      * the character array does not affect the newly created string.
 245      *
 246      * @param  value
 247      *         The initial value of the string
 248      */
 249     public String(char value[]) {
 250         this(value, 0, value.length, null);
 251     }
 252 
 253     /**
 254      * Allocates a new {@code String} that contains characters from a subarray
 255      * of the character array argument. The {@code offset} argument is the
 256      * index of the first character of the subarray and the {@code count}
 257      * argument specifies the length of the subarray. The contents of the
 258      * subarray are copied; subsequent modification of the character array does
 259      * not affect the newly created string.
 260      *
 261      * @param  value
 262      *         Array that is the source of characters
 263      *
 264      * @param  offset
 265      *         The initial offset
 266      *
 267      * @param  count
 268      *         The length
 269      *
 270      * @throws  IndexOutOfBoundsException
 271      *          If {@code offset} is negative, {@code count} is negative, or
 272      *          {@code offset} is greater than {@code value.length - count}
 273      */
 274     public String(char value[], int offset, int count) {
 275         this(value, offset, count, rangeCheck(value, offset, count));
 276     }
 277 
 278     private static Void rangeCheck(char[] value, int offset, int count) {
 279         checkBoundsOffCount(offset, count, value.length);
 280         return null;
 281     }
 282 
 283     /**
 284      * Allocates a new {@code String} that contains characters from a subarray
 285      * of the <a href="Character.html#unicode">Unicode code point</a> array
 286      * argument.  The {@code offset} argument is the index of the first code
 287      * point of the subarray and the {@code count} argument specifies the
 288      * length of the subarray.  The contents of the subarray are converted to
 289      * {@code char}s; subsequent modification of the {@code int} array does not
 290      * affect the newly created string.
 291      *
 292      * @param  codePoints
 293      *         Array that is the source of Unicode code points
 294      *
 295      * @param  offset
 296      *         The initial offset
 297      *
 298      * @param  count
 299      *         The length
 300      *
 301      * @throws  IllegalArgumentException
 302      *          If any invalid Unicode code point is found in {@code
 303      *          codePoints}
 304      *
 305      * @throws  IndexOutOfBoundsException
 306      *          If {@code offset} is negative, {@code count} is negative, or
 307      *          {@code offset} is greater than {@code codePoints.length - count}
 308      *
 309      * @since  1.5
 310      */
 311     public String(int[] codePoints, int offset, int count) {
 312         checkBoundsOffCount(offset, count, codePoints.length);
 313         if (count == 0) {
 314             this.value = "".value;
 315             this.coder = "".coder;
 316             return;
 317         }
 318         if (COMPACT_STRINGS) {
 319             byte[] val = StringLatin1.toBytes(codePoints, offset, count);
 320             if (val != null) {
 321                 this.coder = LATIN1;
 322                 this.value = val;
 323                 return;
 324             }
 325         }
 326         this.coder = UTF16;
 327         this.value = StringUTF16.toBytes(codePoints, offset, count);
 328     }
 329 
 330     /**
 331      * Allocates a new {@code String} constructed from a subarray of an array
 332      * of 8-bit integer values.
 333      *
 334      * <p> The {@code offset} argument is the index of the first byte of the
 335      * subarray, and the {@code count} argument specifies the length of the
 336      * subarray.
 337      *
 338      * <p> Each {@code byte} in the subarray is converted to a {@code char} as
 339      * specified in the method above.
 340      *
 341      * @deprecated This method does not properly convert bytes into characters.
 342      * As of JDK&nbsp;1.1, the preferred way to do this is via the
 343      * {@code String} constructors that take a {@link
 344      * java.nio.charset.Charset}, charset name, or that use the platform's
 345      * default charset.
 346      *
 347      * @param  ascii
 348      *         The bytes to be converted to characters
 349      *
 350      * @param  hibyte
 351      *         The top 8 bits of each 16-bit Unicode code unit
 352      *
 353      * @param  offset
 354      *         The initial offset
 355      * @param  count
 356      *         The length
 357      *
 358      * @throws  IndexOutOfBoundsException
 359      *          If {@code offset} is negative, {@code count} is negative, or
 360      *          {@code offset} is greater than {@code ascii.length - count}
 361      *
 362      * @see  #String(byte[], int)
 363      * @see  #String(byte[], int, int, java.lang.String)
 364      * @see  #String(byte[], int, int, java.nio.charset.Charset)
 365      * @see  #String(byte[], int, int)
 366      * @see  #String(byte[], java.lang.String)
 367      * @see  #String(byte[], java.nio.charset.Charset)
 368      * @see  #String(byte[])
 369      */
 370     @Deprecated(since="1.1")
 371     public String(byte ascii[], int hibyte, int offset, int count) {
 372         checkBoundsOffCount(offset, count, ascii.length);
 373         if (count == 0) {
 374             this.value = "".value;
 375             this.coder = "".coder;
 376             return;
 377         }
 378         if (COMPACT_STRINGS && (byte)hibyte == 0) {
 379             this.value = Arrays.copyOfRange(ascii, offset, offset + count);
 380             this.coder = LATIN1;
 381         } else {
 382             hibyte <<= 8;
 383             byte[] val = StringUTF16.newBytesFor(count);
 384             for (int i = 0; i < count; i++) {
 385                 StringUTF16.putChar(val, i, hibyte | (ascii[offset++] & 0xff));
 386             }
 387             this.value = val;
 388             this.coder = UTF16;
 389         }
 390     }
 391 
 392     /**
 393      * Allocates a new {@code String} containing characters constructed from
 394      * an array of 8-bit integer values. Each character <i>c</i>in the
 395      * resulting string is constructed from the corresponding component
 396      * <i>b</i> in the byte array such that:
 397      *
 398      * <blockquote><pre>
 399      *     <b><i>c</i></b> == (char)(((hibyte &amp; 0xff) &lt;&lt; 8)
 400      *                         | (<b><i>b</i></b> &amp; 0xff))
 401      * </pre></blockquote>
 402      *
 403      * @deprecated  This method does not properly convert bytes into
 404      * characters.  As of JDK&nbsp;1.1, the preferred way to do this is via the
 405      * {@code String} constructors that take a {@link
 406      * java.nio.charset.Charset}, charset name, or that use the platform's
 407      * default charset.
 408      *
 409      * @param  ascii
 410      *         The bytes to be converted to characters
 411      *
 412      * @param  hibyte
 413      *         The top 8 bits of each 16-bit Unicode code unit
 414      *
 415      * @see  #String(byte[], int, int, java.lang.String)
 416      * @see  #String(byte[], int, int, java.nio.charset.Charset)
 417      * @see  #String(byte[], int, int)
 418      * @see  #String(byte[], java.lang.String)
 419      * @see  #String(byte[], java.nio.charset.Charset)
 420      * @see  #String(byte[])
 421      */
 422     @Deprecated(since="1.1")
 423     public String(byte ascii[], int hibyte) {
 424         this(ascii, hibyte, 0, ascii.length);
 425     }
 426 
 427     /**
 428      * Constructs a new {@code String} by decoding the specified subarray of
 429      * bytes using the specified charset.  The length of the new {@code String}
 430      * is a function of the charset, and hence may not be equal to the length
 431      * of the subarray.
 432      *
 433      * <p> The behavior of this constructor when the given bytes are not valid
 434      * in the given charset is unspecified.  The {@link
 435      * java.nio.charset.CharsetDecoder} class should be used when more control
 436      * over the decoding process is required.
 437      *
 438      * @param  bytes
 439      *         The bytes to be decoded into characters
 440      *
 441      * @param  offset
 442      *         The index of the first byte to decode
 443      *
 444      * @param  length
 445      *         The number of bytes to decode
 446 
 447      * @param  charsetName
 448      *         The name of a supported {@linkplain java.nio.charset.Charset
 449      *         charset}
 450      *
 451      * @throws  UnsupportedEncodingException
 452      *          If the named charset is not supported
 453      *
 454      * @throws  IndexOutOfBoundsException
 455      *          If {@code offset} is negative, {@code length} is negative, or
 456      *          {@code offset} is greater than {@code bytes.length - length}
 457      *
 458      * @since  1.1
 459      */
 460     public String(byte bytes[], int offset, int length, String charsetName)
 461             throws UnsupportedEncodingException {
 462         if (charsetName == null)
 463             throw new NullPointerException("charsetName");
 464         checkBoundsOffCount(offset, length, bytes.length);
 465         StringCoding.Result ret =
 466             StringCoding.decode(charsetName, bytes, offset, length);
 467         this.value = ret.value;
 468         this.coder = ret.coder;
 469     }
 470 
 471     /**
 472      * Constructs a new {@code String} by decoding the specified subarray of
 473      * bytes using the specified {@linkplain java.nio.charset.Charset charset}.
 474      * The length of the new {@code String} is a function of the charset, and
 475      * hence may not be equal to the length of the subarray.
 476      *
 477      * <p> This method always replaces malformed-input and unmappable-character
 478      * sequences with this charset's default replacement string.  The {@link
 479      * java.nio.charset.CharsetDecoder} class should be used when more control
 480      * over the decoding process is required.
 481      *
 482      * @param  bytes
 483      *         The bytes to be decoded into characters
 484      *
 485      * @param  offset
 486      *         The index of the first byte to decode
 487      *
 488      * @param  length
 489      *         The number of bytes to decode
 490      *
 491      * @param  charset
 492      *         The {@linkplain java.nio.charset.Charset charset} to be used to
 493      *         decode the {@code bytes}
 494      *
 495      * @throws  IndexOutOfBoundsException
 496      *          If {@code offset} is negative, {@code length} is negative, or
 497      *          {@code offset} is greater than {@code bytes.length - length}
 498      *
 499      * @since  1.6
 500      */
 501     public String(byte bytes[], int offset, int length, Charset charset) {
 502         if (charset == null)
 503             throw new NullPointerException("charset");
 504         checkBoundsOffCount(offset, length, bytes.length);
 505         StringCoding.Result ret =
 506             StringCoding.decode(charset, bytes, offset, length);
 507         this.value = ret.value;
 508         this.coder = ret.coder;
 509     }
 510 
 511     /**
 512      * Constructs a new {@code String} by decoding the specified array of bytes
 513      * using the specified {@linkplain java.nio.charset.Charset charset}.  The
 514      * length of the new {@code String} is a function of the charset, and hence
 515      * may not be equal to the length of the byte array.
 516      *
 517      * <p> The behavior of this constructor when the given bytes are not valid
 518      * in the given charset is unspecified.  The {@link
 519      * java.nio.charset.CharsetDecoder} class should be used when more control
 520      * over the decoding process is required.
 521      *
 522      * @param  bytes
 523      *         The bytes to be decoded into characters
 524      *
 525      * @param  charsetName
 526      *         The name of a supported {@linkplain java.nio.charset.Charset
 527      *         charset}
 528      *
 529      * @throws  UnsupportedEncodingException
 530      *          If the named charset is not supported
 531      *
 532      * @since  1.1
 533      */
 534     public String(byte bytes[], String charsetName)
 535             throws UnsupportedEncodingException {
 536         this(bytes, 0, bytes.length, charsetName);
 537     }
 538 
 539     /**
 540      * Constructs a new {@code String} by decoding the specified array of
 541      * bytes using the specified {@linkplain java.nio.charset.Charset charset}.
 542      * The length of the new {@code String} is a function of the charset, and
 543      * hence may not be equal to the length of the byte array.
 544      *
 545      * <p> This method always replaces malformed-input and unmappable-character
 546      * sequences with this charset's default replacement string.  The {@link
 547      * java.nio.charset.CharsetDecoder} class should be used when more control
 548      * over the decoding process is required.
 549      *
 550      * @param  bytes
 551      *         The bytes to be decoded into characters
 552      *
 553      * @param  charset
 554      *         The {@linkplain java.nio.charset.Charset charset} to be used to
 555      *         decode the {@code bytes}
 556      *
 557      * @since  1.6
 558      */
 559     public String(byte bytes[], Charset charset) {
 560         this(bytes, 0, bytes.length, charset);
 561     }
 562 
 563     /**
 564      * Constructs a new {@code String} by decoding the specified subarray of
 565      * bytes using the platform's default charset.  The length of the new
 566      * {@code String} is a function of the charset, and hence may not be equal
 567      * to the length of the subarray.
 568      *
 569      * <p> The behavior of this constructor when the given bytes are not valid
 570      * in the default charset is unspecified.  The {@link
 571      * java.nio.charset.CharsetDecoder} class should be used when more control
 572      * over the decoding process is required.
 573      *
 574      * @param  bytes
 575      *         The bytes to be decoded into characters
 576      *
 577      * @param  offset
 578      *         The index of the first byte to decode
 579      *
 580      * @param  length
 581      *         The number of bytes to decode
 582      *
 583      * @throws  IndexOutOfBoundsException
 584      *          If {@code offset} is negative, {@code length} is negative, or
 585      *          {@code offset} is greater than {@code bytes.length - length}
 586      *
 587      * @since  1.1
 588      */
 589     public String(byte bytes[], int offset, int length) {
 590         checkBoundsOffCount(offset, length, bytes.length);
 591         StringCoding.Result ret = StringCoding.decode(bytes, offset, length);
 592         this.value = ret.value;
 593         this.coder = ret.coder;
 594     }
 595 
 596     /**
 597      * Constructs a new {@code String} by decoding the specified array of bytes
 598      * using the platform's default charset.  The length of the new {@code
 599      * String} is a function of the charset, and hence may not be equal to the
 600      * length of the byte array.
 601      *
 602      * <p> The behavior of this constructor when the given bytes are not valid
 603      * in the default charset is unspecified.  The {@link
 604      * java.nio.charset.CharsetDecoder} class should be used when more control
 605      * over the decoding process is required.
 606      *
 607      * @param  bytes
 608      *         The bytes to be decoded into characters
 609      *
 610      * @since  1.1
 611      */
 612     public String(byte[] bytes) {
 613         this(bytes, 0, bytes.length);
 614     }
 615 
 616     /**
 617      * Allocates a new string that contains the sequence of characters
 618      * currently contained in the string buffer argument. The contents of the
 619      * string buffer are copied; subsequent modification of the string buffer
 620      * does not affect the newly created string.
 621      *
 622      * @param  buffer
 623      *         A {@code StringBuffer}
 624      */
 625     public String(StringBuffer buffer) {
 626         this(buffer.toString());
 627     }
 628 
 629     /**
 630      * Allocates a new string that contains the sequence of characters
 631      * currently contained in the string builder argument. The contents of the
 632      * string builder are copied; subsequent modification of the string builder
 633      * does not affect the newly created string.
 634      *
 635      * <p> This constructor is provided to ease migration to {@code
 636      * StringBuilder}. Obtaining a string from a string builder via the {@code
 637      * toString} method is likely to run faster and is generally preferred.
 638      *
 639      * @param   builder
 640      *          A {@code StringBuilder}
 641      *
 642      * @since  1.5
 643      */
 644     public String(StringBuilder builder) {
 645         this(builder, null);
 646     }
 647 
 648    /*
 649     * Package private constructor which shares value array for speed.
 650     * this constructor is always expected to be called with share==true.
 651     * a separate constructor is needed because we already have a public
 652     * String(char[]) constructor that makes a copy of the given char[].
 653     */
 654     // TBD: this is kept for package internal use (Thread/System),
 655     // should be removed if they all have a byte[] version
 656     String(char[] val, boolean share) {
 657         // assert share : "unshared not supported";
 658         this(val, 0, val.length, null);
 659     }
 660 
 661     /**
 662      * Returns the length of this string.
 663      * The length is equal to the number of <a href="Character.html#unicode">Unicode
 664      * code units</a> in the string.
 665      *
 666      * @return  the length of the sequence of characters represented by this
 667      *          object.
 668      */
 669     public int length() {
 670         return value.length >> coder();
 671     }
 672 
 673     /**
 674      * Returns {@code true} if, and only if, {@link #length()} is {@code 0}.
 675      *
 676      * @return {@code true} if {@link #length()} is {@code 0}, otherwise
 677      * {@code false}
 678      *
 679      * @since 1.6
 680      */
 681     public boolean isEmpty() {
 682         return value.length == 0;
 683     }
 684 
 685     /**
 686      * Returns the {@code char} value at the
 687      * specified index. An index ranges from {@code 0} to
 688      * {@code length() - 1}. The first {@code char} value of the sequence
 689      * is at index {@code 0}, the next at index {@code 1},
 690      * and so on, as for array indexing.
 691      *
 692      * <p>If the {@code char} value specified by the index is a
 693      * <a href="Character.html#unicode">surrogate</a>, the surrogate
 694      * value is returned.
 695      *
 696      * @param      index   the index of the {@code char} value.
 697      * @return     the {@code char} value at the specified index of this string.
 698      *             The first {@code char} value is at index {@code 0}.
 699      * @exception  IndexOutOfBoundsException  if the {@code index}
 700      *             argument is negative or not less than the length of this
 701      *             string.
 702      */
 703     public char charAt(int index) {
 704         if (isLatin1()) {
 705             return StringLatin1.charAt(value, index);
 706         } else {
 707             return StringUTF16.charAt(value, index);
 708         }
 709     }
 710 
 711     /**
 712      * Returns the character (Unicode code point) at the specified
 713      * index. The index refers to {@code char} values
 714      * (Unicode code units) and ranges from {@code 0} to
 715      * {@link #length()}{@code  - 1}.
 716      *
 717      * <p> If the {@code char} value specified at the given index
 718      * is in the high-surrogate range, the following index is less
 719      * than the length of this {@code String}, and the
 720      * {@code char} value at the following index is in the
 721      * low-surrogate range, then the supplementary code point
 722      * corresponding to this surrogate pair is returned. Otherwise,
 723      * the {@code char} value at the given index is returned.
 724      *
 725      * @param      index the index to the {@code char} values
 726      * @return     the code point value of the character at the
 727      *             {@code index}
 728      * @exception  IndexOutOfBoundsException  if the {@code index}
 729      *             argument is negative or not less than the length of this
 730      *             string.
 731      * @since      1.5
 732      */
 733     public int codePointAt(int index) {
 734         if (isLatin1()) {
 735             checkIndex(index, value.length);
 736             return value[index] & 0xff;
 737         }
 738         int length = value.length >> 1;
 739         checkIndex(index, length);
 740         return StringUTF16.codePointAt(value, index, length);
 741     }
 742 
 743     /**
 744      * Returns the character (Unicode code point) before the specified
 745      * index. The index refers to {@code char} values
 746      * (Unicode code units) and ranges from {@code 1} to {@link
 747      * CharSequence#length() length}.
 748      *
 749      * <p> If the {@code char} value at {@code (index - 1)}
 750      * is in the low-surrogate range, {@code (index - 2)} is not
 751      * negative, and the {@code char} value at {@code (index -
 752      * 2)} is in the high-surrogate range, then the
 753      * supplementary code point value of the surrogate pair is
 754      * returned. If the {@code char} value at {@code index -
 755      * 1} is an unpaired low-surrogate or a high-surrogate, the
 756      * surrogate value is returned.
 757      *
 758      * @param     index the index following the code point that should be returned
 759      * @return    the Unicode code point value before the given index.
 760      * @exception IndexOutOfBoundsException if the {@code index}
 761      *            argument is less than 1 or greater than the length
 762      *            of this string.
 763      * @since     1.5
 764      */
 765     public int codePointBefore(int index) {
 766         int i = index - 1;
 767         if (i < 0 || i >= length()) {
 768             throw new StringIndexOutOfBoundsException(index);
 769         }
 770         if (isLatin1()) {
 771             return (value[i] & 0xff);
 772         }
 773         return StringUTF16.codePointBefore(value, index);
 774     }
 775 
 776     /**
 777      * Returns the number of Unicode code points in the specified text
 778      * range of this {@code String}. The text range begins at the
 779      * specified {@code beginIndex} and extends to the
 780      * {@code char} at index {@code endIndex - 1}. Thus the
 781      * length (in {@code char}s) of the text range is
 782      * {@code endIndex-beginIndex}. Unpaired surrogates within
 783      * the text range count as one code point each.
 784      *
 785      * @param beginIndex the index to the first {@code char} of
 786      * the text range.
 787      * @param endIndex the index after the last {@code char} of
 788      * the text range.
 789      * @return the number of Unicode code points in the specified text
 790      * range
 791      * @exception IndexOutOfBoundsException if the
 792      * {@code beginIndex} is negative, or {@code endIndex}
 793      * is larger than the length of this {@code String}, or
 794      * {@code beginIndex} is larger than {@code endIndex}.
 795      * @since  1.5
 796      */
 797     public int codePointCount(int beginIndex, int endIndex) {
 798         if (beginIndex < 0 || beginIndex > endIndex ||
 799             endIndex > length()) {
 800             throw new IndexOutOfBoundsException();
 801         }
 802         if (isLatin1()) {
 803             return endIndex - beginIndex;
 804         }
 805         return StringUTF16.codePointCount(value, beginIndex, endIndex);
 806     }
 807 
 808     /**
 809      * Returns the index within this {@code String} that is
 810      * offset from the given {@code index} by
 811      * {@code codePointOffset} code points. Unpaired surrogates
 812      * within the text range given by {@code index} and
 813      * {@code codePointOffset} count as one code point each.
 814      *
 815      * @param index the index to be offset
 816      * @param codePointOffset the offset in code points
 817      * @return the index within this {@code String}
 818      * @exception IndexOutOfBoundsException if {@code index}
 819      *   is negative or larger then the length of this
 820      *   {@code String}, or if {@code codePointOffset} is positive
 821      *   and the substring starting with {@code index} has fewer
 822      *   than {@code codePointOffset} code points,
 823      *   or if {@code codePointOffset} is negative and the substring
 824      *   before {@code index} has fewer than the absolute value
 825      *   of {@code codePointOffset} code points.
 826      * @since 1.5
 827      */
 828     public int offsetByCodePoints(int index, int codePointOffset) {
 829         if (index < 0 || index > length()) {
 830             throw new IndexOutOfBoundsException();
 831         }
 832         return Character.offsetByCodePoints(this, index, codePointOffset);
 833     }
 834 
 835     /**
 836      * Copies characters from this string into the destination character
 837      * array.
 838      * <p>
 839      * The first character to be copied is at index {@code srcBegin};
 840      * the last character to be copied is at index {@code srcEnd-1}
 841      * (thus the total number of characters to be copied is
 842      * {@code srcEnd-srcBegin}). The characters are copied into the
 843      * subarray of {@code dst} starting at index {@code dstBegin}
 844      * and ending at index:
 845      * <blockquote><pre>
 846      *     dstBegin + (srcEnd-srcBegin) - 1
 847      * </pre></blockquote>
 848      *
 849      * @param      srcBegin   index of the first character in the string
 850      *                        to copy.
 851      * @param      srcEnd     index after the last character in the string
 852      *                        to copy.
 853      * @param      dst        the destination array.
 854      * @param      dstBegin   the start offset in the destination array.
 855      * @exception IndexOutOfBoundsException If any of the following
 856      *            is true:
 857      *            <ul><li>{@code srcBegin} is negative.
 858      *            <li>{@code srcBegin} is greater than {@code srcEnd}
 859      *            <li>{@code srcEnd} is greater than the length of this
 860      *                string
 861      *            <li>{@code dstBegin} is negative
 862      *            <li>{@code dstBegin+(srcEnd-srcBegin)} is larger than
 863      *                {@code dst.length}</ul>
 864      */
 865     public void getChars(int srcBegin, int srcEnd, char dst[], int dstBegin) {
 866         checkBoundsBeginEnd(srcBegin, srcEnd, length());
 867         checkBoundsOffCount(dstBegin, srcEnd - srcBegin, dst.length);
 868         if (isLatin1()) {
 869             StringLatin1.getChars(value, srcBegin, srcEnd, dst, dstBegin);
 870         } else {
 871             StringUTF16.getChars(value, srcBegin, srcEnd, dst, dstBegin);
 872         }
 873     }
 874 
 875     /**
 876      * Copies characters from this string into the destination byte array. Each
 877      * byte receives the 8 low-order bits of the corresponding character. The
 878      * eight high-order bits of each character are not copied and do not
 879      * participate in the transfer in any way.
 880      *
 881      * <p> The first character to be copied is at index {@code srcBegin}; the
 882      * last character to be copied is at index {@code srcEnd-1}.  The total
 883      * number of characters to be copied is {@code srcEnd-srcBegin}. The
 884      * characters, converted to bytes, are copied into the subarray of {@code
 885      * dst} starting at index {@code dstBegin} and ending at index:
 886      *
 887      * <blockquote><pre>
 888      *     dstBegin + (srcEnd-srcBegin) - 1
 889      * </pre></blockquote>
 890      *
 891      * @deprecated  This method does not properly convert characters into
 892      * bytes.  As of JDK&nbsp;1.1, the preferred way to do this is via the
 893      * {@link #getBytes()} method, which uses the platform's default charset.
 894      *
 895      * @param  srcBegin
 896      *         Index of the first character in the string to copy
 897      *
 898      * @param  srcEnd
 899      *         Index after the last character in the string to copy
 900      *
 901      * @param  dst
 902      *         The destination array
 903      *
 904      * @param  dstBegin
 905      *         The start offset in the destination array
 906      *
 907      * @throws  IndexOutOfBoundsException
 908      *          If any of the following is true:
 909      *          <ul>
 910      *            <li> {@code srcBegin} is negative
 911      *            <li> {@code srcBegin} is greater than {@code srcEnd}
 912      *            <li> {@code srcEnd} is greater than the length of this String
 913      *            <li> {@code dstBegin} is negative
 914      *            <li> {@code dstBegin+(srcEnd-srcBegin)} is larger than {@code
 915      *                 dst.length}
 916      *          </ul>
 917      */
 918     @Deprecated(since="1.1")
 919     public void getBytes(int srcBegin, int srcEnd, byte dst[], int dstBegin) {
 920         checkBoundsBeginEnd(srcBegin, srcEnd, length());
 921         Objects.requireNonNull(dst);
 922         checkBoundsOffCount(dstBegin, srcEnd - srcBegin, dst.length);
 923         if (isLatin1()) {
 924             StringLatin1.getBytes(value, srcBegin, srcEnd, dst, dstBegin);
 925         } else {
 926             StringUTF16.getBytes(value, srcBegin, srcEnd, dst, dstBegin);
 927         }
 928     }
 929 
 930     /**
 931      * Encodes this {@code String} into a sequence of bytes using the named
 932      * charset, storing the result into a new byte array.
 933      *
 934      * <p> The behavior of this method when this string cannot be encoded in
 935      * the given charset is unspecified.  The {@link
 936      * java.nio.charset.CharsetEncoder} class should be used when more control
 937      * over the encoding process is required.
 938      *
 939      * @param  charsetName
 940      *         The name of a supported {@linkplain java.nio.charset.Charset
 941      *         charset}
 942      *
 943      * @return  The resultant byte array
 944      *
 945      * @throws  UnsupportedEncodingException
 946      *          If the named charset is not supported
 947      *
 948      * @since  1.1
 949      */
 950     public byte[] getBytes(String charsetName)
 951             throws UnsupportedEncodingException {
 952         if (charsetName == null) throw new NullPointerException();
 953         return StringCoding.encode(charsetName, coder(), value);
 954     }
 955 
 956     /**
 957      * Encodes this {@code String} into a sequence of bytes using the given
 958      * {@linkplain java.nio.charset.Charset charset}, storing the result into a
 959      * new byte array.
 960      *
 961      * <p> This method always replaces malformed-input and unmappable-character
 962      * sequences with this charset's default replacement byte array.  The
 963      * {@link java.nio.charset.CharsetEncoder} class should be used when more
 964      * control over the encoding process is required.
 965      *
 966      * @param  charset
 967      *         The {@linkplain java.nio.charset.Charset} to be used to encode
 968      *         the {@code String}
 969      *
 970      * @return  The resultant byte array
 971      *
 972      * @since  1.6
 973      */
 974     public byte[] getBytes(Charset charset) {
 975         if (charset == null) throw new NullPointerException();
 976         return StringCoding.encode(charset, coder(), value);
 977      }
 978 
 979     /**
 980      * Encodes this {@code String} into a sequence of bytes using the
 981      * platform's default charset, storing the result into a new byte array.
 982      *
 983      * <p> The behavior of this method when this string cannot be encoded in
 984      * the default charset is unspecified.  The {@link
 985      * java.nio.charset.CharsetEncoder} class should be used when more control
 986      * over the encoding process is required.
 987      *
 988      * @return  The resultant byte array
 989      *
 990      * @since      1.1
 991      */
 992     public byte[] getBytes() {
 993         return StringCoding.encode(coder(), value);
 994     }
 995 
 996     /**
 997      * Compares this string to the specified object.  The result is {@code
 998      * true} if and only if the argument is not {@code null} and is a {@code
 999      * String} object that represents the same sequence of characters as this
1000      * object.
1001      *
1002      * <p>For finer-grained String comparison, refer to
1003      * {@link java.text.Collator}.
1004      *
1005      * @param  anObject
1006      *         The object to compare this {@code String} against
1007      *
1008      * @return  {@code true} if the given object represents a {@code String}
1009      *          equivalent to this string, {@code false} otherwise
1010      *
1011      * @see  #compareTo(String)
1012      * @see  #equalsIgnoreCase(String)
1013      */
1014     public boolean equals(Object anObject) {
1015         if (this == anObject) {
1016             return true;
1017         }
1018         if (anObject instanceof String) {
1019             String aString = (String)anObject;
1020             if (coder() == aString.coder()) {
1021                 return isLatin1() ? StringLatin1.equals(value, aString.value)
1022                                   : StringUTF16.equals(value, aString.value);
1023             }
1024         }
1025         return false;
1026     }
1027 
1028     /**
1029      * Compares this string to the specified {@code StringBuffer}.  The result
1030      * is {@code true} if and only if this {@code String} represents the same
1031      * sequence of characters as the specified {@code StringBuffer}. This method
1032      * synchronizes on the {@code StringBuffer}.
1033      *
1034      * <p>For finer-grained String comparison, refer to
1035      * {@link java.text.Collator}.
1036      *
1037      * @param  sb
1038      *         The {@code StringBuffer} to compare this {@code String} against
1039      *
1040      * @return  {@code true} if this {@code String} represents the same
1041      *          sequence of characters as the specified {@code StringBuffer},
1042      *          {@code false} otherwise
1043      *
1044      * @since  1.4
1045      */
1046     public boolean contentEquals(StringBuffer sb) {
1047         return contentEquals((CharSequence)sb);
1048     }
1049 
1050     private boolean nonSyncContentEquals(AbstractStringBuilder sb) {
1051         int len = length();
1052         if (len != sb.length()) {
1053             return false;
1054         }
1055         byte v1[] = value;
1056         byte v2[] = sb.getValue();
1057         if (coder() == sb.getCoder()) {
1058             int n = v1.length;
1059             for (int i = 0; i < n; i++) {
1060                 if (v1[i] != v2[i]) {
1061                     return false;
1062                 }
1063             }
1064         } else {
1065             if (!isLatin1()) {  // utf16 str and latin1 abs can never be "equal"
1066                 return false;
1067             }
1068             for (int i = 0; i < len; i++) {
1069                 if ((char)(v1[i] & 0xff) != StringUTF16.getChar(v2, i)) {
1070                     return false;
1071                 }
1072             }
1073         }
1074         return true;
1075     }
1076 
1077     /**
1078      * Compares this string to the specified {@code CharSequence}.  The
1079      * result is {@code true} if and only if this {@code String} represents the
1080      * same sequence of char values as the specified sequence. Note that if the
1081      * {@code CharSequence} is a {@code StringBuffer} then the method
1082      * synchronizes on it.
1083      *
1084      * <p>For finer-grained String comparison, refer to
1085      * {@link java.text.Collator}.
1086      *
1087      * @param  cs
1088      *         The sequence to compare this {@code String} against
1089      *
1090      * @return  {@code true} if this {@code String} represents the same
1091      *          sequence of char values as the specified sequence, {@code
1092      *          false} otherwise
1093      *
1094      * @since  1.5
1095      */
1096     public boolean contentEquals(CharSequence cs) {
1097         // Argument is a StringBuffer, StringBuilder
1098         if (cs instanceof AbstractStringBuilder) {
1099             if (cs instanceof StringBuffer) {
1100                 synchronized(cs) {
1101                    return nonSyncContentEquals((AbstractStringBuilder)cs);
1102                 }
1103             } else {
1104                 return nonSyncContentEquals((AbstractStringBuilder)cs);
1105             }
1106         }
1107         // Argument is a String
1108         if (cs instanceof String) {
1109             return equals(cs);
1110         }
1111         // Argument is a generic CharSequence
1112         int n = cs.length();
1113         if (n != length()) {
1114             return false;
1115         }
1116         byte[] val = this.value;
1117         if (isLatin1()) {
1118             for (int i = 0; i < n; i++) {
1119                 if ((val[i] & 0xff) != cs.charAt(i)) {
1120                     return false;
1121                 }
1122             }
1123         } else {
1124             for (int i = 0; i < n; i++) {
1125                 if (StringUTF16.getChar(val, i) != cs.charAt(i)) {
1126                     return false;
1127                 }
1128             }
1129         }
1130         return true;
1131     }
1132 
1133     /**
1134      * Compares this {@code String} to another {@code String}, ignoring case
1135      * considerations.  Two strings are considered equal ignoring case if they
1136      * are of the same length and corresponding characters in the two strings
1137      * are equal ignoring case.
1138      *
1139      * <p> Two characters {@code c1} and {@code c2} are considered the same
1140      * ignoring case if at least one of the following is true:
1141      * <ul>
1142      *   <li> The two characters are the same (as compared by the
1143      *        {@code ==} operator)
1144      *   <li> Calling {@code Character.toLowerCase(Character.toUpperCase(char))}
1145      *        on each character produces the same result
1146      * </ul>
1147      *
1148      * <p>Note that this method does <em>not</em> take locale into account, and
1149      * will result in unsatisfactory results for certain locales.  The
1150      * {@link java.text.Collator} class provides locale-sensitive comparison.
1151      *
1152      * @param  anotherString
1153      *         The {@code String} to compare this {@code String} against
1154      *
1155      * @return  {@code true} if the argument is not {@code null} and it
1156      *          represents an equivalent {@code String} ignoring case; {@code
1157      *          false} otherwise
1158      *
1159      * @see  #equals(Object)
1160      */
1161     public boolean equalsIgnoreCase(String anotherString) {
1162         return (this == anotherString) ? true
1163                 : (anotherString != null)
1164                 && (anotherString.length() == length())
1165                 && regionMatches(true, 0, anotherString, 0, length());
1166     }
1167 
1168     /**
1169      * Compares two strings lexicographically.
1170      * The comparison is based on the Unicode value of each character in
1171      * the strings. The character sequence represented by this
1172      * {@code String} object is compared lexicographically to the
1173      * character sequence represented by the argument string. The result is
1174      * a negative integer if this {@code String} object
1175      * lexicographically precedes the argument string. The result is a
1176      * positive integer if this {@code String} object lexicographically
1177      * follows the argument string. The result is zero if the strings
1178      * are equal; {@code compareTo} returns {@code 0} exactly when
1179      * the {@link #equals(Object)} method would return {@code true}.
1180      * <p>
1181      * This is the definition of lexicographic ordering. If two strings are
1182      * different, then either they have different characters at some index
1183      * that is a valid index for both strings, or their lengths are different,
1184      * or both. If they have different characters at one or more index
1185      * positions, let <i>k</i> be the smallest such index; then the string
1186      * whose character at position <i>k</i> has the smaller value, as
1187      * determined by using the {@code <} operator, lexicographically precedes the
1188      * other string. In this case, {@code compareTo} returns the
1189      * difference of the two character values at position {@code k} in
1190      * the two string -- that is, the value:
1191      * <blockquote><pre>
1192      * this.charAt(k)-anotherString.charAt(k)
1193      * </pre></blockquote>
1194      * If there is no index position at which they differ, then the shorter
1195      * string lexicographically precedes the longer string. In this case,
1196      * {@code compareTo} returns the difference of the lengths of the
1197      * strings -- that is, the value:
1198      * <blockquote><pre>
1199      * this.length()-anotherString.length()
1200      * </pre></blockquote>
1201      *
1202      * <p>For finer-grained String comparison, refer to
1203      * {@link java.text.Collator}.
1204      *
1205      * @param   anotherString   the {@code String} to be compared.
1206      * @return  the value {@code 0} if the argument string is equal to
1207      *          this string; a value less than {@code 0} if this string
1208      *          is lexicographically less than the string argument; and a
1209      *          value greater than {@code 0} if this string is
1210      *          lexicographically greater than the string argument.
1211      */
1212     public int compareTo(String anotherString) {
1213         byte v1[] = value;
1214         byte v2[] = anotherString.value;
1215         if (coder() == anotherString.coder()) {
1216             return isLatin1() ? StringLatin1.compareTo(v1, v2)
1217                               : StringUTF16.compareTo(v1, v2);
1218         }
1219         return isLatin1() ? StringLatin1.compareToUTF16(v1, v2)
1220                           : StringUTF16.compareToLatin1(v1, v2);
1221      }
1222 
1223     /**
1224      * A Comparator that orders {@code String} objects as by
1225      * {@code compareToIgnoreCase}. This comparator is serializable.
1226      * <p>
1227      * Note that this Comparator does <em>not</em> take locale into account,
1228      * and will result in an unsatisfactory ordering for certain locales.
1229      * The {@link java.text.Collator} class provides locale-sensitive comparison.
1230      *
1231      * @see     java.text.Collator
1232      * @since   1.2
1233      */
1234     public static final Comparator<String> CASE_INSENSITIVE_ORDER
1235                                          = new CaseInsensitiveComparator();
1236     private static class CaseInsensitiveComparator
1237             implements Comparator<String>, java.io.Serializable {
1238         // use serialVersionUID from JDK 1.2.2 for interoperability
1239         private static final long serialVersionUID = 8575799808933029326L;
1240 
1241         public int compare(String s1, String s2) {
1242             byte v1[] = s1.value;
1243             byte v2[] = s2.value;
1244             if (s1.coder() == s2.coder()) {
1245                 return s1.isLatin1() ? StringLatin1.compareToCI(v1, v2)
1246                                      : StringUTF16.compareToCI(v1, v2);
1247             }
1248             return s1.isLatin1() ? StringLatin1.compareToCI_UTF16(v1, v2)
1249                                  : StringUTF16.compareToCI_Latin1(v1, v2);
1250         }
1251 
1252         /** Replaces the de-serialized object. */
1253         private Object readResolve() { return CASE_INSENSITIVE_ORDER; }
1254     }
1255 
1256     /**
1257      * Compares two strings lexicographically, ignoring case
1258      * differences. This method returns an integer whose sign is that of
1259      * calling {@code compareTo} with normalized versions of the strings
1260      * where case differences have been eliminated by calling
1261      * {@code Character.toLowerCase(Character.toUpperCase(character))} on
1262      * each character.
1263      * <p>
1264      * Note that this method does <em>not</em> take locale into account,
1265      * and will result in an unsatisfactory ordering for certain locales.
1266      * The {@link java.text.Collator} class provides locale-sensitive comparison.
1267      *
1268      * @param   str   the {@code String} to be compared.
1269      * @return  a negative integer, zero, or a positive integer as the
1270      *          specified String is greater than, equal to, or less
1271      *          than this String, ignoring case considerations.
1272      * @see     java.text.Collator
1273      * @since   1.2
1274      */
1275     public int compareToIgnoreCase(String str) {
1276         return CASE_INSENSITIVE_ORDER.compare(this, str);
1277     }
1278 
1279     /**
1280      * Tests if two string regions are equal.
1281      * <p>
1282      * A substring of this {@code String} object is compared to a substring
1283      * of the argument other. The result is true if these substrings
1284      * represent identical character sequences. The substring of this
1285      * {@code String} object to be compared begins at index {@code toffset}
1286      * and has length {@code len}. The substring of other to be compared
1287      * begins at index {@code ooffset} and has length {@code len}. The
1288      * result is {@code false} if and only if at least one of the following
1289      * is true:
1290      * <ul><li>{@code toffset} is negative.
1291      * <li>{@code ooffset} is negative.
1292      * <li>{@code toffset+len} is greater than the length of this
1293      * {@code String} object.
1294      * <li>{@code ooffset+len} is greater than the length of the other
1295      * argument.
1296      * <li>There is some nonnegative integer <i>k</i> less than {@code len}
1297      * such that:
1298      * {@code this.charAt(toffset + }<i>k</i>{@code ) != other.charAt(ooffset + }
1299      * <i>k</i>{@code )}
1300      * </ul>
1301      *
1302      * <p>Note that this method does <em>not</em> take locale into account.  The
1303      * {@link java.text.Collator} class provides locale-sensitive comparison.
1304      *
1305      * @param   toffset   the starting offset of the subregion in this string.
1306      * @param   other     the string argument.
1307      * @param   ooffset   the starting offset of the subregion in the string
1308      *                    argument.
1309      * @param   len       the number of characters to compare.
1310      * @return  {@code true} if the specified subregion of this string
1311      *          exactly matches the specified subregion of the string argument;
1312      *          {@code false} otherwise.
1313      */
1314     public boolean regionMatches(int toffset, String other, int ooffset, int len) {
1315         byte tv[] = value;
1316         byte ov[] = other.value;
1317         // Note: toffset, ooffset, or len might be near -1>>>1.
1318         if ((ooffset < 0) || (toffset < 0) ||
1319              (toffset > (long)length() - len) ||
1320              (ooffset > (long)other.length() - len)) {
1321             return false;
1322         }
1323         if (coder() == other.coder()) {
1324             if (!isLatin1() && (len > 0)) {
1325                 toffset = toffset << 1;
1326                 ooffset = ooffset << 1;
1327                 len = len << 1;
1328             }
1329             while (len-- > 0) {
1330                 if (tv[toffset++] != ov[ooffset++]) {
1331                     return false;
1332                 }
1333             }
1334         } else {
1335             if (coder() == LATIN1) {
1336                 while (len-- > 0) {
1337                     if (StringLatin1.getChar(tv, toffset++) !=
1338                         StringUTF16.getChar(ov, ooffset++)) {
1339                         return false;
1340                     }
1341                 }
1342             } else {
1343                 while (len-- > 0) {
1344                     if (StringUTF16.getChar(tv, toffset++) !=
1345                         StringLatin1.getChar(ov, ooffset++)) {
1346                         return false;
1347                     }
1348                 }
1349             }
1350         }
1351         return true;
1352     }
1353 
1354     /**
1355      * Tests if two string regions are equal.
1356      * <p>
1357      * A substring of this {@code String} object is compared to a substring
1358      * of the argument {@code other}. The result is {@code true} if these
1359      * substrings represent character sequences that are the same, ignoring
1360      * case if and only if {@code ignoreCase} is true. The substring of
1361      * this {@code String} object to be compared begins at index
1362      * {@code toffset} and has length {@code len}. The substring of
1363      * {@code other} to be compared begins at index {@code ooffset} and
1364      * has length {@code len}. The result is {@code false} if and only if
1365      * at least one of the following is true:
1366      * <ul><li>{@code toffset} is negative.
1367      * <li>{@code ooffset} is negative.
1368      * <li>{@code toffset+len} is greater than the length of this
1369      * {@code String} object.
1370      * <li>{@code ooffset+len} is greater than the length of the other
1371      * argument.
1372      * <li>{@code ignoreCase} is {@code false} and there is some nonnegative
1373      * integer <i>k</i> less than {@code len} such that:
1374      * <blockquote><pre>
1375      * this.charAt(toffset+k) != other.charAt(ooffset+k)
1376      * </pre></blockquote>
1377      * <li>{@code ignoreCase} is {@code true} and there is some nonnegative
1378      * integer <i>k</i> less than {@code len} such that:
1379      * <blockquote><pre>
1380      * Character.toLowerCase(Character.toUpperCase(this.charAt(toffset+k))) !=
1381      Character.toLowerCase(Character.toUpperCase(other.charAt(ooffset+k)))
1382      * </pre></blockquote>
1383      * </ul>
1384      *
1385      * <p>Note that this method does <em>not</em> take locale into account,
1386      * and will result in unsatisfactory results for certain locales when
1387      * {@code ignoreCase} is {@code true}.  The {@link java.text.Collator} class
1388      * provides locale-sensitive comparison.
1389      *
1390      * @param   ignoreCase   if {@code true}, ignore case when comparing
1391      *                       characters.
1392      * @param   toffset      the starting offset of the subregion in this
1393      *                       string.
1394      * @param   other        the string argument.
1395      * @param   ooffset      the starting offset of the subregion in the string
1396      *                       argument.
1397      * @param   len          the number of characters to compare.
1398      * @return  {@code true} if the specified subregion of this string
1399      *          matches the specified subregion of the string argument;
1400      *          {@code false} otherwise. Whether the matching is exact
1401      *          or case insensitive depends on the {@code ignoreCase}
1402      *          argument.
1403      */
1404     public boolean regionMatches(boolean ignoreCase, int toffset,
1405             String other, int ooffset, int len) {
1406         if (!ignoreCase) {
1407             return regionMatches(toffset, other, ooffset, len);
1408         }
1409         // Note: toffset, ooffset, or len might be near -1>>>1.
1410         if ((ooffset < 0) || (toffset < 0)
1411                 || (toffset > (long)length() - len)
1412                 || (ooffset > (long)other.length() - len)) {
1413             return false;
1414         }
1415         byte tv[] = value;
1416         byte ov[] = other.value;
1417         if (coder() == other.coder()) {
1418             return isLatin1()
1419               ? StringLatin1.regionMatchesCI(tv, toffset, ov, ooffset, len)
1420               : StringUTF16.regionMatchesCI(tv, toffset, ov, ooffset, len);
1421         }
1422         return isLatin1()
1423               ? StringLatin1.regionMatchesCI_UTF16(tv, toffset, ov, ooffset, len)
1424               : StringUTF16.regionMatchesCI_Latin1(tv, toffset, ov, ooffset, len);
1425     }
1426 
1427     /**
1428      * Tests if the substring of this string beginning at the
1429      * specified index starts with the specified prefix.
1430      *
1431      * @param   prefix    the prefix.
1432      * @param   toffset   where to begin looking in this string.
1433      * @return  {@code true} if the character sequence represented by the
1434      *          argument is a prefix of the substring of this object starting
1435      *          at index {@code toffset}; {@code false} otherwise.
1436      *          The result is {@code false} if {@code toffset} is
1437      *          negative or greater than the length of this
1438      *          {@code String} object; otherwise the result is the same
1439      *          as the result of the expression
1440      *          <pre>
1441      *          this.substring(toffset).startsWith(prefix)
1442      *          </pre>
1443      */
1444     public boolean startsWith(String prefix, int toffset) {
1445         // Note: toffset might be near -1>>>1.
1446         if (toffset < 0 || toffset > length() - prefix.length()) {
1447             return false;
1448         }
1449         byte ta[] = value;
1450         byte pa[] = prefix.value;
1451         int po = 0;
1452         int pc = pa.length;
1453         if (coder() == prefix.coder()) {
1454             int to = isLatin1() ? toffset : toffset << 1;
1455             while (po < pc) {
1456                 if (ta[to++] != pa[po++]) {
1457                     return false;
1458                 }
1459             }
1460         } else {
1461             if (isLatin1()) {  // && pcoder == UTF16
1462                 return false;
1463             }
1464             // coder == UTF16 && pcoder == LATIN1)
1465             while (po < pc) {
1466                 if (StringUTF16.getChar(ta, toffset++) != (pa[po++] & 0xff)) {
1467                     return false;
1468                }
1469             }
1470         }
1471         return true;
1472     }
1473 
1474     /**
1475      * Tests if this string starts with the specified prefix.
1476      *
1477      * @param   prefix   the prefix.
1478      * @return  {@code true} if the character sequence represented by the
1479      *          argument is a prefix of the character sequence represented by
1480      *          this string; {@code false} otherwise.
1481      *          Note also that {@code true} will be returned if the
1482      *          argument is an empty string or is equal to this
1483      *          {@code String} object as determined by the
1484      *          {@link #equals(Object)} method.
1485      * @since   1.0
1486      */
1487     public boolean startsWith(String prefix) {
1488         return startsWith(prefix, 0);
1489     }
1490 
1491     /**
1492      * Tests if this string ends with the specified suffix.
1493      *
1494      * @param   suffix   the suffix.
1495      * @return  {@code true} if the character sequence represented by the
1496      *          argument is a suffix of the character sequence represented by
1497      *          this object; {@code false} otherwise. Note that the
1498      *          result will be {@code true} if the argument is the
1499      *          empty string or is equal to this {@code String} object
1500      *          as determined by the {@link #equals(Object)} method.
1501      */
1502     public boolean endsWith(String suffix) {
1503         return startsWith(suffix, length() - suffix.length());
1504     }
1505 
1506     /**
1507      * Returns a hash code for this string. The hash code for a
1508      * {@code String} object is computed as
1509      * <blockquote><pre>
1510      * s[0]*31^(n-1) + s[1]*31^(n-2) + ... + s[n-1]
1511      * </pre></blockquote>
1512      * using {@code int} arithmetic, where {@code s[i]} is the
1513      * <i>i</i>th character of the string, {@code n} is the length of
1514      * the string, and {@code ^} indicates exponentiation.
1515      * (The hash value of the empty string is zero.)
1516      *
1517      * @return  a hash code value for this object.
1518      */
1519     public int hashCode() {
1520         int h = hash;
1521         if (h == 0 && value.length > 0) {
1522             hash = h = isLatin1() ? StringLatin1.hashCode(value)
1523                                   : StringUTF16.hashCode(value);
1524         }
1525         return h;
1526     }
1527 
1528     /**
1529      * Returns the index within this string of the first occurrence of
1530      * the specified character. If a character with value
1531      * {@code ch} occurs in the character sequence represented by
1532      * this {@code String} object, then the index (in Unicode
1533      * code units) of the first such occurrence is returned. For
1534      * values of {@code ch} in the range from 0 to 0xFFFF
1535      * (inclusive), this is the smallest value <i>k</i> such that:
1536      * <blockquote><pre>
1537      * this.charAt(<i>k</i>) == ch
1538      * </pre></blockquote>
1539      * is true. For other values of {@code ch}, it is the
1540      * smallest value <i>k</i> such that:
1541      * <blockquote><pre>
1542      * this.codePointAt(<i>k</i>) == ch
1543      * </pre></blockquote>
1544      * is true. In either case, if no such character occurs in this
1545      * string, then {@code -1} is returned.
1546      *
1547      * @param   ch   a character (Unicode code point).
1548      * @return  the index of the first occurrence of the character in the
1549      *          character sequence represented by this object, or
1550      *          {@code -1} if the character does not occur.
1551      */
1552     public int indexOf(int ch) {
1553         return indexOf(ch, 0);
1554     }
1555 
1556     /**
1557      * Returns the index within this string of the first occurrence of the
1558      * specified character, starting the search at the specified index.
1559      * <p>
1560      * If a character with value {@code ch} occurs in the
1561      * character sequence represented by this {@code String}
1562      * object at an index no smaller than {@code fromIndex}, then
1563      * the index of the first such occurrence is returned. For values
1564      * of {@code ch} in the range from 0 to 0xFFFF (inclusive),
1565      * this is the smallest value <i>k</i> such that:
1566      * <blockquote><pre>
1567      * (this.charAt(<i>k</i>) == ch) {@code &&} (<i>k</i> &gt;= fromIndex)
1568      * </pre></blockquote>
1569      * is true. For other values of {@code ch}, it is the
1570      * smallest value <i>k</i> such that:
1571      * <blockquote><pre>
1572      * (this.codePointAt(<i>k</i>) == ch) {@code &&} (<i>k</i> &gt;= fromIndex)
1573      * </pre></blockquote>
1574      * is true. In either case, if no such character occurs in this
1575      * string at or after position {@code fromIndex}, then
1576      * {@code -1} is returned.
1577      *
1578      * <p>
1579      * There is no restriction on the value of {@code fromIndex}. If it
1580      * is negative, it has the same effect as if it were zero: this entire
1581      * string may be searched. If it is greater than the length of this
1582      * string, it has the same effect as if it were equal to the length of
1583      * this string: {@code -1} is returned.
1584      *
1585      * <p>All indices are specified in {@code char} values
1586      * (Unicode code units).
1587      *
1588      * @param   ch          a character (Unicode code point).
1589      * @param   fromIndex   the index to start the search from.
1590      * @return  the index of the first occurrence of the character in the
1591      *          character sequence represented by this object that is greater
1592      *          than or equal to {@code fromIndex}, or {@code -1}
1593      *          if the character does not occur.
1594      */
1595     public int indexOf(int ch, int fromIndex) {
1596         return isLatin1() ? StringLatin1.indexOf(value, ch, fromIndex)
1597                           : StringUTF16.indexOf(value, ch, fromIndex);
1598     }
1599 
1600     /**
1601      * Returns the index within this string of the last occurrence of
1602      * the specified character. For values of {@code ch} in the
1603      * range from 0 to 0xFFFF (inclusive), the index (in Unicode code
1604      * units) returned is the largest value <i>k</i> such that:
1605      * <blockquote><pre>
1606      * this.charAt(<i>k</i>) == ch
1607      * </pre></blockquote>
1608      * is true. For other values of {@code ch}, it is the
1609      * largest value <i>k</i> such that:
1610      * <blockquote><pre>
1611      * this.codePointAt(<i>k</i>) == ch
1612      * </pre></blockquote>
1613      * is true.  In either case, if no such character occurs in this
1614      * string, then {@code -1} is returned.  The
1615      * {@code String} is searched backwards starting at the last
1616      * character.
1617      *
1618      * @param   ch   a character (Unicode code point).
1619      * @return  the index of the last occurrence of the character in the
1620      *          character sequence represented by this object, or
1621      *          {@code -1} if the character does not occur.
1622      */
1623     public int lastIndexOf(int ch) {
1624         return lastIndexOf(ch, length() - 1);
1625     }
1626 
1627     /**
1628      * Returns the index within this string of the last occurrence of
1629      * the specified character, searching backward starting at the
1630      * specified index. For values of {@code ch} in the range
1631      * from 0 to 0xFFFF (inclusive), the index returned is the largest
1632      * value <i>k</i> such that:
1633      * <blockquote><pre>
1634      * (this.charAt(<i>k</i>) == ch) {@code &&} (<i>k</i> &lt;= fromIndex)
1635      * </pre></blockquote>
1636      * is true. For other values of {@code ch}, it is the
1637      * largest value <i>k</i> such that:
1638      * <blockquote><pre>
1639      * (this.codePointAt(<i>k</i>) == ch) {@code &&} (<i>k</i> &lt;= fromIndex)
1640      * </pre></blockquote>
1641      * is true. In either case, if no such character occurs in this
1642      * string at or before position {@code fromIndex}, then
1643      * {@code -1} is returned.
1644      *
1645      * <p>All indices are specified in {@code char} values
1646      * (Unicode code units).
1647      *
1648      * @param   ch          a character (Unicode code point).
1649      * @param   fromIndex   the index to start the search from. There is no
1650      *          restriction on the value of {@code fromIndex}. If it is
1651      *          greater than or equal to the length of this string, it has
1652      *          the same effect as if it were equal to one less than the
1653      *          length of this string: this entire string may be searched.
1654      *          If it is negative, it has the same effect as if it were -1:
1655      *          -1 is returned.
1656      * @return  the index of the last occurrence of the character in the
1657      *          character sequence represented by this object that is less
1658      *          than or equal to {@code fromIndex}, or {@code -1}
1659      *          if the character does not occur before that point.
1660      */
1661     public int lastIndexOf(int ch, int fromIndex) {
1662         return isLatin1() ? StringLatin1.lastIndexOf(value, ch, fromIndex)
1663                           : StringUTF16.lastIndexOf(value, ch, fromIndex);
1664     }
1665 
1666     /**
1667      * Returns the index within this string of the first occurrence of the
1668      * specified substring.
1669      *
1670      * <p>The returned index is the smallest value {@code k} for which:
1671      * <pre>{@code
1672      * this.startsWith(str, k)
1673      * }</pre>
1674      * If no such value of {@code k} exists, then {@code -1} is returned.
1675      *
1676      * @param   str   the substring to search for.
1677      * @return  the index of the first occurrence of the specified substring,
1678      *          or {@code -1} if there is no such occurrence.
1679      */
1680     public int indexOf(String str) {
1681         if (coder() == str.coder()) {
1682             return isLatin1() ? StringLatin1.indexOf(value, str.value)
1683                               : StringUTF16.indexOf(value, str.value);
1684         }
1685         if (coder() == LATIN1) {  // str.coder == UTF16
1686             return -1;
1687         }
1688         return StringUTF16.indexOfLatin1(value, str.value);
1689     }
1690 
1691     /**
1692      * Returns the index within this string of the first occurrence of the
1693      * specified substring, starting at the specified index.
1694      *
1695      * <p>The returned index is the smallest value {@code k} for which:
1696      * <pre>{@code
1697      *     k >= Math.min(fromIndex, this.length()) &&
1698      *                   this.startsWith(str, k)
1699      * }</pre>
1700      * If no such value of {@code k} exists, then {@code -1} is returned.
1701      *
1702      * @param   str         the substring to search for.
1703      * @param   fromIndex   the index from which to start the search.
1704      * @return  the index of the first occurrence of the specified substring,
1705      *          starting at the specified index,
1706      *          or {@code -1} if there is no such occurrence.
1707      */
1708     public int indexOf(String str, int fromIndex) {
1709         return indexOf(value, coder(), length(), str, fromIndex);
1710     }
1711 
1712     /**
1713      * Code shared by String and AbstractStringBuilder to do searches. The
1714      * source is the character array being searched, and the target
1715      * is the string being searched for.
1716      *
1717      * @param   src       the characters being searched.
1718      * @param   srcCoder  the coder of the source string.
1719      * @param   srcCount  length of the source string.
1720      * @param   tgtStr    the characters being searched for.
1721      * @param   fromIndex the index to begin searching from.
1722      */
1723     static int indexOf(byte[] src, byte srcCoder, int srcCount,
1724                        String tgtStr, int fromIndex) {
1725         byte[] tgt    = tgtStr.value;
1726         byte tgtCoder = tgtStr.coder();
1727         int tgtCount  = tgtStr.length();
1728 
1729         if (fromIndex >= srcCount) {
1730             return (tgtCount == 0 ? srcCount : -1);
1731         }
1732         if (fromIndex < 0) {
1733             fromIndex = 0;
1734         }
1735         if (tgtCount == 0) {
1736             return fromIndex;
1737         }
1738         if (srcCoder == tgtCoder) {
1739             return srcCoder == LATIN1
1740                 ? StringLatin1.indexOf(src, srcCount, tgt, tgtCount, fromIndex)
1741                 : StringUTF16.indexOf(src, srcCount, tgt, tgtCount, fromIndex);
1742         }
1743         if (srcCoder == LATIN1) {    //  && tgtCoder == UTF16
1744             return -1;
1745         }
1746         // srcCoder == UTF16 && tgtCoder == LATIN1) {
1747         return StringUTF16.indexOfLatin1(src, srcCount, tgt, tgtCount, fromIndex);
1748     }
1749 
1750     /**
1751      * Returns the index within this string of the last occurrence of the
1752      * specified substring.  The last occurrence of the empty string ""
1753      * is considered to occur at the index value {@code this.length()}.
1754      *
1755      * <p>The returned index is the largest value {@code k} for which:
1756      * <pre>{@code
1757      * this.startsWith(str, k)
1758      * }</pre>
1759      * If no such value of {@code k} exists, then {@code -1} is returned.
1760      *
1761      * @param   str   the substring to search for.
1762      * @return  the index of the last occurrence of the specified substring,
1763      *          or {@code -1} if there is no such occurrence.
1764      */
1765     public int lastIndexOf(String str) {
1766         return lastIndexOf(str, length());
1767     }
1768 
1769     /**
1770      * Returns the index within this string of the last occurrence of the
1771      * specified substring, searching backward starting at the specified index.
1772      *
1773      * <p>The returned index is the largest value {@code k} for which:
1774      * <pre>{@code
1775      *     k <= Math.min(fromIndex, this.length()) &&
1776      *                   this.startsWith(str, k)
1777      * }</pre>
1778      * If no such value of {@code k} exists, then {@code -1} is returned.
1779      *
1780      * @param   str         the substring to search for.
1781      * @param   fromIndex   the index to start the search from.
1782      * @return  the index of the last occurrence of the specified substring,
1783      *          searching backward from the specified index,
1784      *          or {@code -1} if there is no such occurrence.
1785      */
1786     public int lastIndexOf(String str, int fromIndex) {
1787         return lastIndexOf(value, coder(), length(), str, fromIndex);
1788     }
1789 
1790     /**
1791      * Code shared by String and AbstractStringBuilder to do searches. The
1792      * source is the character array being searched, and the target
1793      * is the string being searched for.
1794      *
1795      * @param   src         the characters being searched.
1796      * @param   srcCoder  coder handles the mapping between bytes/chars
1797      * @param   srcCount    count of the source string.
1798      * @param   tgt         the characters being searched for.
1799      * @param   fromIndex   the index to begin searching from.
1800      */
1801     static int lastIndexOf(byte[] src, byte srcCoder, int srcCount,
1802                            String tgtStr, int fromIndex) {
1803         byte[] tgt = tgtStr.value;
1804         byte tgtCoder = tgtStr.coder();
1805         int tgtCount = tgtStr.length();
1806         /*
1807          * Check arguments; return immediately where possible. For
1808          * consistency, don't check for null str.
1809          */
1810         int rightIndex = srcCount - tgtCount;
1811         if (fromIndex < 0) {
1812             return -1;
1813         }
1814         if (fromIndex > rightIndex) {
1815             fromIndex = rightIndex;
1816         }
1817         /* Empty string always matches. */
1818         if (tgtCount == 0) {
1819             return fromIndex;
1820         }
1821         if (srcCoder == tgtCoder) {
1822             return srcCoder == LATIN1
1823                 ? StringLatin1.lastIndexOf(src, srcCount, tgt, tgtCount, fromIndex)
1824                 : StringUTF16.lastIndexOf(src, srcCount, tgt, tgtCount, fromIndex);
1825         }
1826         if (srcCoder == LATIN1) {    // && tgtCoder == UTF16
1827             return -1;
1828         }
1829                                      // srcCoder == UTF16 && tgtCoder == LATIN1
1830         int min = tgtCount - 1;
1831         int i = min + fromIndex;
1832         int strLastIndex = tgtCount - 1;
1833 
1834         char strLastChar = (char)(tgt[strLastIndex] & 0xff);
1835     startSearchForLastChar:
1836         while (true) {
1837             while (i >= min && StringUTF16.getChar(src, i) != strLastChar) {
1838                 i--;
1839             }
1840             if (i < min) {
1841                 return -1;
1842             }
1843             int j = i - 1;
1844             int start = j - strLastIndex;
1845             int k = strLastIndex - 1;
1846             while (j > start) {
1847                 if (StringUTF16.getChar(src, j--) != (tgt[k--] & 0xff)) {
1848                     i--;
1849                     continue startSearchForLastChar;
1850                 }
1851             }
1852             return start + 1;
1853         }
1854     }
1855 
1856     /**
1857      * Returns a string that is a substring of this string. The
1858      * substring begins with the character at the specified index and
1859      * extends to the end of this string. <p>
1860      * Examples:
1861      * <blockquote><pre>
1862      * "unhappy".substring(2) returns "happy"
1863      * "Harbison".substring(3) returns "bison"
1864      * "emptiness".substring(9) returns "" (an empty string)
1865      * </pre></blockquote>
1866      *
1867      * @param      beginIndex   the beginning index, inclusive.
1868      * @return     the specified substring.
1869      * @exception  IndexOutOfBoundsException  if
1870      *             {@code beginIndex} is negative or larger than the
1871      *             length of this {@code String} object.
1872      */
1873     public String substring(int beginIndex) {
1874         if (beginIndex < 0) {
1875             throw new StringIndexOutOfBoundsException(beginIndex);
1876         }
1877         int subLen = length() - beginIndex;
1878         if (subLen < 0) {
1879             throw new StringIndexOutOfBoundsException(subLen);
1880         }
1881         if (beginIndex == 0) {
1882             return this;
1883         }
1884         return isLatin1() ? StringLatin1.newString(value, beginIndex, subLen)
1885                           : StringUTF16.newString(value, beginIndex, subLen);
1886     }
1887 
1888     /**
1889      * Returns a string that is a substring of this string. The
1890      * substring begins at the specified {@code beginIndex} and
1891      * extends to the character at index {@code endIndex - 1}.
1892      * Thus the length of the substring is {@code endIndex-beginIndex}.
1893      * <p>
1894      * Examples:
1895      * <blockquote><pre>
1896      * "hamburger".substring(4, 8) returns "urge"
1897      * "smiles".substring(1, 5) returns "mile"
1898      * </pre></blockquote>
1899      *
1900      * @param      beginIndex   the beginning index, inclusive.
1901      * @param      endIndex     the ending index, exclusive.
1902      * @return     the specified substring.
1903      * @exception  IndexOutOfBoundsException  if the
1904      *             {@code beginIndex} is negative, or
1905      *             {@code endIndex} is larger than the length of
1906      *             this {@code String} object, or
1907      *             {@code beginIndex} is larger than
1908      *             {@code endIndex}.
1909      */
1910     public String substring(int beginIndex, int endIndex) {
1911         int length = length();
1912         checkBoundsBeginEnd(beginIndex, endIndex, length);
1913         int subLen = endIndex - beginIndex;
1914         if (beginIndex == 0 && endIndex == length) {
1915             return this;
1916         }
1917         return isLatin1() ? StringLatin1.newString(value, beginIndex, subLen)
1918                           : StringUTF16.newString(value, beginIndex, subLen);
1919     }
1920 
1921     /**
1922      * Returns a character sequence that is a subsequence of this sequence.
1923      *
1924      * <p> An invocation of this method of the form
1925      *
1926      * <blockquote><pre>
1927      * str.subSequence(begin,&nbsp;end)</pre></blockquote>
1928      *
1929      * behaves in exactly the same way as the invocation
1930      *
1931      * <blockquote><pre>
1932      * str.substring(begin,&nbsp;end)</pre></blockquote>
1933      *
1934      * @apiNote
1935      * This method is defined so that the {@code String} class can implement
1936      * the {@link CharSequence} interface.
1937      *
1938      * @param   beginIndex   the begin index, inclusive.
1939      * @param   endIndex     the end index, exclusive.
1940      * @return  the specified subsequence.
1941      *
1942      * @throws  IndexOutOfBoundsException
1943      *          if {@code beginIndex} or {@code endIndex} is negative,
1944      *          if {@code endIndex} is greater than {@code length()},
1945      *          or if {@code beginIndex} is greater than {@code endIndex}
1946      *
1947      * @since 1.4
1948      * @spec JSR-51
1949      */
1950     public CharSequence subSequence(int beginIndex, int endIndex) {
1951         return this.substring(beginIndex, endIndex);
1952     }
1953 
1954     /**
1955      * Concatenates the specified string to the end of this string.
1956      * <p>
1957      * If the length of the argument string is {@code 0}, then this
1958      * {@code String} object is returned. Otherwise, a
1959      * {@code String} object is returned that represents a character
1960      * sequence that is the concatenation of the character sequence
1961      * represented by this {@code String} object and the character
1962      * sequence represented by the argument string.<p>
1963      * Examples:
1964      * <blockquote><pre>
1965      * "cares".concat("s") returns "caress"
1966      * "to".concat("get").concat("her") returns "together"
1967      * </pre></blockquote>
1968      *
1969      * @param   str   the {@code String} that is concatenated to the end
1970      *                of this {@code String}.
1971      * @return  a string that represents the concatenation of this object's
1972      *          characters followed by the string argument's characters.
1973      */
1974     public String concat(String str) {
1975         int olen = str.length();
1976         if (olen == 0) {
1977             return this;
1978         }
1979         if (coder() == str.coder()) {
1980             byte[] val = this.value;
1981             byte[] oval = str.value;
1982             int len = val.length + oval.length;
1983             byte[] buf = Arrays.copyOf(val, len);
1984             System.arraycopy(oval, 0, buf, val.length, oval.length);
1985             return new String(buf, coder);
1986         }
1987         int len = length();
1988         byte[] buf = StringUTF16.newBytesFor(len + olen);
1989         getBytes(buf, 0, UTF16);
1990         str.getBytes(buf, len, UTF16);
1991         return new String(buf, UTF16);
1992     }
1993 
1994     /**
1995      * Returns a string resulting from replacing all occurrences of
1996      * {@code oldChar} in this string with {@code newChar}.
1997      * <p>
1998      * If the character {@code oldChar} does not occur in the
1999      * character sequence represented by this {@code String} object,
2000      * then a reference to this {@code String} object is returned.
2001      * Otherwise, a {@code String} object is returned that
2002      * represents a character sequence identical to the character sequence
2003      * represented by this {@code String} object, except that every
2004      * occurrence of {@code oldChar} is replaced by an occurrence
2005      * of {@code newChar}.
2006      * <p>
2007      * Examples:
2008      * <blockquote><pre>
2009      * "mesquite in your cellar".replace('e', 'o')
2010      *         returns "mosquito in your collar"
2011      * "the war of baronets".replace('r', 'y')
2012      *         returns "the way of bayonets"
2013      * "sparring with a purple porpoise".replace('p', 't')
2014      *         returns "starring with a turtle tortoise"
2015      * "JonL".replace('q', 'x') returns "JonL" (no change)
2016      * </pre></blockquote>
2017      *
2018      * @param   oldChar   the old character.
2019      * @param   newChar   the new character.
2020      * @return  a string derived from this string by replacing every
2021      *          occurrence of {@code oldChar} with {@code newChar}.
2022      */
2023     public String replace(char oldChar, char newChar) {
2024         if (oldChar != newChar) {
2025             String ret = isLatin1() ? StringLatin1.replace(value, oldChar, newChar)
2026                                     : StringUTF16.replace(value, oldChar, newChar);
2027             if (ret != null) {
2028                 return ret;
2029             }
2030         }
2031         return this;
2032     }
2033 
2034     /**
2035      * Tells whether or not this string matches the given <a
2036      * href="../util/regex/Pattern.html#sum">regular expression</a>.
2037      *
2038      * <p> An invocation of this method of the form
2039      * <i>str</i>{@code .matches(}<i>regex</i>{@code )} yields exactly the
2040      * same result as the expression
2041      *
2042      * <blockquote>
2043      * {@link java.util.regex.Pattern}.{@link java.util.regex.Pattern#matches(String,CharSequence)
2044      * matches(<i>regex</i>, <i>str</i>)}
2045      * </blockquote>
2046      *
2047      * @param   regex
2048      *          the regular expression to which this string is to be matched
2049      *
2050      * @return  {@code true} if, and only if, this string matches the
2051      *          given regular expression
2052      *
2053      * @throws  PatternSyntaxException
2054      *          if the regular expression's syntax is invalid
2055      *
2056      * @see java.util.regex.Pattern
2057      *
2058      * @since 1.4
2059      * @spec JSR-51
2060      */
2061     public boolean matches(String regex) {
2062         return Pattern.matches(regex, this);
2063     }
2064 
2065     /**
2066      * Returns true if and only if this string contains the specified
2067      * sequence of char values.
2068      *
2069      * @param s the sequence to search for
2070      * @return true if this string contains {@code s}, false otherwise
2071      * @since 1.5
2072      */
2073     public boolean contains(CharSequence s) {
2074         return indexOf(s.toString()) >= 0;
2075     }
2076 
2077     /**
2078      * Replaces the first substring of this string that matches the given <a
2079      * href="../util/regex/Pattern.html#sum">regular expression</a> with the
2080      * given replacement.
2081      *
2082      * <p> An invocation of this method of the form
2083      * <i>str</i>{@code .replaceFirst(}<i>regex</i>{@code ,} <i>repl</i>{@code )}
2084      * yields exactly the same result as the expression
2085      *
2086      * <blockquote>
2087      * <code>
2088      * {@link java.util.regex.Pattern}.{@link
2089      * java.util.regex.Pattern#compile compile}(<i>regex</i>).{@link
2090      * java.util.regex.Pattern#matcher(java.lang.CharSequence) matcher}(<i>str</i>).{@link
2091      * java.util.regex.Matcher#replaceFirst replaceFirst}(<i>repl</i>)
2092      * </code>
2093      * </blockquote>
2094      *
2095      *<p>
2096      * Note that backslashes ({@code \}) and dollar signs ({@code $}) in the
2097      * replacement string may cause the results to be different than if it were
2098      * being treated as a literal replacement string; see
2099      * {@link java.util.regex.Matcher#replaceFirst}.
2100      * Use {@link java.util.regex.Matcher#quoteReplacement} to suppress the special
2101      * meaning of these characters, if desired.
2102      *
2103      * @param   regex
2104      *          the regular expression to which this string is to be matched
2105      * @param   replacement
2106      *          the string to be substituted for the first match
2107      *
2108      * @return  The resulting {@code String}
2109      *
2110      * @throws  PatternSyntaxException
2111      *          if the regular expression's syntax is invalid
2112      *
2113      * @see java.util.regex.Pattern
2114      *
2115      * @since 1.4
2116      * @spec JSR-51
2117      */
2118     public String replaceFirst(String regex, String replacement) {
2119         return Pattern.compile(regex).matcher(this).replaceFirst(replacement);
2120     }
2121 
2122     /**
2123      * Replaces each substring of this string that matches the given <a
2124      * href="../util/regex/Pattern.html#sum">regular expression</a> with the
2125      * given replacement.
2126      *
2127      * <p> An invocation of this method of the form
2128      * <i>str</i>{@code .replaceAll(}<i>regex</i>{@code ,} <i>repl</i>{@code )}
2129      * yields exactly the same result as the expression
2130      *
2131      * <blockquote>
2132      * <code>
2133      * {@link java.util.regex.Pattern}.{@link
2134      * java.util.regex.Pattern#compile compile}(<i>regex</i>).{@link
2135      * java.util.regex.Pattern#matcher(java.lang.CharSequence) matcher}(<i>str</i>).{@link
2136      * java.util.regex.Matcher#replaceAll replaceAll}(<i>repl</i>)
2137      * </code>
2138      * </blockquote>
2139      *
2140      *<p>
2141      * Note that backslashes ({@code \}) and dollar signs ({@code $}) in the
2142      * replacement string may cause the results to be different than if it were
2143      * being treated as a literal replacement string; see
2144      * {@link java.util.regex.Matcher#replaceAll Matcher.replaceAll}.
2145      * Use {@link java.util.regex.Matcher#quoteReplacement} to suppress the special
2146      * meaning of these characters, if desired.
2147      *
2148      * @param   regex
2149      *          the regular expression to which this string is to be matched
2150      * @param   replacement
2151      *          the string to be substituted for each match
2152      *
2153      * @return  The resulting {@code String}
2154      *
2155      * @throws  PatternSyntaxException
2156      *          if the regular expression's syntax is invalid
2157      *
2158      * @see java.util.regex.Pattern
2159      *
2160      * @since 1.4
2161      * @spec JSR-51
2162      */
2163     public String replaceAll(String regex, String replacement) {
2164         return Pattern.compile(regex).matcher(this).replaceAll(replacement);
2165     }
2166 
2167     /**
2168      * Replaces each substring of this string that matches the literal target
2169      * sequence with the specified literal replacement sequence. The
2170      * replacement proceeds from the beginning of the string to the end, for
2171      * example, replacing "aa" with "b" in the string "aaa" will result in
2172      * "ba" rather than "ab".
2173      *
2174      * @param  target The sequence of char values to be replaced
2175      * @param  replacement The replacement sequence of char values
2176      * @return  The resulting string
2177      * @since 1.5
2178      */
2179     public String replace(CharSequence target, CharSequence replacement) {
2180         String tgtStr = target.toString();
2181         String replStr = replacement.toString();
2182         int j = indexOf(tgtStr);
2183         if (j < 0) {
2184             return this;
2185         }
2186         int tgtLen = tgtStr.length();
2187         int tgtLen1 = Math.max(tgtLen, 1);
2188         int thisLen = length();
2189 
2190         int newLenHint = thisLen - tgtLen + replStr.length();
2191         if (newLenHint < 0) {
2192             throw new OutOfMemoryError();
2193         }
2194         StringBuilder sb = new StringBuilder(newLenHint);
2195         int i = 0;
2196         do {
2197             sb.append(this, i, j).append(replStr);
2198             i = j + tgtLen;
2199         } while (j < thisLen && (j = indexOf(tgtStr, j + tgtLen1)) > 0);
2200         return sb.append(this, i, thisLen).toString();
2201     }
2202 
2203     /**
2204      * Splits this string around matches of the given
2205      * <a href="../util/regex/Pattern.html#sum">regular expression</a>.
2206      *
2207      * <p> The array returned by this method contains each substring of this
2208      * string that is terminated by another substring that matches the given
2209      * expression or is terminated by the end of the string.  The substrings in
2210      * the array are in the order in which they occur in this string.  If the
2211      * expression does not match any part of the input then the resulting array
2212      * has just one element, namely this string.
2213      *
2214      * <p> When there is a positive-width match at the beginning of this
2215      * string then an empty leading substring is included at the beginning
2216      * of the resulting array. A zero-width match at the beginning however
2217      * never produces such empty leading substring.
2218      *
2219      * <p> The {@code limit} parameter controls the number of times the
2220      * pattern is applied and therefore affects the length of the resulting
2221      * array.  If the limit <i>n</i> is greater than zero then the pattern
2222      * will be applied at most <i>n</i>&nbsp;-&nbsp;1 times, the array's
2223      * length will be no greater than <i>n</i>, and the array's last entry
2224      * will contain all input beyond the last matched delimiter.  If <i>n</i>
2225      * is non-positive then the pattern will be applied as many times as
2226      * possible and the array can have any length.  If <i>n</i> is zero then
2227      * the pattern will be applied as many times as possible, the array can
2228      * have any length, and trailing empty strings will be discarded.
2229      *
2230      * <p> The string {@code "boo:and:foo"}, for example, yields the
2231      * following results with these parameters:
2232      *
2233      * <blockquote><table cellpadding=1 cellspacing=0 summary="Split example showing regex, limit, and result">
2234      * <tr>
2235      *     <th>Regex</th>
2236      *     <th>Limit</th>
2237      *     <th>Result</th>
2238      * </tr>
2239      * <tr><td align=center>:</td>
2240      *     <td align=center>2</td>
2241      *     <td>{@code { "boo", "and:foo" }}</td></tr>
2242      * <tr><td align=center>:</td>
2243      *     <td align=center>5</td>
2244      *     <td>{@code { "boo", "and", "foo" }}</td></tr>
2245      * <tr><td align=center>:</td>
2246      *     <td align=center>-2</td>
2247      *     <td>{@code { "boo", "and", "foo" }}</td></tr>
2248      * <tr><td align=center>o</td>
2249      *     <td align=center>5</td>
2250      *     <td>{@code { "b", "", ":and:f", "", "" }}</td></tr>
2251      * <tr><td align=center>o</td>
2252      *     <td align=center>-2</td>
2253      *     <td>{@code { "b", "", ":and:f", "", "" }}</td></tr>
2254      * <tr><td align=center>o</td>
2255      *     <td align=center>0</td>
2256      *     <td>{@code { "b", "", ":and:f" }}</td></tr>
2257      * </table></blockquote>
2258      *
2259      * <p> An invocation of this method of the form
2260      * <i>str.</i>{@code split(}<i>regex</i>{@code ,}&nbsp;<i>n</i>{@code )}
2261      * yields the same result as the expression
2262      *
2263      * <blockquote>
2264      * <code>
2265      * {@link java.util.regex.Pattern}.{@link
2266      * java.util.regex.Pattern#compile compile}(<i>regex</i>).{@link
2267      * java.util.regex.Pattern#split(java.lang.CharSequence,int) split}(<i>str</i>,&nbsp;<i>n</i>)
2268      * </code>
2269      * </blockquote>
2270      *
2271      *
2272      * @param  regex
2273      *         the delimiting regular expression
2274      *
2275      * @param  limit
2276      *         the result threshold, as described above
2277      *
2278      * @return  the array of strings computed by splitting this string
2279      *          around matches of the given regular expression
2280      *
2281      * @throws  PatternSyntaxException
2282      *          if the regular expression's syntax is invalid
2283      *
2284      * @see java.util.regex.Pattern
2285      *
2286      * @since 1.4
2287      * @spec JSR-51
2288      */
2289     public String[] split(String regex, int limit) {
2290         /* fastpath if the regex is a
2291          (1)one-char String and this character is not one of the
2292             RegEx's meta characters ".$|()[{^?*+\\", or
2293          (2)two-char String and the first char is the backslash and
2294             the second is not the ascii digit or ascii letter.
2295          */
2296         char ch = 0;
2297         if (((regex.length() == 1 &&
2298              ".$|()[{^?*+\\".indexOf(ch = regex.charAt(0)) == -1) ||
2299              (regex.length() == 2 &&
2300               regex.charAt(0) == '\\' &&
2301               (((ch = regex.charAt(1))-'0')|('9'-ch)) < 0 &&
2302               ((ch-'a')|('z'-ch)) < 0 &&
2303               ((ch-'A')|('Z'-ch)) < 0)) &&
2304             (ch < Character.MIN_HIGH_SURROGATE ||
2305              ch > Character.MAX_LOW_SURROGATE))
2306         {
2307             int off = 0;
2308             int next = 0;
2309             boolean limited = limit > 0;
2310             ArrayList<String> list = new ArrayList<>();
2311             while ((next = indexOf(ch, off)) != -1) {
2312                 if (!limited || list.size() < limit - 1) {
2313                     list.add(substring(off, next));
2314                     off = next + 1;
2315                 } else {    // last one
2316                     //assert (list.size() == limit - 1);
2317                     int last = length();
2318                     list.add(substring(off, last));
2319                     off = last;
2320                     break;
2321                 }
2322             }
2323             // If no match was found, return this
2324             if (off == 0)
2325                 return new String[]{this};
2326 
2327             // Add remaining segment
2328             if (!limited || list.size() < limit)
2329                 list.add(substring(off, length()));
2330 
2331             // Construct result
2332             int resultSize = list.size();
2333             if (limit == 0) {
2334                 while (resultSize > 0 && list.get(resultSize - 1).length() == 0) {
2335                     resultSize--;
2336                 }
2337             }
2338             String[] result = new String[resultSize];
2339             return list.subList(0, resultSize).toArray(result);
2340         }
2341         return Pattern.compile(regex).split(this, limit);
2342     }
2343 
2344     /**
2345      * Splits this string around matches of the given <a
2346      * href="../util/regex/Pattern.html#sum">regular expression</a>.
2347      *
2348      * <p> This method works as if by invoking the two-argument {@link
2349      * #split(String, int) split} method with the given expression and a limit
2350      * argument of zero.  Trailing empty strings are therefore not included in
2351      * the resulting array.
2352      *
2353      * <p> The string {@code "boo:and:foo"}, for example, yields the following
2354      * results with these expressions:
2355      *
2356      * <blockquote><table cellpadding=1 cellspacing=0 summary="Split examples showing regex and result">
2357      * <tr>
2358      *  <th>Regex</th>
2359      *  <th>Result</th>
2360      * </tr>
2361      * <tr><td align=center>:</td>
2362      *     <td>{@code { "boo", "and", "foo" }}</td></tr>
2363      * <tr><td align=center>o</td>
2364      *     <td>{@code { "b", "", ":and:f" }}</td></tr>
2365      * </table></blockquote>
2366      *
2367      *
2368      * @param  regex
2369      *         the delimiting regular expression
2370      *
2371      * @return  the array of strings computed by splitting this string
2372      *          around matches of the given regular expression
2373      *
2374      * @throws  PatternSyntaxException
2375      *          if the regular expression's syntax is invalid
2376      *
2377      * @see java.util.regex.Pattern
2378      *
2379      * @since 1.4
2380      * @spec JSR-51
2381      */
2382     public String[] split(String regex) {
2383         return split(regex, 0);
2384     }
2385 
2386     /**
2387      * Returns a new String composed of copies of the
2388      * {@code CharSequence elements} joined together with a copy of
2389      * the specified {@code delimiter}.
2390      *
2391      * <blockquote>For example,
2392      * <pre>{@code
2393      *     String message = String.join("-", "Java", "is", "cool");
2394      *     // message returned is: "Java-is-cool"
2395      * }</pre></blockquote>
2396      *
2397      * Note that if an element is null, then {@code "null"} is added.
2398      *
2399      * @param  delimiter the delimiter that separates each element
2400      * @param  elements the elements to join together.
2401      *
2402      * @return a new {@code String} that is composed of the {@code elements}
2403      *         separated by the {@code delimiter}
2404      *
2405      * @throws NullPointerException If {@code delimiter} or {@code elements}
2406      *         is {@code null}
2407      *
2408      * @see java.util.StringJoiner
2409      * @since 1.8
2410      */
2411     public static String join(CharSequence delimiter, CharSequence... elements) {
2412         Objects.requireNonNull(delimiter);
2413         Objects.requireNonNull(elements);
2414         // Number of elements not likely worth Arrays.stream overhead.
2415         StringJoiner joiner = new StringJoiner(delimiter);
2416         for (CharSequence cs: elements) {
2417             joiner.add(cs);
2418         }
2419         return joiner.toString();
2420     }
2421 
2422     /**
2423      * Returns a new {@code String} composed of copies of the
2424      * {@code CharSequence elements} joined together with a copy of the
2425      * specified {@code delimiter}.
2426      *
2427      * <blockquote>For example,
2428      * <pre>{@code
2429      *     List<String> strings = List.of("Java", "is", "cool");
2430      *     String message = String.join(" ", strings);
2431      *     //message returned is: "Java is cool"
2432      *
2433      *     Set<String> strings =
2434      *         new LinkedHashSet<>(List.of("Java", "is", "very", "cool"));
2435      *     String message = String.join("-", strings);
2436      *     //message returned is: "Java-is-very-cool"
2437      * }</pre></blockquote>
2438      *
2439      * Note that if an individual element is {@code null}, then {@code "null"} is added.
2440      *
2441      * @param  delimiter a sequence of characters that is used to separate each
2442      *         of the {@code elements} in the resulting {@code String}
2443      * @param  elements an {@code Iterable} that will have its {@code elements}
2444      *         joined together.
2445      *
2446      * @return a new {@code String} that is composed from the {@code elements}
2447      *         argument
2448      *
2449      * @throws NullPointerException If {@code delimiter} or {@code elements}
2450      *         is {@code null}
2451      *
2452      * @see    #join(CharSequence,CharSequence...)
2453      * @see    java.util.StringJoiner
2454      * @since 1.8
2455      */
2456     public static String join(CharSequence delimiter,
2457             Iterable<? extends CharSequence> elements) {
2458         Objects.requireNonNull(delimiter);
2459         Objects.requireNonNull(elements);
2460         StringJoiner joiner = new StringJoiner(delimiter);
2461         for (CharSequence cs: elements) {
2462             joiner.add(cs);
2463         }
2464         return joiner.toString();
2465     }
2466 
2467     /**
2468      * Converts all of the characters in this {@code String} to lower
2469      * case using the rules of the given {@code Locale}.  Case mapping is based
2470      * on the Unicode Standard version specified by the {@link java.lang.Character Character}
2471      * class. Since case mappings are not always 1:1 char mappings, the resulting
2472      * {@code String} may be a different length than the original {@code String}.
2473      * <p>
2474      * Examples of lowercase  mappings are in the following table:
2475      * <table border="1" summary="Lowercase mapping examples showing language code of locale, upper case, lower case, and description">
2476      * <tr>
2477      *   <th>Language Code of Locale</th>
2478      *   <th>Upper Case</th>
2479      *   <th>Lower Case</th>
2480      *   <th>Description</th>
2481      * </tr>
2482      * <tr>
2483      *   <td>tr (Turkish)</td>
2484      *   <td>&#92;u0130</td>
2485      *   <td>&#92;u0069</td>
2486      *   <td>capital letter I with dot above -&gt; small letter i</td>
2487      * </tr>
2488      * <tr>
2489      *   <td>tr (Turkish)</td>
2490      *   <td>&#92;u0049</td>
2491      *   <td>&#92;u0131</td>
2492      *   <td>capital letter I -&gt; small letter dotless i </td>
2493      * </tr>
2494      * <tr>
2495      *   <td>(all)</td>
2496      *   <td>French Fries</td>
2497      *   <td>french fries</td>
2498      *   <td>lowercased all chars in String</td>
2499      * </tr>
2500      * <tr>
2501      *   <td>(all)</td>
2502      *   <td><img src="doc-files/capiota.gif" alt="capiota"><img src="doc-files/capchi.gif" alt="capchi">
2503      *       <img src="doc-files/captheta.gif" alt="captheta"><img src="doc-files/capupsil.gif" alt="capupsil">
2504      *       <img src="doc-files/capsigma.gif" alt="capsigma"></td>
2505      *   <td><img src="doc-files/iota.gif" alt="iota"><img src="doc-files/chi.gif" alt="chi">
2506      *       <img src="doc-files/theta.gif" alt="theta"><img src="doc-files/upsilon.gif" alt="upsilon">
2507      *       <img src="doc-files/sigma1.gif" alt="sigma"></td>
2508      *   <td>lowercased all chars in String</td>
2509      * </tr>
2510      * </table>
2511      *
2512      * @param locale use the case transformation rules for this locale
2513      * @return the {@code String}, converted to lowercase.
2514      * @see     java.lang.String#toLowerCase()
2515      * @see     java.lang.String#toUpperCase()
2516      * @see     java.lang.String#toUpperCase(Locale)
2517      * @since   1.1
2518      */
2519     public String toLowerCase(Locale locale) {
2520         return isLatin1() ? StringLatin1.toLowerCase(this, value, locale)
2521                           : StringUTF16.toLowerCase(this, value, locale);
2522     }
2523 
2524     /**
2525      * Converts all of the characters in this {@code String} to lower
2526      * case using the rules of the default locale. This is equivalent to calling
2527      * {@code toLowerCase(Locale.getDefault())}.
2528      * <p>
2529      * <b>Note:</b> This method is locale sensitive, and may produce unexpected
2530      * results if used for strings that are intended to be interpreted locale
2531      * independently.
2532      * Examples are programming language identifiers, protocol keys, and HTML
2533      * tags.
2534      * For instance, {@code "TITLE".toLowerCase()} in a Turkish locale
2535      * returns {@code "t\u005Cu0131tle"}, where '\u005Cu0131' is the
2536      * LATIN SMALL LETTER DOTLESS I character.
2537      * To obtain correct results for locale insensitive strings, use
2538      * {@code toLowerCase(Locale.ROOT)}.
2539      *
2540      * @return  the {@code String}, converted to lowercase.
2541      * @see     java.lang.String#toLowerCase(Locale)
2542      */
2543     public String toLowerCase() {
2544         return toLowerCase(Locale.getDefault());
2545     }
2546 
2547     /**
2548      * Converts all of the characters in this {@code String} to upper
2549      * case using the rules of the given {@code Locale}. Case mapping is based
2550      * on the Unicode Standard version specified by the {@link java.lang.Character Character}
2551      * class. Since case mappings are not always 1:1 char mappings, the resulting
2552      * {@code String} may be a different length than the original {@code String}.
2553      * <p>
2554      * Examples of locale-sensitive and 1:M case mappings are in the following table.
2555      *
2556      * <table border="1" summary="Examples of locale-sensitive and 1:M case mappings. Shows Language code of locale, lower case, upper case, and description.">
2557      * <tr>
2558      *   <th>Language Code of Locale</th>
2559      *   <th>Lower Case</th>
2560      *   <th>Upper Case</th>
2561      *   <th>Description</th>
2562      * </tr>
2563      * <tr>
2564      *   <td>tr (Turkish)</td>
2565      *   <td>&#92;u0069</td>
2566      *   <td>&#92;u0130</td>
2567      *   <td>small letter i -&gt; capital letter I with dot above</td>
2568      * </tr>
2569      * <tr>
2570      *   <td>tr (Turkish)</td>
2571      *   <td>&#92;u0131</td>
2572      *   <td>&#92;u0049</td>
2573      *   <td>small letter dotless i -&gt; capital letter I</td>
2574      * </tr>
2575      * <tr>
2576      *   <td>(all)</td>
2577      *   <td>&#92;u00df</td>
2578      *   <td>&#92;u0053 &#92;u0053</td>
2579      *   <td>small letter sharp s -&gt; two letters: SS</td>
2580      * </tr>
2581      * <tr>
2582      *   <td>(all)</td>
2583      *   <td>Fahrvergn&uuml;gen</td>
2584      *   <td>FAHRVERGN&Uuml;GEN</td>
2585      *   <td></td>
2586      * </tr>
2587      * </table>
2588      * @param locale use the case transformation rules for this locale
2589      * @return the {@code String}, converted to uppercase.
2590      * @see     java.lang.String#toUpperCase()
2591      * @see     java.lang.String#toLowerCase()
2592      * @see     java.lang.String#toLowerCase(Locale)
2593      * @since   1.1
2594      */
2595     public String toUpperCase(Locale locale) {
2596         return isLatin1() ? StringLatin1.toUpperCase(this, value, locale)
2597                           : StringUTF16.toUpperCase(this, value, locale);
2598     }
2599 
2600     /**
2601      * Converts all of the characters in this {@code String} to upper
2602      * case using the rules of the default locale. This method is equivalent to
2603      * {@code toUpperCase(Locale.getDefault())}.
2604      * <p>
2605      * <b>Note:</b> This method is locale sensitive, and may produce unexpected
2606      * results if used for strings that are intended to be interpreted locale
2607      * independently.
2608      * Examples are programming language identifiers, protocol keys, and HTML
2609      * tags.
2610      * For instance, {@code "title".toUpperCase()} in a Turkish locale
2611      * returns {@code "T\u005Cu0130TLE"}, where '\u005Cu0130' is the
2612      * LATIN CAPITAL LETTER I WITH DOT ABOVE character.
2613      * To obtain correct results for locale insensitive strings, use
2614      * {@code toUpperCase(Locale.ROOT)}.
2615      *
2616      * @return  the {@code String}, converted to uppercase.
2617      * @see     java.lang.String#toUpperCase(Locale)
2618      */
2619     public String toUpperCase() {
2620         return toUpperCase(Locale.getDefault());
2621     }
2622 
2623     /**
2624      * Returns a string whose value is this string, with any leading and trailing
2625      * whitespace removed.
2626      * <p>
2627      * If this {@code String} object represents an empty character
2628      * sequence, or the first and last characters of character sequence
2629      * represented by this {@code String} object both have codes
2630      * greater than {@code '\u005Cu0020'} (the space character), then a
2631      * reference to this {@code String} object is returned.
2632      * <p>
2633      * Otherwise, if there is no character with a code greater than
2634      * {@code '\u005Cu0020'} in the string, then a
2635      * {@code String} object representing an empty string is
2636      * returned.
2637      * <p>
2638      * Otherwise, let <i>k</i> be the index of the first character in the
2639      * string whose code is greater than {@code '\u005Cu0020'}, and let
2640      * <i>m</i> be the index of the last character in the string whose code
2641      * is greater than {@code '\u005Cu0020'}. A {@code String}
2642      * object is returned, representing the substring of this string that
2643      * begins with the character at index <i>k</i> and ends with the
2644      * character at index <i>m</i>-that is, the result of
2645      * {@code this.substring(k, m + 1)}.
2646      * <p>
2647      * This method may be used to trim whitespace (as defined above) from
2648      * the beginning and end of a string.
2649      *
2650      * @return  A string whose value is this string, with any leading and trailing white
2651      *          space removed, or this string if it has no leading or
2652      *          trailing white space.
2653      */
2654     public String trim() {
2655         String ret = isLatin1() ? StringLatin1.trim(value)
2656                                 : StringUTF16.trim(value);
2657         return ret == null ? this : ret;
2658     }
2659 
2660     /**
2661      * This object (which is already a string!) is itself returned.
2662      *
2663      * @return  the string itself.
2664      */
2665     public String toString() {
2666         return this;
2667     }
2668 
2669     /**
2670      * Returns a stream of {@code int} zero-extending the {@code char} values
2671      * from this sequence.  Any char which maps to a <a
2672      * href="{@docRoot}/java/lang/Character.html#unicode">surrogate code
2673      * point</a> is passed through uninterpreted.
2674      *
2675      * @return an IntStream of char values from this sequence
2676      * @since 9
2677      */
2678     @Override
2679     public IntStream chars() {
2680         return StreamSupport.intStream(
2681             isLatin1() ? new StringLatin1.CharsSpliterator(value, Spliterator.IMMUTABLE)
2682                        : new StringUTF16.CharsSpliterator(value, Spliterator.IMMUTABLE),
2683             false);
2684     }
2685 
2686 
2687     /**
2688      * Returns a stream of code point values from this sequence.  Any surrogate
2689      * pairs encountered in the sequence are combined as if by {@linkplain
2690      * Character#toCodePoint Character.toCodePoint} and the result is passed
2691      * to the stream. Any other code units, including ordinary BMP characters,
2692      * unpaired surrogates, and undefined code units, are zero-extended to
2693      * {@code int} values which are then passed to the stream.
2694      *
2695      * @return an IntStream of Unicode code points from this sequence
2696      * @since 9
2697      */
2698     @Override
2699     public IntStream codePoints() {
2700         return StreamSupport.intStream(
2701             isLatin1() ? new StringLatin1.CharsSpliterator(value, Spliterator.IMMUTABLE)
2702                        : new StringUTF16.CodePointsSpliterator(value, Spliterator.IMMUTABLE),
2703             false);
2704     }
2705 
2706     /**
2707      * Converts this string to a new character array.
2708      *
2709      * @return  a newly allocated character array whose length is the length
2710      *          of this string and whose contents are initialized to contain
2711      *          the character sequence represented by this string.
2712      */
2713     public char[] toCharArray() {
2714         return isLatin1() ? StringLatin1.toChars(value)
2715                           : StringUTF16.toChars(value);
2716     }
2717 
2718     /**
2719      * Returns a formatted string using the specified format string and
2720      * arguments.
2721      *
2722      * <p> The locale always used is the one returned by {@link
2723      * java.util.Locale#getDefault(java.util.Locale.Category)
2724      * Locale.getDefault(Locale.Category)} with
2725      * {@link java.util.Locale.Category#FORMAT FORMAT} category specified.
2726      *
2727      * @param  format
2728      *         A <a href="../util/Formatter.html#syntax">format string</a>
2729      *
2730      * @param  args
2731      *         Arguments referenced by the format specifiers in the format
2732      *         string.  If there are more arguments than format specifiers, the
2733      *         extra arguments are ignored.  The number of arguments is
2734      *         variable and may be zero.  The maximum number of arguments is
2735      *         limited by the maximum dimension of a Java array as defined by
2736      *         <cite>The Java&trade; Virtual Machine Specification</cite>.
2737      *         The behaviour on a
2738      *         {@code null} argument depends on the <a
2739      *         href="../util/Formatter.html#syntax">conversion</a>.
2740      *
2741      * @throws  java.util.IllegalFormatException
2742      *          If a format string contains an illegal syntax, a format
2743      *          specifier that is incompatible with the given arguments,
2744      *          insufficient arguments given the format string, or other
2745      *          illegal conditions.  For specification of all possible
2746      *          formatting errors, see the <a
2747      *          href="../util/Formatter.html#detail">Details</a> section of the
2748      *          formatter class specification.
2749      *
2750      * @return  A formatted string
2751      *
2752      * @see  java.util.Formatter
2753      * @since  1.5
2754      */
2755     public static String format(String format, Object... args) {
2756         return new Formatter().format(format, args).toString();
2757     }
2758 
2759     /**
2760      * Returns a formatted string using the specified locale, format string,
2761      * and arguments.
2762      *
2763      * @param  l
2764      *         The {@linkplain java.util.Locale locale} to apply during
2765      *         formatting.  If {@code l} is {@code null} then no localization
2766      *         is applied.
2767      *
2768      * @param  format
2769      *         A <a href="../util/Formatter.html#syntax">format string</a>
2770      *
2771      * @param  args
2772      *         Arguments referenced by the format specifiers in the format
2773      *         string.  If there are more arguments than format specifiers, the
2774      *         extra arguments are ignored.  The number of arguments is
2775      *         variable and may be zero.  The maximum number of arguments is
2776      *         limited by the maximum dimension of a Java array as defined by
2777      *         <cite>The Java&trade; Virtual Machine Specification</cite>.
2778      *         The behaviour on a
2779      *         {@code null} argument depends on the
2780      *         <a href="../util/Formatter.html#syntax">conversion</a>.
2781      *
2782      * @throws  java.util.IllegalFormatException
2783      *          If a format string contains an illegal syntax, a format
2784      *          specifier that is incompatible with the given arguments,
2785      *          insufficient arguments given the format string, or other
2786      *          illegal conditions.  For specification of all possible
2787      *          formatting errors, see the <a
2788      *          href="../util/Formatter.html#detail">Details</a> section of the
2789      *          formatter class specification
2790      *
2791      * @return  A formatted string
2792      *
2793      * @see  java.util.Formatter
2794      * @since  1.5
2795      */
2796     public static String format(Locale l, String format, Object... args) {
2797         return new Formatter(l).format(format, args).toString();
2798     }
2799 
2800     /**
2801      * Returns the string representation of the {@code Object} argument.
2802      *
2803      * @param   obj   an {@code Object}.
2804      * @return  if the argument is {@code null}, then a string equal to
2805      *          {@code "null"}; otherwise, the value of
2806      *          {@code obj.toString()} is returned.
2807      * @see     java.lang.Object#toString()
2808      */
2809     public static String valueOf(Object obj) {
2810         return (obj == null) ? "null" : obj.toString();
2811     }
2812 
2813     /**
2814      * Returns the string representation of the {@code char} array
2815      * argument. The contents of the character array are copied; subsequent
2816      * modification of the character array does not affect the returned
2817      * string.
2818      *
2819      * @param   data     the character array.
2820      * @return  a {@code String} that contains the characters of the
2821      *          character array.
2822      */
2823     public static String valueOf(char data[]) {
2824         return new String(data);
2825     }
2826 
2827     /**
2828      * Returns the string representation of a specific subarray of the
2829      * {@code char} array argument.
2830      * <p>
2831      * The {@code offset} argument is the index of the first
2832      * character of the subarray. The {@code count} argument
2833      * specifies the length of the subarray. The contents of the subarray
2834      * are copied; subsequent modification of the character array does not
2835      * affect the returned string.
2836      *
2837      * @param   data     the character array.
2838      * @param   offset   initial offset of the subarray.
2839      * @param   count    length of the subarray.
2840      * @return  a {@code String} that contains the characters of the
2841      *          specified subarray of the character array.
2842      * @exception IndexOutOfBoundsException if {@code offset} is
2843      *          negative, or {@code count} is negative, or
2844      *          {@code offset+count} is larger than
2845      *          {@code data.length}.
2846      */
2847     public static String valueOf(char data[], int offset, int count) {
2848         return new String(data, offset, count);
2849     }
2850 
2851     /**
2852      * Equivalent to {@link #valueOf(char[], int, int)}.
2853      *
2854      * @param   data     the character array.
2855      * @param   offset   initial offset of the subarray.
2856      * @param   count    length of the subarray.
2857      * @return  a {@code String} that contains the characters of the
2858      *          specified subarray of the character array.
2859      * @exception IndexOutOfBoundsException if {@code offset} is
2860      *          negative, or {@code count} is negative, or
2861      *          {@code offset+count} is larger than
2862      *          {@code data.length}.
2863      */
2864     public static String copyValueOf(char data[], int offset, int count) {
2865         return new String(data, offset, count);
2866     }
2867 
2868     /**
2869      * Equivalent to {@link #valueOf(char[])}.
2870      *
2871      * @param   data   the character array.
2872      * @return  a {@code String} that contains the characters of the
2873      *          character array.
2874      */
2875     public static String copyValueOf(char data[]) {
2876         return new String(data);
2877     }
2878 
2879     /**
2880      * Returns the string representation of the {@code boolean} argument.
2881      *
2882      * @param   b   a {@code boolean}.
2883      * @return  if the argument is {@code true}, a string equal to
2884      *          {@code "true"} is returned; otherwise, a string equal to
2885      *          {@code "false"} is returned.
2886      */
2887     public static String valueOf(boolean b) {
2888         return b ? "true" : "false";
2889     }
2890 
2891     /**
2892      * Returns the string representation of the {@code char}
2893      * argument.
2894      *
2895      * @param   c   a {@code char}.
2896      * @return  a string of length {@code 1} containing
2897      *          as its single character the argument {@code c}.
2898      */
2899     public static String valueOf(char c) {
2900         if (COMPACT_STRINGS && StringLatin1.canEncode(c)) {
2901             return new String(StringLatin1.toBytes(c), LATIN1);
2902         }
2903         return new String(StringUTF16.toBytes(c), UTF16);
2904     }
2905 
2906     /**
2907      * Returns the string representation of the {@code int} argument.
2908      * <p>
2909      * The representation is exactly the one returned by the
2910      * {@code Integer.toString} method of one argument.
2911      *
2912      * @param   i   an {@code int}.
2913      * @return  a string representation of the {@code int} argument.
2914      * @see     java.lang.Integer#toString(int, int)
2915      */
2916     public static String valueOf(int i) {
2917         return Integer.toString(i);
2918     }
2919 
2920     /**
2921      * Returns the string representation of the {@code long} argument.
2922      * <p>
2923      * The representation is exactly the one returned by the
2924      * {@code Long.toString} method of one argument.
2925      *
2926      * @param   l   a {@code long}.
2927      * @return  a string representation of the {@code long} argument.
2928      * @see     java.lang.Long#toString(long)
2929      */
2930     public static String valueOf(long l) {
2931         return Long.toString(l);
2932     }
2933 
2934     /**
2935      * Returns the string representation of the {@code float} argument.
2936      * <p>
2937      * The representation is exactly the one returned by the
2938      * {@code Float.toString} method of one argument.
2939      *
2940      * @param   f   a {@code float}.
2941      * @return  a string representation of the {@code float} argument.
2942      * @see     java.lang.Float#toString(float)
2943      */
2944     public static String valueOf(float f) {
2945         return Float.toString(f);
2946     }
2947 
2948     /**
2949      * Returns the string representation of the {@code double} argument.
2950      * <p>
2951      * The representation is exactly the one returned by the
2952      * {@code Double.toString} method of one argument.
2953      *
2954      * @param   d   a {@code double}.
2955      * @return  a  string representation of the {@code double} argument.
2956      * @see     java.lang.Double#toString(double)
2957      */
2958     public static String valueOf(double d) {
2959         return Double.toString(d);
2960     }
2961 
2962     /**
2963      * Returns a canonical representation for the string object.
2964      * <p>
2965      * A pool of strings, initially empty, is maintained privately by the
2966      * class {@code String}.
2967      * <p>
2968      * When the intern method is invoked, if the pool already contains a
2969      * string equal to this {@code String} object as determined by
2970      * the {@link #equals(Object)} method, then the string from the pool is
2971      * returned. Otherwise, this {@code String} object is added to the
2972      * pool and a reference to this {@code String} object is returned.
2973      * <p>
2974      * It follows that for any two strings {@code s} and {@code t},
2975      * {@code s.intern() == t.intern()} is {@code true}
2976      * if and only if {@code s.equals(t)} is {@code true}.
2977      * <p>
2978      * All literal strings and string-valued constant expressions are
2979      * interned. String literals are defined in section 3.10.5 of the
2980      * <cite>The Java&trade; Language Specification</cite>.
2981      *
2982      * @return  a string that has the same contents as this string, but is
2983      *          guaranteed to be from a pool of unique strings.
2984      * @jls 3.10.5 String Literals
2985      */
2986     public native String intern();
2987 
2988     @Override
2989     public String resolveConstant(MethodHandles.Lookup lookup) {
2990         return this;
2991     }
2992 
2993     ////////////////////////////////////////////////////////////////
2994 
2995     /**
2996      * Copy character bytes from this string into dst starting at dstBegin.
2997      * This method doesn't perform any range checking.
2998      *
2999      * Invoker guarantees: dst is in UTF16 (inflate itself for asb), if two
3000      * coders are different, and dst is big enough (range check)
3001      *
3002      * @param dstBegin  the char index, not offset of byte[]
3003      * @param coder     the coder of dst[]
3004      */
3005     void getBytes(byte dst[], int dstBegin, byte coder) {
3006         if (coder() == coder) {
3007             System.arraycopy(value, 0, dst, dstBegin << coder, value.length);
3008         } else {    // this.coder == LATIN && coder == UTF16
3009             StringLatin1.inflate(value, 0, dst, dstBegin, value.length);
3010         }
3011     }
3012 
3013     /*
3014      * Package private constructor. Trailing Void argument is there for
3015      * disambiguating it against other (public) constructors.
3016      *
3017      * Stores the char[] value into a byte[] that each byte represents
3018      * the8 low-order bits of the corresponding character, if the char[]
3019      * contains only latin1 character. Or a byte[] that stores all
3020      * characters in their byte sequences defined by the {@code StringUTF16}.
3021      */
3022     String(char[] value, int off, int len, Void sig) {
3023         if (len == 0) {
3024             this.value = "".value;
3025             this.coder = "".coder;
3026             return;
3027         }
3028         if (COMPACT_STRINGS) {
3029             byte[] val = StringUTF16.compress(value, off, len);
3030             if (val != null) {
3031                 this.value = val;
3032                 this.coder = LATIN1;
3033                 return;
3034             }
3035         }
3036         this.coder = UTF16;
3037         this.value = StringUTF16.toBytes(value, off, len);
3038     }
3039 
3040     /*
3041      * Package private constructor. Trailing Void argument is there for
3042      * disambiguating it against other (public) constructors.
3043      */
3044     String(AbstractStringBuilder asb, Void sig) {
3045         byte[] val = asb.getValue();
3046         int length = asb.length();
3047         if (asb.isLatin1()) {
3048             this.coder = LATIN1;
3049             this.value = Arrays.copyOfRange(val, 0, length);
3050         } else {
3051             if (COMPACT_STRINGS) {
3052                 byte[] buf = StringUTF16.compress(val, 0, length);
3053                 if (buf != null) {
3054                     this.coder = LATIN1;
3055                     this.value = buf;
3056                     return;
3057                 }
3058             }
3059             this.coder = UTF16;
3060             this.value = Arrays.copyOfRange(val, 0, length << 1);
3061         }
3062     }
3063 
3064    /*
3065     * Package private constructor which shares value array for speed.
3066     */
3067     String(byte[] value, byte coder) {
3068         this.value = value;
3069         this.coder = coder;
3070     }
3071 
3072     byte coder() {
3073         return COMPACT_STRINGS ? coder : UTF16;
3074     }
3075 
3076     private boolean isLatin1() {
3077         return COMPACT_STRINGS && coder == LATIN1;
3078     }
3079 
3080     static final byte LATIN1 = 0;
3081     static final byte UTF16  = 1;
3082 
3083     /*
3084      * StringIndexOutOfBoundsException  if {@code index} is
3085      * negative or greater than or equal to {@code length}.
3086      */
3087     static void checkIndex(int index, int length) {
3088         if (index < 0 || index >= length) {
3089             throw new StringIndexOutOfBoundsException("index " + index);
3090         }
3091     }
3092 
3093     /*
3094      * StringIndexOutOfBoundsException  if {@code offset}
3095      * is negative or greater than {@code length}.
3096      */
3097     static void checkOffset(int offset, int length) {
3098         if (offset < 0 || offset > length) {
3099             throw new StringIndexOutOfBoundsException("offset " + offset +
3100                                                       ",length " + length);
3101         }
3102     }
3103 
3104     /*
3105      * Check {@code offset}, {@code count} against {@code 0} and {@code length}
3106      * bounds.
3107      *
3108      * @throws  StringIndexOutOfBoundsException
3109      *          If {@code offset} is negative, {@code count} is negative,
3110      *          or {@code offset} is greater than {@code length - count}
3111      */
3112     static void checkBoundsOffCount(int offset, int count, int length) {
3113         if (offset < 0 || count < 0 || offset > length - count) {
3114             throw new StringIndexOutOfBoundsException(
3115                 "offset " + offset + ", count " + count + ", length " + length);
3116         }
3117     }
3118 
3119     /*
3120      * Check {@code begin}, {@code end} against {@code 0} and {@code length}
3121      * bounds.
3122      *
3123      * @throws  StringIndexOutOfBoundsException
3124      *          If {@code begin} is negative, {@code begin} is greater than
3125      *          {@code end}, or {@code end} is greater than {@code length}.
3126      */
3127     private static void checkBoundsBeginEnd(int begin, int end, int length) {
3128         if (begin < 0 || begin > end || end > length) {
3129             throw new StringIndexOutOfBoundsException(
3130                 "begin " + begin + ", end " + end + ", length " + length);
3131         }
3132     }
3133 }