1 /*
2 * Copyright (c) 1994, 2023, 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.annotation.Native;
31 import java.lang.invoke.MethodHandles;
32 import java.lang.constant.Constable;
33 import java.lang.constant.ConstantDesc;
34 import java.nio.ByteBuffer;
35 import java.nio.CharBuffer;
36 import java.nio.charset.*;
37 import java.util.ArrayList;
38 import java.util.Arrays;
39 import java.util.Comparator;
40 import java.util.Formatter;
41 import java.util.List;
42 import java.util.Locale;
43 import java.util.Objects;
44 import java.util.Optional;
45 import java.util.Spliterator;
46 import java.util.function.Function;
47 import java.util.regex.Pattern;
48 import java.util.regex.PatternSyntaxException;
49 import java.util.stream.Collectors;
50 import java.util.stream.IntStream;
51 import java.util.stream.Stream;
52 import java.util.stream.StreamSupport;
53
54 import jdk.internal.util.ArraysSupport;
55 import jdk.internal.util.Preconditions;
56 import jdk.internal.vm.annotation.ForceInline;
57 import jdk.internal.vm.annotation.IntrinsicCandidate;
58 import jdk.internal.vm.annotation.Stable;
59 import sun.nio.cs.ArrayDecoder;
60 import sun.nio.cs.ArrayEncoder;
61
62 import sun.nio.cs.ISO_8859_1;
63 import sun.nio.cs.US_ASCII;
64 import sun.nio.cs.UTF_8;
65
66 /**
67 * The {@code String} class represents character strings. All
68 * string literals in Java programs, such as {@code "abc"}, are
69 * implemented as instances of this class.
70 * <p>
71 * Strings are constant; their values cannot be changed after they
72 * are created. String buffers support mutable strings.
73 * Because String objects are immutable they can be shared. For example:
74 * <blockquote><pre>
75 * String str = "abc";
76 * </pre></blockquote><p>
77 * is equivalent to:
78 * <blockquote><pre>
79 * char data[] = {'a', 'b', 'c'};
80 * String str = new String(data);
81 * </pre></blockquote><p>
82 * Here are some more examples of how strings can be used:
83 * <blockquote><pre>
84 * System.out.println("abc");
85 * String cde = "cde";
86 * System.out.println("abc" + cde);
87 * String c = "abc".substring(2, 3);
88 * String d = cde.substring(1, 2);
89 * </pre></blockquote>
90 * <p>
91 * The class {@code String} includes methods for examining
92 * individual characters of the sequence, for comparing strings, for
93 * searching strings, for extracting substrings, and for creating a
94 * copy of a string with all characters translated to uppercase or to
95 * lowercase. Case mapping is based on the Unicode Standard version
96 * specified by the {@link java.lang.Character Character} class.
97 * <p>
98 * The Java language provides special support for the string
99 * concatenation operator ( + ), and for conversion of
100 * other objects to strings. For additional information on string
101 * concatenation and conversion, see <i>The Java Language Specification</i>.
102 *
103 * <p> Unless otherwise noted, passing a {@code null} argument to a constructor
104 * or method in this class will cause a {@link NullPointerException} to be
105 * thrown.
106 *
107 * <p>A {@code String} represents a string in the UTF-16 format
108 * in which <em>supplementary characters</em> are represented by <em>surrogate
109 * pairs</em> (see the section <a href="Character.html#unicode">Unicode
110 * Character Representations</a> in the {@code Character} class for
111 * more information).
112 * Index values refer to {@code char} code units, so a supplementary
113 * character uses two positions in a {@code String}.
114 * <p>The {@code String} class provides methods for dealing with
115 * Unicode code points (i.e., characters), in addition to those for
116 * dealing with Unicode code units (i.e., {@code char} values).
117 *
118 * <p>Unless otherwise noted, methods for comparing Strings do not take locale
119 * into account. The {@link java.text.Collator} class provides methods for
120 * finer-grain, locale-sensitive String comparison.
121 *
122 * @implNote The implementation of the string concatenation operator is left to
123 * the discretion of a Java compiler, as long as the compiler ultimately conforms
124 * to <i>The Java Language Specification</i>. For example, the {@code javac} compiler
125 * may implement the operator with {@code StringBuffer}, {@code StringBuilder},
126 * or {@code java.lang.invoke.StringConcatFactory} depending on the JDK version. The
127 * implementation of string conversion is typically through the method {@code toString},
128 * defined by {@code Object} and inherited by all classes in Java.
129 *
130 * @author Lee Boynton
131 * @author Arthur van Hoff
132 * @author Martin Buchholz
133 * @author Ulf Zibis
134 * @see java.lang.Object#toString()
135 * @see java.lang.StringBuffer
136 * @see java.lang.StringBuilder
137 * @see java.nio.charset.Charset
138 * @since 1.0
139 * @jls 15.18.1 String Concatenation Operator +
140 */
141
142 public final class String
143 implements java.io.Serializable, Comparable<String>, CharSequence,
144 Constable, ConstantDesc {
145
146 /**
147 * The value is used for character storage.
148 *
149 * @implNote This field is trusted by the VM, and is a subject to
150 * constant folding if String instance is constant. Overwriting this
151 * field after construction will cause problems.
152 *
153 * Additionally, it is marked with {@link Stable} to trust the contents
154 * of the array. No other facility in JDK provides this functionality (yet).
155 * {@link Stable} is safe here, because value is never null.
156 */
157 @Stable
158 private final byte[] value;
159
160 /**
161 * The identifier of the encoding used to encode the bytes in
162 * {@code value}. The supported values in this implementation are
163 *
164 * LATIN1
165 * UTF16
166 *
167 * @implNote This field is trusted by the VM, and is a subject to
168 * constant folding if String instance is constant. Overwriting this
169 * field after construction will cause problems.
170 */
171 private final byte coder;
172
173 /** Cache the hash code for the string */
174 private int hash; // Default to 0
175
176 /**
177 * Cache if the hash has been calculated as actually being zero, enabling
178 * us to avoid recalculating this.
179 */
180 private boolean hashIsZero; // Default to false;
181
182 /** use serialVersionUID from JDK 1.0.2 for interoperability */
183 @java.io.Serial
184 private static final long serialVersionUID = -6849794470754667710L;
185
186 /**
187 * If String compaction is disabled, the bytes in {@code value} are
188 * always encoded in UTF16.
189 *
190 * For methods with several possible implementation paths, when String
191 * compaction is disabled, only one code path is taken.
192 *
193 * The instance field value is generally opaque to optimizing JIT
194 * compilers. Therefore, in performance-sensitive place, an explicit
195 * check of the static boolean {@code COMPACT_STRINGS} is done first
196 * before checking the {@code coder} field since the static boolean
197 * {@code COMPACT_STRINGS} would be constant folded away by an
198 * optimizing JIT compiler. The idioms for these cases are as follows.
199 *
200 * For code such as:
201 *
202 * if (coder == LATIN1) { ... }
203 *
204 * can be written more optimally as
205 *
206 * if (coder() == LATIN1) { ... }
207 *
208 * or:
209 *
210 * if (COMPACT_STRINGS && coder == LATIN1) { ... }
211 *
212 * An optimizing JIT compiler can fold the above conditional as:
213 *
214 * COMPACT_STRINGS == true => if (coder == LATIN1) { ... }
215 * COMPACT_STRINGS == false => if (false) { ... }
216 *
217 * @implNote
218 * The actual value for this field is injected by JVM. The static
219 * initialization block is used to set the value here to communicate
220 * that this static final field is not statically foldable, and to
221 * avoid any possible circular dependency during vm initialization.
222 */
223 static final boolean COMPACT_STRINGS;
224
225 static {
226 COMPACT_STRINGS = true;
227 }
228
229 /**
230 * Class String is special cased within the Serialization Stream Protocol.
231 *
232 * A String instance is written into an ObjectOutputStream according to
233 * <a href="{@docRoot}/../specs/serialization/protocol.html#stream-elements">
234 * <cite>Java Object Serialization Specification</cite>, Section 6.2, "Stream Elements"</a>
235 */
236 @java.io.Serial
237 private static final ObjectStreamField[] serialPersistentFields =
238 new ObjectStreamField[0];
239
240 /**
241 * Initializes a newly created {@code String} object so that it represents
242 * an empty character sequence. Note that use of this constructor is
243 * unnecessary since Strings are immutable.
244 */
245 public String() {
246 this.value = "".value;
247 this.coder = "".coder;
248 }
249
250 /**
251 * Initializes a newly created {@code String} object so that it represents
252 * the same sequence of characters as the argument; in other words, the
253 * newly created string is a copy of the argument string. Unless an
254 * explicit copy of {@code original} is needed, use of this constructor is
255 * unnecessary since Strings are immutable.
256 *
257 * @param original
258 * A {@code String}
259 */
260 @IntrinsicCandidate
261 public String(String original) {
262 this.value = original.value;
263 this.coder = original.coder;
264 this.hash = original.hash;
265 this.hashIsZero = original.hashIsZero;
266 }
267
268 /**
269 * Allocates a new {@code String} so that it represents the sequence of
270 * characters currently contained in the character array argument. The
271 * contents of the character array are copied; subsequent modification of
272 * the character array does not affect the newly created string.
273 *
274 * @param value
275 * The initial value of the string
276 */
277 public String(char[] value) {
278 this(value, 0, value.length, null);
279 }
280
281 /**
282 * Allocates a new {@code String} that contains characters from a subarray
283 * of the character array argument. The {@code offset} argument is the
284 * index of the first character of the subarray and the {@code count}
285 * argument specifies the length of the subarray. The contents of the
286 * subarray are copied; subsequent modification of the character array does
287 * not affect the newly created string.
288 *
289 * @param value
290 * Array that is the source of characters
291 *
292 * @param offset
293 * The initial offset
294 *
295 * @param count
296 * The length
297 *
298 * @throws IndexOutOfBoundsException
299 * If {@code offset} is negative, {@code count} is negative, or
300 * {@code offset} is greater than {@code value.length - count}
301 */
302 public String(char[] value, int offset, int count) {
303 this(value, offset, count, rangeCheck(value, offset, count));
304 }
305
306 private static Void rangeCheck(char[] value, int offset, int count) {
307 checkBoundsOffCount(offset, count, value.length);
308 return null;
309 }
310
311 /**
312 * Allocates a new {@code String} that contains characters from a subarray
313 * of the <a href="Character.html#unicode">Unicode code point</a> array
314 * argument. The {@code offset} argument is the index of the first code
315 * point of the subarray and the {@code count} argument specifies the
316 * length of the subarray. The contents of the subarray are converted to
317 * {@code char}s; subsequent modification of the {@code int} array does not
318 * affect the newly created string.
319 *
320 * @param codePoints
321 * Array that is the source of Unicode code points
322 *
323 * @param offset
324 * The initial offset
325 *
326 * @param count
327 * The length
328 *
329 * @throws IllegalArgumentException
330 * If any invalid Unicode code point is found in {@code
331 * codePoints}
332 *
333 * @throws IndexOutOfBoundsException
334 * If {@code offset} is negative, {@code count} is negative, or
335 * {@code offset} is greater than {@code codePoints.length - count}
336 *
337 * @since 1.5
338 */
339 public String(int[] codePoints, int offset, int count) {
340 checkBoundsOffCount(offset, count, codePoints.length);
341 if (count == 0) {
342 this.value = "".value;
343 this.coder = "".coder;
344 return;
345 }
346 if (COMPACT_STRINGS) {
347 byte[] val = StringLatin1.toBytes(codePoints, offset, count);
348 if (val != null) {
349 this.coder = LATIN1;
350 this.value = val;
351 return;
352 }
353 }
354 this.coder = UTF16;
355 this.value = StringUTF16.toBytes(codePoints, offset, count);
356 }
357
358 /**
359 * Allocates a new {@code String} constructed from a subarray of an array
360 * of 8-bit integer values.
361 *
362 * <p> The {@code offset} argument is the index of the first byte of the
363 * subarray, and the {@code count} argument specifies the length of the
364 * subarray.
365 *
366 * <p> Each {@code byte} in the subarray is converted to a {@code char} as
367 * specified in the {@link #String(byte[],int) String(byte[],int)} constructor.
368 *
369 * @deprecated This method does not properly convert bytes into characters.
370 * As of JDK 1.1, the preferred way to do this is via the
371 * {@code String} constructors that take a {@link Charset}, charset name,
372 * or that use the {@link Charset#defaultCharset() default charset}.
373 *
374 * @param ascii
375 * The bytes to be converted to characters
376 *
377 * @param hibyte
378 * The top 8 bits of each 16-bit Unicode code unit
379 *
380 * @param offset
381 * The initial offset
382 * @param count
383 * The length
384 *
385 * @throws IndexOutOfBoundsException
386 * If {@code offset} is negative, {@code count} is negative, or
387 * {@code offset} is greater than {@code ascii.length - count}
388 *
389 * @see #String(byte[], int)
390 * @see #String(byte[], int, int, java.lang.String)
391 * @see #String(byte[], int, int, java.nio.charset.Charset)
392 * @see #String(byte[], int, int)
393 * @see #String(byte[], java.lang.String)
394 * @see #String(byte[], java.nio.charset.Charset)
395 * @see #String(byte[])
396 */
397 @Deprecated(since="1.1")
398 public String(byte[] ascii, int hibyte, int offset, int count) {
399 checkBoundsOffCount(offset, count, ascii.length);
400 if (count == 0) {
401 this.value = "".value;
402 this.coder = "".coder;
403 return;
404 }
405 if (COMPACT_STRINGS && (byte)hibyte == 0) {
406 this.value = Arrays.copyOfRange(ascii, offset, offset + count);
407 this.coder = LATIN1;
408 } else {
409 hibyte <<= 8;
410 byte[] val = StringUTF16.newBytesFor(count);
411 for (int i = 0; i < count; i++) {
412 StringUTF16.putChar(val, i, hibyte | (ascii[offset++] & 0xff));
413 }
414 this.value = val;
415 this.coder = UTF16;
416 }
417 }
418
419 /**
420 * Allocates a new {@code String} containing characters constructed from
421 * an array of 8-bit integer values. Each character <i>c</i> in the
422 * resulting string is constructed from the corresponding component
423 * <i>b</i> in the byte array such that:
424 *
425 * <blockquote><pre>
426 * <b><i>c</i></b> == (char)(((hibyte & 0xff) << 8)
427 * | (<b><i>b</i></b> & 0xff))
428 * </pre></blockquote>
429 *
430 * @deprecated This method does not properly convert bytes into
431 * characters. As of JDK 1.1, the preferred way to do this is via the
432 * {@code String} constructors that take a {@link Charset}, charset name,
433 * or that use the {@link Charset#defaultCharset() default charset}.
434 *
435 * @param ascii
436 * The bytes to be converted to characters
437 *
438 * @param hibyte
439 * The top 8 bits of each 16-bit Unicode code unit
440 *
441 * @see #String(byte[], int, int, java.lang.String)
442 * @see #String(byte[], int, int, java.nio.charset.Charset)
443 * @see #String(byte[], int, int)
444 * @see #String(byte[], java.lang.String)
445 * @see #String(byte[], java.nio.charset.Charset)
446 * @see #String(byte[])
447 */
448 @Deprecated(since="1.1")
449 public String(byte[] ascii, int hibyte) {
450 this(ascii, hibyte, 0, ascii.length);
451 }
452
453 /**
454 * Constructs a new {@code String} by decoding the specified subarray of
455 * bytes using the specified charset. The length of the new {@code String}
456 * is a function of the charset, and hence may not be equal to the length
457 * of the subarray.
458 *
459 * <p> The behavior of this constructor when the given bytes are not valid
460 * in the given charset is unspecified. The {@link
461 * java.nio.charset.CharsetDecoder} class should be used when more control
462 * over the decoding process is required.
463 *
464 * @param bytes
465 * The bytes to be decoded into characters
466 *
467 * @param offset
468 * The index of the first byte to decode
469 *
470 * @param length
471 * The number of bytes to decode
472 *
473 * @param charsetName
474 * The name of a supported {@linkplain java.nio.charset.Charset
475 * charset}
476 *
477 * @throws UnsupportedEncodingException
478 * If the named charset is not supported
479 *
480 * @throws IndexOutOfBoundsException
481 * If {@code offset} is negative, {@code length} is negative, or
482 * {@code offset} is greater than {@code bytes.length - length}
483 *
484 * @since 1.1
485 */
486 public String(byte[] bytes, int offset, int length, String charsetName)
487 throws UnsupportedEncodingException {
488 this(lookupCharset(charsetName), bytes, checkBoundsOffCount(offset, length, bytes.length), length);
489 }
490
491 /**
492 * Constructs a new {@code String} by decoding the specified subarray of
493 * bytes using the specified {@linkplain java.nio.charset.Charset charset}.
494 * The length of the new {@code String} is a function of the charset, and
495 * hence may not be equal to the length of the subarray.
496 *
497 * <p> This method always replaces malformed-input and unmappable-character
498 * sequences with this charset's default replacement string. The {@link
499 * java.nio.charset.CharsetDecoder} class should be used when more control
500 * over the decoding process is required.
501 *
502 * @param bytes
503 * The bytes to be decoded into characters
504 *
505 * @param offset
506 * The index of the first byte to decode
507 *
508 * @param length
509 * The number of bytes to decode
510 *
511 * @param charset
512 * The {@linkplain java.nio.charset.Charset charset} to be used to
513 * decode the {@code bytes}
514 *
515 * @throws IndexOutOfBoundsException
516 * If {@code offset} is negative, {@code length} is negative, or
517 * {@code offset} is greater than {@code bytes.length - length}
518 *
519 * @since 1.6
520 */
521 public String(byte[] bytes, int offset, int length, Charset charset) {
522 this(Objects.requireNonNull(charset), bytes, checkBoundsOffCount(offset, length, bytes.length), length);
523 }
524
525 /**
526 * This method does not do any precondition checks on its arguments.
527 * <p>
528 * Important: parameter order of this method is deliberately changed in order to
529 * disambiguate it against other similar methods of this class.
530 */
531 @SuppressWarnings("removal")
532 private String(Charset charset, byte[] bytes, int offset, int length) {
533 if (length == 0) {
534 this.value = "".value;
535 this.coder = "".coder;
536 } else if (charset == UTF_8.INSTANCE) {
537 if (COMPACT_STRINGS) {
538 int dp = StringCoding.countPositives(bytes, offset, length);
539 if (dp == length) {
540 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
541 this.coder = LATIN1;
542 return;
543 }
544 int sl = offset + length;
545 byte[] dst = new byte[length];
546 if (dp > 0) {
547 System.arraycopy(bytes, offset, dst, 0, dp);
548 offset += dp;
549 }
550 while (offset < sl) {
551 int b1 = bytes[offset++];
552 if (b1 >= 0) {
553 dst[dp++] = (byte)b1;
554 continue;
555 }
556 if ((b1 & 0xfe) == 0xc2 && offset < sl) { // b1 either 0xc2 or 0xc3
557 int b2 = bytes[offset];
558 if (b2 < -64) { // continuation bytes are always negative values in the range -128 to -65
559 dst[dp++] = (byte)decode2(b1, b2);
560 offset++;
561 continue;
562 }
563 }
564 // anything not a latin1, including the REPL
565 // we have to go with the utf16
566 offset--;
567 break;
568 }
569 if (offset == sl) {
570 if (dp != dst.length) {
571 dst = Arrays.copyOf(dst, dp);
572 }
573 this.value = dst;
574 this.coder = LATIN1;
575 return;
576 }
577 byte[] buf = new byte[length << 1];
578 StringLatin1.inflate(dst, 0, buf, 0, dp);
579 dst = buf;
580 dp = decodeUTF8_UTF16(bytes, offset, sl, dst, dp, true);
581 if (dp != length) {
582 dst = Arrays.copyOf(dst, dp << 1);
583 }
584 this.value = dst;
585 this.coder = UTF16;
586 } else { // !COMPACT_STRINGS
587 byte[] dst = new byte[length << 1];
588 int dp = decodeUTF8_UTF16(bytes, offset, offset + length, dst, 0, true);
589 if (dp != length) {
590 dst = Arrays.copyOf(dst, dp << 1);
591 }
592 this.value = dst;
593 this.coder = UTF16;
594 }
595 } else if (charset == ISO_8859_1.INSTANCE) {
596 if (COMPACT_STRINGS) {
597 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
598 this.coder = LATIN1;
599 } else {
600 this.value = StringLatin1.inflate(bytes, offset, length);
601 this.coder = UTF16;
602 }
603 } else if (charset == US_ASCII.INSTANCE) {
604 if (COMPACT_STRINGS && !StringCoding.hasNegatives(bytes, offset, length)) {
605 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
606 this.coder = LATIN1;
607 } else {
608 byte[] dst = new byte[length << 1];
609 int dp = 0;
610 while (dp < length) {
611 int b = bytes[offset++];
612 StringUTF16.putChar(dst, dp++, (b >= 0) ? (char) b : REPL);
613 }
614 this.value = dst;
615 this.coder = UTF16;
616 }
617 } else {
618 // (1)We never cache the "external" cs, the only benefit of creating
619 // an additional StringDe/Encoder object to wrap it is to share the
620 // de/encode() method. These SD/E objects are short-lived, the young-gen
621 // gc should be able to take care of them well. But the best approach
622 // is still not to generate them if not really necessary.
623 // (2)The defensive copy of the input byte/char[] has a big performance
624 // impact, as well as the outgoing result byte/char[]. Need to do the
625 // optimization check of (sm==null && classLoader0==null) for both.
626 CharsetDecoder cd = charset.newDecoder();
627 // ArrayDecoder fastpaths
628 if (cd instanceof ArrayDecoder ad) {
629 // ascii
630 if (ad.isASCIICompatible() && !StringCoding.hasNegatives(bytes, offset, length)) {
631 if (COMPACT_STRINGS) {
632 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
633 this.coder = LATIN1;
634 return;
635 }
636 this.value = StringLatin1.inflate(bytes, offset, length);
637 this.coder = UTF16;
638 return;
639 }
640
641 // fastpath for always Latin1 decodable single byte
642 if (COMPACT_STRINGS && ad.isLatin1Decodable()) {
643 byte[] dst = new byte[length];
644 ad.decodeToLatin1(bytes, offset, length, dst);
645 this.value = dst;
646 this.coder = LATIN1;
647 return;
648 }
649
650 int en = scale(length, cd.maxCharsPerByte());
651 cd.onMalformedInput(CodingErrorAction.REPLACE)
652 .onUnmappableCharacter(CodingErrorAction.REPLACE);
653 char[] ca = new char[en];
654 int clen = ad.decode(bytes, offset, length, ca);
655 if (COMPACT_STRINGS) {
656 byte[] bs = StringUTF16.compress(ca, 0, clen);
657 if (bs != null) {
658 value = bs;
659 coder = LATIN1;
660 return;
661 }
662 }
663 coder = UTF16;
664 value = StringUTF16.toBytes(ca, 0, clen);
665 return;
666 }
667
668 // decode using CharsetDecoder
669 int en = scale(length, cd.maxCharsPerByte());
670 cd.onMalformedInput(CodingErrorAction.REPLACE)
671 .onUnmappableCharacter(CodingErrorAction.REPLACE);
672 char[] ca = new char[en];
673 if (charset.getClass().getClassLoader0() != null &&
674 System.getSecurityManager() != null) {
675 bytes = Arrays.copyOfRange(bytes, offset, offset + length);
676 offset = 0;
677 }
678
679 int caLen;
680 try {
681 caLen = decodeWithDecoder(cd, ca, bytes, offset, length);
682 } catch (CharacterCodingException x) {
683 // Substitution is enabled, so this shouldn't happen
684 throw new Error(x);
685 }
686 if (COMPACT_STRINGS) {
687 byte[] bs = StringUTF16.compress(ca, 0, caLen);
688 if (bs != null) {
689 value = bs;
690 coder = LATIN1;
691 return;
692 }
693 }
694 coder = UTF16;
695 value = StringUTF16.toBytes(ca, 0, caLen);
696 }
697 }
698
699 /*
700 * Throws iae, instead of replacing, if malformed or unmappable.
701 *
702 * @param noShare
703 * {@code true} if the resulting string MUST NOT share the byte array,
704 * {@code false} if the byte array can be exclusively used to construct
705 * the string and is not modified or used for any other purpose.
706 */
707 static String newStringUTF8NoRepl(byte[] bytes, int offset, int length, boolean noShare) {
708 checkBoundsOffCount(offset, length, bytes.length);
709 if (length == 0) {
710 return "";
711 }
712 int dp;
713 byte[] dst;
714 if (COMPACT_STRINGS) {
715 dp = StringCoding.countPositives(bytes, offset, length);
716 int sl = offset + length;
717 if (dp == length) {
718 if (noShare || length != bytes.length) {
719 return new String(Arrays.copyOfRange(bytes, offset, offset + length), LATIN1);
720 } else {
721 return new String(bytes, LATIN1);
722 }
723 }
724 dst = new byte[length];
725 System.arraycopy(bytes, offset, dst, 0, dp);
726 offset += dp;
727 while (offset < sl) {
728 int b1 = bytes[offset++];
729 if (b1 >= 0) {
730 dst[dp++] = (byte)b1;
731 continue;
732 }
733 if ((b1 & 0xfe) == 0xc2 && offset < sl) { // b1 either 0xc2 or 0xc3
734 int b2 = bytes[offset];
735 if (b2 < -64) { // continuation bytes are always negative values in the range -128 to -65
736 dst[dp++] = (byte)decode2(b1, b2);
737 offset++;
738 continue;
739 }
740 }
741 // anything not a latin1, including the REPL
742 // we have to go with the utf16
743 offset--;
744 break;
745 }
746 if (offset == sl) {
747 if (dp != dst.length) {
748 dst = Arrays.copyOf(dst, dp);
749 }
750 return new String(dst, LATIN1);
751 }
752 if (dp == 0) {
753 dst = new byte[length << 1];
754 } else {
755 byte[] buf = new byte[length << 1];
756 StringLatin1.inflate(dst, 0, buf, 0, dp);
757 dst = buf;
758 }
759 dp = decodeUTF8_UTF16(bytes, offset, sl, dst, dp, false);
760 } else { // !COMPACT_STRINGS
761 dst = new byte[length << 1];
762 dp = decodeUTF8_UTF16(bytes, offset, offset + length, dst, 0, false);
763 }
764 if (dp != length) {
765 dst = Arrays.copyOf(dst, dp << 1);
766 }
767 return new String(dst, UTF16);
768 }
769
770 static String newStringNoRepl(byte[] src, Charset cs) throws CharacterCodingException {
771 try {
772 return newStringNoRepl1(src, cs);
773 } catch (IllegalArgumentException e) {
774 //newStringNoRepl1 throws IAE with MalformedInputException or CCE as the cause
775 Throwable cause = e.getCause();
776 if (cause instanceof MalformedInputException mie) {
777 throw mie;
778 }
779 throw (CharacterCodingException)cause;
780 }
781 }
782
783 @SuppressWarnings("removal")
784 private static String newStringNoRepl1(byte[] src, Charset cs) {
785 int len = src.length;
786 if (len == 0) {
787 return "";
788 }
789 if (cs == UTF_8.INSTANCE) {
790 return newStringUTF8NoRepl(src, 0, src.length, false);
791 }
792 if (cs == ISO_8859_1.INSTANCE) {
793 if (COMPACT_STRINGS)
794 return new String(src, LATIN1);
795 return new String(StringLatin1.inflate(src, 0, src.length), UTF16);
796 }
797 if (cs == US_ASCII.INSTANCE) {
798 if (!StringCoding.hasNegatives(src, 0, src.length)) {
799 if (COMPACT_STRINGS)
800 return new String(src, LATIN1);
801 return new String(StringLatin1.inflate(src, 0, src.length), UTF16);
802 } else {
803 throwMalformed(src);
804 }
805 }
806
807 CharsetDecoder cd = cs.newDecoder();
808 // ascii fastpath
809 if (cd instanceof ArrayDecoder ad &&
810 ad.isASCIICompatible() &&
811 !StringCoding.hasNegatives(src, 0, src.length)) {
812 if (COMPACT_STRINGS)
813 return new String(src, LATIN1);
814 return new String(src, 0, src.length, ISO_8859_1.INSTANCE);
815 }
816 int en = scale(len, cd.maxCharsPerByte());
817 char[] ca = new char[en];
818 if (cs.getClass().getClassLoader0() != null &&
819 System.getSecurityManager() != null) {
820 src = Arrays.copyOf(src, len);
821 }
822 int caLen;
823 try {
824 caLen = decodeWithDecoder(cd, ca, src, 0, src.length);
825 } catch (CharacterCodingException x) {
826 // throw via IAE
827 throw new IllegalArgumentException(x);
828 }
829 if (COMPACT_STRINGS) {
830 byte[] bs = StringUTF16.compress(ca, 0, caLen);
831 if (bs != null) {
832 return new String(bs, LATIN1);
833 }
834 }
835 return new String(StringUTF16.toBytes(ca, 0, caLen), UTF16);
836 }
837
838 private static final char REPL = '\ufffd';
839
840 // Trim the given byte array to the given length
841 @SuppressWarnings("removal")
842 private static byte[] safeTrim(byte[] ba, int len, boolean isTrusted) {
843 if (len == ba.length && (isTrusted || System.getSecurityManager() == null)) {
844 return ba;
845 } else {
846 return Arrays.copyOf(ba, len);
847 }
848 }
849
850 private static int scale(int len, float expansionFactor) {
851 // We need to perform double, not float, arithmetic; otherwise
852 // we lose low order bits when len is larger than 2**24.
853 return (int)(len * (double)expansionFactor);
854 }
855
856 private static Charset lookupCharset(String csn) throws UnsupportedEncodingException {
857 Objects.requireNonNull(csn);
858 try {
859 return Charset.forName(csn);
860 } catch (UnsupportedCharsetException | IllegalCharsetNameException x) {
861 throw new UnsupportedEncodingException(csn);
862 }
863 }
864
865 private static byte[] encode(Charset cs, byte coder, byte[] val) {
866 if (cs == UTF_8.INSTANCE) {
867 return encodeUTF8(coder, val, true);
868 }
869 if (cs == ISO_8859_1.INSTANCE) {
870 return encode8859_1(coder, val);
871 }
872 if (cs == US_ASCII.INSTANCE) {
873 return encodeASCII(coder, val);
874 }
875 return encodeWithEncoder(cs, coder, val, true);
876 }
877
878 private static byte[] encodeWithEncoder(Charset cs, byte coder, byte[] val, boolean doReplace) {
879 CharsetEncoder ce = cs.newEncoder();
880 int len = val.length >> coder; // assume LATIN1=0/UTF16=1;
881 int en = scale(len, ce.maxBytesPerChar());
882 // fastpath with ArrayEncoder implies `doReplace`.
883 if (doReplace && ce instanceof ArrayEncoder ae) {
884 // fastpath for ascii compatible
885 if (coder == LATIN1 &&
886 ae.isASCIICompatible() &&
887 !StringCoding.hasNegatives(val, 0, val.length)) {
888 return val.clone();
889 }
890 byte[] ba = new byte[en];
891 if (len == 0) {
892 return ba;
893 }
894
895 int blen = (coder == LATIN1) ? ae.encodeFromLatin1(val, 0, len, ba)
896 : ae.encodeFromUTF16(val, 0, len, ba);
897 if (blen != -1) {
898 return safeTrim(ba, blen, true);
899 }
900 }
901
902 byte[] ba = new byte[en];
903 if (len == 0) {
904 return ba;
905 }
906 if (doReplace) {
907 ce.onMalformedInput(CodingErrorAction.REPLACE)
908 .onUnmappableCharacter(CodingErrorAction.REPLACE);
909 }
910 char[] ca = (coder == LATIN1 ) ? StringLatin1.toChars(val)
911 : StringUTF16.toChars(val);
912 ByteBuffer bb = ByteBuffer.wrap(ba);
913 CharBuffer cb = CharBuffer.wrap(ca, 0, len);
914 try {
915 CoderResult cr = ce.encode(cb, bb, true);
916 if (!cr.isUnderflow())
917 cr.throwException();
918 cr = ce.flush(bb);
919 if (!cr.isUnderflow())
920 cr.throwException();
921 } catch (CharacterCodingException x) {
922 if (!doReplace) {
923 throw new IllegalArgumentException(x);
924 } else {
925 throw new Error(x);
926 }
927 }
928 return safeTrim(ba, bb.position(), cs.getClass().getClassLoader0() == null);
929 }
930
931 /*
932 * Throws iae, instead of replacing, if unmappable.
933 */
934 static byte[] getBytesUTF8NoRepl(String s) {
935 return encodeUTF8(s.coder(), s.value(), false);
936 }
937
938 private static boolean isASCII(byte[] src) {
939 return !StringCoding.hasNegatives(src, 0, src.length);
940 }
941
942 /*
943 * Throws CCE, instead of replacing, if unmappable.
944 */
945 static byte[] getBytesNoRepl(String s, Charset cs) throws CharacterCodingException {
946 try {
947 return getBytesNoRepl1(s, cs);
948 } catch (IllegalArgumentException e) {
949 //getBytesNoRepl1 throws IAE with UnmappableCharacterException or CCE as the cause
950 Throwable cause = e.getCause();
951 if (cause instanceof UnmappableCharacterException) {
952 throw (UnmappableCharacterException)cause;
953 }
954 throw (CharacterCodingException)cause;
955 }
956 }
957
958 private static byte[] getBytesNoRepl1(String s, Charset cs) {
959 byte[] val = s.value();
960 byte coder = s.coder();
961 if (cs == UTF_8.INSTANCE) {
962 if (coder == LATIN1 && isASCII(val)) {
963 return val;
964 }
965 return encodeUTF8(coder, val, false);
966 }
967 if (cs == ISO_8859_1.INSTANCE) {
968 if (coder == LATIN1) {
969 return val;
970 }
971 return encode8859_1(coder, val, false);
972 }
973 if (cs == US_ASCII.INSTANCE) {
974 if (coder == LATIN1) {
975 if (isASCII(val)) {
976 return val;
977 } else {
978 throwUnmappable(val);
979 }
980 }
981 }
982 return encodeWithEncoder(cs, coder, val, false);
983 }
984
985 private static byte[] encodeASCII(byte coder, byte[] val) {
986 if (coder == LATIN1) {
987 int positives = StringCoding.countPositives(val, 0, val.length);
988 byte[] dst = val.clone();
989 if (positives < dst.length) {
990 replaceNegatives(dst, positives);
991 }
992 return dst;
993 }
994 int len = val.length >> 1;
995 byte[] dst = new byte[len];
996 int dp = 0;
997 for (int i = 0; i < len; i++) {
998 char c = StringUTF16.getChar(val, i);
999 if (c < 0x80) {
1000 dst[dp++] = (byte)c;
1001 continue;
1002 }
1003 if (Character.isHighSurrogate(c) && i + 1 < len &&
1004 Character.isLowSurrogate(StringUTF16.getChar(val, i + 1))) {
1005 i++;
1006 }
1007 dst[dp++] = '?';
1008 }
1009 if (len == dp) {
1010 return dst;
1011 }
1012 return Arrays.copyOf(dst, dp);
1013 }
1014
1015 private static void replaceNegatives(byte[] val, int fromIndex) {
1016 for (int i = fromIndex; i < val.length; i++) {
1017 if (val[i] < 0) {
1018 val[i] = '?';
1019 }
1020 }
1021 }
1022
1023 private static byte[] encode8859_1(byte coder, byte[] val) {
1024 return encode8859_1(coder, val, true);
1025 }
1026
1027 private static byte[] encode8859_1(byte coder, byte[] val, boolean doReplace) {
1028 if (coder == LATIN1) {
1029 return val.clone();
1030 }
1031 int len = val.length >> 1;
1032 byte[] dst = new byte[len];
1033 int dp = 0;
1034 int sp = 0;
1035 int sl = len;
1036 while (sp < sl) {
1037 int ret = StringCoding.implEncodeISOArray(val, sp, dst, dp, len);
1038 sp = sp + ret;
1039 dp = dp + ret;
1040 if (ret != len) {
1041 if (!doReplace) {
1042 throwUnmappable(sp);
1043 }
1044 char c = StringUTF16.getChar(val, sp++);
1045 if (Character.isHighSurrogate(c) && sp < sl &&
1046 Character.isLowSurrogate(StringUTF16.getChar(val, sp))) {
1047 sp++;
1048 }
1049 dst[dp++] = '?';
1050 len = sl - sp;
1051 }
1052 }
1053 if (dp == dst.length) {
1054 return dst;
1055 }
1056 return Arrays.copyOf(dst, dp);
1057 }
1058
1059 //////////////////////////////// utf8 ////////////////////////////////////
1060
1061 /**
1062 * Decodes ASCII from the source byte array into the destination
1063 * char array. Used via JavaLangAccess from UTF_8 and other charset
1064 * decoders.
1065 *
1066 * @return the number of bytes successfully decoded, at most len
1067 */
1068 /* package-private */
1069 static int decodeASCII(byte[] sa, int sp, char[] da, int dp, int len) {
1070 int count = StringCoding.countPositives(sa, sp, len);
1071 while (count < len) {
1072 if (sa[sp + count] < 0) {
1073 break;
1074 }
1075 count++;
1076 }
1077 StringLatin1.inflate(sa, sp, da, dp, count);
1078 return count;
1079 }
1080
1081 private static boolean isNotContinuation(int b) {
1082 return (b & 0xc0) != 0x80;
1083 }
1084
1085 private static boolean isMalformed3(int b1, int b2, int b3) {
1086 return (b1 == (byte)0xe0 && (b2 & 0xe0) == 0x80) ||
1087 (b2 & 0xc0) != 0x80 || (b3 & 0xc0) != 0x80;
1088 }
1089
1090 private static boolean isMalformed3_2(int b1, int b2) {
1091 return (b1 == (byte)0xe0 && (b2 & 0xe0) == 0x80) ||
1092 (b2 & 0xc0) != 0x80;
1093 }
1094
1095 private static boolean isMalformed4(int b2, int b3, int b4) {
1096 return (b2 & 0xc0) != 0x80 || (b3 & 0xc0) != 0x80 ||
1097 (b4 & 0xc0) != 0x80;
1098 }
1099
1100 private static boolean isMalformed4_2(int b1, int b2) {
1101 return (b1 == 0xf0 && (b2 < 0x90 || b2 > 0xbf)) ||
1102 (b1 == 0xf4 && (b2 & 0xf0) != 0x80) ||
1103 (b2 & 0xc0) != 0x80;
1104 }
1105
1106 private static boolean isMalformed4_3(int b3) {
1107 return (b3 & 0xc0) != 0x80;
1108 }
1109
1110 private static char decode2(int b1, int b2) {
1111 return (char)(((b1 << 6) ^ b2) ^
1112 (((byte) 0xC0 << 6) ^
1113 ((byte) 0x80 << 0)));
1114 }
1115
1116 private static char decode3(int b1, int b2, int b3) {
1117 return (char)((b1 << 12) ^
1118 (b2 << 6) ^
1119 (b3 ^
1120 (((byte) 0xE0 << 12) ^
1121 ((byte) 0x80 << 6) ^
1122 ((byte) 0x80 << 0))));
1123 }
1124
1125 private static int decode4(int b1, int b2, int b3, int b4) {
1126 return ((b1 << 18) ^
1127 (b2 << 12) ^
1128 (b3 << 6) ^
1129 (b4 ^
1130 (((byte) 0xF0 << 18) ^
1131 ((byte) 0x80 << 12) ^
1132 ((byte) 0x80 << 6) ^
1133 ((byte) 0x80 << 0))));
1134 }
1135
1136 private static int decodeUTF8_UTF16(byte[] src, int sp, int sl, byte[] dst, int dp, boolean doReplace) {
1137 while (sp < sl) {
1138 int b1 = src[sp++];
1139 if (b1 >= 0) {
1140 StringUTF16.putChar(dst, dp++, (char) b1);
1141 } else if ((b1 >> 5) == -2 && (b1 & 0x1e) != 0) {
1142 if (sp < sl) {
1143 int b2 = src[sp++];
1144 if (isNotContinuation(b2)) {
1145 if (!doReplace) {
1146 throwMalformed(sp - 1, 1);
1147 }
1148 StringUTF16.putChar(dst, dp++, REPL);
1149 sp--;
1150 } else {
1151 StringUTF16.putChar(dst, dp++, decode2(b1, b2));
1152 }
1153 continue;
1154 }
1155 if (!doReplace) {
1156 throwMalformed(sp, 1); // underflow()
1157 }
1158 StringUTF16.putChar(dst, dp++, REPL);
1159 break;
1160 } else if ((b1 >> 4) == -2) {
1161 if (sp + 1 < sl) {
1162 int b2 = src[sp++];
1163 int b3 = src[sp++];
1164 if (isMalformed3(b1, b2, b3)) {
1165 if (!doReplace) {
1166 throwMalformed(sp - 3, 3);
1167 }
1168 StringUTF16.putChar(dst, dp++, REPL);
1169 sp -= 3;
1170 sp += malformed3(src, sp);
1171 } else {
1172 char c = decode3(b1, b2, b3);
1173 if (Character.isSurrogate(c)) {
1174 if (!doReplace) {
1175 throwMalformed(sp - 3, 3);
1176 }
1177 StringUTF16.putChar(dst, dp++, REPL);
1178 } else {
1179 StringUTF16.putChar(dst, dp++, c);
1180 }
1181 }
1182 continue;
1183 }
1184 if (sp < sl && isMalformed3_2(b1, src[sp])) {
1185 if (!doReplace) {
1186 throwMalformed(sp - 1, 2);
1187 }
1188 StringUTF16.putChar(dst, dp++, REPL);
1189 continue;
1190 }
1191 if (!doReplace) {
1192 throwMalformed(sp, 1);
1193 }
1194 StringUTF16.putChar(dst, dp++, REPL);
1195 break;
1196 } else if ((b1 >> 3) == -2) {
1197 if (sp + 2 < sl) {
1198 int b2 = src[sp++];
1199 int b3 = src[sp++];
1200 int b4 = src[sp++];
1201 int uc = decode4(b1, b2, b3, b4);
1202 if (isMalformed4(b2, b3, b4) ||
1203 !Character.isSupplementaryCodePoint(uc)) { // shortest form check
1204 if (!doReplace) {
1205 throwMalformed(sp - 4, 4);
1206 }
1207 StringUTF16.putChar(dst, dp++, REPL);
1208 sp -= 4;
1209 sp += malformed4(src, sp);
1210 } else {
1211 StringUTF16.putChar(dst, dp++, Character.highSurrogate(uc));
1212 StringUTF16.putChar(dst, dp++, Character.lowSurrogate(uc));
1213 }
1214 continue;
1215 }
1216 b1 &= 0xff;
1217 if (b1 > 0xf4 || sp < sl && isMalformed4_2(b1, src[sp] & 0xff)) {
1218 if (!doReplace) {
1219 throwMalformed(sp - 1, 1); // or 2
1220 }
1221 StringUTF16.putChar(dst, dp++, REPL);
1222 continue;
1223 }
1224 if (!doReplace) {
1225 throwMalformed(sp - 1, 1);
1226 }
1227 sp++;
1228 StringUTF16.putChar(dst, dp++, REPL);
1229 if (sp < sl && isMalformed4_3(src[sp])) {
1230 continue;
1231 }
1232 break;
1233 } else {
1234 if (!doReplace) {
1235 throwMalformed(sp - 1, 1);
1236 }
1237 StringUTF16.putChar(dst, dp++, REPL);
1238 }
1239 }
1240 return dp;
1241 }
1242
1243 private static int decodeWithDecoder(CharsetDecoder cd, char[] dst, byte[] src, int offset, int length)
1244 throws CharacterCodingException {
1245 ByteBuffer bb = ByteBuffer.wrap(src, offset, length);
1246 CharBuffer cb = CharBuffer.wrap(dst, 0, dst.length);
1247 CoderResult cr = cd.decode(bb, cb, true);
1248 if (!cr.isUnderflow())
1249 cr.throwException();
1250 cr = cd.flush(cb);
1251 if (!cr.isUnderflow())
1252 cr.throwException();
1253 return cb.position();
1254 }
1255
1256 private static int malformed3(byte[] src, int sp) {
1257 int b1 = src[sp++];
1258 int b2 = src[sp]; // no need to lookup b3
1259 return ((b1 == (byte)0xe0 && (b2 & 0xe0) == 0x80) ||
1260 isNotContinuation(b2)) ? 1 : 2;
1261 }
1262
1263 private static int malformed4(byte[] src, int sp) {
1264 // we don't care the speed here
1265 int b1 = src[sp++] & 0xff;
1266 int b2 = src[sp++] & 0xff;
1267 if (b1 > 0xf4 ||
1268 (b1 == 0xf0 && (b2 < 0x90 || b2 > 0xbf)) ||
1269 (b1 == 0xf4 && (b2 & 0xf0) != 0x80) ||
1270 isNotContinuation(b2))
1271 return 1;
1272 if (isNotContinuation(src[sp]))
1273 return 2;
1274 return 3;
1275 }
1276
1277 private static void throwMalformed(int off, int nb) {
1278 String msg = "malformed input off : " + off + ", length : " + nb;
1279 throw new IllegalArgumentException(msg, new MalformedInputException(nb));
1280 }
1281
1282 private static void throwMalformed(byte[] val) {
1283 int dp = StringCoding.countPositives(val, 0, val.length);
1284 throwMalformed(dp, 1);
1285 }
1286
1287 private static void throwUnmappable(int off) {
1288 String msg = "malformed input off : " + off + ", length : 1";
1289 throw new IllegalArgumentException(msg, new UnmappableCharacterException(1));
1290 }
1291
1292 private static void throwUnmappable(byte[] val) {
1293 int dp = StringCoding.countPositives(val, 0, val.length);
1294 throwUnmappable(dp);
1295 }
1296
1297 private static byte[] encodeUTF8(byte coder, byte[] val, boolean doReplace) {
1298 if (coder == UTF16) {
1299 return encodeUTF8_UTF16(val, doReplace);
1300 }
1301
1302 if (!StringCoding.hasNegatives(val, 0, val.length)) {
1303 return val.clone();
1304 }
1305
1306 int dp = 0;
1307 byte[] dst = new byte[val.length << 1];
1308 for (byte c : val) {
1309 if (c < 0) {
1310 dst[dp++] = (byte) (0xc0 | ((c & 0xff) >> 6));
1311 dst[dp++] = (byte) (0x80 | (c & 0x3f));
1312 } else {
1313 dst[dp++] = c;
1314 }
1315 }
1316 if (dp == dst.length) {
1317 return dst;
1318 }
1319 return Arrays.copyOf(dst, dp);
1320 }
1321
1322 private static byte[] encodeUTF8_UTF16(byte[] val, boolean doReplace) {
1323 int dp = 0;
1324 int sp = 0;
1325 int sl = val.length >> 1;
1326 byte[] dst = new byte[sl * 3];
1327 while (sp < sl) {
1328 // ascii fast loop;
1329 char c = StringUTF16.getChar(val, sp);
1330 if (c >= '\u0080') {
1331 break;
1332 }
1333 dst[dp++] = (byte)c;
1334 sp++;
1335 }
1336 while (sp < sl) {
1337 char c = StringUTF16.getChar(val, sp++);
1338 if (c < 0x80) {
1339 dst[dp++] = (byte)c;
1340 } else if (c < 0x800) {
1341 dst[dp++] = (byte)(0xc0 | (c >> 6));
1342 dst[dp++] = (byte)(0x80 | (c & 0x3f));
1343 } else if (Character.isSurrogate(c)) {
1344 int uc = -1;
1345 char c2;
1346 if (Character.isHighSurrogate(c) && sp < sl &&
1347 Character.isLowSurrogate(c2 = StringUTF16.getChar(val, sp))) {
1348 uc = Character.toCodePoint(c, c2);
1349 }
1350 if (uc < 0) {
1351 if (doReplace) {
1352 dst[dp++] = '?';
1353 } else {
1354 throwUnmappable(sp - 1);
1355 }
1356 } else {
1357 dst[dp++] = (byte)(0xf0 | ((uc >> 18)));
1358 dst[dp++] = (byte)(0x80 | ((uc >> 12) & 0x3f));
1359 dst[dp++] = (byte)(0x80 | ((uc >> 6) & 0x3f));
1360 dst[dp++] = (byte)(0x80 | (uc & 0x3f));
1361 sp++; // 2 chars
1362 }
1363 } else {
1364 // 3 bytes, 16 bits
1365 dst[dp++] = (byte)(0xe0 | ((c >> 12)));
1366 dst[dp++] = (byte)(0x80 | ((c >> 6) & 0x3f));
1367 dst[dp++] = (byte)(0x80 | (c & 0x3f));
1368 }
1369 }
1370 if (dp == dst.length) {
1371 return dst;
1372 }
1373 return Arrays.copyOf(dst, dp);
1374 }
1375
1376 /**
1377 * Constructs a new {@code String} by decoding the specified array of bytes
1378 * using the specified {@linkplain java.nio.charset.Charset charset}. The
1379 * length of the new {@code String} is a function of the charset, and hence
1380 * may not be equal to the length of the byte array.
1381 *
1382 * <p> The behavior of this constructor when the given bytes are not valid
1383 * in the given charset is unspecified. The {@link
1384 * java.nio.charset.CharsetDecoder} class should be used when more control
1385 * over the decoding process is required.
1386 *
1387 * @param bytes
1388 * The bytes to be decoded into characters
1389 *
1390 * @param charsetName
1391 * The name of a supported {@linkplain java.nio.charset.Charset
1392 * charset}
1393 *
1394 * @throws UnsupportedEncodingException
1395 * If the named charset is not supported
1396 *
1397 * @since 1.1
1398 */
1399 public String(byte[] bytes, String charsetName)
1400 throws UnsupportedEncodingException {
1401 this(lookupCharset(charsetName), bytes, 0, bytes.length);
1402 }
1403
1404 /**
1405 * Constructs a new {@code String} by decoding the specified array of
1406 * bytes using the specified {@linkplain java.nio.charset.Charset charset}.
1407 * The length of the new {@code String} is a function of the charset, and
1408 * hence may not be equal to the length of the byte array.
1409 *
1410 * <p> This method always replaces malformed-input and unmappable-character
1411 * sequences with this charset's default replacement string. The {@link
1412 * java.nio.charset.CharsetDecoder} class should be used when more control
1413 * over the decoding process is required.
1414 *
1415 * @param bytes
1416 * The bytes to be decoded into characters
1417 *
1418 * @param charset
1419 * The {@linkplain java.nio.charset.Charset charset} to be used to
1420 * decode the {@code bytes}
1421 *
1422 * @since 1.6
1423 */
1424 public String(byte[] bytes, Charset charset) {
1425 this(Objects.requireNonNull(charset), bytes, 0, bytes.length);
1426 }
1427
1428 /**
1429 * Constructs a new {@code String} by decoding the specified subarray of
1430 * bytes using the {@link Charset#defaultCharset() default charset}.
1431 * The length of the new {@code String} is a function of the charset,
1432 * and hence may not be equal to the length of the subarray.
1433 *
1434 * <p> The behavior of this constructor when the given bytes are not valid
1435 * in the default charset is unspecified. The {@link
1436 * java.nio.charset.CharsetDecoder} class should be used when more control
1437 * over the decoding process is required.
1438 *
1439 * @param bytes
1440 * The bytes to be decoded into characters
1441 *
1442 * @param offset
1443 * The index of the first byte to decode
1444 *
1445 * @param length
1446 * The number of bytes to decode
1447 *
1448 * @throws IndexOutOfBoundsException
1449 * If {@code offset} is negative, {@code length} is negative, or
1450 * {@code offset} is greater than {@code bytes.length - length}
1451 *
1452 * @since 1.1
1453 */
1454 public String(byte[] bytes, int offset, int length) {
1455 this(Charset.defaultCharset(), bytes, checkBoundsOffCount(offset, length, bytes.length), length);
1456 }
1457
1458 /**
1459 * Constructs a new {@code String} by decoding the specified array of bytes
1460 * using the {@link Charset#defaultCharset() default charset}. The length
1461 * of the new {@code String} is a function of the charset, and hence may not
1462 * be equal to the length of the byte array.
1463 *
1464 * <p> The behavior of this constructor when the given bytes are not valid
1465 * in the default charset is unspecified. The {@link
1466 * java.nio.charset.CharsetDecoder} class should be used when more control
1467 * over the decoding process is required.
1468 *
1469 * @param bytes
1470 * The bytes to be decoded into characters
1471 *
1472 * @since 1.1
1473 */
1474 public String(byte[] bytes) {
1475 this(Charset.defaultCharset(), bytes, 0, bytes.length);
1476 }
1477
1478 /**
1479 * Allocates a new string that contains the sequence of characters
1480 * currently contained in the string buffer argument. The contents of the
1481 * string buffer are copied; subsequent modification of the string buffer
1482 * does not affect the newly created string.
1483 *
1484 * @param buffer
1485 * A {@code StringBuffer}
1486 */
1487 public String(StringBuffer buffer) {
1488 this(buffer.toString());
1489 }
1490
1491 /**
1492 * Allocates a new string that contains the sequence of characters
1493 * currently contained in the string builder argument. The contents of the
1494 * string builder are copied; subsequent modification of the string builder
1495 * does not affect the newly created string.
1496 *
1497 * <p> This constructor is provided to ease migration to {@code
1498 * StringBuilder}. Obtaining a string from a string builder via the {@code
1499 * toString} method is likely to run faster and is generally preferred.
1500 *
1501 * @param builder
1502 * A {@code StringBuilder}
1503 *
1504 * @since 1.5
1505 */
1506 public String(StringBuilder builder) {
1507 this(builder, null);
1508 }
1509
1510 /**
1511 * Returns the length of this string.
1512 * The length is equal to the number of <a href="Character.html#unicode">Unicode
1513 * code units</a> in the string.
1514 *
1515 * @return the length of the sequence of characters represented by this
1516 * object.
1517 */
1518 public int length() {
1519 return value.length >> coder();
1520 }
1521
1522 /**
1523 * Returns {@code true} if, and only if, {@link #length()} is {@code 0}.
1524 *
1525 * @return {@code true} if {@link #length()} is {@code 0}, otherwise
1526 * {@code false}
1527 *
1528 * @since 1.6
1529 */
1530 @Override
1531 public boolean isEmpty() {
1532 return value.length == 0;
1533 }
1534
1535 /**
1536 * Returns the {@code char} value at the
1537 * specified index. An index ranges from {@code 0} to
1538 * {@code length() - 1}. The first {@code char} value of the sequence
1539 * is at index {@code 0}, the next at index {@code 1},
1540 * and so on, as for array indexing.
1541 *
1542 * <p>If the {@code char} value specified by the index is a
1543 * <a href="Character.html#unicode">surrogate</a>, the surrogate
1544 * value is returned.
1545 *
1546 * @param index the index of the {@code char} value.
1547 * @return the {@code char} value at the specified index of this string.
1548 * The first {@code char} value is at index {@code 0}.
1549 * @throws IndexOutOfBoundsException if the {@code index}
1550 * argument is negative or not less than the length of this
1551 * string.
1552 */
1553 public char charAt(int index) {
1554 if (isLatin1()) {
1555 return StringLatin1.charAt(value, index);
1556 } else {
1557 return StringUTF16.charAt(value, index);
1558 }
1559 }
1560
1561 /**
1562 * Returns the character (Unicode code point) at the specified
1563 * index. The index refers to {@code char} values
1564 * (Unicode code units) and ranges from {@code 0} to
1565 * {@link #length()}{@code - 1}.
1566 *
1567 * <p> If the {@code char} value specified at the given index
1568 * is in the high-surrogate range, the following index is less
1569 * than the length of this {@code String}, and the
1570 * {@code char} value at the following index is in the
1571 * low-surrogate range, then the supplementary code point
1572 * corresponding to this surrogate pair is returned. Otherwise,
1573 * the {@code char} value at the given index is returned.
1574 *
1575 * @param index the index to the {@code char} values
1576 * @return the code point value of the character at the
1577 * {@code index}
1578 * @throws IndexOutOfBoundsException if the {@code index}
1579 * argument is negative or not less than the length of this
1580 * string.
1581 * @since 1.5
1582 */
1583 public int codePointAt(int index) {
1584 if (isLatin1()) {
1585 checkIndex(index, value.length);
1586 return value[index] & 0xff;
1587 }
1588 int length = value.length >> 1;
1589 checkIndex(index, length);
1590 return StringUTF16.codePointAt(value, index, length);
1591 }
1592
1593 /**
1594 * Returns the character (Unicode code point) before the specified
1595 * index. The index refers to {@code char} values
1596 * (Unicode code units) and ranges from {@code 1} to {@link
1597 * CharSequence#length() length}.
1598 *
1599 * <p> If the {@code char} value at {@code (index - 1)}
1600 * is in the low-surrogate range, {@code (index - 2)} is not
1601 * negative, and the {@code char} value at {@code (index -
1602 * 2)} is in the high-surrogate range, then the
1603 * supplementary code point value of the surrogate pair is
1604 * returned. If the {@code char} value at {@code index -
1605 * 1} is an unpaired low-surrogate or a high-surrogate, the
1606 * surrogate value is returned.
1607 *
1608 * @param index the index following the code point that should be returned
1609 * @return the Unicode code point value before the given index.
1610 * @throws IndexOutOfBoundsException if the {@code index}
1611 * argument is less than 1 or greater than the length
1612 * of this string.
1613 * @since 1.5
1614 */
1615 public int codePointBefore(int index) {
1616 int i = index - 1;
1617 checkIndex(i, length());
1618 if (isLatin1()) {
1619 return (value[i] & 0xff);
1620 }
1621 return StringUTF16.codePointBefore(value, index);
1622 }
1623
1624 /**
1625 * Returns the number of Unicode code points in the specified text
1626 * range of this {@code String}. The text range begins at the
1627 * specified {@code beginIndex} and extends to the
1628 * {@code char} at index {@code endIndex - 1}. Thus the
1629 * length (in {@code char}s) of the text range is
1630 * {@code endIndex-beginIndex}. Unpaired surrogates within
1631 * the text range count as one code point each.
1632 *
1633 * @param beginIndex the index to the first {@code char} of
1634 * the text range.
1635 * @param endIndex the index after the last {@code char} of
1636 * the text range.
1637 * @return the number of Unicode code points in the specified text
1638 * range
1639 * @throws IndexOutOfBoundsException if the
1640 * {@code beginIndex} is negative, or {@code endIndex}
1641 * is larger than the length of this {@code String}, or
1642 * {@code beginIndex} is larger than {@code endIndex}.
1643 * @since 1.5
1644 */
1645 public int codePointCount(int beginIndex, int endIndex) {
1646 Objects.checkFromToIndex(beginIndex, endIndex, length());
1647 if (isLatin1()) {
1648 return endIndex - beginIndex;
1649 }
1650 return StringUTF16.codePointCount(value, beginIndex, endIndex);
1651 }
1652
1653 /**
1654 * Returns the index within this {@code String} that is
1655 * offset from the given {@code index} by
1656 * {@code codePointOffset} code points. Unpaired surrogates
1657 * within the text range given by {@code index} and
1658 * {@code codePointOffset} count as one code point each.
1659 *
1660 * @param index the index to be offset
1661 * @param codePointOffset the offset in code points
1662 * @return the index within this {@code String}
1663 * @throws IndexOutOfBoundsException if {@code index}
1664 * is negative or larger than the length of this
1665 * {@code String}, or if {@code codePointOffset} is positive
1666 * and the substring starting with {@code index} has fewer
1667 * than {@code codePointOffset} code points,
1668 * or if {@code codePointOffset} is negative and the substring
1669 * before {@code index} has fewer than the absolute value
1670 * of {@code codePointOffset} code points.
1671 * @since 1.5
1672 */
1673 public int offsetByCodePoints(int index, int codePointOffset) {
1674 return Character.offsetByCodePoints(this, index, codePointOffset);
1675 }
1676
1677 /**
1678 * Copies characters from this string into the destination character
1679 * array.
1680 * <p>
1681 * The first character to be copied is at index {@code srcBegin};
1682 * the last character to be copied is at index {@code srcEnd-1}
1683 * (thus the total number of characters to be copied is
1684 * {@code srcEnd-srcBegin}). The characters are copied into the
1685 * subarray of {@code dst} starting at index {@code dstBegin}
1686 * and ending at index:
1687 * <blockquote><pre>
1688 * dstBegin + (srcEnd-srcBegin) - 1
1689 * </pre></blockquote>
1690 *
1691 * @param srcBegin index of the first character in the string
1692 * to copy.
1693 * @param srcEnd index after the last character in the string
1694 * to copy.
1695 * @param dst the destination array.
1696 * @param dstBegin the start offset in the destination array.
1697 * @throws IndexOutOfBoundsException If any of the following
1698 * is true:
1699 * <ul><li>{@code srcBegin} is negative.
1700 * <li>{@code srcBegin} is greater than {@code srcEnd}
1701 * <li>{@code srcEnd} is greater than the length of this
1702 * string
1703 * <li>{@code dstBegin} is negative
1704 * <li>{@code dstBegin+(srcEnd-srcBegin)} is larger than
1705 * {@code dst.length}</ul>
1706 */
1707 public void getChars(int srcBegin, int srcEnd, char[] dst, int dstBegin) {
1708 checkBoundsBeginEnd(srcBegin, srcEnd, length());
1709 checkBoundsOffCount(dstBegin, srcEnd - srcBegin, dst.length);
1710 if (isLatin1()) {
1711 StringLatin1.getChars(value, srcBegin, srcEnd, dst, dstBegin);
1712 } else {
1713 StringUTF16.getChars(value, srcBegin, srcEnd, dst, dstBegin);
1714 }
1715 }
1716
1717 /**
1718 * Copies characters from this string into the destination byte array. Each
1719 * byte receives the 8 low-order bits of the corresponding character. The
1720 * eight high-order bits of each character are not copied and do not
1721 * participate in the transfer in any way.
1722 *
1723 * <p> The first character to be copied is at index {@code srcBegin}; the
1724 * last character to be copied is at index {@code srcEnd-1}. The total
1725 * number of characters to be copied is {@code srcEnd-srcBegin}. The
1726 * characters, converted to bytes, are copied into the subarray of {@code
1727 * dst} starting at index {@code dstBegin} and ending at index:
1728 *
1729 * <blockquote><pre>
1730 * dstBegin + (srcEnd-srcBegin) - 1
1731 * </pre></blockquote>
1732 *
1733 * @deprecated This method does not properly convert characters into
1734 * bytes. As of JDK 1.1, the preferred way to do this is via the
1735 * {@link #getBytes()} method, which uses the {@link Charset#defaultCharset()
1736 * default charset}.
1737 *
1738 * @param srcBegin
1739 * Index of the first character in the string to copy
1740 *
1741 * @param srcEnd
1742 * Index after the last character in the string to copy
1743 *
1744 * @param dst
1745 * The destination array
1746 *
1747 * @param dstBegin
1748 * The start offset in the destination array
1749 *
1750 * @throws IndexOutOfBoundsException
1751 * If any of the following is true:
1752 * <ul>
1753 * <li> {@code srcBegin} is negative
1754 * <li> {@code srcBegin} is greater than {@code srcEnd}
1755 * <li> {@code srcEnd} is greater than the length of this String
1756 * <li> {@code dstBegin} is negative
1757 * <li> {@code dstBegin+(srcEnd-srcBegin)} is larger than {@code
1758 * dst.length}
1759 * </ul>
1760 */
1761 @Deprecated(since="1.1")
1762 public void getBytes(int srcBegin, int srcEnd, byte[] dst, int dstBegin) {
1763 checkBoundsBeginEnd(srcBegin, srcEnd, length());
1764 Objects.requireNonNull(dst);
1765 checkBoundsOffCount(dstBegin, srcEnd - srcBegin, dst.length);
1766 if (isLatin1()) {
1767 StringLatin1.getBytes(value, srcBegin, srcEnd, dst, dstBegin);
1768 } else {
1769 StringUTF16.getBytes(value, srcBegin, srcEnd, dst, dstBegin);
1770 }
1771 }
1772
1773 /**
1774 * Encodes this {@code String} into a sequence of bytes using the named
1775 * charset, storing the result into a new byte array.
1776 *
1777 * <p> The behavior of this method when this string cannot be encoded in
1778 * the given charset is unspecified. The {@link
1779 * java.nio.charset.CharsetEncoder} class should be used when more control
1780 * over the encoding process is required.
1781 *
1782 * @param charsetName
1783 * The name of a supported {@linkplain java.nio.charset.Charset
1784 * charset}
1785 *
1786 * @return The resultant byte array
1787 *
1788 * @throws UnsupportedEncodingException
1789 * If the named charset is not supported
1790 *
1791 * @since 1.1
1792 */
1793 public byte[] getBytes(String charsetName)
1794 throws UnsupportedEncodingException {
1795 return encode(lookupCharset(charsetName), coder(), value);
1796 }
1797
1798 /**
1799 * Encodes this {@code String} into a sequence of bytes using the given
1800 * {@linkplain java.nio.charset.Charset charset}, storing the result into a
1801 * new byte array.
1802 *
1803 * <p> This method always replaces malformed-input and unmappable-character
1804 * sequences with this charset's default replacement byte array. The
1805 * {@link java.nio.charset.CharsetEncoder} class should be used when more
1806 * control over the encoding process is required.
1807 *
1808 * @param charset
1809 * The {@linkplain java.nio.charset.Charset} to be used to encode
1810 * the {@code String}
1811 *
1812 * @return The resultant byte array
1813 *
1814 * @since 1.6
1815 */
1816 public byte[] getBytes(Charset charset) {
1817 if (charset == null) throw new NullPointerException();
1818 return encode(charset, coder(), value);
1819 }
1820
1821 /**
1822 * Encodes this {@code String} into a sequence of bytes using the
1823 * {@link Charset#defaultCharset() default charset}, storing the result
1824 * into a new byte array.
1825 *
1826 * <p> The behavior of this method when this string cannot be encoded in
1827 * the default charset is unspecified. The {@link
1828 * java.nio.charset.CharsetEncoder} class should be used when more control
1829 * over the encoding process is required.
1830 *
1831 * @return The resultant byte array
1832 *
1833 * @since 1.1
1834 */
1835 public byte[] getBytes() {
1836 return encode(Charset.defaultCharset(), coder(), value);
1837 }
1838
1839 /**
1840 * Compares this string to the specified object. The result is {@code
1841 * true} if and only if the argument is not {@code null} and is a {@code
1842 * String} object that represents the same sequence of characters as this
1843 * object.
1844 *
1845 * <p>For finer-grained String comparison, refer to
1846 * {@link java.text.Collator}.
1847 *
1848 * @param anObject
1849 * The object to compare this {@code String} against
1850 *
1851 * @return {@code true} if the given object represents a {@code String}
1852 * equivalent to this string, {@code false} otherwise
1853 *
1854 * @see #compareTo(String)
1855 * @see #equalsIgnoreCase(String)
1856 */
1857 public boolean equals(Object anObject) {
1858 if (this == anObject) {
1859 return true;
1860 }
1861 return (anObject instanceof String aString)
1862 && (!COMPACT_STRINGS || this.coder == aString.coder)
1863 && StringLatin1.equals(value, aString.value);
1864 }
1865
1866 /**
1867 * Compares this string to the specified {@code StringBuffer}. The result
1868 * is {@code true} if and only if this {@code String} represents the same
1869 * sequence of characters as the specified {@code StringBuffer}. This method
1870 * synchronizes on the {@code StringBuffer}.
1871 *
1872 * <p>For finer-grained String comparison, refer to
1873 * {@link java.text.Collator}.
1874 *
1875 * @param sb
1876 * The {@code StringBuffer} to compare this {@code String} against
1877 *
1878 * @return {@code true} if this {@code String} represents the same
1879 * sequence of characters as the specified {@code StringBuffer},
1880 * {@code false} otherwise
1881 *
1882 * @since 1.4
1883 */
1884 public boolean contentEquals(StringBuffer sb) {
1885 return contentEquals((CharSequence)sb);
1886 }
1887
1888 private boolean nonSyncContentEquals(AbstractStringBuilder sb) {
1889 int len = length();
1890 if (len != sb.length()) {
1891 return false;
1892 }
1893 byte[] v1 = value;
1894 byte[] v2 = sb.getValue();
1895 byte coder = coder();
1896 if (coder == sb.getCoder()) {
1897 return v1.length <= v2.length && ArraysSupport.mismatch(v1, v2, v1.length) < 0;
1898 } else {
1899 if (coder != LATIN1) { // utf16 str and latin1 abs can never be "equal"
1900 return false;
1901 }
1902 return StringUTF16.contentEquals(v1, v2, len);
1903 }
1904 }
1905
1906 /**
1907 * Compares this string to the specified {@code CharSequence}. The
1908 * result is {@code true} if and only if this {@code String} represents the
1909 * same sequence of char values as the specified sequence. Note that if the
1910 * {@code CharSequence} is a {@code StringBuffer} then the method
1911 * synchronizes on it.
1912 *
1913 * <p>For finer-grained String comparison, refer to
1914 * {@link java.text.Collator}.
1915 *
1916 * @param cs
1917 * The sequence to compare this {@code String} against
1918 *
1919 * @return {@code true} if this {@code String} represents the same
1920 * sequence of char values as the specified sequence, {@code
1921 * false} otherwise
1922 *
1923 * @since 1.5
1924 */
1925 public boolean contentEquals(CharSequence cs) {
1926 // Argument is a StringBuffer, StringBuilder
1927 if (cs instanceof AbstractStringBuilder) {
1928 if (cs instanceof StringBuffer) {
1929 synchronized(cs) {
1930 return nonSyncContentEquals((AbstractStringBuilder)cs);
1931 }
1932 } else {
1933 return nonSyncContentEquals((AbstractStringBuilder)cs);
1934 }
1935 }
1936 // Argument is a String
1937 if (cs instanceof String) {
1938 return equals(cs);
1939 }
1940 // Argument is a generic CharSequence
1941 int n = cs.length();
1942 if (n != length()) {
1943 return false;
1944 }
1945 byte[] val = this.value;
1946 if (isLatin1()) {
1947 for (int i = 0; i < n; i++) {
1948 if ((val[i] & 0xff) != cs.charAt(i)) {
1949 return false;
1950 }
1951 }
1952 } else {
1953 if (!StringUTF16.contentEquals(val, cs, n)) {
1954 return false;
1955 }
1956 }
1957 return true;
1958 }
1959
1960 /**
1961 * Compares this {@code String} to another {@code String}, ignoring case
1962 * considerations. Two strings are considered equal ignoring case if they
1963 * are of the same length and corresponding Unicode code points in the two
1964 * strings are equal ignoring case.
1965 *
1966 * <p> Two Unicode code points are considered the same
1967 * ignoring case if at least one of the following is true:
1968 * <ul>
1969 * <li> The two Unicode code points are the same (as compared by the
1970 * {@code ==} operator)
1971 * <li> Calling {@code Character.toLowerCase(Character.toUpperCase(int))}
1972 * on each Unicode code point produces the same result
1973 * </ul>
1974 *
1975 * <p>Note that this method does <em>not</em> take locale into account, and
1976 * will result in unsatisfactory results for certain locales. The
1977 * {@link java.text.Collator} class provides locale-sensitive comparison.
1978 *
1979 * @param anotherString
1980 * The {@code String} to compare this {@code String} against
1981 *
1982 * @return {@code true} if the argument is not {@code null} and it
1983 * represents an equivalent {@code String} ignoring case; {@code
1984 * false} otherwise
1985 *
1986 * @see #equals(Object)
1987 * @see #codePoints()
1988 */
1989 public boolean equalsIgnoreCase(String anotherString) {
1990 return (this == anotherString) ? true
1991 : (anotherString != null)
1992 && (anotherString.length() == length())
1993 && regionMatches(true, 0, anotherString, 0, length());
1994 }
1995
1996 /**
1997 * Compares two strings lexicographically.
1998 * The comparison is based on the Unicode value of each character in
1999 * the strings. The character sequence represented by this
2000 * {@code String} object is compared lexicographically to the
2001 * character sequence represented by the argument string. The result is
2002 * a negative integer if this {@code String} object
2003 * lexicographically precedes the argument string. The result is a
2004 * positive integer if this {@code String} object lexicographically
2005 * follows the argument string. The result is zero if the strings
2006 * are equal; {@code compareTo} returns {@code 0} exactly when
2007 * the {@link #equals(Object)} method would return {@code true}.
2008 * <p>
2009 * This is the definition of lexicographic ordering. If two strings are
2010 * different, then either they have different characters at some index
2011 * that is a valid index for both strings, or their lengths are different,
2012 * or both. If they have different characters at one or more index
2013 * positions, let <i>k</i> be the smallest such index; then the string
2014 * whose character at position <i>k</i> has the smaller value, as
2015 * determined by using the {@code <} operator, lexicographically precedes the
2016 * other string. In this case, {@code compareTo} returns the
2017 * difference of the two character values at position {@code k} in
2018 * the two string -- that is, the value:
2019 * <blockquote><pre>
2020 * this.charAt(k)-anotherString.charAt(k)
2021 * </pre></blockquote>
2022 * If there is no index position at which they differ, then the shorter
2023 * string lexicographically precedes the longer string. In this case,
2024 * {@code compareTo} returns the difference of the lengths of the
2025 * strings -- that is, the value:
2026 * <blockquote><pre>
2027 * this.length()-anotherString.length()
2028 * </pre></blockquote>
2029 *
2030 * <p>For finer-grained String comparison, refer to
2031 * {@link java.text.Collator}.
2032 *
2033 * @param anotherString the {@code String} to be compared.
2034 * @return the value {@code 0} if the argument string is equal to
2035 * this string; a value less than {@code 0} if this string
2036 * is lexicographically less than the string argument; and a
2037 * value greater than {@code 0} if this string is
2038 * lexicographically greater than the string argument.
2039 */
2040 public int compareTo(String anotherString) {
2041 byte[] v1 = value;
2042 byte[] v2 = anotherString.value;
2043 byte coder = coder();
2044 if (coder == anotherString.coder()) {
2045 return coder == LATIN1 ? StringLatin1.compareTo(v1, v2)
2046 : StringUTF16.compareTo(v1, v2);
2047 }
2048 return coder == LATIN1 ? StringLatin1.compareToUTF16(v1, v2)
2049 : StringUTF16.compareToLatin1(v1, v2);
2050 }
2051
2052 /**
2053 * A Comparator that orders {@code String} objects as by
2054 * {@link #compareToIgnoreCase(String) compareToIgnoreCase}.
2055 * This comparator is serializable.
2056 * <p>
2057 * Note that this Comparator does <em>not</em> take locale into account,
2058 * and will result in an unsatisfactory ordering for certain locales.
2059 * The {@link java.text.Collator} class provides locale-sensitive comparison.
2060 *
2061 * @see java.text.Collator
2062 * @since 1.2
2063 */
2064 public static final Comparator<String> CASE_INSENSITIVE_ORDER
2065 = new CaseInsensitiveComparator();
2066
2067 /**
2068 * CaseInsensitiveComparator for Strings.
2069 */
2070 private static class CaseInsensitiveComparator
2071 implements Comparator<String>, java.io.Serializable {
2072 // use serialVersionUID from JDK 1.2.2 for interoperability
2073 @java.io.Serial
2074 private static final long serialVersionUID = 8575799808933029326L;
2075
2076 public int compare(String s1, String s2) {
2077 byte[] v1 = s1.value;
2078 byte[] v2 = s2.value;
2079 byte coder = s1.coder();
2080 if (coder == s2.coder()) {
2081 return coder == LATIN1 ? StringLatin1.compareToCI(v1, v2)
2082 : StringUTF16.compareToCI(v1, v2);
2083 }
2084 return coder == LATIN1 ? StringLatin1.compareToCI_UTF16(v1, v2)
2085 : StringUTF16.compareToCI_Latin1(v1, v2);
2086 }
2087
2088 /** Replaces the de-serialized object. */
2089 @java.io.Serial
2090 private Object readResolve() { return CASE_INSENSITIVE_ORDER; }
2091 }
2092
2093 /**
2094 * Compares two strings lexicographically, ignoring case
2095 * differences. This method returns an integer whose sign is that of
2096 * calling {@code compareTo} with case folded versions of the strings
2097 * where case differences have been eliminated by calling
2098 * {@code Character.toLowerCase(Character.toUpperCase(int))} on
2099 * each Unicode code point.
2100 * <p>
2101 * Note that this method does <em>not</em> take locale into account,
2102 * and will result in an unsatisfactory ordering for certain locales.
2103 * The {@link java.text.Collator} class provides locale-sensitive comparison.
2104 *
2105 * @param str the {@code String} to be compared.
2106 * @return a negative integer, zero, or a positive integer as the
2107 * specified String is greater than, equal to, or less
2108 * than this String, ignoring case considerations.
2109 * @see java.text.Collator
2110 * @see #codePoints()
2111 * @since 1.2
2112 */
2113 public int compareToIgnoreCase(String str) {
2114 return CASE_INSENSITIVE_ORDER.compare(this, str);
2115 }
2116
2117 /**
2118 * Tests if two string regions are equal.
2119 * <p>
2120 * A substring of this {@code String} object is compared to a substring
2121 * of the argument other. The result is true if these substrings
2122 * represent identical character sequences. The substring of this
2123 * {@code String} object to be compared begins at index {@code toffset}
2124 * and has length {@code len}. The substring of other to be compared
2125 * begins at index {@code ooffset} and has length {@code len}. The
2126 * result is {@code false} if and only if at least one of the following
2127 * is true:
2128 * <ul><li>{@code toffset} is negative.
2129 * <li>{@code ooffset} is negative.
2130 * <li>{@code toffset+len} is greater than the length of this
2131 * {@code String} object.
2132 * <li>{@code ooffset+len} is greater than the length of the other
2133 * argument.
2134 * <li>There is some nonnegative integer <i>k</i> less than {@code len}
2135 * such that:
2136 * {@code this.charAt(toffset + }<i>k</i>{@code ) != other.charAt(ooffset + }
2137 * <i>k</i>{@code )}
2138 * </ul>
2139 *
2140 * <p>Note that this method does <em>not</em> take locale into account. The
2141 * {@link java.text.Collator} class provides locale-sensitive comparison.
2142 *
2143 * @param toffset the starting offset of the subregion in this string.
2144 * @param other the string argument.
2145 * @param ooffset the starting offset of the subregion in the string
2146 * argument.
2147 * @param len the number of characters to compare.
2148 * @return {@code true} if the specified subregion of this string
2149 * exactly matches the specified subregion of the string argument;
2150 * {@code false} otherwise.
2151 */
2152 public boolean regionMatches(int toffset, String other, int ooffset, int len) {
2153 // Note: toffset, ooffset, or len might be near -1>>>1.
2154 if ((ooffset < 0) || (toffset < 0) ||
2155 (toffset > (long)length() - len) ||
2156 (ooffset > (long)other.length() - len)) {
2157 return false;
2158 }
2159 byte[] tv = value;
2160 byte[] ov = other.value;
2161 byte coder = coder();
2162 if (coder == other.coder()) {
2163 if (coder == UTF16) {
2164 toffset <<= UTF16;
2165 ooffset <<= UTF16;
2166 len <<= UTF16;
2167 }
2168 return ArraysSupport.mismatch(tv, toffset,
2169 ov, ooffset, len) < 0;
2170 } else {
2171 if (coder == LATIN1) {
2172 while (len-- > 0) {
2173 if (StringLatin1.getChar(tv, toffset++) !=
2174 StringUTF16.getChar(ov, ooffset++)) {
2175 return false;
2176 }
2177 }
2178 } else {
2179 while (len-- > 0) {
2180 if (StringUTF16.getChar(tv, toffset++) !=
2181 StringLatin1.getChar(ov, ooffset++)) {
2182 return false;
2183 }
2184 }
2185 }
2186 }
2187 return true;
2188 }
2189
2190 /**
2191 * Tests if two string regions are equal.
2192 * <p>
2193 * A substring of this {@code String} object is compared to a substring
2194 * of the argument {@code other}. The result is {@code true} if these
2195 * substrings represent Unicode code point sequences that are the same,
2196 * ignoring case if and only if {@code ignoreCase} is true.
2197 * The sequences {@code tsequence} and {@code osequence} are compared,
2198 * where {@code tsequence} is the sequence produced as if by calling
2199 * {@code this.substring(toffset, toffset + len).codePoints()} and
2200 * {@code osequence} is the sequence produced as if by calling
2201 * {@code other.substring(ooffset, ooffset + len).codePoints()}.
2202 * The result is {@code true} if and only if all of the following
2203 * are true:
2204 * <ul><li>{@code toffset} is non-negative.
2205 * <li>{@code ooffset} is non-negative.
2206 * <li>{@code toffset+len} is less than or equal to the length of this
2207 * {@code String} object.
2208 * <li>{@code ooffset+len} is less than or equal to the length of the other
2209 * argument.
2210 * <li>if {@code ignoreCase} is {@code false}, all pairs of corresponding Unicode
2211 * code points are equal integer values; or if {@code ignoreCase} is {@code true},
2212 * {@link Character#toLowerCase(int) Character.toLowerCase(}
2213 * {@link Character#toUpperCase(int)}{@code )} on all pairs of Unicode code points
2214 * results in equal integer values.
2215 * </ul>
2216 *
2217 * <p>Note that this method does <em>not</em> take locale into account,
2218 * and will result in unsatisfactory results for certain locales when
2219 * {@code ignoreCase} is {@code true}. The {@link java.text.Collator} class
2220 * provides locale-sensitive comparison.
2221 *
2222 * @param ignoreCase if {@code true}, ignore case when comparing
2223 * characters.
2224 * @param toffset the starting offset of the subregion in this
2225 * string.
2226 * @param other the string argument.
2227 * @param ooffset the starting offset of the subregion in the string
2228 * argument.
2229 * @param len the number of characters (Unicode code units -
2230 * 16bit {@code char} value) to compare.
2231 * @return {@code true} if the specified subregion of this string
2232 * matches the specified subregion of the string argument;
2233 * {@code false} otherwise. Whether the matching is exact
2234 * or case insensitive depends on the {@code ignoreCase}
2235 * argument.
2236 * @see #codePoints()
2237 */
2238 public boolean regionMatches(boolean ignoreCase, int toffset,
2239 String other, int ooffset, int len) {
2240 if (!ignoreCase) {
2241 return regionMatches(toffset, other, ooffset, len);
2242 }
2243 // Note: toffset, ooffset, or len might be near -1>>>1.
2244 if ((ooffset < 0) || (toffset < 0)
2245 || (toffset > (long)length() - len)
2246 || (ooffset > (long)other.length() - len)) {
2247 return false;
2248 }
2249 byte[] tv = value;
2250 byte[] ov = other.value;
2251 byte coder = coder();
2252 if (coder == other.coder()) {
2253 return coder == LATIN1
2254 ? StringLatin1.regionMatchesCI(tv, toffset, ov, ooffset, len)
2255 : StringUTF16.regionMatchesCI(tv, toffset, ov, ooffset, len);
2256 }
2257 return coder == LATIN1
2258 ? StringLatin1.regionMatchesCI_UTF16(tv, toffset, ov, ooffset, len)
2259 : StringUTF16.regionMatchesCI_Latin1(tv, toffset, ov, ooffset, len);
2260 }
2261
2262 /**
2263 * Tests if the substring of this string beginning at the
2264 * specified index starts with the specified prefix.
2265 *
2266 * @param prefix the prefix.
2267 * @param toffset where to begin looking in this string.
2268 * @return {@code true} if the character sequence represented by the
2269 * argument is a prefix of the substring of this object starting
2270 * at index {@code toffset}; {@code false} otherwise.
2271 * The result is {@code false} if {@code toffset} is
2272 * negative or greater than the length of this
2273 * {@code String} object; otherwise the result is the same
2274 * as the result of the expression
2275 * <pre>
2276 * this.substring(toffset).startsWith(prefix)
2277 * </pre>
2278 */
2279 public boolean startsWith(String prefix, int toffset) {
2280 // Note: toffset might be near -1>>>1.
2281 if (toffset < 0 || toffset > length() - prefix.length()) {
2282 return false;
2283 }
2284 byte[] ta = value;
2285 byte[] pa = prefix.value;
2286 int po = 0;
2287 int pc = pa.length;
2288 byte coder = coder();
2289 if (coder == prefix.coder()) {
2290 if (coder == UTF16) {
2291 toffset <<= UTF16;
2292 }
2293 return ArraysSupport.mismatch(ta, toffset,
2294 pa, 0, pc) < 0;
2295 } else {
2296 if (coder == LATIN1) { // && pcoder == UTF16
2297 return false;
2298 }
2299 // coder == UTF16 && pcoder == LATIN1)
2300 while (po < pc) {
2301 if (StringUTF16.getChar(ta, toffset++) != (pa[po++] & 0xff)) {
2302 return false;
2303 }
2304 }
2305 }
2306 return true;
2307 }
2308
2309 /**
2310 * Tests if this string starts with the specified prefix.
2311 *
2312 * @param prefix the prefix.
2313 * @return {@code true} if the character sequence represented by the
2314 * argument is a prefix of the character sequence represented by
2315 * this string; {@code false} otherwise.
2316 * Note also that {@code true} will be returned if the
2317 * argument is an empty string or is equal to this
2318 * {@code String} object as determined by the
2319 * {@link #equals(Object)} method.
2320 * @since 1.0
2321 */
2322 public boolean startsWith(String prefix) {
2323 return startsWith(prefix, 0);
2324 }
2325
2326 /**
2327 * Tests if this string ends with the specified suffix.
2328 *
2329 * @param suffix the suffix.
2330 * @return {@code true} if the character sequence represented by the
2331 * argument is a suffix of the character sequence represented by
2332 * this object; {@code false} otherwise. Note that the
2333 * result will be {@code true} if the argument is the
2334 * empty string or is equal to this {@code String} object
2335 * as determined by the {@link #equals(Object)} method.
2336 */
2337 public boolean endsWith(String suffix) {
2338 return startsWith(suffix, length() - suffix.length());
2339 }
2340
2341 /**
2342 * Returns a hash code for this string. The hash code for a
2343 * {@code String} object is computed as
2344 * <blockquote><pre>
2345 * s[0]*31^(n-1) + s[1]*31^(n-2) + ... + s[n-1]
2346 * </pre></blockquote>
2347 * using {@code int} arithmetic, where {@code s[i]} is the
2348 * <i>i</i>th character of the string, {@code n} is the length of
2349 * the string, and {@code ^} indicates exponentiation.
2350 * (The hash value of the empty string is zero.)
2351 *
2352 * @return a hash code value for this object.
2353 */
2354 public int hashCode() {
2355 // The hash or hashIsZero fields are subject to a benign data race,
2356 // making it crucial to ensure that any observable result of the
2357 // calculation in this method stays correct under any possible read of
2358 // these fields. Necessary restrictions to allow this to be correct
2359 // without explicit memory fences or similar concurrency primitives is
2360 // that we can ever only write to one of these two fields for a given
2361 // String instance, and that the computation is idempotent and derived
2362 // from immutable state
2363 int h = hash;
2364 if (h == 0 && !hashIsZero) {
2365 h = isLatin1() ? StringLatin1.hashCode(value)
2366 : StringUTF16.hashCode(value);
2367 if (h == 0) {
2368 hashIsZero = true;
2369 } else {
2370 hash = h;
2371 }
2372 }
2373 return h;
2374 }
2375
2376 /**
2377 * Returns the index within this string of the first occurrence of
2378 * the specified character. If a character with value
2379 * {@code ch} occurs in the character sequence represented by
2380 * this {@code String} object, then the index (in Unicode
2381 * code units) of the first such occurrence is returned. For
2382 * values of {@code ch} in the range from 0 to 0xFFFF
2383 * (inclusive), this is the smallest value <i>k</i> such that:
2384 * <blockquote><pre>
2385 * this.charAt(<i>k</i>) == ch
2386 * </pre></blockquote>
2387 * is true. For other values of {@code ch}, it is the
2388 * smallest value <i>k</i> such that:
2389 * <blockquote><pre>
2390 * this.codePointAt(<i>k</i>) == ch
2391 * </pre></blockquote>
2392 * is true. In either case, if no such character occurs in this
2393 * string, then {@code -1} is returned.
2394 *
2395 * @param ch a character (Unicode code point).
2396 * @return the index of the first occurrence of the character in the
2397 * character sequence represented by this object, or
2398 * {@code -1} if the character does not occur.
2399 */
2400 public int indexOf(int ch) {
2401 return indexOf(ch, 0);
2402 }
2403
2404 /**
2405 * Returns the index within this string of the first occurrence of the
2406 * specified character, starting the search at the specified index.
2407 * <p>
2408 * If a character with value {@code ch} occurs in the
2409 * character sequence represented by this {@code String}
2410 * object at an index no smaller than {@code fromIndex}, then
2411 * the index of the first such occurrence is returned. For values
2412 * of {@code ch} in the range from 0 to 0xFFFF (inclusive),
2413 * this is the smallest value <i>k</i> such that:
2414 * <blockquote><pre>
2415 * (this.charAt(<i>k</i>) == ch) {@code &&} (<i>k</i> >= fromIndex)
2416 * </pre></blockquote>
2417 * is true. For other values of {@code ch}, it is the
2418 * smallest value <i>k</i> such that:
2419 * <blockquote><pre>
2420 * (this.codePointAt(<i>k</i>) == ch) {@code &&} (<i>k</i> >= fromIndex)
2421 * </pre></blockquote>
2422 * is true. In either case, if no such character occurs in this
2423 * string at or after position {@code fromIndex}, then
2424 * {@code -1} is returned.
2425 *
2426 * <p>
2427 * There is no restriction on the value of {@code fromIndex}. If it
2428 * is negative, it has the same effect as if it were zero: this entire
2429 * string may be searched. If it is greater than the length of this
2430 * string, it has the same effect as if it were equal to the length of
2431 * this string: {@code -1} is returned.
2432 *
2433 * <p>All indices are specified in {@code char} values
2434 * (Unicode code units).
2435 *
2436 * @param ch a character (Unicode code point).
2437 * @param fromIndex the index to start the search from.
2438 * @return the index of the first occurrence of the character in the
2439 * character sequence represented by this object that is greater
2440 * than or equal to {@code fromIndex}, or {@code -1}
2441 * if the character does not occur.
2442 *
2443 * @apiNote
2444 * Unlike {@link #substring(int)}, for example, this method does not throw
2445 * an exception when {@code fromIndex} is outside the valid range.
2446 * Rather, it returns -1 when {@code fromIndex} is larger than the length of
2447 * the string.
2448 * This result is, by itself, indistinguishable from a genuine absence of
2449 * {@code ch} in the string.
2450 * If stricter behavior is needed, {@link #indexOf(int, int, int)}
2451 * should be considered instead.
2452 * On a {@link String} {@code s}, for example,
2453 * {@code s.indexOf(ch, fromIndex, s.length())} would throw if
2454 * {@code fromIndex} were larger than the string length, or were negative.
2455 */
2456 public int indexOf(int ch, int fromIndex) {
2457 return isLatin1() ? StringLatin1.indexOf(value, ch, fromIndex, length())
2458 : StringUTF16.indexOf(value, ch, fromIndex, length());
2459 }
2460
2461 /**
2462 * Returns the index within this string of the first occurrence of the
2463 * specified character, starting the search at {@code beginIndex} and
2464 * stopping before {@code endIndex}.
2465 *
2466 * <p>If a character with value {@code ch} occurs in the
2467 * character sequence represented by this {@code String}
2468 * object at an index no smaller than {@code beginIndex} but smaller than
2469 * {@code endIndex}, then
2470 * the index of the first such occurrence is returned. For values
2471 * of {@code ch} in the range from 0 to 0xFFFF (inclusive),
2472 * this is the smallest value <i>k</i> such that:
2473 * <blockquote><pre>
2474 * (this.charAt(<i>k</i>) == ch) && (beginIndex <= <i>k</i> < endIndex)
2475 * </pre></blockquote>
2476 * is true. For other values of {@code ch}, it is the
2477 * smallest value <i>k</i> such that:
2478 * <blockquote><pre>
2479 * (this.codePointAt(<i>k</i>) == ch) && (beginIndex <= <i>k</i> < endIndex)
2480 * </pre></blockquote>
2481 * is true. In either case, if no such character occurs in this
2482 * string at or after position {@code beginIndex} and before position
2483 * {@code endIndex}, then {@code -1} is returned.
2484 *
2485 * <p>All indices are specified in {@code char} values
2486 * (Unicode code units).
2487 *
2488 * @param ch a character (Unicode code point).
2489 * @param beginIndex the index to start the search from (included).
2490 * @param endIndex the index to stop the search at (excluded).
2491 * @return the index of the first occurrence of the character in the
2492 * character sequence represented by this object that is greater
2493 * than or equal to {@code beginIndex} and less than {@code endIndex},
2494 * or {@code -1} if the character does not occur.
2495 * @throws StringIndexOutOfBoundsException if {@code beginIndex}
2496 * is negative, or {@code endIndex} is larger than the length of
2497 * this {@code String} object, or {@code beginIndex} is larger than
2498 * {@code endIndex}.
2499 * @since 21
2500 */
2501 public int indexOf(int ch, int beginIndex, int endIndex) {
2502 checkBoundsBeginEnd(beginIndex, endIndex, length());
2503 return isLatin1() ? StringLatin1.indexOf(value, ch, beginIndex, endIndex)
2504 : StringUTF16.indexOf(value, ch, beginIndex, endIndex);
2505 }
2506
2507 /**
2508 * Returns the index within this string of the last occurrence of
2509 * the specified character. For values of {@code ch} in the
2510 * range from 0 to 0xFFFF (inclusive), the index (in Unicode code
2511 * units) returned is the largest value <i>k</i> such that:
2512 * <blockquote><pre>
2513 * this.charAt(<i>k</i>) == ch
2514 * </pre></blockquote>
2515 * is true. For other values of {@code ch}, it is the
2516 * largest value <i>k</i> such that:
2517 * <blockquote><pre>
2518 * this.codePointAt(<i>k</i>) == ch
2519 * </pre></blockquote>
2520 * is true. In either case, if no such character occurs in this
2521 * string, then {@code -1} is returned. The
2522 * {@code String} is searched backwards starting at the last
2523 * character.
2524 *
2525 * @param ch a character (Unicode code point).
2526 * @return the index of the last occurrence of the character in the
2527 * character sequence represented by this object, or
2528 * {@code -1} if the character does not occur.
2529 */
2530 public int lastIndexOf(int ch) {
2531 return lastIndexOf(ch, length() - 1);
2532 }
2533
2534 /**
2535 * Returns the index within this string of the last occurrence of
2536 * the specified character, searching backward starting at the
2537 * specified index. For values of {@code ch} in the range
2538 * from 0 to 0xFFFF (inclusive), the index returned is the largest
2539 * value <i>k</i> such that:
2540 * <blockquote><pre>
2541 * (this.charAt(<i>k</i>) == ch) {@code &&} (<i>k</i> <= fromIndex)
2542 * </pre></blockquote>
2543 * is true. For other values of {@code ch}, it is the
2544 * largest value <i>k</i> such that:
2545 * <blockquote><pre>
2546 * (this.codePointAt(<i>k</i>) == ch) {@code &&} (<i>k</i> <= fromIndex)
2547 * </pre></blockquote>
2548 * is true. In either case, if no such character occurs in this
2549 * string at or before position {@code fromIndex}, then
2550 * {@code -1} is returned.
2551 *
2552 * <p>All indices are specified in {@code char} values
2553 * (Unicode code units).
2554 *
2555 * @param ch a character (Unicode code point).
2556 * @param fromIndex the index to start the search from. There is no
2557 * restriction on the value of {@code fromIndex}. If it is
2558 * greater than or equal to the length of this string, it has
2559 * the same effect as if it were equal to one less than the
2560 * length of this string: this entire string may be searched.
2561 * If it is negative, it has the same effect as if it were -1:
2562 * -1 is returned.
2563 * @return the index of the last occurrence of the character in the
2564 * character sequence represented by this object that is less
2565 * than or equal to {@code fromIndex}, or {@code -1}
2566 * if the character does not occur before that point.
2567 */
2568 public int lastIndexOf(int ch, int fromIndex) {
2569 return isLatin1() ? StringLatin1.lastIndexOf(value, ch, fromIndex)
2570 : StringUTF16.lastIndexOf(value, ch, fromIndex);
2571 }
2572
2573 /**
2574 * Returns the index within this string of the first occurrence of the
2575 * specified substring.
2576 *
2577 * <p>The returned index is the smallest value {@code k} for which:
2578 * <pre>{@code
2579 * this.startsWith(str, k)
2580 * }</pre>
2581 * If no such value of {@code k} exists, then {@code -1} is returned.
2582 *
2583 * @param str the substring to search for.
2584 * @return the index of the first occurrence of the specified substring,
2585 * or {@code -1} if there is no such occurrence.
2586 */
2587 public int indexOf(String str) {
2588 byte coder = coder();
2589 if (coder == str.coder()) {
2590 return isLatin1() ? StringLatin1.indexOf(value, str.value)
2591 : StringUTF16.indexOf(value, str.value);
2592 }
2593 if (coder == LATIN1) { // str.coder == UTF16
2594 return -1;
2595 }
2596 return StringUTF16.indexOfLatin1(value, str.value);
2597 }
2598
2599 /**
2600 * Returns the index within this string of the first occurrence of the
2601 * specified substring, starting at the specified index.
2602 *
2603 * <p>The returned index is the smallest value {@code k} for which:
2604 * <pre>{@code
2605 * k >= Math.min(fromIndex, this.length()) &&
2606 * this.startsWith(str, k)
2607 * }</pre>
2608 * If no such value of {@code k} exists, then {@code -1} is returned.
2609 *
2610 * @apiNote
2611 * Unlike {@link #substring(int)}, for example, this method does not throw
2612 * an exception when {@code fromIndex} is outside the valid range.
2613 * Rather, it returns -1 when {@code fromIndex} is larger than the length of
2614 * the string.
2615 * This result is, by itself, indistinguishable from a genuine absence of
2616 * {@code str} in the string.
2617 * If stricter behavior is needed, {@link #indexOf(String, int, int)}
2618 * should be considered instead.
2619 * On {@link String} {@code s} and a non-empty {@code str}, for example,
2620 * {@code s.indexOf(str, fromIndex, s.length())} would throw if
2621 * {@code fromIndex} were larger than the string length, or were negative.
2622 *
2623 * @param str the substring to search for.
2624 * @param fromIndex the index from which to start the search.
2625 * @return the index of the first occurrence of the specified substring,
2626 * starting at the specified index,
2627 * or {@code -1} if there is no such occurrence.
2628 */
2629 public int indexOf(String str, int fromIndex) {
2630 return indexOf(value, coder(), length(), str, fromIndex);
2631 }
2632
2633 /**
2634 * Returns the index of the first occurrence of the specified substring
2635 * within the specified index range of {@code this} string.
2636 *
2637 * <p>This method returns the same result as the one of the invocation
2638 * <pre>{@code
2639 * s.substring(beginIndex, endIndex).indexOf(str) + beginIndex
2640 * }</pre>
2641 * if the index returned by {@link #indexOf(String)} is non-negative,
2642 * and returns -1 otherwise.
2643 * (No substring is instantiated, though.)
2644 *
2645 * @param str the substring to search for.
2646 * @param beginIndex the index to start the search from (included).
2647 * @param endIndex the index to stop the search at (excluded).
2648 * @return the index of the first occurrence of the specified substring
2649 * within the specified index range,
2650 * or {@code -1} if there is no such occurrence.
2651 * @throws StringIndexOutOfBoundsException if {@code beginIndex}
2652 * is negative, or {@code endIndex} is larger than the length of
2653 * this {@code String} object, or {@code beginIndex} is larger than
2654 * {@code endIndex}.
2655 * @since 21
2656 */
2657 public int indexOf(String str, int beginIndex, int endIndex) {
2658 if (str.length() == 1) {
2659 /* Simple optimization, can be omitted without behavioral impact */
2660 return indexOf(str.charAt(0), beginIndex, endIndex);
2661 }
2662 checkBoundsBeginEnd(beginIndex, endIndex, length());
2663 return indexOf(value, coder(), endIndex, str, beginIndex);
2664 }
2665
2666 /**
2667 * Code shared by String and AbstractStringBuilder to do searches. The
2668 * source is the character array being searched, and the target
2669 * is the string being searched for.
2670 *
2671 * @param src the characters being searched.
2672 * @param srcCoder the coder of the source string.
2673 * @param srcCount last index (exclusive) in the source string.
2674 * @param tgtStr the characters being searched for.
2675 * @param fromIndex the index to begin searching from.
2676 */
2677 static int indexOf(byte[] src, byte srcCoder, int srcCount,
2678 String tgtStr, int fromIndex) {
2679 fromIndex = Math.clamp(fromIndex, 0, srcCount);
2680 int tgtCount = tgtStr.length();
2681 if (tgtCount > srcCount - fromIndex) {
2682 return -1;
2683 }
2684 if (tgtCount == 0) {
2685 return fromIndex;
2686 }
2687
2688 byte[] tgt = tgtStr.value;
2689 byte tgtCoder = tgtStr.coder();
2690 if (srcCoder == tgtCoder) {
2691 return srcCoder == LATIN1
2692 ? StringLatin1.indexOf(src, srcCount, tgt, tgtCount, fromIndex)
2693 : StringUTF16.indexOf(src, srcCount, tgt, tgtCount, fromIndex);
2694 }
2695 if (srcCoder == LATIN1) { // && tgtCoder == UTF16
2696 return -1;
2697 }
2698 // srcCoder == UTF16 && tgtCoder == LATIN1) {
2699 return StringUTF16.indexOfLatin1(src, srcCount, tgt, tgtCount, fromIndex);
2700 }
2701
2702 /**
2703 * Returns the index within this string of the last occurrence of the
2704 * specified substring. The last occurrence of the empty string ""
2705 * is considered to occur at the index value {@code this.length()}.
2706 *
2707 * <p>The returned index is the largest value {@code k} for which:
2708 * <pre>{@code
2709 * this.startsWith(str, k)
2710 * }</pre>
2711 * If no such value of {@code k} exists, then {@code -1} is returned.
2712 *
2713 * @param str the substring to search for.
2714 * @return the index of the last occurrence of the specified substring,
2715 * or {@code -1} if there is no such occurrence.
2716 */
2717 public int lastIndexOf(String str) {
2718 return lastIndexOf(str, length());
2719 }
2720
2721 /**
2722 * Returns the index within this string of the last occurrence of the
2723 * specified substring, searching backward starting at the specified index.
2724 *
2725 * <p>The returned index is the largest value {@code k} for which:
2726 * <pre>{@code
2727 * k <= Math.min(fromIndex, this.length()) &&
2728 * this.startsWith(str, k)
2729 * }</pre>
2730 * If no such value of {@code k} exists, then {@code -1} is returned.
2731 *
2732 * @param str the substring to search for.
2733 * @param fromIndex the index to start the search from.
2734 * @return the index of the last occurrence of the specified substring,
2735 * searching backward from the specified index,
2736 * or {@code -1} if there is no such occurrence.
2737 */
2738 public int lastIndexOf(String str, int fromIndex) {
2739 return lastIndexOf(value, coder(), length(), str, fromIndex);
2740 }
2741
2742 /**
2743 * Code shared by String and AbstractStringBuilder to do searches. The
2744 * source is the character array being searched, and the target
2745 * is the string being searched for.
2746 *
2747 * @param src the characters being searched.
2748 * @param srcCoder coder handles the mapping between bytes/chars
2749 * @param srcCount count of the source string.
2750 * @param tgtStr the characters being searched for.
2751 * @param fromIndex the index to begin searching from.
2752 */
2753 static int lastIndexOf(byte[] src, byte srcCoder, int srcCount,
2754 String tgtStr, int fromIndex) {
2755 byte[] tgt = tgtStr.value;
2756 byte tgtCoder = tgtStr.coder();
2757 int tgtCount = tgtStr.length();
2758 /*
2759 * Check arguments; return immediately where possible. For
2760 * consistency, don't check for null str.
2761 */
2762 int rightIndex = srcCount - tgtCount;
2763 if (fromIndex > rightIndex) {
2764 fromIndex = rightIndex;
2765 }
2766 if (fromIndex < 0) {
2767 return -1;
2768 }
2769 /* Empty string always matches. */
2770 if (tgtCount == 0) {
2771 return fromIndex;
2772 }
2773 if (srcCoder == tgtCoder) {
2774 return srcCoder == LATIN1
2775 ? StringLatin1.lastIndexOf(src, srcCount, tgt, tgtCount, fromIndex)
2776 : StringUTF16.lastIndexOf(src, srcCount, tgt, tgtCount, fromIndex);
2777 }
2778 if (srcCoder == LATIN1) { // && tgtCoder == UTF16
2779 return -1;
2780 }
2781 // srcCoder == UTF16 && tgtCoder == LATIN1
2782 return StringUTF16.lastIndexOfLatin1(src, srcCount, tgt, tgtCount, fromIndex);
2783 }
2784
2785 /**
2786 * Returns a string that is a substring of this string. The
2787 * substring begins with the character at the specified index and
2788 * extends to the end of this string. <p>
2789 * Examples:
2790 * <blockquote><pre>
2791 * "unhappy".substring(2) returns "happy"
2792 * "Harbison".substring(3) returns "bison"
2793 * "emptiness".substring(9) returns "" (an empty string)
2794 * </pre></blockquote>
2795 *
2796 * @param beginIndex the beginning index, inclusive.
2797 * @return the specified substring.
2798 * @throws IndexOutOfBoundsException if
2799 * {@code beginIndex} is negative or larger than the
2800 * length of this {@code String} object.
2801 */
2802 public String substring(int beginIndex) {
2803 return substring(beginIndex, length());
2804 }
2805
2806 /**
2807 * Returns a string that is a substring of this string. The
2808 * substring begins at the specified {@code beginIndex} and
2809 * extends to the character at index {@code endIndex - 1}.
2810 * Thus the length of the substring is {@code endIndex-beginIndex}.
2811 * <p>
2812 * Examples:
2813 * <blockquote><pre>
2814 * "hamburger".substring(4, 8) returns "urge"
2815 * "smiles".substring(1, 5) returns "mile"
2816 * </pre></blockquote>
2817 *
2818 * @param beginIndex the beginning index, inclusive.
2819 * @param endIndex the ending index, exclusive.
2820 * @return the specified substring.
2821 * @throws IndexOutOfBoundsException if the
2822 * {@code beginIndex} is negative, or
2823 * {@code endIndex} is larger than the length of
2824 * this {@code String} object, or
2825 * {@code beginIndex} is larger than
2826 * {@code endIndex}.
2827 */
2828 public String substring(int beginIndex, int endIndex) {
2829 int length = length();
2830 checkBoundsBeginEnd(beginIndex, endIndex, length);
2831 if (beginIndex == 0 && endIndex == length) {
2832 return this;
2833 }
2834 int subLen = endIndex - beginIndex;
2835 return isLatin1() ? StringLatin1.newString(value, beginIndex, subLen)
2836 : StringUTF16.newString(value, beginIndex, subLen);
2837 }
2838
2839 /**
2840 * Returns a character sequence that is a subsequence of this sequence.
2841 *
2842 * <p> An invocation of this method of the form
2843 *
2844 * <blockquote><pre>
2845 * str.subSequence(begin, end)</pre></blockquote>
2846 *
2847 * behaves in exactly the same way as the invocation
2848 *
2849 * <blockquote><pre>
2850 * str.substring(begin, end)</pre></blockquote>
2851 *
2852 * @apiNote
2853 * This method is defined so that the {@code String} class can implement
2854 * the {@link CharSequence} interface.
2855 *
2856 * @param beginIndex the begin index, inclusive.
2857 * @param endIndex the end index, exclusive.
2858 * @return the specified subsequence.
2859 *
2860 * @throws IndexOutOfBoundsException
2861 * if {@code beginIndex} or {@code endIndex} is negative,
2862 * if {@code endIndex} is greater than {@code length()},
2863 * or if {@code beginIndex} is greater than {@code endIndex}
2864 *
2865 * @since 1.4
2866 */
2867 public CharSequence subSequence(int beginIndex, int endIndex) {
2868 return this.substring(beginIndex, endIndex);
2869 }
2870
2871 /**
2872 * Concatenates the specified string to the end of this string.
2873 * <p>
2874 * If the length of the argument string is {@code 0}, then this
2875 * {@code String} object is returned. Otherwise, a
2876 * {@code String} object is returned that represents a character
2877 * sequence that is the concatenation of the character sequence
2878 * represented by this {@code String} object and the character
2879 * sequence represented by the argument string.<p>
2880 * Examples:
2881 * <blockquote><pre>
2882 * "cares".concat("s") returns "caress"
2883 * "to".concat("get").concat("her") returns "together"
2884 * </pre></blockquote>
2885 *
2886 * @param str the {@code String} that is concatenated to the end
2887 * of this {@code String}.
2888 * @return a string that represents the concatenation of this object's
2889 * characters followed by the string argument's characters.
2890 */
2891 public String concat(String str) {
2892 if (str.isEmpty()) {
2893 return this;
2894 }
2895 return StringConcatHelper.simpleConcat(this, str);
2896 }
2897
2898 /**
2899 * Returns a string resulting from replacing all occurrences of
2900 * {@code oldChar} in this string with {@code newChar}.
2901 * <p>
2902 * If the character {@code oldChar} does not occur in the
2903 * character sequence represented by this {@code String} object,
2904 * then a reference to this {@code String} object is returned.
2905 * Otherwise, a {@code String} object is returned that
2906 * represents a character sequence identical to the character sequence
2907 * represented by this {@code String} object, except that every
2908 * occurrence of {@code oldChar} is replaced by an occurrence
2909 * of {@code newChar}.
2910 * <p>
2911 * Examples:
2912 * <blockquote><pre>
2913 * "mesquite in your cellar".replace('e', 'o')
2914 * returns "mosquito in your collar"
2915 * "the war of baronets".replace('r', 'y')
2916 * returns "the way of bayonets"
2917 * "sparring with a purple porpoise".replace('p', 't')
2918 * returns "starring with a turtle tortoise"
2919 * "JonL".replace('q', 'x') returns "JonL" (no change)
2920 * </pre></blockquote>
2921 *
2922 * @param oldChar the old character.
2923 * @param newChar the new character.
2924 * @return a string derived from this string by replacing every
2925 * occurrence of {@code oldChar} with {@code newChar}.
2926 */
2927 public String replace(char oldChar, char newChar) {
2928 if (oldChar != newChar) {
2929 String ret = isLatin1() ? StringLatin1.replace(value, oldChar, newChar)
2930 : StringUTF16.replace(value, oldChar, newChar);
2931 if (ret != null) {
2932 return ret;
2933 }
2934 }
2935 return this;
2936 }
2937
2938 /**
2939 * Tells whether or not this string matches the given <a
2940 * href="../util/regex/Pattern.html#sum">regular expression</a>.
2941 *
2942 * <p> An invocation of this method of the form
2943 * <i>str</i>{@code .matches(}<i>regex</i>{@code )} yields exactly the
2944 * same result as the expression
2945 *
2946 * <blockquote>
2947 * {@link java.util.regex.Pattern}.{@link java.util.regex.Pattern#matches(String,CharSequence)
2948 * matches(<i>regex</i>, <i>str</i>)}
2949 * </blockquote>
2950 *
2951 * @param regex
2952 * the regular expression to which this string is to be matched
2953 *
2954 * @return {@code true} if, and only if, this string matches the
2955 * given regular expression
2956 *
2957 * @throws PatternSyntaxException
2958 * if the regular expression's syntax is invalid
2959 *
2960 * @see java.util.regex.Pattern
2961 *
2962 * @since 1.4
2963 */
2964 public boolean matches(String regex) {
2965 return Pattern.matches(regex, this);
2966 }
2967
2968 /**
2969 * Returns true if and only if this string contains the specified
2970 * sequence of char values.
2971 *
2972 * @param s the sequence to search for
2973 * @return true if this string contains {@code s}, false otherwise
2974 * @since 1.5
2975 */
2976 public boolean contains(CharSequence s) {
2977 return indexOf(s.toString()) >= 0;
2978 }
2979
2980 /**
2981 * Replaces the first substring of this string that matches the given <a
2982 * href="../util/regex/Pattern.html#sum">regular expression</a> with the
2983 * given replacement.
2984 *
2985 * <p> An invocation of this method of the form
2986 * <i>str</i>{@code .replaceFirst(}<i>regex</i>{@code ,} <i>repl</i>{@code )}
2987 * yields exactly the same result as the expression
2988 *
2989 * <blockquote>
2990 * <code>
2991 * {@link java.util.regex.Pattern}.{@link
2992 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
2993 * java.util.regex.Pattern#matcher(java.lang.CharSequence) matcher}(<i>str</i>).{@link
2994 * java.util.regex.Matcher#replaceFirst(String) replaceFirst}(<i>repl</i>)
2995 * </code>
2996 * </blockquote>
2997 *
2998 *<p>
2999 * Note that backslashes ({@code \}) and dollar signs ({@code $}) in the
3000 * replacement string may cause the results to be different than if it were
3001 * being treated as a literal replacement string; see
3002 * {@link java.util.regex.Matcher#replaceFirst}.
3003 * Use {@link java.util.regex.Matcher#quoteReplacement} to suppress the special
3004 * meaning of these characters, if desired.
3005 *
3006 * @param regex
3007 * the regular expression to which this string is to be matched
3008 * @param replacement
3009 * the string to be substituted for the first match
3010 *
3011 * @return The resulting {@code String}
3012 *
3013 * @throws PatternSyntaxException
3014 * if the regular expression's syntax is invalid
3015 *
3016 * @see java.util.regex.Pattern
3017 *
3018 * @since 1.4
3019 */
3020 public String replaceFirst(String regex, String replacement) {
3021 return Pattern.compile(regex).matcher(this).replaceFirst(replacement);
3022 }
3023
3024 /**
3025 * Replaces each substring of this string that matches the given <a
3026 * href="../util/regex/Pattern.html#sum">regular expression</a> with the
3027 * given replacement.
3028 *
3029 * <p> An invocation of this method of the form
3030 * <i>str</i>{@code .replaceAll(}<i>regex</i>{@code ,} <i>repl</i>{@code )}
3031 * yields exactly the same result as the expression
3032 *
3033 * <blockquote>
3034 * <code>
3035 * {@link java.util.regex.Pattern}.{@link
3036 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
3037 * java.util.regex.Pattern#matcher(java.lang.CharSequence) matcher}(<i>str</i>).{@link
3038 * java.util.regex.Matcher#replaceAll(String) replaceAll}(<i>repl</i>)
3039 * </code>
3040 * </blockquote>
3041 *
3042 *<p>
3043 * Note that backslashes ({@code \}) and dollar signs ({@code $}) in the
3044 * replacement string may cause the results to be different than if it were
3045 * being treated as a literal replacement string; see
3046 * {@link java.util.regex.Matcher#replaceAll Matcher.replaceAll}.
3047 * Use {@link java.util.regex.Matcher#quoteReplacement} to suppress the special
3048 * meaning of these characters, if desired.
3049 *
3050 * @param regex
3051 * the regular expression to which this string is to be matched
3052 * @param replacement
3053 * the string to be substituted for each match
3054 *
3055 * @return The resulting {@code String}
3056 *
3057 * @throws PatternSyntaxException
3058 * if the regular expression's syntax is invalid
3059 *
3060 * @see java.util.regex.Pattern
3061 *
3062 * @since 1.4
3063 */
3064 public String replaceAll(String regex, String replacement) {
3065 return Pattern.compile(regex).matcher(this).replaceAll(replacement);
3066 }
3067
3068 /**
3069 * Replaces each substring of this string that matches the literal target
3070 * sequence with the specified literal replacement sequence. The
3071 * replacement proceeds from the beginning of the string to the end, for
3072 * example, replacing "aa" with "b" in the string "aaa" will result in
3073 * "ba" rather than "ab".
3074 *
3075 * @param target The sequence of char values to be replaced
3076 * @param replacement The replacement sequence of char values
3077 * @return The resulting string
3078 * @since 1.5
3079 */
3080 public String replace(CharSequence target, CharSequence replacement) {
3081 String trgtStr = target.toString();
3082 String replStr = replacement.toString();
3083 int thisLen = length();
3084 int trgtLen = trgtStr.length();
3085 int replLen = replStr.length();
3086
3087 if (trgtLen > 0) {
3088 if (trgtLen == 1 && replLen == 1) {
3089 return replace(trgtStr.charAt(0), replStr.charAt(0));
3090 }
3091
3092 boolean thisIsLatin1 = this.isLatin1();
3093 boolean trgtIsLatin1 = trgtStr.isLatin1();
3094 boolean replIsLatin1 = replStr.isLatin1();
3095 String ret = (thisIsLatin1 && trgtIsLatin1 && replIsLatin1)
3096 ? StringLatin1.replace(value, thisLen,
3097 trgtStr.value, trgtLen,
3098 replStr.value, replLen)
3099 : StringUTF16.replace(value, thisLen, thisIsLatin1,
3100 trgtStr.value, trgtLen, trgtIsLatin1,
3101 replStr.value, replLen, replIsLatin1);
3102 if (ret != null) {
3103 return ret;
3104 }
3105 return this;
3106
3107 } else { // trgtLen == 0
3108 int resultLen;
3109 try {
3110 resultLen = Math.addExact(thisLen, Math.multiplyExact(
3111 Math.addExact(thisLen, 1), replLen));
3112 } catch (ArithmeticException ignored) {
3113 throw new OutOfMemoryError("Required length exceeds implementation limit");
3114 }
3115
3116 StringBuilder sb = new StringBuilder(resultLen);
3117 sb.append(replStr);
3118 for (int i = 0; i < thisLen; ++i) {
3119 sb.append(charAt(i)).append(replStr);
3120 }
3121 return sb.toString();
3122 }
3123 }
3124
3125 /**
3126 * Splits this string around matches of the given
3127 * <a href="../util/regex/Pattern.html#sum">regular expression</a>.
3128 *
3129 * <p> The array returned by this method contains each substring of this
3130 * string that is terminated by another substring that matches the given
3131 * expression or is terminated by the end of the string. The substrings in
3132 * the array are in the order in which they occur in this string. If the
3133 * expression does not match any part of the input then the resulting array
3134 * has just one element, namely this string.
3135 *
3136 * <p> When there is a positive-width match at the beginning of this
3137 * string then an empty leading substring is included at the beginning
3138 * of the resulting array. A zero-width match at the beginning however
3139 * never produces such empty leading substring.
3140 *
3141 * <p> The {@code limit} parameter controls the number of times the
3142 * pattern is applied and therefore affects the length of the resulting
3143 * array.
3144 * <ul>
3145 * <li><p>
3146 * If the <i>limit</i> is positive then the pattern will be applied
3147 * at most <i>limit</i> - 1 times, the array's length will be
3148 * no greater than <i>limit</i>, and the array's last entry will contain
3149 * all input beyond the last matched delimiter.</p></li>
3150 *
3151 * <li><p>
3152 * If the <i>limit</i> is zero then the pattern will be applied as
3153 * many times as possible, the array can have any length, and trailing
3154 * empty strings will be discarded.</p></li>
3155 *
3156 * <li><p>
3157 * If the <i>limit</i> is negative then the pattern will be applied
3158 * as many times as possible and the array can have any length.</p></li>
3159 * </ul>
3160 *
3161 * <p> The string {@code "boo:and:foo"}, for example, yields the
3162 * following results with these parameters:
3163 *
3164 * <blockquote><table class="plain">
3165 * <caption style="display:none">Split example showing regex, limit, and result</caption>
3166 * <thead>
3167 * <tr>
3168 * <th scope="col">Regex</th>
3169 * <th scope="col">Limit</th>
3170 * <th scope="col">Result</th>
3171 * </tr>
3172 * </thead>
3173 * <tbody>
3174 * <tr><th scope="row" rowspan="3" style="font-weight:normal">:</th>
3175 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">2</th>
3176 * <td>{@code { "boo", "and:foo" }}</td></tr>
3177 * <tr><!-- : -->
3178 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3179 * <td>{@code { "boo", "and", "foo" }}</td></tr>
3180 * <tr><!-- : -->
3181 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-2</th>
3182 * <td>{@code { "boo", "and", "foo" }}</td></tr>
3183 * <tr><th scope="row" rowspan="3" style="font-weight:normal">o</th>
3184 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3185 * <td>{@code { "b", "", ":and:f", "", "" }}</td></tr>
3186 * <tr><!-- o -->
3187 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-2</th>
3188 * <td>{@code { "b", "", ":and:f", "", "" }}</td></tr>
3189 * <tr><!-- o -->
3190 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">0</th>
3191 * <td>{@code { "b", "", ":and:f" }}</td></tr>
3192 * </tbody>
3193 * </table></blockquote>
3194 *
3195 * <p> An invocation of this method of the form
3196 * <i>str.</i>{@code split(}<i>regex</i>{@code ,} <i>n</i>{@code )}
3197 * yields the same result as the expression
3198 *
3199 * <blockquote>
3200 * <code>
3201 * {@link java.util.regex.Pattern}.{@link
3202 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
3203 * java.util.regex.Pattern#split(java.lang.CharSequence,int) split}(<i>str</i>, <i>n</i>)
3204 * </code>
3205 * </blockquote>
3206 *
3207 *
3208 * @param regex
3209 * the delimiting regular expression
3210 *
3211 * @param limit
3212 * the result threshold, as described above
3213 *
3214 * @return the array of strings computed by splitting this string
3215 * around matches of the given regular expression
3216 *
3217 * @throws PatternSyntaxException
3218 * if the regular expression's syntax is invalid
3219 *
3220 * @see java.util.regex.Pattern
3221 *
3222 * @since 1.4
3223 */
3224 public String[] split(String regex, int limit) {
3225 return split(regex, limit, false);
3226 }
3227
3228 /**
3229 * Splits this string around matches of the given regular expression and
3230 * returns both the strings and the matching delimiters.
3231 *
3232 * <p> The array returned by this method contains each substring of this
3233 * string that is terminated by another substring that matches the given
3234 * expression or is terminated by the end of the string.
3235 * Each substring is immediately followed by the subsequence (the delimiter)
3236 * that matches the given expression, <em>except</em> for the last
3237 * substring, which is not followed by anything.
3238 * The substrings in the array and the delimiters are in the order in which
3239 * they occur in the input.
3240 * If the expression does not match any part of the input then the resulting
3241 * array has just one element, namely this string.
3242 *
3243 * <p> When there is a positive-width match at the beginning of this
3244 * string then an empty leading substring is included at the beginning
3245 * of the resulting array. A zero-width match at the beginning however
3246 * never produces such empty leading substring nor the empty delimiter.
3247 *
3248 * <p> The {@code limit} parameter controls the number of times the
3249 * pattern is applied and therefore affects the length of the resulting
3250 * array.
3251 * <ul>
3252 * <li> If the <i>limit</i> is positive then the pattern will be applied
3253 * at most <i>limit</i> - 1 times, the array's length will be
3254 * no greater than 2 × <i>limit</i> - 1, and the array's last
3255 * entry will contain all input beyond the last matched delimiter.</li>
3256 *
3257 * <li> If the <i>limit</i> is zero then the pattern will be applied as
3258 * many times as possible, the array can have any length, and trailing
3259 * empty strings will be discarded.</li>
3260 *
3261 * <li> If the <i>limit</i> is negative then the pattern will be applied
3262 * as many times as possible and the array can have any length.</li>
3263 * </ul>
3264 *
3265 * <p> The input {@code "boo:::and::foo"}, for example, yields the following
3266 * results with these parameters:
3267 *
3268 * <table class="plain" style="margin-left:2em;">
3269 * <caption style="display:none">Split example showing regex, limit, and result</caption>
3270 * <thead>
3271 * <tr>
3272 * <th scope="col">Regex</th>
3273 * <th scope="col">Limit</th>
3274 * <th scope="col">Result</th>
3275 * </tr>
3276 * </thead>
3277 * <tbody>
3278 * <tr><th scope="row" rowspan="3" style="font-weight:normal">:+</th>
3279 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">2</th>
3280 * <td>{@code { "boo", ":::", "and::foo" }}</td></tr>
3281 * <tr><!-- : -->
3282 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3283 * <td>{@code { "boo", ":::", "and", "::", "foo" }}</td></tr>
3284 * <tr><!-- : -->
3285 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-1</th>
3286 * <td>{@code { "boo", ":::", "and", "::", "foo" }}</td></tr>
3287 * <tr><th scope="row" rowspan="3" style="font-weight:normal">o</th>
3288 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3289 * <td>{@code { "b", "o", "", "o", ":::and::f", "o", "", "o", "" }}</td></tr>
3290 * <tr><!-- o -->
3291 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-1</th>
3292 * <td>{@code { "b", "o", "", "o", ":::and::f", "o", "", "o", "" }}</td></tr>
3293 * <tr><!-- o -->
3294 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">0</th>
3295 * <td>{@code { "b", "o", "", "o", ":::and::f", "o", "", "o" }}</td></tr>
3296 * </tbody>
3297 * </table>
3298 *
3299 * @apiNote An invocation of this method of the form
3300 * <i>str.</i>{@code splitWithDelimiters(}<i>regex</i>{@code ,} <i>n</i>{@code )}
3301 * yields the same result as the expression
3302 *
3303 * <blockquote>
3304 * <code>
3305 * {@link java.util.regex.Pattern}.{@link
3306 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
3307 * java.util.regex.Pattern#splitWithDelimiters(CharSequence,int) splitWithDelimiters}(<i>str</i>, <i>n</i>)
3308 * </code>
3309 * </blockquote>
3310 *
3311 * @param regex
3312 * the delimiting regular expression
3313 *
3314 * @param limit
3315 * the result threshold, as described above
3316 *
3317 * @return the array of strings computed by splitting this string
3318 * around matches of the given regular expression, alternating
3319 * substrings and matching delimiters
3320 *
3321 * @since 21
3322 */
3323 public String[] splitWithDelimiters(String regex, int limit) {
3324 return split(regex, limit, true);
3325 }
3326
3327 private String[] split(String regex, int limit, boolean withDelimiters) {
3328 /* fastpath if the regex is a
3329 * (1) one-char String and this character is not one of the
3330 * RegEx's meta characters ".$|()[{^?*+\\", or
3331 * (2) two-char String and the first char is the backslash and
3332 * the second is not the ascii digit or ascii letter.
3333 */
3334 char ch = 0;
3335 if (((regex.length() == 1 &&
3336 ".$|()[{^?*+\\".indexOf(ch = regex.charAt(0)) == -1) ||
3337 (regex.length() == 2 &&
3338 regex.charAt(0) == '\\' &&
3339 (((ch = regex.charAt(1))-'0')|('9'-ch)) < 0 &&
3340 ((ch-'a')|('z'-ch)) < 0 &&
3341 ((ch-'A')|('Z'-ch)) < 0)) &&
3342 (ch < Character.MIN_HIGH_SURROGATE ||
3343 ch > Character.MAX_LOW_SURROGATE))
3344 {
3345 // All the checks above can potentially be constant folded by
3346 // a JIT/AOT compiler when the regex is a constant string.
3347 // That requires method inlining of the checks, which is only
3348 // possible when the actual split logic is in a separate method
3349 // because the large split loop can usually not be inlined.
3350 return split(ch, limit, withDelimiters);
3351 }
3352 Pattern pattern = Pattern.compile(regex);
3353 return withDelimiters
3354 ? pattern.splitWithDelimiters(this, limit)
3355 : pattern.split(this, limit);
3356 }
3357
3358 private String[] split(char ch, int limit, boolean withDelimiters) {
3359 int matchCount = 0;
3360 int off = 0;
3361 int next;
3362 boolean limited = limit > 0;
3363 ArrayList<String> list = new ArrayList<>();
3364 String del = withDelimiters ? String.valueOf(ch) : null;
3365 while ((next = indexOf(ch, off)) != -1) {
3366 if (!limited || matchCount < limit - 1) {
3367 list.add(substring(off, next));
3368 if (withDelimiters) {
3369 list.add(del);
3370 }
3371 off = next + 1;
3372 ++matchCount;
3373 } else { // last one
3374 int last = length();
3375 list.add(substring(off, last));
3376 off = last;
3377 ++matchCount;
3378 break;
3379 }
3380 }
3381 // If no match was found, return this
3382 if (off == 0)
3383 return new String[] {this};
3384
3385 // Add remaining segment
3386 if (!limited || matchCount < limit)
3387 list.add(substring(off, length()));
3388
3389 // Construct result
3390 int resultSize = list.size();
3391 if (limit == 0) {
3392 while (resultSize > 0 && list.get(resultSize - 1).isEmpty()) {
3393 resultSize--;
3394 }
3395 }
3396 String[] result = new String[resultSize];
3397 return list.subList(0, resultSize).toArray(result);
3398 }
3399
3400 /**
3401 * Splits this string around matches of the given <a
3402 * href="../util/regex/Pattern.html#sum">regular expression</a>.
3403 *
3404 * <p> This method works as if by invoking the two-argument {@link
3405 * #split(String, int) split} method with the given expression and a limit
3406 * argument of zero. Trailing empty strings are therefore not included in
3407 * the resulting array.
3408 *
3409 * <p> The string {@code "boo:and:foo"}, for example, yields the following
3410 * results with these expressions:
3411 *
3412 * <blockquote><table class="plain">
3413 * <caption style="display:none">Split examples showing regex and result</caption>
3414 * <thead>
3415 * <tr>
3416 * <th scope="col">Regex</th>
3417 * <th scope="col">Result</th>
3418 * </tr>
3419 * </thead>
3420 * <tbody>
3421 * <tr><th scope="row" style="text-weight:normal">:</th>
3422 * <td>{@code { "boo", "and", "foo" }}</td></tr>
3423 * <tr><th scope="row" style="text-weight:normal">o</th>
3424 * <td>{@code { "b", "", ":and:f" }}</td></tr>
3425 * </tbody>
3426 * </table></blockquote>
3427 *
3428 *
3429 * @param regex
3430 * the delimiting regular expression
3431 *
3432 * @return the array of strings computed by splitting this string
3433 * around matches of the given regular expression
3434 *
3435 * @throws PatternSyntaxException
3436 * if the regular expression's syntax is invalid
3437 *
3438 * @see java.util.regex.Pattern
3439 *
3440 * @since 1.4
3441 */
3442 public String[] split(String regex) {
3443 return split(regex, 0, false);
3444 }
3445
3446 /**
3447 * Returns a new String composed of copies of the
3448 * {@code CharSequence elements} joined together with a copy of
3449 * the specified {@code delimiter}.
3450 *
3451 * <blockquote>For example,
3452 * <pre>{@code
3453 * String message = String.join("-", "Java", "is", "cool");
3454 * // message returned is: "Java-is-cool"
3455 * }</pre></blockquote>
3456 *
3457 * Note that if an element is null, then {@code "null"} is added.
3458 *
3459 * @param delimiter the delimiter that separates each element
3460 * @param elements the elements to join together.
3461 *
3462 * @return a new {@code String} that is composed of the {@code elements}
3463 * separated by the {@code delimiter}
3464 *
3465 * @throws NullPointerException If {@code delimiter} or {@code elements}
3466 * is {@code null}
3467 *
3468 * @see java.util.StringJoiner
3469 * @since 1.8
3470 */
3471 public static String join(CharSequence delimiter, CharSequence... elements) {
3472 var delim = delimiter.toString();
3473 var elems = new String[elements.length];
3474 for (int i = 0; i < elements.length; i++) {
3475 elems[i] = String.valueOf(elements[i]);
3476 }
3477 return join("", "", delim, elems, elems.length);
3478 }
3479
3480 /**
3481 * Designated join routine.
3482 *
3483 * @param prefix the non-null prefix
3484 * @param suffix the non-null suffix
3485 * @param delimiter the non-null delimiter
3486 * @param elements the non-null array of non-null elements
3487 * @param size the number of elements in the array (<= elements.length)
3488 * @return the joined string
3489 */
3490 @ForceInline
3491 static String join(String prefix, String suffix, String delimiter, String[] elements, int size) {
3492 int icoder = prefix.coder() | suffix.coder();
3493 long len = (long) prefix.length() + suffix.length();
3494 if (size > 1) { // when there are more than one element, size - 1 delimiters will be emitted
3495 len += (long) (size - 1) * delimiter.length();
3496 icoder |= delimiter.coder();
3497 }
3498 // assert len > 0L; // max: (long) Integer.MAX_VALUE << 32
3499 // following loop will add max: (long) Integer.MAX_VALUE * Integer.MAX_VALUE to len
3500 // so len can overflow at most once
3501 for (int i = 0; i < size; i++) {
3502 var el = elements[i];
3503 len += el.length();
3504 icoder |= el.coder();
3505 }
3506 byte coder = (byte) icoder;
3507 // long len overflow check, char -> byte length, int len overflow check
3508 if (len < 0L || (len <<= coder) != (int) len) {
3509 throw new OutOfMemoryError("Requested string length exceeds VM limit");
3510 }
3511 byte[] value = StringConcatHelper.newArray(len);
3512
3513 int off = 0;
3514 prefix.getBytes(value, off, coder); off += prefix.length();
3515 if (size > 0) {
3516 var el = elements[0];
3517 el.getBytes(value, off, coder); off += el.length();
3518 for (int i = 1; i < size; i++) {
3519 delimiter.getBytes(value, off, coder); off += delimiter.length();
3520 el = elements[i];
3521 el.getBytes(value, off, coder); off += el.length();
3522 }
3523 }
3524 suffix.getBytes(value, off, coder);
3525 // assert off + suffix.length() == value.length >> coder;
3526
3527 return new String(value, coder);
3528 }
3529
3530 /**
3531 * Returns a new {@code String} composed of copies of the
3532 * {@code CharSequence elements} joined together with a copy of the
3533 * specified {@code delimiter}.
3534 *
3535 * <blockquote>For example,
3536 * <pre>{@code
3537 * List<String> strings = List.of("Java", "is", "cool");
3538 * String message = String.join(" ", strings);
3539 * // message returned is: "Java is cool"
3540 *
3541 * Set<String> strings =
3542 * new LinkedHashSet<>(List.of("Java", "is", "very", "cool"));
3543 * String message = String.join("-", strings);
3544 * // message returned is: "Java-is-very-cool"
3545 * }</pre></blockquote>
3546 *
3547 * Note that if an individual element is {@code null}, then {@code "null"} is added.
3548 *
3549 * @param delimiter a sequence of characters that is used to separate each
3550 * of the {@code elements} in the resulting {@code String}
3551 * @param elements an {@code Iterable} that will have its {@code elements}
3552 * joined together.
3553 *
3554 * @return a new {@code String} that is composed from the {@code elements}
3555 * argument
3556 *
3557 * @throws NullPointerException If {@code delimiter} or {@code elements}
3558 * is {@code null}
3559 *
3560 * @see #join(CharSequence,CharSequence...)
3561 * @see java.util.StringJoiner
3562 * @since 1.8
3563 */
3564 public static String join(CharSequence delimiter,
3565 Iterable<? extends CharSequence> elements) {
3566 Objects.requireNonNull(delimiter);
3567 Objects.requireNonNull(elements);
3568 var delim = delimiter.toString();
3569 var elems = new String[8];
3570 int size = 0;
3571 for (CharSequence cs: elements) {
3572 if (size >= elems.length) {
3573 elems = Arrays.copyOf(elems, elems.length << 1);
3574 }
3575 elems[size++] = String.valueOf(cs);
3576 }
3577 return join("", "", delim, elems, size);
3578 }
3579
3580 /**
3581 * Converts all of the characters in this {@code String} to lower
3582 * case using the rules of the given {@code Locale}. Case mapping is based
3583 * on the Unicode Standard version specified by the {@link java.lang.Character Character}
3584 * class. Since case mappings are not always 1:1 char mappings, the resulting {@code String}
3585 * and this {@code String} may differ in length.
3586 * <p>
3587 * Examples of lowercase mappings are in the following table:
3588 * <table class="plain">
3589 * <caption style="display:none">Lowercase mapping examples showing language code of locale, upper case, lower case, and description</caption>
3590 * <thead>
3591 * <tr>
3592 * <th scope="col">Language Code of Locale</th>
3593 * <th scope="col">Upper Case</th>
3594 * <th scope="col">Lower Case</th>
3595 * <th scope="col">Description</th>
3596 * </tr>
3597 * </thead>
3598 * <tbody>
3599 * <tr>
3600 * <td>tr (Turkish)</td>
3601 * <th scope="row" style="font-weight:normal; text-align:left">\u0130</th>
3602 * <td>\u0069</td>
3603 * <td>capital letter I with dot above -> small letter i</td>
3604 * </tr>
3605 * <tr>
3606 * <td>tr (Turkish)</td>
3607 * <th scope="row" style="font-weight:normal; text-align:left">\u0049</th>
3608 * <td>\u0131</td>
3609 * <td>capital letter I -> small letter dotless i </td>
3610 * </tr>
3611 * <tr>
3612 * <td>(all)</td>
3613 * <th scope="row" style="font-weight:normal; text-align:left">French Fries</th>
3614 * <td>french fries</td>
3615 * <td>lowercased all chars in String</td>
3616 * </tr>
3617 * <tr>
3618 * <td>(all)</td>
3619 * <th scope="row" style="font-weight:normal; text-align:left">
3620 * ΙΧΘΥΣ</th>
3621 * <td>ιχθυσ</td>
3622 * <td>lowercased all chars in String</td>
3623 * </tr>
3624 * </tbody>
3625 * </table>
3626 *
3627 * @param locale use the case transformation rules for this locale
3628 * @return the {@code String}, converted to lowercase.
3629 * @see java.lang.String#toLowerCase()
3630 * @see java.lang.String#toUpperCase()
3631 * @see java.lang.String#toUpperCase(Locale)
3632 * @since 1.1
3633 */
3634 public String toLowerCase(Locale locale) {
3635 return isLatin1() ? StringLatin1.toLowerCase(this, value, locale)
3636 : StringUTF16.toLowerCase(this, value, locale);
3637 }
3638
3639 /**
3640 * Converts all of the characters in this {@code String} to lower
3641 * case using the rules of the default locale. This method is equivalent to
3642 * {@code toLowerCase(Locale.getDefault())}.
3643 *
3644 * @apiNote This method is locale sensitive, and may produce unexpected
3645 * results if used for strings that are intended to be interpreted locale
3646 * independently.
3647 * Examples are programming language identifiers, protocol keys, and HTML
3648 * tags.
3649 * For instance, {@code "TITLE".toLowerCase()} in a Turkish locale
3650 * returns {@code "t\u005Cu0131tle"}, where '\u005Cu0131' is the
3651 * LATIN SMALL LETTER DOTLESS I character.
3652 * To obtain correct results for locale insensitive strings, use
3653 * {@code toLowerCase(Locale.ROOT)}.
3654 *
3655 * @return the {@code String}, converted to lowercase.
3656 * @see java.lang.String#toLowerCase(Locale)
3657 */
3658 public String toLowerCase() {
3659 return toLowerCase(Locale.getDefault());
3660 }
3661
3662 /**
3663 * Converts all of the characters in this {@code String} to upper
3664 * case using the rules of the given {@code Locale}. Case mapping is based
3665 * on the Unicode Standard version specified by the {@link java.lang.Character Character}
3666 * class. Since case mappings are not always 1:1 char mappings, the resulting {@code String}
3667 * and this {@code String} may differ in length.
3668 * <p>
3669 * Examples of locale-sensitive and 1:M case mappings are in the following table:
3670 * <table class="plain">
3671 * <caption style="display:none">Examples of locale-sensitive and 1:M case mappings. Shows Language code of locale, lower case, upper case, and description.</caption>
3672 * <thead>
3673 * <tr>
3674 * <th scope="col">Language Code of Locale</th>
3675 * <th scope="col">Lower Case</th>
3676 * <th scope="col">Upper Case</th>
3677 * <th scope="col">Description</th>
3678 * </tr>
3679 * </thead>
3680 * <tbody>
3681 * <tr>
3682 * <td>tr (Turkish)</td>
3683 * <th scope="row" style="font-weight:normal; text-align:left">\u0069</th>
3684 * <td>\u0130</td>
3685 * <td>small letter i -> capital letter I with dot above</td>
3686 * </tr>
3687 * <tr>
3688 * <td>tr (Turkish)</td>
3689 * <th scope="row" style="font-weight:normal; text-align:left">\u0131</th>
3690 * <td>\u0049</td>
3691 * <td>small letter dotless i -> capital letter I</td>
3692 * </tr>
3693 * <tr>
3694 * <td>(all)</td>
3695 * <th scope="row" style="font-weight:normal; text-align:left">\u00df</th>
3696 * <td>\u0053 \u0053</td>
3697 * <td>small letter sharp s -> two letters: SS</td>
3698 * </tr>
3699 * <tr>
3700 * <td>(all)</td>
3701 * <th scope="row" style="font-weight:normal; text-align:left">Fahrvergnügen</th>
3702 * <td>FAHRVERGNÜGEN</td>
3703 * <td></td>
3704 * </tr>
3705 * </tbody>
3706 * </table>
3707 * @param locale use the case transformation rules for this locale
3708 * @return the {@code String}, converted to uppercase.
3709 * @see java.lang.String#toUpperCase()
3710 * @see java.lang.String#toLowerCase()
3711 * @see java.lang.String#toLowerCase(Locale)
3712 * @since 1.1
3713 */
3714 public String toUpperCase(Locale locale) {
3715 return isLatin1() ? StringLatin1.toUpperCase(this, value, locale)
3716 : StringUTF16.toUpperCase(this, value, locale);
3717 }
3718
3719 /**
3720 * Converts all of the characters in this {@code String} to upper
3721 * case using the rules of the default locale. This method is equivalent to
3722 * {@code toUpperCase(Locale.getDefault())}.
3723 *
3724 * @apiNote This method is locale sensitive, and may produce unexpected
3725 * results if used for strings that are intended to be interpreted locale
3726 * independently.
3727 * Examples are programming language identifiers, protocol keys, and HTML
3728 * tags.
3729 * For instance, {@code "title".toUpperCase()} in a Turkish locale
3730 * returns {@code "T\u005Cu0130TLE"}, where '\u005Cu0130' is the
3731 * LATIN CAPITAL LETTER I WITH DOT ABOVE character.
3732 * To obtain correct results for locale insensitive strings, use
3733 * {@code toUpperCase(Locale.ROOT)}.
3734 *
3735 * @return the {@code String}, converted to uppercase.
3736 * @see java.lang.String#toUpperCase(Locale)
3737 */
3738 public String toUpperCase() {
3739 return toUpperCase(Locale.getDefault());
3740 }
3741
3742 /**
3743 * Returns a string whose value is this string, with all leading
3744 * and trailing space removed, where space is defined
3745 * as any character whose codepoint is less than or equal to
3746 * {@code 'U+0020'} (the space character).
3747 * <p>
3748 * If this {@code String} object represents an empty character
3749 * sequence, or the first and last characters of character sequence
3750 * represented by this {@code String} object both have codes
3751 * that are not space (as defined above), then a
3752 * reference to this {@code String} object is returned.
3753 * <p>
3754 * Otherwise, if all characters in this string are space (as
3755 * defined above), then a {@code String} object representing an
3756 * empty string is returned.
3757 * <p>
3758 * Otherwise, let <i>k</i> be the index of the first character in the
3759 * string whose code is not a space (as defined above) and let
3760 * <i>m</i> be the index of the last character in the string whose code
3761 * is not a space (as defined above). A {@code String}
3762 * object is returned, representing the substring of this string that
3763 * begins with the character at index <i>k</i> and ends with the
3764 * character at index <i>m</i>-that is, the result of
3765 * {@code this.substring(k, m + 1)}.
3766 * <p>
3767 * This method may be used to trim space (as defined above) from
3768 * the beginning and end of a string.
3769 *
3770 * @return a string whose value is this string, with all leading
3771 * and trailing space removed, or this string if it
3772 * has no leading or trailing space.
3773 */
3774 public String trim() {
3775 String ret = isLatin1() ? StringLatin1.trim(value)
3776 : StringUTF16.trim(value);
3777 return ret == null ? this : ret;
3778 }
3779
3780 /**
3781 * Returns a string whose value is this string, with all leading
3782 * and trailing {@linkplain Character#isWhitespace(int) white space}
3783 * removed.
3784 * <p>
3785 * If this {@code String} object represents an empty string,
3786 * or if all code points in this string are
3787 * {@linkplain Character#isWhitespace(int) white space}, then an empty string
3788 * is returned.
3789 * <p>
3790 * Otherwise, returns a substring of this string beginning with the first
3791 * code point that is not a {@linkplain Character#isWhitespace(int) white space}
3792 * up to and including the last code point that is not a
3793 * {@linkplain Character#isWhitespace(int) white space}.
3794 * <p>
3795 * This method may be used to strip
3796 * {@linkplain Character#isWhitespace(int) white space} from
3797 * the beginning and end of a string.
3798 *
3799 * @return a string whose value is this string, with all leading
3800 * and trailing white space removed
3801 *
3802 * @see Character#isWhitespace(int)
3803 *
3804 * @since 11
3805 */
3806 public String strip() {
3807 String ret = isLatin1() ? StringLatin1.strip(value)
3808 : StringUTF16.strip(value);
3809 return ret == null ? this : ret;
3810 }
3811
3812 /**
3813 * Returns a string whose value is this string, with all leading
3814 * {@linkplain Character#isWhitespace(int) white space} removed.
3815 * <p>
3816 * If this {@code String} object represents an empty string,
3817 * or if all code points in this string are
3818 * {@linkplain Character#isWhitespace(int) white space}, then an empty string
3819 * is returned.
3820 * <p>
3821 * Otherwise, returns a substring of this string beginning with the first
3822 * code point that is not a {@linkplain Character#isWhitespace(int) white space}
3823 * up to and including the last code point of this string.
3824 * <p>
3825 * This method may be used to trim
3826 * {@linkplain Character#isWhitespace(int) white space} from
3827 * the beginning of a string.
3828 *
3829 * @return a string whose value is this string, with all leading white
3830 * space removed
3831 *
3832 * @see Character#isWhitespace(int)
3833 *
3834 * @since 11
3835 */
3836 public String stripLeading() {
3837 String ret = isLatin1() ? StringLatin1.stripLeading(value)
3838 : StringUTF16.stripLeading(value);
3839 return ret == null ? this : ret;
3840 }
3841
3842 /**
3843 * Returns a string whose value is this string, with all trailing
3844 * {@linkplain Character#isWhitespace(int) white space} removed.
3845 * <p>
3846 * If this {@code String} object represents an empty string,
3847 * or if all characters in this string are
3848 * {@linkplain Character#isWhitespace(int) white space}, then an empty string
3849 * is returned.
3850 * <p>
3851 * Otherwise, returns a substring of this string beginning with the first
3852 * code point of this string up to and including the last code point
3853 * that is not a {@linkplain Character#isWhitespace(int) white space}.
3854 * <p>
3855 * This method may be used to trim
3856 * {@linkplain Character#isWhitespace(int) white space} from
3857 * the end of a string.
3858 *
3859 * @return a string whose value is this string, with all trailing white
3860 * space removed
3861 *
3862 * @see Character#isWhitespace(int)
3863 *
3864 * @since 11
3865 */
3866 public String stripTrailing() {
3867 String ret = isLatin1() ? StringLatin1.stripTrailing(value)
3868 : StringUTF16.stripTrailing(value);
3869 return ret == null ? this : ret;
3870 }
3871
3872 /**
3873 * Returns {@code true} if the string is empty or contains only
3874 * {@linkplain Character#isWhitespace(int) white space} codepoints,
3875 * otherwise {@code false}.
3876 *
3877 * @return {@code true} if the string is empty or contains only
3878 * {@linkplain Character#isWhitespace(int) white space} codepoints,
3879 * otherwise {@code false}
3880 *
3881 * @see Character#isWhitespace(int)
3882 *
3883 * @since 11
3884 */
3885 public boolean isBlank() {
3886 return indexOfNonWhitespace() == length();
3887 }
3888
3889 /**
3890 * Returns a stream of lines extracted from this string,
3891 * separated by line terminators.
3892 * <p>
3893 * A <i>line terminator</i> is one of the following:
3894 * a line feed character {@code "\n"} (U+000A),
3895 * a carriage return character {@code "\r"} (U+000D),
3896 * or a carriage return followed immediately by a line feed
3897 * {@code "\r\n"} (U+000D U+000A).
3898 * <p>
3899 * A <i>line</i> is either a sequence of zero or more characters
3900 * followed by a line terminator, or it is a sequence of one or
3901 * more characters followed by the end of the string. A
3902 * line does not include the line terminator.
3903 * <p>
3904 * The stream returned by this method contains the lines from
3905 * this string in the order in which they occur.
3906 *
3907 * @apiNote This definition of <i>line</i> implies that an empty
3908 * string has zero lines and that there is no empty line
3909 * following a line terminator at the end of a string.
3910 *
3911 * @implNote This method provides better performance than
3912 * split("\R") by supplying elements lazily and
3913 * by faster search of new line terminators.
3914 *
3915 * @return the stream of lines extracted from this string
3916 *
3917 * @since 11
3918 */
3919 public Stream<String> lines() {
3920 return isLatin1() ? StringLatin1.lines(value) : StringUTF16.lines(value);
3921 }
3922
3923 /**
3924 * Adjusts the indentation of each line of this string based on the value of
3925 * {@code n}, and normalizes line termination characters.
3926 * <p>
3927 * This string is conceptually separated into lines using
3928 * {@link String#lines()}. Each line is then adjusted as described below
3929 * and then suffixed with a line feed {@code "\n"} (U+000A). The resulting
3930 * lines are then concatenated and returned.
3931 * <p>
3932 * If {@code n > 0} then {@code n} spaces (U+0020) are inserted at the
3933 * beginning of each line.
3934 * <p>
3935 * If {@code n < 0} then up to {@code n}
3936 * {@linkplain Character#isWhitespace(int) white space characters} are removed
3937 * from the beginning of each line. If a given line does not contain
3938 * sufficient white space then all leading
3939 * {@linkplain Character#isWhitespace(int) white space characters} are removed.
3940 * Each white space character is treated as a single character. In
3941 * particular, the tab character {@code "\t"} (U+0009) is considered a
3942 * single character; it is not expanded.
3943 * <p>
3944 * If {@code n == 0} then the line remains unchanged. However, line
3945 * terminators are still normalized.
3946 *
3947 * @param n number of leading
3948 * {@linkplain Character#isWhitespace(int) white space characters}
3949 * to add or remove
3950 *
3951 * @return string with indentation adjusted and line endings normalized
3952 *
3953 * @see String#lines()
3954 * @see String#isBlank()
3955 * @see Character#isWhitespace(int)
3956 *
3957 * @since 12
3958 */
3959 public String indent(int n) {
3960 if (isEmpty()) {
3961 return "";
3962 }
3963 Stream<String> stream = lines();
3964 if (n > 0) {
3965 final String spaces = " ".repeat(n);
3966 stream = stream.map(s -> spaces + s);
3967 } else if (n == Integer.MIN_VALUE) {
3968 stream = stream.map(s -> s.stripLeading());
3969 } else if (n < 0) {
3970 stream = stream.map(s -> s.substring(Math.min(-n, s.indexOfNonWhitespace())));
3971 }
3972 return stream.collect(Collectors.joining("\n", "", "\n"));
3973 }
3974
3975 private int indexOfNonWhitespace() {
3976 return isLatin1() ? StringLatin1.indexOfNonWhitespace(value)
3977 : StringUTF16.indexOfNonWhitespace(value);
3978 }
3979
3980 private int lastIndexOfNonWhitespace() {
3981 return isLatin1() ? StringLatin1.lastIndexOfNonWhitespace(value)
3982 : StringUTF16.lastIndexOfNonWhitespace(value);
3983 }
3984
3985 /**
3986 * Returns a string whose value is this string, with incidental
3987 * {@linkplain Character#isWhitespace(int) white space} removed from
3988 * the beginning and end of every line.
3989 * <p>
3990 * Incidental {@linkplain Character#isWhitespace(int) white space}
3991 * is often present in a text block to align the content with the opening
3992 * delimiter. For example, in the following code, dots represent incidental
3993 * {@linkplain Character#isWhitespace(int) white space}:
3994 * <blockquote><pre>
3995 * String html = """
3996 * ..............<html>
3997 * .............. <body>
3998 * .............. <p>Hello, world</p>
3999 * .............. </body>
4000 * ..............</html>
4001 * ..............""";
4002 * </pre></blockquote>
4003 * This method treats the incidental
4004 * {@linkplain Character#isWhitespace(int) white space} as indentation to be
4005 * stripped, producing a string that preserves the relative indentation of
4006 * the content. Using | to visualize the start of each line of the string:
4007 * <blockquote><pre>
4008 * |<html>
4009 * | <body>
4010 * | <p>Hello, world</p>
4011 * | </body>
4012 * |</html>
4013 * </pre></blockquote>
4014 * First, the individual lines of this string are extracted. A <i>line</i>
4015 * is a sequence of zero or more characters followed by either a line
4016 * terminator or the end of the string.
4017 * If the string has at least one line terminator, the last line consists
4018 * of the characters between the last terminator and the end of the string.
4019 * Otherwise, if the string has no terminators, the last line is the start
4020 * of the string to the end of the string, in other words, the entire
4021 * string.
4022 * A line does not include the line terminator.
4023 * <p>
4024 * Then, the <i>minimum indentation</i> (min) is determined as follows:
4025 * <ul>
4026 * <li><p>For each non-blank line (as defined by {@link String#isBlank()}),
4027 * the leading {@linkplain Character#isWhitespace(int) white space}
4028 * characters are counted.</p>
4029 * </li>
4030 * <li><p>The leading {@linkplain Character#isWhitespace(int) white space}
4031 * characters on the last line are also counted even if
4032 * {@linkplain String#isBlank() blank}.</p>
4033 * </li>
4034 * </ul>
4035 * <p>The <i>min</i> value is the smallest of these counts.
4036 * <p>
4037 * For each {@linkplain String#isBlank() non-blank} line, <i>min</i> leading
4038 * {@linkplain Character#isWhitespace(int) white space} characters are
4039 * removed, and any trailing {@linkplain Character#isWhitespace(int) white
4040 * space} characters are removed. {@linkplain String#isBlank() Blank} lines
4041 * are replaced with the empty string.
4042 *
4043 * <p>
4044 * Finally, the lines are joined into a new string, using the LF character
4045 * {@code "\n"} (U+000A) to separate lines.
4046 *
4047 * @apiNote
4048 * This method's primary purpose is to shift a block of lines as far as
4049 * possible to the left, while preserving relative indentation. Lines
4050 * that were indented the least will thus have no leading
4051 * {@linkplain Character#isWhitespace(int) white space}.
4052 * The result will have the same number of line terminators as this string.
4053 * If this string ends with a line terminator then the result will end
4054 * with a line terminator.
4055 *
4056 * @implSpec
4057 * This method treats all {@linkplain Character#isWhitespace(int) white space}
4058 * characters as having equal width. As long as the indentation on every
4059 * line is consistently composed of the same character sequences, then the
4060 * result will be as described above.
4061 *
4062 * @return string with incidental indentation removed and line
4063 * terminators normalized
4064 *
4065 * @see String#lines()
4066 * @see String#isBlank()
4067 * @see String#indent(int)
4068 * @see Character#isWhitespace(int)
4069 *
4070 * @since 15
4071 *
4072 */
4073 public String stripIndent() {
4074 int length = length();
4075 if (length == 0) {
4076 return "";
4077 }
4078 char lastChar = charAt(length - 1);
4079 boolean optOut = lastChar == '\n' || lastChar == '\r';
4080 List<String> lines = lines().toList();
4081 final int outdent = optOut ? 0 : outdent(lines);
4082 return lines.stream()
4083 .map(line -> {
4084 int firstNonWhitespace = line.indexOfNonWhitespace();
4085 int lastNonWhitespace = line.lastIndexOfNonWhitespace();
4086 int incidentalWhitespace = Math.min(outdent, firstNonWhitespace);
4087 return firstNonWhitespace > lastNonWhitespace
4088 ? "" : line.substring(incidentalWhitespace, lastNonWhitespace);
4089 })
4090 .collect(Collectors.joining("\n", "", optOut ? "\n" : ""));
4091 }
4092
4093 private static int outdent(List<String> lines) {
4094 // Note: outdent is guaranteed to be zero or positive number.
4095 // If there isn't a non-blank line then the last must be blank
4096 int outdent = Integer.MAX_VALUE;
4097 for (String line : lines) {
4098 int leadingWhitespace = line.indexOfNonWhitespace();
4099 if (leadingWhitespace != line.length()) {
4100 outdent = Integer.min(outdent, leadingWhitespace);
4101 }
4102 }
4103 String lastLine = lines.get(lines.size() - 1);
4104 if (lastLine.isBlank()) {
4105 outdent = Integer.min(outdent, lastLine.length());
4106 }
4107 return outdent;
4108 }
4109
4110 /**
4111 * Returns a string whose value is this string, with escape sequences
4112 * translated as if in a string literal.
4113 * <p>
4114 * Escape sequences are translated as follows;
4115 * <table class="striped">
4116 * <caption style="display:none">Translation</caption>
4117 * <thead>
4118 * <tr>
4119 * <th scope="col">Escape</th>
4120 * <th scope="col">Name</th>
4121 * <th scope="col">Translation</th>
4122 * </tr>
4123 * </thead>
4124 * <tbody>
4125 * <tr>
4126 * <th scope="row">{@code \u005Cb}</th>
4127 * <td>backspace</td>
4128 * <td>{@code U+0008}</td>
4129 * </tr>
4130 * <tr>
4131 * <th scope="row">{@code \u005Ct}</th>
4132 * <td>horizontal tab</td>
4133 * <td>{@code U+0009}</td>
4134 * </tr>
4135 * <tr>
4136 * <th scope="row">{@code \u005Cn}</th>
4137 * <td>line feed</td>
4138 * <td>{@code U+000A}</td>
4139 * </tr>
4140 * <tr>
4141 * <th scope="row">{@code \u005Cf}</th>
4142 * <td>form feed</td>
4143 * <td>{@code U+000C}</td>
4144 * </tr>
4145 * <tr>
4146 * <th scope="row">{@code \u005Cr}</th>
4147 * <td>carriage return</td>
4148 * <td>{@code U+000D}</td>
4149 * </tr>
4150 * <tr>
4151 * <th scope="row">{@code \u005Cs}</th>
4152 * <td>space</td>
4153 * <td>{@code U+0020}</td>
4154 * </tr>
4155 * <tr>
4156 * <th scope="row">{@code \u005C"}</th>
4157 * <td>double quote</td>
4158 * <td>{@code U+0022}</td>
4159 * </tr>
4160 * <tr>
4161 * <th scope="row">{@code \u005C'}</th>
4162 * <td>single quote</td>
4163 * <td>{@code U+0027}</td>
4164 * </tr>
4165 * <tr>
4166 * <th scope="row">{@code \u005C\u005C}</th>
4167 * <td>backslash</td>
4168 * <td>{@code U+005C}</td>
4169 * </tr>
4170 * <tr>
4171 * <th scope="row">{@code \u005C0 - \u005C377}</th>
4172 * <td>octal escape</td>
4173 * <td>code point equivalents</td>
4174 * </tr>
4175 * <tr>
4176 * <th scope="row">{@code \u005C<line-terminator>}</th>
4177 * <td>continuation</td>
4178 * <td>discard</td>
4179 * </tr>
4180 * </tbody>
4181 * </table>
4182 *
4183 * @implNote
4184 * This method does <em>not</em> translate Unicode escapes such as "{@code \u005cu2022}".
4185 * Unicode escapes are translated by the Java compiler when reading input characters and
4186 * are not part of the string literal specification.
4187 *
4188 * @throws IllegalArgumentException when an escape sequence is malformed.
4189 *
4190 * @return String with escape sequences translated.
4191 *
4192 * @jls 3.10.7 Escape Sequences
4193 *
4194 * @since 15
4195 */
4196 public String translateEscapes() {
4197 if (isEmpty()) {
4198 return "";
4199 }
4200 char[] chars = toCharArray();
4201 int length = chars.length;
4202 int from = 0;
4203 int to = 0;
4204 while (from < length) {
4205 char ch = chars[from++];
4206 if (ch == '\\') {
4207 ch = from < length ? chars[from++] : '\0';
4208 switch (ch) {
4209 case 'b':
4210 ch = '\b';
4211 break;
4212 case 'f':
4213 ch = '\f';
4214 break;
4215 case 'n':
4216 ch = '\n';
4217 break;
4218 case 'r':
4219 ch = '\r';
4220 break;
4221 case 's':
4222 ch = ' ';
4223 break;
4224 case 't':
4225 ch = '\t';
4226 break;
4227 case '\'':
4228 case '\"':
4229 case '\\':
4230 // as is
4231 break;
4232 case '0': case '1': case '2': case '3':
4233 case '4': case '5': case '6': case '7':
4234 int limit = Integer.min(from + (ch <= '3' ? 2 : 1), length);
4235 int code = ch - '0';
4236 while (from < limit) {
4237 ch = chars[from];
4238 if (ch < '0' || '7' < ch) {
4239 break;
4240 }
4241 from++;
4242 code = (code << 3) | (ch - '0');
4243 }
4244 ch = (char)code;
4245 break;
4246 case '\n':
4247 continue;
4248 case '\r':
4249 if (from < length && chars[from] == '\n') {
4250 from++;
4251 }
4252 continue;
4253 default: {
4254 String msg = String.format(
4255 "Invalid escape sequence: \\%c \\\\u%04X",
4256 ch, (int)ch);
4257 throw new IllegalArgumentException(msg);
4258 }
4259 }
4260 }
4261
4262 chars[to++] = ch;
4263 }
4264
4265 return new String(chars, 0, to);
4266 }
4267
4268 /**
4269 * This method allows the application of a function to {@code this}
4270 * string. The function should expect a single String argument
4271 * and produce an {@code R} result.
4272 * <p>
4273 * Any exception thrown by {@code f.apply()} will be propagated to the
4274 * caller.
4275 *
4276 * @param f a function to apply
4277 *
4278 * @param <R> the type of the result
4279 *
4280 * @return the result of applying the function to this string
4281 *
4282 * @see java.util.function.Function
4283 *
4284 * @since 12
4285 */
4286 public <R> R transform(Function<? super String, ? extends R> f) {
4287 return f.apply(this);
4288 }
4289
4290 /**
4291 * This object (which is already a string!) is itself returned.
4292 *
4293 * @return the string itself.
4294 */
4295 public String toString() {
4296 return this;
4297 }
4298
4299 /**
4300 * Returns a stream of {@code int} zero-extending the {@code char} values
4301 * from this sequence. Any char which maps to a {@linkplain
4302 * Character##unicode surrogate code point} is passed through
4303 * uninterpreted.
4304 *
4305 * @return an IntStream of char values from this sequence
4306 * @since 9
4307 */
4308 @Override
4309 public IntStream chars() {
4310 return StreamSupport.intStream(
4311 isLatin1() ? new StringLatin1.CharsSpliterator(value, Spliterator.IMMUTABLE)
4312 : new StringUTF16.CharsSpliterator(value, Spliterator.IMMUTABLE),
4313 false);
4314 }
4315
4316
4317 /**
4318 * Returns a stream of code point values from this sequence. Any surrogate
4319 * pairs encountered in the sequence are combined as if by {@linkplain
4320 * Character#toCodePoint Character.toCodePoint} and the result is passed
4321 * to the stream. Any other code units, including ordinary BMP characters,
4322 * unpaired surrogates, and undefined code units, are zero-extended to
4323 * {@code int} values which are then passed to the stream.
4324 *
4325 * @return an IntStream of Unicode code points from this sequence
4326 * @since 9
4327 */
4328 @Override
4329 public IntStream codePoints() {
4330 return StreamSupport.intStream(
4331 isLatin1() ? new StringLatin1.CharsSpliterator(value, Spliterator.IMMUTABLE)
4332 : new StringUTF16.CodePointsSpliterator(value, Spliterator.IMMUTABLE),
4333 false);
4334 }
4335
4336 /**
4337 * Converts this string to a new character array.
4338 *
4339 * @return a newly allocated character array whose length is the length
4340 * of this string and whose contents are initialized to contain
4341 * the character sequence represented by this string.
4342 */
4343 public char[] toCharArray() {
4344 return isLatin1() ? StringLatin1.toChars(value)
4345 : StringUTF16.toChars(value);
4346 }
4347
4348 /**
4349 * Returns a formatted string using the specified format string and
4350 * arguments.
4351 *
4352 * <p> The locale always used is the one returned by {@link
4353 * java.util.Locale#getDefault(java.util.Locale.Category)
4354 * Locale.getDefault(Locale.Category)} with
4355 * {@link java.util.Locale.Category#FORMAT FORMAT} category specified.
4356 *
4357 * @param format
4358 * A <a href="../util/Formatter.html#syntax">format string</a>
4359 *
4360 * @param args
4361 * Arguments referenced by the format specifiers in the format
4362 * string. If there are more arguments than format specifiers, the
4363 * extra arguments are ignored. The number of arguments is
4364 * variable and may be zero. The maximum number of arguments is
4365 * limited by the maximum dimension of a Java array as defined by
4366 * <cite>The Java Virtual Machine Specification</cite>.
4367 * The behaviour on a
4368 * {@code null} argument depends on the <a
4369 * href="../util/Formatter.html#syntax">conversion</a>.
4370 *
4371 * @throws java.util.IllegalFormatException
4372 * If a format string contains an illegal syntax, a format
4373 * specifier that is incompatible with the given arguments,
4374 * insufficient arguments given the format string, or other
4375 * illegal conditions. For specification of all possible
4376 * formatting errors, see the <a
4377 * href="../util/Formatter.html#detail">Details</a> section of the
4378 * formatter class specification.
4379 *
4380 * @return A formatted string
4381 *
4382 * @see java.util.Formatter
4383 * @since 1.5
4384 */
4385 public static String format(String format, Object... args) {
4386 return new Formatter().format(format, args).toString();
4387 }
4388
4389 /**
4390 * Returns a formatted string using the specified locale, format string,
4391 * and arguments.
4392 *
4393 * @param l
4394 * The {@linkplain java.util.Locale locale} to apply during
4395 * formatting. If {@code l} is {@code null} then no localization
4396 * is applied.
4397 *
4398 * @param format
4399 * A <a href="../util/Formatter.html#syntax">format string</a>
4400 *
4401 * @param args
4402 * Arguments referenced by the format specifiers in the format
4403 * string. If there are more arguments than format specifiers, the
4404 * extra arguments are ignored. The number of arguments is
4405 * variable and may be zero. The maximum number of arguments is
4406 * limited by the maximum dimension of a Java array as defined by
4407 * <cite>The Java Virtual Machine Specification</cite>.
4408 * The behaviour on a
4409 * {@code null} argument depends on the
4410 * <a href="../util/Formatter.html#syntax">conversion</a>.
4411 *
4412 * @throws java.util.IllegalFormatException
4413 * If a format string contains an illegal syntax, a format
4414 * specifier that is incompatible with the given arguments,
4415 * insufficient arguments given the format string, or other
4416 * illegal conditions. For specification of all possible
4417 * formatting errors, see the <a
4418 * href="../util/Formatter.html#detail">Details</a> section of the
4419 * formatter class specification
4420 *
4421 * @return A formatted string
4422 *
4423 * @see java.util.Formatter
4424 * @since 1.5
4425 */
4426 public static String format(Locale l, String format, Object... args) {
4427 return new Formatter(l).format(format, args).toString();
4428 }
4429
4430 /**
4431 * Formats using this string as the format string, and the supplied
4432 * arguments.
4433 *
4434 * @implSpec This method is equivalent to {@code String.format(this, args)}.
4435 *
4436 * @param args
4437 * Arguments referenced by the format specifiers in this string.
4438 *
4439 * @return A formatted string
4440 *
4441 * @see java.lang.String#format(String,Object...)
4442 * @see java.util.Formatter
4443 *
4444 * @since 15
4445 *
4446 */
4447 public String formatted(Object... args) {
4448 return new Formatter().format(this, args).toString();
4449 }
4450
4451 /**
4452 * Returns the string representation of the {@code Object} argument.
4453 *
4454 * @param obj an {@code Object}.
4455 * @return if the argument is {@code null}, then a string equal to
4456 * {@code "null"}; otherwise, the value of
4457 * {@code obj.toString()} is returned.
4458 * @see java.lang.Object#toString()
4459 */
4460 public static String valueOf(Object obj) {
4461 return (obj == null) ? "null" : obj.toString();
4462 }
4463
4464 /**
4465 * Returns the string representation of the {@code char} array
4466 * argument. The contents of the character array are copied; subsequent
4467 * modification of the character array does not affect the returned
4468 * string.
4469 *
4470 * @param data the character array.
4471 * @return a {@code String} that contains the characters of the
4472 * character array.
4473 */
4474 public static String valueOf(char[] data) {
4475 return new String(data);
4476 }
4477
4478 /**
4479 * Returns the string representation of a specific subarray of the
4480 * {@code char} array argument.
4481 * <p>
4482 * The {@code offset} argument is the index of the first
4483 * character of the subarray. The {@code count} argument
4484 * specifies the length of the subarray. The contents of the subarray
4485 * are copied; subsequent modification of the character array does not
4486 * affect the returned string.
4487 *
4488 * @param data the character array.
4489 * @param offset initial offset of the subarray.
4490 * @param count length of the subarray.
4491 * @return a {@code String} that contains the characters of the
4492 * specified subarray of the character array.
4493 * @throws IndexOutOfBoundsException if {@code offset} is
4494 * negative, or {@code count} is negative, or
4495 * {@code offset+count} is larger than
4496 * {@code data.length}.
4497 */
4498 public static String valueOf(char[] data, int offset, int count) {
4499 return new String(data, offset, count);
4500 }
4501
4502 /**
4503 * Equivalent to {@link #valueOf(char[], int, int)}.
4504 *
4505 * @param data the character array.
4506 * @param offset initial offset of the subarray.
4507 * @param count length of the subarray.
4508 * @return a {@code String} that contains the characters of the
4509 * specified subarray of the character array.
4510 * @throws IndexOutOfBoundsException if {@code offset} is
4511 * negative, or {@code count} is negative, or
4512 * {@code offset+count} is larger than
4513 * {@code data.length}.
4514 */
4515 public static String copyValueOf(char[] data, int offset, int count) {
4516 return new String(data, offset, count);
4517 }
4518
4519 /**
4520 * Equivalent to {@link #valueOf(char[])}.
4521 *
4522 * @param data the character array.
4523 * @return a {@code String} that contains the characters of the
4524 * character array.
4525 */
4526 public static String copyValueOf(char[] data) {
4527 return new String(data);
4528 }
4529
4530 /**
4531 * Returns the string representation of the {@code boolean} argument.
4532 *
4533 * @param b a {@code boolean}.
4534 * @return if the argument is {@code true}, a string equal to
4535 * {@code "true"} is returned; otherwise, a string equal to
4536 * {@code "false"} is returned.
4537 */
4538 public static String valueOf(boolean b) {
4539 return b ? "true" : "false";
4540 }
4541
4542 /**
4543 * Returns the string representation of the {@code char}
4544 * argument.
4545 *
4546 * @param c a {@code char}.
4547 * @return a string of length {@code 1} containing
4548 * as its single character the argument {@code c}.
4549 */
4550 public static String valueOf(char c) {
4551 if (COMPACT_STRINGS && StringLatin1.canEncode(c)) {
4552 return new String(StringLatin1.toBytes(c), LATIN1);
4553 }
4554 return new String(StringUTF16.toBytes(c), UTF16);
4555 }
4556
4557 /**
4558 * Returns the string representation of the {@code int} argument.
4559 * <p>
4560 * The representation is exactly the one returned by the
4561 * {@code Integer.toString} method of one argument.
4562 *
4563 * @param i an {@code int}.
4564 * @return a string representation of the {@code int} argument.
4565 * @see java.lang.Integer#toString(int, int)
4566 */
4567 public static String valueOf(int i) {
4568 return Integer.toString(i);
4569 }
4570
4571 /**
4572 * Returns the string representation of the {@code long} argument.
4573 * <p>
4574 * The representation is exactly the one returned by the
4575 * {@code Long.toString} method of one argument.
4576 *
4577 * @param l a {@code long}.
4578 * @return a string representation of the {@code long} argument.
4579 * @see java.lang.Long#toString(long)
4580 */
4581 public static String valueOf(long l) {
4582 return Long.toString(l);
4583 }
4584
4585 /**
4586 * Returns the string representation of the {@code float} argument.
4587 * <p>
4588 * The representation is exactly the one returned by the
4589 * {@code Float.toString} method of one argument.
4590 *
4591 * @param f a {@code float}.
4592 * @return a string representation of the {@code float} argument.
4593 * @see java.lang.Float#toString(float)
4594 */
4595 public static String valueOf(float f) {
4596 return Float.toString(f);
4597 }
4598
4599 /**
4600 * Returns the string representation of the {@code double} argument.
4601 * <p>
4602 * The representation is exactly the one returned by the
4603 * {@code Double.toString} method of one argument.
4604 *
4605 * @param d a {@code double}.
4606 * @return a string representation of the {@code double} argument.
4607 * @see java.lang.Double#toString(double)
4608 */
4609 public static String valueOf(double d) {
4610 return Double.toString(d);
4611 }
4612
4613 /**
4614 * Returns a canonical representation for the string object.
4615 * <p>
4616 * A pool of strings, initially empty, is maintained privately by the
4617 * class {@code String}.
4618 * <p>
4619 * When the intern method is invoked, if the pool already contains a
4620 * string equal to this {@code String} object as determined by
4621 * the {@link #equals(Object)} method, then the string from the pool is
4622 * returned. Otherwise, this {@code String} object is added to the
4623 * pool and a reference to this {@code String} object is returned.
4624 * <p>
4625 * It follows that for any two strings {@code s} and {@code t},
4626 * {@code s.intern() == t.intern()} is {@code true}
4627 * if and only if {@code s.equals(t)} is {@code true}.
4628 * <p>
4629 * All literal strings and string-valued constant expressions are
4630 * interned. String literals are defined in section {@jls 3.10.5} of the
4631 * <cite>The Java Language Specification</cite>.
4632 *
4633 * @return a string that has the same contents as this string, but is
4634 * guaranteed to be from a pool of unique strings.
4635 */
4636 public native String intern();
4637
4638 /**
4639 * Returns a string whose value is the concatenation of this
4640 * string repeated {@code count} times.
4641 * <p>
4642 * If this string is empty or count is zero then the empty
4643 * string is returned.
4644 *
4645 * @param count number of times to repeat
4646 *
4647 * @return A string composed of this string repeated
4648 * {@code count} times or the empty string if this
4649 * string is empty or count is zero
4650 *
4651 * @throws IllegalArgumentException if the {@code count} is
4652 * negative.
4653 *
4654 * @since 11
4655 */
4656 public String repeat(int count) {
4657 if (count < 0) {
4658 throw new IllegalArgumentException("count is negative: " + count);
4659 }
4660 if (count == 1) {
4661 return this;
4662 }
4663 final int len = value.length;
4664 if (len == 0 || count == 0) {
4665 return "";
4666 }
4667 if (Integer.MAX_VALUE / count < len) {
4668 throw new OutOfMemoryError("Required length exceeds implementation limit");
4669 }
4670 if (len == 1) {
4671 final byte[] single = new byte[count];
4672 Arrays.fill(single, value[0]);
4673 return new String(single, coder);
4674 }
4675 final int limit = len * count;
4676 final byte[] multiple = new byte[limit];
4677 System.arraycopy(value, 0, multiple, 0, len);
4678 repeatCopyRest(multiple, 0, limit, len);
4679 return new String(multiple, coder);
4680 }
4681
4682 /**
4683 * Used to perform copying after the initial insertion. Copying is optimized
4684 * by using power of two duplication. First pass duplicates original copy,
4685 * second pass then duplicates the original and the copy yielding four copies,
4686 * third pass duplicates four copies yielding eight copies, and so on.
4687 * Finally, the remainder is filled in with prior copies.
4688 *
4689 * @implNote The technique used here is significantly faster than hand-rolled
4690 * loops or special casing small numbers due to the intensive optimization
4691 * done by intrinsic {@code System.arraycopy}.
4692 *
4693 * @param buffer destination buffer
4694 * @param offset offset in the destination buffer
4695 * @param limit total replicated including what is already in the buffer
4696 * @param copied number of bytes that have already in the buffer
4697 */
4698 static void repeatCopyRest(byte[] buffer, int offset, int limit, int copied) {
4699 // Initial copy is in the buffer.
4700 for (; copied < limit - copied; copied <<= 1) {
4701 // Power of two duplicate.
4702 System.arraycopy(buffer, offset, buffer, offset + copied, copied);
4703 }
4704 // Duplicate remainder.
4705 System.arraycopy(buffer, offset, buffer, offset + copied, limit - copied);
4706 }
4707
4708 ////////////////////////////////////////////////////////////////
4709
4710 /**
4711 * Copy character bytes from this string into dst starting at dstBegin.
4712 * This method doesn't perform any range checking.
4713 *
4714 * Invoker guarantees: dst is in UTF16 (inflate itself for asb), if two
4715 * coders are different, and dst is big enough (range check)
4716 *
4717 * @param dstBegin the char index, not offset of byte[]
4718 * @param coder the coder of dst[]
4719 */
4720 void getBytes(byte[] dst, int dstBegin, byte coder) {
4721 if (coder() == coder) {
4722 System.arraycopy(value, 0, dst, dstBegin << coder, value.length);
4723 } else { // this.coder == LATIN && coder == UTF16
4724 StringLatin1.inflate(value, 0, dst, dstBegin, value.length);
4725 }
4726 }
4727
4728 /**
4729 * Copy character bytes from this string into dst starting at dstBegin.
4730 * This method doesn't perform any range checking.
4731 *
4732 * Invoker guarantees: dst is in UTF16 (inflate itself for asb), if two
4733 * coders are different, and dst is big enough (range check)
4734 *
4735 * @param srcPos the char index, not offset of byte[]
4736 * @param dstBegin the char index to start from
4737 * @param coder the coder of dst[]
4738 * @param length the amount of copied chars
4739 */
4740 void getBytes(byte[] dst, int srcPos, int dstBegin, byte coder, int length) {
4741 if (coder() == coder) {
4742 System.arraycopy(value, srcPos << coder, dst, dstBegin << coder, length << coder);
4743 } else { // this.coder == LATIN && coder == UTF16
4744 StringLatin1.inflate(value, srcPos, dst, dstBegin, length);
4745 }
4746 }
4747
4748 /*
4749 * Package private constructor. Trailing Void argument is there for
4750 * disambiguating it against other (public) constructors.
4751 *
4752 * Stores the char[] value into a byte[] that each byte represents
4753 * the8 low-order bits of the corresponding character, if the char[]
4754 * contains only latin1 character. Or a byte[] that stores all
4755 * characters in their byte sequences defined by the {@code StringUTF16}.
4756 */
4757 String(char[] value, int off, int len, Void sig) {
4758 if (len == 0) {
4759 this.value = "".value;
4760 this.coder = "".coder;
4761 return;
4762 }
4763 if (COMPACT_STRINGS) {
4764 byte[] val = StringUTF16.compress(value, off, len);
4765 if (val != null) {
4766 this.value = val;
4767 this.coder = LATIN1;
4768 return;
4769 }
4770 }
4771 this.coder = UTF16;
4772 this.value = StringUTF16.toBytes(value, off, len);
4773 }
4774
4775 /*
4776 * Package private constructor. Trailing Void argument is there for
4777 * disambiguating it against other (public) constructors.
4778 */
4779 String(AbstractStringBuilder asb, Void sig) {
4780 byte[] val = asb.getValue();
4781 int length = asb.length();
4782 if (asb.isLatin1()) {
4783 this.coder = LATIN1;
4784 this.value = Arrays.copyOfRange(val, 0, length);
4785 } else {
4786 // only try to compress val if some characters were deleted.
4787 if (COMPACT_STRINGS && asb.maybeLatin1) {
4788 byte[] buf = StringUTF16.compress(val, 0, length);
4789 if (buf != null) {
4790 this.coder = LATIN1;
4791 this.value = buf;
4792 return;
4793 }
4794 }
4795 this.coder = UTF16;
4796 this.value = Arrays.copyOfRange(val, 0, length << 1);
4797 }
4798 }
4799
4800 /*
4801 * Package private constructor which shares value array for speed.
4802 */
4803 String(byte[] value, byte coder) {
4804 this.value = value;
4805 this.coder = coder;
4806 }
4807
4808 byte coder() {
4809 return COMPACT_STRINGS ? coder : UTF16;
4810 }
4811
4812 byte[] value() {
4813 return value;
4814 }
4815
4816 boolean isLatin1() {
4817 return COMPACT_STRINGS && coder == LATIN1;
4818 }
4819
4820 @Native static final byte LATIN1 = 0;
4821 @Native static final byte UTF16 = 1;
4822
4823 /*
4824 * StringIndexOutOfBoundsException if {@code index} is
4825 * negative or greater than or equal to {@code length}.
4826 */
4827 static void checkIndex(int index, int length) {
4828 Preconditions.checkIndex(index, length, Preconditions.SIOOBE_FORMATTER);
4829 }
4830
4831 /*
4832 * StringIndexOutOfBoundsException if {@code offset}
4833 * is negative or greater than {@code length}.
4834 */
4835 static void checkOffset(int offset, int length) {
4836 Preconditions.checkFromToIndex(offset, length, length, Preconditions.SIOOBE_FORMATTER);
4837 }
4838
4839 /*
4840 * Check {@code offset}, {@code count} against {@code 0} and {@code length}
4841 * bounds.
4842 *
4843 * @return {@code offset} if the sub-range within bounds of the range
4844 * @throws StringIndexOutOfBoundsException
4845 * If {@code offset} is negative, {@code count} is negative,
4846 * or {@code offset} is greater than {@code length - count}
4847 */
4848 static int checkBoundsOffCount(int offset, int count, int length) {
4849 return Preconditions.checkFromIndexSize(offset, count, length, Preconditions.SIOOBE_FORMATTER);
4850 }
4851
4852 /*
4853 * Check {@code begin}, {@code end} against {@code 0} and {@code length}
4854 * bounds.
4855 *
4856 * @throws StringIndexOutOfBoundsException
4857 * If {@code begin} is negative, {@code begin} is greater than
4858 * {@code end}, or {@code end} is greater than {@code length}.
4859 */
4860 static void checkBoundsBeginEnd(int begin, int end, int length) {
4861 Preconditions.checkFromToIndex(begin, end, length, Preconditions.SIOOBE_FORMATTER);
4862 }
4863
4864 /**
4865 * Returns the string representation of the {@code codePoint}
4866 * argument.
4867 *
4868 * @param codePoint a {@code codePoint}.
4869 * @return a string of length {@code 1} or {@code 2} containing
4870 * as its single character the argument {@code codePoint}.
4871 * @throws IllegalArgumentException if the specified
4872 * {@code codePoint} is not a {@linkplain Character#isValidCodePoint
4873 * valid Unicode code point}.
4874 */
4875 static String valueOfCodePoint(int codePoint) {
4876 if (COMPACT_STRINGS && StringLatin1.canEncode(codePoint)) {
4877 return new String(StringLatin1.toBytes((char)codePoint), LATIN1);
4878 } else if (Character.isBmpCodePoint(codePoint)) {
4879 return new String(StringUTF16.toBytes((char)codePoint), UTF16);
4880 } else if (Character.isSupplementaryCodePoint(codePoint)) {
4881 return new String(StringUTF16.toBytesSupplementary(codePoint), UTF16);
4882 }
4883
4884 throw new IllegalArgumentException(
4885 format("Not a valid Unicode code point: 0x%X", codePoint));
4886 }
4887
4888 /**
4889 * Returns an {@link Optional} containing the nominal descriptor for this
4890 * instance, which is the instance itself.
4891 *
4892 * @return an {@link Optional} describing the {@linkplain String} instance
4893 * @since 12
4894 */
4895 @Override
4896 public Optional<String> describeConstable() {
4897 return Optional.of(this);
4898 }
4899
4900 /**
4901 * Resolves this instance as a {@link ConstantDesc}, the result of which is
4902 * the instance itself.
4903 *
4904 * @param lookup ignored
4905 * @return the {@linkplain String} instance
4906 * @since 12
4907 */
4908 @Override
4909 public String resolveConstantDesc(MethodHandles.Lookup lookup) {
4910 return this;
4911 }
4912
4913 }