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.foreign.MemorySegment;
32 import java.lang.foreign.ValueLayout;
33 import java.lang.invoke.MethodHandles;
34 import java.lang.constant.Constable;
35 import java.lang.constant.ConstantDesc;
36 import java.nio.ByteBuffer;
37 import java.nio.CharBuffer;
38 import java.nio.charset.*;
39 import java.util.ArrayList;
40 import java.util.Arrays;
41 import java.util.Comparator;
42 import java.util.Formatter;
43 import java.util.List;
44 import java.util.Locale;
45 import java.util.Objects;
46 import java.util.Optional;
47 import java.util.Spliterator;
48 import java.util.function.Function;
49 import java.util.regex.Pattern;
50 import java.util.regex.PatternSyntaxException;
51 import java.util.stream.Collectors;
52 import java.util.stream.IntStream;
53 import java.util.stream.Stream;
54 import java.util.stream.StreamSupport;
55
56 import jdk.internal.util.ArraysSupport;
57 import jdk.internal.util.Preconditions;
58 import jdk.internal.vm.annotation.ForceInline;
59 import jdk.internal.vm.annotation.IntrinsicCandidate;
60 import jdk.internal.vm.annotation.Stable;
61 import sun.nio.cs.ArrayDecoder;
62 import sun.nio.cs.ArrayEncoder;
63
64 import sun.nio.cs.ISO_8859_1;
65 import sun.nio.cs.US_ASCII;
66 import sun.nio.cs.UTF_8;
67
68 /**
69 * The {@code String} class represents character strings. All
70 * string literals in Java programs, such as {@code "abc"}, are
71 * implemented as instances of this class.
72 * <p>
73 * Strings are constant; their values cannot be changed after they
74 * are created. String buffers support mutable strings.
75 * Because String objects are immutable they can be shared. For example:
76 * <blockquote><pre>
77 * String str = "abc";
78 * </pre></blockquote><p>
79 * is equivalent to:
80 * <blockquote><pre>
81 * char data[] = {'a', 'b', 'c'};
82 * String str = new String(data);
83 * </pre></blockquote><p>
84 * Here are some more examples of how strings can be used:
85 * <blockquote><pre>
86 * System.out.println("abc");
87 * String cde = "cde";
88 * System.out.println("abc" + cde);
89 * String c = "abc".substring(2, 3);
90 * String d = cde.substring(1, 2);
91 * </pre></blockquote>
92 * <p>
93 * The class {@code String} includes methods for examining
94 * individual characters of the sequence, for comparing strings, for
95 * searching strings, for extracting substrings, and for creating a
96 * copy of a string with all characters translated to uppercase or to
97 * lowercase. Case mapping is based on the Unicode Standard version
98 * specified by the {@link java.lang.Character Character} class.
99 * <p>
100 * The Java language provides special support for the string
101 * concatenation operator ( + ), and for conversion of
102 * other objects to strings. For additional information on string
103 * concatenation and conversion, see <i>The Java Language Specification</i>.
104 *
105 * <p> Unless otherwise noted, passing a {@code null} argument to a constructor
106 * or method in this class will cause a {@link NullPointerException} to be
107 * thrown.
108 *
109 * <p>A {@code String} represents a string in the UTF-16 format
110 * in which <em>supplementary characters</em> are represented by <em>surrogate
111 * pairs</em> (see the section <a href="Character.html#unicode">Unicode
112 * Character Representations</a> in the {@code Character} class for
113 * more information).
114 * Index values refer to {@code char} code units, so a supplementary
115 * character uses two positions in a {@code String}.
116 * <p>The {@code String} class provides methods for dealing with
117 * Unicode code points (i.e., characters), in addition to those for
118 * dealing with Unicode code units (i.e., {@code char} values).
119 *
120 * <p>Unless otherwise noted, methods for comparing Strings do not take locale
121 * into account. The {@link java.text.Collator} class provides methods for
122 * finer-grain, locale-sensitive String comparison.
123 *
124 * @implNote The implementation of the string concatenation operator is left to
125 * the discretion of a Java compiler, as long as the compiler ultimately conforms
126 * to <i>The Java Language Specification</i>. For example, the {@code javac} compiler
127 * may implement the operator with {@code StringBuffer}, {@code StringBuilder},
128 * or {@code java.lang.invoke.StringConcatFactory} depending on the JDK version. The
129 * implementation of string conversion is typically through the method {@code toString},
130 * defined by {@code Object} and inherited by all classes in Java.
131 *
132 * @author Lee Boynton
133 * @author Arthur van Hoff
134 * @author Martin Buchholz
135 * @author Ulf Zibis
136 * @see java.lang.Object#toString()
137 * @see java.lang.StringBuffer
138 * @see java.lang.StringBuilder
139 * @see java.nio.charset.Charset
140 * @since 1.0
141 * @jls 15.18.1 String Concatenation Operator +
142 */
143
144 public final class String
145 implements java.io.Serializable, Comparable<String>, CharSequence,
146 Constable, ConstantDesc {
147
148 /**
149 * The value is used for character storage.
150 *
151 * @implNote This field is trusted by the VM, and is a subject to
152 * constant folding if String instance is constant. Overwriting this
153 * field after construction will cause problems.
154 *
155 * Additionally, it is marked with {@link Stable} to trust the contents
156 * of the array. No other facility in JDK provides this functionality (yet).
157 * {@link Stable} is safe here, because value is never null.
158 */
159 @Stable
160 private final byte[] value;
161
162 /**
163 * The identifier of the encoding used to encode the bytes in
164 * {@code value}. The supported values in this implementation are
165 *
166 * LATIN1
167 * UTF16
168 *
169 * @implNote This field is trusted by the VM, and is a subject to
170 * constant folding if String instance is constant. Overwriting this
171 * field after construction will cause problems.
172 */
173 private final byte coder;
174
175 /** Cache the hash code for the string */
176 private int hash; // Default to 0
177
178 /**
179 * Cache if the hash has been calculated as actually being zero, enabling
180 * us to avoid recalculating this.
181 */
182 private boolean hashIsZero; // Default to false;
183
184 /** use serialVersionUID from JDK 1.0.2 for interoperability */
185 @java.io.Serial
186 private static final long serialVersionUID = -6849794470754667710L;
187
188 /**
189 * If String compaction is disabled, the bytes in {@code value} are
190 * always encoded in UTF16.
191 *
192 * For methods with several possible implementation paths, when String
193 * compaction is disabled, only one code path is taken.
194 *
195 * The instance field value is generally opaque to optimizing JIT
196 * compilers. Therefore, in performance-sensitive place, an explicit
197 * check of the static boolean {@code COMPACT_STRINGS} is done first
198 * before checking the {@code coder} field since the static boolean
199 * {@code COMPACT_STRINGS} would be constant folded away by an
200 * optimizing JIT compiler. The idioms for these cases are as follows.
201 *
202 * For code such as:
203 *
204 * if (coder == LATIN1) { ... }
205 *
206 * can be written more optimally as
207 *
208 * if (coder() == LATIN1) { ... }
209 *
210 * or:
211 *
212 * if (COMPACT_STRINGS && coder == LATIN1) { ... }
213 *
214 * An optimizing JIT compiler can fold the above conditional as:
215 *
216 * COMPACT_STRINGS == true => if (coder == LATIN1) { ... }
217 * COMPACT_STRINGS == false => if (false) { ... }
218 *
219 * @implNote
220 * The actual value for this field is injected by JVM. The static
221 * initialization block is used to set the value here to communicate
222 * that this static final field is not statically foldable, and to
223 * avoid any possible circular dependency during vm initialization.
224 */
225 static final boolean COMPACT_STRINGS;
226
227 static {
228 COMPACT_STRINGS = true;
229 }
230
231 /**
232 * Class String is special cased within the Serialization Stream Protocol.
233 *
234 * A String instance is written into an ObjectOutputStream according to
235 * <a href="{@docRoot}/../specs/serialization/protocol.html#stream-elements">
236 * <cite>Java Object Serialization Specification</cite>, Section 6.2, "Stream Elements"</a>
237 */
238 @java.io.Serial
239 private static final ObjectStreamField[] serialPersistentFields =
240 new ObjectStreamField[0];
241
242 /**
243 * Initializes a newly created {@code String} object so that it represents
244 * an empty character sequence. Note that use of this constructor is
245 * unnecessary since Strings are immutable.
246 */
247 public String() {
248 this.value = "".value;
249 this.coder = "".coder;
250 }
251
252 /**
253 * Initializes a newly created {@code String} object so that it represents
254 * the same sequence of characters as the argument; in other words, the
255 * newly created string is a copy of the argument string. Unless an
256 * explicit copy of {@code original} is needed, use of this constructor is
257 * unnecessary since Strings are immutable.
258 *
259 * @param original
260 * A {@code String}
261 */
262 @IntrinsicCandidate
263 public String(String original) {
264 this.value = original.value;
265 this.coder = original.coder;
266 this.hash = original.hash;
267 this.hashIsZero = original.hashIsZero;
268 }
269
270 /**
271 * Allocates a new {@code String} so that it represents the sequence of
272 * characters currently contained in the character array argument. The
273 * contents of the character array are copied; subsequent modification of
274 * the character array does not affect the newly created string.
275 *
276 * @param value
277 * The initial value of the string
278 */
279 public String(char[] value) {
280 this(value, 0, value.length, null);
281 }
282
283 /**
284 * Allocates a new {@code String} that contains characters from a subarray
285 * of the character array argument. The {@code offset} argument is the
286 * index of the first character of the subarray and the {@code count}
287 * argument specifies the length of the subarray. The contents of the
288 * subarray are copied; subsequent modification of the character array does
289 * not affect the newly created string.
290 *
291 * @param value
292 * Array that is the source of characters
293 *
294 * @param offset
295 * The initial offset
296 *
297 * @param count
298 * The length
299 *
300 * @throws IndexOutOfBoundsException
301 * If {@code offset} is negative, {@code count} is negative, or
302 * {@code offset} is greater than {@code value.length - count}
303 */
304 public String(char[] value, int offset, int count) {
305 this(value, offset, count, rangeCheck(value, offset, count));
306 }
307
308 private static Void rangeCheck(char[] value, int offset, int count) {
309 checkBoundsOffCount(offset, count, value.length);
310 return null;
311 }
312
313 /**
314 * Allocates a new {@code String} that contains characters from a subarray
315 * of the <a href="Character.html#unicode">Unicode code point</a> array
316 * argument. The {@code offset} argument is the index of the first code
317 * point of the subarray and the {@code count} argument specifies the
318 * length of the subarray. The contents of the subarray are converted to
319 * {@code char}s; subsequent modification of the {@code int} array does not
320 * affect the newly created string.
321 *
322 * @param codePoints
323 * Array that is the source of Unicode code points
324 *
325 * @param offset
326 * The initial offset
327 *
328 * @param count
329 * The length
330 *
331 * @throws IllegalArgumentException
332 * If any invalid Unicode code point is found in {@code
333 * codePoints}
334 *
335 * @throws IndexOutOfBoundsException
336 * If {@code offset} is negative, {@code count} is negative, or
337 * {@code offset} is greater than {@code codePoints.length - count}
338 *
339 * @since 1.5
340 */
341 public String(int[] codePoints, int offset, int count) {
342 checkBoundsOffCount(offset, count, codePoints.length);
343 if (count == 0) {
344 this.value = "".value;
345 this.coder = "".coder;
346 return;
347 }
348 if (COMPACT_STRINGS) {
349 byte[] val = StringLatin1.toBytes(codePoints, offset, count);
350 if (val != null) {
351 this.coder = LATIN1;
352 this.value = val;
353 return;
354 }
355 }
356 this.coder = UTF16;
357 this.value = StringUTF16.toBytes(codePoints, offset, count);
358 }
359
360 /**
361 * Allocates a new {@code String} constructed from a subarray of an array
362 * of 8-bit integer values.
363 *
364 * <p> The {@code offset} argument is the index of the first byte of the
365 * subarray, and the {@code count} argument specifies the length of the
366 * subarray.
367 *
368 * <p> Each {@code byte} in the subarray is converted to a {@code char} as
369 * specified in the {@link #String(byte[],int) String(byte[],int)} constructor.
370 *
371 * @deprecated This method does not properly convert bytes into characters.
372 * As of JDK 1.1, the preferred way to do this is via the
373 * {@code String} constructors that take a {@link Charset}, charset name,
374 * or that use the {@link Charset#defaultCharset() default charset}.
375 *
376 * @param ascii
377 * The bytes to be converted to characters
378 *
379 * @param hibyte
380 * The top 8 bits of each 16-bit Unicode code unit
381 *
382 * @param offset
383 * The initial offset
384 * @param count
385 * The length
386 *
387 * @throws IndexOutOfBoundsException
388 * If {@code offset} is negative, {@code count} is negative, or
389 * {@code offset} is greater than {@code ascii.length - count}
390 *
391 * @see #String(byte[], int)
392 * @see #String(byte[], int, int, java.lang.String)
393 * @see #String(byte[], int, int, java.nio.charset.Charset)
394 * @see #String(byte[], int, int)
395 * @see #String(byte[], java.lang.String)
396 * @see #String(byte[], java.nio.charset.Charset)
397 * @see #String(byte[])
398 */
399 @Deprecated(since="1.1")
400 public String(byte[] ascii, int hibyte, int offset, int count) {
401 checkBoundsOffCount(offset, count, ascii.length);
402 if (count == 0) {
403 this.value = "".value;
404 this.coder = "".coder;
405 return;
406 }
407 if (COMPACT_STRINGS && (byte)hibyte == 0) {
408 this.value = Arrays.copyOfRange(ascii, offset, offset + count);
409 this.coder = LATIN1;
410 } else {
411 hibyte <<= 8;
412 byte[] val = StringUTF16.newBytesFor(count);
413 for (int i = 0; i < count; i++) {
414 StringUTF16.putChar(val, i, hibyte | (ascii[offset++] & 0xff));
415 }
416 this.value = val;
417 this.coder = UTF16;
418 }
419 }
420
421 /**
422 * Allocates a new {@code String} containing characters constructed from
423 * an array of 8-bit integer values. Each character <i>c</i> in the
424 * resulting string is constructed from the corresponding component
425 * <i>b</i> in the byte array such that:
426 *
427 * <blockquote><pre>
428 * <b><i>c</i></b> == (char)(((hibyte & 0xff) << 8)
429 * | (<b><i>b</i></b> & 0xff))
430 * </pre></blockquote>
431 *
432 * @deprecated This method does not properly convert bytes into
433 * characters. As of JDK 1.1, the preferred way to do this is via the
434 * {@code String} constructors that take a {@link Charset}, charset name,
435 * or that use the {@link Charset#defaultCharset() default charset}.
436 *
437 * @param ascii
438 * The bytes to be converted to characters
439 *
440 * @param hibyte
441 * The top 8 bits of each 16-bit Unicode code unit
442 *
443 * @see #String(byte[], int, int, java.lang.String)
444 * @see #String(byte[], int, int, java.nio.charset.Charset)
445 * @see #String(byte[], int, int)
446 * @see #String(byte[], java.lang.String)
447 * @see #String(byte[], java.nio.charset.Charset)
448 * @see #String(byte[])
449 */
450 @Deprecated(since="1.1")
451 public String(byte[] ascii, int hibyte) {
452 this(ascii, hibyte, 0, ascii.length);
453 }
454
455 /**
456 * Constructs a new {@code String} by decoding the specified subarray of
457 * bytes using the specified charset. The length of the new {@code String}
458 * is a function of the charset, and hence may not be equal to the length
459 * of the subarray.
460 *
461 * <p> The behavior of this constructor when the given bytes are not valid
462 * in the given charset is unspecified. The {@link
463 * java.nio.charset.CharsetDecoder} class should be used when more control
464 * over the decoding process is required.
465 *
466 * @param bytes
467 * The bytes to be decoded into characters
468 *
469 * @param offset
470 * The index of the first byte to decode
471 *
472 * @param length
473 * The number of bytes to decode
474 *
475 * @param charsetName
476 * The name of a supported {@linkplain java.nio.charset.Charset
477 * charset}
478 *
479 * @throws UnsupportedEncodingException
480 * If the named charset is not supported
481 *
482 * @throws IndexOutOfBoundsException
483 * If {@code offset} is negative, {@code length} is negative, or
484 * {@code offset} is greater than {@code bytes.length - length}
485 *
486 * @since 1.1
487 */
488 public String(byte[] bytes, int offset, int length, String charsetName)
489 throws UnsupportedEncodingException {
490 this(lookupCharset(charsetName), bytes, checkBoundsOffCount(offset, length, bytes.length), length);
491 }
492
493 /**
494 * Constructs a new {@code String} by decoding the specified subarray of
495 * bytes using the specified {@linkplain java.nio.charset.Charset charset}.
496 * The length of the new {@code String} is a function of the charset, and
497 * hence may not be equal to the length of the subarray.
498 *
499 * <p> This method always replaces malformed-input and unmappable-character
500 * sequences with this charset's default replacement string. The {@link
501 * java.nio.charset.CharsetDecoder} class should be used when more control
502 * over the decoding process is required.
503 *
504 * @param bytes
505 * The bytes to be decoded into characters
506 *
507 * @param offset
508 * The index of the first byte to decode
509 *
510 * @param length
511 * The number of bytes to decode
512 *
513 * @param charset
514 * The {@linkplain java.nio.charset.Charset charset} to be used to
515 * decode the {@code bytes}
516 *
517 * @throws IndexOutOfBoundsException
518 * If {@code offset} is negative, {@code length} is negative, or
519 * {@code offset} is greater than {@code bytes.length - length}
520 *
521 * @since 1.6
522 */
523 public String(byte[] bytes, int offset, int length, Charset charset) {
524 this(Objects.requireNonNull(charset), bytes, checkBoundsOffCount(offset, length, bytes.length), length);
525 }
526
527 /**
528 * This method does not do any precondition checks on its arguments.
529 * <p>
530 * Important: parameter order of this method is deliberately changed in order to
531 * disambiguate it against other similar methods of this class.
532 */
533 @SuppressWarnings("removal")
534 private String(Charset charset, byte[] bytes, int offset, int length) {
535 if (length == 0) {
536 this.value = "".value;
537 this.coder = "".coder;
538 } else if (charset == UTF_8.INSTANCE) {
539 if (COMPACT_STRINGS) {
540 int dp = StringCoding.countPositives(bytes, offset, length);
541 if (dp == length) {
542 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
543 this.coder = LATIN1;
544 return;
545 }
546 int sl = offset + length;
547 byte[] dst = new byte[length];
548 if (dp > 0) {
549 System.arraycopy(bytes, offset, dst, 0, dp);
550 offset += dp;
551 }
552 while (offset < sl) {
553 int b1 = bytes[offset++];
554 if (b1 >= 0) {
555 dst[dp++] = (byte)b1;
556 continue;
557 }
558 if ((b1 & 0xfe) == 0xc2 && offset < sl) { // b1 either 0xc2 or 0xc3
559 int b2 = bytes[offset];
560 if (b2 < -64) { // continuation bytes are always negative values in the range -128 to -65
561 dst[dp++] = (byte)decode2(b1, b2);
562 offset++;
563 continue;
564 }
565 }
566 // anything not a latin1, including the REPL
567 // we have to go with the utf16
568 offset--;
569 break;
570 }
571 if (offset == sl) {
572 if (dp != dst.length) {
573 dst = Arrays.copyOf(dst, dp);
574 }
575 this.value = dst;
576 this.coder = LATIN1;
577 return;
578 }
579 byte[] buf = new byte[length << 1];
580 StringLatin1.inflate(dst, 0, buf, 0, dp);
581 dst = buf;
582 dp = decodeUTF8_UTF16(bytes, offset, sl, dst, dp, true);
583 if (dp != length) {
584 dst = Arrays.copyOf(dst, dp << 1);
585 }
586 this.value = dst;
587 this.coder = UTF16;
588 } else { // !COMPACT_STRINGS
589 byte[] dst = new byte[length << 1];
590 int dp = decodeUTF8_UTF16(bytes, offset, offset + length, dst, 0, true);
591 if (dp != length) {
592 dst = Arrays.copyOf(dst, dp << 1);
593 }
594 this.value = dst;
595 this.coder = UTF16;
596 }
597 } else if (charset == ISO_8859_1.INSTANCE) {
598 if (COMPACT_STRINGS) {
599 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
600 this.coder = LATIN1;
601 } else {
602 this.value = StringLatin1.inflate(bytes, offset, length);
603 this.coder = UTF16;
604 }
605 } else if (charset == US_ASCII.INSTANCE) {
606 if (COMPACT_STRINGS && !StringCoding.hasNegatives(bytes, offset, length)) {
607 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
608 this.coder = LATIN1;
609 } else {
610 byte[] dst = new byte[length << 1];
611 int dp = 0;
612 while (dp < length) {
613 int b = bytes[offset++];
614 StringUTF16.putChar(dst, dp++, (b >= 0) ? (char) b : REPL);
615 }
616 this.value = dst;
617 this.coder = UTF16;
618 }
619 } else {
620 // (1)We never cache the "external" cs, the only benefit of creating
621 // an additional StringDe/Encoder object to wrap it is to share the
622 // de/encode() method. These SD/E objects are short-lived, the young-gen
623 // gc should be able to take care of them well. But the best approach
624 // is still not to generate them if not really necessary.
625 // (2)The defensive copy of the input byte/char[] has a big performance
626 // impact, as well as the outgoing result byte/char[]. Need to do the
627 // optimization check of (sm==null && classLoader0==null) for both.
628 CharsetDecoder cd = charset.newDecoder();
629 // ArrayDecoder fastpaths
630 if (cd instanceof ArrayDecoder ad) {
631 // ascii
632 if (ad.isASCIICompatible() && !StringCoding.hasNegatives(bytes, offset, length)) {
633 if (COMPACT_STRINGS) {
634 this.value = Arrays.copyOfRange(bytes, offset, offset + length);
635 this.coder = LATIN1;
636 return;
637 }
638 this.value = StringLatin1.inflate(bytes, offset, length);
639 this.coder = UTF16;
640 return;
641 }
642
643 // fastpath for always Latin1 decodable single byte
644 if (COMPACT_STRINGS && ad.isLatin1Decodable()) {
645 byte[] dst = new byte[length];
646 ad.decodeToLatin1(bytes, offset, length, dst);
647 this.value = dst;
648 this.coder = LATIN1;
649 return;
650 }
651
652 int en = scale(length, cd.maxCharsPerByte());
653 cd.onMalformedInput(CodingErrorAction.REPLACE)
654 .onUnmappableCharacter(CodingErrorAction.REPLACE);
655 char[] ca = new char[en];
656 int clen = ad.decode(bytes, offset, length, ca);
657 if (COMPACT_STRINGS) {
658 byte[] bs = StringUTF16.compress(ca, 0, clen);
659 if (bs != null) {
660 value = bs;
661 coder = LATIN1;
662 return;
663 }
664 }
665 coder = UTF16;
666 value = StringUTF16.toBytes(ca, 0, clen);
667 return;
668 }
669
670 // decode using CharsetDecoder
671 int en = scale(length, cd.maxCharsPerByte());
672 cd.onMalformedInput(CodingErrorAction.REPLACE)
673 .onUnmappableCharacter(CodingErrorAction.REPLACE);
674 char[] ca = new char[en];
675 if (charset.getClass().getClassLoader0() != null &&
676 System.getSecurityManager() != null) {
677 bytes = Arrays.copyOfRange(bytes, offset, offset + length);
678 offset = 0;
679 }
680
681 int caLen;
682 try {
683 caLen = decodeWithDecoder(cd, ca, bytes, offset, length);
684 } catch (CharacterCodingException x) {
685 // Substitution is enabled, so this shouldn't happen
686 throw new Error(x);
687 }
688 if (COMPACT_STRINGS) {
689 byte[] bs = StringUTF16.compress(ca, 0, caLen);
690 if (bs != null) {
691 value = bs;
692 coder = LATIN1;
693 return;
694 }
695 }
696 coder = UTF16;
697 value = StringUTF16.toBytes(ca, 0, caLen);
698 }
699 }
700
701 /*
702 * Throws iae, instead of replacing, if malformed or unmappable.
703 *
704 * @param noShare
705 * {@code true} if the resulting string MUST NOT share the byte array,
706 * {@code false} if the byte array can be exclusively used to construct
707 * the string and is not modified or used for any other purpose.
708 */
709 static String newStringUTF8NoRepl(byte[] bytes, int offset, int length, boolean noShare) {
710 checkBoundsOffCount(offset, length, bytes.length);
711 if (length == 0) {
712 return "";
713 }
714 int dp;
715 byte[] dst;
716 if (COMPACT_STRINGS) {
717 dp = StringCoding.countPositives(bytes, offset, length);
718 int sl = offset + length;
719 if (dp == length) {
720 if (noShare || length != bytes.length) {
721 return new String(Arrays.copyOfRange(bytes, offset, offset + length), LATIN1);
722 } else {
723 return new String(bytes, LATIN1);
724 }
725 }
726 dst = new byte[length];
727 System.arraycopy(bytes, offset, dst, 0, dp);
728 offset += dp;
729 while (offset < sl) {
730 int b1 = bytes[offset++];
731 if (b1 >= 0) {
732 dst[dp++] = (byte)b1;
733 continue;
734 }
735 if ((b1 & 0xfe) == 0xc2 && offset < sl) { // b1 either 0xc2 or 0xc3
736 int b2 = bytes[offset];
737 if (b2 < -64) { // continuation bytes are always negative values in the range -128 to -65
738 dst[dp++] = (byte)decode2(b1, b2);
739 offset++;
740 continue;
741 }
742 }
743 // anything not a latin1, including the REPL
744 // we have to go with the utf16
745 offset--;
746 break;
747 }
748 if (offset == sl) {
749 if (dp != dst.length) {
750 dst = Arrays.copyOf(dst, dp);
751 }
752 return new String(dst, LATIN1);
753 }
754 if (dp == 0) {
755 dst = new byte[length << 1];
756 } else {
757 byte[] buf = new byte[length << 1];
758 StringLatin1.inflate(dst, 0, buf, 0, dp);
759 dst = buf;
760 }
761 dp = decodeUTF8_UTF16(bytes, offset, sl, dst, dp, false);
762 } else { // !COMPACT_STRINGS
763 dst = new byte[length << 1];
764 dp = decodeUTF8_UTF16(bytes, offset, offset + length, dst, 0, false);
765 }
766 if (dp != length) {
767 dst = Arrays.copyOf(dst, dp << 1);
768 }
769 return new String(dst, UTF16);
770 }
771
772 static String newStringNoRepl(byte[] src, Charset cs) throws CharacterCodingException {
773 try {
774 return newStringNoRepl1(src, cs);
775 } catch (IllegalArgumentException e) {
776 //newStringNoRepl1 throws IAE with MalformedInputException or CCE as the cause
777 Throwable cause = e.getCause();
778 if (cause instanceof MalformedInputException mie) {
779 throw mie;
780 }
781 throw (CharacterCodingException)cause;
782 }
783 }
784
785 @SuppressWarnings("removal")
786 private static String newStringNoRepl1(byte[] src, Charset cs) {
787 int len = src.length;
788 if (len == 0) {
789 return "";
790 }
791 if (cs == UTF_8.INSTANCE) {
792 return newStringUTF8NoRepl(src, 0, src.length, false);
793 }
794 if (cs == ISO_8859_1.INSTANCE) {
795 if (COMPACT_STRINGS)
796 return new String(src, LATIN1);
797 return new String(StringLatin1.inflate(src, 0, src.length), UTF16);
798 }
799 if (cs == US_ASCII.INSTANCE) {
800 if (!StringCoding.hasNegatives(src, 0, src.length)) {
801 if (COMPACT_STRINGS)
802 return new String(src, LATIN1);
803 return new String(StringLatin1.inflate(src, 0, src.length), UTF16);
804 } else {
805 throwMalformed(src);
806 }
807 }
808
809 CharsetDecoder cd = cs.newDecoder();
810 // ascii fastpath
811 if (cd instanceof ArrayDecoder ad &&
812 ad.isASCIICompatible() &&
813 !StringCoding.hasNegatives(src, 0, src.length)) {
814 if (COMPACT_STRINGS)
815 return new String(src, LATIN1);
816 return new String(src, 0, src.length, ISO_8859_1.INSTANCE);
817 }
818 int en = scale(len, cd.maxCharsPerByte());
819 char[] ca = new char[en];
820 if (cs.getClass().getClassLoader0() != null &&
821 System.getSecurityManager() != null) {
822 src = Arrays.copyOf(src, len);
823 }
824 int caLen;
825 try {
826 caLen = decodeWithDecoder(cd, ca, src, 0, src.length);
827 } catch (CharacterCodingException x) {
828 // throw via IAE
829 throw new IllegalArgumentException(x);
830 }
831 if (COMPACT_STRINGS) {
832 byte[] bs = StringUTF16.compress(ca, 0, caLen);
833 if (bs != null) {
834 return new String(bs, LATIN1);
835 }
836 }
837 return new String(StringUTF16.toBytes(ca, 0, caLen), UTF16);
838 }
839
840 private static final char REPL = '\ufffd';
841
842 // Trim the given byte array to the given length
843 @SuppressWarnings("removal")
844 private static byte[] safeTrim(byte[] ba, int len, boolean isTrusted) {
845 if (len == ba.length && (isTrusted || System.getSecurityManager() == null)) {
846 return ba;
847 } else {
848 return Arrays.copyOf(ba, len);
849 }
850 }
851
852 private static int scale(int len, float expansionFactor) {
853 // We need to perform double, not float, arithmetic; otherwise
854 // we lose low order bits when len is larger than 2**24.
855 return (int)(len * (double)expansionFactor);
856 }
857
858 private static Charset lookupCharset(String csn) throws UnsupportedEncodingException {
859 Objects.requireNonNull(csn);
860 try {
861 return Charset.forName(csn);
862 } catch (UnsupportedCharsetException | IllegalCharsetNameException x) {
863 throw new UnsupportedEncodingException(csn);
864 }
865 }
866
867 private static byte[] encode(Charset cs, byte coder, byte[] val) {
868 if (cs == UTF_8.INSTANCE) {
869 return encodeUTF8(coder, val, true);
870 }
871 if (cs == ISO_8859_1.INSTANCE) {
872 return encode8859_1(coder, val);
873 }
874 if (cs == US_ASCII.INSTANCE) {
875 return encodeASCII(coder, val);
876 }
877 return encodeWithEncoder(cs, coder, val, true);
878 }
879
880 private static byte[] encodeWithEncoder(Charset cs, byte coder, byte[] val, boolean doReplace) {
881 CharsetEncoder ce = cs.newEncoder();
882 int len = val.length >> coder; // assume LATIN1=0/UTF16=1;
883 int en = scale(len, ce.maxBytesPerChar());
884 // fastpath with ArrayEncoder implies `doReplace`.
885 if (doReplace && ce instanceof ArrayEncoder ae) {
886 // fastpath for ascii compatible
887 if (coder == LATIN1 &&
888 ae.isASCIICompatible() &&
889 !StringCoding.hasNegatives(val, 0, val.length)) {
890 return val.clone();
891 }
892 byte[] ba = new byte[en];
893 if (len == 0) {
894 return ba;
895 }
896
897 int blen = (coder == LATIN1) ? ae.encodeFromLatin1(val, 0, len, ba)
898 : ae.encodeFromUTF16(val, 0, len, ba);
899 if (blen != -1) {
900 return safeTrim(ba, blen, true);
901 }
902 }
903
904 byte[] ba = new byte[en];
905 if (len == 0) {
906 return ba;
907 }
908 if (doReplace) {
909 ce.onMalformedInput(CodingErrorAction.REPLACE)
910 .onUnmappableCharacter(CodingErrorAction.REPLACE);
911 }
912 char[] ca = (coder == LATIN1 ) ? StringLatin1.toChars(val)
913 : StringUTF16.toChars(val);
914 ByteBuffer bb = ByteBuffer.wrap(ba);
915 CharBuffer cb = CharBuffer.wrap(ca, 0, len);
916 try {
917 CoderResult cr = ce.encode(cb, bb, true);
918 if (!cr.isUnderflow())
919 cr.throwException();
920 cr = ce.flush(bb);
921 if (!cr.isUnderflow())
922 cr.throwException();
923 } catch (CharacterCodingException x) {
924 if (!doReplace) {
925 throw new IllegalArgumentException(x);
926 } else {
927 throw new Error(x);
928 }
929 }
930 return safeTrim(ba, bb.position(), cs.getClass().getClassLoader0() == null);
931 }
932
933 /*
934 * Throws iae, instead of replacing, if unmappable.
935 */
936 static byte[] getBytesUTF8NoRepl(String s) {
937 return encodeUTF8(s.coder(), s.value(), false);
938 }
939
940 private static boolean isASCII(byte[] src) {
941 return !StringCoding.hasNegatives(src, 0, src.length);
942 }
943
944 /*
945 * Throws CCE, instead of replacing, if unmappable.
946 */
947 static byte[] getBytesNoRepl(String s, Charset cs) throws CharacterCodingException {
948 try {
949 return getBytesNoRepl1(s, cs);
950 } catch (IllegalArgumentException e) {
951 //getBytesNoRepl1 throws IAE with UnmappableCharacterException or CCE as the cause
952 Throwable cause = e.getCause();
953 if (cause instanceof UnmappableCharacterException) {
954 throw (UnmappableCharacterException)cause;
955 }
956 throw (CharacterCodingException)cause;
957 }
958 }
959
960 private static byte[] getBytesNoRepl1(String s, Charset cs) {
961 byte[] val = s.value();
962 byte coder = s.coder();
963 if (cs == UTF_8.INSTANCE) {
964 if (coder == LATIN1 && isASCII(val)) {
965 return val;
966 }
967 return encodeUTF8(coder, val, false);
968 }
969 if (cs == ISO_8859_1.INSTANCE) {
970 if (coder == LATIN1) {
971 return val;
972 }
973 return encode8859_1(coder, val, false);
974 }
975 if (cs == US_ASCII.INSTANCE) {
976 if (coder == LATIN1) {
977 if (isASCII(val)) {
978 return val;
979 } else {
980 throwUnmappable(val);
981 }
982 }
983 }
984 return encodeWithEncoder(cs, coder, val, false);
985 }
986
987 private static byte[] encodeASCII(byte coder, byte[] val) {
988 if (coder == LATIN1) {
989 int positives = StringCoding.countPositives(val, 0, val.length);
990 byte[] dst = val.clone();
991 if (positives < dst.length) {
992 replaceNegatives(dst, positives);
993 }
994 return dst;
995 }
996 int len = val.length >> 1;
997 byte[] dst = new byte[len];
998 int dp = 0;
999 for (int i = 0; i < len; i++) {
1000 char c = StringUTF16.getChar(val, i);
1001 if (c < 0x80) {
1002 dst[dp++] = (byte)c;
1003 continue;
1004 }
1005 if (Character.isHighSurrogate(c) && i + 1 < len &&
1006 Character.isLowSurrogate(StringUTF16.getChar(val, i + 1))) {
1007 i++;
1008 }
1009 dst[dp++] = '?';
1010 }
1011 if (len == dp) {
1012 return dst;
1013 }
1014 return Arrays.copyOf(dst, dp);
1015 }
1016
1017 private static void replaceNegatives(byte[] val, int fromIndex) {
1018 for (int i = fromIndex; i < val.length; i++) {
1019 if (val[i] < 0) {
1020 val[i] = '?';
1021 }
1022 }
1023 }
1024
1025 private static byte[] encode8859_1(byte coder, byte[] val) {
1026 return encode8859_1(coder, val, true);
1027 }
1028
1029 private static byte[] encode8859_1(byte coder, byte[] val, boolean doReplace) {
1030 if (coder == LATIN1) {
1031 return val.clone();
1032 }
1033 int len = val.length >> 1;
1034 byte[] dst = new byte[len];
1035 int dp = 0;
1036 int sp = 0;
1037 int sl = len;
1038 while (sp < sl) {
1039 int ret = StringCoding.implEncodeISOArray(val, sp, dst, dp, len);
1040 sp = sp + ret;
1041 dp = dp + ret;
1042 if (ret != len) {
1043 if (!doReplace) {
1044 throwUnmappable(sp);
1045 }
1046 char c = StringUTF16.getChar(val, sp++);
1047 if (Character.isHighSurrogate(c) && sp < sl &&
1048 Character.isLowSurrogate(StringUTF16.getChar(val, sp))) {
1049 sp++;
1050 }
1051 dst[dp++] = '?';
1052 len = sl - sp;
1053 }
1054 }
1055 if (dp == dst.length) {
1056 return dst;
1057 }
1058 return Arrays.copyOf(dst, dp);
1059 }
1060
1061 //////////////////////////////// utf8 ////////////////////////////////////
1062
1063 /**
1064 * Decodes ASCII from the source byte array into the destination
1065 * char array. Used via JavaLangAccess from UTF_8 and other charset
1066 * decoders.
1067 *
1068 * @return the number of bytes successfully decoded, at most len
1069 */
1070 /* package-private */
1071 static int decodeASCII(byte[] sa, int sp, char[] da, int dp, int len) {
1072 int count = StringCoding.countPositives(sa, sp, len);
1073 while (count < len) {
1074 if (sa[sp + count] < 0) {
1075 break;
1076 }
1077 count++;
1078 }
1079 StringLatin1.inflate(sa, sp, da, dp, count);
1080 return count;
1081 }
1082
1083 private static boolean isNotContinuation(int b) {
1084 return (b & 0xc0) != 0x80;
1085 }
1086
1087 private static boolean isMalformed3(int b1, int b2, int b3) {
1088 return (b1 == (byte)0xe0 && (b2 & 0xe0) == 0x80) ||
1089 (b2 & 0xc0) != 0x80 || (b3 & 0xc0) != 0x80;
1090 }
1091
1092 private static boolean isMalformed3_2(int b1, int b2) {
1093 return (b1 == (byte)0xe0 && (b2 & 0xe0) == 0x80) ||
1094 (b2 & 0xc0) != 0x80;
1095 }
1096
1097 private static boolean isMalformed4(int b2, int b3, int b4) {
1098 return (b2 & 0xc0) != 0x80 || (b3 & 0xc0) != 0x80 ||
1099 (b4 & 0xc0) != 0x80;
1100 }
1101
1102 private static boolean isMalformed4_2(int b1, int b2) {
1103 return (b1 == 0xf0 && (b2 < 0x90 || b2 > 0xbf)) ||
1104 (b1 == 0xf4 && (b2 & 0xf0) != 0x80) ||
1105 (b2 & 0xc0) != 0x80;
1106 }
1107
1108 private static boolean isMalformed4_3(int b3) {
1109 return (b3 & 0xc0) != 0x80;
1110 }
1111
1112 private static char decode2(int b1, int b2) {
1113 return (char)(((b1 << 6) ^ b2) ^
1114 (((byte) 0xC0 << 6) ^
1115 ((byte) 0x80 << 0)));
1116 }
1117
1118 private static char decode3(int b1, int b2, int b3) {
1119 return (char)((b1 << 12) ^
1120 (b2 << 6) ^
1121 (b3 ^
1122 (((byte) 0xE0 << 12) ^
1123 ((byte) 0x80 << 6) ^
1124 ((byte) 0x80 << 0))));
1125 }
1126
1127 private static int decode4(int b1, int b2, int b3, int b4) {
1128 return ((b1 << 18) ^
1129 (b2 << 12) ^
1130 (b3 << 6) ^
1131 (b4 ^
1132 (((byte) 0xF0 << 18) ^
1133 ((byte) 0x80 << 12) ^
1134 ((byte) 0x80 << 6) ^
1135 ((byte) 0x80 << 0))));
1136 }
1137
1138 private static int decodeUTF8_UTF16(byte[] src, int sp, int sl, byte[] dst, int dp, boolean doReplace) {
1139 while (sp < sl) {
1140 int b1 = src[sp++];
1141 if (b1 >= 0) {
1142 StringUTF16.putChar(dst, dp++, (char) b1);
1143 } else if ((b1 >> 5) == -2 && (b1 & 0x1e) != 0) {
1144 if (sp < sl) {
1145 int b2 = src[sp++];
1146 if (isNotContinuation(b2)) {
1147 if (!doReplace) {
1148 throwMalformed(sp - 1, 1);
1149 }
1150 StringUTF16.putChar(dst, dp++, REPL);
1151 sp--;
1152 } else {
1153 StringUTF16.putChar(dst, dp++, decode2(b1, b2));
1154 }
1155 continue;
1156 }
1157 if (!doReplace) {
1158 throwMalformed(sp, 1); // underflow()
1159 }
1160 StringUTF16.putChar(dst, dp++, REPL);
1161 break;
1162 } else if ((b1 >> 4) == -2) {
1163 if (sp + 1 < sl) {
1164 int b2 = src[sp++];
1165 int b3 = src[sp++];
1166 if (isMalformed3(b1, b2, b3)) {
1167 if (!doReplace) {
1168 throwMalformed(sp - 3, 3);
1169 }
1170 StringUTF16.putChar(dst, dp++, REPL);
1171 sp -= 3;
1172 sp += malformed3(src, sp);
1173 } else {
1174 char c = decode3(b1, b2, b3);
1175 if (Character.isSurrogate(c)) {
1176 if (!doReplace) {
1177 throwMalformed(sp - 3, 3);
1178 }
1179 StringUTF16.putChar(dst, dp++, REPL);
1180 } else {
1181 StringUTF16.putChar(dst, dp++, c);
1182 }
1183 }
1184 continue;
1185 }
1186 if (sp < sl && isMalformed3_2(b1, src[sp])) {
1187 if (!doReplace) {
1188 throwMalformed(sp - 1, 2);
1189 }
1190 StringUTF16.putChar(dst, dp++, REPL);
1191 continue;
1192 }
1193 if (!doReplace) {
1194 throwMalformed(sp, 1);
1195 }
1196 StringUTF16.putChar(dst, dp++, REPL);
1197 break;
1198 } else if ((b1 >> 3) == -2) {
1199 if (sp + 2 < sl) {
1200 int b2 = src[sp++];
1201 int b3 = src[sp++];
1202 int b4 = src[sp++];
1203 int uc = decode4(b1, b2, b3, b4);
1204 if (isMalformed4(b2, b3, b4) ||
1205 !Character.isSupplementaryCodePoint(uc)) { // shortest form check
1206 if (!doReplace) {
1207 throwMalformed(sp - 4, 4);
1208 }
1209 StringUTF16.putChar(dst, dp++, REPL);
1210 sp -= 4;
1211 sp += malformed4(src, sp);
1212 } else {
1213 StringUTF16.putChar(dst, dp++, Character.highSurrogate(uc));
1214 StringUTF16.putChar(dst, dp++, Character.lowSurrogate(uc));
1215 }
1216 continue;
1217 }
1218 b1 &= 0xff;
1219 if (b1 > 0xf4 || sp < sl && isMalformed4_2(b1, src[sp] & 0xff)) {
1220 if (!doReplace) {
1221 throwMalformed(sp - 1, 1); // or 2
1222 }
1223 StringUTF16.putChar(dst, dp++, REPL);
1224 continue;
1225 }
1226 if (!doReplace) {
1227 throwMalformed(sp - 1, 1);
1228 }
1229 sp++;
1230 StringUTF16.putChar(dst, dp++, REPL);
1231 if (sp < sl && isMalformed4_3(src[sp])) {
1232 continue;
1233 }
1234 break;
1235 } else {
1236 if (!doReplace) {
1237 throwMalformed(sp - 1, 1);
1238 }
1239 StringUTF16.putChar(dst, dp++, REPL);
1240 }
1241 }
1242 return dp;
1243 }
1244
1245 private static int decodeWithDecoder(CharsetDecoder cd, char[] dst, byte[] src, int offset, int length)
1246 throws CharacterCodingException {
1247 ByteBuffer bb = ByteBuffer.wrap(src, offset, length);
1248 CharBuffer cb = CharBuffer.wrap(dst, 0, dst.length);
1249 CoderResult cr = cd.decode(bb, cb, true);
1250 if (!cr.isUnderflow())
1251 cr.throwException();
1252 cr = cd.flush(cb);
1253 if (!cr.isUnderflow())
1254 cr.throwException();
1255 return cb.position();
1256 }
1257
1258 private static int malformed3(byte[] src, int sp) {
1259 int b1 = src[sp++];
1260 int b2 = src[sp]; // no need to lookup b3
1261 return ((b1 == (byte)0xe0 && (b2 & 0xe0) == 0x80) ||
1262 isNotContinuation(b2)) ? 1 : 2;
1263 }
1264
1265 private static int malformed4(byte[] src, int sp) {
1266 // we don't care the speed here
1267 int b1 = src[sp++] & 0xff;
1268 int b2 = src[sp++] & 0xff;
1269 if (b1 > 0xf4 ||
1270 (b1 == 0xf0 && (b2 < 0x90 || b2 > 0xbf)) ||
1271 (b1 == 0xf4 && (b2 & 0xf0) != 0x80) ||
1272 isNotContinuation(b2))
1273 return 1;
1274 if (isNotContinuation(src[sp]))
1275 return 2;
1276 return 3;
1277 }
1278
1279 private static void throwMalformed(int off, int nb) {
1280 String msg = "malformed input off : " + off + ", length : " + nb;
1281 throw new IllegalArgumentException(msg, new MalformedInputException(nb));
1282 }
1283
1284 private static void throwMalformed(byte[] val) {
1285 int dp = StringCoding.countPositives(val, 0, val.length);
1286 throwMalformed(dp, 1);
1287 }
1288
1289 private static void throwUnmappable(int off) {
1290 String msg = "malformed input off : " + off + ", length : 1";
1291 throw new IllegalArgumentException(msg, new UnmappableCharacterException(1));
1292 }
1293
1294 private static void throwUnmappable(byte[] val) {
1295 int dp = StringCoding.countPositives(val, 0, val.length);
1296 throwUnmappable(dp);
1297 }
1298
1299 private static byte[] encodeUTF8(byte coder, byte[] val, boolean doReplace) {
1300 if (coder == UTF16) {
1301 return encodeUTF8_UTF16(val, doReplace);
1302 }
1303
1304 if (!StringCoding.hasNegatives(val, 0, val.length)) {
1305 return val.clone();
1306 }
1307
1308 int dp = 0;
1309 byte[] dst = new byte[val.length << 1];
1310 for (byte c : val) {
1311 if (c < 0) {
1312 dst[dp++] = (byte) (0xc0 | ((c & 0xff) >> 6));
1313 dst[dp++] = (byte) (0x80 | (c & 0x3f));
1314 } else {
1315 dst[dp++] = c;
1316 }
1317 }
1318 if (dp == dst.length) {
1319 return dst;
1320 }
1321 return Arrays.copyOf(dst, dp);
1322 }
1323
1324 private static byte[] encodeUTF8_UTF16(byte[] val, boolean doReplace) {
1325 int dp = 0;
1326 int sp = 0;
1327 int sl = val.length >> 1;
1328 byte[] dst = new byte[sl * 3];
1329 while (sp < sl) {
1330 // ascii fast loop;
1331 char c = StringUTF16.getChar(val, sp);
1332 if (c >= '\u0080') {
1333 break;
1334 }
1335 dst[dp++] = (byte)c;
1336 sp++;
1337 }
1338 while (sp < sl) {
1339 char c = StringUTF16.getChar(val, sp++);
1340 if (c < 0x80) {
1341 dst[dp++] = (byte)c;
1342 } else if (c < 0x800) {
1343 dst[dp++] = (byte)(0xc0 | (c >> 6));
1344 dst[dp++] = (byte)(0x80 | (c & 0x3f));
1345 } else if (Character.isSurrogate(c)) {
1346 int uc = -1;
1347 char c2;
1348 if (Character.isHighSurrogate(c) && sp < sl &&
1349 Character.isLowSurrogate(c2 = StringUTF16.getChar(val, sp))) {
1350 uc = Character.toCodePoint(c, c2);
1351 }
1352 if (uc < 0) {
1353 if (doReplace) {
1354 dst[dp++] = '?';
1355 } else {
1356 throwUnmappable(sp - 1);
1357 }
1358 } else {
1359 dst[dp++] = (byte)(0xf0 | ((uc >> 18)));
1360 dst[dp++] = (byte)(0x80 | ((uc >> 12) & 0x3f));
1361 dst[dp++] = (byte)(0x80 | ((uc >> 6) & 0x3f));
1362 dst[dp++] = (byte)(0x80 | (uc & 0x3f));
1363 sp++; // 2 chars
1364 }
1365 } else {
1366 // 3 bytes, 16 bits
1367 dst[dp++] = (byte)(0xe0 | ((c >> 12)));
1368 dst[dp++] = (byte)(0x80 | ((c >> 6) & 0x3f));
1369 dst[dp++] = (byte)(0x80 | (c & 0x3f));
1370 }
1371 }
1372 if (dp == dst.length) {
1373 return dst;
1374 }
1375 return Arrays.copyOf(dst, dp);
1376 }
1377
1378 /**
1379 * Constructs a new {@code String} by decoding the specified array of bytes
1380 * using the specified {@linkplain java.nio.charset.Charset charset}. The
1381 * length of the new {@code String} is a function of the charset, and hence
1382 * may not be equal to the length of the byte array.
1383 *
1384 * <p> The behavior of this constructor when the given bytes are not valid
1385 * in the given charset is unspecified. The {@link
1386 * java.nio.charset.CharsetDecoder} class should be used when more control
1387 * over the decoding process is required.
1388 *
1389 * @param bytes
1390 * The bytes to be decoded into characters
1391 *
1392 * @param charsetName
1393 * The name of a supported {@linkplain java.nio.charset.Charset
1394 * charset}
1395 *
1396 * @throws UnsupportedEncodingException
1397 * If the named charset is not supported
1398 *
1399 * @since 1.1
1400 */
1401 public String(byte[] bytes, String charsetName)
1402 throws UnsupportedEncodingException {
1403 this(lookupCharset(charsetName), bytes, 0, bytes.length);
1404 }
1405
1406 /**
1407 * Constructs a new {@code String} by decoding the specified array of
1408 * bytes using the specified {@linkplain java.nio.charset.Charset charset}.
1409 * The length of the new {@code String} is a function of the charset, and
1410 * hence may not be equal to the length of the byte array.
1411 *
1412 * <p> This method always replaces malformed-input and unmappable-character
1413 * sequences with this charset's default replacement string. The {@link
1414 * java.nio.charset.CharsetDecoder} class should be used when more control
1415 * over the decoding process is required.
1416 *
1417 * @param bytes
1418 * The bytes to be decoded into characters
1419 *
1420 * @param charset
1421 * The {@linkplain java.nio.charset.Charset charset} to be used to
1422 * decode the {@code bytes}
1423 *
1424 * @since 1.6
1425 */
1426 public String(byte[] bytes, Charset charset) {
1427 this(Objects.requireNonNull(charset), bytes, 0, bytes.length);
1428 }
1429
1430 /**
1431 * Constructs a new {@code String} by decoding the specified subarray of
1432 * bytes using the {@link Charset#defaultCharset() default charset}.
1433 * The length of the new {@code String} is a function of the charset,
1434 * and hence may not be equal to the length of the subarray.
1435 *
1436 * <p> The behavior of this constructor when the given bytes are not valid
1437 * in the default charset is unspecified. The {@link
1438 * java.nio.charset.CharsetDecoder} class should be used when more control
1439 * over the decoding process is required.
1440 *
1441 * @param bytes
1442 * The bytes to be decoded into characters
1443 *
1444 * @param offset
1445 * The index of the first byte to decode
1446 *
1447 * @param length
1448 * The number of bytes to decode
1449 *
1450 * @throws IndexOutOfBoundsException
1451 * If {@code offset} is negative, {@code length} is negative, or
1452 * {@code offset} is greater than {@code bytes.length - length}
1453 *
1454 * @since 1.1
1455 */
1456 public String(byte[] bytes, int offset, int length) {
1457 this(Charset.defaultCharset(), bytes, checkBoundsOffCount(offset, length, bytes.length), length);
1458 }
1459
1460 /**
1461 * Constructs a new {@code String} by decoding the specified array of bytes
1462 * using the {@link Charset#defaultCharset() default charset}. The length
1463 * of the new {@code String} is a function of the charset, and hence may not
1464 * be equal to the length of the byte array.
1465 *
1466 * <p> The behavior of this constructor when the given bytes are not valid
1467 * in the default charset is unspecified. The {@link
1468 * java.nio.charset.CharsetDecoder} class should be used when more control
1469 * over the decoding process is required.
1470 *
1471 * @param bytes
1472 * The bytes to be decoded into characters
1473 *
1474 * @since 1.1
1475 */
1476 public String(byte[] bytes) {
1477 this(Charset.defaultCharset(), bytes, 0, bytes.length);
1478 }
1479
1480 /**
1481 * Allocates a new string that contains the sequence of characters
1482 * currently contained in the string buffer argument. The contents of the
1483 * string buffer are copied; subsequent modification of the string buffer
1484 * does not affect the newly created string.
1485 *
1486 * @param buffer
1487 * A {@code StringBuffer}
1488 */
1489 public String(StringBuffer buffer) {
1490 this(buffer.toString());
1491 }
1492
1493 /**
1494 * Allocates a new string that contains the sequence of characters
1495 * currently contained in the string builder argument. The contents of the
1496 * string builder are copied; subsequent modification of the string builder
1497 * does not affect the newly created string.
1498 *
1499 * <p> This constructor is provided to ease migration to {@code
1500 * StringBuilder}. Obtaining a string from a string builder via the {@code
1501 * toString} method is likely to run faster and is generally preferred.
1502 *
1503 * @param builder
1504 * A {@code StringBuilder}
1505 *
1506 * @since 1.5
1507 */
1508 public String(StringBuilder builder) {
1509 this(builder, null);
1510 }
1511
1512 /**
1513 * Returns the length of this string.
1514 * The length is equal to the number of <a href="Character.html#unicode">Unicode
1515 * code units</a> in the string.
1516 *
1517 * @return the length of the sequence of characters represented by this
1518 * object.
1519 */
1520 public int length() {
1521 return value.length >> coder();
1522 }
1523
1524 /**
1525 * Returns {@code true} if, and only if, {@link #length()} is {@code 0}.
1526 *
1527 * @return {@code true} if {@link #length()} is {@code 0}, otherwise
1528 * {@code false}
1529 *
1530 * @since 1.6
1531 */
1532 @Override
1533 public boolean isEmpty() {
1534 return value.length == 0;
1535 }
1536
1537 /**
1538 * Returns the {@code char} value at the
1539 * specified index. An index ranges from {@code 0} to
1540 * {@code length() - 1}. The first {@code char} value of the sequence
1541 * is at index {@code 0}, the next at index {@code 1},
1542 * and so on, as for array indexing.
1543 *
1544 * <p>If the {@code char} value specified by the index is a
1545 * <a href="Character.html#unicode">surrogate</a>, the surrogate
1546 * value is returned.
1547 *
1548 * @param index the index of the {@code char} value.
1549 * @return the {@code char} value at the specified index of this string.
1550 * The first {@code char} value is at index {@code 0}.
1551 * @throws IndexOutOfBoundsException if the {@code index}
1552 * argument is negative or not less than the length of this
1553 * string.
1554 */
1555 public char charAt(int index) {
1556 if (isLatin1()) {
1557 return StringLatin1.charAt(value, index);
1558 } else {
1559 return StringUTF16.charAt(value, index);
1560 }
1561 }
1562
1563 /**
1564 * Returns the character (Unicode code point) at the specified
1565 * index. The index refers to {@code char} values
1566 * (Unicode code units) and ranges from {@code 0} to
1567 * {@link #length()}{@code - 1}.
1568 *
1569 * <p> If the {@code char} value specified at the given index
1570 * is in the high-surrogate range, the following index is less
1571 * than the length of this {@code String}, and the
1572 * {@code char} value at the following index is in the
1573 * low-surrogate range, then the supplementary code point
1574 * corresponding to this surrogate pair is returned. Otherwise,
1575 * the {@code char} value at the given index is returned.
1576 *
1577 * @param index the index to the {@code char} values
1578 * @return the code point value of the character at the
1579 * {@code index}
1580 * @throws IndexOutOfBoundsException if the {@code index}
1581 * argument is negative or not less than the length of this
1582 * string.
1583 * @since 1.5
1584 */
1585 public int codePointAt(int index) {
1586 if (isLatin1()) {
1587 checkIndex(index, value.length);
1588 return value[index] & 0xff;
1589 }
1590 int length = value.length >> 1;
1591 checkIndex(index, length);
1592 return StringUTF16.codePointAt(value, index, length);
1593 }
1594
1595 /**
1596 * Returns the character (Unicode code point) before the specified
1597 * index. The index refers to {@code char} values
1598 * (Unicode code units) and ranges from {@code 1} to {@link
1599 * CharSequence#length() length}.
1600 *
1601 * <p> If the {@code char} value at {@code (index - 1)}
1602 * is in the low-surrogate range, {@code (index - 2)} is not
1603 * negative, and the {@code char} value at {@code (index -
1604 * 2)} is in the high-surrogate range, then the
1605 * supplementary code point value of the surrogate pair is
1606 * returned. If the {@code char} value at {@code index -
1607 * 1} is an unpaired low-surrogate or a high-surrogate, the
1608 * surrogate value is returned.
1609 *
1610 * @param index the index following the code point that should be returned
1611 * @return the Unicode code point value before the given index.
1612 * @throws IndexOutOfBoundsException if the {@code index}
1613 * argument is less than 1 or greater than the length
1614 * of this string.
1615 * @since 1.5
1616 */
1617 public int codePointBefore(int index) {
1618 int i = index - 1;
1619 checkIndex(i, length());
1620 if (isLatin1()) {
1621 return (value[i] & 0xff);
1622 }
1623 return StringUTF16.codePointBefore(value, index);
1624 }
1625
1626 /**
1627 * Returns the number of Unicode code points in the specified text
1628 * range of this {@code String}. The text range begins at the
1629 * specified {@code beginIndex} and extends to the
1630 * {@code char} at index {@code endIndex - 1}. Thus the
1631 * length (in {@code char}s) of the text range is
1632 * {@code endIndex-beginIndex}. Unpaired surrogates within
1633 * the text range count as one code point each.
1634 *
1635 * @param beginIndex the index to the first {@code char} of
1636 * the text range.
1637 * @param endIndex the index after the last {@code char} of
1638 * the text range.
1639 * @return the number of Unicode code points in the specified text
1640 * range
1641 * @throws IndexOutOfBoundsException if the
1642 * {@code beginIndex} is negative, or {@code endIndex}
1643 * is larger than the length of this {@code String}, or
1644 * {@code beginIndex} is larger than {@code endIndex}.
1645 * @since 1.5
1646 */
1647 public int codePointCount(int beginIndex, int endIndex) {
1648 Objects.checkFromToIndex(beginIndex, endIndex, length());
1649 if (isLatin1()) {
1650 return endIndex - beginIndex;
1651 }
1652 return StringUTF16.codePointCount(value, beginIndex, endIndex);
1653 }
1654
1655 /**
1656 * Returns the index within this {@code String} that is
1657 * offset from the given {@code index} by
1658 * {@code codePointOffset} code points. Unpaired surrogates
1659 * within the text range given by {@code index} and
1660 * {@code codePointOffset} count as one code point each.
1661 *
1662 * @param index the index to be offset
1663 * @param codePointOffset the offset in code points
1664 * @return the index within this {@code String}
1665 * @throws IndexOutOfBoundsException if {@code index}
1666 * is negative or larger than the length of this
1667 * {@code String}, or if {@code codePointOffset} is positive
1668 * and the substring starting with {@code index} has fewer
1669 * than {@code codePointOffset} code points,
1670 * or if {@code codePointOffset} is negative and the substring
1671 * before {@code index} has fewer than the absolute value
1672 * of {@code codePointOffset} code points.
1673 * @since 1.5
1674 */
1675 public int offsetByCodePoints(int index, int codePointOffset) {
1676 return Character.offsetByCodePoints(this, index, codePointOffset);
1677 }
1678
1679 /**
1680 * Copies characters from this string into the destination character
1681 * array.
1682 * <p>
1683 * The first character to be copied is at index {@code srcBegin};
1684 * the last character to be copied is at index {@code srcEnd-1}
1685 * (thus the total number of characters to be copied is
1686 * {@code srcEnd-srcBegin}). The characters are copied into the
1687 * subarray of {@code dst} starting at index {@code dstBegin}
1688 * and ending at index:
1689 * <blockquote><pre>
1690 * dstBegin + (srcEnd-srcBegin) - 1
1691 * </pre></blockquote>
1692 *
1693 * @param srcBegin index of the first character in the string
1694 * to copy.
1695 * @param srcEnd index after the last character in the string
1696 * to copy.
1697 * @param dst the destination array.
1698 * @param dstBegin the start offset in the destination array.
1699 * @throws IndexOutOfBoundsException If any of the following
1700 * is true:
1701 * <ul><li>{@code srcBegin} is negative.
1702 * <li>{@code srcBegin} is greater than {@code srcEnd}
1703 * <li>{@code srcEnd} is greater than the length of this
1704 * string
1705 * <li>{@code dstBegin} is negative
1706 * <li>{@code dstBegin+(srcEnd-srcBegin)} is larger than
1707 * {@code dst.length}</ul>
1708 */
1709 public void getChars(int srcBegin, int srcEnd, char[] dst, int dstBegin) {
1710 checkBoundsBeginEnd(srcBegin, srcEnd, length());
1711 checkBoundsOffCount(dstBegin, srcEnd - srcBegin, dst.length);
1712 if (isLatin1()) {
1713 StringLatin1.getChars(value, srcBegin, srcEnd, dst, dstBegin);
1714 } else {
1715 StringUTF16.getChars(value, srcBegin, srcEnd, dst, dstBegin);
1716 }
1717 }
1718
1719 /**
1720 * Copies characters from this string into the destination byte array. Each
1721 * byte receives the 8 low-order bits of the corresponding character. The
1722 * eight high-order bits of each character are not copied and do not
1723 * participate in the transfer in any way.
1724 *
1725 * <p> The first character to be copied is at index {@code srcBegin}; the
1726 * last character to be copied is at index {@code srcEnd-1}. The total
1727 * number of characters to be copied is {@code srcEnd-srcBegin}. The
1728 * characters, converted to bytes, are copied into the subarray of {@code
1729 * dst} starting at index {@code dstBegin} and ending at index:
1730 *
1731 * <blockquote><pre>
1732 * dstBegin + (srcEnd-srcBegin) - 1
1733 * </pre></blockquote>
1734 *
1735 * @deprecated This method does not properly convert characters into
1736 * bytes. As of JDK 1.1, the preferred way to do this is via the
1737 * {@link #getBytes()} method, which uses the {@link Charset#defaultCharset()
1738 * default charset}.
1739 *
1740 * @param srcBegin
1741 * Index of the first character in the string to copy
1742 *
1743 * @param srcEnd
1744 * Index after the last character in the string to copy
1745 *
1746 * @param dst
1747 * The destination array
1748 *
1749 * @param dstBegin
1750 * The start offset in the destination array
1751 *
1752 * @throws IndexOutOfBoundsException
1753 * If any of the following is true:
1754 * <ul>
1755 * <li> {@code srcBegin} is negative
1756 * <li> {@code srcBegin} is greater than {@code srcEnd}
1757 * <li> {@code srcEnd} is greater than the length of this String
1758 * <li> {@code dstBegin} is negative
1759 * <li> {@code dstBegin+(srcEnd-srcBegin)} is larger than {@code
1760 * dst.length}
1761 * </ul>
1762 */
1763 @Deprecated(since="1.1")
1764 public void getBytes(int srcBegin, int srcEnd, byte[] dst, int dstBegin) {
1765 checkBoundsBeginEnd(srcBegin, srcEnd, length());
1766 Objects.requireNonNull(dst);
1767 checkBoundsOffCount(dstBegin, srcEnd - srcBegin, dst.length);
1768 if (isLatin1()) {
1769 StringLatin1.getBytes(value, srcBegin, srcEnd, dst, dstBegin);
1770 } else {
1771 StringUTF16.getBytes(value, srcBegin, srcEnd, dst, dstBegin);
1772 }
1773 }
1774
1775 /**
1776 * Encodes this {@code String} into a sequence of bytes using the named
1777 * charset, storing the result into a new byte array.
1778 *
1779 * <p> The behavior of this method when this string cannot be encoded in
1780 * the given charset is unspecified. The {@link
1781 * java.nio.charset.CharsetEncoder} class should be used when more control
1782 * over the encoding process is required.
1783 *
1784 * @param charsetName
1785 * The name of a supported {@linkplain java.nio.charset.Charset
1786 * charset}
1787 *
1788 * @return The resultant byte array
1789 *
1790 * @throws UnsupportedEncodingException
1791 * If the named charset is not supported
1792 *
1793 * @since 1.1
1794 */
1795 public byte[] getBytes(String charsetName)
1796 throws UnsupportedEncodingException {
1797 return encode(lookupCharset(charsetName), coder(), value);
1798 }
1799
1800 /**
1801 * Encodes this {@code String} into a sequence of bytes using the given
1802 * {@linkplain java.nio.charset.Charset charset}, storing the result into a
1803 * new byte array.
1804 *
1805 * <p> This method always replaces malformed-input and unmappable-character
1806 * sequences with this charset's default replacement byte array. The
1807 * {@link java.nio.charset.CharsetEncoder} class should be used when more
1808 * control over the encoding process is required.
1809 *
1810 * @param charset
1811 * The {@linkplain java.nio.charset.Charset} to be used to encode
1812 * the {@code String}
1813 *
1814 * @return The resultant byte array
1815 *
1816 * @since 1.6
1817 */
1818 public byte[] getBytes(Charset charset) {
1819 if (charset == null) throw new NullPointerException();
1820 return encode(charset, coder(), value);
1821 }
1822
1823 /**
1824 * Encodes this {@code String} into a sequence of bytes using the
1825 * {@link Charset#defaultCharset() default charset}, storing the result
1826 * into a new byte array.
1827 *
1828 * <p> The behavior of this method when this string cannot be encoded in
1829 * the default charset is unspecified. The {@link
1830 * java.nio.charset.CharsetEncoder} class should be used when more control
1831 * over the encoding process is required.
1832 *
1833 * @return The resultant byte array
1834 *
1835 * @since 1.1
1836 */
1837 public byte[] getBytes() {
1838 return encode(Charset.defaultCharset(), coder(), value);
1839 }
1840
1841 boolean bytesCompatible(Charset charset) {
1842 if (isLatin1()) {
1843 if (charset == ISO_8859_1.INSTANCE) {
1844 return true; // ok, same encoding
1845 } else if (charset == UTF_8.INSTANCE || charset == US_ASCII.INSTANCE) {
1846 return !StringCoding.hasNegatives(value, 0, value.length); // ok, if ASCII-compatible
1847 }
1848 }
1849 return false;
1850 }
1851
1852 void copyToSegmentRaw(MemorySegment segment, long offset) {
1853 MemorySegment.copy(value, 0, segment, ValueLayout.JAVA_BYTE, offset, value.length);
1854 }
1855
1856 /**
1857 * Compares this string to the specified object. The result is {@code
1858 * true} if and only if the argument is not {@code null} and is a {@code
1859 * String} object that represents the same sequence of characters as this
1860 * object.
1861 *
1862 * <p>For finer-grained String comparison, refer to
1863 * {@link java.text.Collator}.
1864 *
1865 * @param anObject
1866 * The object to compare this {@code String} against
1867 *
1868 * @return {@code true} if the given object represents a {@code String}
1869 * equivalent to this string, {@code false} otherwise
1870 *
1871 * @see #compareTo(String)
1872 * @see #equalsIgnoreCase(String)
1873 */
1874 public boolean equals(Object anObject) {
1875 if (this == anObject) {
1876 return true;
1877 }
1878 return (anObject instanceof String aString)
1879 && (!COMPACT_STRINGS || this.coder == aString.coder)
1880 && StringLatin1.equals(value, aString.value);
1881 }
1882
1883 /**
1884 * Compares this string to the specified {@code StringBuffer}. The result
1885 * is {@code true} if and only if this {@code String} represents the same
1886 * sequence of characters as the specified {@code StringBuffer}. This method
1887 * synchronizes on the {@code StringBuffer}.
1888 *
1889 * <p>For finer-grained String comparison, refer to
1890 * {@link java.text.Collator}.
1891 *
1892 * @param sb
1893 * The {@code StringBuffer} to compare this {@code String} against
1894 *
1895 * @return {@code true} if this {@code String} represents the same
1896 * sequence of characters as the specified {@code StringBuffer},
1897 * {@code false} otherwise
1898 *
1899 * @since 1.4
1900 */
1901 public boolean contentEquals(StringBuffer sb) {
1902 return contentEquals((CharSequence)sb);
1903 }
1904
1905 private boolean nonSyncContentEquals(AbstractStringBuilder sb) {
1906 int len = length();
1907 if (len != sb.length()) {
1908 return false;
1909 }
1910 byte[] v1 = value;
1911 byte[] v2 = sb.getValue();
1912 byte coder = coder();
1913 if (coder == sb.getCoder()) {
1914 return v1.length <= v2.length && ArraysSupport.mismatch(v1, v2, v1.length) < 0;
1915 } else {
1916 if (coder != LATIN1) { // utf16 str and latin1 abs can never be "equal"
1917 return false;
1918 }
1919 return StringUTF16.contentEquals(v1, v2, len);
1920 }
1921 }
1922
1923 /**
1924 * Compares this string to the specified {@code CharSequence}. The
1925 * result is {@code true} if and only if this {@code String} represents the
1926 * same sequence of char values as the specified sequence. Note that if the
1927 * {@code CharSequence} is a {@code StringBuffer} then the method
1928 * synchronizes on it.
1929 *
1930 * <p>For finer-grained String comparison, refer to
1931 * {@link java.text.Collator}.
1932 *
1933 * @param cs
1934 * The sequence to compare this {@code String} against
1935 *
1936 * @return {@code true} if this {@code String} represents the same
1937 * sequence of char values as the specified sequence, {@code
1938 * false} otherwise
1939 *
1940 * @since 1.5
1941 */
1942 public boolean contentEquals(CharSequence cs) {
1943 // Argument is a StringBuffer, StringBuilder
1944 if (cs instanceof AbstractStringBuilder) {
1945 if (cs instanceof StringBuffer) {
1946 synchronized(cs) {
1947 return nonSyncContentEquals((AbstractStringBuilder)cs);
1948 }
1949 } else {
1950 return nonSyncContentEquals((AbstractStringBuilder)cs);
1951 }
1952 }
1953 // Argument is a String
1954 if (cs instanceof String) {
1955 return equals(cs);
1956 }
1957 // Argument is a generic CharSequence
1958 int n = cs.length();
1959 if (n != length()) {
1960 return false;
1961 }
1962 byte[] val = this.value;
1963 if (isLatin1()) {
1964 for (int i = 0; i < n; i++) {
1965 if ((val[i] & 0xff) != cs.charAt(i)) {
1966 return false;
1967 }
1968 }
1969 } else {
1970 if (!StringUTF16.contentEquals(val, cs, n)) {
1971 return false;
1972 }
1973 }
1974 return true;
1975 }
1976
1977 /**
1978 * Compares this {@code String} to another {@code String}, ignoring case
1979 * considerations. Two strings are considered equal ignoring case if they
1980 * are of the same length and corresponding Unicode code points in the two
1981 * strings are equal ignoring case.
1982 *
1983 * <p> Two Unicode code points are considered the same
1984 * ignoring case if at least one of the following is true:
1985 * <ul>
1986 * <li> The two Unicode code points are the same (as compared by the
1987 * {@code ==} operator)
1988 * <li> Calling {@code Character.toLowerCase(Character.toUpperCase(int))}
1989 * on each Unicode code point produces the same result
1990 * </ul>
1991 *
1992 * <p>Note that this method does <em>not</em> take locale into account, and
1993 * will result in unsatisfactory results for certain locales. The
1994 * {@link java.text.Collator} class provides locale-sensitive comparison.
1995 *
1996 * @param anotherString
1997 * The {@code String} to compare this {@code String} against
1998 *
1999 * @return {@code true} if the argument is not {@code null} and it
2000 * represents an equivalent {@code String} ignoring case; {@code
2001 * false} otherwise
2002 *
2003 * @see #equals(Object)
2004 * @see #codePoints()
2005 */
2006 public boolean equalsIgnoreCase(String anotherString) {
2007 return (this == anotherString) ? true
2008 : (anotherString != null)
2009 && (anotherString.length() == length())
2010 && regionMatches(true, 0, anotherString, 0, length());
2011 }
2012
2013 /**
2014 * Compares two strings lexicographically.
2015 * The comparison is based on the Unicode value of each character in
2016 * the strings. The character sequence represented by this
2017 * {@code String} object is compared lexicographically to the
2018 * character sequence represented by the argument string. The result is
2019 * a negative integer if this {@code String} object
2020 * lexicographically precedes the argument string. The result is a
2021 * positive integer if this {@code String} object lexicographically
2022 * follows the argument string. The result is zero if the strings
2023 * are equal; {@code compareTo} returns {@code 0} exactly when
2024 * the {@link #equals(Object)} method would return {@code true}.
2025 * <p>
2026 * This is the definition of lexicographic ordering. If two strings are
2027 * different, then either they have different characters at some index
2028 * that is a valid index for both strings, or their lengths are different,
2029 * or both. If they have different characters at one or more index
2030 * positions, let <i>k</i> be the smallest such index; then the string
2031 * whose character at position <i>k</i> has the smaller value, as
2032 * determined by using the {@code <} operator, lexicographically precedes the
2033 * other string. In this case, {@code compareTo} returns the
2034 * difference of the two character values at position {@code k} in
2035 * the two string -- that is, the value:
2036 * <blockquote><pre>
2037 * this.charAt(k)-anotherString.charAt(k)
2038 * </pre></blockquote>
2039 * If there is no index position at which they differ, then the shorter
2040 * string lexicographically precedes the longer string. In this case,
2041 * {@code compareTo} returns the difference of the lengths of the
2042 * strings -- that is, the value:
2043 * <blockquote><pre>
2044 * this.length()-anotherString.length()
2045 * </pre></blockquote>
2046 *
2047 * <p>For finer-grained String comparison, refer to
2048 * {@link java.text.Collator}.
2049 *
2050 * @param anotherString the {@code String} to be compared.
2051 * @return the value {@code 0} if the argument string is equal to
2052 * this string; a value less than {@code 0} if this string
2053 * is lexicographically less than the string argument; and a
2054 * value greater than {@code 0} if this string is
2055 * lexicographically greater than the string argument.
2056 */
2057 public int compareTo(String anotherString) {
2058 byte[] v1 = value;
2059 byte[] v2 = anotherString.value;
2060 byte coder = coder();
2061 if (coder == anotherString.coder()) {
2062 return coder == LATIN1 ? StringLatin1.compareTo(v1, v2)
2063 : StringUTF16.compareTo(v1, v2);
2064 }
2065 return coder == LATIN1 ? StringLatin1.compareToUTF16(v1, v2)
2066 : StringUTF16.compareToLatin1(v1, v2);
2067 }
2068
2069 /**
2070 * A Comparator that orders {@code String} objects as by
2071 * {@link #compareToIgnoreCase(String) compareToIgnoreCase}.
2072 * This comparator is serializable.
2073 * <p>
2074 * Note that this Comparator does <em>not</em> take locale into account,
2075 * and will result in an unsatisfactory ordering for certain locales.
2076 * The {@link java.text.Collator} class provides locale-sensitive comparison.
2077 *
2078 * @see java.text.Collator
2079 * @since 1.2
2080 */
2081 public static final Comparator<String> CASE_INSENSITIVE_ORDER
2082 = new CaseInsensitiveComparator();
2083
2084 /**
2085 * CaseInsensitiveComparator for Strings.
2086 */
2087 private static class CaseInsensitiveComparator
2088 implements Comparator<String>, java.io.Serializable {
2089 // use serialVersionUID from JDK 1.2.2 for interoperability
2090 @java.io.Serial
2091 private static final long serialVersionUID = 8575799808933029326L;
2092
2093 public int compare(String s1, String s2) {
2094 byte[] v1 = s1.value;
2095 byte[] v2 = s2.value;
2096 byte coder = s1.coder();
2097 if (coder == s2.coder()) {
2098 return coder == LATIN1 ? StringLatin1.compareToCI(v1, v2)
2099 : StringUTF16.compareToCI(v1, v2);
2100 }
2101 return coder == LATIN1 ? StringLatin1.compareToCI_UTF16(v1, v2)
2102 : StringUTF16.compareToCI_Latin1(v1, v2);
2103 }
2104
2105 /** Replaces the de-serialized object. */
2106 @java.io.Serial
2107 private Object readResolve() { return CASE_INSENSITIVE_ORDER; }
2108 }
2109
2110 /**
2111 * Compares two strings lexicographically, ignoring case
2112 * differences. This method returns an integer whose sign is that of
2113 * calling {@code compareTo} with case folded versions of the strings
2114 * where case differences have been eliminated by calling
2115 * {@code Character.toLowerCase(Character.toUpperCase(int))} on
2116 * each Unicode code point.
2117 * <p>
2118 * Note that this method does <em>not</em> take locale into account,
2119 * and will result in an unsatisfactory ordering for certain locales.
2120 * The {@link java.text.Collator} class provides locale-sensitive comparison.
2121 *
2122 * @param str the {@code String} to be compared.
2123 * @return a negative integer, zero, or a positive integer as the
2124 * specified String is greater than, equal to, or less
2125 * than this String, ignoring case considerations.
2126 * @see java.text.Collator
2127 * @see #codePoints()
2128 * @since 1.2
2129 */
2130 public int compareToIgnoreCase(String str) {
2131 return CASE_INSENSITIVE_ORDER.compare(this, str);
2132 }
2133
2134 /**
2135 * Tests if two string regions are equal.
2136 * <p>
2137 * A substring of this {@code String} object is compared to a substring
2138 * of the argument other. The result is true if these substrings
2139 * represent identical character sequences. The substring of this
2140 * {@code String} object to be compared begins at index {@code toffset}
2141 * and has length {@code len}. The substring of other to be compared
2142 * begins at index {@code ooffset} and has length {@code len}. The
2143 * result is {@code false} if and only if at least one of the following
2144 * is true:
2145 * <ul><li>{@code toffset} is negative.
2146 * <li>{@code ooffset} is negative.
2147 * <li>{@code toffset+len} is greater than the length of this
2148 * {@code String} object.
2149 * <li>{@code ooffset+len} is greater than the length of the other
2150 * argument.
2151 * <li>There is some nonnegative integer <i>k</i> less than {@code len}
2152 * such that:
2153 * {@code this.charAt(toffset + }<i>k</i>{@code ) != other.charAt(ooffset + }
2154 * <i>k</i>{@code )}
2155 * </ul>
2156 *
2157 * <p>Note that this method does <em>not</em> take locale into account. The
2158 * {@link java.text.Collator} class provides locale-sensitive comparison.
2159 *
2160 * @param toffset the starting offset of the subregion in this string.
2161 * @param other the string argument.
2162 * @param ooffset the starting offset of the subregion in the string
2163 * argument.
2164 * @param len the number of characters to compare.
2165 * @return {@code true} if the specified subregion of this string
2166 * exactly matches the specified subregion of the string argument;
2167 * {@code false} otherwise.
2168 */
2169 public boolean regionMatches(int toffset, String other, int ooffset, int len) {
2170 // Note: toffset, ooffset, or len might be near -1>>>1.
2171 if ((ooffset < 0) || (toffset < 0) ||
2172 (toffset > (long)length() - len) ||
2173 (ooffset > (long)other.length() - len)) {
2174 return false;
2175 }
2176 byte[] tv = value;
2177 byte[] ov = other.value;
2178 byte coder = coder();
2179 if (coder == other.coder()) {
2180 if (coder == UTF16) {
2181 toffset <<= UTF16;
2182 ooffset <<= UTF16;
2183 len <<= UTF16;
2184 }
2185 return ArraysSupport.mismatch(tv, toffset,
2186 ov, ooffset, len) < 0;
2187 } else {
2188 if (coder == LATIN1) {
2189 while (len-- > 0) {
2190 if (StringLatin1.getChar(tv, toffset++) !=
2191 StringUTF16.getChar(ov, ooffset++)) {
2192 return false;
2193 }
2194 }
2195 } else {
2196 while (len-- > 0) {
2197 if (StringUTF16.getChar(tv, toffset++) !=
2198 StringLatin1.getChar(ov, ooffset++)) {
2199 return false;
2200 }
2201 }
2202 }
2203 }
2204 return true;
2205 }
2206
2207 /**
2208 * Tests if two string regions are equal.
2209 * <p>
2210 * A substring of this {@code String} object is compared to a substring
2211 * of the argument {@code other}. The result is {@code true} if these
2212 * substrings represent Unicode code point sequences that are the same,
2213 * ignoring case if and only if {@code ignoreCase} is true.
2214 * The sequences {@code tsequence} and {@code osequence} are compared,
2215 * where {@code tsequence} is the sequence produced as if by calling
2216 * {@code this.substring(toffset, toffset + len).codePoints()} and
2217 * {@code osequence} is the sequence produced as if by calling
2218 * {@code other.substring(ooffset, ooffset + len).codePoints()}.
2219 * The result is {@code true} if and only if all of the following
2220 * are true:
2221 * <ul><li>{@code toffset} is non-negative.
2222 * <li>{@code ooffset} is non-negative.
2223 * <li>{@code toffset+len} is less than or equal to the length of this
2224 * {@code String} object.
2225 * <li>{@code ooffset+len} is less than or equal to the length of the other
2226 * argument.
2227 * <li>if {@code ignoreCase} is {@code false}, all pairs of corresponding Unicode
2228 * code points are equal integer values; or if {@code ignoreCase} is {@code true},
2229 * {@link Character#toLowerCase(int) Character.toLowerCase(}
2230 * {@link Character#toUpperCase(int)}{@code )} on all pairs of Unicode code points
2231 * results in equal integer values.
2232 * </ul>
2233 *
2234 * <p>Note that this method does <em>not</em> take locale into account,
2235 * and will result in unsatisfactory results for certain locales when
2236 * {@code ignoreCase} is {@code true}. The {@link java.text.Collator} class
2237 * provides locale-sensitive comparison.
2238 *
2239 * @param ignoreCase if {@code true}, ignore case when comparing
2240 * characters.
2241 * @param toffset the starting offset of the subregion in this
2242 * string.
2243 * @param other the string argument.
2244 * @param ooffset the starting offset of the subregion in the string
2245 * argument.
2246 * @param len the number of characters (Unicode code units -
2247 * 16bit {@code char} value) to compare.
2248 * @return {@code true} if the specified subregion of this string
2249 * matches the specified subregion of the string argument;
2250 * {@code false} otherwise. Whether the matching is exact
2251 * or case insensitive depends on the {@code ignoreCase}
2252 * argument.
2253 * @see #codePoints()
2254 */
2255 public boolean regionMatches(boolean ignoreCase, int toffset,
2256 String other, int ooffset, int len) {
2257 if (!ignoreCase) {
2258 return regionMatches(toffset, other, ooffset, len);
2259 }
2260 // Note: toffset, ooffset, or len might be near -1>>>1.
2261 if ((ooffset < 0) || (toffset < 0)
2262 || (toffset > (long)length() - len)
2263 || (ooffset > (long)other.length() - len)) {
2264 return false;
2265 }
2266 byte[] tv = value;
2267 byte[] ov = other.value;
2268 byte coder = coder();
2269 if (coder == other.coder()) {
2270 return coder == LATIN1
2271 ? StringLatin1.regionMatchesCI(tv, toffset, ov, ooffset, len)
2272 : StringUTF16.regionMatchesCI(tv, toffset, ov, ooffset, len);
2273 }
2274 return coder == LATIN1
2275 ? StringLatin1.regionMatchesCI_UTF16(tv, toffset, ov, ooffset, len)
2276 : StringUTF16.regionMatchesCI_Latin1(tv, toffset, ov, ooffset, len);
2277 }
2278
2279 /**
2280 * Tests if the substring of this string beginning at the
2281 * specified index starts with the specified prefix.
2282 *
2283 * @param prefix the prefix.
2284 * @param toffset where to begin looking in this string.
2285 * @return {@code true} if the character sequence represented by the
2286 * argument is a prefix of the substring of this object starting
2287 * at index {@code toffset}; {@code false} otherwise.
2288 * The result is {@code false} if {@code toffset} is
2289 * negative or greater than the length of this
2290 * {@code String} object; otherwise the result is the same
2291 * as the result of the expression
2292 * <pre>
2293 * this.substring(toffset).startsWith(prefix)
2294 * </pre>
2295 */
2296 public boolean startsWith(String prefix, int toffset) {
2297 // Note: toffset might be near -1>>>1.
2298 if (toffset < 0 || toffset > length() - prefix.length()) {
2299 return false;
2300 }
2301 byte[] ta = value;
2302 byte[] pa = prefix.value;
2303 int po = 0;
2304 int pc = pa.length;
2305 byte coder = coder();
2306 if (coder == prefix.coder()) {
2307 if (coder == UTF16) {
2308 toffset <<= UTF16;
2309 }
2310 return ArraysSupport.mismatch(ta, toffset,
2311 pa, 0, pc) < 0;
2312 } else {
2313 if (coder == LATIN1) { // && pcoder == UTF16
2314 return false;
2315 }
2316 // coder == UTF16 && pcoder == LATIN1)
2317 while (po < pc) {
2318 if (StringUTF16.getChar(ta, toffset++) != (pa[po++] & 0xff)) {
2319 return false;
2320 }
2321 }
2322 }
2323 return true;
2324 }
2325
2326 /**
2327 * Tests if this string starts with the specified prefix.
2328 *
2329 * @param prefix the prefix.
2330 * @return {@code true} if the character sequence represented by the
2331 * argument is a prefix of the character sequence represented by
2332 * this string; {@code false} otherwise.
2333 * Note also that {@code true} will be returned if the
2334 * argument is an empty string or is equal to this
2335 * {@code String} object as determined by the
2336 * {@link #equals(Object)} method.
2337 * @since 1.0
2338 */
2339 public boolean startsWith(String prefix) {
2340 return startsWith(prefix, 0);
2341 }
2342
2343 /**
2344 * Tests if this string ends with the specified suffix.
2345 *
2346 * @param suffix the suffix.
2347 * @return {@code true} if the character sequence represented by the
2348 * argument is a suffix of the character sequence represented by
2349 * this object; {@code false} otherwise. Note that the
2350 * result will be {@code true} if the argument is the
2351 * empty string or is equal to this {@code String} object
2352 * as determined by the {@link #equals(Object)} method.
2353 */
2354 public boolean endsWith(String suffix) {
2355 return startsWith(suffix, length() - suffix.length());
2356 }
2357
2358 /**
2359 * Returns a hash code for this string. The hash code for a
2360 * {@code String} object is computed as
2361 * <blockquote><pre>
2362 * s[0]*31^(n-1) + s[1]*31^(n-2) + ... + s[n-1]
2363 * </pre></blockquote>
2364 * using {@code int} arithmetic, where {@code s[i]} is the
2365 * <i>i</i>th character of the string, {@code n} is the length of
2366 * the string, and {@code ^} indicates exponentiation.
2367 * (The hash value of the empty string is zero.)
2368 *
2369 * @return a hash code value for this object.
2370 */
2371 public int hashCode() {
2372 // The hash or hashIsZero fields are subject to a benign data race,
2373 // making it crucial to ensure that any observable result of the
2374 // calculation in this method stays correct under any possible read of
2375 // these fields. Necessary restrictions to allow this to be correct
2376 // without explicit memory fences or similar concurrency primitives is
2377 // that we can ever only write to one of these two fields for a given
2378 // String instance, and that the computation is idempotent and derived
2379 // from immutable state
2380 int h = hash;
2381 if (h == 0 && !hashIsZero) {
2382 h = isLatin1() ? StringLatin1.hashCode(value)
2383 : StringUTF16.hashCode(value);
2384 if (h == 0) {
2385 hashIsZero = true;
2386 } else {
2387 hash = h;
2388 }
2389 }
2390 return h;
2391 }
2392
2393 /**
2394 * Returns the index within this string of the first occurrence of
2395 * the specified character. If a character with value
2396 * {@code ch} occurs in the character sequence represented by
2397 * this {@code String} object, then the index (in Unicode
2398 * code units) of the first such occurrence is returned. For
2399 * values of {@code ch} in the range from 0 to 0xFFFF
2400 * (inclusive), this is the smallest value <i>k</i> such that:
2401 * <blockquote><pre>
2402 * this.charAt(<i>k</i>) == ch
2403 * </pre></blockquote>
2404 * is true. For other values of {@code ch}, it is the
2405 * smallest value <i>k</i> such that:
2406 * <blockquote><pre>
2407 * this.codePointAt(<i>k</i>) == ch
2408 * </pre></blockquote>
2409 * is true. In either case, if no such character occurs in this
2410 * string, then {@code -1} is returned.
2411 *
2412 * @param ch a character (Unicode code point).
2413 * @return the index of the first occurrence of the character in the
2414 * character sequence represented by this object, or
2415 * {@code -1} if the character does not occur.
2416 */
2417 public int indexOf(int ch) {
2418 return indexOf(ch, 0);
2419 }
2420
2421 /**
2422 * Returns the index within this string of the first occurrence of the
2423 * specified character, starting the search at the specified index.
2424 * <p>
2425 * If a character with value {@code ch} occurs in the
2426 * character sequence represented by this {@code String}
2427 * object at an index no smaller than {@code fromIndex}, then
2428 * the index of the first such occurrence is returned. For values
2429 * of {@code ch} in the range from 0 to 0xFFFF (inclusive),
2430 * this is the smallest value <i>k</i> such that:
2431 * <blockquote><pre>
2432 * (this.charAt(<i>k</i>) == ch) {@code &&} (<i>k</i> >= fromIndex)
2433 * </pre></blockquote>
2434 * is true. For other values of {@code ch}, it is the
2435 * smallest value <i>k</i> such that:
2436 * <blockquote><pre>
2437 * (this.codePointAt(<i>k</i>) == ch) {@code &&} (<i>k</i> >= fromIndex)
2438 * </pre></blockquote>
2439 * is true. In either case, if no such character occurs in this
2440 * string at or after position {@code fromIndex}, then
2441 * {@code -1} is returned.
2442 *
2443 * <p>
2444 * There is no restriction on the value of {@code fromIndex}. If it
2445 * is negative, it has the same effect as if it were zero: this entire
2446 * string may be searched. If it is greater than the length of this
2447 * string, it has the same effect as if it were equal to the length of
2448 * this string: {@code -1} is returned.
2449 *
2450 * <p>All indices are specified in {@code char} values
2451 * (Unicode code units).
2452 *
2453 * @param ch a character (Unicode code point).
2454 * @param fromIndex the index to start the search from.
2455 * @return the index of the first occurrence of the character in the
2456 * character sequence represented by this object that is greater
2457 * than or equal to {@code fromIndex}, or {@code -1}
2458 * if the character does not occur.
2459 *
2460 * @apiNote
2461 * Unlike {@link #substring(int)}, for example, this method does not throw
2462 * an exception when {@code fromIndex} is outside the valid range.
2463 * Rather, it returns -1 when {@code fromIndex} is larger than the length of
2464 * the string.
2465 * This result is, by itself, indistinguishable from a genuine absence of
2466 * {@code ch} in the string.
2467 * If stricter behavior is needed, {@link #indexOf(int, int, int)}
2468 * should be considered instead.
2469 * On a {@link String} {@code s}, for example,
2470 * {@code s.indexOf(ch, fromIndex, s.length())} would throw if
2471 * {@code fromIndex} were larger than the string length, or were negative.
2472 */
2473 public int indexOf(int ch, int fromIndex) {
2474 return isLatin1() ? StringLatin1.indexOf(value, ch, fromIndex, length())
2475 : StringUTF16.indexOf(value, ch, fromIndex, length());
2476 }
2477
2478 /**
2479 * Returns the index within this string of the first occurrence of the
2480 * specified character, starting the search at {@code beginIndex} and
2481 * stopping before {@code endIndex}.
2482 *
2483 * <p>If a character with value {@code ch} occurs in the
2484 * character sequence represented by this {@code String}
2485 * object at an index no smaller than {@code beginIndex} but smaller than
2486 * {@code endIndex}, then
2487 * the index of the first such occurrence is returned. For values
2488 * of {@code ch} in the range from 0 to 0xFFFF (inclusive),
2489 * this is the smallest value <i>k</i> such that:
2490 * <blockquote><pre>
2491 * (this.charAt(<i>k</i>) == ch) && (beginIndex <= <i>k</i> < endIndex)
2492 * </pre></blockquote>
2493 * is true. For other values of {@code ch}, it is the
2494 * smallest value <i>k</i> such that:
2495 * <blockquote><pre>
2496 * (this.codePointAt(<i>k</i>) == ch) && (beginIndex <= <i>k</i> < endIndex)
2497 * </pre></blockquote>
2498 * is true. In either case, if no such character occurs in this
2499 * string at or after position {@code beginIndex} and before position
2500 * {@code endIndex}, then {@code -1} is returned.
2501 *
2502 * <p>All indices are specified in {@code char} values
2503 * (Unicode code units).
2504 *
2505 * @param ch a character (Unicode code point).
2506 * @param beginIndex the index to start the search from (included).
2507 * @param endIndex the index to stop the search at (excluded).
2508 * @return the index of the first occurrence of the character in the
2509 * character sequence represented by this object that is greater
2510 * than or equal to {@code beginIndex} and less than {@code endIndex},
2511 * or {@code -1} if the character does not occur.
2512 * @throws StringIndexOutOfBoundsException if {@code beginIndex}
2513 * is negative, or {@code endIndex} is larger than the length of
2514 * this {@code String} object, or {@code beginIndex} is larger than
2515 * {@code endIndex}.
2516 * @since 21
2517 */
2518 public int indexOf(int ch, int beginIndex, int endIndex) {
2519 checkBoundsBeginEnd(beginIndex, endIndex, length());
2520 return isLatin1() ? StringLatin1.indexOf(value, ch, beginIndex, endIndex)
2521 : StringUTF16.indexOf(value, ch, beginIndex, endIndex);
2522 }
2523
2524 /**
2525 * Returns the index within this string of the last occurrence of
2526 * the specified character. For values of {@code ch} in the
2527 * range from 0 to 0xFFFF (inclusive), the index (in Unicode code
2528 * units) returned is the largest value <i>k</i> such that:
2529 * <blockquote><pre>
2530 * this.charAt(<i>k</i>) == ch
2531 * </pre></blockquote>
2532 * is true. For other values of {@code ch}, it is the
2533 * largest value <i>k</i> such that:
2534 * <blockquote><pre>
2535 * this.codePointAt(<i>k</i>) == ch
2536 * </pre></blockquote>
2537 * is true. In either case, if no such character occurs in this
2538 * string, then {@code -1} is returned. The
2539 * {@code String} is searched backwards starting at the last
2540 * character.
2541 *
2542 * @param ch a character (Unicode code point).
2543 * @return the index of the last occurrence of the character in the
2544 * character sequence represented by this object, or
2545 * {@code -1} if the character does not occur.
2546 */
2547 public int lastIndexOf(int ch) {
2548 return lastIndexOf(ch, length() - 1);
2549 }
2550
2551 /**
2552 * Returns the index within this string of the last occurrence of
2553 * the specified character, searching backward starting at the
2554 * specified index. For values of {@code ch} in the range
2555 * from 0 to 0xFFFF (inclusive), the index returned is the largest
2556 * value <i>k</i> such that:
2557 * <blockquote><pre>
2558 * (this.charAt(<i>k</i>) == ch) {@code &&} (<i>k</i> <= fromIndex)
2559 * </pre></blockquote>
2560 * is true. For other values of {@code ch}, it is the
2561 * largest value <i>k</i> such that:
2562 * <blockquote><pre>
2563 * (this.codePointAt(<i>k</i>) == ch) {@code &&} (<i>k</i> <= fromIndex)
2564 * </pre></blockquote>
2565 * is true. In either case, if no such character occurs in this
2566 * string at or before position {@code fromIndex}, then
2567 * {@code -1} is returned.
2568 *
2569 * <p>All indices are specified in {@code char} values
2570 * (Unicode code units).
2571 *
2572 * @param ch a character (Unicode code point).
2573 * @param fromIndex the index to start the search from. There is no
2574 * restriction on the value of {@code fromIndex}. If it is
2575 * greater than or equal to the length of this string, it has
2576 * the same effect as if it were equal to one less than the
2577 * length of this string: this entire string may be searched.
2578 * If it is negative, it has the same effect as if it were -1:
2579 * -1 is returned.
2580 * @return the index of the last occurrence of the character in the
2581 * character sequence represented by this object that is less
2582 * than or equal to {@code fromIndex}, or {@code -1}
2583 * if the character does not occur before that point.
2584 */
2585 public int lastIndexOf(int ch, int fromIndex) {
2586 return isLatin1() ? StringLatin1.lastIndexOf(value, ch, fromIndex)
2587 : StringUTF16.lastIndexOf(value, ch, fromIndex);
2588 }
2589
2590 /**
2591 * Returns the index within this string of the first occurrence of the
2592 * specified substring.
2593 *
2594 * <p>The returned index is the smallest value {@code k} for which:
2595 * <pre>{@code
2596 * this.startsWith(str, k)
2597 * }</pre>
2598 * If no such value of {@code k} exists, then {@code -1} is returned.
2599 *
2600 * @param str the substring to search for.
2601 * @return the index of the first occurrence of the specified substring,
2602 * or {@code -1} if there is no such occurrence.
2603 */
2604 public int indexOf(String str) {
2605 byte coder = coder();
2606 if (coder == str.coder()) {
2607 return isLatin1() ? StringLatin1.indexOf(value, str.value)
2608 : StringUTF16.indexOf(value, str.value);
2609 }
2610 if (coder == LATIN1) { // str.coder == UTF16
2611 return -1;
2612 }
2613 return StringUTF16.indexOfLatin1(value, str.value);
2614 }
2615
2616 /**
2617 * Returns the index within this string of the first occurrence of the
2618 * specified substring, starting at the specified index.
2619 *
2620 * <p>The returned index is the smallest value {@code k} for which:
2621 * <pre>{@code
2622 * k >= Math.min(fromIndex, this.length()) &&
2623 * this.startsWith(str, k)
2624 * }</pre>
2625 * If no such value of {@code k} exists, then {@code -1} is returned.
2626 *
2627 * @apiNote
2628 * Unlike {@link #substring(int)}, for example, this method does not throw
2629 * an exception when {@code fromIndex} is outside the valid range.
2630 * Rather, it returns -1 when {@code fromIndex} is larger than the length of
2631 * the string.
2632 * This result is, by itself, indistinguishable from a genuine absence of
2633 * {@code str} in the string.
2634 * If stricter behavior is needed, {@link #indexOf(String, int, int)}
2635 * should be considered instead.
2636 * On {@link String} {@code s} and a non-empty {@code str}, for example,
2637 * {@code s.indexOf(str, fromIndex, s.length())} would throw if
2638 * {@code fromIndex} were larger than the string length, or were negative.
2639 *
2640 * @param str the substring to search for.
2641 * @param fromIndex the index from which to start the search.
2642 * @return the index of the first occurrence of the specified substring,
2643 * starting at the specified index,
2644 * or {@code -1} if there is no such occurrence.
2645 */
2646 public int indexOf(String str, int fromIndex) {
2647 return indexOf(value, coder(), length(), str, fromIndex);
2648 }
2649
2650 /**
2651 * Returns the index of the first occurrence of the specified substring
2652 * within the specified index range of {@code this} string.
2653 *
2654 * <p>This method returns the same result as the one of the invocation
2655 * <pre>{@code
2656 * s.substring(beginIndex, endIndex).indexOf(str) + beginIndex
2657 * }</pre>
2658 * if the index returned by {@link #indexOf(String)} is non-negative,
2659 * and returns -1 otherwise.
2660 * (No substring is instantiated, though.)
2661 *
2662 * @param str the substring to search for.
2663 * @param beginIndex the index to start the search from (included).
2664 * @param endIndex the index to stop the search at (excluded).
2665 * @return the index of the first occurrence of the specified substring
2666 * within the specified index range,
2667 * or {@code -1} if there is no such occurrence.
2668 * @throws StringIndexOutOfBoundsException if {@code beginIndex}
2669 * is negative, or {@code endIndex} is larger than the length of
2670 * this {@code String} object, or {@code beginIndex} is larger than
2671 * {@code endIndex}.
2672 * @since 21
2673 */
2674 public int indexOf(String str, int beginIndex, int endIndex) {
2675 if (str.length() == 1) {
2676 /* Simple optimization, can be omitted without behavioral impact */
2677 return indexOf(str.charAt(0), beginIndex, endIndex);
2678 }
2679 checkBoundsBeginEnd(beginIndex, endIndex, length());
2680 return indexOf(value, coder(), endIndex, str, beginIndex);
2681 }
2682
2683 /**
2684 * Code shared by String and AbstractStringBuilder to do searches. The
2685 * source is the character array being searched, and the target
2686 * is the string being searched for.
2687 *
2688 * @param src the characters being searched.
2689 * @param srcCoder the coder of the source string.
2690 * @param srcCount last index (exclusive) in the source string.
2691 * @param tgtStr the characters being searched for.
2692 * @param fromIndex the index to begin searching from.
2693 */
2694 static int indexOf(byte[] src, byte srcCoder, int srcCount,
2695 String tgtStr, int fromIndex) {
2696 fromIndex = Math.clamp(fromIndex, 0, srcCount);
2697 int tgtCount = tgtStr.length();
2698 if (tgtCount > srcCount - fromIndex) {
2699 return -1;
2700 }
2701 if (tgtCount == 0) {
2702 return fromIndex;
2703 }
2704
2705 byte[] tgt = tgtStr.value;
2706 byte tgtCoder = tgtStr.coder();
2707 if (srcCoder == tgtCoder) {
2708 return srcCoder == LATIN1
2709 ? StringLatin1.indexOf(src, srcCount, tgt, tgtCount, fromIndex)
2710 : StringUTF16.indexOf(src, srcCount, tgt, tgtCount, fromIndex);
2711 }
2712 if (srcCoder == LATIN1) { // && tgtCoder == UTF16
2713 return -1;
2714 }
2715 // srcCoder == UTF16 && tgtCoder == LATIN1) {
2716 return StringUTF16.indexOfLatin1(src, srcCount, tgt, tgtCount, fromIndex);
2717 }
2718
2719 /**
2720 * Returns the index within this string of the last occurrence of the
2721 * specified substring. The last occurrence of the empty string ""
2722 * is considered to occur at the index value {@code this.length()}.
2723 *
2724 * <p>The returned index is the largest value {@code k} for which:
2725 * <pre>{@code
2726 * this.startsWith(str, k)
2727 * }</pre>
2728 * If no such value of {@code k} exists, then {@code -1} is returned.
2729 *
2730 * @param str the substring to search for.
2731 * @return the index of the last occurrence of the specified substring,
2732 * or {@code -1} if there is no such occurrence.
2733 */
2734 public int lastIndexOf(String str) {
2735 return lastIndexOf(str, length());
2736 }
2737
2738 /**
2739 * Returns the index within this string of the last occurrence of the
2740 * specified substring, searching backward starting at the specified index.
2741 *
2742 * <p>The returned index is the largest value {@code k} for which:
2743 * <pre>{@code
2744 * k <= Math.min(fromIndex, this.length()) &&
2745 * this.startsWith(str, k)
2746 * }</pre>
2747 * If no such value of {@code k} exists, then {@code -1} is returned.
2748 *
2749 * @param str the substring to search for.
2750 * @param fromIndex the index to start the search from.
2751 * @return the index of the last occurrence of the specified substring,
2752 * searching backward from the specified index,
2753 * or {@code -1} if there is no such occurrence.
2754 */
2755 public int lastIndexOf(String str, int fromIndex) {
2756 return lastIndexOf(value, coder(), length(), str, fromIndex);
2757 }
2758
2759 /**
2760 * Code shared by String and AbstractStringBuilder to do searches. The
2761 * source is the character array being searched, and the target
2762 * is the string being searched for.
2763 *
2764 * @param src the characters being searched.
2765 * @param srcCoder coder handles the mapping between bytes/chars
2766 * @param srcCount count of the source string.
2767 * @param tgtStr the characters being searched for.
2768 * @param fromIndex the index to begin searching from.
2769 */
2770 static int lastIndexOf(byte[] src, byte srcCoder, int srcCount,
2771 String tgtStr, int fromIndex) {
2772 byte[] tgt = tgtStr.value;
2773 byte tgtCoder = tgtStr.coder();
2774 int tgtCount = tgtStr.length();
2775 /*
2776 * Check arguments; return immediately where possible. For
2777 * consistency, don't check for null str.
2778 */
2779 int rightIndex = srcCount - tgtCount;
2780 if (fromIndex > rightIndex) {
2781 fromIndex = rightIndex;
2782 }
2783 if (fromIndex < 0) {
2784 return -1;
2785 }
2786 /* Empty string always matches. */
2787 if (tgtCount == 0) {
2788 return fromIndex;
2789 }
2790 if (srcCoder == tgtCoder) {
2791 return srcCoder == LATIN1
2792 ? StringLatin1.lastIndexOf(src, srcCount, tgt, tgtCount, fromIndex)
2793 : StringUTF16.lastIndexOf(src, srcCount, tgt, tgtCount, fromIndex);
2794 }
2795 if (srcCoder == LATIN1) { // && tgtCoder == UTF16
2796 return -1;
2797 }
2798 // srcCoder == UTF16 && tgtCoder == LATIN1
2799 return StringUTF16.lastIndexOfLatin1(src, srcCount, tgt, tgtCount, fromIndex);
2800 }
2801
2802 /**
2803 * Returns a string that is a substring of this string. The
2804 * substring begins with the character at the specified index and
2805 * extends to the end of this string. <p>
2806 * Examples:
2807 * <blockquote><pre>
2808 * "unhappy".substring(2) returns "happy"
2809 * "Harbison".substring(3) returns "bison"
2810 * "emptiness".substring(9) returns "" (an empty string)
2811 * </pre></blockquote>
2812 *
2813 * @param beginIndex the beginning index, inclusive.
2814 * @return the specified substring.
2815 * @throws IndexOutOfBoundsException if
2816 * {@code beginIndex} is negative or larger than the
2817 * length of this {@code String} object.
2818 */
2819 public String substring(int beginIndex) {
2820 return substring(beginIndex, length());
2821 }
2822
2823 /**
2824 * Returns a string that is a substring of this string. The
2825 * substring begins at the specified {@code beginIndex} and
2826 * extends to the character at index {@code endIndex - 1}.
2827 * Thus the length of the substring is {@code endIndex-beginIndex}.
2828 * <p>
2829 * Examples:
2830 * <blockquote><pre>
2831 * "hamburger".substring(4, 8) returns "urge"
2832 * "smiles".substring(1, 5) returns "mile"
2833 * </pre></blockquote>
2834 *
2835 * @param beginIndex the beginning index, inclusive.
2836 * @param endIndex the ending index, exclusive.
2837 * @return the specified substring.
2838 * @throws IndexOutOfBoundsException if the
2839 * {@code beginIndex} is negative, or
2840 * {@code endIndex} is larger than the length of
2841 * this {@code String} object, or
2842 * {@code beginIndex} is larger than
2843 * {@code endIndex}.
2844 */
2845 public String substring(int beginIndex, int endIndex) {
2846 int length = length();
2847 checkBoundsBeginEnd(beginIndex, endIndex, length);
2848 if (beginIndex == 0 && endIndex == length) {
2849 return this;
2850 }
2851 int subLen = endIndex - beginIndex;
2852 return isLatin1() ? StringLatin1.newString(value, beginIndex, subLen)
2853 : StringUTF16.newString(value, beginIndex, subLen);
2854 }
2855
2856 /**
2857 * Returns a character sequence that is a subsequence of this sequence.
2858 *
2859 * <p> An invocation of this method of the form
2860 *
2861 * <blockquote><pre>
2862 * str.subSequence(begin, end)</pre></blockquote>
2863 *
2864 * behaves in exactly the same way as the invocation
2865 *
2866 * <blockquote><pre>
2867 * str.substring(begin, end)</pre></blockquote>
2868 *
2869 * @apiNote
2870 * This method is defined so that the {@code String} class can implement
2871 * the {@link CharSequence} interface.
2872 *
2873 * @param beginIndex the begin index, inclusive.
2874 * @param endIndex the end index, exclusive.
2875 * @return the specified subsequence.
2876 *
2877 * @throws IndexOutOfBoundsException
2878 * if {@code beginIndex} or {@code endIndex} is negative,
2879 * if {@code endIndex} is greater than {@code length()},
2880 * or if {@code beginIndex} is greater than {@code endIndex}
2881 *
2882 * @since 1.4
2883 */
2884 public CharSequence subSequence(int beginIndex, int endIndex) {
2885 return this.substring(beginIndex, endIndex);
2886 }
2887
2888 /**
2889 * Concatenates the specified string to the end of this string.
2890 * <p>
2891 * If the length of the argument string is {@code 0}, then this
2892 * {@code String} object is returned. Otherwise, a
2893 * {@code String} object is returned that represents a character
2894 * sequence that is the concatenation of the character sequence
2895 * represented by this {@code String} object and the character
2896 * sequence represented by the argument string.<p>
2897 * Examples:
2898 * <blockquote><pre>
2899 * "cares".concat("s") returns "caress"
2900 * "to".concat("get").concat("her") returns "together"
2901 * </pre></blockquote>
2902 *
2903 * @param str the {@code String} that is concatenated to the end
2904 * of this {@code String}.
2905 * @return a string that represents the concatenation of this object's
2906 * characters followed by the string argument's characters.
2907 */
2908 public String concat(String str) {
2909 if (str.isEmpty()) {
2910 return this;
2911 }
2912 return StringConcatHelper.simpleConcat(this, str);
2913 }
2914
2915 /**
2916 * Returns a string resulting from replacing all occurrences of
2917 * {@code oldChar} in this string with {@code newChar}.
2918 * <p>
2919 * If the character {@code oldChar} does not occur in the
2920 * character sequence represented by this {@code String} object,
2921 * then a reference to this {@code String} object is returned.
2922 * Otherwise, a {@code String} object is returned that
2923 * represents a character sequence identical to the character sequence
2924 * represented by this {@code String} object, except that every
2925 * occurrence of {@code oldChar} is replaced by an occurrence
2926 * of {@code newChar}.
2927 * <p>
2928 * Examples:
2929 * <blockquote><pre>
2930 * "mesquite in your cellar".replace('e', 'o')
2931 * returns "mosquito in your collar"
2932 * "the war of baronets".replace('r', 'y')
2933 * returns "the way of bayonets"
2934 * "sparring with a purple porpoise".replace('p', 't')
2935 * returns "starring with a turtle tortoise"
2936 * "JonL".replace('q', 'x') returns "JonL" (no change)
2937 * </pre></blockquote>
2938 *
2939 * @param oldChar the old character.
2940 * @param newChar the new character.
2941 * @return a string derived from this string by replacing every
2942 * occurrence of {@code oldChar} with {@code newChar}.
2943 */
2944 public String replace(char oldChar, char newChar) {
2945 if (oldChar != newChar) {
2946 String ret = isLatin1() ? StringLatin1.replace(value, oldChar, newChar)
2947 : StringUTF16.replace(value, oldChar, newChar);
2948 if (ret != null) {
2949 return ret;
2950 }
2951 }
2952 return this;
2953 }
2954
2955 /**
2956 * Tells whether or not this string matches the given <a
2957 * href="../util/regex/Pattern.html#sum">regular expression</a>.
2958 *
2959 * <p> An invocation of this method of the form
2960 * <i>str</i>{@code .matches(}<i>regex</i>{@code )} yields exactly the
2961 * same result as the expression
2962 *
2963 * <blockquote>
2964 * {@link java.util.regex.Pattern}.{@link java.util.regex.Pattern#matches(String,CharSequence)
2965 * matches(<i>regex</i>, <i>str</i>)}
2966 * </blockquote>
2967 *
2968 * @param regex
2969 * the regular expression to which this string is to be matched
2970 *
2971 * @return {@code true} if, and only if, this string matches the
2972 * given regular expression
2973 *
2974 * @throws PatternSyntaxException
2975 * if the regular expression's syntax is invalid
2976 *
2977 * @see java.util.regex.Pattern
2978 *
2979 * @since 1.4
2980 */
2981 public boolean matches(String regex) {
2982 return Pattern.matches(regex, this);
2983 }
2984
2985 /**
2986 * Returns true if and only if this string contains the specified
2987 * sequence of char values.
2988 *
2989 * @param s the sequence to search for
2990 * @return true if this string contains {@code s}, false otherwise
2991 * @since 1.5
2992 */
2993 public boolean contains(CharSequence s) {
2994 return indexOf(s.toString()) >= 0;
2995 }
2996
2997 /**
2998 * Replaces the first substring of this string that matches the given <a
2999 * href="../util/regex/Pattern.html#sum">regular expression</a> with the
3000 * given replacement.
3001 *
3002 * <p> An invocation of this method of the form
3003 * <i>str</i>{@code .replaceFirst(}<i>regex</i>{@code ,} <i>repl</i>{@code )}
3004 * yields exactly the same result as the expression
3005 *
3006 * <blockquote>
3007 * <code>
3008 * {@link java.util.regex.Pattern}.{@link
3009 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
3010 * java.util.regex.Pattern#matcher(java.lang.CharSequence) matcher}(<i>str</i>).{@link
3011 * java.util.regex.Matcher#replaceFirst(String) replaceFirst}(<i>repl</i>)
3012 * </code>
3013 * </blockquote>
3014 *
3015 *<p>
3016 * Note that backslashes ({@code \}) and dollar signs ({@code $}) in the
3017 * replacement string may cause the results to be different than if it were
3018 * being treated as a literal replacement string; see
3019 * {@link java.util.regex.Matcher#replaceFirst}.
3020 * Use {@link java.util.regex.Matcher#quoteReplacement} to suppress the special
3021 * meaning of these characters, if desired.
3022 *
3023 * @param regex
3024 * the regular expression to which this string is to be matched
3025 * @param replacement
3026 * the string to be substituted for the first match
3027 *
3028 * @return The resulting {@code String}
3029 *
3030 * @throws PatternSyntaxException
3031 * if the regular expression's syntax is invalid
3032 *
3033 * @see java.util.regex.Pattern
3034 *
3035 * @since 1.4
3036 */
3037 public String replaceFirst(String regex, String replacement) {
3038 return Pattern.compile(regex).matcher(this).replaceFirst(replacement);
3039 }
3040
3041 /**
3042 * Replaces each substring of this string that matches the given <a
3043 * href="../util/regex/Pattern.html#sum">regular expression</a> with the
3044 * given replacement.
3045 *
3046 * <p> An invocation of this method of the form
3047 * <i>str</i>{@code .replaceAll(}<i>regex</i>{@code ,} <i>repl</i>{@code )}
3048 * yields exactly the same result as the expression
3049 *
3050 * <blockquote>
3051 * <code>
3052 * {@link java.util.regex.Pattern}.{@link
3053 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
3054 * java.util.regex.Pattern#matcher(java.lang.CharSequence) matcher}(<i>str</i>).{@link
3055 * java.util.regex.Matcher#replaceAll(String) replaceAll}(<i>repl</i>)
3056 * </code>
3057 * </blockquote>
3058 *
3059 *<p>
3060 * Note that backslashes ({@code \}) and dollar signs ({@code $}) in the
3061 * replacement string may cause the results to be different than if it were
3062 * being treated as a literal replacement string; see
3063 * {@link java.util.regex.Matcher#replaceAll Matcher.replaceAll}.
3064 * Use {@link java.util.regex.Matcher#quoteReplacement} to suppress the special
3065 * meaning of these characters, if desired.
3066 *
3067 * @param regex
3068 * the regular expression to which this string is to be matched
3069 * @param replacement
3070 * the string to be substituted for each match
3071 *
3072 * @return The resulting {@code String}
3073 *
3074 * @throws PatternSyntaxException
3075 * if the regular expression's syntax is invalid
3076 *
3077 * @see java.util.regex.Pattern
3078 *
3079 * @since 1.4
3080 */
3081 public String replaceAll(String regex, String replacement) {
3082 return Pattern.compile(regex).matcher(this).replaceAll(replacement);
3083 }
3084
3085 /**
3086 * Replaces each substring of this string that matches the literal target
3087 * sequence with the specified literal replacement sequence. The
3088 * replacement proceeds from the beginning of the string to the end, for
3089 * example, replacing "aa" with "b" in the string "aaa" will result in
3090 * "ba" rather than "ab".
3091 *
3092 * @param target The sequence of char values to be replaced
3093 * @param replacement The replacement sequence of char values
3094 * @return The resulting string
3095 * @since 1.5
3096 */
3097 public String replace(CharSequence target, CharSequence replacement) {
3098 String trgtStr = target.toString();
3099 String replStr = replacement.toString();
3100 int thisLen = length();
3101 int trgtLen = trgtStr.length();
3102 int replLen = replStr.length();
3103
3104 if (trgtLen > 0) {
3105 if (trgtLen == 1 && replLen == 1) {
3106 return replace(trgtStr.charAt(0), replStr.charAt(0));
3107 }
3108
3109 boolean thisIsLatin1 = this.isLatin1();
3110 boolean trgtIsLatin1 = trgtStr.isLatin1();
3111 boolean replIsLatin1 = replStr.isLatin1();
3112 String ret = (thisIsLatin1 && trgtIsLatin1 && replIsLatin1)
3113 ? StringLatin1.replace(value, thisLen,
3114 trgtStr.value, trgtLen,
3115 replStr.value, replLen)
3116 : StringUTF16.replace(value, thisLen, thisIsLatin1,
3117 trgtStr.value, trgtLen, trgtIsLatin1,
3118 replStr.value, replLen, replIsLatin1);
3119 if (ret != null) {
3120 return ret;
3121 }
3122 return this;
3123
3124 } else { // trgtLen == 0
3125 int resultLen;
3126 try {
3127 resultLen = Math.addExact(thisLen, Math.multiplyExact(
3128 Math.addExact(thisLen, 1), replLen));
3129 } catch (ArithmeticException ignored) {
3130 throw new OutOfMemoryError("Required length exceeds implementation limit");
3131 }
3132
3133 StringBuilder sb = new StringBuilder(resultLen);
3134 sb.append(replStr);
3135 for (int i = 0; i < thisLen; ++i) {
3136 sb.append(charAt(i)).append(replStr);
3137 }
3138 return sb.toString();
3139 }
3140 }
3141
3142 /**
3143 * Splits this string around matches of the given
3144 * <a href="../util/regex/Pattern.html#sum">regular expression</a>.
3145 *
3146 * <p> The array returned by this method contains each substring of this
3147 * string that is terminated by another substring that matches the given
3148 * expression or is terminated by the end of the string. The substrings in
3149 * the array are in the order in which they occur in this string. If the
3150 * expression does not match any part of the input then the resulting array
3151 * has just one element, namely this string.
3152 *
3153 * <p> When there is a positive-width match at the beginning of this
3154 * string then an empty leading substring is included at the beginning
3155 * of the resulting array. A zero-width match at the beginning however
3156 * never produces such empty leading substring.
3157 *
3158 * <p> The {@code limit} parameter controls the number of times the
3159 * pattern is applied and therefore affects the length of the resulting
3160 * array.
3161 * <ul>
3162 * <li><p>
3163 * If the <i>limit</i> is positive then the pattern will be applied
3164 * at most <i>limit</i> - 1 times, the array's length will be
3165 * no greater than <i>limit</i>, and the array's last entry will contain
3166 * all input beyond the last matched delimiter.</p></li>
3167 *
3168 * <li><p>
3169 * If the <i>limit</i> is zero then the pattern will be applied as
3170 * many times as possible, the array can have any length, and trailing
3171 * empty strings will be discarded.</p></li>
3172 *
3173 * <li><p>
3174 * If the <i>limit</i> is negative then the pattern will be applied
3175 * as many times as possible and the array can have any length.</p></li>
3176 * </ul>
3177 *
3178 * <p> The string {@code "boo:and:foo"}, for example, yields the
3179 * following results with these parameters:
3180 *
3181 * <blockquote><table class="plain">
3182 * <caption style="display:none">Split example showing regex, limit, and result</caption>
3183 * <thead>
3184 * <tr>
3185 * <th scope="col">Regex</th>
3186 * <th scope="col">Limit</th>
3187 * <th scope="col">Result</th>
3188 * </tr>
3189 * </thead>
3190 * <tbody>
3191 * <tr><th scope="row" rowspan="3" style="font-weight:normal">:</th>
3192 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">2</th>
3193 * <td>{@code { "boo", "and:foo" }}</td></tr>
3194 * <tr><!-- : -->
3195 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3196 * <td>{@code { "boo", "and", "foo" }}</td></tr>
3197 * <tr><!-- : -->
3198 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-2</th>
3199 * <td>{@code { "boo", "and", "foo" }}</td></tr>
3200 * <tr><th scope="row" rowspan="3" style="font-weight:normal">o</th>
3201 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3202 * <td>{@code { "b", "", ":and:f", "", "" }}</td></tr>
3203 * <tr><!-- o -->
3204 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-2</th>
3205 * <td>{@code { "b", "", ":and:f", "", "" }}</td></tr>
3206 * <tr><!-- o -->
3207 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">0</th>
3208 * <td>{@code { "b", "", ":and:f" }}</td></tr>
3209 * </tbody>
3210 * </table></blockquote>
3211 *
3212 * <p> An invocation of this method of the form
3213 * <i>str.</i>{@code split(}<i>regex</i>{@code ,} <i>n</i>{@code )}
3214 * yields the same result as the expression
3215 *
3216 * <blockquote>
3217 * <code>
3218 * {@link java.util.regex.Pattern}.{@link
3219 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
3220 * java.util.regex.Pattern#split(java.lang.CharSequence,int) split}(<i>str</i>, <i>n</i>)
3221 * </code>
3222 * </blockquote>
3223 *
3224 *
3225 * @param regex
3226 * the delimiting regular expression
3227 *
3228 * @param limit
3229 * the result threshold, as described above
3230 *
3231 * @return the array of strings computed by splitting this string
3232 * around matches of the given regular expression
3233 *
3234 * @throws PatternSyntaxException
3235 * if the regular expression's syntax is invalid
3236 *
3237 * @see java.util.regex.Pattern
3238 *
3239 * @since 1.4
3240 */
3241 public String[] split(String regex, int limit) {
3242 return split(regex, limit, false);
3243 }
3244
3245 /**
3246 * Splits this string around matches of the given regular expression and
3247 * returns both the strings and the matching delimiters.
3248 *
3249 * <p> The array returned by this method contains each substring of this
3250 * string that is terminated by another substring that matches the given
3251 * expression or is terminated by the end of the string.
3252 * Each substring is immediately followed by the subsequence (the delimiter)
3253 * that matches the given expression, <em>except</em> for the last
3254 * substring, which is not followed by anything.
3255 * The substrings in the array and the delimiters are in the order in which
3256 * they occur in the input.
3257 * If the expression does not match any part of the input then the resulting
3258 * array has just one element, namely this string.
3259 *
3260 * <p> When there is a positive-width match at the beginning of this
3261 * string then an empty leading substring is included at the beginning
3262 * of the resulting array. A zero-width match at the beginning however
3263 * never produces such empty leading substring nor the empty delimiter.
3264 *
3265 * <p> The {@code limit} parameter controls the number of times the
3266 * pattern is applied and therefore affects the length of the resulting
3267 * array.
3268 * <ul>
3269 * <li> If the <i>limit</i> is positive then the pattern will be applied
3270 * at most <i>limit</i> - 1 times, the array's length will be
3271 * no greater than 2 × <i>limit</i> - 1, and the array's last
3272 * entry will contain all input beyond the last matched delimiter.</li>
3273 *
3274 * <li> If the <i>limit</i> is zero then the pattern will be applied as
3275 * many times as possible, the array can have any length, and trailing
3276 * empty strings will be discarded.</li>
3277 *
3278 * <li> If the <i>limit</i> is negative then the pattern will be applied
3279 * as many times as possible and the array can have any length.</li>
3280 * </ul>
3281 *
3282 * <p> The input {@code "boo:::and::foo"}, for example, yields the following
3283 * results with these parameters:
3284 *
3285 * <table class="plain" style="margin-left:2em;">
3286 * <caption style="display:none">Split example showing regex, limit, and result</caption>
3287 * <thead>
3288 * <tr>
3289 * <th scope="col">Regex</th>
3290 * <th scope="col">Limit</th>
3291 * <th scope="col">Result</th>
3292 * </tr>
3293 * </thead>
3294 * <tbody>
3295 * <tr><th scope="row" rowspan="3" style="font-weight:normal">:+</th>
3296 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">2</th>
3297 * <td>{@code { "boo", ":::", "and::foo" }}</td></tr>
3298 * <tr><!-- : -->
3299 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3300 * <td>{@code { "boo", ":::", "and", "::", "foo" }}</td></tr>
3301 * <tr><!-- : -->
3302 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-1</th>
3303 * <td>{@code { "boo", ":::", "and", "::", "foo" }}</td></tr>
3304 * <tr><th scope="row" rowspan="3" style="font-weight:normal">o</th>
3305 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">5</th>
3306 * <td>{@code { "b", "o", "", "o", ":::and::f", "o", "", "o", "" }}</td></tr>
3307 * <tr><!-- o -->
3308 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">-1</th>
3309 * <td>{@code { "b", "o", "", "o", ":::and::f", "o", "", "o", "" }}</td></tr>
3310 * <tr><!-- o -->
3311 * <th scope="row" style="font-weight:normal; text-align:right; padding-right:1em">0</th>
3312 * <td>{@code { "b", "o", "", "o", ":::and::f", "o", "", "o" }}</td></tr>
3313 * </tbody>
3314 * </table>
3315 *
3316 * @apiNote An invocation of this method of the form
3317 * <i>str.</i>{@code splitWithDelimiters(}<i>regex</i>{@code ,} <i>n</i>{@code )}
3318 * yields the same result as the expression
3319 *
3320 * <blockquote>
3321 * <code>
3322 * {@link java.util.regex.Pattern}.{@link
3323 * java.util.regex.Pattern#compile(String) compile}(<i>regex</i>).{@link
3324 * java.util.regex.Pattern#splitWithDelimiters(CharSequence,int) splitWithDelimiters}(<i>str</i>, <i>n</i>)
3325 * </code>
3326 * </blockquote>
3327 *
3328 * @param regex
3329 * the delimiting regular expression
3330 *
3331 * @param limit
3332 * the result threshold, as described above
3333 *
3334 * @return the array of strings computed by splitting this string
3335 * around matches of the given regular expression, alternating
3336 * substrings and matching delimiters
3337 *
3338 * @since 21
3339 */
3340 public String[] splitWithDelimiters(String regex, int limit) {
3341 return split(regex, limit, true);
3342 }
3343
3344 private String[] split(String regex, int limit, boolean withDelimiters) {
3345 /* fastpath if the regex is a
3346 * (1) one-char String and this character is not one of the
3347 * RegEx's meta characters ".$|()[{^?*+\\", or
3348 * (2) two-char String and the first char is the backslash and
3349 * the second is not the ascii digit or ascii letter.
3350 */
3351 char ch = 0;
3352 if (((regex.length() == 1 &&
3353 ".$|()[{^?*+\\".indexOf(ch = regex.charAt(0)) == -1) ||
3354 (regex.length() == 2 &&
3355 regex.charAt(0) == '\\' &&
3356 (((ch = regex.charAt(1))-'0')|('9'-ch)) < 0 &&
3357 ((ch-'a')|('z'-ch)) < 0 &&
3358 ((ch-'A')|('Z'-ch)) < 0)) &&
3359 (ch < Character.MIN_HIGH_SURROGATE ||
3360 ch > Character.MAX_LOW_SURROGATE))
3361 {
3362 // All the checks above can potentially be constant folded by
3363 // a JIT/AOT compiler when the regex is a constant string.
3364 // That requires method inlining of the checks, which is only
3365 // possible when the actual split logic is in a separate method
3366 // because the large split loop can usually not be inlined.
3367 return split(ch, limit, withDelimiters);
3368 }
3369 Pattern pattern = Pattern.compile(regex);
3370 return withDelimiters
3371 ? pattern.splitWithDelimiters(this, limit)
3372 : pattern.split(this, limit);
3373 }
3374
3375 private String[] split(char ch, int limit, boolean withDelimiters) {
3376 int matchCount = 0;
3377 int off = 0;
3378 int next;
3379 boolean limited = limit > 0;
3380 ArrayList<String> list = new ArrayList<>();
3381 String del = withDelimiters ? String.valueOf(ch) : null;
3382 while ((next = indexOf(ch, off)) != -1) {
3383 if (!limited || matchCount < limit - 1) {
3384 list.add(substring(off, next));
3385 if (withDelimiters) {
3386 list.add(del);
3387 }
3388 off = next + 1;
3389 ++matchCount;
3390 } else { // last one
3391 int last = length();
3392 list.add(substring(off, last));
3393 off = last;
3394 ++matchCount;
3395 break;
3396 }
3397 }
3398 // If no match was found, return this
3399 if (off == 0)
3400 return new String[] {this};
3401
3402 // Add remaining segment
3403 if (!limited || matchCount < limit)
3404 list.add(substring(off, length()));
3405
3406 // Construct result
3407 int resultSize = list.size();
3408 if (limit == 0) {
3409 while (resultSize > 0 && list.get(resultSize - 1).isEmpty()) {
3410 resultSize--;
3411 }
3412 }
3413 String[] result = new String[resultSize];
3414 return list.subList(0, resultSize).toArray(result);
3415 }
3416
3417 /**
3418 * Splits this string around matches of the given <a
3419 * href="../util/regex/Pattern.html#sum">regular expression</a>.
3420 *
3421 * <p> This method works as if by invoking the two-argument {@link
3422 * #split(String, int) split} method with the given expression and a limit
3423 * argument of zero. Trailing empty strings are therefore not included in
3424 * the resulting array.
3425 *
3426 * <p> The string {@code "boo:and:foo"}, for example, yields the following
3427 * results with these expressions:
3428 *
3429 * <blockquote><table class="plain">
3430 * <caption style="display:none">Split examples showing regex and result</caption>
3431 * <thead>
3432 * <tr>
3433 * <th scope="col">Regex</th>
3434 * <th scope="col">Result</th>
3435 * </tr>
3436 * </thead>
3437 * <tbody>
3438 * <tr><th scope="row" style="text-weight:normal">:</th>
3439 * <td>{@code { "boo", "and", "foo" }}</td></tr>
3440 * <tr><th scope="row" style="text-weight:normal">o</th>
3441 * <td>{@code { "b", "", ":and:f" }}</td></tr>
3442 * </tbody>
3443 * </table></blockquote>
3444 *
3445 *
3446 * @param regex
3447 * the delimiting regular expression
3448 *
3449 * @return the array of strings computed by splitting this string
3450 * around matches of the given regular expression
3451 *
3452 * @throws PatternSyntaxException
3453 * if the regular expression's syntax is invalid
3454 *
3455 * @see java.util.regex.Pattern
3456 *
3457 * @since 1.4
3458 */
3459 public String[] split(String regex) {
3460 return split(regex, 0, false);
3461 }
3462
3463 /**
3464 * Returns a new String composed of copies of the
3465 * {@code CharSequence elements} joined together with a copy of
3466 * the specified {@code delimiter}.
3467 *
3468 * <blockquote>For example,
3469 * <pre>{@code
3470 * String message = String.join("-", "Java", "is", "cool");
3471 * // message returned is: "Java-is-cool"
3472 * }</pre></blockquote>
3473 *
3474 * Note that if an element is null, then {@code "null"} is added.
3475 *
3476 * @param delimiter the delimiter that separates each element
3477 * @param elements the elements to join together.
3478 *
3479 * @return a new {@code String} that is composed of the {@code elements}
3480 * separated by the {@code delimiter}
3481 *
3482 * @throws NullPointerException If {@code delimiter} or {@code elements}
3483 * is {@code null}
3484 *
3485 * @see java.util.StringJoiner
3486 * @since 1.8
3487 */
3488 public static String join(CharSequence delimiter, CharSequence... elements) {
3489 var delim = delimiter.toString();
3490 var elems = new String[elements.length];
3491 for (int i = 0; i < elements.length; i++) {
3492 elems[i] = String.valueOf(elements[i]);
3493 }
3494 return join("", "", delim, elems, elems.length);
3495 }
3496
3497 /**
3498 * Designated join routine.
3499 *
3500 * @param prefix the non-null prefix
3501 * @param suffix the non-null suffix
3502 * @param delimiter the non-null delimiter
3503 * @param elements the non-null array of non-null elements
3504 * @param size the number of elements in the array (<= elements.length)
3505 * @return the joined string
3506 */
3507 @ForceInline
3508 static String join(String prefix, String suffix, String delimiter, String[] elements, int size) {
3509 int icoder = prefix.coder() | suffix.coder();
3510 long len = (long) prefix.length() + suffix.length();
3511 if (size > 1) { // when there are more than one element, size - 1 delimiters will be emitted
3512 len += (long) (size - 1) * delimiter.length();
3513 icoder |= delimiter.coder();
3514 }
3515 // assert len > 0L; // max: (long) Integer.MAX_VALUE << 32
3516 // following loop will add max: (long) Integer.MAX_VALUE * Integer.MAX_VALUE to len
3517 // so len can overflow at most once
3518 for (int i = 0; i < size; i++) {
3519 var el = elements[i];
3520 len += el.length();
3521 icoder |= el.coder();
3522 }
3523 byte coder = (byte) icoder;
3524 // long len overflow check, char -> byte length, int len overflow check
3525 if (len < 0L || (len <<= coder) != (int) len) {
3526 throw new OutOfMemoryError("Requested string length exceeds VM limit");
3527 }
3528 byte[] value = StringConcatHelper.newArray(len);
3529
3530 int off = 0;
3531 prefix.getBytes(value, off, coder); off += prefix.length();
3532 if (size > 0) {
3533 var el = elements[0];
3534 el.getBytes(value, off, coder); off += el.length();
3535 for (int i = 1; i < size; i++) {
3536 delimiter.getBytes(value, off, coder); off += delimiter.length();
3537 el = elements[i];
3538 el.getBytes(value, off, coder); off += el.length();
3539 }
3540 }
3541 suffix.getBytes(value, off, coder);
3542 // assert off + suffix.length() == value.length >> coder;
3543
3544 return new String(value, coder);
3545 }
3546
3547 /**
3548 * Returns a new {@code String} composed of copies of the
3549 * {@code CharSequence elements} joined together with a copy of the
3550 * specified {@code delimiter}.
3551 *
3552 * <blockquote>For example,
3553 * <pre>{@code
3554 * List<String> strings = List.of("Java", "is", "cool");
3555 * String message = String.join(" ", strings);
3556 * // message returned is: "Java is cool"
3557 *
3558 * Set<String> strings =
3559 * new LinkedHashSet<>(List.of("Java", "is", "very", "cool"));
3560 * String message = String.join("-", strings);
3561 * // message returned is: "Java-is-very-cool"
3562 * }</pre></blockquote>
3563 *
3564 * Note that if an individual element is {@code null}, then {@code "null"} is added.
3565 *
3566 * @param delimiter a sequence of characters that is used to separate each
3567 * of the {@code elements} in the resulting {@code String}
3568 * @param elements an {@code Iterable} that will have its {@code elements}
3569 * joined together.
3570 *
3571 * @return a new {@code String} that is composed from the {@code elements}
3572 * argument
3573 *
3574 * @throws NullPointerException If {@code delimiter} or {@code elements}
3575 * is {@code null}
3576 *
3577 * @see #join(CharSequence,CharSequence...)
3578 * @see java.util.StringJoiner
3579 * @since 1.8
3580 */
3581 public static String join(CharSequence delimiter,
3582 Iterable<? extends CharSequence> elements) {
3583 Objects.requireNonNull(delimiter);
3584 Objects.requireNonNull(elements);
3585 var delim = delimiter.toString();
3586 var elems = new String[8];
3587 int size = 0;
3588 for (CharSequence cs: elements) {
3589 if (size >= elems.length) {
3590 elems = Arrays.copyOf(elems, elems.length << 1);
3591 }
3592 elems[size++] = String.valueOf(cs);
3593 }
3594 return join("", "", delim, elems, size);
3595 }
3596
3597 /**
3598 * Converts all of the characters in this {@code String} to lower
3599 * case using the rules of the given {@code Locale}. Case mapping is based
3600 * on the Unicode Standard version specified by the {@link java.lang.Character Character}
3601 * class. Since case mappings are not always 1:1 char mappings, the resulting {@code String}
3602 * and this {@code String} may differ in length.
3603 * <p>
3604 * Examples of lowercase mappings are in the following table:
3605 * <table class="plain">
3606 * <caption style="display:none">Lowercase mapping examples showing language code of locale, upper case, lower case, and description</caption>
3607 * <thead>
3608 * <tr>
3609 * <th scope="col">Language Code of Locale</th>
3610 * <th scope="col">Upper Case</th>
3611 * <th scope="col">Lower Case</th>
3612 * <th scope="col">Description</th>
3613 * </tr>
3614 * </thead>
3615 * <tbody>
3616 * <tr>
3617 * <td>tr (Turkish)</td>
3618 * <th scope="row" style="font-weight:normal; text-align:left">\u0130</th>
3619 * <td>\u0069</td>
3620 * <td>capital letter I with dot above -> small letter i</td>
3621 * </tr>
3622 * <tr>
3623 * <td>tr (Turkish)</td>
3624 * <th scope="row" style="font-weight:normal; text-align:left">\u0049</th>
3625 * <td>\u0131</td>
3626 * <td>capital letter I -> small letter dotless i </td>
3627 * </tr>
3628 * <tr>
3629 * <td>(all)</td>
3630 * <th scope="row" style="font-weight:normal; text-align:left">French Fries</th>
3631 * <td>french fries</td>
3632 * <td>lowercased all chars in String</td>
3633 * </tr>
3634 * <tr>
3635 * <td>(all)</td>
3636 * <th scope="row" style="font-weight:normal; text-align:left">
3637 * ΙΧΘΥΣ</th>
3638 * <td>ιχθυσ</td>
3639 * <td>lowercased all chars in String</td>
3640 * </tr>
3641 * </tbody>
3642 * </table>
3643 *
3644 * @param locale use the case transformation rules for this locale
3645 * @return the {@code String}, converted to lowercase.
3646 * @see java.lang.String#toLowerCase()
3647 * @see java.lang.String#toUpperCase()
3648 * @see java.lang.String#toUpperCase(Locale)
3649 * @since 1.1
3650 */
3651 public String toLowerCase(Locale locale) {
3652 return isLatin1() ? StringLatin1.toLowerCase(this, value, locale)
3653 : StringUTF16.toLowerCase(this, value, locale);
3654 }
3655
3656 /**
3657 * Converts all of the characters in this {@code String} to lower
3658 * case using the rules of the default locale. This method is equivalent to
3659 * {@code toLowerCase(Locale.getDefault())}.
3660 *
3661 * @apiNote This method is locale sensitive, and may produce unexpected
3662 * results if used for strings that are intended to be interpreted locale
3663 * independently.
3664 * Examples are programming language identifiers, protocol keys, and HTML
3665 * tags.
3666 * For instance, {@code "TITLE".toLowerCase()} in a Turkish locale
3667 * returns {@code "t\u005Cu0131tle"}, where '\u005Cu0131' is the
3668 * LATIN SMALL LETTER DOTLESS I character.
3669 * To obtain correct results for locale insensitive strings, use
3670 * {@code toLowerCase(Locale.ROOT)}.
3671 *
3672 * @return the {@code String}, converted to lowercase.
3673 * @see java.lang.String#toLowerCase(Locale)
3674 */
3675 public String toLowerCase() {
3676 return toLowerCase(Locale.getDefault());
3677 }
3678
3679 /**
3680 * Converts all of the characters in this {@code String} to upper
3681 * case using the rules of the given {@code Locale}. Case mapping is based
3682 * on the Unicode Standard version specified by the {@link java.lang.Character Character}
3683 * class. Since case mappings are not always 1:1 char mappings, the resulting {@code String}
3684 * and this {@code String} may differ in length.
3685 * <p>
3686 * Examples of locale-sensitive and 1:M case mappings are in the following table:
3687 * <table class="plain">
3688 * <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>
3689 * <thead>
3690 * <tr>
3691 * <th scope="col">Language Code of Locale</th>
3692 * <th scope="col">Lower Case</th>
3693 * <th scope="col">Upper Case</th>
3694 * <th scope="col">Description</th>
3695 * </tr>
3696 * </thead>
3697 * <tbody>
3698 * <tr>
3699 * <td>tr (Turkish)</td>
3700 * <th scope="row" style="font-weight:normal; text-align:left">\u0069</th>
3701 * <td>\u0130</td>
3702 * <td>small letter i -> capital letter I with dot above</td>
3703 * </tr>
3704 * <tr>
3705 * <td>tr (Turkish)</td>
3706 * <th scope="row" style="font-weight:normal; text-align:left">\u0131</th>
3707 * <td>\u0049</td>
3708 * <td>small letter dotless i -> capital letter I</td>
3709 * </tr>
3710 * <tr>
3711 * <td>(all)</td>
3712 * <th scope="row" style="font-weight:normal; text-align:left">\u00df</th>
3713 * <td>\u0053 \u0053</td>
3714 * <td>small letter sharp s -> two letters: SS</td>
3715 * </tr>
3716 * <tr>
3717 * <td>(all)</td>
3718 * <th scope="row" style="font-weight:normal; text-align:left">Fahrvergnügen</th>
3719 * <td>FAHRVERGNÜGEN</td>
3720 * <td></td>
3721 * </tr>
3722 * </tbody>
3723 * </table>
3724 * @param locale use the case transformation rules for this locale
3725 * @return the {@code String}, converted to uppercase.
3726 * @see java.lang.String#toUpperCase()
3727 * @see java.lang.String#toLowerCase()
3728 * @see java.lang.String#toLowerCase(Locale)
3729 * @since 1.1
3730 */
3731 public String toUpperCase(Locale locale) {
3732 return isLatin1() ? StringLatin1.toUpperCase(this, value, locale)
3733 : StringUTF16.toUpperCase(this, value, locale);
3734 }
3735
3736 /**
3737 * Converts all of the characters in this {@code String} to upper
3738 * case using the rules of the default locale. This method is equivalent to
3739 * {@code toUpperCase(Locale.getDefault())}.
3740 *
3741 * @apiNote This method is locale sensitive, and may produce unexpected
3742 * results if used for strings that are intended to be interpreted locale
3743 * independently.
3744 * Examples are programming language identifiers, protocol keys, and HTML
3745 * tags.
3746 * For instance, {@code "title".toUpperCase()} in a Turkish locale
3747 * returns {@code "T\u005Cu0130TLE"}, where '\u005Cu0130' is the
3748 * LATIN CAPITAL LETTER I WITH DOT ABOVE character.
3749 * To obtain correct results for locale insensitive strings, use
3750 * {@code toUpperCase(Locale.ROOT)}.
3751 *
3752 * @return the {@code String}, converted to uppercase.
3753 * @see java.lang.String#toUpperCase(Locale)
3754 */
3755 public String toUpperCase() {
3756 return toUpperCase(Locale.getDefault());
3757 }
3758
3759 /**
3760 * Returns a string whose value is this string, with all leading
3761 * and trailing space removed, where space is defined
3762 * as any character whose codepoint is less than or equal to
3763 * {@code 'U+0020'} (the space character).
3764 * <p>
3765 * If this {@code String} object represents an empty character
3766 * sequence, or the first and last characters of character sequence
3767 * represented by this {@code String} object both have codes
3768 * that are not space (as defined above), then a
3769 * reference to this {@code String} object is returned.
3770 * <p>
3771 * Otherwise, if all characters in this string are space (as
3772 * defined above), then a {@code String} object representing an
3773 * empty string is returned.
3774 * <p>
3775 * Otherwise, let <i>k</i> be the index of the first character in the
3776 * string whose code is not a space (as defined above) and let
3777 * <i>m</i> be the index of the last character in the string whose code
3778 * is not a space (as defined above). A {@code String}
3779 * object is returned, representing the substring of this string that
3780 * begins with the character at index <i>k</i> and ends with the
3781 * character at index <i>m</i>-that is, the result of
3782 * {@code this.substring(k, m + 1)}.
3783 * <p>
3784 * This method may be used to trim space (as defined above) from
3785 * the beginning and end of a string.
3786 *
3787 * @return a string whose value is this string, with all leading
3788 * and trailing space removed, or this string if it
3789 * has no leading or trailing space.
3790 */
3791 public String trim() {
3792 String ret = isLatin1() ? StringLatin1.trim(value)
3793 : StringUTF16.trim(value);
3794 return ret == null ? this : ret;
3795 }
3796
3797 /**
3798 * Returns a string whose value is this string, with all leading
3799 * and trailing {@linkplain Character#isWhitespace(int) white space}
3800 * removed.
3801 * <p>
3802 * If this {@code String} object represents an empty string,
3803 * or if all code points in this string are
3804 * {@linkplain Character#isWhitespace(int) white space}, then an empty string
3805 * is returned.
3806 * <p>
3807 * Otherwise, returns a substring of this string beginning with the first
3808 * code point that is not a {@linkplain Character#isWhitespace(int) white space}
3809 * up to and including the last code point that is not a
3810 * {@linkplain Character#isWhitespace(int) white space}.
3811 * <p>
3812 * This method may be used to strip
3813 * {@linkplain Character#isWhitespace(int) white space} from
3814 * the beginning and end of a string.
3815 *
3816 * @return a string whose value is this string, with all leading
3817 * and trailing white space removed
3818 *
3819 * @see Character#isWhitespace(int)
3820 *
3821 * @since 11
3822 */
3823 public String strip() {
3824 String ret = isLatin1() ? StringLatin1.strip(value)
3825 : StringUTF16.strip(value);
3826 return ret == null ? this : ret;
3827 }
3828
3829 /**
3830 * Returns a string whose value is this string, with all leading
3831 * {@linkplain Character#isWhitespace(int) white space} removed.
3832 * <p>
3833 * If this {@code String} object represents an empty string,
3834 * or if all code points in this string are
3835 * {@linkplain Character#isWhitespace(int) white space}, then an empty string
3836 * is returned.
3837 * <p>
3838 * Otherwise, returns a substring of this string beginning with the first
3839 * code point that is not a {@linkplain Character#isWhitespace(int) white space}
3840 * up to and including the last code point of this string.
3841 * <p>
3842 * This method may be used to trim
3843 * {@linkplain Character#isWhitespace(int) white space} from
3844 * the beginning of a string.
3845 *
3846 * @return a string whose value is this string, with all leading white
3847 * space removed
3848 *
3849 * @see Character#isWhitespace(int)
3850 *
3851 * @since 11
3852 */
3853 public String stripLeading() {
3854 String ret = isLatin1() ? StringLatin1.stripLeading(value)
3855 : StringUTF16.stripLeading(value);
3856 return ret == null ? this : ret;
3857 }
3858
3859 /**
3860 * Returns a string whose value is this string, with all trailing
3861 * {@linkplain Character#isWhitespace(int) white space} removed.
3862 * <p>
3863 * If this {@code String} object represents an empty string,
3864 * or if all characters in this string are
3865 * {@linkplain Character#isWhitespace(int) white space}, then an empty string
3866 * is returned.
3867 * <p>
3868 * Otherwise, returns a substring of this string beginning with the first
3869 * code point of this string up to and including the last code point
3870 * that is not a {@linkplain Character#isWhitespace(int) white space}.
3871 * <p>
3872 * This method may be used to trim
3873 * {@linkplain Character#isWhitespace(int) white space} from
3874 * the end of a string.
3875 *
3876 * @return a string whose value is this string, with all trailing white
3877 * space removed
3878 *
3879 * @see Character#isWhitespace(int)
3880 *
3881 * @since 11
3882 */
3883 public String stripTrailing() {
3884 String ret = isLatin1() ? StringLatin1.stripTrailing(value)
3885 : StringUTF16.stripTrailing(value);
3886 return ret == null ? this : ret;
3887 }
3888
3889 /**
3890 * Returns {@code true} if the string is empty or contains only
3891 * {@linkplain Character#isWhitespace(int) white space} codepoints,
3892 * otherwise {@code false}.
3893 *
3894 * @return {@code true} if the string is empty or contains only
3895 * {@linkplain Character#isWhitespace(int) white space} codepoints,
3896 * otherwise {@code false}
3897 *
3898 * @see Character#isWhitespace(int)
3899 *
3900 * @since 11
3901 */
3902 public boolean isBlank() {
3903 return indexOfNonWhitespace() == length();
3904 }
3905
3906 /**
3907 * Returns a stream of lines extracted from this string,
3908 * separated by line terminators.
3909 * <p>
3910 * A <i>line terminator</i> is one of the following:
3911 * a line feed character {@code "\n"} (U+000A),
3912 * a carriage return character {@code "\r"} (U+000D),
3913 * or a carriage return followed immediately by a line feed
3914 * {@code "\r\n"} (U+000D U+000A).
3915 * <p>
3916 * A <i>line</i> is either a sequence of zero or more characters
3917 * followed by a line terminator, or it is a sequence of one or
3918 * more characters followed by the end of the string. A
3919 * line does not include the line terminator.
3920 * <p>
3921 * The stream returned by this method contains the lines from
3922 * this string in the order in which they occur.
3923 *
3924 * @apiNote This definition of <i>line</i> implies that an empty
3925 * string has zero lines and that there is no empty line
3926 * following a line terminator at the end of a string.
3927 *
3928 * @implNote This method provides better performance than
3929 * split("\R") by supplying elements lazily and
3930 * by faster search of new line terminators.
3931 *
3932 * @return the stream of lines extracted from this string
3933 *
3934 * @since 11
3935 */
3936 public Stream<String> lines() {
3937 return isLatin1() ? StringLatin1.lines(value) : StringUTF16.lines(value);
3938 }
3939
3940 /**
3941 * Adjusts the indentation of each line of this string based on the value of
3942 * {@code n}, and normalizes line termination characters.
3943 * <p>
3944 * This string is conceptually separated into lines using
3945 * {@link String#lines()}. Each line is then adjusted as described below
3946 * and then suffixed with a line feed {@code "\n"} (U+000A). The resulting
3947 * lines are then concatenated and returned.
3948 * <p>
3949 * If {@code n > 0} then {@code n} spaces (U+0020) are inserted at the
3950 * beginning of each line.
3951 * <p>
3952 * If {@code n < 0} then up to {@code n}
3953 * {@linkplain Character#isWhitespace(int) white space characters} are removed
3954 * from the beginning of each line. If a given line does not contain
3955 * sufficient white space then all leading
3956 * {@linkplain Character#isWhitespace(int) white space characters} are removed.
3957 * Each white space character is treated as a single character. In
3958 * particular, the tab character {@code "\t"} (U+0009) is considered a
3959 * single character; it is not expanded.
3960 * <p>
3961 * If {@code n == 0} then the line remains unchanged. However, line
3962 * terminators are still normalized.
3963 *
3964 * @param n number of leading
3965 * {@linkplain Character#isWhitespace(int) white space characters}
3966 * to add or remove
3967 *
3968 * @return string with indentation adjusted and line endings normalized
3969 *
3970 * @see String#lines()
3971 * @see String#isBlank()
3972 * @see Character#isWhitespace(int)
3973 *
3974 * @since 12
3975 */
3976 public String indent(int n) {
3977 if (isEmpty()) {
3978 return "";
3979 }
3980 Stream<String> stream = lines();
3981 if (n > 0) {
3982 final String spaces = " ".repeat(n);
3983 stream = stream.map(s -> spaces + s);
3984 } else if (n == Integer.MIN_VALUE) {
3985 stream = stream.map(s -> s.stripLeading());
3986 } else if (n < 0) {
3987 stream = stream.map(s -> s.substring(Math.min(-n, s.indexOfNonWhitespace())));
3988 }
3989 return stream.collect(Collectors.joining("\n", "", "\n"));
3990 }
3991
3992 private int indexOfNonWhitespace() {
3993 return isLatin1() ? StringLatin1.indexOfNonWhitespace(value)
3994 : StringUTF16.indexOfNonWhitespace(value);
3995 }
3996
3997 private int lastIndexOfNonWhitespace() {
3998 return isLatin1() ? StringLatin1.lastIndexOfNonWhitespace(value)
3999 : StringUTF16.lastIndexOfNonWhitespace(value);
4000 }
4001
4002 /**
4003 * Returns a string whose value is this string, with incidental
4004 * {@linkplain Character#isWhitespace(int) white space} removed from
4005 * the beginning and end of every line.
4006 * <p>
4007 * Incidental {@linkplain Character#isWhitespace(int) white space}
4008 * is often present in a text block to align the content with the opening
4009 * delimiter. For example, in the following code, dots represent incidental
4010 * {@linkplain Character#isWhitespace(int) white space}:
4011 * <blockquote><pre>
4012 * String html = """
4013 * ..............<html>
4014 * .............. <body>
4015 * .............. <p>Hello, world</p>
4016 * .............. </body>
4017 * ..............</html>
4018 * ..............""";
4019 * </pre></blockquote>
4020 * This method treats the incidental
4021 * {@linkplain Character#isWhitespace(int) white space} as indentation to be
4022 * stripped, producing a string that preserves the relative indentation of
4023 * the content. Using | to visualize the start of each line of the string:
4024 * <blockquote><pre>
4025 * |<html>
4026 * | <body>
4027 * | <p>Hello, world</p>
4028 * | </body>
4029 * |</html>
4030 * </pre></blockquote>
4031 * First, the individual lines of this string are extracted. A <i>line</i>
4032 * is a sequence of zero or more characters followed by either a line
4033 * terminator or the end of the string.
4034 * If the string has at least one line terminator, the last line consists
4035 * of the characters between the last terminator and the end of the string.
4036 * Otherwise, if the string has no terminators, the last line is the start
4037 * of the string to the end of the string, in other words, the entire
4038 * string.
4039 * A line does not include the line terminator.
4040 * <p>
4041 * Then, the <i>minimum indentation</i> (min) is determined as follows:
4042 * <ul>
4043 * <li><p>For each non-blank line (as defined by {@link String#isBlank()}),
4044 * the leading {@linkplain Character#isWhitespace(int) white space}
4045 * characters are counted.</p>
4046 * </li>
4047 * <li><p>The leading {@linkplain Character#isWhitespace(int) white space}
4048 * characters on the last line are also counted even if
4049 * {@linkplain String#isBlank() blank}.</p>
4050 * </li>
4051 * </ul>
4052 * <p>The <i>min</i> value is the smallest of these counts.
4053 * <p>
4054 * For each {@linkplain String#isBlank() non-blank} line, <i>min</i> leading
4055 * {@linkplain Character#isWhitespace(int) white space} characters are
4056 * removed, and any trailing {@linkplain Character#isWhitespace(int) white
4057 * space} characters are removed. {@linkplain String#isBlank() Blank} lines
4058 * are replaced with the empty string.
4059 *
4060 * <p>
4061 * Finally, the lines are joined into a new string, using the LF character
4062 * {@code "\n"} (U+000A) to separate lines.
4063 *
4064 * @apiNote
4065 * This method's primary purpose is to shift a block of lines as far as
4066 * possible to the left, while preserving relative indentation. Lines
4067 * that were indented the least will thus have no leading
4068 * {@linkplain Character#isWhitespace(int) white space}.
4069 * The result will have the same number of line terminators as this string.
4070 * If this string ends with a line terminator then the result will end
4071 * with a line terminator.
4072 *
4073 * @implSpec
4074 * This method treats all {@linkplain Character#isWhitespace(int) white space}
4075 * characters as having equal width. As long as the indentation on every
4076 * line is consistently composed of the same character sequences, then the
4077 * result will be as described above.
4078 *
4079 * @return string with incidental indentation removed and line
4080 * terminators normalized
4081 *
4082 * @see String#lines()
4083 * @see String#isBlank()
4084 * @see String#indent(int)
4085 * @see Character#isWhitespace(int)
4086 *
4087 * @since 15
4088 *
4089 */
4090 public String stripIndent() {
4091 int length = length();
4092 if (length == 0) {
4093 return "";
4094 }
4095 char lastChar = charAt(length - 1);
4096 boolean optOut = lastChar == '\n' || lastChar == '\r';
4097 List<String> lines = lines().toList();
4098 final int outdent = optOut ? 0 : outdent(lines);
4099 return lines.stream()
4100 .map(line -> {
4101 int firstNonWhitespace = line.indexOfNonWhitespace();
4102 int lastNonWhitespace = line.lastIndexOfNonWhitespace();
4103 int incidentalWhitespace = Math.min(outdent, firstNonWhitespace);
4104 return firstNonWhitespace > lastNonWhitespace
4105 ? "" : line.substring(incidentalWhitespace, lastNonWhitespace);
4106 })
4107 .collect(Collectors.joining("\n", "", optOut ? "\n" : ""));
4108 }
4109
4110 private static int outdent(List<String> lines) {
4111 // Note: outdent is guaranteed to be zero or positive number.
4112 // If there isn't a non-blank line then the last must be blank
4113 int outdent = Integer.MAX_VALUE;
4114 for (String line : lines) {
4115 int leadingWhitespace = line.indexOfNonWhitespace();
4116 if (leadingWhitespace != line.length()) {
4117 outdent = Integer.min(outdent, leadingWhitespace);
4118 }
4119 }
4120 String lastLine = lines.get(lines.size() - 1);
4121 if (lastLine.isBlank()) {
4122 outdent = Integer.min(outdent, lastLine.length());
4123 }
4124 return outdent;
4125 }
4126
4127 /**
4128 * Returns a string whose value is this string, with escape sequences
4129 * translated as if in a string literal.
4130 * <p>
4131 * Escape sequences are translated as follows;
4132 * <table class="striped">
4133 * <caption style="display:none">Translation</caption>
4134 * <thead>
4135 * <tr>
4136 * <th scope="col">Escape</th>
4137 * <th scope="col">Name</th>
4138 * <th scope="col">Translation</th>
4139 * </tr>
4140 * </thead>
4141 * <tbody>
4142 * <tr>
4143 * <th scope="row">{@code \u005Cb}</th>
4144 * <td>backspace</td>
4145 * <td>{@code U+0008}</td>
4146 * </tr>
4147 * <tr>
4148 * <th scope="row">{@code \u005Ct}</th>
4149 * <td>horizontal tab</td>
4150 * <td>{@code U+0009}</td>
4151 * </tr>
4152 * <tr>
4153 * <th scope="row">{@code \u005Cn}</th>
4154 * <td>line feed</td>
4155 * <td>{@code U+000A}</td>
4156 * </tr>
4157 * <tr>
4158 * <th scope="row">{@code \u005Cf}</th>
4159 * <td>form feed</td>
4160 * <td>{@code U+000C}</td>
4161 * </tr>
4162 * <tr>
4163 * <th scope="row">{@code \u005Cr}</th>
4164 * <td>carriage return</td>
4165 * <td>{@code U+000D}</td>
4166 * </tr>
4167 * <tr>
4168 * <th scope="row">{@code \u005Cs}</th>
4169 * <td>space</td>
4170 * <td>{@code U+0020}</td>
4171 * </tr>
4172 * <tr>
4173 * <th scope="row">{@code \u005C"}</th>
4174 * <td>double quote</td>
4175 * <td>{@code U+0022}</td>
4176 * </tr>
4177 * <tr>
4178 * <th scope="row">{@code \u005C'}</th>
4179 * <td>single quote</td>
4180 * <td>{@code U+0027}</td>
4181 * </tr>
4182 * <tr>
4183 * <th scope="row">{@code \u005C\u005C}</th>
4184 * <td>backslash</td>
4185 * <td>{@code U+005C}</td>
4186 * </tr>
4187 * <tr>
4188 * <th scope="row">{@code \u005C0 - \u005C377}</th>
4189 * <td>octal escape</td>
4190 * <td>code point equivalents</td>
4191 * </tr>
4192 * <tr>
4193 * <th scope="row">{@code \u005C<line-terminator>}</th>
4194 * <td>continuation</td>
4195 * <td>discard</td>
4196 * </tr>
4197 * </tbody>
4198 * </table>
4199 *
4200 * @implNote
4201 * This method does <em>not</em> translate Unicode escapes such as "{@code \u005cu2022}".
4202 * Unicode escapes are translated by the Java compiler when reading input characters and
4203 * are not part of the string literal specification.
4204 *
4205 * @throws IllegalArgumentException when an escape sequence is malformed.
4206 *
4207 * @return String with escape sequences translated.
4208 *
4209 * @jls 3.10.7 Escape Sequences
4210 *
4211 * @since 15
4212 */
4213 public String translateEscapes() {
4214 if (isEmpty()) {
4215 return "";
4216 }
4217 char[] chars = toCharArray();
4218 int length = chars.length;
4219 int from = 0;
4220 int to = 0;
4221 while (from < length) {
4222 char ch = chars[from++];
4223 if (ch == '\\') {
4224 ch = from < length ? chars[from++] : '\0';
4225 switch (ch) {
4226 case 'b':
4227 ch = '\b';
4228 break;
4229 case 'f':
4230 ch = '\f';
4231 break;
4232 case 'n':
4233 ch = '\n';
4234 break;
4235 case 'r':
4236 ch = '\r';
4237 break;
4238 case 's':
4239 ch = ' ';
4240 break;
4241 case 't':
4242 ch = '\t';
4243 break;
4244 case '\'':
4245 case '\"':
4246 case '\\':
4247 // as is
4248 break;
4249 case '0': case '1': case '2': case '3':
4250 case '4': case '5': case '6': case '7':
4251 int limit = Integer.min(from + (ch <= '3' ? 2 : 1), length);
4252 int code = ch - '0';
4253 while (from < limit) {
4254 ch = chars[from];
4255 if (ch < '0' || '7' < ch) {
4256 break;
4257 }
4258 from++;
4259 code = (code << 3) | (ch - '0');
4260 }
4261 ch = (char)code;
4262 break;
4263 case '\n':
4264 continue;
4265 case '\r':
4266 if (from < length && chars[from] == '\n') {
4267 from++;
4268 }
4269 continue;
4270 default: {
4271 String msg = String.format(
4272 "Invalid escape sequence: \\%c \\\\u%04X",
4273 ch, (int)ch);
4274 throw new IllegalArgumentException(msg);
4275 }
4276 }
4277 }
4278
4279 chars[to++] = ch;
4280 }
4281
4282 return new String(chars, 0, to);
4283 }
4284
4285 /**
4286 * This method allows the application of a function to {@code this}
4287 * string. The function should expect a single String argument
4288 * and produce an {@code R} result.
4289 * <p>
4290 * Any exception thrown by {@code f.apply()} will be propagated to the
4291 * caller.
4292 *
4293 * @param f a function to apply
4294 *
4295 * @param <R> the type of the result
4296 *
4297 * @return the result of applying the function to this string
4298 *
4299 * @see java.util.function.Function
4300 *
4301 * @since 12
4302 */
4303 public <R> R transform(Function<? super String, ? extends R> f) {
4304 return f.apply(this);
4305 }
4306
4307 /**
4308 * This object (which is already a string!) is itself returned.
4309 *
4310 * @return the string itself.
4311 */
4312 public String toString() {
4313 return this;
4314 }
4315
4316 /**
4317 * Returns a stream of {@code int} zero-extending the {@code char} values
4318 * from this sequence. Any char which maps to a {@linkplain
4319 * Character##unicode surrogate code point} is passed through
4320 * uninterpreted.
4321 *
4322 * @return an IntStream of char values from this sequence
4323 * @since 9
4324 */
4325 @Override
4326 public IntStream chars() {
4327 return StreamSupport.intStream(
4328 isLatin1() ? new StringLatin1.CharsSpliterator(value, Spliterator.IMMUTABLE)
4329 : new StringUTF16.CharsSpliterator(value, Spliterator.IMMUTABLE),
4330 false);
4331 }
4332
4333
4334 /**
4335 * Returns a stream of code point values from this sequence. Any surrogate
4336 * pairs encountered in the sequence are combined as if by {@linkplain
4337 * Character#toCodePoint Character.toCodePoint} and the result is passed
4338 * to the stream. Any other code units, including ordinary BMP characters,
4339 * unpaired surrogates, and undefined code units, are zero-extended to
4340 * {@code int} values which are then passed to the stream.
4341 *
4342 * @return an IntStream of Unicode code points from this sequence
4343 * @since 9
4344 */
4345 @Override
4346 public IntStream codePoints() {
4347 return StreamSupport.intStream(
4348 isLatin1() ? new StringLatin1.CharsSpliterator(value, Spliterator.IMMUTABLE)
4349 : new StringUTF16.CodePointsSpliterator(value, Spliterator.IMMUTABLE),
4350 false);
4351 }
4352
4353 /**
4354 * Converts this string to a new character array.
4355 *
4356 * @return a newly allocated character array whose length is the length
4357 * of this string and whose contents are initialized to contain
4358 * the character sequence represented by this string.
4359 */
4360 public char[] toCharArray() {
4361 return isLatin1() ? StringLatin1.toChars(value)
4362 : StringUTF16.toChars(value);
4363 }
4364
4365 /**
4366 * Returns a formatted string using the specified format string and
4367 * arguments.
4368 *
4369 * <p> The locale always used is the one returned by {@link
4370 * java.util.Locale#getDefault(java.util.Locale.Category)
4371 * Locale.getDefault(Locale.Category)} with
4372 * {@link java.util.Locale.Category#FORMAT FORMAT} category specified.
4373 *
4374 * @param format
4375 * A <a href="../util/Formatter.html#syntax">format string</a>
4376 *
4377 * @param args
4378 * Arguments referenced by the format specifiers in the format
4379 * string. If there are more arguments than format specifiers, the
4380 * extra arguments are ignored. The number of arguments is
4381 * variable and may be zero. The maximum number of arguments is
4382 * limited by the maximum dimension of a Java array as defined by
4383 * <cite>The Java Virtual Machine Specification</cite>.
4384 * The behaviour on a
4385 * {@code null} argument depends on the <a
4386 * href="../util/Formatter.html#syntax">conversion</a>.
4387 *
4388 * @throws java.util.IllegalFormatException
4389 * If a format string contains an illegal syntax, a format
4390 * specifier that is incompatible with the given arguments,
4391 * insufficient arguments given the format string, or other
4392 * illegal conditions. For specification of all possible
4393 * formatting errors, see the <a
4394 * href="../util/Formatter.html#detail">Details</a> section of the
4395 * formatter class specification.
4396 *
4397 * @return A formatted string
4398 *
4399 * @see java.util.Formatter
4400 * @since 1.5
4401 */
4402 public static String format(String format, Object... args) {
4403 return new Formatter().format(format, args).toString();
4404 }
4405
4406 /**
4407 * Returns a formatted string using the specified locale, format string,
4408 * and arguments.
4409 *
4410 * @param l
4411 * The {@linkplain java.util.Locale locale} to apply during
4412 * formatting. If {@code l} is {@code null} then no localization
4413 * is applied.
4414 *
4415 * @param format
4416 * A <a href="../util/Formatter.html#syntax">format string</a>
4417 *
4418 * @param args
4419 * Arguments referenced by the format specifiers in the format
4420 * string. If there are more arguments than format specifiers, the
4421 * extra arguments are ignored. The number of arguments is
4422 * variable and may be zero. The maximum number of arguments is
4423 * limited by the maximum dimension of a Java array as defined by
4424 * <cite>The Java Virtual Machine Specification</cite>.
4425 * The behaviour on a
4426 * {@code null} argument depends on the
4427 * <a href="../util/Formatter.html#syntax">conversion</a>.
4428 *
4429 * @throws java.util.IllegalFormatException
4430 * If a format string contains an illegal syntax, a format
4431 * specifier that is incompatible with the given arguments,
4432 * insufficient arguments given the format string, or other
4433 * illegal conditions. For specification of all possible
4434 * formatting errors, see the <a
4435 * href="../util/Formatter.html#detail">Details</a> section of the
4436 * formatter class specification
4437 *
4438 * @return A formatted string
4439 *
4440 * @see java.util.Formatter
4441 * @since 1.5
4442 */
4443 public static String format(Locale l, String format, Object... args) {
4444 return new Formatter(l).format(format, args).toString();
4445 }
4446
4447 /**
4448 * Formats using this string as the format string, and the supplied
4449 * arguments.
4450 *
4451 * @implSpec This method is equivalent to {@code String.format(this, args)}.
4452 *
4453 * @param args
4454 * Arguments referenced by the format specifiers in this string.
4455 *
4456 * @return A formatted string
4457 *
4458 * @see java.lang.String#format(String,Object...)
4459 * @see java.util.Formatter
4460 *
4461 * @since 15
4462 *
4463 */
4464 public String formatted(Object... args) {
4465 return new Formatter().format(this, args).toString();
4466 }
4467
4468 /**
4469 * Returns the string representation of the {@code Object} argument.
4470 *
4471 * @param obj an {@code Object}.
4472 * @return if the argument is {@code null}, then a string equal to
4473 * {@code "null"}; otherwise, the value of
4474 * {@code obj.toString()} is returned.
4475 * @see java.lang.Object#toString()
4476 */
4477 public static String valueOf(Object obj) {
4478 return (obj == null) ? "null" : obj.toString();
4479 }
4480
4481 /**
4482 * Returns the string representation of the {@code char} array
4483 * argument. The contents of the character array are copied; subsequent
4484 * modification of the character array does not affect the returned
4485 * string.
4486 *
4487 * @param data the character array.
4488 * @return a {@code String} that contains the characters of the
4489 * character array.
4490 */
4491 public static String valueOf(char[] data) {
4492 return new String(data);
4493 }
4494
4495 /**
4496 * Returns the string representation of a specific subarray of the
4497 * {@code char} array argument.
4498 * <p>
4499 * The {@code offset} argument is the index of the first
4500 * character of the subarray. The {@code count} argument
4501 * specifies the length of the subarray. The contents of the subarray
4502 * are copied; subsequent modification of the character array does not
4503 * affect the returned string.
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 valueOf(char[] data, int offset, int count) {
4516 return new String(data, offset, count);
4517 }
4518
4519 /**
4520 * Equivalent to {@link #valueOf(char[], int, int)}.
4521 *
4522 * @param data the character array.
4523 * @param offset initial offset of the subarray.
4524 * @param count length of the subarray.
4525 * @return a {@code String} that contains the characters of the
4526 * specified subarray of the character array.
4527 * @throws IndexOutOfBoundsException if {@code offset} is
4528 * negative, or {@code count} is negative, or
4529 * {@code offset+count} is larger than
4530 * {@code data.length}.
4531 */
4532 public static String copyValueOf(char[] data, int offset, int count) {
4533 return new String(data, offset, count);
4534 }
4535
4536 /**
4537 * Equivalent to {@link #valueOf(char[])}.
4538 *
4539 * @param data the character array.
4540 * @return a {@code String} that contains the characters of the
4541 * character array.
4542 */
4543 public static String copyValueOf(char[] data) {
4544 return new String(data);
4545 }
4546
4547 /**
4548 * Returns the string representation of the {@code boolean} argument.
4549 *
4550 * @param b a {@code boolean}.
4551 * @return if the argument is {@code true}, a string equal to
4552 * {@code "true"} is returned; otherwise, a string equal to
4553 * {@code "false"} is returned.
4554 */
4555 public static String valueOf(boolean b) {
4556 return b ? "true" : "false";
4557 }
4558
4559 /**
4560 * Returns the string representation of the {@code char}
4561 * argument.
4562 *
4563 * @param c a {@code char}.
4564 * @return a string of length {@code 1} containing
4565 * as its single character the argument {@code c}.
4566 */
4567 public static String valueOf(char c) {
4568 if (COMPACT_STRINGS && StringLatin1.canEncode(c)) {
4569 return new String(StringLatin1.toBytes(c), LATIN1);
4570 }
4571 return new String(StringUTF16.toBytes(c), UTF16);
4572 }
4573
4574 /**
4575 * Returns the string representation of the {@code int} argument.
4576 * <p>
4577 * The representation is exactly the one returned by the
4578 * {@code Integer.toString} method of one argument.
4579 *
4580 * @param i an {@code int}.
4581 * @return a string representation of the {@code int} argument.
4582 * @see java.lang.Integer#toString(int, int)
4583 */
4584 public static String valueOf(int i) {
4585 return Integer.toString(i);
4586 }
4587
4588 /**
4589 * Returns the string representation of the {@code long} argument.
4590 * <p>
4591 * The representation is exactly the one returned by the
4592 * {@code Long.toString} method of one argument.
4593 *
4594 * @param l a {@code long}.
4595 * @return a string representation of the {@code long} argument.
4596 * @see java.lang.Long#toString(long)
4597 */
4598 public static String valueOf(long l) {
4599 return Long.toString(l);
4600 }
4601
4602 /**
4603 * Returns the string representation of the {@code float} argument.
4604 * <p>
4605 * The representation is exactly the one returned by the
4606 * {@code Float.toString} method of one argument.
4607 *
4608 * @param f a {@code float}.
4609 * @return a string representation of the {@code float} argument.
4610 * @see java.lang.Float#toString(float)
4611 */
4612 public static String valueOf(float f) {
4613 return Float.toString(f);
4614 }
4615
4616 /**
4617 * Returns the string representation of the {@code double} argument.
4618 * <p>
4619 * The representation is exactly the one returned by the
4620 * {@code Double.toString} method of one argument.
4621 *
4622 * @param d a {@code double}.
4623 * @return a string representation of the {@code double} argument.
4624 * @see java.lang.Double#toString(double)
4625 */
4626 public static String valueOf(double d) {
4627 return Double.toString(d);
4628 }
4629
4630 /**
4631 * Returns a canonical representation for the string object.
4632 * <p>
4633 * A pool of strings, initially empty, is maintained privately by the
4634 * class {@code String}.
4635 * <p>
4636 * When the intern method is invoked, if the pool already contains a
4637 * string equal to this {@code String} object as determined by
4638 * the {@link #equals(Object)} method, then the string from the pool is
4639 * returned. Otherwise, this {@code String} object is added to the
4640 * pool and a reference to this {@code String} object is returned.
4641 * <p>
4642 * It follows that for any two strings {@code s} and {@code t},
4643 * {@code s.intern() == t.intern()} is {@code true}
4644 * if and only if {@code s.equals(t)} is {@code true}.
4645 * <p>
4646 * All literal strings and string-valued constant expressions are
4647 * interned. String literals are defined in section {@jls 3.10.5} of the
4648 * <cite>The Java Language Specification</cite>.
4649 *
4650 * @return a string that has the same contents as this string, but is
4651 * guaranteed to be from a pool of unique strings.
4652 */
4653 public native String intern();
4654
4655 /**
4656 * Returns a string whose value is the concatenation of this
4657 * string repeated {@code count} times.
4658 * <p>
4659 * If this string is empty or count is zero then the empty
4660 * string is returned.
4661 *
4662 * @param count number of times to repeat
4663 *
4664 * @return A string composed of this string repeated
4665 * {@code count} times or the empty string if this
4666 * string is empty or count is zero
4667 *
4668 * @throws IllegalArgumentException if the {@code count} is
4669 * negative.
4670 *
4671 * @since 11
4672 */
4673 public String repeat(int count) {
4674 if (count < 0) {
4675 throw new IllegalArgumentException("count is negative: " + count);
4676 }
4677 if (count == 1) {
4678 return this;
4679 }
4680 final int len = value.length;
4681 if (len == 0 || count == 0) {
4682 return "";
4683 }
4684 if (Integer.MAX_VALUE / count < len) {
4685 throw new OutOfMemoryError("Required length exceeds implementation limit");
4686 }
4687 if (len == 1) {
4688 final byte[] single = new byte[count];
4689 Arrays.fill(single, value[0]);
4690 return new String(single, coder);
4691 }
4692 final int limit = len * count;
4693 final byte[] multiple = new byte[limit];
4694 System.arraycopy(value, 0, multiple, 0, len);
4695 repeatCopyRest(multiple, 0, limit, len);
4696 return new String(multiple, coder);
4697 }
4698
4699 /**
4700 * Used to perform copying after the initial insertion. Copying is optimized
4701 * by using power of two duplication. First pass duplicates original copy,
4702 * second pass then duplicates the original and the copy yielding four copies,
4703 * third pass duplicates four copies yielding eight copies, and so on.
4704 * Finally, the remainder is filled in with prior copies.
4705 *
4706 * @implNote The technique used here is significantly faster than hand-rolled
4707 * loops or special casing small numbers due to the intensive optimization
4708 * done by intrinsic {@code System.arraycopy}.
4709 *
4710 * @param buffer destination buffer
4711 * @param offset offset in the destination buffer
4712 * @param limit total replicated including what is already in the buffer
4713 * @param copied number of bytes that have already in the buffer
4714 */
4715 static void repeatCopyRest(byte[] buffer, int offset, int limit, int copied) {
4716 // Initial copy is in the buffer.
4717 for (; copied < limit - copied; copied <<= 1) {
4718 // Power of two duplicate.
4719 System.arraycopy(buffer, offset, buffer, offset + copied, copied);
4720 }
4721 // Duplicate remainder.
4722 System.arraycopy(buffer, offset, buffer, offset + copied, limit - copied);
4723 }
4724
4725 ////////////////////////////////////////////////////////////////
4726
4727 /**
4728 * Copy character bytes from this string into dst starting at dstBegin.
4729 * This method doesn't perform any range checking.
4730 *
4731 * Invoker guarantees: dst is in UTF16 (inflate itself for asb), if two
4732 * coders are different, and dst is big enough (range check)
4733 *
4734 * @param dstBegin the char index, not offset of byte[]
4735 * @param coder the coder of dst[]
4736 */
4737 void getBytes(byte[] dst, int dstBegin, byte coder) {
4738 if (coder() == coder) {
4739 System.arraycopy(value, 0, dst, dstBegin << coder, value.length);
4740 } else { // this.coder == LATIN && coder == UTF16
4741 StringLatin1.inflate(value, 0, dst, dstBegin, value.length);
4742 }
4743 }
4744
4745 /**
4746 * Copy character bytes from this string into dst starting at dstBegin.
4747 * This method doesn't perform any range checking.
4748 *
4749 * Invoker guarantees: dst is in UTF16 (inflate itself for asb), if two
4750 * coders are different, and dst is big enough (range check)
4751 *
4752 * @param srcPos the char index, not offset of byte[]
4753 * @param dstBegin the char index to start from
4754 * @param coder the coder of dst[]
4755 * @param length the amount of copied chars
4756 */
4757 void getBytes(byte[] dst, int srcPos, int dstBegin, byte coder, int length) {
4758 if (coder() == coder) {
4759 System.arraycopy(value, srcPos << coder, dst, dstBegin << coder, length << coder);
4760 } else { // this.coder == LATIN && coder == UTF16
4761 StringLatin1.inflate(value, srcPos, dst, dstBegin, length);
4762 }
4763 }
4764
4765 /*
4766 * Package private constructor. Trailing Void argument is there for
4767 * disambiguating it against other (public) constructors.
4768 *
4769 * Stores the char[] value into a byte[] that each byte represents
4770 * the8 low-order bits of the corresponding character, if the char[]
4771 * contains only latin1 character. Or a byte[] that stores all
4772 * characters in their byte sequences defined by the {@code StringUTF16}.
4773 */
4774 String(char[] value, int off, int len, Void sig) {
4775 if (len == 0) {
4776 this.value = "".value;
4777 this.coder = "".coder;
4778 return;
4779 }
4780 if (COMPACT_STRINGS) {
4781 byte[] val = StringUTF16.compress(value, off, len);
4782 if (val != null) {
4783 this.value = val;
4784 this.coder = LATIN1;
4785 return;
4786 }
4787 }
4788 this.coder = UTF16;
4789 this.value = StringUTF16.toBytes(value, off, len);
4790 }
4791
4792 /*
4793 * Package private constructor. Trailing Void argument is there for
4794 * disambiguating it against other (public) constructors.
4795 */
4796 String(AbstractStringBuilder asb, Void sig) {
4797 byte[] val = asb.getValue();
4798 int length = asb.length();
4799 if (asb.isLatin1()) {
4800 this.coder = LATIN1;
4801 this.value = Arrays.copyOfRange(val, 0, length);
4802 } else {
4803 // only try to compress val if some characters were deleted.
4804 if (COMPACT_STRINGS && asb.maybeLatin1) {
4805 byte[] buf = StringUTF16.compress(val, 0, length);
4806 if (buf != null) {
4807 this.coder = LATIN1;
4808 this.value = buf;
4809 return;
4810 }
4811 }
4812 this.coder = UTF16;
4813 this.value = Arrays.copyOfRange(val, 0, length << 1);
4814 }
4815 }
4816
4817 /*
4818 * Package private constructor which shares value array for speed.
4819 */
4820 String(byte[] value, byte coder) {
4821 this.value = value;
4822 this.coder = coder;
4823 }
4824
4825 byte coder() {
4826 return COMPACT_STRINGS ? coder : UTF16;
4827 }
4828
4829 byte[] value() {
4830 return value;
4831 }
4832
4833 boolean isLatin1() {
4834 return COMPACT_STRINGS && coder == LATIN1;
4835 }
4836
4837 @Native static final byte LATIN1 = 0;
4838 @Native static final byte UTF16 = 1;
4839
4840 /*
4841 * StringIndexOutOfBoundsException if {@code index} is
4842 * negative or greater than or equal to {@code length}.
4843 */
4844 static void checkIndex(int index, int length) {
4845 Preconditions.checkIndex(index, length, Preconditions.SIOOBE_FORMATTER);
4846 }
4847
4848 /*
4849 * StringIndexOutOfBoundsException if {@code offset}
4850 * is negative or greater than {@code length}.
4851 */
4852 static void checkOffset(int offset, int length) {
4853 Preconditions.checkFromToIndex(offset, length, length, Preconditions.SIOOBE_FORMATTER);
4854 }
4855
4856 /*
4857 * Check {@code offset}, {@code count} against {@code 0} and {@code length}
4858 * bounds.
4859 *
4860 * @return {@code offset} if the sub-range within bounds of the range
4861 * @throws StringIndexOutOfBoundsException
4862 * If {@code offset} is negative, {@code count} is negative,
4863 * or {@code offset} is greater than {@code length - count}
4864 */
4865 static int checkBoundsOffCount(int offset, int count, int length) {
4866 return Preconditions.checkFromIndexSize(offset, count, length, Preconditions.SIOOBE_FORMATTER);
4867 }
4868
4869 /*
4870 * Check {@code begin}, {@code end} against {@code 0} and {@code length}
4871 * bounds.
4872 *
4873 * @throws StringIndexOutOfBoundsException
4874 * If {@code begin} is negative, {@code begin} is greater than
4875 * {@code end}, or {@code end} is greater than {@code length}.
4876 */
4877 static void checkBoundsBeginEnd(int begin, int end, int length) {
4878 Preconditions.checkFromToIndex(begin, end, length, Preconditions.SIOOBE_FORMATTER);
4879 }
4880
4881 /**
4882 * Returns the string representation of the {@code codePoint}
4883 * argument.
4884 *
4885 * @param codePoint a {@code codePoint}.
4886 * @return a string of length {@code 1} or {@code 2} containing
4887 * as its single character the argument {@code codePoint}.
4888 * @throws IllegalArgumentException if the specified
4889 * {@code codePoint} is not a {@linkplain Character#isValidCodePoint
4890 * valid Unicode code point}.
4891 */
4892 static String valueOfCodePoint(int codePoint) {
4893 if (COMPACT_STRINGS && StringLatin1.canEncode(codePoint)) {
4894 return new String(StringLatin1.toBytes((char)codePoint), LATIN1);
4895 } else if (Character.isBmpCodePoint(codePoint)) {
4896 return new String(StringUTF16.toBytes((char)codePoint), UTF16);
4897 } else if (Character.isSupplementaryCodePoint(codePoint)) {
4898 return new String(StringUTF16.toBytesSupplementary(codePoint), UTF16);
4899 }
4900
4901 throw new IllegalArgumentException(
4902 format("Not a valid Unicode code point: 0x%X", codePoint));
4903 }
4904
4905 /**
4906 * Returns an {@link Optional} containing the nominal descriptor for this
4907 * instance, which is the instance itself.
4908 *
4909 * @return an {@link Optional} describing the {@linkplain String} instance
4910 * @since 12
4911 */
4912 @Override
4913 public Optional<String> describeConstable() {
4914 return Optional.of(this);
4915 }
4916
4917 /**
4918 * Resolves this instance as a {@link ConstantDesc}, the result of which is
4919 * the instance itself.
4920 *
4921 * @param lookup ignored
4922 * @return the {@linkplain String} instance
4923 * @since 12
4924 */
4925 @Override
4926 public String resolveConstantDesc(MethodHandles.Lookup lookup) {
4927 return this;
4928 }
4929
4930 }