1 /*
2 * Copyright (c) 2003, 2015, 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 jdk.internal.math.FloatingDecimal;
29 import java.util.Arrays;
30 import java.util.Spliterator;
31 import java.util.stream.IntStream;
32 import java.util.stream.StreamSupport;
33
34 import static java.lang.String.COMPACT_STRINGS;
35 import static java.lang.String.UTF16;
36 import static java.lang.String.LATIN1;
37 import static java.lang.String.checkIndex;
38 import static java.lang.String.checkOffset;
39
40 /**
41 * A mutable sequence of characters.
42 * <p>
43 * Implements a modifiable string. At any point in time it contains some
44 * particular sequence of characters, but the length and content of the
45 * sequence can be changed through certain method calls.
46 *
47 * <p>Unless otherwise noted, passing a {@code null} argument to a constructor
48 * or method in this class will cause a {@link NullPointerException} to be
49 * thrown.
50 *
51 * @author Michael McCloskey
52 * @author Martin Buchholz
53 * @author Ulf Zibis
54 * @since 1.5
55 */
56 abstract class AbstractStringBuilder implements Appendable, CharSequence {
57 /**
58 * The value is used for character storage.
59 */
60 byte[] value;
61
62 /**
63 * The id of the encoding used to encode the bytes in {@code value}.
64 */
65 byte coder;
66
67 /**
68 * The count is the number of characters used.
69 */
70 int count;
71
72 /**
73 * This no-arg constructor is necessary for serialization of subclasses.
74 */
75 AbstractStringBuilder() {
76 }
77
78 /**
79 * Creates an AbstractStringBuilder of the specified capacity.
80 */
81 AbstractStringBuilder(int capacity) {
82 if (COMPACT_STRINGS) {
83 value = new byte[capacity];
84 coder = LATIN1;
85 } else {
86 value = StringUTF16.newBytesFor(capacity);
87 coder = UTF16;
88 }
89 }
90
91 /**
92 * Returns the length (character count).
93 *
94 * @return the length of the sequence of characters currently
95 * represented by this object
96 */
97 @Override
98 public int length() {
99 return count;
100 }
101
102 /**
103 * Returns the current capacity. The capacity is the amount of storage
104 * available for newly inserted characters, beyond which an allocation
105 * will occur.
106 *
107 * @return the current capacity
108 */
109 public int capacity() {
110 return value.length >> coder;
111 }
112
113 /**
114 * Ensures that the capacity is at least equal to the specified minimum.
115 * If the current capacity is less than the argument, then a new internal
116 * array is allocated with greater capacity. The new capacity is the
117 * larger of:
118 * <ul>
119 * <li>The {@code minimumCapacity} argument.
120 * <li>Twice the old capacity, plus {@code 2}.
121 * </ul>
122 * If the {@code minimumCapacity} argument is nonpositive, this
123 * method takes no action and simply returns.
124 * Note that subsequent operations on this object can reduce the
125 * actual capacity below that requested here.
126 *
127 * @param minimumCapacity the minimum desired capacity.
128 */
129 public void ensureCapacity(int minimumCapacity) {
130 if (minimumCapacity > 0) {
131 ensureCapacityInternal(minimumCapacity);
132 }
133 }
134
135 /**
136 * This method has the same contract as ensureCapacity, but is
137 * never synchronized.
138 */
139 private void ensureCapacityInternal(int minimumCapacity) {
140 // overflow-conscious code
141 int capacity = value.length >> coder;
142 if (minimumCapacity - capacity > 0) {
143 expandCapacity(minimumCapacity);
144 }
145 }
146
147 /**
148 * The maximum size of array to allocate (unless necessary).
149 * Some VMs reserve some header words in an array.
150 * Attempts to allocate larger arrays may result in
151 * OutOfMemoryError: Requested array size exceeds VM limit
152 */
153 private static final int MAX_ARRAY_SIZE = Integer.MAX_VALUE - 8;
154
155 /**
156 * This implements the expansion semantics of ensureCapacity with no
157 * size check or synchronization.
158 */
159 private void expandCapacity(int minimumCapacity) {
160 int newCapacity = (value.length >> coder) * 2 + 2;
161 if (newCapacity - minimumCapacity < 0) {
162 newCapacity = minimumCapacity;
163 }
164
165 final int SAFE_BOUND = (MAX_ARRAY_SIZE >> coder);
166 if (((SAFE_BOUND - newCapacity) | newCapacity) < 0) {
167 final int UNSAFE_BOUND = (Integer.MAX_VALUE >> coder);
168 if (UNSAFE_BOUND - minimumCapacity < 0) { // overflow
169 throw new OutOfMemoryError();
170 }
171 newCapacity = (minimumCapacity > SAFE_BOUND)
172 ? minimumCapacity : SAFE_BOUND;
173 }
174 this.value = Arrays.copyOf(value, newCapacity << coder);
175 }
176
177 /**
178 * If the coder is "isLatin1", this inflates the internal 8-bit storage
179 * to 16-bit <hi=0, low> pair storage.
180 */
181 private void inflate() {
182 if (!isLatin1()) {
183 return;
184 }
185 byte[] buf = StringUTF16.newBytesFor(value.length);
186 StringLatin1.inflate(value, 0, buf, 0, count);
187 this.value = buf;
188 this.coder = UTF16;
189 }
190
191 /**
192 * Attempts to reduce storage used for the character sequence.
193 * If the buffer is larger than necessary to hold its current sequence of
194 * characters, then it may be resized to become more space efficient.
195 * Calling this method may, but is not required to, affect the value
196 * returned by a subsequent call to the {@link #capacity()} method.
197 */
198 public void trimToSize() {
199 int length = count << coder;
200 if (length < value.length) {
201 value = Arrays.copyOf(value, length);
202 }
203 }
204
205 /**
206 * Sets the length of the character sequence.
207 * The sequence is changed to a new character sequence
208 * whose length is specified by the argument. For every nonnegative
209 * index <i>k</i> less than {@code newLength}, the character at
210 * index <i>k</i> in the new character sequence is the same as the
211 * character at index <i>k</i> in the old sequence if <i>k</i> is less
212 * than the length of the old character sequence; otherwise, it is the
213 * null character {@code '\u005Cu0000'}.
214 *
215 * In other words, if the {@code newLength} argument is less than
216 * the current length, the length is changed to the specified length.
217 * <p>
218 * If the {@code newLength} argument is greater than or equal
219 * to the current length, sufficient null characters
220 * ({@code '\u005Cu0000'}) are appended so that
221 * length becomes the {@code newLength} argument.
222 * <p>
223 * The {@code newLength} argument must be greater than or equal
224 * to {@code 0}.
225 *
226 * @param newLength the new length
227 * @throws IndexOutOfBoundsException if the
228 * {@code newLength} argument is negative.
229 */
230 public void setLength(int newLength) {
231 if (newLength < 0) {
232 throw new StringIndexOutOfBoundsException(newLength);
233 }
234 ensureCapacityInternal(newLength);
235 if (count < newLength) {
236 if (isLatin1()) {
237 StringLatin1.fillNull(value, count, newLength);
238 } else {
239 StringUTF16.fillNull(value, count, newLength);
240 }
241 }
242 count = newLength;
243 }
244
245 /**
246 * Returns the {@code char} value in this sequence at the specified index.
247 * The first {@code char} value is at index {@code 0}, the next at index
248 * {@code 1}, and so on, as in array indexing.
249 * <p>
250 * The index argument must be greater than or equal to
251 * {@code 0}, and less than the length of this sequence.
252 *
253 * <p>If the {@code char} value specified by the index is a
254 * <a href="Character.html#unicode">surrogate</a>, the surrogate
255 * value is returned.
256 *
257 * @param index the index of the desired {@code char} value.
258 * @return the {@code char} value at the specified index.
259 * @throws IndexOutOfBoundsException if {@code index} is
260 * negative or greater than or equal to {@code length()}.
261 */
262 @Override
263 public char charAt(int index) {
264 checkIndex(index, count);
265 if (isLatin1()) {
266 return (char)(value[index] & 0xff);
267 }
268 return StringUTF16.charAt(value, index);
269 }
270
271 /**
272 * Returns the character (Unicode code point) at the specified
273 * index. The index refers to {@code char} values
274 * (Unicode code units) and ranges from {@code 0} to
275 * {@link #length()}{@code - 1}.
276 *
277 * <p> If the {@code char} value specified at the given index
278 * is in the high-surrogate range, the following index is less
279 * than the length of this sequence, and the
280 * {@code char} value at the following index is in the
281 * low-surrogate range, then the supplementary code point
282 * corresponding to this surrogate pair is returned. Otherwise,
283 * the {@code char} value at the given index is returned.
284 *
285 * @param index the index to the {@code char} values
286 * @return the code point value of the character at the
287 * {@code index}
288 * @exception IndexOutOfBoundsException if the {@code index}
289 * argument is negative or not less than the length of this
290 * sequence.
291 */
292 public int codePointAt(int index) {
293 checkIndex(index, count);
294 if (isLatin1()) {
295 return value[index] & 0xff;
296 }
297 return StringUTF16.codePointAtSB(value, index, count);
298 }
299
300 /**
301 * Returns the character (Unicode code point) before the specified
302 * index. The index refers to {@code char} values
303 * (Unicode code units) and ranges from {@code 1} to {@link
304 * #length()}.
305 *
306 * <p> If the {@code char} value at {@code (index - 1)}
307 * is in the low-surrogate range, {@code (index - 2)} is not
308 * negative, and the {@code char} value at {@code (index -
309 * 2)} is in the high-surrogate range, then the
310 * supplementary code point value of the surrogate pair is
311 * returned. If the {@code char} value at {@code index -
312 * 1} is an unpaired low-surrogate or a high-surrogate, the
313 * surrogate value is returned.
314 *
315 * @param index the index following the code point that should be returned
316 * @return the Unicode code point value before the given index.
317 * @exception IndexOutOfBoundsException if the {@code index}
318 * argument is less than 1 or greater than the length
319 * of this sequence.
320 */
321 public int codePointBefore(int index) {
322 int i = index - 1;
323 if (i < 0 || i >= count) {
324 throw new StringIndexOutOfBoundsException(index);
325 }
326 if (isLatin1()) {
327 return value[i] & 0xff;
328 }
329 return StringUTF16.codePointBeforeSB(value, index);
330 }
331
332 /**
333 * Returns the number of Unicode code points in the specified text
334 * range of this sequence. The text range begins at the specified
335 * {@code beginIndex} and extends to the {@code char} at
336 * index {@code endIndex - 1}. Thus the length (in
337 * {@code char}s) of the text range is
338 * {@code endIndex-beginIndex}. Unpaired surrogates within
339 * this sequence count as one code point each.
340 *
341 * @param beginIndex the index to the first {@code char} of
342 * the text range.
343 * @param endIndex the index after the last {@code char} of
344 * the text range.
345 * @return the number of Unicode code points in the specified text
346 * range
347 * @exception IndexOutOfBoundsException if the
348 * {@code beginIndex} is negative, or {@code endIndex}
349 * is larger than the length of this sequence, or
350 * {@code beginIndex} is larger than {@code endIndex}.
351 */
352 public int codePointCount(int beginIndex, int endIndex) {
353 if (beginIndex < 0 || endIndex > count || beginIndex > endIndex) {
354 throw new IndexOutOfBoundsException();
355 }
356 if (isLatin1()) {
357 return endIndex - beginIndex;
358 }
359 return StringUTF16.codePointCountSB(value, beginIndex, endIndex);
360 }
361
362 /**
363 * Returns the index within this sequence that is offset from the
364 * given {@code index} by {@code codePointOffset} code
365 * points. Unpaired surrogates within the text range given by
366 * {@code index} and {@code codePointOffset} count as
367 * one code point each.
368 *
369 * @param index the index to be offset
370 * @param codePointOffset the offset in code points
371 * @return the index within this sequence
372 * @exception IndexOutOfBoundsException if {@code index}
373 * is negative or larger then the length of this sequence,
374 * or if {@code codePointOffset} is positive and the subsequence
375 * starting with {@code index} has fewer than
376 * {@code codePointOffset} code points,
377 * or if {@code codePointOffset} is negative and the subsequence
378 * before {@code index} has fewer than the absolute value of
379 * {@code codePointOffset} code points.
380 */
381 public int offsetByCodePoints(int index, int codePointOffset) {
382 if (index < 0 || index > count) {
383 throw new IndexOutOfBoundsException();
384 }
385 return Character.offsetByCodePoints(this,
386 index, codePointOffset);
387 }
388
389 /**
390 * Characters are copied from this sequence into the
391 * destination character array {@code dst}. The first character to
392 * be copied is at index {@code srcBegin}; the last character to
393 * be copied is at index {@code srcEnd-1}. The total number of
394 * characters to be copied is {@code srcEnd-srcBegin}. The
395 * characters are copied into the subarray of {@code dst} starting
396 * at index {@code dstBegin} and ending at index:
397 * <pre>{@code
398 * dstbegin + (srcEnd-srcBegin) - 1
399 * }</pre>
400 *
401 * @param srcBegin start copying at this offset.
402 * @param srcEnd stop copying at this offset.
403 * @param dst the array to copy the data into.
404 * @param dstBegin offset into {@code dst}.
405 * @throws IndexOutOfBoundsException if any of the following is true:
406 * <ul>
407 * <li>{@code srcBegin} is negative
408 * <li>{@code dstBegin} is negative
409 * <li>the {@code srcBegin} argument is greater than
410 * the {@code srcEnd} argument.
411 * <li>{@code srcEnd} is greater than
412 * {@code this.length()}.
413 * <li>{@code dstBegin+srcEnd-srcBegin} is greater than
414 * {@code dst.length}
415 * </ul>
416 */
417 public void getChars(int srcBegin, int srcEnd, char[] dst, int dstBegin)
418 {
419 checkRangeSIOOBE(srcBegin, srcEnd, count); // compatible to old version
420 int n = srcEnd - srcBegin;
421 checkRange(dstBegin, dstBegin + n, dst.length);
422 if (isLatin1()) {
423 StringLatin1.getChars(value, srcBegin, srcEnd, dst, dstBegin);
424 } else {
425 StringUTF16.getChars(value, srcBegin, srcEnd, dst, dstBegin);
426 }
427 }
428
429 /**
430 * The character at the specified index is set to {@code ch}. This
431 * sequence is altered to represent a new character sequence that is
432 * identical to the old character sequence, except that it contains the
433 * character {@code ch} at position {@code index}.
434 * <p>
435 * The index argument must be greater than or equal to
436 * {@code 0}, and less than the length of this sequence.
437 *
438 * @param index the index of the character to modify.
439 * @param ch the new character.
440 * @throws IndexOutOfBoundsException if {@code index} is
441 * negative or greater than or equal to {@code length()}.
442 */
443 public void setCharAt(int index, char ch) {
444 checkIndex(index, count);
445 if (isLatin1() && StringLatin1.canEncode(ch)) {
446 value[index] = (byte)ch;
447 } else {
448 if (isLatin1()) {
449 inflate();
450 }
451 StringUTF16.putCharSB(value, index, ch);
452 }
453 }
454
455 /**
456 * Appends the string representation of the {@code Object} argument.
457 * <p>
458 * The overall effect is exactly as if the argument were converted
459 * to a string by the method {@link String#valueOf(Object)},
460 * and the characters of that string were then
461 * {@link #append(String) appended} to this character sequence.
462 *
463 * @param obj an {@code Object}.
464 * @return a reference to this object.
465 */
466 public AbstractStringBuilder append(Object obj) {
467 return append(String.valueOf(obj));
468 }
469
470 /**
471 * Appends the specified string to this character sequence.
472 * <p>
473 * The characters of the {@code String} argument are appended, in
474 * order, increasing the length of this sequence by the length of the
475 * argument. If {@code str} is {@code null}, then the four
476 * characters {@code "null"} are appended.
477 * <p>
478 * Let <i>n</i> be the length of this character sequence just prior to
479 * execution of the {@code append} method. Then the character at
480 * index <i>k</i> in the new character sequence is equal to the character
481 * at index <i>k</i> in the old character sequence, if <i>k</i> is less
482 * than <i>n</i>; otherwise, it is equal to the character at index
483 * <i>k-n</i> in the argument {@code str}.
484 *
485 * @param str a string.
486 * @return a reference to this object.
487 */
488 public AbstractStringBuilder append(String str) {
489 if (str == null) {
490 return appendNull();
491 }
492 int len = str.length();
493 ensureCapacityInternal(count + len);
494 putStringAt(count, str);
495 count += len;
496 return this;
497 }
498
499 // Documentation in subclasses because of synchro difference
500 public AbstractStringBuilder append(StringBuffer sb) {
501 return this.append((AbstractStringBuilder)sb);
502 }
503
504 /**
505 * @since 1.8
506 */
507 AbstractStringBuilder append(AbstractStringBuilder asb) {
508 if (asb == null) {
509 return appendNull();
510 }
511 int len = asb.length();
512 ensureCapacityInternal(count + len);
513 if (getCoder() != asb.getCoder()) {
514 inflate();
515 }
516 asb.getBytes(value, count, coder);
517 count += len;
518 return this;
519 }
520
521 // Documentation in subclasses because of synchro difference
522 @Override
523 public AbstractStringBuilder append(CharSequence s) {
524 if (s == null) {
525 return appendNull();
526 }
527 if (s instanceof String) {
528 return this.append((String)s);
529 }
530 if (s instanceof AbstractStringBuilder) {
531 return this.append((AbstractStringBuilder)s);
532 }
533 return this.append(s, 0, s.length());
534 }
535
536 private AbstractStringBuilder appendNull() {
537 ensureCapacityInternal(count + 4);
538 int count = this.count;
539 byte[] val = this.value;
540 if (isLatin1()) {
541 val[count++] = 'n';
542 val[count++] = 'u';
543 val[count++] = 'l';
544 val[count++] = 'l';
545 } else {
546 checkOffset(count + 4, val.length >> 1);
547 StringUTF16.putChar(val, count++, 'n');
548 StringUTF16.putChar(val, count++, 'u');
549 StringUTF16.putChar(val, count++, 'l');
550 StringUTF16.putChar(val, count++, 'l');
551 }
552 this.count = count;
553 return this;
554 }
555
556 /**
557 * Appends a subsequence of the specified {@code CharSequence} to this
558 * sequence.
559 * <p>
560 * Characters of the argument {@code s}, starting at
561 * index {@code start}, are appended, in order, to the contents of
562 * this sequence up to the (exclusive) index {@code end}. The length
563 * of this sequence is increased by the value of {@code end - start}.
564 * <p>
565 * Let <i>n</i> be the length of this character sequence just prior to
566 * execution of the {@code append} method. Then the character at
567 * index <i>k</i> in this character sequence becomes equal to the
568 * character at index <i>k</i> in this sequence, if <i>k</i> is less than
569 * <i>n</i>; otherwise, it is equal to the character at index
570 * <i>k+start-n</i> in the argument {@code s}.
571 * <p>
572 * If {@code s} is {@code null}, then this method appends
573 * characters as if the s parameter was a sequence containing the four
574 * characters {@code "null"}.
575 *
576 * @param s the sequence to append.
577 * @param start the starting index of the subsequence to be appended.
578 * @param end the end index of the subsequence to be appended.
579 * @return a reference to this object.
580 * @throws IndexOutOfBoundsException if
581 * {@code start} is negative, or
582 * {@code start} is greater than {@code end} or
583 * {@code end} is greater than {@code s.length()}
584 */
585 @Override
586 public AbstractStringBuilder append(CharSequence s, int start, int end) {
587 if (s == null) {
588 s = "null";
589 }
590 checkRange(start, end, s.length());
591 int len = end - start;
592 ensureCapacityInternal(count + len);
593 appendChars(s, start, end);
594 return this;
595 }
596
597 /**
598 * Appends the string representation of the {@code char} array
599 * argument to this sequence.
600 * <p>
601 * The characters of the array argument are appended, in order, to
602 * the contents of this sequence. The length of this sequence
603 * increases by the length of the argument.
604 * <p>
605 * The overall effect is exactly as if the argument were converted
606 * to a string by the method {@link String#valueOf(char[])},
607 * and the characters of that string were then
608 * {@link #append(String) appended} to this character sequence.
609 *
610 * @param str the characters to be appended.
611 * @return a reference to this object.
612 */
613 public AbstractStringBuilder append(char[] str) {
614 int len = str.length;
615 ensureCapacityInternal(count + len);
616 appendChars(str, 0, len);
617 return this;
618 }
619
620 /**
621 * Appends the string representation of a subarray of the
622 * {@code char} array argument to this sequence.
623 * <p>
624 * Characters of the {@code char} array {@code str}, starting at
625 * index {@code offset}, are appended, in order, to the contents
626 * of this sequence. The length of this sequence increases
627 * by the value of {@code len}.
628 * <p>
629 * The overall effect is exactly as if the arguments were converted
630 * to a string by the method {@link String#valueOf(char[],int,int)},
631 * and the characters of that string were then
632 * {@link #append(String) appended} to this character sequence.
633 *
634 * @param str the characters to be appended.
635 * @param offset the index of the first {@code char} to append.
636 * @param len the number of {@code char}s to append.
637 * @return a reference to this object.
638 * @throws IndexOutOfBoundsException
639 * if {@code offset < 0} or {@code len < 0}
640 * or {@code offset+len > str.length}
641 */
642 public AbstractStringBuilder append(char str[], int offset, int len) {
643 int end = offset + len;
644 checkRange(offset, end, str.length);
645 ensureCapacityInternal(count + len);
646 appendChars(str, offset, end);
647 return this;
648 }
649
650 /**
651 * Appends the string representation of the {@code boolean}
652 * argument to the sequence.
653 * <p>
654 * The overall effect is exactly as if the argument were converted
655 * to a string by the method {@link String#valueOf(boolean)},
656 * and the characters of that string were then
657 * {@link #append(String) appended} to this character sequence.
658 *
659 * @param b a {@code boolean}.
660 * @return a reference to this object.
661 */
662 public AbstractStringBuilder append(boolean b) {
663 ensureCapacityInternal(count + (b ? 4 : 5));
664 int count = this.count;
665 byte[] val = this.value;
666 if (isLatin1()) {
667 if (b) {
668 val[count++] = 't';
669 val[count++] = 'r';
670 val[count++] = 'u';
671 val[count++] = 'e';
672 } else {
673 val[count++] = 'f';
674 val[count++] = 'a';
675 val[count++] = 'l';
676 val[count++] = 's';
677 val[count++] = 'e';
678 }
679 } else {
680 if (b) {
681 checkOffset(count + 4, val.length >> 1);
682 StringUTF16.putChar(val, count++, 't');
683 StringUTF16.putChar(val, count++, 'r');
684 StringUTF16.putChar(val, count++, 'u');
685 StringUTF16.putChar(val, count++, 'e');
686 } else {
687 checkOffset(count + 5, val.length >> 1);
688 StringUTF16.putChar(val, count++, 'f');
689 StringUTF16.putChar(val, count++, 'a');
690 StringUTF16.putChar(val, count++, 'l');
691 StringUTF16.putChar(val, count++, 's');
692 StringUTF16.putChar(val, count++, 'e');
693 }
694 }
695 this.count = count;
696 return this;
697 }
698
699 /**
700 * Appends the string representation of the {@code char}
701 * argument to this sequence.
702 * <p>
703 * The argument is appended to the contents of this sequence.
704 * The length of this sequence increases by {@code 1}.
705 * <p>
706 * The overall effect is exactly as if the argument were converted
707 * to a string by the method {@link String#valueOf(char)},
708 * and the character in that string were then
709 * {@link #append(String) appended} to this character sequence.
710 *
711 * @param c a {@code char}.
712 * @return a reference to this object.
713 */
714 @Override
715 public AbstractStringBuilder append(char c) {
716 ensureCapacityInternal(count + 1);
717 if (isLatin1() && StringLatin1.canEncode(c)) {
718 value[count++] = (byte)c;
719 } else {
720 if (isLatin1()) {
721 inflate();
722 }
723 StringUTF16.putCharSB(value, count++, c);
724 }
725 return this;
726 }
727
728 /**
729 * Appends the string representation of the {@code int}
730 * argument to this sequence.
731 * <p>
732 * The overall effect is exactly as if the argument were converted
733 * to a string by the method {@link String#valueOf(int)},
734 * and the characters of that string were then
735 * {@link #append(String) appended} to this character sequence.
736 *
737 * @param i an {@code int}.
738 * @return a reference to this object.
739 */
740 public AbstractStringBuilder append(int i) {
741 int spaceNeeded = count + Integer.stringSize(i);
742 ensureCapacityInternal(spaceNeeded);
743 if (isLatin1()) {
744 Integer.getChars(i, spaceNeeded, value);
745 } else {
746 byte[] val = this.value;
747 checkOffset(spaceNeeded, val.length >> 1);
748 Integer.getCharsUTF16(i, spaceNeeded, val);
749 }
750 count = spaceNeeded;
751 return this;
752 }
753
754 /**
755 * Appends the string representation of the {@code long}
756 * argument to this sequence.
757 * <p>
758 * The overall effect is exactly as if the argument were converted
759 * to a string by the method {@link String#valueOf(long)},
760 * and the characters of that string were then
761 * {@link #append(String) appended} to this character sequence.
762 *
763 * @param l a {@code long}.
764 * @return a reference to this object.
765 */
766 public AbstractStringBuilder append(long l) {
767 int spaceNeeded = count + Long.stringSize(l);
768 ensureCapacityInternal(spaceNeeded);
769 if (isLatin1()) {
770 Long.getChars(l, spaceNeeded, value);
771 } else {
772 byte[] val = this.value;
773 checkOffset(spaceNeeded, val.length >> 1);
774 Long.getCharsUTF16(l, spaceNeeded, val);
775 }
776 count = spaceNeeded;
777 return this;
778 }
779
780 /**
781 * Appends the string representation of the {@code float}
782 * argument to this sequence.
783 * <p>
784 * The overall effect is exactly as if the argument were converted
785 * to a string by the method {@link String#valueOf(float)},
786 * and the characters of that string were then
787 * {@link #append(String) appended} to this character sequence.
788 *
789 * @param f a {@code float}.
790 * @return a reference to this object.
791 */
792 public AbstractStringBuilder append(float f) {
793 FloatingDecimal.appendTo(f,this);
794 return this;
795 }
796
797 /**
798 * Appends the string representation of the {@code double}
799 * argument to this sequence.
800 * <p>
801 * The overall effect is exactly as if the argument were converted
802 * to a string by the method {@link String#valueOf(double)},
803 * and the characters of that string were then
804 * {@link #append(String) appended} to this character sequence.
805 *
806 * @param d a {@code double}.
807 * @return a reference to this object.
808 */
809 public AbstractStringBuilder append(double d) {
810 FloatingDecimal.appendTo(d,this);
811 return this;
812 }
813
814 /**
815 * Removes the characters in a substring of this sequence.
816 * The substring begins at the specified {@code start} and extends to
817 * the character at index {@code end - 1} or to the end of the
818 * sequence if no such character exists. If
819 * {@code start} is equal to {@code end}, no changes are made.
820 *
821 * @param start The beginning index, inclusive.
822 * @param end The ending index, exclusive.
823 * @return This object.
824 * @throws StringIndexOutOfBoundsException if {@code start}
825 * is negative, greater than {@code length()}, or
826 * greater than {@code end}.
827 */
828 public AbstractStringBuilder delete(int start, int end) {
829 if (end > count) {
830 end = count;
831 }
832 checkRangeSIOOBE(start, end, count);
833 int len = end - start;
834 if (len > 0) {
835 shift(end, -len);
836 count -= len;
837 }
838 return this;
839 }
840
841 /**
842 * Appends the string representation of the {@code codePoint}
843 * argument to this sequence.
844 *
845 * <p> The argument is appended to the contents of this sequence.
846 * The length of this sequence increases by
847 * {@link Character#charCount(int) Character.charCount(codePoint)}.
848 *
849 * <p> The overall effect is exactly as if the argument were
850 * converted to a {@code char} array by the method
851 * {@link Character#toChars(int)} and the character in that array
852 * were then {@link #append(char[]) appended} to this character
853 * sequence.
854 *
855 * @param codePoint a Unicode code point
856 * @return a reference to this object.
857 * @exception IllegalArgumentException if the specified
858 * {@code codePoint} isn't a valid Unicode code point
859 */
860 public AbstractStringBuilder appendCodePoint(int codePoint) {
861 if (Character.isBmpCodePoint(codePoint)) {
862 return append((char)codePoint);
863 }
864 return append(Character.toChars(codePoint));
865 }
866
867 /**
868 * Removes the {@code char} at the specified position in this
869 * sequence. This sequence is shortened by one {@code char}.
870 *
871 * <p>Note: If the character at the given index is a supplementary
872 * character, this method does not remove the entire character. If
873 * correct handling of supplementary characters is required,
874 * determine the number of {@code char}s to remove by calling
875 * {@code Character.charCount(thisSequence.codePointAt(index))},
876 * where {@code thisSequence} is this sequence.
877 *
878 * @param index Index of {@code char} to remove
879 * @return This object.
880 * @throws StringIndexOutOfBoundsException if the {@code index}
881 * is negative or greater than or equal to
882 * {@code length()}.
883 */
884 public AbstractStringBuilder deleteCharAt(int index) {
885 checkIndex(index, count);
886 shift(index + 1, -1);
887 count--;
888 return this;
889 }
890
891 /**
892 * Replaces the characters in a substring of this sequence
893 * with characters in the specified {@code String}. The substring
894 * begins at the specified {@code start} and extends to the character
895 * at index {@code end - 1} or to the end of the
896 * sequence if no such character exists. First the
897 * characters in the substring are removed and then the specified
898 * {@code String} is inserted at {@code start}. (This
899 * sequence will be lengthened to accommodate the
900 * specified String if necessary.)
901 *
902 * @param start The beginning index, inclusive.
903 * @param end The ending index, exclusive.
904 * @param str String that will replace previous contents.
905 * @return This object.
906 * @throws StringIndexOutOfBoundsException if {@code start}
907 * is negative, greater than {@code length()}, or
908 * greater than {@code end}.
909 */
910 public AbstractStringBuilder replace(int start, int end, String str) {
911 if (end > count) {
912 end = count;
913 }
914 checkRangeSIOOBE(start, end, count);
915 int len = str.length();
916 int newCount = count + len - (end - start);
917 ensureCapacityInternal(newCount);
918 shift(end, newCount - count);
919 count = newCount;
920 putStringAt(start, str);
921 return this;
922 }
923
924 /**
925 * Returns a new {@code String} that contains a subsequence of
926 * characters currently contained in this character sequence. The
927 * substring begins at the specified index and extends to the end of
928 * this sequence.
929 *
930 * @param start The beginning index, inclusive.
931 * @return The new string.
932 * @throws StringIndexOutOfBoundsException if {@code start} is
933 * less than zero, or greater than the length of this object.
934 */
935 public String substring(int start) {
936 return substring(start, count);
937 }
938
939 /**
940 * Returns a new character sequence that is a subsequence of this sequence.
941 *
942 * <p> An invocation of this method of the form
943 *
944 * <pre>{@code
945 * sb.subSequence(begin, end)}</pre>
946 *
947 * behaves in exactly the same way as the invocation
948 *
949 * <pre>{@code
950 * sb.substring(begin, end)}</pre>
951 *
952 * This method is provided so that this class can
953 * implement the {@link CharSequence} interface.
954 *
955 * @param start the start index, inclusive.
956 * @param end the end index, exclusive.
957 * @return the specified subsequence.
958 *
959 * @throws IndexOutOfBoundsException
960 * if {@code start} or {@code end} are negative,
961 * if {@code end} is greater than {@code length()},
962 * or if {@code start} is greater than {@code end}
963 * @spec JSR-51
964 */
965 @Override
966 public CharSequence subSequence(int start, int end) {
967 return substring(start, end);
968 }
969
970 /**
971 * Returns a new {@code String} that contains a subsequence of
972 * characters currently contained in this sequence. The
973 * substring begins at the specified {@code start} and
974 * extends to the character at index {@code end - 1}.
975 *
976 * @param start The beginning index, inclusive.
977 * @param end The ending index, exclusive.
978 * @return The new string.
979 * @throws StringIndexOutOfBoundsException if {@code start}
980 * or {@code end} are negative or greater than
981 * {@code length()}, or {@code start} is
982 * greater than {@code end}.
983 */
984 public String substring(int start, int end) {
985 checkRangeSIOOBE(start, end, count);
986 if (isLatin1()) {
987 return StringLatin1.newString(value, start, end - start);
988 }
989 return StringUTF16.newString(value, start, end - start);
990 }
991
992 private void shift(int offset, int n) {
993 System.arraycopy(value, offset << coder,
994 value, (offset + n) << coder, (count - offset) << coder);
995 }
996
997 /**
998 * Inserts the string representation of a subarray of the {@code str}
999 * array argument into this sequence. The subarray begins at the
1000 * specified {@code offset} and extends {@code len} {@code char}s.
1001 * The characters of the subarray are inserted into this sequence at
1002 * the position indicated by {@code index}. The length of this
1003 * sequence increases by {@code len} {@code char}s.
1004 *
1005 * @param index position at which to insert subarray.
1006 * @param str A {@code char} array.
1007 * @param offset the index of the first {@code char} in subarray to
1008 * be inserted.
1009 * @param len the number of {@code char}s in the subarray to
1010 * be inserted.
1011 * @return This object
1012 * @throws StringIndexOutOfBoundsException if {@code index}
1013 * is negative or greater than {@code length()}, or
1014 * {@code offset} or {@code len} are negative, or
1015 * {@code (offset+len)} is greater than
1016 * {@code str.length}.
1017 */
1018 public AbstractStringBuilder insert(int index, char[] str, int offset,
1019 int len)
1020 {
1021 checkOffset(index, count);
1022 checkRangeSIOOBE(offset, offset + len, str.length);
1023 ensureCapacityInternal(count + len);
1024 shift(index, len);
1025 count += len;
1026 putCharsAt(index, str, offset, offset + len);
1027 return this;
1028 }
1029
1030 /**
1031 * Inserts the string representation of the {@code Object}
1032 * argument into this character sequence.
1033 * <p>
1034 * The overall effect is exactly as if the second argument were
1035 * converted to a string by the method {@link String#valueOf(Object)},
1036 * and the characters of that string were then
1037 * {@link #insert(int,String) inserted} into this character
1038 * sequence at the indicated offset.
1039 * <p>
1040 * The {@code offset} argument must be greater than or equal to
1041 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1042 * of this sequence.
1043 *
1044 * @param offset the offset.
1045 * @param obj an {@code Object}.
1046 * @return a reference to this object.
1047 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1048 */
1049 public AbstractStringBuilder insert(int offset, Object obj) {
1050 return insert(offset, String.valueOf(obj));
1051 }
1052
1053 /**
1054 * Inserts the string into this character sequence.
1055 * <p>
1056 * The characters of the {@code String} argument are inserted, in
1057 * order, into this sequence at the indicated offset, moving up any
1058 * characters originally above that position and increasing the length
1059 * of this sequence by the length of the argument. If
1060 * {@code str} is {@code null}, then the four characters
1061 * {@code "null"} are inserted into this sequence.
1062 * <p>
1063 * The character at index <i>k</i> in the new character sequence is
1064 * equal to:
1065 * <ul>
1066 * <li>the character at index <i>k</i> in the old character sequence, if
1067 * <i>k</i> is less than {@code offset}
1068 * <li>the character at index <i>k</i>{@code -offset} in the
1069 * argument {@code str}, if <i>k</i> is not less than
1070 * {@code offset} but is less than {@code offset+str.length()}
1071 * <li>the character at index <i>k</i>{@code -str.length()} in the
1072 * old character sequence, if <i>k</i> is not less than
1073 * {@code offset+str.length()}
1074 * </ul><p>
1075 * The {@code offset} argument must be greater than or equal to
1076 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1077 * of this sequence.
1078 *
1079 * @param offset the offset.
1080 * @param str a string.
1081 * @return a reference to this object.
1082 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1083 */
1084 public AbstractStringBuilder insert(int offset, String str) {
1085 checkOffset(offset, count);
1086 if (str == null) {
1087 str = "null";
1088 }
1089 int len = str.length();
1090 ensureCapacityInternal(count + len);
1091 shift(offset, len);
1092 count += len;
1093 putStringAt(offset, str);
1094 return this;
1095 }
1096
1097 /**
1098 * Inserts the string representation of the {@code char} array
1099 * argument into this sequence.
1100 * <p>
1101 * The characters of the array argument are inserted into the
1102 * contents of this sequence at the position indicated by
1103 * {@code offset}. The length of this sequence increases by
1104 * the length of the argument.
1105 * <p>
1106 * The overall effect is exactly as if the second argument were
1107 * converted to a string by the method {@link String#valueOf(char[])},
1108 * and the characters of that string were then
1109 * {@link #insert(int,String) inserted} into this character
1110 * sequence at the indicated offset.
1111 * <p>
1112 * The {@code offset} argument must be greater than or equal to
1113 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1114 * of this sequence.
1115 *
1116 * @param offset the offset.
1117 * @param str a character array.
1118 * @return a reference to this object.
1119 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1120 */
1121 public AbstractStringBuilder insert(int offset, char[] str) {
1122 checkOffset(offset, count);
1123 int len = str.length;
1124 ensureCapacityInternal(count + len);
1125 shift(offset, len);
1126 count += len;
1127 putCharsAt(offset, str, 0, len);
1128 return this;
1129 }
1130
1131 /**
1132 * Inserts the specified {@code CharSequence} into this sequence.
1133 * <p>
1134 * The characters of the {@code CharSequence} argument are inserted,
1135 * in order, into this sequence at the indicated offset, moving up
1136 * any characters originally above that position and increasing the length
1137 * of this sequence by the length of the argument s.
1138 * <p>
1139 * The result of this method is exactly the same as if it were an
1140 * invocation of this object's
1141 * {@link #insert(int,CharSequence,int,int) insert}(dstOffset, s, 0, s.length())
1142 * method.
1143 *
1144 * <p>If {@code s} is {@code null}, then the four characters
1145 * {@code "null"} are inserted into this sequence.
1146 *
1147 * @param dstOffset the offset.
1148 * @param s the sequence to be inserted
1149 * @return a reference to this object.
1150 * @throws IndexOutOfBoundsException if the offset is invalid.
1151 */
1152 public AbstractStringBuilder insert(int dstOffset, CharSequence s) {
1153 if (s == null) {
1154 s = "null";
1155 }
1156 if (s instanceof String) {
1157 return this.insert(dstOffset, (String)s);
1158 }
1159 return this.insert(dstOffset, s, 0, s.length());
1160 }
1161
1162 /**
1163 * Inserts a subsequence of the specified {@code CharSequence} into
1164 * this sequence.
1165 * <p>
1166 * The subsequence of the argument {@code s} specified by
1167 * {@code start} and {@code end} are inserted,
1168 * in order, into this sequence at the specified destination offset, moving
1169 * up any characters originally above that position. The length of this
1170 * sequence is increased by {@code end - start}.
1171 * <p>
1172 * The character at index <i>k</i> in this sequence becomes equal to:
1173 * <ul>
1174 * <li>the character at index <i>k</i> in this sequence, if
1175 * <i>k</i> is less than {@code dstOffset}
1176 * <li>the character at index <i>k</i>{@code +start-dstOffset} in
1177 * the argument {@code s}, if <i>k</i> is greater than or equal to
1178 * {@code dstOffset} but is less than {@code dstOffset+end-start}
1179 * <li>the character at index <i>k</i>{@code -(end-start)} in this
1180 * sequence, if <i>k</i> is greater than or equal to
1181 * {@code dstOffset+end-start}
1182 * </ul><p>
1183 * The {@code dstOffset} argument must be greater than or equal to
1184 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1185 * of this sequence.
1186 * <p>The start argument must be nonnegative, and not greater than
1187 * {@code end}.
1188 * <p>The end argument must be greater than or equal to
1189 * {@code start}, and less than or equal to the length of s.
1190 *
1191 * <p>If {@code s} is {@code null}, then this method inserts
1192 * characters as if the s parameter was a sequence containing the four
1193 * characters {@code "null"}.
1194 *
1195 * @param dstOffset the offset in this sequence.
1196 * @param s the sequence to be inserted.
1197 * @param start the starting index of the subsequence to be inserted.
1198 * @param end the end index of the subsequence to be inserted.
1199 * @return a reference to this object.
1200 * @throws IndexOutOfBoundsException if {@code dstOffset}
1201 * is negative or greater than {@code this.length()}, or
1202 * {@code start} or {@code end} are negative, or
1203 * {@code start} is greater than {@code end} or
1204 * {@code end} is greater than {@code s.length()}
1205 */
1206 public AbstractStringBuilder insert(int dstOffset, CharSequence s,
1207 int start, int end)
1208 {
1209 if (s == null) {
1210 s = "null";
1211 }
1212 checkOffset(dstOffset, count);
1213 checkRange(start, end, s.length());
1214 int len = end - start;
1215 ensureCapacityInternal(count + len);
1216 shift(dstOffset, len);
1217 count += len;
1218 putCharsAt(dstOffset, s, start, end);
1219 return this;
1220 }
1221
1222 /**
1223 * Inserts the string representation of the {@code boolean}
1224 * argument into this sequence.
1225 * <p>
1226 * The overall effect is exactly as if the second argument were
1227 * converted to a string by the method {@link String#valueOf(boolean)},
1228 * and the characters of that string were then
1229 * {@link #insert(int,String) inserted} into this character
1230 * sequence at the indicated offset.
1231 * <p>
1232 * The {@code offset} argument must be greater than or equal to
1233 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1234 * of this sequence.
1235 *
1236 * @param offset the offset.
1237 * @param b a {@code boolean}.
1238 * @return a reference to this object.
1239 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1240 */
1241 public AbstractStringBuilder insert(int offset, boolean b) {
1242 return insert(offset, String.valueOf(b));
1243 }
1244
1245 /**
1246 * Inserts the string representation of the {@code char}
1247 * argument into this sequence.
1248 * <p>
1249 * The overall effect is exactly as if the second argument were
1250 * converted to a string by the method {@link String#valueOf(char)},
1251 * and the character in that string were then
1252 * {@link #insert(int,String) inserted} into this character
1253 * sequence at the indicated offset.
1254 * <p>
1255 * The {@code offset} argument must be greater than or equal to
1256 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1257 * of this sequence.
1258 *
1259 * @param offset the offset.
1260 * @param c a {@code char}.
1261 * @return a reference to this object.
1262 * @throws IndexOutOfBoundsException if the offset is invalid.
1263 */
1264 public AbstractStringBuilder insert(int offset, char c) {
1265 checkOffset(offset, count);
1266 ensureCapacityInternal(count + 1);
1267 shift(offset, 1);
1268 count += 1;
1269 if (isLatin1() && StringLatin1.canEncode(c)) {
1270 value[offset] = (byte)c;
1271 } else {
1272 if (isLatin1()) {
1273 inflate();
1274 }
1275 StringUTF16.putCharSB(value, offset, c);
1276 }
1277 return this;
1278 }
1279
1280 /**
1281 * Inserts the string representation of the second {@code int}
1282 * argument into this sequence.
1283 * <p>
1284 * The overall effect is exactly as if the second argument were
1285 * converted to a string by the method {@link String#valueOf(int)},
1286 * and the characters of that string were then
1287 * {@link #insert(int,String) inserted} into this character
1288 * sequence at the indicated offset.
1289 * <p>
1290 * The {@code offset} argument must be greater than or equal to
1291 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1292 * of this sequence.
1293 *
1294 * @param offset the offset.
1295 * @param i an {@code int}.
1296 * @return a reference to this object.
1297 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1298 */
1299 public AbstractStringBuilder insert(int offset, int i) {
1300 return insert(offset, String.valueOf(i));
1301 }
1302
1303 /**
1304 * Inserts the string representation of the {@code long}
1305 * argument into this sequence.
1306 * <p>
1307 * The overall effect is exactly as if the second argument were
1308 * converted to a string by the method {@link String#valueOf(long)},
1309 * and the characters of that string were then
1310 * {@link #insert(int,String) inserted} into this character
1311 * sequence at the indicated offset.
1312 * <p>
1313 * The {@code offset} argument must be greater than or equal to
1314 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1315 * of this sequence.
1316 *
1317 * @param offset the offset.
1318 * @param l a {@code long}.
1319 * @return a reference to this object.
1320 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1321 */
1322 public AbstractStringBuilder insert(int offset, long l) {
1323 return insert(offset, String.valueOf(l));
1324 }
1325
1326 /**
1327 * Inserts the string representation of the {@code float}
1328 * argument into this sequence.
1329 * <p>
1330 * The overall effect is exactly as if the second argument were
1331 * converted to a string by the method {@link String#valueOf(float)},
1332 * and the characters of that string were then
1333 * {@link #insert(int,String) inserted} into this character
1334 * sequence at the indicated offset.
1335 * <p>
1336 * The {@code offset} argument must be greater than or equal to
1337 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1338 * of this sequence.
1339 *
1340 * @param offset the offset.
1341 * @param f a {@code float}.
1342 * @return a reference to this object.
1343 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1344 */
1345 public AbstractStringBuilder insert(int offset, float f) {
1346 return insert(offset, String.valueOf(f));
1347 }
1348
1349 /**
1350 * Inserts the string representation of the {@code double}
1351 * argument into this sequence.
1352 * <p>
1353 * The overall effect is exactly as if the second argument were
1354 * converted to a string by the method {@link String#valueOf(double)},
1355 * and the characters of that string were then
1356 * {@link #insert(int,String) inserted} into this character
1357 * sequence at the indicated offset.
1358 * <p>
1359 * The {@code offset} argument must be greater than or equal to
1360 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1361 * of this sequence.
1362 *
1363 * @param offset the offset.
1364 * @param d a {@code double}.
1365 * @return a reference to this object.
1366 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1367 */
1368 public AbstractStringBuilder insert(int offset, double d) {
1369 return insert(offset, String.valueOf(d));
1370 }
1371
1372 /**
1373 * Returns the index within this string of the first occurrence of the
1374 * specified substring.
1375 *
1376 * <p>The returned index is the smallest value {@code k} for which:
1377 * <pre>{@code
1378 * this.toString().startsWith(str, k)
1379 * }</pre>
1380 * If no such value of {@code k} exists, then {@code -1} is returned.
1381 *
1382 * @param str the substring to search for.
1383 * @return the index of the first occurrence of the specified substring,
1384 * or {@code -1} if there is no such occurrence.
1385 */
1386 public int indexOf(String str) {
1387 return indexOf(str, 0);
1388 }
1389
1390 /**
1391 * Returns the index within this string of the first occurrence of the
1392 * specified substring, starting at the specified index.
1393 *
1394 * <p>The returned index is the smallest value {@code k} for which:
1395 * <pre>{@code
1396 * k >= Math.min(fromIndex, this.length()) &&
1397 * this.toString().startsWith(str, k)
1398 * }</pre>
1399 * If no such value of {@code k} exists, then {@code -1} is returned.
1400 *
1401 * @param str the substring to search for.
1402 * @param fromIndex the index from which to start the search.
1403 * @return the index of the first occurrence of the specified substring,
1404 * starting at the specified index,
1405 * or {@code -1} if there is no such occurrence.
1406 */
1407 public int indexOf(String str, int fromIndex) {
1408 return String.indexOf(value, coder, count, str, fromIndex);
1409 }
1410
1411 /**
1412 * Returns the index within this string of the last occurrence of the
1413 * specified substring. The last occurrence of the empty string "" is
1414 * considered to occur at the index value {@code this.length()}.
1415 *
1416 * <p>The returned index is the largest value {@code k} for which:
1417 * <pre>{@code
1418 * this.toString().startsWith(str, k)
1419 * }</pre>
1420 * If no such value of {@code k} exists, then {@code -1} is returned.
1421 *
1422 * @param str the substring to search for.
1423 * @return the index of the last occurrence of the specified substring,
1424 * or {@code -1} if there is no such occurrence.
1425 */
1426 public int lastIndexOf(String str) {
1427 return lastIndexOf(str, count);
1428 }
1429
1430 /**
1431 * Returns the index within this string of the last occurrence of the
1432 * specified substring, searching backward starting at the specified index.
1433 *
1434 * <p>The returned index is the largest value {@code k} for which:
1435 * <pre>{@code
1436 * k <= Math.min(fromIndex, this.length()) &&
1437 * this.toString().startsWith(str, k)
1438 * }</pre>
1439 * If no such value of {@code k} exists, then {@code -1} is returned.
1440 *
1441 * @param str the substring to search for.
1442 * @param fromIndex the index to start the search from.
1443 * @return the index of the last occurrence of the specified substring,
1444 * searching backward from the specified index,
1445 * or {@code -1} if there is no such occurrence.
1446 */
1447 public int lastIndexOf(String str, int fromIndex) {
1448 return String.lastIndexOf(value, coder, count, str, fromIndex);
1449 }
1450
1451 /**
1452 * Causes this character sequence to be replaced by the reverse of
1453 * the sequence. If there are any surrogate pairs included in the
1454 * sequence, these are treated as single characters for the
1455 * reverse operation. Thus, the order of the high-low surrogates
1456 * is never reversed.
1457 *
1458 * Let <i>n</i> be the character length of this character sequence
1459 * (not the length in {@code char} values) just prior to
1460 * execution of the {@code reverse} method. Then the
1461 * character at index <i>k</i> in the new character sequence is
1462 * equal to the character at index <i>n-k-1</i> in the old
1463 * character sequence.
1464 *
1465 * <p>Note that the reverse operation may result in producing
1466 * surrogate pairs that were unpaired low-surrogates and
1467 * high-surrogates before the operation. For example, reversing
1468 * "\u005CuDC00\u005CuD800" produces "\u005CuD800\u005CuDC00" which is
1469 * a valid surrogate pair.
1470 *
1471 * @return a reference to this object.
1472 */
1473 public AbstractStringBuilder reverse() {
1474 byte[] val = this.value;
1475 int count = this.count;
1476 int coder = this.coder;
1477 int n = count - 1;
1478 if (COMPACT_STRINGS && coder == LATIN1) {
1479 for (int j = (n-1) >> 1; j >= 0; j--) {
1480 int k = n - j;
1481 byte cj = val[j];
1482 val[j] = val[k];
1483 val[k] = cj;
1484 }
1485 } else {
1486 checkOffset(count, val.length >> 1);
1487 boolean hasSurrogates = false;
1488 for (int j = (n-1) >> 1; j >= 0; j--) {
1489 int k = n - j;
1490 char cj = StringUTF16.getChar(val, j);
1491 char ck = StringUTF16.getChar(val, k);
1492 StringUTF16.putChar(val, j, ck);
1493 StringUTF16.putChar(val, k, cj);
1494 if (Character.isSurrogate(cj) ||
1495 Character.isSurrogate(ck)) {
1496 hasSurrogates = true;
1497 }
1498 }
1499 if (hasSurrogates) {
1500 reverseAllValidSurrogatePairs(val, count);
1501 }
1502 }
1503 return this;
1504 }
1505
1506 /** Outlined helper method for reverse() */
1507 private void reverseAllValidSurrogatePairs(byte[] val, int count) {
1508 for (int i = 0; i < count - 1; i++) {
1509 char c2 = StringUTF16.getChar(val, i);
1510 if (Character.isLowSurrogate(c2)) {
1511 char c1 = StringUTF16.getChar(val, i + 1);
1512 if (Character.isHighSurrogate(c1)) {
1513 StringUTF16.putChar(val, i++, c1);
1514 StringUTF16.putChar(val, i, c2);
1515 }
1516 }
1517 }
1518 }
1519
1520 /**
1521 * Returns a string representing the data in this sequence.
1522 * A new {@code String} object is allocated and initialized to
1523 * contain the character sequence currently represented by this
1524 * object. This {@code String} is then returned. Subsequent
1525 * changes to this sequence do not affect the contents of the
1526 * {@code String}.
1527 *
1528 * @return a string representation of this sequence of characters.
1529 */
1530 @Override
1531 public abstract String toString();
1532
1533 /**
1534 * {@inheritDoc}
1535 * @since 9
1536 */
1537 @Override
1538 public IntStream chars() {
1539 byte[] val = this.value; int count = this.count; byte coder = this.coder;
1540 checkOffset(count, val.length >> coder);
1541 // Reuse String-based spliterator. This requires a supplier to
1542 // capture the value and count when the terminal operation is executed
1543 return StreamSupport.intStream(
1544 () -> coder == LATIN1 ? new StringLatin1.CharsSpliterator(val, 0, count, 0)
1545 : new StringUTF16.CharsSpliterator(val, 0, count, 0),
1546 Spliterator.ORDERED | Spliterator.SIZED | Spliterator.SUBSIZED,
1547 false);
1548 }
1549
1550 /**
1551 * {@inheritDoc}
1552 * @since 9
1553 */
1554 @Override
1555 public IntStream codePoints() {
1556 byte[] val = this.value; int count = this.count; byte coder = this.coder;
1557 checkOffset(count, val.length >> coder);
1558 // Reuse String-based spliterator. This requires a supplier to
1559 // capture the value and count when the terminal operation is executed
1560 return StreamSupport.intStream(
1561 () -> coder == LATIN1 ? new StringLatin1.CharsSpliterator(val, 0, count, 0)
1562 : new StringUTF16.CodePointsSpliterator(val, 0, count, 0),
1563 Spliterator.ORDERED,
1564 false);
1565 }
1566
1567 /**
1568 * Needed by {@code String} for the contentEquals method.
1569 */
1570 final byte[] getValue() {
1571 return value;
1572 }
1573
1574 /*
1575 * Invoker guarantees it is in UTF16 (inflate itself for asb), if two
1576 * coders are different and the dstBegin has enough space
1577 *
1578 * @param dstBegin the char index, not offset of byte[]
1579 * @param coder the coder of dst[]
1580 */
1581 void getBytes(byte dst[], int dstBegin, byte coder) {
1582 if (this.coder == coder) {
1583 System.arraycopy(value, 0, dst, dstBegin << coder, count << coder);
1584 } else { // this.coder == LATIN && coder == UTF16
1585 StringLatin1.inflate(value, 0, dst, dstBegin, count);
1586 }
1587 }
1588
1589 /* for readObject() */
1590 void initBytes(char[] value, int off, int len) {
1591 if (String.COMPACT_STRINGS) {
1592 this.value = StringUTF16.compress(value, off, len);
1593 if (this.value != null) {
1594 this.coder = LATIN1;
1595 return;
1596 }
1597 }
1598 this.coder = UTF16;
1599 this.value = StringUTF16.toBytes(value, off, len);
1600 }
1601
1602 final byte getCoder() {
1603 return COMPACT_STRINGS ? coder : UTF16;
1604 }
1605
1606 final boolean isLatin1() {
1607 return COMPACT_STRINGS && coder == LATIN1;
1608 }
1609
1610 private final void putCharsAt(int index, char[] s, int off, int end) {
1611 if (isLatin1()) {
1612 byte[] val = this.value;
1613 for (int i = off, j = index; i < end; i++) {
1614 char c = s[i];
1615 if (StringLatin1.canEncode(c)) {
1616 val[j++] = (byte)c;
1617 } else {
1618 inflate();
1619 StringUTF16.putCharsSB(this.value, j, s, i, end);
1620 return;
1621 }
1622 }
1623 } else {
1624 StringUTF16.putCharsSB(this.value, index, s, off, end);
1625 }
1626 }
1627
1628 private final void putCharsAt(int index, CharSequence s, int off, int end) {
1629 if (isLatin1()) {
1630 byte[] val = this.value;
1631 for (int i = off, j = index; i < end; i++) {
1632 char c = s.charAt(i);
1633 if (StringLatin1.canEncode(c)) {
1634 val[j++] = (byte)c;
1635 } else {
1636 inflate();
1637 StringUTF16.putCharsSB(this.value, j, s, i, end);
1638 return;
1639 }
1640 }
1641 } else {
1642 StringUTF16.putCharsSB(this.value, index, s, off, end);
1643 }
1644 }
1645
1646 private final void putStringAt(int index, String str) {
1647 if (getCoder() != str.coder()) {
1648 inflate();
1649 }
1650 str.getBytes(value, index, coder);
1651 }
1652
1653 private final void appendChars(char[] s, int off, int end) {
1654 if (isLatin1()) {
1655 byte[] val = this.value;
1656 for (int i = off, j = count; i < end; i++) {
1657 char c = s[i];
1658 if (StringLatin1.canEncode(c)) {
1659 val[j++] = (byte)c;
1660 } else {
1661 count = j;
1662 inflate();
1663 StringUTF16.putCharsSB(this.value, j, s, i, end);
1664 count += end - i;
1665 return;
1666 }
1667 }
1668 } else {
1669 StringUTF16.putCharsSB(this.value, count, s, off, end);
1670 }
1671 count += end - off;
1672 }
1673
1674 private final void appendChars(CharSequence s, int off, int end) {
1675 if (isLatin1()) {
1676 byte[] val = this.value;
1677 for (int i = off, j = count; i < end; i++) {
1678 char c = s.charAt(i);
1679 if (StringLatin1.canEncode(c)) {
1680 val[j++] = (byte)c;
1681 } else {
1682 count = j;
1683 inflate();
1684 StringUTF16.putCharsSB(this.value, j, s, i, end);
1685 count += end - i;
1686 return;
1687 }
1688 }
1689 } else {
1690 StringUTF16.putCharsSB(this.value, count, s, off, end);
1691 }
1692 count += end - off;
1693 }
1694
1695 /* IndexOutOfBoundsException, if out of bounds */
1696 private static void checkRange(int start, int end, int len) {
1697 if (start < 0 || start > end || end > len) {
1698 throw new IndexOutOfBoundsException(
1699 "start " + start + ", end " + end + ", length " + len);
1700 }
1701 }
1702
1703 /* StringIndexOutOfBoundsException, if out of bounds */
1704 private static void checkRangeSIOOBE(int start, int end, int len) {
1705 if (start < 0 || start > end || end > len) {
1706 throw new StringIndexOutOfBoundsException(
1707 "start " + start + ", end " + end + ", length " + len);
1708 }
1709 }
1710 }