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 sun.misc.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 * This implements the expansion semantics of ensureCapacity with no
149 * size check or synchronization.
150 */
151 private void expandCapacity(int minimumCapacity) {
152 int newCapacity = (value.length >> coder) * 2 + 2;
153 if (newCapacity - minimumCapacity < 0) {
154 newCapacity = minimumCapacity;
155 }
156 if (newCapacity < 0) {
157 if (minimumCapacity < 0) {// overflow
158 throw new OutOfMemoryError();
159 }
160 newCapacity = Integer.MAX_VALUE;
161 }
162 if (coder != LATIN1 && newCapacity > StringUTF16.MAX_LENGTH) {
163 if (minimumCapacity >= StringUTF16.MAX_LENGTH) {
164 throw new OutOfMemoryError();
165 }
166 newCapacity = StringUTF16.MAX_LENGTH;
167 }
168 this.value = Arrays.copyOf(value, newCapacity << coder);
169 }
170
171 /**
172 * If the coder is "isLatin1", this inflates the internal 8-bit storage
173 * to 16-bit <hi=0, low> pair storage.
174 */
175 private void inflate() {
176 if (!isLatin1()) {
177 return;
178 }
179 byte[] buf = StringUTF16.newBytesFor(value.length);
180 StringLatin1.inflateSB(value, buf, 0, count);
181 this.value = buf;
182 this.coder = UTF16;
183 }
184
185 /**
186 * Attempts to reduce storage used for the character sequence.
187 * If the buffer is larger than necessary to hold its current sequence of
188 * characters, then it may be resized to become more space efficient.
189 * Calling this method may, but is not required to, affect the value
190 * returned by a subsequent call to the {@link #capacity()} method.
191 */
192 public void trimToSize() {
193 int length = count << coder;
194 if (length < value.length) {
195 value = Arrays.copyOf(value, length);
196 }
197 }
198
199 /**
200 * Sets the length of the character sequence.
201 * The sequence is changed to a new character sequence
202 * whose length is specified by the argument. For every nonnegative
203 * index <i>k</i> less than {@code newLength}, the character at
204 * index <i>k</i> in the new character sequence is the same as the
205 * character at index <i>k</i> in the old sequence if <i>k</i> is less
206 * than the length of the old character sequence; otherwise, it is the
207 * null character {@code '\u005Cu0000'}.
208 *
209 * In other words, if the {@code newLength} argument is less than
210 * the current length, the length is changed to the specified length.
211 * <p>
212 * If the {@code newLength} argument is greater than or equal
213 * to the current length, sufficient null characters
214 * ({@code '\u005Cu0000'}) are appended so that
215 * length becomes the {@code newLength} argument.
216 * <p>
217 * The {@code newLength} argument must be greater than or equal
218 * to {@code 0}.
219 *
220 * @param newLength the new length
221 * @throws IndexOutOfBoundsException if the
222 * {@code newLength} argument is negative.
223 */
224 public void setLength(int newLength) {
225 if (newLength < 0) {
226 throw new StringIndexOutOfBoundsException(newLength);
227 }
228 ensureCapacityInternal(newLength);
229 if (count < newLength) {
230 if (isLatin1()) {
231 StringLatin1.fillNull(value, count, newLength);
232 } else {
233 StringUTF16.fillNull(value, count, newLength);
234 }
235 }
236 count = newLength;
237 }
238
239 /**
240 * Returns the {@code char} value in this sequence at the specified index.
241 * The first {@code char} value is at index {@code 0}, the next at index
242 * {@code 1}, and so on, as in array indexing.
243 * <p>
244 * The index argument must be greater than or equal to
245 * {@code 0}, and less than the length of this sequence.
246 *
247 * <p>If the {@code char} value specified by the index is a
248 * <a href="Character.html#unicode">surrogate</a>, the surrogate
249 * value is returned.
250 *
251 * @param index the index of the desired {@code char} value.
252 * @return the {@code char} value at the specified index.
253 * @throws IndexOutOfBoundsException if {@code index} is
254 * negative or greater than or equal to {@code length()}.
255 */
256 @Override
257 public char charAt(int index) {
258 checkIndex(index, count);
259 if (isLatin1()) {
260 return (char)(value[index] & 0xff);
261 }
262 return StringUTF16.charAt(value, index);
263 }
264
265 /**
266 * Returns the character (Unicode code point) at the specified
267 * index. The index refers to {@code char} values
268 * (Unicode code units) and ranges from {@code 0} to
269 * {@link #length()}{@code - 1}.
270 *
271 * <p> If the {@code char} value specified at the given index
272 * is in the high-surrogate range, the following index is less
273 * than the length of this sequence, and the
274 * {@code char} value at the following index is in the
275 * low-surrogate range, then the supplementary code point
276 * corresponding to this surrogate pair is returned. Otherwise,
277 * the {@code char} value at the given index is returned.
278 *
279 * @param index the index to the {@code char} values
280 * @return the code point value of the character at the
281 * {@code index}
282 * @exception IndexOutOfBoundsException if the {@code index}
283 * argument is negative or not less than the length of this
284 * sequence.
285 */
286 public int codePointAt(int index) {
287 checkIndex(index, count);
288 if (isLatin1()) {
289 return value[index] & 0xff;
290 }
291 return StringUTF16.codePointAtSB(value, index, count);
292 }
293
294 /**
295 * Returns the character (Unicode code point) before the specified
296 * index. The index refers to {@code char} values
297 * (Unicode code units) and ranges from {@code 1} to {@link
298 * #length()}.
299 *
300 * <p> If the {@code char} value at {@code (index - 1)}
301 * is in the low-surrogate range, {@code (index - 2)} is not
302 * negative, and the {@code char} value at {@code (index -
303 * 2)} is in the high-surrogate range, then the
304 * supplementary code point value of the surrogate pair is
305 * returned. If the {@code char} value at {@code index -
306 * 1} is an unpaired low-surrogate or a high-surrogate, the
307 * surrogate value is returned.
308 *
309 * @param index the index following the code point that should be returned
310 * @return the Unicode code point value before the given index.
311 * @exception IndexOutOfBoundsException if the {@code index}
312 * argument is less than 1 or greater than the length
313 * of this sequence.
314 */
315 public int codePointBefore(int index) {
316 int i = index - 1;
317 if (i < 0 || i >= count) {
318 throw new StringIndexOutOfBoundsException(index);
319 }
320 if (isLatin1()) {
321 return value[i] & 0xff;
322 }
323 return StringUTF16.codePointBeforeSB(value, index);
324 }
325
326 /**
327 * Returns the number of Unicode code points in the specified text
328 * range of this sequence. The text range begins at the specified
329 * {@code beginIndex} and extends to the {@code char} at
330 * index {@code endIndex - 1}. Thus the length (in
331 * {@code char}s) of the text range is
332 * {@code endIndex-beginIndex}. Unpaired surrogates within
333 * this sequence count as one code point each.
334 *
335 * @param beginIndex the index to the first {@code char} of
336 * the text range.
337 * @param endIndex the index after the last {@code char} of
338 * the text range.
339 * @return the number of Unicode code points in the specified text
340 * range
341 * @exception IndexOutOfBoundsException if the
342 * {@code beginIndex} is negative, or {@code endIndex}
343 * is larger than the length of this sequence, or
344 * {@code beginIndex} is larger than {@code endIndex}.
345 */
346 public int codePointCount(int beginIndex, int endIndex) {
347 if (beginIndex < 0 || endIndex > count || beginIndex > endIndex) {
348 throw new IndexOutOfBoundsException();
349 }
350 if (isLatin1()) {
351 return endIndex - beginIndex;
352 }
353 return StringUTF16.codePointCountSB(value, beginIndex, endIndex);
354 }
355
356 /**
357 * Returns the index within this sequence that is offset from the
358 * given {@code index} by {@code codePointOffset} code
359 * points. Unpaired surrogates within the text range given by
360 * {@code index} and {@code codePointOffset} count as
361 * one code point each.
362 *
363 * @param index the index to be offset
364 * @param codePointOffset the offset in code points
365 * @return the index within this sequence
366 * @exception IndexOutOfBoundsException if {@code index}
367 * is negative or larger then the length of this sequence,
368 * or if {@code codePointOffset} is positive and the subsequence
369 * starting with {@code index} has fewer than
370 * {@code codePointOffset} code points,
371 * or if {@code codePointOffset} is negative and the subsequence
372 * before {@code index} has fewer than the absolute value of
373 * {@code codePointOffset} code points.
374 */
375 public int offsetByCodePoints(int index, int codePointOffset) {
376 if (index < 0 || index > count) {
377 throw new IndexOutOfBoundsException();
378 }
379 return Character.offsetByCodePoints(this,
380 index, codePointOffset);
381 }
382
383 /**
384 * Characters are copied from this sequence into the
385 * destination character array {@code dst}. The first character to
386 * be copied is at index {@code srcBegin}; the last character to
387 * be copied is at index {@code srcEnd-1}. The total number of
388 * characters to be copied is {@code srcEnd-srcBegin}. The
389 * characters are copied into the subarray of {@code dst} starting
390 * at index {@code dstBegin} and ending at index:
391 * <pre>{@code
392 * dstbegin + (srcEnd-srcBegin) - 1
393 * }</pre>
394 *
395 * @param srcBegin start copying at this offset.
396 * @param srcEnd stop copying at this offset.
397 * @param dst the array to copy the data into.
398 * @param dstBegin offset into {@code dst}.
399 * @throws IndexOutOfBoundsException if any of the following is true:
400 * <ul>
401 * <li>{@code srcBegin} is negative
402 * <li>{@code dstBegin} is negative
403 * <li>the {@code srcBegin} argument is greater than
404 * the {@code srcEnd} argument.
405 * <li>{@code srcEnd} is greater than
406 * {@code this.length()}.
407 * <li>{@code dstBegin+srcEnd-srcBegin} is greater than
408 * {@code dst.length}
409 * </ul>
410 */
411 public void getChars(int srcBegin, int srcEnd, char[] dst, int dstBegin)
412 {
413 checkRangeSIOOBE(srcBegin, srcEnd, count); // compatible to old version
414 int n = srcEnd - srcBegin;
415 checkRange(dstBegin, dstBegin + n, dst.length);
416 if (isLatin1()) {
417 StringLatin1.getCharsSB(value, srcBegin, srcEnd, dst, dstBegin);
418 } else {
419 StringUTF16.getCharsSB(value, srcBegin, srcEnd, dst, dstBegin);
420 }
421 }
422
423 /**
424 * The character at the specified index is set to {@code ch}. This
425 * sequence is altered to represent a new character sequence that is
426 * identical to the old character sequence, except that it contains the
427 * character {@code ch} at position {@code index}.
428 * <p>
429 * The index argument must be greater than or equal to
430 * {@code 0}, and less than the length of this sequence.
431 *
432 * @param index the index of the character to modify.
433 * @param ch the new character.
434 * @throws IndexOutOfBoundsException if {@code index} is
435 * negative or greater than or equal to {@code length()}.
436 */
437 public void setCharAt(int index, char ch) {
438 checkIndex(index, count);
439 if (isLatin1() && StringLatin1.canEncode(ch)) {
440 value[index] = (byte)ch;
441 } else {
442 if (isLatin1()) {
443 inflate();
444 }
445 StringUTF16.putCharSB(value, index, ch);
446 }
447 }
448
449 /**
450 * Appends the string representation of the {@code Object} argument.
451 * <p>
452 * The overall effect is exactly as if the argument were converted
453 * to a string by the method {@link String#valueOf(Object)},
454 * and the characters of that string were then
455 * {@link #append(String) appended} to this character sequence.
456 *
457 * @param obj an {@code Object}.
458 * @return a reference to this object.
459 */
460 public AbstractStringBuilder append(Object obj) {
461 return append(String.valueOf(obj));
462 }
463
464 /**
465 * Appends the specified string to this character sequence.
466 * <p>
467 * The characters of the {@code String} argument are appended, in
468 * order, increasing the length of this sequence by the length of the
469 * argument. If {@code str} is {@code null}, then the four
470 * characters {@code "null"} are appended.
471 * <p>
472 * Let <i>n</i> be the length of this character sequence just prior to
473 * execution of the {@code append} method. Then the character at
474 * index <i>k</i> in the new character sequence is equal to the character
475 * at index <i>k</i> in the old character sequence, if <i>k</i> is less
476 * than <i>n</i>; otherwise, it is equal to the character at index
477 * <i>k-n</i> in the argument {@code str}.
478 *
479 * @param str a string.
480 * @return a reference to this object.
481 */
482 public AbstractStringBuilder append(String str) {
483 if (str == null) {
484 return appendNull();
485 }
486 int len = str.length();
487 ensureCapacityInternal(count + len);
488 putStringAt(count, str);
489 count += len;
490 return this;
491 }
492
493 // Documentation in subclasses because of synchro difference
494 public AbstractStringBuilder append(StringBuffer sb) {
495 return this.append((AbstractStringBuilder)sb);
496 }
497
498 /**
499 * @since 1.8
500 */
501 AbstractStringBuilder append(AbstractStringBuilder asb) {
502 if (asb == null) {
503 return appendNull();
504 }
505 int len = asb.length();
506 ensureCapacityInternal(count + len);
507 if (getCoder() != asb.getCoder()) {
508 inflate();
509 }
510 asb.getBytes(value, count, coder);
511 count += len;
512 return this;
513 }
514
515 // Documentation in subclasses because of synchro difference
516 @Override
517 public AbstractStringBuilder append(CharSequence s) {
518 if (s == null) {
519 return appendNull();
520 }
521 if (s instanceof String) {
522 return this.append((String)s);
523 }
524 if (s instanceof AbstractStringBuilder) {
525 return this.append((AbstractStringBuilder)s);
526 }
527 return this.append(s, 0, s.length());
528 }
529
530 private AbstractStringBuilder appendNull() {
531 ensureCapacityInternal(count + 4);
532 int count = this.count;
533 byte[] val = this.value;
534 if (isLatin1()) {
535 val[count++] = 'n';
536 val[count++] = 'u';
537 val[count++] = 'l';
538 val[count++] = 'l';
539 } else {
540 checkOffset(count + 4, val.length >> 1);
541 StringUTF16.putChar(val, count++, 'n');
542 StringUTF16.putChar(val, count++, 'u');
543 StringUTF16.putChar(val, count++, 'l');
544 StringUTF16.putChar(val, count++, 'l');
545 }
546 this.count = count;
547 return this;
548 }
549
550 /**
551 * Appends a subsequence of the specified {@code CharSequence} to this
552 * sequence.
553 * <p>
554 * Characters of the argument {@code s}, starting at
555 * index {@code start}, are appended, in order, to the contents of
556 * this sequence up to the (exclusive) index {@code end}. The length
557 * of this sequence is increased by the value of {@code end - start}.
558 * <p>
559 * Let <i>n</i> be the length of this character sequence just prior to
560 * execution of the {@code append} method. Then the character at
561 * index <i>k</i> in this character sequence becomes equal to the
562 * character at index <i>k</i> in this sequence, if <i>k</i> is less than
563 * <i>n</i>; otherwise, it is equal to the character at index
564 * <i>k+start-n</i> in the argument {@code s}.
565 * <p>
566 * If {@code s} is {@code null}, then this method appends
567 * characters as if the s parameter was a sequence containing the four
568 * characters {@code "null"}.
569 *
570 * @param s the sequence to append.
571 * @param start the starting index of the subsequence to be appended.
572 * @param end the end index of the subsequence to be appended.
573 * @return a reference to this object.
574 * @throws IndexOutOfBoundsException if
575 * {@code start} is negative, or
576 * {@code start} is greater than {@code end} or
577 * {@code end} is greater than {@code s.length()}
578 */
579 @Override
580 public AbstractStringBuilder append(CharSequence s, int start, int end) {
581 if (s == null) {
582 s = "null";
583 }
584 checkRange(start, end, s.length());
585 int len = end - start;
586 ensureCapacityInternal(count + len);
587 appendChars(s, start, end);
588 return this;
589 }
590
591 /**
592 * Appends the string representation of the {@code char} array
593 * argument to this sequence.
594 * <p>
595 * The characters of the array argument are appended, in order, to
596 * the contents of this sequence. The length of this sequence
597 * increases by the length of the argument.
598 * <p>
599 * The overall effect is exactly as if the argument were converted
600 * to a string by the method {@link String#valueOf(char[])},
601 * and the characters of that string were then
602 * {@link #append(String) appended} to this character sequence.
603 *
604 * @param str the characters to be appended.
605 * @return a reference to this object.
606 */
607 public AbstractStringBuilder append(char[] str) {
608 int len = str.length;
609 ensureCapacityInternal(count + len);
610 appendChars(str, 0, len);
611 return this;
612 }
613
614 /**
615 * Appends the string representation of a subarray of the
616 * {@code char} array argument to this sequence.
617 * <p>
618 * Characters of the {@code char} array {@code str}, starting at
619 * index {@code offset}, are appended, in order, to the contents
620 * of this sequence. The length of this sequence increases
621 * by the value of {@code len}.
622 * <p>
623 * The overall effect is exactly as if the arguments were converted
624 * to a string by the method {@link String#valueOf(char[],int,int)},
625 * and the characters of that string were then
626 * {@link #append(String) appended} to this character sequence.
627 *
628 * @param str the characters to be appended.
629 * @param offset the index of the first {@code char} to append.
630 * @param len the number of {@code char}s to append.
631 * @return a reference to this object.
632 * @throws IndexOutOfBoundsException
633 * if {@code offset < 0} or {@code len < 0}
634 * or {@code offset+len > str.length}
635 */
636 public AbstractStringBuilder append(char str[], int offset, int len) {
637 int end = offset + len;
638 checkRange(offset, end, str.length);
639 ensureCapacityInternal(count + len);
640 appendChars(str, offset, end);
641 return this;
642 }
643
644 /**
645 * Appends the string representation of the {@code boolean}
646 * argument to the sequence.
647 * <p>
648 * The overall effect is exactly as if the argument were converted
649 * to a string by the method {@link String#valueOf(boolean)},
650 * and the characters of that string were then
651 * {@link #append(String) appended} to this character sequence.
652 *
653 * @param b a {@code boolean}.
654 * @return a reference to this object.
655 */
656 public AbstractStringBuilder append(boolean b) {
657 ensureCapacityInternal(count + (b ? 4 : 5));
658 int count = this.count;
659 byte[] val = this.value;
660 if (isLatin1()) {
661 if (b) {
662 val[count++] = 't';
663 val[count++] = 'r';
664 val[count++] = 'u';
665 val[count++] = 'e';
666 } else {
667 val[count++] = 'f';
668 val[count++] = 'a';
669 val[count++] = 'l';
670 val[count++] = 's';
671 val[count++] = 'e';
672 }
673 } else {
674 if (b) {
675 checkOffset(count + 4, val.length >> 1);
676 StringUTF16.putChar(val, count++, 't');
677 StringUTF16.putChar(val, count++, 'r');
678 StringUTF16.putChar(val, count++, 'u');
679 StringUTF16.putChar(val, count++, 'e');
680 } else {
681 checkOffset(count + 5, val.length >> 1);
682 StringUTF16.putChar(val, count++, 'f');
683 StringUTF16.putChar(val, count++, 'a');
684 StringUTF16.putChar(val, count++, 'l');
685 StringUTF16.putChar(val, count++, 's');
686 StringUTF16.putChar(val, count++, 'e');
687 }
688 }
689 this.count = count;
690 return this;
691 }
692
693 /**
694 * Appends the string representation of the {@code char}
695 * argument to this sequence.
696 * <p>
697 * The argument is appended to the contents of this sequence.
698 * The length of this sequence increases by {@code 1}.
699 * <p>
700 * The overall effect is exactly as if the argument were converted
701 * to a string by the method {@link String#valueOf(char)},
702 * and the character in that string were then
703 * {@link #append(String) appended} to this character sequence.
704 *
705 * @param c a {@code char}.
706 * @return a reference to this object.
707 */
708 @Override
709 public AbstractStringBuilder append(char c) {
710 ensureCapacityInternal(count + 1);
711 if (isLatin1() && StringLatin1.canEncode(c)) {
712 value[count++] = (byte)c;
713 } else {
714 if (isLatin1()) {
715 inflate();
716 }
717 StringUTF16.putCharSB(value, count++, c);
718 }
719 return this;
720 }
721
722 /**
723 * Appends the string representation of the {@code int}
724 * argument to this sequence.
725 * <p>
726 * The overall effect is exactly as if the argument were converted
727 * to a string by the method {@link String#valueOf(int)},
728 * and the characters of that string were then
729 * {@link #append(String) appended} to this character sequence.
730 *
731 * @param i an {@code int}.
732 * @return a reference to this object.
733 */
734 public AbstractStringBuilder append(int i) {
735 if (i == Integer.MIN_VALUE) {
736 append("-2147483648");
737 return this;
738 }
739 int appendedLength = (i < 0) ? Integer.stringSize(-i) + 1
740 : Integer.stringSize(i);
741 int spaceNeeded = count + appendedLength;
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 if (l == Long.MIN_VALUE) {
768 append("-9223372036854775808");
769 return this;
770 }
771 int appendedLength = (l < 0) ? Long.stringSize(-l) + 1
772 : Long.stringSize(l);
773 int spaceNeeded = count + appendedLength;
774 ensureCapacityInternal(spaceNeeded);
775 if (isLatin1()) {
776 Long.getChars(l, spaceNeeded, value);
777 } else {
778 byte[] val = this.value;
779 checkOffset(spaceNeeded, val.length >> 1);
780 Long.getCharsUTF16(l, spaceNeeded, val);
781 }
782 count = spaceNeeded;
783 return this;
784 }
785
786 /**
787 * Appends the string representation of the {@code float}
788 * argument to this sequence.
789 * <p>
790 * The overall effect is exactly as if the argument were converted
791 * to a string by the method {@link String#valueOf(float)},
792 * and the characters of that string were then
793 * {@link #append(String) appended} to this character sequence.
794 *
795 * @param f a {@code float}.
796 * @return a reference to this object.
797 */
798 public AbstractStringBuilder append(float f) {
799 FloatingDecimal.appendTo(f,this);
800 return this;
801 }
802
803 /**
804 * Appends the string representation of the {@code double}
805 * argument to this sequence.
806 * <p>
807 * The overall effect is exactly as if the argument were converted
808 * to a string by the method {@link String#valueOf(double)},
809 * and the characters of that string were then
810 * {@link #append(String) appended} to this character sequence.
811 *
812 * @param d a {@code double}.
813 * @return a reference to this object.
814 */
815 public AbstractStringBuilder append(double d) {
816 FloatingDecimal.appendTo(d,this);
817 return this;
818 }
819
820 /**
821 * Removes the characters in a substring of this sequence.
822 * The substring begins at the specified {@code start} and extends to
823 * the character at index {@code end - 1} or to the end of the
824 * sequence if no such character exists. If
825 * {@code start} is equal to {@code end}, no changes are made.
826 *
827 * @param start The beginning index, inclusive.
828 * @param end The ending index, exclusive.
829 * @return This object.
830 * @throws StringIndexOutOfBoundsException if {@code start}
831 * is negative, greater than {@code length()}, or
832 * greater than {@code end}.
833 */
834 public AbstractStringBuilder delete(int start, int end) {
835 if (end > count) {
836 end = count;
837 }
838 checkRangeSIOOBE(start, end, count);
839 int len = end - start;
840 if (len > 0) {
841 shift(end, -len);
842 count -= len;
843 }
844 return this;
845 }
846
847 /**
848 * Appends the string representation of the {@code codePoint}
849 * argument to this sequence.
850 *
851 * <p> The argument is appended to the contents of this sequence.
852 * The length of this sequence increases by
853 * {@link Character#charCount(int) Character.charCount(codePoint)}.
854 *
855 * <p> The overall effect is exactly as if the argument were
856 * converted to a {@code char} array by the method
857 * {@link Character#toChars(int)} and the character in that array
858 * were then {@link #append(char[]) appended} to this character
859 * sequence.
860 *
861 * @param codePoint a Unicode code point
862 * @return a reference to this object.
863 * @exception IllegalArgumentException if the specified
864 * {@code codePoint} isn't a valid Unicode code point
865 */
866 public AbstractStringBuilder appendCodePoint(int codePoint) {
867 if (Character.isBmpCodePoint(codePoint)) {
868 return append((char)codePoint);
869 }
870 return append(Character.toChars(codePoint));
871 }
872
873 /**
874 * Removes the {@code char} at the specified position in this
875 * sequence. This sequence is shortened by one {@code char}.
876 *
877 * <p>Note: If the character at the given index is a supplementary
878 * character, this method does not remove the entire character. If
879 * correct handling of supplementary characters is required,
880 * determine the number of {@code char}s to remove by calling
881 * {@code Character.charCount(thisSequence.codePointAt(index))},
882 * where {@code thisSequence} is this sequence.
883 *
884 * @param index Index of {@code char} to remove
885 * @return This object.
886 * @throws StringIndexOutOfBoundsException if the {@code index}
887 * is negative or greater than or equal to
888 * {@code length()}.
889 */
890 public AbstractStringBuilder deleteCharAt(int index) {
891 checkIndex(index, count);
892 shift(index + 1, -1);
893 count--;
894 return this;
895 }
896
897 /**
898 * Replaces the characters in a substring of this sequence
899 * with characters in the specified {@code String}. The substring
900 * begins at the specified {@code start} and extends to the character
901 * at index {@code end - 1} or to the end of the
902 * sequence if no such character exists. First the
903 * characters in the substring are removed and then the specified
904 * {@code String} is inserted at {@code start}. (This
905 * sequence will be lengthened to accommodate the
906 * specified String if necessary.)
907 *
908 * @param start The beginning index, inclusive.
909 * @param end The ending index, exclusive.
910 * @param str String that will replace previous contents.
911 * @return This object.
912 * @throws StringIndexOutOfBoundsException if {@code start}
913 * is negative, greater than {@code length()}, or
914 * greater than {@code end}.
915 */
916 public AbstractStringBuilder replace(int start, int end, String str) {
917 if (end > count) {
918 end = count;
919 }
920 checkRangeSIOOBE(start, end, count);
921 int len = str.length();
922 int newCount = count + len - (end - start);
923 ensureCapacityInternal(newCount);
924 shift(end, newCount - count);
925 count = newCount;
926 putStringAt(start, str);
927 return this;
928 }
929
930 /**
931 * Returns a new {@code String} that contains a subsequence of
932 * characters currently contained in this character sequence. The
933 * substring begins at the specified index and extends to the end of
934 * this sequence.
935 *
936 * @param start The beginning index, inclusive.
937 * @return The new string.
938 * @throws StringIndexOutOfBoundsException if {@code start} is
939 * less than zero, or greater than the length of this object.
940 */
941 public String substring(int start) {
942 return substring(start, count);
943 }
944
945 /**
946 * Returns a new character sequence that is a subsequence of this sequence.
947 *
948 * <p> An invocation of this method of the form
949 *
950 * <pre>{@code
951 * sb.subSequence(begin, end)}</pre>
952 *
953 * behaves in exactly the same way as the invocation
954 *
955 * <pre>{@code
956 * sb.substring(begin, end)}</pre>
957 *
958 * This method is provided so that this class can
959 * implement the {@link CharSequence} interface.
960 *
961 * @param start the start index, inclusive.
962 * @param end the end index, exclusive.
963 * @return the specified subsequence.
964 *
965 * @throws IndexOutOfBoundsException
966 * if {@code start} or {@code end} are negative,
967 * if {@code end} is greater than {@code length()},
968 * or if {@code start} is greater than {@code end}
969 * @spec JSR-51
970 */
971 @Override
972 public CharSequence subSequence(int start, int end) {
973 return substring(start, end);
974 }
975
976 /**
977 * Returns a new {@code String} that contains a subsequence of
978 * characters currently contained in this sequence. The
979 * substring begins at the specified {@code start} and
980 * extends to the character at index {@code end - 1}.
981 *
982 * @param start The beginning index, inclusive.
983 * @param end The ending index, exclusive.
984 * @return The new string.
985 * @throws StringIndexOutOfBoundsException if {@code start}
986 * or {@code end} are negative or greater than
987 * {@code length()}, or {@code start} is
988 * greater than {@code end}.
989 */
990 public String substring(int start, int end) {
991 checkRangeSIOOBE(start, end, count);
992 if (isLatin1()) {
993 return StringLatin1.newString(value, start, end - start);
994 }
995 return StringUTF16.newStringSB(value, start, end - start);
996 }
997
998 private void shift(int offset, int n) {
999 System.arraycopy(value, offset << coder,
1000 value, (offset + n) << coder, (count - offset) << coder);
1001 }
1002
1003 /**
1004 * Inserts the string representation of a subarray of the {@code str}
1005 * array argument into this sequence. The subarray begins at the
1006 * specified {@code offset} and extends {@code len} {@code char}s.
1007 * The characters of the subarray are inserted into this sequence at
1008 * the position indicated by {@code index}. The length of this
1009 * sequence increases by {@code len} {@code char}s.
1010 *
1011 * @param index position at which to insert subarray.
1012 * @param str A {@code char} array.
1013 * @param offset the index of the first {@code char} in subarray to
1014 * be inserted.
1015 * @param len the number of {@code char}s in the subarray to
1016 * be inserted.
1017 * @return This object
1018 * @throws StringIndexOutOfBoundsException if {@code index}
1019 * is negative or greater than {@code length()}, or
1020 * {@code offset} or {@code len} are negative, or
1021 * {@code (offset+len)} is greater than
1022 * {@code str.length}.
1023 */
1024 public AbstractStringBuilder insert(int index, char[] str, int offset,
1025 int len)
1026 {
1027 checkOffset(index, count);
1028 checkRangeSIOOBE(offset, offset + len, str.length);
1029 ensureCapacityInternal(count + len);
1030 shift(index, len);
1031 count += len;
1032 putCharsAt(index, str, offset, offset + len);
1033 return this;
1034 }
1035
1036 /**
1037 * Inserts the string representation of the {@code Object}
1038 * argument into this character sequence.
1039 * <p>
1040 * The overall effect is exactly as if the second argument were
1041 * converted to a string by the method {@link String#valueOf(Object)},
1042 * and the characters of that string were then
1043 * {@link #insert(int,String) inserted} into this character
1044 * sequence at the indicated offset.
1045 * <p>
1046 * The {@code offset} argument must be greater than or equal to
1047 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1048 * of this sequence.
1049 *
1050 * @param offset the offset.
1051 * @param obj an {@code Object}.
1052 * @return a reference to this object.
1053 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1054 */
1055 public AbstractStringBuilder insert(int offset, Object obj) {
1056 return insert(offset, String.valueOf(obj));
1057 }
1058
1059 /**
1060 * Inserts the string into this character sequence.
1061 * <p>
1062 * The characters of the {@code String} argument are inserted, in
1063 * order, into this sequence at the indicated offset, moving up any
1064 * characters originally above that position and increasing the length
1065 * of this sequence by the length of the argument. If
1066 * {@code str} is {@code null}, then the four characters
1067 * {@code "null"} are inserted into this sequence.
1068 * <p>
1069 * The character at index <i>k</i> in the new character sequence is
1070 * equal to:
1071 * <ul>
1072 * <li>the character at index <i>k</i> in the old character sequence, if
1073 * <i>k</i> is less than {@code offset}
1074 * <li>the character at index <i>k</i>{@code -offset} in the
1075 * argument {@code str}, if <i>k</i> is not less than
1076 * {@code offset} but is less than {@code offset+str.length()}
1077 * <li>the character at index <i>k</i>{@code -str.length()} in the
1078 * old character sequence, if <i>k</i> is not less than
1079 * {@code offset+str.length()}
1080 * </ul><p>
1081 * The {@code offset} argument must be greater than or equal to
1082 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1083 * of this sequence.
1084 *
1085 * @param offset the offset.
1086 * @param str a string.
1087 * @return a reference to this object.
1088 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1089 */
1090 public AbstractStringBuilder insert(int offset, String str) {
1091 checkOffset(offset, count);
1092 if (str == null) {
1093 str = "null";
1094 }
1095 int len = str.length();
1096 ensureCapacityInternal(count + len);
1097 shift(offset, len);
1098 count += len;
1099 putStringAt(offset, str);
1100 return this;
1101 }
1102
1103 /**
1104 * Inserts the string representation of the {@code char} array
1105 * argument into this sequence.
1106 * <p>
1107 * The characters of the array argument are inserted into the
1108 * contents of this sequence at the position indicated by
1109 * {@code offset}. The length of this sequence increases by
1110 * the length of the argument.
1111 * <p>
1112 * The overall effect is exactly as if the second argument were
1113 * converted to a string by the method {@link String#valueOf(char[])},
1114 * and the characters of that string were then
1115 * {@link #insert(int,String) inserted} into this character
1116 * sequence at the indicated offset.
1117 * <p>
1118 * The {@code offset} argument must be greater than or equal to
1119 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1120 * of this sequence.
1121 *
1122 * @param offset the offset.
1123 * @param str a character array.
1124 * @return a reference to this object.
1125 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1126 */
1127 public AbstractStringBuilder insert(int offset, char[] str) {
1128 checkOffset(offset, count);
1129 int len = str.length;
1130 ensureCapacityInternal(count + len);
1131 shift(offset, len);
1132 count += len;
1133 putCharsAt(offset, str, 0, len);
1134 return this;
1135 }
1136
1137 /**
1138 * Inserts the specified {@code CharSequence} into this sequence.
1139 * <p>
1140 * The characters of the {@code CharSequence} argument are inserted,
1141 * in order, into this sequence at the indicated offset, moving up
1142 * any characters originally above that position and increasing the length
1143 * of this sequence by the length of the argument s.
1144 * <p>
1145 * The result of this method is exactly the same as if it were an
1146 * invocation of this object's
1147 * {@link #insert(int,CharSequence,int,int) insert}(dstOffset, s, 0, s.length())
1148 * method.
1149 *
1150 * <p>If {@code s} is {@code null}, then the four characters
1151 * {@code "null"} are inserted into this sequence.
1152 *
1153 * @param dstOffset the offset.
1154 * @param s the sequence to be inserted
1155 * @return a reference to this object.
1156 * @throws IndexOutOfBoundsException if the offset is invalid.
1157 */
1158 public AbstractStringBuilder insert(int dstOffset, CharSequence s) {
1159 if (s == null) {
1160 s = "null";
1161 }
1162 if (s instanceof String) {
1163 return this.insert(dstOffset, (String)s);
1164 }
1165 return this.insert(dstOffset, s, 0, s.length());
1166 }
1167
1168 /**
1169 * Inserts a subsequence of the specified {@code CharSequence} into
1170 * this sequence.
1171 * <p>
1172 * The subsequence of the argument {@code s} specified by
1173 * {@code start} and {@code end} are inserted,
1174 * in order, into this sequence at the specified destination offset, moving
1175 * up any characters originally above that position. The length of this
1176 * sequence is increased by {@code end - start}.
1177 * <p>
1178 * The character at index <i>k</i> in this sequence becomes equal to:
1179 * <ul>
1180 * <li>the character at index <i>k</i> in this sequence, if
1181 * <i>k</i> is less than {@code dstOffset}
1182 * <li>the character at index <i>k</i>{@code +start-dstOffset} in
1183 * the argument {@code s}, if <i>k</i> is greater than or equal to
1184 * {@code dstOffset} but is less than {@code dstOffset+end-start}
1185 * <li>the character at index <i>k</i>{@code -(end-start)} in this
1186 * sequence, if <i>k</i> is greater than or equal to
1187 * {@code dstOffset+end-start}
1188 * </ul><p>
1189 * The {@code dstOffset} argument must be greater than or equal to
1190 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1191 * of this sequence.
1192 * <p>The start argument must be nonnegative, and not greater than
1193 * {@code end}.
1194 * <p>The end argument must be greater than or equal to
1195 * {@code start}, and less than or equal to the length of s.
1196 *
1197 * <p>If {@code s} is {@code null}, then this method inserts
1198 * characters as if the s parameter was a sequence containing the four
1199 * characters {@code "null"}.
1200 *
1201 * @param dstOffset the offset in this sequence.
1202 * @param s the sequence to be inserted.
1203 * @param start the starting index of the subsequence to be inserted.
1204 * @param end the end index of the subsequence to be inserted.
1205 * @return a reference to this object.
1206 * @throws IndexOutOfBoundsException if {@code dstOffset}
1207 * is negative or greater than {@code this.length()}, or
1208 * {@code start} or {@code end} are negative, or
1209 * {@code start} is greater than {@code end} or
1210 * {@code end} is greater than {@code s.length()}
1211 */
1212 public AbstractStringBuilder insert(int dstOffset, CharSequence s,
1213 int start, int end)
1214 {
1215 if (s == null) {
1216 s = "null";
1217 }
1218 checkOffset(dstOffset, count);
1219 checkRange(start, end, s.length());
1220 int len = end - start;
1221 ensureCapacityInternal(count + len);
1222 shift(dstOffset, len);
1223 count += len;
1224 putCharsAt(dstOffset, s, start, end);
1225 return this;
1226 }
1227
1228 /**
1229 * Inserts the string representation of the {@code boolean}
1230 * argument into this sequence.
1231 * <p>
1232 * The overall effect is exactly as if the second argument were
1233 * converted to a string by the method {@link String#valueOf(boolean)},
1234 * and the characters of that string were then
1235 * {@link #insert(int,String) inserted} into this character
1236 * sequence at the indicated offset.
1237 * <p>
1238 * The {@code offset} argument must be greater than or equal to
1239 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1240 * of this sequence.
1241 *
1242 * @param offset the offset.
1243 * @param b a {@code boolean}.
1244 * @return a reference to this object.
1245 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1246 */
1247 public AbstractStringBuilder insert(int offset, boolean b) {
1248 return insert(offset, String.valueOf(b));
1249 }
1250
1251 /**
1252 * Inserts the string representation of the {@code char}
1253 * argument into this sequence.
1254 * <p>
1255 * The overall effect is exactly as if the second argument were
1256 * converted to a string by the method {@link String#valueOf(char)},
1257 * and the character in that string were then
1258 * {@link #insert(int,String) inserted} into this character
1259 * sequence at the indicated offset.
1260 * <p>
1261 * The {@code offset} argument must be greater than or equal to
1262 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1263 * of this sequence.
1264 *
1265 * @param offset the offset.
1266 * @param c a {@code char}.
1267 * @return a reference to this object.
1268 * @throws IndexOutOfBoundsException if the offset is invalid.
1269 */
1270 public AbstractStringBuilder insert(int offset, char c) {
1271 checkOffset(offset, count);
1272 ensureCapacityInternal(count + 1);
1273 shift(offset, 1);
1274 count += 1;
1275 if (isLatin1() && StringLatin1.canEncode(c)) {
1276 value[offset] = (byte)c;
1277 } else {
1278 if (isLatin1()) {
1279 inflate();
1280 }
1281 StringUTF16.putCharSB(value, offset, c);
1282 }
1283 return this;
1284 }
1285
1286 /**
1287 * Inserts the string representation of the second {@code int}
1288 * argument into this sequence.
1289 * <p>
1290 * The overall effect is exactly as if the second argument were
1291 * converted to a string by the method {@link String#valueOf(int)},
1292 * and the characters of that string were then
1293 * {@link #insert(int,String) inserted} into this character
1294 * sequence at the indicated offset.
1295 * <p>
1296 * The {@code offset} argument must be greater than or equal to
1297 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1298 * of this sequence.
1299 *
1300 * @param offset the offset.
1301 * @param i an {@code int}.
1302 * @return a reference to this object.
1303 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1304 */
1305 public AbstractStringBuilder insert(int offset, int i) {
1306 return insert(offset, String.valueOf(i));
1307 }
1308
1309 /**
1310 * Inserts the string representation of the {@code long}
1311 * argument into this sequence.
1312 * <p>
1313 * The overall effect is exactly as if the second argument were
1314 * converted to a string by the method {@link String#valueOf(long)},
1315 * and the characters of that string were then
1316 * {@link #insert(int,String) inserted} into this character
1317 * sequence at the indicated offset.
1318 * <p>
1319 * The {@code offset} argument must be greater than or equal to
1320 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1321 * of this sequence.
1322 *
1323 * @param offset the offset.
1324 * @param l a {@code long}.
1325 * @return a reference to this object.
1326 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1327 */
1328 public AbstractStringBuilder insert(int offset, long l) {
1329 return insert(offset, String.valueOf(l));
1330 }
1331
1332 /**
1333 * Inserts the string representation of the {@code float}
1334 * argument into this sequence.
1335 * <p>
1336 * The overall effect is exactly as if the second argument were
1337 * converted to a string by the method {@link String#valueOf(float)},
1338 * and the characters of that string were then
1339 * {@link #insert(int,String) inserted} into this character
1340 * sequence at the indicated offset.
1341 * <p>
1342 * The {@code offset} argument must be greater than or equal to
1343 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1344 * of this sequence.
1345 *
1346 * @param offset the offset.
1347 * @param f a {@code float}.
1348 * @return a reference to this object.
1349 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1350 */
1351 public AbstractStringBuilder insert(int offset, float f) {
1352 return insert(offset, String.valueOf(f));
1353 }
1354
1355 /**
1356 * Inserts the string representation of the {@code double}
1357 * argument into this sequence.
1358 * <p>
1359 * The overall effect is exactly as if the second argument were
1360 * converted to a string by the method {@link String#valueOf(double)},
1361 * and the characters of that string were then
1362 * {@link #insert(int,String) inserted} into this character
1363 * sequence at the indicated offset.
1364 * <p>
1365 * The {@code offset} argument must be greater than or equal to
1366 * {@code 0}, and less than or equal to the {@linkplain #length() length}
1367 * of this sequence.
1368 *
1369 * @param offset the offset.
1370 * @param d a {@code double}.
1371 * @return a reference to this object.
1372 * @throws StringIndexOutOfBoundsException if the offset is invalid.
1373 */
1374 public AbstractStringBuilder insert(int offset, double d) {
1375 return insert(offset, String.valueOf(d));
1376 }
1377
1378 /**
1379 * Returns the index within this string of the first occurrence of the
1380 * specified substring.
1381 *
1382 * <p>The returned index is the smallest value {@code k} for which:
1383 * <pre>{@code
1384 * this.toString().startsWith(str, k)
1385 * }</pre>
1386 * If no such value of {@code k} exists, then {@code -1} is returned.
1387 *
1388 * @param str the substring to search for.
1389 * @return the index of the first occurrence of the specified substring,
1390 * or {@code -1} if there is no such occurrence.
1391 */
1392 public int indexOf(String str) {
1393 return indexOf(str, 0);
1394 }
1395
1396 /**
1397 * Returns the index within this string of the first occurrence of the
1398 * specified substring, starting at the specified index.
1399 *
1400 * <p>The returned index is the smallest value {@code k} for which:
1401 * <pre>{@code
1402 * k >= Math.min(fromIndex, this.length()) &&
1403 * this.toString().startsWith(str, k)
1404 * }</pre>
1405 * If no such value of {@code k} exists, then {@code -1} is returned.
1406 *
1407 * @param str the substring to search for.
1408 * @param fromIndex the index from which to start the search.
1409 * @return the index of the first occurrence of the specified substring,
1410 * starting at the specified index,
1411 * or {@code -1} if there is no such occurrence.
1412 */
1413 public int indexOf(String str, int fromIndex) {
1414 return String.indexOf(value, coder, count, str, fromIndex);
1415 }
1416
1417 /**
1418 * Returns the index within this string of the last occurrence of the
1419 * specified substring. The last occurrence of the empty string "" is
1420 * considered to occur at the index value {@code this.length()}.
1421 *
1422 * <p>The returned index is the largest value {@code k} for which:
1423 * <pre>{@code
1424 * this.toString().startsWith(str, k)
1425 * }</pre>
1426 * If no such value of {@code k} exists, then {@code -1} is returned.
1427 *
1428 * @param str the substring to search for.
1429 * @return the index of the last occurrence of the specified substring,
1430 * or {@code -1} if there is no such occurrence.
1431 */
1432 public int lastIndexOf(String str) {
1433 return lastIndexOf(str, count);
1434 }
1435
1436 /**
1437 * Returns the index within this string of the last occurrence of the
1438 * specified substring, searching backward starting at the specified index.
1439 *
1440 * <p>The returned index is the largest value {@code k} for which:
1441 * <pre>{@code
1442 * k <= Math.min(fromIndex, this.length()) &&
1443 * this.toString().startsWith(str, k)
1444 * }</pre>
1445 * If no such value of {@code k} exists, then {@code -1} is returned.
1446 *
1447 * @param str the substring to search for.
1448 * @param fromIndex the index to start the search from.
1449 * @return the index of the last occurrence of the specified substring,
1450 * searching backward from the specified index,
1451 * or {@code -1} if there is no such occurrence.
1452 */
1453 public int lastIndexOf(String str, int fromIndex) {
1454 return String.lastIndexOf(value, coder, count, str, fromIndex);
1455 }
1456
1457 /**
1458 * Causes this character sequence to be replaced by the reverse of
1459 * the sequence. If there are any surrogate pairs included in the
1460 * sequence, these are treated as single characters for the
1461 * reverse operation. Thus, the order of the high-low surrogates
1462 * is never reversed.
1463 *
1464 * Let <i>n</i> be the character length of this character sequence
1465 * (not the length in {@code char} values) just prior to
1466 * execution of the {@code reverse} method. Then the
1467 * character at index <i>k</i> in the new character sequence is
1468 * equal to the character at index <i>n-k-1</i> in the old
1469 * character sequence.
1470 *
1471 * <p>Note that the reverse operation may result in producing
1472 * surrogate pairs that were unpaired low-surrogates and
1473 * high-surrogates before the operation. For example, reversing
1474 * "\u005CuDC00\u005CuD800" produces "\u005CuD800\u005CuDC00" which is
1475 * a valid surrogate pair.
1476 *
1477 * @return a reference to this object.
1478 */
1479 public AbstractStringBuilder reverse() {
1480 byte[] val = this.value;
1481 int count = this.count;
1482 int coder = this.coder;
1483 int n = count - 1;
1484 if (COMPACT_STRINGS && coder == LATIN1) {
1485 for (int j = (n-1) >> 1; j >= 0; j--) {
1486 int k = n - j;
1487 byte cj = val[j];
1488 val[j] = val[k];
1489 val[k] = cj;
1490 }
1491 } else {
1492 checkOffset(count, val.length >> 1);
1493 boolean hasSurrogates = false;
1494 for (int j = (n-1) >> 1; j >= 0; j--) {
1495 int k = n - j;
1496 char cj = StringUTF16.getChar(val, j);
1497 char ck = StringUTF16.getChar(val, k);
1498 StringUTF16.putChar(val, j, ck);
1499 StringUTF16.putChar(val, k, cj);
1500 if (Character.isSurrogate(cj) ||
1501 Character.isSurrogate(ck)) {
1502 hasSurrogates = true;
1503 }
1504 }
1505 if (hasSurrogates) {
1506 reverseAllValidSurrogatePairs(val, count);
1507 }
1508 }
1509 return this;
1510 }
1511
1512 /** Outlined helper method for reverse() */
1513 private void reverseAllValidSurrogatePairs(byte[] val, int count) {
1514 for (int i = 0; i < count - 1; i++) {
1515 char c2 = StringUTF16.getChar(val, i);
1516 if (Character.isLowSurrogate(c2)) {
1517 char c1 = StringUTF16.getChar(val, i + 1);
1518 if (Character.isHighSurrogate(c1)) {
1519 StringUTF16.putChar(val, i++, c1);
1520 StringUTF16.putChar(val, i, c2);
1521 }
1522 }
1523 }
1524 }
1525
1526 /**
1527 * Returns a string representing the data in this sequence.
1528 * A new {@code String} object is allocated and initialized to
1529 * contain the character sequence currently represented by this
1530 * object. This {@code String} is then returned. Subsequent
1531 * changes to this sequence do not affect the contents of the
1532 * {@code String}.
1533 *
1534 * @return a string representation of this sequence of characters.
1535 */
1536 @Override
1537 public abstract String toString();
1538
1539 /**
1540 * {@inheritDoc}
1541 * @since 1.9
1542 */
1543 @Override
1544 public IntStream chars() {
1545 byte[] val = this.value; int count = this.count; byte coder = this.coder;
1546 checkOffset(count, val.length >> coder);
1547 // Reuse String-based spliterator. This requires a supplier to
1548 // capture the value and count when the terminal operation is executed
1549 return StreamSupport.intStream(
1550 () -> coder == LATIN1 ? new StringLatin1.CharsSpliterator(val, 0, count, 0)
1551 : new StringUTF16.CharsSpliterator(val, 0, count, 0),
1552 Spliterator.ORDERED | Spliterator.SIZED | Spliterator.SUBSIZED,
1553 false);
1554 }
1555
1556 /**
1557 * {@inheritDoc}
1558 * @since 1.9
1559 */
1560 @Override
1561 public IntStream codePoints() {
1562 byte[] val = this.value; int count = this.count; byte coder = this.coder;
1563 checkOffset(count, val.length >> coder);
1564 // Reuse String-based spliterator. This requires a supplier to
1565 // capture the value and count when the terminal operation is executed
1566 return StreamSupport.intStream(
1567 () -> coder == LATIN1 ? new StringLatin1.CharsSpliterator(val, 0, count, 0)
1568 : new StringUTF16.CodePointsSpliterator(val, 0, count, 0),
1569 Spliterator.ORDERED,
1570 false);
1571 }
1572
1573 /**
1574 * Needed by {@code String} for the contentEquals method.
1575 */
1576 final byte[] getValue() {
1577 return value;
1578 }
1579
1580 /*
1581 * Invoker guarantees it is in UTF16 (inflate itself for asb), if two
1582 * coders are different and the dstBegin has enough space
1583 *
1584 * @param dstBegin the char index, not offset of byte[]
1585 * @param coder the coder of dst[]
1586 */
1587 protected void getBytes(byte dst[], int dstBegin, byte coder) {
1588 if (this.coder == coder) {
1589 System.arraycopy(value, 0, dst, dstBegin << coder, count << coder);
1590 } else { // this.coder == LATIN && coder == UTF16
1591 StringLatin1.inflateSB(value, dst, dstBegin, count);
1592 }
1593 }
1594
1595 /* for readObject() */
1596 protected void initBytes(char[] value, int off, int len) {
1597 if (String.COMPACT_STRINGS) {
1598 this.value = StringUTF16.compress(value, off, len);
1599 if (this.value != null) {
1600 this.coder = LATIN1;
1601 return;
1602 }
1603 }
1604 this.coder = UTF16;
1605 this.value = StringUTF16.toBytes(value, off, len);
1606 }
1607
1608 final byte getCoder() {
1609 return COMPACT_STRINGS ? coder : UTF16;
1610 }
1611
1612 final boolean isLatin1() {
1613 return COMPACT_STRINGS && coder == LATIN1;
1614 }
1615
1616 private final void putCharsAt(int index, char[] s, int off, int end) {
1617 if (isLatin1()) {
1618 byte[] val = this.value;
1619 for (int i = off, j = index; i < end; i++) {
1620 char c = s[i];
1621 if (StringLatin1.canEncode(c)) {
1622 val[j++] = (byte)c;
1623 } else {
1624 inflate();
1625 StringUTF16.putCharsSB(this.value, j, s, i, end);
1626 return;
1627 }
1628 }
1629 } else {
1630 StringUTF16.putCharsSB(this.value, index, s, off, end);
1631 }
1632 }
1633
1634 private final void putCharsAt(int index, CharSequence s, int off, int end) {
1635 if (isLatin1()) {
1636 byte[] val = this.value;
1637 for (int i = off, j = index; i < end; i++) {
1638 char c = s.charAt(i);
1639 if (StringLatin1.canEncode(c)) {
1640 val[j++] = (byte)c;
1641 } else {
1642 inflate();
1643 StringUTF16.putCharsSB(this.value, j, s, i, end);
1644 return;
1645 }
1646 }
1647 } else {
1648 StringUTF16.putCharsSB(this.value, index, s, off, end);
1649 }
1650 }
1651
1652 private final void putStringAt(int index, String str) {
1653 if (getCoder() != str.coder()) {
1654 inflate();
1655 }
1656 byte[] val = this.value;
1657 byte coder = this.coder;
1658 checkOffset(index + str.length(), val.length >> coder);
1659 str.getBytes(val, index, coder);
1660 }
1661
1662 private final void appendChars(char[] s, int off, int end) {
1663 if (isLatin1()) {
1664 byte[] val = this.value;
1665 for (int i = off, j = count; i < end; i++) {
1666 char c = s[i];
1667 if (StringLatin1.canEncode(c)) {
1668 val[j++] = (byte)c;
1669 } else {
1670 count = j;
1671 inflate();
1672 StringUTF16.putCharsSB(this.value, j, s, i, end);
1673 count += end - i;
1674 return;
1675 }
1676 }
1677 } else {
1678 StringUTF16.putCharsSB(this.value, count, s, off, end);
1679 }
1680 count += end - off;
1681 }
1682
1683 private final void appendChars(CharSequence s, int off, int end) {
1684 if (isLatin1()) {
1685 byte[] val = this.value;
1686 for (int i = off, j = count; i < end; i++) {
1687 char c = s.charAt(i);
1688 if (StringLatin1.canEncode(c)) {
1689 val[j++] = (byte)c;
1690 } else {
1691 count = j;
1692 inflate();
1693 StringUTF16.putCharsSB(this.value, j, s, i, end);
1694 count += end - i;
1695 return;
1696 }
1697 }
1698 } else {
1699 StringUTF16.putCharsSB(this.value, count, s, off, end);
1700 }
1701 count += end - off;
1702 }
1703
1704 /* IndexOutOfBoundsException, if out of bounds */
1705 private static void checkRange(int start, int end, int len) {
1706 if (start < 0 || start > end || end > len) {
1707 throw new IndexOutOfBoundsException(
1708 "start " + start + ", end " + end + ", length " + len);
1709 }
1710 }
1711
1712 /* StringIndexOutOfBoundsException, if out of bounds */
1713 private static void checkRangeSIOOBE(int start, int end, int len) {
1714 if (start < 0 || start > end || end > len) {
1715 throw new StringIndexOutOfBoundsException(
1716 "start " + start + ", end " + end + ", length " + len);
1717 }
1718 }
1719 }