1 /* 2 * Copyright (C) 2014 The Android Open Source Project 3 * Copyright (c) 2000, 2020, Oracle and/or its affiliates. All rights reserved. 4 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 5 * 6 * This code is free software; you can redistribute it and/or modify it 7 * under the terms of the GNU General Public License version 2 only, as 8 * published by the Free Software Foundation. Oracle designates this 9 * particular file as subject to the "Classpath" exception as provided 10 * by Oracle in the LICENSE file that accompanied this code. 11 * 12 * This code is distributed in the hope that it will be useful, but WITHOUT 13 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 14 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 15 * version 2 for more details (a copy is included in the LICENSE file that 16 * accompanied this code). 17 * 18 * You should have received a copy of the GNU General Public License version 19 * 2 along with this work; if not, write to the Free Software Foundation, 20 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 21 * 22 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 23 * or visit www.oracle.com if you need additional information or have any 24 * questions. 25 */ 26 27 // -- This file was mechanically generated: Do not edit! -- // 28 29 package java.nio.charset; 30 31 import java.nio.Buffer; 32 import java.nio.ByteBuffer; 33 import java.nio.CharBuffer; 34 import java.nio.BufferOverflowException; 35 import java.nio.BufferUnderflowException; 36 import java.lang.ref.WeakReference; 37 import java.nio.charset.CoderMalfunctionError; // javadoc 38 import java.util.Arrays; 39 import java.util.Objects; 40 41 42 /** 43 * An engine that can transform a sequence of sixteen-bit Unicode characters into a sequence of 44 * bytes in a specific charset. 45 * 46 * <a id="steps"></a> 47 * 48 * <p> The input character sequence is provided in a character buffer or a series 49 * of such buffers. The output byte sequence is written to a byte buffer 50 * or a series of such buffers. An encoder should always be used by making 51 * the following sequence of method invocations, hereinafter referred to as an 52 * <i>encoding operation</i>: 53 * 54 * <ol> 55 * 56 * <li><p> Reset the encoder via the {@link #reset reset} method, unless it 57 * has not been used before; </p></li> 58 * 59 * <li><p> Invoke the {@link #encode encode} method zero or more times, as 60 * long as additional input may be available, passing {@code false} for the 61 * {@code endOfInput} argument and filling the input buffer and flushing the 62 * output buffer between invocations; </p></li> 63 * 64 * <li><p> Invoke the {@link #encode encode} method one final time, passing 65 * {@code true} for the {@code endOfInput} argument; and then </p></li> 66 * 67 * <li><p> Invoke the {@link #flush flush} method so that the encoder can 68 * flush any internal state to the output buffer. </p></li> 69 * 70 * </ol> 71 * 72 * Each invocation of the {@link #encode encode} method will encode as many 73 * characters as possible from the input buffer, writing the resulting bytes 74 * to the output buffer. The {@link #encode encode} method returns when more 75 * input is required, when there is not enough room in the output buffer, or 76 * when an encoding error has occurred. In each case a {@link CoderResult} 77 * object is returned to describe the reason for termination. An invoker can 78 * examine this object and fill the input buffer, flush the output buffer, or 79 * attempt to recover from an encoding error, as appropriate, and try again. 80 * 81 * <a id="ce"></a> 82 * 83 * <p> There are two general types of encoding errors. If the input character 84 * sequence is not a legal sixteen-bit Unicode sequence then the input is considered <i>malformed</i>. If 85 * the input character sequence is legal but cannot be mapped to a valid 86 * byte sequence in the given charset then an <i>unmappable character</i> has been encountered. 87 * 88 * <a id="cae"></a> 89 * 90 * <p> How an encoding error is handled depends upon the action requested for 91 * that type of error, which is described by an instance of the {@link 92 * CodingErrorAction} class. The possible error actions are to {@linkplain 93 * CodingErrorAction#IGNORE ignore} the erroneous input, {@linkplain 94 * CodingErrorAction#REPORT report} the error to the invoker via 95 * the returned {@link CoderResult} object, or {@linkplain CodingErrorAction#REPLACE 96 * replace} the erroneous input with the current value of the 97 * replacement byte array. The replacement 98 * 99 100 * is initially set to the encoder's default replacement, which often 101 * (but not always) has the initial value <code>{</code> <code>(byte)'?'</code> <code>}</code>; 102 103 104 105 106 * 107 * its value may be changed via the {@link #replaceWith(byte[]) 108 * replaceWith} method. 109 * 110 * <p> The default action for malformed-input and unmappable-character errors 111 * is to {@linkplain CodingErrorAction#REPORT report} them. The 112 * malformed-input error action may be changed via the {@link 113 * #onMalformedInput(CodingErrorAction) onMalformedInput} method; the 114 * unmappable-character action may be changed via the {@link 115 * #onUnmappableCharacter(CodingErrorAction) onUnmappableCharacter} method. 116 * 117 * <p> This class is designed to handle many of the details of the encoding 118 * process, including the implementation of error actions. An encoder for a 119 * specific charset, which is a concrete subclass of this class, need only 120 * implement the abstract {@link #encodeLoop encodeLoop} method, which 121 * encapsulates the basic encoding loop. A subclass that maintains internal 122 * state should, additionally, override the {@link #implFlush implFlush} and 123 * {@link #implReset implReset} methods. 124 * 125 * <p> Instances of this class are not safe for use by multiple concurrent 126 * threads. </p> 127 * 128 * 129 * @author Mark Reinhold 130 * @author JSR-51 Expert Group 131 * @since 1.4 132 * 133 * @see ByteBuffer 134 * @see CharBuffer 135 * @see Charset 136 * @see CharsetDecoder 137 */ 138 139 public abstract class CharsetEncoder { 140 141 private final Charset charset; 142 private final float averageBytesPerChar; 143 private final float maxBytesPerChar; 144 145 private byte[] replacement; 146 private CodingErrorAction malformedInputAction 147 = CodingErrorAction.REPORT; 148 private CodingErrorAction unmappableCharacterAction 149 = CodingErrorAction.REPORT; 150 151 // Internal states 152 // 153 private static final int ST_RESET = 0; 154 private static final int ST_CODING = 1; 155 private static final int ST_END = 2; 156 private static final int ST_FLUSHED = 3; 157 158 private int state = ST_RESET; 159 160 private static String stateNames[] 161 = { "RESET", "CODING", "CODING_END", "FLUSHED" }; 162 163 164 /** 165 * Initializes a new encoder. The new encoder will have the given 166 * bytes-per-char and replacement values. 167 * 168 * @param cs 169 * The charset that created this encoder 170 * 171 * @param averageBytesPerChar 172 * A positive float value indicating the expected number of 173 * bytes that will be produced for each input character 174 * 175 * @param maxBytesPerChar 176 * A positive float value indicating the maximum number of 177 * bytes that will be produced for each input character 178 * 179 * @param replacement 180 * The initial replacement; must not be {@code null}, must have 181 * non-zero length, must not be longer than maxBytesPerChar, 182 * and must be {@linkplain #isLegalReplacement legal} 183 * 184 * @throws IllegalArgumentException 185 * If the preconditions on the parameters do not hold 186 */ 187 protected CharsetEncoder(Charset cs, float averageBytesPerChar, float maxBytesPerChar, byte[] replacement)188 CharsetEncoder(Charset cs, 189 float averageBytesPerChar, 190 float maxBytesPerChar, 191 byte[] replacement) 192 { 193 194 // BEGIN Android-added: A hidden constructor for the CharsetEncoderICU subclass. 195 this(cs, averageBytesPerChar, maxBytesPerChar, replacement, false); 196 } 197 198 /** 199 * This constructor is for subclasses to specify whether {@code replacement} can be used as it 200 * is ("trusted"). If it is trusted, {@link #replaceWith(byte[])} and 201 * {@link #implReplaceWith(byte[])} will not be called. 202 * @hide 203 */ CharsetEncoder(Charset cs, float averageBytesPerChar, float maxBytesPerChar, byte[] replacement, boolean trusted)204 protected CharsetEncoder(Charset cs, float averageBytesPerChar, float maxBytesPerChar, byte[] replacement, 205 boolean trusted) 206 { 207 // END Android-added: A hidden constructor for the CharsetEncoderICU subclass. 208 209 this.charset = cs; 210 // Use !(a > 0.0f) rather than (a <= 0.0f) to exclude NaN values 211 if (!(averageBytesPerChar > 0.0f)) 212 throw new IllegalArgumentException("Non-positive " 213 + "averageBytesPerChar"); 214 // Use !(a > 0.0f) rather than (a <= 0.0f) to exclude NaN values 215 if (!(maxBytesPerChar > 0.0f)) 216 throw new IllegalArgumentException("Non-positive " 217 + "maxBytesPerChar"); 218 if (averageBytesPerChar > maxBytesPerChar) 219 throw new IllegalArgumentException("averageBytesPerChar" 220 + " exceeds " 221 + "maxBytesPerChar"); 222 this.replacement = replacement; 223 this.averageBytesPerChar = averageBytesPerChar; 224 this.maxBytesPerChar = maxBytesPerChar; 225 226 227 228 229 230 231 // BEGIN Android-changed: Avoid calling replaceWith() for trusted subclasses. 232 // replaceWith(replacement); 233 if (!trusted) { 234 replaceWith(replacement); 235 } 236 // END Android-changed: Avoid calling replaceWith() for trusted subclasses. 237 238 } 239 240 /** 241 * Initializes a new encoder. The new encoder will have the given 242 * bytes-per-char values and its replacement will be the 243 * byte array <code>{</code> <code>(byte)'?'</code> <code>}</code>. 244 * 245 * @param cs 246 * The charset that created this encoder 247 * 248 * @param averageBytesPerChar 249 * A positive float value indicating the expected number of 250 * bytes that will be produced for each input character 251 * 252 * @param maxBytesPerChar 253 * A positive float value indicating the maximum number of 254 * bytes that will be produced for each input character 255 * 256 * @throws IllegalArgumentException 257 * If the preconditions on the parameters do not hold 258 */ CharsetEncoder(Charset cs, float averageBytesPerChar, float maxBytesPerChar)259 protected CharsetEncoder(Charset cs, 260 float averageBytesPerChar, 261 float maxBytesPerChar) 262 { 263 this(cs, 264 averageBytesPerChar, maxBytesPerChar, 265 new byte[] { (byte)'?' }); 266 } 267 268 /** 269 * Returns the charset that created this encoder. 270 * 271 * @return This encoder's charset 272 */ charset()273 public final Charset charset() { 274 return charset; 275 } 276 277 /** 278 * Returns this encoder's replacement value. 279 * 280 * @return This encoder's current replacement, 281 * which is never {@code null} and is never empty 282 */ replacement()283 public final byte[] replacement() { 284 285 286 287 288 return Arrays.copyOf(replacement, replacement.length); 289 290 } 291 292 /** 293 * Changes this encoder's replacement value. 294 * 295 * <p> This method invokes the {@link #implReplaceWith implReplaceWith} 296 * method, passing the new replacement, after checking that the new 297 * replacement is acceptable. </p> 298 * 299 * @param newReplacement The new replacement; must not be 300 * {@code null}, must have non-zero length, 301 302 303 304 305 306 * must not be longer than the value returned by the 307 * {@link #maxBytesPerChar() maxBytesPerChar} method, and 308 * must be {@link #isLegalReplacement legal} 309 310 * 311 * @return This encoder 312 * 313 * @throws IllegalArgumentException 314 * If the preconditions on the parameter do not hold 315 */ replaceWith(byte[] newReplacement)316 public final CharsetEncoder replaceWith(byte[] newReplacement) { 317 if (newReplacement == null) 318 throw new IllegalArgumentException("Null replacement"); 319 int len = newReplacement.length; 320 if (len == 0) 321 throw new IllegalArgumentException("Empty replacement"); 322 if (len > maxBytesPerChar) 323 throw new IllegalArgumentException("Replacement too long"); 324 325 326 327 328 if (!isLegalReplacement(newReplacement)) 329 throw new IllegalArgumentException("Illegal replacement"); 330 this.replacement = Arrays.copyOf(newReplacement, newReplacement.length); 331 332 implReplaceWith(this.replacement); 333 return this; 334 } 335 336 /** 337 * Reports a change to this encoder's replacement value. 338 * 339 * <p> The default implementation of this method does nothing. This method 340 * should be overridden by encoders that require notification of changes to 341 * the replacement. </p> 342 * 343 * @param newReplacement The replacement value 344 */ implReplaceWith(byte[] newReplacement)345 protected void implReplaceWith(byte[] newReplacement) { 346 } 347 348 349 350 private WeakReference<CharsetDecoder> cachedDecoder = null; 351 352 /** 353 * Tells whether or not the given byte array is a legal replacement value 354 * for this encoder. 355 * 356 * <p> A replacement is legal if, and only if, it is a legal sequence of 357 * bytes in this encoder's charset; that is, it must be possible to decode 358 * the replacement into one or more sixteen-bit Unicode characters. 359 * 360 * <p> The default implementation of this method is not very efficient; it 361 * should generally be overridden to improve performance. </p> 362 * 363 * @param repl The byte array to be tested 364 * 365 * @return {@code true} if, and only if, the given byte array 366 * is a legal replacement value for this encoder 367 */ isLegalReplacement(byte[] repl)368 public boolean isLegalReplacement(byte[] repl) { 369 WeakReference<CharsetDecoder> wr = cachedDecoder; 370 CharsetDecoder dec = null; 371 if ((wr == null) || ((dec = wr.get()) == null)) { 372 dec = charset().newDecoder(); 373 dec.onMalformedInput(CodingErrorAction.REPORT); 374 dec.onUnmappableCharacter(CodingErrorAction.REPORT); 375 cachedDecoder = new WeakReference<CharsetDecoder>(dec); 376 } else { 377 dec.reset(); 378 } 379 ByteBuffer bb = ByteBuffer.wrap(repl); 380 CharBuffer cb = CharBuffer.allocate((int)(bb.remaining() 381 * dec.maxCharsPerByte())); 382 CoderResult cr = dec.decode(bb, cb, true); 383 return !cr.isError(); 384 } 385 386 387 388 /** 389 * Returns this encoder's current action for malformed-input errors. 390 * 391 * @return The current malformed-input action, which is never {@code null} 392 */ malformedInputAction()393 public CodingErrorAction malformedInputAction() { 394 return malformedInputAction; 395 } 396 397 /** 398 * Changes this encoder's action for malformed-input errors. 399 * 400 * <p> This method invokes the {@link #implOnMalformedInput 401 * implOnMalformedInput} method, passing the new action. </p> 402 * 403 * @param newAction The new action; must not be {@code null} 404 * 405 * @return This encoder 406 * 407 * @throws IllegalArgumentException 408 * If the precondition on the parameter does not hold 409 */ onMalformedInput(CodingErrorAction newAction)410 public final CharsetEncoder onMalformedInput(CodingErrorAction newAction) { 411 if (newAction == null) 412 throw new IllegalArgumentException("Null action"); 413 malformedInputAction = newAction; 414 implOnMalformedInput(newAction); 415 return this; 416 } 417 418 /** 419 * Reports a change to this encoder's malformed-input action. 420 * 421 * <p> The default implementation of this method does nothing. This method 422 * should be overridden by encoders that require notification of changes to 423 * the malformed-input action. </p> 424 * 425 * @param newAction The new action 426 */ implOnMalformedInput(CodingErrorAction newAction)427 protected void implOnMalformedInput(CodingErrorAction newAction) { } 428 429 /** 430 * Returns this encoder's current action for unmappable-character errors. 431 * 432 * @return The current unmappable-character action, which is never 433 * {@code null} 434 */ unmappableCharacterAction()435 public CodingErrorAction unmappableCharacterAction() { 436 return unmappableCharacterAction; 437 } 438 439 /** 440 * Changes this encoder's action for unmappable-character errors. 441 * 442 * <p> This method invokes the {@link #implOnUnmappableCharacter 443 * implOnUnmappableCharacter} method, passing the new action. </p> 444 * 445 * @param newAction The new action; must not be {@code null} 446 * 447 * @return This encoder 448 * 449 * @throws IllegalArgumentException 450 * If the precondition on the parameter does not hold 451 */ onUnmappableCharacter(CodingErrorAction newAction)452 public final CharsetEncoder onUnmappableCharacter(CodingErrorAction 453 newAction) 454 { 455 if (newAction == null) 456 throw new IllegalArgumentException("Null action"); 457 unmappableCharacterAction = newAction; 458 implOnUnmappableCharacter(newAction); 459 return this; 460 } 461 462 /** 463 * Reports a change to this encoder's unmappable-character action. 464 * 465 * <p> The default implementation of this method does nothing. This method 466 * should be overridden by encoders that require notification of changes to 467 * the unmappable-character action. </p> 468 * 469 * @param newAction The new action 470 */ implOnUnmappableCharacter(CodingErrorAction newAction)471 protected void implOnUnmappableCharacter(CodingErrorAction newAction) { } 472 473 /** 474 * Returns the average number of bytes that will be produced for each 475 * character of input. This heuristic value may be used to estimate the size 476 * of the output buffer required for a given input sequence. 477 * 478 * @return The average number of bytes produced 479 * per character of input 480 */ averageBytesPerChar()481 public final float averageBytesPerChar() { 482 return averageBytesPerChar; 483 } 484 485 /** 486 * Returns the maximum number of bytes that will be produced for each 487 * character of input. This value may be used to compute the worst-case size 488 * of the output buffer required for a given input sequence. This value 489 * accounts for any necessary content-independent prefix or suffix 490 491 * bytes, such as byte-order marks. 492 493 494 495 496 * 497 * @return The maximum number of bytes that will be produced per 498 * character of input 499 */ maxBytesPerChar()500 public final float maxBytesPerChar() { 501 return maxBytesPerChar; 502 } 503 504 // Android-changed: Keep compat behavior. Document NPE thrown for null arguments. 505 /** 506 * Encodes as many characters as possible from the given input buffer, 507 * writing the results to the given output buffer. 508 * 509 * <p> The buffers are read from, and written to, starting at their current 510 * positions. At most {@link Buffer#remaining in.remaining()} characters 511 * will be read and at most {@link Buffer#remaining out.remaining()} 512 * bytes will be written. The buffers' positions will be advanced to 513 * reflect the characters read and the bytes written, but their marks and 514 * limits will not be modified. 515 * 516 * <p> In addition to reading characters from the input buffer and writing 517 * bytes to the output buffer, this method returns a {@link CoderResult} 518 * object to describe its reason for termination: 519 * 520 * <ul> 521 * 522 * <li><p> {@link CoderResult#UNDERFLOW} indicates that as much of the 523 * input buffer as possible has been encoded. If there is no further 524 * input then the invoker can proceed to the next step of the 525 * <a href="#steps">encoding operation</a>. Otherwise this method 526 * should be invoked again with further input. </p></li> 527 * 528 * <li><p> {@link CoderResult#OVERFLOW} indicates that there is 529 * insufficient space in the output buffer to encode any more characters. 530 * This method should be invoked again with an output buffer that has 531 * more {@linkplain Buffer#remaining remaining} bytes. This is 532 * typically done by draining any encoded bytes from the output 533 * buffer. </p></li> 534 * 535 * <li><p> A {@linkplain CoderResult#malformedForLength 536 * malformed-input} result indicates that a malformed-input 537 * error has been detected. The malformed characters begin at the input 538 * buffer's (possibly incremented) position; the number of malformed 539 * characters may be determined by invoking the result object's {@link 540 * CoderResult#length() length} method. This case applies only if the 541 * {@linkplain #onMalformedInput malformed action} of this encoder 542 * is {@link CodingErrorAction#REPORT}; otherwise the malformed input 543 * will be ignored or replaced, as requested. </p></li> 544 * 545 * <li><p> An {@linkplain CoderResult#unmappableForLength 546 * unmappable-character} result indicates that an 547 * unmappable-character error has been detected. The characters that 548 * encode the unmappable character begin at the input buffer's (possibly 549 * incremented) position; the number of such characters may be determined 550 * by invoking the result object's {@link CoderResult#length() length} 551 * method. This case applies only if the {@linkplain #onUnmappableCharacter 552 * unmappable action} of this encoder is {@link 553 * CodingErrorAction#REPORT}; otherwise the unmappable character will be 554 * ignored or replaced, as requested. </p></li> 555 * 556 * </ul> 557 * 558 * In any case, if this method is to be reinvoked in the same encoding 559 * operation then care should be taken to preserve any characters remaining 560 * in the input buffer so that they are available to the next invocation. 561 * 562 * <p> The {@code endOfInput} parameter advises this method as to whether 563 * the invoker can provide further input beyond that contained in the given 564 * input buffer. If there is a possibility of providing additional input 565 * then the invoker should pass {@code false} for this parameter; if there 566 * is no possibility of providing further input then the invoker should 567 * pass {@code true}. It is not erroneous, and in fact it is quite 568 * common, to pass {@code false} in one invocation and later discover that 569 * no further input was actually available. It is critical, however, that 570 * the final invocation of this method in a sequence of invocations always 571 * pass {@code true} so that any remaining unencoded input will be treated 572 * as being malformed. 573 * 574 * <p> This method works by invoking the {@link #encodeLoop encodeLoop} 575 * method, interpreting its results, handling error conditions, and 576 * reinvoking it as necessary. </p> 577 * 578 * 579 * @param in 580 * The input character buffer 581 * 582 * @param out 583 * The output byte buffer 584 * 585 * @param endOfInput 586 * {@code true} if, and only if, the invoker can provide no 587 * additional input characters beyond those in the given buffer 588 * 589 * @return A coder-result object describing the reason for termination 590 * 591 * @throws IllegalStateException 592 * If an encoding operation is already in progress and the previous 593 * step was an invocation neither of the {@link #reset reset} 594 * method, nor of this method with a value of {@code false} for 595 * the {@code endOfInput} parameter, nor of this method with a 596 * value of {@code true} for the {@code endOfInput} parameter 597 * but a return value indicating an incomplete encoding operation 598 * 599 * @throws CoderMalfunctionError 600 * If an invocation of the encodeLoop method threw 601 * an unexpected exception 602 * 603 * @throws NullPointerException if input or output buffer is null 604 */ encode(CharBuffer in, ByteBuffer out, boolean endOfInput)605 public final CoderResult encode(CharBuffer in, ByteBuffer out, 606 boolean endOfInput) 607 { 608 // Android-added: Keep compat behavior. libcore throws NPE for null arguments. 609 Objects.requireNonNull(in, "in"); 610 Objects.requireNonNull(out, "out"); 611 612 int newState = endOfInput ? ST_END : ST_CODING; 613 if ((state != ST_RESET) && (state != ST_CODING) 614 && !(endOfInput && (state == ST_END))) 615 throwIllegalStateException(state, newState); 616 state = newState; 617 618 for (;;) { 619 620 CoderResult cr; 621 try { 622 cr = encodeLoop(in, out); 623 } catch (RuntimeException x) { 624 throw new CoderMalfunctionError(x); 625 } 626 627 if (cr.isOverflow()) 628 return cr; 629 630 if (cr.isUnderflow()) { 631 if (endOfInput && in.hasRemaining()) { 632 cr = CoderResult.malformedForLength(in.remaining()); 633 // Fall through to malformed-input case 634 } else { 635 return cr; 636 } 637 } 638 639 CodingErrorAction action = null; 640 if (cr.isMalformed()) 641 action = malformedInputAction; 642 else if (cr.isUnmappable()) 643 action = unmappableCharacterAction; 644 else 645 assert false : cr.toString(); 646 647 if (action == CodingErrorAction.REPORT) 648 return cr; 649 650 if (action == CodingErrorAction.REPLACE) { 651 if (out.remaining() < replacement.length) 652 return CoderResult.OVERFLOW; 653 out.put(replacement); 654 } 655 656 if ((action == CodingErrorAction.IGNORE) 657 || (action == CodingErrorAction.REPLACE)) { 658 // Skip erroneous input either way 659 in.position(in.position() + cr.length()); 660 continue; 661 } 662 663 assert false; 664 } 665 666 } 667 668 /** 669 * Flushes this encoder. 670 * 671 * <p> Some encoders maintain internal state and may need to write some 672 * final bytes to the output buffer once the overall input sequence has 673 * been read. 674 * 675 * <p> Any additional output is written to the output buffer beginning at 676 * its current position. At most {@link Buffer#remaining out.remaining()} 677 * bytes will be written. The buffer's position will be advanced 678 * appropriately, but its mark and limit will not be modified. 679 * 680 * <p> If this method completes successfully then it returns {@link 681 * CoderResult#UNDERFLOW}. If there is insufficient room in the output 682 * buffer then it returns {@link CoderResult#OVERFLOW}. If this happens 683 * then this method must be invoked again, with an output buffer that has 684 * more room, in order to complete the current <a href="#steps">encoding 685 * operation</a>. 686 * 687 * <p> If this encoder has already been flushed then invoking this method 688 * has no effect. 689 * 690 * <p> This method invokes the {@link #implFlush implFlush} method to 691 * perform the actual flushing operation. </p> 692 * 693 * @param out 694 * The output byte buffer 695 * 696 * @return A coder-result object, either {@link CoderResult#UNDERFLOW} or 697 * {@link CoderResult#OVERFLOW} 698 * 699 * @throws IllegalStateException 700 * If the previous step of the current encoding operation was an 701 * invocation neither of the {@link #flush flush} method nor of 702 * the three-argument {@link 703 * #encode(CharBuffer,ByteBuffer,boolean) encode} method 704 * with a value of {@code true} for the {@code endOfInput} 705 * parameter 706 */ flush(ByteBuffer out)707 public final CoderResult flush(ByteBuffer out) { 708 if (state == ST_END) { 709 CoderResult cr = implFlush(out); 710 if (cr.isUnderflow()) 711 state = ST_FLUSHED; 712 return cr; 713 } 714 715 if (state != ST_FLUSHED) 716 throwIllegalStateException(state, ST_FLUSHED); 717 718 return CoderResult.UNDERFLOW; // Already flushed 719 } 720 721 /** 722 * Flushes this encoder. 723 * 724 * <p> The default implementation of this method does nothing, and always 725 * returns {@link CoderResult#UNDERFLOW}. This method should be overridden 726 * by encoders that may need to write final bytes to the output buffer 727 * once the entire input sequence has been read. </p> 728 * 729 * @param out 730 * The output byte buffer 731 * 732 * @return A coder-result object, either {@link CoderResult#UNDERFLOW} or 733 * {@link CoderResult#OVERFLOW} 734 */ implFlush(ByteBuffer out)735 protected CoderResult implFlush(ByteBuffer out) { 736 return CoderResult.UNDERFLOW; 737 } 738 739 /** 740 * Resets this encoder, clearing any internal state. 741 * 742 * <p> This method resets charset-independent state and also invokes the 743 * {@link #implReset() implReset} method in order to perform any 744 * charset-specific reset actions. </p> 745 * 746 * @return This encoder 747 * 748 */ reset()749 public final CharsetEncoder reset() { 750 implReset(); 751 state = ST_RESET; 752 return this; 753 } 754 755 /** 756 * Resets this encoder, clearing any charset-specific internal state. 757 * 758 * <p> The default implementation of this method does nothing. This method 759 * should be overridden by encoders that maintain internal state. </p> 760 */ implReset()761 protected void implReset() { } 762 763 /** 764 * Encodes one or more characters into one or more bytes. 765 * 766 * <p> This method encapsulates the basic encoding loop, encoding as many 767 * characters as possible until it either runs out of input, runs out of room 768 * in the output buffer, or encounters an encoding error. This method is 769 * invoked by the {@link #encode encode} method, which handles result 770 * interpretation and error recovery. 771 * 772 * <p> The buffers are read from, and written to, starting at their current 773 * positions. At most {@link Buffer#remaining in.remaining()} characters 774 * will be read, and at most {@link Buffer#remaining out.remaining()} 775 * bytes will be written. The buffers' positions will be advanced to 776 * reflect the characters read and the bytes written, but their marks and 777 * limits will not be modified. 778 * 779 * <p> This method returns a {@link CoderResult} object to describe its 780 * reason for termination, in the same manner as the {@link #encode encode} 781 * method. Most implementations of this method will handle encoding errors 782 * by returning an appropriate result object for interpretation by the 783 * {@link #encode encode} method. An optimized implementation may instead 784 * examine the relevant error action and implement that action itself. 785 * 786 * <p> An implementation of this method may perform arbitrary lookahead by 787 * returning {@link CoderResult#UNDERFLOW} until it receives sufficient 788 * input. </p> 789 * 790 * @param in 791 * The input character buffer 792 * 793 * @param out 794 * The output byte buffer 795 * 796 * @return A coder-result object describing the reason for termination 797 */ encodeLoop(CharBuffer in, ByteBuffer out)798 protected abstract CoderResult encodeLoop(CharBuffer in, 799 ByteBuffer out); 800 801 // Android-changed: Document CoderMalfunctionError and NPE thrown for null input buffer. 802 /** 803 * Convenience method that encodes the remaining content of a single input 804 * character buffer into a newly-allocated byte buffer. 805 * 806 * <p> This method implements an entire <a href="#steps">encoding 807 * operation</a>; that is, it resets this encoder, then it encodes the 808 * characters in the given character buffer, and finally it flushes this 809 * encoder. This method should therefore not be invoked if an encoding 810 * operation is already in progress. </p> 811 * 812 * @param in 813 * The input character buffer 814 * 815 * @return A newly-allocated byte buffer containing the result of the 816 * encoding operation. The buffer's position will be zero and its 817 * limit will follow the last byte written. 818 * 819 * @throws IllegalStateException 820 * If an encoding operation is already in progress 821 * 822 * @throws MalformedInputException 823 * If the character sequence starting at the input buffer's current 824 * position is not a legal sixteen-bit Unicode sequence and the current malformed-input action 825 * is {@link CodingErrorAction#REPORT} 826 * 827 * @throws UnmappableCharacterException 828 * If the character sequence starting at the input buffer's current 829 * position cannot be mapped to an equivalent byte sequence and 830 * the current unmappable-character action is {@link 831 * CodingErrorAction#REPORT} 832 * 833 * @throws CoderMalfunctionError 834 * If an invocation of the encodeLoop method threw 835 * an unexpected exception 836 * 837 * @throws NullPointerException if input buffer is null 838 */ encode(CharBuffer in)839 public final ByteBuffer encode(CharBuffer in) 840 throws CharacterCodingException 841 { 842 int n = (int)(in.remaining() * averageBytesPerChar()); 843 ByteBuffer out = ByteBuffer.allocate(n); 844 845 if ((n == 0) && (in.remaining() == 0)) 846 return out; 847 reset(); 848 for (;;) { 849 CoderResult cr = in.hasRemaining() ? 850 encode(in, out, true) : CoderResult.UNDERFLOW; 851 if (cr.isUnderflow()) 852 cr = flush(out); 853 854 if (cr.isUnderflow()) 855 break; 856 if (cr.isOverflow()) { 857 n = 2*n + 1; // Ensure progress; n might be 0! 858 ByteBuffer o = ByteBuffer.allocate(n); 859 out.flip(); 860 o.put(out); 861 out = o; 862 continue; 863 } 864 cr.throwException(); 865 } 866 out.flip(); 867 return out; 868 } 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 canEncode(CharBuffer cb)948 private boolean canEncode(CharBuffer cb) { 949 if (state == ST_FLUSHED) 950 reset(); 951 else if (state != ST_RESET) 952 throwIllegalStateException(state, ST_CODING); 953 954 // BEGIN Android-added: Fast path handling for empty buffers. 955 // Empty buffers can always be "encoded". 956 if (!cb.hasRemaining()) { 957 return true; 958 } 959 // END Android-added: Fast path handling for empty buffers. 960 961 CodingErrorAction ma = malformedInputAction(); 962 CodingErrorAction ua = unmappableCharacterAction(); 963 try { 964 onMalformedInput(CodingErrorAction.REPORT); 965 onUnmappableCharacter(CodingErrorAction.REPORT); 966 encode(cb); 967 } catch (CharacterCodingException x) { 968 return false; 969 } finally { 970 onMalformedInput(ma); 971 onUnmappableCharacter(ua); 972 reset(); 973 } 974 return true; 975 } 976 977 /** 978 * Tells whether or not this encoder can encode the given character. 979 * 980 * <p> This method returns {@code false} if the given character is a 981 * surrogate character; such characters can be interpreted only when they 982 * are members of a pair consisting of a high surrogate followed by a low 983 * surrogate. The {@link #canEncode(java.lang.CharSequence) 984 * canEncode(CharSequence)} method may be used to test whether or not a 985 * character sequence can be encoded. 986 * 987 * <p> This method may modify this encoder's state; it should therefore not 988 * be invoked if an <a href="#steps">encoding operation</a> is already in 989 * progress. 990 * 991 * <p> The default implementation of this method is not very efficient; it 992 * should generally be overridden to improve performance. </p> 993 * 994 * @param c 995 * The given character 996 * 997 * @return {@code true} if, and only if, this encoder can encode 998 * the given character 999 * 1000 * @throws IllegalStateException 1001 * If an encoding operation is already in progress 1002 */ canEncode(char c)1003 public boolean canEncode(char c) { 1004 CharBuffer cb = CharBuffer.allocate(1); 1005 cb.put(c); 1006 cb.flip(); 1007 return canEncode(cb); 1008 } 1009 1010 /** 1011 * Tells whether or not this encoder can encode the given character 1012 * sequence. 1013 * 1014 * <p> If this method returns {@code false} for a particular character 1015 * sequence then more information about why the sequence cannot be encoded 1016 * may be obtained by performing a full <a href="#steps">encoding 1017 * operation</a>. 1018 * 1019 * <p> This method may modify this encoder's state; it should therefore not 1020 * be invoked if an encoding operation is already in progress. 1021 * 1022 * <p> The default implementation of this method is not very efficient; it 1023 * should generally be overridden to improve performance. </p> 1024 * 1025 * @param cs 1026 * The given character sequence 1027 * 1028 * @return {@code true} if, and only if, this encoder can encode 1029 * the given character without throwing any exceptions and without 1030 * performing any replacements 1031 * 1032 * @throws IllegalStateException 1033 * If an encoding operation is already in progress 1034 */ canEncode(CharSequence cs)1035 public boolean canEncode(CharSequence cs) { 1036 CharBuffer cb; 1037 if (cs instanceof CharBuffer) 1038 cb = ((CharBuffer)cs).duplicate(); 1039 else 1040 // Android-removed: An unnecessary call to toString(). 1041 // cb = CharBuffer.wrap(cs.toString()); 1042 cb = CharBuffer.wrap(cs); 1043 1044 return canEncode(cb); 1045 } 1046 1047 1048 1049 throwIllegalStateException(int from, int to)1050 private void throwIllegalStateException(int from, int to) { 1051 throw new IllegalStateException("Current state = " + stateNames[from] 1052 + ", new state = " + stateNames[to]); 1053 } 1054 1055 } 1056