沉淀再出发:java中的equals()辨析
一、前言
关于java中的equals，我们可能非常奇怪，在Object中定义了这个函数，其他的很多类中都重载了它，导致了我们对于辨析其中的内涵有了混淆，再加上和“==”的比较，就显得更加的复杂了。
二、java中的equals()
2.1、Object.java中的equals()
让我们来看一下Object.java中的equals()。
首先是Object的定义：
1 /* 2 * Copyright (c) 1994, 2012, Oracle and/or its affiliates. All rights reserved. 3 * ORACLE PROPRIETARY/CONFIDENTIAL. Use is subject to license terms. 4 * 5 * 6 * 7 * 8 * 9 * 10 * 11 * 12 * 13 * 14 * 15 * 16 * 17 * 18 * 19 * 20 * 21 * 22 * 23 * 24 */ 25 26 package java.lang; 27 28 /** 29 * Class {@code Object} is the root of the class hierarchy. 30 * Every class has {@code Object} as a superclass. All objects, 31 * including arrays, implement the methods of this class. 32 * 33 * @author unascribed 34 * @see java.lang.Class 35 * @since JDK1.0 36 */ 37 public class Object { 38 39 private static native void registerNatives(); 40 static { 41 registerNatives(); 42 } 43 44 /** 45 * Returns the runtime class of this {@code Object}. The returned 46 * {@code Class} object is the object that is locked by {@code 47 * static synchronized} methods of the represented class. 48 * 49 * <p><b>The actual result type is {@code Class<? extends |X|>} 50 * where {@code |X|} is the erasure of the static type of the 51 * expression on which {@code getClass} is called.</b> For 52 * example, no cast is required in this code fragment:</p> 53 * 54 * <p> 55 * {@code Number n = 0; }<br> 56 * {@code Class<? extends Number> c = n.getClass(); } 57 * </p> 58 * 59 * @return The {@code Class} object that represents the runtime 60 * class of this object. 61 * @jls 15.8.2 Class Literals 62 */ 63 public final native Class<?> getClass(); 64 65 /** 66 * Returns a hash code value for the object. This method is 67 * supported for the benefit of hash tables such as those provided by 68 * {@link java.util.HashMap}. 69 * <p> 70 * The general contract of {@code hashCode} is: 71 * <ul> 72 * <li>Whenever it is invoked on the same object more than once during 73 * an execution of a Java application, the {@code hashCode} method 74 * must consistently return the same integer, provided no information 75 * used in {@code equals} comparisons on the object is modified. 76 * This integer need not remain consistent from one execution of an 77 * application to another execution of the same application. 78 * <li>If two objects are equal according to the {@code equals(Object)} 79 * method, then calling the {@code hashCode} method on each of 80 * the two objects must produce the same integer result. 81 * <li>It is <em>not</em> required that if two objects are unequal 82 * according to the {@link java.lang.Object#equals(java.lang.Object)} 83 * method, then calling the {@code hashCode} method on each of the 84 * two objects must produce distinct integer results. However, the 85 * programmer should be aware that producing distinct integer results 86 * for unequal objects may improve the performance of hash tables. 87 * </ul> 88 * <p> 89 * As much as is reasonably practical, the hashCode method defined by 90 * class {@code Object} does return distinct integers for distinct 91 * objects. (This is typically implemented by converting the internal 92 * address of the object into an integer, but this implementation 93 * technique is not required by the 94 * Java™ programming language.) 95 * 96 * @return a hash code value for this object. 97 * @see java.lang.Object#equals(java.lang.Object) 98 * @see java.lang.System#identityHashCode 99 */ 100 public native int hashCode(); 101 102 /** 103 * Indicates whether some other object is "equal to" this one. 104 * <p> 105 * The {@code equals} method implements an equivalence relation 106 * on non-null object references: 107 * <ul> 108 * <li>It is <i>reflexive</i>: for any non-null reference value 109 * {@code x}, {@code x.equals(x)} should return 110 * {@code true}. 111 * <li>It is <i>symmetric</i>: for any non-null reference values 112 * {@code x} and {@code y}, {@code x.equals(y)} 113 * should return {@code true} if and only if 114 * {@code y.equals(x)} returns {@code true}. 115 * <li>It is <i>transitive</i>: for any non-null reference values 116 * {@code x}, {@code y}, and {@code z}, if 117 * {@code x.equals(y)} returns {@code true} and 118 * {@code y.equals(z)} returns {@code true}, then 119 * {@code x.equals(z)} should return {@code true}. 120 * <li>It is <i>consistent</i>: for any non-null reference values 121 * {@code x} and {@code y}, multiple invocations of 122 * {@code x.equals(y)} consistently return {@code true} 123 * or consistently return {@code false}, provided no 124 * information used in {@code equals} comparisons on the 125 * objects is modified. 126 * <li>For any non-null reference value {@code x}, 127 * {@code x.equals(null)} should return {@code false}. 128 * </ul> 129 * <p> 130 * The {@code equals} method for class {@code Object} implements 131 * the most discriminating possible equivalence relation on objects; 132 * that is, for any non-null reference values {@code x} and 133 * {@code y}, this method returns {@code true} if and only 134 * if {@code x} and {@code y} refer to the same object 135 * ({@code x == y} has the value {@code true}). 136 * <p> 137 * Note that it is generally necessary to override the {@code hashCode} 138 * method whenever this method is overridden, so as to maintain the 139 * general contract for the {@code hashCode} method, which states 140 * that equal objects must have equal hash codes. 141 * 142 * @param obj the reference object with which to compare. 143 * @return {@code true} if this object is the same as the obj 144 * argument; {@code false} otherwise. 145 * @see #hashCode() 146 * @see java.util.HashMap 147 */ 148 public boolean equals(Object obj) { 149 return (this == obj); 150 } 151 152 /** 153 * Creates and returns a copy of this object. The precise meaning 154 * of "copy" may depend on the class of the object. The general 155 * intent is that, for any object {@code x}, the expression: 156 * <blockquote> 157 * <pre> 158 * x.clone() != x</pre></blockquote> 159 * will be true, and that the expression: 160 * <blockquote> 161 * <pre> 162 * x.clone().getClass() == x.getClass()</pre></blockquote> 163 * will be {@code true}, but these are not absolute requirements. 164 * While it is typically the case that: 165 * <blockquote> 166 * <pre> 167 * x.clone().equals(x)</pre></blockquote> 168 * will be {@code true}, this is not an absolute requirement. 169 * <p> 170 * By convention, the returned object should be obtained by calling 171 * {@code super.clone}. If a class and all of its superclasses (except 172 * {@code Object}) obey this convention, it will be the case that 173 * {@code x.clone().getClass() == x.getClass()}. 174 * <p> 175 * By convention, the object returned by this method should be independent 176 * of this object (which is being cloned). To achieve this independence, 177 * it may be necessary to modify one or more fields of the object returned 178 * by {@code super.clone} before returning it. Typically, this means 179 * copying any mutable objects that comprise the internal "deep structure" 180 * of the object being cloned and replacing the references to these 181 * objects with references to the copies. If a class contains only 182 * primitive fields or references to immutable objects, then it is usually 183 * the case that no fields in the object returned by {@code super.clone} 184 * need to be modified. 185 * <p> 186 * The method {@code clone} for class {@code Object} performs a 187 * specific cloning operation. First, if the class of this object does 188 * not implement the interface {@code Cloneable}, then a 189 * {@code CloneNotSupportedException} is thrown. Note that all arrays 190 * are considered to implement the interface {@code Cloneable} and that 191 * the return type of the {@code clone} method of an array type {@code T[]} 192 * is {@code T[]} where T is any reference or primitive type. 193 * Otherwise, this method creates a new instance of the class of this 194 * object and initializes all its fields with exactly the contents of 195 * the corresponding fields of this object, as if by assignment; the 196 * contents of the fields are not themselves cloned. Thus, this method 197 * performs a "shallow copy" of this object, not a "deep copy" operation. 198 * <p> 199 * The class {@code Object} does not itself implement the interface 200 * {@code Cloneable}, so calling the {@code clone} method on an object 201 * whose class is {@code Object} will result in throwing an 202 * exception at run time. 203 * 204 * @return a clone of this instance. 205 * @throws CloneNotSupportedException if the object's class does not 206 * support the {@code Cloneable} interface. Subclasses 207 * that override the {@code clone} method can also 208 * throw this exception to indicate that an instance cannot 209 * be cloned. 210 * @see java.lang.Cloneable 211 */ 212 protected native Object clone() throws CloneNotSupportedException; 213 214 /** 215 * Returns a string representation of the object. In general, the 216 * {@code toString} method returns a string that 217 * "textually represents" this object. The result should 218 * be a concise but informative representation that is easy for a 219 * person to read. 220 * It is recommended that all subclasses override this method. 221 * <p> 222 * The {@code toString} method for class {@code Object} 223 * returns a string consisting of the name of the class of which the 224 * object is an instance, the at-sign character `{@code @}', and 225 * the unsigned hexadecimal representation of the hash code of the 226 * object. In other words, this method returns a string equal to the 227 * value of: 228 * <blockquote> 229 * <pre> 230 * getClass().getName() + '@' + Integer.toHexString(hashCode()) 231 * </pre></blockquote> 232 * 233 * @return a string representation of the object. 234 */ 235 public String toString() { 236 return getClass().getName() + "@" + Integer.toHexString(hashCode()); 237 } 238 239 /** 240 * Wakes up a single thread that is waiting on this object's 241 * monitor. If any threads are waiting on this object, one of them 242 * is chosen to be awakened. The choice is arbitrary and occurs at 243 * the discretion of the implementation. A thread waits on an object's 244 * monitor by calling one of the {@code wait} methods. 245 * <p> 246 * The awakened thread will not be able to proceed until the current 247 * thread relinquishes the lock on this object. The awakened thread will 248 * compete in the usual manner with any other threads that might be 249 * actively competing to synchronize on this object; for example, the 250 * awakened thread enjoys no reliable privilege or disadvantage in being 251 * the next thread to lock this object. 252 * <p> 253 * This method should only be called by a thread that is the owner 254 * of this object's monitor. A thread becomes the owner of the 255 * object's monitor in one of three ways: 256 * <ul> 257 * <li>By executing a synchronized instance method of that object. 258 * <li>By executing the body of a {@code synchronized} statement 259 * that synchronizes on the object. 260 * <li>For objects of type {@code Class,} by executing a 261 * synchronized static method of that class. 262 * </ul> 263 * <p> 264 * Only one thread at a time can own an object's monitor. 265 * 266 * @throws IllegalMonitorStateException if the current thread is not 267 * the owner of this object's monitor. 268 * @see java.lang.Object#notifyAll() 269 * @see java.lang.Object#wait() 270 */ 271 public final native void notify(); 272 273 /** 274 * Wakes up all threads that are waiting on this object's monitor. A 275 * thread waits on an object's monitor by calling one of the 276 * {@code wait} methods. 277 * <p> 278 * The awakened threads will not be able to proceed until the current 279 * thread relinquishes the lock on this object. The awakened threads 280 * will compete in the usual manner with any other threads that might 281 * be actively competing to synchronize on this object; for example, 282 * the awakened threads enjoy no reliable privilege or disadvantage in 283 * being the next thread to lock this object. 284 * <p> 285 * This method should only be called by a thread that is the owner 286 * of this object's monitor. See the {@code notify} method for a 287 * description of the ways in which a thread can become the owner of 288 * a monitor. 289 * 290 * @throws IllegalMonitorStateException if the current thread is not 291 * the owner of this object's monitor. 292 * @see java.lang.Object#notify() 293 * @see java.lang.Object#wait() 294 */ 295 public final native void notifyAll(); 296 297 /** 298 * Causes the current thread to wait until either another thread invokes the 299 * {@link java.lang.Object#notify()} method or the 300 * {@link java.lang.Object#notifyAll()} method for this object, or a 301 * specified amount of time has elapsed. 302 * <p> 303 * The current thread must own this object's monitor. 304 * <p> 305 * This method causes the current thread (call it <var>T</var>) to 306 * place itself in the wait set for this object and then to relinquish 307 * any and all synchronization claims on this object. Thread <var>T</var> 308 * becomes disabled for thread scheduling purposes and lies dormant 309 * until one of four things happens: 310 * <ul> 311 * <li>Some other thread invokes the {@code notify} method for this 312 * object and thread <var>T</var> happens to be arbitrarily chosen as 313 * the thread to be awakened. 314 * <li>Some other thread invokes the {@code notifyAll} method for this 315 * object. 316 * <li>Some other thread {@linkplain Thread#interrupt() interrupts} 317 * thread <var>T</var>. 318 * <li>The specified amount of real time has elapsed, more or less. If 319 * {@code timeout} is zero, however, then real time is not taken into 320 * consideration and the thread simply waits until notified. 321 * </ul> 322 * The thread <var>T</var> is then removed from the wait set for this 323 * object and re-enabled for thread scheduling. It then competes in the 324 * usual manner with other threads for the right to synchronize on the 325 * object; once it has gained control of the object, all its 326 * synchronization claims on the object are restored to the status quo 327 * ante - that is, to the situation as of the time that the {@code wait} 328 * method was invoked. Thread <var>T</var> then returns from the 329 * invocation of the {@code wait} method. Thus, on return from the 330 * {@code wait} method, the synchronization state of the object and of 331 * thread {@code T} is exactly as it was when the {@code wait} method 332 * was invoked. 333 * <p> 334 * A thread can also wake up without being notified, interrupted, or 335 * timing out, a so-called <i>spurious wakeup</i>. While this will rarely 336 * occur in practice, applications must guard against it by testing for 337 * the condition that should have caused the thread to be awakened, and 338 * continuing to wait if the condition is not satisfied. In other words, 339 * waits should always occur in loops, like this one: 340 * <pre> 341 * synchronized (obj) { 342 * while (<condition does not hold>) 343 * obj.wait(timeout); 344 * ... // Perform action appropriate to condition 345 * } 346 * </pre> 347 * (For more information on this topic, see Section 3.2.3 in Doug Lea's 348 * "Concurrent Programming in Java (Second Edition)" (Addison-Wesley, 349 * 2000), or Item 50 in Joshua Bloch's "Effective Java Programming 350 * Language Guide" (Addison-Wesley, 2001). 351 * 352 * <p>If the current thread is {@linkplain java.lang.Thread#interrupt() 353 * interrupted} by any thread before or while it is waiting, then an 354 * {@code InterruptedException} is thrown. This exception is not 355 * thrown until the lock status of this object has been restored as 356 * described above. 357 * 358 * <p> 359 * Note that the {@code wait} method, as it places the current thread 360 * into the wait set for this object, unlocks only this object; any 361 * other objects on which the current thread may be synchronized remain 362 * locked while the thread waits. 363 * <p> 364 * This method should only be called by a thread that is the owner 365 * of this object's monitor. See the {@code notify} method for a 366 * description of the ways in which a thread can become the owner of 367 * a monitor. 368 * 369 * @param timeout the maximum time to wait in milliseconds. 370 * @throws IllegalArgumentException if the value of timeout is 371 * negative. 372 * @throws IllegalMonitorStateException if the current thread is not 373 * the owner of the object's monitor. 374 * @throws InterruptedException if any thread interrupted the 375 * current thread before or while the current thread 376 * was waiting for a notification. The <i>interrupted 377 * status</i> of the current thread is cleared when 378 * this exception is thrown. 379 * @see java.lang.Object#notify() 380 * @see java.lang.Object#notifyAll() 381 */ 382 public final native void wait(long timeout) throws InterruptedException; 383 384 /** 385 * Causes the current thread to wait until another thread invokes the 386 * {@link java.lang.Object#notify()} method or the 387 * {@link java.lang.Object#notifyAll()} method for this object, or 388 * some other thread interrupts the current thread, or a certain 389 * amount of real time has elapsed. 390 * <p> 391 * This method is similar to the {@code wait} method of one 392 * argument, but it allows finer control over the amount of time to 393 * wait for a notification before giving up. The amount of real time, 394 * measured in nanoseconds, is given by: 395 * <blockquote> 396 * <pre> 397 * 1000000*timeout+nanos</pre></blockquote> 398 * <p> 399 * In all other respects, this method does the same thing as the 400 * method {@link #wait(long)} of one argument. In particular, 401 * {@code wait(0, 0)} means the same thing as {@code wait(0)}. 402 * <p> 403 * The current thread must own this object's monitor. The thread 404 * releases ownership of this monitor and waits until either of the 405 * following two conditions has occurred: 406 * <ul> 407 * <li>Another thread notifies threads waiting on this object's monitor 408 * to wake up either through a call to the {@code notify} method 409 * or the {@code notifyAll} method. 410 * <li>The timeout period, specified by {@code timeout} 411 * milliseconds plus {@code nanos} nanoseconds arguments, has 412 * elapsed. 413 * </ul> 414 * <p> 415 * The thread then waits until it can re-obtain ownership of the 416 * monitor and resumes execution. 417 * <p> 418 * As in the one argument version, interrupts and spurious wakeups are 419 * possible, and this method should always be used in a loop: 420 * <pre> 421 * synchronized (obj) { 422 * while (<condition does not hold>) 423 * obj.wait(timeout, nanos); 424 * ... // Perform action appropriate to condition 425 * } 426 * </pre> 427 * This method should only be called by a thread that is the owner 428 * of this object's monitor. See the {@code notify} method for a 429 * description of the ways in which a thread can become the owner of 430 * a monitor. 431 * 432 * @param timeout the maximum time to wait in milliseconds. 433 * @param nanos additional time, in nanoseconds range 434 * 0-999999. 435 * @throws IllegalArgumentException if the value of timeout is 436 * negative or the value of nanos is 437 * not in the range 0-999999. 438 * @throws IllegalMonitorStateException if the current thread is not 439 * the owner of this object's monitor. 440 * @throws InterruptedException if any thread interrupted the 441 * current thread before or while the current thread 442 * was waiting for a notification. The <i>interrupted 443 * status</i> of the current thread is cleared when 444 * this exception is thrown. 445 */ 446 public final void wait(long timeout, int nanos) throws InterruptedException { 447 if (timeout < 0) { 448 throw new IllegalArgumentException("timeout value is negative"); 449 } 450 451 if (nanos < 0 || nanos > 999999) { 452 throw new IllegalArgumentException( 453 "nanosecond timeout value out of range"); 454 } 455 456 if (nanos > 0) { 457 timeout++; 458 } 459 460 wait(timeout); 461 } 462 463 /** 464 * Causes the current thread to wait until another thread invokes the 465 * {@link java.lang.Object#notify()} method or the 466 * {@link java.lang.Object#notifyAll()} method for this object. 467 * In other words, this method behaves exactly as if it simply 468 * performs the call {@code wait(0)}. 469 * <p> 470 * The current thread must own this object's monitor. The thread 471 * releases ownership of this monitor and waits until another thread 472 * notifies threads waiting on this object's monitor to wake up 473 * either through a call to the {@code notify} method or the 474 * {@code notifyAll} method. The thread then waits until it can 475 * re-obtain ownership of the monitor and resumes execution. 476 * <p> 477 * As in the one argument version, interrupts and spurious wakeups are 478 * possible, and this method should always be used in a loop: 479 * <pre> 480 * synchronized (obj) { 481 * while (<condition does not hold>) 482 * obj.wait(); 483 * ... // Perform action appropriate to condition 484 * } 485 * </pre> 486 * This method should only be called by a thread that is the owner 487 * of this object's monitor. See the {@code notify} method for a 488 * description of the ways in which a thread can become the owner of 489 * a monitor. 490 * 491 * @throws IllegalMonitorStateException if the current thread is not 492 * the owner of the object's monitor. 493 * @throws InterruptedException if any thread interrupted the 494 * current thread before or while the current thread 495 * was waiting for a notification. The <i>interrupted 496 * status</i> of the current thread is cleared when 497 * this exception is thrown. 498 * @see java.lang.Object#notify() 499 * @see java.lang.Object#notifyAll() 500 */ 501 public final void wait() throws InterruptedException { 502 wait(0); 503 } 504 505 /** 506 * Called by the garbage collector on an object when garbage collection 507 * determines that there are no more references to the object. 508 * A subclass overrides the {@code finalize} method to dispose of 509 * system resources or to perform other cleanup. 510 * <p> 511 * The general contract of {@code finalize} is that it is invoked 512 * if and when the Java™ virtual 513 * machine has determined that there is no longer any 514 * means by which this object can be accessed by any thread that has 515 * not yet died, except as a result of an action taken by the 516 * finalization of some other object or class which is ready to be 517 * finalized. The {@code finalize} method may take any action, including 518 * making this object available again to other threads; the usual purpose 519 * of {@code finalize}, however, is to perform cleanup actions before 520 * the object is irrevocably discarded. For example, the finalize method 521 * for an object that represents an input/output connection might perform 522 * explicit I/O transactions to break the connection before the object is 523 * permanently discarded. 524 * <p> 525 * The {@code finalize} method of class {@code Object} performs no 526 * special action; it simply returns normally. Subclasses of 527 * {@code Object} may override this definition. 528 * <p> 529 * The Java programming language does not guarantee which thread will 530 * invoke the {@code finalize} method for any given object. It is 531 * guaranteed, however, that the thread that invokes finalize will not 532 * be holding any user-visible synchronization locks when finalize is 533 * invoked. If an uncaught exception is thrown by the finalize method, 534 * the exception is ignored and finalization of that object terminates. 535 * <p> 536 * After the {@code finalize} method has been invoked for an object, no 537 * further action is taken until the Java virtual machine has again 538 * determined that there is no longer any means by which this object can 539 * be accessed by any thread that has not yet died, including possible 540 * actions by other objects or classes which are ready to be finalized, 541 * at which point the object may be discarded. 542 * <p> 543 * The {@code finalize} method is never invoked more than once by a Java 544 * virtual machine for any given object. 545 * <p> 546 * Any exception thrown by the {@code finalize} method causes 547 * the finalization of this object to be halted, but is otherwise 548 * ignored. 549 * 550 * @throws Throwable the {@code Exception} raised by this method 551 * @see java.lang.ref.WeakReference 552 * @see java.lang.ref.PhantomReference 553 * @jls 12.6 Finalization of Class Instances 554 */ 555 protected void finalize() throws Throwable { } 556 }