以下源碼基于 Android SDK 23， 與JDK中略有差別，但基本相同；整體源碼由 構(gòu)造、添加(add)、設(shè)置(set)、獲取(get)、移除(remove)、迭代器(iterator) 和序列化(Serializable)組成，最后我還會(huì)把里邊一些不常用的方法舉例說(shuō)明下作用，下面我們就一一探究其實(shí)現(xiàn)原理。
概述
ArrayList 存儲(chǔ)的實(shí)質(zhì)是操作一個(gè)數(shù)組，這個(gè)數(shù)組可以根據(jù)內(nèi)容自動(dòng)擴(kuò)容，所以讓 ArrayList看起來(lái)像一個(gè)無(wú)限大小的容器一樣。
屬性
/**
* The minimum amount by which the capacity of an ArrayList will increase.
* This tuning parameter controls a time-space tradeoff. This value (12)
* gives empirically good results and is arguably consistent with the
* RI's specified default initial capacity of 10: instead of 10, we start
* with 0 (sans allocation) and jump to 12.
*/
private static final int MIN_CAPACITY_INCREMENT = 12;
/**
* The number of elements in this list.
*/
int size;
/**
* The elements in this list, followed by nulls.
*/
transient Object[] array;
- 既然
ArrayList可以自動(dòng)擴(kuò)容，那么就要有一個(gè)描述每次擴(kuò)容的基準(zhǔn)，MIN_CAPACITY_INCREMENT就是這個(gè)基準(zhǔn)，默認(rèn)值是12。 -
array是ArrayList的核心，所有數(shù)據(jù)均存儲(chǔ)在array這個(gè)數(shù)組中，發(fā)生自動(dòng)擴(kuò)容時(shí)，array會(huì)指向新的數(shù)組首地址，但注意了，transient表示它不會(huì)參與序列化過(guò)程。 -
size始終描述ArrayList中實(shí)際的大小。
構(gòu)造方法
/**
* Constructs a new instance of {@code ArrayList} with the specified
* initial capacity.
*
* @param capacity
* the initial capacity of this {@code ArrayList}.
*/
public ArrayList(int capacity) {
if (capacity < 0) {
throw new IllegalArgumentException("capacity < 0: " + capacity);
}
array = (capacity == 0 ? EmptyArray.OBJECT : new Object[capacity]);
}
/**
* Constructs a new {@code ArrayList} instance with zero initial capacity.
*/
public ArrayList() {
array = EmptyArray.OBJECT;
}
/**
* Constructs a new instance of {@code ArrayList} containing the elements of
* the specified collection.
*
* @param collection
* the collection of elements to add.
*/
public ArrayList(Collection<? extends E> collection) {
if (collection == null) {
throw new NullPointerException("collection == null");
}
Object[] a = collection.toArray();
if (a.getClass() != Object[].class) {
Object[] newArray = new Object[a.length];
System.arraycopy(a, 0, newArray, 0, a.length);
a = newArray;
}
array = a;
size = a.length;
}
ArrayList共含有3個(gè)構(gòu)造方法，EmptyArray.OBJECT是一個(gè)length為0的空數(shù)組new Object[0]，new ArrayList() 則會(huì)創(chuàng)建一個(gè)大小為0的數(shù)組；你也可以去指定初始的容量capacity，new ArrayList(int capacity) ，避免ArrayList 第一次add 或者其他操作就進(jìn)行擴(kuò)容；第三個(gè)構(gòu)造可以傳入一個(gè)集合，這里要提一下Collection，你可以認(rèn)為它是 List 、Queue、Set的始祖，這里只要在它們內(nèi)部實(shí)現(xiàn)了 toArray 方法，并且返回一個(gè)Object[]類型的數(shù)據(jù)，就可以成功初始化到 ArrayList中。
添加(add / addAll)
/**
* Adds the specified object at the end of this {@code ArrayList}.
*
* @param object
* the object to add.
* @return always true
*/
@Override public boolean add(E object) {
Object[] a = array;
int s = size;
if (s == a.length) {
Object[] newArray = new Object[s +
(s < (MIN_CAPACITY_INCREMENT / 2) ?
MIN_CAPACITY_INCREMENT : s >> 1)];
System.arraycopy(a, 0, newArray, 0, s);
array = a = newArray;
}
a[s] = object;
size = s + 1;
modCount++;
return true;
}
/**
* Inserts the specified object into this {@code ArrayList} at the specified
* location. The object is inserted before any previous element at the
* specified location. If the location is equal to the size of this
* {@code ArrayList}, the object is added at the end.
*
* @param index
* the index at which to insert the object.
* @param object
* the object to add.
* @throws IndexOutOfBoundsException
* when {@code location < 0 || location > size()}
*/
@Override public void add(int index, E object) {
Object[] a = array;
int s = size;
if (index > s || index < 0) {
throwIndexOutOfBoundsException(index, s);
}
if (s < a.length) {
System.arraycopy(a, index, a, index + 1, s - index);
} else {
// assert s == a.length;
Object[] newArray = new Object[newCapacity(s)];
System.arraycopy(a, 0, newArray, 0, index);
System.arraycopy(a, index, newArray, index + 1, s - index);
array = a = newArray;
}
a[index] = object;
size = s + 1;
modCount++;
}
/**
* This method controls the growth of ArrayList capacities. It represents
* a time-space tradeoff: we don't want to grow lists too frequently
* (which wastes time and fragments storage), but we don't want to waste
* too much space in unused excess capacity.
*
* NOTE: This method is inlined into {@link #add(Object)} for performance.
* If you change the method, change it there too!
*/
private static int newCapacity(int currentCapacity) {
int increment = (currentCapacity < (MIN_CAPACITY_INCREMENT / 2) ?
MIN_CAPACITY_INCREMENT : currentCapacity >> 1);
return currentCapacity + increment;
}
/**
* Adds the objects in the specified collection to this {@code ArrayList}.
*
* @param collection
* the collection of objects.
* @return {@code true} if this {@code ArrayList} is modified, {@code false}
* otherwise.
*/
@Override public boolean addAll(Collection<? extends E> collection) {
Object[] newPart = collection.toArray();
int newPartSize = newPart.length;
if (newPartSize == 0) {
return false;
}
Object[] a = array;
int s = size;
int newSize = s + newPartSize; // If add overflows, arraycopy will fail
if (newSize > a.length) {
int newCapacity = newCapacity(newSize - 1); // ~33% growth room
Object[] newArray = new Object[newCapacity];
System.arraycopy(a, 0, newArray, 0, s);
array = a = newArray;
}
System.arraycopy(newPart, 0, a, s, newPartSize);
size = newSize;
modCount++;
return true;
}
/**
* Inserts the objects in the specified collection at the specified location
* in this List. The objects are added in the order they are returned from
* the collection's iterator.
*
* @param index
* the index at which to insert.
* @param collection
* the collection of objects.
* @return {@code true} if this {@code ArrayList} is modified, {@code false}
* otherwise.
* @throws IndexOutOfBoundsException
* when {@code location < 0 || location > size()}
*/
@Override
public boolean addAll(int index, Collection<? extends E> collection) {
int s = size;
if (index > s || index < 0) {
throwIndexOutOfBoundsException(index, s);
}
Object[] newPart = collection.toArray();
int newPartSize = newPart.length;
if (newPartSize == 0) {
return false;
}
Object[] a = array;
int newSize = s + newPartSize; // If add overflows, arraycopy will fail
if (newSize <= a.length) {
System.arraycopy(a, index, a, index + newPartSize, s - index);
} else {
int newCapacity = newCapacity(newSize - 1); // ~33% growth room
Object[] newArray = new Object[newCapacity];
System.arraycopy(a, 0, newArray, 0, index);
System.arraycopy(a, index, newArray, index + newPartSize, s-index);
array = a = newArray;
}
System.arraycopy(newPart, 0, a, index, newPartSize);
size = newSize;
modCount++;
return true;
}
這里有必要先看一個(gè)方法，System.arraycopy()
public static native void arraycopy(Object src, int srcPos,
Object dst, int dstPos, int length);
這是一個(gè) native方法，負(fù)責(zé)數(shù)組拷貝，從 src 的 srcPos 開(kāi)始，將 length長(zhǎng)度的數(shù)據(jù)拷貝到 dst 中，dstPos中的數(shù)據(jù)是srcPos位置的數(shù)據(jù)。
public boolean add(E object)
這是最簡(jiǎn)單的一個(gè)add操作，里邊會(huì)進(jìn)行擴(kuò)容判斷，如果當(dāng)前ArrayList.size與array.length相同，則進(jìn)行擴(kuò)容，擴(kuò)容的策略是s < (MIN_CAPACITY_INCREMENT / 2) ? MIN_CAPACITY_INCREMENT : s >> 1，即 s < 6 ? 6 : s * 2， 最終擴(kuò)容的大小為 (s + s < 6 ? 6 : s * 2)；newCapacity(int currentCapacity) 方法也是這個(gè)作用，返回最終擴(kuò)容后的大小。
public void add(int index, E object)
這個(gè)方法的作用是將 object 插入至 index位置，這里也會(huì)有擴(kuò)容判斷，既然是插入一個(gè)值，那么size就會(huì) +1，所以 ArrayList.size 小于 array.length是一種情況，數(shù)組可以直接從 index處 后移一位，再將 object 放入 index的位置；若是大于等于，則原array需要擴(kuò)容，擴(kuò)容后現(xiàn)將old array 數(shù)據(jù) 復(fù)制到 new array中，再進(jìn)行后移，最終把object插入到index位置。
public boolean addAll(Collection<? extends E> collection)
public boolean addAll(int index, Collection<? extends E> collection)
這兩個(gè)方法只是批量操作，內(nèi)部邏輯與add 是一樣的，都要先判斷 ArrayList.size與array.length的大小關(guān)系進(jìn)行擴(kuò)容，之后通過(guò) System.arraycopy去操作array。
注：這里你有可能會(huì)發(fā)現(xiàn)有個(gè)變量
modCount，它用來(lái)表達(dá)ArrayList的修改次數(shù)（add、remove），是它導(dǎo)致ArrayList不是線程安全的，等講到迭代器iterator的時(shí)候再來(lái)說(shuō)說(shuō)這個(gè)變量。
設(shè)置
/**
* Replaces the element at the specified location in this {@code ArrayList}
* with the specified object.
*
* @param index
* the index at which to put the specified object.
* @param object
* the object to add.
* @return the previous element at the index.
* @throws IndexOutOfBoundsException
* when {@code location < 0 || location >= size()}
*/
@Override public E set(int index, E object) {
Object[] a = array;
if (index >= size) {
throwIndexOutOfBoundsException(index, size);
}
@SuppressWarnings("unchecked") E result = (E) a[index];
a[index] = object;
return result;
}
這個(gè)方法沒(méi)什么，就是把array[index] 替換，并且把原來(lái)的數(shù)據(jù)返回。
獲取
@SuppressWarnings("unchecked")
@Override
public E get(int index) {
if (index >= size) {
throwIndexOutOfBoundsException(index, size);
}
return (E) array[index];
}
這個(gè)方法也不多說(shuō)，將array[index]返回。
移除
/**
* Removes the object at the specified location from this list.
*
* @param index
* the index of the object to remove.
* @return the removed object.
* @throws IndexOutOfBoundsException
* when {@code location < 0 || location >= size()}
*/
@Override public E remove(int index) {
Object[] a = array;
int s = size;
if (index >= s) {
throwIndexOutOfBoundsException(index, s);
}
@SuppressWarnings("unchecked") E result = (E) a[index];
System.arraycopy(a, index + 1, a, index, --s - index);
a[s] = null; // Prevent memory leak
size = s;
modCount++;
return result;
}
@Override public boolean remove(Object object) {
Object[] a = array;
int s = size;
if (object != null) {
for (int i = 0; i < s; i++) {
if (object.equals(a[i])) {
System.arraycopy(a, i + 1, a, i, --s - i);
a[s] = null; // Prevent memory leak
size = s;
modCount++;
return true;
}
}
} else {
for (int i = 0; i < s; i++) {
if (a[i] == null) {
System.arraycopy(a, i + 1, a, i, --s - i);
a[s] = null; // Prevent memory leak
size = s;
modCount++;
return true;
}
}
}
return false;
}
@Override protected void removeRange(int fromIndex, int toIndex) {
if (fromIndex == toIndex) {
return;
}
Object[] a = array;
int s = size;
if (fromIndex >= s) {
throw new IndexOutOfBoundsException("fromIndex " + fromIndex
+ " >= size " + size);
}
if (toIndex > s) {
throw new IndexOutOfBoundsException("toIndex " + toIndex
+ " > size " + size);
}
if (fromIndex > toIndex) {
throw new IndexOutOfBoundsException("fromIndex " + fromIndex
+ " > toIndex " + toIndex);
}
System.arraycopy(a, toIndex, a, fromIndex, s - toIndex);
int rangeSize = toIndex - fromIndex;
Arrays.fill(a, s - rangeSize, s, null);
size = s - rangeSize;
modCount++;
}
add方法已經(jīng)進(jìn)行了詳細(xì)的講解，想必大家都能猜到，remove操作就是講 index 或者 range的一段數(shù)據(jù)從array中移除，然后再通過(guò)System.arraycopy拷貝之后的數(shù)據(jù)前移補(bǔ)充空位，下圖以移除單個(gè)為例，將步驟分解：
迭代器
@Override public Iterator<E> iterator() {
return new ArrayListIterator();
}
private class ArrayListIterator implements Iterator<E> {
/** Number of elements remaining in this iteration */
private int remaining = size;
/** Index of element that remove() would remove, or -1 if no such elt */
private int removalIndex = -1;
/** The expected modCount value */
private int expectedModCount = modCount;
public boolean hasNext() {
return remaining != 0;
}
@SuppressWarnings("unchecked") public E next() {
ArrayList<E> ourList = ArrayList.this;
int rem = remaining;
if (ourList.modCount != expectedModCount) {
throw new ConcurrentModificationException();
}
if (rem == 0) {
throw new NoSuchElementException();
}
remaining = rem - 1;
return (E) ourList.array[removalIndex = ourList.size - rem];
}
public void remove() {
Object[] a = array;
int removalIdx = removalIndex;
if (modCount != expectedModCount) {
throw new ConcurrentModificationException();
}
if (removalIdx < 0) {
throw new IllegalStateException();
}
System.arraycopy(a, removalIdx + 1, a, removalIdx, remaining);
a[--size] = null; // Prevent memory leak
removalIndex = -1;
expectedModCount = ++modCount;
}
}
迭代器，一個(gè)很重要的概念，它的作用就是便利整個(gè)ArrayList， for each 的原理其實(shí)就是迭代器的使用，上文說(shuō)到了modCount與迭代器相關(guān)，
if (ourList.modCount != expectedModCount) {
throw new ConcurrentModificationException();
}
expectedModCount是iterator初始化時(shí)賦予的值，值為modCount，而modCount會(huì)根據(jù)add或者remove進(jìn)行++操作，這就表明，當(dāng)iterator創(chuàng)建好后，只要使用這個(gè)iterator實(shí)例去進(jìn)行遍歷，就不能使用ArrayList.add或者ArrayList.remove操作，因?yàn)槿绻褂昧耍?code>modCount會(huì)發(fā)生變化，這樣在next()的時(shí)候就會(huì)拋出異常ConcurrentModificationException，這也進(jìn)一步說(shuō)明ArrayList不是線程安全的。那么在遍歷中如何移除元素呢，就是下邊實(shí)現(xiàn)的remove方法了，remove過(guò)程與之前類似，關(guān)鍵在于expectedModCount = ++modCount;，remove需要使modCount 遞增，那么我讓expectedModCount重新賦值，即可完成刪除操作。
序列化
private static final long serialVersionUID = 8683452581122892189L;
private void writeObject(ObjectOutputStream stream) throws IOException {
stream.defaultWriteObject();
stream.writeInt(array.length);
for (int i = 0; i < size; i++) {
stream.writeObject(array[i]);
}
}
private void readObject(ObjectInputStream stream) throws IOException, ClassNotFoundException {
stream.defaultReadObject();
int cap = stream.readInt();
if (cap < size) {
throw new InvalidObjectException(
"Capacity: " + cap + " < size: " + size);
}
array = (cap == 0 ? EmptyArray.OBJECT : new Object[cap]);
for (int i = 0; i < size; i++) {
array[i] = stream.readObject();
}
}
這是在代碼末尾了，ArrayList是通過(guò)stream.writeObject 連續(xù)寫入 array的內(nèi)容。
其他
public boolean contains(Object object)
利用 object 的 equals方法判斷ArrayList中是否包含object對(duì)象。
public int indexOf(Object object)
public int lastIndexOf(Object object)
這兩個(gè)方法都是獲取 object 在 ArrayList中的位置，第一個(gè)是正序遍歷，找到的第一個(gè)返回的index；第二個(gè)是倒序遍歷，找到第一個(gè)返回的index。
/**
* Sets the capacity of this {@code ArrayList} to be the same as the current
* size.
*
* @see #size
*/
public void trimToSize() {
int s = size;
if (s == array.length) {
return;
}
if (s == 0) {
array = EmptyArray.OBJECT;
} else {
Object[] newArray = new Object[s];
System.arraycopy(array, 0, newArray, 0, s);
array = newArray;
}
modCount++;
}
這個(gè)方法是將當(dāng)前的array “精簡(jiǎn)”一下，比如 array.length 是10，但里邊的size是 5個(gè)，那么就將 array.length變?yōu)?5，把數(shù)據(jù)通過(guò) System.arraycopy 拷貝到新的 array中。
@Override public boolean equals(Object o) {
if (o == this) {
return true;
}
if (!(o instanceof List)) {
return false;
}
List<?> that = (List<?>) o;
int s = size;
if (that.size() != s) {
return false;
}
Object[] a = array;
if (that instanceof RandomAccess) {
for (int i = 0; i < s; i++) {
Object eThis = a[i];
Object ethat = that.get(i);
if (eThis == null ? ethat != null : !eThis.equals(ethat)) {
return false;
}
}
} else { // Argument list is not random access; use its iterator
Iterator<?> it = that.iterator();
for (int i = 0; i < s; i++) {
Object eThis = a[i];
Object eThat = it.next();
if (eThis == null ? eThat != null : !eThis.equals(eThat)) {
return false;
}
}
}
return true;
}
再來(lái)看下這個(gè)長(zhǎng)長(zhǎng)的equals方法，非常好懂，但是乍眼一看有個(gè) RandomAccess，這是什么？尋找了一下它的實(shí)現(xiàn)類，發(fā)現(xiàn)ArrayList就是它的實(shí)現(xiàn)類，再看下這個(gè)if(...){}else{}，如果是RandomAccess的實(shí)現(xiàn)類，那么直接使用get(index)獲取元素，否則需要使用迭代器iterator。以下是對(duì)于RandomAccess的一段摘錄：
jdk中有個(gè)RandomAccess接口，這是一個(gè)標(biāo)記接口(Marker)，它沒(méi)有任何方法，這個(gè)接口被List的實(shí)現(xiàn)類（子類）使用。如果List子類實(shí)現(xiàn)了RandomAccess接口，那就表示它能夠快速隨機(jī)訪問(wèn)存儲(chǔ)的元素。RandomAccess接口的意義在于：在對(duì)列表進(jìn)行隨機(jī)或順序訪問(wèn)的時(shí)候，訪問(wèn)算法能夠選擇性能最佳方式。一般的列表訪問(wèn)算法在訪問(wèn)列表元素之前，都被建議先使用instanceof關(guān)鍵字檢查一下列表是否是一個(gè)RandomAccess子類，然后再?zèng)Q定采用隨機(jī)還是順序方式訪問(wèn)列表中的元素，這樣可以保證訪問(wèn)算法擁有最佳的性能。對(duì)于List的子類，如果：
for (int i=0, n=list.size(); i < n; i++)
list.get(i);
的訪問(wèn)方式比
for (Iterator i=list.iterator(); i.hasNext(); )
i.next();
快，那么它應(yīng)該實(shí)現(xiàn)RandomAccess接口。
