先看图
从类图结构可以了解 java.util包下的2个大类:
1、Collecton:可以理解为主要存放的是单个对象
2、Map:可以理解为主要存储key-value类型的对象
一、Collection
Collection继承了Iterate接口,Iterate用于集合内迭代器抽象接口,其子类均实现接口中方法,看下ArrayList下实现:
1 /** 2 * Returns an iterator over the elements in this list in proper sequence. 3 * 4 *The returned iterator is fail-fast.
5 * 6 * @return an iterator over the elements in this list in proper sequence 7 */ 8 public Iteratoriterator() { 9 return new Itr(); // 返回内部类实例 10 } 11 12 /** 13 * An optimized version of AbstractList.Itr 14 */ 15 private class Itr implements Iterator { 16 int cursor; // index of next element to return 指向下一个位置索引id 17 int lastRet = -1; // index of last element returned; -1 if no such 指向上一个位置索引id 18 int expectedModCount = modCount; 19 20 public boolean hasNext() { 21 return cursor != size; 22 } 23 24 @SuppressWarnings("unchecked") 25 public E next() { 26 checkForComodification(); 27 int i = cursor; 28 if (i >= size) 29 throw new NoSuchElementException(); 30 Object[] elementData = ArrayList.this.elementData; 31 if (i >= elementData.length) 32 throw new ConcurrentModificationException(); 33 cursor = i + 1; 34 return (E) elementData[lastRet = i]; 35 } 36 37 public void remove() { 38 if (lastRet < 0) 39 throw new IllegalStateException(); 40 checkForComodification(); 41 42 try { 43 ArrayList.this.remove(lastRet); 44 cursor = lastRet; 45 lastRet = -1; 46 expectedModCount = modCount; 47 } catch (IndexOutOfBoundsException ex) { 48 throw new ConcurrentModificationException(); 49 } 50 } 51 52 @Override 53 @SuppressWarnings("unchecked") 54 public void forEachRemaining(Consumer super E> consumer) { 55 Objects.requireNonNull(consumer); 56 final int size = ArrayList.this.size; 57 int i = cursor; 58 if (i >= size) { 59 return; 60 } 61 final Object[] elementData = ArrayList.this.elementData; 62 if (i >= elementData.length) { 63 throw new ConcurrentModificationException(); 64 } 65 while (i != size && modCount == expectedModCount) { 66 consumer.accept((E) elementData[i++]); 67 } 68 // update once at end of iteration to reduce heap write traffic 69 cursor = i; 70 lastRet = i - 1; 71 checkForComodification(); 72 } 73 74 final void checkForComodification() { 75 if (modCount != expectedModCount) 76 throw new ConcurrentModificationException(); 77 } 78 }
1、List
特点:有序结果、顺序遍历、索引、允许有重复值
(1) ArrayList
以上特点实现:
transient Object[] elementData; // List内部存储对象数组结果 public boolean add(E e) { ensureCapacityInternal(size + 1); // Increments modCount!! elementData[size++] = e; return true; } // 添加对象时先识别是否越界,没有越界则数组对象当前索引值的下一个添 // 添加对象时,不识别重复,所以有序允许重复值 /** * Removes all of the elements from this list. The list will * be empty after this call returns. */ public void clear() { modCount++; // clear to let GC do its work for (int i = 0; i < size; i++) elementData[i] = null; size = 0; } // 清空List时顺序遍历值置为null
public E get(int index) {
rangeCheck(index);
return elementData(index);
}
E elementData(int index) {
return (E) elementData[index];
}
其中 remove方法 :
public E remove(int index) { // 按索引删除对象 rangeCheck(index); // 校验输入索引id是否越界,若越界则抛出运行时异常 IndexOutOfBoundsException modCount++; E oldValue = elementData(index); int numMoved = size - index - 1; // 定位到索引的下一位 if (numMoved > 0) System.arraycopy(elementData, index+1, elementData, index, numMoved); //调用native方法实现数组位置左移 elementData[--size] = null; // clear to let GC do its work // 末尾元素置空 return oldValue; } public boolean remove(Object o) {// 按对象删除 if (o == null) { for (int index = 0; index < size; index++) if (elementData[index] == null) { fastRemove(index); return true; } } else { for (int index = 0; index < size; index++) if (o.equals(elementData[index])) { // 识别对象相等使用equals方法,使用时注意重写equals方法 fastRemove(index); return true; } } return false; } /* * Private remove method that skips bounds checking and does not * return the value removed. */ private void fastRemove(int index) { modCount++; // 删除元素时,modCount值变更 int numMoved = size - index - 1; if (numMoved > 0) System.arraycopy(elementData, index+1, elementData, index, numMoved); elementData[--size] = null; // clear to let GC do its work }
可以看到ArrayList中对数组进行,操作时常用到System.arraycopy
java.lang.System下
public static native void arraycopy(Object src, int srcPos, Object dest, int destPos, int length);
还有在 java.util.Arrays下数组copy方法,最终也是调用System.arraycopy方法
1 public staticT[] copyOf(T[] original, int newLength) { 2 return (T[]) copyOf(original, newLength, original.getClass()); 3 } 4 5 public static T[] copyOf(U[] original, int newLength, Class extends T[]> newType) { 6 @SuppressWarnings("unchecked") 7 T[] copy = ((Object)newType == (Object)Object[].class) 8 ? (T[]) new Object[newLength] 9 : (T[]) Array.newInstance(newType.getComponentType(), newLength); 10 System.arraycopy(original, 0, copy, 0, 11 Math.min(original.length, newLength)); 12 return copy; 13 }
示例:
可以看到出现异常:ConcurrentModificationException,出现该异常原因是:
“快速失败”也就是fail-fast,它是Java集合的一种错误检测机制。当创建Iterator后,在Iterator使用还没有结束时,改变(删除或增添新项)集合元素就会出现上面的错误
看看ArrayList的排序方法:sort(Comparator super E> c)
1 public void sort(Comparator super E> c) { 2 final int expectedModCount = modCount; 3 Arrays.sort((E[]) elementData, 0, size, c); 4 if (modCount != expectedModCount) { 5 throw new ConcurrentModificationException(); 6 } 7 modCount++; 8 } 9 10 public staticvoid sort(T[] a, int fromIndex, int toIndex, 11 Comparator super T> c) { 12 if (c == null) { 13 sort(a, fromIndex, toIndex); 14 } else { 15 rangeCheck(a.length, fromIndex, toIndex); 16 if (LegacyMergeSort.userRequested) 17 legacyMergeSort(a, fromIndex, toIndex, c); 18 else 19 TimSort.sort(a, fromIndex, toIndex, c, null, 0, 0); 20 } 21 }
示例:
(2)LinkedList
1 public class LinkedList2 extends AbstractSequentialList 3 implements List , Deque , Cloneable, java.io.Serializable 4 // 实现了Deque接口,可以做队列使用 5 6 /** 7 * Pointer to first node. 8 * Invariant: (first == null && last == null) || 9 * (first.prev == null && first.item != null) 10 */ 11 transient Node first; 12 13 /** 14 * Pointer to last node. 15 * Invariant: (first == null && last == null) || 16 * (last.next == null && last.item != null) 17 */ 18 transient Node last; 19 20 /** 21 * Constructs an empty list. 22 */ 23 public LinkedList() { 24 } 25 26 // 集合对象存储结构,通过当前节点的前后节点,维护顺序集合(双向链表结构) 27 private static class Node { 28 E item; 29 Node next; 30 Node prev; 31 32 Node(Node prev, E element, Node next) { 33 this.item = element; 34 this.next = next; 35 this.prev = prev; 36 } 37 }
以上为 LinkedList的内部存储结构,以Node存储。
在看下集合元素插入、删除及获取方法实现:
1 public boolean add(E e) { 2 linkLast(e); 3 return true; 4 } 5 6 /** 7 * Links e as last element. 8 */ 9 void linkLast(E e) { 10 final Nodel = last; // 保存最后个节点 11 final Node newNode = new Node<>(l, e, null);// 新增节点 12 last = newNode; // 将新节点置为最后节点 13 if (l == null) 14 first = newNode; 15 else 16 l.next = newNode; 17 size++; 18 modCount++; 19 } 20 21 public E remove() { 22 return removeFirst(); // 去掉首节点 23 } 24 25 public E removeFirst() { 26 final Node f = first; 27 if (f == null) 28 throw new NoSuchElementException(); 29 return unlinkFirst(f); 30 } 31 32 private E unlinkFirst(Node f) { 33 // assert f == first && f != null; 34 final E element = f.item; 35 final Node next = f.next; 36 f.item = null; 37 f.next = null; // help GC 38 first = next; 39 if (next == null) 40 last = null; 41 else 42 next.prev = null; 43 size--; 44 modCount++; 45 return element; 46 } 47 48 // 入栈方法 49 public void push(E e) { 50 addFirst(e); 51 } 52 // 出栈方法 53 public E pop() { 54 return removeFirst(); 55 } 56 57 // 入队 58 public boolean offer(E e) { 59 return add(e); 60 } 61 public boolean add(E e) { 62 linkLast(e); 63 return true; 64 } 65 // 出队 66 public E poll() { 67 final Node f = first; 68 return (f == null) ? null : unlinkFirst(f); 69 } 70 71 //随机访问集合对象 72 public E get(int index) { 73 checkElementIndex(index); 74 return node(index).item; 75 } 76 77 /** 78 * Returns the (non-null) Node at the specified element index. 79 */ 80 Node node(int index) { 81 // assert isElementIndex(index); 82 // 识别 index id离首节点近还是尾节点近,减少遍历 83 if (index < (size >> 1)) { 84 Node x = first; 85 for (int i = 0; i < index; i++) // 0(i) 86 x = x.next; 87 return x; 88 } else { 89 Node x = last; 90 for (int i = size - 1; i > index; i--) // 0(i) 91 x = x.prev; 92 return x; 93 } 94 }
通过以上源码理解 ArrayList 和 LinkedList 区别类似数据结构中 数组及链表结构区别 ,新增、删除 和 随机访问存在 效率上的差别:
ArrayList是最常用的集合,其内部实现是一个数组,ArrayList的大小是可以动态扩充的。对于元素的随机访问效率高,其访问的时间复杂度为O(1)
,对于数据的插入与删除,从尾部操作效率高,时间复杂度和随机访问一样是O(1)
,若是从头部操作则效率会比较低,因为从头部插入或删除时需要移动后面所有元素,其时间复杂度为O(n-i)
(n表示元素个数,i表示元素位置)
LinkList对于随机访问效率是比较低的,因为它需要从头开始索引,所以其时间复杂度为O(i)
。但是对于元素的增删,LinkList效率高,因为只需要修改前后指针即可,其时间复杂度为O(1)
。
(3)Vector
与ArrayList类型,内部也是使用数据来存储对象,但是线程安全的,因为实现方法重写的时候,全部加上了同步关键字:synchronized;(一般不建议使用,性能消耗)
(4)Stack
public class Stackextends Vector { /** * Creates an empty Stack. */ public Stack() { }
2、Queue
遵循FIFO(先入先出规则),内部出栈入栈方法,主要区别在于是否是阻塞入队或出队
3、Set
public class HashSetextends AbstractSet implements Set , Cloneable, java.io.Serializable /** * Constructs a new, empty set; the backing HashMap instance has * default initial capacity (16) and load factor (0.75). */ public HashSet() { map = new HashMap<>(); } // Set 集合对象存储在 Map的 key中 public boolean contains(Object o) { return map.containsKey(o); } // 添加对象到Set集合中 public boolean add(E e) { return map.put(e, PRESENT)==null; } // 删除Set集合中对象 public boolean remove(Object o) { return map.remove(o)==PRESENT; }
实际Set集合的实现依赖于Map的实现,通过Map的 key值唯一性来实现
二、Map
1、HashMap:基于Map接口实现、允许null 键值、无序、非同步
一起看下HashMap的实现
// map 内部对象链表存储结构 static class Nodeimplements Map.Entry { final int hash; final K key; V value; Node next; // 下一节点 Node(int hash, K key, V value, Node next) { this.hash = hash; this.key = key; this.value = value; this.next = next; } public final K getKey() { return key; } public final V getValue() { return value; } public final String toString() { return key + "=" + value; } // 重写hashCode方法 public final int hashCode() { return Objects.hashCode(key) ^ Objects.hashCode(value); } public final V setValue(V newValue) { V oldValue = value; value = newValue; return oldValue; } // 重写 equals方法 public final boolean equals(Object o) { if (o == this) return true; if (o instanceof Map.Entry) { Map.Entry,?> e = (Map.Entry,?>)o; if (Objects.equals(key, e.getKey()) && Objects.equals(value, e.getValue())) return true; } return false; } } transient Node [] table; // 用数组保存多条链表的首节点 // 获取 key所对应的存储JNode的 value值 public V get(Object key) { Node e; return (e = getNode(hash(key), key)) == null ? null : e.value; } // 识别是否存在key所对应的 Node public boolean containsKey(Object key) { return getNode(hash(key), key) != null; }
// map 内部对象链表存储结构 static class Nodeimplements Map.Entry { final int hash; final K key; V value; Node next; // 下一节点 Node(int hash, K key, V value, Node next) { this.hash = hash; this.key = key; this.value = value; this.next = next; } public final K getKey() { return key; } public final V getValue() { return value; } public final String toString() { return key + "=" + value; } // 重写hashCode方法 public final int hashCode() { return Objects.hashCode(key) ^ Objects.hashCode(value); } public final V setValue(V newValue) { V oldValue = value; value = newValue; return oldValue; } // 重写 equals方法 public final boolean equals(Object o) { if (o == this) return true; if (o instanceof Map.Entry) { Map.Entry,?> e = (Map.Entry,?>)o; if (Objects.equals(key, e.getKey()) && Objects.equals(value, e.getValue())) return true; } return false; } } transient Node [] table; // 用数组保存多条链表的首节点 // 获取 key所对应的存储JNode的 value值 public V get(Object key) { Node e; return (e = getNode(hash(key), key)) == null ? null : e.value; } // 识别是否存在key所对应的 Node public boolean containsKey(Object key) { return getNode(hash(key), key) != null; }
// 插入 对象 public V put(K key, V value) { return putVal(hash(key), key, value, false, true); } // 调用的内部方法, final V putVal(int hash, K key, V value, boolean onlyIfAbsent, boolean evict) { // tab 为map内首节点集合 Node[] tab; Node p; int n, i; // 先识别table是否为空,为空则初始化,hashmap内存存储延迟加载在这里体现 if ((tab = table) == null || (n = tab.length) == 0) n = (tab = resize()).length; // 通过hash值以长度做按位与,识别读取元素的存储在tab中的位置 if ((p = tab[i = (n - 1) & hash]) == null) // 若tab所在链表首节点为空,则直接构造新节点 tab[i] = newNode(hash, key, value, null); else { // tab所在链表首节点不为空,则遍历p所在链表或红黑树,找到可以存储的位置 Node e; K k; if (p.hash == hash && ((k = p.key) == key || (key != null && key.equals(k)))) e = p; else if (p instanceof TreeNode) e = ((TreeNode )p).putTreeVal(this, tab, hash, key, value); else { for (int binCount = 0; ; ++binCount) { if ((e = p.next) == null) { p.next = newNode(hash, key, value, null); if (binCount >= TREEIFY_THRESHOLD - 1) // -1 for 1st treeifyBin(tab, hash); break; } if (e.hash == hash && ((k = e.key) == key || (key != null && key.equals(k)))) break; p = e; } } if (e != null) { // existing mapping for key V oldValue = e.value; if (!onlyIfAbsent || oldValue == null) e.value = value; afterNodeAccess(e); return oldValue; } } ++modCount; if (++size > threshold) resize(); afterNodeInsertion(evict); return null; }
可以看到,在HashMap中存储的结构下, Node类型的数组保存头部节点(单链表)或根节点(红黑树),先以 Node的key的hash值与数组长度做位与运算(hash碰撞),初始
时使用单链表存储新插入对象(newNode),当链表长度超过8时,会将链表结构转为红黑树结构存储(treeifyBin方法)
final void treeifyBin(Node[] tab, int hash) { int n, index; Node e; if (tab == null || (n = tab.length) < MIN_TREEIFY_CAPACITY) resize(); else if ((e = tab[index = (n - 1) & hash]) != null) { // 找到需要转换的 单链表 e,遍历单链表,转换为TreeNode,保存前后节点关系 TreeNode hd = null, tl = null; do { TreeNode p = replacementTreeNode(e, null); if (tl == null) hd = p; else { p.prev = tl; tl.next = p; } tl = p; } while ((e = e.next) != null); //让桶的第一个元素指向新建的红黑树头结点,以后这个桶里的元素就是红黑树而不是链表了 if ((tab[index] = hd) != null) hd.treeify(tab); } }
先将单链表转换为 treenode,在调用 treeify方法构造红黑树
2、LinkedHashMap
继承HashMap,HashMap是无序集合,而LinkedHashMap为有序集合
public class LinkedHashMapextends HashMap implements Map
构造LinkedHashMap.EntreyNode
static class Entryextends HashMap.Node { Entry before, after; // before 保存前置节点,after保存后置节点 Entry(int hash, K key, V value, Node next) { super(hash, key, value, next); } } /** * The head (eldest) of the doubly linked list. */ transient LinkedHashMap.Entry head; // 头节点 /** * The tail (youngest) of the doubly linked list. */ transient LinkedHashMap.Entry tail; // 尾部节点 // 重写HashMap的 创建新节点方法 Node newNode(int hash, K key, V value, Node e) { LinkedHashMap.Entry p = new LinkedHashMap.Entry (hash, key, value, e); linkNodeLast(p); // 将新节点放到尾部节点,从而保证顺序 return p; } // link at the end of list private void linkNodeLast(LinkedHashMap.Entry p) { LinkedHashMap.Entry last = tail; tail = p; if (last == null) head = p; else { p.before = last; last.after = p; } }
3、TreeMap
TreeMap直接使用红黑树结构存储集合元素,根据键 做排序,排序规则按内部 comparator 对象的实例对象的排序规则,若comparator为空,则按自然排序
1 public class TreeMap2 extends AbstractMap 3 implements NavigableMap , Cloneable, java.io.Serializable 4 { 5 /** 6 * The comparator used to maintain order in this tree map, or 7 * null if it uses the natural ordering of its keys. 8 * 9 * @serial 10 */ 11 private final Comparator super K> comparator; // 对象比较接口 12 13 private transient Entry root; // 根节点
root的实现逻辑为 TreeMap.Entrey
static final class Entryimplements Map.Entry { K key; V value; Entry left; Entry right; Entry parent; boolean color = BLACK;
所以TreeMap put和get方法就是以键先进行红黑树的查找后操作
4、HashTable
HashTable和HashMap数据结构类似,主要区别为HashTable中操作集合元素对象的方法都加上了 同步关键字(synchronized), 所以说线程安全的及集合