The collection is one of many of our daily programming technologies may use high frequency of use may usually will know how to use but the relationship between the set differs from the not very clear about their underlying principle even more so wrote the lyrics articles so we have a deeper understanding
A collection is a huge family today First is that these ArrayList, LinkedList, Vector
ArrayList because of its underlying array of arrays we all know it's a high efficiency ArrayList query modification is also true but why lower modify the query low efficiency high efficiency insertion and deletion of data structure it which is saying a relationship it took Let's look at it ArrayList data structure model
Insert, Delete: if a number of its 100 steps we want to insert the collection is a copy of the number 32 and then moved back behind the number in the first set to the position you want to insert we not only want to copy the data but also To copy the data behind the data, move it backward if a lot of data, then this collection is that efficiency will be very low, then delete the data to be like the previous one while moving efficiency is very low
Query, Modify: Modify the query if we can locate only need to pass an array of index data so the actual development of high efficiency, we query the data so most are very extensive use ArrayList
Everything has two sides, whether another set (because the programming is the thing people invented) in order to solve this short board ArrayList clever programmers to use in life or work program is equally applicable
ArrayList CRUD source
From source we can see whether it is time to insert and delete elements are copied ArrayList array operation which also led to its efficiency is not high
1 //查询元素
2 public E get(int index) {
3 //检查元素是否越界
4 rangeCheck(index);
5
6 return elementData(index);
7 }
8
9
10 //按顺序添加元素
11 public boolean add(E e) {
12 //确认开启扩容机制
13 ensureCapacityInternal(size + 1); // Increments modCount!!
14 elementData[size++] = e;
15 return true;
16 }
17
18 //在指定位置插入元素
19 public void add(int index, E element) {
20 //检查索引是否越界
21 rangeCheckForAdd(index);
22 //确认开启扩容机制
23 ensureCapacityInternal(size + 1); // Increments modCount!!
24 //复制数组
25 System.arraycopy(elementData, index, elementData, index + 1,
26 size - index);
27 //替换元素
28 elementData[index] = element;
29 size++;
30 }
31
32
33
34
35 //移除某个元素
36 public E remove(int index) {
37 rangeCheck(index);
38
39 modCount++;
40 E oldValue = elementData(index);
41
42 int numMoved = size - index - 1;
43 if (numMoved > 0)
//复制数组
44 System.arraycopy(elementData, index+1, elementData, index,
45 numMoved);
46 elementData[--size] = null; // clear to let GC do its work
47
48 return oldValue;
49 }
50
51
52
LinkedList
LinkedList它的底层是双向链表实现的非线程安全的集合，它是一个链表结构，不能像数组一样随机访问，必须是每个元素依次遍历直到找到元素为止。其结构的特殊性导致它查询数据慢。 接下来我们来看看它的结构模型
插入、删除 ：因为是链表结构 所以它的插入效率很高 (如果在14 和 18之间插入一个33 的话，链表直接会将连接到18的链子断开 然后连接上33所在的前节点 数据18的前节点再连接上33的后节点 如图2所示) 也就是说 插入一个数字我们只需要将(14 和 18 之间的)链表断开 再将14和33之间的链表连上即可 比ArrayList的数组复制效率高
查询、修改 ：LinkedList 查询速度慢 因为它要遍历整个整个集合 直到找到元素为止 如果集合数组多的话 消耗的资源就多 而ArrayList是通过数组下标定位速度快 同样他也是线程不安全的
linkedList
在执行查询时 先判断元素是靠近头部还是尾部 如果是头部 若靠近头部，则从头部开始依次查询判断
执行插入时 判断是插入到中间还是尾部 如果插入到尾部 直接将尾节点的下一个指针指向新增节点。如果插入到中间 获取到当前节点的上一个节点（D） 并将D节点的后指针指向新的节点头指针 然后新增节点的下一个指针指向当前节点。
1 //查询元素
2 public E get(int index) {
3 //检查所引是否越界
4 checkElementIndex(index);
5 return node(index).item;
6 }
7
8 // 返回指定索引处的节点
9 Node<E> node(int index) {
10 // 指定的索引值与链表大小右移一位，及除以 2 进行比较
11 if (index < (size >> 1)) { // 索引小，则从首节点向后扫描，直到索引值处
12 Node<E> x = first;
13 for (int i = 0; i < index; i++)
14 x = x.next;
15 return x;
16 } else { // 索引大，则从尾节点向前扫描，直到索引值处
17 Node<E> x = last;
18 for (int i = size - 1; i > index; i--)
19 x = x.prev;
20 return x;
21 }
22 }
23
24
25 //移除指定元素
26 public E remove(int index) {
27 checkElementIndex(index);
28 return unlink(node(index));
29 }
30
31 //在指定位置添加元素
32 public void add(int index, E element) {
33 //检查所引是否越界
34 checkPositionIndex(index);
35 // 在链表末尾天添加
36 if (index == size)
37 linkLast(element);
38 else
39 linkBefore(element, node(index));
40 }
41
42 private static class Node<E> {
43 E item;
44 //头节点
45 Node<E> next;
46 //尾节点
47 Node<E> prev;
48 Node(Node<E> prev, E element, Node<E> next) {
49 this.item = element;
50 this.next = next;
51 this.prev = prev;
52 }
53 }
54
55 /**
56 * Links e as last element.
57 */
58 void linkLast(E e) {
59 //用l来临时保存未插入前的last节点
60 final Node<E> l = last;
61 //创建一个值为e的新节点 添加第一个元素时 l = null
62 final Node<E> newNode = new Node<>(l, e, null);
63 //将新节点赋值的last
64 last = newNode;
65 if (l == null)
66 first = newNode;
67 else
68 l.next = newNode;
69 size++;
70 modCount++;
71 }
Vector
Vector的数据结构和使用方法 跟ArrayList相同 不同之处在于Vector是线程安全的 几乎所有的对数据操作的方法都被synchronized关键字修饰 synchronized是线程同步的 当一个线程获得Vector对象锁的时候 其它的线程必须等到它执行完毕之后（锁被释放）才能执行
总结
1.ArrayList 它的底层是一个数组 查询修改数据快(通过下标定位) 但是插入删除数据比较慢 (插入数据慢是因为复制数组耗时) 为了改进这个缺点 于是就有了LinkedList数组 它是一个链表结构 插入和删除数据很快（只需要修改指针引用） 但是查询和修改数效率低（他要查询到整个链表从第一个开始寻找 一直找到为止）
2.ArrayList 和LinkedList都是线程不安全的
3.Vector是线程安全的 但是效率低 当我们执行单个线程的时候ArrayList的效率高于Vector