本文基于JDK1.8 >读完本文预计须要25分钟(因有大量源代码,电脑屏观看体验较佳)
摘要
HashMap置信这是呈现频率最高的面试点之一,应该是面试问到烂的面试题之一,同时也是Java中用于解决键值对最罕用的数据类型。那么咱们就针对JDK8的HashMap独特学习一下!
次要办法
要害变量:
/** * The default initial capacity - MUST be a power of two. * 初始容量大小 必须是2的次幂 */ static final int DEFAULT_INITIAL_CAPACITY = 1 << 4; // aka 16 /** * The maximum capacity, used if a higher value is implicitly specified * by either of the constructors with arguments. * MUST be a power of two <= 1<<30. * 最大的容量大小 * 超过这个值就将threshold批改为Integer.MAX_VALUE,数组不进行扩容 */ static final int MAXIMUM_CAPACITY = 1 << 30; /** * The load factor used when none specified in constructor. * 负载因子 为什么是0.75?因为统计学中hash抵触合乎泊松散布,7-8之间抵触最小 */ static final float DEFAULT_LOAD_FACTOR = 0.75f; /** * The bin count threshold for using a tree rather than list for a * bin. Bins are converted to trees when adding an element to a * bin with at least this many nodes. The value must be greater * than 2 and should be at least 8 to mesh with assumptions in * tree removal about conversion back to plain bins upon * shrinkage. * 链表大于这个值就会树化 * 留神:树化并不是整个map链表,而是某一个大于此阈值的链表 */ static final int TREEIFY_THRESHOLD = 8; /** * The bin count threshold for untreeifying a (split) bin during a * resize operation. Should be less than TREEIFY_THRESHOLD, and at * most 6 to mesh with shrinkage detection under removal. * 小于这个值就会反树化 */ static final int UNTREEIFY_THRESHOLD = 6;
四个构造方法:
//构造方法1//指定初始容量大小,负载因子public HashMap(int initialCapacity, float loadFactor) { if (initialCapacity < 0) throw new IllegalArgumentException("Illegal initial capacity: " + initialCapacity); if (initialCapacity > MAXIMUM_CAPACITY) //1<<30 最大容量是 Integer.MAX_VALUE; initialCapacity = MAXIMUM_CAPACITY; if (loadFactor <= 0 || Float.isNaN(loadFactor)) throw new IllegalArgumentException("Illegal load factor: " + loadFactor); this.loadFactor = loadFactor; //tableSizeFor这个办法用于找到大于等于initialCapacity的最小的2的幂 this.threshold = tableSizeFor(initialCapacity); }//构造方法2//其实调用了上边的构造方法1 负载因子给的默认值0.75public HashMap(int initialCapacity) { this(initialCapacity, DEFAULT_LOAD_FACTOR); }//构造方法3//空参结构,均应用默认值public HashMap() { this.loadFactor = DEFAULT_LOAD_FACTOR; // all other fields defaulted } //构造方法4//与其余三个相比,初始化了public HashMap(Map<? extends K, ? extends V> m) { this.loadFactor = DEFAULT_LOAD_FACTOR;//0.75f //调用了putVal办法,而putVal办法中有resize办法,有初始化 putMapEntries(m, false); }
对四个构造方法简略总结一下:
1、前三个构造函数并没有初始化,都是用到的时候去初始化2、构造方法4相当于用到了put办法,所以初始化了
hashmap->hash()
/** * Computes key.hashCode() and spreads (XORs) higher bits of hash * to lower. Because the table uses power-of-two masking, sets of * hashes that vary only in bits above the current mask will * always collide. (Among known examples are sets of Float keys * holding consecutive whole numbers in small tables.) So we * apply a transform that spreads the impact of higher bits * downward. There is a tradeoff between speed, utility, and * quality of bit-spreading. Because many common sets of hashes * are already reasonably distributed (so don't benefit from * spreading), and because we use trees to handle large sets of * collisions in bins, we just XOR some shifted bits in the * cheapest possible way to reduce systematic lossage, as well as * to incorporate impact of the highest bits that would otherwise * never be used in index calculations because of table bounds. * 计算key.hashCode()并将哈希的较高位(XOR)扩大为较低。 * 因为该表应用2的幂次掩码,因而仅在以后掩码上方的位中发生变化的哈希集将始终发生冲突。 * (家喻户晓的示例是在小表中蕴含间断整数的Float键集。)因而,咱们利用了一种变换, * 将向下扩大较高位的影响。 在速度,实用性和位扩大品质之间须要衡量。 * 因为许多常见的哈希集曾经正当散布(因而无奈从扩大中受害), * 并且因为咱们应用树来解决容器中的大量抵触,因而咱们仅以最便宜的形式对一些移位后的位进行XOR, * 以缩小零碎损失, 以及合并最高位的影响,否则因为表范畴的限度,这些位将永远不会在索引计算中应用。 */ static final int hash(Object key) { int h; //为什么^ (h >>> 16) 更加散列,性能考量,不便位运算 //不便在put,get办法中(n - 1) & hash计算数组下标 return (key == null) ? 0 : (h = key.hashCode()) ^ (h >>> 16); //HashMap中key值能够为null, 看到0那么咱们能够判断null值肯定存储在数组的第一个地位 }
hashmap->put()
次要逻辑:
以下是源代码(带正文):
/** * Associates the specified value with the specified key in this map. * If the map previously contained a mapping for the key, the old * value is replaced. * //将指定的值与此映射中的指定键关联。如果该映射先前蕴含了该键的映射,则旧值将被替换。 * * @param key key with which the specified value is to be associated * @param value value to be associated with the specified key * @return the previous value associated with <tt>key</tt>, or * <tt>null</tt> if there was no mapping for <tt>key</tt>. * (A <tt>null</tt> return can also indicate that the map * previously associated <tt>null</tt> with <tt>key</tt>.) */ public V put(K key, V value) { //把key先去hash一下拿到hash值 return putVal(hash(key), key, value, false, true); } /** * Implements Map.put and related methods. * * @param hash hash for key * @param key the key * @param value the value to put * @param onlyIfAbsent if true, don't change existing value //if true 则不扭转原来存在的值 * @param evict if false, the table is in creation mode.//if false 则表处于创立模式 * @return previous value, or null if none */ final V putVal(int hash, K key, V value, boolean onlyIfAbsent, boolean evict) { //数组+链表+红黑树,链表型(Node泛型)数组,每一个元素代表一条链表,则每个元素称为桶 //HashMap 的每一个元素,都是链表的一个节点(Entry<K,V>)这里也就是Node<K,V> //tab:桶 p:桶 n:哈希表数组大小 i:数组下标(桶的地位) Node<K,V>[] tab; Node<K,V> p; int n, i; //1.判断以后桶是否为空,空的就调用resize()办法(resize 中会判断是否进行初始化) if ((tab = table) == null || (n = tab.length) == 0) n = (tab = resize()).length; //2.判断是否有hash抵触,依据入参key与key的hash值找到具体的桶并判空,空则无抵触 间接新建桶 //?为什么采纳(n - 1) & hash计算数组下标,感兴趣的能够深刻理解 if ((p = tab[i = (n - 1) & hash]) == null) tab[i] = newNode(hash, key, value, null); //3.以下示意有抵触,解决hash抵触 else { Node<K,V> e; K k;//均为长期变量 //4.判断以后桶的key是否与入参key统一,统一则存在,把以后桶p赋值给e,笼罩原 value 在步骤10进行 if (p.hash == hash && ((k = p.key) == key || (key != null && key.equals(k)))) e = p; //5.如果以后的桶为红黑树,用putTreeVal()办法写入 赋值给e else if (p instanceof TreeNode) e = ((TreeNode<K,V>)p).putTreeVal(this, tab, hash, key, value); //6.则以后的桶是链表 遍历链表 else { for (int binCount = 0; ; ++binCount) { if ((e = p.next) == null) { //7.尾插法,链表下一个节点是null(链表开端),就new一个新节点写入到以后链表节点的前面 p.next = newNode(hash, key, value, null); //8.判断是否大于阈值,须要链表转红黑树 if (binCount >= TREEIFY_THRESHOLD - 1) // -1 for 1st //binCount从0开始的, 所以当binCount为7时,链表长度为8(算上数组槽位开始的那个节点,总长度为9),则须要树化桶 treeifyBin(tab, hash); break; } //9.与步骤4统一,如果链表中key存在则间接跳出 步骤10笼罩原值 if (e.hash == hash && ((k = e.key) == key || (key != null && key.equals(k)))) break; p = e; } } //10.存在雷同的key的Node节点,则笼罩原value if (e != null) { // existing mapping for key V oldValue = e.value; //onlyIfAbsent为true:不扭转原来的值 ;false: 扭转原来的值 if (!onlyIfAbsent || oldValue == null) e.value = value; //LinkedHashMap用到的回调办法 afterNodeAccess(e); return oldValue; } } /*记录批改次数标识 用于fast-fail,因为HashMap非线程平安,在对HashMap进行迭代时, 如果期间其余线程的参加导致HashMap的构造发生变化了(比方put,remove等操作), 须要抛出异样ConcurrentModificationException */ ++modCount; //11.容量超过阈值,扩容 if (++size > threshold) resize(); //LinkedHashMap用到的回调办法 afterNodeInsertion(evict); return null; }
那么咱们来总结一下put办法:
1、开始,入参key、value2、判断以后table是否为空或者length=0? 是,去扩容,(resize()办法中有判断是否初始化) 否,依据key算出hash值并失去插入的数组的索引 判断找到的这个table[i]是否为空? 是,直接插入,再到步骤 否,判断key是否存在? 是,间接笼罩对应的value,再到步骤3 否,去判断以后这个table[i]是不是treeNode? 是,应用红黑树的形式插入key、value 否,开始遍历链表筹备插入 判断链表长度是不是大于8? 是,链表转红黑树插入key、value 否,以链表的形式插入key、value如果key存在就间接笼罩对应value3、判断map的size()是否大于阈值? 是 就去扩容resize()4、完结
hashmap->get()
public V get(Object key) { Node<K,V> e; return (e = getNode(hash(key), key)) == null ? null : e.value; } /** * Implements Map.get and related methods. * * @param hash hash for key * @param key the key * @return the node, or null if none */ final Node<K,V> getNode(int hash, Object key) { Node<K,V>[] tab; Node<K,V> first, e; int n; K k; //1、判断以后数组不为空并长度大于0 && 由key的hash值找到对应数组下的桶(可能是红黑树或者链表) if ((tab = table) != null && (n = tab.length) > 0 && (first = tab[(n - 1) & hash]) != null) { //2、先判断桶的第一个节点 如果key统一 返回 if (first.hash == hash && // always check first node ((k = first.key) == key || (key != null && key.equals(k)))) return first; //3、再判空下一个节点不为空 && 判断是红黑树还算链表 if ((e = first.next) != null) { //4、如果是红黑树 则按红黑树形式取值 if (first instanceof TreeNode) return ((TreeNode<K,V>)first).getTreeNode(hash, key); //否则就是链表,遍历取值 do { if (e.hash == hash && ((k = e.key) == key || (key != null && key.equals(k)))) return e; } while ((e = e.next) != null); } } return null; }
那么咱们来总结一下get()办法:
1、开始,入参key2、判断以后的数组长度不为空&&length>0 是 return null; 否,去判断第一个节点,如果key合乎,返回 再去判断下一个节点是否为空 是,return null 否,判断否是红黑树? 是,按红黑树的形式取值 否,遍历链表取值3、完结
hashmap->resize()
/** * Initializes or doubles table size. If null, allocates in * accord with initial capacity target held in field threshold. * Otherwise, because we are using power-of-two expansion, the * elements from each bin must either stay at same index, or move * with a power of two offset in the new table. *初始化或减少表大小。 如果为空,则依据字段阈值中放弃的初始容量指标进行调配。 * 否则,因为咱们应用的是2的幂,所以每个bin中的元素必须放弃雷同的索引,或者在新表中以2的幂偏移。 * @return the table */ final Node<K,V>[] resize() { Node<K,V>[] oldTab = table; int oldCap = (oldTab == null) ? 0 : oldTab.length; int oldThr = threshold; int newCap, newThr = 0; //1、原数组扩容 if (oldCap > 0) { //如果原数组长度大于最大容量,把阈值调最大,return if (oldCap >= MAXIMUM_CAPACITY) { threshold = Integer.MAX_VALUE; return oldTab; } //把原数组大小、阈值都扩充一倍 else if ((newCap = oldCap << 1) < MAXIMUM_CAPACITY && oldCap >= DEFAULT_INITIAL_CAPACITY) newThr = oldThr << 1; // double threshold } //应用了指定initialCapacity的构造方法,则用原阈值作为新容量 else if (oldThr > 0) // initial capacity was placed in threshold newCap = oldThr; //应用空参结构,用默认值 else { // zero initial threshold signifies using defaults newCap = DEFAULT_INITIAL_CAPACITY;//16 newThr = (int)(DEFAULT_LOAD_FACTOR * DEFAULT_INITIAL_CAPACITY);//0.75*16=12 } //应用了指定initialCapacity的构造方法,新阈值为0,则计算新的阈值 if (newThr == 0) { float ft = (float)newCap * loadFactor; newThr = (newCap < MAXIMUM_CAPACITY && ft < (float)MAXIMUM_CAPACITY ? (int)ft : Integer.MAX_VALUE); } threshold = newThr; @SuppressWarnings({"rawtypes","unchecked"}) //2、用新的数组容量大小初始化数组 Node<K,V>[] newTab = (Node<K,V>[])new Node[newCap]; //如果仅仅是初始化过程,到此结束 return newTab table = newTab; //3、开始扩容的次要工作,数据迁徙 if (oldTab != null) { //遍历原数组开始复制旧数据 for (int j = 0; j < oldCap; ++j) { Node<K,V> e; if ((e = oldTab[j]) != null) { oldTab[j] = null;//革除旧表引有 //原数组中单个元素,间接复制到新表 if (e.next == null) newTab[e.hash & (newCap - 1)] = e; //如果该元素类型是红黑树,按红黑树形式解决 else if (e instanceof TreeNode) ((TreeNode<K,V>)e).split(this, newTab, j, oldCap); else { // preserve order //这段代码设计奇妙,环环相扣啊 //先定义了两种类型的链表 以及头尾节点 高位链表与低位链表 Node<K,V> loHead = null, loTail = null; Node<K,V> hiHead = null, hiTail = null; Node<K,V> next; //按程序遍历原链表的节点 do { next = e.next; //这是一个外围的判断条件,感兴趣的能够深刻理解?为什么这么做 //=0则放到低位链表 if ((e.hash & oldCap) == 0) { if (loTail == null) loHead = e; else loTail.next = e; loTail = e; } //否则放到高位链表 else { if (hiTail == null) hiHead = e; else hiTail.next = e; hiTail = e; } //以上实际上就是对原来链表拆分成了两个高下位链表 } while ((e = next) != null); //把整个低位链表放到新数组的j地位 if (loTail != null) { loTail.next = null; newTab[j] = loHead; } //把整个高位链表放到新数组的j+oldCap地位上 if (hiTail != null) { hiTail.next = null; newTab[j + oldCap] = hiHead; } } } } } return newTab; }
总结一下resize()办法:
1、开始,拿到原数组2、对原数组扩容 2.1如果原数组中的容量到最大,不再扩容,return原数组 2.2把原数组容量大小与阈值都扩充一倍3、如初始化用的指定initialCapacity的构造方法,则用原阈值作为新容量4、如初始化时候用的空参结构,用默认容量与默认阈值5、如初始化用的指定initialCapacity的构造方法,阈值=0,计算新的阈值6、用新的容量初始化数组,如果是初始化,完结返回新数组7、开始扩容,做数据迁徙 7.1遍历原数组copy数据到新数组 7.1.1如数组中只有一个元素,则间接复制 7.1.2如元素是红黑树数类型,则按红黑树的形式解决 7.1.3对原数组的链表进行解决 定义一个高位链表、一个低位链表(对原链表拆分) 开启一个循环,遍历原链表 判断条件e.hash & oldCap == 0? 是,把这些链表节点放到低位链表 否,放到高位链表 循环完结,遍历链表实现 把整个低位链表放到新数组j地位 把整个高位链表放到新数组j+oldCap地位 7.2循环完结,遍历旧数组实现8、返回新数组
我想这会浏览完结之后应该对HashMap有了肯定的意识,心愿能在面试或者工作中帮到您!
Hongchen闲聊
看起来是不是有点酷,它的名字叫”茑屋“,北京温榆河公园。
上周跟女友在家待闷了出门转转,于是咱们找了个公园,秋冬的公园还是比拟萧瑟。
年底了,疫情零零散散,心愿早点过来!peaceeeee
感激浏览
满腹经纶,如有你发现有谬误的中央,能够在后盾或者评论提出来,我会加以批改。
感谢您的浏览,欢送并感激关注!