java8 使用Map中的computeIfAbsent方法构建本地缓存，提高程序效率

一、概念及使用介绍 

     在JAVA8的Map接口中，增加了一个方法computeIfAbsent，此方法签名如下：

public V computeIfAbsent(K key, Function<? super K,? extends V> mappingFunction)

     Map接口的实现类如HashMap,ConcurrentHashMap,HashTable等继承了此方法，通过此方法可以构建JAVA本地缓存，降低程序的计算量，程序的复杂度，使代码简洁，易懂。

     此方法首先判断缓存MAP中是否存在指定key的值，如果不存在，会自动调用mappingFunction(key)计算key的value，然后将key = value放入到缓存Map,java8会使用thread-safe的方式从cache中存取记录。

    如果mappingFunction(key)返回的值为null或抛出异常，则不会有记录存入map

二 代码样例

import java.util.HashMap;import java.util.HashSet;import java.util.Map;import java.util.concurrent.ConcurrentHashMap;import java.util.concurrent.ExecutorService;import java.util.concurrent.Executors;import java.util.concurrent.TimeUnit;public class Main {static Map<Integer, Integer> cache = new ConcurrentHashMap<>();public static void main(String[] args) throws InterruptedException {cache.put(0, 0);cache.put(1, 1);// 普通方式System.out.println("Fibonacci(7) = " + fibonacci(7));// 采用java7的同步线程方式及java8的本地缓存的方式System.out.println("FibonacciJava8(7) = " + fibonacciJava8(7));System.out.println("FibonacciJava7(7) = " + fibonacciJava7(7));// 构建多值Map样例代码Map<String, HashSet<String>> map1 = new HashMap<>();map1.computeIfAbsent("fruits", k -> genValue(k)).add("apple");map1.computeIfAbsent("fruits", k -> genValue(k)).add("orange");map1.computeIfAbsent("fruits", k -> genValue(k)).add("pear");map1.computeIfAbsent("fruits", k -> genValue(k)).add("banana");map1.computeIfAbsent("fruits", k -> genValue(k)).add("water");System.out.println(map1);//测试多线程并发处理，是否同步操作Map<String, String> map2 = new ConcurrentHashMap<>();ExecutorService exec = Executors.newCachedThreadPool();for (int i = 0; i < 5; i++) {exec.execute(() -> {map2.computeIfAbsent("name", k -> genValue2(k));map2.computeIfAbsent("addr", k -> genValue2(k));map2.computeIfAbsent("email", k -> genValue2(k));map2.computeIfAbsent("mobile", k -> genValue2(k));});}exec.shutdown();exec.awaitTermination(1, TimeUnit.SECONDS);System.out.println(map2);}static HashSet<String> genValue(String str) {return new HashSet<String>();}static String genValue2(String str) {System.out.println("===");return str + "2";}/** * 普通的实现方式 普通方式使用大量的计算，存在性能问题. 并且计算量随着n的增加呈指数级增加，需要用到一些缓存策略，并且是线程安全的. *  * @param n * @return */static int fibonacci(int n) {if (n == 0 || n == 1)return n;System.out.println("calculating Fibonacci(" + n + ")");return fibonacci(n - 2) + fibonacci(n - 1);}/** * 采用java8的本地缓存方式 如果缓存MAP中不存在指定key的值，会自动调用mappingFunction(key)计算key的value * 然后将key = value放入到缓存Map,java8会使用thread-safe的方式从cache中存取记录 *  * @param n * @return */static int fibonacciJava8(int n) {return cache.computeIfAbsent(n, (key) -> {System.out.println("calculating FibonacciJava8 " + n);return fibonacciJava8(n - 2) + fibonacciJava8(n - 1);});}/** * 在java7中的实现方式 * 在java7中，通过synchronized进行线程同步，检查缓存是否存在key对应的值，如果不存在才进行计算并放入缓存中 * 为了更好的性能，需要使用 double-checked locking，那样代码会更复杂 *  * @param n * @return */static int fibonacciJava7(int n) {if (n == 0 || n == 1)return n;Integer result = cache.get(n);if (result == null) {synchronized (cache) {result = cache.get(n);if (result == null) {System.out.println("calculating FibonacciJava7(" + n + ")");result = fibonacciJava7(n - 2) + fibonacciJava7(n - 1);cache.put(n, result);}}}return result;}}

三，程序运行结果

calculating Fibonacci(7)calculating Fibonacci(5)calculating Fibonacci(3)calculating Fibonacci(2)calculating Fibonacci(4)calculating Fibonacci(2)calculating Fibonacci(3)calculating Fibonacci(2)calculating Fibonacci(6)calculating Fibonacci(4)calculating Fibonacci(2)calculating Fibonacci(3)calculating Fibonacci(2)calculating Fibonacci(5)calculating Fibonacci(3)calculating Fibonacci(2)calculating Fibonacci(4)calculating Fibonacci(2)calculating Fibonacci(3)calculating Fibonacci(2)Fibonacci(7) = 13calculating FibonacciJava8 7calculating FibonacciJava8 5calculating FibonacciJava8 3calculating FibonacciJava8 2calculating FibonacciJava8 4calculating FibonacciJava8 6FibonacciJava8(7) = 13FibonacciJava7(7) = 13{fruits=[orange, banana, apple, pear, water]}============{name=name2, mobile=mobile2, addr=addr2, email=email2}

四、参考

http://www.java8.org/caching-with-ConcurrentHashMap-in-java-8.html

JDK8 API

http://stackoverflow.com/questions/19278443/how-do-i-use-the-new-computeifabsent-function