在 Java 8 中,Stream.reduce()
合并流的元素并产生单个值。
使用 for 循环的简单求和运算。
int[] numbers = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}; int sum = 0; for (int i : numbers) { sum += i; } System.out.println("sum : " + sum); // 55
相当于 Stream.reduce()
int[] numbers = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}; // 1st argument, init value = 0 int sum = Arrays.stream(numbers).reduce(0, (a, b) -> a + b); System.out.println("sum : " + sum); // 55
或方法引用 Integer::sum
int sum = Arrays.stream(numbers).reduce(0, Integer::sum); // 55
Integer.java
/** * Adds two integers together as per the + operator. * * @param a the first operand * @param b the second operand * @return the sum of {@code a} and {@code b} * @see java.util.function.BinaryOperator * @since 1.8 */ public static int sum(int a, int b) { return a + b; }
1. 方法签名
1.1 查看Stream.reduce()
方法签名:
Stream.java
T reduce(T identity, BinaryOperator<T> accumulator);
IntStream.java
int reduce(int identity, IntBinaryOperator op);
LongStream.java
long reduce(int identity, LongBinaryOperator op);
- identity = 默认值或初始值。
- BinaryOperator = 函数式接口,取两个值并产生一个新值。(注: java Function 函数中的 BinaryOperator 接口用于执行 lambda 表达式并返回一个 T 类型的返回值)
1.2 如果缺少identity
参数,则没有默认值或初始值,并且它返回 optional。
Stream.java
Optional<T> reduce(BinaryOperator<T> accumulator);
2. 更多例子
2.1 数学运算。
int[] numbers = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}; int sum = Arrays.stream(numbers).reduce(0, (a, b) -> a + b); // 55int sum2 = Arrays.stream(numbers).reduce(0, Integer::sum); // 55 int sum3 = Arrays.stream(numbers).reduce(0, (a, b) -> a - b); // -55int sum4 = Arrays.stream(numbers).reduce(0, (a, b) -> a * b); // 0, initial is 0, 0 * whatever = 0int sum5 = Arrays.stream(numbers).reduce(0, (a, b) -> a / b); // 0
2.2 最大和最小
int[] numbers = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}; int max = Arrays.stream(numbers).reduce(0, (a, b) -> a > b ? a : b); // 10int max1 = Arrays.stream(numbers).reduce(0, Integer::max); // 10 int min = Arrays.stream(numbers).reduce(0, (a, b) -> a < b ? a : b); // 0int min1 = Arrays.stream(numbers).reduce(0, Integer::min); // 0
2.3 连接字符串。
String[] strings = {"a", "b", "c", "d", "e"}; // |a|b|c|d|e , the initial | join is not what we want String reduce = Arrays.stream(strings).reduce("", (a, b) -> a + "|" + b); // a|b|c|d|e, filter the initial "" empty string String reduce2 = Arrays.stream(strings).reduce("", (a, b) -> { if (!"".equals(a)) { return a + "|" + b; } else { return b; } }); // a|b|c|d|e , better uses the Java 8 String.join :) (最好使用 Java 8 的 String.join) String join = String.join("|", strings);
3. Map & Reduce
一个简单的 map 和 reduce 示例,用于从发票 List 中求 BigDecimal 的和。
JavaReduce.java
package com.mkyong; import java.math.BigDecimal;import java.math.RoundingMode;import java.util.Arrays;import java.util.List; public class JavaReduce { public static void main(String[] args) { // 发票集合 List<Invoice> invoices = Arrays.asList( new Invoice("A01", BigDecimal.valueOf(9.99), BigDecimal.valueOf(1)), new Invoice("A02", BigDecimal.valueOf(19.99), BigDecimal.valueOf(1.5)), new Invoice("A03", BigDecimal.valueOf(4.99), BigDecimal.valueOf(2)) ); BigDecimal sum = invoices.stream() .map(x -> x.getQty().multiply(x.getPrice())) // map,对集合中的元素进行操作 .reduce(BigDecimal.ZERO, BigDecimal::add); // reduce,将上一步得到的结果进行合并得到最终的结果 System.out.println(sum); // 49.955 System.out.println(sum.setScale(2, RoundingMode.HALF_UP)); // 49.96 使用setScale方法进行四舍五入 } } class Invoice { // 发票号码 String invoiceNo; // 价格 BigDecimal price; // 数量 BigDecimal qty; // getters, stters n constructor}
输出
49.955
49.96