文章目录
前言
之前记录过：stream的三个常用方式（toMap,groupingBy,findFirst）。
这里继续记录下groupingBy的几个高级用法。
个人认为，最有分歧的是：实现分组，且每个分组也进行排序
这里要特别说明下：
从方法上看，分组之后，是不能对每个分组进行比较的（也就无法排序）
Collectors.groupingBy(Function<? super T, ? extends K> classifier,
Collector<? super T, A, D> downstream)
这里有两个思路：
1:提前排序，再进行分组
2:分组后，对Map进行处理，对其每个value排序
这主要列举了常用的：summingInt、counting、mapping、分组生序排序，分组降序排序
一、groupingBy高级用法
import com.alibaba.fastjson.JSONObject;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Comparator;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;
public class Test {
public static void main(String[] args) {
Student student1 = new Student(1, 1);
Student student2 = new Student(1, 1);
Student student3 = new Student(2, 2);
Student student4 = new Student(2, 3);
Student student5 = new Student(3, 3);
Student student6 = new Student(3, 4);
Student student7 = new Student(4, 1);
Student student8 = new Student(4, 1);
Student student9 = new Student(4, 2);
Student student10 = new Student(4, 1);
List<Student> list = Arrays.asList(student1, student2, student3, student4, student5, student6, student7, student8, student9, student10);
System.out.println("--------- 根据字段分组，求每个分组的sum ----------");
Map<Integer, Integer> collect =
list.stream().collect(Collectors.groupingBy(Student::getId,
Collectors.summingInt(Student::getScore)));
System.out.println(collect.toString());
System.out.println("--------- 根据字段分组，求每个分组的count ----------");
Map<Integer, Long> countMap = list.stream().collect(Collectors.groupingBy(Student::getId, Collectors.counting()));
System.out.println(countMap.toString());
System.out.println("--------- 根据字段分组，每个分组为：对象的指定字段 ----------");
Map<Integer, List<Integer>> groupMap = list.stream().collect(Collectors.groupingBy(Student::getId, Collectors.mapping(Student::getScore, Collectors.toCollection(ArrayList::new))));
System.out.println(groupMap.toString());
System.out.println("--------- 根据字段分组,默认分组 ----------");
Map<Integer, List<Student>> defaultGroupMap = list.stream().collect(Collectors.groupingBy(Student::getId));
System.out.println(JSONObject.toJSONString(defaultGroupMap));
System.out.println("--------- 根据字段分组，每个分组按照指定字段进行生序排序 ----------");
Map<Integer, List<Student>> sortGroupMap = list.stream().sorted(Comparator.comparing(Student::getScore))
.collect(Collectors.groupingBy(Student::getId));
System.out.println(JSONObject.toJSONString(sortGroupMap));
System.out.println("--------- 先排序，再分组 ----------");
Map<Integer, List<Student>> reversedSortGroupMap =list.stream().sorted(Comparator.comparing(Student::getScore).reversed())
.collect(Collectors.groupingBy(Student::getId));
System.out.println(JSONObject.toJSONString(reversedSortGroupMap));
}
}
class Student {
private Integer id;
private Integer score;
public Integer getId() {
return id;
}
public void setId(Integer id) {
this.id = id;
}
public Integer getScore() {
return score;
}
public void setScore(Integer score) {
this.score = score;
}
public Student(Integer id, Integer score) {
this.id = id;
this.score = score;
}
}
执行结果
--------- 根据字段分组，求每个分组的sum ----------
{
1=2, 2=5, 3=7, 4=5}
--------- 根据字段分组，求每个分组的count ----------
{
1=2, 2=2, 3=2, 4=4}
--------- 根据字段分组，每个分组为：对象的指定字段 ----------
{
1=[1, 1], 2=[2, 3], 3=[3, 4], 4=[1, 1, 2, 1]}
--------- 根据字段分组,默认分组 ----------
{
1:[{
"id":1,"score":1},{
"id":1,"score":1}],2:[{
"id":2,"score":2},{
"id":2,"score":3}],3:[{
"id":3,"score":3},{
"id":3,"score":4}],4:[{
"id":4,"score":1},{
"id":4,"score":1},{
"id":4,"score":2},{
"id":4,"score":1}]}
--------- 根据字段分组，每个分组按照指定字段进行生序排序 ----------
{
1:[{
"id":1,"score":1},{
"id":1,"score":1}],2:[{
"id":2,"score":2},{
"id":2,"score":3}],3:[{
"id":3,"score":3},{
"id":3,"score":4}],4:[{
"id":4,"score":1},{
"id":4,"score":1},{
"id":4,"score":1},{
"id":4,"score":2}]}
--------- 先排序，再分组 ----------
{
1:[{
"id":1,"score":1},{
"id":1,"score":1}],2:[{
"id":2,"score":3},{
"id":2,"score":2}],3:[{
"id":3,"score":4},{
"id":3,"score":3}],4:[{
"id":4,"score":2},{
"id":4,"score":1},{
"id":4,"score":1},{
"id":4,"score":1}]}
二、先分组，再排序
先将数据进行groupingBy，得到Map<Long, List>。
在通过forEach对每个List进行排序
Map<Long, List<StreamMapEntity>> map = getMap();
System.out.println("------ flatMap 处理Map<Long,List<StreamMapEntity>> ------");
System.out.println();
System.out.println("初始数据：" + JSONObject.toJSONString(map));
System.out.println();
map.keySet().forEach(key -> map.computeIfPresent(key, (k, v) -> v.stream().sorted(Comparator.comparing(StreamMapEntity::getId).reversed()).collect(Collectors.toList())));
System.out.println("处理map：对map内的每个list进行排序：");
System.out.println(JSONObject.toJSONString(map));
执行结果：
------ flatMap 处理Map<Long,List<StreamMapEntity>> ------
初始数据：{
1111:[{
"id":1,"name":"aa"},{
"id":2,"name":"bb"},{
"id":3,"name":"cc"},{
"id":4,"name":"dd"}],2222:[{
"id":5,"name":"ee"},{
"id":6,"name":"ff"},{
"id":7,"name":"gg"},{
"id":8,"name":"hh"}]}
处理map：对map内的每个list进行排序：
{
1111:[{
"id":4,"name":"dd"},{
"id":3,"name":"cc"},{
"id":2,"name":"bb"},{
"id":1,"name":"aa"}],2222:[{
"id":8,"name":"hh"},{
"id":7,"name":"gg"},{
"id":6,"name":"ff"},{
"id":5,"name":"ee"}]}
总结
关于分组排序的两种方式：先排序再分组；先分组再排序。
效率上肯定是先排序再分组快。