如何更优雅的使用 Java8 进行分组（尤其是多个字段的组合分组）。

在SQL中经常会用到分组，我们也常常遇到一些组合分组的场景。

有下面的一个User类

import lombok.Builder;
import lombok.Data;
import java.time.LocalDateTime;@Data
@Builder
public class User {private String name;private int id;private String city;private String sex;private LocalDateTime birthDay;
}

java8分组传统写法（单个字段分组）

场景：根据城市进行分组

使用的是方法引用：User::getCity 来完成分组

import java.time.LocalDateTime;
import java.time.format.DateTimeFormatter;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;public class Demo2 {public static void main(String[] args) {DateTimeFormatter df = DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss");// data listList<User> userList = Arrays.asList(User.builder().id(123456).name("Zhang, San").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(777777).name("Zhang, San").city("ShangHai").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(888888).name("Li, Si").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(999999).name("Zhan, San").city("HangZhou").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(555555).name("Li, Si").city("NaJin").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build());Map<String, List<User>> groupMap = userList.stream().collect(Collectors.groupingBy(User::getCity));groupMap.forEach((k, v) -> {System.out.println(k);System.out.println(v);});}
}

java8分组传统写法（多个字段分组）`①`

场景：根据 `城市,性别`进行分组

一般的写法会是下面的这种写法，通过lambda表达式将key的生成逻辑传入进去：u -> u.getCity() + "|" + u.getSex() 来实现分组的效果。

import java.time.LocalDateTime;
import java.time.format.DateTimeFormatter;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;public class Demo2 {public static void main(String[] args) {DateTimeFormatter df = DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss");// data listList<User> userList = Arrays.asList(User.builder().id(123456).name("Zhang, San").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(777777).name("Zhang, San").city("ShangHai").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(888888).name("Li, Si").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(999999).name("Zhan, San").city("HangZhou").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(555555).name("Li, Si").city("NaJin").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build());Map<String, List<User>> groupMap = userList.stream().collect(Collectors.groupingBy(u -> u.getCity() + "|" + u.getSex()));groupMap.forEach((k, v) -> {System.out.println(k);System.out.println(v);});}
}

分析：`多个分组条件与单个分组条件两种写法`

单个条件的分组用的比较多，userList.stream().collect(Collectors.groupingBy(User::getCity));
这种方法引用的方式看起来很清爽。
在我们遇到多个字段的分组的时候，我并不太想使用前面那种传统的写法①。

我在想，既然单个字段的分组写法是：
userList.stream().collect(Collectors.groupingBy(User::getCity));
那么多个字段的写法可否是下面这种( 类推 )，传入多个方法引用！
userList.stream().collect(Collectors.groupingBy(User::getCity,User::getSex));

很可惜 jdk 类库中Collectors 没有提供这种写法

多个字段的优雅写法

因为jdk没有提供这种写法，于是自己就想写了一个Util来帮助我们使用多个方法引用的方式完成组合分组

MyBeanUtil groupingBy(userList, User::getSex, User::getCity);

Demo：

import java.time.LocalDateTime;
import java.time.format.DateTimeFormatter;
import java.util.*;
import java.util.function.Function;
import java.util.stream.Collectors;public class MyBeanUtil {public static void main(String[] args) {DateTimeFormatter df = DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss");// data listList<User> userList = Arrays.asList(User.builder().id(123456).name("Zhang, San").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(777777).name("Zhang, San").city("ShangHai").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(888888).name("Li, Si").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(999999).name("Zhan, San").city("HangZhou").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(555555).name("Li, Si").city("NaJin").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build());// 进行分组，根据名字和城市分组Map<String, List<User>> groupMap = groupingBy(userList, User::getSex, User::getCity);//打印分组结果groupMap.forEach((k, v) -> {System.out.println(k);System.out.println(v);});}/*** 将数据分组，根据方法引用（bean的get方法）** @param list      为分组的数据* @param functions get方法数组*/@SafeVarargspublic static <T, R> Map<String, List<T>> groupingBy(List<T> list, Function<T, R>... functions) {return list.stream().collect(Collectors.groupingBy(t -> groupingBy(t, functions)));}/*** 分组工具根据函数式接口使用分组，将数据根据分组结果进行拆分*/@SafeVarargspublic static <T, R> String groupingBy(T t, Function<T, R>... functions) {if (functions == null || functions.length == 0) {throw new NullPointerException("functions数组不可以为空");} else if (functions.length == 1) {return functions[0].apply(t).toString();} else {return Arrays.stream(functions).map(fun -> fun.apply(t).toString()).reduce((str1, str2) -> str1 + "|" + str2).get();}}
}

再度优化

依然不是很满足这种写法，因为这种写法需要借助 Util 类，不够接地气！
我更希望是下面这种接地气的写法：能够完全集成在jdk类库中

userList.stream().collect(Collectors.groupingBy(User::getCity,User::getSex));

为了达到上述的效果，那么显然我们是需要修改jdk源代码的；
于是我就将java.util.stream.Collectors源码完整copy出来，然后加入下面3个方法

    public static <T, K> Collector<T, ?, HashMap<K, List<T>>>groupingBy(Function<? super T, ? extends K>... classifier) {return groupingBy("|", classifier);}public static <T, K> Collector<T, ?, HashMap<K, List<T>>>groupingBy(String split, Function<? super T, ? extends K>... classifier) {return groupingBy(split, classifier, HashMap::new, toList());}public static <T, K, D, A, M extends Map<? super K, D>>Collector<T, ?, M> groupingBy(String split,Function<? super T, ? extends K>[] classifierArr,Supplier<M> mapFactory,Collector<? super T, A, D> downstream) {Supplier<A> downstreamSupplier = downstream.supplier();BiConsumer<A, ? super T> downstreamAccumulator = downstream.accumulator();BiConsumer<Map<K, A>, T> accumulator = (m, t) -> {String key = Arrays.stream(classifierArr).map(classifier -> Objects.requireNonNull(classifier.apply(t))).map(String::valueOf).reduce((s1, s2) -> s1 + split + s2).get();A container = m.computeIfAbsent((K) key, k -> downstreamSupplier.get());downstreamAccumulator.accept(container, t);};BinaryOperator<Map<K, A>> merger = Collectors.<K, A, Map<K, A>>mapMerger(downstream.combiner());@SuppressWarnings("unchecked")Supplier<Map<K, A>> mangledFactory = (Supplier<Map<K, A>>) mapFactory;if (downstream.characteristics().contains(Collector.Characteristics.IDENTITY_FINISH)) {return new CollectorImpl<>(mangledFactory, accumulator, merger, CH_ID);} else {@SuppressWarnings("unchecked")Function<A, A> downstreamFinisher = (Function<A, A>) downstream.finisher();Function<Map<K, A>, M> finisher = intermediate -> {intermediate.replaceAll((k, v) -> downstreamFinisher.apply(v));@SuppressWarnings("unchecked")M castResult = (M) intermediate;return castResult;};return new CollectorImpl<>(mangledFactory, accumulator, merger, finisher, CH_NOID);}}

就达到了我们预期的效果，为了方便大家也一起体验一下，我已经将demo完整的放到了github上

源码地址：https://github.com/1015770492/CollectorsDemo
下载好源码后，找到下面这个类

Demo：

import java.io.Serializable;
import java.time.LocalDateTime;
import java.time.format.DateTimeFormatter;
import java.util.*;public class MultiGroupByDemo {public static void main(String[] args) {DateTimeFormatter df = DateTimeFormatter.ofPattern("yyyy-MM-dd HH:mm:ss");// data listList<User> userList = Arrays.asList(User.builder().id(123456).name("Zhang, San").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(777777).name("Zhang, San").city("ShangHai").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(888888).name("Li, Si").city("ShangHai").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(999999).name("Zhan, San").city("HangZhou").sex("woman").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build(),User.builder().id(555555).name("Li, Si").city("NaJin").sex("man").birthDay(LocalDateTime.parse("2022-07-01 12:00:00", df)).build());/** maybe we can*/// 1.Use the default vertical separatorSystem.out.println("Use the default vertical separator:");HashMap<String, List<User>> defaultSpilt = userList.stream().collect(Collectors.groupingBy(User::getName, User::getCity));printMap(defaultSpilt);System.out.println();// 2.Use custom delimitersSystem.out.println("Use custom delimiters:");userList.stream().collect(Collectors.groupingBy("--", User::getName, User::getCity, User::getId));HashMap<? extends Serializable, List<User>> collect = userList.stream().collect(Collectors.groupingBy("--", User::getName, User::getCity, User::getId));printMap(collect);System.out.println();// 3.Use custom delimitersSystem.out.println("Use custom delimiters:");userList.stream().collect(Collectors.groupingBy("--", User::getName, User::getCity, User::getId));HashMap<? extends Serializable, List<User>> collect2 = userList.stream().collect(Collectors.groupingBy(User::getName, User::getCity, User::getBirthDay));printMap(collect2);}public static <T> void printMap(Map<? extends Serializable, List<T>> map){map.forEach((k, v) -> {System.out.println(k);System.out.println(v);});}}

最后我希望这个特性能被JDK所吸收，这样可以方便大家更好的使用这些好用的特性