Java lambda表达式实现List转Map

将List分组转成Map是日常开发中常见的需求，我们就来总结一下常见的几种写法。

有如下代码：

@Data
class Person {
    private String uuid;
    private String name;
    private String gender;
    private int age;

    public Person(String name, String gender, int age) {
        this.uuid = UUID.randomUUID().toString();
        this.name = name;
        this.gender = gender;
        this.age = age;
    }
}

和一个List集合：

List<Person> persons = new ArrayList<>();
persons.add(new Person("张三", "男", 27));
persons.add(new Person("李四", "男", 14));
persons.add(new Person("王五", "女", 17));
persons.add(new Person("赵六", "女", 34));

1. partitioningBy 分两组

partitioningBy要求传入一个Predicate，会按照满足条件和不满足条件分成两组，得到的结果是Map<Boolean, List<T>>结构，比如我们按是否未成年分成两组：

Map<Boolean, List<Person>> personsByAge = persons.stream()
	.collect(Collectors.partitioningBy(p -> p.getAge() > 18));
System.out.println(JSON.toJSONString(personsByAge));

Output：

{
	false: [{
		"age": 14,
		"gender": "男",
		"name": "李四",
		"uuid": "9fc3be98-f676-42a4-9f02-ebdab328103a"
	}, {
		"age": 17,
		"gender": "女",
		"name": "王五",
		"uuid": "3621044d-25a1-4946-a765-57b074f63f26"
	}],
	true: [{
		"age": 27,
		"gender": "男",
		"name": "张三",
		"uuid": "3f87ec59-29a1-4137-b95b-ae755f0e06ca"
	}, {
		"age": 34,
		"gender": "女",
		"name": "赵六",
		"uuid": "04ed8e9f-545b-49f5-a28b-ce0cccd15663"
	}]
}

2. groupingBy 分多组

比如按照性别进行分组，得到的是Map<String, List<T>>结构：

Map<String, List<Person>> personByGender = persons.stream()
	.collect(Collectors.groupingBy(Person::getGender));
System.out.println(JSON.toJSONString(personByGender));

Output：

{
	"女": [{
		"age": 17,
		"gender": "女",
		"name": "王五",
		"uuid": "feb8ca82-789f-445e-9e85-c14aa1d70546"
	}, {
		"age": 34,
		"gender": "女",
		"name": "赵六",
		"uuid": "6402b5ec-03cd-45d1-aa6d-7134509ca670"
	}],
	"男": [{
		"age": 27,
		"gender": "男",
		"name": "张三",
		"uuid": "e2c5ec58-5767-4807-8470-56a016dbc5eb"
	}, {
		"age": 14,
		"gender": "男",
		"name": "李四",
		"uuid": "d10aad57-038b-4ff8-8b36-86045d657c5a"
	}]
}

3. toMap 自定义<key, value>

前面介绍的partitioningBy和groupingBy返回Map的value部分都是List<T>结构的，有时我们需要value是对象的一个属性，比如我们想构造一个uuid到name的映射，以方便通过uuid快速获取人员的名字：

Map<String, String> uuidNameMap = persons.stream()
	.collect(Collectors.toMap(Person::getUuid, Person::getName));
System.out.println(JSON.toJSONString(uuidNameMap));

Output：

{
	"7a021022-fa62-4f57-bf33-873b8e030cc3": "王五",
	"e0bad9e6-2c3c-417e-9d27-3b321312421a": "张三",
	"895b0f95-b4fd-481e-ba6c-33f0b636e6cf": "李四",
	"fcb6f403-8489-4853-98c5-6f41341165ba": "赵六"
}

实际情况有可能同一个key会对应多个value，就有可能抛Duplicate key异常。这时可以传入第三个参数决定重复时如何选择，比如我们想构造<name, uuid>的映射，但是考虑可能有重名的可能，就可以这么做：

Map<String, String> nameUuidMap = persons.stream()
	.collect(Collectors.toMap(Person::getName, Person::getUuid, (p1, p2) -> p1));
System.out.println(JSON.toJSONString(nameUuidMap));

这里(p1, p2) -> p1表示如果重复则取前者。