「JDK8」stream api流式编程
2022/1/22 21:09:46
本文主要是介绍「JDK8」stream api流式编程,对大家解决编程问题具有一定的参考价值,需要的程序猿们随着小编来一起学习吧!
自带常用的函数式接口
- Predicate boolean test(T t) 传入一个参数返回boolean值
- Consumer void accept(T t) 传入一个参数,无返回值
- Function<T,R> R apply(T t) 传入一个参数,返回另一个类型
准备数据:
//计算机俱乐部 private static List<Student> computerClub = Arrays.asList( new Student("2015134001", "小明", 15, "1501"), new Student("2015134003", "小王", 14, "1503"), new Student("2015134006", "小张", 15, "1501"), new Student("2015134008", "小梁", 17, "1505") ); //篮球俱乐部 private static List<Student> basketballClub = Arrays.asList( new Student("2015134012", "小c", 13, "1503"), new Student("2015134013", "小s", 14, "1503"), new Student("2015134015", "小d", 15, "1504"), new Student("2015134018", "小y", 16, "1505") ); //乒乓球俱乐部 private static List<Student> pingpongClub = Arrays.asList( new Student("2015134022", "小u", 16, "1502"), new Student("2015134021", "小i", 14, "1502"), new Student("2015134026", "小m", 17, "1504"), new Student("2015134027", "小n", 16, "1504") ); private static List<List<Student>> allClubStu = new ArrayList<>(); allClubStu.add(computerClub); allClubStu.add(basketballClub); allClubStu.add(pingpongClub);
常用的stream三种创建方式
- 集合 Collection.stream()
- 静态方法 Stream.of
- 数组 Arrays.stream
//1.集合 Stream<Student> stream = basketballClub.stream(); //2.静态方法 Stream<String> stream2 = Stream.of("a", "b", "c"); //3.数组 String[] arr = {"a","b","c"}; Stream<String> stream3 = Arrays.stream(arr);
Stream的终止操作
- foreach(Consumer c) 遍历操作
- collect(Collector) 将流转化为其他形式
- max(Comparator) 返回流中最大值
- min(Comparator) 返回流中最小值
- count 返回流中元素数量
Collectors 具体方法
- toList List 把流中元素收集到List
- toSet Set 把流中元素收集到Set
- toCollection Coolection 把流中元素收集到Collection中
- groupingBy Map<K,List> 根据K属性对流进行分组
- partitioningBy Map<boolean, List> 根据boolean值进行分组
//此处只是演示 此类需求直接用List构造器即可 List<Student> collect = computerClub.stream().collect(Collectors.toList()); Set<Student> collect1 = pingpongClub.stream().collect(Collectors.toSet()); //注意key必须是唯一的 如果不是唯一的会报错而不是像普通map那样覆盖 Map<String, String> collect2 = pingpongClub.stream() .collect(Collectors.toMap(Student::getIdNum, Student::getName)); //分组 类似于数据库中的group by Map<String, List<Student>> collect3 = pingpongClub.stream() .collect(Collectors.groupingBy(Student::getClassNum)); //字符串拼接 第一个参数是分隔符 第二个参数是前缀 第三个参数是后缀 String collect4 = pingpongClub.stream().map(Student::getName).collect(Collectors.joining(",", "【", "】")); //【小u,小i,小m,小n】 //三个俱乐部符合年龄要求的按照班级分组 Map<String, List<Student>> collect5 = Stream.of(basketballClub, pingpongClub, computerClub) .flatMap(e -> e.stream().filter(s -> s.getAge() < 17)) .collect(Collectors.groupingBy(Student::getClassNum)); //按照是否年龄>16进行分组 key为true和false ConcurrentMap<Boolean, List<Student>> collect6 = Stream.of(basketballClub, pingpongClub, computerClub) .flatMap(Collection::stream) .collect(Collectors.groupingByConcurrent(s -> s.getAge() > 16));
Stream的中间操作
1、filter(Predicate) 筛选流中某些元素
//筛选1501班的学生 computerClub.stream().filter(e -> e.getClassNum().equals("1501")).forEach(System.out::println); //筛选年龄大于15的学生 List<Student> collect = computerClub.stream().filter(e -> e.getAge() > 15).collect(Collectors.toList());
2、map(Function f) 接收流中元素,并且将其映射成为新元素,例如从student对象中取name属性
//篮球俱乐部所有成员名 + 暂时住上商标^_^,并且获取所有队员名 List<String> collect1 = basketballClub.stream() .map(e -> e.getName() + "^_^") .collect(Collectors.toList()); collect1.forEach(System.out::println); //小c^_^^_^ //小s^_^^_^ //小d^_^^_^ //小y^_^^_^
3、flatMap(Function f) 将所有流中的元素并到一起连接成一个流
//获取年龄大于15的所有俱乐部成员 List<Student> collect2 = Stream.of(basketballClub, computerClub, pingpongClub) .flatMap(e -> e.stream().filter(s -> s.getAge() > 15)) .collect(Collectors.toList()); collect2.forEach(System.out::println); //用双层list获取所有年龄大于15的俱乐部成员 List<Student> collect3 = allClubStu.stream() .flatMap(e -> e.stream().filter(s -> s.getAge() > 15)) .collect(Collectors.toList()); collect3.forEach(System.out::println);
4、peek(Consumer c) 获取流中元素,操作流中元素,与foreach不同的是不会截断流,可继续操作流
//篮球俱乐部所有成员名 + 赞助商商标^_^,并且获取所有队员详细内容 List<Student> collect = basketballClub.stream() .peek(e -> e.setName(e.getName() + "^_^")) .collect(Collectors.toList()); collect.forEach(System.out::println); //Student{idNum='2015134012', name='小c^_^', age=13, classNum='1503'} //Student{idNum='2015134013', name='小s^_^', age=14, classNum='1503'} //Student{idNum='2015134015', name='小d^_^', age=15, classNum='1504'} //Student{idNum='2015134018', name='小y^_^', age=16, classNum='1505'}
5、distinct() 通过流所生成元素的equals和hashCode去重
6、limit(long val) 截断流,取流中前val个元素
7、sorted(Comparator) 产生一个新流,按照比较器规则排序
8、sorted() 产生一个新流,按照自然顺序排序
List<String> list = Arrays.asList("b","b","c","a"); list.forEach(System.out::print); //bbca List<String> collect = list.stream().distinct().sorted().collect(Collectors.toList()); collect.forEach(System.out::print);//abc //获取list中排序后的top2 即截断取前两个 List<String> collect1 = list.stream().distinct().sorted().limit(2).collect(Collectors.toList()); collect1.forEach(System.out::print);//ab
匹配
- booelan allMatch(Predicate) 都符合
- .boolean anyMatch(Predicate) 任一元素符合
- boolean noneMatch(Predicate) 都不符合
boolean b = basketballClub.stream().allMatch(e -> e.getAge() < 20); boolean b1 = basketballClub.stream().anyMatch(e -> e.getAge() < 20); boolean b2 = basketballClub.stream().noneMatch(e -> e.getAge() < 20);
寻找元素
- findFirst——返回第一个元素
- findAny——返回当前流中的任意元素
Optional<Student> first = basketballClub.stream().findFirst(); if (first.isPresent()) { Student student = first.get(); System.out.println(student); } Optional<Student> any = basketballClub.stream().findAny(); if (any.isPresent()) { Student student2 = any.get(); System.out.println(student2); } Optional<Student> any1 = basketballClub.stream().parallel().findAny(); System.out.println(any1);
计数和极值
- count——返回流中元素的总个数
- max——返回流中最大值
- min——返回流中最小值
long count = basketballClub.stream().count(); Optional<Student> max = basketballClub.stream().max(Comparator.comparing(Student::getAge)); if (max.isPresent()) { Student student = max.get(); } Optional<Student> min = basketballClub.stream().min(Comparator.comparingInt(Student::getAge)); if (min.isPresent()) { Student student = min.get(); }
总结
吃水不忘挖井人:
|
这篇关于「JDK8」stream api流式编程的文章就介绍到这儿,希望我们推荐的文章对大家有所帮助,也希望大家多多支持为之网!
- 2024-11-23Springboot应用的多环境打包入门
- 2024-11-23Springboot应用的生产发布入门教程
- 2024-11-23Python编程入门指南
- 2024-11-23Java创业入门:从零开始的编程之旅
- 2024-11-23Java创业入门:新手必读的Java编程与创业指南
- 2024-11-23Java对接阿里云智能语音服务入门详解
- 2024-11-23Java对接阿里云智能语音服务入门教程
- 2024-11-23JAVA对接阿里云智能语音服务入门教程
- 2024-11-23Java副业入门:初学者的简单教程
- 2024-11-23JAVA副业入门:初学者的实战指南