Java流 – Java并行流
流可以是顺序的或并行的。
顺序流上的操作由一个线程串行处理。
使用多个线程并行处理并行流上的操作。
默认情况下,Streams API中的大多数方法都会生成顺序流。要从集合(如列表或集合)创建并行流,调用Collection接口的parallelStream()方法。
对流使用parallel()方法将顺序流转换为并行流。
对流使用sequential()方法将并行流转换为顺序流。
例子
以下代码显示如何对流进行串行处理:
import java.time.LocalDate; import java.time.Month; import java.util.Arrays; import java.util.List; import java.util.stream.Collectors; public class Main { public static void main(String[] args) { String names = Employee.persons() .stream() .filter(Employee::isMale) .map(Employee::getName) .collect(Collectors.joining(", ")); System.out.println(names); names = Employee.persons() .parallelStream() .filter(Employee::isMale) .map(Employee::getName) .collect(Collectors.joining(", ")); System.out.println(names); } } class Employee { public static enum Gender { MALE, FEMALE } private long id; private String name; private Gender gender; private LocalDate dob; private double income; public Employee(long id, String name, Gender gender, LocalDate dob, double income) { this.id = id; this.name = name; this.gender = gender; this.dob = dob; this.income = income; } public String getName() { return name; } public boolean isMale() { return this.gender == Gender.MALE; } public static List<Employee> persons() { Employee p1 = new Employee(1, "Jake", Gender.MALE, LocalDate.of(1971, Month.JANUARY, 1), 2343.0); Employee p2 = new Employee(2, "Jack", Gender.MALE, LocalDate.of(1972, Month.JULY, 21), 7100.0); Employee p3 = new Employee(3, "Jane", Gender.FEMALE, LocalDate.of(1973, Month.MAY, 29), 5455.0); Employee p4 = new Employee(4, "Jode", Gender.MALE, LocalDate.of(1974, Month.OCTOBER, 16), 1800.0); Employee p5 = new Employee(5, "Jeny", Gender.FEMALE, LocalDate.of(1975, Month.DECEMBER, 13), 1234.0); Employee p6 = new Employee(6, "Jason", Gender.MALE, LocalDate.of(1976, Month.JUNE, 9), 3211.0); List<Employee> persons = Arrays.asList(p1, p2, p3, p4, p5, p6); return persons; } @Override public String toString() { String str = String.format("(%s, %s, %s, %s, %.2f)\n", id, name, gender, dob, income); return str; } }
上面的代码生成以下结果。
例2
以下代码显示如何混合串行和并行流:
import java.time.LocalDate; import java.time.Month; import java.util.Arrays; import java.util.List; import java.util.stream.Collectors; public class Main { public static void main(String[] args) { String names = Employee.persons() // The data source .stream() // Produces a sequential stream .filter(Employee::isMale) // Processed in serial .parallel() // Produces a parallel stream .map(Employee::getName) // Processed in parallel .collect(Collectors.joining(", ")); // Processed in parallel System.out.println(names); } } class Employee { public static enum Gender { MALE, FEMALE } private long id; private String name; private Gender gender; private LocalDate dob; private double income; public Employee(long id, String name, Gender gender, LocalDate dob, double income) { this.id = id; this.name = name; this.gender = gender; this.dob = dob; this.income = income; } public String getName() { return name; } public boolean isMale() { return this.gender == Gender.MALE; } public static List<Employee> persons() { Employee p1 = new Employee(1, "Jake", Gender.MALE, LocalDate.of(1971, Month.JANUARY, 1), 2343.0); Employee p2 = new Employee(2, "Jack", Gender.MALE, LocalDate.of(1972, Month.JULY, 21), 7100.0); Employee p3 = new Employee(3, "Jane", Gender.FEMALE, LocalDate.of(1973, Month.MAY, 29), 5455.0); Employee p4 = new Employee(4, "Jode", Gender.MALE, LocalDate.of(1974, Month.OCTOBER, 16), 1800.0); Employee p5 = new Employee(5, "Jeny", Gender.FEMALE, LocalDate.of(1975, Month.DECEMBER, 13), 1234.0); Employee p6 = new Employee(6, "Jason", Gender.MALE, LocalDate.of(1976, Month.JUNE, 9), 3211.0); List<Employee> persons = Arrays.asList(p1, p2, p3, p4, p5, p6); return persons; } }
上面的代码生成以下结果。