很难想象有Java开发人员不曾使用过Collection框架。在Collection框架中,主要使用的类是来自List接口中的ArrayList,以及来自Set接口的HashSet、TreeSet,我们经常处理这些Collections的排序。
在本文中,我将主要关注排序Collection的ArrayList、HashSet、TreeSet,以及最后但并非最不重要的数组。
让我们看看如何对给定的整数集合(5,10,0,-1)进行排序:
private void sortNumbersInArrayList() {
List<Integer> integers = new ArrayList<>();
integers.add(5);
integers.add(10);
integers.add(0);
integers.add(-1);
System.out.println("Original list: " +integers);
Collections.sort(integers);
System.out.println("Sorted list: "+integers);
Collections.sort(integers, Collections.reverseOrder());
System.out.println("Reversed List: " +integers);
}
输出:
Original list: [5, 10, 0, -1]
Sorted list: [-1, 0, 5, 10]
Reversed List: [10, 5, 0, -1]
private void sortNumbersInHashSet() {
Set<Integer> integers = new HashSet<>();
integers.add(5);
integers.add(10);
integers.add(0);
integers.add(-1);
System.out.println("Original set: " +integers);
// Collections.sort(integers); This throws error since sort method accepts list not collection
List list = new ArrayList(integers);
Collections.sort(list);
System.out.println("Sorted set: "+list);
Collections.sort(list, Collections.reverseOrder());
System.out.println("Reversed set: " +list);
}
输出:
Original set: [0, -1, 5, 10]
Sorted set: [-1, 0, 5, 10]
Reversed set: [10, 5, 0, -1]
在这个例子中(数据(整数)存储在HashSet中),我们看到HashSet被转换为ArrayList进行排序。在不转换为ArrayList的情况下,可以通过使用TreeSet来实现排序。TreeSet是Set的另一个实现,并且在使用默认构造函数创建Set时,使用自然排序进行排序。
private void sortNumbersInTreeSet() {
Set<Integer> integers = new TreeSet<>();
integers.add(5);
integers.add(10);
integers.add(0);
integers.add(-1);
System.out.println("Original set: " + integers);
System.out.println("Sorted set: "+ integers);
Set<Integer> reversedIntegers = new TreeSet(Collections.reverseOrder());
reversedIntegers.add(5);
reversedIntegers.add(10);
reversedIntegers.add(0);
reversedIntegers.add(-1);
System.out.println("Reversed set: " + reversedIntegers);
}
输出:
Original set: [-1, 0, 5, 10]
Sorted set: [-1, 0, 5, 10]
Reversed set: [10, 5, 0, -1]
在这种情况下,“Original set:”和“Sorted set:”两者相同,因为我们已经使用了按排序顺序存储数据的TreeSet,所以在插入后不用排序。
到目前为止,一切都如期工作,排序似乎是一件轻而易举的事。现在让我们尝试在各个Collection中存储自定义对象(比如Student),并查看排序是如何工作的。
private void sortStudentInArrayList() {
List<Student> students = new ArrayList<>();
Student student1 = createStudent("Biplab", 3);
students.add(student1);
Student student2 = createStudent("John", 1);
students.add(student2);
Student student3 = createStudent("Pal", 5);
students.add(student3);
Student student4 = createStudent("Biplab", 2);
students.add(student4);
System.out.println("Original students list: " + students);
Collections.sort(students);// Error here
System.out.println("Sorted students list: " + students);
Collections.sort(students, Collections.reverseOrder());
System.out.println("Reversed students list: " + students);
}
private Student createStudent(String name, int no) {
Student student = new Student();
student.setName(name);
student.setNo(no);
return student;
}
public class Student {
String name;
int no;
public String getName() {
return name;
}
public int getNo() {
return no;
}
public void setName(String name) {
this.name = name;
}
public void setNo(int no) {
this.no = no;
}
@Override
public String toString() {
return "Student{" +
"name='" + name + '\'' +
", no=" + no +
'}';
}
}
这会抛出编译时错误,并显示以下错误消息:
sort(java.util.List<T>)
in Collections cannot be applied
to (java.util.List<com.example.demo.dto.Student>)
reason: no instance(s) of type variable(s) T exist so that Student
conforms to Comparable<? Super T>
为了解决这个问题,要么Student类需要实现Comparable,要么需要在调用Collections.sort时传递Comparator对象。在整型情况下,排序方法没有错误,因为Integer类实现了Comparable。让我们看看,实现Comparable或传递Comparator如何解决这个问题,以及排序方法如何实现Collection排序。
package com.example.demo.dto;
public class Student implements Comparable{
String name;
int no;
public String getName() {
return name;
}
public int getNo() {
return no;
}
public void setName(String name) {
this.name = name;
}
public void setNo(int no) {
this.no = no;
}
@Override
public String toString() {
return "Student{" +
"name='" + name + '\'' +
", no=" + no +
'}';
}
@Override
public int compareTo(Object o) {
return this.getName().compareTo(((Student) o).getName());
}
}
输出:
Original students list: [Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}, Student{name='Biplab', no=2}]
Sorted students list: [Student{name='Biplab', no=3}, Student{name='Biplab', no=2}, Student{name='John', no=1}, Student{name='Pal', no=5}]
Reversed students list: [Student{name='Pal', no=5}, Student{name='John', no=1}, Student{name='Biplab', no=3}, Student{name='Biplab', no=2}]
在所有示例中,为了颠倒顺序,我们使用“Collections.sort(students, Collections.reverseOrder()”,相反的,它可以通过改变compareTo(..)方法的实现而达成目标,且compareTo(…) 的实现看起来像这样 :
@Override
public int compareTo(Object o) {
return (((Student) o).getName()).compareTo(this.getName());
}
输出:
Original students list: [Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}, Student{name='Biplab', no=2}]
Sorted students list: [Student{name='Pal', no=5}, Student{name='John', no=1}, Student{name='Biplab', no=3}, Student{name='Biplab', no=2}]
Reversed students list: [Student{name='Biplab', no=3}, Student{name='Biplab', no=2}, Student{name='John', no=1}, Student{name='Pal', no=5}]
如果我们观察输出结果,我们可以看到“Sorted students list:”以颠倒的顺序(按学生name)输出学生信息。
到目前为止,对学生的排序是根据学生的“name”而非“no”来完成的。如果我们想按“no”排序,我们只需要更改Student类的compareTo(Object o)实现,如下所示:
@Override
public int compareTo(Object o) {
return (this.getNo() < ((Student) o).getNo() ? -1 : (this.getNo() == ((Student) o).getNo() ? 0 : 1));
}
输出:
Original students list: [Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}, Student{name='Biplab', no=2}]
Sorted students list: [Student{name='John', no=1}, Student{name='Biplab', no=2}, Student{name='Biplab', no=3}, Student{name='Pal', no=5}]
Reversed students list: [Student{name='Pal', no=5}, Student{name='Biplab', no=3}, Student{name='Biplab', no=2}, Student{name='John', no=1}]
在上面的输出中,我们可以看到“no”2和3的两名学生具有相同的名字“Biplab”。
现在假设我们首先需要按“name”对这些学生进行排序,如果超过1名学生具有相同姓名的话,则这些学生需要按“no”排序。为了实现这一点,我们需要改变compareTo(…)方法的实现,如下所示:
@Override
public int compareTo(Object o) {
int result = this.getName().compareTo(((Student) o).getName());
if(result == 0) {
result = (this.getNo() < ((Student) o).getNo() ? -1 : (this.getNo() == ((Student) o).getNo() ? 0 : 1));
}
return result;
}
输出:
Original students list: [Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}, Student{name='Biplab', no=2}]
Sorted students list: [Student{name='Biplab', no=2}, Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}]
Reversed students list: [Student{name='Pal', no=5}, Student{name='John', no=1}, Student{name='Biplab', no=3}, Student{name='Biplab', no=2}]
为了按照“name”对Students进行排序,我们将添加一个Comparator并将其传递给排序方法:
public class Sorting {
private void sortStudentInArrayList() {
List<Student> students = new ArrayList<>();
Student student1 = createStudent("Biplab", 3);
students.add(student1);
Student student2 = createStudent("John", 1);
students.add(student2);
Student student3 = createStudent("Pal", 5);
students.add(student3);
Student student4 = createStudent("Biplab", 2);
students.add(student4);
System.out.println("Original students list: " + students);
Collections.sort(integers, new NameComparator());
System.out.println("Sorted students list: " + students);
}
}
public class Student {
String name;
int no;
public String getName() {
return name;
}
public int getNo() {
return no;
}
public void setName(String name) {
this.name = name;
}
public void setNo(int no) {
this.no = no;
}
@Override
public String toString() {
return "Student{" +
"name='" + name + '\'' +
", no=" + no +
'}';
}
}
class NameComparator implements Comparator<Student> {
@Override
public int compare(Student o1, Student o2) {
return o1.getName().compareTo(o2.getName());
}
}
输出:
Original students list: [Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}, Student{name='Biplab', no=2}]
Sorted students list: [Student{name='Biplab', no=3}, Student{name='Biplab', no=2}, Student{name='John', no=1}, Student{name='Pal', no=5}]
同样,如果我们想按照“no”对Students排序,那么可以再添加一个Comparator(NoComparator.java),并将其传递给排序方法,然后数据将按“no”排序。
现在,如果我们想通过“name”然后“no”对学生进行排序,那么可以在compare(…)内结合两种逻辑来实现。
class NameNoComparator implements Comparator<Student> {
@Override
public int compare(Student o1, Student o2) {
int result = o1.getName().compareTo(o2.getName());
if(result == 0) {
result = o1.getNo() < o2.getNo() ? -1 : o1.getNo() == o2.getNo() ? 0 : 1;
}
return result;
}
}
输出:
Original students list: [Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}, Student{name='Biplab', no=2}]
Sorted students list: [Student{name='Biplab', no=2}, Student{name='Biplab', no=3}, Student{name='John', no=1}, Student{name='Pal', no=5}]
在Set的这个情况下,我们需要将HashSet转换为ArrayList,或使用TreeSet对数据进行排序。此外,我们知道要使Set工作,equals(…)和hashCode()方法需要被覆盖。下面是基于“no”字段覆盖equals和hashcode的例子,且这些是IDE自动生成的代码。与Comparable或Comparator相关的其他代码与ArrayList相同。
@Override
public boolean equals(Object o) {
if (this == o) return true;
if (o == null || getClass() != o.getClass()) return false;
Student student = (Student) o;
return no == student.no;
}
@Override
public int hashCode() {
return Objects.hash(no);
}
为了对数组排序,我们需要做与排序ArrayList相同的事情(要么执行Comparable要么传递Comparable给sort方法)。在这种情况下,sort方法是“Arrays.sort(Object[] a )”而非“Collections.sort(..)”。
private void sortStudentInArray() {
Student [] students = new Student[4];
Student student1 = createStudent("Biplab", 3);
students[0] = student1;
Student student2 = createStudent("John", 1);
students[1] = student2;
Student student3 = createStudent("Pal", 5);
students[2] = student3;
Student student4 = createStudent("Biplab", 2);
students[3] = student4;
System.out.print("Original students list: ");
for (Student student: students) {
System.out.print( student + " ,");
}
Arrays.sort(students);
System.out.print("\nSorted students list: ");
for (Student student: students) {
System.out.print( student +" ,");
}
Arrays.sort(students, Collections.reverseOrder());
System.out.print("\nReversed students list: " );
for (Student student: students) {
System.out.print( student +" ,");
}
}
//Student class
// All properties goes here
@Override
public int compareTo(Object o) {
int result =this.getName().compareTo(((Student)o).getName());
if(result ==0) {
result = (this.getNo() < ((Student) o).getNo() ? -1 : (this.getNo() == ((Student) o).getNo() ? 0 : 1));
}
return result;
}
输出:
Original students list: Student{name='Biplab', no=3} ,Student{name='John', no=1} ,Student{name='Pal', no=5} ,Student{name='Biplab', no=2} ,
Sorted students list: Student{name='Biplab', no=2} ,Student{name='Biplab', no=3} ,Student{name='John', no=1} ,Student{name='Pal', no=5} ,
Reversed students list: Student{name='Pal', no=5} ,Student{name='John', no=1} ,Student{name='Biplab', no=3} ,Student{name='Biplab', no=2} ,
我们经常对Comparable/Comparator的使用以及何时使用哪个感到困惑。下面是我总结的Comparable/Comparator的使用场景。
Comparator:
Comparable:
应该在定义类时知道排序的顺序时使用,并且不会有其他任何需要使用Collection /数组来根据其他字段排序的情况。
注意:我没有介绍Set / List的细节。我假设读者已经了解了这些内容。此外,没有提供使用Set / array进行排序的详细示例,因为实现与ArrayList非常相似,而ArrayList我已经详细给出了示例。