[java] Usage of Comparator

1. Why write?

comparator is an interface in javase, located in the java.util package, the interface is highly abstract, it is necessary to master the use of this interface
Most articles tell you that comparator is used for sorting, but I would like to say that sorting is one of the functions that comparator can achieve, it is not limited to sorting

2. Interface function

This interface represents a comparator, and comparators are comparable! Most of the articles write how to use the comparator to sort, because the sort method provided in the javase array tool class and the collection tool class uses the Comparator interface to handle sorting. Everyone has seen the Comparator interface for a long time. From the abstraction of java, if the interface is born for sorting, it should be called Sortable, Sortor or something like that! Here are some places where javase uses the Comparator interface:

Arrays.sort(T[],Comparator<? super T> c);
Collections.sort(List<T> list,Comparator<? super T> c);

3. Usage scenarios

What scene needs to be compared, then what scene is the use of the Comparator interface, I summarize two scenarios:
1. Sorting, you need to compare who is in the front and who is in the back (sorting can also make the class implement the Comparable interface , instances of this class also have sorting capabilities after implementation).
2. Grouping, you need to compare whether two objects belong to the same group.
3. To be added

4. Take a chestnut

1. If the objects sorted
in the List or array do not implement the Comparable interface, then the caller needs to set a Compartor for the array or List to be sorted. The compare method of the Compartor is used to tell the code how to compare the two instances, and then Sort by comparison result

talk is cheap show me the code

package com.java.demo;

import java.util.ArrayList;
import java.util.Collections;
import java.util.Comparator;
import java.util.List;
/**
 * @author puyf
 */
public class SortTest {
    class Dog{
    public int age;
    public String name;
    public Dog(int age, String name) {
        super();
        this.age = age;
        this.name = name;
    }
    @Override
    public String toString() {
        return "Dog [age=" + age + ", name=" + name + "]";
    }
    }
    public static void main(String[] args) {
    List<Dog> list= new ArrayList<>();
    list.add(new SortTest().new Dog(5, "DogA"));
    list.add(new SortTest().new Dog(6, "DogB"));
    list.add(new SortTest().new Dog(7, "DogC"));
    Collections.sort(list, new Comparator<Dog>() {

        @Override
        public int compare(Dog o1, Dog o2) {
        return o2.age - o1.age;
        }
    });
    System.out.println("给狗狗按照年龄倒序："+list);
    Collections.sort(list, new Comparator<Dog>() {

        @Override
        public int compare(Dog o1, Dog o2) {
        return o1.name.compareTo(o2.name);
        }
    });
    System.out.println("给狗狗按名字字母顺序排序："+list);
    }
}

2.分组
使用Comparator和for循环处理列表，来进行分类；通过调用者实现Comparator接口的比较逻辑，来告诉程序应该怎么比较，通过比较之后得结果来进行分组。比如生活中的拳击比赛，会有公斤级的概念，那么程序中应该实现的处理逻辑是只要两个人的体重在同一个区间则为同一组公斤级的选手。下面例子中分别按照狗狗的颜色和体重级别两个维度来进行分组，因此分组的核心逻辑其实就是比较逻辑。相面我抽了一个工具方法：dividerList,第一个参数为需要处理的数据源，第二参数是分组时的比较逻辑。

package com.java.demo;

import java.util.ArrayList;
import java.util.Comparator;
import java.util.List;
/**
 * @author puyf
 */
public class GroupTest {
    class Apple {
    public String color;
    public int weight;

    public Apple(String color, int weight) {
        super();
        this.color = color;
        this.weight = weight;
    }

    @Override
    public String toString() {
        return "Apple [color=" + color + ", weight=" + weight + "]";
    }
    }

    /**
     * @author puyf
     * @Description:按条件分组
     * @param datas
     * @param c
     *            是否为同一组的判断标准
     * @return
     */
    public static <T> List<List<T>> divider(Collection<T> datas, Comparator<? super T> c) {
        List<List<T>> result = new ArrayList<List<T>>();
        for (T t : datas) {
            boolean isSameGroup = false;
            for (int j = 0; j < result.size(); j++) {
                if (c.compare(t, result.get(j).get(0)) == 0) {
                    isSameGroup = true;
                    result.get(j).add(t);
                    break;
                }
            }
            if (!isSameGroup) {
                // 创建
                List<T> innerList = new ArrayList<T>();
                result.add(innerList);
                innerList.add(t);
            }
        }
        return result;
    }

    public static void main(String[] args) {
    List<Apple> list = new ArrayList<>();
    list.add(new GroupTest().new Apple("红", 205));
    list.add(new GroupTest().new Apple("红", 131));
    list.add(new GroupTest().new Apple("绿", 248));
    list.add(new GroupTest().new Apple("绿", 153));
    list.add(new GroupTest().new Apple("黄", 119));
    list.add(new GroupTest().new Apple("黄", 224));
    List<List<Apple>> byColors = divider(list, new Comparator<Apple>() {

        @Override
        public int compare(Apple o1, Apple o2) {
        // 按颜色分组
        return o1.color.compareTo(o2.color);
        }
    });
    System.out.println("按颜色分组" + byColors);
    List<List<Apple>> byWeight = divider(list, new Comparator<Apple>() {

        @Override
        public int compare(Apple o1, Apple o2) {
        // 按重量级

        return (o1.weight / 100 == o2.weight / 100) ? 0 : 1;
        }
    });
    System.out.println("按重量级分组" + byWeight);
    }
}

结果如下（为了方便看，手动回车换行格式化了下）：
按颜色分组
[
[
Apple [color=红, weight=205],
Apple [color=红, weight=131]
],
[
Apple [color=绿, weight=248],
Apple [color=绿, weight=153]
],
[
Apple [color=黄, weight=119],
Apple [color=黄, weight=224]
]
]

按重量级分组
[
[
Apple [color=红, weight=205],
Apple [color=绿, weight=248],
Apple [color=黄, weight=224]
],
[
Apple [color=红, weight=131],
Apple [color=绿, weight=153],
Apple [color=黄, weight=119]
]
]

5.总结

一般需要做比较的逻辑都可以使用的上Comparator，最常用的场景就是排序和分组，排序常使用Arrays和Collections的sort方法，而分组则可以使用上面提供的divider方法。

排序和分组的区别在于:
排序时，两个对象比较的结果有三种：大于，等于，小于。
分组时，两个对象比较的结果只有两种：等于(两个对象属于同一组)，不等于(两个对象属于不同组)