使用流删除一个列表中的元素（如果存在于另一个列表中）

Question

I didn't find any thread here on this Question. I am trying to Remove elements (Cars in my case) of one list (cars1) if present in another list (cars2) using java stream. I tried using removeIf but then it felt like it works more appropriately with List of Strings, etc.

    Car c1 = new Car();
    c1.id = 1;
    c1.name = "C1";

    Car c2 = new Car();
    c2.id = 2;
    c2.name = "C2";

    List<Car> cars1 = new ArrayList<Car>();
    cars1.add(c1);
    cars1.add(c2);

    List<Car> cars2 = new ArrayList<Car>();
    cars2.add(c2);

    // TODO : Remove all the cars from cars1 list that are in cars2 list using java streams

Answer 1

If methods hashCode and equals are properly implemented in class Car , the stream-based solutions may look as follows:

• filter out the values, collect into a new list

// Predicate.not added in Java 11
List<Car> notJava11 = cars1.stream()
                        .filter(Predicate.not(cars2::contains))
                        .collect(Collectors.toList());

List<Car> notIn2 = cars1.stream()
                        .filter(car -> !cars2.contains(car))
                        .collect(Collectors.toList());

• Use forEach for cars2 (affecting cars1 ):

cars2.forEach(cars1::remove); 
// no need to call cars2.stream().forEach(cars1::remove);

Here the first occurrence of Car instance is removed in cars1

• removeIf should work also

cars1.removeIf(cars2::contains);

---

If you due to some reason equals / hashCode are not overridden in class Car , the following solution may be offered:

List<Car> notIn2 = cars1
        .stream()
        .filter(c1 -> cars2
            .stream()
            .noneMatch(c2 -> 
                 c1.getId() == c2.getId()
                 && Objects.equals(c1.getName(), c2.getName())
            )
        )
        .collect(Collectors.toList());

• removeIf :

cars1.removeIf(c1 -> cars2
    .stream()
    .anyMatch(c2 -> c1.getId() == c2.getId() 
        && Objects.equals(c1.getName(), c2.getName())
    )
);

Answer 2

It can be done simply by using removeAll() method. You have to implement hashcode and equals in Car class. Then you can cars1.removeAll(cars2); . this statement would leave only c1 in cars1 list.

Answer 3

The recommended solution is to override toString and hasCode() methods in the class, Car and then you can, as already comment below the question, do something like

cars1.removeAll(cars2);

or

cars1.removeIf(cars2::contains);

However, if these two methods are not overridden in the class, Car , you will have to compare each attribute within the nested loops. This approach is only useful to learn nested loops and is never recommended for production use.

Demo:

import java.util.ArrayList;
import java.util.List;

class Car {
    int id;
    String name;

    @Override
    public String toString() {
        return "Car [id=" + id + ", name=" + name + "]";
    }
}

public class Main {
    public static void main(String[] args) {
        Car c1 = new Car();
        c1.id = 1;
        c1.name = "C1";

        Car c2 = new Car();
        c2.id = 2;
        c2.name = "C2";

        List<Car> cars1 = new ArrayList<Car>();
        cars1.add(c1);
        cars1.add(c2);

        List<Car> cars2 = new ArrayList<Car>();
        cars2.add(c2);

        for (int i = 0; i < cars1.size(); i++) {
            Car car1 = cars1.get(i);
            for (int j = 0; j < cars2.size(); j++) {
                Car car2 = cars2.get(j);
                if (car2.id == car1.id && car2.name.equals(car1.name))
                    cars1.remove(i);
            }
        }

        System.out.println(cars1);
    }
}

Output:

[Car [id=1, name=C1]]