Java HashSet vs HashMap

I understand that HashSet is based on HashMap implementation but is used when you need unique set of elements. So why in the next code when putting same objects into the map and set we have size of both collections equals to 1? Shouldn't map size be 2? Because if size of both collection is equal I don't see any difference of using this two collections.

    Set testSet = new HashSet<SimpleObject>();
    Map testMap = new HashMap<Integer, SimpleObject>(); 

    SimpleObject simpleObject1 = new SimpleObject("Igor", 1);
    SimpleObject simplObject2 = new SimpleObject("Igor", 1);
    testSet.add(simpleObject1);
    testSet.add(simplObject2);


    Integer key = new Integer(10);

    testMap.put(key, simpleObject1);
    testMap.put(key, simplObject2);

    System.out.println(testSet.size());
    System.out.println(testMap.size());

The output is 1 and 1.

SimpleObject code

public class SimpleObject {

private String dataField1;
private int dataField2;

public SimpleObject(){}

public SimpleObject(String data1, int data2){
    this.dataField1 = data1;
    this.dataField2 = data2;
}

public String getDataField1() {
    return dataField1;
}

public int getDataField2() {
    return dataField2;
}

@Override
public int hashCode() {
    final int prime = 31;
    int result = 1;
    result = prime * result
            + ((dataField1 == null) ? 0 : dataField1.hashCode());
    result = prime * result + dataField2;
    return result;
}

@Override
public boolean equals(Object obj) {
    if (this == obj)
        return true;
    if (obj == null)
        return false;
    if (getClass() != obj.getClass())
        return false;
    SimpleObject other = (SimpleObject) obj;
    if (dataField1 == null) {
        if (other.dataField1 != null)
            return false;
    } else if (!dataField1.equals(other.dataField1))
        return false;
    if (dataField2 != other.dataField2)
        return false;
    return true;
 }
}

Solution 1:

The map holds unique keys. When you invoke put with a key that exists in the map, the object under that key is replaced with the new object. Hence the size 1.

The difference between the two should be obvious:

• in a Map you store key-value pairs
• in a Set you store only the keys

In fact, a HashSet has a HashMap field, and whenever add(obj) is invoked, the put method is invoked on the underlying map map.put(obj, DUMMY) - where the dummy object is a private static final Object DUMMY = new Object(). So the map is populated with your object as key, and a value that is of no interest.

Solution 2:

A key in a Map can only map to a single value. So the second time you put in to the map with the same key, it overwrites the first entry.

Solution 3:

In case of the HashSet, adding the same object will be more or less a no-op. In case of a HashMap, putting a new key,value pair with an existing key will overwrite the existing value to set a new value for that key. Below I've added equals() checks to your code:

SimpleObject simpleObject1 = new SimpleObject("Igor", 1);
SimpleObject simplObject2 = new SimpleObject("Igor", 1);
//If the below prints true, the 2nd add will not add anything
System.out.println("Are the objects equal? " , (simpleObject1.equals(simpleObject2));
testSet.add(simpleObject1);
testSet.add(simplObject2);


Integer key = new Integer(10);
//This is a no-brainer as you've the exact same key, but lets keep it consistent
//If this returns true, the 2nd put will overwrite the 1st key-value pair.
testMap.put(key, simpleObject1);
testMap.put(key, simplObject2);
System.out.println("Are the keys equal? ", (key.equals(key));
System.out.println(testSet.size());
System.out.println(testMap.size());