I have always wondered how would you implement a Set in Java. Can we implement it just like we implement a HashMap using a LinkedList and an object(Cell) which holds a Key and Value? How would you handle the uniqueness part?
Asked
Active
Viewed 2.8k times
5 Answers
5
Basically, a Set is just a Map that only holds keys. So you should inform yourself about mappingalgorithms. Note: the HashSet for example is actually just an adapter for the HashMap. the add-method of HashSet simply uses HashMap.put(value , SomeDummyValue).
1
Below is a code snippet to explain above answers
public HashSet() { map = new HashMap<>(); }
private transient HashMap<E,Object> map;
// Dummy value to associate with an Object in the backing Map
private static final Object PRESENT = new Object();
public boolean add(E e) {
return map.put(e, PRESENT)==null;
}
// Since PRESENT is a constant, for all keys we have same value in backup HashMap called map.
public Iterator<E> iterator() {
return map.keySet().iterator();
}

bitsabhi
- 728
- 8
- 12
1
Short answer: Map
Basics first
What are the features of Set datastructures?
- Each element get stored only once.
- Query for an item in Set should be fastest, i.e. in
O(1)
. - If an item to be inserted is already present in Set, then don't insert. If it's not present then insert.
If we look closely, which datastructure can implement it?
Answer is Map
, because we just keep a record of which value present. And we can query in O(1)
time.

Om Sao
- 7,064
- 2
- 47
- 61
-2
class HashSetBasicImpl<K> {
static class Entry<K> {
private K key;
private Entry<K> next;
Entry(K key) {
key = key;
next = null;
}
}
private Entry<K>[] entries;
public HashSetBasicImpl() {
// fixed size
map =new Entry[100];
}
public boolean contains(K key) {
int idx = key.hashCode();
Entry<K> start = entries[idx];
while(start != null) {
if(start.key == key) {
return true;
}
start = start.next;
}
return false;
}
public boolean add(K key) {
Entry<K> entry = new Entry(key);
int idx = key.hashCode();
// check if entry exists
if(contains(key)) {
return false;
}
// add as first entry
start = entries[idx];
if(start == null) {
entries[idx]= new Entry(key);
return true;
}
// add as nth entry
Entry prev = null;
while(start != null) {
prev = start;
start = start.next;
}
prev.next = new Entry(key);
return true;
}
}