what kind of data structure do i need to implement a hashlist for storing coordinates?

Answer 1

I'd approach this problem like this by just creating a set of positions. A set models a collection of objects that can only occur once. In comparison a map structure stores a set of key / value associations. From my reading of your question, I think a set structure makes most sense.

// You might just be able to use an existing Point depending on what you
// want to do with the position
class Position {
  int x;
  int y;

  // implementations of hashCode() + equals()
  }
}

You need to implement hashCode() so that items can be distributed uniformly in the set and equals() so that objects can be compared. See here for more information.

Set<Position> positions = new HashSet<Position>();
positions.add(new Position(3,4));
positions.add(new Position(5,6)); // and so on

Make sure you define equals / hashCode appropriately (there's plenty of links for this)

You can now test whether a point is in the set using the contains method such as:

positions.contains(new Point(2,1)); // returns false
positions.contains(new Point(3,4)); // returns true

Answer 2

I would suggest using google-collection's MultiMap. This is effectively a managed type for Map<K, Collection<V>>.

Also of interest may be the Multimaps class, which gives you Multimap<K,V> invertFrom(Multimap<V,K>)

Then you could end up with:

public boolean isPositionOccupied(int x, int y) {
    return occupiedPositionsMap.get(x).contains(y);
}

See? Wow! no need for null checks or other nonsense.

Note: This is relatively optimal from a performance point of view, but depending on your other needs, you may want to use Point objects as noted in other answers.