Custom key in Hashtable (or HashMap)

Question

I have built a custom object to be used as a key in a Hashtable. However, even though I provided an implementation of hashcode() and equals(), two key objects that compare equally do NOT locate the same value in the Hashtable. I either do not understand how this is supposed to work or there is a bug. I have JDK 1.3.1.
I hate it when things don't work the way I think it should! Why doesn't this work? ANy ideas how to accomplish this?
public class TransitionKey {
private String stateID;
private String nextStateID;
private String hashKey;
/** Creates new TransitionKey */
public TransitionKey() {
public TransitionKey(String stateID, String nextStateID) {
this.stateID = stateID;
this.nextStateID = nextStateID;
hashKey = stateID+nextStateID;
public TransitionKey(int stateID, int nextStateID) {
this(String.valueOf(stateID), String.valueOf(nextStateID));
public String getStateID() {
return stateID;
public String getNextStateID() {
return nextStateID;
public boolean equals(Object obj) {
TransitionKey other;
if (obj instanceof TransitionKey)
other = (TransitionKey) obj;
else
throw new IllegalArgumentException("Only objects of type " + this.getClass() + " are supported.");
return (other.getStateID().equals(this.getStateID()) && other.getNextStateID().equals(this.getNextStateID()));
public int hashcode() {
return hashKey.hashCode();
public String toString() {
return "(" + getStateID() + "," + getNextStateID() + ")";
public static void main (String [] args) {
TransitionKey one, two;
one = new TransitionKey(10,8);
two = new TransitionKey(10,8);
System.out.println("Objects are: '" + one + "' and '" + two + "'.");
System.out.println("Keys " + (one.equals(two) ? "" : "do NOT ") + "match");
System.out.println("Hashcode test: " + (one.hashcode() == two.hashcode() ? "PASS" : "FAIL"));
java.util.Hashtable tbl = new java.util.Hashtable();
tbl.put(one,"You found me!");
String output = (String) tbl.get(one);
System.out.println("Result using original key: " + output);
output = (String) tbl.get(two);
System.out.println("Result using duplicate key: " + output);
boolean keyTest = tbl.containsKey(two);
System.out.println("Test for duplicate kay in hashtable: " + keyTest);

Answer 1

I believe you don't quite have the right understanding
of how this works. It's not the contents of an object
that allow it to be found in the hashtable, but the
exact object itself.
You have created objects one and two. They have the
same content but are unique objects. You will not be
able to find the string associated with one by trying
to get two.
You can see this by changing one line, instead of
two = new TransitionKey(10,8);try ...
two = one;After that, you'll see that you have found the string
associated with one when you get two. That's because
one and two are now the same object.
So, to get a string associated with two, where two is
unique, you'll have to put two into the hashtable.
I hope this is clear.This is absolutely false. Below is an example using strings that shows this. I am a little confused though, because I have been playing with your class and it is not working the way that I understand maps to work. I'll post again when I have played around some more.
import java.util.HashMap;
public class StringTest {
     public static void main (String [] args) {
          String one = new String("a");
          String two = new String("a");
          HashMap map = new HashMap();
          System.out.println(1);
          map.put(one,"You found one");
          System.out.println(2);
          System.out.println("Result using duplicate key: " + map.get(two));
          System.out.println(3);
          System.out.println("Result using original key: " + map.get(one));
}