print a HashSet of objects

[Gil Shoam]

Hi,
sorry if this should be in Java for beginners. I wasn't sure..

Anyway, I'm writing a program that deals with graphs of any primitive types.
I have 3 classes.
One represents the Nodes:

public class Node {
private Object _object;
public Node(Object obj) {
_object = new Object();
}
}

Another the Edges

public class Edge {
private Node _source;
private Node _dest;
public Edge(Node source, Node dest) {
_source = source;
_dest = dest;
}
}

And finally the Graph:

public class Graph {
private HashSet<Node> _nodes;
private HashSet<Edge> _edges;
public Graph() {
_nodes = new HashSet<Node>();
_edges = new HashSet<Edge>();
}
public Graph(ArrayList<Node> nodeList, ArrayList<Edge> edgeList) {
_nodes = new HashSet<Node>(nodeList);
_edges = new HashSet<Edge>(edgeList);
}
public void addNode(Node node) {
_nodes.add(node);
}
public void removeNode(Node node) {
_nodes.remove(node);
}
public void addEdge(Edge edge) {
_edges.add(edge);
}
public void removeEdge(Edge edge) {
_edges.remove(edge);
}
public boolean isNode(Node node) {
return _nodes.contains(node);
}
public boolean isEdge(Edge edge) {
return _edges.contains(edge);
}
public boolean equals(Graph graph) {
if (_nodes.equals(graph._nodes) && _edges.equals(graph._edges))
return true;
return false;
}

Just to test it I ran this Main:

public static void main(String[] args) {
//new list of nodes
ArrayList<Node> nodes = new ArrayList<Node>();
//creates new nodes
Node nodeA = new Node('A');
Node nodeB = new Node('B');
Node nodeC = new Node('C');
Node nodeD = new Node('D');
//add nodes to the list
nodes.add(nodeA);
nodes.add(nodeB);
nodes.add(nodeC);
nodes.add(nodeD);
ArrayList<Edge> edges = new ArrayList<Edge>();
Edge edge1 = new Edge(nodeA, nodeB);
Edge edge2 = new Edge(nodeA, nodeC);
Edge edge3 = new Edge(nodeB, nodeC);
Graph graph = new Graph(nodes, edges);
Iterator iterator = graph._nodes.iterator();
while (iterator.hasNext())
System.out.print(iterator.next() + ", ");
System.out.println();
}

Problem is as you can see, it prints the object address.
I actually need to write a String representation of the graph inc the nodes and edges.
i.e the output should be:
Nodes: {A,,B,C,D}
Edges: {(A,B),(A,C),(B,C)


(here I tried printing only the nodes to see if it works, but no luck)
Since the actual node is an object how can I downcast it to the type I initialized it with?
Or in general- How do I override object toString() method for the graph class?

Thanks in advance

Gil

[Seetharaman Venkatasamy]

Welcome to JavaRanch Gil

So do you want to see Node's instance variable _object ? then override Object's toString method in Node class

[Gil Shoam]

Fast reply

Not exactly, that was just a test to see if I can access the char from graph._nodes.

What I want is to write a String representation of the graph inc the nodes and edges.
i.e the output for the main (above) should be:
Nodes: {A,,B,C,D}
Edges: {(A,B),(A,C),(B,C)

[Jeff Verdegan]

 

Gil Shoam wrote:Fast reply

Not exactly, that was just a test to see if I can access the char from graph._nodes.

What I want is to write a String representation of the graph inc the nodes and edges.

Then your Graph class needs to override toString() to pring out the nodes and edges how you want. And if you have classes for Node and/or Edge, they need to override toString to display themselves appropriately.

[Gil Shoam]

 

Jeff Verdegan wrote:

Gil Shoam wrote:Fast reply

Not exactly, that was just a test to see if I can access the char from graph._nodes.

What I want is to write a String representation of the graph inc the nodes and edges.

Then your Graph class needs to override toString() to pring out the nodes and edges how you want. And if you have classes for Node and/or Edge, they need to override toString to display themselves appropriately.

That I know..

How do I do it? Since I don't know In advance what types the graph has.
i tried overriding the toString() in the Node Class using the StringBuffer but it does not make sense...
if I don't yet know what type it is how can I write a string method for it?

[Jeff Verdegan]

 

Gil Shoam wrote:

Jeff Verdegan wrote:

Gil Shoam wrote:Fast reply

Not exactly, that was just a test to see if I can access the char from graph._nodes.

What I want is to write a String representation of the graph inc the nodes and edges.

Then your Graph class needs to override toString() to pring out the nodes and edges how you want. And if you have classes for Node and/or Edge, they need to override toString to display themselves appropriately.

That I know..

How do I do it?

That's a really broad question. You should be able to find tons of examples of overriding toString() with a simple web search. What did you try and where exactly did you get stuck?

Since I don't know In advance what types the graph has.

You don't need to. Graph's toString() just says something like:

return node1.toString() + ", " + node2.toString();

except, of course, you'd do that in a loop that walks over the nodes.

The individual nodes are then responsible for their own toString(). If a node is just any old class, then it's up to the author of that class to provide a reasonable toString(). If, on the other hand, you have a Node class that contains data and pointers to other Nodes, then you could have its toString() simply return the data's toString().

So you could have Graph's toString() which calls Node's toString() which calls it's data's toString(), with each layer adding whatever structural stuff it knows about/is responsible for (commas, brackets, etc.).

i tried overriding the toString() in the Node Class using the StringBuffer but it does not make sense...

Not sure why you'd use SB. And without seeing what you tried and what problems it caused, I can't really tell you where you went wrong.

if I don't yet know what type it is how can I write a string method for it?

You don't. And you don't need to. It provides its own.

[Gil Shoam]

 

So you could have Graph's toString() which calls Node's toString() which calls it's data's toString(), with each layer adding whatever structural stuff it knows about/is responsible for (commas, brackets, etc.).

I tired this in the node class and the output I got is the object address not the actual char:

public class Node {
private Object _object;
public Node(Object obj) {
_object = new Object();
}
public String toString() {
String res = _object.toString();
return res;
}
public static void main(String[] args) {
//Character character = new Character('A');
Node node = new Node('A');
System.out.println(node.toString());
}
}

I also tried wrapping it with the Character class but still the same problem...

Any suggestions?
Thanks

[Jeff Verdegan]

 

Gil Shoam wrote:I tired this in the node class and the output I got is the object address not the actual char:

Because you're not storing the Character you pass in.

public Node(Object obj) {
_object = new Object();
}

This says, "Totally ignore the obj argument that was passed to me, and instead create a new Object, and store that."

What you're getting it Object's hashCode, because you have stored an Object that you just created, not a Character that you passed in.

[Gil Shoam]

I got it!

In the constructor I didn't use the obj I got. This solved it

public Node(Object obj) {
_object= new Object();
_object= obj;
}

Thanks for your help.

If I have more Questions related to this project should I keep posting on this thread or open a new one?

Thanks again,

Gil

[Jeff Verdegan]

_object= new Object();
_object= obj;

The first line is totally pointless. It creates a new object, and stores a reference to that object in the _object variable. Then, on the next line, you copy the reference value from the obj parameter into the _object variable, and that new Object that you created and never used becomes eligible for GC.

Just get rid of the first line.

[Jeff Verdegan]

 

Gil Shoam wrote:ore Questions related to this project should I keep posting on this thread or open a new one?

New one, unless it's further clarification on a point that's already come up in this discussion.