How to get multiple return value

It all depends on exactly what you're using the String[] and String[][] for. You could possible extend the String[] to a String[][] with only 1 initial element (so instead of doing String[x] you would do String[0][x] on that one), and then put both of the String[][] in an ArrayList... but depending on what you're doing with them once you have them, that may not be the best solution.

technodolta at 2007-7-14 20:41:08 >

Generally speaking, when you need to return multiple values from a method, it falls into one of two situations:

1) the multiple values correspond with each other in some way, for example "customer name" and "customer phone number". In this situation, it's usually best to return an object that encapsulates this pairing (e.g., a Customer object). It's also a red flag that maybe the caller code should be folded into the class of the called code if they're not already part of the same class -- it could be a violation of encapsulation.

2) the multiple values are multiple instances of the same kind of information, for example "customer #1 name", "customer #2 name", "customer #3 name". In this case, it's usually best to return a java.util.Collection holding multiple objects.

But as Norweed pointed out, it sounds like you're doing something the wrong way.What you're doing sounds unduly complicated to begin with. But also when you're creating complicated array structures, it's usually a sign that you're doing something wrong, especially in an OO language like Java.

paulcwa at 2007-7-14 20:41:08 >

I still don't see the need for a 2-d array.

You're defining a graph.You're referring to the nodes as "boxes" but the standard term is "node".

Create a Node object to represent nodes.

Now you want vertices between nodes. ("vertex" or "edge" is also a standard term; you've used the word "joint".) You can do this a couple ways. One way is to have a java.util.Collection of other Node objects within a given Node object. Each Node within the collection represents an edge.Alternatively, you can create an Edge class, each instance of which contains a Node for the source and a Node for the target. A Node can contain a collection of Edges.

Once you have your graph set up, you can either identify a single Node from which all other nodes can be reached, and just use that as a handle onto your graph; or you could create a Collection of all node objects.

Basically, I'm saying that the first dimension of your array can be the overall list of nodes, but the second dimension is data that really ought to be folded into each object.

paulcwa at 2007-7-14 20:41:08 >