MVC Design – Should DAO Be Called from Controller or Model?

There are no commonly-agreed-upon (i.e. defacto) design guidelines for MVC. It is not that entirely hard to do it yourself but requires some planning on your classes and lots of time and patience.

The reason there is no definite solution is because there are multiple ways to do MVC, all with their pros and cons. So just be smart about it and do what suits you the best.

To answer your question, you actually want to decouple the controller from the view as well (so you can use the same business rule logic for both a Swing app and in the console app). In the Swing example you want to decouple the controller from the JWindow and whatever widget in Swing. The way I used to do (before using actual frameworks) is to create an interface for the view that the controller uses:

public interface PersonView {
    void setPersons(Collection<Person> persons);
}

public class PersonController {

    private PersonView view;
    private PersonModel model;

    public PersonController(PersonView view, PersonModel model) {
        this.view = view;
        this.model = model;
    }
    // ... methods to affect the model etc. 
    // such as refreshing and sort:

    public void refresh() {
        this.view.setPersons(model.getAsList());
    }

    public void sortByName(boolean descending) {
       // do your sorting through the model.
       this.view.setPersons(model.getSortedByName());
    }

}

For this solution during startup you need to register the controller into the view.

public class PersonWindow extends JWindow implements PersonView {

    PersonController controller;
    Model model;

    // ... Constructor etc.

    public void initialize() {
        this.controller = new PersonController(this, this.model);

        // do all the other swing stuff

        this.controller.refresh();
    }

    public void setPersons(Collection<Person> persons) {
        // TODO: set the JList (in case that's you are using) 
        // to use the given parameter
    }

}

It might be a good idea to create an IoC-container to do all the setup for you instead.

Anyhow, this way you can implement console-only views, using the same controllers:

public class PersonConsole implements PersonView {

    PersonController controller;
    Model model;

    public static void main(String[] args) {
        new PersonConsole().run();
    }

    public void run() {
        this.model = createModel();
        this.controller = new PersonController(this, this.model);

        this.controller.refresh();
    }

    public void setPersons(Collection<Person> persons) {
        // just output the collection to the console

        StringBuffer output = new StringBuffer();
        for(Person p : persons) {
            output.append(String.format("%s%n", p.getName()));
        }

        System.out.println(output);
    }

    public void createModel() {
        // TODO: create this.model
    }

    // this could be expanded with simple console menu with keyboard
    // input and other console specific stuff

}    

The fun part is how to do event handling. I implemented this by letting the view register itself to the controller using an interface, this is done using the Observer pattern (if you're using .NET you'd be using event handlers instead). Here is an example of a simple "document observer", that signals when the a document has been saved or loaded.

public interface DocumentObserver {
    void onDocumentSave(DocModel saved);
    void onDocumentLoad(DocModel loaded);
}

// in your controller you implement register/unregister methods
private List<DocumentObserver> observers;

// register observer in to the controller
public void addObserver(DocumentObserver o) {
    this.observers.add(o);
}

// unregisters observer from the controller
public void removeObserver(DocumentObserver o) {
    this.observers.remove(o);
}

public saveDoc() {
    DocModel model = model.save();
    for (DocumentObserver o : observers) {
        o.onDocumentSave(model);
    }
}

public loadDoc(String path) {
    DocModel model = model.load(path);
    for (DocumentObserver o : observers) {
        o.onDocumentLoad(model);
    }        
}

That way, the view can properly update itself since it is subscribing to the document updates. All it has to do is to implement the DocumentObserver interface:

public class DocumentWindow extends JWindow 
        implements DocView, DocumentObserver {

    //... all swing stuff

    public void onDocumentSave(DocModel saved) {
        // No-op
    }

    public void onDocumentLoad(DocModel loaded) {
        // do what you need with the loaded model to the
        // swing components, or let the controller do it on
        // the view interface
    }

    // ...

}

I hope these motivating examples gives you some ideas on how to do it yourself. However I strongly advise you to consider using frameworks in Java that do most of the things for you, or else you'll end up having a lot of boilerplate code that takes a long time to write. There are a couple of Rich Client Platforms (RCP) that you could use that implement some of the basic functionality that you most likely need, such as application-wide document handling and a lot of basic event handling.

There are a couple I can think of from my head: Eclipse and Netbeans RCP's.

You still need to develop controllers and models for yourself, but that's why you use an ORM. Example would be Hibernate.

IoC containers are cool'n all, but there are frameworks for this as well. Such as Spring (which also does data handling as well among other things).

The first question you need to answer is whether you intend to create a truly generic data access framework, or if you want it tailored to the particular project you're working on.

A generic framework has the benefit of being portable and fairly loosely coupled to the underlying DB + schema, but at the cost of pushing schema level information further up in the framework. The more generic you make the framework, the more of this knowledge you push up.

A tailored framework alleviates the higher levels of the application from worrying about the schema or DB details, but at the cost of cost of being less re-usable in other environments. As a corollary, the more specific you make the framework, the less easy it is to extend elsewhere.

If you're not sure on which of those to pick, the generic framework makes sense if you suspect there will be changes at the DB layer or the schema. Likewise, if those are pretty well locked down then the tailored framework is the better approach. Ultimately, you're optimizing the system to more easily handle where you think the change will occur.

As to whether or not you should refactor - the answer is No unless you're encountering a specific problem.

Worrying about the degree of coupling is nice, I suppose, but doesn't necessarily address a specific problem that you may be seeing. Refactoring based upon presumed issues within coupling is actually more likely to create a problem for you than just leaving it alone.

Likewise with expressing the state. If the current mechanism of an enum is sufficient and doesn't present an apparent problem then don't change what you've got.

Overall, it sounds like you've designed a reasonably robust data access method for the domain you're working worth. Your question doesn't list any specific issues that are holding back your development, so I'd call it good and move to the next stage of development.