如何用Java编写可靠的Pure Aggregation(组合)游戏对象? [英] How to write solid Pure Aggregation (composition) Game Objects in Java?

查看:112
本文介绍了如何用Java编写可靠的Pure Aggregation(组合)游戏对象?的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!

问题描述

所以我刚开始用Java编写游戏,我正在编写我的游戏对象。现在我已经在。播放器将委托依赖于 Movement 对象的所有移动操作和它不关心它必须使用的具体类型的 Movement 对象。


So I am just at the beginning of writing a game in Java and I am writing my game objects. Now I have read here in Evolve Your Hierarchy that you should build your games as compositions and not as a big class hierarchy. As this image from the previous link shows:

However, when actually getting down to the implementation I have one small question about where to apply the interfaces.

Lets say you have a class called Player and the interfaces Moveable and Renderable. Do you implement this using public interface variables:

class Player {
    public Moveable moveable;
    public Renderable renderable;
}

class GenericMoveable implements Moveable {
    // Function implementations
}

class PlayerRenderable implements Renderable {
    // Function implementations
}   

Or do you try and do this by applying the interfaces directly to the object:

class Player implements Moveable, Renderable {
    private GenericMoveable genericMoveable;

    // Non-direct Implementation of Moveable
    void someMoveFunc(double x, double y) {
        genericMoveable.someMoveFunc(x, y);
    }

    // Direct implementation of Renderable
    void someRenderableFunction() {
        // Player class specific code
    }
}

class GenericMoveable implements Moveable {
    // Function implementations
}

Now currently I am feeling that the second method is better. The main reason for that is because I can create the following lists:

List<Renderable> renderObjects; // use this to draw all of the objects to the screen
List<Moveable> moveObjects; // use this to move all objects at once

I really just want confirmation that I am on the right track. I am pretty sure that when creating game objects by composition for games in Java you want to apply the interfaces directly to the Game Object in question and then use private interfaces if you need them and direct implementations of those functions for the rest. Is this correct? And what are the positives and negatives of what I am doing?

Though, while we are here, what else do I need to look out for in the implementation of Compositional Game Objects in Java? What should I do differently? Thanks for any responses in advance, I appreciate any and all help.

解决方案

That's an interesting link. I gather that what you have in mind for your Player class is what that article calls a component container. For that, applying the interfaces to the class sounds to me like the way to go. It's the only route that will lead you to a pure aggregation structure (where there is no longer a Player class per se in your system.

By the way, the structure you're proposing is basically an application of the delegation design pattern. A Player relies delegates to a Movement object all movement operations and it doesn't care what specific kind of Movement object it has to work with.

这篇关于如何用Java编写可靠的Pure Aggregation(组合)游戏对象?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!

查看全文
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆