java - 如何让Rhino运行自定义的javascript setTimeout函数，无尽的eval

Question

我发现 Gameboy 脚本是用 javascript 编写的，我正在尝试让它在 java 中运行。 (我不是在寻找任何 java gameboy 模拟器)。

我有一个包含这两个文件的 JavaScript

function cout(e, t) {
  java.lang.System.out.println("e = " + e + " t = " + t);
}

function setTimeout(expr, msec) { 
    if (typeof expr == "function") { 
        // save the global object and trailing args for later apply 
        var gobj = this; 
        var args = [].concat(arguments).slice(2); 
        var o = {actionPerformed: function(){expr.apply(gobj, args)}}; 
    } else { 
        var o = {actionPerformed: function(){eval(expr)}}; 
    } 
    var al = new java.awt.event.ActionListener(o); 
    var t = new javax.swing.Timer(msec, al); 
    t.start(); 
}

我在另一个 javascript 文件中有类似这样的内容

var frames = 0;
function tick() {
  var a = (new Date().getTime() - tstart) - ttime;
  while (a > settings[6]) {
    ttime += settings[6];
    a = (new Date().getTime() - tstart) - ttime;

    gameboy.run()
    frames++;
  }
  setTimeout(tick, settings[6])
}

下面是我的 Java Rhino 代码修改后的示例之一。

public class GameBoyJS {

    /**
     * Main entry point.
     *
     * Process arguments as would a normal Java program. Also
     * create a new Context and associate it with the current thread.
     * Then set up the execution environment and begin to
     * execute scripts.
     */
    public GameBoyJS()
    {
        Context context = Context.enter();
        try {
            Scriptable globalScope = context.initStandardObjects();

        Reader base64LibReader = new InputStreamReader(getClass().getResourceAsStream("js/base64.js"));
        //Reader msgpackLibReader = new InputStreamReader(getClass().getResourceAsStream("js/msgpack.js"));
        Reader json2LibReader = new InputStreamReader(getClass().getResourceAsStream("js/json2.js"));
        Reader terminalLibReader = new InputStreamReader(getClass().getResourceAsStream("js/terminal.js"));
        //Reader resizeLibReader = new InputStreamReader(getClass().getResourceAsStream("js/resize.js"));
        Reader GameBoyIOLibReader = new InputStreamReader(getClass().getResourceAsStream("js/GameBoyIO.js"));
        Reader GameBoyCoreLibReader = new InputStreamReader(getClass().getResourceAsStream("js/GameBoyCore.js"));
        //Reader XAudioServerLibReader = new InputStreamReader(getClass().getResourceAsStream("js/XAudioServer.js"));

        //Replace this with your current gameboy rom file.
        Reader RomLibReader = new InputStreamReader(getClass().getResourceAsStream("js/roms/crystal.js"));

            //Put the files into javascript engine.

        context.evaluateReader(globalScope, base64LibReader, "base64.js", 1, null);
        //context.evaluateReader(globalScope, msgpackLibReader, "msgpack.js", 1, null);
        context.evaluateReader(globalScope, json2LibReader, "json2.js", 1, null);
        context.evaluateReader(globalScope, terminalLibReader, "terminal.js", 1, null);
        //context.evaluateReader(globalScope, resizeLibReader, "resize.js", 1, null);
        context.evaluateReader(globalScope, GameBoyIOLibReader, "GameBoyIO.js", 1, null);
        context.evaluateReader(globalScope, GameBoyCoreLibReader, "GameBoyCore.js", 1, null);
        //context.evaluateReader(globalScope, XAudioServerLibReader, "XAudioServer.js", 1, null);
        context.evaluateReader(globalScope, RomLibReader, "crystal.js", 1, null);

        // Add a global variable out that is a JavaScript reflection of the System.out variable:
        Object wrappedOut = Context.javaToJS(System.out, globalScope);
        ScriptableObject.putProperty(globalScope, "out", wrappedOut);

        String code = "cout('Gameboy frame finished' + frames);";
        context.evaluateString(globalScope, code, "<mem>", 1, null);

        // Tried a hack here (this below is stupid but just to prove my suspensions)
        for(int i = 0;i <=50; i++) {
             context.evaluateString(globalScope, code, "<mem>", 1, null);
             System.out.println("okay next frame");
        }
    } catch(IOException ioe) {
        ioe.printStackTrace();
        } finally {
            Context.exit();
        }
    }    
}

我试图按顺序检索帧计数器，但每次输出都是frame=1..这意味着脚本会在我想要的时候从头开始重新启动，以保持不间断并保留引用以前的变量，老实说，它甚至不应该重新启动，它应该继续使用 setTimeout() 函数

另外，如果问为什么 Rhino 总是占用我的 50% 的 CPU 和 100 MB 的内存，而它甚至没有做任何事情，那么当它到达 setTimeout 无限循环时，可能应该趋于平稳。 (如果可能的话)。

我想我需要 context.compileString 只是无法弄清楚。

Answer 1

找到的解决方案只需将其放入任何 JavaScript 中，然后再运行 rhino。

还有一个无法找到 JavaAdapter 的错误，请确保您使用的 Rhino 1.7R5 已修复此错误。

只需谷歌搜索rhino-1.7R5-SNAPSHOT.jar

(function(global) {
    var timer = new java.util.Timer();
    var counter = 1;
    var ids = {};

    global.setTimeout = function(fn, delay) {
        var id = counter;
        counter += 1;
        ids[id] = new JavaAdapter(java.util.TimerTask, { run : fn });
        timer.schedule(ids[id], delay);
        return id;
    };

    global.clearTimeout = function(id) {
        ids[id].cancel();
        timer.purge();
        delete ids[id];
    };

    global.setInterval = function(fn, delay) {
        var id = counter;
        counter += 1;
        ids[id] = new JavaAdapter(java.util.TimerTask, { run : fn });
        timer.schedule(ids[id], delay, delay);
        return id;
    };

    global.clearInterval = global.clearTimeout;

    // exports object in case of "isCommonJS"
    global.exports = {};

})(this);