graph plotting

[Sirine Sirin]

I'm new here and impressed from your forum. I'm quite new in Java and i need some help regarding the Frame and Graphics classes in java.

I have written 3 classes (function, parabola and polynomial), which represent functions with a variable . I want to extend my function class with a method named public void plot(double xmin, double xmax), which opens a frame with the graph of determined function. the parameter xmin and xmax give the x-area, where the graph should be drawn. A labeld coordinate axis must also appear in the frame. Can you help me?

Here are my classes :

/**
 * Class to respresent parabola functions
 */
public class Function{
	
    /**
     * some static constants 
     */
    public final static int n = 1000;
    public final static double eps = 0.000000001;
    public final static Function NULL=new Function(){
        public double map(double x){return 0.0; }
        public String toString() {return "0.0"; }
    };
    
    
    /**
     * Defines the mapping of the function. The default mapping is not-defined
     * so the return value is Double.NaN for instances of this class. Subclass
     * should override this method to define the mapping
     * @param x
     * @return
     */
    public double map(double x){return Double.NaN;}
    
    /**
     * Evaluates the function at each arguemt of the array
     * @param x array of function values
     * @return
     */
    public double[] map(double[] x){
        double[] y=new double[x.length];
        for (int i=0; i<x.length; i++) y[i]=map(x[i]);
        return y;
    }
    
    /**
     * Access to the derivation of the Function instances. Subclass should
     * override this method. Use Double.NaN at non-differntiable positions
     * @return derivative of this function instance
     */
    public Function derivate(){
        return this;
    }

    /**
     * a
     * approximate the integral of this function using n boxes
     * @param a stating of the integration
     * @param b ending of the integration
     * @param n number of boxes (equal width)
     * @return approximation of the integral over [a,b] of this function
     */    
    public double integrate(double a, double b, int n){
        double delta=(b-a)/(double)n;
        double area=0;
        for (double i=0; i<n; i++) area+=map(a+i*delta);
        return area*delta;
    }

    /**
     * If possible this method should return the exact integral value. 
     * Subclasses should override this method.
     * The default is to call the approximation integrate(a,b,n).
     * @param a stating of the integration
     * @param b ending of the integration
     * @param n number of boxes (equal width)
     * @return integral over [a,b] of this function
     */
    public double integrate(double a, double b){
        return this.integrate(a,b,this.n);
    }

    /** 
     * Computes a zero-crossing using the newton iteration
     * @param x_0 initial argument
     * @return Zero-Crossing or Double.NaN
     */
    public double newtonIteration(double x_0){
        double d=42.0;
        Function f_1=this.derivate();
        for (int i=0; ((++i<this.n) && (Math.abs(d)>eps)); x_0+=d)
            d=-map(x_0)/f_1.map(x_0);
        return (Math.abs(d)>eps) ? Double.NaN : x_0;
    }
    
    /**
     * String representation of this function
     * @return "undefined"
     */
    public String toString(){
        return "undefined";
    }
}

/**
 * Class to respresent parabola functions
 *
 */
public class Parabola extends Polynomial{
    
    /**
     * Constructor for a parabola object ax? + bx + c
     * @param a coefficient for x^2
     * @param b coefficient for x^1
     * @param c coefficient for x^0
     */
    public Parabola(double a, double b, double c){        
        super(new double[] {c,b,a});        
    }
    
    /**
     * Compute the Zero Crossings of the parabola
     * @return zero-crossings, i.e. x with f(x)=0
     */
    public double[] zeroCrossing(){
        double c=map(0);
        double b=map(0.5)-map(-0.5);
        double a=map(1)-b-c;

        //Diskriminante
        double d = b*b - 4.0*c*a;
        
        if (this.getDegree()<=1){
            if (b==0) return new double[0];
            else return new double[] {-c/b};        
        }
                
        if (d==0) return new double[] {-0.5*b/a};
        if (d<0) return new double[0];
        
        return new double[] { 0.5*(-b-Math.sqrt(d))/a, 0.5*(-b+Math.sqrt(d))/a};
    }

}

/**
 * Class to respresent polynomial functions
 *
 * 
 * Bug in limPosInf() for constant functions repaired
 * 
 */
public class Polynomial extends Function{

    // Koeffizienten des Polynoms
    private double[] a=new double[0];
    
    /**
     * Constructor for a polynomial function, sum a[i] * x^i
     * @param a coeffients of the polynomial function
     */
    public Polynomial(double[] a){
        int deg = a.length-1;
        for (int i=deg; ((i>=0) & (a[i]==0.0)) ; i--) deg--;
        this.a = new double[deg+1];
        for (int i=0; i<=deg; i++) this.a[i]=a[i];
    }
    
    /**
     * Access to the degree of the polynomial, i.e. the maximal "nonzero" power
     * @return degree of the polynomial
     */
    public int getDegree(){
        return a.length-1;
    }
    
    /**
     * Access to the derivative of the polynomial
     * @return the derivative of the polynomial
     */
    public Function derivate(){
        if (this.getDegree()==0) return Function.NULL;
        double[] b=new double[this.a.length-1];
        for (int i=0; i<b.length; )
            b[i]=((double)++i)*this.a[i];
        return new Polynomial(b);
    }
    
    /**
     * Computes the integral of the polynomial over an interval
     * @param a begining of the integration interval
     * @param b ending of the integration interval
     * @return integral over [a,b] of polynomial
     */
    public double integrate(double a, double b){
        double[] a_=new double[this.a.length+1];
        a_[0]=0.0;
        for (int i=1; i<a_.length; i++)
            a_[i]=this.a[i-1]/((double)i);
        Polynomial F = new Polynomial(a_);
        return F.map(b)-F.map(a);
    }
    
    /**
     * function value of the polynomial at x
     * @param x argument of the function
     * @return value f(x) of the function
     */
    public double map(double x){
        double y=0.0;
        for (int i=getDegree(); i>=0; i--) y=y*x+a[i];
        return y;
    }
    
    /**
     * computes the limit for x->oo
     * @return limit for x->oo
     */
    public double limPosInf(){
        if (getDegree()==0) return map(0);
        return (a[getDegree()]>0)
                ? Double.POSITIVE_INFINITY
                : Double.NEGATIVE_INFINITY;
    }

    /**
     * computes the limit for x->-oo
     * @return limit for x->-oo
     */
    public double limNegInf(){
        return (getDegree()%2==0)
                ?  limPosInf()
                : -limPosInf();
    }
    
    /** String representation
     * @return awesome string representation
     */
    public String toString(){
        //sorry for this bad code ;)
        String s="";
        String vz="";
        for (int i=0; i<a.length; i++)
            if (a[i]!=0) {
                s=((i==0) ? a[i] : (a[i]==1) ? "" : (a[i]==-1) ? "-" : ""+a[i])
                 +((i==0) ? "" :(i==1) ? "x" : "x^"+i)//aktuelle Potenz
                 +vz                                  //zusätzliches Vorzeichen
                 +s;                                  //Kleinere Potenzen
                
                //Berechnung des aktuellen Vorzeichens
                if (a[i]>0 ) vz="+"; else vz="";
            }            

        return s;
    }
}

[pete stein]

Some useful FAQ links that may have some bearing on your post:

ShowSomeEffort
DoYourOwnHomework
BeForthrightWhenCrossPostingToOtherSites
the last refers to this link and possibly others: graph-plotting.html

and the general FAQ: HowToAskQuestionsOnJavaRanch

Hope this helps.

[Ulf Dittmer]

The PlotTest and PlotSineCurve examples in the CodeBarn may provide some inspiration: http://faq.javaranch.com/java/CodeBarnIntermediateJava

[Sirine Sirin]

i thought this forum is to help beginners I can find myself such codes on internet, it's not difficult Anyway, thanks for your help. I think, your view about helping is only showing the existing codes which are actually not concerned with my questions.

[Ulf Dittmer]

i thought this forum is to help beginners

It is. But that does not include going through 200+ lines of code, and trying to figure out how they might be adapted to what you're asking. That's simply too much to ask for. You'll have better success if you describe what thoughts you have had on the problem, and where exactly you're stuck implementing those.

The examples codes mentioned above do much of what you're asking. Have you looked at them, and thought about how they might be adapted to your case?

[Sirine Sirin]

Ulf Dittmer wrote:

i thought this forum is to help beginners

It is. But that does not include going through 200+ lines of code, and trying to figure out how they might be adapted to what you're asking. That's simply too much to ask for. You'll have better success if you describe what thoughts you have had on the problem, and where exactly you're stuck implementing those.

The examples codes mentioned above do much of what you're asking. Have you looked at them, and thought about how they might be adapted to your case?

if i had known how it's working, i hadn't asked for it here. I don't figure out, how i can fit the example code to plot(double xmin, double xmax). That's my problem.

[Rob Spoor]

sirine,

Please read your private messages regarding an important announcement.

Thank you,

Rob

[Maneesh Godbole]

Darn Rob,
You beat me to it.

[Sirine Sirin]

a short explanation : my name is Sirine (means sweety, in turkish) and my surname is sirin (means sweet, also in turkish)

My parents gave me this name cos they found me too sweety. But sirine is a quite normally turkish name I don't know why my name violates the naming policies

and concerning my problem i would like to ask some further questions.

How can i read the x-values from the function class (with the map-methode) in the plot-method?

[Rob Spoor]

Sirine Sirin wrote:a short explanation : my name is Sirine (means sweety, in turkish) and my surname is sirin (means sweet, also in turkish)

My parents gave me this name cos they found me too sweety. But sirine is a quite normally turkish name I don't know why my name violates the naming policies

As I told you per PM, it just looked so very suspicious. But it's all resolved now.

[Steve Luke]

Sirine Sirin wrote:How can i read the x-values from the function class (with the map-methode) in the plot-method?

The x values are generally the independent variables, from which the y values are calculated. So you don't read the x values, you generate them. So you might do something like:

//Generate values for x from 0.0 to 20.0 in 1.0 increments
for (double xVal = 0.0; xVal < 20.0; xVal += 1.0)
{
    //get the value for y for the given x
    double yVal = function.map(xVal);
    //then graph the results
}

You may have to put some thought into what range of x values you want to plot.