阶段性复习与应用——复数的四则运算

看到这篇博文的同学们,大多数都学习了之前的博文了,那么,现在,我们通过 “复数的实现” 这一项目,来复习一下之前所有博文中的重要知识点!

首先,我们现在来构建一个类,来存储这个复数的 实部虚部:

package com.mec.complex;

public class Complex {
	private double real;
	private double vir;	
}

现在,我们来构造方法,并且编写 Getter() 和 Setter() 方法:

package com.mec.complex;

public class Complex {
	private double real;
	private double vir;
	
	public Complex(double real, double vir) {
		this.real = real;
		this.vir = vir;
	}
	
	public Complex() {
		this(0.0, 0.0);
	}

	public Complex(Complex c) {
		this(c.real, c.vir);
	}

	public double getReal() {
		return real;
	}

	public void setReal(double real) {
		this.real = real;
	}

	public double getVir() {
		return vir;
	}

	public void setVir(double vir) {
		this.vir = vir;
	}

}

基本的录入、输出方法,我们就编写完成了。

现在我们来编写加、减、乘、除方法:

1.方法:

public Complex add(Complex c) {
	this.real += c.real;
	this.vir += c.vir;
		
	return this;
}

public Complex add(Complex one, Complex another) {
	return new Complex (one).add(another);
}

我们在这里,定义了两种加法的实现方法,以满足用户不同的输入,或者我们之后的代码的需求。

2.取相反数方法:

private static Complex opposite(Complex c) {
	return new Complex(-c.real, -c.vir);
}

(因为这个方法,我们是为了完成减法操作才创造的,并不想外类调用,所以用private修饰)

3.方法:

public Complex sub(Complex c) {
	return this.add(opposite(c));	
}
	
public static Complex sub(Complex one, Complex another) {
	return new Complex(one).add(opposite(another));	
}

4.方法:

public Complex mul(Complex one) {
	double real = this.real;
		
	this.real = real * one.real - this.vir * one.vir;
	this.vir = real * one.vir - this.vir * one.real;
		
	return this;
}

public static Complex mul(Complex one, Complex another) {
	return new Complex (one).mul(another);
}

5.取倒数方法:

private static Complex reciprocal(Complex c) {	//这里我们通过参考公式可得:倒数=原复数*共轭数/模长
	double model = c.real * c.real + c.vir * c.vir; //复数的模长
		
	if(Math.abs(model) < 1e-6) {
		return null;
	}	//因为double的精度为1e-6,所以,小于1e-6时,可以忽略不计
		
	return new Complex(c.real / model, -c.vir / model);
}

(因为这个方法,我们是为了完成除法操作才创造的,并不想外类调用,所以用private修饰)

6.方法:

public static Complex div(Complex one, Complex another) {
	Complex rec = reciprocal(another);
	return rec == null ? null : new Complex(one).mul(rec);
}

注意:本人上面的四则运算方法,有的有 static修饰符 修饰,有的却没有,这里做下解释:
(1)有static修饰符修饰的方法,是我们之后的 Demo.java类 (即:测试类)所要调用的方法,而且这些方法的调用也更符合我们的认知(即:四则运算 的 参数是两个 “复数”);
(2)没有 static修饰符 修饰的方法,则是为了辅助我们完成 有 static修饰符 修饰的方法。

那么,我们总结下所有成员 和 方法:
Complex.java:

package com.mec.complex;

public class Complex {
	private double real;
	private double vir;
	
	public Complex(double real, double vir) {
		this.real = real;
		this.vir = vir;
	}
	
	public Complex() {
		this(0.0, 0.0);
	}

	public Complex(Complex c) {
		this(c.real, c.vir);
	}

	public double getReal() {
		return real;
	}

	public void setReal(double real) {
		this.real = real;
	}

	public double getVir() {
		return vir;
	}

	public void setVir(double vir) {
		this.vir = vir;
	}
	
	public Complex add(Complex c) {
		this.real += c.real;
		this.vir += c.vir;
		
		return this;
	}
	
	public static Complex add(Complex one, Complex another) {
		return new Complex (one).add(another);
	}
	
	private static Complex opposite(Complex c) {
		return new Complex(-c.real, -c.vir);
	}
	
	private static Complex reciprocal(Complex c) {
		double model = c.real * c.real + c.vir * c.vir;
		
		if(Math.abs(model) < 1e-6) {
			return null;
		}
		
		return new Complex(c.real / model, -c.vir / model);
	}
	
	public Complex sub(Complex c) {
		return this.add(opposite(c));
	}
	
	public static Complex sub(Complex one, Complex another) {
		return new Complex(one).add(opposite(another));
	}
	
	public Complex mul(Complex one) {
		double real = this.real;
		
		this.real = real * one.real - this.vir * one.vir;
		this.vir = real * one.vir - this.vir * one.real;
		
		return this;
	}

	public static Complex mul(Complex one, Complex another) {
		return new Complex (one).mul(another);
	}
	
	public static Complex div(Complex one, Complex another) {
		Complex rec = reciprocal(another);
		return rec == null ? null : new Complex(one).mul(rec);
	}
	
	@Override
	public String toString() {
		return "(" + real + "," + vir + ")";
	}
	
	@Override
	public boolean equals(Object obj) {
		if(null == obj) {
			return false;
		}
		if(this == obj) {
			return true;
		}
		if(!(obj instanceof Complex)) {
			return false;
		}
		Complex c = (Complex) obj;
		return Math.abs(this.real - c.real) < 1e-6
				&& Math.abs(this.vir - c.vir) < 1e-6;
	}
	
}

现在,我们来编写一个主函数的类来调用这些“工具”吧:
Demo.java:

package com.mec.complex;

public class Demo {

	public static void main(String[] args) {
		Complex num1 = new Complex(1.0, 2.0);
		Complex num2 = new Complex(3.0, 4.0);
		
		System.out.println("num1:" + num1);
		System.out.println("num2:" + num2);

		Complex num3 = Complex.add(num1, num2);
		System.out.println("the add result :" + num3);
		
		Complex num4 = Complex.sub(num1, num2);
		System.out.println("the sub result :" + num4);
		
		Complex num5 = Complex.mul(num1, num2);
		System.out.println("the mul result :" + num5);
		
		Complex num6 = Complex.div(num2, num1);
		System.out.println("the div result :" + num6);
		
	}

}

那么,测试结果如下:
在这里插入图片描述
那么,对于复数的基本操作的“工具”我们就做好了,我们之前所学习的重要知识点,差不多到此,就全部复习完了。

posted @ 2020-03-04 21:51  在下右转,有何贵干  阅读(547)  评论(0编辑  收藏  举报