最近由于项目的原因，需要写一个货币数字转换中文的算法，先在网了找了一下，结果发现无一列外都是用(Replace)替换的方式来实现的，所以想写个另外的算法；因为本人是学数学出身的，所以用纯数学的方法实现。
注意：本文中的算法支持小于1023 (也就是9999亿兆)货币数字转化。
货币中文说明： 在说明代码之前，首先让我们回顾一下货币的读法。
10020002.23 读为 壹仟零贰万零贰元贰角叁分 1020 读为 壹仟零贰拾元整。 100000 读为 拾万元整 0.13 读为 壹角叁分
代码：
测试工程 static void Main(string[] args){
Console.WriteLine("请输入金额");string inputNum = Console.ReadLine();while (inputNum != "exit"){
//货币数字转化类NumCast nc = new NumCast();
if (nc.IsValidated(inputNum)){try{string chineseCharacter = nc.Converttochinese(inputNum);Console.WriteLine(chineseCharacter);}catch (Exception er){Console.WriteLine(er.Message);}}else{Console.WriteLine("不合法的数字或格式");}
Console.WriteLine("\n请输入金额");inputNum = Console.ReadLine();
}Console.ReadLine();}
测试结果如下：
货币转化类（NumCast类）功能介绍
1 常量的规定 /// /// 数位/// public enum NumLevel { Cent,Chiao,Yuan,Ten,Hundred,Thousand,TenThousand,hundredMillon,Trillion };
/// /// 数位的指数/// private int[] NumLevelExponent = new int[] { -2,-1,1,2,3,4,8,12 };
/// /// 数位的中文字符/// private string[] NumLeverChineseSign = new string[] { "分","角","元","拾","佰","仟","万","亿","兆" };
/// /// 大写字符/// private string[] NumChineseCharacter = new string[] {"零","壹","贰","叁","肆","伍","陆","柒","捌","玖"};
/// /// 整(当没有 角分 时)///
private const string EndOfInt = "整";
2：数字合法性验证，采用正则表达式验证 /// /// 正则表达验证数字是否合法/// /// /// public bool IsValidated(T Num){Regex reg = new Regex(@"^(([0])|([1-9]\d{0,23}))(\.\d{1,2})?$");if (reg.IsMatch(Num.ToString())){return true;
}return false;}
3： 获取数位 例如 1000的数位为 NumLevel.Thousand
/// /// 获取数字的数位 使用log/// /// /// private NumLevel GetNumLevel(double Num){double numLevelLength;NumLevel NLvl = new NumLevel();if (Num > 0){numLevelLength = Math.Floor(Math.Log10(Num));for (int i = NumLevelExponent.Length - 1; i >= 0; i--){if (numLevelLength >= NumLevelExponent[i]){NLvl = (NumLevel)i;break;}}}else{NLvl = NumLevel.Yuan;}return NLvl;
}
4：判断数字之间是否有跳位，也就是中文中间是否要加零，例如1020 就应该加零。 /// /// 是否跳位/// /// private bool IsDumpLevel(double Num){ if (Num > 0){NumLevel? currentLevel = GetNumLevel(Num);NumLevel? nextLevel = null;int numExponent = this.NumLevelExponent[(int)currentLevel];
double postfixNun = Math.Round(Num % (Math.Pow(10,numExponent)),2);if(postfixNun> 0)nextLevel = GetNumLevel(postfixNun);if (currentLevel != null && nextLevel != null){if (currentLevel > nextLevel + 1){return true;}}}return false;
}
5 把长数字分割为两个较小的数字数组，例如把9999亿兆，分割为9999亿和0兆，
因为计算机不支持过长的数字。 /// /// 是否大于兆，如果大于就把字符串分为两部分，/// 一部分是兆以前的数字/// 另一部分是兆以后的数字/// /// /// private bool IsBigThanTillion(string Num){bool isBig = false;if (Num.IndexOf('.') != -1){//如果大于兆if (Num.IndexOf('.') > NumLevelExponent[(int)NumLevel.Trillion]){isBig = true;}}else{//如果大于兆if (Num.Length > NumLevelExponent[(int)NumLevel.Trillion]){isBig = true;}}return isBig;}
/// /// 把数字字符串由‘兆’分开两个/// /// private double[] SplitNum(string Num){//兆的开始位double[] TillionLevelNums = new double[2];int trillionLevelLength;if (Num.IndexOf('.') == -1)trillionLevelLength = Num.Length - NumLevelExponent[(int)NumLevel.Trillion];elsetrillionLevelLength = Num.IndexOf('.') - NumLevelExponent[(int)NumLevel.Trillion];//兆以上的数字TillionLevelNums[0] = Convert.ToDouble(Num.Substring(0,trillionLevelLength));//兆以下的数字TillionLevelNums[1] = Convert.ToDouble(Num.Substring(trillionLevelLength ));return TillionLevelNums;}
6 是否以“壹拾”开头，如果是就可以把它变为“拾” bool isStartOfTen = false;while (Num >=10){if (Num == 10){isStartOfTen = true;break;}//Num的数位NumLevel currentLevel = GetNumLevel(Num);
int numExponent = this.NumLevelExponent[(int)currentLevel];Num = Convert.ToInt32(Math.Floor(Num / Math.Pow(10,numExponent)));if (currentLevel == NumLevel.Ten && Num == 1){isStartOfTen = true;break;}}return isStartOfTen;
7 合并大于兆连个数组转化成的货币字符串
/// /// 合并分开的数组中文货币字符/// /// /// private string ContactNumChinese(double[] tillionNums){string uptillionStr = CalculateChineseSign(tillionNums[0],NumLevel.Trillion,true,IsStartOfTen(tillionNums[0]));string downtrillionStr = CalculateChineseSign(tillionNums[1],null,false);string chineseCharactor = string.Empty;//分开后的字符是否有跳位if (GetNumLevel(tillionNums[1] * 10) == NumLevel.Trillion){chineseCharactor = uptillionStr + NumLeverChineseSign[(int)NumLevel.Trillion] + downtrillionStr;}else{chineseCharactor = uptillionStr + NumLeverChineseSign[(int)NumLevel.Trillion];if (downtrillionStr != "零元整"){chineseCharactor += NumChineseCharacter[0] + downtrillionStr;}else{chineseCharactor += "元整";}}return chineseCharactor;
}
8：递归计算货币数字的中文 /// /// 计算中文字符串/// /// /// /// /// 中文大写public string CalculateChineseSign(double Num,NumLevel? NL,bool IsDump,bool IsExceptTen){Num = Math.Round(Num,2);bool isDump = false;//Num的数位NumLevel? currentLevel = GetNumLevel(Num);int numExponent = this.NumLevelExponent[(int)currentLevel];
string Result = string.Empty;
//整除后的结果int prefixNum;//余数 当为小数的时候 分子分母各乘100double postfixNun ;if (Num >= 1){prefixNum = Convert.ToInt32(Math.Floor(Num / Math.Pow(10,numExponent)));postfixNun = Math.Round(Num % (Math.Pow(10,2);}else{prefixNum = Convert.ToInt32(Math.Floor(Num*100 / Math.Pow(10,numExponent+2)));postfixNun = Math.Round(Num * 100 % (Math.Pow(10,numExponent + 2)),2);postfixNun *= 0.01;}
if (prefixNum < 10 ){//避免以‘壹拾’开头if (!(NumChineseCharacter[(int)prefixNum] == NumChineseCharacter[1]&& currentLevel == NumLevel.Ten && IsExceptTen)){Result += NumChineseCharacter[(int)prefixNum];}else{IsExceptTen = false;}//加上单位if (currentLevel == NumLevel.Yuan ){////当为 “元” 位不为零时 加“元”。if (NL == null){Result += NumLeverChineseSign[(int)currentLevel];//当小数点后为零时 加 "整"if (postfixNun == 0){Result += EndOfInt;}}}else{Result += NumLeverChineseSign[(int)currentLevel];} //当真正的个位为零时 加上“元”if (NL == null && postfixNun < 1 && currentLevel > NumLevel.Yuan && postfixNun > 0){Result += NumLeverChineseSign[(int)NumLevel.Yuan];
}
}else{//当 前缀数字未被除尽时， 递归下去NumLevel? NextNL = null;if ((int)currentLevel >= (int)(NumLevel.TenThousand))NextNL = currentLevel;
Result += CalculateChineseSign((double)prefixNum,NextNL,isDump,IsExceptTen);if ((int)currentLevel >= (int)(NumLevel.TenThousand)){Result += NumLeverChineseSign[(int)currentLevel];}}
//是否跳位// 判断是否加零， 比如302 就要给三百 后面加零，变为 三百零二。if (IsDumpLevel(Num)){Result += NumChineseCharacter[0];isDump = true;
}
//余数是否需要递归if (postfixNun > 0){Result += CalculateChineseSign(postfixNun,NL,false);}else if (postfixNun == 0 && currentLevel > NumLevel.Yuan ){//当数字是以零元结尾的加上 元整 比如1000000一百万元整if (NL == null){Result += NumLeverChineseSign[(int)NumLevel.Yuan];Result += EndOfInt;}}
return Result;}
9：外部调用的转换方法。
/// /// 外部调用的转换方法/// /// /// public string Converttochinese(string Num){
if (!IsValidated(Num)){throw new OverflowException("数值格式不正确，请输入小于9999亿兆的数字且最多精确的分的金额！");}string chineseCharactor = string.Empty;if (IsBigThanTillion(Num)){double[] tillionNums = SplitNum(Num);chineseCharactor = ContactNumChinese(tillionNums);}else{double dNum = Convert.ToDouble(Num);chineseCharactor = CalculateChineseSign(dNum,IsStartOfTen(dNum));}return chineseCharactor;}
小结：
个人认为程序的灵魂是算法，大到一个系统中的业务逻辑，小到一个货币数字转中文的算法，处处都体现一种逻辑思想。
是否能把需求抽象成一个好的数学模型，直接关系到程序的实现的复杂度和稳定性。在一些常用功能中想些不一样的算法，对我们开拓思路很有帮助。 本文转载自‘中国IT实验室’
