我们经常在文本检查的时候要查找其中的数字，但是可能文本是中文数字字符串“两亿三千万”等等，但是要转化为阿拉伯数字才能被程序识别，本文提供了一种将中文或者英文转化为阿拉伯数字的一种方法。
其实中文和英文只是我只知道这两种语言，但是从程序设计的角度来看，该方法是用于多种语种，只要将数据库配置好即可
中文转化为阿拉伯数字： 可以通用于不要分隔符区分单词的语种 ，例如 中文、日文、韩文等（配置相应数据库即可）
英文转化为阿拉伯数字： 可以通用语用空格分隔单词的语种，例如 英文、法文、德文等（配置相应的数据库即可）
1、 设计语种对应表
|
t_SYS_Num |
|||
|
fid |
ftext |
fvalue |
ftype |
|
1 |
零 |
0 |
1 |
|
2 |
一 |
1 |
1 |
|
3 |
二 |
2 |
1 |
|
4 |
三 |
3 |
1 |
|
5 |
四 |
4 |
1 |
|
6 |
五 |
5 |
1 |
|
7 |
六 |
6 |
1 |
|
8 |
七 |
7 |
1 |
|
9 |
八 |
8 |
1 |
|
10 |
九 |
9 |
1 |
|
11 |
壹 |
1 |
1 |
|
12 |
贰 |
2 |
1 |
|
13 |
叁 |
3 |
1 |
|
14 |
肆 |
4 |
1 |
|
15 |
伍 |
5 |
1 |
|
16 |
陆 |
6 |
1 |
|
17 |
柒 |
7 |
1 |
|
18 |
捌 |
8 |
1 |
|
19 |
玖 |
9 |
1 |
|
22 |
one |
1 |
0 |
|
23 |
two |
2 |
0 |
|
24 |
three |
3 |
0 |
|
25 |
four |
4 |
0 |
|
26 |
five |
5 |
0 |
|
27 |
six |
6 |
0 |
|
28 |
seven |
7 |
0 |
|
29 |
eight |
8 |
0 |
|
30 |
nine |
9 |
0 |
|
31 |
ten |
10 |
0 |
|
32 |
eleven |
11 |
0 |
|
33 |
twelve |
12 |
0 |
|
34 |
thirteen |
13 |
0 |
|
35 |
fourteen |
14 |
0 |
|
36 |
fifteen |
15 |
0 |
|
37 |
sixteen |
16 |
0 |
|
38 |
seventeen |
17 |
0 |
|
39 |
eighteen |
18 |
0 |
|
40 |
nineteen |
19 |
0 |
|
41 |
twenty |
20 |
0 |
|
42 |
twenty-one |
21 |
0 |
|
43 |
twenty-two |
22 |
0 |
|
44 |
twenty-three |
23 |
0 |
|
45 |
twenty-four |
24 |
0 |
|
46 |
twenty-five |
25 |
0 |
|
47 |
twenty-six |
26 |
0 |
|
48 |
twenty-seven |
27 |
0 |
|
49 |
twenty-eight |
28 |
0 |
|
50 |
twenty-nine |
29 |
0 |
|
51 |
thirty |
30 |
0 |
|
52 |
thirty-one |
31 |
0 |
|
53 |
thirty-two |
32 |
0 |
|
54 |
thirty-three |
33 |
0 |
|
55 |
thirty-four |
34 |
0 |
|
56 |
thirty-five |
35 |
0 |
|
57 |
thirty-six |
36 |
0 |
|
58 |
thirty-seven |
37 |
0 |
|
59 |
thirty-eight |
38 |
0 |
|
60 |
thirty-nine |
39 |
0 |
|
61 |
forty |
40 |
0 |
|
62 |
forty-one |
41 |
0 |
|
63 |
forty-two |
42 |
0 |
|
64 |
forty-three |
43 |
0 |
|
65 |
forty-four |
44 |
0 |
|
66 |
forty-five |
45 |
0 |
|
67 |
forty-six |
46 |
0 |
|
68 |
forty-seven |
47 |
0 |
|
69 |
forty-eight |
48 |
0 |
|
70 |
forty-nine |
49 |
0 |
|
71 |
fifty |
50 |
0 |
|
72 |
fifty-one |
51 |
0 |
|
73 |
fifty-two |
52 |
0 |
|
74 |
fifty-three |
53 |
0 |
|
75 |
fifty-four |
54 |
0 |
|
76 |
fifty-five |
55 |
0 |
|
77 |
fifty-six |
56 |
0 |
|
78 |
fifty-seven |
57 |
0 |
|
79 |
fifty-eight |
58 |
0 |
|
80 |
fifty-nine |
59 |
0 |
|
81 |
sixty |
60 |
0 |
|
82 |
sixty-one |
61 |
0 |
|
83 |
sixty-two |
62 |
0 |
|
84 |
sixty-three |
63 |
0 |
|
85 |
sixty-four |
64 |
0 |
|
86 |
sixty-five |
65 |
0 |
|
87 |
sixty-six |
66 |
0 |
|
88 |
sixty-seven |
67 |
0 |
|
89 |
sixty-eight |
68 |
0 |
|
90 |
sixty-nine |
69 |
0 |
|
91 |
seventy |
70 |
0 |
|
92 |
seventy-one |
71 |
0 |
|
93 |
seventy-two |
72 |
0 |
|
94 |
seventy-three |
73 |
0 |
|
95 |
seventy-four |
74 |
0 |
|
96 |
seventy-five |
75 |
0 |
|
97 |
seventy-six |
76 |
0 |
|
98 |
seventy-seven |
77 |
0 |
|
99 |
seventy-eight |
78 |
0 |
|
100 |
seventy-nine |
79 |
0 |
|
101 |
eighty |
80 |
0 |
|
102 |
eighty-one |
81 |
0 |
|
103 |
eighty-two |
82 |
0 |
|
104 |
eighty-three |
83 |
0 |
|
105 |
eighty-four |
84 |
0 |
|
106 |
eighty-five |
85 |
0 |
|
107 |
eighty-six |
86 |
0 |
|
108 |
eighty-seven |
87 |
0 |
|
109 |
eighty-eight |
88 |
0 |
|
110 |
eighty-nine |
89 |
0 |
|
111 |
ninety |
90 |
0 |
|
112 |
ninety-one |
91 |
0 |
|
113 |
ninety-two |
92 |
0 |
|
114 |
ninety-three |
93 |
0 |
|
115 |
ninety-four |
94 |
0 |
|
116 |
ninety-five |
95 |
0 |
|
117 |
ninety-six |
96 |
0 |
|
118 |
ninety-seven |
97 |
0 |
|
119 |
ninety-eight |
98 |
0 |
|
120 |
ninety-nine |
99 |
0 |
|
121 |
十 |
10 |
1 |
|
122 |
拾 |
10 |
1 |
fId：主键序号
fText：语种字符串
fValue：对应值
fType：语种分类
|
T_SYS_Unit |
|||||
|
fid |
funit |
fvalue |
flevel |
fisspace |
ftype |
|
1 |
十 |
10 |
1 |
No |
1 |
|
2 |
百 |
100 |
2 |
No |
1 |
|
3 |
千 |
1000 |
3 |
No |
1 |
|
4 |
万 |
10000 |
4 |
Yes |
1 |
|
5 |
亿 |
100000000 |
5 |
Yes |
1 |
|
6 |
拾 |
10 |
1 |
No |
1 |
|
7 |
佰 |
100 |
2 |
No |
1 |
|
8 |
仟 |
1000 |
3 |
No |
1 |
|
9 |
萬 |
10000 |
4 |
Yes |
1 |
|
10 |
億 |
100000000 |
5 |
Yes |
1 |
|
11 |
hundred |
100 |
1 |
No |
0 |
|
12 |
thousand |
1000 |
2 |
No |
0 |
|
13 |
million |
1000000 |
3 |
Yes |
0 |
|
14 |
billion |
1000000000 |
4 |
Yes |
0 |
|
15 |
and |
1 |
0 |
No |
0 |
|
16 |
hundreds |
100 |
1 |
No |
0 |
|
17 |
thousands |
1000 |
2 |
No |
0 |
|
18 |
millions |
1000000 |
3 |
Yes |
0 |
|
19 |
billions |
1000000000 |
4 |
Yes |
0 |
fId：主键序号
fUnit：单位
fValue：对应的倍数
fLevel：单位排序级别
fIsspace：是否作为分隔符
fType：语种分类
2、 搜索符合原文中的数字字符串
a) 在文本中查找符合表t_SYS_Num 和t_SYS_Unit 中text的最长的字符串
3、 用递归法分割数字字符串
a) 首先将按照fLevel最大的text进行分割
i. 例如中文： 以“億”、“亿” 进行分割
ii. “三千二百万零二十一亿零五千” ——被分离为 “三千二百万零二十一” 和“五千”
iii. 然后按照下一级的单位进行分离，直到所有的fisspace为True的分离完毕
4、 将万以为的数字字符串转化为数字
5、 数字*以前分离出来的单位
6、 将所有的分离开的数字加起来得到最后的数值
代码实现：
/// <summary>
/// 数字字符串信息
/// </summary>
public struct NumString {
public int Paragraph; public int Start; public int Length; public string numstring; public long Num; } /// <summary> /// 语种信息 /// </summary> public enum LanguageType { English = 0, Chinese = 1, }
private ArrayList GetNumString(string[] content, LanguageType type) { switch (type)
{
case LanguageType.Chinese: return GetNumString_CH(content); case LanguageType.English: return GetNumString_EN(content); default: return GetNumString_EN(content);
} } /// <summary> /// 查找字符串数组中英文字符串 /// </summary> /// <param name="content">文本数组</param> /// <returns>数字字符串数组</returns> private ArrayList GetNumString_EN(string[] content) { string sql = "select * from t_SYS_Num where ftype =0"; DataSet ds = DCBase.ExecuteQuery(sql); sql = "select * from t_SYS_Unit where ftype =0"; DataSet ds1 = DCBase.ExecuteQuery(sql); DataTable dt1 = ds.Tables[0]; DataTable dt2 = ds1.Tables[0]; DataView dv1 = dt1.DefaultView; DataView dv2 = dt2.DefaultView; ArrayList nubmstring = new ArrayList(); for (int i = 0; i < content.Length; i++) { if (content[i] == null || content[i] == "") continue; string[] content_str = content[i].Split(' '); int start = 0; for (int j = 0; j < content_str.Length; j++) { int length = 0; int step = 0; dv1.RowFilter = "ftext='" + content_str[j] + "'"; //该单词不是数字 if (dv1.Count == 0) { start += content_str[j].Length; start++; continue; } else { length += content_str[j].Length; for (int n = j + 1; n < content_str.Length; n++) { dv1.RowFilter = "ftext='" + content_str[n] + "'"; dv2.RowFilter = "funit='" + content_str[n] + "'"; if (dv1.Count == 0 && dv2.Count == 0) { break; } else { length += content_str[n].Length + 1; step++; } } } if (length > 0) { NumString ns = new NumString(); ns.Length = length; ns.Start = start; ns.Paragraph = i; ns.numstring = content[i].Substring(start, length); ns.Num = Splite(ns.numstring, LanguageType.English); nubmstring.Add(ns); j += step; start += length + 1; } else { start += content_str[j].Length; start++; } } } return nubmstring; } /// <summary> /// 查找字符串数组中中文字符串 /// </summary> /// <param name="content">文本数组</param> /// <returns>数字字符串数组</returns> private ArrayList GetNumString_CH(string[] content) { string sql = "select * from t_SYS_Num where ftype =1"; DataSet ds = DCBase.ExecuteQuery(sql); sql = "select * from t_SYS_Unit where ftype =1"; DataSet ds1 = DCBase.ExecuteQuery(sql); DataTable dt1 = ds.Tables[0]; DataTable dt2 = ds1.Tables[0]; DataView dv1 = dt1.DefaultView; DataView dv2 = dt2.DefaultView; ArrayList nubmstring = new ArrayList(); for (int i = 0; i < content.Length; i++) { if (content[i] == null || content[i] == "") continue; string content_str = content[i]; for (int j = 0; j < content_str.Length; j++) { int length = 0; dv1.RowFilter = "ftext='" + content_str[j] + "'"; if (dv1.Count == 0) continue; else { length++; for (int n = j + 1; n < content_str.Length; n++) { dv1.RowFilter = "ftext='" + content_str[n] + "'"; dv2.RowFilter = "funit='" + content_str[n] + "'"; if (dv1.Count == 0 && dv2.Count == 0) break; else { length++; } } } if (length > 0) { NumString ns = new NumString(); ns.Length = length; ns.Start = j; ns.Paragraph = i; ns.numstring = content_str.Substring(j, length); ns.Num = Splite(ns.numstring, LanguageType.Chinese); nubmstring.Add(ns); j += length; } } } return nubmstring; }
/// <summary> /// 将中文字符串数组转化为阿拉伯数组 /// </summary> /// <param name="number_CH"></param> /// <returns></returns> private long[] CH2A(ArrayList number_CH) { long[] num = new long[number_CH.Count]; for (int i = 0; i < number_CH.Count; i++) { NumString ns = (NumString)number_CH[i]; string numberstring = ns.numstring; num[i] = Splite(numberstring, LanguageType.Chinese); } return num; } /// <summary> /// 分解中文或者英文数字字符串 /// </summary> /// <param name="numstring"></param> /// <param name="type"></param> /// <returns></returns> private long Splite(string numstring, LanguageType type) { string sql = "select max(flevel) from t_SYS_Unit where fisspace=true and ftype=" + (int)type; object obj = DCBase.ExecuteScalar(sql); int maxlevel = 0; if (obj != null && obj != DBNull.Value) maxlevel = (int)obj; return Splite(numstring, type, maxlevel); } /// <summary> /// 分解中文或者英文数字字符串 /// </summary> /// <param name="numstring"></param> /// <param name="type"></param> /// <returns></returns> private long Splite(string numstring, LanguageType type, int maxLevel) { string sql = "select flevel from t_SYS_Unit where fisspace=true and ftype =" + (int)type + " and flevel<=" + maxLevel + " order by flevel desc"; object obj = DCBase.ExecuteScalar(sql); int maxlevel = 0; if (obj != null && obj != DBNull.Value) { maxlevel = (int)obj; sql = "select * from t_SYS_Unit where fisspace=true and ftype =" + (int)type + " and flevel=" + maxlevel + " order by flevel desc"; DataSet ds = DCBase.ExecuteQuery(sql); numstring = numstring.Trim(); if (ds.Tables[0].Rows.Count > 0) { string numstring2 = numstring; foreach (DataRow dr in ds.Tables[0].Rows) { string splitestr = ""; if (numstring.ToLower().IndexOf(dr["funit"].ToString()) > -1) { splitestr = dr["funit"].ToString(); numstring2 = numstring.ToLower().Replace(splitestr, "♀"); break; } } string[] num = numstring2.Split('♀'); if (num.Length == 1) { return Splite(numstring, type, maxLevel - 1); } else { return Splite(num[0], type, maxLevel - 1) * (int)ds.Tables[0].Rows[0]["fvalue"] + Splite(num[1], type, maxLevel - 1); } } else { return Tran2A(numstring, type); } } else { return Tran2A(numstring, type); } }
/// <summary> /// 万以内的数字(英文) /// </summary> /// <param name="numstring"></param> /// <returns></returns> private int EN2A(string numstring) { string sql = "select * from t_SYS_Num where ftype =0"; DataSet ds = DCBase.ExecuteQuery(sql); sql = "select * from t_SYS_Unit where fisspace=false and ftype =0 order by flevel desc"; DataSet ds1 = DCBase.ExecuteQuery(sql); DataTable dt1 = ds.Tables[0]; DataTable dt2 = ds1.Tables[0]; int result = 0; int index = -1; int value = 0; int maxindex = -1;//查找的最优一个单位出现的位置 string[] numstr = numstring.Split(' '); foreach (DataRow dr in dt2.Rows) { index = -1; value = 0; for (int i = 0; i < numstr.Length; i++) { if (numstr[i].ToLower() == dr["funit"].ToString()) { index = i; break; } } if (index > 0) { maxindex = index; foreach (DataRow row in dt1.Rows) { value = 0; if (numstr[index - 1].ToString() == row["ftext"].ToString()) { value = (int)row["fvalue"]; break; } } } result += value * (int)dr["fvalue"]; } //个位结尾的情况 value = 0; //处理个位数 if (maxindex != numstr.Length - 1) { foreach (DataRow row in dt1.Rows) { if (numstr[maxindex + 1].ToString() == row["ftext"].ToString()) { value = (int)row["fvalue"]; break; } } } return result + value; }
/// <summary> /// 万以内的数字(中文) /// </summary> /// <param name="numstring"></param> /// <returns></returns> private long CH2A(string numstring) { string sql = "select * from t_SYS_Num where ftype =1"; DataSet ds = DCBase.ExecuteQuery(sql); sql = "select * from t_SYS_Unit where fisspace=false and ftype =1 order by flevel desc"; DataSet ds1 = DCBase.ExecuteQuery(sql); DataTable dt1 = ds.Tables[0]; DataTable dt2 = ds1.Tables[0]; long result = 0; int index = -1; long value = 0; int maxindex = -1;//查找的最优一个单位出现的位置 foreach (DataRow dr in dt2.Rows) { value = 0; index = numstring.IndexOf(dr["funit"].ToString()); //如果数字是以单位开头，就在该数字前加“一”，例如“十四” 变成“一十四”； if (index == 0) { numstring = "一" + numstring; index = numstring.IndexOf(dr["funit"].ToString()); } if (index > 0) { maxindex = index; foreach (DataRow row in dt1.Rows) { value = 0; if (numstring[index - 1].ToString() == row["ftext"].ToString()) { value = (int)row["fvalue"]; break; } } } else continue; result += value * (int)dr["fvalue"]; } //个位结尾的情况 value = 0; //处理个位数 if (maxindex != numstring.Length - 1) { foreach (DataRow row in dt1.Rows) { if (numstring[maxindex + 1].ToString() == row["ftext"].ToString()) { value = (int)row["fvalue"]; break; } } } return result + value; }