用C#实现汉字转化为拼音

转载 2006年06月19日 13:15:00

 

using System;
using System.Collections.Generic;
using System.Text;
using System.Text.RegularExpressions;

namespace SplitWord
{
    
/// 用C#实现汉字转化为拼音 
    
/// 实现的原理就是先将汉字转化成为内码,然后通过内码和拼音的对照来查找。 

    public class chs2py
    
{
        
/**/
        
/**/
        
/**/
        
/// <summary> 
        
///将汉字转换成为拼音 
        
///作者:
        
///完成于:
        
/// </summary> 

        static Regex MyRegex = new Regex("^[一-龥]$");         //汉字的正则表达式.eg: if(MyRegex.IsMatch(chrstr.ToString()))
        private static int[] pyvalue = new int[]{-20319,-20317,-20304,-20295,-20292,-20283,-20265,-20257,-20242,-20230,-20051,-20036,-20032,-20026
                                        
-20002,-19990,-19986,-19982,-19976,-19805,-19784,-19775,-19774,-19763,-19756,-19751,-19746,-19741,-19739,-19728
                                        
-19725,-19715,-19540,-19531,-19525,-19515,-19500,-19484,-19479,-19467,-19289,-19288,-19281,-19275,-19270,-19263
                                        
-19261,-19249,-19243,-19242,-19238,-19235,-19227,-19224,-19218,-19212,-19038,-19023,-19018,-19006,-19003,-18996
                                        
-18977,-18961,-18952,-18783,-18774,-18773,-18763,-18756,-18741,-18735,-18731,-18722,-18710,-18697,-18696,-18526
                                        
-18518,-18501,-18490,-18478,-18463,-18448,-18447,-18446,-18239,-18237,-18231,-18220,-18211,-18201,-18184,-18183
                                        
-18181,-18012,-17997,-17988,-17970,-17964,-17961,-17950,-17947,-17931,-17928,-17922,-17759,-17752,-17733,-17730
                                        
-17721,-17703,-17701,-17697,-17692,-17683,-17676,-17496,-17487,-17482,-17468,-17454,-17433,-17427,-17417,-17202
                                        
-17185,-16983,-16970,-16942,-16915,-16733,-16708,-16706,-16689,-16664,-16657,-16647,-16474,-16470,-16465,-16459
                                        
-16452,-16448,-16433,-16429,-16427,-16423,-16419,-16412,-16407,-16403,-16401,-16393,-16220,-16216,-16212,-16205
                                        
-16202,-16187,-16180,-16171,-16169,-16158,-16155,-15959,-15958,-15944,-15933,-15920,-15915,-15903,-15889,-15878
                                        
-15707,-15701,-15681,-15667,-15661,-15659,-15652,-15640,-15631,-15625,-15454,-15448,-15436,-15435,-15419,-15416
                                        
-15408,-15394,-15385,-15377,-15375,-15369,-15363,-15362,-15183,-15180,-15165,-15158,-15153,-15150,-15149,-15144
                                        
-15143,-15141,-15140,-15139,-15128,-15121,-15119,-15117,-15110,-15109,-14941,-14937,-14933,-14930,-14929,-14928
                                        
-14926,-14922,-14921,-14914,-14908,-14902,-14894,-14889,-14882,-14873,-14871,-14857,-14678,-14674,-14670,-14668
                                        
-14663,-14654,-14645,-14630,-14594,-14429,-14407,-14399,-14384,-14379,-14368,-14355,-14353,-14345,-14170,-14159
                                        
-14151,-14149,-14145,-14140,-14137,-14135,-14125,-14123,-14122,-14112,-14109,-14099,-14097,-14094,-14092,-14090
                                        
-14087,-14083,-13917,-13914,-13910,-13907,-13906,-13905,-13896,-13894,-13878,-13870,-13859,-13847,-13831,-13658
                                        
-13611,-13601,-13406,-13404,-13400,-13398,-13395,-13391,-13387,-13383,-13367,-13359,-13356,-13343,-13340,-13329
                                        
-13326,-13318,-13147,-13138,-13120,-13107,-13096,-13095,-13091,-13076,-13068,-13063,-13060,-12888,-12875,-12871
                                        
-12860,-12858,-12852,-12849,-12838,-12831,-12829,-12812,-12802,-12607,-12597,-12594,-12585,-12556,-12359,-12346
                                        
-12320,-12300,-12120,-12099,-12089,-12074,-12067,-12058,-12039,-11867,-11861,-11847,-11831,-11798,-11781,-11604
                                        
-11589,-11536,-11358,-11340,-11339,-11324,-11303,-11097,-11077,-11067,-11055,-11052,-11045,-11041,-11038,-11024
                                        
-11020,-11019,-11018,-11014,-10838,-10832,-10815,-10800,-10790,-10780,-10764,-10587,-10544,-10533,-10519,-10331
                                        
-10329,-10328,-10322,-10315,-10309,-10307,-10296,-10281,-10274,-10270,-10262,-10260,-10256,-10254}
;
        
private static string[] pystr = new string[]{"a","ai","an","ang","ao","ba","bai","ban","bang","bao","bei","ben","beng","bi","bian","biao"
                                       
"bie","bin","bing","bo","bu","ca","cai","can","cang","cao","ce","ceng","cha","chai","chan","chang","chao","che","chen"
                                       
"cheng","chi","chong","chou","chu","chuai","chuan","chuang","chui","chun","chuo","ci","cong","cou","cu","cuan","cui"
                                       
"cun","cuo","da","dai","dan","dang","dao","de","deng","di","dian","diao","die","ding","diu","dong","dou","du","duan"
                                       
"dui","dun","duo","e","en","er","fa","fan","fang","fei","fen","feng","fo","fou","fu","ga","gai","gan","gang","gao"
                                       
"ge","gei","gen","geng","gong","gou","gu","gua","guai","guan","guang","gui","gun","guo","ha","hai","han","hang"
                                       
"hao","he","hei","hen","heng","hong","hou","hu","hua","huai","huan","huang","hui","hun","huo","ji","jia","jian"
                                       
"jiang","jiao","jie","jin","jing","jiong","jiu","ju","juan","jue","jun","ka","kai","kan","kang","kao","ke","ken"
                                       
"keng","kong","kou","ku","kua","kuai","kuan","kuang","kui","kun","kuo","la","lai","lan","lang","lao","le","lei"
                                       
"leng","li","lia","lian","liang","liao","lie","lin","ling","liu","long","lou","lu","lv","luan","lue","lun","luo"
                                       
"ma","mai","man","mang","mao","me","mei","men","meng","mi","mian","miao","mie","min","ming","miu","mo","mou","mu"
                                       
"na","nai","nan","nang","nao","ne","nei","nen","neng","ni","nian","niang","niao","nie","nin","ning","niu","nong"
                                       
"nu","nv","nuan","nue","nuo","o","ou","pa","pai","pan","pang","pao","pei","pen","peng","pi","pian","piao","pie"
                                       
"pin","ping","po","pu","qi","qia","qian","qiang","qiao","qie","qin","qing","qiong","qiu","qu","quan","que","qun"
                                       
"ran","rang","rao","re","ren","reng","ri","rong","rou","ru","ruan","rui","run","ruo","sa","sai","san","sang"
                                       
"sao","se","sen","seng","sha","shai","shan","shang","shao","she","shen","sheng","shi","shou","shu","shua"
                                       
"shuai","shuan","shuang","shui","shun","shuo","si","song","sou","su","suan","sui","sun","suo","ta","tai"
                                       
"tan","tang","tao","te","teng","ti","tian","tiao","tie","ting","tong","tou","tu","tuan","tui","tun","tuo"
                                       
"wa","wai","wan","wang","wei","wen","weng","wo","wu","xi","xia","xian","xiang","xiao","xie","xin","xing"
                                       
"xiong","xiu","xu","xuan","xue","xun","ya","yan","yang","yao","ye","yi","yin","ying","yo","yong","you"
                                       
"yu","yuan","yue","yun","za","zai","zan","zang","zao","ze","zei","zen","zeng","zha","zhai","zhan","zhang"
                                       
"zhao","zhe","zhen","zheng","zhi","zhong","zhou","zhu","zhua","zhuai","zhuan","zhuang","zhui","zhun","zhuo"
                                       
"zi","zong","zou","zu","zuan","zui","zun","zuo"}
;
        
/// <summary>
        
/// 转化汉字的全部拼音
        
/// </summary>
        
/// <param name="chrstr"></param>
        
/// <returns></returns>

        public static string Convert(string chrstr)
        
{
            Regex MyRegex 
= new Regex("^[一-龥]$");
            
byte[] array = new byte[2];
            
string returnstr = "";
            
int chrasc = 0;
            
int i1 = 0;
            
int i2 = 0;
            
char[] nowchar = chrstr.ToCharArray();
            
for (int j = 0; j < nowchar.Length; j++)
            
{
                
if (MyRegex.IsMatch(nowchar[j].ToString()))
                
{
                    array 
= System.Text.Encoding.Default.GetBytes(nowchar[j].ToString());
                    i1 
= (short)(array[0]);
                    i2 
= (short)(array[1]);
                    chrasc 
= i1 * 256 + i2 - 65536;
                    
if (chrasc > 0 && chrasc < 160)
                    
{
                        returnstr 
+= nowchar[j];
                    }

                    
else
                    
{
                        
for (int i = (pyvalue.Length - 1); i >= 0; i--)
                        
{
                            
if (pyvalue[i] <= chrasc)
                            
{
                                returnstr 
+= pystr[i];
                                
break;
                            }

                        }

                    }

                }

                
else
                
{
                    returnstr 
+= nowchar[j].ToString();
                }

            }

            
return returnstr;
        }

        
//ChineseToPY
        /// <summary>
        
/// 只转换汉字首字母
        
/// </summary>
        
/// <param name="strText"></param>
        
/// <returns></returns>

        public static string GetChineseSpell(string strText)
        
{
            
int len = strText.Length;
            
string myStr = "";
            
for (int i = 0; i < len; i++)
            
{
                myStr 
+= getSpell(strText.Substring(i, 1));
            }

            
return myStr;
        }



        
public static string getSpell(string cnChar)
        
{
            
byte[] arrCN = Encoding.Default.GetBytes(cnChar);
            
if (arrCN.Length > 1)
            
{
                
int area = (short)arrCN[0];
                
int pos = (short)arrCN[1];
                
int code = (area << 8+ pos;
                
int[] areacode = 4521745253457614631846826470104729747614481194811949062493244989650371506145062250906513875144652218526985269852698529805368954481 };
                
for (int i = 0; i < 26; i++)
                
{
                    
int max = 55290;
                    
if (i != 25) max = areacode[i + 1];
                    
if (areacode[i] <= code && code < max)
                    
{
                        
return Encoding.Default.GetString(new byte[] { (byte)(65 + i) });
                    }

                }

                
return "*";
            }

            
else return cnChar;
        }

    }

}

相关文章推荐

C#实现汉字转化为拼音

其实现的原理就是先将汉字转化成为内码,然后通过内码和拼音的对照来查找。 详细的代码如下所示: public class chs2py {private static int[] pyvalue=new...

c#汉字转化为拼音

/// ///汉字转拼音静态类 /// public class LanguageHelper { public LanguageHelper() { // ...

C#汉字转化拼音简写

汉字转化拼音简写 /// /// 汉字转拼音缩写 /// /// 要转换的汉字字符串 /// 拼音缩写 ...

C#汉字转化为拼音

using System; using System.Collections.Generic; using System.Text; using System.Text.RegularExpre...

C#实现汉字自动转拼音码

/// /// 生成拼音简码 /// /// Unicode编码字符串 /// 拼音...
  • z0582
  • z0582
  • 2012年08月08日 15:23
  • 1570

按汉字的拼音排序(c#实现)

原文:http://www.cnblogs.com/Joetao/articles/2022500.html 按汉字的拼音排序(c#实现) 如何按汉字的拼音排序建筑物名称呢...

C#实现汉字拼音简码

using System;using System.Collections.Generic;using System.Text;using Microsoft.VisualBasic;  namesp...

汉字转化拼音asp.net C#

  • 2010年11月01日 14:18
  • 3KB
  • 下载

Asp.net 实现汉字转化为拼音

Asp.net 实现汉字转化为拼音
  • y_h_t
  • y_h_t
  • 2011年01月14日 18:28
  • 537
内容举报
返回顶部
收藏助手
不良信息举报
您举报文章:用C#实现汉字转化为拼音
举报原因:
原因补充:

(最多只允许输入30个字)