基于STL的字符串最大长度匹配算法

MMString 的源码如下，使用MMString ，可以轻松的实现手机号段查询，短信业务中的业务代码模糊匹配。

由于MMString 是string的子类，使用起来和string完全一样，非常方便。同时通过模板参数来控制匹配过程是是否对大小写敏感。

////////////////////////////////////////////////////////////////////////////////////////
//                                                                                    //
// mmstring.h   - 最长匹配string                              //
//                                                                                    //
// Larrin Workshop                                                    //
//                                                                                    //
// email: larrin2002@msn.com                              //
//                                                                                    //
// HomePage : http://www.larrin.net                       //
//                                                                                    //
////////////////////////////////////////////////////////////////////////////////////////

#ifndef LARRINWORKSHOP_MAX_LENGTH_MATCH_STRING__H
#define LARRINWORKSHOP_MAX_LENGTH_MATCH_STRING__H

#include <string>
using std::string ;

#ifndef WIN32
#define strnicmp strnicasecmp
#endif

template <bool ignorecase=false>
class MMString : public string
{
 bool m_isKey ;    
 string m_nextlongger_str; 
 string m_shortest_str ;  
 bool m_ignorecase;

public:

 MMString() : string() , m_isKey(true) , m_ignorecase(ignorecase)
 {
 }

 MMString(char * str) : string(str) , m_isKey(true) , m_ignorecase(ignorecase)
 {
 }

 MMString(bool key,char * str) : string(str) , m_isKey(key) , m_ignorecase(ignorecase)
 {
 }

 MMString(char * str, int len):string(str,len) , m_isKey(true) , m_ignorecase(ignorecase)
 {
 }

 MMString(size_type Size, value_type Character):string(Size,Character) , m_isKey(true) , m_ignorecase(ignorecase)
 {
 }

 MMString(const MMString &dstr)
 {
  this->m_isKey = dstr.m_isKey ;
  this->m_nextlongger_str = dstr.m_nextlongger_str;
  this->m_shortest_str = dstr.m_shortest_str;
  this->m_ignorecase = dstr.m_ignorecase;
  assign(dstr.c_str());
 }

 virtual ~MMString(){}

 MMString operator=(const MMString &val)
 {
  this->m_isKey = val.m_isKey ;
  this->m_nextlongger_str = val.m_nextlongger_str;
  this->m_shortest_str = val.m_shortest_str;
  this->m_ignorecase = val.m_ignorecase;
  assign(val.c_str());
  return *this ;
 }

 bool isKey()
 {
  return this->m_isKey ;
 }

 bool operator < (const MMString &s) const
 {
  int len1 = this->length();
  int len2 = s.length();

  int ret;
  if(m_ignorecase)
   ret = strnicmp(this->c_str(),s.c_str(),len1 < len2 ? len1 : len2);
  else
   ret = strncmp(this->c_str(),s.c_str(),len1 < len2 ? len1 : len2);

  if(this->m_isKey && s.m_isKey)
  {
   if(len1 > len2)  
   {
    if(ret == 0)
    {
     if(m_shortest_str.empty() || m_shortest_str.length() > len2)
     {
      string * str = const_cast<string *>(&(m_shortest_str));
      *str = s.c_str();
     }

     if(s.m_nextlongger_str.length() < len1)
     {
      string * str = const_cast<string *>(&(s.m_nextlongger_str));
      *str = c_str();
     }
    }
    return ret <= 0 ;
   }
   else
   {
    if(ret == 0)
    {
     if(m_nextlongger_str.length() < len2)
     {
      string * str = const_cast<string *>(&(this->m_nextlongger_str));
      *str = s.c_str();
     }

     if(s.m_shortest_str.empty() || s.m_shortest_str.length() > len1)
     {
      string * str = const_cast<string *>(&(s.m_shortest_str));
      *str = c_str();
     }
    }
    return ret < 0 ;
   }
  }
  else
  {
   if(this->m_isKey) 
   {
    if(len1 > len2)
     return ret <= 0 ;
    else
    {
     if(ret == 0)
      return false ; 
     else
     {
      if(ret > 0)
      {
       int len3 = this->m_shortest_str.length();
       if(len3 > 0)
       {
        if(!this->m_ignorecase)
        {
         if(strncmp(this->c_str(),s.c_str(),len3) == 0)
          return true ;
        }
        else
        {
         if(strnicmp(this->c_str(),s.c_str(),len3) == 0)
          return true ;
        }
       }
      }

      return ret < 0 ;
     }
    }
   }
   else
   {
    if(len1 < len2)                                             
     return ret < 0 ;
    else
    {
     if(ret == 0)
     {
      if(s.m_nextlongger_str.length() > 0)
      {
       int len3 = s.m_nextlongger_str.length() ;
       if(len3 > len1)
        return false ;

       if(!this->m_ignorecase)
       {
        if(strncmp(s.m_nextlongger_str.c_str(),this->c_str(),len3) == 0)
         return true ;
       }
       else
       {
        if(strnicmp(s.m_nextlongger_str.c_str(),this->c_str(),len3) == 0)
         return true ;
       }
       

       return false ;
      }

      return false ;
     }
     else
      return ret < 0 ;
    }
   }
  }
 }
};

typedef MMString<false> mmstring ;
typedef MMString<true> immstring;

#endif //LARRINWORKSHOP_MAX_LENGTH_MATCH_STRING__H

下面是测试代码:

#include "mmstring.h"
#include <iostream>
#include <set>

using namespace std;

int main()
{
 typedef set<mmstring> mmstring_set;
 mmstring_set temp_set;

 temp_set.insert("130");
 temp_set.insert("13");
 temp_set.insert("13034");
 temp_set.insert("13033");
 temp_set.insert("1303401");
 temp_set.insert("130340");
 temp_set.insert("1301");
 temp_set.insert("1302");
 temp_set.insert("13021");
 temp_set.insert("TQ");
 temp_set.insert("TQB");
 temp_set.insert("TQBA");
 temp_set.insert("T1QB");
 temp_set.insert("T2QB");
 temp_set.insert("T2Q");

 int s = temp_set.size();

 mmstring_set::iterator it ;
 for(it = temp_set.begin() ; it != temp_set.end() ; it++)
 {
  cout << (*it) << endl ;
 }

 char match_srt[][10]={ "T2QA123",
      "T1QA123",
      "T1QB123",
      "TQA19",
      "TQB19",
      "TQB1",
      "T",
      "TQC19",
      "TQ",
      "13034014",
      "130213",
      "1",
      "13",
      "1303",
      "13034axa",
     "TQ755",
     "tq755",
      "END"};

 cout << "=======================" << endl ;

 for(int i = 0 ; ; i++)
 {
  if(strcmp(match_srt[i],"END")==0) break ;
  it = temp_set.find(mmstring(false,match_srt[i]));
  if(it != temp_set.end())
  {
   cout << match_srt[i] << " match " << *it << endl ;
  }
  else
  {
   cout << "no match for " << match_srt[i] << endl;
  }
 }

 return 0 ;
}

运行结果如下:

1301
13021
1302
13033
1303401
130340
13034
130
13
T1QB
T2QB
T2Q
TQBA
TQB
TQ
=======================
T2QA123 match T2Q
no match for T1QA123
T1QB123 match T1QB
TQA19 match TQ
TQB19 match TQB
TQB1 match TQB
no match for T
TQC19 match TQ
TQ match TQ
13034014 match 1303401
130213 match 13021
no match for 1
13 match 13
1303 match 130
13034axa match 13034
TQ755 match TQ
tq755 match TQ