【字符串问题】求一个字符串中重复出现的最长的子串

2013-09-14 15:34:16

用后缀数组求一个字符串中重复出现的最长的子串。

1. 用C++中的string类可以很方便地进行操作，需将后缀数组保存在vector<string>，如下面代码中的string版本所示，但这样就会因为<string>有很大的开销；
2. 直接用字符指针指向后缀字符串的首地址，可以节省很大的空间，如下面代码中的char *版本所示.
3. 注意使用char *版本时，用qsort函数最后缀字符串数组排序，需要提供comp函数，该函数的写法如下：

int pStrcmp(const void *p,const void *q)
{
    return ( strcmp( *(char **)p , *(char **)q ) );
}

更多关于该函数的说明，详见博文http://www.cnblogs.com/youngforever/articles/3321469.html。

代码（string版本）：

#include <iostream>
#include <cassert>
#include <string>        //用string类模板，必须包含该头文件
#include <vector>        //用vector模板，必须包含该头文件
#include <algorithm>   //用sort函数，必须包含该头文件
using namespace std;

//获取两个字符串的最长公共子串
size_t GetLCS(const string &str1,const string &str2)
{
    size_t len1 = str1.length();
    size_t len2 = str2.length();

    size_t end  = len1 < len2 ? len1 : len2;

    size_t index = 0;
    size_t commenLen = 0;

    for (index = 0;index < end;++index)
    {
        if (str1.at(index) == str2.at(index))
        {
            ++commenLen;
        }
        else
        {
            break;
        }
    }

    return commenLen;
}

//获取字符串中重复出现且最长的子串
size_t FindLongeStringApearTwice(const string &srcStr,string &subStr)
{
    vector<string> vecStr;

    size_t index;
    size_t end = srcStr.length();

    string tmpStr;
    size_t tmpLen = 0;
    size_t maxLen = 0;

    for ( index = 0;index < end;++index)   //生成后缀数组
    {
        tmpStr = srcStr.substr(index,(end - 1 - index));
        vecStr.push_back(tmpStr);
        //tmpStr.clear();  //此处的清楚是不必要的，可以删去
    }

    sort(vecStr.begin(),vecStr.end());  //对后缀字符串数组排序

    vector<string>::iterator iter;

    for (iter = vecStr.begin();(iter + 1) != vecStr.end();++iter)  //求相邻string的LCS
    {
        tmpLen = GetLCS(*iter,*(iter + 1));

        if(tmpLen > maxLen)
        {
            maxLen = tmpLen;
            subStr = (*iter).substr(0,maxLen);
        }
    }

    return maxLen;
}




//测试FindLongeStringApearTwice
void TestDriver()
{
    string strArray[] = {"0123456","yyabcdabjcabceg","abcbcbcabc","hello,li mei! hello,li lei!"};
    size_t arrayLength = 4;

    string srcStr;
    string subStr;
    size_t maxLen = 0;

    for (size_t index = 0;index < arrayLength;++index)
    {
        //srcStr.clear();  //此处的清楚是不必要的，可以删去
        srcStr = strArray[index];
        maxLen = FindLongeStringApearTwice(srcStr,subStr);

        cout<<"the source string is : "<<srcStr<<endl;
        cout<<"the longest sub string is : "<<subStr<<endl;
        cout<<"the max length is : "<<maxLen<<endl;
        cout<<endl;
    }
}

int main()
{
    TestDriver();
    return 0;
}

代码（char *版本）：

#include <iostream>
#include <cassert>
#include <string>
#include <vector>
#include <algorithm>
using namespace std;

size_t GetLCS(const char *str1,const char *str2)
{
    assert(str1 != NULL && str2 != NULL);

    size_t len1 = strlen(str1);
    size_t len2 = strlen(str2);

    size_t end  = len1 < len2 ? len1 : len2;

    size_t index = 0;
    size_t commenLen = 0;

    for (index = 0;index < end;++index)
    {
        if ( *(str1 + index) == *(str2 + index) )
        {
            ++commenLen;
        }
        else
        {
            break;
        }
    }

    return commenLen;
}

typedef char *  pCHAR;

int pStrcmp(const void *p,const void *q)
{
    return ( strcmp( *(char **)p , *(char **)q ) );
}

void DisplayString(const char *pStr)
{
    size_t index = 0;

    while (*(pStr + index))
    {
        cout<<*(pStr + index);
        ++index;
    }
    cout<<endl;
}

size_t FindLongestStringApearTwice(const char *pSrcStr,char *&pSubStr)
{
    assert(pSrcStr != NULL && pSubStr != NULL);

    const size_t lenOfSrcStr  = strlen(pSrcStr);
    pCHAR *pSuffixArray = new pCHAR[lenOfSrcStr];

    size_t index;
    size_t end = strlen(pSrcStr);

    size_t tmpLen = 0;
    size_t maxLen = 0;

    for ( index = 0;index < end;++index)
    {
        pSuffixArray[index] = (char *)pSrcStr + index;
        //DisplayString(pSuffixArray[index]);
    }

    qsort(pSuffixArray,lenOfSrcStr,sizeof(pCHAR),pStrcmp);
/*
    for ( index = 0;index < end;++index )
    {
        DisplayString(pSuffixArray[index]);
    }*/

    for (index = 0;index + 1 < end;++index) 
    {
        tmpLen = GetLCS(pSuffixArray[index],pSuffixArray[index + 1]);

        if(tmpLen > maxLen)
        {
            maxLen = tmpLen;
            pSubStr = pSuffixArray[index];
        }
    }

    return maxLen;
}


void TestDriver()
{
    pCHAR strArray[] = {"0123456","yyabcdabjcabceg","abcbcbcabc","hello,li mei! hello,li lei!"};
    size_t arrayLength = 4;

    pCHAR srcStr;
    pCHAR subStr;
    size_t maxLen = 0;

    for (size_t index = 0;index < arrayLength;++index)
    {
        srcStr = strArray[index];
        maxLen = FindLongestStringApearTwice(srcStr,subStr);

        cout<<"the source string is : "<<srcStr<<endl;
        cout<<"the longest sub string is : ";

        for (size_t i = 0;i < maxLen;++i)
        {
            cout<<*(subStr + i);
        }
        cout<<endl;

        cout<<"the max length is : "<<maxLen<<endl;
        cout<<endl;
    }
}


int main()
{
    TestDriver();
    return 0;
}

测试结果：

the source string is : 0123456
the longest sub string is :
the max length is : 0

the source string is : yyabcdabjcabceg
the longest sub string is : abc
the max length is : 3

the source string is : abcbcbcabc
the longest sub string is : bcbc
the max length is : 4

the source string is : hello,li mei! hello,li lei!
the longest sub string is : hello,li
the max length is : 9

请按任意键继续. . .