PCRE、PCRE2 以及PCRE++ 使用教程

1.准备工作 ：请自行编译

PCRE、PCRE2 以及PCRE++ 库的源码，PCER2 参考我的博客。  

2.VS2008 建立一个win32 程序 ，工程名PcreDemo。



3.将所需的DLL 和 lib 拷贝进工程相应目录，并进行相关设置



4.  pcre 简单使用

 pcre  *re; 

 const char *error; 

 int  erroffset; 

 int  ovector[OVECCOUNT]; 

 int  rc;

 string strOrg = "I am hk!";

 string strRex = "\\b\\w+\\b";

 printf("前言：以下例子只是非常简单的应用，仅供参考！\n   pcreD.dll   pcre2.dll  pcre++.dll  pcreD.lib   pcre2.lib  pcre++.lib 请按照文档网址自行下载并编译！\n");

 printf("\n\n\n");

 printf(" PCRE 1 库   应用实例:\n\n");

 printf("字符串是 : %s\n", strOrg.c_str());

 printf("正则表达式是: %s\n\n\n", strRex.c_str());

 char *p = (char *) strOrg.c_str();

 re = pcre_compile(strRex.c_str(),       // pattern, 输入参数，将要被编译的字符串形式的正则表达式,

  PCRE_CASELESS|PCRE_GLOBAL,// options, 输入参数，用来指定编译时的一些选项

  &error,       // errptr, 输出参数，用来输出错误信息

  &erroffset,   // erroffset, 输出参数，pattern中出错位置的偏移量

  NULL);        // tableptr, 输入参数，用来指定字符表，一般情况用NULL

 // 返回值：被编译好的正则表达式的pcre内部表示结构

 if (re == NULL) {                 //如果编译失败，返回错误信息

  printf("PCRE compilation failed at offset %d: %s\n", erroffset, error);

  return false;

 }

 rc = pcre_exec(re, NULL, p, strlen(p), 0, 0, ovector, OVECCOUNT);

 if (rc < 0) {                     //如果没有匹配，返回错误信息

  if (rc == PCRE_ERROR_NOMATCH)

   printf("Sorry, no match ...\n");

  else

   printf("Matching error %d\n", rc);

  pcre_free(re);

  return false;

 }

 while ( ( rc = pcre_exec(re, NULL, p, strlen(p), 0, 0, ovector, OVECCOUNT)) != PCRE_ERROR_NOMATCH )

 {

  char *substring_start = p + ovector[0];

  int substring_length = ovector[1] - ovector[0];

  char matched[1024];

  memset( matched, 0, 1024 );

  strncpy( matched, substring_start, substring_length );

  //strDes = matched;

  printf( "全部的匹配结果有:%s\n", matched );

  p += ovector[1]; 

  if (*p == '\0')

  {

   break;

  }

 }

 pcre_free(re);                     // 编译正则表达式re 释放内存

5 .pcre2 简单使用

 printf(" PCRE 2 库  应用实例:\n\n");

 int error_code = 0;

 PCRE2_SIZE error_offset = 0;

 int nOptions =  0; 

// const char *pError; 

 string strOrg2 = "I am hk!";

 string strRex2 = "\\b\\w+\\b";

 printf("字符串是 : %s\n", strOrg2.c_str());

 printf("正则表达式是: %s\n\n\n", strRex2.c_str());

 //going to compile the regular expression pattern, and handle  any errors that are detected

  pcre2_code *recm = pcre2_compile(reinterpret_cast<PCRE2_SPTR>(strRex2.c_str()), PCRE2_ZERO_TERMINATED, 0 , &error_code, &error_offset, NULL);

  //Compilation failed: print the error message and exit

  if (re == NULL)

  {

   PCRE2_UCHAR buffer[256];

   pcre2_get_error_message(error_code, buffer, sizeof(buffer));

   printf("PCRE2 compilation failed at offset %d: %s\n", (int)error_offset,buffer);

   return 1;

  }

 // Using this function ensures that the block is exactly the right size for the number of capturing parentheses in the pattern.

  pcre2_match_data *match_data = pcre2_match_data_create_from_pattern(recm, NULL);

  int rc2 = 0;

  int start_offset = 0;

  unsigned int match_index = 0;

  rc2= pcre2_match(recm, reinterpret_cast<PCRE2_SPTR>(strOrg2.c_str()), strOrg2.length(), start_offset, 0, match_data, NULL);

  //if(rc < 0)

  //{

  // 

  //}

  //else

  //{

  // int pos = start_offset;

 

  // while ((rc = pcre2_match(re, reinterpret_cast<PCRE2_SPTR>(strStruff.c_str()), strStruff.length(), start_offset, 0, match_data, NULL)) > 0)

  // {

  //  pos = start_offset;

  // }

  //}

  if (rc2 < 0)

    {

    switch(rc2)

   {

   case PCRE2_ERROR_NOMATCH: printf("No match\n"); break;

   /*

   Handle other special cases if you like

   */

   default: printf("Matching error %d\n", rc2); break;

   }

    pcre2_match_data_free(match_data);   /* Release memory used for the match */

    pcre2_code_free(recm);                 /* data and the compiled pattern. */

    return 1;

    }

  //going to compile the regular expression pattern, and handle  any errors that are detected

  while ((rc2 = pcre2_match(recm, reinterpret_cast<PCRE2_SPTR>(strOrg2.c_str()), strOrg2.length(), start_offset, 0, match_data, NULL)) > 0)

  {

    //Match succeded. Get a pointer to the output vector, where string offsets are stored.

    PCRE2_SIZE *ovector = pcre2_get_ovector_pointer(match_data);

    int i = 0;

    for (i = 0; i < rc2; i++)

    {

     char *substring_start = (char *)strOrg2.c_str() + ovector[2*i];

     int substring_length = ovector[2*i+1] - ovector[2*i];

     char matched[1024];

     memset( matched, 0, 1024 );

     strncpy( matched, substring_start, substring_length );

     printf( "全部匹配结果有  :%s\n",matched );

    }

   start_offset = ovector[2*(i-1) + 1];

   if ( strOrg2[start_offset] == '\0')

   {

    break;

   }

  }

  pcre2_match_data_free(match_data);

  pcre2_code_free(recm);

6.pcre++简单使用

  printf(" PCRE++库 应用实例：\n\n");

 /*

     * define a string with a regular expression

     */

  printf("官方例子如下：\n");

    string expression = "(\\b\\w+\\b)";

    /*

     * this is the string in which we want to search

     */

    string stuff = "I am hk!";

    cout << "  searching in \"" << stuff << "\" for regex \""

  << expression << "\":" << endl;

    /*

     * Create a new Pcre object, search case-insensitive ("i")

     */

 pcrepp::Pcre reg(expression, "i");

   

    /*

     * see if the expression matched

     */

    if(reg.search(stuff) == true) {

      /*

       * see if the expression generated any substrings

       */

      if(reg.matches() >= 1) {

 /*

  * print out the number of substrings

  */

 cout << "  generated " << reg.matches() << " substrings:" << endl;

  

 /*

  * iterate over the matched sub strings

  */

 for(int pos=0; pos < reg.matches(); pos++) {

   /* print out each substring */

   cout << "  substring " << pos << ": " << reg[pos];   // also possible: reg.get_match(pos);

   /* print out the start/end offset of the current substring

    * within the searched string(stuff)

    */

   cout << " (start: " << reg.get_match_start(pos) << ", end: "

        << reg.get_match_end(pos) << ")" << endl;

 }

      }

      else {

 /*

  * we had a match, but it generated no substrings, for whatever reason

  */

 cout << "   it matched, but there where no substrings." << endl;

      }

    }

    else {

      /*

       * no match at all

       */

      cout << "   didn't match." << endl;

    }

 printf("我自己做相关封装如下：\n");

 pcrepp::Pcre regg(expression);

 int npos = 0;

 if ( !regg.search(stuff, npos))

 {

  return false;

 }

 while(regg.search(stuff, npos))

 {

  if(regg.matches() >= 1)

  {

   string strMatched = regg.get_match(0);

   printf("全部匹配结果有：%s\n",strMatched.c_str());

   npos = regg.get_match_end(0) + 1;

  }

 }

7. OK!完成啦，有什么问题大家可以给我留言哦！