Question

我正在编译一个启用了utf8标志的PCRE模式，并尝试匹配utf8 char*字符串，但它不匹配，pcre_exec返回负数。我将主题长度为65传递给pcre_exec，这是字符串中的字符数。我相信它期望字节数，所以我尝试将参数增加到70但仍然得到相同的结果。我不知道还有什么让比赛失败。在我开枪之前请帮忙。

（但是如果我尝试没有标志PCRE_UTF8，它匹配，但偏移矢量[1]是30，这是我的输入字符串中的unicode字符之前的字符的索引）

#include "stdafx.h"
#include "pcre.h"
#include <pcre.h>               /* PCRE lib        NONE  */
#include <stdio.h>              /* I/O lib         C89   */
#include <stdlib.h>             /* Standard Lib    C89   */
#include <string.h>             /* Strings         C89   */
#include <iostream>

int main(int argc, char *argv[]) 
{
   pcre *reCompiled;

   int pcreExecRet;
   int subStrVec[30];
   const char *pcreErrorStr;
   int pcreErrorOffset; 
   char* aStrRegex = "(\\?\\w+\\?\\s*=)?\\s*(call|exec|execute)\\s+(?<spName>\\w+)(" 
                                     // params can be an empty pair of parenthesis or have parameters inside them as well.
                                     "\\(\\s*(?<params>[?\\w,]+)\\s*\\)"
                                     // paramList along with its parenthesis is optional below so a SP call can be just "exec sp_name" for a stored proc call without any parameters.
                                     ")?";
    reCompiled = pcre_compile(aStrRegex, 0, &pcreErrorStr, &pcreErrorOffset, NULL);
    if(reCompiled == NULL) {
      printf("ERROR: Could not compile '%s': %s\n", aStrRegex, pcreErrorStr);
      exit(1);
    } 

    char* line = "?rt?=call SqlTxFunctionTesting(?înFîéld?,?outField?,?inOutField?)";
    pcreExecRet = pcre_exec(reCompiled,
                            NULL,
                            line, 
                            65,  // length of string
                            0,                      // Start looking at this point
                            0,                      // OPTIONS
                            subStrVec,
                            30);                    // Length of subStrVec

   printf("\nret=%d",pcreExecRet);

   //int substrLen = pcre_get_substring(line, subStrVec, pcreExecRet, 1, &mantissa);

}

Answer 1

1）

char * q= "î";
printf("%d, %s", q[0], q);

输出：
63，？

2）您必须使用PCRE_BUILD_PCRE16（或32）和PCRE_SUPPORT_UTF重建PCRE。并使用pcre16.lib和/或pcre16.dll。然后你可以试试这段代码：

  pcre16 *reCompiled;
  int pcreExecRet;
  int subStrVec[30];
  const char *pcreErrorStr;
  int pcreErrorOffset;  
  wchar_t* aStrRegex = L"(\\?\\w+\\?\\s*=)?\\s*(call|exec|execute)\\s+(?<spName>\\w+)(" 
                                     // params can be an empty pair of paranthesis or have parameters inside them as well.
                                     L"\\(\\s*(?<params>[?,\\w\\p{L}]+)\\s*\\)"
                                     // paramList along with its paranthesis is optional below so a SP call can be just "exec sp_name" for a stored proc call without any parameters.
                                     L")?";
   reCompiled = pcre16_compile((PCRE_SPTR16)aStrRegex, PCRE_UTF8, &pcreErrorStr, &pcreErrorOffset, NULL);
   if(reCompiled == NULL) {
    printf("ERROR: Could not compile '%s': %s\n", aStrRegex, pcreErrorStr);
    exit(1);
   } 

  const wchar_t* line = L"?rt?=call SqlTxFunctionTesting(  ?inField?,?outField?,?inOutField?,?fd?  )";
  const wchar_t* mantissa=new wchar_t[wcslen(line)];
  pcreExecRet = pcre16_exec(reCompiled,
                            NULL,
                            (PCRE_SPTR16)line, 
                            wcslen(line),  // length of string
                            0,                      // Start looking at this point
                            0,                      // OPTIONS
                            subStrVec,
                            30);                    // Length of subStrVec

 printf("\nret=%d",pcreExecRet);
 for (int i=0;i<pcreExecRet;i++){
     int substrLen = pcre16_get_substring((PCRE_SPTR16)line, subStrVec, pcreExecRet, i, (PCRE_SPTR16 *)&mantissa);
     wprintf(L"\nret string=%s, length=%i\n",mantissa,substrLen);
 }

3）\ w = [0-9A-Z_a-z]。它不包含unicode符号 4）这确实有帮助：http://answers.oreilly.com/topic/215-how-to-use-unicode-code-points-properties-blocks-and-scripts-in-regular-expressions/
5）来自PCRE 8.33源（pcre_exec.c：2251）

/* Find out if the previous and current characters are "word" characters.
It takes a bit more work in UTF-8 mode. Characters > 255 are assumed to
be "non-word" characters. Remember the earliest consulted character for
partial matching. */

PCRE与utf8字符不匹配

1 个答案: