C语言的正则表达式

.匹配正则表达式

#include <sys/types.h>
#include <regex.h>

typedef struct{
     int              reg_magic;
     size_t           re_nsub;
     const   char     *re_endp'
     struct  re_guts  *re_g;
} regex_t;

int regcomp(regex_t *preg, const char *pattern, int cflags)

pattern    正则表达式字符串
preg       regex_t类型的结构变量, 调用放置编译结果的地方
cflags     下面一外或多个进行"或"操作的位掩码
           REG_EXTENDED   使用POSIX延伸表示语法，否则使用POSIX基本语法
           REG_NOSPEC     禁用所有的元字符，即不认为模式字符具有特殊意义
           REG_ICASE      忽略大小写
           REG_NOSUB      忽略参数nmatch和pmatch
           REG_NEWLINE    不对特殊字符进行比较如^ . * ? + | \ [ ] ( ) < >和$
           REG_PEND       编译模式，在遇到第一个空字节时，正则表达式不会结束，
                          正则表达式将在preg->re_endp指向的字节之前结束。
                          这就允许正则表达式中引入空字节

.匹配正则表达式

#include <sys/types.h>
#include <regex.h>

/*
* 成功返回0
* 失败返回错误代码
*/
int regexec(const regex_t *preg,    //指向以前编译过的而且由regcomp(3)函数初始化的正则表达式
            const char *string,     //要匹配的字符串
            size_t nmatch,          //和pmatch把参数用于把匹配的模式返回给调用程序
            regmatch_t pmatch[],
            int eflags); //包含由0个或以上标志的"或"操作 (REG_NOTBO, REG_NOTEOL, REG_STARTEND)

typedef struct{
     regoff_t  rm_so;    //符合条件的开始位置,如果值为-1，则代表此结构并未让regexec()使用
     regoff_t  rm_eo;    //符合条件的结束位置  end of match offset
} regmatch_t;

参数eflags有两种可能，可使用OR(|)组合:
      REG_NOTBOL      让特殊字符^无作用
      REG_NOTEOL      让特殊字符＄无作用

.正则表达式错误, 用来取得regcomp()或regexec()的错误原因

/**
* errorcode   为由regcomp()或regexec()返回的错误代码
* preg        指向pattern buffer的结构指针
* errbuf      指向欲存放错误字符串的缓冲区， errbuf_size为缓冲区大小
*
* 返回错误字符串的长度
*/
#include <sys/types.h>
#include <regex.h>
size_t regerror(int errorcode, const regex_t *preg,
                char *errbuf, size_t errbuf_size);

.释放正则表达式

#include <sys/types.h>
#include <regex.h>

void regfree(regex_t  *preg);
释放regcomp已经编译的正则表达式

.API示例

int z;          //Error code
regex_t reg;    //Compiled regexpr
char ebuf[128]; //Error message buffer

z = regcomp(&reg, pattern, REG_EXTENDED);

if(z != 0 ){
    regerror(z, &reg, buf, sizeof(ebuf));
    printf("%s: regcomp(3)\n", ebuf);
    exit(1);
}
//.....
regfree(&reg);

示例代码：

#include <stdio.h>
#include <sys/types.h>
#include <regex.h>
#include <string.h>
#include <unistd.h>
#include <stdlib.h>

int cns_reg(const char *str, const char *pattern)
{
    int          z;            //status
    int          cflags = 0;   //compile flags
    regex_t      reg;          //compiled regular expression
    char         ebuf[128];    //error buffer

    regmatch_t   pm[10];       //pattern matches 0-9
    const size_t nmatch = 10;  //The size of array pm[]

    /** //编译正则表达式
     * @param const char*  pattern         将要被编译的正则表达式
     * @param regex_t*     reg             用来保存编译结果
     * @param int          cflags          决定正则表达式将如何被处理的细节
     *
     * @return  success    int        0    并把编译结果填充到reg结构中
     *          fail       int        非0
     */
    z = regcomp(&reg, pattern, cflags);

    if(z != 0){
        regerror(z, &reg, ebuf, sizeof(ebuf));
        fprintf(stderr, "%s: pattern '%s'\n", ebuf, pattern);
        return 1;
    }

    //report the number of subexpressions
    if(!(cflags & REG_NOSUB))
        printf("There were %d subexpression.\n", reg.re_nsub);

    /**
     * reg     指向编译后的正则表达式
     * str     指向将要进行匹配的字符串
     * pm      str字符串中可能有多处和正则表达式相匹配， pm数组用来保存这些位置
     * nmacth  指定pm数组最多可以存放的匹配位置数
     *
     * @return 函数匹配成功后，str+pm[0].rm_so到str+pm[0].rm_eo是第一个匹配的子串
     *                            str+pm[1].rm_so到str+pm[1].rm_eo是第二个匹配的子串
     *                            ....
     */
    z = regexec(&reg, str, nmatch, pm, 0);

    if(z == REG_NOMATCH)
        return 1;
    else if(z != 0){
        regerror(z, &reg, ebuf, sizeof(ebuf));
        fprintf(stderr, "%s: regcomp('%s')\n", ebuf, str);
        return 2;
    }

    regfree(&reg);
    return 0;
}

int main(int argc, char **argv)
{
    printf("%d\n",  cns_reg(argv[1], argv[2]));
    return 0;
}

.运行

[root@localhost sockets]# ./reg "I love you" "^I.*"
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "I love you" "I.*"
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "I love you" "dI.*"
There were 0 subexpression.
1
[root@localhost sockets]# ./reg "I love you" "//"
There were 0 subexpression.
1
[root@localhost sockets]# ./reg "" "//"
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "" "http://"
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "" "fttp://"
There were 0 subexpression.
1
[root@localhost sockets]# ./reg "" ""
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "" "."
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "" ".."
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "" "\.\."
There were 0 subexpression.
1
[root@localhost sockets]# ./reg "" ".."
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "" "..*"
There were 0 subexpression.
0
[root@localhost sockets]# ./reg "" "[a-z]3"
There were 0 subexpression.
1
[root@localhost sockets]# ./reg "" "[a-z]3.*"