Domain sampling

[derk]

Established a perl programme ^https?:\\/\\/([a-z0-9\\.\\-]+\\.[a-z0-9\\-]{2,12})/.*but it needs to be applied in the C-programme.

1. I can't figure out how to do this relay to get the domen out of the link. Used code

       re=pcre_compile(regexp,0,&error,&erroffset,NULL);
       if (!re) continue;
       if (pcre_exec(re,NULL,link,strlen(link),0,0,ovector,sizeof(ovector))>0)
       {
        printf("%s\n",ovector);
       }
       pcre_free(re);
   

2. I doubt the correctness of the regular. Maybe he didn't.

[Avante]

I used some kind of rfc2068. http://ru.wikipedia.org/wiki/URI

#define SPLIT_URI_REGEX \
  "^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\\?([^#]*))?(#(.*))?"

(not pcre) expression.

Example:

...
struct uri {
  char            *scheme,
    *net_loc,
    *path,
    *query,
    *fragment;
};
...
static regex_t reguri;

static pthread_mutex_t  reglock = PTHREAD_ERRORCHECK_MUTEX_INITIALIZER_NP;

static int     inireg = 0;
#define SPLIT_URI_REGEX 

"^(([^:/?#]+):)?(//([^/?#]))?([^?#])(\?([^#]))?(#(.))?"

#define ISCHEME   2

#define INETLOC   4

#define IPATH     5

#define IQUERY    7

#define IFRAGMENT 8
static char *

mkuri_comp (char *uri, regmatch_t *p)

{

if (p->rm_so < 0)

return NULL;
int  l = p->rm_eo - p->rm_so;

char buf[4096];

strncpy(buf, uri+p->rm_so, l);

buf[l] = 0;

return strdup(buf);

}
void

split_uri (char *uri, struct uri *suri)

{

if (!inireg) {

pthread_mutex_lock(&reglock);

if (!inireg) {

regcomp (&reguri, SPLIT_URI_REGEX, REG_EXTENDED);

inireg = 1;

}

pthread_mutex_unlock(&reglock);

}

regmatch_t match[reguri.re_nsub];

if (regexec (&reguri, uri, reguri.re_nsub, match, 0) == 0) { // OK

suri->scheme = mkuri_comp (uri,&match[ISCHEME]);

suri->net_loc = mkuri_comp (uri,&match[INETLOC]);

suri->path = mkuri_comp (uri,&match[IPATH]);

suri->query = mkuri_comp (uri,&match[IQUERY]);

suri->fragment = mkuri_comp (uri,&match[IFRAGMENT]);

}

}

Pre-emptive field cancellation struct uri challenge split_uri()♪