Programming Wiki: [[Regex Regular Expression]] Examples

Programming Wiki : RegexExamples

WikiHomePage :: Categories :: PageIndex :: RecentChanges :: RecentlyCommented :: Login/Register

Regular Expression Examples


Here are two simple grep implementations in C, one for the POSIX API, and one using PCRE.

POSIX <regex.h> C program -- regrep.c
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <sys/types.h>
#include <regex.h>

char enter_reverse_mode[] = "\33[7m";
char exit_reverse_mode[] = "\33[0m";

int main(int argc, char **argv)
{
  const char *pattern;
  int ec;
  regex_t expr;
  regmatch_t rm;
  char buf[12];
  size_t offset, length;
  int flags;
  assert(argc == 2);
  pattern = argv[1];
  if ((ec = regcomp(&expr, pattern, 0)) != 0) {
    char str[256];
    regerror(ec, &expr, str, sizeof str);
    fprintf(stderr, "%s: %s\n", pattern, str);
    return EXIT_FAILURE;
  }
  flags = 0;
  while (fgets(buf, sizeof buf, stdin)) {
    /* Find the end of the buffer */
    length = strcspn(buf, "\n");
    /* Check for beginning and end of line. */
    if (flags & REG_NOTEOL) {
            /* If the last line read was a partial line, then we are
             * not at the beginning of the line. */

      flags |= REG_NOTBOL;
      if (buf[length] == '\n')
        flags &= ~REG_NOTEOL;
    }
    else if (buf[length] != '\n') {
      /* We've read a partial line. */
      flags = REG_NOTEOL;
    }
    else {
      /* We have a complete line. */
      flags = 0;
    }
    /* get rid of any newline character */
    buf[length] = '\0';
    /* start at beginning of the buffer */
    offset = 0;
    while (regexec(&expr, buf + offset, 1, &rm, flags) == 0) {
      assert(rm.rm_so >= 0);
      /* we're not smart enough to support empty matches. */
      assert(rm.rm_eo > rm.rm_so);
      /* print the portion which precedes the match, then the match */
      printf("%.*s%s%.*s%s",
        rm.rm_so, buf + offset,
        enter_reverse_mode,
        rm.rm_eo - rm.rm_so, buf + offset + rm.rm_so,
        exit_reverse_mode);
      /* start next match at the end of this one. */
      offset += rm.rm_eo;
      /* we're no longer at the beginning of the line */
      flags |= REG_NOTBOL;
    }
    /* print remainder of the line */
    printf("%s", buf + offset);
    /* print a newline if we're at the end of a line */
    if (!(flags & REG_NOTEOL))
      putchar('\n');
  }
  return EXIT_SUCCESS;
}


PCRE <pcre.h> C program -- pcregrep.c
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <pcre.h>

char enter_reverse_mode[] = "\33[7m";
char exit_reverse_mode[] = "\33[0m";

int main(int argc, char **argv)
{
  const char *pattern;
  const char *errstr;
  int erroffset;
  pcre *expr;
  char line[512];
  assert(argc == 2); /* XXX fixme */
  pattern = argv[1];
  if (!(expr = pcre_compile(pattern, 0, &errstr, &erroffset, 0))) {
    fprintf(stderr, "%s: %s\n", pattern, errstr);
    return EXIT_FAILURE;
  }
  while (fgets(line, sizeof line, stdin)) {
    size_t len = strcspn(line, "\n");
    int matches[2];
    int offset = 0;
    int flags = 0;
    line[len] = '\0';
    while (0 < pcre_exec(expr, 0, line, len, offset, flags, matches, 2)) {
      printf("%.*s%s%.*s%s",
        matches[0] - offset, line + offset,
        enter_reverse_mode,
        matches[1] - matches[0], line + matches[0],
        exit_reverse_mode);
      offset = matches[1];
      flags |= PCRE_NOTBOL;
    }
    printf("%s\n", line + offset);
  }
  return EXIT_SUCCESS;
}
 Comments [Hide comments/form]
These look interesting. I will have to mess around with them. What is each suppose to do?
-- ScottRippee (2005-12-02 14:15:43)
Powered by Wikka Wakka Wiki 1.1.6.0
Page was generated in 0.0223 seconds