How do you capture a group with regex?

Here’s a code example that demonstrates capturing multiple groups.

You can see that group ‘0’ is the whole match, and subsequent groups are the parts within parentheses.

Note that this will only capture the first match in the source string. Here’s a version that captures multiple groups in multiple matches.

#include <stdio.h>
#include <string.h>
#include <regex.h>

int main ()
{
  char * source = "___ abc123def ___ ghi456 ___";
  char * regexString = "[a-z]*([0-9]+)([a-z]*)";
  size_t maxGroups = 3;

  regex_t regexCompiled;
  regmatch_t groupArray[maxGroups];

  if (regcomp(&regexCompiled, regexString, REG_EXTENDED))
    {
      printf("Could not compile regular expression.\n");
      return 1;
    };

  if (regexec(&regexCompiled, source, maxGroups, groupArray, 0) == 0)
    {
      unsigned int g = 0;
      for (g = 0; g < maxGroups; g++)
        {
          if (groupArray[g].rm_so == (size_t)-1)
            break;  // No more groups

          char sourceCopy[strlen(source) + 1];
          strcpy(sourceCopy, source);
          sourceCopy[groupArray[g].rm_eo] = 0;
          printf("Group %u: [%2u-%2u]: %s\n",
                 g, groupArray[g].rm_so, groupArray[g].rm_eo,
                 sourceCopy + groupArray[g].rm_so);
        }
    }

  regfree(&regexCompiled);

  return 0;
}

Output:

Group 0: [ 4-13]: abc123def
Group 1: [ 7-10]: 123
Group 2: [10-13]: def