Daily Archives: January 1, 2013

Implement strtok in C

To implement the C function ‘strtok’, one doesn’t need to allocate extra memory for the input string before modifying it, since the caller function is supposed to make copy of the input string and deallocate the original input string. The function only needs to define static variables to store the starting of the string.

The caller function provides a string as the input during the first call, and ‘null’ for the following function calls. For details about how this function works, please refer to http://www.cplusplus.com/reference/cstring/strtok/.

char* sp = NULL; /* the start position of the string */

char* strtok1(char* str, const char* delimiters) {

	int i = 0;
	int len = strlen(delimiters);

	/* check in the delimiters */
	if(len == 0)
		printf("delimiters are empty\n");

	/* if the original string has nothing left */
	if(!str && !sp)
		return NULL;

	/* initialize the sp during the first call */
	if(str && !sp)
		sp = str;

	/* find the start of the substring, skip delimiters */
	char* p_start = sp;
	while(true) {
		for(i = 0; i < len; i ++) {
			if(*p_start == delimiters[i]) {
				p_start ++;
				break;
			}
		}

		if(i == len) {
		       sp = p_start;
		       break;
		}
	}

	/* return NULL if nothing left */
	if(*sp == '\0') {
		sp = NULL;
		return sp;
	}

	/* find the end of the substring, and
        replace the delimiter with null */
	while(*sp != '\0') {
		for(i = 0; i < len; i ++) {
			if(*sp == delimiters[i]) {
				*sp = '\0';
				break;
			}
		}

		sp ++;
		if (i < len)
			break;
	}

	return p_start;
}

This following is the test code from cplusplus.com.

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

int main ()
{
  char str[] ="- This, a sample string.";
  char * pch;
  printf ("Splitting string \"%s\" into tokens:\n",str);
  pch = strtok1(str," ,.-");
  while (pch != NULL)
  {
    printf ("%s\n",pch);
    pch = strtok1(NULL, " ,.-");
  }
  return 0;
}

————————-
As pointed out by Ian in the following comments [9.25.2014], there is a much simpler solution to this problem. Feel free to test it and let us know if it has any bugs.

char* strtok1(char *str, const char* delim) {
	static char* _buffer;
	if(str != NULL) _buffer = str;
	if(_buffer[0] == '\0') return NULL;

	char *ret = _buffer, *b;
	const char *d;

	for(b = _buffer; *b !='\0'; b++) {
		for(d = delim; *d != '\0'; d++) {
			if(*b == *d) {
				*b = '\0';
				_buffer = b+1;

				// skip the beginning delimiters
				if(b == ret) { 
					ret++; 
					continue; 
				}
				return ret;
			}
		}
	}

	return ret;
}