tokstr

Routine for splitting strings by a delimiter
(replacement for the bad strtok)
written by Jamie Stevens 2006

Definition:
char *tokstr(char **curr_delim,char *delim,char *part)

Source:
tokstr.c
tokstr.h

This routine works in a similar fashion to strtok.
It takes three args:

• curr_delim - a pointer to a string which contains the string to parse.
• delim - a string containing the delimiter to break on
• part - a string which, upon return, will contain the next string token. This must be sent as a fully allocated char array.

Usage:
Begin by setting curr_delim to point to the beginning of the string you want to parse, and then call the tokstr function. Upon return, curr_delim will point to the letter directly after the first delimiter character, part will contain a the text between the start of the string and the delimiter (although not including the delimiter), and delim will not have changed. The value of curr_delim will also be the return value of the function. Subsequent calls to the function, using the curr_delim values returned from the function, will progress through the string. When no more text exists after curr_delim, the function will return NULL, and curr_delim will be set to NULL.

Example:
Consider the string "This is a short sentence."
To get each word in this string, the following code will work:

#include <stdio.h>
#include "tokstr.h"

#define BUFSIZE 256

void main(){
  char sentence[BUFSIZE],*curr_delim,part[BUFSIZE];
  int i;

  strcpy(sentence,"This is a short sentence.");
  for (i=1,curr_delim=sentence;
       tokstr(&curr_delim," ",part)!=NULL;i++)
    printf("word %d: %s\n",i,part);
}

This program would output:

word 1: This
word 2: is
word 3: a
word 4: short
word 5: sentence.