Description:
------------
The existing string-tokenizing function (strtok) could be improved by being made to take an additional parameter that would tell the function to treat multiple adjacent delimiters as one for the purposes of splitting up the string.

It appears that "preg_split" can accomplish the task but that would involve concocting the right regular expressions -- which is a pain that most people would prefer to avoid. Besides, the chatter on the documentation threads is that messing with regular expressions slows down execution.

Expected result:
----------------
Say, we have the string (omit the quotes):

"abc@def.com,   ghi@jkl.com mno@pqr.com;   stu@vw.com"

And suppose we want to parse it to extract the email addresses (imagine we have a mailing list system where we add people's addresses in bulk with all sorts of crazy formatting in between the addresses that we don't have the time to neaten up).

So, the output of tokenizing the above string by the delimiters: ", ;" should in my vision be an array of just FOUR elements, namely:

abc@def.com
ghi@jkl.com
mno@pqr.com
stu@vw.com

It is easy to see that one can find a lot of uses for this particular tokenization algorithm.

Actual result:
--------------
The way things are now with strtok, I'd get an array of more than the above-listed 4 elements. In addition to them I'd get a whole bunch of empty elements in the array corresponding to all occurences of two delimiters occuring side-by-side in the string to be tokenized.

Please feel free to concact me for any clarifications. In case the improvement suggestion is integrated into PHP, credits will be welcomed.