Rolling my own markdown parser in JavaScript

For several days I was messing around trying to create a markdown parser. This is the best I could come up without using regex. I was wondering if I'm on the right track.(I only covered the *,** case )

var str = "**f*";

function parser() {


    var i;
    var string = ' ';

    for (i = 0; i < str.length; i++) {

        if (token(str[i]) !== -1) {

            var data = trim(str[i], i);

            string += data[0];

            if (data[1] === -1) {
                string += str[i];
            } else {
                i = data[1]; //skip if necessary
            }



        } else {
            string += str[i];
        }


    }


    console.log(string);
}



function trim(chr, pos) {

    var inner_string = '';
    var inner_pos = pos;

    if (str[inner_pos] === chr && str[inner_pos + 1] === chr && str[inner_pos + 1] !== undefined) // consecutive tokens
    {
        var next = pair(chr, pos);

        if (next !== -1) {
            console.log(next);

            inner_string = str.substring(pos + 2, next);
            inner_pos = next;
        } else {

            inner_pos++;
        }
    } else {

        var next = str.indexOf(chr, pos + 1);


        if (next > 0) {

            console.log(next);
            inner_string = str.substring(pos + 1, next);
            inner_pos = next;
            inner_pos = next;
        } else {

            inner_pos = next;
        }

    }

    return [inner_string, inner_pos];
}


function pair(char, pos) {


    var iter = pos;

    while (iter !== str.length - 1) {
        iter++;
        if (str[iter] === char && str[iter + 1] === char && str[iter + 1] !== undefined) {

            return iter;
        }

    }

    return -1;

}




function token(char) {

    if (char === '*') {
        return 1;
    } else return -1;
}

parser();