The key first step would be to "tokenize" the commas that are outside of parentheses. The second step would be to toss away or ignore the equals sign (if someone used one).
Edit: Or as a second step, add an equals sign if missing and continue parsing normally. I like the other way better though.