It depends. If you want to support trivial names (the ones that are used most of the time for interesting molecules), the logic is simple but you need a good database and that has to come from somewhere. If you want to parse systematic (IUPAC) names, you could in principle write code to do that, but it's not simple considering the large amount of rules involved. It would be neat to have that, but it's not in the near-term to-do list.