A quick google tells me that it supports all context-free grammars. Almost by definition that’s the wrong thing to do, I fear. You don’t want multiple parse-trees, nor even multiple derivations coexisting during the parse, only for one to emerge at the end.
ETA: I mentioned LL(k). And what I specifically wonder about, is ANTLR. I haven’t used it, but it seems like the right approach, and at some point I’m going to have to try it out.