>>Can you give a more precise sample of the problem ? > > > The classic from the Dragon book: > > (a|b)*abb$ [...] > Of course if you ANALYSE the regexp, you will get a > finite state machine and you can always write an > LL1 grammar for that.
Yes I understand what you want to say here. Regexp does not naturaly map to LL(1) unless you analyse it. Fortunately, most languages syntax can be easily expressed in terms of LL(1) and LL(k) simply a bit the parser rules. Nicolas -- Neko : One VM to run them all (http://nekovm.org)
