Joel E. Denny wrote:
I haven't had a chance to look through your actual patch, and it might be a while. I believe it needs to wait until after 2.2 is out anyway. If no one else gets to it first, I will look at it eventually.
So far, so good. Anyway, you might prefer to implement a variant of Tomita's algorithm so that the hidden left-recursion case ceases to be a problem. The best known variant seems to be described in Nozohoor-Fashi, R.: _GLR_parsing_for_epsilon-grammars_. In Tomita, M., ed.: Generalized LR parsing. Kluwer Academic Publishers, Netherlands (1991) 60--75 Another interesting variant seems to be described in Scott, E., Johnstone, A.: _Reducing_non-determinism_in_reduction_ _modified_GLR_parsers_. Acta Informatica 40:6-7 (2004) 459--489 -- Regards, Sylvain
