On Fri, Oct 28, 2016 at 9:21 AM, Sérgio Medeiros <sqmedei...@gmail.com> wrote:
> This would be approach B, and my point is that for some LR(k) CFGs > there are no equivalent LL(k) CFGs. > Grammar transformations beyond the simple refactoring of direct left recursion are out of the question in my case. More complex transformations would lose the relation between the original grammar and the parsing process from the user's perspective. Parse traces would become meaningless to the user, and the traces are indispensable when debugging a grammar with, f.i., >400 rules. A grammar G is a program that (hopefully) recognizes language L. In practical applications, G must be debuggable. Cheers, -- Juancarlo *Añez*
_______________________________________________ PEG mailing list PEG@lists.csail.mit.edu https://lists.csail.mit.edu/mailman/listinfo/peg
