FYI AST post-processing is what is done in c2ast.pl<https://github.com/jddurand/MarpaX-Languages-C-AST/blob/master/bin/c2ast.pl>, and what will be done in the incoming MarpaX::Languages::C::Scan<https://github.com/jddurand/MarpaX-Languages-C-Scan>(an alternative to C::Scan providing the same methods - except the constructor up to now). >From what i am concerned I am fined with the blessed arrays, the rest, OO-like or seems decoration on Marpa::R2 - not a bad idea on itself -; Thanks / JD.
Le samedi 28 décembre 2013 23:39:41 UTC+1, Jeffrey Kegler a écrit : > > One approach that might be a "best of both worlds" approach might be to > "wrapper" Marpa::R2 and post-process its AST's according to your best > lights. > On 12/28/2013 02:01 PM, amon wrote: > > I am of the opinion that it might have been a better design choice to > invoke the appropriate constructor instead of "bless"ing a reference (which > would have had the side effect of forcing Jean-Damien to write handy > accessors for MarpaX::Languages::C::AST�). > > -- You received this message because you are subscribed to the Google Groups "marpa parser" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
