Re: Some random reflections on tpr05a

[Rick Klement]

[EMAIL PROTECTED] wrote:
> 
> If anyone has a working tpr05a solution using Parse::RecDescent
> or other CPAN module can you please post it to the list, as I
> am also interested in comparing that solution to the others.

You have only to ask:

This one uses Parse::Yapp.

Here's a rpn.y:

%left '+' '-'
%left '*' '/'

%%
expr:     NUMBER
        | '-' NUMBER      { "-$_[2]" }
        | expr '+' expr   { "$_[1] $_[3] +" }
        | expr '-' expr   { "$_[1] $_[3] -" }
        | expr '*' expr   { "$_[1] $_[3] *" }
        | expr '/' expr   { "$_[1] $_[3] /" }
        | '(' expr ')'    { $_[2] }
        ;
%%

my $s;          # warning - not re-entrant

sub yylex { $s=~/\G\d+/cg?('NUMBER',$&):$s=~/\G./cg?$&:'' }

sub parse
        {
        my $self = shift()->new(yylex=>\&yylex,
                yyerror=>sub{die "Syntax Error"});
        $s = shift;
        $self->YYParse;
        }

package main;           ###############################################

$_ = <>; s/\s+//g; print rpn->parse($_), "\n";



and a Makefile for it:

all: rpn.pl

rpn.pl: rpn.y
        yapp rpn.y
        echo '#!/usr/bin/perl' >rpn.pl
        cat rpn.pm >>rpn.pl
        chmod +x rpn.pl
        rm -f rpn.pm


that produces a rpn.pl that passes v8 tests: rpn.pl: 2807.26 strokes (ok)


Also, instead of using Parse::RecDescent, I did my own recursive
descent parser, and submitted it in the "Artistic" section at 141.36 (though
as the comment says, it's really a 137).


#!perl -p
sub o{print$"x$|++,shift}
sub f{s/^-*\d+//?o$&:e(s/.//)+s/.//}
sub t{f;o$&x1,f while s+^[*/]++}
sub e{t;o$&x1,t while s/^[+-]//}
e y/    //d


This would have scored in the middle of the pack, instead of the
actual depressing position I finished in.

-- 
Rick Klement