Last active
January 27, 2016 03:29
-
-
Save BenGoldberg1/b4ca523738ff64b2a1c9 to your computer and use it in GitHub Desktop.
Array based replacement for PloverHash.pl
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
use strict; | |
use warnings; | |
do 'PloverOO.pl'; | |
package NFA; | |
# The NFA object is simply an arrayref. | |
# The starting state is index 0, the final state | |
# is one past the end. | |
# Whenever we push a new element onto the end, | |
# we can assume that whenever the previous part of the NFA | |
# succeeds, it will transition to our new element. | |
# Elements of the arrayref are hashrefs; each hashref | |
# has keys which are length-one strings, or '', which | |
# represents the lambda transitions. Values are always | |
# arrayrefs of states, which are integers. | |
sub CONCAT::push_NFA { | |
my ($self, $so_far) = @_; | |
$_->push_NFA($so_far) for @{$self}[ 0 .. $#$self-1 ]; | |
return $self->[-1]->push_NFA($so_far); | |
} | |
sub PLUS::push_NFA { | |
my ($self, $so_far) = @_; | |
my $start = @$so_far; | |
my @e = $self->[0]->push_NFA( $so_far ); | |
for my $e ( @e ) { | |
next if grep $_ == $start, @$e; | |
push @$e, $start; | |
} | |
@e; | |
} | |
sub STAR::push_NFA { | |
my ($self, $so_far) = @_; | |
my $start = @$so_far; | |
my @e = PLUS::push_NFA( $self, $so_far ); | |
my $first = ($so_far->[$start]{''} ||= []); | |
unless( grep $_ == $first, @e ) { | |
push @$first, scalar @$so_far; | |
unshift @e, $first; | |
} | |
@e; | |
} | |
sub LITERAL::push_NFA { | |
my ($self, $so_far) = @_; | |
my %t; | |
push @$so_far, \%t; | |
$t{$self->[0]} = [ scalar @$so_far ]; | |
values %t; | |
} | |
sub CLASS::d::push_NFA { | |
my ($self, $so_far) = @_; | |
my %t; | |
push @$so_far, \%t; | |
$t{$_} = [ scalar @$so_far ] for 0 .. 9; | |
values %t; | |
} | |
# If our regex is a|b|c, then the parse tree will be: | |
# bless [bless ['a'], LITERAL, bless ['b'], LITERAL, bless ['c'], LITERAL], ALTERN. | |
# The transition tables produced should be: | |
# { 'a' => [3], '' => [1, 2] }, { 'b' => [3] }, { 'c' => [3] }. | |
# If the regex is a|bc|d, the tables should be: | |
# { 'a' => [4], '' => [1, 3] }, { 'b' => [2] }, { 'c' => [4] }, { 'd' => [4] }. | |
sub ALTERN::push_NFA { | |
my ($self, $so_far) = @_; | |
my (@s, @e0, @end_groups, $end); | |
for my $altern ( @$self ) { | |
push @s, scalar @$so_far; | |
@end_groups = $altern->push_NFA( $so_far ); | |
$end = @$so_far; | |
last if @s == @$self; | |
for my $table ( @end_groups ) { | |
@$table = grep $end != $_, @$table; | |
} | |
push @e0, @end_groups; | |
} | |
my $start = shift @s; | |
if( $so_far->[$start]{''} ) { | |
my %seen; | |
@seen{ @{$so_far->[$start]{''}} } = (); | |
@s = grep !exists $seen{$_}, @s; | |
} | |
push @{$so_far->[$start]{''}}, @s; | |
for my $table ( @e0 ) { | |
next if grep $end == $_, @$table; | |
push @$table, $end; | |
} | |
return @e0, @end_groups; | |
} | |
sub new { | |
my ($class, $tree) = @_; | |
my @self; | |
$tree->push_NFA(\@self); | |
bless \@self, $class; | |
} | |
sub start_state { | |
return 0; | |
} | |
sub is_end_state { | |
my ($self, $state) = @_; | |
return $state == @$self; | |
} | |
sub transitions_for_symbol { | |
my ($self, $state, $symbol) = @_; | |
my $s = $self->[$state] or return; | |
my $n = $s->{$symbol} or return; | |
@$n; | |
} | |
if( !caller ) { | |
NFA->self_test; | |
#NFA->self_test_two; | |
exit 0; | |
} | |
1; | |
__END__ |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment