use strict;
use warnings;
package Set::CartesianProduct::Lazy;
$Set::CartesianProduct::Lazy::VERSION = '0.004';
{
  $Set::CartesianProduct::Lazy::DIST = 'Set-CartesianProduct-Lazy';
}

# ABSTRACT: lazily calculate the tuples of a cartesian-product

use List::Util qw( reduce );
use Scalar::Util qw( reftype );
use Data::Dumper;
use Carp;

sub new {
  my $class = shift;
  my %opts = map { %$_ } grep { reftype $_ eq 'HASH' } @_;
  my @sets = grep { reftype $_ eq 'ARRAY' } @_;

  my $getsub = __make_getsub(\%opts, \@sets);

  return bless { %opts, sets => \@sets, getsub => $getsub }, $class;
}

sub __make_getsub {
  my ($opts, $sets_ref) = @_;

  # TODO: eval code to produce the sub so the code isn't duplicated

  if ($opts->{less_lazy}) {

    # pre-calculate some things, so we don't have to every time.
    my @sets = @$sets_ref;
    my @info = map {
      [ $_, (scalar @{$sets[$_]}), reduce { $a * @$b } 1, @sets[$_ + 1 .. $#sets] ];
    } 0 .. $#sets;

    return sub {
      my ($n) = @_;

      my @tuple = map {
        my ($set_num, $set_size, $factor) = @$_;
        $sets[ $set_num ][ int( $n / $factor ) % $set_size ];
      } @info;

      return wantarray ? @tuple : \@tuple;
    };
  }

  return sub {
    my ($n) = @_;

    my @tuple;

    my @sets = @$sets_ref;

    for my $set_num (0 .. $#sets) {
      my $set_size = @{ $sets[$set_num] };
      my $factor   = reduce { $a * @$b } 1, @sets[$set_num + 1 .. $#sets];
      my $idx      = int( $n / $factor ) % $set_size;

      push @tuple, $sets[$set_num][$idx];
    }

    return wantarray ? @tuple : \@tuple;
  };
}


sub get {
  my $self = shift;
  croak "no value passed to get method\n" unless defined $_[0];
  $self->{getsub}->(@_);
}


#sub get_faster { my $self = shift; $self->{getsub}->(@_); }

{
  no warnings 'once';
  sub count { return reduce { $a * @$b } 1, @{ shift->{sets} || [] } }
}

sub last_idx { return shift->count - 1 }


1 && q{a set in time saves nine};

__END__

=pod

=head1 NAME

Set::CartesianProduct::Lazy - lazily calculate the tuples of a cartesian-product

=head1 VERSION

version 0.004

=head1 SYNOPSIS

  my @a   = qw( foo bar baz bah );
  my @b   = qw( wibble wobble weeble );
  my @c   = qw( nip nop );

  my $cpl = Set::CartesianProduct::Lazy->new( \@a, \@b, \@c );

  my $tuple;

  $tuple = $cpl->get(0);   # [ qw( foo wibble nip ) ]

  $tuple = $cpl->get(21);  # [ qw( bah wobble nop ) ]

  $tuple = $cpl->get(7);   # [ qw( bar wobble nip ) ]

  $cpl->count;             # 24
  $cpl->last_idx;          # 23

=head1 DESCRIPTION

If you have some number of arrays, say like this:

  @a = qw( foo bar baz bah );
  @b = qw( wibble wobble weeble );
  @c = qw( nip nop );

And you want all the combinations of one element from each array, like this:

  @cp = (
    [qw( foo wibble nip )],
    [qw( foo wibble nop )],
    [qw( foo wobble nip )],
    [qw( foo wobble nop )],
    [qw( foo weeble nip )],
    # ...
    [qw( bah wobble nop )],
    [qw( bah weeble nip )],
    [qw( bah weeble nop )],
  )

What you want is a Cartesian Product (also called a Cross Product, but my
mathy friends insist that Cartesian is correct)

Yes, there are already a lot of other modules on the CPAN that do this.
I won't claim that this module does this calculation any better or faster,
but it does do it I<differently>, as far as I can tell.

Nothing else seemed to offer a specific feature - I needed to pick random
individual tuples from the Cartesian Product, I<without> iterating over
the whole set and I<without> calculating any tuples until they were
asked for. Bonus points for not making a copy of the original input sets.

I needed the calculation to be lazy, and I needed random-access with O(1)
(well, O(n) for the persnickety but n is so small it might as well be 1)
retrieval time, even if that meant a slower implementation overall. And
I didn't want to use RAM unnecessarily by creating copies of the original
arrays, since the data I was working with was of a significant size.

=head1 METHODS

=head2 new

Construct a new object. Takes the following arguments:

=over 4

=item *

options

A hashref of options that modify the way the object works.
If you don't want to specify any options, simply omit this
argument.

=over 4

=item less_lazy

Makes the get method slightly faster, at the expense
of not being able to account for any modifications made
to the original input arrays, and using more memory.
If you modify one of the arrays used to consruct the object,
the results of all the other methods are B<undefined>.
You might get the wrong answer. You might trigger an
exception, you might get mutations that give you super-powers
at the expense of never being able to touch another human
being without killing them.

=back

=item *

sets

A list of arrayrefs from which to compute the cartesian product.
You can list as many as you want.

=back

Some examples:

  my $cpl = Set::CartesianProduct::Lazy->new(\@a, [qw(foo bar baz)], \@b);
  my $cpl = Set::CartesianProduct::Lazy->new( { less_lazy => 1 }, \@a, \@b, \@c);

=head2 get

Return the tuple at the given "position" in the cartesian product.
The positions, like array indices, are based at 0.

If called in scalar context, an arrayref is returned. If called in list
context, a list is returned.

If you ask for a position that exceeds the bounds of the array defining the
cartesian product the result will be... interesting. I won't make any
guarantees, but if it's useful, let me know.

Examples:

  my @tuple  = $cpl->get(12);                # list context
  my $tuple2 = $cpl->get( $cpl->count / 2 ); # scalar context

  my $fail  = $cpl->get( $cpl->count ); # probably equal to ->get(0)
  my $fail2 = $cpl->get( -1 );          # who knows

=head2 count

Return the count of tuples that would be in the cartesian
product if it had been generated.

Example:

  my $count = $cpl->count;

=head2 last_idx

Return the index of the last tuple that would be in the cartesian
product if it had been generated. This is just for conveniece
so you don't have to write code like this:

  for my $i ( 0 .. $cpl->count - 1 ) { ... }

And you can do this instead:

  for my $i ( 0 .. $cpl->last_idx ) { ... }

Which I feel is more readable.

=for :stopwords cpan testmatrix url annocpan anno bugtracker rt cpants kwalitee diff irc mailto metadata placeholders metacpan

=head1 SUPPORT

=head2 Bugs / Feature Requests

Please report any bugs or feature requests by email to C<bug-set-cartesianproduct-lazy at rt.cpan.org>, or through
the web interface at L<https://rt.cpan.org/Public/Bug/Report.html?Queue=Set-CartesianProduct-Lazy>. You will be automatically notified of any
progress on the request by the system.

=head2 Source Code

The code is open to the world, and available for you to hack on. Please feel free to browse it and play
with it, or whatever. If you want to contribute patches, please send me a diff or prod me to pull
from your repository :)

L<https://github.com/hercynium/Set-CartesianProduct-Lazy>

  git clone https://github.com/hercynium/Set-CartesianProduct-Lazy.git

=head1 AUTHOR

Stephen R. Scaffidi <sscaffidi@cpan.org>

=head1 COPYRIGHT AND LICENSE

This software is copyright (c) 2019 by Stephen R. Scaffidi.

This is free software; you can redistribute it and/or modify it under
the same terms as the Perl 5 programming language system itself.

=cut