—

              
package List::Rubyish;
use strict;
use warnings;
use 5.008001;
use overload
    '<<'     => sub {
        my($self, $add) = @_;
        $add = [ $add ] unless ref($add) eq 'ARRAY' || ref($add) eq ref($self);
        $self->push(@$add);
    },
    '>>'     => sub {
        my($self, $add) = @_;
        if (ref($self) eq ref($add)) {
            my $tmp = $self; $self = $add; $add = $tmp;
        } elsif (ref($add) ne 'ARRAY') {
            $add = [ $add ];
        }
        $self->unshift(@$add);
    },
    '+'      => sub {
        my($self, $add, $flag) = @_;
        $add = [ $add ] unless ref($add) eq 'ARRAY' || ref($add) eq ref($self);
        $self->add($add, $flag);
    },
    fallback => 1;
our $VERSION = '0.03';
use Carp qw/croak/;
use List::Util ();
use List::MoreUtils ();
sub new {
    my $class = shift;
    $class = ref $class if ref $class;
    my $array =  @_ > 0 ? (@_ == 1 && ref($_[0]) eq 'ARRAY') ? shift : [ @_ ] : [];
    bless $array, $class;
}
sub push {
    my $self = shift;
    push @$self, @_;
    $self;
}
sub unshift {
    my $self = shift;
    unshift @$self, @_;
    $self;
}
sub shift {
    shift @{$_[0]};
}
sub pop {
    pop @{$_[0]};
}
sub first {
    my ($self, $num) = @_;
    if (defined $num) {
        return $self->slice(0, $num - 1);
    } else {
        return $self->[0];
    }
}
sub last {
    my ($self, $num) = @_;
    if (defined $num) {
        return $self->slice($self->_last_index - $num + 1, $self->_last_index);
    } else {
        return $self->[-1];
    }
}
sub slice {
    my $self = CORE::shift;
    my ($start, $end) = @_;
    my $last = $#{$self};
    if (defined $end) {
        if ($start == 0 && $last <= $end) {
            return $self;
        } else {
            $end = $last if $last < $end;
            return $self->new([ @$self[ $start .. $end ] ]);
        }
    } elsif (defined $start && 0 < $start && $last <= $start) {
        return $self->new([]);
    } else {
        return $self;
    }
}
sub dump {
    my $self = CORE::shift;
    require Data::Dumper;
    Data::Dumper->new([ $self->to_a ])->Purity(1)->Terse(1)->Dump;
}
sub zip {
    my $self = CORE::shift;
    my $array = \@_;
    my $index = 0;
    $self->collect(sub { 
         my $ary = $self->new([$_]);
         $ary->push($_->[$index]) for @$array;
         $index++;
         $ary;
    });
}
sub delete {
    my ($self, $value, $code) = @_;
    my $found = 0;
    if (ref $value eq 'CODE') {
        do { my $item = $self->shift; $value->($item) ? $found = 1 : $self->push($item) } for (0..$self->_last_index);
    } else {
        do { my $item = $self->shift; $item eq $value ? $found = 1 : $self->push($item) } for (0..$self->_last_index);
    }
    $found ? $value 
           : ref $code eq 'CODE' ? do { local $_ = $value; return $code->($_) }
                                 : return ;
}
sub delete_at {
    my ($self, $pos) = @_;
    my $last_index = $self->_last_index;
    return if $pos > $last_index ;
    my $result;
    $_ == $pos ? $result = $self->shift 
               : $self->push($self->shift) for 0..$last_index;
    return $result;
}
sub delete_if {
    my ($self, $code) = @_;
    croak "Argument must be a code" unless ref $code eq 'CODE';
    my $last_index = $self->_last_index;
    for (0..$last_index) {
        my $item = $self->shift;
        local $_ = $item;
        $self->push($item) unless $code->($_);
    }
    return $self;
}
sub reject {
    my ($self, $code) = @_;
    return $self->dup->delete_if($code);
}
sub inject {
    my ($self, $result, $code) = @_;
    croak "Argument must be a code" unless ref $code eq 'CODE';
    $result = $code->($result, $_) for @{$self->dup};
    return $result;
}
sub join {
    my ($self, $delimiter) = @_;
    join $delimiter, @$self;
}
sub each_index {
    my ($self, $code) = @_;
    $self->new([ 0..$self->_last_index ])->each($code);
}
sub _last_index {
    my $self = CORE::shift;
    $self->length ? $self->length - 1 : 0;
};
sub concat {
    my ($self, $array) = @_;
    $self->push(@$array);
    $self;
}
*append = \&concat;
sub prepend {
    my ($self, $array) = @_;
    $self->unshift(@$array);
    $self;
}
sub _append_undestructive {
    my ($self, $array) = @_;
    $self->dup->push(@$array);
}
sub _prepend_undestructive {
    my ($self, $array) = @_;
    $self->dup->unshift(@$array);
}
sub add {
    my ($self, $array, $bool) = @_;
    $bool ? $self->_prepend_undestructive($array)
          : $self->_append_undestructive($array);
}
sub each {
    my ($self, $code) = @_;
    croak "Argument must be a code" unless ref $code eq 'CODE';
    $code->($_) for @{$self->dup};
    $self;
}
sub collect {
    my ($self, $code) = @_;
    croak "Argument must be a code" unless ref $code eq 'CODE';
    my @collected = CORE::map &$code, @{$self->dup};
    wantarray ? @collected : $self->new(\@collected);
}
*map = \&collect;
sub grep {
    my ($self, $code) = @_;
    $code or return;
    my @grepped;
    if (!ref($code)) {
        for (@$self) {
            if (ref($_) eq 'HASH') {
                CORE::push @grepped, $_ if $_->{$code};
            } else {
                CORE::push @grepped, $_ if $_->$code;
            }
        }
    } elsif (ref $code eq 'CODE') {
        @grepped = CORE::grep &$code, @$self;
    } else {
        croak "Invalid code";
    }
    wantarray ? @grepped : $self->new(\@grepped);
}
sub find {
    my ($self, $cond) = @_;
    my $code = (ref $cond and ref $cond eq 'CODE')
        ? $cond
        : sub { $_ eq $cond };
    for (@$self) { &$code and return $_ }
    return;
}
*detect = \&find;
sub select {
    my ($self, $code) = @_;
    croak "Argument must be a code" unless ref $code eq 'CODE';
    return $self unless $self->size;
    my $last_index = $self->_last_index;
    my $new = $self->dup;
    for (0..$last_index) {
        my $item = $new->shift;
        local $_ = $item;
        $new->push($item) if $code->($_);
    }
    return $new;
}
*find_all = \&select;
sub index_of {
    my ($self, $target) = @_;
    my $code = (ref $target eq 'CODE') ? $target : sub { CORE::shift eq $target };
    for (my $i = 0; $i < $self->length; $i++) {
        $code->($self->[$i]) and return $i;
    }
    return;
}
sub sort {
    my ($self, $code) = @_;
    my @sorted = $code ? CORE::sort { $code->($a, $b) } @$self : CORE::sort @$self;
    wantarray ? @sorted : $self->new(\@sorted);
}
sub sort_by {
    my ($self, $code, $cmp) = @_;
    my @sorted = $cmp ?
        CORE::map { $_->[1] } CORE::sort { $cmp->($a->[0], $b->[0]) } CORE::map { [$code->($_), $_] } @$self :
        CORE::map { $_->[1] } CORE::sort { $a->[0] <=> $b->[0] } CORE::map { [$code->($_), $_] } @$self;
    wantarray ? @sorted : $self->new(\@sorted);
}
sub compact {
    CORE::shift->grep(sub { defined  });
}
sub length {
    scalar @{$_[0]};
}
*size = \&length;
sub flatten {
    my $self = CORE::shift;
    $self->collect(sub { _flatten($_)  });
}
sub _flatten {
    my $element = CORE::shift;
    (ref $element and ref $element eq 'ARRAY')
        ? CORE::map { _flatten($_) } @$element
        : $element;
}
sub is_empty {
    !$_[0]->length;
}
sub uniq {
    my $self = CORE::shift;
    $self->new([ List::MoreUtils::uniq(@$self) ]);
}
sub reduce {
    my ($self, $code) = @_;
    croak "Argument must be a code" unless ref $code eq 'CODE';
    List::Util::reduce { $code->($a, $b) } @$self;
}
sub to_a {
    my @unblessed = @{$_[0]};
    \@unblessed;
}
sub as_list { # for Template::Iterator
    CORE::shift;
}
sub dup {
    __PACKAGE__->new($_[0]->to_a);
}
sub reverse {
    my $self = CORE::shift;
    $self->new([ reverse @$self ]);
}
sub sum {
    List::Util::sum @{$_[0]};
}
1;
__END__
HideShow 341 lines of Pod
=head1 NAME
List::Rubyish - Array iterator like the Ruby
=head1 SYNOPSIS
  my $array_ref = [
    {name => 'jkondo'},
    {name => 'cinnamon'}
  ];
  my $list = List::Rubyish->new($array_ref);
  $list->size;              #=> 2
  my $first = $list->shift; #=> {name => 'jkondo'}
  $list->push($first);      #=> [{name => 'cinnamon'}, {name => 'jkondo'}];
  # List::Rubyish provides much more useful methods. For more
  # details, see the sections below.
=head1 OVERVIEW
L<DBIx::MoCo::List> is very useful, However installation is complex.
List::Rubyish was made in order to enable use of L<DBIx::MoCo::List> independently.
=head1 METHODS
=over 4
=item dump ()
Dump the content of C<$self> using L<Data::Dumper>.
=item push ( I<@array> )
=item unshift ( I<@array> )
Sets the argument into C<$self>, a refernce to an array blessed by
List::Rubyish, like the same name functions provided by Perl core,
then returns a List::Rubyish object.
  my $list = List::Rubyish->new([qw(1 2 3)]);
  $list->push(4, 5); #=> [1, 2, 3, 4, 5]
  $list->unshift(0); #=> [0, 1, 2, 3, 4, 5]
=item concat ( I<\@array> )
=item prepend ( I<\@array> )
They're almost the same as C<push()>/C<unshift()> described above
except that the argument shoud be a reference to an array.
  my $list = List::Rubyish->new([1, 2, 3]);
  $list->concat([4, 5]); #=> [1, 2, 3, 4, 5]
  $list->prepend([0]);   #=> [0, 1, 2, 3, 4, 5]
=item shift ()
=item pop ()
Pulls out the first/last element from C<$self>, a refernce to an array
blessed by List::Rubyish, then returns it like the same name
functions in Perl core.
  $list = List::Rubyish->new([1, 2, 3]);
  $list->shift; #=> 1
  $list->pop;   #=> 3
  $list->dump   #=> [2]
=item first ()
=item last ()
Returns the first/last element of C<$self>, a refernce to an array
blessed by List::Rubyish. These methods aren't destructive contrary
to C<shift()>/C<pop()> method.
  $list = List::Rubyish->new([1, 2, 3]);
  $list->first; #=> 1
  $list->last;  #=> 3
  $list->dump   #=> [1, 2, 3]
=item slice ( I<$start>, I<$end> )
Returns the elements whose indexes are between C<$start> and C<$end>
as a List::Rubyish object.
  $list = List::Rubyish->new([qw(1 2 3 4)]);
  $list->slice(1, 2) #=> [2, 3]
=item zip ( I<\@array1>, I<\@array2>, ... )
Bundles up the elements in each arguments into an array or a
List::Rubyish object along with the context.
  my $list = List::Rubyish->new([1, 2, 3]);
  $list->zip([4, 5, 6], [7, 8, 9]);
      #=> [[1, 4, 7], [2, 5, 8], [3, 6, 9]]
  # When the numbers of each list are different...
  $list = List::Rubyish->new([1, 2, 3]);
  $list->zip([4, 5], [7, 8, 9]);
      #=> [[1, 4, 7], [2, 5, 8], [3, undef, 9]]
  my $list   = List::Rubyish->new([1, 2]);
  $list->zip([4, 5], [7, 8, 9]);
      #=> [[1, 4, 7], [2, 5, 8]]
=item delete ( I<$value>, I<$code> )
Deletes the same values as C<$value> in C<$self>, a refernce to an
array blessed by List::Rubyish, and returns the value if found. If
the value is not found in C<$self> and C<$code> is passed in, the code
is executed using the value as an argument to find the value to be
deleted.
  $list = List::Rubyish->new([1, 2, 3, 2, 1]);
  $list->delete(2); #=> 2
  $list->dump       #=> [1, 3, 1]
=item delete_at ( I<$pos> )
Deletes the element at C<$pos> and returns it.
  $list = List::Rubyish->new([1, 2, 3, 2, 1]);
  $list->delete_at(3); #=> 2
  $list->dump          #=> [1, 2, 3, 1]
=item delete_if ( I<$code> )
Deletes the elements if C<$code> returns true value with each element
as an argument.
  $list = List::Rubyish->new([1, 2, 3, 4]);
  $list->delete_if(sub { ($_ % 2) == 0) });
  $list->dump #=> [1, 3]
=item inject ( I<$result>, I<$code> )
Executes folding calculation using C<$code> through each element and
returns the result.
  $list = List::Rubyish->new([1, 2, 3, 4]);
  $list->inject(0, sub { $_[0] + $_[1] }); #=> 10
=item join ( I<$delimiter> )
Joins all the elements by C<$delimiter>.
  $list = List::Rubyish->new([0 1 2 3]);
  $list->join(', ') #=> '0, 1, 2, 3'
=item each_index ( I<$code> )
Executes C<$code> with each index of C<$self>, a refernce to an array
blessed by List::Rubyish.
  $list = List::Rubyish->new([1, 2, 3]);
  $list->each_index(sub { do_something($_) });
=item each ( I<$code> )
Executes C<$code> with each value of C<$self>, a refernce to an array
blessed by List::Rubyish.
  $list = List::Rubyish->new([1, 2, 3]);
  $list->each(sub { do_something($_) });
=item collect ( I<$code> )
Executes C<$code> with each element of C<$self>, a refernce to an
array blessed by List::Rubyish using CORE::map() and returns the
results as a list or List::Rubyish object along with the context.
  $list = List::Rubyish->new([1, 2, 3]);
  $list->map(sub { $_ * 2 }); #=> [2, 4, 6]
=item map ( I<$code> )
An alias of C<collect()> method described above.
=item grep ( I<$code> )
Executes C<$code> with each element of C<$self>, a refernce to an
array blessed by List::Rubyish using CORE::grep() and returns the
results as a list or List::Rubyish object along with the context.
  $list = List::Rubyish->new([qw(1 2 3 4)]);
  $list->grep(sub { ($_ % 2) == 0 }); #=> [2, 4]
=item find ( I<$code> )
Returns the first value found in C<$self>, a refernce to an array
blessed by List::Rubyish, as a result of C<$code>..
  $list = List::Rubyish->new([1, 2, 3, 4]);
  $list->find(sub { ($_ % 2) == 0 }); #=> 2
=item select ( I<$code> )
Returns the values found in C<$self>, a refernce to an array
blessed by List::Rubyish, as a result of C<$code>..
  $list = List::Rubyish->new([1, 2, 3, 4]);
  $list->select(sub { ($_ % 2) == 0 }); #=> 2, 4
=item index_of ( I<$arg> )
Returns index of given target or given code returns true.
  $list = List::Rubyish->new([qw(foo bar baz)]);
  $list->index_of('bar');                  #=> 1
  $list->index_of(sub { shift eq 'bar' }); #=> 1
=item sort ( I<$code> )
Sorts out each element and returns the result as a list or
List::Rubyish object along with the context.
  $list = List::Rubyish->new([qw(3 2 4 1]);
  $list->sort;                          #=> [1, 2, 3, 4]
  $list->sort(sub { $_[1] <=> $_[0] }); #=> [4, 3, 2, 1]
=item sort_by ( I<$code>, I<$cmp> )
Sorts out each element with Schwartzian transform returns the result as a list or
List::Rubyish object along with the context.
  $list = List::Rubyish->new([ [3], [2], [4], [1]]);
  $list->sort_by(sub { $_->[0] }); #=> [[1], [2], [3], [4]]
  $list->sort_by(sub { $_->[0] }, sub { $_[1} <=> $_[0] } ); #=> [[4], [3], [2], [1]]
=item compact ()
Eliminates undefined values in C<$self>, a refernce to an array
blessed by List::Rubyish.
  $list = List::Rubyish->new([1, 2, undef, 3, undef, 4]);
  $list->compact; #=> [1, 2, 3, 4]
=item length ()
Returns the length of C<$self>, a refernce to an array blessed by
List::Rubyish.
  $list = List::Rubyish->new([qw(1 2 3 4)]);
  $list->length; #=> 4
=item size ()
An alias of C<length()> method described above.
=item flatten ()
Returns a list or List::Rubyish object which is recursively
flattened out.
  $list = List::Rubyish->new([1, [2, 3, [4], 5]]);
  $list->flattern; #=> [1, 2, 3, 4, 5]
=item is_empty ()
Returns true if C<$self>, a refernce to an array blessed by
List::Rubyish, is empty.
=item uniq ()
Uniquifies the elements in C<$self>, a refernce to an array blessed by
List::Rubyish, and returns the result.
  $list = List::Rubyish->new([1, 2, 2, 3, 3, 4])
  $list->uniq; #=> [1, 2, 3, 4]
=item reduce ( I<$code> )
Reduces the list by C<$code>.
  # finds the maximum value
  $list = List::Rubyish->new([4, 1, 3, 2])
  $list->reduce(sub { $_[0] > $_[1] ? $_[0] : $_[1] }); #=> 4
See L<List::Util> to get to know about details of C<reduce()>.
=item reverse ()
Returns an reversely ordered C<$self>, a refernce to an array blessed
by List::Rubyish.
  $list = List::Rubyish->new([4, 1, 3, 2])
  $list->reverse; #=> [2, 3, 1, 4]
=item dup ()
Returns a duplicated C<$self>, a refernce to an array blessed by
List::Rubyish.
=item sum ()
Returns the sum of each element in C<$self>, a refernce to an array
blessed by List::Rubyish.
  $list = List::Rubyish->new([1, 2, 3, 4]);
  $list->sum; #=> 10
=back
=head1 SEE ALSO
L<DBIx::MoCo::List>, L<List::Util>, L<List::MoreUtils>, L<http://github.com/naoya/list-rubylike>, L<http://d.hatena.ne.jp/naoya/20080419/1208579525>, L<http://www.ruby-lang.org/ja/man/html/Enumerable.html>
=head1 AUTHOR
Junya Kondo, E<lt>jkondo@hatena.comE<gt>,
Naoya Ito, E<lt>naoya@hatena.ne.jpE<gt>,
Kentaro Kuribayashi, E<lt>kentarok@gmail.comE<gt>,
Yuichi Tateno, E<lt>secondlife at hatena ne jp<gt>,
Kazuhiro Osawa E<lt>yappo <at> shibuya <döt> plE<gt>
=head2 thanks to
naoya, kentaro, tokuhirom, kan, lopnor
=head1 REPOSITORY
  svn co http://svn.coderepos.org/share/lang/perl/List-Rubyish/trunk List-Rubyish
List::Rubyish is Subversion repository is hosted at L<http://coderepos.org/share/>.
patches and collaborators are welcome.
=head1 LICENSE
This library is free software; you can redistribute it and/or modify
it under the same terms as Perl itself.
=head1 DBIx::MoCo::List's COPYRIGHT
Copyright (C) Hatena Inc. All  Rights  Reserved.
=cut