/ 
autobox-Transform-1.009
River stage two • 2 direct dependents • 15 total dependents
/ autobox::Transform

NAME

autobox::Transform - Autobox methods to transform Arrays and Hashes

SYNOPSIS

Examples

use autobox::Core;  # uniq, sort, join, sum, etc.
use autobox::Transform;

$books->map_by("genre");
$books->map_by(price_with_tax => [$tax_pct]);

$books->grep_by("is_sold_out");
$books->grep_by(is_in_library => [$library]);

$books->uniq_by("id");

$books->group_by("title");
$books->group_by_count("genre")
$books->group_by_array("genre")

$authors->map_by("books")->flat;

my $order_authors = $order->books
    ->uniq_by("isbn")
    ->map_by("author")
    ->map_by("name")->uniq->sort->join(", ");

my $total_order_amount = $order->books
    ->grep_by(not_covered_by_vouchers => [ $vouchers ])
    ->map_by(price_with_tax => [ $tax_pct ])
    ->sum;

Comparison of vanilla Perl and autobox version

### map_by - method call: $books are Book objects
my @genres = map { $_->genre() } @$books;
my @genres = $books->map_by("genre");

my $genres = [ map { $_->genre() } @$books ];
my $genres = $books->map_by("genre");

# With sum from autobox::Core / List::AllUtils
my $book_order_total = sum(
    map { $_->price_with_tax($tax_pct) } @{$order->books}
);
my $book_order_total = $order->books
    ->map_by(price_with_tax => [$tax_pct])->sum;

### map_by - hash key: $books are book hashrefs
my @genres = map { $_->{genre} } @$books;
my @genres = $books->map_by("genre");



### grep_by - method call: $books are Book objects
my $sold_out_books = [ grep { $_->is_sold_out } @$books ];
my $sold_out_books = $books->grep_by("is_sold_out");

my $books_in_library = [ grep { $_->is_in_library($library) } @$books ];
my $books_in_library = $books->grep_by(is_in_library => [$library]);

### grep_by - hash key: $books are book hashrefs
my $sold_out_books = [ grep { $_->{is_sold_out} } @$books ];
my $sold_out_books = $books->grep_by("is_sold_out");



### uniq_by - method call: $books are Book objects
my %seen; my $distinct_books = [ grep { ! %seen{ $_->id // "" }++ } @$books ];
my $distinct_books = $books->uniq_by("id");

### uniq_by - hash key: $books are book hashrefs
my %seen; my $distinct_books = [ grep { ! %seen{ $_->{id} // "" }++ } @$books ];
my $distinct_books = $books->uniq_by("id");



### group_by - method call: $books are Book objects

$books->group_by("title"),
# {
#     "Leviathan Wakes"       => $books->[0],
#     "Caliban's War"         => $books->[1],
#     "The Tree-Body Problem" => $books->[2],
#     "The Name of the Wind"  => $books->[3],
# },

$authors->group_by(publisher_affiliation => ["with"]),
# {
#     'James A. Corey with Orbit'     => $authors->[0],
#     'Cixin Liu with Head of Zeus'   => $authors->[1],
#     'Patrick Rothfuss with Gollanz' => $authors->[2],
# },

$books->group_by_count("genre"),
# {
#     "Sci-fi"  => 3,
#     "Fantasy" => 1,
# },

my $genre_books = $books->group_by_array("genre");
# {
#     "Sci-fi"  => [ $sf_book_1, $sf_book_2, $sf_book_3 ],
#     "Fantasy" => [ $fantasy_book_1 ],
# },


### group_by - hash key: $books are book hashrefs
$books->group_by("title"), # $books are hashrefs



#### flat - $author->books returns an arrayref of Books
my $author_books = [ map { @{$_->books} } @$authors ]
my $author_books = $authors->map_by("books")->flat

DESCRIPTION

High level autobox methods you can call on arrays, arrayrefs, hashes and hashrefs.

  • $array->map_by()

  • $array->grep_by()

  • $array->uniq_by()

  • $array->group_by()

  • $array->group_by_count()

  • $array->group_by_array()

  • $array->flat()

Raison d'etre

autobox::Core is awesome, for a variety of reasons.

  • It cuts down on dereferencing punctuation clutter.

  • It makes map and grep transforms read in the same direction it's executed.

  • It makes it easier to write those things in a natural order. No need to move the cursor around a lot just to fix dereferencing, order of operations etc.

autobox::Transform provides a few higher level methods for mapping, greping and sorting common cases which are easier to read and write.

Since they are at a slightly higher semantic level, once you know them they also provide a more specific meaning than just "map" or "grep".

(Compare the difference between seeing a "map" and seeing a "foreach" loop. Just seeing the word "map" hints at what type of thing is going on here: transforming a list into another list).

The methods of autobox::Transform are not suitable for all cases, but when used appropriately they will lead to much more clear, succinct and direct code, especially in conjunction with autobox::Core.

Transforming lists of objects vs list of hashrefs

map_by, grep_by etc are called the same way regardless of whether the list contains objects or hashrefs. The items in the list must all be either objects or hashrefs.

If the array contains objects, a method is called on each object (possibly with the arguments provided).

If the array contains hashrefs, the hash key is looked up on each item.

List and Scalar Context

All of the methods below are context sensitive, i.e. they return a list in list context and an arrayref in scalar context, just like autobox::Core.

Beware: you might be in list context when you need an arrayref.

When in doubt, assume they work like map and grep, and convert the return value to references where you might have an unobvious list context. E.g.

$self->my_method(
    # Wrong, this is list context and wouldn't return an arrayref
    books => $books->grep_by("is_published"),
),

$self->my_method(
    # Correct, convert the list to an arrayref
    books => [ $books->grep_by("is_published") ],
),
$self->my_method(
    # Correct, ensure scalar context i.e. an array ref
    books => scalar $books->grep_by("is_published"),
),

AUTOBOX ARRAY METHODS

@array->map_by($accessor, @$args?) : @array | @$array

Call the $accessor on each object in @array, or get the hash key value on each hashref in the list. Like:

map { $_->$accessor() }
# or
map { $_->{$accessor} }

Examples:

my @ahthor_names = $authors->map_by("name");
my $author_names = @publishers->map_by("authors")->map_by("name");

Optionally pass in @$args in the method call. Like:

map { $_->$accessor(@$args) }

Examples:

my @prices_including_tax = $books->map_by("price_with_tax", [ $tax_pct ]);
my $prices_including_tax = $books->map_by(price_with_tax => [ $tax_pct ]);

Or get the hash key value. Examples:

my @review_scores = $reviews->map_by("score");

@array->grep_by($accessor, @$args?) : @array | @$array

Call the $accessor on each object in the list, or get the hash key value on each hashref in the list. Like:

grep { $_->$accessor() }
grep { $_->{$accessor} }

Examples:

my @prolific_authors = $authors->grep_by("is_prolific");

Optionally pass in @$args in the method call. Like:

grep { $_->$accessor(@$args) }

Examples:

my @books_to_charge_for = $books->grep_by("price_with_tax", [ $tax_pct ]);

@array->uniq_by($accessor, @$args?) : @array | @$array

Call the $accessor on each object in the list, or get the hash key value on each hashref in the list. Return list of items wich have a unique set of return values. The order is preserved. On duplicates, keep the first occurrence.

Examples:

# You have gathered multiple Author objects with duplicate ids
my @authors = $authors->uniq_by("author_id");

Optionally pass in @$args in the method call.

Examples:

my @example_book_at_price_point = $books->uniq_by("price_with_tax", [ $tax_pct ]);

@array->group_by($accessor, @$args = [], $value_sub = object) : %key_value | %$key_value

Call ->$accessor(@$args) on each object in the array, or get the hash key for each hashref in the array (just like ->map_by) and group the return values as keys in a hashref.

The default $value_sub puts each object in the list as the hash value. If the key is repeated, the value is overwritten with the last object.

Example:

my $title_book = $books->group_by("title");
# {
#     "Leviathan Wakes"       => $books->[0],
#     "Caliban's War"         => $books->[1],
#     "The Tree-Body Problem" => $books->[2],
#     "The Name of the Wind"  => $books->[3],
# },

The $value_sub

This is a bit tricky to use, so the most common thing would probably be to use one of the more specific group_by-methods (see below). It should be capable enough to achieve what you need though, so here's how it works:

The hash key is whatever is returned from $object->$accessor(@$args).

The hash value is whatever is returned from

my $new_value = $value_sub->($current_value, $object, $key);

where:

  • $current value is the current hash value for this key (or undef if the first one).

  • $object is the current item in the list. The current $_ is also set to this.

  • $key is the key returned by $object->$accessor(@$args)

@array->group_by_count($accessor, @$args = []) : %key_count | %$key_count

Just like group_by, but the hash values are the the number of instances each $accessor value occurs in the list.

Example:

$books->group_by_count("genre"),
# {
#     "Sci-fi"  => 3,
#     "Fantasy" => 1,
# },

$book->genre() returns the genre string. There are three books counted for the "Sci-fi" key.

@array->group_by_array($accessor, @$args = []) : %key_objects | %$key_objects

Just like group_by, but the hash values are arrayrefs containing the objects which has each $accessor value.

Example:

my $genre_books = $books->group_by_array("genre");
# {
#     "Sci-fi"  => [ $sf_book_1, $sf_book_2, $sf_book_3 ],
#     "Fantasy" => [ $fantasy_book_1 ],
# },

$book->genre() returns the genre string. The three Sci-fi book objects are collected under the Sci-fi key.

@array->flat() : @array | @$array

Return a (one level) flattened array, assuming the array items themselves are array refs. I.e.

[
    [ 1, 2, 3 ],
    [ "a", "b" ],
    [ [ 1, 2 ], { 3 => 4 } ]
]->flat

returns

[ 1, 2, 3, "a", "b ", [ 1, 2 ], { 3 => 4 } ]

This is useful if e.g. a map_by("some_method") returns arrayrefs of objects which you want to do further method calls on. Example:

# ->books returns an arrayref of Book objects with a ->title
$authors->map_by("books")->flat->map_by("title")

Note: This is different from autobox::Core's ->flatten, which reurns a list rather than an array and therefore can't be used in this way.

AUTOBOX HASH METHODS

DEVELOPMENT

Author

Johan Lindstrom, <johanl [AT] cpan.org>

Source code

https://github.com/jplindstrom/p5-autobox-Transform

Bug reports

Please report any bugs or feature requests on GitHub:

https://github.com/jplindstrom/p5-autobox-Transform/issues.

COPYRIGHT & LICENSE

Copyright 2016- Johan Lindstrom, All Rights Reserved.

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