5 Moose::Manual::Attributes - Object attributes with Moose
9 Moose attributes have many properties, and attributes are probably the
10 single most powerful and flexible part of Moose. You can create a
11 powerful class simply by declaring attributes. In fact, it's possible
12 to have classes that consist solely of attribute declarations.
14 An attribute is a property that every member of a class has. For
15 example, we might say that "every C<Person> object has a first name and
16 last name". Attributes can be optional, so that we can say "some C<Person>
17 objects have a social security number (and some don't)".
19 At its simplest, an attribute can be thought of as a named value (as
20 in a hash) that can be read and set. However, attributes can also have
21 defaults, type constraints, delegation and much more.
23 In other languages, attributes are also referred to as slots or
26 =head1 ATTRIBUTE OPTIONS
28 Use the C<has> function to declare an attribute:
34 has 'first_name' => ( is => 'rw' );
36 This says that all C<Person> objects have an optional read-write
37 "first_name" attribute.
39 =head2 Read-write vs. read-only
41 The options passed to C<has> define the properties of the
42 attribute. There are many options, but in the simplest form you just
43 need to set C<is>, which can be either C<rw> (read-write) or C<ro>
46 (In fact, you could even omit C<is>, but that gives you an attribute
47 that has no accessors, which is pointless unless you're doing some
50 =head2 Accessor methods
52 Each attribute has one or more accessor methods. An accessor lets you
53 read and write the value of that attribute for an object.
55 By default, the accessor method has the same name as the attribute. If
56 you declared your attribute as C<ro> then your accessor will be
57 read-only. If you declared it read-write, you get a read-write
60 Given our C<Person> example above, we now have a single C<first_name>
61 accessor that can read or write a C<Person> object's C<first_name>
64 If you want, you can also explicitly specify the method names to be
65 used for reading and writing an attribute's value. This is
66 particularly handy when you'd like an attribute to be publicly
67 readable, but only privately settable. For example:
71 writer => '_set_weight',
74 This might be useful if weight is calculated based on other methods.
75 For example, every time the C<eat> method is called, we might adjust
76 weight. This lets us hide the implementation details of weight
77 changes, but still provide the weight value to users of the class.
79 Some people might prefer to have distinct methods for reading and
80 writing. In I<Perl Best Practices>, Damian Conway recommends that
81 reader methods start with "get_" and writer methods start with "set_".
83 We can do exactly that by providing names for both the C<reader> and
88 reader => 'get_weight',
89 writer => 'set_weight',
92 If you're thinking that doing this over and over would be insanely
93 tedious, you're right! Fortunately, Moose provides a powerful
94 extension system that lets you override the default naming
95 conventions. See L<Moose::Manual::MooseX> for more details.
97 =head2 Predicate and clearer methods
99 Moose allows you to explicitly distinguish between a false or
100 undefined attribute value and an attribute which has not been set. If
101 you want to access this information, you must define clearer and
102 predicate methods for an attribute.
104 A predicate method tells you whether or not a given attribute is
105 currently set. Note that an attribute can be explicitly set to
106 C<undef> or some other false value, but the predicate will return
109 The clearer method unsets the attribute. This is I<not> the
110 same as setting the value to C<undef>, but you can only distinguish
111 between them if you define a predicate method!
113 Here's some code to illustrate the relationship between an accessor,
114 predicate, and clearer method.
122 clearer => 'clear_ssn',
123 predicate => 'has_ssn',
128 my $person = Person->new();
129 $person->has_ssn; # false
132 $person->ssn; # returns undef
133 $person->has_ssn; # true
136 $person->ssn; # returns undef
137 $person->has_ssn; # false
139 $person->ssn('123-45-6789');
140 $person->ssn; # returns '123-45-6789'
141 $person->has_ssn; # true
143 my $person2 = Person->new( ssn => '111-22-3333');
144 $person2->has_ssn; # true
146 By default, Moose does not make a predicate or clearer for you. You
147 must explicitly provide names for them.
149 =head2 Required or not?
151 By default, all attributes are optional, and do not need to be
152 provided at object construction time. If you want to make an attribute
153 required, simply set the C<required> option to true:
160 There are a couple caveats worth mentioning in regards to what
161 "required" actually means.
163 Basically, all it says is that this attribute (C<name>) must be provided to
164 the constructor, or be lazy with either a default or a builder. It does not
165 say anything about its value, so it could be C<undef>.
167 If you define a clearer method on a required attribute, the clearer
168 I<will> work, so even a required attribute can be unset after object
171 This means that if you do make an attribute required, providing a
172 clearer doesn't make much sense. In some cases, it might be handy to
173 have a I<private> C<clearer> and C<predicate> for a required
176 =head2 Default and builder methods
178 Attributes can have default values, and Moose provides two ways to
179 specify that default.
181 In the simplest form, you simply provide a non-reference scalar value
182 for the C<default> option:
187 predicate => 'has_size',
190 If the size attribute is not provided to the constructor, then it ends
191 up being set to C<medium>:
193 my $person = Person->new();
194 $person->size; # medium
195 $person->has_size; # true
197 You can also provide a subroutine reference for C<default>. This
198 reference will be called as a method on the object.
203 sub { ( 'small', 'medium', 'large' )[ int( rand 3 ) ] },
204 predicate => 'has_size',
207 This is dumb example, but it illustrates the point that the subroutine
208 will be called for every new object created.
210 When you provide a C<default> subroutine reference, it is called as a
211 method on the object, with no additional parameters:
218 return $self->height > 200 ? 'big' : 'average';
222 When the C<default> is called during object construction, it may be
223 called before other attributes have been set. If your default is
224 dependent on other parts of the object's state, you can make the
225 attribute C<lazy>. Laziness is covered in the next section.
227 If you want to use a reference of any sort as the default value, you
228 must return it from a subroutine. This is necessary because otherwise
229 Perl would instantiate the reference exactly once, and it would be
230 shared by all objects:
234 default => {}, # wrong!
237 Moose will throw an error if you pass a bare non-subroutine reference
240 If Moose allowed this then the default mapping attribute could easily
241 end up shared across many objects. Instead, wrap it in a subroutine
246 default => sub { {} }, # right!
249 This is a bit awkward, but it's just the way Perl works.
251 As an alternative to using a subroutine reference, you can instead
252 supply a C<builder> method for your attribute:
256 builder => '_build_size',
257 predicate => 'has_size',
261 return ( 'small', 'medium', 'large' )[ int( rand 3 ) ];
264 This has several advantages. First, it moves a chunk of code to its
265 own named method, which improves readability and code
268 We strongly recommend that you use a C<builder> instead of a
269 C<default> for anything beyond the most trivial default.
271 A C<builder>, just like a C<default>, is called as a method on the
272 object with no additional parameters.
274 =head3 Builders allow subclassing
276 Because the C<builder> is called I<by name>, it goes through Perl's
277 method resolution. This means that builder methods are both
278 inheritable and overridable.
280 If we subclass our C<Person> class, we can override C<_build_size>:
287 sub _build_size { return 'small' }
289 =head3 Builders can be composed from roles
291 Because builders are called by name, they work well with roles. For
292 example, a role could provide an attribute but require that the
293 consuming class provide the C<builder>:
298 requires '_build_size';
303 builder => '_build_size',
311 sub _build_size { return 'small' }
313 Roles are covered in L<Moose::Manual::Roles>.
315 =head2 Laziness and C<lazy_build>
317 Moose lets you defer attribute population by making an attribute
323 builder => '_build_size',
326 When C<lazy> is true, the default is not generated until the reader
327 method is called, rather than at object construction time. There are
328 several reasons you might choose to do this.
330 First, if the default value for this attribute depends on some other
331 attributes, then the attribute I<must> be C<lazy>. During object
332 construction, defaults are not generated in a predictable order, so
333 you cannot count on some other attribute being populated when
334 generating a default.
336 Second, there's often no reason to calculate a default before it's
337 needed. Making an attribute C<lazy> lets you defer the cost until the
338 attribute is needed. If the attribute is I<never> needed, you save
341 We recommend that you make any attribute with a builder or non-trivial
342 default C<lazy> as a matter of course.
344 To facilitate this, you can simply specify the C<lazy_build> attribute
345 option. This bundles up a number of options together:
352 This is the same as specifying all of these options:
357 builder => '_build_size',
358 clearer => 'clear_size',
359 predicate => 'has_size',
362 If your attribute name starts with an underscore (C<_>), then the clearer
363 and predicate will as well:
375 builder => '_build__size',
376 clearer => '_clear_size',
377 predicate => '_has_size',
380 Note the doubled underscore in the builder name. Internally, Moose
381 simply prepends the attribute name with "_build_" to come up with the
384 If you don't like the names that C<lazy_build> generates, you can
385 always provide your own:
390 clearer => '_clear_size',
393 Options that you explicitly provide are always used in favor of
394 Moose's internal defaults.
396 =head2 Constructor parameters (C<init_arg>)
398 By default, each attribute can be passed by name to the class's
399 constructor. On occasion, you may want to use a different name for
400 the constructor parameter. You may also want to make an attribute
401 unsettable via the constructor.
403 Both of these goals can be accomplished with the C<init_arg> option:
410 Now we have an attribute named "bigness", but we pass C<size> to the
413 Even more useful is the ability to disable setting an attribute via
414 the constructor. This is particularly handy for private attributes:
416 has '_genetic_code' => (
422 By setting the C<init_arg> to C<undef>, we make it impossible to set
423 this attribute when creating a new object.
425 =head2 Weak references
427 Moose has built-in support for weak references. If you set the
428 C<weak_ref> option to a true value, then it will call
429 C<Scalar::Util::weaken> whenever the attribute is set:
436 $node->parent($parent_node);
438 This is very useful when you're building objects that may contain
443 A C<trigger> is a subroutine that is called whenever the attribute is
448 trigger => \&_size_set,
452 my ( $self, $size ) = @_;
454 warn $self->name, " size is now $size\n";
457 The trigger is called as a method, and receives the new value as its argument.
458 The trigger is called I<after> the value is set.
460 This differs from an after method modifier in two ways. First, a
461 trigger is only called when the attribute is set, as opposed to
462 whenever the accessor method is called (for reading or
463 writing). Second, it is also called when an attribute's value is
464 passed to the constructor.
466 However, triggers are I<not> called when an attribute is populated
467 from a C<default> or C<builder>
469 =head2 Attribute types
471 Attributes can be restricted to only accept certain types:
473 has 'first_name' => (
478 This says that the C<first_name> attribute must be a string.
480 Moose also provides a shortcut for specifying that an attribute only
481 accepts objects that do a certain role:
485 does => 'MyApp::Weapon',
488 See the L<Moose::Manual::Types> documentation for a complete
489 discussion of Moose's type system.
493 Attributes can define methods which simply delegate to their values:
495 has 'hair_color' => (
497 isa => 'Graphics::Color::RGB',
498 handles => { hair_color_hex => 'as_hex_string' },
501 This adds a new method, C<hair_color_hex>. When someone calls
502 C<hair_color_hex>, internally, the object just calls C<<
503 $self->hair_color->as_hex_string >>.
505 See L<Moose::Manual::Delegation> for documentation on how to set up
508 =head2 Metaclass and traits
510 One of Moose's best features is that it can be extended in all sorts
511 of ways through the use of custom metaclasses and metaclass traits.
513 When declaring an attribute, you can declare a metaclass or a set of
514 traits for the attribute:
516 use MooseX::AttributeHelpers;
519 metaclass => 'Collection::Hash',
521 default => sub { {} },
524 In this case, the metaclass C<Collection::Hash> really refers to
525 L<MooseX::AttributeHelpers::Collection::Hash>.
527 You can also apply one or more traits to an attribute:
529 use MooseX::MetaDescription;
533 traits => ['MooseX::MetaDescription::Meta::Trait'],
535 html_widget => 'text_input',
536 serialize_as => 'element',
540 The advantage of traits is that you can mix more than one of them
541 together easily (in fact, a trait is just a role under the hood).
543 There are a number of MooseX modules on CPAN which provide useful
544 attribute metaclasses and traits. See L<Moose::Manual::MooseX> for
545 some examples. You can also write your own metaclasses and traits. See
546 the "Meta" and "Extending" recipes in L<Moose::Cookbook> for examples.
548 =head1 ATTRIBUTE INHERITANCE
550 By default, a child inherits all of its parent class(es)' attributes
551 as-is. However, you can explicitly change some aspects of the
552 inherited attribute in the child class.
554 The options that can be overridden in a subclass are:
564 =item * documentation
580 To override an attribute, you simply prepend its name with a plus sign
589 has '+first_name' => (
594 Now the C<first_name> attribute in C<LazyPerson> is lazy, and defaults
597 We recommend that you exercise caution when changing the type (C<isa>)
598 of an inherited attribute.
600 =head1 MORE ON ATTRIBUTES
602 Moose attributes are a big topic, and this document glosses over a few
603 aspects. We recommend that you read the L<Moose::Manual::Delegation>
604 and L<Moose::Manual::Types> documents to get a more complete
605 understanding of attribute features.
607 =head1 A FEW MORE OPTIONS
609 Moose has lots of attribute options. The ones listed below are
610 superseded by some more modern features, but are covered for the sake
613 =head2 The C<documentation> option
615 You can provide a piece of documentation as a string for an attribute:
617 has 'first_name' => (
619 documentation => q{The person's first (personal) name},
622 Moose does absolutely nothing with this information other than store
625 =head2 The C<auto_deref> option
627 If your attribute is an array reference or hash reference, the
628 C<auto_deref> option will make Moose dereference the value when it is
629 returned from the reader method:
631 my %map = $object->mapping;
633 This option only works if your attribute is explicitly typed as an
634 C<ArrayRef> or C<HashRef>.
636 However, we recommend that you use L<MooseX::AttributeHelpers> for
637 these types of attributes, which gives you much more control over how
638 they are accessed and manipulated.
642 Moose provides an attribute option called C<initializer>. This is
643 similar to C<builder>, except that it is I<only> called during object
646 This option is inherited from L<Class::MOP>, but we recommend that you
647 use a C<builder> (which is Moose-only) instead.
651 Dave Rolsky E<lt>autarch@urth.orgE<gt>
653 =head1 COPYRIGHT AND LICENSE
655 Copyright 2009 by Infinity Interactive, Inc.
657 L<http://www.iinteractive.com>
659 This library is free software; you can redistribute it and/or modify
660 it under the same terms as Perl itself.