1 package Moose::Manual::FAQ;
3 # ABSTRACT: Frequently asked questions about Moose
10 =head1 FREQUENTLY ASKED QUESTIONS
12 =head2 Module Stability
14 =head3 Is Moose "production ready"?
16 Yes! Many sites with household names are using Moose to build
17 high-traffic services. Countless others are using Moose in production.
19 As of this writing, Moose is a dependency of several hundred CPAN
20 modules. L<http://cpants.perl.org/dist/used_by/Moose>
22 =head3 Is Moose's API stable?
24 Yes. The sugary API, the one 95% of users will interact with, is
25 B<very stable>. Any changes will be B<100% backwards compatible>.
27 The meta API is less set in stone. We reserve the right to tweak
28 parts of it to improve efficiency or consistency. This will not be
29 done lightly. We do perform deprecation cycles. We I<really>
30 do not like making ourselves look bad by breaking your code.
31 Submitting test cases is the best way to ensure that your code is not
32 inadvertently broken by refactoring.
34 =head3 I heard Moose is slow, is this true?
36 Again, this one is tricky, so Yes I<and> No.
38 Firstly, I<nothing> in life is free, and some Moose features do cost
39 more than others. It is also the policy of Moose to B<only charge you
40 for the features you use>, and to do our absolute best to not place
41 any extra burdens on the execution of your code for features you are
42 not using. Of course using Moose itself does involve some overhead,
43 but it is mostly compile time. At this point we do have some options
44 available for getting the speed you need.
46 Currently we provide the option of making your classes immutable as a
47 means of boosting speed. This will mean a slightly larger compile time
48 cost, but the runtime speed increase (especially in object
49 construction) is pretty significant. This can be done with the
52 MyClass->meta->make_immutable();
54 We are regularly converting the hotspots of L<Class::MOP> to XS.
55 Florian Ragwitz and Yuval Kogman are currently working on a way to
56 compile your accessors and instances directly into C, so that everyone
57 can enjoy blazing fast OO.
59 =head3 When will Moose 1.0 be ready?
61 Moose is ready now! Stevan Little declared 0.18, released in March
62 2007, to be "ready to use".
66 =head3 How do I write custom constructors with Moose?
68 Ideally, you should never write your own C<new> method, and should use
69 Moose's other features to handle your specific object construction
70 needs. Here are a few scenarios, and the Moose way to solve them;
72 If you need to call initialization code post instance construction,
73 then use the C<BUILD> method. This feature is taken directly from Perl
74 6. Every C<BUILD> method in your inheritance chain is called (in the
75 correct order) immediately after the instance is constructed. This
76 allows you to ensure that all your superclasses are initialized
77 properly as well. This is the best approach to take (when possible)
78 because it makes subclassing your class much easier.
80 If you need to affect the constructor's parameters prior to the
81 instance actually being constructed, you have a number of options.
83 To change the parameter processing as a whole, you can use the
84 C<BUILDARGS> method. The default implementation accepts key/value
85 pairs or a hash reference. You can override it to take positional
86 args, or any other format
88 To change the handling of individual parameters, there are
89 I<coercions> (See the L<Moose::Cookbook::Basics::Recipe5> for a
90 complete example and explanation of coercions). With coercions it is
91 possible to morph argument values into the correct expected
92 types. This approach is the most flexible and robust, but does have a
93 slightly higher learning curve.
95 =head3 How do I make non-Moose constructors work with Moose?
97 Usually the correct approach to subclassing a non-Moose class is
98 delegation. Moose makes this easy using the C<handles> keyword,
99 coercions, and C<lazy_build>, so subclassing is often not the ideal
102 That said, if you really need to inherit from a non-Moose class, see
103 L<Moose::Cookbook::Basics::Recipe11> for an example of how to do it,
104 or take a look at L<Moose::Manual::MooseX/"MooseX::NonMoose">.
108 =head3 How do I tell Moose to use get/set accessors?
110 The easiest way to accomplish this is to use the C<reader> and
111 C<writer> attribute options:
119 Moose will still take advantage of type constraints, triggers, etc.
120 when creating these methods.
122 If you do not like this much typing, and wish it to be a default for
123 your classes, please see L<MooseX::FollowPBP>. This extension will
131 Moose will create separate C<get_bar> and C<set_bar> methods instead
132 of a single C<bar> method.
134 If you like C<bar> and C<set_bar>, see
135 L<MooseX::SemiAffordanceAccessor>.
137 NOTE: This B<cannot> be set globally in Moose, as that would break
138 other classes which are built with Moose. You can still save on typing
139 by defining a new L<MyApp::Moose> that exports Moose's sugar and then
140 turns on L<MooseX::FollowPBP>. See
141 L<Moose::Cookbook::Extending::Recipe4>.
143 =head3 How can I inflate/deflate values in accessors?
145 Well, the first question to ask is if you actually need both inflate
148 If you only need to inflate, then we suggest using coercions. Here is
149 some basic sample code for inflating a L<DateTime> object:
151 class_type 'DateTime';
155 => via { DateTime::Format::MySQL->parse_datetime($_) };
157 has 'timestamp' => (is => 'rw', isa => 'DateTime', coerce => 1);
159 This creates a custom type for L<DateTime> objects, then attaches
160 a coercion to that type. The C<timestamp> attribute is then told
161 to expect a C<DateTime> type, and to try to coerce it. When a C<Str>
162 type is given to the C<timestamp> accessor, it will attempt to
163 coerce the value into a C<DateTime> object using the code in found
166 For a more comprehensive example of using coercions, see the
167 L<Moose::Cookbook::Basics::Recipe5>.
169 If you need to deflate your attribute's value, the current best
170 practice is to add an C<around> modifier to your accessor:
172 # a timestamp which stores as
173 # seconds from the epoch
174 has 'timestamp' => (is => 'rw', isa => 'Int');
176 around 'timestamp' => sub {
180 return $self->$next unless @_;
182 # assume we get a DateTime object ...
183 my $timestamp = shift;
184 return $self->$next( $timestamp->epoch );
187 It is also possible to do deflation using coercion, but this tends to
188 get quite complex and require many subtypes. An example of this is
189 outside the scope of this document, ask on #moose or send a mail to
192 Still another option is to write a custom attribute metaclass, which
193 is also outside the scope of this document, but we would be happy to
194 explain it on #moose or the mailing list.
196 =head3 I created an attribute, where are my accessors?
198 Accessors are B<not> created implicitly, you B<must> ask Moose to
199 create them for you. My guess is that you have this:
201 has 'foo' => (isa => 'Bar');
203 when what you really want to say is:
205 has 'foo' => (isa => 'Bar', is => 'rw');
207 The reason this is so is because it is a perfectly valid use case to
208 I<not> have an accessor. The simplest one is that you want to write
209 your own. If Moose created one automatically, then because of the
210 order in which classes are constructed, Moose would overwrite your
211 custom accessor. You wouldn't want that would you?
213 =head2 Method Modifiers
215 =head3 How can I affect the values in C<@_> using C<before>?
217 You can't, actually: C<before> only runs before the main method, and
218 it cannot easily affect the method's execution.
220 You similarly can't use C<after> to affect the return value of a
223 We limit C<before> and C<after> because this lets you write more
224 concise code. You do not have to worry about passing C<@_> to the
225 original method, or forwarding its return value (being careful to
228 The C<around> method modifier has neither of these limitations, but is
229 a little more verbose.
231 =head3 Can I use C<before> to stop execution of a method?
233 Yes, but only if you throw an exception. If this is too drastic a
234 measure then we suggest using C<around> instead. The C<around> method
235 modifier is the only modifier which can gracefully prevent execution
236 of the main method. Here is an example:
238 around 'baz' => sub {
240 my ($self, %options) = @_;
241 unless ($options->{bar} eq 'foo') {
244 $self->$next(%options);
247 By choosing not to call the C<$next> method, you can stop the
248 execution of the main method.
250 =head3 Why can't I see return values in an C<after> modifier?
252 As with the C<before> modifier, the C<after> modifier is simply called
253 I<after> the main method. It is passed the original contents of C<@_>
254 and B<not> the return values of the main method.
256 Again, the arguments are too lengthy as to why this has to be. And as
257 with C<before> I recommend using an C<around> modifier instead. Here
260 around 'foo' => sub {
262 my ($self, @args) = @_;
263 my @rv = $next->($self, @args);
264 # do something silly with the return values
268 =head2 Type Constraints
270 =head3 How can I provide a custom error message for a type constraint?
272 Use the C<message> option when building the subtype:
274 subtype 'NaturalLessThanTen'
277 => message { "This number ($_) is not less than ten!" };
279 This C<message> block will be called when a value fails to pass the
280 C<NaturalLessThanTen> constraint check.
282 =head3 Can I turn off type constraint checking?
284 Not yet. This option may come in a future release.
286 =head3 My coercions stopped working with recent Moose, why did you break it?
288 Moose 0.76 fixed a case where Coercions were being applied even if the original constraint passed. This has caused some edge cases to fail where people were doing something like
290 subtype Address => as 'Str';
291 coerce Address => from Str => via { get_address($_) };
293 Which is not what they intended. The Type Constraint C<Address> is too loose in this case, it is saying that all Strings are Addresses, which is obviously not the case. The solution is to provide a where clause that properly restricts the Type Constraint.
295 subtype Address => as Str => where { looks_like_address($_) };
297 This will allow the coercion to apply only to strings that fail to look like an Address.
301 =head3 Why is BUILD not called for my composed roles?
303 C<BUILD> is never called in composed roles. The primary reason is that
304 roles are B<not> order sensitive. Roles are composed in such a way
305 that the order of composition does not matter (for information on the
306 deeper theory of this read the original traits papers here
307 L<http://www.iam.unibe.ch/~scg/Research/Traits/>).
309 Because roles are essentially unordered, it would be impossible to
310 determine the order in which to execute the C<BUILD> methods.
312 As for alternate solutions, there are a couple.
318 Using a combination of lazy and default in your attributes to defer
319 initialization (see the Binary Tree example in the cookbook for a good
320 example of lazy/default usage L<Moose::Cookbook::Basics::Recipe3>)
324 Use attribute triggers, which fire after an attribute is set, to
325 facilitate initialization. These are described in the L<Moose> docs,
326 and examples can be found in the test suite.
330 In general, roles should not I<require> initialization; they should
331 either provide sane defaults or should be documented as needing
332 specific initialization. One such way to "document" this is to have a
333 separate attribute initializer which is required for the role. Here is
334 an example of how to do this:
349 requires 'init_height';
351 In this example, the role will not compose successfully unless the
352 class provides a C<init_height> method.
354 If none of those solutions work, then it is possible that a role is
355 not the best tool for the job, and you really should be using
356 classes. Or, at the very least, you should reduce the amount of
357 functionality in your role so that it does not require initialization.
359 =head3 What are Traits, and how are they different from Roles?
361 In Moose, a trait is almost exactly the same thing as a role, except
362 that traits typically register themselves, which allows you to refer
363 to them by a short name ("Big" vs "MyApp::Role::Big").
365 In Moose-speak, a I<Role> is usually composed into a I<class> at
366 compile time, whereas a I<Trait> is usually composed into an instance
367 of a class at runtime to add or modify the behavior of B<just that
370 Outside the context of Moose, traits and roles generally mean exactly
371 the same thing. The original paper called them Traits, however Perl 6
372 will call them Roles.
374 =head2 Moose and Subroutine Attributes
376 =head3 Why don't subroutine attributes I inherited from a superclass work?
378 Currently when you subclass a module, this is done at runtime with the
379 C<extends> keyword but attributes are checked at compile time by
380 Perl. To make attributes work, you must place C<extends> in a C<BEGIN>
381 block so that the attribute handlers will be available at compile time
384 BEGIN { extends qw/Foo/ }
386 Note that we're talking about Perl's subroutine attributes here, not
389 sub foo : Bar(27) { ... }