Commit | Line | Data |
e67a0fca |
1 | |
2 | =pod |
3 | |
4 | =head1 NAME |
5 | |
6 | Moose::Cookbook::FAQ - Frequenty asked questions about Moose |
7 | |
8 | =head1 FREQUENTLY ASKED QUESTIONS |
9 | |
2a0f3bd3 |
10 | =head2 Module Stability |
11 | |
12 | =head3 Is Moose "production ready"? |
13 | |
734d1752 |
14 | Yes. I have two medium-to-large-ish web applications in |
15 | production using Moose, they have been running without |
16 | issue now for almost a year. |
17 | |
18 | At $work we are re-writing our core offering to use Moose, |
19 | so it's continued development is assured. |
20 | |
21 | Several other people on #moose either have apps in production |
22 | which use Moose, or are in the process of deploying sites |
23 | which use Moose. |
2a0f3bd3 |
24 | |
25 | =head3 Is Moose's API stable? |
26 | |
27 | Yes and No. The external API, the one 90% of users will interact |
28 | with, is B<very stable> and any changes B<will be 100% backwards |
29 | compatible>. The introspection API is I<mostly> stable, I still |
30 | reserve the right to tweak that if needed, but I will do my |
31 | absolute best to maintain backwards comptability here as well. |
32 | |
734d1752 |
33 | =head3 I heard Moose is slow, is this true? |
2a0f3bd3 |
34 | |
35 | Again, this one is tricky, so Yes I<and> No. |
36 | |
37 | First let me say that I<nothing> in life is free, and that some |
38 | Moose features do cost more than others. It is also the |
39 | policy of Moose to B<only charge you for the features you use>, |
40 | and to do our absolute best to not place any extra burdens on |
d44714be |
41 | the execution of your code for features you are not using. Of |
42 | course using Moose itself does involve some overhead, but it |
43 | is mostly compile time. At this point we do have some options |
44 | available for getting the speed you need. |
2a0f3bd3 |
45 | |
734d1752 |
46 | Currently we have the option of making your classes immutable |
47 | as a means of boosting speed. This will mean a larger compile |
48 | time cost, but the runtime speed increase (especially in object |
d44714be |
49 | construction) is pretty signifigant. This is not very well |
50 | documented yet, so please ask on the list of on #moose for more |
51 | information. |
52 | |
53 | We are also discussing and experimenting with L<Module::Compile>, |
54 | and the idea of compiling highly optimized C<.pmc> files. And |
55 | we have also mapped out some core methods as canidates for |
56 | conversion to XS. |
2a0f3bd3 |
57 | |
58 | =head3 When will Moose be 1.0 ready? |
59 | |
734d1752 |
60 | I had originally said it would be end of 2006, but various bits |
61 | of $work kept me too busy. At this point, I think we are getting |
62 | pretty close and I will likely declare 1.0 within the next few |
63 | releases. |
2a0f3bd3 |
64 | |
734d1752 |
65 | When will that be? Hard to say really, but honestly, it is ready |
66 | to use now, the difference between now and 1.0 will be pretty |
67 | minimal. |
2a0f3bd3 |
68 | |
e67a0fca |
69 | =head2 Constructors |
70 | |
71 | =head3 How do I write custom constructors with Moose? |
72 | |
73 | Ideally, you should never write your own C<new> method, and should |
74 | use Moose's other features to handle your specific object construction |
75 | needs. Here are a few scenarios, and the Moose way to solve them; |
76 | |
77 | If you need to call initializtion code post instance construction, |
78 | then use the C<BUILD> method. This feature is taken directly from |
79 | Perl 6. Every C<BUILD> method in your inheritence chain is called |
80 | (in the correct order) immediately after the instance is constructed. |
81 | This allows you to ensure that all your superclasses are initialized |
82 | properly as well. This is the best approach to take (when possible) |
83 | because it makes subclassing your class much easier. |
84 | |
85 | If you need to affect the constructor's parameters prior to the |
86 | instance actually being constructed, you have a number of options. |
87 | |
88 | First, there are I<coercions> (See the L<Moose::Cookbook::Recipe5> |
89 | for a complete example and explaination of coercions). With |
90 | coercions it is possible to morph argument values into the correct |
91 | expected types. This approach is the most flexible and robust, but |
92 | does have a slightly higher learning curve. |
93 | |
94 | Second, using an C<around> method modifier on C<new> can be an |
95 | effective way to affect the contents of C<@_> prior to letting |
96 | Moose deal with it. This carries with it the extra burden for |
97 | your subclasses, in that they have to be sure to explicitly |
98 | call your C<new> and/or work around your C<new> to get to the |
99 | version from L<Moose::Object>. |
100 | |
101 | The last approach is to use the standard Perl technique of calling |
102 | the C<SUPER::new> within your own custom version of C<new>. This |
103 | of course brings with it all the issues of the C<around> solution |
104 | along with any issues C<SUPER::> might add as well. |
105 | |
106 | In short, try to use C<BUILD> and coercions, they are your best |
107 | bets. |
108 | |
109 | =head3 How do I make non-Moose constuctors work with Moose? |
110 | |
111 | Moose provides it's own constructor, but it does it by making all |
112 | Moose-based classes inherit from L<Moose::Object>. When inheriting |
113 | from a non-Moose class, the inheritence chain to L<Moose::Object> |
114 | is broken. The simplest way to fix this is to simply explicitly |
115 | inherit from L<Moose::Object> yourself. However, this does not |
116 | always fix the issue of a constructor. Here is a basic example of |
117 | how this can be worked around: |
118 | |
119 | package My::HTML::Template; |
120 | use Moose; |
121 | |
122 | # explict inheritence |
123 | extends 'HTML::Template', 'Moose::Object'; |
124 | |
125 | # explicit constructor |
126 | sub new { |
127 | my $class = shift; |
128 | # call HTML::Template's constructor |
129 | my $obj = $class->SUPER::new(@_); |
130 | return $class->meta->new_object( |
131 | # pass in the constructed object |
132 | # using the special key __INSTANCE__ |
133 | __INSTANCE__ => $obj, @_ |
134 | ); |
135 | } |
136 | |
137 | Of course this only works if both your Moose class, and the |
138 | inherited non-Moose class use the same instance type (typically |
139 | HASH refs). |
140 | |
141 | Other techniques can be used as well, such as creating the object |
142 | using C<Moose::Object::new>, but calling the inherited non-Moose |
143 | class's initializtion methods (if available). |
144 | |
145 | It is also entirely possible to just rely on HASH autovivification |
146 | to create the slot's needed for Moose based attributes. Although |
147 | this does somewhat restrict use of construction time attribute |
148 | features. |
149 | |
150 | In short, there are several ways to go about this, it is best to |
151 | evaluate each case based on the class you wish to extend, and the |
152 | features you wish to employ. As always, both IRC and the mailing |
153 | list are great ways to get help finding the best approach. |
154 | |
155 | =head2 Accessors |
156 | |
157 | =head3 How do I tell Moose to use get/set accessors? |
158 | |
159 | The easiest way to accomplish this is to use the C<reader> and |
160 | C<writer> attribute options. Here is some example code: |
161 | |
162 | has 'bar' => ( |
163 | isa => 'Baz', |
164 | reader => 'get_bar', |
165 | writer => 'set_bar', |
166 | ); |
167 | |
168 | Moose will still take advantage of type constraints, triggers, etc. |
169 | when creating these methods. |
170 | |
171 | If you do not like this much typing, and wish it to be a default for |
172 | your class. Please see L<Moose::Policy>, and more specifically the |
173 | L<Moose::Policy::FollowPBP>. This will allow you to write this: |
174 | |
175 | has 'bar' => ( |
176 | isa => 'Baz', |
177 | is => 'rw', |
178 | ); |
179 | |
180 | And have Moose create C<get_bar> and C<set_bar> instead of the usual |
181 | C<bar>. |
182 | |
183 | NOTE: This B<cannot> be set globally in Moose, as this would break |
184 | other classes which are built with Moose. |
185 | |
186 | =head3 How can I get Moose to inflate/deflate values in the accessor? |
187 | |
188 | Well, the first question to ask is if you actually need both inflate |
189 | and deflate. |
190 | |
191 | If you only need to inflate, then I suggest using coercions. Here is |
192 | some basic sample code for inflating a L<DateTime> object. |
193 | |
194 | subtype 'DateTime' |
195 | => as 'Object' |
196 | => where { $_->isa('DateTime') }; |
197 | |
198 | coerce 'DateTime' |
199 | => from 'Str' |
200 | => via { DateTime::Format::MySQL->parse_datetime($_) }; |
201 | |
202 | has 'timestamp' => (is => 'rw', isa => 'DateTime', coerce => 1); |
203 | |
204 | This creates a custom subtype for L<DateTime> objects, then attaches |
205 | a coercion to that subtype. The C<timestamp> attribute is then told |
206 | to expect a C<DateTime> type, and to try and coerce it. When a C<Str> |
207 | type is given to the C<timestamp> accessor, it will attempt to |
208 | coerce the value into a C<DateTime> object using the code in found |
209 | in the C<via> block. |
210 | |
211 | For a more detailed and complete example of coercions, see the |
212 | L<Moose::Cookbook::Recipe5>. |
213 | |
214 | If you need to deflate your attribute, the current best practice is to |
215 | add an C<around> modifier to your accessor. Here is some example code: |
216 | |
217 | # a timestamp which stores as |
218 | # seconds from the epoch |
219 | has 'timestamp' => (is => 'rw', isa => 'Int'); |
220 | |
221 | around 'timestamp' => sub { |
222 | my $next = shift; |
223 | my ($self, $timestamp) = @_; |
224 | # assume we get a DateTime object ... |
225 | $next->($self, $timestamp->epoch); |
226 | }; |
227 | |
228 | It is also possible to do deflation using coercion, but this tends |
229 | to get quite complex and require many subtypes. An example of this |
230 | is outside the scope of this document, ask on #moose or send a mail |
231 | to the list. |
232 | |
233 | Still another option is to write a custom attribute metaclass, which |
234 | is also outside the scope of this document, but I would be happy to |
235 | explain it on #moose or the mailing list. |
236 | |
237 | =head2 Method Modfiers |
238 | |
239 | =head3 How can I affect the values in C<@_> using C<before>? |
240 | |
241 | You can't actually, C<before> only runs before the main method, |
242 | and it cannot easily affect the execution of it. What you want is |
243 | an C<around> method. |
244 | |
245 | =head3 Can I use C<before> to stop execution of a method? |
246 | |
247 | Yes, but only if you throw an exception. If this is too drastic a |
248 | measure then I suggest using C<around> instead. The C<around> method |
249 | modifier is the only modifier which can actually stop the execution |
250 | of the main method. Here is an example: |
251 | |
252 | around 'baz' => sub { |
253 | my $next = shift; |
254 | my ($self, %options) = @_; |
255 | if ($options{bar} eq 'foo') { |
256 | $next->($self, %options); |
257 | } |
258 | else { |
259 | return 'bar'; |
260 | } |
261 | }; |
262 | |
263 | By choosing not to call the C<$next> method, you can stop the |
264 | execution of the main method. |
265 | |
266 | =head2 Type Constraints |
267 | |
268 | =head3 How can I have a custom error message for a type constraint? |
269 | |
270 | Use the C<message> option when building the subtype. Like so: |
271 | |
272 | subtype 'NaturalLessThanTen' |
273 | => as 'Natural' |
274 | => where { $_ < 10 } |
275 | => message { "This number ($_) is not less than ten!" }; |
276 | |
277 | This will be called when a value fails to pass the C<NaturalLessThanTen> |
278 | constraint check. |
279 | |
734d1752 |
280 | =head3 Can I turn type constraint checking off? |
281 | |
282 | Not yet, but soon. This option will likely be coming in the next |
283 | release. |
284 | |
e67a0fca |
285 | =head1 AUTHOR |
286 | |
287 | Stevan Little E<lt>stevan@iinteractive.comE<gt> |
288 | |
289 | =head1 COPYRIGHT AND LICENSE |
290 | |
b77fdbed |
291 | Copyright 2006, 2007 by Infinity Interactive, Inc. |
e67a0fca |
292 | |
293 | L<http://www.iinteractive.com> |
294 | |
295 | This library is free software; you can redistribute it and/or modify |
296 | it under the same terms as Perl itself. |
297 | |
298 | =cut |