aboutsummaryrefslogtreecommitdiffstats
path: root/activerecord/lib/active_record/callbacks.rb
blob: eb44887e1804d0f5d2f07ff3b830a233f44ef969 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
module ActiveRecord
  # = Active Record \Callbacks
  #
  # \Callbacks are hooks into the life cycle of an Active Record object that allow you to trigger logic
  # before or after an alteration of the object state. This can be used to make sure that associated and
  # dependent objects are deleted when {ActiveRecord::Base#destroy}[rdoc-ref:Persistence#destroy] is called (by overwriting +before_destroy+) or
  # to massage attributes before they're validated (by overwriting +before_validation+).
  # As an example of the callbacks initiated, consider the {ActiveRecord::Base#save}[rdoc-ref:Persistence#save] call for a new record:
  #
  # * (-) <tt>save</tt>
  # * (-) <tt>valid</tt>
  # * (1) <tt>before_validation</tt>
  # * (-) <tt>validate</tt>
  # * (2) <tt>after_validation</tt>
  # * (3) <tt>before_save</tt>
  # * (4) <tt>before_create</tt>
  # * (-) <tt>create</tt>
  # * (5) <tt>after_create</tt>
  # * (6) <tt>after_save</tt>
  # * (7) <tt>after_commit</tt>
  #
  # Also, an <tt>after_rollback</tt> callback can be configured to be triggered whenever a rollback is issued.
  # Check out ActiveRecord::Transactions for more details about <tt>after_commit</tt> and
  # <tt>after_rollback</tt>.
  #
  # Additionally, an <tt>after_touch</tt> callback is triggered whenever an
  # object is touched.
  #
  # Lastly an <tt>after_find</tt> and <tt>after_initialize</tt> callback is triggered for each object that
  # is found and instantiated by a finder, with <tt>after_initialize</tt> being triggered after new objects
  # are instantiated as well.
  #
  # There are nineteen callbacks in total, which give you immense power to react and prepare for each state in the
  # Active Record life cycle. The sequence for calling {ActiveRecord::Base#save}[rdoc-ref:Persistence#save] for an existing record is similar,
  # except that each <tt>_create</tt> callback is replaced by the corresponding <tt>_update</tt> callback.
  #
  # Examples:
  #   class CreditCard < ActiveRecord::Base
  #     # Strip everything but digits, so the user can specify "555 234 34" or
  #     # "5552-3434" and both will mean "55523434"
  #     before_validation(on: :create) do
  #       self.number = number.gsub(/[^0-9]/, "") if attribute_present?("number")
  #     end
  #   end
  #
  #   class Subscription < ActiveRecord::Base
  #     before_create :record_signup
  #
  #     private
  #       def record_signup
  #         self.signed_up_on = Date.today
  #       end
  #   end
  #
  #   class Firm < ActiveRecord::Base
  #     # Disables access to the system, for associated clients and people when the firm is destroyed
  #     before_destroy { |record| Person.where(firm_id: record.id).update_all(access: 'disabled')   }
  #     before_destroy { |record| Client.where(client_of: record.id).update_all(access: 'disabled') }
  #   end
  #
  # == Inheritable callback queues
  #
  # Besides the overwritable callback methods, it's also possible to register callbacks through the
  # use of the callback macros. Their main advantage is that the macros add behavior into a callback
  # queue that is kept intact down through an inheritance hierarchy.
  #
  #   class Topic < ActiveRecord::Base
  #     before_destroy :destroy_author
  #   end
  #
  #   class Reply < Topic
  #     before_destroy :destroy_readers
  #   end
  #
  # Now, when <tt>Topic#destroy</tt> is run only +destroy_author+ is called. When <tt>Reply#destroy</tt> is
  # run, both +destroy_author+ and +destroy_readers+ are called. Contrast this to the following situation
  # where the +before_destroy+ method is overridden:
  #
  #   class Topic < ActiveRecord::Base
  #     def before_destroy() destroy_author end
  #   end
  #
  #   class Reply < Topic
  #     def before_destroy() destroy_readers end
  #   end
  #
  # In that case, <tt>Reply#destroy</tt> would only run +destroy_readers+ and _not_ +destroy_author+.
  # So, use the callback macros when you want to ensure that a certain callback is called for the entire
  # hierarchy, and use the regular overwritable methods when you want to leave it up to each descendant
  # to decide whether they want to call +super+ and trigger the inherited callbacks.
  #
  # *IMPORTANT:* In order for inheritance to work for the callback queues, you must specify the
  # callbacks before specifying the associations. Otherwise, you might trigger the loading of a
  # child before the parent has registered the callbacks and they won't be inherited.
  #
  # == Types of callbacks
  #
  # There are four types of callbacks accepted by the callback macros: Method references (symbol), callback objects,
  # inline methods (using a proc), and inline eval methods (using a string). Method references and callback objects
  # are the recommended approaches, inline methods using a proc are sometimes appropriate (such as for
  # creating mix-ins), and inline eval methods are deprecated.
  #
  # The method reference callbacks work by specifying a protected or private method available in the object, like this:
  #
  #   class Topic < ActiveRecord::Base
  #     before_destroy :delete_parents
  #
  #     private
  #       def delete_parents
  #         self.class.delete_all "parent_id = #{id}"
  #       end
  #   end
  #
  # The callback objects have methods named after the callback called with the record as the only parameter, such as:
  #
  #   class BankAccount < ActiveRecord::Base
  #     before_save      EncryptionWrapper.new
  #     after_save       EncryptionWrapper.new
  #     after_initialize EncryptionWrapper.new
  #   end
  #
  #   class EncryptionWrapper
  #     def before_save(record)
  #       record.credit_card_number = encrypt(record.credit_card_number)
  #     end
  #
  #     def after_save(record)
  #       record.credit_card_number = decrypt(record.credit_card_number)
  #     end
  #
  #     alias_method :after_initialize, :after_save
  #
  #     private
  #       def encrypt(value)
  #         # Secrecy is committed
  #       end
  #
  #       def decrypt(value)
  #         # Secrecy is unveiled
  #       end
  #   end
  #
  # So you specify the object you want messaged on a given callback. When that callback is triggered, the object has
  # a method by the name of the callback messaged. You can make these callbacks more flexible by passing in other
  # initialization data such as the name of the attribute to work with:
  #
  #   class BankAccount < ActiveRecord::Base
  #     before_save      EncryptionWrapper.new("credit_card_number")
  #     after_save       EncryptionWrapper.new("credit_card_number")
  #     after_initialize EncryptionWrapper.new("credit_card_number")
  #   end
  #
  #   class EncryptionWrapper
  #     def initialize(attribute)
  #       @attribute = attribute
  #     end
  #
  #     def before_save(record)
  #       record.send("#{@attribute}=", encrypt(record.send("#{@attribute}")))
  #     end
  #
  #     def after_save(record)
  #       record.send("#{@attribute}=", decrypt(record.send("#{@attribute}")))
  #     end
  #
  #     alias_method :after_initialize, :after_save
  #
  #     private
  #       def encrypt(value)
  #         # Secrecy is committed
  #       end
  #
  #       def decrypt(value)
  #         # Secrecy is unveiled
  #       end
  #   end
  #
  # == <tt>before_validation*</tt> returning statements
  #
  # If the +before_validation+ callback throws +:abort+, the process will be
  # aborted and {ActiveRecord::Base#save}[rdoc-ref:Persistence#save] will return +false+.
  # If {ActiveRecord::Base#save!}[rdoc-ref:Persistence#save!] is called it will raise an ActiveRecord::RecordInvalid exception.
  # Nothing will be appended to the errors object.
  #
  # == Canceling callbacks
  #
  # If a <tt>before_*</tt> callback throws +:abort+, all the later callbacks and
  # the associated action are cancelled.
  # Callbacks are generally run in the order they are defined, with the exception of callbacks defined as
  # methods on the model, which are called last.
  #
  # == Ordering callbacks
  #
  # Sometimes the code needs that the callbacks execute in a specific order. For example, a +before_destroy+
  # callback (+log_children+ in this case) should be executed before the children get destroyed by the
  # <tt>dependent: :destroy</tt> option.
  #
  # Let's look at the code below:
  #
  #   class Topic < ActiveRecord::Base
  #     has_many :children, dependent: :destroy
  #
  #     before_destroy :log_children
  #
  #     private
  #       def log_children
  #         # Child processing
  #       end
  #   end
  #
  # In this case, the problem is that when the +before_destroy+ callback is executed, the children are not available
  # because the {ActiveRecord::Base#destroy}[rdoc-ref:Persistence#destroy] callback gets executed first.
  # You can use the +prepend+ option on the +before_destroy+ callback to avoid this.
  #
  #   class Topic < ActiveRecord::Base
  #     has_many :children, dependent: :destroy
  #
  #     before_destroy :log_children, prepend: true
  #
  #     private
  #       def log_children
  #         # Child processing
  #       end
  #   end
  #
  # This way, the +before_destroy+ gets executed before the <tt>dependent: :destroy</tt> is called, and the data is still available.
  #
  # Also, there are cases when you want several callbacks of the same type to
  # be executed in order.
  #
  # For example:
  #
  #   class Topic
  #     has_many :children
  #
  #     after_save :log_children
  #     after_save :do_something_else
  #
  #     private
  #
  #     def log_chidren
  #       # Child processing
  #     end
  #
  #     def do_something_else
  #       # Something else
  #     end
  #   end
  #
  # In this case the +log_children+ gets executed before +do_something_else+.
  # The same applies to all non-transactional callbacks.
  #
  # In case there are multiple transactional callbacks as seen below, the order
  # is reversed.
  #
  # For example:
  #
  #   class Topic
  #     has_many :children
  #
  #     after_commit :log_children
  #     after_commit :do_something_else
  #
  #     private
  #
  #     def log_chidren
  #       # Child processing
  #     end
  #
  #     def do_something_else
  #       # Something else
  #     end
  #   end
  #
  # In this case the +do_something_else+ gets executed before +log_children+.
  #
  # == \Transactions
  #
  # The entire callback chain of a {#save}[rdoc-ref:Persistence#save], {#save!}[rdoc-ref:Persistence#save!],
  # or {#destroy}[rdoc-ref:Persistence#destroy] call runs within a transaction. That includes <tt>after_*</tt> hooks.
  # If everything goes fine a COMMIT is executed once the chain has been completed.
  #
  # If a <tt>before_*</tt> callback cancels the action a ROLLBACK is issued. You
  # can also trigger a ROLLBACK raising an exception in any of the callbacks,
  # including <tt>after_*</tt> hooks. Note, however, that in that case the client
  # needs to be aware of it because an ordinary {#save}[rdoc-ref:Persistence#save] will raise such exception
  # instead of quietly returning +false+.
  #
  # == Debugging callbacks
  #
  # The callback chain is accessible via the <tt>_*_callbacks</tt> method on an object. Active Model \Callbacks support
  # <tt>:before</tt>, <tt>:after</tt> and <tt>:around</tt> as values for the <tt>kind</tt> property. The <tt>kind</tt> property
  # defines what part of the chain the callback runs in.
  #
  # To find all callbacks in the before_save callback chain:
  #
  #   Topic._save_callbacks.select { |cb| cb.kind.eql?(:before) }
  #
  # Returns an array of callback objects that form the before_save chain.
  #
  # To further check if the before_save chain contains a proc defined as <tt>rest_when_dead</tt> use the <tt>filter</tt> property of the callback object:
  #
  #   Topic._save_callbacks.select { |cb| cb.kind.eql?(:before) }.collect(&:filter).include?(:rest_when_dead)
  #
  # Returns true or false depending on whether the proc is contained in the before_save callback chain on a Topic model.
  #
  module Callbacks
    extend ActiveSupport::Concern

    CALLBACKS = [
      :after_initialize, :after_find, :after_touch, :before_validation, :after_validation,
      :before_save, :around_save, :after_save, :before_create, :around_create,
      :after_create, :before_update, :around_update, :after_update,
      :before_destroy, :around_destroy, :after_destroy, :after_commit, :after_rollback
    ]

    def destroy #:nodoc:
      @_destroy_callback_already_called ||= false
      return if @_destroy_callback_already_called
      @_destroy_callback_already_called = true
      _run_destroy_callbacks { super }
    rescue RecordNotDestroyed => e
      @_association_destroy_exception = e
      false
    ensure
      @_destroy_callback_already_called = false
    end

    def touch(*) #:nodoc:
      _run_touch_callbacks { super }
    end

  private

    def create_or_update(*)
      _run_save_callbacks { super }
    end

    def _create_record
      _run_create_callbacks { super }
    end

    def _update_record(*)
      _run_update_callbacks { super }
    end
  end
end