/
calculations.rb
674 lines (594 loc) · 23.5 KB
/
calculations.rb
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
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
# frozen_string_literal: true
require "active_support/core_ext/enumerable"
module ActiveRecord
# = Active Record \Calculations
module Calculations
class ColumnAliasTracker # :nodoc:
def initialize(connection)
@connection = connection
@aliases = Hash.new(0)
end
def alias_for(field)
aliased_name = column_alias_for(field)
if @aliases[aliased_name] == 0
@aliases[aliased_name] = 1
aliased_name
else
# Update the count
count = @aliases[aliased_name] += 1
"#{truncate(aliased_name)}_#{count}"
end
end
private
# Converts the given field to the value that the database adapter returns as
# a usable column name:
#
# column_alias_for("users.id") # => "users_id"
# column_alias_for("sum(id)") # => "sum_id"
# column_alias_for("count(distinct users.id)") # => "count_distinct_users_id"
# column_alias_for("count(*)") # => "count_all"
def column_alias_for(field)
column_alias = +field
column_alias.gsub!(/\*/, "all")
column_alias.gsub!(/\W+/, " ")
column_alias.strip!
column_alias.gsub!(/ +/, "_")
@connection.table_alias_for(column_alias)
end
def truncate(name)
name.slice(0, @connection.table_alias_length - 2)
end
end
# Count the records.
#
# Person.count
# # => the total count of all people
#
# Person.count(:age)
# # => returns the total count of all people whose age is present in database
#
# Person.count(:all)
# # => performs a COUNT(*) (:all is an alias for '*')
#
# Person.distinct.count(:age)
# # => counts the number of different age values
#
# If #count is used with {Relation#group}[rdoc-ref:QueryMethods#group],
# it returns a Hash whose keys represent the aggregated column,
# and the values are the respective amounts:
#
# Person.group(:city).count
# # => { 'Rome' => 5, 'Paris' => 3 }
#
# If #count is used with {Relation#group}[rdoc-ref:QueryMethods#group] for multiple columns, it returns a Hash whose
# keys are an array containing the individual values of each column and the value
# of each key would be the #count.
#
# Article.group(:status, :category).count
# # => {["draft", "business"]=>10, ["draft", "technology"]=>4, ["published", "technology"]=>2}
#
# If #count is used with {Relation#select}[rdoc-ref:QueryMethods#select], it will count the selected columns:
#
# Person.select(:age).count
# # => counts the number of different age values
#
# Note: not all valid {Relation#select}[rdoc-ref:QueryMethods#select] expressions are valid #count expressions. The specifics differ
# between databases. In invalid cases, an error from the database is thrown.
#
# When given a block, loads all records in the relation, if the relation
# hasn't been loaded yet. Calls the block with each record in the relation.
# Returns the number of records for which the block returns a truthy value.
#
# Person.count { |person| person.age > 21 }
# # => counts the number of people older that 21
#
# Note: If there are a lot of records in the relation, loading all records
# could result in performance issues.
def count(column_name = nil)
if block_given?
unless column_name.nil?
raise ArgumentError, "Column name argument is not supported when a block is passed."
end
super()
else
calculate(:count, column_name)
end
end
# Same as #count, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_count(column_name = nil)
async.count(column_name)
end
# Calculates the average value on a given column. Returns +nil+ if there's
# no row. See #calculate for examples with options.
#
# Person.average(:age) # => 35.8
def average(column_name)
calculate(:average, column_name)
end
# Same as #average, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_average(column_name)
async.average(column_name)
end
# Calculates the minimum value on a given column. The value is returned
# with the same data type of the column, or +nil+ if there's no row. See
# #calculate for examples with options.
#
# Person.minimum(:age) # => 7
def minimum(column_name)
calculate(:minimum, column_name)
end
# Same as #minimum, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_minimum(column_name)
async.minimum(column_name)
end
# Calculates the maximum value on a given column. The value is returned
# with the same data type of the column, or +nil+ if there's no row. See
# #calculate for examples with options.
#
# Person.maximum(:age) # => 93
def maximum(column_name)
calculate(:maximum, column_name)
end
# Same as #maximum, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_maximum(column_name)
async.maximum(column_name)
end
# Calculates the sum of values on a given column. The value is returned
# with the same data type of the column, +0+ if there's no row. See
# #calculate for examples with options.
#
# Person.sum(:age) # => 4562
#
# When given a block, loads all records in the relation, if the relation
# hasn't been loaded yet. Calls the block with each record in the relation.
# Returns the sum of +initial_value_or_column+ and the block return
# values:
#
# Person.sum { |person| person.age } # => 4562
# Person.sum(1000) { |person| person.age } # => 5562
#
# Note: If there are a lot of records in the relation, loading all records
# could result in performance issues.
def sum(initial_value_or_column = 0, &block)
if block_given?
map(&block).sum(initial_value_or_column)
else
calculate(:sum, initial_value_or_column)
end
end
# Same as #sum, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_sum(identity_or_column = nil)
async.sum(identity_or_column)
end
# This calculates aggregate values in the given column. Methods for #count, #sum, #average,
# #minimum, and #maximum have been added as shortcuts.
#
# Person.calculate(:count, :all) # The same as Person.count
# Person.average(:age) # SELECT AVG(age) FROM people...
#
# # Selects the minimum age for any family without any minors
# Person.group(:last_name).having("min(age) > 17").minimum(:age)
#
# Person.sum("2 * age")
#
# There are two basic forms of output:
#
# * Single aggregate value: The single value is type cast to Integer for COUNT, Float
# for AVG, and the given column's type for everything else.
#
# * Grouped values: This returns an ordered hash of the values and groups them. It
# takes either a column name, or the name of a belongs_to association.
#
# values = Person.group('last_name').maximum(:age)
# puts values["Drake"]
# # => 43
#
# drake = Family.find_by(last_name: 'Drake')
# values = Person.group(:family).maximum(:age) # Person belongs_to :family
# puts values[drake]
# # => 43
#
# values.each do |family, max_age|
# ...
# end
def calculate(operation, column_name)
operation = operation.to_s.downcase
if @none
case operation
when "count", "sum"
result = group_values.any? ? Hash.new : 0
return @async ? Promise::Complete.new(result) : result
when "average", "minimum", "maximum"
result = group_values.any? ? Hash.new : nil
return @async ? Promise::Complete.new(result) : result
end
end
if has_include?(column_name)
relation = apply_join_dependency
if operation == "count"
unless distinct_value || distinct_select?(column_name || select_for_count)
relation.distinct!
relation.select_values = Array(klass.primary_key || table[Arel.star])
end
# PostgreSQL: ORDER BY expressions must appear in SELECT list when using DISTINCT
relation.order_values = [] if group_values.empty?
end
relation.calculate(operation, column_name)
else
perform_calculation(operation, column_name)
end
end
# Use #pluck as a shortcut to select one or more attributes without
# loading an entire record object per row.
#
# Person.pluck(:name)
#
# instead of
#
# Person.all.map(&:name)
#
# Pluck returns an Array of attribute values type-casted to match
# the plucked column names, if they can be deduced. Plucking an SQL fragment
# returns String values by default.
#
# Person.pluck(:name)
# # SELECT people.name FROM people
# # => ['David', 'Jeremy', 'Jose']
#
# Person.pluck(:id, :name)
# # SELECT people.id, people.name FROM people
# # => [[1, 'David'], [2, 'Jeremy'], [3, 'Jose']]
#
# Person.distinct.pluck(:role)
# # SELECT DISTINCT role FROM people
# # => ['admin', 'member', 'guest']
#
# Person.where(age: 21).limit(5).pluck(:id)
# # SELECT people.id FROM people WHERE people.age = 21 LIMIT 5
# # => [2, 3]
#
# Comment.joins(:person).pluck(:id, person: [:id])
# # SELECT comments.id, people.id FROM comments INNER JOIN people on comments.person_id = people.id
# # => [[1, 2], [2, 2]]
#
# Person.pluck(Arel.sql('DATEDIFF(updated_at, created_at)'))
# # SELECT DATEDIFF(updated_at, created_at) FROM people
# # => ['0', '27761', '173']
#
# See also #ids.
def pluck(*column_names)
if @none
if @async
return Promise::Complete.new([])
else
return []
end
end
if loaded? && all_attributes?(column_names)
result = records.pluck(*column_names)
if @async
return Promise::Complete.new(result)
else
return result
end
end
if has_include?(column_names.first)
relation = apply_join_dependency
relation.pluck(*column_names)
else
klass.disallow_raw_sql!(flattened_args(column_names))
columns = arel_columns(column_names)
relation = spawn
relation.select_values = columns
result = skip_query_cache_if_necessary do
if where_clause.contradiction?
ActiveRecord::Result.empty(async: @async)
else
klass.with_connection do |c|
c.select_all(relation.arel, "#{klass.name} Pluck", async: @async)
end
end
end
result.then do |result|
type_cast_pluck_values(result, columns)
end
end
end
# Same as #pluck, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_pluck(*column_names)
async.pluck(*column_names)
end
# Pick the value(s) from the named column(s) in the current relation.
# This is short-hand for <tt>relation.limit(1).pluck(*column_names).first</tt>, and is primarily useful
# when you have a relation that's already narrowed down to a single row.
#
# Just like #pluck, #pick will only load the actual value, not the entire record object, so it's also
# more efficient. The value is, again like with pluck, typecast by the column type.
#
# Person.where(id: 1).pick(:name)
# # SELECT people.name FROM people WHERE id = 1 LIMIT 1
# # => 'David'
#
# Person.where(id: 1).pick(:name, :email_address)
# # SELECT people.name, people.email_address FROM people WHERE id = 1 LIMIT 1
# # => [ 'David', 'david@loudthinking.com' ]
def pick(*column_names)
if loaded? && all_attributes?(column_names)
result = records.pick(*column_names)
return @async ? Promise::Complete.new(result) : result
end
limit(1).pluck(*column_names).then(&:first)
end
# Same as #pick, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_pick(*column_names)
async.pick(*column_names)
end
# Returns the base model's ID's for the relation using the table's primary key
#
# Person.ids # SELECT people.id FROM people
# Person.joins(:company).ids # SELECT people.id FROM people INNER JOIN companies ON companies.id = people.company_id
def ids
primary_key_array = Array(primary_key)
if loaded?
result = records.map do |record|
if primary_key_array.one?
record._read_attribute(primary_key_array.first)
else
primary_key_array.map { |column| record._read_attribute(column) }
end
end
return @async ? Promise::Complete.new(result) : result
end
if has_include?(primary_key)
relation = apply_join_dependency.group(*primary_key_array)
return relation.ids
end
columns = arel_columns(primary_key_array)
relation = spawn
relation.select_values = columns
result = if relation.where_clause.contradiction?
ActiveRecord::Result.empty
else
skip_query_cache_if_necessary do
klass.with_connection do |c|
c.select_all(relation, "#{klass.name} Ids", async: @async)
end
end
end
result.then { |result| type_cast_pluck_values(result, columns) }
end
# Same as #ids, but performs the query asynchronously and returns an
# ActiveRecord::Promise.
def async_ids
async.ids
end
private
def all_attributes?(column_names)
(column_names.map(&:to_s) - @klass.attribute_names - @klass.attribute_aliases.keys).empty?
end
def has_include?(column_name)
eager_loading? || (includes_values.present? && column_name && column_name != :all)
end
def perform_calculation(operation, column_name)
operation = operation.to_s.downcase
# If #count is used with #distinct (i.e. `relation.distinct.count`) it is
# considered distinct.
distinct = distinct_value
if operation == "count"
column_name ||= select_for_count
if column_name == :all
if !distinct
distinct = distinct_select?(select_for_count) if group_values.empty?
elsif group_values.any? || select_values.empty? && order_values.empty?
column_name = primary_key
end
elsif distinct_select?(column_name)
distinct = nil
end
end
if group_values.any?
execute_grouped_calculation(operation, column_name, distinct)
else
execute_simple_calculation(operation, column_name, distinct)
end
end
def distinct_select?(column_name)
column_name.is_a?(::String) && /\bDISTINCT[\s(]/i.match?(column_name)
end
def aggregate_column(column_name)
return column_name if Arel::Expressions === column_name
arel_column(column_name.to_s) do |name|
column_name == :all ? Arel.sql("*", retryable: true) : Arel.sql(name)
end
end
def operation_over_aggregate_column(column, operation, distinct)
operation == "count" ? column.count(distinct) : column.public_send(operation)
end
def execute_simple_calculation(operation, column_name, distinct) # :nodoc:
if build_count_subquery?(operation, column_name, distinct)
# Shortcut when limit is zero.
return 0 if limit_value == 0
relation = self
query_builder = build_count_subquery(spawn, column_name, distinct)
else
# PostgreSQL doesn't like ORDER BY when there are no GROUP BY
relation = unscope(:order).distinct!(false)
column = aggregate_column(column_name)
select_value = operation_over_aggregate_column(column, operation, distinct)
select_value.distinct = true if operation == "sum" && distinct
relation.select_values = [select_value]
query_builder = relation.arel
end
query_result = if relation.where_clause.contradiction?
ActiveRecord::Result.empty
else
skip_query_cache_if_necessary do
@klass.with_connection do |c|
c.select_all(query_builder, "#{@klass.name} #{operation.capitalize}", async: @async)
end
end
end
query_result.then do |result|
if operation != "count"
type = column.try(:type_caster) ||
lookup_cast_type_from_join_dependencies(column_name.to_s) || Type.default_value
type = type.subtype if Enum::EnumType === type
end
type_cast_calculated_value(result.cast_values.first, operation, type)
end
end
def execute_grouped_calculation(operation, column_name, distinct) # :nodoc:
group_fields = group_values
group_fields = group_fields.uniq if group_fields.size > 1
if group_fields.size == 1 && group_fields.first.respond_to?(:to_sym)
association = klass._reflect_on_association(group_fields.first)
associated = association && association.belongs_to? # only count belongs_to associations
group_fields = Array(association.foreign_key) if associated
end
group_fields = arel_columns(group_fields)
@klass.with_connection do |connection|
column_alias_tracker = ColumnAliasTracker.new(connection)
group_aliases = group_fields.map { |field|
field = connection.visitor.compile(field) if Arel.arel_node?(field)
column_alias_tracker.alias_for(field.to_s.downcase)
}
group_columns = group_aliases.zip(group_fields)
column = aggregate_column(column_name)
column_alias = column_alias_tracker.alias_for("#{operation} #{column_name.to_s.downcase}")
select_value = operation_over_aggregate_column(column, operation, distinct)
select_value.as(adapter_class.quote_column_name(column_alias))
select_values = [select_value]
select_values += self.select_values unless having_clause.empty?
select_values.concat group_columns.map { |aliaz, field|
aliaz = adapter_class.quote_column_name(aliaz)
if field.respond_to?(:as)
field.as(aliaz)
else
"#{field} AS #{aliaz}"
end
}
relation = except(:group).distinct!(false)
relation.group_values = group_fields
relation.select_values = select_values
result = skip_query_cache_if_necessary do
connection.select_all(relation.arel, "#{@klass.name} #{operation.capitalize}", async: @async)
end
result.then do |calculated_data|
if association
key_ids = calculated_data.collect { |row| row[group_aliases.first] }
key_records = association.klass.base_class.where(association.klass.base_class.primary_key => key_ids)
key_records = key_records.index_by(&:id)
end
key_types = group_columns.each_with_object({}) do |(aliaz, col_name), types|
types[aliaz] = col_name.try(:type_caster) ||
type_for(col_name) do
calculated_data.column_types.fetch(aliaz, Type.default_value)
end
end
hash_rows = calculated_data.cast_values(key_types).map! do |row|
calculated_data.columns.each_with_object({}).with_index do |(col_name, hash), i|
hash[col_name] = row[i]
end
end
if operation != "count"
type = column.try(:type_caster) ||
lookup_cast_type_from_join_dependencies(column_name.to_s) || Type.default_value
type = type.subtype if Enum::EnumType === type
end
hash_rows.each_with_object({}) do |row, result|
key = group_aliases.map { |aliaz| row[aliaz] }
key = key.first if key.size == 1
key = key_records[key] if associated
result[key] = type_cast_calculated_value(row[column_alias], operation, type)
end
end
end
end
def type_for(field, &block)
field_name = field.respond_to?(:name) ? field.name.to_s : field.to_s.split(".").last
@klass.type_for_attribute(field_name, &block)
end
def lookup_cast_type_from_join_dependencies(name, join_dependencies = build_join_dependencies)
each_join_dependencies(join_dependencies) do |join|
type = join.base_klass.attribute_types.fetch(name, nil)
return type if type
end
nil
end
def type_cast_pluck_values(result, columns)
cast_types = if result.columns.size != columns.size
klass.attribute_types
else
join_dependencies = nil
columns.map.with_index do |column, i|
column.try(:type_caster) ||
klass.attribute_types.fetch(name = result.columns[i]) do
join_dependencies ||= build_join_dependencies
lookup_cast_type_from_join_dependencies(name, join_dependencies) ||
result.column_types[name] || Type.default_value
end
end
end
result.cast_values(cast_types)
end
def type_cast_calculated_value(value, operation, type)
case operation
when "count"
value.to_i
when "sum"
type.deserialize(value || 0)
when "average"
case type.type
when :integer, :decimal
value&.to_d
else
type.deserialize(value)
end
else # "minimum", "maximum"
type.deserialize(value)
end
end
def select_for_count
if select_values.present?
return select_values.first if select_values.one?
select_values.map do |field|
column = arel_column(field.to_s) do |attr_name|
Arel.sql(attr_name)
end
if column.is_a?(Arel::Nodes::SqlLiteral)
column
else
"#{adapter_class.quote_table_name(column.relation.name)}.#{adapter_class.quote_column_name(column.name)}"
end
end.join(", ")
else
:all
end
end
def build_count_subquery?(operation, column_name, distinct)
# SQLite and older MySQL does not support `COUNT DISTINCT` with `*` or
# multiple columns, so we need to use subquery for this.
operation == "count" &&
(((column_name == :all || select_values.many?) && distinct) || has_limit_or_offset?)
end
def build_count_subquery(relation, column_name, distinct)
if column_name == :all
column_alias = Arel.star
relation.select_values = [ Arel.sql(FinderMethods::ONE_AS_ONE) ] unless distinct
else
column_alias = Arel.sql("count_column")
relation.select_values = [ aggregate_column(column_name).as(column_alias) ]
end
subquery_alias = Arel.sql("subquery_for_count", retryable: true)
select_value = operation_over_aggregate_column(column_alias, "count", false)
relation.build_subquery(subquery_alias, select_value)
end
end
end