From 25cee1f0373aa3b1d893413a959375480e0ac684 Mon Sep 17 00:00:00 2001 From: Sina Siadat Date: Sat, 18 Jul 2015 20:44:13 +0430 Subject: Add ActiveRecord::Relation#in_batches `in_batches` yields Relation objects if a block is given, otherwise it returns an instance of `BatchEnumerator`. The existing `find_each` and `find_in_batches` methods work with batches of records. The new API allows working with relation batches as well. Examples: Person.in_batches.each_record(&:party_all_night!) Person.in_batches.update_all(awesome: true) Person.in_batches.delete_all Person.in_batches.map do |relation| relation.delete_all sleep 10 # Throttles the delete queries end --- activerecord/lib/active_record/querying.rb | 2 +- activerecord/lib/active_record/relation.rb | 7 ++ activerecord/lib/active_record/relation/batches.rb | 98 ++++++++++++++++++++-- .../relation/batches/batch_enumerator.rb | 67 +++++++++++++++ 4 files changed, 164 insertions(+), 10 deletions(-) create mode 100644 activerecord/lib/active_record/relation/batches/batch_enumerator.rb (limited to 'activerecord/lib') diff --git a/activerecord/lib/active_record/querying.rb b/activerecord/lib/active_record/querying.rb index 4e597590e9..87a1988f2f 100644 --- a/activerecord/lib/active_record/querying.rb +++ b/activerecord/lib/active_record/querying.rb @@ -6,7 +6,7 @@ module ActiveRecord delegate :find_or_create_by, :find_or_create_by!, :find_or_initialize_by, to: :all delegate :find_by, :find_by!, to: :all delegate :destroy, :destroy_all, :delete, :delete_all, :update, :update_all, to: :all - delegate :find_each, :find_in_batches, to: :all + delegate :find_each, :find_in_batches, :in_batches, to: :all delegate :select, :group, :order, :except, :reorder, :limit, :offset, :joins, :or, :where, :rewhere, :preload, :eager_load, :includes, :from, :lock, :readonly, :having, :create_with, :uniq, :distinct, :references, :none, :unscope, to: :all diff --git a/activerecord/lib/active_record/relation.rb b/activerecord/lib/active_record/relation.rb index 3ed04dee3b..e47b7b1ed9 100644 --- a/activerecord/lib/active_record/relation.rb +++ b/activerecord/lib/active_record/relation.rb @@ -667,6 +667,13 @@ module ActiveRecord "#<#{self.class.name} [#{entries.join(', ')}]>" end + protected + + def load_records(records) + @records = records + @loaded = true + end + private def exec_queries diff --git a/activerecord/lib/active_record/relation/batches.rb b/activerecord/lib/active_record/relation/batches.rb index e07580a563..beb8fa511c 100644 --- a/activerecord/lib/active_record/relation/batches.rb +++ b/activerecord/lib/active_record/relation/batches.rb @@ -1,3 +1,5 @@ +require "active_record/relation/batches/batch_enumerator" + module ActiveRecord module Batches # Looping through a collection of records from the database @@ -122,24 +124,102 @@ module ActiveRecord end end + in_batches(of: batch_size, begin_at: begin_at, end_at: end_at, load: true) do |batch| + yield batch.to_a + end + end + + # Yields ActiveRecord::Relation objects to work with a batch of records. + # + # Person.where("age > 21").in_batches do |relation| + # relation.delete_all + # sleep(10) # Throttle the delete queries + # end + # + # If you do not provide a block to #in_batches, it will return a + # BatchEnumerator which is enumerable. + # + # Person.in_batches.with_index do |relation, batch_index| + # puts "Processing relation ##{batch_index}" + # relation.each { |relation| relation.delete_all } + # end + # + # Examples of calling methods on the returned BatchEnumerator object: + # + # Person.in_batches.delete_all + # Person.in_batches.update_all(awesome: true) + # Person.in_batches.each_record(&:party_all_night!) + # + # ==== Options + # * :of - Specifies the size of the batch. Default to 1000. + # * :load - Specifies if the relation should be loaded. Default to false. + # * :begin_at - Specifies the primary key value to start from, inclusive of the value. + # * :end_at - Specifies the primary key value to end at, inclusive of the value. + # + # This is especially useful if you want to work with the + # ActiveRecord::Relation object instead of the array of records, or if + # you want multiple workers dealing with the same processing queue. You can + # make worker 1 handle all the records between id 0 and 10,000 and worker 2 + # handle from 10,000 and beyond (by setting the +:begin_at+ and +:end_at+ + # option on each worker). + # + # # Let's process the next 2000 records + # Person.in_batches(of: 2000, begin_at: 2000).update_all(awesome: true) + # + # An example of calling where query method on the relation: + # + # Person.in_batches.each do |relation| + # relation.update_all('age = age + 1') + # relation.where('age > 21').update_all(should_party: true) + # relation.where('age <= 21').delete_all + # end + # + # NOTE: If you are going to iterate through each record, you should call + # #each_record on the yielded BatchEnumerator: + # + # Person.in_batches.each_record(&:party_all_night!) + # + # NOTE: It's not possible to set the order. That is automatically set to + # ascending on the primary key ("id ASC") to make the batch ordering + # consistent. Therefore the primary key must be orderable, e.g an integer + # or a string. + # + # NOTE: You can't set the limit either, that's used to control the batch + # sizes. + def in_batches(of: 1000, begin_at: nil, end_at: nil, load: false) + relation = self + unless block_given? + return BatchEnumerator.new(of: of, begin_at: begin_at, end_at: end_at, relation: self) + end + if logger && (arel.orders.present? || arel.taken.present?) logger.warn("Scoped order and limit are ignored, it's forced to be batch order and batch size") end - relation = relation.reorder(batch_order).limit(batch_size) + relation = relation.reorder(batch_order).limit(of) relation = apply_limits(relation, begin_at, end_at) - records = relation.to_a + batch_relation = relation + + loop do + if load + records = batch_relation.to_a + ids = records.map(&:id) + yielded_relation = self.where(primary_key => ids) + yielded_relation.load_records(records) + else + ids = batch_relation.pluck(primary_key) + yielded_relation = self.where(primary_key => ids) + end - while records.any? - records_size = records.size - primary_key_offset = records.last.id - raise "Primary key not included in the custom select clause" unless primary_key_offset + break if ids.empty? - yield records + primary_key_offset = ids.last + raise ArgumentError.new("Primary key not included in the custom select clause") unless primary_key_offset - break if records_size < batch_size + yield yielded_relation - records = relation.where(table[primary_key].gt(primary_key_offset)).to_a + break if ids.length < of + batch_relation = relation.where(table[primary_key].gt(primary_key_offset)) end end diff --git a/activerecord/lib/active_record/relation/batches/batch_enumerator.rb b/activerecord/lib/active_record/relation/batches/batch_enumerator.rb new file mode 100644 index 0000000000..153aae9584 --- /dev/null +++ b/activerecord/lib/active_record/relation/batches/batch_enumerator.rb @@ -0,0 +1,67 @@ +module ActiveRecord + module Batches + class BatchEnumerator + include Enumerable + + def initialize(of: 1000, begin_at: nil, end_at: nil, relation:) #:nodoc: + @of = of + @relation = relation + @begin_at = begin_at + @end_at = end_at + end + + # Looping through a collection of records from the database (using the + # +all+ method, for example) is very inefficient since it will try to + # instantiate all the objects at once. + # + # In that case, batch processing methods allow you to work with the + # records in batches, thereby greatly reducing memory consumption. + # + # Person.in_batches.each_record do |person| + # person.do_awesome_stuff + # end + # + # Person.where("age > 21").in_batches(of: 10).each_record do |person| + # person.party_all_night! + # end + # + # If you do not provide a block to #each_record, it will return an Enumerator + # for chaining with other methods: + # + # Person.in_batches.each_record.with_index do |person, index| + # person.award_trophy(index + 1) + # end + def each_record + return to_enum(:each_record) unless block_given? + + @relation.to_enum(:in_batches, of: @of, begin_at: @begin_at, end_at: @end_at, load: true).each do |relation| + relation.to_a.each { |record| yield record } + end + end + + # Delegates #delete_all, #update_all, #destroy_all methods to each batch. + # + # People.in_batches.delete_all + # People.in_batches.destroy_all('age < 10') + # People.in_batches.update_all('age = age + 1') + [:delete_all, :update_all, :destroy_all].each do |method| + define_method(method) do |*args, &block| + @relation.to_enum(:in_batches, of: @of, begin_at: @begin_at, end_at: @end_at, load: false).each do |relation| + relation.send(method, *args, &block) + end + end + end + + # Yields an ActiveRecord::Relation object for each batch of records. + # + # Person.in_batches.each do |relation| + # relation.update_all(awesome: true) + # end + def each + enum = @relation.to_enum(:in_batches, of: @of, begin_at: @begin_at, end_at: @end_at, load: false) + return enum.each { |relation| yield relation } if block_given? + enum + end + end + end +end -- cgit v1.2.3