activerecord_batch_update 0.0.3

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 614292d3e09e070b8d7aff48134d8cee8104dc9f00e96e9deb33b0195bf04858
+  data.tar.gz: 2c4f608cea0f5ae569c76df93c3f3aafed9bc0911c13253d7a2cb02b9c24332d
+SHA512:
+  metadata.gz: d5be0abbf08a712b5b5a138b15e37928b9b50c138851651877ac56837f18b884e82b226eaa1c96c1dd1198c8c017bd665c3ac57b6c7e78ecbbfb70f8d0f22443
+  data.tar.gz: ba349d81455f2908dbcddff5da11c4016b8660e13af081dcdfd4cfd55043c35b73b22bd6e221f95638eb6a23e45a6042ae310e8b38de85f7d73e20bf9a677ea8

data/lib/activerecord_batch_update.rb

@@ -0,0 +1,118 @@
+# frozen_string_literal: true
+
+require 'active_support'
+
+module ActiveRecordBatchUpdate
+  extend ::ActiveSupport::Concern
+
+  # Given an array of records with changes,
+  # perform the minimal amount of queries to update
+  # all the records without including unchanged attributes
+  # in the UPDATE statement.
+  #
+  # Do this without INSERT ... ON DUPLICATE KEY UPDATE
+  # which will re-insert the objects if they were deleted in another thread
+
+  module ClassMethods
+    def batch_update(entries, columns:, batch_size: 100, validate: true)
+      columns = column_names if columns == :all
+      columns = (Array.wrap(columns).map(&:to_s) + %w[updated_at]).uniq
+
+      entries = entries.select { columns.intersect?(_1.changed) }
+      entries.each { _1.updated_at = Time.current } if has_attribute?('updated_at')
+      entries.each(&:validate!) if validate
+
+      primary_keys = Array.wrap(primary_key).map(&:to_s)
+
+      updated_count = batch_update_statements(
+        entries.map do |entry|
+          (primary_keys + (entry.changed & columns)).to_h { [_1, entry.read_attribute(_1)] }
+        end,
+        update_on: primary_keys,
+        batch_size: batch_size
+      ).sum do |sql|
+        connection.exec_update(sql)
+      end
+
+      connection.clear_query_cache if connection.query_cache_enabled
+
+      updated_count
+    end
+
+    def batch_update_statements(entries, update_on: :id, batch_size: 100)
+      update_on = Array.wrap(update_on).map(&:to_s)
+
+      entries.map(&:stringify_keys).group_by { _1.keys.sort! }.sort.flat_map do |(keys, items)|
+        next [] if keys.empty?
+
+        where_clause = where_statement(update_on)
+        update_clause = update_statement(keys - update_on)
+
+        items.each_slice(batch_size).map do |slice|
+          [
+            "WITH \"#{cte_table.name}\" (#{keys.join(', ')})",
+            "AS ( #{values_statement(slice, keys)} )",
+            update_clause,
+            "FROM \"#{cte_table.name}\"",
+            "WHERE #{where_clause}"
+          ].join(' ')
+        end
+      end
+    end
+
+    private
+
+    def cte_table
+      @cte_table ||= Arel::Table.new('batch_updates')
+    end
+
+    def values_statement(items, cols)
+      first, *rest = items
+
+      rows = [
+        values_list_casted_item(first, cols),
+        *rest.map { values_list_other_item(_1, cols) }
+      ]
+
+      "VALUES #{rows.map { "(#{_1.map(&:to_sql).join(', ')})" }.join(', ')}"
+    end
+
+    def values_list_casted_item(item, cols)
+      cols.map do |col|
+        Arel::Nodes::NamedFunction.new(
+          'CAST',
+          [
+            Arel::Nodes.build_quoted(item[col], arel_table[col]).as(columns_hash[col].sql_type_metadata.sql_type)
+          ]
+        )
+      end
+    end
+
+    def values_list_other_item(item, cols)
+      cols.map do |col|
+        Arel::Nodes.build_quoted(item[col], arel_table[col])
+      end
+    end
+
+    def update_statement(cols)
+      Arel::UpdateManager.new(arel_table).tap do |um|
+        um.set(
+          cols.map do |col|
+            [
+              arel_table[col],
+              cte_table[col]
+            ]
+          end
+        )
+      end.to_sql
+    end
+
+    def where_statement(primary_keys)
+      primary_keys.map { arel_table[_1].eq(cte_table[_1]) }.reduce(:and).to_sql
+    end
+  end
+end
+
+ActiveSupport.on_load(:active_record) do
+  include(ActiveRecordBatchUpdate)
+end

metadata

@@ -0,0 +1,72 @@
+--- !ruby/object:Gem::Specification
+name: activerecord_batch_update
+version: !ruby/object:Gem::Version
+  version: 0.0.3
+platform: ruby
+authors:
+- Quentin de Metz
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2024-10-19 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: activerecord
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '7.0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '7.0'
+- !ruby/object:Gem::Dependency
+  name: activesupport
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '7.0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '7.0'
+description: ''
+email: quentin@pennylane.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/activerecord_batch_update.rb
+homepage: https://rubygems.org/gems/activerecord_batch_update
+licenses:
+- MIT
+metadata:
+  rubygems_mfa_required: 'true'
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 3.3.4
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.5.11
+signing_key:
+specification_version: 4
+summary: Update multiple records with different values in a small number of queries
+test_files: []