activerecord-import 0.27.0 → 1.4.1

data/lib/activerecord-import/import.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require "ostruct"
 
 module ActiveRecord::Import::ConnectionAdapters; end
@@ -24,33 +26,66 @@ module ActiveRecord::Import #:nodoc:
   end
 
   class Validator
-    def initialize(options = {})
+    def initialize(klass, options = {})
       @options = options
+      @validator_class = klass
+      init_validations(klass)
+    end
+
+    def init_validations(klass)
+      @validate_callbacks = klass._validate_callbacks.dup
+
+      @validate_callbacks.each_with_index do |callback, i|
+        filter = callback.respond_to?(:raw_filter) ? callback.raw_filter : callback.filter
+        next unless filter.class.name =~ /Validations::PresenceValidator/ ||
+                    (!@options[:validate_uniqueness] &&
+                     filter.is_a?(ActiveRecord::Validations::UniquenessValidator))
+
+        callback = callback.dup
+        filter = filter.dup
+        attrs = filter.instance_variable_get(:@attributes).dup
+
+        if filter.is_a?(ActiveRecord::Validations::UniquenessValidator)
+          attrs = []
+        else
+          associations = klass.reflect_on_all_associations(:belongs_to)
+          associations.each do |assoc|
+            if (index = attrs.index(assoc.name))
+              key = assoc.foreign_key.is_a?(Array) ? assoc.foreign_key.map(&:to_sym) : assoc.foreign_key.to_sym
+              attrs[index] = key unless attrs.include?(key)
+            end
+          end
+        end
+
+        filter.instance_variable_set(:@attributes, attrs.flatten)
+
+        if @validate_callbacks.respond_to?(:chain, true)
+          @validate_callbacks.send(:chain).tap do |chain|
+            callback.instance_variable_set(:@filter, filter)
+            chain[i] = callback
+          end
+        else
+          callback.raw_filter = filter
+          callback.filter = callback.send(:_compile_filter, filter)
+          @validate_callbacks[i] = callback
+        end
+      end
     end
 
     def valid_model?(model)
+      init_validations(model.class) unless model.class == @validator_class
+
       validation_context = @options[:validate_with_context]
       validation_context ||= (model.new_record? ? :create : :update)
-
       current_context = model.send(:validation_context)
+
       begin
         model.send(:validation_context=, validation_context)
         model.errors.clear
 
-        validate_callbacks = model._validate_callbacks.dup
-        associations = model.class.reflect_on_all_associations(:belongs_to).map(&:name)
-
-        model._validate_callbacks.each do |callback|
-          filter = callback.raw_filter
-          if (!@options[:validate_uniqueness] && filter.is_a?(ActiveRecord::Validations::UniquenessValidator)) ||
-             (defined?(ActiveRecord::Validations::PresenceValidator) && filter.is_a?(ActiveRecord::Validations::PresenceValidator) && associations.include?(filter.attributes.first))
-            validate_callbacks.delete(callback)
-          end
-        end
-
         model.run_callbacks(:validation) do
           if defined?(ActiveSupport::Callbacks::Filters::Environment) # ActiveRecord >= 4.1
-            runner = validate_callbacks.compile
+            runner = @validate_callbacks.compile
             env = ActiveSupport::Callbacks::Filters::Environment.new(model, false, nil)
             if runner.respond_to?(:call) # ActiveRecord < 5.1
               runner.call(env)
@@ -69,10 +104,10 @@ module ActiveRecord::Import #:nodoc:
               runner.invoke_before(env)
               runner.invoke_after(env)
             end
-          elsif validate_callbacks.method(:compile).arity == 0 # ActiveRecord = 4.0
-            model.instance_eval validate_callbacks.compile
+          elsif @validate_callbacks.method(:compile).arity == 0 # ActiveRecord = 4.0
+            model.instance_eval @validate_callbacks.compile
           else # ActiveRecord 3.x
-            model.instance_eval validate_callbacks.compile(nil, model)
+            model.instance_eval @validate_callbacks.compile(nil, model)
           end
         end
 
@@ -212,16 +247,17 @@ class ActiveRecord::Associations::CollectionAssociation
   alias import bulk_import unless respond_to? :import
 end
 
+module ActiveRecord::Import::Connection
+  def establish_connection(args = nil)
+    conn = super(args)
+    ActiveRecord::Import.load_from_connection_pool connection_pool
+    conn
+  end
+end
+
 class ActiveRecord::Base
   class << self
-    def establish_connection_with_activerecord_import(*args)
-      conn = establish_connection_without_activerecord_import(*args)
-      ActiveRecord::Import.load_from_connection_pool connection_pool
-      conn
-    end
-
-    alias establish_connection_without_activerecord_import establish_connection
-    alias establish_connection establish_connection_with_activerecord_import
+    prepend ActiveRecord::Import::Connection
 
     # Returns true if the current database connection adapter
     # supports import functionality, otherwise returns false.
@@ -498,7 +534,7 @@ class ActiveRecord::Base
         import_helper(*args)
       end
     end
-    alias import bulk_import unless respond_to? :import
+    alias import bulk_import unless ActiveRecord::Base.respond_to? :import
 
     # Imports a collection of values if all values are valid. Import fails at the
     # first encountered validation error and raises ActiveRecord::RecordInvalid
@@ -510,17 +546,17 @@ class ActiveRecord::Base
 
       bulk_import(*args, options)
     end
-    alias import! bulk_import! unless respond_to? :import!
+    alias import! bulk_import! unless ActiveRecord::Base.respond_to? :import!
 
     def import_helper( *args )
-      options = { validate: true, timestamps: true }
+      options = { model: self, validate: true, timestamps: true, track_validation_failures: false }
       options.merge!( args.pop ) if args.last.is_a? Hash
       # making sure that current model's primary key is used
       options[:primary_key] = primary_key
       options[:locking_column] = locking_column if attribute_names.include?(locking_column)
 
       is_validating = options[:validate_with_context].present? ? true : options[:validate]
-      validator = ActiveRecord::Import::Validator.new(options)
+      validator = ActiveRecord::Import::Validator.new(self, options)
 
       # assume array of model objects
       if args.last.is_a?( Array ) && args.last.first.is_a?(ActiveRecord::Base)
@@ -544,13 +580,12 @@ class ActiveRecord::Base
           end
         end
 
-        default_values = column_defaults
-        stored_attrs = respond_to?(:stored_attributes) ? stored_attributes : {}
-        serialized_attrs = if defined?(ActiveRecord::Type::Serialized)
-          attrs = column_names.select { |c| type_for_attribute(c.to_s).class == ActiveRecord::Type::Serialized }
-          Hash[attrs.map { |a| [a, nil] }]
-        else
-          serialized_attributes
+        update_attrs = if record_timestamps && options[:timestamps]
+          if respond_to?(:timestamp_attributes_for_update, true)
+            send(:timestamp_attributes_for_update).map(&:to_sym)
+          else
+            allocate.send(:timestamp_attributes_for_update_in_model)
+          end
         end
 
         array_of_attributes = []
@@ -566,12 +601,12 @@ class ActiveRecord::Base
           end
 
           array_of_attributes << column_names.map do |name|
-            if stored_attrs.key?(name.to_sym) ||
-               serialized_attrs.key?(name) ||
-               default_values.key?(name.to_s)
-              model.read_attribute(name.to_s)
+            if model.persisted? &&
+               update_attrs && update_attrs.include?(name.to_sym) &&
+               !model.send("#{name}_changed?")
+              nil
             else
-              model.read_attribute_before_type_cast(name.to_s)
+              model.read_attribute(name.to_s)
             end
           end
         end
@@ -598,7 +633,7 @@ class ActiveRecord::Base
         end
         # supports empty array
       elsif args.last.is_a?( Array ) && args.last.empty?
-        return ActiveRecord::Import::Result.new([], 0, [])
+        return ActiveRecord::Import::Result.new([], 0, [], [])
         # supports 2-element array and array
       elsif args.size == 2 && args.first.is_a?( Array ) && args.last.is_a?( Array )
 
@@ -655,7 +690,7 @@ class ActiveRecord::Base
       timestamps = {}
 
       # record timestamps unless disabled in ActiveRecord::Base
-      if record_timestamps && options.delete( :timestamps )
+      if record_timestamps && options[:timestamps]
         timestamps = add_special_rails_stamps column_names, array_of_attributes, options
       end
 
@@ -670,14 +705,18 @@ class ActiveRecord::Base
             # keep track of the instance and the position it is currently at. if this fails
             # validation we'll use the index to remove it from the array_of_attributes
             arr.each_with_index do |hsh, i|
-              model = new
-              hsh.each_pair { |k, v| model[k] = v }
+              # utilize block initializer syntax to prevent failure when 'mass_assignment_sanitizer = :strict'
+              model = new do |m|
+                hsh.each_pair { |k, v| m[k] = v }
+              end
+
               next if validator.valid_model?(model)
               raise(ActiveRecord::RecordInvalid, model) if options[:raise_error]
+
               array_of_attributes[i] = nil
               failure = model.dup
               failure.errors.send(:initialize_dup, model.errors)
-              failed_instances << failure
+              failed_instances << (options[:track_validation_failures] ? [i, failure] : failure )
             end
             array_of_attributes.compact!
           end
@@ -697,7 +736,10 @@ class ActiveRecord::Base
         set_attributes_and_mark_clean(models, return_obj, timestamps, options)
 
         # if there are auto-save associations on the models we imported that are new, import them as well
-        import_associations(models, options.dup) if options[:recursive]
+        if options[:recursive]
+          options[:on_duplicate_key_update] = on_duplicate_key_update unless on_duplicate_key_update.nil?
+          import_associations(models, options.dup.merge(validate: false))
+        end
       end
 
       return_obj
@@ -738,21 +780,22 @@ class ActiveRecord::Base
       unless scope_columns.blank?
         scope_columns.zip(scope_values).each do |name, value|
           name_as_sym = name.to_sym
-          next if column_names.include?(name_as_sym)
-
-          is_sti = (name_as_sym == inheritance_column.to_sym && self < base_class)
-          value = Array(value).first if is_sti
-
+          next if column_names.include?(name_as_sym) || name_as_sym == inheritance_column.to_sym
           column_names << name_as_sym
           array_of_attributes.each { |attrs| attrs << value }
         end
       end
 
+      if finder_needs_type_condition?
+        unless column_names.include?(inheritance_column.to_sym)
+          column_names << inheritance_column.to_sym
+          array_of_attributes.each { |attrs| attrs << sti_name }
+        end
+      end
+
       columns = column_names.each_with_index.map do |name, i|
         column = columns_hash[name.to_s]
-
         raise ActiveRecord::Import::MissingColumnError.new(name.to_s, i) if column.nil?
-
         column
       end
 
@@ -768,17 +811,29 @@ class ActiveRecord::Base
       if supports_import?
         # generate the sql
         post_sql_statements = connection.post_sql_statements( quoted_table_name, options )
+        import_size = values_sql.size
+
+        batch_size = options[:batch_size] || import_size
+        run_proc = options[:batch_size].to_i.positive? && options[:batch_progress].respond_to?( :call )
+        progress_proc = options[:batch_progress]
+        current_batch = 0
+        batches = (import_size / batch_size.to_f).ceil
 
-        batch_size = options[:batch_size] || values_sql.size
         values_sql.each_slice(batch_size) do |batch_values|
+          batch_started_at = Time.now.to_i
+
           # perform the inserts
           result = connection.insert_many( [insert_sql, post_sql_statements].flatten,
             batch_values,
             options,
-            "#{model_name} Create Many Without Validations Or Callbacks" )
+            "#{model_name} Create Many" )
+
           number_inserted += result.num_inserts
           ids += result.ids
           results += result.results
+          current_batch += 1
+
+          progress_proc.call(import_size, batches, current_batch, Time.now.to_i - batch_started_at) if run_proc
         end
       else
         transaction(requires_new: true) do
@@ -804,29 +859,56 @@ class ActiveRecord::Base
           model.id = id
 
           timestamps.each do |attr, value|
-            model.send(attr + "=", value)
+            model.send(attr + "=", value) if model.send(attr).nil?
           end
         end
       end
 
-      if models.size == import_result.results.size
-        columns = Array(options[:returning])
-        single_column = "#{columns.first}=" if columns.size == 1
-        import_result.results.each_with_index do |result, index|
+      deserialize_value = lambda do |column, value|
+        column = columns_hash[column]
+        return value unless column
+        if respond_to?(:type_caster)
+          type = type_for_attribute(column.name)
+          type.deserialize(value)
+        elsif column.respond_to?(:type_cast_from_database)
+          column.type_cast_from_database(value)
+        else
+          value
+        end
+      end
+
+      set_value = lambda do |model, column, value|
+        val = deserialize_value.call(column, value)
+        if model.attribute_names.include?(column)
+          model.send("#{column}=", val)
+        else
+          attributes = attributes_builder.build_from_database(model.attributes.merge(column => val))
+          model.instance_variable_set(:@attributes, attributes)
+        end
+      end
+
+      columns = Array(options[:returning_columns])
+      results = Array(import_result.results)
+      if models.size == results.size
+        single_column = columns.first if columns.size == 1
+        results.each_with_index do |result, index|
           model = models[index]
 
           if single_column
-            model.send(single_column, result)
+            set_value.call(model, single_column, result)
           else
             columns.each_with_index do |column, col_index|
-              model.send("#{column}=", result[col_index])
+              set_value.call(model, column, result[col_index])
             end
           end
         end
       end
 
       models.each do |model|
-        if model.respond_to?(:clear_changes_information) # Rails 4.0 and higher
+        if model.respond_to?(:changes_applied) # Rails 4.1.8 and higher
+          model.changes_internally_applied if model.respond_to?(:changes_internally_applied) # legacy behavior for Rails 5.1
+          model.changes_applied
+        elsif model.respond_to?(:clear_changes_information) # Rails 4.0 and higher
           model.clear_changes_information
         else # Rails 3.2
           model.instance_variable_get(:@changed_attributes).clear
@@ -837,16 +919,22 @@ class ActiveRecord::Base
 
     # Sync belongs_to association ids with foreign key field
     def load_association_ids(model)
+      changed_columns = model.changed
       association_reflections = model.class.reflect_on_all_associations(:belongs_to)
       association_reflections.each do |association_reflection|
-        column_name = association_reflection.foreign_key
         next if association_reflection.options[:polymorphic]
-        association = model.association(association_reflection.name)
-        association = association.target
-        next if association.blank? || model.public_send(column_name).present?
 
-        association_primary_key = association_reflection.association_primary_key
-        model.public_send("#{column_name}=", association.send(association_primary_key))
+        column_names = Array(association_reflection.foreign_key).map(&:to_s)
+        column_names.each_with_index do |column_name, column_index|
+          next if changed_columns.include?(column_name)
+
+          association = model.association(association_reflection.name)
+          association = association.target
+          next if association.blank? || model.public_send(column_name).present?
+
+          association_primary_key = Array(association_reflection.association_primary_key)[column_index]
+          model.public_send("#{column_name}=", association.send(association_primary_key))
+        end
       end
     end
 
@@ -859,8 +947,10 @@ class ActiveRecord::Base
       associated_objects_by_class = {}
       models.each { |model| find_associated_objects_for_import(associated_objects_by_class, model) }
 
-      # :on_duplicate_key_update not supported for associations
-      options.delete(:on_duplicate_key_update)
+      # :on_duplicate_key_update only supported for all fields
+      options.delete(:on_duplicate_key_update) unless options[:on_duplicate_key_update] == :all
+      # :returning not supported for associations
+      options.delete(:returning)
 
       associated_objects_by_class.each_value do |associations|
         associations.each_value do |associated_records|
@@ -891,8 +981,13 @@ class ActiveRecord::Base
         changed_objects.each do |child|
           child.public_send("#{association_reflection.foreign_key}=", model.id)
           # For polymorphic associations
+          association_name = if model.class.respond_to?(:polymorphic_name)
+            model.class.polymorphic_name
+          else
+            model.class.base_class
+          end
           association_reflection.type.try do |type|
-            child.public_send("#{type}=", model.class.base_class.name)
+            child.public_send("#{type}=", association_name)
           end
         end
         associated_objects_by_class[model.class.name][association_reflection.name].concat changed_objects
@@ -919,7 +1014,7 @@ class ActiveRecord::Base
           elsif column
             if respond_to?(:type_caster)                                         # Rails 5.0 and higher
               type = type_for_attribute(column.name)
-              val = type.type == :boolean ? type.cast(val) : type.serialize(val)
+              val = !type.respond_to?(:subtype) && type.type == :boolean ? type.cast(val) : type.serialize(val)
               connection_memo.quote(val)
             elsif column.respond_to?(:type_cast_from_user)                       # Rails 4.2
               connection_memo.quote(column.type_cast_from_user(val), column)
@@ -928,9 +1023,11 @@ class ActiveRecord::Base
                 val = serialized_attributes[column.name].dump(val)
               end
               # Fixes #443 to support binary (i.e. bytea) columns on PG
-              val = column.type_cast(val) unless column.type.to_sym == :binary
+              val = column.type_cast(val) unless column.type && column.type.to_sym == :binary
               connection_memo.quote(val, column)
             end
+          else
+            raise ArgumentError, "Number of values (#{arr.length}) exceeds number of columns (#{columns.length})"
           end
         end
         "(#{my_values.join(',')})"
@@ -945,13 +1042,18 @@ class ActiveRecord::Base
         timestamp_columns[:create] = timestamp_attributes_for_create_in_model
         timestamp_columns[:update] = timestamp_attributes_for_update_in_model
       else
-        instance = new
+        instance = allocate
         timestamp_columns[:create] = instance.send(:timestamp_attributes_for_create_in_model)
         timestamp_columns[:update] = instance.send(:timestamp_attributes_for_update_in_model)
       end
 
       # use tz as set in ActiveRecord::Base
-      timestamp = ActiveRecord::Base.default_timezone == :utc ? Time.now.utc : Time.now
+      default_timezone = if ActiveRecord.respond_to?(:default_timezone)
+        ActiveRecord.default_timezone
+      else
+        ActiveRecord::Base.default_timezone
+      end
+      timestamp = default_timezone == :utc ? Time.now.utc : Time.now
 
       [:create, :update].each do |action|
         timestamp_columns[action].each do |column|
@@ -961,7 +1063,7 @@ class ActiveRecord::Base
           index = column_names.index(column) || column_names.index(column.to_sym)
           if index
             # replace every instance of the array of attributes with our value
-            array_of_attributes.each { |arr| arr[index] = timestamp if arr[index].nil? || action == :update }
+            array_of_attributes.each { |arr| arr[index] = timestamp if arr[index].nil? }
           else
             column_names << column
             array_of_attributes.each { |arr| arr << timestamp }

data/lib/activerecord-import/mysql2.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 warn <<-MSG
 [DEPRECATION] loading activerecord-import via 'require "activerecord-import/<adapter-name>"'
   is deprecated. Update to autorequire using 'require "activerecord-import"'. See

data/lib/activerecord-import/postgresql.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 warn <<-MSG
 [DEPRECATION] loading activerecord-import via 'require "activerecord-import/<adapter-name>"'
   is deprecated. Update to autorequire using 'require "activerecord-import"'. See

data/lib/activerecord-import/sqlite3.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 warn <<-MSG
 [DEPRECATION] loading activerecord-import via 'require "activerecord-import/<adapter-name>"'
   is deprecated. Update to autorequire using 'require "activerecord-import"'. See

data/lib/activerecord-import/synchronize.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module ActiveRecord # :nodoc:
   class Base # :nodoc:
     # Synchronizes the passed in ActiveRecord instances with data
@@ -39,8 +41,8 @@ module ActiveRecord # :nodoc:
 
         next unless matched_instance
 
-        instance.send :clear_aggregation_cache
-        instance.send :clear_association_cache
+        instance.instance_variable_set :@association_cache, {}
+        instance.send :clear_aggregation_cache if instance.respond_to?(:clear_aggregation_cache, true)
         instance.instance_variable_set :@attributes, matched_instance.instance_variable_get(:@attributes)
 
         if instance.respond_to?(:clear_changes_information)

data/lib/activerecord-import/value_sets_parser.rb

@@ -1,3 +1,7 @@
+# frozen_string_literal: true
+
+require 'active_support/core_ext/array'
+
 module ActiveRecord::Import
   class ValueSetTooLargeError < StandardError
     attr_reader :size

data/lib/activerecord-import/version.rb

@@ -1,5 +1,7 @@
+# frozen_string_literal: true
+
 module ActiveRecord
   module Import
-    VERSION = "0.27.0".freeze
+    VERSION = "1.4.1"
   end
 end

data/lib/activerecord-import.rb

@@ -1,4 +1,6 @@
-# rubocop:disable Style/FileName
+# rubocop:disable Naming/FileName
+# frozen_string_literal: true
+
 require "active_support/lazy_load_hooks"
 
 ActiveSupport.on_load(:active_record) do

data/test/adapters/jdbcmysql.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "jdbcmysql"

data/test/adapters/jdbcpostgresql.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "jdbcpostgresql"

data/test/adapters/jdbcsqlite3.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "jdbcsqlite3"

data/test/adapters/makara_postgis.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "postgis"

data/test/adapters/mysql2.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "mysql2"

data/test/adapters/mysql2_makara.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "mysql2_makara"

data/test/adapters/mysql2spatial.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "mysql2spatial"

data/test/adapters/postgis.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "postgis"

data/test/adapters/postgresql.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "postgresql"

data/test/adapters/postgresql_makara.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "postgresql"

data/test/adapters/seamless_database_pool.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "seamless_database_pool"

data/test/adapters/spatialite.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "spatialite"

data/test/adapters/sqlite3.rb

@@ -1 +1,3 @@
+# frozen_string_literal: true
+
 ENV["ARE_DB"] = "sqlite3"

data/test/{travis → github}/database.yml

@@ -1,7 +1,8 @@
 common: &common
   username: root
-  password:
+  password: root
   encoding: utf8
+  collation: utf8_general_ci
   host:     localhost
   database: activerecord_import_test
 
@@ -37,6 +38,7 @@ oracle:
 postgresql: &postgresql
   <<: *common
   username: postgres
+  password: postgres
   adapter: postgresql
   min_messages: warning