activerecord-import 0.19.0 → 1.0.0

data/lib/activerecord-import/import.rb

@@ -3,7 +3,7 @@ require "ostruct"
 module ActiveRecord::Import::ConnectionAdapters; end
 
 module ActiveRecord::Import #:nodoc:
-  Result = Struct.new(:failed_instances, :num_inserts, :ids)
+  Result = Struct.new(:failed_instances, :num_inserts, :ids, :results)
 
   module ImportSupport #:nodoc:
     def supports_import? #:nodoc:
@@ -24,27 +24,63 @@ module ActiveRecord::Import #:nodoc:
   end
 
   class Validator
-    def initialize(options = {})
+    def initialize(klass, options = {})
       @options = options
+      init_validations(klass)
+    end
+
+    def init_validations(klass)
+      @validate_callbacks = klass._validate_callbacks.dup
+
+      @validate_callbacks.each_with_index do |callback, i|
+        filter = callback.raw_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.to_sym
+              attrs[index] = key unless attrs.include?(key)
+            end
+          end
+        end
+
+        filter.instance_variable_set(:@attributes, attrs)
+
+        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)
       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
-        validate_callbacks.each do |callback|
-          validate_callbacks.delete(callback) if callback.raw_filter.is_a? ActiveRecord::Validations::UniquenessValidator
-        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)
@@ -63,10 +99,10 @@ 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
 
@@ -79,13 +115,14 @@ module ActiveRecord::Import #:nodoc:
 end
 
 class ActiveRecord::Associations::CollectionProxy
-  def import(*args, &block)
-    @association.import(*args, &block)
+  def bulk_import(*args, &block)
+    @association.bulk_import(*args, &block)
   end
+  alias import bulk_import unless respond_to? :import
 end
 
 class ActiveRecord::Associations::CollectionAssociation
-  def import(*args, &block)
+  def bulk_import(*args, &block)
     unless owner.persisted?
       raise ActiveRecord::RecordNotSaved, "You cannot call import unless the parent is saved"
     end
@@ -94,16 +131,21 @@ class ActiveRecord::Associations::CollectionAssociation
 
     model_klass = reflection.klass
     symbolized_foreign_key = reflection.foreign_key.to_sym
-    symbolized_column_names = model_klass.column_names.map(&:to_sym)
 
-    owner_primary_key = owner.class.primary_key
+    symbolized_column_names = if model_klass.connection.respond_to?(:supports_virtual_columns?) && model_klass.connection.supports_virtual_columns?
+      model_klass.columns.reject(&:virtual?).map { |c| c.name.to_sym }
+    else
+      model_klass.column_names.map(&:to_sym)
+    end
+
+    owner_primary_key = reflection.active_record_primary_key.to_sym
     owner_primary_key_value = owner.send(owner_primary_key)
 
     # assume array of model objects
     if args.last.is_a?( Array ) && args.last.first.is_a?(ActiveRecord::Base)
       if args.length == 2
         models = args.last
-        column_names = args.first
+        column_names = args.first.dup
       else
         models = args.first
         column_names = symbolized_column_names
@@ -118,7 +160,46 @@ class ActiveRecord::Associations::CollectionAssociation
         m.public_send "#{reflection.type}=", owner.class.name if reflection.type
       end
 
-      return model_klass.import column_names, models, options
+      return model_klass.bulk_import column_names, models, options
+
+    # supports array of hash objects
+    elsif args.last.is_a?( Array ) && args.last.first.is_a?(Hash)
+      if args.length == 2
+        array_of_hashes = args.last
+        column_names = args.first.dup
+        allow_extra_hash_keys = true
+      else
+        array_of_hashes = args.first
+        column_names = array_of_hashes.first.keys
+        allow_extra_hash_keys = false
+      end
+
+      symbolized_column_names = column_names.map(&:to_sym)
+      unless symbolized_column_names.include?(symbolized_foreign_key)
+        column_names << symbolized_foreign_key
+      end
+
+      if reflection.type && !symbolized_column_names.include?(reflection.type.to_sym)
+        column_names << reflection.type.to_sym
+      end
+
+      array_of_attributes = array_of_hashes.map do |h|
+        error_message = model_klass.send(:validate_hash_import, h, symbolized_column_names, allow_extra_hash_keys)
+
+        raise ArgumentError, error_message if error_message
+
+        column_names.map do |key|
+          if key == symbolized_foreign_key
+            owner_primary_key_value
+          elsif reflection.type && key == reflection.type.to_sym
+            owner.class.name
+          else
+            h[key]
+          end
+        end
+      end
+
+      return model_klass.bulk_import column_names, array_of_attributes, options
 
     # supports empty array
     elsif args.last.is_a?( Array ) && args.last.empty?
@@ -127,7 +208,12 @@ class ActiveRecord::Associations::CollectionAssociation
     # supports 2-element array and array
     elsif args.size == 2 && args.first.is_a?( Array ) && args.last.is_a?( Array )
       column_names, array_of_attributes = args
-      symbolized_column_names = column_names.map(&:to_s)
+
+      # dup the passed args so we don't modify unintentionally
+      column_names = column_names.dup
+      array_of_attributes = array_of_attributes.map(&:dup)
+
+      symbolized_column_names = column_names.map(&:to_sym)
 
       if symbolized_column_names.include?(symbolized_foreign_key)
         index = symbolized_column_names.index(symbolized_foreign_key)
@@ -138,19 +224,35 @@ class ActiveRecord::Associations::CollectionAssociation
       end
 
       if reflection.type
-        column_names << reflection.type
-        array_of_attributes.each { |attrs| attrs << owner.class.name }
+        symbolized_type = reflection.type.to_sym
+        if symbolized_column_names.include?(symbolized_type)
+          index = symbolized_column_names.index(symbolized_type)
+          array_of_attributes.each { |attrs| attrs[index] = owner.class.name }
+        else
+          column_names << symbolized_type
+          array_of_attributes.each { |attrs| attrs << owner.class.name }
+        end
       end
 
-      return model_klass.import column_names, array_of_attributes, options
+      return model_klass.bulk_import column_names, array_of_attributes, options
     else
       raise ArgumentError, "Invalid arguments!"
     end
   end
+  alias import bulk_import unless respond_to? :import
 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
+
     # Returns true if the current database connection adapter
     # supports import functionality, otherwise returns false.
     def supports_import?(*args)
@@ -161,14 +263,14 @@ class ActiveRecord::Base
     # supports on duplicate key update functionality, otherwise
     # returns false.
     def supports_on_duplicate_key_update?
-      connection.supports_on_duplicate_key_update?
+      connection.respond_to?(:supports_on_duplicate_key_update?) && connection.supports_on_duplicate_key_update?
     end
 
     # returns true if the current database connection adapter
     # supports setting the primary key of bulk imported models, otherwise
     # returns false
-    def support_setting_primary_key_of_imported_objects?
-      connection.respond_to?(:support_setting_primary_key_of_imported_objects?) && connection.support_setting_primary_key_of_imported_objects?
+    def supports_setting_primary_key_of_imported_objects?
+      connection.respond_to?(:supports_setting_primary_key_of_imported_objects?) && connection.supports_setting_primary_key_of_imported_objects?
     end
 
     # Imports a collection of values to the database.
@@ -214,18 +316,21 @@ class ActiveRecord::Base
     #
     # == Options
     # * +validate+ - true|false, tells import whether or not to use
-    #    ActiveRecord validations. Validations are enforced by default.
+    #   ActiveRecord validations. Validations are enforced by default.
+    #   It skips the uniqueness validation for performance reasons.
+    #   You can find more details here:
+    #   https://github.com/zdennis/activerecord-import/issues/228
     # * +ignore+ - true|false, an alias for on_duplicate_key_ignore.
     # * +on_duplicate_key_ignore+ - true|false, tells import to discard
-    #    records that contain duplicate keys. For Postgres 9.5+ it adds
-    #    ON CONFLICT DO NOTHING, for MySQL it uses INSERT IGNORE, and for
-    #    SQLite it uses INSERT OR IGNORE. Cannot be enabled on a
-    #    recursive import. For database adapters that normally support
-    #    setting primary keys on imported objects, this option prevents
-    #    that from occurring.
-    # * +on_duplicate_key_update+ - an Array or Hash, tells import to
-    #    use MySQL's ON DUPLICATE KEY UPDATE or Postgres 9.5+ ON CONFLICT
-    #    DO UPDATE ability. See On Duplicate Key Update below.
+    #   records that contain duplicate keys. For Postgres 9.5+ it adds
+    #   ON CONFLICT DO NOTHING, for MySQL it uses INSERT IGNORE, and for
+    #   SQLite it uses INSERT OR IGNORE. Cannot be enabled on a
+    #   recursive import. For database adapters that normally support
+    #   setting primary keys on imported objects, this option prevents
+    #   that from occurring.
+    # * +on_duplicate_key_update+ - :all, an Array, or Hash, tells import to
+    #   use MySQL's ON DUPLICATE KEY UPDATE or Postgres/SQLite ON CONFLICT
+    #   DO UPDATE ability. See On Duplicate Key Update below.
     # * +synchronize+ - an array of ActiveRecord instances for the model
     #   that you are currently importing data into. This synchronizes
     #   existing model instances in memory with updates from the import.
@@ -247,6 +352,9 @@ class ActiveRecord::Base
     #  BlogPost.import posts
     #
     #  # Example using array_of_hash_objects
+    #  # NOTE: column_names will be determined by using the keys of the first hash in the array. If later hashes in the
+    #  # array have different keys an exception will be raised. If you have hashes to import with different sets of keys
+    #  # we recommend grouping these into batches before importing.
     #  values = [ {author_name: 'zdennis', title: 'test post'} ], [ {author_name: 'jdoe', title: 'another test post'} ] ]
     #  BlogPost.import values
     #
@@ -280,7 +388,15 @@ class ActiveRecord::Base
     #
     # == On Duplicate Key Update (MySQL)
     #
-    # The :on_duplicate_key_update option can be either an Array or a Hash.
+    # The :on_duplicate_key_update option can be either :all, an Array, or a Hash.
+    #
+    # ==== Using :all
+    #
+    # The :on_duplicate_key_update option can be set to :all. All columns
+    # other than the primary key are updated. If a list of column names is
+    # supplied, only those columns will be updated. Below is an example:
+    #
+    #   BlogPost.import columns, values, on_duplicate_key_update: :all
     #
     # ==== Using an Array
     #
@@ -299,11 +415,19 @@ class ActiveRecord::Base
     #
     #   BlogPost.import columns, attributes, on_duplicate_key_update: { title: :title }
     #
-    # == On Duplicate Key Update (Postgres 9.5+)
+    # == On Duplicate Key Update (Postgres 9.5+ and SQLite 3.24+)
     #
-    # The :on_duplicate_key_update option can be an Array or a Hash with up to
+    # The :on_duplicate_key_update option can be :all, an Array, or a Hash with up to
     # three attributes, :conflict_target (and optionally :index_predicate) or
-    # :constraint_name, and :columns.
+    # :constraint_name (Postgres), and :columns.
+    #
+    # ==== Using :all
+    #
+    # The :on_duplicate_key_update option can be set to :all. All columns
+    # other than the primary key are updated. If a list of column names is
+    # supplied, only those columns will be updated. Below is an example:
+    #
+    #   BlogPost.import columns, values, on_duplicate_key_update: :all
     #
     # ==== Using an Array
     #
@@ -323,7 +447,7 @@ class ActiveRecord::Base
     # conflicting constraint to be explicitly specified. Using this option
     # allows you to specify a constraint other than the primary key.
     #
-    # ====== :conflict_target
+    # ===== :conflict_target
     #
     # The :conflict_target attribute specifies the columns that make up the
     # conflicting unique constraint and can be a single column or an array of
@@ -333,7 +457,7 @@ class ActiveRecord::Base
     #
     #   BlogPost.import columns, values, on_duplicate_key_update: { conflict_target: [ :author_id, :slug ], columns: [ :date_modified ] }
     #
-    # ====== :index_predicate
+    # ===== :index_predicate
     #
     # The :index_predicate attribute optionally specifies a WHERE condition
     # on :conflict_target, which is required for matching against partial
@@ -342,7 +466,7 @@ class ActiveRecord::Base
     #
     #   BlogPost.import columns, values, on_duplicate_key_update: { conflict_target: [ :author_id, :slug ], index_predicate: 'status <> 0', columns: [ :date_modified ] }
     #
-    # ====== :constraint_name
+    # ===== :constraint_name
     #
     # The :constraint_name attribute explicitly identifies the conflicting
     # unique index by name. Postgres documentation discourages using this method
@@ -350,7 +474,7 @@ class ActiveRecord::Base
     #
     #   BlogPost.import columns, values, on_duplicate_key_update: { constraint_name: :blog_posts_pkey, columns: [ :date_modified ] }
     #
-    # ====== :condition
+    # ===== :condition
     #
     # The :condition attribute optionally specifies a WHERE condition
     # on :conflict_action. Only rows for which this expression returns true will be updated.
@@ -358,12 +482,20 @@ class ActiveRecord::Base
     # Below is an example:
     #
     #   BlogPost.import columns, values, on_duplicate_key_update: { conflict_target: [ :author_id ], condition: "blog_posts.title NOT LIKE '%sample%'", columns: [ :author_name ] }
-
-    # ====== :columns
     #
-    # The :columns attribute can be either an Array or a Hash.
+    # ===== :columns
+    #
+    # The :columns attribute can be either :all, an Array, or a Hash.
+    #
+    # ===== Using :all
+    #
+    # The :columns attribute can be :all. All columns other than the primary key will be updated.
+    # If a list of column names is supplied, only those columns will be updated.
+    # Below is an example:
+    #
+    #   BlogPost.import columns, values, on_duplicate_key_update: { conflict_target: :slug, columns: :all }
     #
-    # ======== Using an Array
+    # ===== Using an Array
     #
     # The :columns attribute can be an array of column names. The column names
     # are the only fields that are updated if a duplicate record is found.
@@ -371,7 +503,7 @@ class ActiveRecord::Base
     #
     #   BlogPost.import columns, values, on_duplicate_key_update: { conflict_target: :slug, columns: [ :date_modified, :content, :author ] }
     #
-    # ========  Using a Hash
+    # =====  Using a Hash
     #
     # The :columns option can be a hash of column names to model attribute name
     # mappings. This gives you finer grained control over what fields are updated
@@ -384,7 +516,8 @@ class ActiveRecord::Base
     # * failed_instances - an array of objects that fails validation and were not committed to the database. An empty array if no validation is performed.
     # * num_inserts - the number of insert statements it took to import the data
     # * ids - the primary keys of the imported ids if the adapter supports it, otherwise an empty array.
-    def import(*args)
+    # * results - import results if the adapter supports it, otherwise an empty array.
+    def bulk_import(*args)
       if args.first.is_a?( Array ) && args.first.first.is_a?(ActiveRecord::Base)
         options = {}
         options.merge!( args.pop ) if args.last.is_a?(Hash)
@@ -395,31 +528,29 @@ class ActiveRecord::Base
         import_helper(*args)
       end
     end
+    alias import bulk_import unless 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
     # with the failed instance.
-    def import!(*args)
+    def bulk_import!(*args)
       options = args.last.is_a?( Hash ) ? args.pop : {}
       options[:validate] = true
       options[:raise_error] = true
 
-      import(*args, options)
+      bulk_import(*args, options)
     end
+    alias import! bulk_import! unless respond_to? :import!
 
     def import_helper( *args )
       options = { validate: true, timestamps: true }
       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)
 
-      # Don't modify incoming arguments
-      if options[:on_duplicate_key_update] && options[:on_duplicate_key_update].duplicable?
-        options[:on_duplicate_key_update] = options[:on_duplicate_key_update].dup
-      end
-
-      is_validating = options[:validate]
-      is_validating = true unless options[:validate_with_context].nil?
+      is_validating = options[:validate_with_context].present? ? true : options[:validate]
+      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)
@@ -428,11 +559,19 @@ class ActiveRecord::Base
           column_names = args.first.dup
         else
           models = args.first
-          column_names = self.column_names.dup
+          column_names = if connection.respond_to?(:supports_virtual_columns?) && connection.supports_virtual_columns?
+            columns.reject(&:virtual?).map(&:name)
+          else
+            self.column_names.dup
+          end
         end
 
-        if models.first.id.nil? && column_names.include?(primary_key) && columns_hash[primary_key].type == :uuid
-          column_names.delete(primary_key)
+        if models.first.id.nil?
+          Array(primary_key).each do |c|
+            if column_names.include?(c) && columns_hash[c].type == :uuid
+              column_names.delete(c)
+            end
+          end
         end
 
         default_values = column_defaults
@@ -444,11 +583,34 @@ class ActiveRecord::Base
           serialized_attributes
         end
 
-        array_of_attributes = models.map do |model|
-          column_names.map do |name|
-            if stored_attrs.key?(name.to_sym) ||
-               serialized_attrs.key?(name) ||
-               default_values.key?(name.to_s)
+        update_attrs = if record_timestamps && options[:timestamps]
+          if respond_to?(:timestamp_attributes_for_update, true)
+            send(:timestamp_attributes_for_update).map(&:to_sym)
+          else
+            new.send(:timestamp_attributes_for_update_in_model)
+          end
+        end
+
+        array_of_attributes = []
+
+        models.each do |model|
+          if supports_setting_primary_key_of_imported_objects?
+            load_association_ids(model)
+          end
+
+          if is_validating && !validator.valid_model?(model)
+            raise(ActiveRecord::RecordInvalid, model) if options[:raise_error]
+            next
+          end
+
+          array_of_attributes << column_names.map do |name|
+            if model.persisted? &&
+               update_attrs && update_attrs.include?(name.to_sym) &&
+               !model.send("#{name}_changed?")
+              nil
+            elsif stored_attrs.key?(name.to_sym) ||
+                  serialized_attrs.key?(name.to_s) ||
+                  default_values[name.to_s]
               model.read_attribute(name.to_s)
             else
               model.read_attribute_before_type_cast(name.to_s)
@@ -460,12 +622,18 @@ class ActiveRecord::Base
         if args.length == 2
           array_of_hashes = args.last
           column_names = args.first.dup
+          allow_extra_hash_keys = true
         else
           array_of_hashes = args.first
           column_names = array_of_hashes.first.keys
+          allow_extra_hash_keys = false
         end
 
         array_of_attributes = array_of_hashes.map do |h|
+          error_message = validate_hash_import(h, column_names, allow_extra_hash_keys)
+
+          raise ArgumentError, error_message if error_message
+
           column_names.map do |key|
             h[key]
           end
@@ -497,47 +665,78 @@ class ActiveRecord::Base
 
       if !symbolized_primary_key.to_set.subset?(symbolized_column_names.to_set) && connection.prefetch_primary_key? && sequence_name
         column_count = column_names.size
-        column_names.concat(primary_key).uniq!
+        column_names.concat(Array(primary_key)).uniq!
         columns_added = column_names.size - column_count
         new_fields = Array.new(columns_added)
         array_of_attributes.each { |a| a.concat(new_fields) }
       end
 
+      # Don't modify incoming arguments
+      on_duplicate_key_update = options[:on_duplicate_key_update]
+      if on_duplicate_key_update
+        updatable_columns = symbolized_column_names.reject { |c| symbolized_primary_key.include? c }
+        options[:on_duplicate_key_update] = if on_duplicate_key_update.is_a?(Hash)
+          on_duplicate_key_update.each_with_object({}) do |(k, v), duped_options|
+            duped_options[k] = if k == :columns && v == :all
+              updatable_columns
+            elsif v.duplicable?
+              v.dup
+            else
+              v
+            end
+          end
+        elsif on_duplicate_key_update == :all
+          updatable_columns
+        elsif on_duplicate_key_update.duplicable?
+          on_duplicate_key_update.dup
+        else
+          on_duplicate_key_update
+        end
+      end
+
       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
 
       return_obj = if is_validating
-        if models
-          import_with_validations( column_names, array_of_attributes, options ) do |validator, failed|
-            models.each_with_index do |model, i|
-              model = model.dup if options[:recursive]
-              next if validator.valid_model? model
+        import_with_validations( column_names, array_of_attributes, options ) do |failed_instances|
+          if models
+            models.each { |m| failed_instances << m if m.errors.any? }
+          else
+            # create instances for each of our column/value sets
+            arr = validations_array_for_column_names_and_attributes( column_names, array_of_attributes )
+
+            # 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 }
+              next if validator.valid_model?(model)
               raise(ActiveRecord::RecordInvalid, model) if options[:raise_error]
               array_of_attributes[i] = nil
-              failed << model
+              failure = model.dup
+              failure.errors.send(:initialize_dup, model.errors)
+              failed_instances << failure
             end
+            array_of_attributes.compact!
           end
-        else
-          import_with_validations( column_names, array_of_attributes, options )
         end
       else
-        (num_inserts, ids) = import_without_validations_or_callbacks( column_names, array_of_attributes, options )
-        ActiveRecord::Import::Result.new([], num_inserts, ids)
+        import_without_validations_or_callbacks( column_names, array_of_attributes, options )
       end
 
       if options[:synchronize]
-        sync_keys = options[:synchronize_keys] || [primary_key]
+        sync_keys = options[:synchronize_keys] || Array(primary_key)
         synchronize( options[:synchronize], sync_keys)
       end
       return_obj.num_inserts = 0 if return_obj.num_inserts.nil?
 
       # if we have ids, then set the id on the models and mark the models as clean.
-      if models && support_setting_primary_key_of_imported_objects?
-        set_attributes_and_mark_clean(models, return_obj, timestamps)
+      if models && supports_setting_primary_key_of_imported_objects?
+        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]
@@ -556,36 +755,14 @@ class ActiveRecord::Base
     def import_with_validations( column_names, array_of_attributes, options = {} )
       failed_instances = []
 
-      validator = ActiveRecord::Import::Validator.new(options)
-
-      if block_given?
-        yield validator, failed_instances
-      else
-        # create instances for each of our column/value sets
-        arr = validations_array_for_column_names_and_attributes( column_names, array_of_attributes )
-
-        # 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
-        model = new
-        arr.each_with_index do |hsh, i|
-          hsh.each_pair { |k, v| model[k] = v }
-          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
-        end
-      end
-
-      array_of_attributes.compact!
+      yield failed_instances if block_given?
 
-      num_inserts, ids = if array_of_attributes.empty? || options[:all_or_none] && failed_instances.any?
-        [0, []]
+      result = if options[:all_or_none] && failed_instances.any?
+        ActiveRecord::Import::Result.new([], 0, [], [])
       else
         import_without_validations_or_callbacks( column_names, array_of_attributes, options )
       end
-      ActiveRecord::Import::Result.new(failed_instances, num_inserts, ids)
+      ActiveRecord::Import::Result.new(failed_instances, result.num_inserts, result.ids, result.results)
     end
 
     # Imports the passed in +column_names+ and +array_of_attributes+
@@ -595,6 +772,8 @@ class ActiveRecord::Base
     # information on +column_names+, +array_of_attributes_ and
     # +options+.
     def import_without_validations_or_callbacks( column_names, array_of_attributes, options = {} )
+      return ActiveRecord::Import::Result.new([], 0, [], []) if array_of_attributes.empty?
+
       column_names = column_names.map(&:to_sym)
       scope_columns, scope_values = scope_attributes.to_a.transpose
 
@@ -604,7 +783,7 @@ class ActiveRecord::Base
           next if column_names.include?(name_as_sym)
 
           is_sti = (name_as_sym == inheritance_column.to_sym && self < base_class)
-          value = value.first if is_sti
+          value = Array(value).first if is_sti
 
           column_names << name_as_sym
           array_of_attributes.each { |attrs| attrs << value }
@@ -627,6 +806,7 @@ class ActiveRecord::Base
 
       number_inserted = 0
       ids = []
+      results = []
       if supports_import?
         # generate the sql
         post_sql_statements = connection.post_sql_statements( quoted_table_name, options )
@@ -637,9 +817,10 @@ class ActiveRecord::Base
           result = connection.insert_many( [insert_sql, post_sql_statements].flatten,
             batch_values,
             options,
-            "#{self.class.name} Create Many Without Validations Or Callbacks" )
-          number_inserted += result[0]
-          ids += result[1]
+            "#{model_name} Create Many Without Validations Or Callbacks" )
+          number_inserted += result.num_inserts
+          ids += result.ids
+          results += result.results
         end
       else
         transaction(requires_new: true) do
@@ -649,22 +830,14 @@ class ActiveRecord::Base
           end
         end
       end
-      [number_inserted, ids]
+      ActiveRecord::Import::Result.new([], number_inserted, ids, results)
     end
 
     private
 
-    def set_attributes_and_mark_clean(models, import_result, timestamps)
+    def set_attributes_and_mark_clean(models, import_result, timestamps, options)
       return if models.nil?
       models -= import_result.failed_instances
-      models.each do |model|
-        if 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
-        end
-        model.instance_variable_set(:@new_record, false)
-      end
 
       # if ids were returned for all models we know all were updated
       if models.size == import_result.ids.size
@@ -677,6 +850,49 @@ class ActiveRecord::Base
           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|
+          model = models[index]
+
+          if single_column
+            model.send(single_column, result)
+          else
+            columns.each_with_index do |column, col_index|
+              model.send("#{column}=", result[col_index])
+            end
+          end
+        end
+      end
+
+      models.each do |model|
+        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
+        end
+        model.instance_variable_set(:@new_record, false)
+      end
+    end
+
+    # Sync belongs_to association ids with foreign key field
+    def load_association_ids(model)
+      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))
+      end
     end
 
     def import_associations(models, options)
@@ -693,7 +909,7 @@ class ActiveRecord::Base
 
       associated_objects_by_class.each_value do |associations|
         associations.each_value do |associated_records|
-          associated_records.first.class.import(associated_records, options) unless associated_records.empty?
+          associated_records.first.class.bulk_import(associated_records, options) unless associated_records.empty?
         end
       end
     end
@@ -702,6 +918,7 @@ class ActiveRecord::Base
     # of class => objects to import.
     def find_associated_objects_for_import(associated_objects_by_class, model)
       associated_objects_by_class[model.class.name] ||= {}
+      return associated_objects_by_class unless model.id
 
       association_reflections =
         model.class.reflect_on_all_associations(:has_one) +
@@ -740,8 +957,10 @@ class ActiveRecord::Base
           column = columns[j]
 
           # be sure to query sequence_name *last*, only if cheaper tests fail, because it's costly
-          if val.nil? && column.name == primary_key && !sequence_name.blank?
+          if val.nil? && Array(primary_key).first == column.name && !sequence_name.blank?
             connection_memo.next_value_for_sequence(sequence_name)
+          elsif val.respond_to?(:to_sql)
+            "(#{val.to_sql})"
           elsif column
             if respond_to?(:type_caster)                                         # Rails 5.0 and higher
               type = type_for_attribute(column.name)
@@ -753,7 +972,9 @@ class ActiveRecord::Base
               if serialized_attributes.include?(column.name)
                 val = serialized_attributes[column.name].dump(val)
               end
-              connection_memo.quote(column.type_cast(val), column)
+              # Fixes #443 to support binary (i.e. bytea) columns on PG
+              val = column.type_cast(val) unless column.type.to_sym == :binary
+              connection_memo.quote(val, column)
             end
           end
         end
@@ -785,7 +1006,7 @@ 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 }
@@ -804,5 +1025,42 @@ class ActiveRecord::Base
     def validations_array_for_column_names_and_attributes( column_names, array_of_attributes ) # :nodoc:
       array_of_attributes.map { |values| Hash[column_names.zip(values)] }
     end
+
+    # Checks that the imported hash has the required_keys, optionally also checks that the hash has
+    # no keys beyond those required when `allow_extra_keys` is false.
+    # returns `nil` if validation passes, or an error message if it fails
+    def validate_hash_import(hash, required_keys, allow_extra_keys) # :nodoc:
+      extra_keys = allow_extra_keys ? [] : hash.keys - required_keys
+      missing_keys = required_keys - hash.keys
+
+      return nil if extra_keys.empty? && missing_keys.empty?
+
+      if allow_extra_keys
+        <<-EOS
+Hash key mismatch.
+
+When importing an array of hashes with provided columns_names, each hash must contain keys for all column_names.
+
+Required keys: #{required_keys}
+Missing keys: #{missing_keys}
+
+Hash: #{hash}
+        EOS
+      else
+        <<-EOS
+Hash key mismatch.
+
+When importing an array of hashes, all hashes must have the same keys.
+If you have records that are missing some values, we recommend you either set default values
+for the missing keys or group these records into batches by key set before importing.
+
+Required keys: #{required_keys}
+Extra keys: #{extra_keys}
+Missing keys: #{missing_keys}
+
+Hash: #{hash}
+        EOS
+      end
+    end
   end
 end