# frozen-string-literal: true module Sequel class Model # Associations are used in order to specify relationships between model classes # that reflect relations between tables in the database using foreign keys. module Associations # Map of association type symbols to association reflection classes. ASSOCIATION_TYPES = {} # Set an empty association reflection hash in the model def self.apply(model) model.instance_exec do @association_reflections = {} @autoreloading_associations = {} @cache_associations = true @default_eager_limit_strategy = true @default_association_options = {} @default_association_type_options = {} @dataset_module_class = DatasetModule end end # The dataset module to use for classes using the associations plugin. class DatasetModule < Model::DatasetModule def_dataset_caching_method(self, :eager) end # AssociationReflection is a Hash subclass that keeps information on Sequel::Model associations. It # provides methods to reduce internal code duplication. It should not # be instantiated by the user. class AssociationReflection < Hash include Sequel::Inflections # Name symbol for the _add internal association method def _add_method self[:_add_method] end # Name symbol for the _remove_all internal association method def _remove_all_method self[:_remove_all_method] end # Name symbol for the _remove internal association method def _remove_method self[:_remove_method] end # Name symbol for the _setter association method def _setter_method self[:_setter_method] end # Name symbol for the add association method def add_method self[:add_method] end # Name symbol for association method, the same as the name of the association. def association_method self[:name] end # The class associated to the current model class via this association def associated_class cached_fetch(:class) do begin constantize(self[:class_name]) rescue NameError => e raise NameError, "#{e.message} (this happened when attempting to find the associated class for #{inspect})", e.backtrace end end end # The dataset associated via this association, with the non-instance specific # changes already applied. This will be a joined dataset if the association # requires joining tables. def associated_dataset cached_fetch(:_dataset){apply_dataset_changes(_associated_dataset)} end # Apply all non-instance specific changes to the given dataset and return it. def apply_dataset_changes(ds) ds = ds.with_extend(AssociationDatasetMethods).clone(:association_reflection => self) if exts = self[:reverse_extend] ds = ds.with_extend(*exts) end ds = ds.select(*select) if select if c = self[:conditions] ds = (c.is_a?(Array) && !Sequel.condition_specifier?(c)) ? ds.where(*c) : ds.where(c) end ds = ds.order(*self[:order]) if self[:order] ds = ds.limit(*self[:limit]) if self[:limit] ds = ds.limit(1).skip_limit_check if limit_to_single_row? ds = ds.eager(self[:eager]) if self[:eager] ds = ds.distinct if self[:distinct] ds end # Apply all non-instance specific changes and the eager_block option to the given # dataset and return it. def apply_eager_dataset_changes(ds) ds = apply_dataset_changes(ds) if block = self[:eager_block] ds = block.call(ds) end ds end # Apply the eager graph limit strategy to the dataset to graph into the current dataset, or return # the dataset unmodified if no SQL limit strategy is needed. def apply_eager_graph_limit_strategy(strategy, ds) case strategy when :distinct_on apply_distinct_on_eager_limit_strategy(ds.order_prepend(*self[:order])) when :window_function apply_window_function_eager_limit_strategy(ds.order_prepend(*self[:order])).select(*ds.columns) else ds end end # Apply an eager limit strategy to the dataset, or return the dataset # unmodified if it doesn't need an eager limit strategy. def apply_eager_limit_strategy(ds, strategy=eager_limit_strategy, limit_and_offset=limit_and_offset()) case strategy when :distinct_on apply_distinct_on_eager_limit_strategy(ds) when :window_function apply_window_function_eager_limit_strategy(ds, limit_and_offset) else ds end end # Use DISTINCT ON and ORDER BY clauses to limit the results to the first record with matching keys. def apply_distinct_on_eager_limit_strategy(ds) keys = predicate_key ds.distinct(*keys).order_prepend(*keys) end # Use a window function to limit the results of the eager loading dataset. def apply_window_function_eager_limit_strategy(ds, limit_and_offset=limit_and_offset()) rn = ds.row_number_column limit, offset = limit_and_offset ds = ds.unordered.select_append{|o| o.row_number.function.over(:partition=>predicate_key, :order=>ds.opts[:order]).as(rn)}.from_self ds = ds.order(rn) if ds.db.database_type == :mysql ds = if !returns_array? ds.where(rn => offset ? offset+1 : 1) elsif offset offset += 1 if limit ds.where(rn => (offset...(offset+limit))) else ds.where{SQL::Identifier.new(rn) >= offset} end else ds.where{SQL::Identifier.new(rn) <= limit} end end # If the ruby eager limit strategy is being used, slice the array using the slice # range to return the object(s) at the correct offset/limit. def apply_ruby_eager_limit_strategy(rows, limit_and_offset = limit_and_offset()) name = self[:name] if returns_array? range = slice_range(limit_and_offset) rows.each{|o| o.associations[name] = o.associations[name][range] || []} elsif sr = slice_range(limit_and_offset) offset = sr.begin rows.each{|o| o.associations[name] = o.associations[name][offset]} end end # Whether the associations cache should use an array when storing the # associated records during eager loading. def assign_singular? !returns_array? end # Whether this association can have associated objects, given the current # object. Should be false if obj cannot have associated objects because # the necessary key columns are NULL. def can_have_associated_objects?(obj) true end # Whether you are able to clone from the given association type to the current # association type, true by default only if the types match. def cloneable?(ref) ref[:type] == self[:type] end # Name symbol for the dataset association method def dataset_method self[:dataset_method] end # Whether the dataset needs a primary key to function, true by default. def dataset_need_primary_key? true end # Return the symbol used for the row number column if the window function # eager limit strategy is being used, or nil otherwise. def delete_row_number_column(ds=associated_dataset) if eager_limit_strategy == :window_function ds.row_number_column end end # Return an dataset that will load the appropriate associated objects for # the given object using this association. def association_dataset_for(object) condition = if can_have_associated_objects?(object) predicate_keys.zip(predicate_key_values(object)) else false end associated_dataset.where(condition) end ASSOCIATION_DATASET_PROC = proc{|r| r.association_dataset_for(self)} # Proc used to create the association dataset method. def association_dataset_proc ASSOCIATION_DATASET_PROC end # The eager_graph limit strategy to use for this dataset def eager_graph_limit_strategy(strategy) if self[:limit] || !returns_array? strategy = strategy[self[:name]] if strategy.is_a?(Hash) case strategy when true true_eager_graph_limit_strategy when Symbol strategy else if returns_array? || offset :ruby end end end end # The eager limit strategy to use for this dataset. def eager_limit_strategy cached_fetch(:_eager_limit_strategy) do if self[:limit] || !returns_array? case s = cached_fetch(:eager_limit_strategy){default_eager_limit_strategy} when true true_eager_limit_strategy else s end end end end # Eager load the associated objects using the hash of eager options, # yielding each row to the block. def eager_load_results(eo, &block) rows = eo[:rows] initialize_association_cache(rows) unless eo[:initialize_rows] == false if eo[:id_map] ids = eo[:id_map].keys return ids if ids.empty? end strategy = eager_limit_strategy cascade = eo[:associations] eager_limit = nil if eo[:eager_block] || eo[:loader] == false ds = eager_loading_dataset(eo) strategy = ds.opts[:eager_limit_strategy] || strategy eager_limit = if el = ds.opts[:eager_limit] raise Error, "The :eager_limit dataset option is not supported for associations returning a single record" unless returns_array? strategy ||= true_eager_graph_limit_strategy if el.is_a?(Array) el else [el, nil] end else limit_and_offset end strategy = true_eager_graph_limit_strategy if strategy == :union # Correlated subqueries are not supported for regular eager loading strategy = :ruby if strategy == :correlated_subquery strategy = nil if strategy == :ruby && assign_singular? objects = apply_eager_limit_strategy(ds, strategy, eager_limit).all elsif strategy == :union objects = [] ds = associated_dataset loader = union_eager_loader joiner = " UNION ALL " ids.each_slice(subqueries_per_union).each do |slice| objects.concat(ds.with_sql(slice.map{|k| loader.sql(*k)}.join(joiner)).to_a) end ds = ds.eager(cascade) if cascade ds.send(:post_load, objects) else loader = placeholder_eager_loader loader = loader.with_dataset{|dataset| dataset.eager(cascade)} if cascade objects = loader.all(ids) end objects.each(&block) if strategy == :ruby apply_ruby_eager_limit_strategy(rows, eager_limit || limit_and_offset) end end # The key to use for the key hash when eager loading def eager_loader_key self[:eager_loader_key] end # By default associations do not need to select a key in an associated table # to eagerly load. def eager_loading_use_associated_key? false end # Whether to eagerly graph a lazy dataset, true by default. If this # is false, the association won't respect the :eager_graph option # when loading the association for a single record. def eager_graph_lazy_dataset? true end # Whether additional conditions should be added when using the filter # by associations support. def filter_by_associations_add_conditions? self[:conditions] || self[:eager_block] || self[:limit] end # The expression to use for the additional conditions to be added for # the filter by association support, when the association itself is # filtered. Works by using a subquery to test that the objects passed # also meet the association filter criteria. def filter_by_associations_conditions_expression(obj) ds = filter_by_associations_conditions_dataset.where(filter_by_associations_conditions_subquery_conditions(obj)) {filter_by_associations_conditions_key=>ds} end # Finalize the association by first attempting to populate the thread-safe cache, # and then transfering the thread-safe cache value to the association itself, # so that a mutex is not needed to get the value. def finalize return unless cache = self[:cache] finalize_settings.each do |meth, key| next if has_key?(key) # Allow calling private methods to make sure caching is done appropriately send(meth) self[key] = cache.delete(key) if cache.has_key?(key) end nil end # Map of methods to cache keys used for finalizing associations. FINALIZE_SETTINGS = { :associated_class=>:class, :associated_dataset=>:_dataset, :associated_eager_dataset=>:associated_eager_dataset, :eager_limit_strategy=>:_eager_limit_strategy, :filter_by_associations_conditions_dataset=>:filter_by_associations_conditions_dataset, :placeholder_loader=>:placeholder_loader, :predicate_key=>:predicate_key, :predicate_keys=>:predicate_keys, :reciprocal=>:reciprocal, }.freeze def finalize_settings FINALIZE_SETTINGS end # Whether to handle silent modification failure when adding/removing # associated records, false by default. def handle_silent_modification_failure? false end # Initialize the associations cache for the current association for the given objects. def initialize_association_cache(objects) name = self[:name] if assign_singular? objects.each{|object| object.associations[name] = nil} else objects.each{|object| object.associations[name] = []} end end # Show which type of reflection this is, and a guess at what code was used to create the # association. def inspect o = self[:orig_opts].dup o.delete(:class) o.delete(:class_name) o.delete(:block) unless o[:block] o[:class] = self[:orig_class] if self[:orig_class] "#<#{self.class} #{self[:model]}.#{self[:type]} #{self[:name].inspect}#{", #{o.inspect[1...-1]}" unless o.empty?}>" end # The limit and offset for this association (returned as a two element array). def limit_and_offset if (v = self[:limit]).is_a?(Array) v else [v, nil] end end # Whether the associated object needs a primary key to be added/removed, # false by default. def need_associated_primary_key? false end # A placeholder literalizer that can be used to lazily load the association. If # one can't be used, returns nil. def placeholder_loader if use_placeholder_loader? cached_fetch(:placeholder_loader) do Sequel::Dataset::PlaceholderLiteralizer.loader(associated_dataset) do |pl, ds| ds.where(Sequel.&(*predicate_keys.map{|k| SQL::BooleanExpression.new(:'=', k, pl.arg)})) end end end end # The keys to use for loading of the regular dataset, as an array. def predicate_keys cached_fetch(:predicate_keys){Array(predicate_key)} end # The values that predicate_keys should match for objects to be associated. def predicate_key_values(object) predicate_key_methods.map{|k| object.get_column_value(k)} end # Qualify +col+ with the given table name. def qualify(table, col) transform(col) do |k| case k when Symbol, SQL::Identifier SQL::QualifiedIdentifier.new(table, k) else Sequel::Qualifier.new(table).transform(k) end end end # Qualify col with the associated model's table name. def qualify_assoc(col) qualify(associated_class.table_name, col) end # Qualify col with the current model's table name. def qualify_cur(col) qualify(self[:model].table_name, col) end # Returns the reciprocal association variable, if one exists. The reciprocal # association is the association in the associated class that is the opposite # of the current association. For example, Album.many_to_one :artist and # Artist.one_to_many :albums are reciprocal associations. This information is # to populate reciprocal associations. For example, when you do this_artist.add_album(album) # it sets album.artist to this_artist. def reciprocal cached_fetch(:reciprocal) do possible_recips = [] associated_class.all_association_reflections.each do |assoc_reflect| if reciprocal_association?(assoc_reflect) possible_recips << assoc_reflect end end if possible_recips.length == 1 cached_set(:reciprocal_type, possible_recips.first[:type]) if ambiguous_reciprocal_type? possible_recips.first[:name] end end end # Whether the reciprocal of this association returns an array of objects instead of a single object, # true by default. def reciprocal_array? true end # Name symbol for the remove_all_ association method def remove_all_method self[:remove_all_method] end # Whether associated objects need to be removed from the association before # being destroyed in order to preserve referential integrity. def remove_before_destroy? true end # Name symbol for the remove_ association method def remove_method self[:remove_method] end # Whether to check that an object to be disassociated is already associated to this object, false by default. def remove_should_check_existing? false end # Whether this association returns an array of objects instead of a single object, # true by default. def returns_array? true end # The columns to select when loading the association. def select self[:select] end # Whether to set the reciprocal association to self when loading associated # records, false by default. def set_reciprocal_to_self? false end # Name symbol for the setter association method def setter_method self[:setter_method] end # The range used for slicing when using the :ruby eager limit strategy. def slice_range(limit_and_offset = limit_and_offset()) limit, offset = limit_and_offset if limit || offset (offset||0)..(limit ? (offset||0)+limit-1 : -1) end end private # If the key exists in the reflection hash, return it. # If the key doesn't exist and association reflections are uncached, then yield to get the value. # If the key doesn't exist and association reflection are cached, check the cache and return # the value if present, or yield to get the value, cache the value, and return it. def cached_fetch(key) fetch(key) do return yield unless h = self[:cache] Sequel.synchronize{return h[key] if h.has_key?(key)} value = yield Sequel.synchronize{h[key] = value} end end # Cache the value at the given key if caching. def cached_set(key, value) return unless h = self[:cache] Sequel.synchronize{h[key] = value} end # The base dataset used for the association, before any order/conditions # options have been applied. def _associated_dataset associated_class.dataset end # Whether for the reciprocal type for the given association cannot be # known in advantage, false by default. def ambiguous_reciprocal_type? false end # Apply a limit strategy to the given dataset so that filter by # associations works with a limited dataset. def apply_filter_by_associations_limit_strategy(ds) case filter_by_associations_limit_strategy when :distinct_on apply_filter_by_associations_distinct_on_limit_strategy(ds) when :window_function apply_filter_by_associations_window_function_limit_strategy(ds) else ds end end # Apply a distinct on eager limit strategy using IN with a subquery # that uses DISTINCT ON to ensure only the first matching record for # each key is included. def apply_filter_by_associations_distinct_on_limit_strategy(ds) k = filter_by_associations_limit_key ds.where(k=>apply_distinct_on_eager_limit_strategy(associated_eager_dataset.select(*k))) end # Apply a distinct on eager limit strategy using IN with a subquery # that uses a filter on the row_number window function to ensure # that only rows inside the limit are returned. def apply_filter_by_associations_window_function_limit_strategy(ds) ds.where(filter_by_associations_limit_key=>apply_window_function_eager_limit_strategy(associated_eager_dataset.select(*filter_by_associations_limit_alias_key)).select(*filter_by_associations_limit_aliases)) end # The associated_dataset with the eager_block callback already applied. def associated_eager_dataset cached_fetch(:associated_eager_dataset) do ds = associated_dataset.unlimited if block = self[:eager_block] ds = block.call(ds) end ds end end # The dataset to use for eager loading associated objects for multiple current objects, # given the hash passed to the eager loader. def eager_loading_dataset(eo=OPTS) ds = eo[:dataset] || associated_eager_dataset if id_map = eo[:id_map] ds = ds.where(eager_loading_predicate_condition(id_map.keys)) end if associations = eo[:associations] ds = ds.eager(associations) end if block = eo[:eager_block] orig_ds = ds ds = block.call(ds) end if eager_loading_use_associated_key? ds = if ds.opts[:eager_graph] && !orig_ds.opts[:eager_graph] block.call(orig_ds.select_append(*associated_key_array)) else ds.select_append(*associated_key_array) end end if self[:eager_graph] raise(Error, "cannot eagerly load a #{self[:type]} association that uses :eager_graph") if eager_loading_use_associated_key? ds = ds.eager_graph(self[:eager_graph]) end ds end # The default eager limit strategy to use for this association def default_eager_limit_strategy self[:model].default_eager_limit_strategy || :ruby end # The predicate condition to use for the eager_loader. def eager_loading_predicate_condition(keys) {predicate_key=>keys} end # Add conditions to the dataset to not include NULL values for # the associated keys, and select those keys. def filter_by_associations_add_conditions_dataset_filter(ds) k = filter_by_associations_conditions_associated_keys ds.select(*k).where(Sequel.negate(k.zip([]))) end # The conditions to add to the filter by associations conditions # subquery to restrict it to to the object(s) that was used as the # filter value. def filter_by_associations_conditions_subquery_conditions(obj) key = qualify(associated_class.table_name, associated_class.primary_key) case obj when Array {key=>obj.map(&:pk)} when Sequel::Dataset {key=>obj.select(*Array(qualify(associated_class.table_name, associated_class.primary_key)))} else Array(key).zip(Array(obj.pk)) end end # The base dataset to use for the filter by associations conditions # subquery, regardless of the objects that are passed in as filter # values. def filter_by_associations_conditions_dataset cached_fetch(:filter_by_associations_conditions_dataset) do ds = associated_eager_dataset.unordered ds = filter_by_associations_add_conditions_dataset_filter(ds) ds = apply_filter_by_associations_limit_strategy(ds) ds end end # The strategy to use to filter by a limited association def filter_by_associations_limit_strategy v = fetch(:filter_limit_strategy, self[:eager_limit_strategy]) if v || self[:limit] || !returns_array? case v ||= self[:model].default_eager_limit_strategy when true, :union, :ruby # Can't use a union or ruby-based strategy for filtering by associations, switch to default eager graph limit # strategy. true_eager_graph_limit_strategy when Symbol v end end end # Whether to limit the associated dataset to a single row. def limit_to_single_row? !returns_array? end # Any offset to use for this association (or nil if there is no offset). def offset limit_and_offset.last end # A placeholder literalizer used to speed up eager loading. def placeholder_eager_loader cached_fetch(:placeholder_eager_loader) do Sequel::Dataset::PlaceholderLiteralizer.loader(associated_dataset) do |pl, ds| apply_eager_limit_strategy(eager_loading_dataset.where(predicate_key=>pl.arg), eager_limit_strategy) end end end # The reciprocal type as an array, should be overridden in reflection subclasses that # have ambiguous reciprocal types. def possible_reciprocal_types [reciprocal_type] end # Whether the given association reflection is possible reciprocal # association for the current association reflection. def reciprocal_association?(assoc_reflect) possible_reciprocal_types.include?(assoc_reflect[:type]) && (begin; assoc_reflect.associated_class; rescue NameError; end) == self[:model] && assoc_reflect[:conditions].nil? && assoc_reflect[:block].nil? end # The number of subqueries to use in each union query, used to eagerly load # limited associations. Defaults to 40, the optimal number depends on the # latency between the database and the application. def subqueries_per_union self[:subqueries_per_union] || 40 end # If +s+ is an array, map +s+ over the block. Otherwise, just call the # block with +s+. def transform(s) s.is_a?(Array) ? s.map(&Proc.new) : (yield s) end # What eager limit strategy should be used when true is given as the value, # defaults to UNION as that is the fastest strategy if the appropriate keys are indexed. def true_eager_limit_strategy if self[:eager_graph] || (offset && !associated_dataset.supports_offsets_in_correlated_subqueries?) # An SQL-based approach won't work if you are also eager graphing, # so use a ruby based approach in that case. :ruby else :union end end # The eager_graph limit strategy used when true is given as the value, choosing the # best strategy based on what the database supports. def true_eager_graph_limit_strategy if associated_class.dataset.supports_window_functions? :window_function else :ruby end end # A placeholder literalizer used to speed up the creation of union queries when eager # loading a limited association. def union_eager_loader cached_fetch(:union_eager_loader) do Sequel::Dataset::PlaceholderLiteralizer.loader(associated_dataset) do |pl, ds| ds = self[:eager_block].call(ds) if self[:eager_block] keys = predicate_keys ds = ds.where(keys.map{pl.arg}.zip(keys)) if eager_loading_use_associated_key? ds = ds.select_append(*associated_key_array) end ds.from_self end end end # Whether the placeholder loader can be used to load the association. def use_placeholder_loader? !self[:instance_specific] && !self[:eager_graph] end end class ManyToOneAssociationReflection < AssociationReflection ASSOCIATION_TYPES[:many_to_one] = self # many_to_one associations can only have associated objects if none of # the :keys options have a nil value. def can_have_associated_objects?(obj) !self[:keys].any?{|k| obj.get_column_value(k).nil?} end # Whether the dataset needs a primary key to function, false for many_to_one associations. def dataset_need_primary_key? false end # Default foreign key name symbol for foreign key in current model's table that points to # the given association's table's primary key. def default_key :"#{self[:name]}_id" end # Whether to eagerly graph a lazy dataset, true for many_to_one associations # only if the key is nil. def eager_graph_lazy_dataset? self[:key].nil? end # many_to_one associations don't need an eager_graph limit strategy def eager_graph_limit_strategy(_) nil end # many_to_one associations don't need an eager limit strategy def eager_limit_strategy nil end # many_to_one associations don't need a filter by associations limit strategy def filter_by_associations_limit_strategy nil end FINALIZE_SETTINGS = superclass::FINALIZE_SETTINGS.merge( :primary_key=>:primary_key, :primary_keys=>:primary_keys, :primary_key_method=>:primary_key_method, :primary_key_methods=>:primary_key_methods, :qualified_primary_key=>:qualified_primary_key, :reciprocal_type=>:reciprocal_type ).freeze def finalize_settings FINALIZE_SETTINGS end # The expression to use on the left hand side of the IN lookup when eager loading def predicate_key cached_fetch(:predicate_key){qualified_primary_key} end # The column(s) in the associated table that the key in the current table references (either a symbol or an array). def primary_key cached_fetch(:primary_key){associated_class.primary_key || raise(Error, "no primary key specified for #{associated_class.inspect}")} end # The columns in the associated table that the key in the current table references (always an array). def primary_keys cached_fetch(:primary_keys){Array(primary_key)} end alias associated_object_keys primary_keys # The method symbol or array of method symbols to call on the associated object # to get the value to use for the foreign keys. def primary_key_method cached_fetch(:primary_key_method){primary_key} end # The array of method symbols to call on the associated object # to get the value to use for the foreign keys. def primary_key_methods cached_fetch(:primary_key_methods){Array(primary_key_method)} end # #primary_key qualified by the associated table def qualified_primary_key cached_fetch(:qualified_primary_key){self[:qualify] == false ? primary_key : qualify_assoc(primary_key)} end # True only if the reciprocal is a one_to_many association. def reciprocal_array? !set_reciprocal_to_self? end # Whether this association returns an array of objects instead of a single object, # false for a many_to_one association. def returns_array? false end # True only if the reciprocal is a one_to_one association. def set_reciprocal_to_self? reciprocal reciprocal_type == :one_to_one end private # Reciprocals of many_to_one associations could be either one_to_many or one_to_one, # and which is not known in advance. def ambiguous_reciprocal_type? true end def filter_by_associations_conditions_associated_keys qualify(associated_class.table_name, primary_keys) end def filter_by_associations_conditions_key qualify(self[:model].table_name, self[:key_column]) end # many_to_one associations do not need to be limited to a single row if they # explicitly do not have a key. def limit_to_single_row? super && self[:key] end def predicate_key_methods self[:keys] end # The reciprocal type of a many_to_one association is either # a one_to_many or a one_to_one association. def possible_reciprocal_types [:one_to_many, :one_to_one] end # Whether the given association reflection is possible reciprocal def reciprocal_association?(assoc_reflect) super && self[:keys] == assoc_reflect[:keys] && primary_key == assoc_reflect.primary_key end # The reciprocal type of a many_to_one association is either # a one_to_many or a one_to_one association, look in the associated class # to try to figure out which. def reciprocal_type cached_fetch(:reciprocal_type) do possible_recips = [] associated_class.all_association_reflections.each do |assoc_reflect| if reciprocal_association?(assoc_reflect) possible_recips << assoc_reflect end end if possible_recips.length == 1 possible_recips.first[:type] else possible_reciprocal_types end end end end class OneToManyAssociationReflection < AssociationReflection ASSOCIATION_TYPES[:one_to_many] = self # Support a correlated subquery limit strategy when using eager_graph. def apply_eager_graph_limit_strategy(strategy, ds) case strategy when :correlated_subquery apply_correlated_subquery_limit_strategy(ds) else super end end # The keys in the associated model's table related to this association def associated_object_keys self[:keys] end # one_to_many associations can only have associated objects if none of # the :keys options have a nil value. def can_have_associated_objects?(obj) !self[:primary_keys].any?{|k| obj.get_column_value(k).nil?} end # one_to_many and one_to_one associations can be clones def cloneable?(ref) ref[:type] == :one_to_many || ref[:type] == :one_to_one end # Default foreign key name symbol for key in associated table that points to # current table's primary key. def default_key :"#{underscore(demodulize(self[:model].name))}_id" end FINALIZE_SETTINGS = superclass::FINALIZE_SETTINGS.merge( :qualified_primary_key=>:qualified_primary_key ).freeze def finalize_settings FINALIZE_SETTINGS end # Handle silent failure of add/remove methods if raise_on_save_failure is false. def handle_silent_modification_failure? self[:raise_on_save_failure] == false end # The hash key to use for the eager loading predicate (left side of IN (1, 2, 3)) def predicate_key cached_fetch(:predicate_key){qualify_assoc(self[:key])} end alias qualified_key predicate_key # The column in the current table that the key in the associated table references. def primary_key self[:primary_key] end # #primary_key qualified by the current table def qualified_primary_key cached_fetch(:qualified_primary_key){qualify_cur(primary_key)} end # Whether the reciprocal of this association returns an array of objects instead of a single object, # false for a one_to_many association. def reciprocal_array? false end # Destroying one_to_many associated objects automatically deletes the foreign key. def remove_before_destroy? false end # The one_to_many association needs to check that an object to be removed already is associated. def remove_should_check_existing? true end # One to many associations set the reciprocal to self when loading associated records. def set_reciprocal_to_self? true end private # Use a correlated subquery to limit the dataset. Note that this will not # work correctly if the associated dataset uses qualified identifers in the WHERE clause, # as they would reference the containing query instead of the subquery. def apply_correlated_subquery_limit_strategy(ds) table = ds.first_source_table table_alias = ds.first_source_alias primary_key = associated_class.primary_key key = self[:key] cs_alias = :t1 cs = associated_dataset. from(Sequel.as(table, :t1)). select(*qualify(cs_alias, primary_key)). where(Array(qualify(cs_alias, key)).zip(Array(qualify(table_alias, key)))). limit(*limit_and_offset) ds.where(qualify(table_alias, primary_key)=>cs) end # Support correlated subquery strategy when filtering by limited associations. def apply_filter_by_associations_limit_strategy(ds) case filter_by_associations_limit_strategy when :correlated_subquery apply_correlated_subquery_limit_strategy(ds) else super end end def filter_by_associations_conditions_associated_keys qualify(associated_class.table_name, self[:keys]) end def filter_by_associations_conditions_key qualify(self[:model].table_name, self[:primary_key_column]) end def filter_by_associations_limit_alias_key Array(filter_by_associations_limit_key) end def filter_by_associations_limit_aliases filter_by_associations_limit_alias_key.map(&:column) end def filter_by_associations_limit_key qualify(associated_class.table_name, associated_class.primary_key) end def predicate_key_methods self[:primary_keys] end def reciprocal_association?(assoc_reflect) super && self[:keys] == assoc_reflect[:keys] && primary_key == assoc_reflect.primary_key end # The reciprocal type of a one_to_many association is a many_to_one association. def reciprocal_type :many_to_one end # Support automatic use of correlated subqueries if :ruby option is best available option, # the database supports them, and either the associated class has a non-composite primary key # or the database supports multiple columns in IN. def true_eager_graph_limit_strategy r = super ds = associated_dataset if r == :ruby && ds.supports_limits_in_correlated_subqueries? && (Array(associated_class.primary_key).length == 1 || ds.supports_multiple_column_in?) && (!offset || ds.supports_offsets_in_correlated_subqueries?) :correlated_subquery else r end end end # Methods that turn an association that returns multiple objects into an association that # returns a single object. module SingularAssociationReflection # Singular associations do not assign singular if they are using the ruby eager limit strategy # and have a slice range, since they need to store the array of associated objects in order to # pick the correct one with an offset. def assign_singular? super && (eager_limit_strategy != :ruby || !slice_range) end # Add conditions when filtering by singular associations with orders, since the # underlying relationship is probably not one-to-one. def filter_by_associations_add_conditions? super || self[:order] || self[:eager_limit_strategy] || self[:filter_limit_strategy] end # Make sure singular associations always have 1 as the limit def limit_and_offset r = super if r.first == 1 r else [1, r[1]] end end # Singular associations always return a single object, not an array. def returns_array? false end private # Only use a eager limit strategy by default if there is an offset or an order. def default_eager_limit_strategy super if self[:order] || offset end # Use a strategy for filtering by associations if there is an order or an offset, # or a specific limiting strategy has been specified. def filter_by_associations_limit_strategy super if self[:order] || offset || self[:eager_limit_strategy] || self[:filter_limit_strategy] end # Use the DISTINCT ON eager limit strategy for true if the database supports it. def true_eager_graph_limit_strategy if associated_class.dataset.supports_ordered_distinct_on? && !offset :distinct_on else super end end end class OneToOneAssociationReflection < OneToManyAssociationReflection ASSOCIATION_TYPES[:one_to_one] = self include SingularAssociationReflection end class ManyToManyAssociationReflection < AssociationReflection ASSOCIATION_TYPES[:many_to_many] = self # The alias to use for the associated key when eagerly loading def associated_key_alias self[:left_key_alias] end # Array of associated keys used when eagerly loading. def associated_key_array cached_fetch(:associated_key_array) do if self[:uses_left_composite_keys] associated_key_alias.zip(predicate_keys).map{|a, k| SQL::AliasedExpression.new(k, a)} else [SQL::AliasedExpression.new(predicate_key, associated_key_alias)] end end end # The column to use for the associated key when eagerly loading def associated_key_column self[:left_key] end # Alias of right_primary_keys def associated_object_keys right_primary_keys end # many_to_many associations can only have associated objects if none of # the :left_primary_keys options have a nil value. def can_have_associated_objects?(obj) !self[:left_primary_keys].any?{|k| obj.get_column_value(k).nil?} end # one_through_one and many_to_many associations can be clones def cloneable?(ref) ref[:type] == :many_to_many || ref[:type] == :one_through_one end # The default associated key alias(es) to use when eager loading # associations via eager. def default_associated_key_alias self[:uses_left_composite_keys] ? (0...self[:left_keys].length).map{|i| :"x_foreign_key_#{i}_x"} : :x_foreign_key_x end # The default eager loader used if the user doesn't override it. Extracted # to a method so the code can be shared with the many_through_many plugin. def default_eager_loader(eo) h = eo[:id_map] assign_singular = assign_singular? delete_rn = delete_row_number_column uses_lcks = self[:uses_left_composite_keys] left_key_alias = self[:left_key_alias] name = self[:name] self[:model].eager_load_results(self, eo) do |assoc_record| assoc_record.values.delete(delete_rn) if delete_rn hash_key = if uses_lcks left_key_alias.map{|k| assoc_record.values.delete(k)} else assoc_record.values.delete(left_key_alias) end next unless objects = h[hash_key] if assign_singular objects.each do |object| object.associations[name] ||= assoc_record end else objects.each do |object| object.associations[name].push(assoc_record) end end end end # Default name symbol for the join table. def default_join_table [self[:class_name], self[:model].name].map{|i| underscore(pluralize(demodulize(i)))}.sort.join('_').to_sym end # Default foreign key name symbol for key in join table that points to # current table's primary key (or :left_primary_key column). def default_left_key :"#{underscore(demodulize(self[:model].name))}_id" end # Default foreign key name symbol for foreign key in join table that points to # the association's table's primary key (or :right_primary_key column). def default_right_key :"#{singularize(self[:name])}_id" end FINALIZE_SETTINGS = superclass::FINALIZE_SETTINGS.merge( :associated_key_array=>:associated_key_array, :qualified_right_key=>:qualified_right_key, :join_table_source=>:join_table_source, :join_table_alias=>:join_table_alias, :qualified_right_primary_key=>:qualified_right_primary_key, :right_primary_key=>:right_primary_key, :right_primary_keys=>:right_primary_keys, :right_primary_key_method=>:right_primary_key_method, :right_primary_key_methods=>:right_primary_key_methods, :select=>:select ).freeze def finalize_settings FINALIZE_SETTINGS end # The hash key to use for the eager loading predicate (left side of IN (1, 2, 3)). # The left key qualified by the join table. def predicate_key cached_fetch(:predicate_key){qualify(join_table_alias, self[:left_key])} end alias qualified_left_key predicate_key # The right key qualified by the join table. def qualified_right_key cached_fetch(:qualified_right_key){qualify(join_table_alias, self[:right_key])} end # many_to_many associations need to select a key in an associated table to eagerly load def eager_loading_use_associated_key? true end # The source of the join table. This is the join table itself, unless it # is aliased, in which case it is the unaliased part. def join_table_source cached_fetch(:join_table_source){split_join_table_alias[0]} end # The join table itself, unless it is aliased, in which case this # is the alias. def join_table_alias cached_fetch(:join_table_alias) do s, a = split_join_table_alias a || s end end alias associated_key_table join_table_alias # Whether the associated object needs a primary key to be added/removed, # true for many_to_many associations. def need_associated_primary_key? true end # #right_primary_key qualified by the associated table def qualified_right_primary_key cached_fetch(:qualified_right_primary_key){qualify_assoc(right_primary_key)} end # The primary key column(s) to use in the associated table (can be symbol or array). def right_primary_key cached_fetch(:right_primary_key){associated_class.primary_key || raise(Error, "no primary key specified for #{associated_class.inspect}")} end # The primary key columns to use in the associated table (always array). def right_primary_keys cached_fetch(:right_primary_keys){Array(right_primary_key)} end # The method symbol or array of method symbols to call on the associated objects # to get the foreign key values for the join table. def right_primary_key_method cached_fetch(:right_primary_key_method){right_primary_key} end # The array of method symbols to call on the associated objects # to get the foreign key values for the join table. def right_primary_key_methods cached_fetch(:right_primary_key_methods){Array(right_primary_key_method)} end # The columns to select when loading the association, associated_class.table_name.* by default. def select cached_fetch(:select){default_select} end private def _associated_dataset super.inner_join(self[:join_table], self[:right_keys].zip(right_primary_keys), :qualify=>:deep) end # The default selection for associations that require joins. These do not use the default # model selection unless all entries in the select are explicitly qualified identifiers, as # other it can include unqualified columns which would be made ambiguous by joining. def default_select if (sel = associated_class.dataset.opts[:select]) && sel.all?{|c| selection_is_qualified?(c)} sel else Sequel::SQL::ColumnAll.new(associated_class.table_name) end end def filter_by_associations_conditions_associated_keys qualify(join_table_alias, self[:left_keys]) end def filter_by_associations_conditions_key qualify(self[:model].table_name, self[:left_primary_key_column]) end def filter_by_associations_limit_alias_key aliaz = 'a' filter_by_associations_limit_key.map{|c| c.as(Sequel.identifier(aliaz = aliaz.next))} end def filter_by_associations_limit_aliases filter_by_associations_limit_alias_key.map(&:alias) end def filter_by_associations_limit_key qualify(join_table_alias, self[:left_keys]) + Array(qualify(associated_class.table_name, associated_class.primary_key)) end def predicate_key_methods self[:left_primary_keys] end def reciprocal_association?(assoc_reflect) super && assoc_reflect[:left_keys] == self[:right_keys] && assoc_reflect[:right_keys] == self[:left_keys] && assoc_reflect[:join_table] == self[:join_table] && right_primary_keys == assoc_reflect[:left_primary_key_columns] && self[:left_primary_key_columns] == assoc_reflect.right_primary_keys end def reciprocal_type :many_to_many end # Whether the given expression represents a qualified identifier. Used to determine if it is # OK to use directly when joining. def selection_is_qualified?(c) case c when Symbol Sequel.split_symbol(c)[0] when Sequel::SQL::QualifiedIdentifier true when Sequel::SQL::AliasedExpression selection_is_qualified?(c.expression) else false end end # Split the join table into source and alias parts. def split_join_table_alias associated_class.dataset.split_alias(self[:join_table]) end end class OneThroughOneAssociationReflection < ManyToManyAssociationReflection ASSOCIATION_TYPES[:one_through_one] = self include SingularAssociationReflection # one_through_one associations should not singularize the association name when # creating the foreign key. def default_right_key :"#{self[:name]}_id" end # one_through_one associations have no reciprocals def reciprocal nil end end # This module contains methods added to all association datasets module AssociationDatasetMethods # The model object that created the association dataset def model_object @opts[:model_object] end # The association reflection related to the association dataset def association_reflection @opts[:association_reflection] end private def non_sql_option?(key) super || key == :model_object || key == :association_reflection end end # Each kind of association adds a number of instance methods to the model class which # are specialized according to the association type and optional parameters # given in the definition. Example: # # class Project < Sequel::Model # many_to_one :portfolio # # or: one_to_one :portfolio # one_to_many :milestones # # or: many_to_many :milestones # end # # The project class now has the following instance methods: # portfolio :: Returns the associated portfolio. # portfolio=(obj) :: Sets the associated portfolio to the object, # but the change is not persisted until you save the record (for many_to_one associations). # portfolio_dataset :: Returns a dataset that would return the associated # portfolio, only useful in fairly specific circumstances. # milestones :: Returns an array of associated milestones # add_milestone(obj) :: Associates the passed milestone with this object. # remove_milestone(obj) :: Removes the association with the passed milestone. # remove_all_milestones :: Removes associations with all associated milestones. # milestones_dataset :: Returns a dataset that would return the associated # milestones, allowing for further filtering/limiting/etc. # # If you want to override the behavior of the add_/remove_/remove_all_/ methods # or the association setter method, use the :adder, :remover, :clearer, and/or :setter # options. These options override the default behavior. # # By default the classes for the associations are inferred from the association # name, so for example the Project#portfolio will return an instance of # Portfolio, and Project#milestones will return an array of Milestone # instances. You can use the :class option to change which class is used. # # Association definitions are also reflected by the class, e.g.: # # Project.associations # => [:portfolio, :milestones] # Project.association_reflection(:portfolio) # => # # # Associations should not have the same names as any of the columns in the # model's current table they reference. If you are dealing with an existing schema that # has a column named status, you can't name the association status, you'd # have to name it foo_status or something else. If you give an association the same name # as a column, you will probably end up with an association that doesn't work, or a SystemStackError. # # For a more in depth general overview, as well as a reference guide, # see the {Association Basics guide}[rdoc-ref:doc/association_basics.rdoc]. # For examples of advanced usage, see the {Advanced Associations guide}[rdoc-ref:doc/advanced_associations.rdoc]. module ClassMethods # All association reflections defined for this model (default: {}). attr_reader :association_reflections # Hash with column symbol keys and arrays of many_to_one # association symbols that should be cleared when the column # value changes. attr_reader :autoreloading_associations # Whether association metadata should be cached in the association reflection. If not cached, it will be computed # on demand. In general you only want to set this to false when using code reloading. When using code reloading, # setting this will make sure that if an associated class is removed or modified, this class will not have a reference to # the previous class. attr_accessor :cache_associations # The default options to use for all associations. This hash is merged into the association reflection hash for # all association reflections. attr_accessor :default_association_options # The default options to use for all associations of a given type. This is a hash keyed by association type # symbol. If there is a value for the association type symbol key, the resulting hash will be merged into the # association reflection hash for all association reflections of that type. attr_accessor :default_association_type_options # The default :eager_limit_strategy option to use for limited or offset associations (default: true, causing Sequel # to use what it considers the most appropriate strategy). attr_accessor :default_eager_limit_strategy # Array of all association reflections for this model class def all_association_reflections association_reflections.values end # Associates a related model with the current model. The following types are # supported: # # :many_to_one :: Foreign key in current model's table points to # associated model's primary key. Each associated model object can # be associated with more than one current model objects. Each current # model object can be associated with only one associated model object. # :one_to_many :: Foreign key in associated model's table points to this # model's primary key. Each current model object can be associated with # more than one associated model objects. Each associated model object # can be associated with only one current model object. # :one_through_one :: Similar to many_to_many in terms of foreign keys, but only one object # is associated to the current object through the association. # Provides only getter methods, no setter or modification methods. # :one_to_one :: Similar to one_to_many in terms of foreign keys, but # only one object is associated to the current object through the # association. The methods created are similar to many_to_one, except # that the one_to_one setter method saves the passed object. # :many_to_many :: A join table is used that has a foreign key that points # to this model's primary key and a foreign key that points to the # associated model's primary key. Each current model object can be # associated with many associated model objects, and each associated # model object can be associated with many current model objects. # # The following options can be supplied: # === Multiple Types # :adder :: Proc used to define the private _add_* method for doing the database work # to associate the given object to the current object (*_to_many assocations). # :after_add :: Symbol, Proc, or array of both/either specifying a callback to call # after a new item is added to the association. # :after_load :: Symbol, Proc, or array of both/either specifying a callback to call # after the associated record(s) have been retrieved from the database. # :after_remove :: Symbol, Proc, or array of both/either specifying a callback to call # after an item is removed from the association. # :after_set :: Symbol, Proc, or array of both/either specifying a callback to call # after an item is set using the association setter method. # :allow_eager :: If set to false, you cannot load the association eagerly # via eager or eager_graph # :before_add :: Symbol, Proc, or array of both/either specifying a callback to call # before a new item is added to the association. # :before_remove :: Symbol, Proc, or array of both/either specifying a callback to call # before an item is removed from the association. # :before_set :: Symbol, Proc, or array of both/either specifying a callback to call # before an item is set using the association setter method. # :cartesian_product_number :: the number of joins completed by this association that could cause more # than one row for each row in the current table (default: 0 for # many_to_one, one_to_one, and one_through_one associations, 1 # for one_to_many and many_to_many associations). # :class :: The associated class or its name as a string or symbol. If not # given, uses the association's name, which is camelized (and # singularized unless the type is :many_to_one, :one_to_one, or one_through_one). If this is specified # as a string or symbol, you must specify the full class name (e.g. "::SomeModule::MyModel"). # :class_namespace :: If :class is given as a string or symbol, sets the default namespace in which to look for # the class. class: 'Foo', class_namespace: 'Bar' looks for ::Bar::Foo.) # :clearer :: Proc used to define the private _remove_all_* method for doing the database work # to remove all objects associated to the current object (*_to_many assocations). # :clone :: Merge the current options and block into the options and block used in defining # the given association. Can be used to DRY up a bunch of similar associations that # all share the same options such as :class and :key, while changing the order and block used. # :conditions :: The conditions to use to filter the association, can be any argument passed to where. # This option is not respected when using eager_graph or association_join, unless it # is hash or array of two element arrays. Consider also specifying the :graph_block # option if the value for this option is not a hash or array of two element arrays # and you plan to use this association in eager_graph or association_join. # :dataset :: A proc that is instance_execed to get the base dataset to use (before the other # options are applied). If the proc accepts an argument, it is passed the related # association reflection. # :distinct :: Use the DISTINCT clause when selecting associating object, both when # lazy loading and eager loading via .eager (but not when using .eager_graph). # :eager :: The associations to eagerly load via +eager+ when loading the associated object(s). # :eager_block :: If given, use the block instead of the default block when # eagerly loading. To not use a block when eager loading (when one is used normally), # set to nil. # :eager_graph :: The associations to eagerly load via +eager_graph+ when loading the associated object(s). # many_to_many associations with this option cannot be eagerly loaded via +eager+. # :eager_grapher :: A proc to use to implement eager loading via +eager_graph+, overriding the default. # Takes an options hash with at least the entries :self (the receiver of the eager_graph call), # :table_alias (the alias to use for table to graph into the association), and :implicit_qualifier # (the alias that was used for the current table). # Should return a copy of the dataset with the association graphed into it. # :eager_limit_strategy :: Determines the strategy used for enforcing limits and offsets when eager loading # associations via the +eager+ method. # :eager_loader :: A proc to use to implement eager loading, overriding the default. Takes a single hash argument, # with at least the keys: :rows, which is an array of current model instances, :associations, # which is a hash of dependent associations, :self, which is the dataset doing the eager loading, # :eager_block, which is a dynamic callback that should be called with the dataset, and :id_map, # which is a mapping of key values to arrays of current model instances. In the proc, the # associated records should be queried from the database and the associations cache for each # record should be populated. # :eager_loader_key :: A symbol for the key column to use to populate the key_hash # for the eager loader. Can be set to nil to not populate the key_hash. # :extend :: A module or array of modules to extend the dataset with. # :filter_limit_strategy :: Determines the strategy used for enforcing limits and offsets when filtering by # limited associations. Possible options are :window_function, :distinct_on, or # :correlated_subquery depending on association type and database type. # :graph_alias_base :: The base name to use for the table alias when eager graphing. Defaults to the name # of the association. If the alias name has already been used in the query, Sequel will create # a unique alias by appending a numeric suffix (e.g. alias_0, alias_1, ...) until the alias is # unique. # :graph_block :: The block to pass to join_table when eagerly loading # the association via +eager_graph+. # :graph_conditions :: The additional conditions to use on the SQL join when eagerly loading # the association via +eager_graph+. Should be a hash or an array of two element arrays. If not # specified, the :conditions option is used if it is a hash or array of two element arrays. # :graph_join_type :: The type of SQL join to use when eagerly loading the association via # eager_graph. Defaults to :left_outer. # :graph_only_conditions :: The conditions to use on the SQL join when eagerly loading # the association via +eager_graph+, instead of the default conditions specified by the # foreign/primary keys. This option causes the :graph_conditions option to be ignored. # :graph_order :: Over the order to use when using eager_graph, instead of the default order. This should be used # in the case where :order contains an identifier qualified by the table's name, which may not match # the alias used when eager graphing. By setting this to the unqualified identifier, it will be # automatically qualified when using eager_graph. # :graph_select :: A column or array of columns to select from the associated table # when eagerly loading the association via +eager_graph+. Defaults to all # columns in the associated table. # :limit :: Limit the number of records to the provided value. Use # an array with two elements for the value to specify a # limit (first element) and an offset (second element). # :methods_module :: The module that methods the association creates will be placed into. Defaults # to the module containing the model's columns. # :order :: the column(s) by which to order the association dataset. Can be a # singular column symbol or an array of column symbols. # :order_eager_graph :: Whether to add the association's order to the graphed dataset's order when graphing # via +eager_graph+. Defaults to true, so set to false to disable. # :read_only :: Do not add a setter method (for many_to_one or one_to_one associations), # or add_/remove_/remove_all_ methods (for one_to_many and many_to_many associations). # :reciprocal :: the symbol name of the reciprocal association, # if it exists. By default, Sequel will try to determine it by looking at the # associated model's assocations for a association that matches # the current association's key(s). Set to nil to not use a reciprocal. # :remover :: Proc used to define the private _remove_* method for doing the database work # to remove the association between the given object and the current object (*_to_many assocations). # :select :: the columns to select. Defaults to the associated class's table_name.* in an association # that uses joins, which means it doesn't include the attributes from the # join table. If you want to include the join table attributes, you can # use this option, but beware that the join table attributes can clash with # attributes from the model table, so you should alias any attributes that have # the same name in both the join table and the associated table. # :setter :: Proc used to define the private _*= method for doing the work to setup the assocation # between the given object and the current object (*_to_one associations). # :subqueries_per_union :: The number of subqueries to use in each UNION query, for eager # loading limited associations using the default :union strategy. # :validate :: Set to false to not validate when implicitly saving any associated object. # === :many_to_one # :key :: foreign key in current model's table that references # associated model's primary key, as a symbol. Defaults to :"#{name}_id". Can use an # array of symbols for a composite key association. # :key_column :: Similar to, and usually identical to, :key, but :key refers to the model method # to call, where :key_column refers to the underlying column. Should only be # used if the model method differs from the foreign key column, in conjunction # with defining a model alias method for the key column. # :primary_key :: column in the associated table that :key option references, as a symbol. # Defaults to the primary key of the associated table. Can use an # array of symbols for a composite key association. # :primary_key_method :: the method symbol or array of method symbols to call on the associated # object to get the foreign key values. Defaults to :primary_key option. # :qualify :: Whether to use qualified primary keys when loading the association. The default # is true, so you must set to false to not qualify. Qualification rarely causes # problems, but it's necessary to disable in some cases, such as when you are doing # a JOIN USING operation on the column on Oracle. # === :one_to_many and :one_to_one # :key :: foreign key in associated model's table that references # current model's primary key, as a symbol. Defaults to # :"#{self.name.underscore}_id". Can use an # array of symbols for a composite key association. # :key_method :: the method symbol or array of method symbols to call on the associated # object to get the foreign key values. Defaults to :key option. # :primary_key :: column in the current table that :key option references, as a symbol. # Defaults to primary key of the current table. Can use an # array of symbols for a composite key association. # :primary_key_column :: Similar to, and usually identical to, :primary_key, but :primary_key refers # to the model method call, where :primary_key_column refers to the underlying column. # Should only be used if the model method differs from the primary key column, in # conjunction with defining a model alias method for the primary key column. # :raise_on_save_failure :: Do not raise exceptions for hook or validation failures when saving associated # objects in the add/remove methods (return nil instead) [one_to_many only]. # === :many_to_many and :one_through_one # :graph_join_table_block :: The block to pass to +join_table+ for # the join table when eagerly loading the association via +eager_graph+. # :graph_join_table_conditions :: The additional conditions to use on the SQL join for # the join table when eagerly loading the association via +eager_graph+. # Should be a hash or an array of two element arrays. # :graph_join_table_join_type :: The type of SQL join to use for the join table when eagerly # loading the association via +eager_graph+. Defaults to the # :graph_join_type option or :left_outer. # :graph_join_table_only_conditions :: The conditions to use on the SQL join for the join # table when eagerly loading the association via +eager_graph+, # instead of the default conditions specified by the # foreign/primary keys. This option causes the # :graph_join_table_conditions option to be ignored. # :join_table :: name of table that includes the foreign keys to both # the current model and the associated model, as a symbol. Defaults to the name # of current model and name of associated model, pluralized, # underscored, sorted, and joined with '_'. # :join_table_block :: proc that can be used to modify the dataset used in the add/remove/remove_all # methods. Should accept a dataset argument and return a modified dataset if present. # :left_key :: foreign key in join table that points to current model's # primary key, as a symbol. Defaults to :"#{self.name.underscore}_id". # Can use an array of symbols for a composite key association. # :left_primary_key :: column in current table that :left_key points to, as a symbol. # Defaults to primary key of current table. Can use an # array of symbols for a composite key association. # :left_primary_key_column :: Similar to, and usually identical to, :left_primary_key, but :left_primary_key refers to # the model method to call, where :left_primary_key_column refers to the underlying column. Should only # be used if the model method differs from the left primary key column, in conjunction # with defining a model alias method for the left primary key column. # :right_key :: foreign key in join table that points to associated # model's primary key, as a symbol. Defaults to :"#{name.to_s.singularize}_id". # Can use an array of symbols for a composite key association. # :right_primary_key :: column in associated table that :right_key points to, as a symbol. # Defaults to primary key of the associated table. Can use an # array of symbols for a composite key association. # :right_primary_key_method :: the method symbol or array of method symbols to call on the associated # object to get the foreign key values for the join table. # Defaults to :right_primary_key option. # :uniq :: Adds a after_load callback that makes the array of objects unique. def associate(type, name, opts = OPTS, &block) raise(Error, 'invalid association type') unless assoc_class = Sequel.synchronize{ASSOCIATION_TYPES[type]} raise(Error, 'Model.associate name argument must be a symbol') unless name.is_a?(Symbol) # dup early so we don't modify opts orig_opts = opts.dup if opts[:clone] cloned_assoc = association_reflection(opts[:clone]) orig_opts = cloned_assoc[:orig_opts].merge(orig_opts) end opts = Hash[default_association_options] if type_options = default_association_type_options[type] opts.merge!(type_options) end opts.merge!(orig_opts) opts.merge!(:type => type, :name => name, :cache=>({} if cache_associations), :model => self) opts[:block] = block if block if !opts.has_key?(:instance_specific) && (block || orig_opts[:block] || orig_opts[:dataset]) # It's possible the association is instance specific, in that it depends on # values other than the foreign key value. This needs to be checked for # in certain places to disable optimizations. opts[:instance_specific] = true end opts = assoc_class.new.merge!(opts) if opts[:clone] && !opts.cloneable?(cloned_assoc) raise(Error, "cannot clone an association to an association of different type (association #{name} with type #{type} cloning #{opts[:clone]} with type #{cloned_assoc[:type]})") end opts[:eager_block] = opts[:block] unless opts.include?(:eager_block) opts[:graph_join_type] ||= :left_outer opts[:order_eager_graph] = true unless opts.include?(:order_eager_graph) conds = opts[:conditions] opts[:graph_alias_base] ||= name opts[:graph_conditions] = conds if !opts.include?(:graph_conditions) and Sequel.condition_specifier?(conds) opts[:graph_conditions] = opts.fetch(:graph_conditions, []).to_a opts[:graph_select] = Array(opts[:graph_select]) if opts[:graph_select] [:before_add, :before_remove, :after_add, :after_remove, :after_load, :before_set, :after_set].each do |cb_type| opts[cb_type] = Array(opts[cb_type]) if opts[cb_type] end if opts[:extend] opts[:extend] = Array(opts[:extend]) opts[:reverse_extend] = opts[:extend].reverse end late_binding_class_option(opts, opts.returns_array? ? singularize(name) : name) # Remove :class entry if it exists and is nil, to work with cached_fetch opts.delete(:class) unless opts[:class] send(:"def_#{type}", opts) def_association_instance_methods(opts) orig_opts.delete(:clone) opts[:orig_class] = orig_opts[:class] || orig_opts[:class_name] orig_opts.merge!(:class_name=>opts[:class_name], :class=>opts[:class], :block=>opts[:block]) opts[:orig_opts] = orig_opts # don't add to association_reflections until we are sure there are no errors association_reflections[name] = opts end # The association reflection hash for the association of the given name. def association_reflection(name) association_reflections[name] end # Array of association name symbols def associations association_reflections.keys end # Eager load the association with the given eager loader options. def eager_load_results(opts, eo, &block) opts.eager_load_results(eo, &block) end # Freeze association related metadata when freezing model class. def freeze @association_reflections.freeze.each_value(&:freeze) @autoreloading_associations.freeze.each_value(&:freeze) @default_association_options.freeze @default_association_type_options.freeze @default_association_type_options.each_value(&:freeze) super end # Finalize all associations such that values that are looked up # dynamically in associated classes are set statically. # As this modifies the associations, it must be done before # calling freeze. def finalize_associations @association_reflections.each_value(&:finalize) end # Shortcut for adding a many_to_many association, see #associate def many_to_many(name, opts=OPTS, &block) associate(:many_to_many, name, opts, &block) end # Shortcut for adding a many_to_one association, see #associate def many_to_one(name, opts=OPTS, &block) associate(:many_to_one, name, opts, &block) end # Shortcut for adding a one_through_one association, see #associate def one_through_one(name, opts=OPTS, &block) associate(:one_through_one, name, opts, &block) end # Shortcut for adding a one_to_many association, see #associate def one_to_many(name, opts=OPTS, &block) associate(:one_to_many, name, opts, &block) end # Shortcut for adding a one_to_one association, see #associate def one_to_one(name, opts=OPTS, &block) associate(:one_to_one, name, opts, &block) end Plugins.inherited_instance_variables(self, :@association_reflections=>:dup, :@autoreloading_associations=>:hash_dup, :@default_association_options=>:dup, :@default_association_type_options=>:hash_dup, :@cache_associations=>nil, :@default_eager_limit_strategy=>nil) Plugins.def_dataset_methods(self, [:eager, :eager_graph, :eager_graph_with_options, :association_join, :association_full_join, :association_inner_join, :association_left_join, :association_right_join]) private # The module to use for the association's methods. Defaults to # the overridable_methods_module. def association_module(opts=OPTS) opts.fetch(:methods_module, overridable_methods_module) end # Add a method to the module included in the class, so the method # can be easily overridden in the class itself while allowing for # super to be called. def association_module_def(name, opts=OPTS, &block) association_module(opts).module_eval{define_method(name, &block)} end # Add a private method to the module included in the class. def association_module_private_def(name, opts=OPTS, &block) association_module_def(name, opts, &block) association_module(opts).send(:private, name) end # Adds the association method to the association methods module. def def_association_method(opts) association_module_def(opts.association_method, opts) do |dynamic_opts=OPTS, &block| load_associated_objects(opts, dynamic_opts, &block) end end # Define all of the association instance methods for this association. def def_association_instance_methods(opts) # Always set the method names in the association reflection, even if they # are not used, for backwards compatibility. opts[:dataset_method] = :"#{opts[:name]}_dataset" if opts.returns_array? sname = singularize(opts[:name]) opts[:_add_method] = :"_add_#{sname}" opts[:add_method] = :"add_#{sname}" opts[:_remove_method] = :"_remove_#{sname}" opts[:remove_method] = :"remove_#{sname}" opts[:_remove_all_method] = :"_remove_all_#{opts[:name]}" opts[:remove_all_method] = :"remove_all_#{opts[:name]}" else opts[:_setter_method] = :"_#{opts[:name]}=" opts[:setter_method] = :"#{opts[:name]}=" end association_module_def(opts.dataset_method, opts){_dataset(opts)} def_association_method(opts) return if opts[:read_only] if opts[:setter] && opts[:_setter] # This is backwards due to backwards compatibility association_module_private_def(opts[:_setter_method], opts, &opts[:setter]) association_module_def(opts[:setter_method], opts, &opts[:_setter]) end if adder = opts[:adder] association_module_private_def(opts[:_add_method], opts, &adder) association_module_def(opts[:add_method], opts){|o,*args| add_associated_object(opts, o, *args)} end if remover = opts[:remover] association_module_private_def(opts[:_remove_method], opts, &remover) association_module_def(opts[:remove_method], opts){|o,*args| remove_associated_object(opts, o, *args)} end if clearer = opts[:clearer] association_module_private_def(opts[:_remove_all_method], opts, &clearer) association_module_def(opts[:remove_all_method], opts){|*args| remove_all_associated_objects(opts, *args)} end end # Configures many_to_many and one_through_one association reflection and adds the related association methods def def_many_to_many(opts) one_through_one = opts[:type] == :one_through_one left = (opts[:left_key] ||= opts.default_left_key) lcks = opts[:left_keys] = Array(left) right = (opts[:right_key] ||= opts.default_right_key) rcks = opts[:right_keys] = Array(right) left_pk = (opts[:left_primary_key] ||= self.primary_key) opts[:eager_loader_key] = left_pk unless opts.has_key?(:eager_loader_key) lcpks = opts[:left_primary_keys] = Array(left_pk) lpkc = opts[:left_primary_key_column] ||= left_pk lpkcs = opts[:left_primary_key_columns] ||= Array(lpkc) raise(Error, "mismatched number of left keys: #{lcks.inspect} vs #{lcpks.inspect}") unless lcks.length == lcpks.length if opts[:right_primary_key] rcpks = Array(opts[:right_primary_key]) raise(Error, "mismatched number of right keys: #{rcks.inspect} vs #{rcpks.inspect}") unless rcks.length == rcpks.length end opts[:uses_left_composite_keys] = lcks.length > 1 opts[:uses_right_composite_keys] = rcks.length > 1 opts[:cartesian_product_number] ||= one_through_one ? 0 : 1 join_table = (opts[:join_table] ||= opts.default_join_table) opts[:left_key_alias] ||= opts.default_associated_key_alias opts[:graph_join_table_join_type] ||= opts[:graph_join_type] if opts[:uniq] opts[:after_load] ||= [] opts[:after_load].unshift(:array_uniq!) end opts[:dataset] ||= opts.association_dataset_proc opts[:eager_loader] ||= opts.method(:default_eager_loader) join_type = opts[:graph_join_type] select = opts[:graph_select] use_only_conditions = opts.include?(:graph_only_conditions) only_conditions = opts[:graph_only_conditions] conditions = opts[:graph_conditions] graph_block = opts[:graph_block] graph_jt_conds = opts[:graph_join_table_conditions] = opts.fetch(:graph_join_table_conditions, []).to_a use_jt_only_conditions = opts.include?(:graph_join_table_only_conditions) jt_only_conditions = opts[:graph_join_table_only_conditions] jt_join_type = opts[:graph_join_table_join_type] jt_graph_block = opts[:graph_join_table_block] opts[:eager_grapher] ||= proc do |eo| ds = eo[:self] egls = eo[:limit_strategy] if egls && egls != :ruby associated_key_array = opts.associated_key_array orig_egds = egds = eager_graph_dataset(opts, eo) egds = egds. inner_join(join_table, rcks.zip(opts.right_primary_keys) + graph_jt_conds, :qualify=>:deep). select_all(egds.first_source). select_append(*associated_key_array) egds = opts.apply_eager_graph_limit_strategy(egls, egds) ds.graph(egds, associated_key_array.map(&:alias).zip(lpkcs) + conditions, :qualify=>:deep, :table_alias=>eo[:table_alias], :implicit_qualifier=>eo[:implicit_qualifier], :join_type=>eo[:join_type]||join_type, :from_self_alias=>eo[:from_self_alias], :join_only=>eo[:join_only], :select=>select||orig_egds.columns, &graph_block) else ds = ds.graph(join_table, use_jt_only_conditions ? jt_only_conditions : lcks.zip(lpkcs) + graph_jt_conds, :select=>false, :table_alias=>ds.unused_table_alias(join_table, [eo[:table_alias]]), :join_type=>eo[:join_type]||jt_join_type, :join_only=>eo[:join_only], :implicit_qualifier=>eo[:implicit_qualifier], :qualify=>:deep, :from_self_alias=>eo[:from_self_alias], &jt_graph_block) ds.graph(eager_graph_dataset(opts, eo), use_only_conditions ? only_conditions : opts.right_primary_keys.zip(rcks) + conditions, :select=>select, :table_alias=>eo[:table_alias], :qualify=>:deep, :join_type=>eo[:join_type]||join_type, :join_only=>eo[:join_only], &graph_block) end end return if opts[:read_only] if one_through_one opts[:setter] ||= proc do |o| h = {} lh = lcks.zip(lcpks.map{|k| get_column_value(k)}) jtds = _join_table_dataset(opts).where(lh) checked_transaction do current = jtds.first if o new_values = [] rcks.zip(opts.right_primary_key_methods).each{|k, pk| new_values << (h[k] = o.get_column_value(pk))} end if current current_values = rcks.map{|k| current[k]} jtds = jtds.where(rcks.zip(current_values)) if o if current_values != new_values jtds.update(h) end else jtds.delete end elsif o lh.each{|k,v| h[k] = v} jtds.insert(h) end end end opts[:_setter] = proc{|o| set_one_through_one_associated_object(opts, o)} else opts[:adder] ||= proc do |o| h = {} lcks.zip(lcpks).each{|k, pk| h[k] = get_column_value(pk)} rcks.zip(opts.right_primary_key_methods).each{|k, pk| h[k] = o.get_column_value(pk)} _join_table_dataset(opts).insert(h) end opts[:remover] ||= proc do |o| _join_table_dataset(opts).where(lcks.zip(lcpks.map{|k| get_column_value(k)}) + rcks.zip(opts.right_primary_key_methods.map{|k| o.get_column_value(k)})).delete end opts[:clearer] ||= proc do _join_table_dataset(opts).where(lcks.zip(lcpks.map{|k| get_column_value(k)})).delete end end end # Configures many_to_one association reflection and adds the related association methods def def_many_to_one(opts) name = opts[:name] opts[:key] = opts.default_key unless opts.has_key?(:key) key = opts[:key] opts[:eager_loader_key] = key unless opts.has_key?(:eager_loader_key) cks = opts[:graph_keys] = opts[:keys] = Array(key) opts[:key_column] ||= key opts[:graph_keys] = opts[:key_columns] = Array(opts[:key_column]) opts[:qualified_key] = opts.qualify_cur(key) if opts[:primary_key] cpks = Array(opts[:primary_key]) raise(Error, "mismatched number of keys: #{cks.inspect} vs #{cpks.inspect}") unless cks.length == cpks.length end uses_cks = opts[:uses_composite_keys] = cks.length > 1 opts[:cartesian_product_number] ||= 0 if !opts.has_key?(:many_to_one_pk_lookup) && (opts[:dataset] || opts[:conditions] || opts[:block] || opts[:select] || (opts.has_key?(:key) && opts[:key] == nil)) opts[:many_to_one_pk_lookup] = false end auto_assocs = @autoreloading_associations cks.each do |k| (auto_assocs[k] ||= []) << name end opts[:dataset] ||= opts.association_dataset_proc opts[:eager_loader] ||= proc do |eo| h = eo[:id_map] pk_meths = opts.primary_key_methods eager_load_results(opts, eo) do |assoc_record| hash_key = uses_cks ? pk_meths.map{|k| assoc_record.get_column_value(k)} : assoc_record.get_column_value(opts.primary_key_method) if objects = h[hash_key] objects.each{|object| object.associations[name] = assoc_record} end end end join_type = opts[:graph_join_type] select = opts[:graph_select] use_only_conditions = opts.include?(:graph_only_conditions) only_conditions = opts[:graph_only_conditions] conditions = opts[:graph_conditions] graph_block = opts[:graph_block] graph_cks = opts[:graph_keys] opts[:eager_grapher] ||= proc do |eo| ds = eo[:self] ds.graph(eager_graph_dataset(opts, eo), use_only_conditions ? only_conditions : opts.primary_keys.zip(graph_cks) + conditions, Hash[eo].merge!(:select=>select, :join_type=>eo[:join_type]||join_type, :qualify=>:deep, :from_self_alias=>eo[:from_self_alias]), &graph_block) end return if opts[:read_only] opts[:setter] ||= proc{|o| cks.zip(opts.primary_key_methods).each{|k, pk| set_column_value(:"#{k}=", (o.get_column_value(pk) if o))}} opts[:_setter] = proc{|o| set_associated_object(opts, o)} end # Configures one_to_many and one_to_one association reflections and adds the related association methods def def_one_to_many(opts) one_to_one = opts[:type] == :one_to_one name = opts[:name] key = (opts[:key] ||= opts.default_key) km = opts[:key_method] ||= opts[:key] cks = opts[:keys] = Array(key) opts[:key_methods] = Array(opts[:key_method]) primary_key = (opts[:primary_key] ||= self.primary_key) opts[:eager_loader_key] = primary_key unless opts.has_key?(:eager_loader_key) cpks = opts[:primary_keys] = Array(primary_key) pkc = opts[:primary_key_column] ||= primary_key pkcs = opts[:primary_key_columns] ||= Array(pkc) raise(Error, "mismatched number of keys: #{cks.inspect} vs #{cpks.inspect}") unless cks.length == cpks.length uses_cks = opts[:uses_composite_keys] = cks.length > 1 opts[:dataset] ||= opts.association_dataset_proc opts[:eager_loader] ||= proc do |eo| h = eo[:id_map] reciprocal = opts.reciprocal assign_singular = opts.assign_singular? delete_rn = opts.delete_row_number_column eager_load_results(opts, eo) do |assoc_record| assoc_record.values.delete(delete_rn) if delete_rn hash_key = uses_cks ? km.map{|k| assoc_record.get_column_value(k)} : assoc_record.get_column_value(km) next unless objects = h[hash_key] if assign_singular objects.each do |object| unless object.associations[name] object.associations[name] = assoc_record assoc_record.associations[reciprocal] = object if reciprocal end end else objects.each do |object| object.associations[name].push(assoc_record) assoc_record.associations[reciprocal] = object if reciprocal end end end end join_type = opts[:graph_join_type] select = opts[:graph_select] use_only_conditions = opts.include?(:graph_only_conditions) only_conditions = opts[:graph_only_conditions] conditions = opts[:graph_conditions] opts[:cartesian_product_number] ||= one_to_one ? 0 : 1 graph_block = opts[:graph_block] opts[:eager_grapher] ||= proc do |eo| ds = eo[:self] ds = ds.graph(opts.apply_eager_graph_limit_strategy(eo[:limit_strategy], eager_graph_dataset(opts, eo)), use_only_conditions ? only_conditions : cks.zip(pkcs) + conditions, Hash[eo].merge!(:select=>select, :join_type=>eo[:join_type]||join_type, :qualify=>:deep, :from_self_alias=>eo[:from_self_alias]), &graph_block) # We only load reciprocals for one_to_many associations, as other reciprocals don't make sense ds.opts[:eager_graph][:reciprocals][eo[:table_alias]] = opts.reciprocal ds end return if opts[:read_only] save_opts = {:validate=>opts[:validate]} ck_nil_hash ={} cks.each{|k| ck_nil_hash[k] = nil} if one_to_one opts[:setter] ||= proc do |o| up_ds = _apply_association_options(opts, opts.associated_dataset.where(cks.zip(cpks.map{|k| get_column_value(k)}))) if o up_ds = up_ds.exclude(o.pk_hash) unless o.new? cks.zip(cpks).each{|k, pk| o.set_column_value(:"#{k}=", get_column_value(pk))} end checked_transaction do up_ds.skip_limit_check.update(ck_nil_hash) o.save(save_opts) || raise(Sequel::Error, "invalid associated object, cannot save") if o end end opts[:_setter] = proc{|o| set_one_to_one_associated_object(opts, o)} else save_opts[:raise_on_failure] = opts[:raise_on_save_failure] != false opts[:adder] ||= proc do |o| cks.zip(cpks).each{|k, pk| o.set_column_value(:"#{k}=", get_column_value(pk))} o.save(save_opts) end opts[:remover] ||= proc do |o| cks.each{|k| o.set_column_value(:"#{k}=", nil)} o.save(save_opts) end opts[:clearer] ||= proc do _apply_association_options(opts, opts.associated_dataset.where(cks.zip(cpks.map{|k| get_column_value(k)}))).update(ck_nil_hash) end end end # Alias of def_many_to_many, since they share pretty much the same code. def def_one_through_one(opts) def_many_to_many(opts) end # Alias of def_one_to_many, since they share pretty much the same code. def def_one_to_one(opts) def_one_to_many(opts) end # Return dataset to graph into given the association reflection, applying the :callback option if set. def eager_graph_dataset(opts, eager_options) ds = opts.associated_class.dataset if cb = eager_options[:callback] ds = cb.call(ds) end ds end # If not caching associations, reload the database schema by default, # ignoring any cached values. def reload_db_schema? !@cache_associations end end # Instance methods used to implement the associations support. module InstanceMethods # The currently cached associations. A hash with the keys being the # association name symbols and the values being the associated object # or nil (many_to_one), or the array of associated objects (*_to_many). def associations @associations ||= {} end # Freeze the associations cache when freezing the object. Note that # retrieving associations after freezing will still work in most cases, # but the associations will not be cached in the association cache. def freeze associations super associations.freeze self end private # Apply the association options such as :order and :limit to the given dataset, returning a modified dataset. def _apply_association_options(opts, ds) unless ds.kind_of?(AssociationDatasetMethods) ds = opts.apply_dataset_changes(ds) end ds = ds.clone(:model_object => self) ds = ds.eager_graph(opts[:eager_graph]) if opts[:eager_graph] && opts.eager_graph_lazy_dataset? ds = instance_exec(ds, &opts[:block]) if opts[:block] ds end # Return a dataset for the association after applying any dynamic callback. def _associated_dataset(opts, dynamic_opts) ds = public_send(opts.dataset_method) if callback = dynamic_opts[:callback] ds = callback.call(ds) end ds end # A placeholder literalizer that can be used to load the association, or nil to not use one. def _associated_object_loader(opts, dynamic_opts) if !dynamic_opts[:callback] && (loader = opts.placeholder_loader) loader end end # Return an association dataset for the given association reflection def _dataset(opts) raise(Sequel::Error, "model object #{inspect} does not have a primary key") if opts.dataset_need_primary_key? && !pk ds = if opts[:dataset].arity == 1 instance_exec(opts, &opts[:dataset]) else instance_exec(&opts[:dataset]) end _apply_association_options(opts, ds) end # Dataset for the join table of the given many to many association reflection def _join_table_dataset(opts) ds = model.db.from(opts.join_table_source) opts[:join_table_block] ? opts[:join_table_block].call(ds) : ds end # Return the associated single object for the given association reflection and dynamic options # (or nil if no associated object). def _load_associated_object(opts, dynamic_opts) _load_associated_object_array(opts, dynamic_opts).first end # Return the associated single object using a primary key lookup on the associated class. def _load_associated_object_via_primary_key(opts) opts.associated_class.send(:primary_key_lookup, ((fk = opts[:key]).is_a?(Array) ? fk.map{|c| get_column_value(c)} : get_column_value(fk))) end # Load the associated objects for the given association reflection and dynamic options # as an array. def _load_associated_object_array(opts, dynamic_opts) if loader = _associated_object_loader(opts, dynamic_opts) loader.all(*opts.predicate_key_values(self)) else _associated_dataset(opts, dynamic_opts).all end end # Return the associated objects from the dataset, without association callbacks, reciprocals, and caching. # Still apply the dynamic callback if present. def _load_associated_objects(opts, dynamic_opts=OPTS) if opts.can_have_associated_objects?(self) if opts.returns_array? _load_associated_object_array(opts, dynamic_opts) elsif load_with_primary_key_lookup?(opts, dynamic_opts) _load_associated_object_via_primary_key(opts) else _load_associated_object(opts, dynamic_opts) end elsif opts.returns_array? [] end end # Clear the associations cache when refreshing def _refresh_set_values(hash) @associations.clear if @associations super end # Add the given associated object to the given association def add_associated_object(opts, o, *args) o = make_add_associated_object(opts, o) raise(Sequel::Error, "model object #{inspect} does not have a primary key") if opts.dataset_need_primary_key? && !pk ensure_associated_primary_key(opts, o, *args) return if run_association_callbacks(opts, :before_add, o) == false # Allow calling private _add method return if !send(opts[:_add_method], o, *args) && opts.handle_silent_modification_failure? if array = associations[opts[:name]] and !array.include?(o) array.push(o) end add_reciprocal_object(opts, o) run_association_callbacks(opts, :after_add, o) o end # Add/Set the current object to/as the given object's reciprocal association. def add_reciprocal_object(opts, o) return if o.frozen? return unless reciprocal = opts.reciprocal if opts.reciprocal_array? if array = o.associations[reciprocal] and !array.include?(self) array.push(self) end else o.associations[reciprocal] = self end end # Call uniq! on the given array. This is used by the :uniq option, # and is an actual method for memory reasons. def array_uniq!(a) a.uniq! end # If a foreign key column value changes, clear the related # cached associations. def change_column_value(column, value) if assocs = model.autoreloading_associations[column] vals = @values if new? # Do deeper checking for new objects, so that associations are # not deleted when values do not change. This code is run at # a higher level for existing objects. if value == (c = vals[column]) && value.class == c.class # If the value is the same, there is no reason to delete # the related associations, so exit early in that case. return super end only_delete_nil = c.nil? elsif vals[column].nil? only_delete_nil = true end if only_delete_nil # If the current foreign key value is nil, but the association # is already present in the cache, it was probably added to the # cache for a reason, and we do not want to delete it in that case. # However, we still want to delete associations with nil values # to remove the cached false negative. assocs.each{|a| associations.delete(a) if associations[a].nil?} else assocs.each{|a| associations.delete(a)} end end super end # Save the associated object if the associated object needs a primary key # and the associated object is new and does not have one. Raise an error if # the object still does not have a primary key def ensure_associated_primary_key(opts, o, *args) if opts.need_associated_primary_key? o.save(:validate=>opts[:validate]) if o.new? raise(Sequel::Error, "associated object #{o.inspect} does not have a primary key") unless o.pk end end # Duplicate the associations hash when duplicating the object. def initialize_copy(other) super @associations = Hash[@associations] if @associations self end # If a block is given, assign it as the :callback option in the hash, and return the hash. def load_association_objects_options(dynamic_opts, &block) if block dynamic_opts = Hash[dynamic_opts] dynamic_opts[:callback] = block end dynamic_opts end # Load the associated objects using the dataset, handling callbacks, reciprocals, and caching. def load_associated_objects(opts, dynamic_opts, &block) dynamic_opts = load_association_objects_options(dynamic_opts, &block) name = opts[:name] if associations.include?(name) && !dynamic_opts[:callback] && !dynamic_opts[:reload] associations[name] else objs = _load_associated_objects(opts, dynamic_opts) if opts.set_reciprocal_to_self? if opts.returns_array? objs.each{|o| add_reciprocal_object(opts, o)} elsif objs add_reciprocal_object(opts, objs) end end # If the current object is frozen, you can't update the associations # cache. This can cause issues for after_load procs that expect # the objects to be already cached in the associations, but # unfortunately that case cannot be handled. associations[name] = objs unless frozen? run_association_callbacks(opts, :after_load, objs) frozen? ? objs : associations[name] end end # Whether to use a simple primary key lookup on the associated class when loading. def load_with_primary_key_lookup?(opts, dynamic_opts) opts[:type] == :many_to_one && !dynamic_opts[:callback] && opts.send(:cached_fetch, :many_to_one_pk_lookup){opts.primary_key == opts.associated_class.primary_key} end # Convert the input of the add_* association method into an associated object. For # hashes, this creates a new object using the hash. For integers, strings, and arrays, # assume the value specifies a primary key, and lookup an existing object with that primary key. # Otherwise, if the object is not already an instance of the class, raise an exception. def make_add_associated_object(opts, o) klass = opts.associated_class case o when Hash klass.new(o) when Integer, String, Array klass.with_pk!(o) when klass o else raise(Sequel::Error, "associated object #{o.inspect} not of correct type #{klass}") end end # Remove all associated objects from the given association def remove_all_associated_objects(opts, *args) raise(Sequel::Error, "model object #{inspect} does not have a primary key") if opts.dataset_need_primary_key? && !pk # Allow calling private _remove_all method send(opts[:_remove_all_method], *args) ret = associations[opts[:name]].each{|o| remove_reciprocal_object(opts, o)} if associations.include?(opts[:name]) associations[opts[:name]] = [] ret end # Remove the given associated object from the given association def remove_associated_object(opts, o, *args) klass = opts.associated_class if o.is_a?(Integer) || o.is_a?(String) || o.is_a?(Array) o = remove_check_existing_object_from_pk(opts, o, *args) elsif !o.is_a?(klass) raise(Sequel::Error, "associated object #{o.inspect} not of correct type #{klass}") elsif opts.remove_should_check_existing? && public_send(opts.dataset_method).where(o.pk_hash).empty? raise(Sequel::Error, "associated object #{o.inspect} is not currently associated to #{inspect}") end raise(Sequel::Error, "model object #{inspect} does not have a primary key") if opts.dataset_need_primary_key? && !pk raise(Sequel::Error, "associated object #{o.inspect} does not have a primary key") if opts.need_associated_primary_key? && !o.pk return if run_association_callbacks(opts, :before_remove, o) == false # Allow calling private _remove method return if !send(opts[:_remove_method], o, *args) && opts.handle_silent_modification_failure? associations[opts[:name]].delete_if{|x| o === x} if associations.include?(opts[:name]) remove_reciprocal_object(opts, o) run_association_callbacks(opts, :after_remove, o) o end # Check that the object from the associated table specified by the primary key # is currently associated to the receiver. If it is associated, return the object, otherwise # raise an error. def remove_check_existing_object_from_pk(opts, o, *args) key = o pkh = opts.associated_class.qualified_primary_key_hash(key) raise(Sequel::Error, "no object with key(s) #{key.inspect} is currently associated to #{inspect}") unless o = public_send(opts.dataset_method).first(pkh) o end # Remove/unset the current object from/as the given object's reciprocal association. def remove_reciprocal_object(opts, o) return unless reciprocal = opts.reciprocal if opts.reciprocal_array? if array = o.associations[reciprocal] array.delete_if{|x| self === x} end else o.associations[reciprocal] = nil end end # Run the callback for the association with the object. def run_association_callbacks(reflection, callback_type, object) return unless cbs = reflection[callback_type] begin cbs.each do |cb| case cb when Symbol # Allow calling private methods in association callbacks send(cb, object) when Proc cb.call(self, object) else raise Error, "callbacks should either be Procs or Symbols" end end rescue HookFailed # The reason we automatically set raise_error for singular associations is that # assignment in ruby always returns the argument instead of the result of the # method, so we can't return nil to signal that the association callback prevented # the modification return false unless raise_on_save_failure || !reflection.returns_array? raise end end # Set the given object as the associated object for the given *_to_one association reflection def _set_associated_object(opts, o) a = associations[opts[:name]] reciprocal = opts.reciprocal if set_associated_object_if_same? if reciprocal remove_reciprocal = a && (a != o || a.associations[reciprocal] != self) add_reciprocal = o && o.associations[reciprocal] != self end else return if a && a == o if reciprocal remove_reciprocal = a add_reciprocal = o end end run_association_callbacks(opts, :before_set, o) remove_reciprocal_object(opts, a) if remove_reciprocal # Allow calling private _setter method send(opts[:_setter_method], o) associations[opts[:name]] = o add_reciprocal_object(opts, o) if add_reciprocal run_association_callbacks(opts, :after_set, o) o end # Whether run the associated object setter code if passed the same object as the one already # cached in the association. Usually not set (so nil), can be set on a per-object basis # if necessary. def set_associated_object_if_same? @set_associated_object_if_same end # Set the given object as the associated object for the given many_to_one association reflection def set_associated_object(opts, o) raise(Error, "associated object #{o.inspect} does not have a primary key") if o && !o.pk _set_associated_object(opts, o) end # Set the given object as the associated object for the given one_through_one association reflection def set_one_through_one_associated_object(opts, o) raise(Error, "object #{inspect} does not have a primary key") unless pk raise(Error, "associated object #{o.inspect} does not have a primary key") if o && !o.pk _set_associated_object(opts, o) end # Set the given object as the associated object for the given one_to_one association reflection def set_one_to_one_associated_object(opts, o) raise(Error, "object #{inspect} does not have a primary key") unless pk _set_associated_object(opts, o) end end # Eager loading makes it so that you can load all associated records for a # set of objects in a single query, instead of a separate query for each object. # # Two separate implementations are provided. +eager+ should be used most of the # time, as it loads associated records using one query per association. However, # it does not allow you the ability to filter or order based on columns in associated tables. +eager_graph+ loads # all records in a single query using JOINs, allowing you to filter or order based on columns in associated # tables. However, +eager_graph+ is usually slower than +eager+, especially if multiple # one_to_many or many_to_many associations are joined. # # You can cascade the eager loading (loading associations on associated objects) # with no limit to the depth of the cascades. You do this by passing a hash to +eager+ or +eager_graph+ # with the keys being associations of the current model and values being # associations of the model associated with the current model via the key. # # The arguments can be symbols or hashes with symbol keys (for cascaded # eager loading). Examples: # # Album.eager(:artist).all # Album.eager_graph(:artist).all # Album.eager(:artist, :genre).all # Album.eager_graph(:artist, :genre).all # Album.eager(:artist).eager(:genre).all # Album.eager_graph(:artist).eager_graph(:genre).all # Artist.eager(albums: :tracks).all # Artist.eager_graph(albums: :tracks).all # Artist.eager(albums: {tracks: :genre}).all # Artist.eager_graph(albums: {tracks: :genre}).all # # You can also pass a callback as a hash value in order to customize the dataset being # eager loaded at query time, analogous to the way the :eager_block association option # allows you to customize it at association definition time. For example, # if you wanted artists with their albums since 1990: # # Artist.eager(albums: proc{|ds| ds.where{year > 1990}}) # # Or if you needed albums and their artist's name only, using a single query: # # Albums.eager_graph(artist: proc{|ds| ds.select(:name)}) # # To cascade eager loading while using a callback, you substitute the cascaded # associations with a single entry hash that has the proc callback as the key and # the cascaded associations as the value. This will load artists with their albums # since 1990, and also the tracks on those albums and the genre for those tracks: # # Artist.eager(albums: {proc{|ds| ds.where{year > 1990}}=>{tracks: :genre}}) module DatasetMethods %w'inner left right full'.each do |type| class_eval(<<-END, __FILE__, __LINE__+1) def association_#{type}_join(*associations) _association_join(:#{type}, associations) end END end # Adds one or more INNER JOINs to the existing dataset using the keys and conditions # specified by the given association(s). Take the same arguments as eager_graph, and # operates similarly, but only adds the joins as opposed to making the other changes # (such as adding selected columns and setting up eager loading). # # The following methods also exist for specifying a different type of JOIN: # # association_full_join :: FULL JOIN # association_inner_join :: INNER JOIN # association_left_join :: LEFT JOIN # association_right_join :: RIGHT JOIN # # Examples: # # # For each album, association_join load the artist # Album.association_join(:artist).all # # SELECT * # # FROM albums # # INNER JOIN artists AS artist ON (artists.id = albums.artist_id) # # # For each album, association_join load the artist, using a specified alias # Album.association_join(Sequel[:artist].as(:a)).all # # SELECT * # # FROM albums # # INNER JOIN artists AS a ON (a.id = albums.artist_id) # # # For each album, association_join load the artist and genre # Album.association_join(:artist, :genre).all # Album.association_join(:artist).association_join(:genre).all # # SELECT * # # FROM albums # # INNER JOIN artists AS artist ON (artist.id = albums.artist_id) # # INNER JOIN genres AS genre ON (genre.id = albums.genre_id) # # # For each artist, association_join load albums and tracks for each album # Artist.association_join(albums: :tracks).all # # SELECT * # # FROM artists # # INNER JOIN albums ON (albums.artist_id = artists.id) # # INNER JOIN tracks ON (tracks.album_id = albums.id) # # # For each artist, association_join load albums, tracks for each album, and genre for each track # Artist.association_join(albums: {tracks: :genre}).all # # SELECT * # # FROM artists # # INNER JOIN albums ON (albums.artist_id = artists.id) # # INNER JOIN tracks ON (tracks.album_id = albums.id) # # INNER JOIN genres AS genre ON (genre.id = tracks.genre_id) # # # For each artist, association_join load albums with year > 1990 # Artist.association_join(albums: proc{|ds| ds.where{year > 1990}}).all # # SELECT * # # FROM artists # # INNER JOIN ( # # SELECT * FROM albums WHERE (year > 1990) # # ) AS albums ON (albums.artist_id = artists.id) # # # For each artist, association_join load albums and tracks 1-10 for each album # Artist.association_join(albums: {tracks: proc{|ds| ds.where(number: 1..10)}}).all # # SELECT * # # FROM artists # # INNER JOIN albums ON (albums.artist_id = artists.id) # # INNER JOIN ( # # SELECT * FROM tracks WHERE ((number >= 1) AND (number <= 10)) # # ) AS tracks ON (tracks.albums_id = albums.id) # # # For each artist, association_join load albums with year > 1990, and tracks for those albums # Artist.association_join(albums: {proc{|ds| ds.where{year > 1990}}=>:tracks}).all # # SELECT * # # FROM artists # # INNER JOIN ( # # SELECT * FROM albums WHERE (year > 1990) # # ) AS albums ON (albums.artist_id = artists.id) # # INNER JOIN tracks ON (tracks.album_id = albums.id) def association_join(*associations) association_inner_join(*associations) end # If the expression is in the form x = y where +y+ is a Sequel::Model # instance, array of Sequel::Model instances, or a Sequel::Model dataset, # assume +x+ is an association symbol and look up the association reflection # via the dataset's model. From there, return the appropriate SQL based on the type of # association and the values of the foreign/primary keys of +y+. For most association # types, this is a simple transformation, but for +many_to_many+ associations this # creates a subquery to the join table. def complex_expression_sql_append(sql, op, args) r = args[1] if (((op == :'=' || op == :'!=') && r.is_a?(Sequel::Model)) || (multiple = ((op == :IN || op == :'NOT IN') && ((is_ds = r.is_a?(Sequel::Dataset)) || (r.respond_to?(:all?) && r.all?{|x| x.is_a?(Sequel::Model)}))))) l = args[0] if ar = model.association_reflections[l] if multiple klass = ar.associated_class if is_ds if r.respond_to?(:model) unless r.model <= klass # A dataset for a different model class, could be a valid regular query return super end else # Not a model dataset, could be a valid regular query return super end else unless r.all?{|x| x.is_a?(klass)} raise Sequel::Error, "invalid association class for one object for association #{l.inspect} used in dataset filter for model #{model.inspect}, expected class #{klass.inspect}" end end elsif !r.is_a?(ar.associated_class) raise Sequel::Error, "invalid association class #{r.class.inspect} for association #{l.inspect} used in dataset filter for model #{model.inspect}, expected class #{ar.associated_class.inspect}" end if exp = association_filter_expression(op, ar, r) literal_append(sql, exp) else raise Sequel::Error, "invalid association type #{ar[:type].inspect} for association #{l.inspect} used in dataset filter for model #{model.inspect}" end elsif multiple && (is_ds || r.empty?) # Not a query designed for this support, could be a valid regular query super else raise Sequel::Error, "invalid association #{l.inspect} used in dataset filter for model #{model.inspect}" end else super end end # The preferred eager loading method. Loads all associated records using one # query for each association. # # The basic idea for how it works is that the dataset is first loaded normally. # Then it goes through all associations that have been specified via +eager+. # It loads each of those associations separately, then associates them back # to the original dataset via primary/foreign keys. Due to the necessity of # all objects being present, you need to use +all+ to use eager loading, as it # can't work with +each+. # # This implementation avoids the complexity of extracting an object graph out # of a single dataset, by building the object graph out of multiple datasets, # one for each association. By using a separate dataset for each association, # it avoids problems such as aliasing conflicts and creating cartesian product # result sets if multiple one_to_many or many_to_many eager associations are requested. # # One limitation of using this method is that you cannot filter the current dataset # based on values of columns in an associated table, since the associations are loaded # in separate queries. To do that you need to load all associations in the # same query, and extract an object graph from the results of that query. If you # need to filter based on columns in associated tables, look at +eager_graph+ # or join the tables you need to filter on manually. # # Each association's order, if defined, is respected. # If the association uses a block or has an :eager_block argument, it is used. # # To modify the associated dataset that will be used for the eager load, you should use a # hash for the association, with the key being the association name symbol, and the value being # a callable object that is called with the associated dataset and should return a modified # dataset. If that association also has dependent associations, instead of a callable object, # use a hash with the callable object being the key, and the dependent association(s) as the value. # # Examples: # # # For each album, eager load the artist # Album.eager(:artist).all # # SELECT * FROM albums # # SELECT * FROM artists WHERE (id IN (...)) # # # For each album, eager load the artist and genre # Album.eager(:artist, :genre).all # Album.eager(:artist).eager(:genre).all # # SELECT * FROM albums # # SELECT * FROM artists WHERE (id IN (...)) # # SELECT * FROM genres WHERE (id IN (...)) # # # For each artist, eager load albums and tracks for each album # Artist.eager(albums: :tracks).all # # SELECT * FROM artists # # SELECT * FROM albums WHERE (artist_id IN (...)) # # SELECT * FROM tracks WHERE (album_id IN (...)) # # # For each artist, eager load albums, tracks for each album, and genre for each track # Artist.eager(albums: {tracks: :genre}).all # # SELECT * FROM artists # # SELECT * FROM albums WHERE (artist_id IN (...)) # # SELECT * FROM tracks WHERE (album_id IN (...)) # # SELECT * FROM genre WHERE (id IN (...)) # # # For each artist, eager load albums with year > 1990 # Artist.eager(albums: proc{|ds| ds.where{year > 1990}}).all # # SELECT * FROM artists # # SELECT * FROM albums WHERE ((year > 1990) AND (artist_id IN (...))) # # # For each artist, eager load albums and tracks 1-10 for each album # Artist.eager(albums: {tracks: proc{|ds| ds.where(number: 1..10)}}).all # # SELECT * FROM artists # # SELECT * FROM albums WHERE (artist_id IN (...)) # # SELECT * FROM tracks WHERE ((number >= 1) AND (number <= 10) AND (album_id IN (...))) # # # For each artist, eager load albums with year > 1990, and tracks for those albums # Artist.eager(albums: {proc{|ds| ds.where{year > 1990}}=>:tracks}).all # # SELECT * FROM artists # # SELECT * FROM albums WHERE ((year > 1990) AND (artist_id IN (...))) # # SELECT * FROM albums WHERE (artist_id IN (...)) def eager(*associations) opts = @opts[:eager] association_opts = eager_options_for_associations(associations) opts = opts ? Hash[opts].merge!(association_opts) : association_opts clone(:eager=>opts.freeze) end # The secondary eager loading method. Loads all associations in a single query. This # method should only be used if you need to filter or order based on columns in associated tables, # or if you have done comparative benchmarking it and determined it is faster. # # This method uses Dataset#graph to create appropriate aliases for columns in all the # tables. Then it uses the graph's metadata to build the associations from the single hash, and # finally replaces the array of hashes with an array model objects inside all. # # Be very careful when using this with multiple one_to_many or many_to_many associations, as you can # create large cartesian products. If you must graph multiple one_to_many and many_to_many associations, # make sure your filters are narrow if the datasets are large. # # Each association's order, if defined, is respected. +eager_graph+ probably # won't work correctly on a limited dataset, unless you are # only graphing many_to_one, one_to_one, and one_through_one associations. # # Does not use the block defined for the association, since it does a single query for # all objects. You can use the :graph_* association options to modify the SQL query. # # Like +eager+, you need to call +all+ on the dataset for the eager loading to work. If you just # call +each+, it will yield plain hashes, each containing all columns from all the tables. # # To modify the associated dataset that will be joined to the current dataset, you should use a # hash for the association, with the key being the association name symbol, and the value being # a callable object that is called with the associated dataset and should return a modified # dataset. If that association also has dependent associations, instead of a callable object, # use a hash with the callable object being the key, and the dependent association(s) as the value. # # You can specify an alias by providing a Sequel::SQL::AliasedExpression object instead of # an a Symbol for the assocation name. # # Examples: # # # For each album, eager_graph load the artist # Album.eager_graph(:artist).all # # SELECT ... # # FROM albums # # LEFT OUTER JOIN artists AS artist ON (artists.id = albums.artist_id) # # # For each album, eager_graph load the artist, using a specified alias # Album.eager_graph(Sequel[:artist].as(:a)).all # # SELECT ... # # FROM albums # # LEFT OUTER JOIN artists AS a ON (a.id = albums.artist_id) # # # For each album, eager_graph load the artist and genre # Album.eager_graph(:artist, :genre).all # Album.eager_graph(:artist).eager_graph(:genre).all # # SELECT ... # # FROM albums # # LEFT OUTER JOIN artists AS artist ON (artist.id = albums.artist_id) # # LEFT OUTER JOIN genres AS genre ON (genre.id = albums.genre_id) # # # For each artist, eager_graph load albums and tracks for each album # Artist.eager_graph(albums: :tracks).all # # SELECT ... # # FROM artists # # LEFT OUTER JOIN albums ON (albums.artist_id = artists.id) # # LEFT OUTER JOIN tracks ON (tracks.album_id = albums.id) # # # For each artist, eager_graph load albums, tracks for each album, and genre for each track # Artist.eager_graph(albums: {tracks: :genre}).all # # SELECT ... # # FROM artists # # LEFT OUTER JOIN albums ON (albums.artist_id = artists.id) # # LEFT OUTER JOIN tracks ON (tracks.album_id = albums.id) # # LEFT OUTER JOIN genres AS genre ON (genre.id = tracks.genre_id) # # # For each artist, eager_graph load albums with year > 1990 # Artist.eager_graph(albums: proc{|ds| ds.where{year > 1990}}).all # # SELECT ... # # FROM artists # # LEFT OUTER JOIN ( # # SELECT * FROM albums WHERE (year > 1990) # # ) AS albums ON (albums.artist_id = artists.id) # # # For each artist, eager_graph load albums and tracks 1-10 for each album # Artist.eager_graph(albums: {tracks: proc{|ds| ds.where(number: 1..10)}}).all # # SELECT ... # # FROM artists # # LEFT OUTER JOIN albums ON (albums.artist_id = artists.id) # # LEFT OUTER JOIN ( # # SELECT * FROM tracks WHERE ((number >= 1) AND (number <= 10)) # # ) AS tracks ON (tracks.albums_id = albums.id) # # # For each artist, eager_graph load albums with year > 1990, and tracks for those albums # Artist.eager_graph(albums: {proc{|ds| ds.where{year > 1990}}=>:tracks}).all # # SELECT ... # # FROM artists # # LEFT OUTER JOIN ( # # SELECT * FROM albums WHERE (year > 1990) # # ) AS albums ON (albums.artist_id = artists.id) # # LEFT OUTER JOIN tracks ON (tracks.album_id = albums.id) def eager_graph(*associations) eager_graph_with_options(associations) end # Run eager_graph with some options specific to just this call. Unlike eager_graph, this takes # the associations as a single argument instead of multiple arguments. # # Options: # # :join_type :: Override the join type specified in the association # :limit_strategy :: Use a strategy for handling limits on associations. # Appropriate :limit_strategy values are: # true :: Pick the most appropriate based on what the database supports # :distinct_on :: Force use of DISTINCT ON stategy (*_one associations only) # :correlated_subquery :: Force use of correlated subquery strategy (one_to_* associations only) # :window_function :: Force use of window function strategy # :ruby :: Don't modify the SQL, implement limits/offsets with array slicing # # This can also be a hash with association name symbol keys and one of the above values, # to use different strategies per association. # # The default is the :ruby strategy. Choosing a different strategy can make your code # significantly slower in some cases (perhaps even the majority of cases), so you should # only use this if you have benchmarked that it is faster for your use cases. def eager_graph_with_options(associations, opts=OPTS) opts = opts.dup unless opts.frozen? associations = [associations] unless associations.is_a?(Array) ds = if eg = @opts[:eager_graph] eg = eg.dup [:requirements, :reflections, :reciprocals, :limits].each{|k| eg[k] = eg[k].dup} eg[:local] = opts ds = clone(:eager_graph=>eg) ds.eager_graph_associations(ds, model, ds.opts[:eager_graph][:master], [], *associations) else # Each of the following have a symbol key for the table alias, with the following values: # :reciprocals :: the reciprocal value to use for this association # :reflections :: AssociationReflection instance related to this association # :requirements :: array of requirements for this association # :limits :: Any limit/offset array slicing that need to be handled in ruby land after loading opts = {:requirements=>{}, :master=>alias_symbol(first_source), :reflections=>{}, :reciprocals=>{}, :limits=>{}, :local=>opts, :cartesian_product_number=>0, :row_proc=>row_proc} ds = clone(:eager_graph=>opts) ds = ds.eager_graph_associations(ds, model, ds.opts[:eager_graph][:master], [], *associations).naked end ds.opts[:eager_graph].freeze ds.opts[:eager_graph].each_value{|v| v.freeze if v.is_a?(Hash)} ds end # If the dataset is being eagerly loaded, default to calling all # instead of each. def as_hash(key_column=nil, value_column=nil, opts=OPTS) if (@opts[:eager_graph] || @opts[:eager]) && !opts.has_key?(:all) opts = Hash[opts] opts[:all] = true end super end # If the dataset is being eagerly loaded, default to calling all # instead of each. def to_hash_groups(key_column, value_column=nil, opts=OPTS) if (@opts[:eager_graph] || @opts[:eager]) && !opts.has_key?(:all) opts = Hash[opts] opts[:all] = true end super end # Do not attempt to split the result set into associations, # just return results as simple objects. This is useful if you # want to use eager_graph as a shortcut to have all of the joins # and aliasing set up, but want to do something else with the dataset. def ungraphed ds = super.clone(:eager_graph=>nil) if (eg = @opts[:eager_graph]) && (rp = eg[:row_proc]) ds = ds.with_row_proc(rp) end ds end protected # Call graph on the association with the correct arguments, # update the eager_graph data structure, and recurse into # eager_graph_associations if there are any passed in associations # (which would be dependencies of the current association) # # Arguments: # ds :: Current dataset # model :: Current Model # ta :: table_alias used for the parent association # requirements :: an array, used as a stack for requirements # r :: association reflection for the current association, or an SQL::AliasedExpression # with the reflection as the expression and the alias base as the aliaz. # *associations :: any associations dependent on this one def eager_graph_association(ds, model, ta, requirements, r, *associations) if r.is_a?(SQL::AliasedExpression) alias_base = r.alias r = r.expression else alias_base = r[:graph_alias_base] end assoc_table_alias = ds.unused_table_alias(alias_base) loader = r[:eager_grapher] if !associations.empty? if associations.first.respond_to?(:call) callback = associations.first associations = {} elsif associations.length == 1 && (assocs = associations.first).is_a?(Hash) && assocs.length == 1 && (pr_assoc = assocs.to_a.first) && pr_assoc.first.respond_to?(:call) callback, assoc = pr_assoc associations = assoc.is_a?(Array) ? assoc : [assoc] end end local_opts = ds.opts[:eager_graph][:local] limit_strategy = r.eager_graph_limit_strategy(local_opts[:limit_strategy]) if r[:conditions] && !Sequel.condition_specifier?(r[:conditions]) && !r[:orig_opts].has_key?(:graph_conditions) && !r[:orig_opts].has_key?(:graph_only_conditions) && !r.has_key?(:graph_block) raise Error, "Cannot eager_graph association when :conditions specified and not a hash or an array of pairs. Specify :graph_conditions, :graph_only_conditions, or :graph_block for the association. Model: #{r[:model]}, association: #{r[:name]}" end ds = loader.call(:self=>ds, :table_alias=>assoc_table_alias, :implicit_qualifier=>(ta == ds.opts[:eager_graph][:master]) ? first_source : qualifier_from_alias_symbol(ta, first_source), :callback=>callback, :join_type=>local_opts[:join_type], :join_only=>local_opts[:join_only], :limit_strategy=>limit_strategy, :from_self_alias=>ds.opts[:eager_graph][:master]) if r[:order_eager_graph] && (order = r.fetch(:graph_order, r[:order])) ds = ds.order_append(*qualified_expression(order, assoc_table_alias)) end eager_graph = ds.opts[:eager_graph] eager_graph[:requirements][assoc_table_alias] = requirements.dup eager_graph[:reflections][assoc_table_alias] = r if limit_strategy == :ruby eager_graph[:limits][assoc_table_alias] = r.limit_and_offset end eager_graph[:cartesian_product_number] += r[:cartesian_product_number] || 2 ds = ds.eager_graph_associations(ds, r.associated_class, assoc_table_alias, requirements + [assoc_table_alias], *associations) unless associations.empty? ds end # Check the associations are valid for the given model. # Call eager_graph_association on each association. # # Arguments: # ds :: Current dataset # model :: Current Model # ta :: table_alias used for the parent association # requirements :: an array, used as a stack for requirements # *associations :: the associations to add to the graph def eager_graph_associations(ds, model, ta, requirements, *associations) return ds if associations.empty? associations.flatten.each do |association| ds = case association when Symbol, SQL::AliasedExpression ds.eager_graph_association(ds, model, ta, requirements, eager_graph_check_association(model, association)) when Hash association.each do |assoc, assoc_assocs| ds = ds.eager_graph_association(ds, model, ta, requirements, eager_graph_check_association(model, assoc), assoc_assocs) end ds else raise(Sequel::Error, 'Associations must be in the form of a symbol or hash') end end ds end # Replace the array of plain hashes with an array of model objects will all eager_graphed # associations set in the associations cache for each object. def eager_graph_build_associations(hashes) hashes.replace(_eager_graph_build_associations(hashes, eager_graph_loader)) end private # Return a new dataset with JOINs of the given type added, using the tables and # conditions specified by the associations. def _association_join(type, associations) clone(:join=>clone(:graph_from_self=>false).eager_graph_with_options(associations, :join_type=>type, :join_only=>true).opts[:join]) end # Process the array of hashes using the eager graph loader to return an array # of model objects with the associations set. def _eager_graph_build_associations(hashes, egl) egl.load(hashes) end # If the association has conditions itself, then it requires additional filters be # added to the current dataset to ensure that the passed in object would also be # included by the association's conditions. def add_association_filter_conditions(ref, obj, expr) if expr != SQL::Constants::FALSE && ref.filter_by_associations_add_conditions? Sequel[ref.filter_by_associations_conditions_expression(obj)] else expr end end # Process the array of associations arguments (Symbols, Arrays, and Hashes), # and return a hash of options suitable for cascading. def eager_options_for_associations(associations) opts = {} associations.flatten.each do |association| case association when Symbol check_association(model, association) opts[association] = nil when Hash association.keys.each{|assoc| check_association(model, assoc)} opts.merge!(association) else raise(Sequel::Error, 'Associations must be in the form of a symbol or hash') end end opts end # Return an expression for filtering by the given association reflection and associated object. def association_filter_expression(op, ref, obj) meth = :"#{ref[:type]}_association_filter_expression" # Allow calling private association specific method to get filter expression send(meth, op, ref, obj) if respond_to?(meth, true) end # Handle inversion for association filters by returning an inverted expression, # plus also handling cases where the referenced columns are NULL. def association_filter_handle_inversion(op, exp, cols) if op == :'!=' || op == :'NOT IN' if exp == SQL::Constants::FALSE ~exp else ~exp | Sequel::SQL::BooleanExpression.from_value_pairs(cols.zip([]), :OR) end else exp end end # Return an expression for making sure that the given keys match the value of # the given methods for either the single object given or for any of the objects # given if +obj+ is an array. def association_filter_key_expression(keys, meths, obj) vals = if obj.is_a?(Sequel::Dataset) {(keys.length == 1 ? keys.first : keys)=>obj.select(*meths).exclude(Sequel::SQL::BooleanExpression.from_value_pairs(meths.zip([]), :OR))} else vals = Array(obj).reject{|o| !meths.all?{|m| o.get_column_value(m)}} return SQL::Constants::FALSE if vals.empty? if obj.is_a?(Array) if keys.length == 1 meth = meths.first {keys.first=>vals.map{|o| o.get_column_value(meth)}} else {keys=>vals.map{|o| meths.map{|m| o.get_column_value(m)}}} end else keys.zip(meths.map{|k| obj.get_column_value(k)}) end end SQL::BooleanExpression.from_value_pairs(vals) end # Make sure the association is valid for this model, and return the related AssociationReflection. def check_association(model, association) raise(Sequel::UndefinedAssociation, "Invalid association #{association} for #{model.name}") unless reflection = model.association_reflection(association) raise(Sequel::Error, "Eager loading is not allowed for #{model.name} association #{association}") if reflection[:allow_eager] == false reflection end # Allow associations that are eagerly graphed to be specified as an SQL::AliasedExpression, for # per-call determining of the alias base. def eager_graph_check_association(model, association) if association.is_a?(SQL::AliasedExpression) SQL::AliasedExpression.new(check_association(model, association.expression), association.alias) else check_association(model, association) end end # The EagerGraphLoader instance used for converting eager_graph results. def eager_graph_loader unless egl = cache_get(:_model_eager_graph_loader) egl = cache_set(:_model_eager_graph_loader, EagerGraphLoader.new(self)) end egl.dup end # Eagerly load all specified associations def eager_load(a, eager_assoc=@opts[:eager]) return if a.empty? # Key is foreign/primary key name symbol. # Value is hash with keys being foreign/primary key values (generally integers) # and values being an array of current model objects with that specific foreign/primary key key_hash = {} # Reflections for all associations to eager load reflections = eager_assoc.keys.map{|assoc| model.association_reflection(assoc) || (raise Sequel::UndefinedAssociation, "Model: #{self}, Association: #{assoc}")} # Populate the key_hash entry for each association being eagerly loaded reflections.each do |r| if key = r.eager_loader_key # key_hash for this key has already been populated, # skip populating again so that duplicate values # aren't added. unless id_map = key_hash[key] id_map = key_hash[key] = Hash.new{|h,k| h[k] = []} # Supporting both single (Symbol) and composite (Array) keys. a.each do |rec| case key when Array if (k = key.map{|k2| rec.get_column_value(k2)}) && k.all? id_map[k] << rec end when Symbol if k = rec.get_column_value(key) id_map[k] << rec end else raise Error, "unhandled eager_loader_key #{key.inspect} for association #{r[:name]}" end end end else id_map = nil end loader = r[:eager_loader] associations = eager_assoc[r[:name]] if associations.respond_to?(:call) eager_block = associations associations = OPTS elsif associations.is_a?(Hash) && associations.length == 1 && (pr_assoc = associations.to_a.first) && pr_assoc.first.respond_to?(:call) eager_block, associations = pr_assoc end loader.call(:key_hash=>key_hash, :rows=>a, :associations=>associations, :self=>self, :eager_block=>eager_block, :id_map=>id_map) a.each{|object| object.send(:run_association_callbacks, r, :after_load, object.associations[r[:name]])} if r[:after_load] end end # Return a subquery expression for filering by a many_to_many association def many_to_many_association_filter_expression(op, ref, obj) lpks, lks, rks = ref.values_at(:left_primary_key_columns, :left_keys, :right_keys) jt = ref.join_table_alias lpks = lpks.first if lpks.length == 1 lpks = ref.qualify(model.table_name, lpks) meths = if obj.is_a?(Sequel::Dataset) ref.qualify(obj.model.table_name, ref.right_primary_keys) else ref.right_primary_key_methods end expr = association_filter_key_expression(ref.qualify(jt, rks), meths, obj) unless expr == SQL::Constants::FALSE expr = SQL::BooleanExpression.from_value_pairs(lpks=>model.db.from(ref[:join_table]).select(*ref.qualify(jt, lks)).where(expr).exclude(SQL::BooleanExpression.from_value_pairs(ref.qualify(jt, lks).zip([]), :OR))) expr = add_association_filter_conditions(ref, obj, expr) end association_filter_handle_inversion(op, expr, Array(lpks)) end alias one_through_one_association_filter_expression many_to_many_association_filter_expression # Return a simple equality expression for filering by a many_to_one association def many_to_one_association_filter_expression(op, ref, obj) keys = ref.qualify(model.table_name, ref[:key_columns]) meths = if obj.is_a?(Sequel::Dataset) ref.qualify(obj.model.table_name, ref.primary_keys) else ref.primary_key_methods end expr = association_filter_key_expression(keys, meths, obj) expr = add_association_filter_conditions(ref, obj, expr) association_filter_handle_inversion(op, expr, keys) end # Return a simple equality expression for filering by a one_to_* association def one_to_many_association_filter_expression(op, ref, obj) keys = ref.qualify(model.table_name, ref[:primary_key_columns]) meths = if obj.is_a?(Sequel::Dataset) ref.qualify(obj.model.table_name, ref[:keys]) else ref[:key_methods] end expr = association_filter_key_expression(keys, meths, obj) expr = add_association_filter_conditions(ref, obj, expr) association_filter_handle_inversion(op, expr, keys) end alias one_to_one_association_filter_expression one_to_many_association_filter_expression def non_sql_option?(key) super || key == :eager || key == :eager_graph end # Build associations from the graph if #eager_graph was used, # and/or load other associations if #eager was used. def post_load(all_records) eager_graph_build_associations(all_records) if @opts[:eager_graph] eager_load(all_records) if @opts[:eager] && (row_proc || @opts[:eager_graph]) super end end # This class is the internal implementation of eager_graph. It is responsible for taking an array of plain # hashes and returning an array of model objects with all eager_graphed associations already set in the # association cache. class EagerGraphLoader # Hash with table alias symbol keys and after_load hook values attr_reader :after_load_map # Hash with table alias symbol keys and association name values attr_reader :alias_map # Hash with table alias symbol keys and subhash values mapping column_alias symbols to the # symbol of the real name of the column attr_reader :column_maps # Recursive hash with table alias symbol keys mapping to hashes with dependent table alias symbol keys. attr_reader :dependency_map # Hash with table alias symbol keys and [limit, offset] values attr_reader :limit_map # The table alias symbol for the primary model attr_reader :master # Hash with table alias symbol keys and primary key symbol values (or arrays of primary key symbols for # composite key tables) attr_reader :primary_keys # Hash with table alias symbol keys and reciprocal association symbol values, # used for setting reciprocals for one_to_many associations. attr_reader :reciprocal_map # Hash with table alias symbol keys and subhash values mapping primary key symbols (or array of symbols) # to model instances. Used so that only a single model instance is created for each object. attr_reader :records_map # Hash with table alias symbol keys and AssociationReflection values attr_reader :reflection_map # Hash with table alias symbol keys and callable values used to create model instances attr_reader :row_procs # Hash with table alias symbol keys and true/false values, where true means the # association represented by the table alias uses an array of values instead of # a single value (i.e. true => *_many, false => *_to_one). attr_reader :type_map # Initialize all of the data structures used during loading. def initialize(dataset) opts = dataset.opts eager_graph = opts[:eager_graph] @master = eager_graph[:master] requirements = eager_graph[:requirements] reflection_map = @reflection_map = eager_graph[:reflections] reciprocal_map = @reciprocal_map = eager_graph[:reciprocals] limit_map = @limit_map = eager_graph[:limits] @unique = eager_graph[:cartesian_product_number] > 1 alias_map = @alias_map = {} type_map = @type_map = {} after_load_map = @after_load_map = {} reflection_map.each do |k, v| alias_map[k] = v[:name] after_load_map[k] = v[:after_load] if v[:after_load] type_map[k] = if v.returns_array? true elsif (limit_and_offset = limit_map[k]) && !limit_and_offset.last.nil? :offset end end after_load_map.freeze alias_map.freeze type_map.freeze # Make dependency map hash out of requirements array for each association. # This builds a tree of dependencies that will be used for recursion # to ensure that all parts of the object graph are loaded into the # appropriate subordinate association. dependency_map = @dependency_map = {} # Sort the associations by requirements length, so that # requirements are added to the dependency hash before their # dependencies. requirements.sort_by{|a| a[1].length}.each do |ta, deps| if deps.empty? dependency_map[ta] = {} else deps = deps.dup hash = dependency_map[deps.shift] deps.each do |dep| hash = hash[dep] end hash[ta] = {} end end freezer = lambda do |h| h.freeze h.each_value(&freezer) end freezer.call(dependency_map) datasets = opts[:graph][:table_aliases].to_a.reject{|ta,ds| ds.nil?} column_aliases = opts[:graph][:column_aliases] primary_keys = {} column_maps = {} models = {} row_procs = {} datasets.each do |ta, ds| models[ta] = ds.model primary_keys[ta] = [] column_maps[ta] = {} row_procs[ta] = ds.row_proc end column_aliases.each do |col_alias, tc| ta, column = tc column_maps[ta][col_alias] = column end column_maps.each do |ta, h| pk = models[ta].primary_key if pk.is_a?(Array) primary_keys[ta] = [] h.select{|ca, c| primary_keys[ta] << ca if pk.include?(c)} else h.select{|ca, c| primary_keys[ta] = ca if pk == c} end end @column_maps = column_maps.freeze @primary_keys = primary_keys.freeze @row_procs = row_procs.freeze # For performance, create two special maps for the master table, # so you can skip a hash lookup. @master_column_map = column_maps[master] @master_primary_keys = primary_keys[master] # Add a special hash mapping table alias symbols to 5 element arrays that just # contain the data in other data structures for that table alias. This is # used for performance, to get all values in one hash lookup instead of # separate hash lookups for each data structure. ta_map = {} alias_map.each_key do |ta| ta_map[ta] = [row_procs[ta], alias_map[ta], type_map[ta], reciprocal_map[ta]].freeze end @ta_map = ta_map.freeze freeze end # Return an array of primary model instances with the associations cache prepopulated # for all model objects (both primary and associated). def load(hashes) # This mapping is used to make sure that duplicate entries in the # result set are mapped to a single record. For example, using a # single one_to_many association with 10 associated records, # the main object column values appear in the object graph 10 times. # We map by primary key, if available, or by the object's entire values, # if not. The mapping must be per table, so create sub maps for each table # alias. @records_map = records_map = {} alias_map.keys.each{|ta| records_map[ta] = {}} master = master() # Assign to local variables for speed increase rp = row_procs[master] rm = records_map[master] = {} dm = dependency_map records_map.freeze # This will hold the final record set that we will be replacing the object graph with. records = [] hashes.each do |h| unless key = master_pk(h) key = hkey(master_hfor(h)) end unless primary_record = rm[key] primary_record = rm[key] = rp.call(master_hfor(h)) # Only add it to the list of records to return if it is a new record records.push(primary_record) end # Build all associations for the current object and it's dependencies _load(dm, primary_record, h) end # Remove duplicate records from all associations if this graph could possibly be a cartesian product # Run after_load procs if there are any post_process(records, dm) if @unique || !after_load_map.empty? || !limit_map.empty? records_map.each_value(&:freeze) freeze records end private # Recursive method that creates associated model objects and associates them to the current model object. def _load(dependency_map, current, h) dependency_map.each do |ta, deps| unless key = pk(ta, h) ta_h = hfor(ta, h) unless ta_h.values.any? assoc_name = alias_map[ta] unless (assoc = current.associations).has_key?(assoc_name) assoc[assoc_name] = type_map[ta] ? [] : nil end next end key = hkey(ta_h) end rp, assoc_name, tm, rcm = @ta_map[ta] rm = records_map[ta] # Check type map for all dependencies, and use a unique # object if any are dependencies for multiple objects, # to prevent duplicate objects from showing up in the case # the normal duplicate removal code is not being used. if !@unique && !deps.empty? && deps.any?{|dep_key,_| @ta_map[dep_key][2]} key = [current.object_id, key] end unless rec = rm[key] rec = rm[key] = rp.call(hfor(ta, h)) end if tm unless (assoc = current.associations).has_key?(assoc_name) assoc[assoc_name] = [] end assoc[assoc_name].push(rec) rec.associations[rcm] = current if rcm else current.associations[assoc_name] ||= rec end # Recurse into dependencies of the current object _load(deps, rec, h) unless deps.empty? end end # Return the subhash for the specific table alias +ta+ by parsing the values out of the main hash +h+ def hfor(ta, h) out = {} @column_maps[ta].each{|ca, c| out[c] = h[ca]} out end # Return a suitable hash key for any subhash +h+, which is an array of values by column order. # This is only used if the primary key cannot be used. def hkey(h) h.sort_by{|x| x[0]} end # Return the subhash for the master table by parsing the values out of the main hash +h+ def master_hfor(h) out = {} @master_column_map.each{|ca, c| out[c] = h[ca]} out end # Return a primary key value for the master table by parsing it out of the main hash +h+. def master_pk(h) x = @master_primary_keys if x.is_a?(Array) unless x == [] x = x.map{|ca| h[ca]} x if x.all? end else h[x] end end # Return a primary key value for the given table alias by parsing it out of the main hash +h+. def pk(ta, h) x = primary_keys[ta] if x.is_a?(Array) unless x == [] x = x.map{|ca| h[ca]} x if x.all? end else h[x] end end # If the result set is the result of a cartesian product, then it is possible that # there are multiple records for each association when there should only be one. # In that case, for each object in all associations loaded via +eager_graph+, run # uniq! on the association to make sure no duplicate records show up. # Note that this can cause legitimate duplicate records to be removed. def post_process(records, dependency_map) records.each do |record| dependency_map.each do |ta, deps| assoc_name = alias_map[ta] list = record.public_send(assoc_name) rec_list = if type_map[ta] list.uniq! if lo = limit_map[ta] limit, offset = lo offset ||= 0 if type_map[ta] == :offset [record.associations[assoc_name] = list[offset]] else list.replace(list[(offset)..(limit ? (offset)+limit-1 : -1)] || []) end else list end elsif list [list] else [] end record.send(:run_association_callbacks, reflection_map[ta], :after_load, list) if after_load_map[ta] post_process(rec_list, deps) if !rec_list.empty? && !deps.empty? end end end end end end end