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_variable_set(:@association_reflections, {})
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
:"_add_#{singularize(self[:name])}"
end
# Name symbol for the _dataset association method
def _dataset_method
:"_#{self[:name]}_dataset"
end
# Name symbol for the _remove_all internal association method
def _remove_all_method
:"_remove_all_#{self[:name]}"
end
# Name symbol for the _remove internal association method
def _remove_method
:"_remove_#{singularize(self[:name])}"
end
# Name symbol for the _setter association method
def _setter_method
:"_#{self[:name]}="
end
# Name symbol for the add association method
def add_method
:"add_#{singularize(self[:name])}"
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
self[:class] ||= constantize(self[:class_name])
end
# Name symbol for the dataset association method
def dataset_method
:"#{self[:name]}_dataset"
end
# Name symbol for the _helper internal association method
def dataset_helper_method
:"_#{self[:name]}_dataset_helper"
end
# Whether the dataset needs a primary key to function, true by default.
def dataset_need_primary_key?
true
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 the associated object needs a primary key to be added/removed,
# false by default.
def need_associated_primary_key?
false
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
return self[:reciprocal] if include?(:reciprocal)
r_type = reciprocal_type
key = self[:key]
associated_class.all_association_reflections.each do |assoc_reflect|
if assoc_reflect[:type] == r_type && assoc_reflect[:key] == key && assoc_reflect.associated_class == self[:model]
return self[:reciprocal] = assoc_reflect[:name]
end
end
self[:reciprocal] = nil
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
:"remove_all_#{self[:name]}"
end
# Name symbol for the remove_ association method
def remove_method
:"remove_#{singularize(self[:name])}"
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, nil by default.
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[:name]}="
end
end
class ManyToOneAssociationReflection < AssociationReflection
ASSOCIATION_TYPES[:many_to_one] = self
# 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
# The key to use for the key hash when eager loading
def eager_loader_key
self[:key]
end
# The column in the associated table that the key in the current table references.
def primary_key
self[:primary_key] ||= associated_class.primary_key
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
private
# The reciprocal type of a many_to_one association is a one_to_many association.
def reciprocal_type
:one_to_many
end
end
class OneToManyAssociationReflection < AssociationReflection
ASSOCIATION_TYPES[:one_to_many] = self
# 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
# The column in the current table that the key in the associated table references.
def primary_key
self[:primary_key] ||= self[:model].primary_key
end
alias eager_loader_key primary_key
# One to many associations set the reciprocal to self when loading associated records.
def set_reciprocal_to_self?
true
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
private
# The reciprocal type of a one_to_many association is a many_to_one association.
def reciprocal_type
:many_to_one
end
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
# The column to use for the associated key when eagerly loading
def associated_key_column
self[:left_key]
end
# The table containing the column to use for the associated key when eagerly loading
def associated_key_table
self[:join_table]
end
# The default associated key alias
def default_associated_key_alias
:x_foreign_key_x
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
# The key to use for the key hash when eager loading
def eager_loader_key
self[:left_primary_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
# 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
# Returns the reciprocal association symbol, if one exists.
def reciprocal
return self[:reciprocal] if include?(:reciprocal)
left_key = self[:left_key]
right_key = self[:right_key]
join_table = self[:join_table]
associated_class.all_association_reflections.each do |assoc_reflect|
if assoc_reflect[:type] == :many_to_many && assoc_reflect[:left_key] == right_key &&
assoc_reflect[:right_key] == left_key && assoc_reflect[:join_table] == join_table &&
assoc_reflect.associated_class == self[:model]
return self[:reciprocal] = assoc_reflect[:name]
end
end
self[:reciprocal] = nil
end
# The primary key column to use in the associated table.
def right_primary_key
self[:right_primary_key] ||= associated_class.primary_key
end
# The columns to select when loading the association, associated_class.table_name.* by default.
def select
return self[:select] if include?(:select)
self[:select] ||= Sequel::SQL::ColumnAll.new(associated_class.table_name)
end
end
# This module contains methods added to all association datasets
module AssociationDatasetMethods
# The model object that created the association dataset
attr_accessor :model_object
# The association reflection related to the association dataset
attr_accessor :association_reflection
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
# 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.
# * 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,
# there are private instance methods created that a prepended with an
# underscore (e.g. _add_milestone). The private instance methods can be
# easily overridden, but you shouldn't override the public instance methods without
# calling super, as they deal with callbacks and caching.
#
# 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.
#
# Association definitions are also reflected by the class, e.g.:
#
# Project.associations
# => [:portfolio, :milestones]
# Project.association_reflection(:portfolio)
# => {:type => :many_to_one, :name => :portfolio, :class_name => "Portfolio"}
module ClassMethods
# All association reflections defined for this model (default: none).
attr_reader :association_reflections
# 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.
# * :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.
#
# A one to one relationship can be set up with a many_to_one association
# on the table with the foreign key, and a one_to_many association with the
# :one_to_one option specified on the table without the foreign key. The
# two associations will operate similarly, except that the many_to_one
# association setter doesn't update the database until you call save manually.
# Also, in most cases you need to specify the plural association name when using
# one_to_many with the :one_to_one option.
#
# The following options can be supplied:
# * *ALL types*:
# - :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. Not called
# when eager loading via eager_graph, but called when eager loading via eager.
# - :after_remove - Symbol, Proc, or array of both/either specifying a callback to call
# after an item is removed from the association.
# - :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.
# - :cartesian_product_number - he 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 associations,
# 1 for *_to_many associations).
# - :class - The associated class or its name. If not
# given, uses the association's name, which is camelized (and
# singularized unless the type is :many_to_one)
# - :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 filter.
# - :dataset - A proc that is instance_evaled to get the base dataset
# to use for the _dataset method (before the other options are applied).
# - :eager - The associations to eagerly load via #eager when loading the associated object(s).
# For many_to_one associations, this is ignored unless this association is
# being eagerly loaded, as it doesn't save queries unless multiple objects
# can be loaded at once.
# - :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).
# For many_to_one associations, this is ignored unless this association is
# being eagerly loaded, as it doesn't save queries unless multiple objects
# can be loaded at once.
# - :eager_grapher - A proc to use to implement eager loading via eager graph, overriding the default.
# Takes three arguments, a dataset, an alias to use for the table to graph for this association,
# and the alias that was used for the current table (since you can cascade associations),
# Should return a copy of the dataset with the association graphed into it.
# - :eager_loader - A proc to use to implement eager loading, overriding the default. Takes three arguments,
# a key hash (used solely to enhance performance), an array of records,
# and a hash of dependent associations. The associated records should
# be queried from the database and the associations cache for each
# record should be populated for this to work correctly.
# - :extend - A module or array of modules to extend the dataset with.
# - :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 all two pairs. If not
# specified, the :conditions option is used if it is a hash or array of all two pairs.
# - :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_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 arguments for the value to specify a limit and an offset.
# - :order - the column(s) by which to order the association dataset. Can be a
# singular column or an array.
# - :order_eager_graph - Whether to add the order to the 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_many with :one_to_one),
# or add_/remove_/remove_all_ methods (for one_to_many, many_to_many)
# - :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.
# - :select - the attributes to select. Defaults to the associated class's
# table_name.* in a many_to_many association, 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.
# * :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".
# - :primary_key - column in the associated table that :key option references, as a symbol.
# Defaults to the primary key of the associated table.
# * :one_to_many:
# - :key - foreign key in associated model's table that references
# current model's primary key, as a symbol. Defaults to
# :"#{self.name.underscore}_id".
# - :one_to_one: Create a getter and setter similar to those of many_to_one
# associations. The getter returns a singular matching record, or raises an
# error if multiple records match. The setter updates the record given and removes
# associations with all other records. When this option is used, the other
# association methods usually added are either removed or made private,
# so using this is similar to using many_to_one, in terms of the methods
# it adds, the main difference is that the foreign key is in the associated
# table instead of the current table. Note that using this option still requires
# you to use a plural name when creating and using the association (e.g. for reflections, eager loading, etc.).
# - :primary_key - column in the current table that :key option references, as a symbol.
# Defaults to primary key of the current table.
# * :many_to_many:
# - :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 all two pairs.
# - :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 '_'.
# - :left_key - foreign key in join table that points to current model's
# primary key, as a symbol. Defaults to :"#{self.name.underscore}_id".
# - :left_primary_key - column in current table that :left_key points to, as a symbol.
# Defaults to primary key of current table.
# - :right_key - foreign key in join table that points to associated
# model's primary key, as a symbol. Defaults to Defaults to :"#{name.to_s.singularize}_id".
# - :right_primary_key - column in associated table that :right_key points to, as a symbol.
# Defaults to primary key of the associated table.
# - :uniq - Adds a after_load callback that makes the array of objects unique.
def associate(type, name, opts = {}, &block)
raise(Error, 'invalid association type') unless assoc_class = ASSOCIATION_TYPES[type]
raise(Error, 'Model.associate name argument must be a symbol') unless Symbol === name
# merge early so we don't modify opts
orig_opts = opts.dup
orig_opts = association_reflection(opts[:clone])[:orig_opts].merge(orig_opts) if opts[:clone]
opts = orig_opts.merge(:type => type, :name => name, :cache => true, :model => self)
opts[:block] = block if block
opts = assoc_class.new.merge!(opts)
opts[:eager_block] = 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_conditions] = conds if !opts.include?(:graph_conditions) and Sequel.condition_specifier?(conds)
opts[:graph_conditions] = opts[:graph_conditions] ? opts[: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, :extend].each do |cb_type|
opts[cb_type] = Array(opts[cb_type])
end
# find class
case opts[:class]
when String, Symbol
# Delete :class to allow late binding
opts[:class_name] ||= opts.delete(:class).to_s
when Class
opts[:class_name] ||= opts[:class].name
end
opts[:class_name] ||= ((self.name || '').split("::")[0..-2] + [camelize(opts.returns_array? ? singularize(name) : name)]).join('::')
send(:"def_#{type}", opts)
orig_opts.delete(:clone)
orig_opts.merge!(:class_name=>opts[:class_name], :class=>opts[:class], :block=>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
# Modify and return eager loading dataset based on association options. Options:
def eager_loading_dataset(opts, ds, select, associations)
ds = ds.select(*select) if select
if c = opts[:conditions]
ds = (c.is_a?(Array) && !Sequel.condition_specifier?(c)) ? ds.filter(*c) : ds.filter(c)
end
ds = ds.order(*opts[:order]) if opts[:order]
ds = ds.eager(opts[:eager]) if opts[:eager]
if opts[:eager_graph]
ds = ds.eager_graph(opts[:eager_graph])
ds = ds.add_graph_aliases(opts.associated_key_alias=>[opts.associated_class.table_name, opts.associated_key_alias, SQL::QualifiedIdentifier.new(opts.associated_key_table, opts.associated_key_column)]) if opts.eager_loading_use_associated_key?
else
ds.select_more(SQL::AliasedExpression.new(SQL::QualifiedIdentifier.new(opts.associated_key_table, opts.associated_key_column), opts.associated_key_alias)) if opts.eager_loading_use_associated_key?
end
ds = ds.eager(associations) unless Array(associations).empty?
ds = opts[:eager_block].call(ds) if opts[:eager_block]
ds
end
# Copy the association reflections to the subclass
def inherited(subclass)
super
subclass.instance_variable_set(:@association_reflections, @association_reflections.dup)
end
# Shortcut for adding a many_to_many association, see associate
def many_to_many(*args, &block)
associate(:many_to_many, *args, &block)
end
# Shortcut for adding a many_to_one association, see associate
def many_to_one(*args, &block)
associate(:many_to_one, *args, &block)
end
# Shortcut for adding a one_to_many association, see associate
def one_to_many(*args, &block)
associate(:one_to_many, *args, &block)
end
private
# 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, &block)
overridable_methods_module.module_eval{define_method(name, &block)}
end
# Add a private method to the module included in the class.
def association_module_private_def(name, &block)
association_module_def(name, &block)
overridable_methods_module.send(:private, name)
end
# Add the add_ instance method
def def_add_method(opts)
association_module_def(opts.add_method){|o,*args| add_associated_object(opts, o, *args)}
end
# Adds methods related to the association's dataset to the module included in the class.
def def_association_dataset_methods(opts)
# If a block is given, define a helper method for it, because it takes
# an argument. This is unnecessary in Ruby 1.9, as that has instance_exec.
association_module_private_def(opts.dataset_helper_method, &opts[:block]) if opts[:block]
association_module_private_def(opts._dataset_method, &opts[:dataset])
association_module_def(opts.dataset_method){_dataset(opts)}
def_association_method(opts)
end
# Adds method for retrieving the associated objects to the module included in the class.
def def_association_method(opts)
association_module_def(opts.association_method){|*reload| load_associated_objects(opts, reload[0])}
end
# Adds many_to_many association instance methods
def def_many_to_many(opts)
name = opts[:name]
model = self
left = (opts[:left_key] ||= opts.default_left_key)
right = (opts[:right_key] ||= opts.default_right_key)
left_pk = (opts[:left_primary_key] ||= self.primary_key)
opts[:cartesian_product_number] ||= 1
join_table = (opts[:join_table] ||= opts.default_join_table)
left_key_alias = opts[:left_key_alias] ||= opts.default_associated_key_alias
graph_jt_conds = opts[:graph_join_table_conditions] = opts[:graph_join_table_conditions] ? opts[:graph_join_table_conditions].to_a : []
opts[:graph_join_table_join_type] ||= opts[:graph_join_type]
opts[:after_load].unshift(:array_uniq!) if opts[:uniq]
opts[:dataset] ||= proc{opts.associated_class.inner_join(join_table, [[right, opts.right_primary_key], [left, send(left_pk)]])}
database = db
opts[:eager_loader] ||= proc do |key_hash, records, associations|
h = key_hash[left_pk]
records.each{|object| object.associations[name] = []}
model.eager_loading_dataset(opts, opts.associated_class.inner_join(join_table, [[right, opts.right_primary_key], [left, h.keys]]), Array(opts.select), associations).all do |assoc_record|
next unless objects = h[assoc_record.values.delete(left_key_alias)]
objects.each{|object| object.associations[name].push(assoc_record)}
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]
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 |ds, assoc_alias, table_alias|
ds = ds.graph(join_table, use_jt_only_conditions ? jt_only_conditions : [[left, left_pk]] + graph_jt_conds, :select=>false, :table_alias=>ds.send(:eager_unique_table_alias, ds, join_table), :join_type=>jt_join_type, :implicit_qualifier=>table_alias, &jt_graph_block)
ds.graph(opts.associated_class, use_only_conditions ? only_conditions : [[opts.right_primary_key, right]] + conditions, :select=>select, :table_alias=>assoc_alias, :join_type=>join_type, &graph_block)
end
def_association_dataset_methods(opts)
return if opts[:read_only]
association_module_private_def(opts._add_method) do |o|
database.dataset.from(join_table).insert(left=>send(left_pk), right=>o.send(opts.right_primary_key))
end
association_module_private_def(opts._remove_method) do |o|
database.dataset.from(join_table).filter([[left, send(left_pk)], [right, o.send(opts.right_primary_key)]]).delete
end
association_module_private_def(opts._remove_all_method) do
database.dataset.from(join_table).filter(left=>send(left_pk)).delete
end
def_add_method(opts)
def_remove_methods(opts)
end
# Adds many_to_one association instance methods
def def_many_to_one(opts)
name = opts[:name]
model = self
opts[:key] = opts.default_key unless opts.include?(:key)
key = opts[:key]
opts[:cartesian_product_number] ||= 0
opts[:dataset] ||= proc do
klass = opts.associated_class
klass.filter(SQL::QualifiedIdentifier.new(klass.table_name, opts.primary_key)=>send(key))
end
opts[:eager_loader] ||= proc do |key_hash, records, associations|
h = key_hash[key]
keys = h.keys
# Default the cached association to nil, so any object that doesn't have it
# populated will have cached the negative lookup.
records.each{|object| object.associations[name] = nil}
# Skip eager loading if no objects have a foreign key for this association
unless keys.empty?
klass = opts.associated_class
model.eager_loading_dataset(opts, klass.filter(SQL::QualifiedIdentifier.new(klass.table_name, opts.primary_key)=>keys), opts.select, associations).all do |assoc_record|
next unless objects = h[assoc_record.send(opts.primary_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]
opts[:eager_grapher] ||= proc do |ds, assoc_alias, table_alias|
ds.graph(opts.associated_class, use_only_conditions ? only_conditions : [[opts.primary_key, key]] + conditions, :select=>select, :table_alias=>assoc_alias, :join_type=>join_type, :implicit_qualifier=>table_alias, &graph_block)
end
def_association_dataset_methods(opts)
return if opts[:read_only]
association_module_private_def(opts._setter_method){|o| send(:"#{key}=", (o.send(opts.primary_key) if o))}
association_module_def(opts.setter_method){|o| set_associated_object(opts, o)}
end
# Adds one_to_many association instance methods
def def_one_to_many(opts)
name = opts[:name]
model = self
key = (opts[:key] ||= opts.default_key)
primary_key = (opts[:primary_key] ||= self.primary_key)
opts[:dataset] ||= proc do
klass = opts.associated_class
klass.filter(SQL::QualifiedIdentifier.new(klass.table_name, key) => send(primary_key))
end
opts[:eager_loader] ||= proc do |key_hash, records, associations|
h = key_hash[primary_key]
records.each{|object| object.associations[name] = []}
reciprocal = opts.reciprocal
klass = opts.associated_class
model.eager_loading_dataset(opts, klass.filter(SQL::QualifiedIdentifier.new(klass.table_name, key)=>h.keys), opts.select, associations).all do |assoc_record|
next unless objects = h[assoc_record[key]]
objects.each do |object|
object.associations[name].push(assoc_record)
assoc_record.associations[reciprocal] = object if reciprocal
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] ||= 1
graph_block = opts[:graph_block]
opts[:eager_grapher] ||= proc do |ds, assoc_alias, table_alias|
ds = ds.graph(opts.associated_class, use_only_conditions ? only_conditions : [[key, primary_key]] + conditions, :select=>select, :table_alias=>assoc_alias, :join_type=>join_type, :implicit_qualifier=>table_alias, &graph_block)
# We only load reciprocals for one_to_many associations, as other reciprocals don't make sense
ds.opts[:eager_graph][:reciprocals][assoc_alias] = opts.reciprocal
ds
end
def_association_dataset_methods(opts)
unless opts[:read_only]
association_module_private_def(opts._add_method) do |o|
o.send(:"#{key}=", send(primary_key))
o.save || raise(Sequel::Error, "invalid associated object, cannot save")
end
def_add_method(opts)
unless opts[:one_to_one]
association_module_private_def(opts._remove_method) do |o|
o.send(:"#{key}=", nil)
o.save || raise(Sequel::Error, "invalid associated object, cannot save")
end
association_module_private_def(opts._remove_all_method) do
opts.associated_class.filter(key=>send(primary_key)).update(key=>nil)
end
def_remove_methods(opts)
end
end
if opts[:one_to_one]
overridable_methods_module.send(:private, opts.association_method, opts.dataset_method)
n = singularize(name).to_sym
raise(Sequel::Error, "one_to_many association names should still be plural even when using the :one_to_one option") if n == name
association_module_def(n) do |*o|
objs = send(name, *o)
raise(Sequel::Error, "multiple values found for a one-to-one relationship") if objs.length > 1
objs.first
end
unless opts[:read_only]
overridable_methods_module.send(:private, opts.add_method)
association_module_def(:"#{n}=") do |o|
klass = opts.associated_class
update_database = lambda do
send(opts.add_method, o)
klass.filter(Sequel::SQL::BooleanExpression.new(:AND, {key=>send(primary_key)}, SQL::BooleanExpression.new(:'!=', klass.primary_key, o.pk))).update(key=>nil)
end
use_transactions ? db.transaction(opts){update_database.call} : update_database.call
end
end
end
end
# Add the remove_ and remove_all instance methods
def def_remove_methods(opts)
association_module_def(opts.remove_method){|o,*args| remove_associated_object(opts, o, *args)}
association_module_def(opts.remove_all_method){|*args| remove_all_associated_objects(opts, *args)}
end
end
# Private instance methods used to implement the associations support.
module InstanceMethods
# Used internally by the associations code, like pk but doesn't raise
# an Error if the model has no primary key.
def pk_or_nil
key = primary_key
key.is_a?(Array) ? key.map{|k| @values[k]} : @values[key]
end
private
# Backbone behind association dataset methods
def _dataset(opts)
raise(Sequel::Error, "model object #{model} does not have a primary key") if opts.dataset_need_primary_key? && !pk
ds = send(opts._dataset_method)
ds.extend(AssociationDatasetMethods)
ds.model_object = self
ds.association_reflection = opts
opts[:extend].each{|m| ds.extend(m)}
ds = ds.select(*opts.select) if opts.select
if c = opts[:conditions]
ds = (c.is_a?(Array) && !Sequel.condition_specifier?(c)) ? ds.filter(*c) : ds.filter(c)
end
ds = ds.order(*opts[:order]) if opts[:order]
ds = ds.limit(*opts[:limit]) if opts[:limit]
ds = ds.eager(*opts[:eager]) if opts[:eager]
ds = ds.eager_graph(opts[:eager_graph]) if opts[:eager_graph] && opts.eager_graph_lazy_dataset?
ds = send(opts.dataset_helper_method, ds) if opts[:block]
ds
end
# Return the associated objects from the dataset, without callbacks, reciprocals, and caching.
def _load_associated_objects(opts)
if opts.returns_array?
send(opts.dataset_method).all
else
if !opts[:key]
send(opts.dataset_method).all.first
elsif send(opts[:key])
send(opts.dataset_method).first
end
end
end
# Add the given associated object to the given association
def add_associated_object(opts, o, *args)
raise(Sequel::Error, "model object #{model} does not have a primary key") unless pk
raise(Sequel::Error, "associated object #{o.model} does not have a primary key") if opts.need_associated_primary_key? && !o.pk
return if run_association_callbacks(opts, :before_add, o) == false
send(opts._add_method, o, *args)
associations[opts[:name]].push(o) if associations.include?(opts[:name])
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 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
# Load the associated objects using the dataset, handling callbacks, reciprocals, and caching.
def load_associated_objects(opts, reload=false)
name = opts[:name]
if associations.include?(name) and !reload
associations[name]
else
objs = _load_associated_objects(opts)
run_association_callbacks(opts, :after_load, objs)
objs.each{|o| add_reciprocal_object(opts, o)} if opts.set_reciprocal_to_self?
associations[name] = objs
end
end
# Remove all associated objects from the given association
def remove_all_associated_objects(opts, *args)
raise(Sequel::Error, "model object #{model} does not have a primary key") unless pk
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)
raise(Sequel::Error, "model object #{model} does not have a primary key") unless pk
raise(Sequel::Error, "associated object #{o.model} does not have a primary key") if opts.need_associated_primary_key? && !o.pk
return if run_association_callbacks(opts, :before_remove, o) == false
send(opts._remove_method, o, *args)
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
# 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)
raise_error = raise_on_save_failure || !reflection.returns_array?
stop_on_false = [:before_add, :before_remove].include?(callback_type)
reflection[callback_type].each do |cb|
res = case cb
when Symbol
send(cb, object)
when Proc
cb.call(self, object)
else
raise Error, "callbacks should either be Procs or Symbols"
end
if res == false and stop_on_false
raise(BeforeHookFailed, "Unable to modify association for record: one of the #{callback_type} hooks returned false") if raise_error
return false
end
end
end
# Set the given object as the associated object for the given association
def set_associated_object(opts, o)
raise(Sequel::Error, "model object #{model} does not have a primary key") if o && !o.pk
old_val = send(opts.association_method)
return o if old_val == o
return if old_val and run_association_callbacks(opts, :before_remove, old_val) == false
return if o and run_association_callbacks(opts, :before_add, o) == false
send(opts._setter_method, o)
associations[opts[:name]] = o
remove_reciprocal_object(opts, old_val) if old_val
if o
add_reciprocal_object(opts, o)
run_association_callbacks(opts, :after_add, o)
end
run_association_callbacks(opts, :after_remove, old_val) if old_val
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 based on columns in associated tables. #eager_graph loads
# all records in one query. Using #eager_graph you can filter based on columns in associated
# tables. However, #eager_graph can be slower than #eager, especially if multiple
# *_to_many associations are joined.
#
# You can cascade the eager loading (loading associations' associations)
# 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(: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
module DatasetMethods
# Add the #eager! and #eager_graph! mutation methods to the dataset.
def self.extended(obj)
obj.def_mutation_method(:eager, :eager_graph)
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 *_to_many eager associations are requested.
#
# One limitation of using this method is that you cannot filter the 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. Eager also works
# on a limited dataset, but does not use any :limit options for associations.
# If the association uses a block or has an :eager_block argument, it is used.
def eager(*associations)
opt = @opts[:eager]
opt = opt ? opt.dup : {}
associations.flatten.each do |association|
case association
when Symbol
check_association(model, association)
opt[association] = nil
when Hash
association.keys.each{|assoc| check_association(model, assoc)}
opt.merge!(association)
else raise(Sequel::Error, 'Associations must be in the form of a symbol or hash')
end
end
clone(:eager=>opt)
end
# The secondary eager loading method. Loads all associations in a single query. This
# method should only be used if you need to filter based on columns in associated tables.
#
# This method builds an object graph using Dataset#graph. Then it uses the graph
# to build the associations, and finally replaces the graph with a simple array
# of model objects.
#
# Be very careful when using this with multiple *_to_many associations, as you can
# create large cartesian products. If you must graph multiple *_to_many associations,
# make sure your filters are specific if you have a large database.
#
# Each association's order, if definied, is respected. #eager_graph probably
# won't work correctly on a limited dataset, unless you are
# only graphing many_to_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, you will get a normal graphed result back (a hash with model object values).
def eager_graph(*associations)
table_name = model.table_name
ds = if @opts[:eager_graph]
self
else
# Each of the following have a symbol key for the table alias, with the following values:
# :reciprocals - the reciprocal instance variable to use for this association
# :requirements - array of requirements for this association
# :alias_association_type_map - the type of association for this association
# :alias_association_name_map - the name of the association for this association
clone(:eager_graph=>{:requirements=>{}, :master=>model.table_name, :alias_association_type_map=>{}, :alias_association_name_map=>{}, :reciprocals=>{}, :cartesian_product_number=>0})
end
ds.eager_graph_associations(ds, model, table_name, [], *associations)
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
super.clone(:eager_graph=>nil)
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
# * *associations - any associations dependent on this one
def eager_graph_association(ds, model, ta, requirements, r, *associations)
klass = r.associated_class
assoc_name = r[:name]
assoc_table_alias = ds.eager_unique_table_alias(ds, assoc_name)
ds = r[:eager_grapher].call(ds, assoc_table_alias, ta)
ds = ds.order_more(*Array(r[:order]).map{|c| eager_graph_qualify_order(assoc_table_alias, c)}) if r[:order] and r[:order_eager_graph]
eager_graph = ds.opts[:eager_graph]
eager_graph[:requirements][assoc_table_alias] = requirements.dup
eager_graph[:alias_association_name_map][assoc_table_alias] = assoc_name
eager_graph[:alias_association_type_map][assoc_table_alias] = r.returns_array?
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
ds.eager_graph_association(ds, model, ta, requirements, check_association(model, association))
when Hash
association.each do |assoc, assoc_assocs|
ds = ds.eager_graph_association(ds, model, ta, requirements, 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
# Build associations out of the array of returned object graphs.
def eager_graph_build_associations(record_graphs)
eager_graph = @opts[:eager_graph]
master = eager_graph[:master]
requirements = eager_graph[:requirements]
alias_map = eager_graph[:alias_association_name_map]
type_map = eager_graph[:alias_association_type_map]
reciprocal_map = eager_graph[:reciprocals]
# 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 = {}
# 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
# 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 will 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 = {master=>{}}
alias_map.keys.each{|ta| records_map[ta] = {}}
# This will hold the final record set that we will be replacing the object graph with.
records = []
record_graphs.each do |record_graph|
primary_record = record_graph[master]
key = primary_record.pk_or_nil || primary_record.values.sort_by{|x| x[0].to_s}
if cached_pr = records_map[master][key]
primary_record = cached_pr
else
records_map[master][key] = primary_record
# 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
eager_graph_build_associations_graph(dependency_map, alias_map, type_map, reciprocal_map, records_map, primary_record, record_graph)
end
# Remove duplicate records from all associations if this graph could possibly be a cartesian product
eager_graph_make_associations_unique(records, dependency_map, alias_map, type_map) if eager_graph[:cartesian_product_number] > 1
# Replace the array of object graphs with an array of model objects
record_graphs.replace(records)
end
# Creates a unique table alias that hasn't already been used in the query.
# Will either be the table_alias itself or table_alias_N for some integer
# N (starting at 0 and increasing until an unused one is found).
def eager_unique_table_alias(ds, table_alias)
used_aliases = ds.opts[:from]
used_aliases += ds.opts[:join].map{|j| j.table_alias || j.table} if ds.opts[:join]
graph = ds.opts[:graph]
used_aliases += graph[:table_aliases].keys if graph
if used_aliases.include?(table_alias)
i = 0
loop do
ta = :"#{table_alias}_#{i}"
return ta unless used_aliases.include?(ta)
i += 1
end
end
table_alias
end
private
# Make sure the association is valid for this model, and return the related AssociationReflection.
def check_association(model, association)
raise(Sequel::Error, 'Invalid association') 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
# Build associations for the current object. This is called recursively
# to build all dependencies.
def eager_graph_build_associations_graph(dependency_map, alias_map, type_map, reciprocal_map, records_map, current, record_graph)
return if dependency_map.empty?
# Don't clobber the instance variable array for *_to_many associations if it has already been setup
dependency_map.keys.each do |ta|
assoc_name = alias_map[ta]
current.associations[assoc_name] = type_map[ta] ? [] : nil unless current.associations.include?(assoc_name)
end
dependency_map.each do |ta, deps|
next unless rec = record_graph[ta]
key = rec.pk_or_nil || rec.values.sort_by{|x| x[0].to_s}
if cached_rec = records_map[ta][key]
rec = cached_rec
else
records_map[ta][key] = rec
end
assoc_name = alias_map[ta]
if type_map[ta]
current.associations[assoc_name].push(rec)
if reciprocal = reciprocal_map[ta]
rec.associations[reciprocal] = current
end
else
current.associations[assoc_name] = rec
end
# Recurse into dependencies of the current object
eager_graph_build_associations_graph(deps, alias_map, type_map, reciprocal_map, records_map, rec, record_graph)
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 eager_graph_make_associations_unique(records, dependency_map, alias_map, type_map)
records.each do |record|
dependency_map.each do |ta, deps|
list = record.send(alias_map[ta])
list = if type_map[ta]
list.uniq!
else
[list] if list
end
# Recurse into dependencies
eager_graph_make_associations_unique(list, deps, alias_map, type_map) if list
end
end
end
# Qualify the given expression if necessary. The only expressions which are qualified are
# unqualified symbols and identifiers, either of which may by sorted.
def eager_graph_qualify_order(table_alias, expression)
case expression
when Symbol
table, column, aliaz = split_symbol(expression)
raise(Sequel::Error, "Can't use an aliased expression in the :order option") if aliaz
table ? expression : Sequel::SQL::QualifiedIdentifier.new(table_alias, expression)
when Sequel::SQL::Identifier
Sequel::SQL::QualifiedIdentifier.new(table_alias, expression)
when Sequel::SQL::OrderedExpression
Sequel::SQL::OrderedExpression.new(eager_graph_qualify_order(table_alias, expression.expression), expression.descending)
else
expression
end
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.collect{|assoc| model.association_reflection(assoc)}
# Populate keys to monitor
reflections.each{|reflection| key_hash[reflection.eager_loader_key] ||= Hash.new{|h,k| h[k] = []}}
# Associate each object with every key being monitored
a.each do |rec|
key_hash.each do |key, id_map|
id_map[rec[key]] << rec if rec[key]
end
end
reflections.each do |r|
r[:eager_loader].call(key_hash, a, eager_assoc[r[:name]])
a.each{|object| object.send(:run_association_callbacks, r, :after_load, object.associations[r[:name]])} unless r[:after_load].empty?
end
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]
super
end
end
end
end
end