When I create a new pet, the pet is associated with a User through habtm. These are the classes:
class Pet < ActiveRecord::Base has_and_belongs_to_many :users end
class User < ActiveRecord::Base has_and_belongs_to_many :pets validates_uniqueness_of :screen_name, :email end
This is the code in the controller for adding the new pet, which updates the join table:
@pet = Pet.new(params[:pet]) @user = User.find_by_id(user_id) @pet.users << @user @pet.save
This makes a lot of calls to the database:
SHOW FIELDS FROM `users` SELECT * FROM `users` WHERE (`users`.`id` = '11') LIMIT 1 BEGIN COMMIT BEGIN (these next two lines are validations) SELECT * FROM `users` WHERE (users.screen_name = 'chairs' AND users.id <> 11) LIMIT 1 SELECT * FROM `users` WHERE (users.email = 'chairs@foo.com' AND users.id <> 11) LIMIT 1 INSERT INTO `pets` (`name`, `updated_at`, `description`, `created_at`) VALUES('Kitty Kitty', '2008-02-28 18:08:36', 'Lost cat in Washington.', '2008-02-28 18:08:36')e INSERT INTO users_pets (`user_id`, `pet_id`) VALUES (11, 2)e COMMIT
Is there any way of reducing the number of calls to the database? Is there a way to suppress the validation calls? I also don't understand why it makes the first BEGIN/COMMIT calls. Or is there a better way to do this that is more efficient? I could do custom SQL commands, but I was hoping to do it the rails way.
