how to enforce inclusion rules in ActiveRecord associations - ruby-on-rails

I'm new to rails and I have a question on how best to enforce custom rules on my model associations.
For example, suppose I have:
class Person < ActiveRecord::Base
belongs_to :organization
end
class Organization < ActiveRecord::Base
has_many :people
end
and now suppose that I only want to allow the Organization.people << Person.new(...) command to succeed if the new Person object is compatible with the other people that were previously added to the Organization. This would entail running a validation check across all the existing elements of Organization.people and deciding whether the new Person could be added or not.
It seems to me that I can do this by overriding all the Organization.people assignment operators (such as << and =) and putting my validation logic in the override routine.
Is this best way to accomplish this?
Thanks!

I think you could put a validation in the Person class. It would run a test against the other people in self.organiation.people. I don't know if I would override the << on the has many relationship only because if you decide to create a person like Person.new(:organization => some_org) your << override would not get used. If the validation lives on the Person class, it would get exercises no matter how you create the person.

Related

Defining attributes at runtime based on data from related object

I'm building an application where users are part of an Organisation. An organisation has many Lists, which in turn have many ListItems.
Now, I would like for admin users to be able to specify which attributes are available on list items, based on the organisation they belong to (or rather, on the organisation their list belongs to), without having to touch any code.
So far, when defining attributes that are not bound to a specific column in the database, I have used document_serializable, a nifty little gem (based on virtus) which serializes virtual attributes to a JSONB column in the db. I like this approach, because I get all of virtus' goodies (types, coercion, validations, etc.), and because data ends up sitting in a JSONB column, meaning it can be loaded quickly, indexed, and searched through with relative ease.
I would like to keep using this approach when adding these user-defined attributes on the fly. So I'd like to do something like:
class ListItem < ApplicationRecord
belongs_to :list
delegate :organisation, to: :list
organisation.list_attributes.each do |a, t|
attribute a, t
end
end
Where Organisation#list_attributes returns the user-defined hash of attribute names and their associated types, which, for example, might look like:
{
name: String,
age: Integer
}
As you might have guessed, this does not work, because organisation.list_attributes.each actually runs in the context of ListItem, which is an instance of Class, and Class doesn't have an #organisation method. I hope that's worded in a way that makes sense1.
I've tried using after_initialize, but at that point in the object's lifecycle, #attribute is owned by ActiveRecord::AttributeMethods::Read and not DocumentSerializable::ClassMethods, so it's an entirely different method and I can't figure out wether I can still access the one I need, and wether that would even work.
Another alternative would be to find the organisation in question in some explicit way, Organisation#find-style, but I honestly don't know where I should store the information necessary to do so.
So, my question: at the moment of instantiating (initializing or loading2) a record, is there a way I can retrieve a hash stored in a database column of one of its relations? Or am I trying to build this in a completely misguided way, and if so, how else should I go about it?
1 To clarify, if I were to use the hash directly like so:
class ListItem < ApplicationRecord
belongs_to :list
delegate :organisation, to: :list
{
name: String,
age: Integer
}.each do |a, t|
attribute a, t
end
end
it would work, my issue is solely with getting a record's relation at this earlier point in time.
2 My understanding is that Rails runs a model's code whenever a record of that type is created or loaded from the database, meaning the virtual attributes are defined anew every time this happens, which is why I'm asking how to do this in both cases.
at the moment of instantiating (initializing or loading) a record, is
there a way I can retrieve a hash stored in a database column of one
of its relations?
Yes. This is fairly trivial as long as your relations are setup correctly / simply. Lets say we have these three models:
class ListItem < ApplicationRecord
belongs_to :list
end
class List < ApplicationRecord
belongs_to :organisation
has_many :list_items
end
class Organisation < ApplicationRecord
has_many :lists
end
We can instantiate a ListItem and then retrieve data from anyone of its parents.
#list_item = ListItem.find(5) # assume that the proper inherited
foreign_keys exist for this and
its parent
#list = #list_item.list
#hash = #list.organisation.special_hash_of_org
And if we wanted to do this at every instance of a ListItem, we can use Active Record Callbacks like this:
class ListItem < ApplicationRecord
belongs_to :list
# this is called on ListItem.new and whenever we pull from our DB
after_initialize do |list_item|
puts "You have initialized a ListItem!"
list = list_item.list
hash = list.organisation.special_hash_of_org
end
end
But after_initialize feels like a strange usage for this kind of thing. Maybe a helper method would be a better option!

Can I eager load attributes from one-to-one relationship for delegation without instantiating the related model?

Imagine this:
class House < ActiveRecord::Base
belongs_to :ground
delegate :elevation_in_meters, to: :ground
# attributes: stories, roof_type
end
class Ground < ActiveRecord::Base
has_one :house
# attributes: elevation_in_meters, geo_data
end
Then to eager load ground so that house.elevation_in_meters can be called without loading Ground I can do:
houses=House.includes(:ground).first(3)
The problem with this is, that the entire Ground object is actually instantiated with all attributes including the geo_data attribute - which I don't need in this case. The reason why I care is, that the query needs to be VERY performant, and geo_data is a pretty huge text field. I only need to read the delegated attributes, not write to them.
What approach could I take on eager loading the elevation_in_meters attribute from Ground without loading everything from Ground?
I'm on rails 4.1 btw
NOTE: Preferably I would like to have this eager loading behaviour by default for House, so that I do not need to specify it every time.
First off write a scope for the model you want to partially get and select the fields you like. Notice that I used the full name (with table name) and a string for the select. I'm not sure if you could just select(:elevation_in_meters,:geo_data) since I've copied this from our production example, and we use some joins with this scope that wont work without the table name. Just try it yourself.
class Ground < ActiveRecord::Base
has_one :house
attributes: elevation_in_meters, geo_data
scope :reduced, -> {
select('grounds.elevation_in_meters, grounds.geo_data')
}
end
With the scope present you can make a second belongs_to relation (don't be scared that it messes up your first one, since rails relations are basically just methods that are created for you), that calls the scope on your Ground model.
class House < ActiveRecord::Base
belongs_to :ground
belongs_to :ground_reduced, ->(_o) { reduced },
class_name: 'Ground', foreign_key: 'ground_id'
delegate :elevation_in_meters, to: :ground_reduced
# go for an additional delegation if
# you also need this with the full object sometimes
end
In the end you can just call your query like this:
houses = House.includes(:ground_reduced).first(3)
Technically it is not the proper answer to your question, since the Ground object is still instantiated. But the instance will only have the data you wanted and the other fields will be nil, so it should do the trick.
UPDATE:
As I just saw that you want to preferably have this behaviour as default, just add a scope for your House:
scope :reduced, -> { includes(:ground_reduced) }
You could then add this as your default scope, since your original relation will be untouched by this.
I know it's been a while but I just stumbled across this.
If you're only interested in the singular attribute you can also use a joins combined with a select and the attribute will magically be added to your House instance.
res = House.joins(:ground).select('houses.*, grounds.elevation_in_meters').first
res.elevation_in_meters # attribute is available on the object
To always have this attribute present, make it the default_scope for House, like so:
default_scope { joins(:ground).select('houses.*, grounds.elevation_in_meters') }
Depending on the nature of the tables you're joining you may need a distinct also.

rails: create scaffold for models to inherit from superclass?

I'm new to Rails, still getting my feet wet, so please pardon me if this is either trivial or "the wrong way" to do things.
I'd like to create a superclass for some scaffolded models. For example, I'd like to create a scaffold for Men and for Women, but I want them both to inherit from a People superclass; Men and Women would inherit fields like height and weight from the People class.
Where/how do I define this People superclass? How do I define the subclasses Men and Women via scaffolding?
Usually I do something like:
rails g scaffold People type:string name:string birth:date height:integer
class People < ActiveRecord::Base
end
Important use the reserved word 'type'! That's where the table will keep which type the class is. Run the migration.
So, for the the subclasses you can do:
rails g scaffold Men --parent=People
resulting Men:
class Men < People
end
Same for Women:
rails g scaffold Women --parent=People
Resulting
class Women < People
end
No migration will be generated for the subclasses.
I'm not sure but this approach only works for STI.
Hope it, helps!
This is something I've thought about doing with my application. I haven't done it yet, and I wouldn't recommend it if you are new to rails. I would either make separate models entirely, or make one model, and have the attribute gender, which should be either a 0 or a 1, and then make a method that returns the string for the corresponding gender.
EDIT
So I opened up the rails console, and from what I could see, it is possible totally possible, all you need to do is declare the class, and if you want to use different tables, set_table_name
class This < That
set_table_name :this
end
class There < This
set_table_name :there
end
Or you could use one table, but if your trying to stay DRY, I would use two.
If you want to use the scaffold generator, you will have to run the typical rails g scaffold Men for each class you want views for (men and women). The model that this generates inherits from the ActiveRecord::Base class. The inheritance marker is the less than symbol (<).
# THESE WILL BE THE DEFAULT GENERATED MODELS
class Men < ActiveRecord::Base
end
class Women < ActiveRecord::Base
end
You will then manually create the super class User
class User < ActiveRecord::Base
end
and then edit the Men and Women models to inherit from User
# men.rb
class Men < User
end
# women.rb
class Women < User
end
lets say you wanted to subclass with one table, you could would right the migrations for that table, and then add the attr_accessible to the appropriate subclass.
attr_accessible is a rails security feature. It determines which attributes may be set in mass assignment. Anything related to security, site rank, etc. should not be accessible.
Example:
attr_accessible :favorite_food, :interests, :password, :email # THIS IS GOOD
attr_accessible :admin, :has_access_to_missile_launch_codes # THIS IS BAD
because then someone could undermine your security system by passing
params => { :man => { :admin => true }}
The main point is that using these attr_accessible will determine which type of user can set what. Obviously you can DRY this up by putting shared features in the super-class. Hope this helps
You should also read about the super keyword, and the self keyword. If your running an inherited setup you will eventually want to use these.
AFAIK you'd need to tweak the existing scaffolding templates, I don't believe there's a means to specify the controller base class. That said, I think in Rails 3 you can copy the templates into $ROOT/lib/templates/rails/... where ... depends on which you want to change.
That said, what's the real goal in doing this in a scaffold? In general, models will (a) only rarely be subclasses, and (b) even more rarely be the same subclass.
Just edit them by hand.
watch this screencast on single table inheritance.
http://railscasts.com/episodes/394-sti-and-polymorphic-associations
Single table inheritance and where to use it in Rails

How to implement a last_modified_by (person) attribute on two unrelated models - Rails

I have a Record model and in order to edit this model, you must be logged in as an instance of Admin. I would like to have a column called last_modified_by which points to the Admin who last modified the Record. In the database, I was thinking it would be good in the records table to add a column that holds the Admin's id; however, the only way I know how to do that is with an association. These two models are not associated with each other so an association doesn't make a lot of sense. Is there any other way I might be able to accomplish this task without resorting to associations? Any advice would be much appreciated!
Hmm, I think the association is a good tool here. You might want to try to hack it somehow but I think nothing you can conjure up will ever be as good as an association via a foreign_key(also so fast). But perhaps you would like to name your association and do something like:
class Record < ActiveRecord::Base
belongs_to :culprit, :class_name => 'Admin', :foreign_key => 'last_modified_by'
end
or give it some more senseful naming?
You could create an Active Record before_save callback. The callback would save the admin's id into the last_modified_column. This would make sure the admin id is saved/updated each time there is a change to the model.
For example, assuming admin is #admin:
class Record < ActiveRecord::Base
before_save :save_last_modified
def save_last_modified
self.last_modified_column = #admin.id
end
As for getting #admin, you could employ a method similar to this, and set #admin = Admin.current (like User.current in the link) somewhere in the Record model.

Single table inheritance with relations

I'm trying to build a student portal in Rails 3, but I'm having some problem.
The idea is to have a users table that contains all basic data for a given person. See the UML/E-R below for example attributes.
A user can be both an Assistant and a Student at the same time.
Assistant and Student should inherit from User.
The idea was to inherit directly from the User, like this.
class User < ActiveRecord::Base
# ...
def awesome?
[true, false].sample
end
# ...
end
class Student < User
has_one :student
has_many :registered_courses, through: :students
end
Student.new.awesome?
This makes the relations in the student model very strange.
has_many :registered_courses, through: :students
I want to be able to do something like this in the end.
student.full_name
student.pin_code
student.registered_courses
One solution would be to implementing the method by hand, like this
class Student < User
has_one :student
def pin_number
student.pin_number
end
end
But it looks really strange to refer to a student object inside the student model.
Is there a clearer, better way of doing this?
Here is an example UML/E-R. I've tried to keep this example clean by removing non relevant attributes. That is why there are so few attributes in the registered course entity.
STI is not a good choice for this the way that you have articulated it here, since users can be both students and assistants. When you are using STI, you generally add a type column to specify which subclass the record really belongs to. If both Student and Assistant inherit from User, then that really isn't an option, since you'd be forced to create duplicate User records for someone who is both an Assistant and a Student.
I think you'd be better off simply having Student and Assistant rows that belong_to a Student, and then delegating the elements that are contained in User back to the User object.
I feel like Inheritance is a bad move here. If you're going to have STI like this it HAS to be one or the other.
Instead throw all your logic into the User model, all your data is there anyway. Plus since Student & Assistant aren't mutually exclusive there shouldn't be any methods that will override each other.
Why not STI?
STI is mainly meant for objects that contain the same data, but does different things with them.
For example, I have a specification that contains multiple processes(ex. build and test). So I have a order that contains processes.
process_1:
order_id: 1
specification: foo
type: build
process_2:
order_id: 1
specification: foo
type: test
In this example the only thing that changes in the data is the type, but because the type changes I know what process to perform from the specification.

Resources