Lets assume I have the following model:
class Computer < ActiveRecord::Base
end
and I retrieve all computers in my controllers like this:
#computers = Computer.all
Now I add a feature to deactivate certain computers and filter the deactivated computer like this:
class Computer < ActiveRecord::Base
scope :not_deactivated, -> { where('deactivated IS NULL OR deactivated = ?', false) }
end
and in my controllers:
#computers = Computer.all.not_deactivated
The issue with this method is that I have to add not_deactivated in all my controllers.
Is it possible to do this filter in the model so I don't have to touch the controllers?
Easy and common thing to do in controller:
before_action :not_deactivated # , only [:index...]
private
def not_deactivated
#computers = Computer.where(your code)
end
Since the controller handles the view you must initiate the object somehow anyway. By filtering like that you can achieve what you are trying to do now with the model filter.
you can do it by declaring default_scope to run this every time you make a call to your Computer model
class Computer < ActiveRecord::Base
default_scope { where('deactivated IS NULL OR deactivated = ?', false) }
end
So, when you run Computer.all the result you get is Computer.where('deactivated IS NULL OR deactivated = ?', false)
I'm used to MySQL but trying to use Ruby on Rails right now. In MySQL, I would have two tables, with one containing a reference to another ("posts" referring to "topic"). A MySQL query doing what I want would be similar to "SELECT * FROM Posts WHERE posts.topic="topic" ("topic" here is a variable).
However, trying to work with the Ruby model stuff has me confused. The variables being passed between the controller and view are null because they are empty tables.
In my controller:
def topic
#topic = Topic.where(params[:topic])
#posts = Post.where(topic: #topic.object_id)
end
I don't know how to select the posts which have the topic defined by the "topic" variable.
In the view:
<% #posts.each do |post| %>
<p><%= post.title %></p>
<% end %>
The migration files:
class CreatePosts < ActiveRecord::Migration
def change
create_table :posts do |t|
t.string :title
t.string :text
t.references :topic
t.timestamps
end
end
end
class CreateTopics < ActiveRecord::Migration
def change
create_table :topics do |t|
t.string :topic
t.timestamps
end
end
end
Given that Post and Topic are related, according to your migrations at least, in the models you should be stating"
class Topic
has_many :posts
and
class Post
belongs_to :topic
Given that you then have an instance of Topic, #topic, you can retrieve all the related records with:
#posts = #topic.posts
I think those methods you put in your controller are fine where they are, but keep in mind that the Rails way is "fat models, skinny controllers." If you put that logic in the model as a method, it's much easier to read in the controller. Also, you should look into scopes, as they'll help you with queries like this down the line too.
In any case, you should stick the following in your Topic model:
scope :by_name, ->(name) { where(topic: name) }
That's essentially the same as doing the following:
def self.by_name(name)
where(topic: name)
end
On your posts model, you'd be able to do the following:
scope :by_topic, ->(topic) { where(topic_id: topic) }
The other problem with what you've stuck in your controller is that when you use scopes, or a "where", it returns an array that contains all of the different records that match your query terms. So, when you call #topic = Topic.where(params[:topic]), you're getting back an array of objects. Therefore, when you do a #topic.id, you're trying to get back the id of an array instead of one object.
Based off of what I showed you before, it makes much more sense for you to do this:
def topic
#topic = Topic.by_name(params[:topic]).first #this returns the first record
#post = Post.by_topic(#topic.id)
end
That will return an array of posts that match the first topic name that you query for.
Alright, first a primer on how database design and how Rails (really, ActiveRecord) works. Basically, you should be connecting posts.topic_id = topic.id, not posts.topic = topic.topic.
Your migration is correct as is, create_table automatically includes an :id PRIMARY KEY column. That said you should know that these are all equivalent:
t.references :topic
t.belongs_to :topic
t.integer :topic_id
In your view, instead of embedding topic.topic and passing that to the controller when the form is submitted, embed topic.id (the documentation for the select helper has a good example of this) and in your controller:
#topic = Topic.find params[:id]
#posts = #topic.posts
I am using rails version 4.2 and ruby version 2.2.0. I am trying to save a record to lollypops table. No exceptions indicating reasons.
TASK: As soon as a member is created and saved, I want to populate the lollypops table by calling the create_lollypop(#member.id) in members controller's create method like this:
# POST /members
# POST /members.json
def create
#member = Member.create(members_params)
return unless request.post?
#member.save!
self.current_user = #member
c = Country.find(#member.country_id)
#member.update_attributes(
:country_code=>c.code)
create_lollypop(#member.id) #From here I want to create lollypop
MemberMailer.signup_notification(#member).deliver_now
redirect_to(:controller => '/admin/members', :action => 'show',
:id=> #member.id)
flash[:notice] = "Thanks for signing up! Check your email now to
confirm that your email is correct!"
rescue ActiveRecord::RecordInvalid
load_data
render :action => 'new'
end
def create_lollypop(member_id)
#member = Member.find(member_id)
Lollypop.create(
:member_id=>#member.id,
:product_name=>'lollypop',
:product_price=>100,
:email=>#member.email,
:house_flat => #member.house_flat,
:street=>#member.street,
:city_town=>#member.city_town,
:country =>#member.country,
:postcode_index=>#member.postcode_index,
:name=>#member.name)
end
The 'member' is created but the 'lollypops' table is not populated. The associations are:
MEMBER model:
has_one :lollypop, :dependent=>:destroy
LOLLYPOP model
belongs_to :member
If I use generic SQL command then the lollypops table gets populated but I do not want to do that:
def self.create_lollypop(member_id)
member = Member.find(member_id)
ActiveRecord::Base.connection.execute("insert into lollypops (member_id,product_name,product_price,email,house_flat,street,city_town,country,postcode_index,name)
values(#{member.id},'lollypop',#{100},'#{member.email}','#{member.house_flat}','#{member.street}','#{member.city_town}','#{member.country_code}','#{member.postcode_index}','#{member.name}')")
end
Any advice would be welcomed. Thank you.
In your create_lollypop(), You are not defining #member.
def create_lollypop(member_id)
#member = Member.find member_id
Lollypop.create!(
:member_id=>#member.id,
:product_name=>'lollypop',
:product_price=>100,
:email=>#member.email,
:house_flat => #member.house_flat,
:street=>#member.street,
:city_town=>#member.city_town,
:country =>#member.country,
:postcode_index=>#member.postcode_index,
:name=>#member.name
)
end
Also use create! so in case any validation failed then it will raise exception. So it will help you sort out issue.
For the moment try to create lollypop using the association method create_lollypop directly in your controller. use this code in you create controller method, note that create_lollypop method will fill (member_id field automatically):
#member = Member.create(members_params)
return unless request.post?
#member.save!
self.current_user = #member
c = Country.find(#member.country_id)
#member.update_attributes(
:country_code=>c.code)
#From here I want to create lollypop
#member.create_lollypop(
:product_name=>'lollypop',
:product_price=>100,
:email=>#member.email,
:house_flat => #member.house_flat,
:street=>#member.street,
:city_town=>#member.city_town,
:country =>#member.country,
:postcode_index=>#member.postcode_index,
:name=>#member.name
)
MemberMailer.signup_notification(#member).deliver_now
redirect_to(:controller => '/admin/members', :action => 'show',
:id=> #member.id)
flash[:notice] = "Thanks for signing up! Check your email now to
confirm that your email is correct!"
rescue ActiveRecord::RecordInvalid
load_data
render :action => 'new'
This is not exactly an answer, more like tips and notes, it's a little long and I hope you don't mind.
return unless request.post?
This is more of a php thing not a rails thing, in rails already the routing is checking this, so you don't need to do this check inside the controller, if it isn't a post it will be routed elsewhere.
#member = Member.create(members_params)
return unless request.post?
#member.save!
Saving after creating is meaningless, because create already saves the data, if you are doing it for the sake of the bang save!, then you could use the create with bang create!, not to mention that you do the redirection check after the member's create, so if this did work, it would leave you with stray members.
c = Country.find(#member.country_id)
#member.update_attributes(:country_code=>c.code)
If you have your assocciations correctly, you don't need to save the code like this, because the member knows that this country_id belongs to a country.
So add this to the member model
class Member < ActiveRecord::Base
has_one :lollypop, dependent: :destroy
belongs_to :country
end
This way you could always call #member.country to return the country object, then the code could come from there, like #member.country.code, or you could just write a method to shorten that up
def country_code
country.code
end
this way will get the code through an extra query, but it has an advantage, if you for any reason change a country's code, you don't need to loop on all members who have that country and update their codes too, you could also shorten this up even more using #delegate
#member.save!
#member.update_attributes(:country_code=>c.code)
Here you are updating the attributes of member after saving the member, which is kinda a waste, because you are doing 2 queries for what could be done with 1 query, programmatically it is correct and it will work, but it's bad for scaling, when more users start using your app the database will be more busy and the responses will be slower.
Instead i would recommend to postpone the creation of member till you have all the data you want
#member = Member.new(members_params) # this won't save to the database yet
#memeber.code = Country.find(#member.country_id).code
#member.save
This will only do 1 query at the end when all data is ready to be saved.
redirect_to(:controller => '/admin/members', :action => 'show', :id=> #member.id)
This is ok, but you probably have a better shorter path name in your routes, something like members_admin_path, check your routes name by doing a bin/rake routes in your terminal.
redirect_to members_admin_path(id: #member)
redirect_to ...
flash[:notice] = "message"
I'm not sure this will work, because the redirection needs to be returned, but when you added the flash after it, either the redirection will happen without the flash, or the flash will be set and returned as it's the last statement, but the redirection won't happen, not sure which will happen, to fix it you can simply swap the two statements, create the flash first and then redirect, or use the more convenient way of setting the flash while redirecting, cause that's supported
redirect_to ....., notice: 'my message'
rescue ActiveRecord::RecordInvalid
load_data
render :action => 'new'
This will do the job, but it isn't conventional, people tend to use the soft save and then do an if condition on the return value, either true or false, here's a short layout
# prepare #member's data
if #member.save
# set flash and redirect
else
load_data
render :new
end
The lollypop creation
Now there's a few things about this, first you have the method in the controller, which is bad cause it shouldn't be the controller's concern, the second method the self.create_lollypop is better cause it's created on the model level, but it's a class method, then the better way is creating it as a member method, this way the member who creates the lollypop already knows the data because it's his own self, notice i don't need to call #member because i am already inside member, so simple calls like id, email will return the member's data
# inside member.rb
def create_lollypop
Lollypop.create!(
member_id: id,
product_name: 'lollypop',
product_price: 100,
email: email,
house_flat: house_flat,
street: street,
city_town: city_town,
country: country,
postcode_index: postcode_index,
name: name
)
end
if you want you can also add this as an after create callback
after_create :create_lollypop
ps: This method name will probably conflict with the ActiveRecords create_lollypop method, so maybe you should pick a different name for this method.
As Mohammad had suggested to me, I changed Lollypop.create to Lollypop.create! and
while running my code, one validation error popped up. After correcting it and
altering my code to:
Lollypop.create!(
:member_id=> #member.id,
:product_name=>'lollypop',
:product_price=>100,
:email=>#member.email,
:house_flat => #member.house_flat,
:street=>#member.street,
:city_town=>#member.city_town,
:country =>#member.country_code,
:postcode_index=>#member.postcode_index,
:name=>#member.name
)
The 'lollypops' table got populated.
I have searched through all the active model serializer (v 0.9.0) documentation and SO questions I can find, but can't figure this out.
I have objects which can be marked as "published" or "draft". When they aren't published, only the user who created the object should be able to see it. I can obviously set permissions for "show" in my controller, but I also want to remove these objects from the json my "index" action returns unless it is the correct user. Is there a way to remove this object from the json returned by the serializer completely?
In my activemodel serializer, I am able to user filter(keys) and overloaded attributes to remove the data, as shown using my code below, but I can't just delete the entire object (I'm left having to return an empty {} in my json, trying to return nil breaks the serializer).
I'm probably missing something simple. Any help would be much appreciated!
class CompleteExampleSerializer < ExampleSerializer
attributes :id, :title
has_many :children
def attributes
data = super
(object.published? || object.user == scope || scope.admin?) ? data : {}
end
def filter(keys)
keys = super
(object.published? || object.user == scope || scope.admin?) ? keys : {}
end
end
That looks correct, try returning an array instead of a hash when you dont want any keys. Also, I don't think calling super is necessary b/c the filter takes in the keys.
Also, I don't think defining an attributes method is necessary.
I have chapters that can either be published or unpublished. They're owned by a story so I ended doing something like below.
has_many :unpublished_chapters, -> { where published: false }, :class_name => "Chapter", dependent: :destroy
has_many :published_chapters, -> { where published: true }, :class_name => "Chapter", dependent: :destroy
Inside of my serializer, I choose to include unpublished_chapters only if the current_user is the owner of those chapters. In ams 0.8.0 the syntax is like so.
def include_associations!
include! :published_chapters if ::Authorization::Story.include_published_chapters?(current_user,object,#options)
include! :unpublished_chapters if ::Authorization::Story.include_unpublished_chapters?(current_user,object,#options)
end
In my case, it's not so bad to differentiate the two and it saves me the trouble of dealing with it on the client. Our situations are similar but say you want to get all of the chapters by visiting the chapters index route. This doesn't make much sense in my app but you could go to that controller and render a query on that table.
I have built a user and friend relationship model but the problem is that with those associations I can friend myself. I have successfully suppressed it in my views and controller, but logically it should be suppressed in the model because I could still create the friendship from the console which I want to avoid.
User model
has_many :user_friendships
has_many :friends, through: :user_friendships,
conditions: { user_friendships: { state: 'accepted' } }
User_friendship model
belongs_to :user
belongs_to :friend, class_name: 'User', foreign_key: 'friend_id'
Everything else is working perfectly like adding, blocking, deleting, requesting a friend the only problem with my model is that I can also friend myself which I want to avoid.
Add a validation to UserFriendship:
validate :cannot_friend_self
def cannot_friend_self
errors.add(:friend_id, "cannot friend self") unless user_id != friend_id
end
This issue is a little problematic because we want to remain RESTful, separate the different tasks (MVC,) and take into account of weird race conditions (Thread Safety.)
Try using validations#exclusions (http://guides.rubyonrails.org/active_record_validations_callbacks.html#exclusion)
class ApplicationController < ActionController::Base
...
before_filter do |c|
User.current_user = User.find(c.session[:user]) unless c.session[:user].nil?
end
...
end
class User < ActiveRecord::Base
...
cattr_accessor :current_user
...
end
class Friends < ActiveRecord::Base
...
validates :friend_id, :exclusion => { :in => %w(User.current_user.id),
:message => "I don't think you really want to friend yourself" }
...
end
If you want to be safe, please refer to (http://nhw.pl/wp/2011/11/30/passing-current-user-id-to-rails-models)
Disclaimer:
I wrote this possible solution without testing it (aka pulled it out of the thin air with little reference)
I have not thread with Ruby on Rails.
You probably want to throw in a validation
Such as
validate :cannot_friend_self
def cannot_friend_self
current_user.id != friend.id
end
This code may not be exactly what you want, but should point you in the right direction.
Full guide here http://guides.rubyonrails.org/active_record_validations_callbacks.html#custom-methods