Monday 29 December 2014

Use logger.debug rather than puts!

It is often tempting to use puts in the ruby code to debug the informations.

I like logging instead because:
-allows the filtering capability
-allows formatting abilities
-with a little tweak it takes the same amount of effort than puts
-it is an already invented and well tested system

It is much better in short. So whenever is a logger system available I use that.

logger.debug 'doing this and that'

It seem to me too much to type at first, therefore I created a sublime logger snippet which accelerated typing. I created another one for those cases where I need to implicitly reference the Rails.logger.

;)

Sunday 28 December 2014

API Development in Ruby On Rails, Nested Entities

There are times when there is one-to-many association amongst the entities. And the mobile developer needs to show all of them in one screen. In these cases it is very easy to fell into the bad habit to execute n+1 queries from mobile side toward the server or from the server to the sql-server.

Here are the entities:

class DetailedGameEntity < Grape::Entity
  ...
  expose :reviews, using: ReviewEntity do |game, options|
    game.reviews
  end
end

class ReviewPresenter < Grape::Entity
  expose :user_id, as: :owner_id
  expose :comment
  expose :rating

  expose :image_urls do |review, options|
    review.images.map{|image| image.image}
  end
end

They are designed to return everything the mobile screen needs. The games contains their reviews. And the reviews contain their images. So in one query the whole mobile screen can be populated.

Let's see how a programmer in a hurry develops all this API:

games = Game.actual.limit(500)
present games, with: DetailedGameEntity

By developing the API this way the call will result in these queries:
...
  Review Load (0.5ms)  SELECT "reviews".* FROM "reviews"  WHERE "reviews"."reviewable_id" = $1 AND "reviews"."reviewable_type" = $2  [["reviewable_id", 2359], ["reviewable_type", "Game"]]
  Review Load (0.4ms)  SELECT "reviews".* FROM "reviews"  WHERE "reviews"."reviewable_id" = $1 AND "reviews"."reviewable_type" = $2  [["reviewable_id", 2358], ["reviewable_type", "Game"]]
...

As you can see for each returned game game another select is executed to fetch the belonging reviews. The same is done to fetch the belonging user and to fetch the images belonging to the review. This is a very ineffective way because as the review count grows more and more queries need to be executed. This problem is called n+1 query problem.

I do this in my code in these cases:

games = Game.actual.limit(500)
games = games.includes(reviews: [:images, :user] )
present games, with: DetailedGameEntity

The above code will do this sql query:

 Review Load (1.0ms)  SELECT "reviews".* FROM "reviews"  WHERE "reviews"."reviewable_type" = 'Schedule' AND "reviews"."reviewable_id" IN (2195, 2198, 2197, 9567, 9572, 9573, 9574, 9575, 9576, 9571, 9570, 9569, 9568, 2196, 2204, 2210, 2211, ...

So, the underlying Rails code will execute only one query per entity.

It is nice, isn't it?

Please use it ;)

Saturday 27 December 2014

API Development in Ruby On Rails, Entities

In my last blog post I presented in general the features I like the most about the grape gem. Now I will present the grape entities which are related to the presentation of the returned data. I like these entities a lot because they are an OO way to present data and they help a lot to keep my presentation layer DRY.

The last statement of every grape call is returned to the caller as JSON.
Despite this simple efficiency I like to have more control over the returned data and I use the grape entities to format the returned data wherever I can. I can also rspec them and I am doing it extensively because I am a Test Driven Guy ;)

Let me present some examples of entities their usage and how do I spec them:

Here is the spec for the entity:

describe GameEntity do
  describe 'fields' do
    subject(:subject) { GameEntity }
    it { is_expected.to represent(:id) }
    it { is_expected.to represent(:user_id).as(:owner_id) }
...


Here is the entity itself:

class GameEntity < Grape::Entity
  expose :id
  expose :user_id, as: :owner_id
...

The part of the grape API:

  namespace :games do
    desc "Retrieve all the games"
    params do
      optional :include_reviews, type: Boolean, default: true, desc: 'Accepts: true/false.'
    end
    get do
      games = Game.actual.limit(500)
      present games, with: GameEntity

This was simple so far. But I can use a more sophisticated GameEntity in case I want to present the included reviews. In all other cases I will use my old simple GameEntity to present the Game.

Let's see that case too:

I modify the grape API to this:

    ...
    if params[:include_reviews]
      games = games.includes(:reviews)
      present games, with: DetailedGameEntity
    else
      present games, with: GameEntity
    end

I also spec the new entity:

describe DetailedGameEntity do
  describe 'special fields' do
      subject(:detailed_game) { DetailedGameEntity(game).as_json }
      specify { expect(detailed_game[:reviews]).to be_an(Array) }
  end

Of course the above was a very simple example.

In real world scenarios we can find ourselves that we are reusing the entities like this:

class DetailedGameEntity < Grape::Entity
  expose :game, using: GameEntity do |game, options|
    game
  end

  expose :venue, using: VenueEntity do |game, options|
    game.the_venue
  end

  expose :visitor_team, using: TeamEntity do |game, options|
    game.visitor_team
  end

  expose :host_team, using: TeamEntity do |game, options|
    game.host_team
  end
end

As you can see we are keeping our entity codebase DRY by calling the TeamEntity twice in the DetailedGameEntity. This is a typical use of delegation of the Entities.

Let me show a more complex example of our entities:

The usage of the entities:

  desc "Retrieve all the reviews belonging to a team"
    get do
      ...
      present reviews, :with => ReviewPresenter, user: current_user
    end

The entity itself:
  class ReviewPresenter < Grape::Entity
    expose :current_user_likes do |review, options|
      review.likes.pluck(:user_id).include? options[:user].id if options[:user].present?
    end
  end

And finally the spec for the entity:
describe ReviewPresenter do
  describe 'special fields' do
    context 'without passing a user' do
      subject(:presented_review) { ReviewPresenter.new(image.review).as_json }
      specify { expect(presented_review[:current_user_likes]).to be_nil }
    end

    subject(:presented_review) { ReviewPresenter.new(image.review, user: current_user).as_json }
    ...
      context 'user is the current_user' do
        specify { expect(presented_review[:current_user_likes]).to eq(true) }
      end
    end
    ...
  end
end

What I was doing here is to populate the field 'current_user_likes'
 to true or false depending if the current_user likes or not the presented review.

As you can see the presentation layer of any grape API could became more DRY and versatile by using the grape entities.

Enjoy using them ;)

Friday 26 December 2014

API Development in Ruby On Rails

I met a technology and a set of gems which were designed to develop any API in Ruby On Rails about half years ago.

The set of gems were:


At first glance it seemed that it adds complexity to the Rails application compared to standard Rails JSON returns. After I played a bit with it I suddenly realized that its API formulating DSL is much more superior. So I started to like it :)

Let me list the elements of DSL I like most:

  • ability to mount API endpoints. This can be very handy if I want to unmount certain unused parts of the API.

      mount Caesars::Ping
      mount Caesars::PingProtected

  • ability to describe the API endpoint. This description will appear in the Swagger documentation. It is a help for the API user.
  • ability to specify the parameters (type, mandatory or required).
      params do
        optional :user_ids, desc: "JSON array of the user ids to be deleted."
        requires :group_id, type: Integer, desc: "The id of the group"
      end

By specifying the type our parameters will be converted automagically to the specified type. It is easier this way for us developers to write our core logic.
  • a complex DSL to organize the API hierarchy and use route params. I consider this the most powerful feature. I can easily organize the API to look more like Facebook graph API or Philips Hue API than standard Rails restfull API.
      namespace :teams do
        route_param :id do
          desc 'Retrieve all games belonging to teams'
          get 'scores' do
            #DO logic here
          end
        end
      end

The grape features are many more including versioning. Take a look for a complete description at grape site ;)

Thursday 25 December 2014

Hi,

I have news for you. My daughter had been born in august :) With Her time passed very fast.
I even missed the new macbook release date which was in July. But I don't feel disappointed because according to the rumors the newer MacBook will have a much better CPU. So I will wait for the next generation of MacBook pro :)

Here is a picture about us: