Offline Jobs

Last exercise we focused on removing code from the controller and moved it out into callbacks and/or observers. This helped with the manageability of the code but it failed to alleviate the problem of forcing the user to wait until the request was done sending out posts to all their services that they selected.

Imagine what the experience would be like if one of the services went offline or became delayed. The user would wait endlessly for the page to finish loading even though their work was done.

Imagine that the site became popular and all of sudden you had a large number of users wanting to post to all their services. Your servers would soon be crippled by your own success.

Exercise

This week we are going to look at a simple Rails application that allows a user to create an account and then tie different social network accounts to their account. Users are able to create posts and have them echo'd to all the services that they have set up.

The goal of this week's exercise is to move the code that echoes their post to their specified services into an offline job.

Goal

Understand the code.

Move the code that publishes content to Twitter and Facebook out of controller and move it into it's own job.

Reading

Ruby Meta-programming - Friday

Ryan Bates - Setting up Resque

The Little Redis Book

Read about another alternative to Resque called Delayed Job and how it is implemented. Ryan Bates - Delayed Job

Peep Code - Aaron Patterson - Play by Play

Further Exercise

• After a user authorizes a service, enqueue a job that will retrieve content from that service.

  From Twitter or Facebook you may retrieve their timeline, friends, likes, or hobbies.

• Store that content within the database

  Modeling the content here, you may want to extract only certain pieces of data or you may want to store all the data. Ensure that you do not store duplicate data for the user.

• Use that content to personalize a service

  From their list of hobbies or likes, randomly provide them a suggestion for the post.

• Visualize how you might continually import data on a schedule

  How could you continually enqueue a job to import data at a specified interval?

Exploration

Resque is a wrapper for Redis. It offers a layer on top of a Redis client to perform tasks of enqueuing and dequeuing tasks.

• Launch IRB and connect to your Redis Server

require "redis"

redis = Redis.new

redis.set "foo", "bar" # => "OK"

redis.get "foo" # => "bar"

redis.sadd "users", "albert" # => true

redis.sadd "users", "bernard" # => true

redis.sadd "users", "charles" # => true

• Refer to the documentation and learn about the various different data types and commands.

  http://redis.io/commands