/Cabiri

An easy and intuitive Ruby job queue for working with parallel processes.

Primary LanguageRubyMIT LicenseMIT

Introduction ↑

Cabiri is an easy and intuitive Ruby job queue for working with parallel processes. This gem requires Ruby 1.9.3 or greater.

Getting Started ↑

Install the gem at the command prompt:

gem install cabiri

If it complains about Ruby version 1.9.3 not being installed, you can do the following:

rvm get head
rvm reload
rvm install 1.9.3

Examples ↑

Cabiri is rather straightforward.

# myapp.rb
require 'cabiri'

queue = Cabiri::JobQueue.new

queue.add('my first job') do
  # define your first job here
  result = 0
    100000000.times do |t|
    result += t
  end
  result
end

queue.add('my second job') do
  # define your second job here
  result = 0
    200000000.times do |t|
    result += t
  end
  result
end

queue.add('my third job') do
  # define your third job here
  result = 0
    300000000.times do |t|
    result += t
  end
  result
end

# start the queue and specify how many processes you want the queue to run in parallel.
# This command blocks until all jobs in the queue have been finished.
queue.start(2)

# get results from all three jobs
result_1st_job = queue.finished_jobs['my first job'].result
result_2nd_job = queue.finished_jobs['my second job'].result
result_3rd_job = queue.finished_jobs['my third job'].result

Alternatively, if having a blocking call is something you’d like to avoid, you can also do as shown below.

# myapp.rb
require 'cabiri'

queue = Cabiri::JobQueue.new

queue.add('my first job') do
  # define your first job here
  result = 0
    100000000.times do |t|
    result += t
  end
  result
end

queue.add('my second job') do
  # define your second job here
  result = 0
    200000000.times do |t|
    result += t
  end
  result
end

queue.add('my third job') do
  # define your third job here
  result = 0
    300000000.times do |t|
    result += t
  end
  result
end

# start the queue and specify how many processes you want the queue to run in parallel.
# This command blocks until all jobs in the queue have been finished.
# You can start the queue in a different thread to get around this.
Thread.new do
  queue.start(2)
end

while !queue.finished?
  # do something useful
  sleep 1
end

# get results from all three jobs
result_1st_job = queue.finished_jobs['my first job'].result
result_2nd_job = queue.finished_jobs['my second job'].result
result_3rd_job = queue.finished_jobs['my third job'].result