README.md

Experiment Ruby SDK

Amplitude Ruby Server SDK for Experiment.

⚠️ Local evaluation process fork safety: When using a LocalEvaluationClient and forking the process, you must call #start after forking to update the flag configuration state on the new process. For example, in Puma, this means calling #start in the on_worker_boot hook.

Installation

Into Gemfile from rubygems.org:

gem 'amplitude-experiment'

Into environment gems from rubygems.org:

gem install 'amplitude-experiment'

To install beta versions:

gem install amplitude-experiment --pre

Remote Evaluation Quick Start

require 'amplitude-experiment'

# (1) Get your deployment's API key
apiKey = 'YOUR-API-KEY'

# (2) Initialize the experiment client
experiment = AmplitudeExperiment.initialize_remote(api_key)

# (3) Fetch variants for a user
user = AmplitudeExperiment::User.new(user_id: '[email protected]', device_id: 'abcezas123', user_properties: {'premium' => true})

# (4) Lookup a flag's variant
# 
# To fetch asynchronous
experiment.fetch_async(user) do |_, variants|
  variant = variants['YOUR-FLAG-KEY']
  unless variant.nil?
    if variant.value == 'on'
      # Flag is on
    else
      # Flag is off
    end
  end
end

# To fetch synchronous
variants = experiment.fetch(user)
variant = variants['YOUR-FLAG-KEY']
unless variant.nil?
  if variant.value == 'on'
    # Flag is on
  else
    # Flag is off
  end
end

Local Evaluation Quick Start

require 'amplitude-experiment'

# (1) Get your deployment's API key
apiKey = 'YOUR-API-KEY'

# (2) Initialize the experiment client
experiment = AmplitudeExperiment.initialize_local(api_key)

# (3) Start the local evaluation client
experiment.start

# (4) Evaluate a user
user = AmplitudeExperiment::User.new(user_id: '[email protected]', device_id: 'abcezas123', user_properties: {'premium' => true})
variants = experiment.evaluate(user)
variant = variants['YOUR-FLAG-KEY']
unless variant.nil?
  if variant.value == 'on'
    # Flag is on
  else
    # Flag is off
  end
end

More Information

Please visit our 💯Developer Center for more instructions on using our the SDK.

See our Experiment Ruby SDK Docs for a list and description of all available SDK methods.

See our Local Evaluation Repo for more details about local evaluation binary logic.

Need Help?

If you have any problems or issues over our SDK, feel free to create a github issue or submit a request on Amplitude Help.