ripple is a rich Ruby toolkit for Riak, Basho’s distributed database. It consists of three gems:

  • riak-client (Riak namespace) contains a basic wrapper around typical operations, including bucket manipulation, object CRUD, link-walking, and map-reduce.
  • ripple (Ripple namespace) contains an ActiveModel-compatible modeling layer that is inspired by ActiveRecord, DataMapper, and MongoMapper.
  • riak-sessions contains session stores for Rack and Rails 3 applications.


riak-client requires i18n and either json or yajl-ruby. For higher performance on HTTP requests, install the 'curb’ or 'excon’ gems. The cache store implementation requires ActiveSupport 3 or later.

ripple requires Ruby 1.8.7 or later and versions 3 or above of ActiveModel and ActiveSupport (and their dependencies, including i18n).

riak-sessions requires Rack (any version > 1.0), and Rails 3.0 if you want the Rails-specific session store.

Development dependencies are handled with bundler. Install bundler (gem install bundler) and run this command in each sub-project to get started:

<notextile><pre>$ bundle install</pre></notextile>

Run the RSpec suite using bundle exec:

<notextile><pre>$ bundle exec rake spec</pre></notextile>

Basic Example

<notextile><pre>require 'riak’

  1. Create a client interface

client =

  1. Create a client interface that uses Excon

client = => :Excon)

  1. Retrieve a bucket

bucket = client.bucket(“doc”) # a Riak::Bucket

  1. Get an object from the bucket

object = bucket.get(“index.html”) # a Riak::RObject

  1. Change the object’s data and save = “<html><body>Hello, world!</body></html>”

  1. Reload an object you already have

object.reload # Works if you have the key and vclock, using conditional GET
object.reload :force => true # Reloads whether you have the vclock or not

  1. Access more like a hash, client[bucket][key]

client['doc’]['index.html’] # the Riak::RObject

  1. Create a new object

new_one =, “application.js”)
new_one.content_type = “application/javascript” # You must set the content type. = “alert('Hello, World!’)”</pre></notextile>

Map-Reduce Example


  1. Assuming you’ve already instantiated a client, get the album titles for The Beatles

results = add(“artists”,“Beatles”). link(:bucket => “albums”). map(“function(v){ return [JSON.parse(]; }”, :keep => true).run

p results # => [“Please Please Me”, “With The Beatles”, “A Hard Day’s Night”, # “Beatles For Sale”, “Help!”, “Rubber Soul”, # “Revolver”, “Sgt. Pepper’s Lonely Hearts Club Band”, “Magical Mystery Tour”, # “The Beatles”, “Yellow Submarine”, “Abbey Road”, “Let It Be”]</pre></notextile>

Riak Search Examples

For more information about Riak Search, see the Basho wiki.

require 'riak/search’ # optional riak_search additions

  1. Create a client, specifying the Solr-compatible endpoint

client = :solr => “/solr”

  1. Search the default index for documents

result =“title:Yesterday”) # Returns a vivified JSON object # containing 'responseHeaders’ and 'response’ keys
result['response’]['numFound’] # total number of results
result['response’]['start’] # offset into the total result set
result['response’]['docs’] # the list of indexed documents

  1. Search the 'users’ index for documents“users”, “name:Sean”)

  1. Add a document to an index

client.index(“users”, {:id => “”, :name => “Sean Cribbs”}) # adds to the 'users’ index

client.index({:id => “index.html”, :content => “Hello, world!”}) # adds to the default index

client.index({:id => 1, :name => “one”}, {:id => 2, :name => “two”}) # adds multiple docs

  1. Remove document(s) from an index

client.remove({:id => 1}) # removes the document with ID 1
client.remove({:query => “archived”}) # removes all documents matching query
client.remove({:id => 1}, {:id => 5}) # removes multiple docs

client.remove(“users”, {:id => “”}) # removes from the 'users’ index

  1. Seed MapReduce with search results search(“users”,“email:basho”). map(“Riak.mapValuesJson”, :keep => true). run

  1. Detect whether a bucket has auto-indexing


  1. Enable auto-indexing on a bucket


  1. Disable auto-indexing on a bucket


Document Model Examples

require 'ripple’

  1. Documents are stored as JSON objects in Riak but have rich
  2. semantics, including validations and associations.

class Email include Ripple::Document property :from, String, :presence => true property :to, String, :presence => true property :sent, Time, :default => proc { } property :body, String

email = Email.find(“37458abc752f8413e”) # GET /riak/emails/37458abc752f8413e
email.from = “” # PUT /riak/emails/37458abc752f8413e

reply =
reply.from = “” = “”
reply.body = “Riak is a good fit for scalable Ruby apps.” # POST /riak/emails (Riak-assigned key)

  1. Documents can contain embedded documents, and link to other standalone documents
  2. via associations using the many and one class methods.

class Person include Ripple::Document property :name, String many :addresses many :friends, :class_name => “Person” one :account

  1. Account and Address are embeddable documents

class Account include Ripple::EmbeddedDocument property :paid_until, Time embedded_in :person # Adds “person” method to get parent document

class Address include Ripple::EmbeddedDocument property :street, String property :city, String property :state, String property :zip, String

person = Person.find(“adamhunter”)
person.friends << Person.find(“seancribbs”) # Links to people/seancribbs with tag “friend”
person.addresses << => “100 Main Street”) # Adds an embedded address
person.account.paid_until = 3.months.from_now

Configuration Example

When using Ripple with Rails 3, config/ripple.yml should contain your Riak connection information:

development: port: 8098 host: localhost

production: port: 8098 host:

require 'ripple/railtie' from the top of your config/application.rb file to turn this on.

How to Contribute

  • Fork the project on Github. If you have already forked, use git pull --rebase to reapply your changes on top of the mainline. Example:
<notextile><pre>$ git checkout master $ git pull —rebase seancribbs master</pre></notextile>
  • Create a topic branch. If you’ve already created a topic branch, rebase it on top of changes from the mainline “master” branch. Examples:
    • New branch:
<pre>$ git checkout -b topic</pre>
  • Existing branch:
<pre>$ git rebase master</pre>
  • Write an RSpec example or set of examples that demonstrate the necessity and validity of your changes. Patches without specs will most often be ignored. Just do it, you’ll thank me later. Documentation patches need no specs, of course.
  • Make your feature addition or bug fix. Make your specs and stories pass (green).
  • Run the suite using multiruby or rvm to ensure cross-version compatibility.
  • Cleanup any trailing whitespace in your code (try whitespace-mode in Emacs, or “Remove Trailing Spaces in Document” in the “Text” bundle in Textmate).
  • Commit, do not mess with Rakefile or VERSION. If related to an existing issue in the tracker, include “Closes #X” in the commit message (where X is the issue number).
  • Send me a pull request.

License & Copyright

Copyright &copy;2010 Sean Cribbs, Sonian Inc., and Basho Technologies, Inc.

Licensed under the Apache License, Version 2.0 (the “License”); you may not use this file except in compliance with the License. You may obtain a copy of the License at

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an “AS IS” BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Auxillary Licenses

The included photo (spec/fixtures/cat.jpg) is Copyright &copy;2009 Sean Cribbs, and is licensed under the Creative Commons Attribution Non-Commercial 3.0 license.

The “Poor Man’s Fibers” implementation (lib/riak/util/fiber1.8.rb) is Copyright &copy;2008 Aman Gupta.