Riak Ruby Client (riak-client) Build Status

riak-client is a rich Ruby client/toolkit for Riak, Basho's distributed database that contains a basic wrapper around typical operations, including bucket manipulation, object CRUD, link-walking, and map-reduce.

Dependencies

riak-client requires i18n, builder, beefcake, and multi_json. For higher performance on HTTP requests, install the 'excon' gem. The cache store implementation requires ActiveSupport 3 or later.

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

$ bundle install

Run the RSpec suite using bundle exec:

$ bundle exec rake

Basic Example

require 'riak'

# Create a client interface
client = Riak::Client.new

# Create a client interface that uses Excon
client = Riak::Client.new(:http_backend => :Excon)

# Create a client that uses Protocol Buffers
client = Riak::Client.new(:protocol => "pbc")

# Automatically balance between multiple nodes
client = Riak::Client.new(:nodes => [
  {:host => '10.0.0.1'},
  {:host => '10.0.0.2', :pb_port => 1234},
  {:host => '10.0.0.3', :http_port => 5678}
])

# Retrieve a bucket
bucket = client.bucket("doc")  # a Riak::Bucket

# Get an object from the bucket
object = bucket.get_or_new("index.html")   # a Riak::RObject

# Change the object's data and save
object.raw_data = "<html><body>Hello, world!</body></html>"
object.content_type = "text/html"
object.store

# 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

# Access more like a hash, client[bucket][key]
client['doc']['index.html']   # the Riak::RObject

# Create a new object
new_one = Riak::RObject.new(bucket, "application.js")
new_one.content_type = "application/javascript" # You must set the content type.
new_one.raw_data = "alert('Hello, World!')"
new_one.store

Map-Reduce Example

# Assuming you've already instantiated a client, get the album titles for The Beatles
results = Riak::MapReduce.new(client).
                add("artists","Beatles").
                link(:bucket => "albums").
                map("function(v){ return [JSON.parse(v.values[0].data).title]; }", :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"]

Riak Search Examples

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

# Create a client, specifying the Solr-compatible endpoint
# When connecting to Riak 0.14 and later, the Solr endpoint configuration option is not necessary.
client = Riak::Client.new :solr => "/solr"

# Search the default index for documents
result = client.search("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

# Search the 'users' index for documents
client.search("users", "name:Sean")

# Add a document to an index
client.index("users", {:id => "sean@basho.com", :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

# 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 => "sean@basho.com"}) # removes from the 'users' index

# Seed MapReduce with search results
Riak::MapReduce.new(client).
        search("users","email:basho").
        map("Riak.mapValuesJson", :keep => true).
        run

# Detect whether a bucket has auto-indexing
client['users'].is_indexed?

# Enable auto-indexing on a bucket
client['users'].enable_index!

# Disable auto-indexing on a bucket
client['users'].disable_index!

Secondary Index Examples

Riak supports secondary indexes. Secondary indexing, or "2i," gives you the ability to tag objects with multiple queryable values at write time, and then query them later.

Tagging Objects

Objects are tagged with a hash kept behind the indexes method. Secondary index storage logic is in lib/riak/rcontent.rb.

object = bucket.get_or_new 'cobb.salad'

# Indexes end with the "_bin" suffix to indicate they're binary or string 
# indexes. They can have multiple values.
object.indexes[:ingredients_bin] = %w{lettuce tomato bacon egg chives}

# Indexes ending with the "_int" suffix are indexed as integers. They can
# have multiple values too.
object.indexes[:calories_int] = [220]

# You must re-save the object to store indexes.
object.store

Finding Objects

Secondary index queries return a list of keys exactly matching a scalar or within a range.

# The Bucket#get_index method allows querying by scalar...
bucket.get_index 'calories_int', 220 # => ['cobb.salad']

# or range.
bucket.get_index 'calories_int', 100..300 # => ['cobb.salad']

# Binary indexes also support both ranges and scalars.
bucket.get_index 'ingredients_bin', 'tomata'..'tomatz' # => ['cobb.salad']

# The collection from #get_index also provides a continuation for pagination:
c = bucket.get_index 'ingredients_bin', 'lettuce', max_results: 5
c.length # => 5
c.continuation # => "g2gCbQAAA="

# You can use that continuation to get the next page of results:
c2 = bucket.get_index 'ingredients_bin', 'lettuce', max_results: 5, continuation: c.continuation

# More complicated operations may benefit by using the `SecondaryIndex` object:
q = Riak::SecondaryIndex.new bucket, 'ingredients_bin', 'lettuce', max_results: 5

# SecondaryIndex objects give you access to the keys...
q.keys # => ['cobb.salad', 'wedge.salad', 'buffalo_chicken.wrap', ...]

# but can also fetch values for you in parallel.
q.values # => [<RObject {recipes,cobb.salad} ...>, <RObject {recipes,wedge...

# They also provide simpler pagination:
q.has_next_page? # => true
q2 = q.next_page

How to Contribute

License & Copyright

Copyright ©2010-2012 Sean Cribbs 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

http://www.apache.org/licenses/LICENSE-2.0

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 ©2009 Sean Cribbs, and is licensed under the Creative Commons Attribution Non-Commercial 3.0 license. "Creative Commons"