Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


This project needs a new maintainer! Please contact @andrewvc if you'd like to take over!

Gem Version Build Status Code Climate Coverage Status

A concise, fast ElasticSearch Ruby client designed to reflect the actual elastic search API as closely as possible. Elastic search's API is complex, and mostly documented on the Elastic Search Guide. This client tries to stay out of your way more than others making advanced techniques easier to implement, and making debugging Elastic Search's sometimes cryptic errors easier. Stretcher is currently in production use by Pose, Get Satisfaction, Reverb, and many others.


  • Cleanly matches up to elastic search's JSON api
  • Efficiently re-uses connections on a per-server object basis (via excon)
  • Supports efficient bulk indexing operations
  • Returns most responses in convenient Hashie::Mash form
  • Configurable logging
  • Logs curl commandline statements in debug mode
  • Pure, threadsafe, ruby
  • Easily swap HTTP clients via Faraday
  • Tested against Ruby 2.0,1.9, Jruby, and Rubinius
  • Semantically versioned


Add this line to your application's Gemfile:

gem 'stretcher'


Getting Started

  1. Basic Usage
  2. Multi Search
  3. Bulk Indexing
  4. Percolate
  5. Logging
  6. Rails Integration


Stretcher Strives to provide adequate documentation for all methods. Many methods not described in the 'Getting Started' section can be found in the full rdocs.

Basic Usage

# First Create a server
server ='http://localhost:9200')
# Delete an index (in case you already have this one)
server.index(:foo).delete rescue nil
# Create an index
server.index(:foo).create(mappings: {tweet: {properties: {text: {type: 'string'}}}})
# Add some documents
30.times {|t| server.index(:foo).type(:tweet).put(t, {text: "Hello #{t}"}) }
# Retrieve a document
# => #<Hashie::Mash text="Hello 3">
# Perform a search (Returns a Stretcher::SearchResults instance)
res = server.index(:foo).search(size: 12, query: {match_all: {}})
res.class     # Stretcher::SearchResults     # => 30
res.documents # => [#<Hashie::Mash _id="4" text="Hello 4">, ...]
res.facets    # => nil
res.raw       # => #<Hashie::Mash ...> Raw JSON from the search
res.raw_plain # => #<Hash ...> Non-Hashie Raw JSON from the search (fastest)
# use an alias
alias = server.index(:foo).alias(:my_alias)
alias.create({ filter: { term: { user_id: 1 } } }){ query: { match_all: {} } })
# or get some cluster health information # Hashie::Mash

Block Syntax

# A nested block syntax is also supported.
# with_server takes the same args as #new, but is amenable to blocks
Stretcher::Server.with_server('http://localhost:9200') {|srv|
  srv.index(:foo) {|idx|
    idx.type(:tweet) {|t| {exists: t.exists?, mapping: t.get_mapping} }
# => {:exists=>true, :mapping=>#<Hashie::Mash tweet=...>}

Multi Search

# Within a single index
server.index(:foo).msearch([{query: {match_all: {}}}])
# => Returns an array of Stretcher::SearchResults
# Across multiple indexes
server.msearch([{index: :foo}, {query: {match_all: {}}}])
# => Returns an array of Stretcher::SearchResults

Bulk Indexing

docs = [{"_type" => "tweet", "_id" => 91011, "text" => "Bulked"}]


Implements the ElasticSearch Percolate API

# Register a new percolate query
server.index(:foo).register_percolator_query(query_name, query_hash)
# Check a document against an index, returns an array of query matches
# Delete a percolate query


Pass in the :log_level parameter to set logging verbosity. cURL statements are surfaced at the :debug log level. For instance:'http://localhost:9200', :log_level => :debug)

You can also pass any Logger style object with the :logger option. For instance:'http://localhost:9200', :logger => Rails.logger)

Rails Integration

Stretcher is a low level-client, but it was built as a part of a full suite of Rails integration tools. While not yet open-sourced, you can read our detailed blog post: integrating Stretcher with Rails.

Running Specs

Running the specs requires an operational Elastic Search server on http://localhost:9200. The test suite is not to be trusted, don't count on your indexes staying around!

Specs may be run with rake spec

Email or tweet @andrewvc if you'd like to be added to this list!



  1. Fork it
  2. Create your feature branch (git checkout -b my-new-feature)
  3. Commit your changes (git commit -am 'Add some feature')
  4. Push to the branch (git push origin my-new-feature)
  5. Create new Pull Request


Fast, Elegant, ElasticSearch client




No packages published