Octokit
Ruby toolkit for the GitHub API.
Octokit 2.0 is out, check the Upgrade Guide before upgrading from 1.x.
Philosophy
API wrappers should reflect the idioms of the language in which they were written. Octokit.rb wraps the GitHub API in a flat API client that follows Ruby conventions and requires little knowledge of REST. Most methods have positional arguments for required input and an options hash for optional parameters, headers, or other options:
# Fetch a README with Accept header for HTML format
Octokit.readme 'al3x/sovereign', :accept => 'application/vnd.github.html'
Quick start
Install via Rubygems
gem install octokit
... or add to your Gemfile
gem "octokit", "~> 2.0"
Making requests
API methods are available as module methods (consuming module-level configuration) or as client instance methods.
# Provide authentication credentials
Octokit.configure do |c|
c.login 'defunkt'
c.password 'c0d3b4ssssss!'
end
# Fetch the current user
Octokit.user
or
# Provide authentication credentials
client = Octokit::Client.new :login => 'defunkt', :password => 'c0d3b4ssssss!'
# Fetch the current user
client.user
Consuming resources
Most methods return a Resource
object which provides dot notation and []
access for fields returned in the API response.
# Fetch a user
user = Octokit.user 'jbarnette'
puts user.name
# => "John Barnette"
puts user.fields
# => <Set: {:login, :id, :gravatar_id, :type, :name, :company, :blog, :location, :email, :hireable, :bio, :public_repos, :followers, :following, :created_at, :updated_at, :public_gists}>
puts user[:company]
# => "GitHub"
user.rels[:gists].href
# => "https://api.github.com/users/jbarnette/gists"
Note: URL fields are culled into a separate .rels
collection for easier
Hypermedia support.
Accessing HTTP responses
While most methods return a Resource
object or a Boolean, sometimes you may
need access to the raw HTTP response headers. You can access the last HTTP
response with Client#last_response
:
user = Octokit.user 'andrewpthorp'
response = Octokit.last_response
etag = response.headers[:etag]
Authentication
Octokit supports the various authentication methods supported by the GitHub API:
Basic Authentication
Using your GitHub username and password is the easiest way to get started making authenticated requests:
client = Octokit::Client.new \
:login => 'defunkt',
:password => 'c0d3b4ssssss!'
user = client.user
user.login
# => "defunkt"
While Basic Authentication makes it easy to get started quickly, OAuth access tokens are the preferred way to authenticate on behalf of users.
OAuth access tokens
OAuth access tokens provide two main benefits over using your username and password:
- Revokable access. Access tokens can be revoked, removing access for just that token without having to change your password everywhere.
- Limited access. Access tokens have access scopes which allow for more granular access to API resources. For instance, you can grant a third party access to your gists but not your private repositories.
To use an access token with the Octokit client, just pass it in lieu of your username and password:
client = Octokit::Client.new :access_token => "<your 40 char token>"
user = client.user
user.login
# => "defunkt"
You can use .create_authorization
to create a token using Basic Authorization
that you can use for subsequent calls.
Using a .netrc file
Octokit supports reading credentials from a netrc file (defaulting to
~/.netrc
). Given these lines in your netrc:
machine api.github.com
login defunkt
password c0d3b4ssssss!
You can now create a client with those credentials:
client = Octokit::Client.new :netrc => true
client.login
# => "defunkt"
But I want to use OAuth you say. Since the GitHub API supports using an OAuth token as a Basic password, you totally can:
machine api.github.com
login defunkt
password <your 40 char token>
Note: Support for netrc requires adding the netrc gem to your Gemfile
or .gemspec
.
Application authentication
Octokit also supports application-only authentication using OAuth application client credentials. Using application credentials will result in making anonymous API calls on behalf of an application in order to take advantage of the higher rate limit.
client = Octokit::Client.new \
:client_id => "<your 20 char id>",
:client_secret => "<your 40 char secret>"
user = client.users 'defunkt'
Configuration and defaults
While Octokit::Client
accepts a range of options when creating a new client
instance, Octokit's configuration API allows you to set your configuration
options at the module level. This is particularly handy if you're creating a
number of client instances based on some shared defaults.
Configuring module defaults
Every writable attribute in Octokit::Configurable can be set one at a time:
Octokit.api_endpoint = 'http://api.github.dev'
Octokit.web_endpoint = 'http://github.dev'
or in batch:
Octokit.configure do |c|
c.api_endpoint = 'http://api.github.dev'
c.web_endpoint = 'http://github.dev'
end
Using ENV variables
Default configuration values are specified in Octokit::Default. Many attributes will look for a default value from the ENV before returning Octokit's default.
# Given $OCTOKIT_API_ENDPOINT is "http://api.github.dev"
Octokit.api_endpoint
# => "http://api.github.dev"
Hypermedia agent
Starting in version 2.0, Octokit is hypermedia-enabled. Under the hood, Octokit::Client uses Sawyer, a hypermedia client built on Faraday.
Hypermedia in Octokit
Resources returned by Octokit methods contain not only data but hypermedia link relations:
user = Octokit.user 'technoweenie'
# Get the repos rel, returned from the API
# as repos_url in the resource
user.rels[:repos].href
# => "https://api.github.com/users/technoweenie/repos"
repos = user.rels[:repos].get.data
repos.last.name
# => "faraday-zeromq"
When processing API responses, all *_url
attributes are culled in to the link
relations collection. Any url
attribute becomes .rels[:self]
.
URI templates
You might notice many link relations have variable placeholders. Octokit supports URI Templates for parameterized URI expansion:
repo = Octokit.repo 'pengwynn/pingwynn'
rel = repo.rels[:issues]
# => #<Sawyer::Relation: issues: get https://api.github.com/repos/pengwynn/pingwynn/issues{/number}>
# Get a page of issues
repo.rels[:issues].get.data
# Get issue #2
repo.rels[:issues].get(:uri => {:number => 2}).data
The Full Hypermedia Experience™
If you want to use Octokit as a pure hypermedia API client, you can start at the API root and and follow link relations from there:
root = Octokit.root
root.rels[:repository].get :uri => {:owner => "octokit", :repo => "octokit.rb" }
Octokit 3.0 aims to be hypermedia-driven, removing the internal URL construction currently used throughout the client.
Upgrading guide
Version 2.0 includes a completely rewritten Client
factory that now memoizes
client instances based on unique configuration options. Breaking changes also
include:
:oauth_token
is now:access_token
Hashie::Mash
has been removed. Responses now return aSawyer::Resource
object. This new type behaves mostly like a RubyHash
, but does not fully support theHashie::Mash
API.- Two new client error types are raised where appropriate:
Octokit::TooManyRequests
andOctokit::TooManyLoginAttempts
- The
search_*
methods from v1.x are now found atlegacy_search_*
- Support for netrc requires including the netrc gem in your Gemfile or gemspec.
Advanced usage
Since Octokit employs Faraday under the hood, some behavior can be extended via middleware.
Debugging
Often, it helps to know what Octokit is doing under the hood. Faraday makes it easy to peek into the underlying HTTP traffic:
stack = Faraday::Builder.new do |builder|
builder.response :logger
builder.use Octokit::Response::RaiseError
builder.adapter Faraday.default_adapter
end
Octokit.middleware = stack
Octokit.user 'pengwynn'
I, [2013-08-22T15:54:38.583300 #88227] INFO -- : get https://api.github.com/users/pengwynn
D, [2013-08-22T15:54:38.583401 #88227] DEBUG -- request: Accept: "application/vnd.github.beta+json"
User-Agent: "Octokit Ruby Gem 2.0.0.rc4"
I, [2013-08-22T15:54:38.843313 #88227] INFO -- Status: 200
D, [2013-08-22T15:54:38.843459 #88227] DEBUG -- response: server: "GitHub.com"
date: "Thu, 22 Aug 2013 20:54:40 GMT"
content-type: "application/json; charset=utf-8"
transfer-encoding: "chunked"
connection: "close"
status: "200 OK"
x-ratelimit-limit: "60"
x-ratelimit-remaining: "39"
x-ratelimit-reset: "1377205443"
...
See the Faraday README for more middleware magic.
Caching
If you want to boost performance, stretch your API rate limit, or avoid paying the hypermedia tax, you can use Faraday Http Cache.
Add the gem to your Gemfile
gem 'faraday-http-cache'
Next, construct your own Faraday middleware:
stack = Faraday::Builder.new do |builder|
builder.use Faraday::HttpCache
builder.use Octokit::Response::RaiseError
builder.adapter Faraday.default_adapter
end
Octokit.middleware = stack
Once configured, the middleware will store responses in cache based on ETag
fingerprint and serve those back up for future 304
responses for the same
resource. See the project README for advanced usage.
Hacking on Octokit.rb
If you want to hack on Octokit locally, we try to make bootstrapping the project as painless as possible. Just clone and run:
script/bootstrap
This will install project dependencies and get you up and running. If you want to run a Ruby console to poke on Octokit, you can crank one up with:
script/console
Using the scripts in ./scripts
instead of bundle exec rspec
, bundle
console
, etc. ensures your dependencies are up-to-date.
Running and writing new tests
Octokit uses VCR for recording and playing back API fixtures during test
runs. These fixtures are part of the Git project in the spec/cassettes
folder. For the most part, tests use an authenticated client, using a token
stored in ENV['OCTOKIT_TEST_GITHUB_TOKEN']
. If you're not recording new
cassettes, you don't need to have this set. If you do need to record new
cassettes, this token can be any GitHub API token because the test suite strips
the actual token from the cassette output before storing to disk.
Since we periodically refresh our cassettes, please keep some points in mind when writing new specs.
- Specs should be idempotent. The HTTP calls made during a spec should be able to be run over and over. This means deleting a known resource prior to creating it if the name has to be unique.
- Specs should be able to be run in random order. If a spec depends on another resource as a fixture, make sure that's created in the scope of the spec and not depend on a previous spec to create the data needed.
- Do not depend on authenticated user info. Instead of asserting actual values in resources, try to assert the existence of a key or that a response is an Array. We're testing the client, not the API.
Supported Ruby Versions
This library aims to support and is tested against the following Ruby implementations:
- Ruby 1.9.2
- Ruby 1.9.3
- Ruby 2.0.0
If something doesn't work on one of these Ruby versions, it's a bug.
This library may inadvertently work (or seem to work) on other Ruby implementations, however support will only be provided for the versions listed above.
If you would like this library to support another Ruby version, you may volunteer to be a maintainer. Being a maintainer entails making sure all tests run and pass on that implementation. When something breaks on your implementation, you will be responsible for providing patches in a timely fashion. If critical issues for a particular implementation exist at the time of a major release, support for that Ruby version may be dropped.
Versioning
This library aims to adhere to Semantic Versioning 2.0.0. Violations of this scheme should be reported as bugs. Specifically, if a minor or patch version is released that breaks backward compatibility, that version should be immediately yanked and/or a new version should be immediately released that restores compatibility. Breaking changes to the public API will only be introduced with new major versions. As a result of this policy, you can (and should) specify a dependency on this gem using the Pessimistic Version Constraint with two digits of precision. For example:
spec.add_dependency 'octokit', '~> 2.0'