Rack::LTI exposes LTI launch and config URLs in your Rack application, handling authorization, storing launch parameters, and generating config information for consumers.
Add this line to your application's Gemfile:
gem 'rack-lti'
Rack::LTI should work with any Rack-based app. This means Rails 3.x and Sinatra, and probably whatever wonky framework you happen to be using.
Rack::LTI is tested on MRI Ruby 1.9 and 2.0, and the 1.9 branches of JRuby and Rubinius. It will not work on any flavor of 1.8; upgrade already.
Add Rack::LTI to your config/application.rb
:
class Application < Rails::Application
config.middleware.use Rack::LTI,
consumer_key: ->(key, consumer_id) { 'key' },
consumer_secret: ->(key, consumer_id) { 'secret' }
app_path: '/',
config_path: '/lti/config.xml',
launch_path: '/lti/launch',
redirect: true,
title: 'My LTI App',
description: 'My LTI App description',
nonce_validator: ->(nonce) { !FakeNonceStore.include?(nonce) },
success: ->(lti_params, request, response) {
request.session['launch_params'] = lti_params
response.headers['X-Custom-Header'] = 'value'
},
time_limit: 60*60,
extensions: {
'canvas.instructure.com' => {
course_navigation: {
default: 'enabled',
text: 'My LTI App'
}
}
},
custom_params: {
preferred_name: 'El Tigre Chino'
}
end
Add Rack::LTI to your app:
class Application < Sinatra::Base
use Rack::LTI,
consumer_key: 'my_key',
consumer_secret: 'my_secret',
app_path: '/',
config_path: '/lti/config.xml',
launch_path: '/lti/launch',
redirect: true,
title: 'My LTI App',
description: 'My LTI App description',
nonce_validator: ->(nonce) { !FakeNonceStore.include?(nonce) },
success: ->(lti_params, request, response) {
request.session['launch_params'] = lti_params
response.headers['X-Custom-Header'] = 'value'
},
time_limit: 60*60,
extensions: {
'canvas.instructure.com' => {
course_navigation: {
default: 'enabled',
text: 'My LTI App'
}
}
},
custom_params: {
preferred_name: 'El Tigre Chino'
}
end
Rack::LTI takes either a configuration hash or block at initialization. Allowed values are:
consumer_key
The consumer_key to check against the key given at launch. This value can be a string or a lambda. If a lambda, it is passed the key used by the consumer as well as their tool_consumer_instance_guid.consumer_secret
The consumer_secret to check against the secret given at launch. Like the consumer key, this value can be a string or a lambda. If a lambda, it is passed the key and tool_consumer_instance_guid of the consumer.app_path
The path to redirect to on a successful launch. This should be the main page of your application. Defaults to '/'.config_path
The path to serve LTI config XML from. Defaults to '/lti/config.xml'.launch_path
The path to receive LTI launch requests at. Defaults to '/lti/launch'.redirect
If true, redirect to theapp_path
. If false, pass the launch request through to the application. If false, app_path is not used.title
The title of your LTI application.description
The description of your LTI application.nonce_validator
A lambda used to validate the current request's nonce. It is passed the nonce to verify. If not provided, all nonces are allowed.time_limit
The time limit, in seconds, to consider requests valid within. If not passed, the default is 3600 seconds (one hour).success
A lambda called on successful launch. It is passed the launch params as a hash, the Rack Request, and the Rack Response. Can be used to cache params for the current user, find the current user, etc. By default, the launch params are stored in the 'launch_params' key of the session.extensions
A hash of extension information to include with the config. Format is platform -> option -> properties. See usage examples above for more detail.custom_params
A hash of custom parameters to accept from the client. See usage examples above for more detail.
Interested in learning more about LTI? Here are some links to get you started:
- Fork it
- Create your feature branch (
git checkout -b my-new-feature
) - Commit your changes (
git commit -am 'Add some feature'
) - Push to the branch (
git push origin my-new-feature
) - Create new Pull Request