Copy the sample configuration file to the configuration directory. This is used as a template for the final configuration.
mkdir -p config
cp datacultures/config/.env_conf.yml.example config/.env_conf.yml
The LTI standard specifies that an embedded LTI provider share an application key and application secret for authentication. They are configured from the Canvas UI.
The database and Canvas information must be updated. Do this for each environment you wish to configure (production, qa, staging, development, etc):
From the sample file:
production:
assets:
served_by_app_server: false # set to false where Apache is available
databases:
database: 'datacultures_production'
host: 'localhost'
port: 5432
api:
server: 'http://localhost:3100/'
course: '1'
api_key: 'FOOBARBAZ'
Edit the configuration file (../config/.env_conf.yml relative the project root, directory created and file copied as above) Configure the 'databases' section with appropriate values:
databases:
database: 'lti_app_live'
host: 'https://some.database.host'
port: 9000
and also configure the last section. This section is for the Canvas instance in which the app is embedded. If the instructor's API key is 'uq0b5QOsjTfpAjYjXUFSUur7lkX2JkgKJlDNKc+FA3Q3gwmQYTo1MOIpAlt112pfjBeuG0N7bvqP9YGrXWPTmnQ', it might look like this:
api:
server: 'https://datacultures-canvas.instructure.com/'
course: '30765'
api_key: 'uq0b5QOsjTfpAjYjXUFSUur7lkX2JkgKJlDNKc+FA3Q3gwmQYTo1MOIpAlt112pfjBeuG0N7bvqP9YGrXWPTmnQ'