uva-study-progress

StudyProgress Dashboard for the University of Amsterdam, 2ndYD

Development

1. As the local source code is mounted into the container(s) at runtime, any dependencies should be installed in the local folder: composer install --ignore-platform-reqs.
2. If you don't have an APP_KEY yet you should run php artisan key:generate --show and add the value to .env.development.local.
3. Run composer run docker-build to build the local Docker images.
4. Start the development environment with the docker-compose up command.
5. Optionally, run composer run docker-migrate and composer run docker-seed to prepare your database.
6. Build the browser bundle: npm run watch.

By default, the following services are accessible from your browser:

• http://localhost:8888
• http://localhost:8888/api
• http://localhost:8889, phpMyAdmin

docker-compose down will stop the services. Next time, you'll only have to run docker-compose up again to start developing.

Deploy to production

1. Check out a copy of the repository.
2. Build the browser bundle: npm run prod.
3. Build the Docker images: composer run docker-build.
4. Deploy the production stack: docker stack deploy --compose-file docker-stack.yml uva-study-progress.
5. If necessary, run migrate and seed scripts on an fpm container.

Roles

The following roles are used in the application and should be provided by the LTI tool consumer:

1. Administrator: feb_dashboard_admin (was: rens_1)
2. Study Adviser: feb_dashboard_studyadviser (was: rens_2)
3. Student: urn:lti:role:ims/lis/Learner

LTI setup

For Canvas LMS, an XML is available for easily integrating the tool at /lti/canvas.xml.

In general, the following LTI parameters are required:

• Launch URL: https://path.to.tool/lti/launch.
• Custom parameter**: custom_student_number should provide a unique, display-friendly student identifier.

Optional parameters:

• Custom parameter**: custom_role_membership provides a way to override standard LTI roles. Useful when a tool consumer / LMS doesn't support the required roles.

** Note: Blackboard LMS seems to automatically prefix custom parameters with custom_.

Most tool consumers will automatically provide all other parameters necessary. The LTI documentation can be consulted for instructions on creating a complete request.