Bridge for Adaptivity

About

An application that interfaces with an adaptive engine to dynamically serve content in MOOCs based on real time student activity.

The ALOSI adaptivity ecosystem

The Bridge for Adaptivity is designed to work with three external systems to enable adaptivity in a course. These are:

• LMS (Learning Management System), for example edX, Open edX, Canvas or other LTI consumers
• Content Source - contains the content (problems, html content) to serve dynamically. Examples of a content source system might be Open edX or other LTI providers.
• Adaptive Engine - Provides activity recommendations based on student activity. An example of an adaptive engine application is the ALOSI adaptive engine.

More information

Visit our github wiki or the ALOSI Labs site for more information about our group and our work.

Containers list

• Container with Bridge for Adaptivity application

• Container with postgressql database

• Container with celery worker

• Container with rabbitmq message queue

• Container with nginx (doesn't exist for local deployment)

Getting started

Deployment

Deployment is based on the Docker containers. There are three config files docker-compose_local.yml, docker-compose-stage.yml and docker-compose.yml for local , stage and production deployments respectively.

Docker and Docker Compose are required to be installed before start the deploying.

Clone project.

Before running deployment configure secure.py settings in the bridge_adaptivity/config/settings/ directory (see secure.py.example).

Local deployment

Local deployment can be started by the make command in the console in bridge_adaptivity directory:

[sudo] make run-local

Volume "pgs" is added to the the database container.

Note: Development server available on localhost:8008

Running tests

You can run tests locally (directly on your host), or on the docker machine.

• to run tests locally:
  • install requirements with command pip install -r requirements_local.txt
  • run tests: python manage.py test --settings config.settings.test or just pytest. Both commands are equal.
• to run tests in docker container:
  • create docker container: make run-local
  • run tests: docker exec -it BFA pytest
    • if you see an error:

      import file mismatch:
      which is not the same as the test file we want to collect:
      /bridge_adaptivity/config/settings/test.py
      HINT: remove __pycache__ / .pyc files and/or use a unique basename for your test file modules
      

      you should run: find . | grep -E "(__pycache__|\.pyc|\.pyo$)" | xargs rm -rf and after that retry running the tests: docker exec -it BFA pytest

Staging deployment

Please ensure that file in nginx/stage/bridge.conf exists and is configured in proper way.

Run make command to start staging deployment:

[sudo] make run-stage

Production deployment

Please ensure that file in nginx/prod/bridge.conf exists and is configured in proper way.

Run make command to start production deployment:

[sudo] make run

Additional notes

• if requirements changes were made containers rebuilding needed:

  [sudo] make docker-build

• For run migration(production, stage and local):

  [sudo] make migrate

  [sudo] make migrate-stage

  [sudo] make migrate-local