Seeding

Subato provides a YAML-based interface for the initial import of courses, terms, exercises, languages, task pools, and other entities. Run the import with SeedRunner:

gradle bootRun --args='seed:seed ./seed'

The second argument is a directory containing the YAML files, see demo/seed for a working example. For what can be imported, check the code in de.hsrm.sls.subato.seed.model. YAML attributes map to fields using snake_case, and references to other entities can be made by using YAML slugs. For example, a term with the slug ws2324 can be referenced as follows when defining a course:

ws2324:
  name: WS2324
  lectures_start_at: 2023-10-16
  lectures_end_at: 2024-02-02

oose:
  name: Objektorientierte Softwareentwicklung
  short_name: OOSE
  term: ws2324

Not idempotent

Subato generates a unique ID for most entities at seed time, and re-running a seed does not check whether an entity with the same name already exists. Seeding is meant to run once against an empty database, running it again against the same database will create duplicates.

When linking tasks from a task-pool to an exercise in exercises.yaml (through the task_instances attribute), the attribute tracking_id has to refer to the trackingId field in the meta.xml of the target task. See the STEF specification for more information on how tasks can be defined.