The following document summarizes some core concepts around the development of the class scheduler, and should be read before contributing to the project. Suggestions of change to this document are always welcome.
Installation of a compatible clingo solver and
all of the Python dependencies can be done using
docker and docker
compose. There are currently two relevant
services, dev and test, that can be used to run a interactive shell in the
preconfigured environment and to run the constraints unit test routine,
respectively.
The following bash snippet runs a bash shell in the development environment:
docker compose run --rm devThe following bash snippet runs all tests for the project:
docker compose run --rm testWhile we are not yet done with the class scheduler, there is a proof of
concept (POC) version of the expected final product, without some of the
required constraints. The POC can be run using by starting a interactive bash
session within the development environment (see the section
above) and running the script
class-scheduler.sh. The expected output is a pretty table with a valid
schedule.