https://github.com/nwalters512/cs411-prairielearn-demo

Work on a demo of how PrairieLearn could be used for CS 411 assignments
https://github.com/nwalters512/cs411-prairielearn-demo

Last synced: 3 days ago
JSON representation

Work on a demo of how PrairieLearn could be used for CS 411 assignments

• Host: GitHub
• URL: https://github.com/nwalters512/cs411-prairielearn-demo
• Owner: nwalters512
• Created: 2019-02-06T21:15:49.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2019-03-11T23:37:03.000Z (over 5 years ago)
• Last Synced: 2024-10-10T03:10:51.948Z (27 days ago)
• Language: Python
• Size: 15.6 KB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# cs411-prairielearn-demo

Work on a demo of how PrairieLearn could be used for CS 411 assignments

## Question Structure

Questions are set up to require as little work as possible to write a new question.

A given question must specify two files in the `tests` directory:

* `setup.sql`: creates any necessary tables and populates them with data; can contain

  multiple queries

* `solution.sql`: contains a single query that defines the "solution" to this question.

When a question is graded, two databases are created and populated with `setup.py`.

The query in `solution.sql` is run on one of the databases, and the query submitted

by the student is run on the other. The resulting relations are compared, and the

student gets credit if the relation they return matches the solution exactly.

With those two files, it's possible to create a question that tests `SELECT` queries.

But what if you want to test insertion, updating, or deletion? Those queries won't

return relations; instead, you really want to diff the contents of a database after

the query completes. To do that, you can specify an optional third file `dump.sql`.

This contains a query that will be used to diff the contents of the databases after

the solution and student queries are run on their respective databases.

This allows question authors to largely avoid hardcoding answers. An author must simply

specify the input data and the query that should operate on it, and the student submission

can be automatically checked for correctness.

## Running locally

The Docker image with MySQL in it isn't published anywhere yet. To run this with a local

copy of PrairieLearn, you'll need to first build the Docker image:

```

docker build docker -t mysql-test

```