This is very rough tool I've used to check student work against a collection of previous submissions and solutions online.
Over a couple of years, it became a little too hacky for me and I started the process of re-factoring it.
I have not used it since, but I believe it still works if used carefully.
A directory which contains any distribution code for each problem you want to check.
A directory which contains a subdirectory for each problem (named for the problem). Each of those directories should have subdirectories each of which contains a solution to the problem (named for the problem).
It should make sense once you have tried to use this.
I use a git repo which I forked from another maintained by dkiang and I store this in a directory that is a sibling of this directory. These solutions SHOULD NOT be published online! I keep them in a directory which is a sibling of this directloy and I highly recommend doing that. If you don't, you'll need to adjust your settings in settings.py
An empty directory which will contain your students' code. You will use the setup_students.py
script here to clone your students' repos to this directory.
- Ensure
settings.py
is configured for your directory and path names - Ensure each of your students has submitted at least one problem using
submit50
. - Add the GitHub usernames for each of your students to
students.txt
one per line. - Run
setup_students.py
using Python 3
- Esure you've added any distribution code to the correct directory.
- Run
check_problem.py
passing thesubmit50
slug as the only argument - The script produces a lot of output, some(much?) of which is impossible to read.
- You should find a directory named
f{problem_name}_results
- Navigate to that folder and launch
index.html
to view results