CanvasFS

This is currently a toy project to experiment with mounting assignments from Canvas as a file system. The main idea is that it lets me inspect files students hand in and the state of the assignments without having to download them with a browser.

To get an idea why: the students in the current course I'm teaching have handed in 994 files so far. About half of them zip files with source code.

The file system exposes the revision history for handed in assignments to make it easier to focus on improvements the students have made.

Currently, it only focuses on assignments.

Some of the code here might be merged with the https://github.com/edvardpedersen/canvashelper project.

Requirements

It has only been tested with Linux, but fusepy should work with Mac as well. On Windows, it may work with WSL2 which now has support for Fuse filesystems.

• Python 3.8+ (sorry, I just had to play with the Walrus operator)
• fusepy : https://github.com/fusepy/fusepy
• canvasapi : https://github.com/ucfopen/canvasapi

Setting up

You need an API key for Canvas. The easiest way to do this is to go to Account/Settings in Canvas. There you can create a new access token.

You need to set up a config.toml file for some config values. The file is not included to avoid accidental mistakes, but this can be used as a template:

base_url = "<url>"
course_id = <numeric id>
api_key = "<key>"
include_completed = <true/false>

The contents are as follows:

• base_url is the the URL for your canvas installation.
• course_id is the numeric id of the course you want to browse.
• api_key is the access token / API key for Canvas.
• include_completed a true or false value. True if you want to view submissions after a course is closed, or if you want to view students that have withdrawn from the course.

Using

Fetching the metadata from canvas takes a while. To make it easier to experiment, the metadata is downloaded as follows:

python3 get-submission-info.py

The data is stored in the .cache subdirectory.

Mounting the filesystem is as simple as:

python3 canvasfs.py tmp

Where tmp is the directory where you want to mount the assignments.

Metadata for directories and the files inside the directories is available as a json file called .meta.

Files that students have handed in can be read in the file system normally. To avoid bothering Canvas too much, the files are cached in the .cache directory.

Example: using ls to find the newest hand-ins:

ls -ltr t/Assignment\ 1\ -\ Breakout/

Future

Caching downloaded information will probably be removed (or made optional) in the future to avoid storing too much data on the computer. If you want to be on the safe side, you can remove the .cache when unmounting.