DiggiCamp

Scrape your digicampus (that use a very specific single sign-on)

Installation

Self-contained inside virtual environment:

If you don't want to install anything, just use diggicamp-venv.sh that automatically loads dependencies inside a virtual environment. Just symlink it into your desired PATH folder and everything will be taken care of. If you have any problems with the virtual environment, the script intercepts the install argument and reinstalls all dependencies again if supplied.

Other

If you feel the need to polute your global python package space, you can install dependencies manually by running pip install -r requirements.txt. You should probably not do this.

Usage

This... thing... exposes a couple of ways to interact with it:

CLI

to use the cli, link the diggicamp.py script anywhere to your path (alias is dgc or something like that) and then use it like this:

Usage: dgc [<flags>] <command> [<args>] [--cfg <path>]

Download files for courses from digicampus.

flags:

     -v                      verbose mode - more output


other args:

     --cfg <path>            specify a config file path (default is dgc.json)


commands:

    init [<url>] --user <username> [--pass <password>]
                             initialize a new config file. if no password is 
                             specified, you will be prompted, everytime it is
                             required (not very often).
    fetch [--threads <threadcount>]
                             refresh semester, courses, folders and files. Use
                             <threadcount> threads for this (default 32)
    show [--all]             show courses for the current (or all) semesters
    show <semester>          show courses in a specific semester
    show <course>            show files in a specific course from the current
                             semester
    show <semester> <course> show files in a specific course from a specific
                             semester
    clean                    cleans cached folders and files, deletes all 
                             download rules. This can be used to easily clean
                             up all state after the semester is completed

handling downloads: ('downloads' can be shortened to 'dl')

    downloads [show]         list all entries in the sync-list
    downloads add <course> [<folder>] <target> [--sem <semster>] [--regex <regex>]
                             add a folder (or all folders) of a course to the 
                             sync-list (it will sync with 'dgc pull'). If no 
                             semester is specified, the current semester is 
                             assumed. If a regex is specified, only files 
                             matching it will be downloaded
    downloads remove <id>    remove download with specified id
    pull [-f|--fetch] [--threads <threadcount>]
                             download all files from the folders on the
                             sync-list to their destinations. If not specified,
                             32 concurrent downloads are started

Python package

You can also just import the diggicamp package in python3 and use it that way. Although documentation is scarce.

from diggicamp import *

## Basics
# open an existing conf
dgc = d_open('path-to-your-dgc.json') #defaults to 'dgc.json'

# fetch from server
fetch(dgc, threads=16)

# create new dgc instance and save the dgc.json file
dgc = new('path-to-your-dgc.json') #defaults to 'dgc.json'
# add username and password to auth
dgc.conf.add_auth('plain', username="", password="")

# add a download config
add_download(dgc, folder_id, target_path) # optional pattern arg

# download all configured folders
pull(dgc, threads=64)

## Helpers:

# find semesters
semester_by_name(dgc, 'SS 2019')

# find courses 
course = course_by_name(dgc, 'Analysis II') # optional semster_title

# find folders
folder_by_name(dgc, 'Script', course)

# find courses by id:
course_by_id(dgc, 'a0808df98009adfe98af')

## Printing

# print courses
print_courses(dgc, all=True)

# print folders in a course
print_curses(dgc, course)

structure of the config file

The config file is json. It has the following entries:

• version: a version string representing the version of diggicamp
• baseurl: the base url of the digicampus website
• credentials: the credentials. Format for this is:
  • mode: the credential mode used. Available modes:
    • plain: The credentials are stored in plaintext in the fields username and password
    • prompt: Ask the user for his password every time login is required (this does not happen often, as the session is saved between executions)
• courses: an array of the courses. Sorted in descending order, so the first element is the current semester
• files: file list, structured like this:
  • files.course_id.folder_id:
    • id: folder id
    • name: file name
    • files: files contained in this folder, each file has the following fields:
      • id: file id
      • name: file name (displayed, no file ending)
      • fname: file name (real, with file ending)
• downloads: A list of downlaod rules with these fields:
  • folder: id of the folder
  • target: path where to download
  • regex: (optional) a regex files not matching will be ignored
• cookies: session cookies, used internally
• downloaded_version: The versions of the downloaded files to prevent double downloads

TO-DO:

• multithreaded downloads
• multithreaded sync
• pull with fetch in one command
• cli autocomplete
• multiple auth options (pass, prompt, etc)
• store session between calls
• compare regex before downloading