Home
Hello and welcome to the ISIS Data Fetcher Project.
ISIS stands for Information System for Instructors and Student from the TU Berlin.
As far as I know ISIS bases on Moodle and has a Shibboleth Login in front.
So the project is dedicated to all TU Students but everyone is invited to help. I can't provide a test account for non TU students so if you want to use/test any code you have to have an ISIS account or the credentials from someone else.
I want a programm which downloads all PDFs, ZIPs and so on from all your courses from the ISIS website.
Currently I programm in python and @Jokabe in bash.
##Programm Sections
###Login
The login from ISIS is done with Shibboleth.
The following image shows how it works.
You have to access the login page from the ISIS main page to get the initial cookie.
My approach is very simply with Selenium which just follows the redirections.
If everything went well you should be on your ISIS page(https://isis.tu-berlin.de/my/) and have several cookies which you can use from now on to access all your pages.
This is done in shell and python.
###Finding Courses, Weeks and Data Now we want to extract at first the courses which should be done with XPaths in my opinion. The XPaths for the most important are/will be on another wiki page.
The first step should be to collect all your courses, the second to collect the weeks from your courses and the last to get all file links from a week.
So in the end we will get a tree structure with the user, the courses as first level, weeks as second and files as third.
That means for example:
user
|-course1
||-week1
|||-pdffile1.pdf(link)
|||-zip2.zip(link)
||-week2
|||-video.mp4(link)
||-week3
|-course2
.
.
.
###Fetching Data Last but not least everything needs to be downloaded and copied in the right folder structure of your PC.
##Approach Right now everything is made with Selenium but the latest idea has been to make the login with Selenium and give the cookies to spyder so he can crawl for all files.