Made this and currently using this for a course where my professor does not release lecture recordings on leccap until right before the exams. The intention of this policy is to force students to come to lecture, but several classmates have expressed that they would appreciate having recordings just to review and learn what the professor maybe went over too fast in lecture, and I just happened to find a way to retrieve the recording URLs for reference.
One video was "leaked" to leccap (the first lecture), and from the url I got for that one (which I got from inspecting element enough times), I found the parts of the recording urls which correspond to the class and to the room in which the lecture was recorded in. Take a look in the research folder if you want to see my "research."
The rest of the url is just a timestamp (corresponding to the start time of the lecture, but may be off by ~10 seconds) and a random 3-digit code, so my script just generates all possible URLs and sends HEAD requests until it hits something.
Feel free to send a pull request to edit the script to make it more generalizable. Right now a lot of things are hard-coded in guesser.py. I also realize using the ThreadPoolExecutor may not be the most efficient idea, but I have not experimented with this.
To run the script, download this repo, create a python virtual environment, and change the constants in guesser.py (for now), run the script, and wait :)