saman222 / skyroom-record

Automate skyroom classes recording.

Skyroom recording automation

This repository is a dockerized environment to record classes of vclass.sharif.edu and skyroom.online.

How to use?

It's simple. Install docker and then create a downloads directory and use:

docker run --rm -v "$(pwd)/downloads:/opt/downloads" atofighi/skyroom-record:latest -u VLASS_URL -d CLASS_DURATION -n test-class -e encoding

Notes:

• VCLASS_URL must be the url of class with https://.

• CLASS_DURATION must be the duration of recording in minutes. like 90

• Your recorded video will be saved on ./downloads/test-class/NOW/video.webm.

• To update your docker image you can just run: docker pull atofighi/skyroom-record.

• Encoding quality -e

  This option converts the .webm file to a .mp4 file. It has encoding presets that should be defined otherwise no conversion would occur.

  high -best quality, slower and higher in size

  medium -best choice for uploading, medium quality, medium speed, medium size

  low - chat text becomes unreadable but low in size and fast

  no-encoding -default best quality, medium size, no conversion

• With the -v option you can specify a file to encode without needing to record. Please note that when using this option you should provide only -e and -v otherwise it wouldn't work.

Development

The main file is src/main.py. source code is not clean enough! sorry! ;)

There is also a simple flask app in src/app.py for viewing current window and can do clicks for test. You can expose port 5000 for viewing current screen for example with this command:

docker run --rm -v "$(pwd)/downloads:/opt/downloads" -p 5000:5000 atofighi/skyroom-record:latest -u VLASS_URL -d CLASS_DURATION -n test-class

And see http://localhost:5000. (Note: If you click on this page. App will click on docker window on that point.)

Roadmap

Done:

• Handling exceptions and retry when something went wrong.
• Logging for failure

TODO:

• Login with CW and vclass user/pass.
• Add cronjob for schedule recording
• Better failure detection (currently use screenshot similarity)
• Refactor code and make it clean

This work is licensed under a Creative Commons Attribution 4.0 International License, and it is offered “as-is”, without warranty, and we disclaim liability for any type of damages or problems resulting from using or abusing the project.