These obsolete scripts may still be useful as long as they work. They may still be useful in case of emergency (ie. getting a fatal error in a pinch). The main program above was based on these originally. They are very limited, and should still mostly work, apart from the Youtube throttling problem which requires advanced pre-processing of the download URLs such as computing token signature (thanks Youtube for breaking our things on purpose!).
This was originally a gist, forked from this script by @cheadrian. The script is called manual_download.py.
While it should work fine, it is very basic and not very convenient to use.
For this you have to be logged into Youtube, grab the links from your web browser's developer tools, and manually replace the vid_link and sound_link variables every time.
After the script has downloaded all the segments, use pre_merge.py.
This script is a simple compatibility bridge for the merge script located in the youtube_stream_capture. It simply emulates renaming the downloaded segments by creating symbolic links to the files located in stream_capture_IDHASH/aud and stream_capture_IDHASH/vid into one directory called segments_IDHASH and with appropriate filenames.
- Open a Youtube live stream in your browser
- Get the video and audio links from the developer tools (CTRL+I)
- Copy them in
vid_link=andaudio_link=variables $> python manual_download.py- Once chunks have been downloaded in
stream_capture_IDHASH, call$> python pre_merge.py - Call youtube_stream_capture/merge.py like this
$> merge.py https://www.youtube.com/watch?v=IDHASH - If everything worked, you have the final video file generated. You can remove the
segments_IDHASHdirectory containing the symbolic links, as well as the source chunks
This script is currently obsolete, but could be fixed by calling ffmpeg the same way as the above decent script. I can't be bothered to rewrite this for now.