Newsboat, yt-dlp and mpv

If you’re a fan of RSS at all, or just tired of how terrible the Youtube experience is on the web. Newsboat + yt-dlp may be a decent alternative for you. I’ve been using it now for several weeks and it’s been great. Since switching from youtube-dl. Since Youtube has been hosing it’s speeds, yt-dlp doesn’t seem to run into the same bandwidth issues (at least for now, knock on wood).

With a couple very simple configurations, you can turn Newsboat into a decent youtube front end, without any of the ads and piping directly to mpv for viewing videos.

Installation

Newsboat and mpv should both be in the regular Arch repos, however yt-dlp is only found in the AUR’s for now, so you will need something like yay to install it.

yay -S newsboat mpv yt-dlp

Configuration

Below is my Newsboat configuration. It’s quite basic, but it works quite nicely for me.

~/.newsboat/config

auto-reload no
browser firefox
macro v set browser "setsid -f mpv --really-quiet --no-terminal" ; open-in-browser ; set browser firefox

, will toggle the macro mode of Newsboat, and as you can see in the configuration above we mapped it to v (for video, but you can choose any other key that you like).

, + v will open youtube links in mpv.

Finding youtube channel RSS links

Next you will need some RSS links to the youtube channels that you want to watch. You can snag this by peeking at the source of the youtube channels. Grabbing the ID from the URL bar and building your link manually, or this short command line script below.

yt_snag_rss_link.py

import argparse
import requests

# snag the rss link from the youtube channel
def snag_rss_link(channel_id):
    response = requests.get(f'https://www.youtube.com/channel/{channel_id}')
    # alternatively if you don't want to actually hit youtube and parse the source
    # you can just build the link and template the channel_id in there.
    #return f'https://www.youtube.com/feeds/videos.xml?channel_id={channel_id}'
    return response.text.split('"rssUrl":"')[1].split('"')[0]

if __name__ == '__main__':
    parser = argparse.ArgumentParser(description='Get the rss link from a youtube channel')
    parser.add_argument('channel_id', help='the channel id')
    channel_id = parser.parse_args().channel_id

    # channel_id = 'UCBJycsmduvYEL83R_U4JriQ'
    print(snag_rss_link(channel_id))

Here is how to use the script:

$ python yt_snag_rss_link.py UCBJycsmduvYEL83R_U4JriQ

https://www.youtube.com/feeds/videos.xml?channel_id=UCBJycsmduvYEL83R_U4JriQ

From there you can add the URLs to your Newsboat urls file.

~/.newsboat/urls

"  "
"---HN---"
https://news.ycombinator.com/rss "news"

"  "
"---Python---"
https://reddit.com/r/python/.rss "python"
https://reddit.com/r/learnpython/.rss "python"
https://reddit.com/r/pythontips/.rss "python"
https://reddit.com/r/pygame/.rss "python"

"  "
"---People---"
https://www.mcgillij.dev/feeds/all.atom.xml "me"
https://fasterthanli.me/index.xml

"  "
"---Youtube---"
https://www.youtube.com/feeds/videos.xml?channel_id=UCOWcZ6Wicl-1N34H0zZe38w "linux" #Level1Tech Linux
https://www.youtube.com/feeds/videos.xml?channel_id=UCSAXsBMga3Y2wYSPwFI5f5w #gnif
https://www.youtube.com/feeds/videos.xml?channel_id=UChz00vupzP_mNPIYD8GSmBw "overlanding" #dahl
https://www.youtube.com/feeds/videos.xml?channel_id=UCUMSHXWczvxHy9e8silnVNw "gloriouseggroll" #glorious eggroll
https://www.youtube.com/feeds/videos.xml?channel_id=UCBJycsmduvYEL83R_U4JriQ "marquee"