Video Archiver

A python script for building your own personal little archive of internet videos, built on the back of youtube-dl yt-dlp. Designed for YouTube, but should work with any popular video sharing site.

Dependencies

• Python 3 with the following packages
  • yt-dlp
  • click
  • requests
• ffmpeg
• Pushover (optional)

Usage

Downloading

The main script is archive.py. In it's simplest form, you can download a new video just by passing the URL as an argument:

./archive.py https://www.youtube.com/watch?v=dQw4w9WgXcQ

This will download the video and place it in a directory called Archives wherever you run the script from. Information about the download will be logged to archive.log. A text file called download archive.txt will also be created, which will prevent the video from being downloaded again.

You can also pass a URL to a playlist or youtube channel:

./archive.py https://www.youtube.com/playlist?list=PLqs5ohhass_RugObMuXClrZh7dP0g4e5l
./archive.py https://www.youtube.com/@videogamedunkey/videos

Note that omitting the /videos from a channel URL can cause unexpected results.

The archiver can also take a file containing a list of URLs using the -f option:

./archive.py -f channels.txt

This file should have one link per line. You can specify a limit to the number of videos to get after each link, separated by a semicolon. For more advanced filtering, you can also specify specific yt-dlp options, separated by two colons. Lines starting with a # are ignored.

# Example channels.txt
https://www.youtube.com/user/freddiew/videos;5
https://www.youtube.com/user/marcrebillet/videos;5::{"match_filter":"duration < 1800"}

Path Options

By default, the script uses a temporary directory downloads to download videos, and then moves them to the Archives directory. If you would like to customize this behaviour, you can pass a different downloads directory with -o, a different archives directory with -a, or have files remain in the downloads directory with -k.

./archive.py -f channels.txt -o "temporary" -a "My Archives"
./archive.py -f channels.txt -k -o "~/Downloads"

Notifications

The script can optionally notify you using Pushover when it has completed an update. To enable this functionality:

1. Create a Pushover account
2. Generate a Pushover application token
3. Set the PUSHOVER_USER environment variable to your user token
4. set the PUSHOVER_TOKEN environment variable to your new application API token

Now when the script is run, it will send a summary of its work and any errors it encounters. Passing the -n flag to the script will send notifications for each video that is downloaded, rather than one notification for the whole run.

More

All the options that the script supports can be seen by running:

./archive.py --help

For more details on how I run my setup, check out my blog post.

Add-ons

Included along with the script are two additional components that I find useful.

Archives.hazelrules is a pair of rules for the excellent Mac utility Hazel that I use for keeping my Archives directory organized.

oneoffs.sh is a shell script that enables me to automatically download videos that I save to my watch later list in the app Play. When run, it saves the IDs of any videos that are tagged with Download, downloads them, and then removes them from the watch later queue. To interface with Play it relies on two Shortcuts, which you can get here:

• Get Download IDs
• Clear Download IDs You'll need to pass a temporary file path to the script, and configure the first shortcut to use that same path.