shellTube is a collection of (pretty POSIX) shell scripts to browse YouTube quickly, efficiently, and without the bloat most command-line clients require.

shellTube is written in pure shell; its only dependencies are any modern shell (pdksh, bash, zsh), curl/wget, and vlc/mplayer/kaffeine.

Even these dependencies, though, can easily be changed. If, for example, you don't have vlc, you can just modify a single line and be good-to-go using another player.

Also, shellTube doesn't use the YouTube API at all. This avoids the annoying red-tapey stuff that goes with it -- IDs, registration, quotas -- but has some disadvantages. We'll power through the downsides, though! :)

Usage

shellTube is quite simple to use; this tutorial will go over the usage of the wrapper script, shelltube. When running the script, you'll see a prompt:

 >>

In this prompt you can type any of the following commands:

(short) command syntax	description
(!) about	View the about page.
(cls) clear	Clear the screen.
(dl) download [URL]	Download the selected/specified video.
exit	Exit Shelltube.
(?) help	Display this message.
(md) metadata [URL]	Display selected/specified video's metadata.
(//) playlist URL playlist ID	View playlist based on URL or ID.
(/) search TERM	Perform a search for TERM.
(str) stream [URL]	Stream the selected/specified video.
(sel) video URL video ID	Select video based on URL or ID.
(lq) queue	Lists all videos in the queue.
(sq) squeue	Streams all videos in the queue.
(dq) dqueue	Downloads all videos in the queue.

In [brackets], optional arguments are written.

You can use shellTube in one of two ways (or both):

1. By selecting a video and then doing something with it
2. By doing something and specifying the video

Method A entails using either the video or search command to select a video, which will then be displayed before the prompt like so:

$VIDEO-ID-HERE >>

When a video is selected, you use the download, stream, or metadata commands without arguments to act on the video.

Method B entails just using the download, stream, or metadata commands while using a URL or video ID as an argument.

For example:

 >> download $VIDEO-ID-HERE

You can also download or stream multiple videos in a row without user input with the queue. The queue is a list of video IDs in the file /tmp/yt-queue You can add to the queue manually or while looking at search results, playlist videos, or channel videos by inputting * rather than a number when prompted. This adds every video listed to the queue. When you're back at the main prompt, you can enter lq, dq, or sq to list, download, or stream the queued videos.

You could opt to not use this interactive wrapper script and instead just use the yt-down, yt-search, and yt-metadata scripts on their own.

They are each pretty simple, and you can read their USAGE messages at the top of each script.

Licensing

All of ST is released under the ISC license.

Except for the y-down script, which is released under the GPLv2.