datahoarder

I do something very similar. Thanks for sharing cause there's always something to learn by seeing how someone else solved a problem. I'll share mine here too if that's cool:

#!/usr/bin/env bash

music() {
  case $(uname -s) in
    Darwin)
      yt-dlp --format bestaudio --extract-audio --audio-format mp3 \
        --postprocessor-args "-strict experimental" "$1" ;;
    Linux)
      yt-dlp --format bestaudio --extract-audio --audio-format mp3 "$1" ;;
  esac
}

video() {
  yt-dlp \
    --format "bestvideo+bestaudio[ext=m4a]/bestvideo+bestaudio/best" \
    --merge-output-format mp4 \
    -o "%(title)s.%(ext)s" "$1"
}

while getopts ':hm:v:' flag; do
  case "$flag" in
    h) echo "Usage: youtube [-m(usic) <url> | -v(ideo) <url>]" ; exit 0 ;;
    m) music "$OPTARG" ;;
    v) video "$OPTARG" ;;
    *) echo "Invalid argument." >&2 ; exit 1 ;;
  esac
done

You really shouldn't build these from scratch, but use existing time-tested configs, like

https://github.com/TheFrenchGhosty/TheFrenchGhostys-Ultimate-YouTube-DL-Scripts-Collection/

You know you can just put your options in a config file, right? https://github.com/yt-dlp/yt-dlp?tab=readme-ov-file#configuration

I have two depending on what I'm grabbing is part of a playlist (where I want to maintain order) or not:

--download-archive archive.txt --write-auto-subs --sub-langs en --embed-subs -o "%(upload_date)s_%(title)s_%(id)s.%(ext)s" -S res,vcodec:h264,acodec:m4a

or

--download-archive archive.txt --write-auto-subs --sub-langs en --embed-subs -o "%(playlist_title)s/%(playlist_index)s_%(upload_date)s_%(title)s_%(id)s.%(ext)s" -S res,vcodec:h264,acodec:m4a

that --download-archive archive.txt is a godsend for when I rediscover something I've already grabbed. I often move the files to better locations after, but archive.txt doesn't care. Embedding the subtitles, forcing h264/m4a (because more and more things are webp it seems), and renaming the file to the title + youtube ID are what make up the rest.

nice