Download a YouTube Playlist, extract audio in best quality with youtube-dl and normalize volume with r128gain (ReplayGain 2.0)

getmusic.sh

#!/bin/bash

# helper function parameters: 0 or empty = print only, 1 = rename files!
renamemp3() {
local s="${1:-0}"
for f in *.mp3; do
  if [[ "$f" == *[\"\`]* ]]
    then
      filename=${f%.*}
      filename=${filename//\"}
      filename=${filename//\`}
      extension=${f##*.}
      [ "$s" == 0 ] && echo -e "$f 🠪 ${LIGHTGREEN}$filename${NC}.$extension" && i=$((i+1))
      [ "$s" == 1 ] && mv -v "$f" "$filename.$extension"
  fi
done
}

CYAN='\e[0;36m'
RED='\e[0;31m'
LIGHTGREEN='\e[1;32m'
YELLOW='\e[0;33m'
NC='\e[0m'              # No Color

history=".youtuberc"
defaultkey=""
dateafter=""
noplaylist=0
echo -e "Full URL format: ${CYAN}https://www.youtube.com/playlist?list=${RED}PLAYLIST${NC}"

default=""
echo -ne "YouTube playlist ID (${RED}PLAYLIST${NC} argument only!)"

if [[ -f "$history" ]]
 then
  defaultkey=$(awk '{print $1; exit}' $history)
fi

if [ "$defaultkey" == "playlist" ]
 then
  default=$(awk '{print $2; exit}' $history)
  read -p " [$(echo -e "${YELLOW}$default${NC}")]: " PLAYLIST
  PLAYLIST=${PLAYLIST:-"$default"}
 else
  read -p ": " PLAYLIST
  noplaylist=1
fi

URL="https://www.youtube.com/playlist?list=$PLAYLIST"

if wget -q --method=HEAD "$URL";
 then
  echo -e "[${LIGHTGREEN}OK${NC}] The playlist is reachable!"
 else
  exit 0
fi

if [[ ! -f "$history" ]]
 then
  echo -e "[${YELLOW}NOTE${NC}] $history does not exist."
  echo
  echo -n "Would you like to download the entire playlist? [Y/n] "
  read input1
  case $input1 in
   [yY][eE][sS]|[yY]|"")
    echo -e "[${LIGHTGREEN}OK${NC}] Downloading entire playlist..."
    ;;
   [nN])
    echo -ne "Upload date of the newest item within the playlist (${RED}dateafter${NC})? (eg. YYYYMMDD, now-1week) "
    read dateafter
    dateafter="--dateafter $dateafter"
    echo -e "[${LIGHTGREEN}OK${NC}] Processing a partial download..."
    ;;
  esac
 else
  echo -e "[${LIGHTGREEN}OK${NC}] Updating your download history file with new items only..."
fi

if [ "$dateafter" == "" ]
 then
  youtube-dl "$URL" \
   --quiet \
   --ignore-errors \
   --extract-audio \
   --output '%(playlist_index)s - %(title)s.%(ext)s' \
   --metadata-from-title "%(artist)s - %(title)s" \
   --audio-quality 0 \
   --audio-format mp3 \
   --embed-thumbnail \
   --download-archive $history \
   --exec 'echo && r128gain {} > /dev/null'
 else
  youtube-dl "$URL" \
   --quiet \
   --ignore-errors \
   --extract-audio \
   --output '%(playlist_index)s - %(title)s.%(ext)s' \
   --metadata-from-title "%(artist)s - %(title)s" \
   --audio-quality 0 \
   --audio-format mp3 \
   --embed-thumbnail \
   $dateafter \
   --download-archive $history \
   --exec 'echo && r128gain {} > /dev/null'
fi
   
if [ $noplaylist == 1 ]
  then
    # Add the playlist ID to the first line of the dowload history file.
    sed -i "1s/^/playlist $PLAYLIST\n/" "$history"
fi

echo 
i=0
renamemp3 0
echo 

fsize=$(du -d 0 -Sh . | head -n 1 | tr -s ' ' | cut -d$'\t' -f1)
ls -p *.mp3 | grep -v / | echo -e "There are ${CYAN}$(wc -l)${NC} (${CYAN}${fsize}${NC}) songs available."

if [ "$i" -gt 0 ]
  then
    echo -n "Rename all files automatically? [Y/n] "
    read input1
    case $input1 in
    [yY][eE][sS]|[yY]|"")
        renamemp3 1
        echo
        echo -e "We're all set. Enjoy your music! 😊"
        echo
        ;;
    [nN])
        echo "Renaming files omitted."
        echo "Done."
        ;;
    esac
fi

Great thing about this script is when frequently used for the following scenario:

1. You initially download your playlist within the empty folder by running this script and entering the playlist's ID
2. After some time, you'll probably update your playlist on YouTube with many new items and so you'd like to download the new items.
3. All you have to do is to run the script again and it will remember the playlist's ID by default, so you don't need to copy/paste it again (just hit Enter). It will also download only new items that were not downloaded with the previous run of the script.
4. Unlike other scripts, this script uses r128gain, not mp3gain. With r128gain you could benefit from the new ReplayGain 2.0 (ReplayGain 2.0 targets the same loudness as ReplayGain 1.0 specification, but utilizes the improved ITU BS.1770 algorithm for measuring original tracks. mp3gain is based on ReplayGain 1.0 specification.)

NOTE: Always use one folder for a single playlist i.e. if you have 5 playlists, you should use 5 different folders.