Download a Music Playlist from YouTube

download.sh

#!/bin/bash

set -Eeuo pipefail

if [ $# -ne 1 ]; then
  echo "Usage: $0 https://youtube.com/playlist?list=..."
  exit 1
fi

PLAYLIST_URL="$1"

declare -a YTDLP_PARAMS=(
    # Select the MP4 Audio format (m4a) to prevent re-encoding
    -f m4a/bestaudio
    # Extract audio track
    -x
    # Embed metadata and thumbnail
    --embed-thumbnail
    --embed-metadata
    # The YouTube chain becomes the artist name
    --parse-metadata "%(uploader|)s:%(artist)s"
    # If the artist name is in the title extract only the song name
    --parse-metadata "title:([^-–.,:：]+ ?[-–.,:：] )?(?P<title>.*)"
    # Cleanup title
    --replace-in-metadata "title" "( -)? ?[([](([Tt]he )?[Oo]fficial.*|OFFICIAL.*|([Cc]lip )?[Oo]fficiel.*|Remastered|[Vv]ideo)[)\]]" ""
    --replace-in-metadata "title" " ?[-|] ?(Officiel|Warner Vault)$" ""
    --replace-in-metadata "title" " ?[([]HD[)\]] ?" ""
    # Remove common prefixes in artist name
    --replace-in-metadata "artist" "^([Oo]fficial) ?" ""
    # Cleanup artist
    --replace-in-metadata "artist" " ?[(]?[Oo]fficial.*[)]?" ""
    # Remove common suffixes in artist name
    --replace-in-metadata "artist" "(officialVEVO|VEVO| ?- Topic|Music|TV| ?Channel)$" ""
    # Add spaces in a CamelCase word (ie. CamelCase => Camel Case)
    --replace-in-metadata "artist" "KoRn" "Korn" # Exception
    --replace-in-metadata "artist" "P!NK" "Pink" # Exception
    --replace-in-metadata "artist" "([a-z])([A-Z]+[a-z])" "\\1 \\2"
    # Sanitize the title and artist name (remove path component separators)
    --replace-in-metadata "title,artist" "[/\\\\]" "-"
    # Output file name
    --output "%(artist)s - %(title)s.%(ext)s"
    # Throttling parameters
    --sleep-requests 1
    --min-sleep-interval 2
    --max-sleep-interval 5
    # Partial download
    #--playlist-start 1
    #--playlist-end 10
)

# Download the complete playlist
yt-dlp "${YTDLP_PARAMS[@]}" --yes-playlist "$PLAYLIST_URL"

PLAYLIST_FILE="playlist.m3u"
PLAYLIST_NAME="YouTube Playlist"
# M3U Header
cat > "$PLAYLIST_FILE" <<EOF
#EXTM3U
#EXTENC:UTF-8
#PLAYLIST:$PLAYLIST_NAME
EOF

# Generate M3U file content
for file in *.m4a; do
  duration="$(mp4info "$file" |sed -r 's/^.* ([0-9]+).[0-9]+ secs.*/\1/;t;d')"
  basename="$(basename "$file" .m4a)"
  cat >> "$PLAYLIST_FILE" <<EOF
#EXTINF:$duration,$basename
$file
EOF
done

# Convert M3U to CRLF line ends to have a wider compatibility
todos "$PLAYLIST_FILE"