Varaverk/Media/lidarr_cleanup.sh

#!/bin/bash
# -----------------------------------------------------------------------------------------------
# --------------------------------- Lidarr Cleanup Script --------------------------------------
# -----------------------------------------------------------------------------------------------
# Removes orphaned music files from the library that Lidarr no longer tracks.
# Uses the Lidarr API to build a complete list of tracked file paths then compares
# against what exists on disk — anything not tracked and older than LIDARR_ORPHAN_AGE
# days is considered an orphan and deleted.
#
# File classification:
#   TRACKED   — Lidarr API knows about this exact file path → leave it alone
#   PROTECTED — matches LIDARR_PROTECTED_PATTERNS → never delete (cover art, .nfo etc.)
#   ORPHAN    — music file, not tracked, older than LIDARR_ORPHAN_AGE days → delete
#   JUNK      — not a music extension, not protected → delete regardless of age
#   RECENT    — not tracked, under LIDARR_ORPHAN_AGE days old → skip (may be mid-import)
#
# Why protected patterns matter:
#   Lidarr generates cover art (*.jpg), metadata (*.nfo) and lyrics (*.lrc) but does
#   not include these in its tracked file API response. Without protection these would
#   be classified as orphans and deleted — breaking Lidarr and Emby metadata display.
#
# Lidarr runs on HOST1 only — music library is HOST1's source of truth.
# If run on HOST2 this script exits cleanly with no action.
# All configuration in Master.conf under Arr Cleanup section.
# Supports --dry-run to preview what would be deleted without making changes.
# -----------------------------------------------------------------------------------------------

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"

source "$SCRIPT_DIR/../Master.conf"
source "$SCRIPT_DIR/../common.sh"

parse_args "$@"

# -----------------------------------------------------------------------------------------------
# ━━━ $ICON_GEAR Setup ━━━
# -----------------------------------------------------------------------------------------------
echo ""
echo "━━━ $ICON_GEAR Setup ━━━"

if [[ "$EUID" -ne 0 ]]; then
    error "Must be run as root"
    exit 1
fi

success "Running as root"

if ! command -v curl >/dev/null 2>&1; then
    error "curl not found — required for Lidarr API calls"
    exit 1
fi

if ! command -v jq >/dev/null 2>&1; then
    error "jq not found — required for JSON parsing"
    notify "Lidarr cleanup failed on $(hostname) — jq not installed" "Lidarr Cleanup" "warning"
    exit 1
fi

# Lidarr runs on HOST1 only — exit cleanly if running on HOST2
detect_hosts

if [[ "$LOCAL_SERVER_NAME" != "$HOST1" ]]; then
    info "Lidarr runs on $HOST1 only — skipping on $LOCAL_SERVER_NAME"
    exit 0
fi

LIDARR_URL="$HOST1_LIDARR_URL"
LIDARR_API_KEY="$HOST1_LIDARR_API_KEY"
LIDARR_MUSIC_ROOT="$HOST1_LIDARR_MUSIC_ROOT"

info "Lidarr instance: $LOCAL_SERVER_NAME → $LIDARR_URL"
info "Music root: $LIDARR_MUSIC_ROOT"

require_var LIDARR_URL
require_var LIDARR_API_KEY
require_var LIDARR_MUSIC_ROOT

if [[ ! -d "$LIDARR_MUSIC_ROOT" ]]; then
    error "Music root not found: $LIDARR_MUSIC_ROOT"
    notify "Lidarr cleanup failed on $(hostname) — music root not found: $LIDARR_MUSIC_ROOT" "Lidarr Cleanup" "warning"
    exit 1
fi

acquire_lock "wait"

success "Config validated"

# -----------------------------------------------------------------------------------------------
# ━━━ $ICON_SUMMARY Status ━━━
# -----------------------------------------------------------------------------------------------
if [[ "$SHOW_STATUS" == true ]]; then
    echo ""
    echo "━━━━━ $ICON_SUMMARY STATUS ━━━━━"
    echo "$ICON_GEAR  Lidarr URL:         $LIDARR_URL"
    echo "$ICON_GEAR  Music root:         $LIDARR_MUSIC_ROOT"
    echo "$ICON_TIME  Orphan age:         ${LIDARR_ORPHAN_AGE} days"
    echo "$ICON_GEAR  Extensions:         ${LIDARR_EXTENSIONS[*]}"
    echo "$ICON_GEAR  Protected patterns: ${LIDARR_PROTECTED_PATTERNS[*]}"
    echo "$ICON_GEAR  Dry Run:            $DRY_RUN"
    echo "━━━━━━━━━━━━━━━━━━━━━━━"
    exit 0
fi

[[ "$DRY_RUN" == true ]] && warn "DRY RUN — no files will be deleted"

# -----------------------------------------------------------------------------------------------
# HELPERS
# -----------------------------------------------------------------------------------------------

# Makes a GET request to the Lidarr API — exits on connection failure
lidarr_api() {
    local endpoint="$1"
    local response http_code body

    response=$(curl -sf \
        --max-time 30 \
        -H "X-Api-Key: $LIDARR_API_KEY" \
        -w "\n%{http_code}" \
        "${LIDARR_URL}/api/v1/${endpoint}" 2>/dev/null)

    http_code=$(echo "$response" | tail -1)
    body=$(echo "$response" | head -n -1)

    if [[ "$http_code" != "200" ]]; then
        error "Lidarr API returned HTTP $http_code for endpoint: $endpoint"
        return 1
    fi

    echo "$body"
}

# Returns 0 if file extension matches LIDARR_EXTENSIONS list
is_music_file() {
    local ext="${1##*.}"
    ext="${ext,,}"
    for valid_ext in "${LIDARR_EXTENSIONS[@]}"; do
        [[ "$ext" == "$valid_ext" ]] && return 0
    done
    return 1
}

# Returns 0 if filename matches any pattern in LIDARR_PROTECTED_PATTERNS
is_protected_file() {
    local filename
    filename=$(basename "$1")
    for pattern in "${LIDARR_PROTECTED_PATTERNS[@]}"; do
        # shellcheck disable=SC2254
        case "$filename" in
            $pattern) return 0 ;;
        esac
    done
    return 1
}

# -----------------------------------------------------------------------------------------------
# ━━━ $ICON_SYNC Fetching Lidarr Tracked Files ━━━
# -----------------------------------------------------------------------------------------------
echo ""
echo "━━━ $ICON_SYNC Fetching Lidarr Tracked Files ━━━"

check_api "$LIDARR_URL" "Lidarr" || {
    notify "Lidarr cleanup aborted on $(hostname) — API unreachable" "Lidarr Cleanup" "warning"
    exit 1
}

info "Querying Lidarr API: $LIDARR_URL"

# Lidarr requires artistId parameter for trackFile endpoint
# Step 1: Get all artist IDs
# Step 2: For each artist, get their track files
ARTIST_RESPONSE=$(lidarr_api "artist") || {
    error "Failed to fetch artists from Lidarr — check URL and API key"
    notify "Lidarr cleanup failed on $(hostname) — could not fetch artists" "Lidarr Cleanup" "warning"
    exit 1
}

ARTIST_IDS=$(echo "$ARTIST_RESPONSE" | jq -r '.[].id' 2>/dev/null)
ARTIST_COUNT=$(echo "$ARTIST_IDS" | grep -c "[0-9]" 2>/dev/null || echo 0)

if [[ "$ARTIST_COUNT" -eq 0 ]]; then
    warn "No artists found in Lidarr — library may be empty"
    exit 0
fi

info "Found $ARTIST_COUNT artists — fetching track files..."

TMP_DIR="/tmp/lidarr_cleanup_$$"
mkdir -p "$TMP_DIR"
trap "rm -rf $TMP_DIR" EXIT

TRACKED_FILE="$TMP_DIR/tracked_paths.txt"
> "$TRACKED_FILE"

while IFS= read -r artist_id; do
    [[ -z "$artist_id" ]] && continue
    ARTIST_TRACKS=$(lidarr_api "trackFile?artistId=${artist_id}" 2>/dev/null)
    if [[ -n "$ARTIST_TRACKS" ]]; then
        echo "$ARTIST_TRACKS" | jq -r '.[].path' 2>/dev/null >> "$TRACKED_FILE"
    fi
done <<< "$ARTIST_IDS"

sort -u "$TRACKED_FILE" -o "$TRACKED_FILE"

TRACKED_COUNT=$(wc -l < "$TRACKED_FILE")
success "Lidarr tracks $TRACKED_COUNT files across $ARTIST_COUNT artists"

if [[ "$TRACKED_COUNT" -eq 0 ]]; then
    warn "No tracked files returned — Lidarr may not have scanned yet or library is empty"
    warn "Aborting to prevent mass deletion"
    notify "Lidarr cleanup aborted on $(hostname) — no tracked files returned from API" "Lidarr Cleanup" "warning"
    exit 1
fi

# -----------------------------------------------------------------------------------------------
# ━━━ $ICON_CLEAN Scanning Music Root ━━━
# -----------------------------------------------------------------------------------------------
echo ""
echo "━━━ $ICON_CLEAN Scanning Music Root ━━━"
info "Root:              $LIDARR_MUSIC_ROOT"
info "Orphan age:        ${LIDARR_ORPHAN_AGE} days"
info "Protected:         ${LIDARR_PROTECTED_PATTERNS[*]}"
echo ""

START=$(date +%s)
ORPHAN_COUNT=0
JUNK_COUNT=0
RECENT_COUNT=0
PROTECTED_COUNT=0
ORPHAN_BYTES=0
JUNK_BYTES=0

AGE_SECONDS=$(( LIDARR_ORPHAN_AGE * 86400 ))
NOW=$(date +%s)

while IFS= read -r filepath; do
    [[ -z "$filepath" ]] && continue

    # TRACKED — Lidarr knows about this file
    if grep -qF "$filepath" "$TRACKED_FILE" 2>/dev/null; then
        log "TRACKED: $filepath"
        continue
    fi

    # PROTECTED — never delete regardless of tracking status
    if is_protected_file "$filepath"; then
        log "PROTECTED: $filepath"
        ((PROTECTED_COUNT++))
        continue
    fi

    FILE_SIZE=$(stat -c%s "$filepath" 2>/dev/null || echo 0)

    if is_music_file "$filepath"; then
        # Music file not tracked — check age
        FILE_MTIME=$(stat -c %Y "$filepath" 2>/dev/null || echo 0)
        FILE_AGE=$(( NOW - FILE_MTIME ))

        if [[ "$FILE_AGE" -lt "$AGE_SECONDS" ]]; then
            log "RECENT (skipping): $filepath"
            ((RECENT_COUNT++))
            continue
        fi

        # ORPHAN — old enough to delete
        warn "$ICON_TRASH ORPHAN: $filepath"
        if [[ "$DRY_RUN" == false ]]; then
            rm -f "$filepath" && {
                ((ORPHAN_COUNT++))
                ORPHAN_BYTES=$((ORPHAN_BYTES + FILE_SIZE))
            } || error "Failed to delete: $filepath"
        else
            ((ORPHAN_COUNT++))
            ORPHAN_BYTES=$((ORPHAN_BYTES + FILE_SIZE))
        fi
    else
        # JUNK — not a music extension, not protected
        log "JUNK: $filepath"
        if [[ "$DRY_RUN" == false ]]; then
            rm -f "$filepath" && {
                ((JUNK_COUNT++))
                JUNK_BYTES=$((JUNK_BYTES + FILE_SIZE))
            } || error "Failed to delete: $filepath"
        else
            ((JUNK_COUNT++))
            JUNK_BYTES=$((JUNK_BYTES + FILE_SIZE))
        fi
    fi

done < <(find "$LIDARR_MUSIC_ROOT" -type f 2>/dev/null)

# Empty folder cleanup
if [[ "$DRY_RUN" == false ]]; then
    echo ""
    info "Cleaning up empty folders..."
    find "$LIDARR_MUSIC_ROOT" -mindepth 1 -type d -empty -delete 2>/dev/null
    success "Empty folders removed"
fi

END=$(date +%s)

# Format bytes helper
format_bytes() {
    local bytes=$1
    if (( bytes > 1073741824 )); then
        awk "BEGIN {printf \"%.1fGB\", $bytes / 1073741824}"
    elif (( bytes > 1048576 )); then
        awk "BEGIN {printf \"%.1fMB\", $bytes / 1048576}"
    else
        echo "${bytes}B"
    fi
}

ORPHAN_HUMAN=$(format_bytes $ORPHAN_BYTES)
JUNK_HUMAN=$(format_bytes $JUNK_BYTES)
TOTAL_REMOVED=$(( ORPHAN_COUNT + JUNK_COUNT ))

# -----------------------------------------------------------------------------------------------
# ━━━ $ICON_SUMMARY Summary ━━━
# -----------------------------------------------------------------------------------------------
echo ""
echo "━━━━━ $ICON_SUMMARY LIDARR CLEANUP SUMMARY ━━━━━"
echo "$ICON_SYNC   Tracked by Lidarr: $TRACKED_COUNT files"
echo "$ICON_SHIELD Protected:         $PROTECTED_COUNT files (cover art, metadata)"
echo "$ICON_TRASH  Orphans removed:   $ORPHAN_COUNT files ($ORPHAN_HUMAN)"
echo "$ICON_TRASH  Junk removed:      $JUNK_COUNT files ($JUNK_HUMAN)"
echo "$ICON_TIME   Recent skipped:    $RECENT_COUNT files (under ${LIDARR_ORPHAN_AGE} days)"
echo "$ICON_TIME   Duration:          $(format_duration $((END - START)))"
echo ""
if [[ "$DRY_RUN" == true ]]; then
    echo "$ICON_WARN  Status: DRY RUN — no files deleted"
elif [[ "$TOTAL_REMOVED" -eq 0 ]]; then
    echo "$ICON_DONE  Status: $ICON_SUCCESS CLEAN — nothing to remove"
    notify "Lidarr cleanup complete on $(hostname) — library is clean" "Lidarr Cleanup" "normal"
else
    echo "$ICON_DONE  Status: $ICON_SUCCESS DONE — $TOTAL_REMOVED files removed"
    notify "Lidarr cleanup on $(hostname) — removed $TOTAL_REMOVED files (orphans: $ORPHAN_HUMAN junk: $JUNK_HUMAN)" "Lidarr Cleanup" "normal"
fi
echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"