Some more

2026-07-01 20:07:49 +02:00
parent b448e076bc
commit d52ce9508e
4 changed files with 4279 additions and 0 deletions
@@ -0,0 +1,135 @@
+package matcher
+
+import (
+	"regexp"
+	"strings"
+)
+
+const MatchThreshold = 0.7
+
+var (
+	punctuationRe = regexp.MustCompile(`[^\w\s]`)
+	yearRe        = regexp.MustCompile(`\b(19|20)\d{2}\b`)
+	whitespaceRe  = regexp.MustCompile(`\s+`)
+	noiseWords    = map[string]struct{}{
+		"remastered": {}, "remaster": {}, "deluxe": {}, "edition": {},
+		"expanded": {}, "anniversary": {}, "special": {}, "limited": {},
+		"collectors": {}, "collector": {}, "bonus": {}, "tracks": {},
+		"standard": {}, "super": {},
+		"flac": {}, "mp3": {}, "aac": {}, "ogg": {}, "wav": {},
+		"24bit": {}, "16bit": {}, "48khz": {}, "44khz": {}, "96khz": {}, "192khz": {},
+		"cd": {}, "vinyl": {}, "lp": {}, "web": {}, "sacd": {},
+		"box": {}, "set": {}, "japan": {}, "shm": {},
+		"singles": {}, "single": {},
+		"pmedia": {},
+	}
+)
+
+func normalize(s string) string {
+	s = strings.ToLower(s)
+	s = punctuationRe.ReplaceAllString(s, " ")
+	s = yearRe.ReplaceAllString(s, " ")
+	s = whitespaceRe.ReplaceAllString(s, " ")
+	return strings.TrimSpace(s)
+}
+
+func tokenize(s string) []string {
+	var tokens []string
+	for _, word := range strings.Fields(s) {
+		if _, noise := noiseWords[word]; !noise && len(word) > 1 {
+			tokens = append(tokens, word)
+		}
+	}
+	return tokens
+}
+
+func AlbumMatchScore(albumTitle, torrentTitle, artistName string) float64 {
+	normAlbum := normalize(albumTitle)
+	normTorrent := normalize(torrentTitle)
+
+	if artistName != "" {
+		normArtist := normalize(artistName)
+		normTorrent = strings.Replace(normTorrent, normArtist, "", 1)
+		normTorrent = strings.TrimSpace(whitespaceRe.ReplaceAllString(normTorrent, " "))
+	}
+
+	if normAlbum == "" {
+		return 0
+	}
+
+	if strings.Contains(normTorrent, normAlbum) {
+		return 1.0
+	}
+
+	albumTokens := tokenize(normAlbum)
+	if len(albumTokens) == 0 {
+		return 0
+	}
+
+	torrentTokens := make(map[string]struct{}, len(albumTokens))
+	for _, t := range tokenize(normTorrent) {
+		torrentTokens[t] = struct{}{}
+	}
+
+	var matches int
+	for _, token := range albumTokens {
+		if _, found := torrentTokens[token]; found {
+			matches++
+			continue
+		}
+		for tt := range torrentTokens {
+			if levenshteinClose(token, tt) {
+				matches++
+				break
+			}
+		}
+	}
+
+	return float64(matches) / float64(len(albumTokens))
+}
+
+func levenshteinClose(a, b string) bool {
+	if len(a) < 3 || len(b) < 3 {
+		return false
+	}
+	maxDist := 1
+	if len(a) > 6 {
+		maxDist = 2
+	}
+	return levenshtein(a, b) <= maxDist
+}
+
+func levenshtein(a, b string) int {
+	la, lb := len(a), len(b)
+	if la == 0 {
+		return lb
+	}
+	if lb == 0 {
+		return la
+	}
+
+	prev := make([]int, lb+1)
+	curr := make([]int, lb+1)
+
+	for j := 0; j <= lb; j++ {
+		prev[j] = j
+	}
+
+	for i := 1; i <= la; i++ {
+		curr[0] = i
+		for j := 1; j <= lb; j++ {
+			cost := 1
+			if a[i-1] == b[j-1] {
+				cost = 0
+			}
+			curr[j] = min(curr[j-1]+1, min(prev[j]+1, prev[j-1]+cost))
+		}
+		prev, curr = curr, prev
+	}
+
+	return prev[lb]
+}
+
+func IsAlbumMatch(albumTitle, torrentTitle, artistName string) bool {
+	return AlbumMatchScore(albumTitle, torrentTitle, artistName) >= MatchThreshold
+}
@@ -0,0 +1,95 @@
+package matcher
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestAlbumMatchScore(t *testing.T) {
+	tests := []struct {
+		album   string
+		torrent string
+		artist  string
+		wantMin float64
+		wantMax float64
+		match   bool
+	}{
+		{
+			album:   "Ain't My Bitch",
+			torrent: "Metallica - Ain't My Bitch (2025) [FLAC]",
+			artist:  "Metallica",
+			wantMin: 0.9, wantMax: 1.0, match: true,
+		},
+		{
+			album:   "Ain't My Bitch",
+			torrent: "Metallica - Aint My Bitch Remastered 2025 FLAC",
+			artist:  "Metallica",
+			wantMin: 0.9, wantMax: 1.0, match: true,
+		},
+		{
+			album:   "72 Seasons",
+			torrent: "Metallica - 72 Seasons (2023) [24Bit-48kHz] FLAC [PMEDIA] ⭐️",
+			artist:  "Metallica",
+			wantMin: 0.9, wantMax: 1.0, match: true,
+		},
+		{
+			album:   "Kill 'Em All",
+			torrent: "Metallica - Kill Em All [1983] FLAC",
+			artist:  "Metallica",
+			wantMin: 0.9, wantMax: 1.0, match: true,
+		},
+		{
+			album:   "...And Justice for All",
+			torrent: "Metallica - And Justice For All (Remastered) FLAC",
+			artist:  "Metallica",
+			wantMin: 0.7, wantMax: 1.0, match: true,
+		},
+		{
+			album:   "Master of Puppets",
+			torrent: "Metallica - Master of Puppets (Remastered Deluxe Box Set) FLAC",
+			artist:  "Metallica",
+			wantMin: 0.9, wantMax: 1.0, match: true,
+		},
+		{
+			album:   "Ain't My Bitch",
+			torrent: "Metallica - Load 1996 (Japan SHM-CD)",
+			artist:  "Metallica",
+			wantMin: 0.0, wantMax: 0.3, match: false,
+		},
+		{
+			album:   "Ain't My Bitch",
+			torrent: "Metallica - The Memory Remains (Singles, 2026 remastered)",
+			artist:  "Metallica",
+			wantMin: 0.0, wantMax: 0.3, match: false,
+		},
+		{
+			album:   "Ain't My Bitch",
+			torrent: "Metallica - 72 Seasons (2023) [24Bit-48kHz] FLAC",
+			artist:  "Metallica",
+			wantMin: 0.0, wantMax: 0.3, match: false,
+		},
+		{
+			album:   "Hardwired...to Self-Destruct",
+			torrent: "Metallica - Hardwired To Self-Destruct (Deluxe 2016) [FLAC]",
+			artist:  "Metallica",
+			wantMin: 0.7, wantMax: 1.0, match: true,
+		},
+		{
+			album:   "S&M",
+			torrent: "Metallica - S&M (1999) [FLAC]",
+			artist:  "Metallica",
+			wantMin: 0.9, wantMax: 1.0, match: true,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.album+" vs "+tt.torrent, func(t *testing.T) {
+			score := AlbumMatchScore(tt.album, tt.torrent, tt.artist)
+			assert.GreaterOrEqual(t, score, tt.wantMin, "score too low: %.2f", score)
+			assert.LessOrEqual(t, score, tt.wantMax, "score too high: %.2f", score)
+			assert.Equal(t, tt.match, IsAlbumMatch(tt.album, tt.torrent, tt.artist),
+				"match mismatch for score %.2f", score)
+		})
+	}
+}