cover_handler.py

@@ -48,6 +48,7 @@ def _image_ok(path: Path) -> bool:
 
 def find_local_cover(image_files: List[Path]) -> Optional[Path]:
     priority = ("folder", "front", "cover", "album")
+    # Sort by priority keyword, then size descending
     def key(p: Path):
         name = p.name.lower()
         score = next((i for i, kw in enumerate(priority) if kw in name), len(priority))
@@ -60,41 +61,6 @@ def find_local_cover(image_files: List[Path]) -> Optional[Path]:
     return None
 
 
-def normalize_cover_to_folder_jpg(cover_path: Path) -> Path:
-    """
-    Stellt sicher dass das Cover als folder.jpg (JPEG) im Album-Verzeichnis liegt.
-    - Ist es bereits folder.jpg → unverändert zurückgeben.
-    - Ist es eine andere JPEG → umbenennen.
-    - Ist es WebP oder PNG → zu JPEG konvertieren, Original löschen.
-    Gibt den Pfad zur folder.jpg zurück.
-    """
-    dest = cover_path.parent / "folder.jpg"
-    if cover_path.resolve() == dest.resolve():
-        return dest
-
-    suffix = cover_path.suffix.lower()
-    try:
-        if suffix in (".jpg", ".jpeg"):
-            cover_path.rename(dest)
-        elif HAS_PIL:
-            import io
-            with cover_path.open("rb") as f:
-                raw = f.read()
-            with Image.open(io.BytesIO(raw)) as img:
-                buf = io.BytesIO()
-                img.convert("RGB").save(buf, format="JPEG", quality=92)
-                dest.write_bytes(buf.getvalue())
-            cover_path.unlink()
-        else:
-            # PIL nicht verfügbar: einfach umbenennen, auch wenn es kein JPEG ist
-            cover_path.rename(dest)
-        print(f"  🖼️  Cover normalisiert → folder.jpg ({cover_path.name})")
-    except Exception as e:
-        print(f"  ⚠️ Cover-Normalisierung fehlgeschlagen: {e}", file=sys.stderr)
-        return cover_path
-    return dest
-
-
 def _mb_cover_url(release_mbid: str) -> Optional[str]:
     url = f"https://coverartarchive.org/release/{release_mbid}/front"
     if not HAS_REQUESTS:
@@ -221,72 +187,14 @@ def embed_cover(audio_path: Path, cover_path: Path) -> bool:
     return False
 
 
-def _discogs_cover_url(artist: Optional[str], album: Optional[str]) -> Optional[str]:
-    """Sucht auf Discogs nach artist+album und gibt die primäre Image-URL zurück."""
-    if not HAS_REQUESTS or not artist or not album:
-        return None
-    import os
-    token = os.getenv("DISCOGS_TOKEN", "")
-    headers = {"User-Agent": "MusicMetadataEnricher/1.0"}
-    if token:
-        headers["Authorization"] = f"Discogs token={token}"
-    try:
-        r = requests.get(
-            "https://api.discogs.com/database/search",
-            params={"artist": artist, "release_title": album, "type": "release", "per_page": 3},
-            headers=headers,
-            timeout=10,
-        )
-        if r.status_code != 200:
-            return None
-        results = r.json().get("results", [])
-        for result in results:
-            cover = result.get("cover_image") or result.get("thumb")
-            if cover and "spacer" not in cover:
-                return cover
-    except Exception as e:
-        print(f"  ⚠️ Discogs-Suchfehler: {e}", file=sys.stderr)
-    return None
-
-
-def download_discogs_cover(artist: Optional[str], album: Optional[str], dest_dir: Path) -> Optional[Path]:
-    url = _discogs_cover_url(artist, album)
-    if not url:
-        return None
-    dest = dest_dir / "folder.jpg"
-    try:
-        r = requests.get(url, timeout=15, headers={"User-Agent": "MusicMetadataEnricher/1.0"})
-        if r.status_code != 200:
-            return None
-        ct = r.headers.get("content-type", "")
-        if ("png" in ct or url.lower().endswith(".png")) and HAS_PIL:
-            import io
-            with Image.open(io.BytesIO(r.content)) as img:
-                buf = io.BytesIO()
-                img.convert("RGB").save(buf, format="JPEG", quality=92)
-                dest.write_bytes(buf.getvalue())
-        else:
-            dest.write_bytes(r.content)
-        if _image_ok(dest):
-            return dest
-        dest.unlink(missing_ok=True)
-    except Exception as e:
-        print(f"  ⚠️ Discogs-Cover-Fehler: {e}", file=sys.stderr)
-        dest.unlink(missing_ok=True)
-    return None
-
-
 def resolve_cover(
     image_files: List[Path],
     release_mbid: Optional[str],
     album_dir: Path,
-    artist: Optional[str] = None,
-    album: Optional[str] = None,
 ) -> tuple[Optional[Path], Optional[str]]:
     """Returns (cover_path, source_label)."""
     local = find_local_cover(image_files)
     if local:
-        local = normalize_cover_to_folder_jpg(local)
         return local, "local"
 
     if release_mbid:
@@ -294,9 +202,4 @@ def resolve_cover(
         if downloaded:
             return downloaded, "musicbrainz"
 
-    if artist or album:
-        downloaded = download_discogs_cover(artist, album, album_dir)
-        if downloaded:
-            return downloaded, "discogs"
-
     return None, None

executor.py

@@ -42,52 +42,21 @@ def _safe_name(s: str) -> str:
     return re.sub(r"\s+", "_", s).strip("._-")
 
 
-_CLASSICAL_GENRE_KEYWORDS = {
-    "classical", "klassik", "baroque", "barock", "romantic", "romantik",
-    "opera", "oper", "operetta", "operette", "chamber", "kammermusik",
-    "symphon", "concerto", "oratorio", "sacred", "kirchenmusik",
-    "renaissance", "medieval", "contemporary classical",
-}
-
-_CLASSICAL_COMPOSER_KEYWORDS = {
-    # Bekannte Komponisten als Signal (Nachname reicht)
-    "bach", "beethoven", "mozart", "handel", "haydn", "schubert", "brahms",
-    "chopin", "liszt", "schumann", "wagner", "verdi", "puccini", "vivaldi",
-    "telemann", "buxtehude", "monteverdi", "palestrina", "purcell",
-    "mahler", "bruckner", "dvorak", "tchaikovsky", "tschaikowski",
-    "debussy", "ravel", "satie", "strauss", "sibelius", "grieg",
-}
-
-
 def _is_classical(albumartist: str, track_artist: str, genre: str) -> bool:
     """
-    Klassik-Schema (Performer_-_Komponist_-_Werk) wird angewendet wenn:
+    Classical schema applies when performer (albumartist) ≠ composer (track_artist),
-    1. Genre explizit klassisch ist, ODER
+    which covers both 'real' classical music and jazz-on-classical-themes albums.
-    2. track_artist ist ein bekannter Komponist (und ≠ albumartist), ODER
+    Genre keyword matching is used as additional signal but not required.
-    3. albumartist ≠ track_artist UND beide sind bekannte Komponistennamen.
-    Reine Performer≠Komponist-Heuristik ohne Genre-Bestätigung ist abgeschaltet
-    (zu viele Falschpositive bei Samplern, Jazz, Volksmusik).
     """
     aa = (albumartist or "").casefold().strip()
     ta = (track_artist or "").casefold().strip()
-    g = (genre or "").casefold().strip()
-
     if not aa or aa in ("various artists", "unknown artist", "unknown"):
         return False
     if not ta or ta in ("unknown artist", "unknown"):
-        return False
+        return False  # placeholder, not a real composer
     if aa == ta:
         return False
-
+    return True  # performer ≠ composer → classical naming
-    # Primäres Signal: Genre-Keyword
-    if any(kw in g for kw in _CLASSICAL_GENRE_KEYWORDS):
-        return True
-
-    # Sekundäres Signal: track_artist enthält bekannten Komponistennamen
-    if any(kw in ta for kw in _CLASSICAL_COMPOSER_KEYWORDS):
-        return True
-
-    return False
 
 
 def _proposed_filename(

metadata_resolver.py

@@ -33,63 +33,6 @@ try:
 except ImportError:
     HAS_ANTHROPIC = False
 
-# ---------------------------------------------------------------------------
-# Genre normalization
-# ---------------------------------------------------------------------------
-
-_GENRE_MAP: Dict[str, str] = {
-    # Deutsch → Englisch (Jellyfin-Standardbegriffe)
-    "volksmusik": "Folk",
-    "volkslieder": "Folk",
-    "volkslied": "Folk",
-    "heimatlieder": "Folk",
-    "schlager": "Schlager",
-    "deutsche schlager": "Schlager",
-    "marsch": "March",
-    "marschmusik": "March",
-    "militaermusik": "March",
-    "militärmusik": "March",
-    "kirchenmusik": "Sacred",
-    "chormusik": "Choral",
-    "kinderlieder": "Children",
-    "weihnachtslieder": "Christmas",
-    "weihnachtsmusik": "Christmas",
-    "blasmusik": "Brass Band",
-    "operette": "Operetta",
-    "oper": "Opera",
-    "kammermusik": "Chamber Music",
-    "klassik": "Classical",
-    "classic": "Classical",
-    "klassische musik": "Classical",
-    "barock": "Baroque",
-    "romantik": "Romantic",
-    # Englische Varianten vereinheitlichen
-    "rhythm and blues": "R&B",
-    "rhythmic soul": "R&B",
-    "rock and roll": "Rock 'n' Roll",
-    "rock & roll": "Rock 'n' Roll",
-    "easy listening": "Easy Listening",
-    "vocal pop": "Pop",
-    "adult contemporary": "Pop",
-    "big band": "Big Band",
-    "swing music": "Swing",
-    "latin jazz": "Latin Jazz",
-    "bossa nova": "Bossa Nova",
-    "nueva cancion": "Nueva Canción",
-}
-
-
-def normalize_genre(genre: Optional[str]) -> Optional[str]:
-    if not genre:
-        return genre
-    key = genre.strip().lower()
-    normalized = _GENRE_MAP.get(key)
-    if normalized:
-        return normalized
-    # Titlcase wenn nicht in der Map (verhindert ALL CAPS oder all lowercase)
-    return genre.strip().title() if genre == genre.upper() or genre == genre.lower() else genre.strip()
-
-
 _MB_RATE_LIMIT = 1.1   # seconds between MusicBrainz requests
 _last_mb_call = 0.0
 ACOUSTID_API_KEY  = os.getenv("ACOUSTID_API_KEY", "")
@@ -562,7 +505,7 @@ def resolve(
         album=album,
         albumartist=albumartist,
         date=year,
-        genre=normalize_genre(genre),
+        genre=genre,
         label=label,
         mbid=release_mbid,
         cover_path=None,

music_enricher.py

@@ -148,8 +148,6 @@ def process_album(
             hints.cover_images,
             proposal.mbid,
             album_dir,
-            artist=proposal.albumartist,
-            album=proposal.album,
         )
         if cover_path and not args.no_cover:
             proposal.cover_path = cover_path
@@ -230,71 +228,6 @@ def process_album(
     return stats
 
 
-def _print_status(args: argparse.Namespace) -> None:
-    """Scannt die Bibliothek und zeigt Alben mit fehlenden/schlechten Metadaten."""
-    from mutagen import File as MutagenFile
-
-    IMAGE_EXTS = {".jpg", ".jpeg", ".png", ".webp"}
-    AUDIO_EXTS = {".mp3", ".flac", ".m4a", ".wav", ".ogg", ".opus"}
-
-    album_dirs: List[Path] = []
-    if args.album:
-        album_dirs.append(args.album.expanduser().resolve())
-    for raw in args.paths:
-        root = Path(raw).expanduser().resolve()
-        if root.is_dir():
-            album_dirs.extend(collect_album_dirs(root))
-
-    no_cover, bad_tags, ok = [], [], []
-
-    for album_dir in sorted(album_dirs):
-        has_cover = any(
-            f.suffix.lower() in IMAGE_EXTS
-            for f in album_dir.rglob("*") if f.is_file()
-        )
-        audio_files = [
-            f for f in sorted(album_dir.rglob("*"))
-            if f.is_file() and f.suffix.lower() in AUDIO_EXTS
-        ]
-        missing_tags = []
-        for af in audio_files[:3]:  # nur erste 3 prüfen (schnell)
-            try:
-                tags = MutagenFile(str(af), easy=True)
-                if tags is None:
-                    missing_tags.append(af.name)
-                    continue
-                title = (tags.get("title") or [""])[0].strip()
-                artist = (tags.get("artist") or [""])[0].strip()
-                if not title or title.lower() in ("unknown", "audiotrack", "") \
-                        or not artist or artist.lower() in ("unknown", ""):
-                    missing_tags.append(af.name)
-            except Exception:
-                missing_tags.append(af.name)
-
-        problems = []
-        if not has_cover:
-            problems.append("kein Cover")
-        if missing_tags:
-            problems.append(f"schlechte Tags ({len(missing_tags)}/{min(3,len(audio_files))} geprüft)")
-
-        if problems:
-            bad_tags.append((album_dir, problems))
-        else:
-            ok.append(album_dir)
-
-    print(f"\n{'=' * 60}")
-    print(f"📊 Bibliotheksstatus — {len(album_dirs)} Alben")
-    print(f"{'=' * 60}")
-    print(f"  ✅ In Ordnung:   {len(ok)}")
-    print(f"  ⚠️  Mit Problemen: {len(bad_tags)}")
-    print()
-    for album_dir, problems in bad_tags:
-        print(f"  💿 {album_dir.name}")
-        for p in problems:
-            print(f"       → {p}")
-    print("=" * 60)
-
-
 def main() -> None:
     parser = argparse.ArgumentParser(
         description="KI-gestützter Musik-Metadaten-Enricher für Jellyfin",
@@ -329,17 +262,9 @@ def main() -> None:
     parser.add_argument("--playlist-generator", type=Path, dest="playlist_generator",
                         help="Pfad zu jellyfin_playlist_generator.py\n"
                              "(Standard: ../Jellyfin_Playlist_Generator/jellyfin_playlist_generator.py)")
-    parser.add_argument("--status", action="store_true",
-                        help="Bibliotheksstatus anzeigen (fehlende Cover, schlechte Tags) — nichts schreiben")
 
     args = parser.parse_args()
 
-    if args.status:
-        if not args.paths and not args.album:
-            parser.error("--status benötigt mindestens einen Pfad.")
-        _print_status(args)
-        return
-
     if not args.album and not args.paths:
         parser.error("Mindestens ein Pfad oder --album erforderlich.")
 

test_suite_enricher.py

@@ -188,140 +188,6 @@ def test_extract_hints_multi_disc() -> str:
     return f"disc numbers detected: {disc_nums}"
 
 
-# ---------------------------------------------------------------------------
-# Vertical tracklist parser Tests
-# ---------------------------------------------------------------------------
-
-def test_vertical_tracklist_basic() -> str:
-    from hint_extractor import _normalize_vertical_tracklist
-    text = "1\nKatka dovádí\n3:22\n2\nZáludná\n2:15\n3\nPolka pro trubku\n4:01"
-    result = _normalize_vertical_tracklist(text)
-    assert result is not None, "should recognize vertical format"
-    assert "1. Katka" in result, f"got: {result!r}"
-    assert "2. Záludná" in result, f"got: {result!r}"
-    return f"normalized: {result[:60]!r}"
-
-
-def test_vertical_tracklist_without_duration() -> str:
-    from hint_extractor import _normalize_vertical_tracklist
-    text = "1\nFirst Song\n2\nSecond Song\n3\nThird Song"
-    result = _normalize_vertical_tracklist(text)
-    assert result is not None, "should work without durations"
-    assert "1. First Song" in result, f"got: {result!r}"
-    return f"no-duration OK: {result[:60]!r}"
-
-
-def test_vertical_tracklist_not_triggered_for_normal() -> str:
-    from hint_extractor import _normalize_vertical_tracklist
-    text = "1. Dancing Queen\n2. Waterloo\n3. Fernando"
-    result = _normalize_vertical_tracklist(text)
-    assert result is None, f"should return None for normal format, got: {result!r}"
-    return "correctly returns None for standard format"
-
-
-# ---------------------------------------------------------------------------
-# Single-CD disc handling Tests
-# ---------------------------------------------------------------------------
-
-def test_single_cd_tracklist_match() -> str:
-    """Track-Nummer-Match darf nicht disc_num erfordern (Single-CD hat disc=None)."""
-    from hint_extractor import _parse_tracklist
-    from models import TrackHints, AlbumHints, AlbumScan
-    from pathlib import Path
-    import tempfile
-
-    with tempfile.TemporaryDirectory() as tmpdir:
-        root = Path(tmpdir) / "Tufaranka_-_Katka_dovadi"
-        root.mkdir()
-        (root / "01_-_Tufaranka_-_AudioTrack_01.mp3").write_bytes(b"\x00" * 100)
-        (root / "tracklist.txt").write_text("1\nKatka dovádí\n3:22\n2\nZáludná\n2:15\n3\nPolka\n4:01")
-
-        from scanner import scan_album
-        from hint_extractor import extract_hints
-        scan = scan_album(root)
-        hints = extract_hints(scan, use_ocr=False)
-        track = hints.tracks[0]
-        assert track.title == "Katka dovádí", f"expected tracklist title, got: {track.title!r}"
-    return f"single-CD match OK: title={track.title!r}"
-
-
-# ---------------------------------------------------------------------------
-# Genre normalization Tests
-# ---------------------------------------------------------------------------
-
-def test_genre_normalize_german() -> str:
-    from metadata_resolver import normalize_genre
-    assert normalize_genre("volksmusik") == "Folk", "volksmusik → Folk"
-    assert normalize_genre("klassik") == "Classical", "klassik → Classical"
-    assert normalize_genre("marschmusik") == "March", "marschmusik → March"
-    return "German genres normalized correctly"
-
-
-def test_genre_normalize_english_variants() -> str:
-    from metadata_resolver import normalize_genre
-    assert normalize_genre("rhythm and blues") == "R&B"
-    assert normalize_genre("rock and roll") == "Rock 'n' Roll"
-    return "English variants normalized correctly"
-
-
-def test_genre_normalize_titlecase() -> str:
-    from metadata_resolver import normalize_genre
-    assert normalize_genre("JAZZ") == "Jazz", f"got: {normalize_genre('JAZZ')!r}"
-    assert normalize_genre("folk") == "Folk", f"got: {normalize_genre('folk')!r}"
-    assert normalize_genre("Big Band") == "Big Band"  # unchanged
-    return "Titlecase normalization OK"
-
-
-# ---------------------------------------------------------------------------
-# _is_classical() Tests
-# ---------------------------------------------------------------------------
-
-def test_is_classical_by_genre() -> str:
-    from executor import _is_classical
-    assert _is_classical("Gardiner", "Bach", "Classical"), "Classical genre should trigger"
-    assert _is_classical("Herreweghe", "Handel", "Baroque"), "Baroque should trigger"
-    return "genre-based detection OK"
-
-
-def test_is_classical_by_composer() -> str:
-    from executor import _is_classical
-    assert _is_classical("Gardiner", "Bach", ""), "Bach as track_artist should trigger"
-    assert _is_classical("Hurford", "beethoven", ""), "beethoven should trigger"
-    return "composer-name detection OK"
-
-
-def test_is_classical_false_for_pop() -> str:
-    from executor import _is_classical
-    assert not _is_classical("Trini Lopez", "Trini Lopez", "Pop"), "same artist = not classical"
-    assert not _is_classical("ABBA", "ABBA", "Pop"), "ABBA is not classical"
-    assert not _is_classical("Trini Lopez", "", "R&B"), "empty track_artist = not classical"
-    return "pop albums correctly not classical"
-
-
-def test_is_classical_false_for_folk() -> str:
-    from executor import _is_classical
-    assert not _is_classical("Tufaranka", "Tufaranka", "Folk"), "Folk is not classical"
-    return "Folk correctly not classical"
-
-
-# ---------------------------------------------------------------------------
-# cover normalize Tests
-# ---------------------------------------------------------------------------
-
-def test_normalize_cover_renames_front_jpg() -> str:
-    from cover_handler import normalize_cover_to_folder_jpg
-    import tempfile, shutil
-    with tempfile.TemporaryDirectory() as tmpdir:
-        root = Path(tmpdir)
-        front = root / "Front.jpg"
-        front.write_bytes(b"\xff\xd8" + b"\x00" * 200)
-        result = normalize_cover_to_folder_jpg(front)
-        assert result.name == "folder.jpg", f"expected folder.jpg, got {result.name!r}"
-        assert (root / "folder.jpg").exists(), "folder.jpg should exist"
-        assert not front.exists(), "Front.jpg should be gone"
-    return "Front.jpg → folder.jpg rename OK"
-
-
 # ---------------------------------------------------------------------------
 # executor Tests
 # ---------------------------------------------------------------------------
@@ -386,19 +252,6 @@ def main() -> None:
         ("UNIT_15_proposed_filename_single_disc",     test_proposed_filename_single_disc),
         ("UNIT_16_proposed_filename_multi_disc",      test_proposed_filename_multi_disc),
         ("UNIT_17_proposed_filename_sanitizes_chars", test_proposed_filename_sanitizes_chars),
-        # Neue Tests
-        ("UNIT_18_vertical_tracklist_basic",          test_vertical_tracklist_basic),
-        ("UNIT_19_vertical_tracklist_no_duration",    test_vertical_tracklist_without_duration),
-        ("UNIT_20_vertical_tracklist_no_false_pos",   test_vertical_tracklist_not_triggered_for_normal),
-        ("UNIT_21_single_cd_tracklist_match",         test_single_cd_tracklist_match),
-        ("UNIT_22_genre_normalize_german",            test_genre_normalize_german),
-        ("UNIT_23_genre_normalize_english",           test_genre_normalize_english_variants),
-        ("UNIT_24_genre_normalize_titlecase",         test_genre_normalize_titlecase),
-        ("UNIT_25_is_classical_by_genre",             test_is_classical_by_genre),
-        ("UNIT_26_is_classical_by_composer",          test_is_classical_by_composer),
-        ("UNIT_27_is_classical_false_pop",            test_is_classical_false_for_pop),
-        ("UNIT_28_is_classical_false_folk",           test_is_classical_false_for_folk),
-        ("UNIT_29_normalize_cover_renames",           test_normalize_cover_renames_front_jpg),
     ]
 
     for test_id, fn in cases: