Fix audio normalization not working, add online metadata helper that uses MusicBrainz, Cover Art Archive and LRCLIB to gather as much data as possible

utils/constants.py

@@ -6,11 +6,13 @@ from arcade.gui.widgets.slider import UISliderStyle
 menu_background_color = (17, 17, 17)
 log_dir = 'logs'
 discord_presence_id = 1368277020332523530
-
 audio_extensions = ["mp3", "m4a", "aac", "flac", "ogg", "opus", "wav"]
-
 view_modes = ["files", "playlist"]
 
+MUSICBRAINZ_PROJECT_NAME = "csd4ni3l/music-player"
+MUSCIBRAINZ_VERSION = "git"
+MUSICBRAINZ_CONTACT = "csd4ni3l@proton.me"
+
 DARK_GRAY = Color(45, 45, 45)
 GRAY = Color(70, 70, 70)
 LIGHT_GRAY = Color(150, 150, 150)

utils/music_handling.py

@@ -27,6 +27,7 @@ def extract_metadata_and_thumbnail(file_path: str, thumb_resolution: tuple) -> t
     sample_rate = 0
     last_played = 0
     play_count = 0
+    upload_year = 0
 
     basename = os.path.basename(file_path)
     name_only = os.path.splitext(basename)[0]
@@ -37,6 +38,7 @@ def extract_metadata_and_thumbnail(file_path: str, thumb_resolution: tuple) -> t
         try:
             artist = str(thumb_audio["artist"][0])
             title = str(thumb_audio["title"][0])
+            upload_year = int(thumb_audio["date"][0])
         except KeyError:
             artist_title_match = re.search(r'^.+\s*-\s*.+$', title)
             if artist_title_match:
@@ -111,6 +113,7 @@ def extract_metadata_and_thumbnail(file_path: str, thumb_resolution: tuple) -> t
         "file_size": file_size,
         "last_played": last_played,
         "play_count": play_count,
+        "upload_year": upload_year,
         "sample_rate": sample_rate,
         "uploader_url": uploader_url,
         "source_url": source_url,
@@ -120,6 +123,12 @@ def extract_metadata_and_thumbnail(file_path: str, thumb_resolution: tuple) -> t
     }
 
 def adjust_volume(input_path, volume):
+    audio = AudioSegment.from_file(input_path)
+    change = volume - audio.dBFS
+
+    if abs(change) < 1.0:
+        return
+
     try:
         easy_tags = EasyID3(input_path)
         tags = dict(easy_tags)
@@ -143,11 +152,8 @@ def adjust_volume(input_path, volume):
     except Exception as e:
         cover_path = None
 
-    audio = AudioSegment.from_file(input_path)
-    
-    if int(audio.dBFS) == volume:
-        return
-    
+    audio = audio.apply_gain(change)
+
     export_args = {
         "format": "mp3",
         "tags": tags
@@ -155,8 +161,6 @@ def adjust_volume(input_path, volume):
     if cover_path:
         export_args["cover"] = cover_path
 
-    change = volume - audio.dBFS
-    audio.apply_gain(change)
     audio.export(input_path, **export_args)
 
 def update_last_play_statistics(filepath):
@@ -186,4 +190,4 @@ def convert_timestamp_to_time_ago(timestamp):
     if not timestamp == 0:
         return convert_seconds_to_date(elapsed_time) + ' ago'
     else:
-        return "Never"
+        return "Never"

utils/online_metadata.py

@@ -0,0 +1,134 @@
+import musicbrainzngs as music_api
+from iso3166 import countries
+import urllib.request, json
+from utils.constants import MUSICBRAINZ_PROJECT_NAME, MUSICBRAINZ_CONTACT, MUSCIBRAINZ_VERSION
+
+WORD_BLACKLIST = ["compilation", "remix", "vs", "cover"]
+LRCLIB_BASE_URL = "https://lrclib.net/api/search"
+
+def check_blacklist(text, blacklist):
+    return any(word in text for word in blacklist)
+
+def finalize_blacklist(title):
+    blacklist = WORD_BLACKLIST[:]
+
+    for word in WORD_BLACKLIST:
+        if word in title:
+            blacklist.remove(word)
+
+    return blacklist
+
+def is_release_valid(release_id):
+    try:
+        release_data = music_api.get_release_by_id(release_id, includes=["release-groups"])
+        rg = release_data.get("release", {}).get("release-group", {})
+        if rg.get("primary-type", "").lower() == "album":
+            return True
+    except music_api.ResponseError:
+        pass
+    return False
+
+def get_country(country_code):
+    try:
+        country = countries.get(country_code)
+    except KeyError:
+        country = None
+
+    return country.name if country else None
+
+def get_music_metadata(artist, title):
+    music_api.set_useragent(MUSICBRAINZ_PROJECT_NAME, MUSCIBRAINZ_VERSION, MUSICBRAINZ_CONTACT)
+
+    if artist:
+        results = music_api.search_recordings(query=f"{artist} - {title}", limit=100)["recording-list"]
+    else:
+        results = music_api.search_recordings(query=title, limit=100)["recording-list"]
+
+    finalized_blacklist = finalize_blacklist(title)
+
+    for r in results:
+        if not r.get("title") or not r.get("isrc-list"):
+            continue
+
+        if check_blacklist(r["title"].lower(), finalized_blacklist) or check_blacklist(r.get("disambiguation", "").lower(), finalized_blacklist):
+            continue
+
+        recording_id = r["id"]
+
+        try:
+            detailed = music_api.get_recording_by_id(
+                recording_id,
+                includes=["artists", "releases", "isrcs", "tags", "ratings"]
+            )["recording"]
+        except music_api.ResponseError:
+            continue
+
+        release = None
+        for rel in detailed.get("release-list", []):
+            release_title = rel.get("title", "").lower()
+
+            if any(word in release_title for word in ["single", "ep", "maxi"]):
+                continue
+
+            if rel.get("status") == "Official" and is_release_valid(rel["id"]): # Only do it if the album is official, skipping many API calls
+                release = rel
+
+        metadata = {
+            "musicbrainz_id": recording_id,
+            "isrc": detailed["isrc-list"][0] if "isrc-list" in detailed else "Unknown",
+            "musicbrainz_album_id": release.get("id") if release else "Unknown",
+            "album_name": release.get("title") if release else "Unknown",
+            "album_date": release.get("date") if release else "Unknown",
+            "album_country": (get_country(release.get("country")) or "Worldwide") if release else "Unknown",
+            "recording_length": int(detailed["length"]) if "length" in detailed else "Unknown",
+            "musicbrainz_rating": detailed["rating"]["rating"] if "rating" in detailed else "Unknown",
+            "tags": [tag["name"] for tag in detailed.get("tag-list", [])]
+        }
+
+        return metadata
+
+    return None
+
+def get_artist_metadata(artist):
+    music_api.set_useragent(MUSICBRAINZ_PROJECT_NAME, MUSCIBRAINZ_VERSION, MUSICBRAINZ_CONTACT)
+
+    result = music_api.search_artists(query=artist, limit=10)
+
+    for r in result["artist-list"]:
+        if not r["type"] == "Person":
+            continue
+            
+        return {
+            "musicbrainz_id": r["id"],
+            "gender": r.get("gender", "Unknown"),
+            "country": get_country(r.get("country")) or "Unknown",
+            "ipi-list": r.get("ipi-list", "None"),
+            "isni-list": r.get("isni-list", "None"),
+            "born": r.get("life-span", {}).get("begin", "Unknown"),
+            "dead": r.get("life-span", {}).get("ended").lower() == "true",
+            "comment": r["disambiguation"]
+        }
+
+def get_lyrics(artist, title):
+    if artist:
+        query = f"{artist} - {title}"
+    else:
+        query = title
+    
+    query_string = urllib.parse.urlencode({"q": query})
+    full_url = f"{LRCLIB_BASE_URL}?{query_string}"
+
+    with urllib.request.urlopen(full_url) as request:
+        data = json.loads(request.read().decode("utf-8"))
+    
+    for result in data:
+        if result.get("plainLyrics"):
+            return result["plainLyrics"]
+        
+    return "Unknown"
+
+def get_album_cover_art(musicbrainz_album_id):
+    cover_art_bytes = music_api.get_image_front(musicbrainz_album_id)
+    with open("music_cover_art.jpg", "wb") as file:
+        file.write(cover_art_bytes)
+