akarpov/music/services/spotify.py

@@ -33,11 +33,7 @@ def get_spotdl_client():
         spot_settings = {
             "simple_tui": True,
             "log_level": "ERROR",
-            "lyrics_providers": ["genius", "azlyrics", "musixmatch"],
-            "threads": 6,
-            "format": "mp3",
-            "ffmpeg": "ffmpeg",
-            "sponsor_block": True,
+            "lyrics_providers": ["genius", "musixmatch"],
         }
         thread_local.spotdl_client = Spotdl(
             client_id=settings.MUSIC_SPOTIFY_ID,

akarpov/music/services/youtube.py

@@ -10,7 +10,6 @@
 from PIL import Image
 from pydub import AudioSegment
 from pytube import Search, YouTube
-from spotdl.providers.audio import YouTubeMusic
 
 from akarpov.music.models import Song
 from akarpov.music.services.db import load_track
@@ -19,28 +18,22 @@
 final_filename = None
 
 
-ytmusic = YouTubeMusic()
+ydl_opts = {
+    "format": "m4a/bestaudio/best",
+    "postprocessors": [
+        {  # Extract audio using ffmpeg
+            "key": "FFmpegExtractAudio",
+            "preferredcodec": "m4a",
+        }
+    ],
+    "outtmpl": f"{settings.MEDIA_ROOT}/%(uploader)s_%(title)s.%(ext)s",
+}
 
 
 def download_file(url):
-    ydl_opts = {
-        "format": "bestaudio/best",
-        "outtmpl": f"{settings.MEDIA_ROOT}/%(uploader)s_%(title)s.%(ext)s",
-        "postprocessors": [
-            {"key": "SponsorBlock"},  # Skip sponsor segments
-            {
-                "key": "FFmpegExtractAudio",
-                "preferredcodec": "mp3",
-                "preferredquality": "192",
-            },  # Extract audio
-            {"key": "EmbedThumbnail"},  # Embed Thumbnail
-            {"key": "FFmpegMetadata"},  # Apply correct metadata
-        ],
-    }
     with yt_dlp.YoutubeDL(ydl_opts) as ydl:
-        info = ydl.extract_info(url, download=True)
-        filename = ydl.prepare_filename(info)
-    return os.path.splitext(filename)[0] + ".mp3"
+        info = ydl.extract_info(url)
+    return info["requested_downloads"][0]["_filename"]
 
 
 def parse_description(description: str) -> list:
@@ -74,7 +67,7 @@ def parse_description(description: str) -> list:
 def download_from_youtube_link(link: str, user_id: int) -> Song:
     song = None
 
-    with yt_dlp.YoutubeDL({"ignoreerrors": True, "extract_flat": True}) as ydl:
+    with yt_dlp.YoutubeDL(ydl_opts) as ydl:
         info_dict = ydl.extract_info(link, download=False)
         title = info_dict.get("title", None)
         description = info_dict.get("description", None)
@@ -89,9 +82,6 @@ def download_from_youtube_link(link: str, user_id: int) -> Song:
         + slugify(orig_path.split("/")[-1].split(".")[0])
         + ".mp3"
     )
-    if orig_path.endswith(".mp3"):
-        os.rename(orig_path, path)
-    else:
     AudioSegment.from_file(orig_path).export(path)
     if orig_path != path:
         os.remove(orig_path)

akarpov/music/tasks.py

@@ -3,13 +3,14 @@
 import pylast
 import spotipy
 import structlog
-import ytmusicapi
+import yt_dlp
 from asgiref.sync import async_to_sync
 from celery import shared_task
 from channels.layers import get_channel_layer
 from django.conf import settings
 from django.utils import timezone
 from django.utils.timezone import now
+from pytube import Channel, Playlist
 from spotipy import SpotifyClientCredentials
 
 from akarpov.music.api.serializers import SongSerializer
@@ -39,30 +40,19 @@ def list_tracks(url, user_id):
         yandex.load_playlist(url, user_id)
     if "youtube.com" in url:
         if "channel" in url or "/c/" in url:
-            ytmusic = ytmusicapi.YTMusic()
-            channel_id = url.split("/")[-1]
-            channel_songs = ytmusic.get_artist(channel_id)["songs"]["results"]
-            print(channel_songs)
-
-            for song in channel_songs:
+            channel = Channel(url)
+            for video in channel.videos:
+                with yt_dlp.YoutubeDL({}) as ydl:
+                    info = ydl.extract_info(video, download=False)
+                    if info.get("category") == "Music":
                         process_yb.apply_async(
-                    kwargs={
-                        "url": f"https://youtube.com/watch?v={song['videoId']}",
-                        "user_id": user_id,
-                    }
+                            kwargs={"url": video.watch_url, "user_id": user_id}
                         )
-
         elif "playlist" in url or "&list=" in url:
-            ytmusic = ytmusicapi.YTMusic()
-            playlist_id = url.split("=")[-1]
-            playlist_songs = ytmusic.get_playlist(playlist_id)["tracks"]["results"]
-
-            for song in playlist_songs:
+            playlist = Playlist(url)
+            for video in playlist.videos:
                 process_yb.apply_async(
-                    kwargs={
-                        "url": f"https://music.youtube.com/watch?v={song['videoId']}",
-                        "user_id": user_id,
-                    }
+                    kwargs={"url": video.watch_url, "user_id": user_id}
                 )
         else:
             process_yb.apply_async(kwargs={"url": url, "user_id": user_id})

compose/local/django/Dockerfile

@@ -33,13 +33,6 @@ RUN apt-get update && \
     apt-get purge -y --auto-remove -o APT:AutoRemove:RecommendsImportant=false && \
     rm -rf /var/lib/apt/lists/*
 
-# Make ssh dir
-RUN mkdir -p /root/.ssh/
-
-# Create known_hosts and add github to it
-RUN touch /root/.ssh/known_hosts
-RUN ssh-keyscan -t rsa github.com >> /root/.ssh/known_hosts
-
 RUN pip install "poetry==$POETRY_VERSION"
 RUN python -m venv /venv
 

pyproject.toml

@@ -119,7 +119,7 @@ spotdl = "^4.2.4"
 fuzzywuzzy = "^0.18.0"
 python-levenshtein = "^0.23.0"
 pylast = "^5.2.0"
-textract = {git = "https://github.com/Alexander-D-Karpov/textract", branch = "master"}
+textract = {git = "https://github.com/Alexander-D-Karpov/textract.git", branch = "master"}
 librosa = "^0.10.1"