feat: added detection to autoscann pages

.vscode/settings.json

@@ -27,6 +27,7 @@
         "Gitea",
         "iframe",
         "isrc",
+        "itemprop",
         "levenshtein",
         "metallum",
         "MUSICBRAINZ",

development/actual_donwload.py

@@ -6,9 +6,10 @@ logging.getLogger().setLevel(logging.DEBUG)
 
 if __name__ == "__main__":
     commands = [
-        "s: #a I'm in a coffin",
+        "s: #a Crystal F",
-        "0",
+        "10",
-        "d: 0",
+        "1",
+        "3",
     ]
 
     

music_kraken/audio/metadata.py

@@ -93,6 +93,10 @@ def write_metadata_to_target(metadata: Metadata, target: Target, song: Song):
             # resize the image to the preferred resolution
             img.thumbnail((main_settings["preferred_artwork_resolution"], main_settings["preferred_artwork_resolution"]))
 
+            # https://stackoverflow.com/a/59476938/16804841
+            if img.mode != 'RGB':
+                img = img.convert('RGB')
+
             img.save(converted_target.file_path, "JPEG")
 
         # https://stackoverflow.com/questions/70228440/mutagen-how-can-i-correctly-embed-album-art-into-mp3-file-so-that-i-can-see-t

music_kraken/connection/cache.py

@@ -6,6 +6,7 @@ from typing import List, Optional
 from functools import lru_cache
 import logging
 
+from ..utils import output, BColors
 from ..utils.config import main_settings
 from ..utils.string_processing import fit_to_file_system
 
@@ -136,13 +137,13 @@ class Cache:
         )
         self._write_attribute(cache_attribute)
 
-        cache_path = fit_to_file_system(Path(module_path, name), hidden_ok=True)
+        cache_path = fit_to_file_system(Path(module_path, name.replace("/", "_")), hidden_ok=True)
         with cache_path.open("wb") as content_file:
             self.logger.debug(f"writing cache to {cache_path}")
             content_file.write(content)
 
     def get(self, name: str) -> Optional[CacheResult]:
-        path = fit_to_file_system(Path(self._dir, self.module, name), hidden_ok=True)
+        path = fit_to_file_system(Path(self._dir, self.module, name.replace("/", "_")), hidden_ok=True)
 
         if not path.is_file():
             return None
@@ -165,7 +166,7 @@ class Cache:
             if ca.name == "":
                 continue
 
-            file = fit_to_file_system(Path(self._dir, ca.module, ca.name), hidden_ok=True)
+            file = fit_to_file_system(Path(self._dir, ca.module, ca.name.replace("/", "_")), hidden_ok=True)
 
             if not ca.is_valid:
                 self.logger.debug(f"deleting cache {ca.id}")
@@ -204,9 +205,12 @@ class Cache:
         for path in self._dir.iterdir():
             if path.is_dir():
                 for file in path.iterdir():
+                    output(f"Deleting file {file}", color=BColors.GREY)
                     file.unlink()
+                output(f"Deleting folder {path}", color=BColors.HEADER)
                 path.rmdir()
             else:
+                output(f"Deleting folder {path}", color=BColors.HEADER)
                 path.unlink()
 
         self.cached_attributes.clear()

music_kraken/download/page_attributes.py

@@ -30,11 +30,12 @@ from ..utils.exception import MKMissingNameException
 from ..utils.exception.download import UrlNotFoundException
 from ..utils.shared import DEBUG_PAGES
 
-from ..pages import Page, EncyclopaediaMetallum, Musify, YouTube, YoutubeMusic, Bandcamp, INDEPENDENT_DB_OBJECTS
+from ..pages import Page, EncyclopaediaMetallum, Musify, YouTube, YoutubeMusic, Bandcamp, Genius, INDEPENDENT_DB_OBJECTS
 
 
 ALL_PAGES: Set[Type[Page]] = {
     # EncyclopaediaMetallum,
+    Genius,
     Musify,
     YoutubeMusic,
     Bandcamp

music_kraken/objects/artwork.py

@@ -59,4 +59,6 @@ class Artwork:
                 self._variant_mapping[key] = value
 
     def __eq__(self, other: Artwork) -> bool:
+        if not isinstance(other, Artwork):
+            return False
         return any(a == b for a, b in zip(self._variant_mapping.keys(), other._variant_mapping.keys()))

music_kraken/objects/formatted_text.py

@@ -37,11 +37,19 @@ class FormattedText:
     @property
     def markdown(self) -> str:
         return md(self.html).strip()
+    
+    @markdown.setter
+    def markdown(self, value: str) -> None:
+        self.html = mistune.markdown(value)
 
     @property
     def plain(self) -> str:
         md = self.markdown
         return md.replace("\n\n", "\n")
+    
+    @plain.setter
+    def plain(self, value: str) -> None:
+        self.html = mistune.markdown(plain_to_markdown(value))
 
     def __str__(self) -> str:
         return self.markdown

music_kraken/pages/__init__.py

@@ -1,7 +1,56 @@
-from .encyclopaedia_metallum import EncyclopaediaMetallum
+from typing import Type, Generator, Set, Dict, List
-from .musify import Musify
+from collections import defaultdict
-from .youtube import YouTube
-from .youtube_music import YoutubeMusic
-from .bandcamp import Bandcamp
 
-from .abstract import Page, INDEPENDENT_DB_OBJECTS
+from ._encyclopaedia_metallum import EncyclopaediaMetallum
+from ._musify import Musify
+from ._youtube import YouTube
+from ._youtube_music import YoutubeMusic
+from ._bandcamp import Bandcamp
+from ._genius import Genius
+from ._abstract import Page, INDEPENDENT_DB_OBJECTS
+
+
+_registered_pages: Dict[Type[Page], Set[Page]] = defaultdict(set)
+
+
+def get_pages(*page_types: List[Type[Page]]) -> Generator[Page, None, None]:
+    if len(page_types) == 0:
+        page_types = _registered_pages.keys()
+
+    for page_type in page_types:
+        yield from _registered_pages[page_type]
+
+
+def register_page(page_type: Type[Page], **kwargs):
+    if page_type in _registered_pages:
+        return
+
+    _registered_pages[page_type].add(page_type(**kwargs))
+
+
+def deregister_page(page_type: Type[Page]):
+    if page_type not in _registered_pages:
+        return
+
+    for p in _registered_pages[page_type]:
+        p.__del__()
+    del _registered_pages[page_type]
+
+def scan_for_pages():
+    # assuming the wanted pages are the leaf classes of the interface
+    leaf_classes = []
+
+    _class_list = [Page]
+    while len(_class_list):
+        _class = _class_list.pop()
+        _class_subclasses = _class.__subclasses__()
+
+        if len(_class_subclasses) == 0:
+            if _class.REGISTER:
+                leaf_classes.append(_class)
+        else:
+            _class_list.extend(_class_subclasses)
+
+    print(leaf_classes)
+    for leaf_class in leaf_classes:
+        register_page(leaf_class)

music_kraken/pages/_abstract.py

@@ -48,12 +48,12 @@ class DownloadOptions:
     process_metadata_if_found: bool = True
 
 class Page:
+    REGISTER = True
     SOURCE_TYPE: SourceType
     LOGGER: logging.Logger
 
     def __new__(cls, *args, **kwargs):
         cls.LOGGER = logging.getLogger(cls.__name__)
-
         return super().__new__(cls)
 
     def __init__(self, download_options: DownloadOptions = None, fetch_options: FetchOptions = None):
@@ -62,6 +62,9 @@ class Page:
         self.download_options: DownloadOptions = download_options or DownloadOptions()
         self.fetch_options: FetchOptions = fetch_options or FetchOptions()
 
+    def __del__(self):
+        self.SOURCE_TYPE.deregister_page(self)
+
     def _search_regex(self, pattern, string, default=None, fatal=True, flags=0, group=None):
         """
         Perform a regex search on the given string, using a single or a list of

music_kraken/pages/_bandcamp.py

@@ -6,7 +6,7 @@ from bs4 import BeautifulSoup
 import pycountry
 
 from ..objects import Source, DatabaseObject
-from .abstract import Page
+from ._abstract import Page
 from ..objects import (
     Artist,
     Source,

music_kraken/pages/_encyclopaedia_metallum.py

@@ -6,7 +6,7 @@ from urllib.parse import urlparse, urlencode
 
 from ..connection import Connection
 from ..utils.config import logging_settings
-from .abstract import Page
+from ._abstract import Page
 from ..utils.enums import SourceType, ALL_SOURCE_TYPES
 from ..utils.enums.album import AlbumType
 from ..utils.support_classes.query import Query
@@ -207,6 +207,7 @@ def create_grid(
 
 
 class EncyclopaediaMetallum(Page):
+    REGISTER = False
     SOURCE_TYPE = ALL_SOURCE_TYPES.ENCYCLOPAEDIA_METALLUM
     LOGGER = logging_settings["metal_archives_logger"]
     

music_kraken/pages/_genius.py

@@ -0,0 +1,297 @@
+from typing import List, Optional, Type
+from urllib.parse import urlparse, urlunparse, urlencode
+import json
+from enum import Enum
+from bs4 import BeautifulSoup
+import pycountry
+
+from ..objects import Source, DatabaseObject
+from ._abstract import Page
+from ..objects import (
+    Artist,
+    Source,
+    SourceType,
+    Song,
+    Album,
+    Label,
+    Target,
+    Contact,
+    ID3Timestamp,
+    Lyrics,
+    FormattedText,
+    Artwork,
+)
+from ..connection import Connection
+from ..utils import dump_to_file, traverse_json_path
+from ..utils.enums import SourceType, ALL_SOURCE_TYPES
+from ..utils.support_classes.download_result import DownloadResult
+from ..utils.string_processing import clean_song_title
+from ..utils.config import main_settings, logging_settings
+from ..utils.shared import DEBUG
+
+if DEBUG:
+    from ..utils import dump_to_file
+
+
+class Genius(Page):
+    SOURCE_TYPE = ALL_SOURCE_TYPES.GENIUS
+    HOST = "genius.com"
+
+    def __init__(self, *args, **kwargs):
+        self.connection: Connection = Connection(
+            host="https://genius.com/",
+            logger=self.LOGGER,
+            module="genius",
+        )
+
+        super().__init__(*args, **kwargs)
+
+    def get_source_type(self, source: Source) -> Optional[Type[DatabaseObject]]:
+        path = source.parsed_url.path.replace("/", "")
+        
+        if path.startswith("artists"):
+            return Artist
+        if path.startswith("albums"):
+            return Album
+
+        return Song
+
+    def add_to_artwork(self, artwork: Artwork, url: str):
+        if url is None:
+            return
+        
+        url_frags = url.split(".")
+        if len(url_frags) < 2:
+            artwork.append(url=url)
+            return
+
+        dimensions = url_frags[-2].split("x")
+        if len(dimensions) < 2:
+            artwork.append(url=url)
+            return
+
+        if len(dimensions) == 3:
+            dimensions = dimensions[:-1]
+        
+        try:
+            artwork.append(url=url, width=int(dimensions[0]), height=int(dimensions[1]))
+        except ValueError:
+            artwork.append(url=url)
+
+    def parse_api_object(self, data: dict) -> Optional[DatabaseObject]:
+        if data is None:
+            return None
+        object_type = data.get("_type")
+
+        artwork = Artwork()
+        self.add_to_artwork(artwork, data.get("header_image_url"))
+        self.add_to_artwork(artwork, data.get("image_url"))
+        
+        additional_sources: List[Source] = []
+        source: Source = Source(self.SOURCE_TYPE, data.get("url"), additional_data={
+            "id": data.get("id"),
+            "slug": data.get("slug"),
+            "api_path": data.get("api_path"),
+        })
+
+        notes = FormattedText()
+        description = data.get("description") or {}
+        if "html" in description:
+            notes.html = description["html"]
+        elif "markdown" in description:
+            notes.markdown = description["markdown"]
+        elif "description_preview" in data:
+            notes.plaintext = data["description_preview"]
+
+        if source.url is None:
+            return None
+
+        if object_type == "artist":
+            if data.get("instagram_name") is not None:
+                additional_sources.append(Source(ALL_SOURCE_TYPES.INSTAGRAM, f"https://www.instagram.com/{data['instagram_name']}/"))
+            if data.get("facebook_name") is not None:
+                additional_sources.append(Source(ALL_SOURCE_TYPES.FACEBOOK, f"https://www.facebook.com/{data['facebook_name']}/"))
+            if data.get("twitter_name") is not None:
+                additional_sources.append(Source(ALL_SOURCE_TYPES.TWITTER, f"https://x.com/{data['twitter_name']}/"))
+
+            return Artist(
+                name=data["name"].strip() if data.get("name") is not None else None,
+                source_list=[source],
+                artwork=artwork,
+                notes=notes,
+            )
+
+        if object_type == "album":
+            self.add_to_artwork(artwork, data.get("cover_art_thumbnail_url"))
+            self.add_to_artwork(artwork, data.get("cover_art_url"))
+
+            for cover_art in data.get("cover_arts", []):
+                self.add_to_artwork(artwork, cover_art.get("image_url"))
+                self.add_to_artwork(artwork, cover_art.get("thumbnail_image_url"))
+
+            return Album(
+                title=data.get("name").strip(),
+                source_list=[source],
+                artist_list=[self.parse_api_object(data.get("artist"))],
+                artwork=artwork,
+                date=ID3Timestamp(**data.get("release_date_components", {})),
+            )
+
+        if object_type == "song":
+            self.add_to_artwork(artwork, data.get("song_art_image_thumbnail_url"))
+            self.add_to_artwork(artwork, data.get("song_art_image_url"))
+
+            main_artist_list = []
+            featured_artist_list = []
+
+            _artist_name = None
+            primary_artist = self.parse_api_object(data.get("primary_artist"))
+            if primary_artist is not None:
+                _artist_name = primary_artist.name
+                main_artist_list.append(primary_artist)
+            for feature_artist in (*(data.get("featured_artists") or []), *(data.get("producer_artists") or []), *(data.get("writer_artists") or [])):
+                artist = self.parse_api_object(feature_artist)
+                if artist is not None:
+                    featured_artist_list.append(artist)
+
+            return Song(
+                title=clean_song_title(data.get("title"), artist_name=_artist_name),
+                source_list=[source],
+                artwork=artwork,
+                feature_artist_list=featured_artist_list,
+                artist_list=main_artist_list,
+            )
+
+        return None
+
+    def general_search(self, search_query: str, **kwargs) -> List[DatabaseObject]:
+        results = []
+
+        search_params = {
+            "q": search_query,
+        }
+
+        r = self.connection.get("https://genius.com/api/search/multi?" + urlencode(search_params), name=f"search_{search_query}")
+        if r is None:
+            return results
+
+        dump_to_file("search_genius.json", r.text, is_json=True, exit_after_dump=False)
+        data = r.json()
+
+        for elements in traverse_json_path(data, "response.sections", default=[]):
+            hits = elements.get("hits", [])
+            for hit in hits:
+                parsed = self.parse_api_object(hit.get("result"))
+                if parsed is not None:
+                    results.append(parsed)
+
+        return results
+
+    def fetch_artist(self, source: Source, stop_at_level: int = 1) -> Artist:
+        artist: Artist = Artist()
+        # https://genius.com/api/artists/24527/albums?page=1
+
+        r = self.connection.get(source.url, name=source.url)
+        if r is None:
+            return artist
+        soup = self.get_soup_from_response(r)
+
+        # find the content attribute in the meta tag which is contained in the head
+        data_container = soup.find("meta", {"itemprop": "page_data"})
+        if data_container is not None:
+            content = data_container["content"]
+            dump_to_file("genius_itemprop_artist.json", content, is_json=True, exit_after_dump=False)
+            data = json.loads(content)
+
+            artist = self.parse_api_object(data.get("artist"))
+            
+            for e in (data.get("artist_albums") or []):
+                r = self.parse_api_object(e)
+                if not isinstance(r, Album):
+                    continue
+
+                artist.album_collection.append(r)
+            
+            for e in (data.get("artist_songs") or []):
+                r = self.parse_api_object(e)
+                if not isinstance(r, Song):
+                    continue
+
+                """
+                TODO
+                fetch the album for these songs, because the api doesn't 
+                return them
+                """
+
+                artist.album_collection.extend(r.album_collection)
+
+        artist.source_collection.append(source)
+
+        return artist
+
+    def fetch_album(self, source: Source, stop_at_level: int = 1) -> Album:
+        album: Album = Album()
+        # https://genius.com/api/artists/24527/albums?page=1
+
+        r = self.connection.get(source.url, name=source.url)
+        if r is None:
+            return album
+        soup = self.get_soup_from_response(r)
+
+        # find the content attribute in the meta tag which is contained in the head
+        data_container = soup.find("meta", {"itemprop": "page_data"})
+        if data_container is not None:
+            content = data_container["content"]
+            dump_to_file("genius_itemprop_album.json", content, is_json=True, exit_after_dump=False)
+            data = json.loads(content)
+
+            album = self.parse_api_object(data.get("album"))
+
+            for e in data.get("album_appearances", []):
+                r = self.parse_api_object(e.get("song"))
+                if not isinstance(r, Song):
+                    continue
+
+                album.song_collection.append(r)
+
+        album.source_collection.append(source)
+        
+        return album
+
+    def get_json_content_from_response(self, response, start: str, end: str) -> Optional[str]:
+        content = response.text
+        start_index = content.find(start)
+        if start_index < 0:
+            return None
+        start_index += len(start)
+        end_index = content.find(end, start_index)
+        if end_index < 0:
+            return None
+        return content[start_index:end_index]
+
+    def fetch_song(self, source: Source, stop_at_level: int = 1) -> Song:
+        song: Song = Song()
+
+        r = self.connection.get(source.url, name=source.url)
+        if r is None:
+            return song
+
+        # get the contents that are between `JSON.parse('` and `');`
+        content = self.get_json_content_from_response(r, start="window.__PRELOADED_STATE__ = JSON.parse('", end="');\n      window.__APP_CONFIG__ = ")
+        if content is not None:
+            content = content.replace("\\\\", "\\").replace('\\"', '"').replace("\\'", "'")
+            data = json.loads(content)
+
+            lyrics_html = traverse_json_path(data, "songPage.lyricsData.body.html", default=None)
+            if lyrics_html is not None:
+                song.lyrics_collection.append(Lyrics(FormattedText(html=lyrics_html)))
+
+            dump_to_file("genius_song_script_json.json", content, is_json=True, exit_after_dump=False)
+
+        soup = self.get_soup_from_response(r)
+        for lyrics in soup.find_all("div", {"data-lyrics-container": "true"}):
+            lyrics_object = Lyrics(FormattedText(html=lyrics.prettify()))
+            song.lyrics_collection.append(lyrics_object)
+
+        song.source_collection.append(source)
+        return song

music_kraken/pages/_musify.py

@@ -8,7 +8,7 @@ import pycountry
 from bs4 import BeautifulSoup
 
 from ..connection import Connection
-from .abstract import Page
+from ._abstract import Page
 from ..utils.enums import SourceType, ALL_SOURCE_TYPES
 from ..utils.enums.album import AlbumType, AlbumStatus
 from ..objects import (

music_kraken/pages/_youtube.py

@@ -5,7 +5,7 @@ from enum import Enum
 import python_sponsorblock
 
 from ..objects import Source, DatabaseObject, Song, Target
-from .abstract import Page
+from ._abstract import Page
 from ..objects import (
     Artist,
     Source,
@@ -22,7 +22,7 @@ from ..utils.enums import SourceType, ALL_SOURCE_TYPES
 from ..utils.support_classes.download_result import DownloadResult
 from ..utils.config import youtube_settings, main_settings, logging_settings
 
-from .youtube_music.super_youtube import SuperYouTube, YouTubeUrl, get_invidious_url, YouTubeUrlType
+from ._youtube_music.super_youtube import SuperYouTube, YouTubeUrl, get_invidious_url, YouTubeUrlType
 
 
 """
@@ -38,7 +38,7 @@ def get_piped_url(path: str = "", params: str = "", query: str = "", fragment: s
 
 
 class YouTube(SuperYouTube):
-    # CHANGE
+    REGISTER = youtube_settings["use_youtube_alongside_youtube_music"]
     SOURCE_TYPE = ALL_SOURCE_TYPES.YOUTUBE
 
     def __init__(self, *args, **kwargs):

music_kraken/pages/_youtube_music/_list_render.py

@@ -3,7 +3,6 @@ from enum import Enum
 
 from ...utils.config import logging_settings
 from ...objects import Source, DatabaseObject
-from ..abstract import Page
 from ...objects import (
     Artist,
     Source,

music_kraken/pages/_youtube_music/_music_object_render.py

@@ -6,7 +6,6 @@ from ...utils.string_processing import clean_song_title
 from ...utils.enums import SourceType, ALL_SOURCE_TYPES
 
 from ...objects import Source, DatabaseObject
-from ..abstract import Page
 from ...objects import (
     Artist,
     Source,

music_kraken/pages/_youtube_music/super_youtube.py

@@ -6,7 +6,7 @@ import requests
 import python_sponsorblock
 
 from ...objects import Source, DatabaseObject, Song, Target
-from ..abstract import Page
+from .._abstract import Page
 from ...objects import (
     Artist,
     Source,

music_kraken/pages/_youtube_music/youtube_music.py

@@ -22,7 +22,7 @@ from ...utils import get_current_millis, traverse_json_path
 
 from ...utils import dump_to_file
 
-from ..abstract import Page
+from .._abstract import Page
 from ...objects import (
     DatabaseObject as DataObject,
     Source,
@@ -549,6 +549,11 @@ class YoutubeMusic(SuperYouTube):
         return album
 
     def fetch_lyrics(self, video_id: str, playlist_id: str = None) -> str:
+        """
+        1. fetches the tabs of a song, to get the browse id
+        2. finds the browse id of the lyrics
+        3. fetches the lyrics with the browse id
+        """
         request_data = {
             "context": {**self.credentials.context, "adSignalsInfo": {"params": []}},
             "videoId": video_id,
@@ -575,7 +580,8 @@ class YoutubeMusic(SuperYouTube):
             pageType = traverse_json_path(tab, "tabRenderer.endpoint.browseEndpoint.browseEndpointContextSupportedConfigs.browseEndpointContextMusicConfig.pageType", default="")
             if pageType in ("MUSIC_TAB_TYPE_LYRICS", "MUSIC_PAGE_TYPE_TRACK_LYRICS") or "lyrics" in pageType.lower():
                 browse_id = traverse_json_path(tab, "tabRenderer.endpoint.browseEndpoint.browseId", default=None)
-                break
+                if browse_id is not None:
+                    break
 
         if browse_id is None:
             return None
@@ -721,7 +727,6 @@ class YoutubeMusic(SuperYouTube):
 
         self.download_values_by_url[source.url] = {
             "url": _best_format.get("url"),
-            "chunk_size": _best_format.get("downloader_options", {}).get("http_chunk_size", main_settings["chunk_size"]),
             "headers": _best_format.get("http_headers", {}),
         }
 

music_kraken/utils/enums/__init__.py

@@ -17,6 +17,9 @@ class SourceType:
     def register_page(self, page: Page):
         self.page = page
 
+    def deregister_page(self):
+        self.page = None
+
     def __hash__(self):
         return hash(self.name)
 

music_kraken/utils/shared.py

@@ -15,11 +15,11 @@ __stage__ = os.getenv("STAGE", "prod")
 DEBUG = (__stage__ == "dev") and True
 DEBUG_LOGGING = DEBUG and False
 DEBUG_TRACE = DEBUG and True
-DEBUG_OBJECT_TRACE = DEBUG and True
+DEBUG_OBJECT_TRACE = DEBUG and False
 DEBUG_OBJECT_TRACE_CALLSTACK = DEBUG_OBJECT_TRACE and False
 DEBUG_YOUTUBE_INITIALIZING = DEBUG and False
 DEBUG_PAGES = DEBUG and False
-DEBUG_DUMP = DEBUG and False
+DEBUG_DUMP = DEBUG and True
 DEBUG_PRINT_ID = DEBUG and True
 
 if DEBUG: