feat: added hooks for collection on append

.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",
-        "0",
-        "d: 0",
+        "s: #a Crystal F",
+        "10",
+        "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

@@ -137,13 +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
@@ -166,7 +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}")

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

@@ -1,18 +1,14 @@
 from __future__ import annotations
 
-from typing import List, Optional, Dict, Tuple, Type, Union, TypedDict
-
-from .collection import Collection
-from .metadata import (
-    Mapping as id3Mapping,
-    ID3Timestamp,
-    Metadata
-)
-from ..utils.string_processing import unify, hash_url
-
-from .parents import OuterProxy as Base
+from typing import Dict, List, Optional, Set, Tuple, Type, TypedDict, Union
 
 from ..utils.config import main_settings
+from ..utils.string_processing import hash_url, unify
+from .collection import Collection
+from .metadata import ID3Timestamp
+from .metadata import Mapping as id3Mapping
+from .metadata import Metadata
+from .parents import OuterProxy as Base
 
 
 class ArtworkVariant(TypedDict):
@@ -23,7 +19,9 @@ class ArtworkVariant(TypedDict):
 
 
 class Artwork:
-    def __init__(self, *variants: List[ArtworkVariant]) -> None:
+    def __init__(self, *variants: List[ArtworkVariant], parent_artworks: Set[Artwork] = None) -> None:
+        self.parent_artworks: Set[Artwork] = parent_artworks or set()
+        
         self._variant_mapping: Dict[str, ArtworkVariant] = {}
 
         for variant in variants:
@@ -44,19 +42,36 @@ class Artwork:
             "deviation": self._calculate_deviation(width, height),
         }
 
+    @property
+    def flat_empty(self) -> bool:
+        return len(self._variant_mapping.keys()) <= 0
+
+    def _get_best_from_list(self, artwork_variants: List[ArtworkVariant]) -> Optional[ArtworkVariant]:
+        return min(artwork_variants, key=lambda x: x["deviation"])
+
     @property
     def best_variant(self) -> ArtworkVariant:
-        if len(self._variant_mapping.keys()) <= 0:
-            return None
-        return min(self._variant_mapping.values(), key=lambda x: x["deviation"])
+        if self.flat_empty:
+            return self._get_best_from_list([parent.best_variant for parent in self.parent_artworks])
+        return self._get_best_from_list(self._variant_mapping.values())
 
     def get_variant_name(self, variant: ArtworkVariant) -> str:
         return f"artwork_{variant['width']}x{variant['height']}_{hash_url(variant['url']).replace('/', '_')}"
 
     def __merge__(self, other: Artwork, **kwargs) -> None:
+        self.parent_artworks.update(other.parent_artworks)
+
         for key, value in other._variant_mapping.items():
             if key not in self._variant_mapping:
                 self._variant_mapping[key] = value
 
+    def __hash__(self) -> int:
+        return id(self)
+
     def __eq__(self, other: Artwork) -> bool:
+        if hash(self) == hash(other):
+            return True
+
+        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/collection.py

@@ -1,16 +1,44 @@
 from __future__ import annotations
 
-from collections import defaultdict
-from typing import TypeVar, Generic, Dict, Optional, Iterable, List, Iterator, Tuple, Generator, Union, Any, Set
 import copy
+from collections import defaultdict
+from dataclasses import dataclass
+from typing import (Any, Callable, Dict, Generator, Generic, Iterable,
+                    Iterator, List, Optional, Set, Tuple, TypeVar, Union)
 
-from .parents import OuterProxy
-from ..utils import object_trace
-from ..utils import output, BColors
+from ..utils import BColors, object_trace, output
+from .parents import InnerData, OuterProxy
 
 T = TypeVar('T', bound=OuterProxy)
 
 
+@dataclass
+class AppendHookArguments:
+    """
+    This class is used to store the arguments for the append hook.
+    The best explanation is with an examples:
+    
+    ```
+    # this is the action that triggers the append hook
+    album = Album()
+    song = Song()
+    album.song_collection.append(song)
+    ```
+
+    In this case, the append hook is triggered with the following arguments:
+    ```
+    AppendHookArguments(
+        collection=album.song_collection,
+        new_object=song,
+        collection_root_objects=[album]
+    )
+    ```
+    """
+
+    collection: Collection
+    new_object: T
+    collection_root_objects: Set[InnerData]
+
 class Collection(Generic[T]):
     __is_collection__ = True
 
@@ -27,6 +55,7 @@ class Collection(Generic[T]):
             sync_on_append: Dict[str, Collection] = None,
             append_object_to_attribute: Dict[str, T] = None,
             extend_object_to_attribute: Dict[str, Collection] = None,
+            append_callbacks: List[Callable[[AppendHookArguments], None]] = None,
     ) -> None:
         self._collection_for: dict = dict()
 
@@ -41,6 +70,7 @@ class Collection(Generic[T]):
         self.sync_on_append: Dict[str, Collection] = sync_on_append or {}
         self.pull_from: List[Collection] = []
         self.push_to: List[Collection] = []
+        self.append_callbacks: List[Callable[[AppendHookArguments], None]] = append_callbacks or []
 
         # This is to cleanly unmap previously mapped items by their id
         self._indexed_from_id: Dict[int, Dict[str, Any]] = defaultdict(dict)
@@ -141,6 +171,14 @@ class Collection(Generic[T]):
         for attribute, new_object in self.append_object_to_attribute.items():
             other.__getattribute__(attribute).append(new_object, **kwargs)
 
+        append_hook_args = AppendHookArguments(
+            collection=self,
+            new_object=other,
+            collection_root_objects=self._collection_for.keys(),
+        )
+        for callback in self.append_callbacks:
+            callback(append_hook_args)
+
     def append(self, other: Optional[T], **kwargs):
         """
         If an object, that represents the same entity exists in a relevant collection,

music_kraken/objects/formatted_text.py

@@ -38,11 +38,19 @@ class FormattedText:
     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/objects/song.py

@@ -477,6 +477,8 @@ class Artist(Base):
     general_genre: str
     unformatted_location: str
 
+    artwork: List[Artwork]
+
     source_collection: SourceCollection
     contact_collection: Collection[Contact]
 
@@ -493,6 +495,8 @@ class Artist(Base):
         "lyrical_themes": list,
         "general_genre": lambda: "",
 
+        "artwork": list,
+
         "source_collection": SourceCollection,
         "album_collection": Collection,
         "contact_collection": Collection,
@@ -511,6 +515,7 @@ class Artist(Base):
         notes: FormattedText = None, 
         lyrical_themes: List[str] = None,
         general_genre: str = None, 
+        artwork: List[Artwork] = None,
         unformatted_location: str = None, 
         source_list: List[Source] = None,
         contact_list: List[Contact] = None, 

music_kraken/pages/__init__.py

@@ -3,5 +3,6 @@ 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

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") or {})),
+            )
+
+        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

@@ -485,7 +485,7 @@ class Musify(Page):
             
             track_name = list_points[4].text.strip()
 
-        # artwork
+        # album artwork
         artwork: Artwork = Artwork()
         album_image_element_list: List[BeautifulSoup] = soup.find_all("img", {"class": "album-img"})
         for album_image_element in album_image_element_list:
@@ -918,7 +918,8 @@ class Musify(Page):
             name=name,
             country=country,
             source_list=source_list,
-            notes=notes
+            notes=notes,
+            artwork=self._fetch_artist_artwork(soup, **kwargs)
         )
 
     def _parse_album_card(self, album_card: BeautifulSoup, artist_name: str = None, **kwargs) -> Album:
@@ -1056,6 +1057,20 @@ class Musify(Page):
 
             artist.album_collection.append(album)
 
+    def _fetch_artist_artwork(self, soup: BeautifulSoup, **kwargs):
+        # artist artwork
+        artist_artwork: List[Artwork] = Artwork()
+        artist_a_element_list: List[BeautifulSoup] = soup.find_all("a")
+        for artist_a_element in artist_a_element_list:
+            if artist_a_element.find_all("img", {"class": "artist-img"}).count() > 0:
+                artwork_gallery = self.connection.get(artist_a_element("data-src", artist_a_element.get("href")))
+                if artwork_gallery is not None:
+                    gallery_image_element_list: List[BeautifulSoup] = artwork_gallery.find_all("img", {"class": "artist-img"})
+                    for gallery_image_element in gallery_image_element_list:
+                        artist_artwork.push(Artwork(url=gallery_image_element.get("data-src", gallery_image_element.get("src"))))
+
+        return artist_artwork
+
     def fetch_artist(self, source: Source, **kwargs) -> Artist:
         """
         TODO
@@ -1068,7 +1083,7 @@ class Musify(Page):
 
         artist = self._fetch_initial_artist(url, source=source, **kwargs)
         self._fetch_artist_discography(artist, url, artist.name, **kwargs)
-        
+        self._fetch_artist_artwork(artist, **kwargs)
         return artist
 
     def fetch_label(self, source: Source, stop_at_level: int = 1) -> Label:

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: