Merge pull request 'feature/artwork_gallery' (#41) from feature/artwork_gallery into experimental

Reviewed-on: #41

.vscode/settings.json

@@ -20,6 +20,7 @@
         "APIC",
         "Bandcamp",
         "bitrate",
+        "CALLSTACK",
         "DEEZER",
         "dotenv",
         "encyclopaedia",

development/actual_donwload.py

@@ -1,15 +1,13 @@
+import logging
+
 import music_kraken
 
-import logging
 print("Setting logging-level to DEBUG")
 logging.getLogger().setLevel(logging.DEBUG)
 
 if __name__ == "__main__":
     commands = [
-        "s: #a Crystal F",
-        "10",
-        "1",
-        "3",
+        "s: #a Ghost Bath",
     ]
 
     

music_kraken/audio/metadata.py

@@ -1,21 +1,21 @@
-import mutagen
-from mutagen.id3 import ID3, Frame, APIC, USLT
+import logging
 from pathlib import Path
 from typing import List
-import logging
+
+import mutagen
+from mutagen.id3 import APIC, ID3, USLT, Frame
 from PIL import Image
 
-from ..utils.config import logging_settings, main_settings
-from ..objects import Song, Target, Metadata
-from ..objects.metadata import Mapping
 from ..connection import Connection
+from ..objects import Metadata, Song, Target
+from ..objects.metadata import Mapping
+from ..utils.config import logging_settings, main_settings
 
 LOGGER = logging_settings["tagging_logger"]
 
 
 artwork_connection: Connection = Connection()
 
-
 class AudioMetadata:
     def __init__(self, file_location: str = None) -> None:
         self._file_location = None
@@ -67,13 +67,14 @@ def write_metadata_to_target(metadata: Metadata, target: Target, song: Song):
     id3_object = AudioMetadata(file_location=target.file_path)
 
     LOGGER.info(str(metadata))
-
-    if song.artwork.best_variant is not None:
-        best_variant = song.artwork.best_variant
+    ## REWRITE COMPLETLY !!!!!!!!!!!!
+    if len(song.artwork._data) != 0:
+        variants = song.artwork._data.__getitem__(0)
+        best_variant = variants.variants.__getitem__(0)
 
         r = artwork_connection.get(
-            url=best_variant["url"],
-            name=song.artwork.get_variant_name(best_variant),
+            url=best_variant.url,
+            name=best_variant.url,
         )
 
         temp_target: Target = Target.temp()
@@ -107,7 +108,7 @@ def write_metadata_to_target(metadata: Metadata, target: Target, song: Song):
                 mime="image/jpeg",
                 type=3,
                 desc=u"Cover",
-                data=converted_target.read_bytes(),
+                data=converted_target.raw_content,
             )
         )
         id3_object.frames.delall("USLT")

music_kraken/connection/connection.py

@@ -1,12 +1,12 @@
 from __future__ import annotations
 
+import copy
+import inspect
 import logging
 import threading
 import time
-from typing import List, Dict, Optional, Set
-from urllib.parse import urlparse, urlunsplit, ParseResult
-import copy
-import inspect
+from typing import TYPE_CHECKING, Dict, List, Optional, Set
+from urllib.parse import ParseResult, urlparse, urlunsplit
 
 import requests
 import responses
@@ -14,12 +14,15 @@ from tqdm import tqdm
 
 from .cache import Cache
 from .rotating import RotatingProxy
+
+if TYPE_CHECKING:
     from ..objects import Target
+
 from ..utils import request_trace
-from ..utils.string_processing import shorten_display_url
 from ..utils.config import main_settings
-from ..utils.support_classes.download_result import DownloadResult
 from ..utils.hacking import merge_args
+from ..utils.string_processing import shorten_display_url
+from ..utils.support_classes.download_result import DownloadResult
 
 
 class Connection:

music_kraken/download/page_attributes.py

@@ -3,6 +3,7 @@ from collections import defaultdict
 from pathlib import Path
 import re
 import logging
+import subprocess
 
 from PIL import Image
 
@@ -36,8 +37,6 @@ from ..connection import Connection
 
 from ..pages import Page, EncyclopaediaMetallum, Musify, YouTube, YoutubeMusic, Bandcamp, Genius, INDEPENDENT_DB_OBJECTS
 
-artwork_connection: Connection = Connection()
-
 ALL_PAGES: Set[Type[Page]] = {
     # EncyclopaediaMetallum,
     Genius,
@@ -92,7 +91,8 @@ class Pages:
             exclude_pages = exclude_pages.union(SHADY_PAGES)
 
         if not exclude_pages.issubset(ALL_PAGES):
-            raise ValueError(f"The excluded pages have to be a subset of all pages: {exclude_pages} | {ALL_PAGES}")
+            raise ValueError(
+                f"The excluded pages have to be a subset of all pages: {exclude_pages} | {ALL_PAGES}")
 
         def _set_to_tuple(page_set: Set[Type[Page]]) -> Tuple[Type[Page], ...]:
             return tuple(sorted(page_set, key=lambda page: page.__name__))
@@ -100,11 +100,14 @@ class Pages:
         self._pages_set: Set[Type[Page]] = ALL_PAGES.difference(exclude_pages)
         self.pages: Tuple[Type[Page], ...] = _set_to_tuple(self._pages_set)
 
-        self._audio_pages_set: Set[Type[Page]] = self._pages_set.intersection(AUDIO_PAGES)
-        self.audio_pages: Tuple[Type[Page], ...] = _set_to_tuple(self._audio_pages_set)
+        self._audio_pages_set: Set[Type[Page]
+                                   ] = self._pages_set.intersection(AUDIO_PAGES)
+        self.audio_pages: Tuple[Type[Page], ...] = _set_to_tuple(
+            self._audio_pages_set)
 
         for page_type in self.pages:
-            self._page_instances[page_type] = page_type(fetch_options=self.fetch_options, download_options=self.download_options)
+            self._page_instances[page_type] = page_type(
+                fetch_options=self.fetch_options, download_options=self.download_options)
             self._source_to_page[page_type.SOURCE_TYPE] = page_type
 
     def _get_page_from_enum(self, source_page: SourceType) -> Page:
@@ -118,7 +121,8 @@ class Pages:
         for page_type in self.pages:
             result.add(
                 page=page_type,
-                search_result=self._page_instances[page_type].search(query=query)
+                search_result=self._page_instances[page_type].search(
+                    query=query)
             )
 
         return result
@@ -131,7 +135,8 @@ class Pages:
         for source in data_object.source_collection.get_sources(source_type_sorting={
             "only_with_page": True,
         }):
-            new_data_object = self.fetch_from_source(source=source, stop_at_level=stop_at_level)
+            new_data_object = self.fetch_from_source(
+                source=source, stop_at_level=stop_at_level)
             if new_data_object is not None:
                 data_object.merge(new_data_object)
 
@@ -167,64 +172,38 @@ class Pages:
 
         return False
 
-    def download_artwork_variant_to_target(self, artwork_variant: ArtworkVariant, target: Target):
-        
-        r = artwork_connection.get(
-            url=artwork_variant["url"],
-            name=artwork_variant["url"],
-        )
-
-        
-        temp_target: Target = Target.temp()
-        with temp_target.open("wb") as f:
-            f.write(r.content)
-
-        converted_target: Target = Target.temp(file_extension=main_settings["image_format"])
-        with Image.open(temp_target.file_path) as img:
-            # crop the image if it isn't square in the middle with minimum data loss
-            width, height = img.size
-            if width != height:
-                if width > height:
-                    img = img.crop((width // 2 - height // 2, 0, width // 2 + height // 2, height))
-                else:
-                    img = img.crop((0, height // 2 - width // 2, width, height // 2 + width // 2))
-
-            # 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(target.file_path, main_settings["image_format"])
-
-
-
     def _fetch_artist_artwork(self, artist: Artist, naming: dict):
         naming: Dict[str, List[str]] = defaultdict(list, naming)
         naming["artist"].append(artist.name)
-        naming["label"].extend([l.title_value for l in artist.label_collection])
+        naming["label"].extend(
+            [l.title_value for l in artist.label_collection])
         # removing duplicates from the naming, and process the strings
         for key, value in naming.items():
             # https://stackoverflow.com/a/17016257
             naming[key] = list(dict.fromkeys(value))
 
-        artwork: Artwork = artist.artwork
-        for image_number, variant in enumerate(artwork):
+        artwork_collection: ArtworkCollection = artist.artwork
+        artwork_collection.compile()
+        for image_number, artwork in enumerate(artwork_collection):
+            for artwork_variant in artwork.variants:
                 naming["image_number"] = [str(image_number)]
-
-            url: str = variant["url"]
-
                 target = Target(
                     relative_to_music_dir=True,
-                file_path=Path(self._parse_path_template(main_settings["artist_artwork_path"], naming=naming))
+                    file_path=Path(self._parse_path_template(
+                        main_settings["artist_artwork_path"], naming=naming))
                 )
-            self.download_artwork_variant_to_target(variant, target)
+                if not target.file_path.parent.exists():
+                    target.create_path()
+                    subprocess.Popen(["gio", "set", target.file_path.parent, "metadata::custom-icon", "file://"+str(target.file_path)])
+                with Image.open(artwork_variant.target.file_path) as img:
+                    img.save(target.file_path, main_settings["image_format"])
+                    artwork_variant.target = Target
 
     def download(self, data_object: DataObject, genre: str, **kwargs) -> DownloadResult:
         # fetch the given object
         self.fetch_details(data_object)
-        output(f"\nDownloading {data_object.option_string}...", color=BColors.BOLD)
+        output(
+            f"\nDownloading {data_object.option_string}...", color=BColors.BOLD)
 
         # fetching all parent objects (e.g. if you only download a song)
         if not kwargs.get("fetched_upwards", False):
@@ -280,13 +259,15 @@ class Pages:
         return set(re.findall(r"{([^}]+)}", path_template))
 
     def _parse_path_template(self, path_template: str, naming: Dict[str, List[str]]) -> str:
-        field_names: Set[str] = self._extract_fields_from_template(path_template)
+        field_names: Set[str] = self._extract_fields_from_template(
+            path_template)
 
         for field in field_names:
             if len(naming[field]) == 0:
                 raise MKMissingNameException(f"Missing field for {field}.")
 
-            path_template = path_template.replace(f"{{{field}}}", naming[field][0])
+            path_template = path_template.replace(
+                f"{{{field}}}", naming[field][0])
 
         return path_template
 
@@ -305,7 +286,8 @@ class Pages:
         naming["song"].append(song.title_value)
         naming["isrc"].append(song.isrc)
         naming["album"].extend(a.title_value for a in song.album_collection)
-        naming["album_type"].extend(a.album_type.value for a in song.album_collection)
+        naming["album_type"].extend(
+            a.album_type.value for a in song.album_collection)
         naming["artist"].extend(a.name for a in song.artist_collection)
         naming["artist"].extend(a.name for a in song.feature_artist_collection)
         for a in song.album_collection:
@@ -322,13 +304,16 @@ class Pages:
         song.target_collection.append(Target(
             relative_to_music_dir=True,
             file_path=Path(
-                self._parse_path_template(main_settings["download_path"], naming=naming), 
-                self._parse_path_template(main_settings["download_file"], naming=naming),
+                self._parse_path_template(
+                    main_settings["download_path"], naming=naming),
+                self._parse_path_template(
+                    main_settings["download_file"], naming=naming),
             )
         ))
         for target in song.target_collection:
             if target.exists:
-                output(f'{target.file_path} {BColors.OKGREEN.value}[already exists]', color=BColors.GREY)
+                output(
+                    f'{target.file_path} {BColors.OKGREEN.value}[already exists]', color=BColors.GREY)
                 r.found_on_disk += 1
 
                 if not self.download_options.download_again_if_found:
@@ -349,8 +334,10 @@ class Pages:
                 break
 
             used_source = source
-            streaming_results = source.page.download_song_to_target(source=source, target=tmp, desc="download")
-            skip_intervals = source.page.get_skip_intervals(song=song, source=source)
+            streaming_results = source.page.download_song_to_target(
+                source=source, target=tmp, desc="download")
+            skip_intervals = source.page.get_skip_intervals(
+                song=song, source=source)
 
             # if something has been downloaded but it somehow failed, delete the file
             if streaming_results.is_fatal_error and tmp.exists:
@@ -374,7 +361,8 @@ class Pages:
             used_source.page.post_process_hook(song=song, temp_target=tmp)
 
         if not found_on_disk or self.download_options.process_metadata_if_found:
-            write_metadata_to_target(metadata=song.metadata, target=tmp, song=song)
+            write_metadata_to_target(
+                metadata=song.metadata, target=tmp, song=song)
 
         # copy the tmp target to the final locations
         for target in song.target_collection:
@@ -392,5 +380,3 @@ class Pages:
         _actual_page = self._source_to_page[source.source_type]
 
         return _actual_page, self._page_instances[_actual_page].fetch_object_from_source(source=source, stop_at_level=stop_at_level)
-
-    

music_kraken/objects/artwork.py

@@ -1,83 +1,243 @@
 from __future__ import annotations
 
+from copy import copy
+from dataclasses import dataclass, field
+from functools import cached_property
 from typing import Dict, List, Optional, Set, Tuple, Type, TypedDict, Union
 
+from ..connection import Connection
+from ..utils import create_dataclass_instance, custom_hash
 from ..utils.config import main_settings
+from ..utils.enums import PictureType
 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
+from .target import Target
+from PIL import Image
+
+import imagehash
+
+artwork_connection: Connection = Connection(module="artwork")
 
 
-class ArtworkVariant(TypedDict):
+@dataclass
+class ArtworkVariant:
     url: str
-    width: int
-    height: int
-    deviation: float
+    width: Optional[int] = None
+    heigth: Optional[int] = None
+    image_format: Optional[str] = None
+
+    def __hash__(self) -> int:
+        return custom_hash(self.url)
+
+    def __eq__(self, other: ArtworkVariant) -> bool:
+        return hash(self) == hash(other)
+
+    def __contains__(self, other: str) -> bool:
+        return custom_hash(other) == hash(self.url)
+
+    def __merge__(self, other: ArtworkVariant) -> None:
+        for key, value in other.__dict__.items():
+            if value is None:
+                continue
+
+            if getattr(self, key) is None:
+                setattr(self, key, value)
+
+    @cached_property
+    def target(self) -> Target:
+        return Target.temp()
+
+    def fetch(self) -> None:
+        global artwork_connection
+
+        r = artwork_connection.get(self.url, name=hash_url(self.url))
+        if r is None:
+            return
+
+        self.target.raw_content = r.content
+
+@dataclass
+class Artwork:
+    variants: List[ArtworkVariant] = field(default_factory=list)
+
+    artwork_type: PictureType = PictureType.OTHER
+
+    def search_variant(self, url: str) -> Optional[ArtworkVariant]:
+        if url is None: 
+            return None
+
+        for variant in self.variants:
+            if url in variant:
+                return variant
+
+        return None
+
+    def __contains__(self, other: str) -> bool:
+        return self.search_variant(other) is not None
+
+    def add_data(self, **kwargs) -> None:
+        variant = self.search_variant(kwargs.get("url"))
+
+        if variant is None:
+            variant, kwargs = create_dataclass_instance(ArtworkVariant, kwargs)
+            self.variants.append(variant)
+
+        variant.__dict__.update(kwargs)
+
+    @property
+    def url(self) -> Optional[str]:
+        if len(self.variants) <= 0:
+            return None
+        return self.variants[0].url
+
+    def fetch(self) -> None:
+        for variant in self.variants:
+            variant.fetch()
 
 
 class ArtworkCollection:
-    def __init__(self, *variants: List[ArtworkVariant], parent_artworks: Set[ArtworkCollection] = None, crop_images: bool = True) -> None:
-        self.crop_images: bool = crop_images
+    """
+    Stores all the images/artworks for one data object.
+    
+    There could be duplicates before calling ArtworkCollection.compile()  
+    _this is called before one object is downloaded automatically._
+    """
+
+    artwork_type: PictureType = PictureType.OTHER
+
+    def __init__(
+        self, 
+        *data: List[Artwork], 
+        parent_artworks: Set[ArtworkCollection] = None, 
+        crop_images: bool = True,
+    ) -> None:
+        # this is used for the song artwork, to fall back to the song artwork
         self.parent_artworks: Set[ArtworkCollection] = parent_artworks or set()
+        self.crop_images: bool = crop_images
         
-        self._variant_mapping: Dict[str, ArtworkVariant] = {}
+        self._data = []
+        self.extend(data)
 
-        for variant in variants:
-            self.append(**variant)
+    def search_artwork(self, url: str) -> Optional[ArtworkVariant]:
+        for artwork in self._data:
+            if url in artwork:
+                return artwork
 
-    @staticmethod
-    def _calculate_deviation(*dimensions: List[int]) -> float:
-        return sum(abs(d - main_settings["preferred_artwork_resolution"]) for d in dimensions) / len(dimensions)
+        return None
     
-    def append(self, url: str, width: int = main_settings["preferred_artwork_resolution"], height: int = main_settings["preferred_artwork_resolution"], **kwargs) -> None:
-        if url is None:
+    def __contains__(self, other: str) -> bool:
+        return self.search_artwork(other) is not None
+
+    def _create_new_artwork(self, **kwargs) -> Tuple[Artwork, dict]:
+        kwargs["artwork_type"] = kwargs.get("artwork_type", self.artwork_type)
+
+        return create_dataclass_instance(Artwork, dict(**kwargs))
+
+    def add_data(self, url: str, **kwargs) -> Artwork:
+        kwargs["url"] = url
+
+        artwork = self.search_artwork(url)
+
+        if artwork is None:
+            artwork, kwargs = self._create_new_artwork(**kwargs)
+            self._data.append(artwork)
+
+        artwork.add_data(**kwargs)
+        return artwork
+
+    def append(self, value: Union[Artwork, ArtworkVariant, dict], **kwargs):
+        """
+        You can append the types Artwork, ArtworkVariant or dict 
+        the best option would be to use Artwork and avoid the other options.
+        """
+        if isinstance(value, dict):
+            kwargs.update(value)
+            value, kwargs = create_dataclass_instance(ArtworkVariant, kwargs)
+        
+        if isinstance(value, ArtworkVariant):
+            kwargs["variants"] = [value]
+            value, kwargs = create_dataclass_instance(Artwork, kwargs)
+
+        if isinstance(value, Artwork):
+            self._data.append(value)
             return
     
-        self._variant_mapping[hash_url(url=url)] = {
-            "url": url,
-            "width": width,
-            "height": height,
-            "deviation": self._calculate_deviation(width, height),
-        }
+    def extend(self, values: List[Union[Artwork, ArtworkVariant, dict]], **kwargs):
+        for value in values:
+            self.append(value, **kwargs)
 
-    @property
-    def flat_empty(self) -> bool:
-        return len(self._variant_mapping.keys()) <= 0
+    def compile(self, **kwargs) -> None:
+        """
+        This will make the artworks ready for download and delete duplicates.
+        """
+        artwork_hashes: list = list()
+        artwork_urls: list = list()
+        for artwork in self._data:
+            index = 0
+            for artwork_variant in artwork.variants:
+                r = artwork_connection.get(
+                    url=artwork_variant.url,
+                    name=artwork_variant.url,
+                )
 
-    def _get_best_from_list(self, artwork_variants: List[ArtworkVariant]) -> Optional[ArtworkVariant]:
-        return min(artwork_variants, key=lambda x: x["deviation"])
+                if artwork_variant.url in artwork_urls:
+                    artwork.variants.pop(index)
+                    continue
+                artwork_urls.append(artwork_variant.url)
+
+                target: Target = artwork_variant.target
+                with target.open("wb") as f:
+                    f.write(r.content)
+
+                with Image.open(target.file_path) as img:
+                    # https://stackoverflow.com/a/59476938/16804841
+                    if img.mode != 'RGB':
+                        img = img.convert('RGB')
+                    
+                    try:
+                        image_hash = imagehash.crop_resistant_hash(img)
+                    except Exception as e:
+                        continue
+
+                    if image_hash in artwork_hashes:
+                        artwork.variants.pop(index)
+                        target.delete()
+                        continue
+                    artwork_hashes.append(image_hash)
+                    width, height = img.size
+                    if width != height:
+                        if width > height:
+                            img = img.crop((width // 2 - height // 2, 0, width // 2 + height // 2, height))
+                        else:
+                            img = img.crop((0, height // 2 - width // 2, width, height // 2 + width // 2))
+
+                        # resize the image to the preferred resolution
+                        img.thumbnail((main_settings["preferred_artwork_resolution"], main_settings["preferred_artwork_resolution"]))
+                        index =+ 1
                     
-    @property
-    def best_variant(self) -> ArtworkVariant:
-        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: ArtworkCollection, **kwargs) -> None:
         self.parent_artworks.update(other.parent_artworks)
+        for other_artwork in other._data:
+            for other_variant in other_artwork.variants:
+                if self.__contains__(other_variant.url):
+                    continue
+                self.append(ArtworkVariant(other_variant.url))
 
-        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: ArtworkCollection) -> bool:
-        if hash(self) == hash(other):
-            return True
+    def __iter__(self) -> Generator[Artwork, None, None]:
+        yield from self._data
+
+    def get_urls(self) -> Generator[str, None, None]:
+        yield from (artwork.url for artwork in self._data if artwork.url is not None)
 
-        if not isinstance(other, ArtworkCollection):
-            return False
-        return any(a == b for a, b in zip(self._variant_mapping.keys(), other._variant_mapping.keys()))
-
-    def __iter__(self) -> Generator[ArtworkVariant, None, None]:
-        yield from self._variant_mapping.values()
     
     

music_kraken/objects/song.py

@@ -184,6 +184,10 @@ class Song(Base):
             self.album_collection.extend(object_list)
             return
 
+    def _compile(self):
+        self.artwork.compile()
+
+
     INDEX_DEPENDS_ON = ("title", "isrc", "source_collection")
 
     @property

music_kraken/objects/target.py

@@ -1,17 +1,17 @@
 from __future__ import annotations
 
-from pathlib import Path
-from typing import List, Tuple, TextIO, Union, Optional
 import logging
 import random
+from pathlib import Path
+from typing import List, Optional, TextIO, Tuple, Union
+
 import requests
 from tqdm import tqdm
 
-from .parents import OuterProxy
+from ..utils.config import logging_settings, main_settings
 from ..utils.shared import HIGHEST_ID
-from ..utils.config import main_settings, logging_settings
 from ..utils.string_processing import fit_to_file_system
-
+from .parents import OuterProxy
 
 LOGGER = logging.getLogger("target")
 
@@ -31,7 +31,8 @@ class Target(OuterProxy):
     }
 
     @classmethod
-    def temp(cls, name: str = str(random.randint(0, HIGHEST_ID)), file_extension: Optional[str] = None) -> P:
+    def temp(cls, name: str = None, file_extension: Optional[str] = None) -> P:
+        name = name or str(random.randint(0, HIGHEST_ID))
         if file_extension is not None:
             name = f"{name}.{file_extension}"
 
@@ -117,3 +118,11 @@ class Target(OuterProxy):
 
     def read_bytes(self) -> bytes:
         return self.file_path.read_bytes()
+
+    @property
+    def raw_content(self) -> bytes:
+        return self.file_path.read_bytes()
+
+    @raw_content.setter
+    def raw_content(self, content: bytes):
+        self.file_path.write_bytes(content)

music_kraken/pages/bandcamp.py

@@ -231,7 +231,7 @@ class Bandcamp(Page):
         # artist artwork
         artist_artwork: BeautifulSoup = soup.find("img", {"class":"band-photo"})
         if artist_artwork is not None:
-            artist.artwork.append(artist_artwork.get("data-src", artist_artwork.get("src")))
+            artist.artwork.add_data(artist_artwork.get("data-src", artist_artwork.get("src")))
         
         for i, data_blob_soup in enumerate(soup.find_all("div", {"id": ["pagedata", "collectors-data"]})):
             data_blob = data_blob_soup["data-blob"]
@@ -308,12 +308,12 @@ class Bandcamp(Page):
 
         _artwork_url = _get_artwork_url(data)
         if _artwork_url is not None:
-            artwork.append(url=_artwork_url, width=350, height=350)
+            artwork.add_data(url=_artwork_url, width=350, height=350)
         else:
             for album_release in data.get("albumRelease", []):
                 _artwork_url = _get_artwork_url(album_release)
                 if _artwork_url is not None:
-                    artwork.append(url=_artwork_url, width=350, height=350)
+                    artwork.add_data(url=_artwork_url, width=350, height=350)
                     break
 
         for i, track_json in enumerate(data.get("track", {}).get("itemListElement", [])):

music_kraken/pages/genius.py

@@ -1,4 +1,5 @@
-import json
+import simplejson as json
+from json_unescape import escape_json, unescape_json
 from enum import Enum
 from typing import List, Optional, Type
 from urllib.parse import urlencode, urlparse, urlunparse
@@ -51,21 +52,21 @@ class Genius(Page):
         
         url_frags = url.split(".")
         if len(url_frags) < 2:
-            artwork.append(url=url)
+            artwork.add_data(url=url)
             return
 
         dimensions = url_frags[-2].split("x")
         if len(dimensions) < 2:
-            artwork.append(url=url)
+            artwork.add_data(url=url)
             return
 
         if len(dimensions) == 3:
             dimensions = dimensions[:-1]
         
         try:
-            artwork.append(url=url, width=int(dimensions[0]), height=int(dimensions[1]))
+            artwork.add_data(url=url, width=int(dimensions[0]), height=int(dimensions[1]))
         except ValueError:
-            artwork.append(url=url)
+            artwork.add_data(url=url)
 
     def parse_api_object(self, data: dict) -> Optional[DatabaseObject]:
         if data is None:
@@ -268,7 +269,8 @@ class Genius(Page):
         # 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("\\'", "'")
+            #IMPLEMENT FIX FROM HAZEL
+            content = escape_json(content)
             data = json.loads(content) 
 
             lyrics_html = traverse_json_path(data, "songPage.lyricsData.body.html", default=None)

music_kraken/pages/musify.py

@@ -8,9 +8,10 @@ import pycountry
 from bs4 import BeautifulSoup
 
 from ..connection import Connection
-from ..objects import (Album, Artist, ArtworkCollection, DatabaseObject,
+from ..objects import (Album, Artist, DatabaseObject,
                        FormattedText, ID3Timestamp, Label, Lyrics, Song,
                        Source, Target)
+from ..objects.artwork import (Artwork, ArtworkVariant, ArtworkCollection)
 from ..utils import shared, string_processing
 from ..utils.config import logging_settings, main_settings
 from ..utils.enums import ALL_SOURCE_TYPES, SourceType
@@ -479,7 +480,7 @@ class Musify(Page):
         artwork: ArtworkCollection = ArtworkCollection()
         album_image_element_list: List[BeautifulSoup] = soup.find_all("img", {"class": "album-img"})
         for album_image_element in album_image_element_list:
-            artwork.append(url=album_image_element.get("data-src", album_image_element.get("src")))
+            artwork.add_data(url=album_image_element.get("data-src", album_image_element.get("src")))
         
         # lyrics
         lyrics_container: List[BeautifulSoup] = soup.find_all("div", {"id": "tabLyrics"})
@@ -748,7 +749,7 @@ class Musify(Page):
         album_artwork: ArtworkCollection = ArtworkCollection()
         album_artwork_list: List[BeautifulSoup] = soup.find_all("img", {"class":"artist-img"})
         for album_artwork in album_artwork_list:
-            album_artwork.append(url=album_artwork.get("data-src", album_artwork.get("src")))
+            album_artwork.add_data(url=album_artwork.get("data-src", album_artwork.get("src")))
 
         return Album(
             title=name,
@@ -917,7 +918,7 @@ class Musify(Page):
         main_artist_artwork: ArtworkCollection = ArtworkCollection()
         artist_image_element_list: List[BeautifulSoup] = soup.find_all("img", {"class":"artist-img"})
         for artist_image_element in artist_image_element_list:
-            main_artist_artwork.append(url=artist_image_element.get("data-src", artist_image_element.get("src")))
+            main_artist_artwork.add_data(url=artist_image_element.get("data-src", artist_image_element.get("src")))
 
         return Artist(
             name=name,
@@ -1069,7 +1070,7 @@ class Musify(Page):
             gallery_body_content: BeautifulSoup = artwork_gallery.find(id="bodyContent")
             gallery_image_element_list: List[BeautifulSoup] = gallery_body_content.find_all("img")
             for gallery_image_element in gallery_image_element_list:
-                artist.artwork.append(url=gallery_image_element.get("data-src", gallery_image_element.get("src")), width=247, heigth=247)
+                artist.artwork.append(ArtworkVariant(url=gallery_image_element.get("data-src", gallery_image_element.get("src")), width=247, heigth=247))
 
 
     def fetch_artist(self, source: Source, **kwargs) -> Artist:

music_kraken/pages/youtube_music/youtube_music.py

@@ -441,7 +441,7 @@ class YoutubeMusic(SuperYouTube):
         # fetch artist artwork
         artist_thumbnails = musicImmersiveHeaderRenderer.get("thumbnail", {}).get("musicThumbnailRenderer", {}).get("thumbnail", {}).get("thumbnails", {})
         for artist_thumbnail in artist_thumbnails:
-            artist.artwork.append(**artist_thumbnail)
+            artist.artwork.append(artist_thumbnail)
 
         if DEBUG:
             for i, content in enumerate(renderer_list):
@@ -493,7 +493,7 @@ class YoutubeMusic(SuperYouTube):
         # album artwork
         album_thumbnails = musicDetailHeaderRenderer.get("thumbnail", {}).get("croppedSquareThumbnailRenderer", {}).get("thumbnail", {}).get("thumbnails", {})
         for album_thumbnail in album_thumbnails:
-            album.artwork.append(**album_thumbnail)
+            album.artwork.append(value=album_thumbnail)
 
         title_runs: List[dict] = musicDetailHeaderRenderer.get("title", {}).get("runs", [])
         subtitle_runs: List[dict] = musicDetailHeaderRenderer.get("subtitle", {}).get("runs", [])
@@ -683,7 +683,7 @@ class YoutubeMusic(SuperYouTube):
             for album in song.album_list:
                 album.album_type = AlbumType.LIVE_ALBUM
         for thumbnail in video_details.get("thumbnails", []):
-            song.artwork.append(**thumbnail)
+            song.artwork.add_data(**thumbnail)
 
         song.lyrics_collection.append(self.fetch_lyrics(browse_id, playlist_id=request_data.get("playlistId")))
 

music_kraken/utils/__init__.py

@@ -1,15 +1,18 @@
-from datetime import datetime
-from pathlib import Path
+import inspect
 import json
 import logging
-import inspect
-from typing import List, Union
+from datetime import datetime
+from functools import lru_cache
+from pathlib import Path
+from typing import Any, List, Union
 
-from .shared import DEBUG, DEBUG_LOGGING, DEBUG_DUMP, DEBUG_TRACE, DEBUG_OBJECT_TRACE, DEBUG_OBJECT_TRACE_CALLSTACK
 from .config import config, read_config, write_config
 from .enums.colors import BColors
-from .path_manager import LOCATIONS
 from .hacking import merge_args
+from .path_manager import LOCATIONS
+from .shared import (DEBUG, DEBUG_DUMP, DEBUG_LOGGING, DEBUG_OBJECT_TRACE,
+                     DEBUG_OBJECT_TRACE_CALLSTACK, DEBUG_TRACE, URL_PATTERN)
+from .string_processing import hash_url, is_url, unify
 
 """
 IO functions
@@ -126,3 +129,34 @@ def get_current_millis() -> int:
 
 def get_unix_time() -> int:
     return int(datetime.now().timestamp())
+
+
+@lru_cache
+def custom_hash(value: Any) -> int:
+    if is_url(value):
+        value = hash_url(value)
+    elif isinstance(value, str):
+        try:
+            value = int(value)
+        except ValueError:
+            value = unify(value)
+    
+    return hash(value)
+
+
+def create_dataclass_instance(t, data: dict):
+    """Creates an instance of a dataclass with the given data.
+    It filters out all data key, which has no attribute in the dataclass.
+
+    Args:
+        t (Type): The dataclass type class
+        data (dict): the attribute to pass into the constructor
+    
+    Returns:
+        Tuple[Type, dict]: The created instance and a dict, containing the data, which was not used in the creation
+    """
+    
+    needed_data = {k: v for k, v in data.items() if k in t.__dataclass_fields__}
+    removed_data = {k: v for k, v in data.items() if k not in t.__dataclass_fields__}
+
+    return t(**needed_data), removed_data

music_kraken/utils/enums/__init__.py

@@ -1,7 +1,11 @@
 from __future__ import annotations
 
 from dataclasses import dataclass
-from typing import Optional, TYPE_CHECKING, Type
+from enum import Enum
+from typing import TYPE_CHECKING, Optional, Type
+
+from mutagen.id3 import PictureType
+
 if TYPE_CHECKING:
     from ...pages.abstract import Page
 
@@ -52,3 +56,73 @@ class ALL_SOURCE_TYPES:
     MANUAL = SourceType(name="manual")
     
     PRESET = SourceType(name="preset")
+
+
+class PictureType(Enum):
+    """Enumeration of image types defined by the ID3 standard for the APIC
+    frame, but also reused in WMA/FLAC/VorbisComment.
+
+    This is copied from mutagen.id3.PictureType
+    """
+
+    OTHER = 0
+
+    FILE_ICON = 1
+    """32x32 pixels 'file icon' (PNG only)"""
+
+    OTHER_FILE_ICON = 2
+    """Other file icon"""
+
+    COVER_FRONT = 3
+    """Cover (front)"""
+
+    COVER_BACK = 4
+    """Cover (back)"""
+
+    LEAFLET_PAGE = 5
+    """Leaflet page"""
+
+    MEDIA = 6
+    """Media (e.g. label side of CD)"""
+
+    LEAD_ARTIST = 7
+    """Lead artist/lead performer/soloist"""
+
+    ARTIST = 8
+    """Artist/performer"""
+
+    CONDUCTOR = 9
+    """Conductor"""
+
+    BAND = 10
+    """Band/Orchestra"""
+
+    COMPOSER = 11
+    """Composer"""
+
+    LYRICIST = 12
+    """Lyricist/text writer"""
+
+    RECORDING_LOCATION = 13
+    """Recording Location"""
+
+    DURING_RECORDING = 14
+    """During recording"""
+
+    DURING_PERFORMANCE = 15
+    """During performance"""
+
+    SCREEN_CAPTURE = 16
+    """Movie/video screen capture"""
+
+    FISH = 17
+    """A bright colored fish"""
+
+    ILLUSTRATION = 18
+    """Illustration"""
+
+    BAND_LOGOTYPE = 19
+    """Band/artist logotype"""
+
+    PUBLISHER_LOGOTYPE = 20
+    """Publisher/Studio logotype"""

music_kraken/utils/string_processing.py

@@ -1,13 +1,15 @@
-from typing import Tuple, Union, Optional
-from pathlib import Path
+import re
 import string
 from functools import lru_cache
+from pathlib import Path
+from typing import Any, Optional, Tuple, Union
+from urllib.parse import ParseResult, parse_qs, urlparse
 
-from transliterate.exceptions import LanguageDetectionError
-from transliterate import translit
 from pathvalidate import sanitize_filename
-from urllib.parse import urlparse, ParseResult, parse_qs
+from transliterate import translit
+from transliterate.exceptions import LanguageDetectionError
 
+from .shared import URL_PATTERN
 
 COMMON_TITLE_APPENDIX_LIST: Tuple[str, ...] = (
     "(official video)",
@@ -229,3 +231,13 @@ def shorten_display_url(url: str, max_length: int = 150, chars_at_end: int = 4,
         return url
     
     return url[:max_length] + shorten_string + url[-chars_at_end:]
+
+def is_url(value: Any) -> bool:
+    if isinstance(value, ParseResult):
+        return True
+    
+    if not isinstance(value, str):
+        return True
+        
+    # value has to be a string
+    return re.match(URL_PATTERN, value) is not None

music_kraken/utils/support_classes/download_result.py

@@ -1,10 +1,14 @@
-from dataclasses import dataclass, field
-from typing import List, Tuple
+from __future__ import annotations
 
-from ...utils.config import main_settings, logging_settings
-from ...utils.enums.colors import BColors
+from dataclasses import dataclass, field
+from typing import TYPE_CHECKING, List, Tuple
+
+if TYPE_CHECKING:
     from ...objects import Target
 
+from ...utils.config import logging_settings, main_settings
+from ...utils.enums.colors import BColors
+
 UNIT_PREFIXES: List[str] = ["", "k", "m", "g", "t"]
 UNIT_DIVISOR = 1024