music-kraken-core/music_kraken/download/__init__.py

from __future__ import annotations

import logging
import random
import re
from collections import defaultdict
from copy import copy
from dataclasses import dataclass, field
from pathlib import Path
from string import Formatter
from typing import (TYPE_CHECKING, Any, Callable, Dict, Generator, List,
                    Optional, Set, Tuple, Type, TypedDict, Union)

import requests
from bs4 import BeautifulSoup

from ..audio import correct_codec, write_metadata_to_target
from ..connection import Connection
from ..objects import Album, Artist, Collection
from ..objects import DatabaseObject as DataObject
from ..objects import Label, Options, Song, Source, Target
from ..utils import BColors, limit_generator, output, trace
from ..utils.config import main_settings, youtube_settings
from ..utils.enums import ALL_SOURCE_TYPES, SourceType
from ..utils.enums.album import AlbumType
from ..utils.exception import MKComposeException, MKMissingNameException
from ..utils.exception.download import UrlNotFoundException
from ..utils.path_manager import LOCATIONS
from ..utils.shared import DEBUG_PAGES
from ..utils.string_processing import fit_to_file_system
from ..utils.support_classes.download_result import DownloadResult
from ..utils.support_classes.query import Query
from .results import SearchResults


@dataclass
class FetchOptions:
    download_all: bool = False
    album_type_blacklist: Set[AlbumType] = field(default_factory=lambda: set(AlbumType(a) for a in main_settings["album_type_blacklist"]))


@dataclass
class DownloadOptions:
    download_all: bool = False
    album_type_blacklist: Set[AlbumType] = field(default_factory=lambda: set(AlbumType(a) for a in main_settings["album_type_blacklist"]))

    download_again_if_found: bool = False
    process_audio_if_found: bool = False
    process_metadata_if_found: bool = True


fetch_map = {
    Song: "fetch_song",
    Album: "fetch_album",
    Artist: "fetch_artist",
    Label: "fetch_label",
}


class Downloader:
    def __init__(
        self, 
        auto_register_pages: bool = True, 
        download_options: DownloadOptions = None, 
        fetch_options: FetchOptions = None, 
        **kwargs
    ):
        self.LOGGER = logging.getLogger("download")
        
        self.download_options: DownloadOptions = download_options or DownloadOptions()
        self.fetch_options: FetchOptions = fetch_options or FetchOptions()

        self._registered_pages: Dict[Type[Page], Set[Page]] = defaultdict(set)
        if auto_register_pages:
            self.scan_for_pages(**kwargs)

    # manage which pages to use

    def register_page(self, page_type: Type[Page], **kwargs):
        if page_type in self._registered_pages:
            return

        self._registered_pages[page_type].add(page_type(
            download_options=self.download_options, 
            fetch_options=self.fetch_options, 
            **kwargs
        ))

    def deregister_page(self, page_type: Type[Page]):
        if page_type not in _registered_pages:
            return

        for p in self._registered_pages[page_type]:
            p.__del__()
        del self._registered_pages[page_type]

    def scan_for_pages(self, **kwargs):
        # assuming the wanted pages are the leaf classes of the interface
        from .. import pages
        
        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)

        if Page in leaf_classes:
            self.LOGGER.warn("couldn't find any data source")
            return
        for leaf_class in leaf_classes:
            self.register_page(leaf_class, **kwargs)

    def get_pages(self, *page_types: List[Type[Page]]) -> Generator[Page, None, None]:
        if len(page_types) == 0:
            page_types = self._registered_pages.keys()

        for page_type in page_types:
            yield from self._registered_pages[page_type]

    # fetching/downloading data

    def search(self, query: Query) -> Generator[DataObject, None, None]:
        """Yields all data objects that were found by the query.
        Other than `Downloader.search_yield_pages`, this function just yields all data objects.
        This looses the data, where th objects were searched originally, so this might not be the best choice.

        Args:
            query (Query): The query to search for.

        Yields:
            Generator[DataObject, None, None]: A generator that yields all found data objects.
        """

        for page in self.get_pages():
            yield from page.search(query=query)

    def search_yield_pages(self, query: Query, results_per_page: Optional[int] = None) -> Generator[Tuple[Page, Generator[DataObject, None, None]], None, None]:
        """Yields all data objects that were found by the query, grouped by the page they were found on.
        every yield is a tuple of the page and a generator that yields the data objects.
        So this could be how it is used:

        ```python
        for page, data_objects in downloader.search_yield_pages(query):  
            print(f"Found on {page}:")  
            for data_object in data_objects:  
                print(data_object)  
        ```

        Args:
            query (Query): The query to search for.
            results_per_page (Optional[int], optional): If this is set, the generators only yield this amount of data objects per page.

        Yields:
            Generator[Tuple[Page, Generator[DataObject, None, None]], None, None]: yields the page and a generator that yields the data objects.
        """
        
        for page in self.get_pages():
            yield page, limit_generator(page.search(query=query), limit=results_per_page)

    def fetch_details(self, data_object: DataObject, stop_at_level: int = 1, **kwargs) -> DataObject:
        source: Source
        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)
            if new_data_object is not None:
                data_object.merge(new_data_object)

        return data_object

    def fetch_from_source(self, source: Source, **kwargs) -> Optional[DataObject]:
        if not source.has_page:
            return None
        
        source_type = source.page.get_source_type(source=source)
        if source_type is None:
            self.LOGGER.debug(f"Could not determine source type for {source}.")
            return None

        func = getattr(source.page, fetch_map[source_type])
        
        # fetching the data object and marking it as fetched
        data_object: DataObject = func(source=source, **kwargs)
        data_object.mark_as_fetched(source.hash_url)
        return data_object

    def fetch_from_url(self, url: str) -> Optional[DataObject]:
        source = Source.match_url(url, ALL_SOURCE_TYPES.MANUAL)
        if source is None:
            return None
        
        return self.fetch_from_source(source=source)
    
    def _skip_object(self, data_object: DataObject) -> bool:
        if isinstance(data_object, Album):
            if not self.download_options.download_all and data_object.album_type in self.download_options.album_type_blacklist:
                return True
        
        return False

    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)
        
        # fetching all parent objects (e.g. if you only download a song)
        if not kwargs.get("fetched_upwards", False):
            to_fetch: List[DataObject] = [data_object]

            while len(to_fetch) > 0:
                new_to_fetch = []
                for d in to_fetch:
                    if self._skip_object(d):
                        continue

                    self.fetch_details(d)

                    for c in d.get_parent_collections():
                        new_to_fetch.extend(c)

                to_fetch = new_to_fetch
            
            kwargs["fetched_upwards"] = True
        
        # download all children
        download_result: DownloadResult = DownloadResult()
        for c in data_object.get_child_collections():
            for d in c:
                if self._skip_object(d):
                    continue

                download_result.merge(self.download(d, genre, **kwargs))

        # actually download if the object is a song
        if isinstance(data_object, Song):
            """
            TODO
            add the traced artist and album to the naming.
            I am able to do that, because duplicate values are removed later on.
            """

            self._download_song(data_object, naming={
                "genre": [genre],
                "audio_format": [main_settings["audio_format"]],
            })

        return download_result

    def _extract_fields_from_template(self, path_template: str) -> Set[str]:
        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)
        
        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])

        return path_template

    def _download_song(self, song: Song, naming: dict) -> DownloadOptions:
        """
        TODO
        Search the song in the file system.
        """
        r = DownloadResult(total=1)
        
        # pre process the data recursively
        song.compile()
        
        # manage the naming
        naming: Dict[str, List[str]] = defaultdict(list, naming)
        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["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:
            naming["label"].extend([l.title_value for l in a.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))
        song.genre = naming["genre"][0]

        # manage the targets
        tmp: Target = Target.temp(file_extension=main_settings["audio_format"])

        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),
            )
        ))
        for target in song.target_collection:
            if target.exists:
                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:
                    target.copy_content(tmp)
            else:
                target.create_path()
                output(f'{target.file_path}', color=BColors.GREY)

        # this streams from every available source until something succeeds, setting the skip intervals to the values of the according source
        used_source: Optional[Source] = None
        skip_intervals: List[Tuple[float, float]] = []
        for source in song.source_collection.get_sources(source_type_sorting={
            "only_with_page": True,
            "sort_key": lambda page: page.download_priority,
            "reverse": True,
        }):
            if tmp.exists:
                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)

            # if something has been downloaded but it somehow failed, delete the file
            if streaming_results.is_fatal_error and tmp.exists:
                tmp.delete()

        # if everything went right, the file should exist now
        if not tmp.exists:
            if used_source is None:
                r.error_message = f"No source found for {song.option_string}."
            else:
                r.error_message = f"Something went wrong downloading {song.option_string}."
            return r

        # post process the audio
        found_on_disk = used_source is None
        if not found_on_disk or self.download_options.process_audio_if_found:
            correct_codec(target=tmp, skip_intervals=skip_intervals)
            r.sponsor_segments = len(skip_intervals)

        if used_source is not None:
            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)

        # copy the tmp target to the final locations
        for target in song.target_collection:
            tmp.copy_content(target)

        tmp.delete()
        return r

    def fetch_url(self, url: str, **kwargs) -> DataObject:
        source = Source.match_url(url, ALL_SOURCE_TYPES.MANUAL)
        
        if source is None or source.page is None:
            raise UrlNotFoundException(url=url)
        
        return source.page.fetch_object_from_source(source=source, **kwargs)


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)

    @classmethod
    def is_leaf_page(cls) -> bool:
        return len(cls.__subclasses__()) == 0

    def __init__(self, download_options: DownloadOptions = None, fetch_options: FetchOptions = None, **kwargs):
        self.SOURCE_TYPE.register_page(self)
        
        self.download_options: DownloadOptions = download_options or DownloadOptions()
        self.fetch_options: FetchOptions = fetch_options or FetchOptions()

    def __del__(self):
        self.SOURCE_TYPE.deregister_page()

    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
        patterns returning the first matching group.
        In case of failure return a default value or raise a WARNING or a
        RegexNotFoundError, depending on fatal, specifying the field name.
        """

        if isinstance(pattern, str):
            mobj = re.search(pattern, string, flags)
        else:
            for p in pattern:
                mobj = re.search(p, string, flags)
                if mobj:
                    break

        if mobj:
            if group is None:
                # return the first matching group
                return next(g for g in mobj.groups() if g is not None)
            elif isinstance(group, (list, tuple)):
                return tuple(mobj.group(g) for g in group)
            else:
                return mobj.group(group)

        return default

    def get_source_type(self, source: Source) -> Optional[Type[DataObject]]:
        return None

    def get_soup_from_response(self, r: requests.Response) -> BeautifulSoup:
        return BeautifulSoup(r.content, "html.parser")

    # to search stuff
    def search(self, query: Query) -> List[DataObject]:
        music_object = query.music_object

        search_functions = {
            Song: self.song_search,
            Album: self.album_search,
            Artist: self.artist_search,
            Label: self.label_search
        }

        if type(music_object) in search_functions:
            r = search_functions[type(music_object)](music_object)
            if r is not None and len(r) > 0:
                return r

        r = []
        for default_query in query.default_search:
            for single_option in self.general_search(default_query):
                r.append(single_option)

        return r

    def general_search(self, search_query: str) -> List[DataObject]:
        return []

    def label_search(self, label: Label) -> List[Label]:
        return []

    def artist_search(self, artist: Artist) -> List[Artist]:
        return []

    def album_search(self, album: Album) -> List[Album]:
        return []

    def song_search(self, song: Song) -> List[Song]:
        return []

    # to fetch stuff
    def fetch_song(self, source: Source, stop_at_level: int = 1) -> Song:
        return Song()

    def fetch_album(self, source: Source, stop_at_level: int = 1) -> Album:
        return Album()

    def fetch_artist(self, source: Source, stop_at_level: int = 1) -> Artist:
        return Artist()

    def fetch_label(self, source: Source, stop_at_level: int = 1) -> Label:
        return Label()

    # to download stuff
    def get_skip_intervals(self, song: Song, source: Source) -> List[Tuple[float, float]]:
        return []

    def post_process_hook(self, song: Song, temp_target: Target, **kwargs):
        pass

    def download_song_to_target(self, source: Source, target: Target, desc: str = None) -> DownloadResult:
        return DownloadResult()