streamlined/server/entertainment_decider/extractors/collection/youtube.py

from __future__ import annotations

from datetime import datetime, timedelta
import logging
import re
from typing import Dict

from pony import orm # TODO remove
import youtubesearchpython

from ...models import MediaCollection
from ..generic import ExtractedData, ExtractionError, SuitableLevel
from ..media import media_extract_uri
from .base import CollectionExtractor


class YouTubeCollectionExtractor(CollectionExtractor[Dict]):

    __uri_regex = re.compile(r"^https?://(www\.)?youtube\.com/(channel/|playlist\?list=)(?P<id>[^/&?]+)")

    @classmethod
    def __get_id(cls, uri: str) -> str:
        m = cls.__uri_regex.search(uri)
        if not m:
            raise Exception(f"Failed to parse Youtube collection uri {uri!r}")
        return m.group("id")

    @staticmethod
    def __is_channel_id(collection_id: str) -> bool:
        return collection_id.startswith("UC") or collection_id.startswith("UU")

    @staticmethod
    def __convert_channel_id(channel_id: str) -> str:
        if channel_id.startswith("UU"):
            return channel_id
        if channel_id.startswith("UC"):
            return f"UU{channel_id[2:]}"
        raise Exception(f"Got not valid channel id: {channel_id!r}")

    @classmethod
    def __convert_if_required(cls, collection_id: str) -> str:
        if cls.__is_channel_id(collection_id):
            return cls.__convert_channel_id(collection_id)
        return collection_id

    def __init__(self):
        super().__init__("youtube")

    def uri_suitable(self, uri: str) -> SuitableLevel:
        return SuitableLevel.ALWAYS if self.__uri_regex.match(uri) else SuitableLevel.NO

    def can_extract_offline(self, uri: str) -> bool:
        return True

    def _cache_expired(self, date: datetime) -> bool:
        return (datetime.now() - date) < timedelta(hours=4)

    def _extract_offline(self, uri: str) -> ExtractedData[Dict]:
        playlist_id = self.__convert_if_required(self.__get_id(uri))
        return ExtractedData(
            extractor_name=self.name,
            object_key=playlist_id,
            object_uri=uri,
        )

    def _extract_online(self, uri: str) -> ExtractedData[Dict]:
        playlist_id = self.__convert_if_required(self.__get_id(uri))
        playlist_link = f"https://www.youtube.com/playlist?list={playlist_id}"
        logging.info(f"Request Youtube playlist {playlist_link!r}")
        playlist = youtubesearchpython.Playlist(playlist_link)
        while playlist.hasMoreVideos:
            playlist.getNextVideos()
        logging.debug(f"Retrieved {len(playlist.videos)} videos from playlist {playlist_link!r}")
        return ExtractedData(
            extractor_name=self.name,
            object_key=playlist_id,
            object_uri=uri,
            data={
                "info": playlist.info["info"],
                "videos": playlist.videos,
            },
        )

    def _update_object_raw(self, object: MediaCollection, data: Dict):
        info = data["info"]
        is_channel = self.__is_channel_id(info["id"])
        object.title = f"[channel] [{self.name}] {info['channel']['name']}" if is_channel else f"{info['title']} ({info['channel']['name']})"
        object.add_uris((info["link"],))
        video_list = data["videos"]
        if object.watch_in_order_auto:
            object.watch_in_order = not is_channel
        len_video_list = len(video_list)
        if is_channel:
            video_list = reversed(video_list)
        for index, video in enumerate(video_list):
            video_url = f"https://www.youtube.com/watch?v={video['id']}"
            other_urls = [
                f"https://youtube.com/watch?v={video['id']}",
                f"https://youtu.be/{video['id']}",
            ]
            logging.debug(f"[youtube] Add to collection {object.title!r} video {video_url!r} ({index+1} of {len_video_list})")
            try:
                element = media_extract_uri(video_url)
                element.add_uris(other_urls)
                object.add_episode(element, episode=index+1)
                orm.commit() # so progress is stored
            except ExtractionError:
                logging.warning(f"Failed while extracting media {video_url!r}", exc_info=True)