import csv
import json
import logging
import subprocess
from dataclasses import dataclass

from personal_data.data import DeduplicateMode, Scraper
from ..util import safe_del

logger = logging.getLogger(__name__)

PLAYLIST_ID='PLAfDVJvDKCvOMvfoTL7eW8GkWNJwd90eV'
#PLAYLIST_ID='LL'

@dataclass(frozen=True)
class YoutubeFavoritesScraper(Scraper):
    dataset_name: str = 'youtube_favorites'
    deduplicate_mode: DeduplicateMode = DeduplicateMode.BY_ALL_COLUMNS
    deduplicate_ignore_columns = []

    def scrape(self) -> list[dict]:
        """Use yt-dlp to fetch the list of favorited videos. This is a placeholder for invoking yt-dlp and parsing its output."""
        result = subprocess.run(
            [
                'yt-dlp',
                '--flat-playlist',
                '--dump-json',
                f'https://www.youtube.com/playlist?list={PLAYLIST_ID}',
            ],
            capture_output=True,
            text=True,
        )

        if result.returncode != 0:
            raise RuntimeError(f'Non-zero returncode in command: {result.returncode}\n\n{result.stderr}')

        for line in result.stdout.splitlines():
            data = json.loads(line)
            data['thumbnail'] = data['thumbnails'][-1]['url']
            safe_del(data, '_type', '_version', 'thumbnails')
            yield data