62 lines
2.0 KiB
Python
62 lines
2.0 KiB
Python
import csv
|
|
import json
|
|
import logging
|
|
import subprocess
|
|
import datetime
|
|
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 = []
|
|
watch_history: bool = False
|
|
|
|
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."""
|
|
if self.watch_history:
|
|
url = 'https://www.youtube.com/feed/history'
|
|
ytdlp_args = [
|
|
'yt-dlp',
|
|
'--dump-json',
|
|
url,
|
|
]
|
|
else:
|
|
url = f'https://www.youtube.com/playlist?list={PLAYLIST_ID}'
|
|
ytdlp_args = [
|
|
'yt-dlp',
|
|
'--flat-playlist',
|
|
'--dump-json',
|
|
url,
|
|
]
|
|
result = subprocess.run(
|
|
ytdlp_args,
|
|
capture_output=True,
|
|
text=True,
|
|
)
|
|
|
|
if result.returncode != 0:
|
|
raise RuntimeError(f'Non-zero returncode in command: {result.returncode}\n\n{result.stderr}')
|
|
|
|
output = []
|
|
for line in result.stdout.splitlines():
|
|
data = json.loads(line)
|
|
if self.watch_history:
|
|
if 'thumbnails' in data and data['thumbnails']:
|
|
data['thumbnail'] = data['thumbnails'][-1]['url']
|
|
if 'timestamp' in data:
|
|
data['watch_datetime'] = datetime.datetime.fromtimestamp(int(data['timestamp'])).isoformat()
|
|
else:
|
|
data['thumbnail'] = data['thumbnails'][-1]['url']
|
|
safe_del(data, '_type', '_version', 'thumbnails')
|
|
output.append(data)
|
|
return output
|