0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-29 01:51:06 +00:00
yt-dlp/yt_dlp/extractor/newsy.py
sepro add96eb9f8
[cleanup] Add more ruff rules (#10149)
Authored by: seproDev

Reviewed-by: bashonly <88596187+bashonly@users.noreply.github.com>
Reviewed-by: Simon Sawicki <contact@grub4k.xyz>
2024-06-12 01:09:58 +02:00

47 lines
1.8 KiB
Python

from .common import InfoExtractor
from ..utils import (
js_to_json,
merge_dicts,
)
class NewsyIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?newsy\.com/stories/(?P<id>[^/?#$&]+)'
_TESTS = [{
'url': 'https://www.newsy.com/stories/nft-trend-leads-to-fraudulent-art-auctions/',
'info_dict': {
'id': '609d65125b086c24fb529312',
'ext': 'mp4',
'title': 'NFT Art Auctions Have A Piracy Problem',
'description': 'md5:971e52ab8bc97e50305475cde8284c83',
'display_id': 'nft-trend-leads-to-fraudulent-art-auctions',
'timestamp': 1621339200,
'duration': 339630,
'thumbnail': 'https://cdn.newsy.com/images/videos/x/1620927824_xyrrP4.jpg',
'upload_date': '20210518',
},
'params': {'skip_download': True},
}]
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
data_json = self._parse_json(self._html_search_regex(
r'data-video-player\s?=\s?"({[^"]+})">', webpage, 'data'), display_id, js_to_json)
ld_json = self._search_json_ld(webpage, display_id, fatal=False)
formats, subtitles = [], {}
if data_json.get('stream'):
fmts, subs = self._extract_m3u8_formats_and_subtitles(data_json['stream'], display_id)
formats.extend(fmts)
subtitles = self._merge_subtitles(subtitles, subs)
return merge_dicts(ld_json, {
'id': data_json['id'],
'display_id': display_id,
'title': data_json.get('headline'),
'duration': data_json.get('duration'),
'thumbnail': data_json.get('image'),
'formats': formats,
'subtitles': subtitles,
})