0
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-30 03:33:02 +00:00
yt-dlp/yt_dlp/extractor/tvc.py
2022-11-17 11:40:17 +05:30

97 lines
3.6 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

from .common import InfoExtractor
from ..utils import (
clean_html,
int_or_none,
)
class TVCIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?tvc\.ru/video/iframe/id/(?P<id>\d+)'
_EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>(?:http:)?//(?:www\.)?tvc\.ru/video/iframe/id/[^"]+)\1']
_TEST = {
'url': 'http://www.tvc.ru/video/iframe/id/74622/isPlay/false/id_stat/channel/?acc_video_id=/channel/brand/id/17/show/episodes/episode_id/39702',
'md5': 'bbc5ff531d1e90e856f60fc4b3afd708',
'info_dict': {
'id': '74622',
'ext': 'mp4',
'title': 'События. "События". Эфир от 22.05.2015 14:30',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 1122,
},
}
def _real_extract(self, url):
video_id = self._match_id(url)
video = self._download_json(
'http://www.tvc.ru/video/json/id/%s' % video_id, video_id)
formats = []
for info in video.get('path', {}).get('quality', []):
video_url = info.get('url')
if not video_url:
continue
format_id = self._search_regex(
r'cdnvideo/([^/]+?)(?:-[^/]+?)?/', video_url,
'format id', default=None)
formats.append({
'url': video_url,
'format_id': format_id,
'width': int_or_none(info.get('width')),
'height': int_or_none(info.get('height')),
'tbr': int_or_none(info.get('bitrate')),
})
return {
'id': video_id,
'title': video['title'],
'thumbnail': video.get('picture'),
'duration': int_or_none(video.get('duration')),
'formats': formats,
}
class TVCArticleIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?tvc\.ru/(?!video/iframe/id/)(?P<id>[^?#]+)'
_TESTS = [{
'url': 'http://www.tvc.ru/channel/brand/id/29/show/episodes/episode_id/39702/',
'info_dict': {
'id': '74622',
'ext': 'mp4',
'title': 'События. "События". Эфир от 22.05.2015 14:30',
'description': 'md5:ad7aa7db22903f983e687b8a3e98c6dd',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 1122,
},
}, {
'url': 'http://www.tvc.ru/news/show/id/69944',
'info_dict': {
'id': '75399',
'ext': 'mp4',
'title': 'Эксперты: в столице встал вопрос о максимально безопасных остановках',
'description': 'md5:f2098f71e21f309e89f69b525fd9846e',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 278,
},
}, {
'url': 'http://www.tvc.ru/channel/brand/id/47/show/episodes#',
'info_dict': {
'id': '2185',
'ext': 'mp4',
'title': 'Ещё не поздно. Эфир от 03.08.2013',
'description': 'md5:51fae9f3f8cfe67abce014e428e5b027',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 3316,
},
}]
def _real_extract(self, url):
webpage = self._download_webpage(url, self._match_id(url))
return {
'_type': 'url_transparent',
'ie_key': 'TVC',
'url': self._og_search_video_url(webpage),
'title': clean_html(self._og_search_title(webpage)),
'description': clean_html(self._og_search_description(webpage)),
'thumbnail': self._og_search_thumbnail(webpage),
}