You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
yt-dlp/yt_dlp/extractor/gazeta.py

45 lines
1.9 KiB
Python

This file contains ambiguous Unicode characters!

This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.

from .common import InfoExtractor
class GazetaIE(InfoExtractor):
_WORKING = False
_VALID_URL = r'(?P<url>https?://(?:www\.)?gazeta\.ru/(?:[^/]+/)?video/(?:main/)*(?:\d{4}/\d{2}/\d{2}/)?(?P<id>[A-Za-z0-9-_.]+)\.s?html)'
_TESTS = [{
'url': 'http://www.gazeta.ru/video/main/zadaite_vopros_vladislavu_yurevichu.shtml',
'md5': 'd49c9bdc6e5a7888f27475dc215ee789',
'info_dict': {
'id': '205566',
'ext': 'mp4',
'title': '«7080 процентов гражданских в Донецке на грани голода»',
'description': 'md5:38617526050bd17b234728e7f9620a71',
'thumbnail': r're:^https?://.*\.jpg',
},
'skip': 'video not found',
}, {
'url': 'http://www.gazeta.ru/lifestyle/video/2015/03/08/master-klass_krasivoi_byt._delaem_vesennii_makiyazh.shtml',
'only_matching': True,
}, {
'url': 'http://www.gazeta.ru/video/main/main/2015/06/22/platit_ili_ne_platit_po_isku_yukosa.shtml',
'md5': '37f19f78355eb2f4256ee1688359f24c',
'info_dict': {
'id': '252048',
'ext': 'mp4',
'title': '"Если по иску ЮКОСа придется платить, это будет большой удар по бюджету"',
},
'add_ie': ['EaglePlatform'],
}]
def _real_extract(self, url):
mobj = self._match_valid_url(url)
display_id = mobj.group('id')
embed_url = '{}?p=embed'.format(mobj.group('url'))
embed_page = self._download_webpage(
embed_url, display_id, 'Downloading embed page')
video_id = self._search_regex(
r'<div[^>]*?class="eagleplayer"[^>]*?data-id="([^"]+)"', embed_page, 'video id')
return self.url_result(
f'eagleplatform:gazeta.media.eagleplatform.com:{video_id}', 'EaglePlatform')