- webpage = self._download_webpage(url, video_id)
-
- iframe_url = self._html_search_regex(
- r'<iframe[^>]+src="((?:https?:)?//(?:www\.youtube\.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"',
- webpage, 'video iframe', default=None)
- if iframe_url:
- return self.url_result(iframe_url)
-
- config = self._parse_json(self._html_search_regex(
- r'data-(?:player-)?config="([^"]+)"', webpage,
- 'data player config', default='{}'),
- video_id)
-
- if config.get('source_type') == 'vine':
- return self.url_result(config['player_url'], 'Vine')
-
- periscope_url = PeriscopeIE._extract_url(webpage)
- if periscope_url:
- return self.url_result(periscope_url, PeriscopeIE.ie_key())
-
- video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source')
-
- if video_url:
- if determine_ext(video_url) == 'm3u8':
- formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls'))
+ urls = [url]
+ if path.startswith('cards/'):
+ urls.append('https://twitter.com/i/videos/' + video_id)
+
+ for u in urls:
+ webpage = self._download_webpage(
+ u, video_id, headers={'Referer': 'https://twitter.com/'})
+
+ iframe_url = self._html_search_regex(
+ r'<iframe[^>]+src="((?:https?:)?//(?:www\.youtube\.com/embed/[^"]+|(?:www\.)?vine\.co/v/\w+/card))"',
+ webpage, 'video iframe', default=None)
+ if iframe_url:
+ return self.url_result(iframe_url)
+
+ config = self._parse_json(self._html_search_regex(
+ r'data-(?:player-)?config="([^"]+)"', webpage,
+ 'data player config', default='{}'),
+ video_id)
+
+ if config.get('source_type') == 'vine':
+ return self.url_result(config['player_url'], 'Vine')
+
+ periscope_url = PeriscopeIE._extract_url(webpage)
+ if periscope_url:
+ return self.url_result(periscope_url, PeriscopeIE.ie_key())
+
+ video_url = config.get('video_url') or config.get('playlist', [{}])[0].get('source')
+
+ if video_url:
+ if determine_ext(video_url) == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls'))
+ else:
+ f = {
+ 'url': video_url,
+ }
+
+ self._search_dimensions_in_video_url(f, video_url)
+
+ formats.append(f)
+
+ vmap_url = config.get('vmapUrl') or config.get('vmap_url')
+ if vmap_url:
+ formats.extend(
+ self._extract_formats_from_vmap_url(vmap_url, video_id))
+
+ media_info = None
+
+ for entity in config.get('status', {}).get('entities', []):
+ if 'mediaInfo' in entity:
+ media_info = entity['mediaInfo']
+
+ if media_info:
+ formats.extend(self._parse_media_info(media_info, video_id))
+ duration = float_or_none(media_info.get('duration', {}).get('nanos'), scale=1e9)
+
+ username = config.get('user', {}).get('screen_name')
+ if username:
+ formats.extend(self._extract_mobile_formats(username, video_id))
+
+ if formats:
+ title = self._search_regex(r'<title>([^<]+)</title>', webpage, 'title')
+ thumbnail = config.get('posterImageUrl') or config.get('image_src')
+ duration = float_or_none(config.get('duration'), scale=1000) or duration
+ break
+
+ if not formats:
+ headers = {
+ 'Authorization': 'Bearer AAAAAAAAAAAAAAAAAAAAAPYXBAAAAAAACLXUNDekMxqa8h%2F40K4moUkGsoc%3DTYfbDKbT3jJPCEVnMYqilB28NHfOPqkca3qaAxGfsyKCs0wRbw',
+ 'Referer': url,
+ }
+ ct0 = self._get_cookies(url).get('ct0')
+ if ct0:
+ headers['csrf_token'] = ct0.value
+ guest_token = self._download_json(
+ '%s/guest/activate.json' % self._API_BASE, video_id,
+ 'Downloading guest token', data=b'',
+ headers=headers)['guest_token']
+ headers['x-guest-token'] = guest_token
+ self._set_cookie('api.twitter.com', 'gt', guest_token)
+ config = self._download_json(
+ '%s/videos/tweet/config/%s.json' % (self._API_BASE, video_id),
+ video_id, headers=headers)
+ track = config['track']
+ vmap_url = track.get('vmapUrl')
+ if vmap_url:
+ formats = self._extract_formats_from_vmap_url(vmap_url, video_id)
Rapsys Git / youtubedl / blobdiff