- mobj = re.match(self._VALID_URL, url)
- video_id = mobj.group('id')
- if video_id is not None:
- all_info = self._download_xml('http://www.nbcnews.com/id/%s/displaymode/1219' % video_id, video_id)
- info = all_info.find('video')
-
- return {
- 'id': video_id,
- 'title': info.find('headline').text,
- 'ext': 'flv',
- 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
- 'description': compat_str(info.find('caption').text),
- 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
- }
- else:
- # "feature" and "nightly-news" pages use theplatform.com
- title = mobj.group('title')
- webpage = self._download_webpage(url, title)
- bootstrap_json = self._search_regex(
- r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
- webpage, 'bootstrap json', flags=re.MULTILINE)
- bootstrap = self._parse_json(bootstrap_json, video_id)
- info = bootstrap['results'][0]['video']
- mpxid = info['mpxId']
-
- base_urls = [
- info['fallbackPlaylistUrl'],
- info['associatedPlaylistUrl'],
- ]
-
- for base_url in base_urls:
- if not base_url:
- continue
- playlist_url = base_url + '?form=MPXNBCNewsAPI'
-
- try:
- all_videos = self._download_json(playlist_url, title)
- except ExtractorError as ee:
- if isinstance(ee.cause, compat_HTTPError):
- continue
- raise
-
- if not all_videos or 'videos' not in all_videos:
- continue
-
- try:
- info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
- break
- except StopIteration:
- continue
-
- if info is None:
- raise ExtractorError('Could not find video in playlists')
-
- return {
- '_type': 'url',
- # We get the best quality video
- 'url': info['videoAssets'][-1]['publicUrl'],
- 'ie_key': 'ThePlatform',
- }
+ video_id = self._match_id(url)
+ if not video_id.isdigit():
+ webpage = self._download_webpage(url, video_id)
+
+ data = self._parse_json(self._search_regex(
+ r'window\.__data\s*=\s*({.+});', webpage,
+ 'bootstrap json'), video_id)
+ video_id = data['article']['content'][0]['primaryMedia']['video']['mpxMetadata']['id']
+
+ return {
+ '_type': 'url_transparent',
+ 'id': video_id,
+ # http://feed.theplatform.com/f/2E2eJC/nbcnews also works
+ 'url': update_url_query('http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews', {'byId': video_id}),
+ 'ie_key': 'ThePlatformFeed',
+ }
+
+
+class NBCOlympicsIE(InfoExtractor):
+ IE_NAME = 'nbcolympics'
+ _VALID_URL = r'https?://www\.nbcolympics\.com/video/(?P<id>[a-z-]+)'
+
+ _TEST = {
+ # Geo-restricted to US
+ 'url': 'http://www.nbcolympics.com/video/justin-roses-son-leo-was-tears-after-his-dad-won-gold',
+ 'md5': '54fecf846d05429fbaa18af557ee523a',
+ 'info_dict': {
+ 'id': 'WjTBzDXx5AUq',
+ 'display_id': 'justin-roses-son-leo-was-tears-after-his-dad-won-gold',
+ 'ext': 'mp4',
+ 'title': 'Rose\'s son Leo was in tears after his dad won gold',
+ 'description': 'Olympic gold medalist Justin Rose gets emotional talking to the impact his win in men\'s golf has already had on his children.',
+ 'timestamp': 1471274964,
+ 'upload_date': '20160815',
+ 'uploader': 'NBCU-SPORTS',
+ },
+ }
+
+ def _real_extract(self, url):
+ display_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, display_id)
+
+ drupal_settings = self._parse_json(self._search_regex(
+ r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
+ webpage, 'drupal settings'), display_id)
+
+ iframe_url = drupal_settings['vod']['iframe_url']
+ theplatform_url = iframe_url.replace(
+ 'vplayer.nbcolympics.com', 'player.theplatform.com')
+
+ return {
+ '_type': 'url_transparent',
+ 'url': theplatform_url,
+ 'ie_key': ThePlatformIE.ie_key(),
+ 'display_id': display_id,
+ }
+
+
+class NBCOlympicsStreamIE(AdobePassIE):
+ IE_NAME = 'nbcolympics:stream'
+ _VALID_URL = r'https?://stream\.nbcolympics\.com/(?P<id>[0-9a-z-]+)'
+ _TEST = {
+ 'url': 'http://stream.nbcolympics.com/2018-winter-olympics-nbcsn-evening-feb-8',
+ 'info_dict': {
+ 'id': '203493',
+ 'ext': 'mp4',
+ 'title': 're:Curling, Alpine, Luge [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+ },
+ 'params': {
+ # m3u8 download
+ 'skip_download': True,
+ },
+ }
+ _DATA_URL_TEMPLATE = 'http://stream.nbcolympics.com/data/%s_%s.json'
+
+ def _real_extract(self, url):
+ display_id = self._match_id(url)
+ webpage = self._download_webpage(url, display_id)
+ pid = self._search_regex(r'pid\s*=\s*(\d+);', webpage, 'pid')
+ resource = self._search_regex(
+ r"resource\s*=\s*'(.+)';", webpage,
+ 'resource').replace("' + pid + '", pid)
+ event_config = self._download_json(
+ self._DATA_URL_TEMPLATE % ('event_config', pid),
+ pid)['eventConfig']
+ title = self._live_title(event_config['eventTitle'])
+ source_url = self._download_json(
+ self._DATA_URL_TEMPLATE % ('live_sources', pid),
+ pid)['videoSources'][0]['sourceUrl']
+ media_token = self._extract_mvpd_auth(
+ url, pid, event_config.get('requestorId', 'NBCOlympics'), resource)
+ formats = self._extract_m3u8_formats(self._download_webpage(
+ 'http://sp.auth.adobe.com/tvs/v1/sign', pid, query={
+ 'cdn': 'akamai',
+ 'mediaToken': base64.b64encode(media_token.encode()),
+ 'resource': base64.b64encode(resource.encode()),
+ 'url': source_url,
+ }), pid, 'mp4')
+ self._sort_formats(formats)
+
+ return {
+ 'id': pid,
+ 'display_id': display_id,
+ 'title': title,
+ 'formats': formats,
+ 'is_live': True,
+ }
Rapsys Git / youtubedl / blobdiff