- 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': info.find('caption').text,
- 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
- }
- else:
- # "feature" and "nightly-news" pages use theplatform.com
- video_id = mobj.group('mpx_id')
- if not video_id.isdigit():
- webpage = self._download_webpage(url, video_id)
- info = None
- bootstrap_json = self._search_regex(
- [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$',
- r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'],
- webpage, 'bootstrap json', default=None)
- bootstrap = self._parse_json(
- bootstrap_json, video_id, transform_source=unescapeHTML)
- if 'results' in bootstrap:
- info = bootstrap['results'][0]['video']
- elif 'video' in bootstrap:
- info = bootstrap['video']
- else:
- info = bootstrap
- video_id = info['mpxId']
-
- return {
- '_type': 'url_transparent',
- 'id': video_id,
- # http://feed.theplatform.com/f/2E2eJC/nbcnews also works
- 'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byId=%s' % video_id,
- 'ie_key': 'ThePlatformFeed',
- }
+ video_id = self._match_id(url)
+ webpage = self._download_webpage(url, video_id)
+
+ data = self._parse_json(self._search_regex(
+ r'window\.__data\s*=\s*({.+});', webpage,
+ 'bootstrap json'), video_id, js_to_json)
+ video_data = try_get(data, lambda x: x['video']['current'], dict)
+ if not video_data:
+ video_data = data['article']['content'][0]['primaryMedia']['video']
+ title = video_data['headline']['primary']
+
+ formats = []
+ for va in video_data.get('videoAssets', []):
+ public_url = va.get('publicUrl')
+ if not public_url:
+ continue
+ if '://link.theplatform.com/' in public_url:
+ public_url = update_url_query(public_url, {'format': 'redirect'})
+ format_id = va.get('format')
+ if format_id == 'M3U':
+ formats.extend(self._extract_m3u8_formats(
+ public_url, video_id, 'mp4', 'm3u8_native',
+ m3u8_id=format_id, fatal=False))
+ continue
+ tbr = int_or_none(va.get('bitrate'), 1000)
+ if tbr:
+ format_id += '-%d' % tbr
+ formats.append({
+ 'format_id': format_id,
+ 'url': public_url,
+ 'width': int_or_none(va.get('width')),
+ 'height': int_or_none(va.get('height')),
+ 'tbr': tbr,
+ 'ext': 'mp4',
+ })
+ self._sort_formats(formats)
+
+ subtitles = {}
+ closed_captioning = video_data.get('closedCaptioning')
+ if closed_captioning:
+ for cc_url in closed_captioning.values():
+ if not cc_url:
+ continue
+ subtitles.setdefault('en', []).append({
+ 'url': cc_url,
+ })
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'description': try_get(video_data, lambda x: x['description']['primary']),
+ 'thumbnail': try_get(video_data, lambda x: x['primaryImage']['url']['primary']),
+ 'duration': parse_duration(video_data.get('duration')),
+ 'timestamp': unified_timestamp(video_data.get('datePublished')),
+ 'formats': formats,
+ 'subtitles': subtitles,
+ }
+
+
+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