+ webpage = self._download_webpage(url, playlist_id)
+
+ json_ld_info = self._search_json_ld(webpage, playlist_id, default={})
+ timestamp = json_ld_info.get('timestamp')
+
+ playlist_title = json_ld_info.get('title')
+ if not playlist_title:
+ playlist_title = self._og_search_title(
+ webpage, default=None) or self._html_search_regex(
+ r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
+ if playlist_title:
+ playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
+
+ playlist_description = json_ld_info.get(
+ 'description') or self._og_search_description(webpage, default=None)
+
+ if not timestamp:
+ timestamp = parse_iso8601(self._search_regex(
+ [r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"',
+ r'itemprop="datePublished"[^>]+datetime="([^"]+)"',
+ r'"datePublished":\s*"([^"]+)'],
+ webpage, 'date', default=None))
+
+ entries = []
+
+ # article with multiple videos embedded with playlist.sxml (e.g.
+ # http://www.bbc.com/sport/0/football/34475836)
+ playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
+ playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
+ if playlists:
+ entries = [
+ self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
+ for playlist_url in playlists]
+
+ # news article with multiple videos embedded with data-playable
+ data_playables = re.findall(r'data-playable=(["\'])({.+?})\1', webpage)
+ if data_playables:
+ for _, data_playable_json in data_playables:
+ data_playable = self._parse_json(
+ unescapeHTML(data_playable_json), playlist_id, fatal=False)
+ if not data_playable:
+ continue
+ settings = data_playable.get('settings', {})
+ if settings:
+ # data-playable with video vpid in settings.playlistObject.items (e.g.
+ # http://www.bbc.com/news/world-us-canada-34473351)
+ playlist_object = settings.get('playlistObject', {})
+ if playlist_object:
+ items = playlist_object.get('items')
+ if items and isinstance(items, list):
+ title = playlist_object['title']
+ description = playlist_object.get('summary')
+ duration = int_or_none(items[0].get('duration'))
+ programme_id = items[0].get('vpid')
+ formats, subtitles = self._download_media_selector(programme_id)
+ self._sort_formats(formats)
+ entries.append({
+ 'id': programme_id,
+ 'title': title,
+ 'description': description,
+ 'timestamp': timestamp,
+ 'duration': duration,
+ 'formats': formats,
+ 'subtitles': subtitles,
+ })
+ else:
+ # data-playable without vpid but with a playlist.sxml URLs
+ # in otherSettings.playlist (e.g.
+ # http://www.bbc.com/turkce/multimedya/2015/10/151010_vid_ankara_patlama_ani)
+ playlist = data_playable.get('otherSettings', {}).get('playlist', {})
+ if playlist:
+ entry = None
+ for key in ('streaming', 'progressiveDownload'):
+ playlist_url = playlist.get('%sUrl' % key)
+ if not playlist_url:
+ continue
+ try:
+ info = self._extract_from_playlist_sxml(
+ playlist_url, playlist_id, timestamp)
+ if not entry:
+ entry = info
+ else:
+ entry['title'] = info['title']
+ entry['formats'].extend(info['formats'])
+ except Exception as e:
+ # Some playlist URL may fail with 500, at the same time
+ # the other one may work fine (e.g.
+ # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu)
+ if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
+ continue
+ raise
+ if entry:
+ self._sort_formats(entry['formats'])
+ entries.append(entry)
+
+ if entries:
+ return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
+
+ # http://www.bbc.co.uk/learningenglish/chinese/features/lingohack/ep-181227
+ group_id = self._search_regex(
+ r'<div[^>]+\bclass=["\']video["\'][^>]+\bdata-pid=["\'](%s)' % self._ID_REGEX,
+ webpage, 'group id', default=None)
+ if playlist_id:
+ return self.url_result(
+ 'https://www.bbc.co.uk/programmes/%s' % group_id,
+ ie=BBCCoUkIE.ie_key())
+
+ # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
+ programme_id = self._search_regex(
+ [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
+ r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
+ r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
+ webpage, 'vpid', default=None)
+
+ if programme_id:
+ formats, subtitles = self._download_media_selector(programme_id)
+ self._sort_formats(formats)
+ # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star)
+ digital_data = self._parse_json(
+ self._search_regex(
+ r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'),
+ programme_id, fatal=False)
+ page_info = digital_data.get('page', {}).get('pageInfo', {})
+ title = page_info.get('pageName') or self._og_search_title(webpage)
+ description = page_info.get('description') or self._og_search_description(webpage)
+ timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp
+ return {
+ 'id': programme_id,
+ 'title': title,
+ 'description': description,
+ 'timestamp': timestamp,
+ 'formats': formats,
+ 'subtitles': subtitles,
+ }
+
+ # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975)
+ # There are several setPayload calls may be present but the video
+ # seems to be always related to the first one
+ morph_payload = self._parse_json(
+ self._search_regex(
+ r'Morph\.setPayload\([^,]+,\s*({.+?})\);',
+ webpage, 'morph payload', default='{}'),
+ playlist_id, fatal=False)
+ if morph_payload:
+ components = try_get(morph_payload, lambda x: x['body']['components'], list) or []
+ for component in components:
+ if not isinstance(component, dict):
+ continue
+ lead_media = try_get(component, lambda x: x['props']['leadMedia'], dict)
+ if not lead_media:
+ continue
+ identifiers = lead_media.get('identifiers')
+ if not identifiers or not isinstance(identifiers, dict):
+ continue
+ programme_id = identifiers.get('vpid') or identifiers.get('playablePid')
+ if not programme_id:
+ continue
+ title = lead_media.get('title') or self._og_search_title(webpage)