- {
- 'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
- 'only_matching': True,
- },
- {
- 'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
- 'only_matching': True,
- }
- ]
-
- def _real_extract(self, url):
- video_id = self._match_id(url)
-
- webpage = self._download_webpage_no_ff(
- 'https://www.dailymotion.com/video/%s' % video_id, video_id)
-
- age_limit = self._rta_search(webpage)
-
- description = self._og_search_description(webpage) or self._html_search_meta(
- 'description', webpage, 'description')
-
- view_count_str = self._search_regex(
- (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
- r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
- webpage, 'view count', fatal=False)
- if view_count_str:
- view_count_str = re.sub(r'\s', '', view_count_str)
- view_count = str_to_int(view_count_str)
- comment_count = int_or_none(self._search_regex(
- r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
- webpage, 'comment count', fatal=False))
-
- player_v5 = self._search_regex(
- [r'buildPlayer\(({.+?})\);\n', # See https://github.com/rg3/youtube-dl/issues/7826
- r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
- r'buildPlayer\(({.+?})\);'],
- webpage, 'player v5', default=None)
- if player_v5:
- player = self._parse_json(player_v5, video_id)
- metadata = player['metadata']
-
- self._check_error(metadata)
-
- formats = []
- for quality, media_list in metadata['qualities'].items():
- for media in media_list:
- media_url = media.get('url')
- if not media_url:
- continue
- type_ = media.get('type')
- if type_ == 'application/vnd.lumberjack.manifest':
- continue
- ext = determine_ext(media_url)
- if type_ == 'application/x-mpegURL' or ext == 'm3u8':
- formats.extend(self._extract_m3u8_formats(
- media_url, video_id, 'mp4', preference=-1,
- m3u8_id='hls', fatal=False))
- elif type_ == 'application/f4m' or ext == 'f4m':
- formats.extend(self._extract_f4m_formats(
- media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
- else:
- f = {
- 'url': media_url,
- 'format_id': 'http-%s' % quality,
- }
- m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
- if m:
- f.update({
- 'width': int(m.group('width')),
- 'height': int(m.group('height')),
- })
- formats.append(f)
- self._sort_formats(formats)
-
- title = metadata['title']
- duration = int_or_none(metadata.get('duration'))
- timestamp = int_or_none(metadata.get('created_time'))
- thumbnail = metadata.get('poster_url')
- uploader = metadata.get('owner', {}).get('screenname')
- uploader_id = metadata.get('owner', {}).get('id')
-
- subtitles = {}
- subtitles_data = metadata.get('subtitles', {}).get('data', {})
- if subtitles_data and isinstance(subtitles_data, dict):
- for subtitle_lang, subtitle in subtitles_data.items():
- subtitles[subtitle_lang] = [{
- 'ext': determine_ext(subtitle_url),
- 'url': subtitle_url,
- } for subtitle_url in subtitle.get('urls', [])]
-
- return {
- 'id': video_id,
- 'title': title,
- 'description': description,
- 'thumbnail': thumbnail,
- 'duration': duration,
- 'timestamp': timestamp,
- 'uploader': uploader,
- 'uploader_id': uploader_id,
- 'age_limit': age_limit,
- 'view_count': view_count,
- 'comment_count': comment_count,
- 'formats': formats,
- 'subtitles': subtitles,
- }
-
- # vevo embed
- vevo_id = self._search_regex(
- r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
- webpage, 'vevo embed', default=None)
- if vevo_id:
- return self.url_result('vevo:%s' % vevo_id, 'Vevo')
-
- # fallback old player
- embed_page = self._download_webpage_no_ff(
- 'https://www.dailymotion.com/embed/video/%s' % video_id,
- video_id, 'Downloading embed page')
-
- timestamp = parse_iso8601(self._html_search_meta(
- 'video:release_date', webpage, 'upload date'))