+
+class DailymotionBaseInfoExtractor(InfoExtractor):
+ @staticmethod
+ def _build_request(url):
+ """Build a request with the family filter disabled"""
+ request = sanitized_Request(url)
+ request.add_header('Cookie', 'family_filter=off; ff=off')
+ return request
+
+ def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
+ request = self._build_request(url)
+ return self._download_webpage_handle(request, *args, **kwargs)
+
+ def _download_webpage_no_ff(self, url, *args, **kwargs):
+ request = self._build_request(url)
+ return self._download_webpage(request, *args, **kwargs)
+
+
+class DailymotionIE(DailymotionBaseInfoExtractor):
+ _VALID_URL = r'(?i)https?://(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|#)/)?video|swf)/(?P<id>[^/?_]+)'
+ IE_NAME = 'dailymotion'
+
+ _FORMATS = [
+ ('stream_h264_ld_url', 'ld'),
+ ('stream_h264_url', 'standard'),
+ ('stream_h264_hq_url', 'hq'),
+ ('stream_h264_hd_url', 'hd'),
+ ('stream_h264_hd1080_url', 'hd180'),
+ ]
+
+ _TESTS = [{
+ 'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
+ 'md5': '074b95bdee76b9e3654137aee9c79dfe',
+ 'info_dict': {
+ 'id': 'x5kesuj',
+ 'ext': 'mp4',
+ 'title': 'Office Christmas Party Review – Jason Bateman, Olivia Munn, T.J. Miller',
+ 'description': 'Office Christmas Party Review - Jason Bateman, Olivia Munn, T.J. Miller',
+ 'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
+ 'duration': 187,
+ 'timestamp': 1493651285,
+ 'upload_date': '20170501',
+ 'uploader': 'Deadline',
+ 'uploader_id': 'x1xm8ri',
+ 'age_limit': 0,
+ },
+ }, {
+ 'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
+ 'md5': '2137c41a8e78554bb09225b8eb322406',
+ 'info_dict': {
+ 'id': 'x2iuewm',
+ 'ext': 'mp4',
+ 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
+ 'description': 'Several come bundled with the Steam Controller.',
+ 'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
+ 'duration': 74,
+ 'timestamp': 1425657362,
+ 'upload_date': '20150306',
+ 'uploader': 'IGN',
+ 'uploader_id': 'xijv66',
+ 'age_limit': 0,
+ 'view_count': int,
+ },
+ 'skip': 'video gone',
+ }, {
+ # Vevo video
+ 'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
+ 'info_dict': {
+ 'title': 'Roar (Official)',
+ 'id': 'USUV71301934',
+ 'ext': 'mp4',
+ 'uploader': 'Katy Perry',
+ 'upload_date': '20130905',
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ 'skip': 'VEVO is only available in some countries',
+ }, {
+ # age-restricted video
+ 'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
+ 'md5': '0d667a7b9cebecc3c89ee93099c4159d',
+ 'info_dict': {
+ 'id': 'xyh2zz',
+ 'ext': 'mp4',
+ 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
+ 'uploader': 'HotWaves1012',
+ 'age_limit': 18,
+ },
+ 'skip': 'video gone',
+ }, {
+ # geo-restricted, player v5
+ 'url': 'http://www.dailymotion.com/video/xhza0o',
+ 'only_matching': True,
+ }, {
+ # with subtitles
+ '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,
+ }, {
+ 'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
+ 'only_matching': True,
+ }]
+
+ @staticmethod
+ def _extract_urls(webpage):
+ # Look for embedded Dailymotion player
+ matches = re.findall(
+ r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
+ return list(map(lambda m: unescapeHTML(m[1]), matches))
+
+ 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, default=None) 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', default=None)
+ 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', default=None))
+
+ 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\(({.+?})\);',
+ r'var\s+config\s*=\s*({.+?});',
+ # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580)
+ r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
+ webpage, 'player v5', default=None)
+ if player_v5:
+ player = self._parse_json(player_v5, video_id, fatal=False) or {}
+ metadata = try_get(player, lambda x: x['metadata'], dict)
+ if not metadata:
+ metadata_url = url_or_none(try_get(
+ player, lambda x: x['context']['metadata_template_url1']))
+ if metadata_url:
+ metadata_url = metadata_url.replace(':videoId', video_id)
+ else:
+ metadata_url = update_url_query(
+ 'https://www.dailymotion.com/player/metadata/video/%s'
+ % video_id, {
+ 'embedder': url,
+ 'integration': 'inline',
+ 'GK_PV5_NEON': '1',
+ })
+ metadata = self._download_json(
+ metadata_url, video_id, 'Downloading metadata JSON')
+
+ if try_get(metadata, lambda x: x['error']['type']) == 'password_protected':
+ password = self._downloader.params.get('videopassword')
+ if password:
+ r = int(metadata['id'][1:], 36)
+ us64e = lambda x: base64.urlsafe_b64encode(x).decode().strip('=')
+ t = ''.join(random.choice(string.ascii_letters) for i in range(10))
+ n = us64e(compat_struct_pack('I', r))
+ i = us64e(hashlib.md5(('%s%d%s' % (password, r, t)).encode()).digest())
+ metadata = self._download_json(
+ 'http://www.dailymotion.com/player/metadata/video/p' + i + t + n, video_id)
+
+ 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 = mimetype2ext(type_) or determine_ext(media_url)
+ if ext == 'm3u8':
+ m3u8_formats = self._extract_m3u8_formats(
+ media_url, video_id, 'mp4', preference=-1,
+ m3u8_id='hls', fatal=False)
+ for f in m3u8_formats:
+ f['url'] = f['url'].split('#')[0]
+ formats.append(f)
+ elif 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,
+ 'ext': ext,
+ }
+ 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'))
+
+ info = self._parse_json(
+ self._search_regex(
+ r'var info = ({.*?}),$', embed_page,
+ 'video info', flags=re.MULTILINE),
+ video_id)
+
+ self._check_error(info)
+
+ formats = []
+ for (key, format_id) in self._FORMATS:
+ video_url = info.get(key)
+ if video_url is not None:
+ m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
+ if m_size is not None:
+ width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
+ else:
+ width, height = None, None
+ formats.append({
+ 'url': video_url,
+ 'ext': 'mp4',
+ 'format_id': format_id,
+ 'width': width,
+ 'height': height,
+ })
+ self._sort_formats(formats)
+
+ # subtitles
+ video_subtitles = self.extract_subtitles(video_id, webpage)
+
+ title = self._og_search_title(webpage, default=None)
+ if title is None:
+ title = self._html_search_regex(
+ r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
+ 'title')
+
+ return {
+ 'id': video_id,
+ 'formats': formats,
+ 'uploader': info['owner.screenname'],
+ 'timestamp': timestamp,
+ 'title': title,
+ 'description': description,
+ 'subtitles': video_subtitles,
+ 'thumbnail': info['thumbnail_url'],
+ 'age_limit': age_limit,
+ 'view_count': view_count,
+ 'duration': info['duration']
Rapsys Git / youtubedl / blobdiff