- # Retrieve video webpage to extract further information
- request = self._build_request(url)
- webpage = self._download_webpage(request, video_id)
-
- # Extract URL, uploader and title from webpage
- self.report_extraction(video_id)
-
- # It may just embed a vevo video:
- m_vevo = re.search(
- r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?videoId=(?P<id>[\w]*)',
- webpage)
- if m_vevo is not None:
- vevo_id = m_vevo.group('id')
- self.to_screen(u'Vevo video detected: %s' % vevo_id)
- return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo')
-
- video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
- # Looking for official user
- r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
- webpage, 'video uploader')
-
- video_upload_date = None
- mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
- if mobj is not None:
- video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
-
- embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
- embed_page = self._download_webpage(embed_url, video_id,
- u'Downloading embed page')
- info = self._search_regex(r'var info = ({.*?}),$', embed_page,
- 'video info', flags=re.MULTILINE)
- info = json.loads(info)
- if info.get('error') is not None:
- msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
- raise ExtractorError(msg, expected=True)
-
- # TODO: support choosing qualities
-
- for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
- 'stream_h264_hq_url','stream_h264_url',
- 'stream_h264_ld_url']:
- if info.get(key):#key in info and info[key]:
- max_quality = key
- self.to_screen(u'Using %s' % key)
- break
- else:
- raise ExtractorError(u'Unable to extract video URL')
- video_url = info[max_quality]
+ 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/ytdl-org/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/ytdl-org/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)
Rapsys Git / youtubedl / blobdiff