yt_dlp/extractor/mtv.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_str
   5 from ..utils import (
   6     ExtractorError,
   7     find_xpath_attr,
   8     fix_xml_ampersands,
   9     float_or_none,
  10     HEADRequest,
  11     int_or_none,
  12     join_nonempty,
  13     RegexNotFoundError,
  14     sanitized_Request,
  15     strip_or_none,
  16     timeconvert,
  17     try_get,
  18     unescapeHTML,
  19     update_url_query,
  20     url_basename,
  21     xpath_text,
  22 )
  23
  24
  25 def _media_xml_tag(tag):
  26     return '{http://search.yahoo.com/mrss/}%s' % tag
  27
  28
  29 class MTVServicesInfoExtractor(InfoExtractor):
  30     _MOBILE_TEMPLATE = None
  31     _LANG = None
  32
  33     @staticmethod
  34     def _id_from_uri(uri):
  35         return uri.split(':')[-1]
  36
  37     @staticmethod
  38     def _remove_template_parameter(url):
  39         # Remove the templates, like &device={device}
  40         return re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', url)
  41
  42     def _get_feed_url(self, uri, url=None):
  43         return self._FEED_URL
  44
  45     def _get_thumbnail_url(self, uri, itemdoc):
  46         search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
  47         thumb_node = itemdoc.find(search_path)
  48         if thumb_node is None:
  49             return None
  50         return thumb_node.get('url') or thumb_node.text or None
  51
  52     def _extract_mobile_video_formats(self, mtvn_id):
  53         webpage_url = self._MOBILE_TEMPLATE % mtvn_id
  54         req = sanitized_Request(webpage_url)
  55         # Otherwise we get a webpage that would execute some javascript
  56         req.add_header('User-Agent', 'curl/7')
  57         webpage = self._download_webpage(req, mtvn_id,
  58                                          'Downloading mobile page')
  59         metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
  60         req = HEADRequest(metrics_url)
  61         response = self._request_webpage(req, mtvn_id, 'Resolving url')
  62         url = response.geturl()
  63         # Transform the url to get the best quality:
  64         url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
  65         return [{'url': url, 'ext': 'mp4'}]
  66
  67     def _extract_video_formats(self, mdoc, mtvn_id, video_id):
  68         if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', mdoc.find('.//src').text) is not None:
  69             if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
  70                 self.to_screen('The normal version is not available from your '
  71                                'country, trying with the mobile version')
  72                 return self._extract_mobile_video_formats(mtvn_id)
  73             raise ExtractorError('This video is not available from your country.',
  74                                  expected=True)
  75
  76         formats = []
  77         for rendition in mdoc.findall('.//rendition'):
  78             if rendition.get('method') == 'hls':
  79                 hls_url = rendition.find('./src').text
  80                 formats.extend(self._extract_m3u8_formats(
  81                     hls_url, video_id, ext='mp4', entry_protocol='m3u8_native',
  82                     m3u8_id='hls', fatal=False))
  83             else:
  84                 # fms
  85                 try:
  86                     _, _, ext = rendition.attrib['type'].partition('/')
  87                     rtmp_video_url = rendition.find('./src').text
  88                     if 'error_not_available.swf' in rtmp_video_url:
  89                         raise ExtractorError(
  90                             '%s said: video is not available' % self.IE_NAME,
  91                             expected=True)
  92                     if rtmp_video_url.endswith('siteunavail.png'):
  93                         continue
  94                     formats.extend([{
  95                         'ext': 'flv' if rtmp_video_url.startswith('rtmp') else ext,
  96                         'url': rtmp_video_url,
  97                         'format_id': join_nonempty(
  98                             'rtmp' if rtmp_video_url.startswith('rtmp') else None,
  99                             rendition.get('bitrate')),
 100                         'width': int(rendition.get('width')),
 101                         'height': int(rendition.get('height')),
 102                     }])
 103                 except (KeyError, TypeError):
 104                     raise ExtractorError('Invalid rendition field.')
 105         if formats:
 106             self._sort_formats(formats)
 107         return formats
 108
 109     def _extract_subtitles(self, mdoc, mtvn_id):
 110         subtitles = {}
 111         for transcript in mdoc.findall('.//transcript'):
 112             if transcript.get('kind') != 'captions':
 113                 continue
 114             lang = transcript.get('srclang')
 115             for typographic in transcript.findall('./typographic'):
 116                 sub_src = typographic.get('src')
 117                 if not sub_src:
 118                     continue
 119                 ext = typographic.get('format')
 120                 if ext == 'cea-608':
 121                     ext = 'scc'
 122                 subtitles.setdefault(lang, []).append({
 123                     'url': compat_str(sub_src),
 124                     'ext': ext
 125                 })
 126         return subtitles
 127
 128     def _get_video_info(self, itemdoc, use_hls=True):
 129         uri = itemdoc.find('guid').text
 130         video_id = self._id_from_uri(uri)
 131         self.report_extraction(video_id)
 132         content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
 133         mediagen_url = self._remove_template_parameter(content_el.attrib['url'])
 134         mediagen_url = mediagen_url.replace('device={device}', '')
 135         if 'acceptMethods' not in mediagen_url:
 136             mediagen_url += '&' if '?' in mediagen_url else '?'
 137             mediagen_url += 'acceptMethods='
 138             mediagen_url += 'hls' if use_hls else 'fms'
 139
 140         mediagen_doc = self._download_xml(
 141             mediagen_url, video_id, 'Downloading video urls', fatal=False)
 142
 143         if mediagen_doc is False:
 144             return None
 145
 146         item = mediagen_doc.find('./video/item')
 147         if item is not None and item.get('type') == 'text':
 148             message = '%s returned error: ' % self.IE_NAME
 149             if item.get('code') is not None:
 150                 message += '%s - ' % item.get('code')
 151             message += item.text
 152             raise ExtractorError(message, expected=True)
 153
 154         description = strip_or_none(xpath_text(itemdoc, 'description'))
 155
 156         timestamp = timeconvert(xpath_text(itemdoc, 'pubDate'))
 157
 158         title_el = None
 159         if title_el is None:
 160             title_el = find_xpath_attr(
 161                 itemdoc, './/{http://search.yahoo.com/mrss/}category',
 162                 'scheme', 'urn:mtvn:video_title')
 163         if title_el is None:
 164             title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
 165         if title_el is None:
 166             title_el = itemdoc.find('.//title')
 167             if title_el.text is None:
 168                 title_el = None
 169
 170         title = title_el.text
 171         if title is None:
 172             raise ExtractorError('Could not find video title')
 173         title = title.strip()
 174
 175         series = find_xpath_attr(
 176             itemdoc, './/{http://search.yahoo.com/mrss/}category',
 177             'scheme', 'urn:mtvn:franchise')
 178         season = find_xpath_attr(
 179             itemdoc, './/{http://search.yahoo.com/mrss/}category',
 180             'scheme', 'urn:mtvn:seasonN')
 181         episode = find_xpath_attr(
 182             itemdoc, './/{http://search.yahoo.com/mrss/}category',
 183             'scheme', 'urn:mtvn:episodeN')
 184         series = series.text if series is not None else None
 185         season = season.text if season is not None else None
 186         episode = episode.text if episode is not None else None
 187         if season and episode:
 188             # episode number includes season, so remove it
 189             episode = re.sub(r'^%s' % season, '', episode)
 190
 191         # This a short id that's used in the webpage urls
 192         mtvn_id = None
 193         mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
 194                                        'scheme', 'urn:mtvn:id')
 195         if mtvn_id_node is not None:
 196             mtvn_id = mtvn_id_node.text
 197
 198         formats = self._extract_video_formats(mediagen_doc, mtvn_id, video_id)
 199
 200         # Some parts of complete video may be missing (e.g. missing Act 3 in
 201         # http://www.southpark.de/alle-episoden/s14e01-sexual-healing)
 202         if not formats:
 203             return None
 204
 205         self._sort_formats(formats)
 206
 207         return {
 208             'title': title,
 209             'formats': formats,
 210             'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
 211             'id': video_id,
 212             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
 213             'description': description,
 214             'duration': float_or_none(content_el.attrib.get('duration')),
 215             'timestamp': timestamp,
 216             'series': series,
 217             'season_number': int_or_none(season),
 218             'episode_number': int_or_none(episode),
 219         }
 220
 221     def _get_feed_query(self, uri):
 222         data = {'uri': uri}
 223         if self._LANG:
 224             data['lang'] = self._LANG
 225         return data
 226
 227     def _get_videos_info(self, uri, use_hls=True, url=None):
 228         video_id = self._id_from_uri(uri)
 229         feed_url = self._get_feed_url(uri, url)
 230         info_url = update_url_query(feed_url, self._get_feed_query(uri))
 231         return self._get_videos_info_from_url(info_url, video_id, use_hls)
 232
 233     def _get_videos_info_from_url(self, url, video_id, use_hls=True):
 234         idoc = self._download_xml(
 235             url, video_id,
 236             'Downloading info', transform_source=fix_xml_ampersands)
 237
 238         title = xpath_text(idoc, './channel/title')
 239         description = xpath_text(idoc, './channel/description')
 240
 241         entries = []
 242         for item in idoc.findall('.//item'):
 243             info = self._get_video_info(item, use_hls)
 244             if info:
 245                 entries.append(info)
 246
 247         # TODO: should be multi-video
 248         return self.playlist_result(
 249             entries, playlist_title=title, playlist_description=description)
 250
 251     def _extract_triforce_mgid(self, webpage, data_zone=None, video_id=None):
 252         triforce_feed = self._parse_json(self._search_regex(
 253             r'triforceManifestFeed\s*=\s*({.+?})\s*;\s*\n', webpage,
 254             'triforce feed', default='{}'), video_id, fatal=False)
 255
 256         data_zone = self._search_regex(
 257             r'data-zone=(["\'])(?P<zone>.+?_lc_promo.*?)\1', webpage,
 258             'data zone', default=data_zone, group='zone')
 259
 260         feed_url = try_get(
 261             triforce_feed, lambda x: x['manifest']['zones'][data_zone]['feed'],
 262             compat_str)
 263         if not feed_url:
 264             return
 265
 266         feed = self._download_json(feed_url, video_id, fatal=False)
 267         if not feed:
 268             return
 269
 270         return try_get(feed, lambda x: x['result']['data']['id'], compat_str)
 271
 272     @staticmethod
 273     def _extract_child_with_type(parent, t):
 274         for c in parent['children']:
 275             if c.get('type') == t:
 276                 return c
 277
 278     def _extract_mgid(self, webpage):
 279         try:
 280             # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
 281             # or http://media.mtvnservices.com/{mgid}
 282             og_url = self._og_search_video_url(webpage)
 283             mgid = url_basename(og_url)
 284             if mgid.endswith('.swf'):
 285                 mgid = mgid[:-4]
 286         except RegexNotFoundError:
 287             mgid = None
 288
 289         if mgid is None or ':' not in mgid:
 290             mgid = self._search_regex(
 291                 [r'data-mgid="(.*?)"', r'swfobject\.embedSWF\(".*?(mgid:.*?)"'],
 292                 webpage, 'mgid', default=None)
 293
 294         if not mgid:
 295             sm4_embed = self._html_search_meta(
 296                 'sm4:video:embed', webpage, 'sm4 embed', default='')
 297             mgid = self._search_regex(
 298                 r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid', default=None)
 299
 300         if not mgid:
 301             mgid = self._extract_triforce_mgid(webpage)
 302
 303         if not mgid:
 304             data = self._parse_json(self._search_regex(
 305                 r'__DATA__\s*=\s*({.+?});', webpage, 'data'), None)
 306             main_container = self._extract_child_with_type(data, 'MainContainer')
 307             ab_testing = self._extract_child_with_type(main_container, 'ABTesting')
 308             video_player = self._extract_child_with_type(ab_testing or main_container, 'VideoPlayer')
 309             if video_player:
 310                 mgid = try_get(video_player, lambda x: x['props']['media']['video']['config']['uri'])
 311             else:
 312                 flex_wrapper = self._extract_child_with_type(ab_testing or main_container, 'FlexWrapper')
 313                 auth_suite_wrapper = self._extract_child_with_type(flex_wrapper, 'AuthSuiteWrapper')
 314                 player = self._extract_child_with_type(auth_suite_wrapper or flex_wrapper, 'Player')
 315                 if player:
 316                     mgid = try_get(player, lambda x: x['props']['videoDetail']['mgid'])
 317
 318         if not mgid:
 319             raise ExtractorError('Could not extract mgid')
 320
 321         return mgid
 322
 323     def _real_extract(self, url):
 324         title = url_basename(url)
 325         webpage = self._download_webpage(url, title)
 326         mgid = self._extract_mgid(webpage)
 327         videos_info = self._get_videos_info(mgid, url=url)
 328         return videos_info
 329
 330
 331 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
 332     IE_NAME = 'mtvservices:embedded'
 333     _VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)'
 334
 335     _TEST = {
 336         # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
 337         'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906',
 338         'md5': 'cb349b21a7897164cede95bd7bf3fbb9',
 339         'info_dict': {
 340             'id': '1043906',
 341             'ext': 'mp4',
 342             'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds',
 343             'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.',
 344             'timestamp': 1400126400,
 345             'upload_date': '20140515',
 346         },
 347     }
 348
 349     @staticmethod
 350     def _extract_url(webpage):
 351         mobj = re.search(
 352             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media\.mtvnservices\.com/embed/.+?)\1', webpage)
 353         if mobj:
 354             return mobj.group('url')
 355
 356     def _get_feed_url(self, uri, url=None):
 357         video_id = self._id_from_uri(uri)
 358         config = self._download_json(
 359             'http://media.mtvnservices.com/pmt/e1/access/index.html?uri=%s&configtype=edge' % uri, video_id)
 360         return self._remove_template_parameter(config['feedWithQueryParams'])
 361
 362     def _real_extract(self, url):
 363         mobj = self._match_valid_url(url)
 364         mgid = mobj.group('mgid')
 365         return self._get_videos_info(mgid)
 366
 367
 368 class MTVIE(MTVServicesInfoExtractor):
 369     IE_NAME = 'mtv'
 370     _VALID_URL = r'https?://(?:www\.)?mtv\.com/(?:video-clips|(?:full-)?episodes)/(?P<id>[^/?#.]+)'
 371     _FEED_URL = 'http://www.mtv.com/feeds/mrss/'
 372
 373     _TESTS = [{
 374         'url': 'http://www.mtv.com/video-clips/vl8qof/unlocking-the-truth-trailer',
 375         'md5': '1edbcdf1e7628e414a8c5dcebca3d32b',
 376         'info_dict': {
 377             'id': '5e14040d-18a4-47c4-a582-43ff602de88e',
 378             'ext': 'mp4',
 379             'title': 'Unlocking The Truth|July 18, 2016|1|101|Trailer',
 380             'description': '"Unlocking the Truth" premieres August 17th at 11/10c.',
 381             'timestamp': 1468846800,
 382             'upload_date': '20160718',
 383         },
 384     }, {
 385         'url': 'http://www.mtv.com/full-episodes/94tujl/unlocking-the-truth-gates-of-hell-season-1-ep-101',
 386         'only_matching': True,
 387     }, {
 388         'url': 'http://www.mtv.com/episodes/g8xu7q/teen-mom-2-breaking-the-wall-season-7-ep-713',
 389         'only_matching': True,
 390     }]
 391
 392
 393 class MTVJapanIE(MTVServicesInfoExtractor):
 394     IE_NAME = 'mtvjapan'
 395     _VALID_URL = r'https?://(?:www\.)?mtvjapan\.com/videos/(?P<id>[0-9a-z]+)'
 396
 397     _TEST = {
 398         'url': 'http://www.mtvjapan.com/videos/prayht/fresh-info-cadillac-escalade',
 399         'info_dict': {
 400             'id': 'bc01da03-6fe5-4284-8880-f291f4e368f5',
 401             'ext': 'mp4',
 402             'title': '【Fresh Info】Cadillac ESCALADE Sport Edition',
 403         },
 404         'params': {
 405             'skip_download': True,
 406         },
 407     }
 408     _GEO_COUNTRIES = ['JP']
 409     _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
 410
 411     def _get_feed_query(self, uri):
 412         return {
 413             'arcEp': 'mtvjapan.com',
 414             'mgid': uri,
 415         }
 416
 417
 418 class MTVVideoIE(MTVServicesInfoExtractor):
 419     IE_NAME = 'mtv:video'
 420     _VALID_URL = r'''(?x)^https?://
 421         (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
 422            m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))'''
 423
 424     _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/'
 425
 426     _TESTS = [
 427         {
 428             'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
 429             'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
 430             'info_dict': {
 431                 'id': '853555',
 432                 'ext': 'mp4',
 433                 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
 434                 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
 435                 'timestamp': 1352610000,
 436                 'upload_date': '20121111',
 437             },
 438         },
 439     ]
 440
 441     def _get_thumbnail_url(self, uri, itemdoc):
 442         return 'http://mtv.mtvnimages.com/uri/' + uri
 443
 444     def _real_extract(self, url):
 445         mobj = self._match_valid_url(url)
 446         video_id = mobj.group('videoid')
 447         uri = mobj.groupdict().get('mgid')
 448         if uri is None:
 449             webpage = self._download_webpage(url, video_id)
 450
 451             # Some videos come from Vevo.com
 452             m_vevo = re.search(
 453                 r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage)
 454             if m_vevo:
 455                 vevo_id = m_vevo.group(1)
 456                 self.to_screen('Vevo video detected: %s' % vevo_id)
 457                 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
 458
 459             uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
 460         return self._get_videos_info(uri)
 461
 462
 463 class MTVDEIE(MTVServicesInfoExtractor):
 464     IE_NAME = 'mtv.de'
 465     _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:musik/videoclips|folgen|news)/(?P<id>[0-9a-z]+)'
 466     _TESTS = [{
 467         'url': 'http://www.mtv.de/musik/videoclips/2gpnv7/Traum',
 468         'info_dict': {
 469             'id': 'd5d472bc-f5b7-11e5-bffd-a4badb20dab5',
 470             'ext': 'mp4',
 471             'title': 'Traum',
 472             'description': 'Traum',
 473         },
 474         'params': {
 475             # rtmp download
 476             'skip_download': True,
 477         },
 478         'skip': 'Blocked at Travis CI',
 479     }, {
 480         # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
 481         'url': 'http://www.mtv.de/folgen/6b1ylu/teen-mom-2-enthuellungen-S5-F1',
 482         'info_dict': {
 483             'id': '1e5a878b-31c5-11e7-a442-0e40cf2fc285',
 484             'ext': 'mp4',
 485             'title': 'Teen Mom 2',
 486             'description': 'md5:dc65e357ef7e1085ed53e9e9d83146a7',
 487         },
 488         'params': {
 489             # rtmp download
 490             'skip_download': True,
 491         },
 492         'skip': 'Blocked at Travis CI',
 493     }, {
 494         'url': 'http://www.mtv.de/news/glolix/77491-mtv-movies-spotlight--pixels--teil-3',
 495         'info_dict': {
 496             'id': 'local_playlist-4e760566473c4c8c5344',
 497             'ext': 'mp4',
 498             'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
 499             'description': 'MTV Movies Supercut',
 500         },
 501         'params': {
 502             # rtmp download
 503             'skip_download': True,
 504         },
 505         'skip': 'Das Video kann zur Zeit nicht abgespielt werden.',
 506     }]
 507     _GEO_COUNTRIES = ['DE']
 508     _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
 509
 510     def _get_feed_query(self, uri):
 511         return {
 512             'arcEp': 'mtv.de',
 513             'mgid': uri,
 514         }
 515
 516
 517 class MTVItaliaIE(MTVServicesInfoExtractor):
 518     IE_NAME = 'mtv.it'
 519     _VALID_URL = r'https?://(?:www\.)?mtv\.it/(?:episodi|video|musica)/(?P<id>[0-9a-z]+)'
 520     _TESTS = [{
 521         'url': 'http://www.mtv.it/episodi/24bqab/mario-una-serie-di-maccio-capatonda-cavoli-amario-episodio-completo-S1-E1',
 522         'info_dict': {
 523             'id': '0f0fc78e-45fc-4cce-8f24-971c25477530',
 524             'ext': 'mp4',
 525             'title': 'Cavoli amario (episodio completo)',
 526             'description': 'md5:4962bccea8fed5b7c03b295ae1340660',
 527             'series': 'Mario - Una Serie Di Maccio Capatonda',
 528             'season_number': 1,
 529             'episode_number': 1,
 530         },
 531         'params': {
 532             'skip_download': True,
 533         },
 534     }]
 535     _GEO_COUNTRIES = ['IT']
 536     _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
 537
 538     def _get_feed_query(self, uri):
 539         return {
 540             'arcEp': 'mtv.it',
 541             'mgid': uri,
 542         }
 543
 544
 545 class MTVItaliaProgrammaIE(MTVItaliaIE):
 546     IE_NAME = 'mtv.it:programma'
 547     _VALID_URL = r'https?://(?:www\.)?mtv\.it/(?:programmi|playlist)/(?P<id>[0-9a-z]+)'
 548     _TESTS = [{
 549         # program page: general
 550         'url': 'http://www.mtv.it/programmi/s2rppv/mario-una-serie-di-maccio-capatonda',
 551         'info_dict': {
 552             'id': 'a6f155bc-8220-4640-aa43-9b95f64ffa3d',
 553             'title': 'Mario - Una Serie Di Maccio Capatonda',
 554             'description': 'md5:72fbffe1f77ccf4e90757dd4e3216153',
 555         },
 556         'playlist_count': 2,
 557         'params': {
 558             'skip_download': True,
 559         },
 560     }, {
 561         # program page: specific season
 562         'url': 'http://www.mtv.it/programmi/d9ncjf/mario-una-serie-di-maccio-capatonda-S2',
 563         'info_dict': {
 564             'id': '4deeb5d8-f272-490c-bde2-ff8d261c6dd1',
 565             'title': 'Mario - Una Serie Di Maccio Capatonda - Stagione 2',
 566         },
 567         'playlist_count': 34,
 568         'params': {
 569             'skip_download': True,
 570         },
 571     }, {
 572         # playlist page + redirect
 573         'url': 'http://www.mtv.it/playlist/sexy-videos/ilctal',
 574         'info_dict': {
 575             'id': 'dee8f9ee-756d-493b-bf37-16d1d2783359',
 576             'title': 'Sexy Videos',
 577         },
 578         'playlist_mincount': 145,
 579         'params': {
 580             'skip_download': True,
 581         },
 582     }]
 583     _GEO_COUNTRIES = ['IT']
 584     _FEED_URL = 'http://www.mtv.it/feeds/triforce/manifest/v8'
 585
 586     def _get_entries(self, title, url):
 587         while True:
 588             pg = self._search_regex(r'/(\d+)$', url, 'entries', '1')
 589             entries = self._download_json(url, title, 'page %s' % pg)
 590             url = try_get(
 591                 entries, lambda x: x['result']['nextPageURL'], compat_str)
 592             entries = try_get(
 593                 entries, (
 594                     lambda x: x['result']['data']['items'],
 595                     lambda x: x['result']['data']['seasons']),
 596                 list)
 597             for entry in entries or []:
 598                 if entry.get('canonicalURL'):
 599                     yield self.url_result(entry['canonicalURL'])
 600             if not url:
 601                 break
 602
 603     def _real_extract(self, url):
 604         query = {'url': url}
 605         info_url = update_url_query(self._FEED_URL, query)
 606         video_id = self._match_id(url)
 607         info = self._download_json(info_url, video_id).get('manifest')
 608
 609         redirect = try_get(
 610             info, lambda x: x['newLocation']['url'], compat_str)
 611         if redirect:
 612             return self.url_result(redirect)
 613
 614         title = info.get('title')
 615         video_id = try_get(
 616             info, lambda x: x['reporting']['itemId'], compat_str)
 617         parent_id = try_get(
 618             info, lambda x: x['reporting']['parentId'], compat_str)
 619
 620         playlist_url = current_url = None
 621         for z in (info.get('zones') or {}).values():
 622             if z.get('moduleName') in ('INTL_M304', 'INTL_M209'):
 623                 info_url = z.get('feed')
 624             if z.get('moduleName') in ('INTL_M308', 'INTL_M317'):
 625                 playlist_url = playlist_url or z.get('feed')
 626             if z.get('moduleName') in ('INTL_M300',):
 627                 current_url = current_url or z.get('feed')
 628
 629         if not info_url:
 630             raise ExtractorError('No info found')
 631
 632         if video_id == parent_id:
 633             video_id = self._search_regex(
 634                 r'([^\/]+)/[^\/]+$', info_url, 'video_id')
 635
 636         info = self._download_json(info_url, video_id, 'Show infos')
 637         info = try_get(info, lambda x: x['result']['data'], dict)
 638         title = title or try_get(
 639             info, (
 640                 lambda x: x['title'],
 641                 lambda x: x['headline']),
 642             compat_str)
 643         description = try_get(info, lambda x: x['content'], compat_str)
 644
 645         if current_url:
 646             season = try_get(
 647                 self._download_json(playlist_url, video_id, 'Seasons info'),
 648                 lambda x: x['result']['data'], dict)
 649             current = try_get(
 650                 season, lambda x: x['currentSeason'], compat_str)
 651             seasons = try_get(
 652                 season, lambda x: x['seasons'], list) or []
 653
 654             if current in [s.get('eTitle') for s in seasons]:
 655                 playlist_url = current_url
 656
 657         title = re.sub(
 658             r'[-|]\s*(?:mtv\s*italia|programma|playlist)',
 659             '', title, flags=re.IGNORECASE).strip()
 660
 661         return self.playlist_result(
 662             self._get_entries(title, playlist_url),
 663             video_id, title, description)