yt_dlp/extractor/mediaset.py

   1 import functools
   2 import re
   3
   4 from .theplatform import ThePlatformBaseIE
   5 from ..utils import (
   6     ExtractorError,
   7     GeoRestrictedError,
   8     OnDemandPagedList,
   9     int_or_none,
  10     try_get,
  11     update_url_query,
  12     urljoin,
  13 )
  14
  15
  16 class MediasetIE(ThePlatformBaseIE):
  17     _TP_TLD = 'eu'
  18     _GUID_RE = r'F[0-9A-Z]{15}'
  19     _VALID_URL = rf'''(?x)
  20                     (?:
  21                         mediaset:|
  22                         https?://
  23                             (?:\w+\.)+mediaset\.it/
  24                             (?:
  25                                 (?:video|on-demand|movie)/(?:[^/]+/)+[^/]+_|
  26                                 player/(?:v\d+/)?index\.html\?\S*?\bprogramGuid=
  27                             )
  28                     )(?P<id>{_GUID_RE})
  29                     '''
  30
  31     _EMBED_REGEX = [
  32         rf'<iframe[^>]+src=[\'"](?P<url>(?:https?:)?//(?:\w+\.)+mediaset\.it/player/(?:v\d+/)?index\.html\?\S*?programGuid={_GUID_RE})[\'"&]'
  33     ]
  34     _TESTS = [{
  35         # full episode
  36         'url': 'https://mediasetinfinity.mediaset.it/video/mrwronglezionidamore/episodio-1_F310575103000102',
  37         'md5': 'a7e75c6384871f322adb781d3bd72c26',
  38         'info_dict': {
  39             'id': 'F310575103000102',
  40             'ext': 'mp4',
  41             'title': 'Episodio 1',
  42             'description': 'md5:e8017b7d7194e9bfb75299c2b8d81e02',
  43             'thumbnail': r're:^https?://.*\.jpg$',
  44             'duration': 2682.0,
  45             'upload_date': '20210530',
  46             'series': 'Mr Wrong - Lezioni d\'amore',
  47             'timestamp': 1622413946,
  48             'uploader': 'Canale 5',
  49             'uploader_id': 'C5',
  50             'season': 'Season 1',
  51             'episode': 'Episode 1',
  52             'season_number': 1,
  53             'episode_number': 1,
  54             'chapters': [{'start_time': 0.0, 'end_time': 439.88}, {'start_time': 439.88, 'end_time': 1685.84}, {'start_time': 1685.84, 'end_time': 2682.0}],
  55         },
  56     }, {
  57         'url': 'https://mediasetinfinity.mediaset.it/video/matrix/puntata-del-25-maggio_F309013801000501',
  58         'md5': '1276f966ac423d16ba255ce867de073e',
  59         'info_dict': {
  60             'id': 'F309013801000501',
  61             'ext': 'mp4',
  62             'title': 'Puntata del 25 maggio',
  63             'description': 'md5:ee2e456e3eb1dba5e814596655bb5296',
  64             'thumbnail': r're:^https?://.*\.jpg$',
  65             'duration': 6565.008,
  66             'upload_date': '20200903',
  67             'series': 'Matrix',
  68             'timestamp': 1599172492,
  69             'uploader': 'Canale 5',
  70             'uploader_id': 'C5',
  71             'season': 'Season 5',
  72             'episode': 'Episode 5',
  73             'season_number': 5,
  74             'episode_number': 5,
  75             'chapters': [{'start_time': 0.0, 'end_time': 3409.08}, {'start_time': 3409.08, 'end_time': 6565.008}],
  76             'categories': ['Informazione'],
  77         },
  78     }, {
  79         # DRM
  80         'url': 'https://mediasetinfinity.mediaset.it/movie/selvaggi/selvaggi_F006474501000101',
  81         'info_dict': {
  82             'id': 'F006474501000101',
  83             'ext': 'mp4',
  84             'title': 'Selvaggi',
  85             'description': 'md5:cfdedbbfdd12d4d0e5dcf1fa1b75284f',
  86             'thumbnail': r're:^https?://.*\.jpg$',
  87             'duration': 5233.01,
  88             'upload_date': '20210729',
  89             'timestamp': 1627594716,
  90             'uploader': 'Cine34',
  91             'uploader_id': 'B6',
  92             'chapters': [{'start_time': 0.0, 'end_time': 1938.56}, {'start_time': 1938.56, 'end_time': 5233.01}],
  93         },
  94         'params': {
  95             'ignore_no_formats_error': True,
  96         },
  97         'expected_warnings': [
  98             'None of the available releases match the specified AssetType, ProtectionScheme, and/or Format preferences',
  99             'Content behind paywall and DRM',
 100         ],
 101         'skip': True,
 102     }, {
 103         # old domain
 104         'url': 'https://www.mediasetplay.mediaset.it/video/mrwronglezionidamore/episodio-1_F310575103000102',
 105         'only_matching': True,
 106     }, {
 107         # iframe
 108         'url': 'https://static3.mediasetplay.mediaset.it/player/index.html?appKey=5ad3966b1de1c4000d5cec48&programGuid=FAFU000000665924&id=665924',
 109         'only_matching': True,
 110     }, {
 111         'url': 'mediaset:FAFU000000665924',
 112         'only_matching': True,
 113     }]
 114     _WEBPAGE_TESTS = [{
 115         # Mediaset embed
 116         'url': 'http://www.tgcom24.mediaset.it/politica/serracchiani-voglio-vivere-in-una-societa-aperta-reazioni-sproporzionate-_3071354-201702a.shtml',
 117         'info_dict': {
 118             'id': 'FD00000000004929',
 119             'ext': 'mp4',
 120             'title': 'Serracchiani: "Voglio vivere in una società aperta, con tutela del patto di fiducia"',
 121             'duration': 67.013,
 122             'thumbnail': r're:^https?://.*\.jpg$',
 123             'uploader': 'Mediaset Play',
 124             'uploader_id': 'QY',
 125             'upload_date': '20201005',
 126             'timestamp': 1601866168,
 127             'chapters': [],
 128         },
 129         'params': {
 130             'skip_download': True,
 131         },
 132         'skip': 'Dead link',
 133     }, {
 134         # WittyTV embed
 135         'url': 'https://www.wittytv.it/mauriziocostanzoshow/ultima-puntata-venerdi-25-novembre/',
 136         'info_dict': {
 137             'id': 'F312172801000801',
 138             'ext': 'mp4',
 139             'title': 'Ultima puntata - Venerdì 25 novembre',
 140             'description': 'Una serata all\'insegna della musica e del buonumore ma non priva di spunti di riflessione',
 141             'duration': 6203.01,
 142             'thumbnail': r're:^https?://.*\.jpg$',
 143             'uploader': 'Canale 5',
 144             'uploader_id': 'C5',
 145             'upload_date': '20221126',
 146             'timestamp': 1669428689,
 147             'chapters': list,
 148             'series': 'Maurizio Costanzo Show',
 149             'season': 'Season 12',
 150             'season_number': 12,
 151             'episode': 'Episode 8',
 152             'episode_number': 8,
 153             'categories': ['Intrattenimento'],
 154         },
 155         'params': {
 156             'skip_download': True,
 157         }
 158     }]
 159
 160     def _parse_smil_formats_and_subtitles(
 161             self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
 162         for video in smil.findall(self._xpath_ns('.//video', namespace)):
 163             video.attrib['src'] = re.sub(r'(https?://vod05)t(-mediaset-it\.akamaized\.net/.+?.mpd)\?.+', r'\1\2', video.attrib['src'])
 164         return super(MediasetIE, self)._parse_smil_formats_and_subtitles(
 165             smil, smil_url, video_id, namespace, f4m_params, transform_rtmp_url)
 166
 167     def _check_drm_formats(self, tp_formats, video_id):
 168         has_nondrm, drm_manifest = False, ''
 169         for f in tp_formats:
 170             if '_sampleaes/' in (f.get('manifest_url') or ''):
 171                 drm_manifest = drm_manifest or f['manifest_url']
 172                 f['has_drm'] = True
 173             if not f.get('has_drm') and f.get('manifest_url'):
 174                 has_nondrm = True
 175
 176         nodrm_manifest = re.sub(r'_sampleaes/(\w+)_fp_', r'/\1_no_', drm_manifest)
 177         if has_nondrm or nodrm_manifest == drm_manifest:
 178             return
 179
 180         tp_formats.extend(self._extract_m3u8_formats(
 181             nodrm_manifest, video_id, m3u8_id='hls', fatal=False) or [])
 182
 183     def _real_extract(self, url):
 184         guid = self._match_id(url)
 185         tp_path = f'PR1GhC/media/guid/2702976343/{guid}'
 186         info = self._extract_theplatform_metadata(tp_path, guid)
 187
 188         formats = []
 189         subtitles = {}
 190         first_e = geo_e = None
 191         asset_type = 'geoNo:HD,browser,geoIT|geoNo:HD,geoIT|geoNo:SD,browser,geoIT|geoNo:SD,geoIT|geoNo|HD|SD'
 192         # TODO: fixup ISM+none manifest URLs
 193         for f in ('MPEG4', 'MPEG-DASH', 'M3U'):
 194             try:
 195                 tp_formats, tp_subtitles = self._extract_theplatform_smil(
 196                     update_url_query(f'http://link.theplatform.{self._TP_TLD}/s/{tp_path}', {
 197                         'mbr': 'true',
 198                         'formats': f,
 199                         'assetTypes': asset_type,
 200                     }), guid, f'Downloading {f.split("+")[0]} SMIL data')
 201             except ExtractorError as e:
 202                 if e.orig_msg == 'None of the available releases match the specified AssetType, ProtectionScheme, and/or Format preferences':
 203                     e.orig_msg = 'This video is DRM protected'
 204                 if not geo_e and isinstance(e, GeoRestrictedError):
 205                     geo_e = e
 206                 if not first_e:
 207                     first_e = e
 208                 continue
 209             self._check_drm_formats(tp_formats, guid)
 210             formats.extend(tp_formats)
 211             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
 212
 213         # check for errors and report them
 214         if (first_e or geo_e) and not formats:
 215             raise geo_e or first_e
 216
 217         feed_data = self._download_json(
 218             f'https://feed.entertainment.tv.theplatform.eu/f/PR1GhC/mediaset-prod-all-programs-v2/guid/-/{guid}',
 219             guid, fatal=False)
 220         if feed_data:
 221             publish_info = feed_data.get('mediasetprogram$publishInfo') or {}
 222             thumbnails = feed_data.get('thumbnails') or {}
 223             thumbnail = None
 224             for key, value in thumbnails.items():
 225                 if key.startswith('image_keyframe_poster-'):
 226                     thumbnail = value.get('url')
 227                     break
 228
 229             info.update({
 230                 'description': info.get('description') or feed_data.get('description') or feed_data.get('longDescription'),
 231                 'uploader': publish_info.get('description'),
 232                 'uploader_id': publish_info.get('channel'),
 233                 'view_count': int_or_none(feed_data.get('mediasetprogram$numberOfViews')),
 234                 'thumbnail': thumbnail,
 235             })
 236
 237             if feed_data.get('programType') == 'episode':
 238                 info.update({
 239                     'episode_number': int_or_none(
 240                         feed_data.get('tvSeasonEpisodeNumber')),
 241                     'season_number': int_or_none(
 242                         feed_data.get('tvSeasonNumber')),
 243                     'series': feed_data.get('mediasetprogram$brandTitle'),
 244                 })
 245
 246         info.update({
 247             'id': guid,
 248             'formats': formats,
 249             'subtitles': subtitles,
 250         })
 251         return info
 252
 253
 254 class MediasetShowIE(MediasetIE):  # XXX: Do not subclass from concrete IE
 255     _VALID_URL = r'''(?x)
 256                     (?:
 257                         https?://
 258                             (\w+\.)+mediaset\.it/
 259                             (?:
 260                                 (?:fiction|programmi-tv|serie-tv|kids)/(?:.+?/)?
 261                                     (?:[a-z-]+)_SE(?P<id>\d{12})
 262                                     (?:,ST(?P<st>\d{12}))?
 263                                     (?:,sb(?P<sb>\d{9}))?$
 264                             )
 265                     )
 266                     '''
 267     _TESTS = [{
 268         # TV Show webpage (general webpage)
 269         'url': 'https://mediasetinfinity.mediaset.it/programmi-tv/leiene/leiene_SE000000000061',
 270         'info_dict': {
 271             'id': '000000000061',
 272             'title': 'Le Iene 2022/2023',
 273         },
 274         'playlist_mincount': 6,
 275     }, {
 276         # TV Show webpage (specific season)
 277         'url': 'https://mediasetinfinity.mediaset.it/programmi-tv/leiene/leiene_SE000000000061,ST000000002763',
 278         'info_dict': {
 279             'id': '000000002763',
 280             'title': 'Le Iene 2021/2022',
 281         },
 282         'playlist_mincount': 7,
 283     }, {
 284         # TV Show specific playlist (with multiple pages)
 285         'url': 'https://mediasetinfinity.mediaset.it/programmi-tv/leiene/iservizi_SE000000000061,ST000000002763,sb100013375',
 286         'info_dict': {
 287             'id': '100013375',
 288             'title': 'I servizi',
 289         },
 290         'playlist_mincount': 50,
 291     }]
 292
 293     _BY_SUBBRAND = 'https://feed.entertainment.tv.theplatform.eu/f/PR1GhC/mediaset-prod-all-programs-v2?byCustomValue={subBrandId}{%s}&sort=:publishInfo_lastPublished|desc,tvSeasonEpisodeNumber|desc&range=%d-%d'
 294     _PAGE_SIZE = 25
 295
 296     def _fetch_page(self, sb, page):
 297         lower_limit = page * self._PAGE_SIZE + 1
 298         upper_limit = lower_limit + self._PAGE_SIZE - 1
 299         content = self._download_json(
 300             self._BY_SUBBRAND % (sb, lower_limit, upper_limit), sb)
 301         for entry in content.get('entries') or []:
 302             yield self.url_result(
 303                 'mediaset:' + entry['guid'],
 304                 playlist_title=entry['mediasetprogram$subBrandDescription'])
 305
 306     def _real_extract(self, url):
 307         playlist_id, st, sb = self._match_valid_url(url).group('id', 'st', 'sb')
 308         if not sb:
 309             page = self._download_webpage(url, st or playlist_id)
 310             entries = [self.url_result(urljoin('https://mediasetinfinity.mediaset.it', url))
 311                        for url in re.findall(r'href="([^<>=]+SE\d{12},ST\d{12},sb\d{9})">[^<]+<', page)]
 312             title = self._html_extract_title(page).split('|')[0].strip()
 313             return self.playlist_result(entries, st or playlist_id, title)
 314
 315         entries = OnDemandPagedList(
 316             functools.partial(self._fetch_page, sb),
 317             self._PAGE_SIZE)
 318         title = try_get(entries, lambda x: x[0]['playlist_title'])
 319
 320         return self.playlist_result(entries, sb, title)