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     int_or_none,
   9     OnDemandPagedList,
  10     parse_qs,
  11     try_get,
  12     urljoin,
  13     update_url_query,
  14 )
  15
  16
  17 class MediasetIE(ThePlatformBaseIE):
  18     _TP_TLD = 'eu'
  19     _VALID_URL = r'''(?x)
  20                     (?:
  21                         mediaset:|
  22                         https?://
  23                             (?:\w+\.)+mediaset\.it/
  24                             (?:
  25                                 (?:video|on-demand|movie)/(?:[^/]+/)+[^/]+_|
  26                                 player/(?:v\d+/)?index\.html\?.*?\bprogramGuid=
  27                             )
  28                     )(?P<id>[0-9A-Z]{16,})
  29                     '''
  30     _TESTS = [{
  31         # full episode
  32         'url': 'https://www.mediasetplay.mediaset.it/video/mrwronglezionidamore/episodio-1_F310575103000102',
  33         'md5': 'a7e75c6384871f322adb781d3bd72c26',
  34         'info_dict': {
  35             'id': 'F310575103000102',
  36             'ext': 'mp4',
  37             'title': 'Episodio 1',
  38             'description': 'md5:e8017b7d7194e9bfb75299c2b8d81e02',
  39             'thumbnail': r're:^https?://.*\.jpg$',
  40             'duration': 2682.0,
  41             'upload_date': '20210530',
  42             'series': 'Mr Wrong - Lezioni d\'amore',
  43             'timestamp': 1622413946,
  44             'uploader': 'Canale 5',
  45             'uploader_id': 'C5',
  46             'season': 'Season 1',
  47             'episode': 'Episode 1',
  48             'season_number': 1,
  49             'episode_number': 1,
  50             'chapters': [{'start_time': 0.0, 'end_time': 439.88}, {'start_time': 439.88, 'end_time': 1685.84}, {'start_time': 1685.84, 'end_time': 2682.0}],
  51         },
  52     }, {
  53         'url': 'https://www.mediasetplay.mediaset.it/video/matrix/puntata-del-25-maggio_F309013801000501',
  54         'md5': '1276f966ac423d16ba255ce867de073e',
  55         'info_dict': {
  56             'id': 'F309013801000501',
  57             'ext': 'mp4',
  58             'title': 'Puntata del 25 maggio',
  59             'description': 'md5:ee2e456e3eb1dba5e814596655bb5296',
  60             'thumbnail': r're:^https?://.*\.jpg$',
  61             'duration': 6565.008,
  62             'upload_date': '20200903',
  63             'series': 'Matrix',
  64             'timestamp': 1599172492,
  65             'uploader': 'Canale 5',
  66             'uploader_id': 'C5',
  67             'season': 'Season 5',
  68             'episode': 'Episode 5',
  69             'season_number': 5,
  70             'episode_number': 5,
  71             'chapters': [{'start_time': 0.0, 'end_time': 3409.08}, {'start_time': 3409.08, 'end_time': 6565.008}],
  72         },
  73     }, {
  74         'url': 'https://www.mediasetplay.mediaset.it/video/cameracafe5/episodio-69-pezzo-di-luna_F303843101017801',
  75         'md5': 'd1650ac9ff944f185556126a736df148',
  76         'info_dict': {
  77             'id': 'F303843101017801',
  78             'ext': 'mp4',
  79             'title': 'Episodio 69 - Pezzo di luna',
  80             'description': 'md5:7c32c8ec4118b72588b9412f11353f73',
  81             'thumbnail': r're:^https?://.*\.jpg$',
  82             'duration': 263.008,
  83             'upload_date': '20200902',
  84             'series': 'Camera Café 5',
  85             'timestamp': 1599064700,
  86             'uploader': 'Italia 1',
  87             'uploader_id': 'I1',
  88             'season': 'Season 5',
  89             'episode': 'Episode 178',
  90             'season_number': 5,
  91             'episode_number': 178,
  92             'chapters': [{'start_time': 0.0, 'end_time': 261.88}, {'start_time': 261.88, 'end_time': 263.008}],
  93         },
  94     }, {
  95         'url': 'https://www.mediasetplay.mediaset.it/video/cameracafe5/episodio-51-tu-chi-sei_F303843107000601',
  96         'md5': '567e9ad375b7a27a0e370650f572a1e3',
  97         'info_dict': {
  98             'id': 'F303843107000601',
  99             'ext': 'mp4',
 100             'title': 'Episodio 51 - Tu chi sei?',
 101             'description': 'md5:42ef006e56824cc31787a547590923f4',
 102             'thumbnail': r're:^https?://.*\.jpg$',
 103             'duration': 367.021,
 104             'upload_date': '20200902',
 105             'series': 'Camera Café 5',
 106             'timestamp': 1599069817,
 107             'uploader': 'Italia 1',
 108             'uploader_id': 'I1',
 109             'season': 'Season 5',
 110             'episode': 'Episode 6',
 111             'season_number': 5,
 112             'episode_number': 6,
 113             'chapters': [{'start_time': 0.0, 'end_time': 358.68}, {'start_time': 358.68, 'end_time': 367.021}],
 114         },
 115     }, {
 116         # movie
 117         'url': 'https://www.mediasetplay.mediaset.it/movie/selvaggi/selvaggi_F006474501000101',
 118         'md5': '720440187a2ae26af8148eb9e6b901ed',
 119         'info_dict': {
 120             'id': 'F006474501000101',
 121             'ext': 'mp4',
 122             'title': 'Selvaggi',
 123             'description': 'md5:cfdedbbfdd12d4d0e5dcf1fa1b75284f',
 124             'thumbnail': r're:^https?://.*\.jpg$',
 125             'duration': 5233.01,
 126             'upload_date': '20210729',
 127             'timestamp': 1627594716,
 128             'uploader': 'Cine34',
 129             'uploader_id': 'B6',
 130             'chapters': [{'start_time': 0.0, 'end_time': 1938.56}, {'start_time': 1938.56, 'end_time': 5233.01}],
 131         },
 132     }, {
 133         # clip
 134         'url': 'https://www.mediasetplay.mediaset.it/video/gogglebox/un-grande-classico-della-commedia-sexy_FAFU000000661680',
 135         'only_matching': True,
 136     }, {
 137         # iframe simple
 138         'url': 'https://static3.mediasetplay.mediaset.it/player/index.html?appKey=5ad3966b1de1c4000d5cec48&programGuid=FAFU000000665924&id=665924',
 139         'only_matching': True,
 140     }, {
 141         # iframe twitter (from http://www.wittytv.it/se-prima-mi-fidavo-zero/)
 142         'url': 'https://static3.mediasetplay.mediaset.it/player/index.html?appKey=5ad3966b1de1c4000d5cec48&programGuid=FAFU000000665104&id=665104',
 143         'only_matching': True,
 144     }, {
 145         'url': 'mediaset:FAFU000000665924',
 146         'only_matching': True,
 147     }, {
 148         'url': 'https://www.mediasetplay.mediaset.it/video/mediasethaacuoreilfuturo/palmieri-alicudi-lisola-dei-tre-bambini-felici--un-decreto-per-alicudi-e-tutte-le-microscuole_FD00000000102295',
 149         'only_matching': True,
 150     }, {
 151         'url': 'https://www.mediasetplay.mediaset.it/video/cherryseason/anticipazioni-degli-episodi-del-23-ottobre_F306837101005C02',
 152         'only_matching': True,
 153     }, {
 154         'url': 'https://www.mediasetplay.mediaset.it/video/tg5/ambiente-onda-umana-per-salvare-il-pianeta_F309453601079D01',
 155         'only_matching': True,
 156     }, {
 157         'url': 'https://www.mediasetplay.mediaset.it/video/grandefratellovip/benedetta-una-doccia-gelata_F309344401044C135',
 158         'only_matching': True,
 159     }, {
 160         'url': 'https://www.mediasetplay.mediaset.it/movie/herculeslaleggendahainizio/hercules-la-leggenda-ha-inizio_F305927501000102',
 161         'only_matching': True,
 162     }, {
 163         'url': 'https://mediasetinfinity.mediaset.it/video/braveandbeautiful/episodio-113_F310948005000402',
 164         'only_matching': True,
 165     }, {
 166         'url': 'https://static3.mediasetplay.mediaset.it/player/v2/index.html?partnerId=wittytv&configId=&programGuid=FD00000000153323',
 167         'only_matching': True,
 168     }]
 169
 170     def _extract_from_webpage(self, url, webpage):
 171         def _qs(url):
 172             return parse_qs(url)
 173
 174         def _program_guid(qs):
 175             return qs.get('programGuid', [None])[0]
 176
 177         entries = []
 178         for mobj in re.finditer(
 179                 r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//(?:www\.)?video\.mediaset\.it/player/playerIFrame(?:Twitter)?\.shtml.*?)\1',
 180                 webpage):
 181             embed_url = mobj.group('url')
 182             embed_qs = _qs(embed_url)
 183             program_guid = _program_guid(embed_qs)
 184             if program_guid:
 185                 entries.append(embed_url)
 186                 continue
 187             video_id = embed_qs.get('id', [None])[0]
 188             if not video_id:
 189                 continue
 190             urlh = self._request_webpage(embed_url, video_id, note='Following embed URL redirect')
 191             embed_url = urlh.geturl()
 192             program_guid = _program_guid(_qs(embed_url))
 193             if program_guid:
 194                 entries.append(embed_url)
 195         return entries
 196
 197     def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
 198         for video in smil.findall(self._xpath_ns('.//video', namespace)):
 199             video.attrib['src'] = re.sub(r'(https?://vod05)t(-mediaset-it\.akamaized\.net/.+?.mpd)\?.+', r'\1\2', video.attrib['src'])
 200         return super(MediasetIE, self)._parse_smil_formats(smil, smil_url, video_id, namespace, f4m_params, transform_rtmp_url)
 201
 202     def _check_drm_formats(self, tp_formats, video_id):
 203         has_nondrm, drm_manifest = False, ''
 204         for f in tp_formats:
 205             if '_sampleaes/' in (f.get('manifest_url') or ''):
 206                 drm_manifest = drm_manifest or f['manifest_url']
 207                 f['has_drm'] = True
 208             if not f.get('has_drm') and f.get('manifest_url'):
 209                 has_nondrm = True
 210
 211         nodrm_manifest = re.sub(r'_sampleaes/(\w+)_fp_', r'/\1_no_', drm_manifest)
 212         if has_nondrm or nodrm_manifest == drm_manifest:
 213             return
 214
 215         tp_formats.extend(self._extract_m3u8_formats(
 216             nodrm_manifest, video_id, m3u8_id='hls', fatal=False) or [])
 217
 218     def _real_extract(self, url):
 219         guid = self._match_id(url)
 220         tp_path = 'PR1GhC/media/guid/2702976343/' + guid
 221         info = self._extract_theplatform_metadata(tp_path, guid)
 222
 223         formats = []
 224         subtitles = {}
 225         first_e = geo_e = None
 226         asset_type = 'geoNo:HD,browser,geoIT|geoNo:HD,geoIT|geoNo:SD,browser,geoIT|geoNo:SD,geoIT|geoNo|HD|SD'
 227         # TODO: fixup ISM+none manifest URLs
 228         for f in ('MPEG4', 'M3U'):
 229             try:
 230                 tp_formats, tp_subtitles = self._extract_theplatform_smil(
 231                     update_url_query('http://link.theplatform.%s/s/%s' % (self._TP_TLD, tp_path), {
 232                         'mbr': 'true',
 233                         'formats': f,
 234                         'assetTypes': asset_type,
 235                     }), guid, 'Downloading %s SMIL data' % (f.split('+')[0]))
 236             except ExtractorError as e:
 237                 if not geo_e and isinstance(e, GeoRestrictedError):
 238                     geo_e = e
 239                 if not first_e:
 240                     first_e = e
 241                 continue
 242             self._check_drm_formats(tp_formats, guid)
 243             formats.extend(tp_formats)
 244             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
 245
 246         # check for errors and report them
 247         if (first_e or geo_e) and not formats:
 248             raise geo_e or first_e
 249
 250         self._sort_formats(formats)
 251
 252         feed_data = self._download_json(
 253             'https://feed.entertainment.tv.theplatform.eu/f/PR1GhC/mediaset-prod-all-programs-v2/guid/-/' + guid,
 254             guid, fatal=False)
 255         if feed_data:
 256             publish_info = feed_data.get('mediasetprogram$publishInfo') or {}
 257             thumbnails = feed_data.get('thumbnails') or {}
 258             thumbnail = None
 259             for key, value in thumbnails.items():
 260                 if key.startswith('image_keyframe_poster-'):
 261                     thumbnail = value.get('url')
 262                     break
 263
 264             info.update({
 265                 'description': info.get('description') or feed_data.get('description') or feed_data.get('longDescription'),
 266                 'uploader': publish_info.get('description'),
 267                 'uploader_id': publish_info.get('channel'),
 268                 'view_count': int_or_none(feed_data.get('mediasetprogram$numberOfViews')),
 269                 'thumbnail': thumbnail,
 270             })
 271
 272             if feed_data.get('programType') == 'episode':
 273                 info.update({
 274                     'episode_number': int_or_none(
 275                         feed_data.get('tvSeasonEpisodeNumber')),
 276                     'season_number': int_or_none(
 277                         feed_data.get('tvSeasonNumber')),
 278                     'series': feed_data.get('mediasetprogram$brandTitle'),
 279                 })
 280
 281         info.update({
 282             'id': guid,
 283             'formats': formats,
 284             'subtitles': subtitles,
 285         })
 286         return info
 287
 288
 289 class MediasetShowIE(MediasetIE):
 290     _VALID_URL = r'''(?x)
 291                     (?:
 292                         https?://
 293                             (\w+\.)+mediaset\.it/
 294                             (?:
 295                                 (?:fiction|programmi-tv|serie-tv|kids)/(?:.+?/)?
 296                                     (?:[a-z-]+)_SE(?P<id>\d{12})
 297                                     (?:,ST(?P<st>\d{12}))?
 298                                     (?:,sb(?P<sb>\d{9}))?$
 299                             )
 300                     )
 301                     '''
 302     _TESTS = [{
 303         # TV Show webpage (general webpage)
 304         'url': 'https://www.mediasetplay.mediaset.it/programmi-tv/leiene/leiene_SE000000000061',
 305         'info_dict': {
 306             'id': '000000000061',
 307             'title': 'Le Iene',
 308         },
 309         'playlist_mincount': 7,
 310     }, {
 311         # TV Show webpage (specific season)
 312         'url': 'https://www.mediasetplay.mediaset.it/programmi-tv/leiene/leiene_SE000000000061,ST000000002763',
 313         'info_dict': {
 314             'id': '000000002763',
 315             'title': 'Le Iene',
 316         },
 317         'playlist_mincount': 7,
 318     }, {
 319         # TV Show specific playlist (with multiple pages)
 320         'url': 'https://www.mediasetplay.mediaset.it/programmi-tv/leiene/iservizi_SE000000000061,ST000000002763,sb100013375',
 321         'info_dict': {
 322             'id': '100013375',
 323             'title': 'I servizi',
 324         },
 325         'playlist_mincount': 50,
 326     }]
 327
 328     _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'
 329     _PAGE_SIZE = 25
 330
 331     def _fetch_page(self, sb, page):
 332         lower_limit = page * self._PAGE_SIZE + 1
 333         upper_limit = lower_limit + self._PAGE_SIZE - 1
 334         content = self._download_json(
 335             self._BY_SUBBRAND % (sb, lower_limit, upper_limit), sb)
 336         for entry in content.get('entries') or []:
 337             yield self.url_result(
 338                 'mediaset:' + entry['guid'],
 339                 playlist_title=entry['mediasetprogram$subBrandDescription'])
 340
 341     def _real_extract(self, url):
 342         playlist_id, st, sb = self._match_valid_url(url).group('id', 'st', 'sb')
 343         if not sb:
 344             page = self._download_webpage(url, st or playlist_id)
 345             entries = [self.url_result(urljoin('https://www.mediasetplay.mediaset.it', url))
 346                        for url in re.findall(r'href="([^<>=]+SE\d{12},ST\d{12},sb\d{9})">[^<]+<', page)]
 347             title = (self._html_search_regex(r'(?s)<h1[^>]*>(.+?)</h1>', page, 'title', default=None)
 348                      or self._og_search_title(page))
 349             return self.playlist_result(entries, st or playlist_id, title)
 350
 351         entries = OnDemandPagedList(
 352             functools.partial(self._fetch_page, sb),
 353             self._PAGE_SIZE)
 354         title = try_get(entries, lambda x: x[0]['playlist_title'])
 355
 356         return self.playlist_result(entries, sb, title)