yt_dlp/extractor/rai.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_str,
   6     compat_urlparse,
   7 )
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     filter_dict,
  12     find_xpath_attr,
  13     fix_xml_ampersands,
  14     GeoRestrictedError,
  15     HEADRequest,
  16     int_or_none,
  17     join_nonempty,
  18     parse_duration,
  19     remove_start,
  20     strip_or_none,
  21     traverse_obj,
  22     try_get,
  23     unified_strdate,
  24     unified_timestamp,
  25     update_url_query,
  26     urljoin,
  27     xpath_text,
  28 )
  29
  30
  31 class RaiBaseIE(InfoExtractor):
  32     _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
  33     _GEO_COUNTRIES = ['IT']
  34     _GEO_BYPASS = False
  35
  36     def _extract_relinker_info(self, relinker_url, video_id, audio_only=False):
  37         if not re.match(r'https?://', relinker_url):
  38             return {'formats': [{'url': relinker_url}]}
  39
  40         formats = []
  41         geoprotection = None
  42         is_live = None
  43         duration = None
  44
  45         for platform in ('mon', 'flash', 'native'):
  46             relinker = self._download_xml(
  47                 relinker_url, video_id,
  48                 note='Downloading XML metadata for platform %s' % platform,
  49                 transform_source=fix_xml_ampersands,
  50                 query={'output': 45, 'pl': platform},
  51                 headers=self.geo_verification_headers())
  52
  53             if not geoprotection:
  54                 geoprotection = xpath_text(
  55                     relinker, './geoprotection', default=None) == 'Y'
  56
  57             if not is_live:
  58                 is_live = xpath_text(
  59                     relinker, './is_live', default=None) == 'Y'
  60             if not duration:
  61                 duration = parse_duration(xpath_text(
  62                     relinker, './duration', default=None))
  63
  64             url_elem = find_xpath_attr(relinker, './url', 'type', 'content')
  65             if url_elem is None:
  66                 continue
  67
  68             media_url = url_elem.text
  69
  70             # This does not imply geo restriction (e.g.
  71             # http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html)
  72             if '/video_no_available.mp4' in media_url:
  73                 continue
  74
  75             ext = determine_ext(media_url)
  76             if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
  77                 continue
  78
  79             if ext == 'mp3':
  80                 formats.append({
  81                     'url': media_url,
  82                     'vcodec': 'none',
  83                     'acodec': 'mp3',
  84                     'format_id': 'http-mp3',
  85                 })
  86                 break
  87             elif ext == 'm3u8' or 'format=m3u8' in media_url or platform == 'mon':
  88                 formats.extend(self._extract_m3u8_formats(
  89                     media_url, video_id, 'mp4', 'm3u8_native',
  90                     m3u8_id='hls', fatal=False))
  91             elif ext == 'f4m' or platform == 'flash':
  92                 manifest_url = update_url_query(
  93                     media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
  94                     {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
  95                 formats.extend(self._extract_f4m_formats(
  96                     manifest_url, video_id, f4m_id='hds', fatal=False))
  97             else:
  98                 bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
  99                 formats.append({
 100                     'url': media_url,
 101                     'tbr': bitrate if bitrate > 0 else None,
 102                     'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
 103                 })
 104
 105         if not formats and geoprotection is True:
 106             self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True)
 107
 108         if not audio_only:
 109             formats.extend(self._create_http_urls(relinker_url, formats))
 110
 111         return filter_dict({
 112             'is_live': is_live,
 113             'duration': duration,
 114             'formats': formats,
 115         })
 116
 117     def _create_http_urls(self, relinker_url, fmts):
 118         _RELINKER_REG = r'https?://(?P<host>[^/]+?)/(?:i/)?(?P<extra>[^/]+?)/(?P<path>.+?)/(?P<id>\w+)(?:_(?P<quality>[\d\,]+))?(?:\.mp4|/playlist\.m3u8).+?'
 119         _MP4_TMPL = '%s&overrideUserAgentRule=mp4-%s'
 120         _QUALITY = {
 121             # tbr: w, h
 122             '250': [352, 198],
 123             '400': [512, 288],
 124             '700': [512, 288],
 125             '800': [700, 394],
 126             '1200': [736, 414],
 127             '1800': [1024, 576],
 128             '2400': [1280, 720],
 129             '3200': [1440, 810],
 130             '3600': [1440, 810],
 131             '5000': [1920, 1080],
 132             '10000': [1920, 1080],
 133         }
 134
 135         def test_url(url):
 136             resp = self._request_webpage(
 137                 HEADRequest(url), None, headers={'User-Agent': 'Rai'},
 138                 fatal=False, errnote=False, note=False)
 139
 140             if resp is False:
 141                 return False
 142
 143             if resp.code == 200:
 144                 return False if resp.url == url else resp.url
 145             return None
 146
 147         # filter out audio-only formats
 148         fmts = [f for f in fmts if not f.get('vcodec') == 'none']
 149
 150         def get_format_info(tbr):
 151             import math
 152             br = int_or_none(tbr)
 153             if len(fmts) == 1 and not br:
 154                 br = fmts[0].get('tbr')
 155             if br > 300:
 156                 tbr = compat_str(math.floor(br / 100) * 100)
 157             else:
 158                 tbr = '250'
 159
 160             # try extracting info from available m3u8 formats
 161             format_copy = None
 162             for f in fmts:
 163                 if f.get('tbr'):
 164                     br_limit = math.floor(br / 100)
 165                     if br_limit - 1 <= math.floor(f['tbr'] / 100) <= br_limit + 1:
 166                         format_copy = f.copy()
 167             return {
 168                 'width': format_copy.get('width'),
 169                 'height': format_copy.get('height'),
 170                 'tbr': format_copy.get('tbr'),
 171                 'vcodec': format_copy.get('vcodec'),
 172                 'acodec': format_copy.get('acodec'),
 173                 'fps': format_copy.get('fps'),
 174                 'format_id': 'https-%s' % tbr,
 175             } if format_copy else {
 176                 'width': _QUALITY[tbr][0],
 177                 'height': _QUALITY[tbr][1],
 178                 'format_id': 'https-%s' % tbr,
 179                 'tbr': int(tbr),
 180             }
 181
 182         loc = test_url(_MP4_TMPL % (relinker_url, '*'))
 183         if not isinstance(loc, compat_str):
 184             return []
 185
 186         mobj = re.match(
 187             _RELINKER_REG,
 188             test_url(relinker_url) or '')
 189         if not mobj:
 190             return []
 191
 192         available_qualities = mobj.group('quality').split(',') if mobj.group('quality') else ['*']
 193         available_qualities = [i for i in available_qualities if i]
 194
 195         formats = []
 196         for q in available_qualities:
 197             fmt = {
 198                 'url': _MP4_TMPL % (relinker_url, q),
 199                 'protocol': 'https',
 200                 'ext': 'mp4',
 201             }
 202             fmt.update(get_format_info(q))
 203             formats.append(fmt)
 204         return formats
 205
 206     @staticmethod
 207     def _extract_subtitles(url, video_data):
 208         STL_EXT = 'stl'
 209         SRT_EXT = 'srt'
 210         subtitles = {}
 211         subtitles_array = video_data.get('subtitlesArray') or []
 212         for k in ('subtitles', 'subtitlesUrl'):
 213             subtitles_array.append({'url': video_data.get(k)})
 214         for subtitle in subtitles_array:
 215             sub_url = subtitle.get('url')
 216             if sub_url and isinstance(sub_url, compat_str):
 217                 sub_lang = subtitle.get('language') or 'it'
 218                 sub_url = urljoin(url, sub_url)
 219                 sub_ext = determine_ext(sub_url, SRT_EXT)
 220                 subtitles.setdefault(sub_lang, []).append({
 221                     'ext': sub_ext,
 222                     'url': sub_url,
 223                 })
 224                 if STL_EXT == sub_ext:
 225                     subtitles[sub_lang].append({
 226                         'ext': SRT_EXT,
 227                         'url': sub_url[:-len(STL_EXT)] + SRT_EXT,
 228                     })
 229         return subtitles
 230
 231
 232 class RaiPlayIE(RaiBaseIE):
 233     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>%s))\.(?:html|json)' % RaiBaseIE._UUID_RE
 234     _TESTS = [{
 235         'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
 236         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 237         'info_dict': {
 238             'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
 239             'ext': 'mp4',
 240             'title': 'Report del 07/04/2014',
 241             'alt_title': 'St 2013/14 - Report - Espresso nel caffè - 07/04/2014',
 242             'description': 'md5:d730c168a58f4bb35600fc2f881ec04e',
 243             'thumbnail': r're:^https?://.*\.jpg$',
 244             'uploader': 'Rai Gulp',
 245             'duration': 6160,
 246             'series': 'Report',
 247             'season': '2013/14',
 248             'subtitles': {
 249                 'it': 'count:4',
 250             },
 251         },
 252         'params': {
 253             'skip_download': True,
 254         },
 255     }, {
 256         # 1080p direct mp4 url
 257         'url': 'https://www.raiplay.it/video/2021/11/Blanca-S1E1-Senza-occhi-b1255a4a-8e72-4a2f-b9f3-fc1308e00736.html',
 258         'md5': 'aeda7243115380b2dd5e881fd42d949a',
 259         'info_dict': {
 260             'id': 'b1255a4a-8e72-4a2f-b9f3-fc1308e00736',
 261             'ext': 'mp4',
 262             'title': 'Blanca - S1E1 - Senza occhi',
 263             'alt_title': 'St 1 Ep 1 - Blanca - Senza occhi',
 264             'description': 'md5:75f95d5c030ec8bac263b1212322e28c',
 265             'thumbnail': r're:^https?://.*\.jpg$',
 266             'uploader': 'Rai 1',
 267             'duration': 6493,
 268             'series': 'Blanca',
 269             'season': 'Season 1',
 270         },
 271     }, {
 272         'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
 273         'only_matching': True,
 274     }, {
 275         # subtitles at 'subtitlesArray' key (see #27698)
 276         'url': 'https://www.raiplay.it/video/2020/12/Report---04-01-2021-2e90f1de-8eee-4de4-ac0e-78d21db5b600.html',
 277         'only_matching': True,
 278     }, {
 279         # DRM protected
 280         'url': 'https://www.raiplay.it/video/2020/09/Lo-straordinario-mondo-di-Zoey-S1E1-Lo-straordinario-potere-di-Zoey-ed493918-1d32-44b7-8454-862e473d00ff.html',
 281         'only_matching': True,
 282     }]
 283
 284     def _real_extract(self, url):
 285         base, video_id = self._match_valid_url(url).groups()
 286
 287         media = self._download_json(
 288             base + '.json', video_id, 'Downloading video JSON')
 289
 290         if not self.get_param('allow_unplayable_formats'):
 291             if try_get(
 292                     media,
 293                     (lambda x: x['rights_management']['rights']['drm'],
 294                      lambda x: x['program_info']['rights_management']['rights']['drm']),
 295                     dict):
 296                 self.report_drm(video_id)
 297
 298         title = media['name']
 299         video = media['video']
 300
 301         relinker_info = self._extract_relinker_info(video['content_url'], video_id)
 302         self._sort_formats(relinker_info['formats'])
 303
 304         thumbnails = []
 305         for _, value in media.get('images', {}).items():
 306             if value:
 307                 thumbnails.append({
 308                     'url': urljoin(url, value),
 309                 })
 310
 311         date_published = media.get('date_published')
 312         time_published = media.get('time_published')
 313         if date_published and time_published:
 314             date_published += ' ' + time_published
 315
 316         subtitles = self._extract_subtitles(url, video)
 317
 318         program_info = media.get('program_info') or {}
 319         season = media.get('season')
 320
 321         alt_title = join_nonempty(media.get('subtitle'), media.get('toptitle'), delim=' - ')
 322
 323         info = {
 324             'id': remove_start(media.get('id'), 'ContentItem-') or video_id,
 325             'display_id': video_id,
 326             'title': title,
 327             'alt_title': strip_or_none(alt_title),
 328             'description': media.get('description'),
 329             'uploader': strip_or_none(media.get('channel')),
 330             'creator': strip_or_none(media.get('editor') or None),
 331             'duration': parse_duration(video.get('duration')),
 332             'timestamp': unified_timestamp(date_published),
 333             'thumbnails': thumbnails,
 334             'series': program_info.get('name'),
 335             'season_number': int_or_none(season),
 336             'season': season if (season and not season.isdigit()) else None,
 337             'episode': media.get('episode_title'),
 338             'episode_number': int_or_none(media.get('episode')),
 339             'subtitles': subtitles,
 340             'release_year': traverse_obj(media, ('track_info', 'edit_year')),
 341         }
 342
 343         info.update(relinker_info)
 344         return info
 345
 346
 347 class RaiPlayLiveIE(RaiPlayIE):
 348     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+))'
 349     _TESTS = [{
 350         'url': 'http://www.raiplay.it/dirette/rainews24',
 351         'info_dict': {
 352             'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
 353             'display_id': 'rainews24',
 354             'ext': 'mp4',
 355             'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 356             'description': 'md5:4d00bcf6dc98b27c6ec480de329d1497',
 357             'uploader': 'Rai News 24',
 358             'creator': 'Rai News 24',
 359             'is_live': True,
 360         },
 361         'params': {
 362             'skip_download': True,
 363         },
 364     }]
 365
 366
 367 class RaiPlayPlaylistIE(InfoExtractor):
 368     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/programmi/(?P<id>[^/?#&]+))(?:/(?P<extra_id>[^?#&]+))?'
 369     _TESTS = [{
 370         'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/',
 371         'info_dict': {
 372             'id': 'nondirloalmiocapo',
 373             'title': 'Non dirlo al mio capo',
 374             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 375         },
 376         'playlist_mincount': 12,
 377     }, {
 378         'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/episodi/stagione-2/',
 379         'info_dict': {
 380             'id': 'nondirloalmiocapo',
 381             'title': 'Non dirlo al mio capo - Stagione 2',
 382             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 383         },
 384         'playlist_mincount': 12,
 385     }]
 386
 387     def _real_extract(self, url):
 388         base, playlist_id, extra_id = self._match_valid_url(url).groups()
 389
 390         program = self._download_json(
 391             base + '.json', playlist_id, 'Downloading program JSON')
 392
 393         if extra_id:
 394             extra_id = extra_id.upper().rstrip('/')
 395
 396         playlist_title = program.get('name')
 397         entries = []
 398         for b in (program.get('blocks') or []):
 399             for s in (b.get('sets') or []):
 400                 if extra_id:
 401                     if extra_id != join_nonempty(
 402                             b.get('name'), s.get('name'), delim='/').replace(' ', '-').upper():
 403                         continue
 404                     playlist_title = join_nonempty(playlist_title, s.get('name'), delim=' - ')
 405
 406                 s_id = s.get('id')
 407                 if not s_id:
 408                     continue
 409                 medias = self._download_json(
 410                     '%s/%s.json' % (base, s_id), s_id,
 411                     'Downloading content set JSON', fatal=False)
 412                 if not medias:
 413                     continue
 414                 for m in (medias.get('items') or []):
 415                     path_id = m.get('path_id')
 416                     if not path_id:
 417                         continue
 418                     video_url = urljoin(url, path_id)
 419                     entries.append(self.url_result(
 420                         video_url, ie=RaiPlayIE.ie_key(),
 421                         video_id=RaiPlayIE._match_id(video_url)))
 422
 423         return self.playlist_result(
 424             entries, playlist_id, playlist_title,
 425             try_get(program, lambda x: x['program_info']['description']))
 426
 427
 428 class RaiPlaySoundIE(RaiBaseIE):
 429     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplaysound\.it/.+?-(?P<id>%s))\.(?:html|json)' % RaiBaseIE._UUID_RE
 430     _TESTS = [{
 431         'url': 'https://www.raiplaysound.it/audio/2021/12/IL-RUGGITO-DEL-CONIGLIO-1ebae2a7-7cdb-42bb-842e-fe0d193e9707.html',
 432         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 433         'info_dict': {
 434             'id': '1ebae2a7-7cdb-42bb-842e-fe0d193e9707',
 435             'ext': 'mp3',
 436             'title': 'Il Ruggito del Coniglio del 10/12/2021',
 437             'description': 'md5:2a17d2107e59a4a8faa0e18334139ee2',
 438             'thumbnail': r're:^https?://.*\.jpg$',
 439             'uploader': 'rai radio 2',
 440             'duration': 5685,
 441             'series': 'Il Ruggito del Coniglio',
 442         },
 443         'params': {
 444             'skip_download': True,
 445         },
 446     }]
 447
 448     def _real_extract(self, url):
 449         base, audio_id = self._match_valid_url(url).group('base', 'id')
 450         media = self._download_json(f'{base}.json', audio_id, 'Downloading audio JSON')
 451         uid = try_get(media, lambda x: remove_start(remove_start(x['uniquename'], 'ContentItem-'), 'Page-'))
 452
 453         info = {}
 454         formats = []
 455         relinkers = set(traverse_obj(media, (('downloadable_audio', 'audio', ('live', 'cards', 0, 'audio')), 'url')))
 456         for r in relinkers:
 457             info = self._extract_relinker_info(r, audio_id, True)
 458             formats.extend(info.get('formats'))
 459
 460         date_published = try_get(media, (lambda x: f'{x["create_date"]} {x.get("create_time") or ""}',
 461                                          lambda x: x['live']['create_date']))
 462
 463         podcast_info = traverse_obj(media, 'podcast_info', ('live', 'cards', 0)) or {}
 464         thumbnails = [{
 465             'url': urljoin(url, thumb_url),
 466         } for thumb_url in (podcast_info.get('images') or {}).values() if thumb_url]
 467
 468         return {
 469             **info,
 470             'id': uid or audio_id,
 471             'display_id': audio_id,
 472             'title': traverse_obj(media, 'title', 'episode_title'),
 473             'alt_title': traverse_obj(media, ('track_info', 'media_name')),
 474             'description': media.get('description'),
 475             'uploader': traverse_obj(media, ('track_info', 'channel'), expected_type=strip_or_none),
 476             'creator': traverse_obj(media, ('track_info', 'editor'), expected_type=strip_or_none),
 477             'timestamp': unified_timestamp(date_published),
 478             'thumbnails': thumbnails,
 479             'series': podcast_info.get('title'),
 480             'season_number': int_or_none(media.get('season')),
 481             'episode': media.get('episode_title'),
 482             'episode_number': int_or_none(media.get('episode')),
 483             'formats': formats,
 484         }
 485
 486
 487 class RaiPlaySoundLiveIE(RaiPlaySoundIE):
 488     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplaysound\.it/(?P<id>[^/?#&]+)$)'
 489     _TESTS = [{
 490         'url': 'https://www.raiplaysound.it/radio2',
 491         'info_dict': {
 492             'id': 'b00a50e6-f404-4af6-8f8c-ff3b9af73a44',
 493             'display_id': 'radio2',
 494             'ext': 'mp4',
 495             'title': 'Rai Radio 2',
 496             'uploader': 'rai radio 2',
 497             'creator': 'raiplaysound',
 498             'is_live': True,
 499         },
 500         'params': {
 501             'skip_download': 'live',
 502         },
 503     }]
 504
 505
 506 class RaiPlaySoundPlaylistIE(InfoExtractor):
 507     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplaysound\.it/(?:programmi|playlist|audiolibri)/(?P<id>[^/?#&]+))(?:/(?P<extra_id>[^?#&]+))?'
 508     _TESTS = [{
 509         'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio',
 510         'info_dict': {
 511             'id': 'ilruggitodelconiglio',
 512             'title': 'Il Ruggito del Coniglio',
 513             'description': 'md5:1bbaf631245a7ab1ec4d9fbb3c7aa8f3',
 514         },
 515         'playlist_mincount': 65,
 516     }, {
 517         'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio/puntate/prima-stagione-1995',
 518         'info_dict': {
 519             'id': 'ilruggitodelconiglio_puntate_prima-stagione-1995',
 520             'title': 'Prima Stagione 1995',
 521         },
 522         'playlist_count': 1,
 523     }]
 524
 525     def _real_extract(self, url):
 526         base, playlist_id, extra_id = self._match_valid_url(url).group('base', 'id', 'extra_id')
 527         url = f'{base}.json'
 528         program = self._download_json(url, playlist_id, 'Downloading program JSON')
 529
 530         if extra_id:
 531             extra_id = extra_id.rstrip('/')
 532             playlist_id += '_' + extra_id.replace('/', '_')
 533             path = next(c['path_id'] for c in program.get('filters') or [] if extra_id in c.get('weblink'))
 534             program = self._download_json(
 535                 urljoin('https://www.raiplaysound.it', path), playlist_id, 'Downloading program secondary JSON')
 536
 537         entries = [
 538             self.url_result(urljoin(base, c['path_id']), ie=RaiPlaySoundIE.ie_key())
 539             for c in traverse_obj(program, 'cards', ('block', 'cards')) or []
 540             if c.get('path_id')]
 541
 542         return self.playlist_result(entries, playlist_id, program.get('title'),
 543                                     traverse_obj(program, ('podcast_info', 'description')))
 544
 545
 546 class RaiIE(RaiBaseIE):
 547     _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
 548     _TESTS = [{
 549         # var uniquename = "ContentItem-..."
 550         # data-id="ContentItem-..."
 551         'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 552         'info_dict': {
 553             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 554             'ext': 'mp4',
 555             'title': 'TG PRIMO TEMPO',
 556             'thumbnail': r're:^https?://.*\.jpg$',
 557             'duration': 1758,
 558             'upload_date': '20140612',
 559         },
 560         'skip': 'This content is available only in Italy',
 561     }, {
 562         # with ContentItem in many metas
 563         'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
 564         'info_dict': {
 565             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 566             'ext': 'mp4',
 567             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 568             'description': 'I film in uscita questa settimana.',
 569             'thumbnail': r're:^https?://.*\.png$',
 570             'duration': 833,
 571             'upload_date': '20161103',
 572         }
 573     }, {
 574         # with ContentItem in og:url
 575         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 576         'md5': '06345bd97c932f19ffb129973d07a020',
 577         'info_dict': {
 578             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 579             'ext': 'mp4',
 580             'title': 'TG1 ore 20:00 del 03/11/2016',
 581             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 582             'thumbnail': r're:^https?://.*\.jpg$',
 583             'duration': 2214,
 584             'upload_date': '20161103',
 585         }
 586     }, {
 587         # initEdizione('ContentItem-...'
 588         'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
 589         'info_dict': {
 590             'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
 591             'ext': 'mp4',
 592             'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
 593             'duration': 2274,
 594             'upload_date': '20170401',
 595         },
 596         'skip': 'Changes daily',
 597     }, {
 598         # HLS live stream with ContentItem in og:url
 599         'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
 600         'info_dict': {
 601             'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
 602             'ext': 'mp4',
 603             'title': 'La diretta di Rainews24',
 604         },
 605         'params': {
 606             'skip_download': True,
 607         },
 608     }, {
 609         # Direct MMS URL
 610         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
 611         'only_matching': True,
 612     }, {
 613         'url': 'https://www.rainews.it/tgr/marche/notiziari/video/2019/02/ContentItem-6ba945a2-889c-4a80-bdeb-8489c70a8db9.html',
 614         'only_matching': True,
 615     }]
 616
 617     def _extract_from_content_id(self, content_id, url):
 618         media = self._download_json(
 619             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
 620             content_id, 'Downloading video JSON')
 621
 622         title = media['name'].strip()
 623
 624         media_type = media['type']
 625         if 'Audio' in media_type:
 626             relinker_info = {
 627                 'formats': [{
 628                     'format_id': media.get('formatoAudio'),
 629                     'url': media['audioUrl'],
 630                     'ext': media.get('formatoAudio'),
 631                 }]
 632             }
 633         elif 'Video' in media_type:
 634             relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
 635         else:
 636             raise ExtractorError('not a media file')
 637
 638         self._sort_formats(relinker_info['formats'])
 639
 640         thumbnails = []
 641         for image_type in ('image', 'image_medium', 'image_300'):
 642             thumbnail_url = media.get(image_type)
 643             if thumbnail_url:
 644                 thumbnails.append({
 645                     'url': compat_urlparse.urljoin(url, thumbnail_url),
 646                 })
 647
 648         subtitles = self._extract_subtitles(url, media)
 649
 650         info = {
 651             'id': content_id,
 652             'title': title,
 653             'description': strip_or_none(media.get('desc')),
 654             'thumbnails': thumbnails,
 655             'uploader': media.get('author'),
 656             'upload_date': unified_strdate(media.get('date')),
 657             'duration': parse_duration(media.get('length')),
 658             'subtitles': subtitles,
 659         }
 660
 661         info.update(relinker_info)
 662
 663         return info
 664
 665     def _real_extract(self, url):
 666         video_id = self._match_id(url)
 667
 668         webpage = self._download_webpage(url, video_id)
 669
 670         content_item_id = None
 671
 672         content_item_url = self._html_search_meta(
 673             ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
 674              'twitter:player', 'jsonlink'), webpage, default=None)
 675         if content_item_url:
 676             content_item_id = self._search_regex(
 677                 r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
 678                 'content item id', default=None)
 679
 680         if not content_item_id:
 681             content_item_id = self._search_regex(
 682                 r'''(?x)
 683                     (?:
 684                         (?:initEdizione|drawMediaRaiTV)\(|
 685                         <(?:[^>]+\bdata-id|var\s+uniquename)=|
 686                         <iframe[^>]+\bsrc=
 687                     )
 688                     (["\'])
 689                     (?:(?!\1).)*\bContentItem-(?P<id>%s)
 690                 ''' % self._UUID_RE,
 691                 webpage, 'content item id', default=None, group='id')
 692
 693         content_item_ids = set()
 694         if content_item_id:
 695             content_item_ids.add(content_item_id)
 696         if video_id not in content_item_ids:
 697             content_item_ids.add(video_id)
 698
 699         for content_item_id in content_item_ids:
 700             try:
 701                 return self._extract_from_content_id(content_item_id, url)
 702             except GeoRestrictedError:
 703                 raise
 704             except ExtractorError:
 705                 pass
 706
 707         relinker_url = self._proto_relative_url(self._search_regex(
 708             r'''(?x)
 709                 (?:
 710                     var\s+videoURL|
 711                     mediaInfo\.mediaUri
 712                 )\s*=\s*
 713                 ([\'"])
 714                 (?P<url>
 715                     (?:https?:)?
 716                     //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
 717                     (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
 718             ''',
 719             webpage, 'relinker URL', group='url'))
 720
 721         relinker_info = self._extract_relinker_info(
 722             urljoin(url, relinker_url), video_id)
 723         self._sort_formats(relinker_info['formats'])
 724
 725         title = self._search_regex(
 726             r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
 727             webpage, 'title', group='title',
 728             default=None) or self._og_search_title(webpage)
 729
 730         info = {
 731             'id': video_id,
 732             'title': title,
 733         }
 734
 735         info.update(relinker_info)
 736
 737         return info