yt_dlp/extractor/rai.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..networking import HEADRequest
   5 from ..utils import (
   6     clean_html,
   7     determine_ext,
   8     ExtractorError,
   9     filter_dict,
  10     GeoRestrictedError,
  11     int_or_none,
  12     join_nonempty,
  13     parse_duration,
  14     remove_start,
  15     strip_or_none,
  16     traverse_obj,
  17     try_get,
  18     unified_strdate,
  19     unified_timestamp,
  20     update_url_query,
  21     urljoin,
  22     xpath_text,
  23 )
  24
  25
  26 class RaiBaseIE(InfoExtractor):
  27     _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
  28     _GEO_COUNTRIES = ['IT']
  29     _GEO_BYPASS = False
  30
  31     def _extract_relinker_info(self, relinker_url, video_id, audio_only=False):
  32         def fix_cdata(s):
  33             # remove \r\n\t before and after <![CDATA[ ]]> to avoid
  34             # polluted text with xpath_text
  35             s = re.sub(r'(\]\]>)[\r\n\t]+(</)', '\\1\\2', s)
  36             return re.sub(r'(>)[\r\n\t]+(<!\[CDATA\[)', '\\1\\2', s)
  37
  38         if not re.match(r'https?://', relinker_url):
  39             return {'formats': [{'url': relinker_url}]}
  40
  41         # set User-Agent to generic 'Rai' to avoid quality filtering from
  42         # the media server and get the maximum qualities available
  43         relinker = self._download_xml(
  44             relinker_url, video_id, note='Downloading XML metadata',
  45             transform_source=fix_cdata, query={'output': 64},
  46             headers={**self.geo_verification_headers(), 'User-Agent': 'Rai'})
  47
  48         if xpath_text(relinker, './license_url', default='{}') != '{}':
  49             self.report_drm(video_id)
  50
  51         is_live = xpath_text(relinker, './is_live', default='N') == 'Y'
  52         duration = parse_duration(xpath_text(relinker, './duration', default=None))
  53         media_url = xpath_text(relinker, './url[@type="content"]', default=None)
  54
  55         if not media_url:
  56             self.raise_no_formats('The relinker returned no media url')
  57
  58         # geo flag is a bit unreliable and not properly set all the time
  59         geoprotection = xpath_text(relinker, './geoprotection', default='N') == 'Y'
  60
  61         ext = determine_ext(media_url)
  62         formats = []
  63
  64         if ext == 'mp3':
  65             formats.append({
  66                 'url': media_url,
  67                 'vcodec': 'none',
  68                 'acodec': 'mp3',
  69                 'format_id': 'https-mp3',
  70             })
  71         elif ext == 'm3u8' or 'format=m3u8' in media_url:
  72             formats.extend(self._extract_m3u8_formats(
  73                 media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
  74         elif ext == 'f4m':
  75             # very likely no longer needed. Cannot find any url that uses it.
  76             manifest_url = update_url_query(
  77                 media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
  78                 {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
  79             formats.extend(self._extract_f4m_formats(
  80                 manifest_url, video_id, f4m_id='hds', fatal=False))
  81         elif ext == 'mp4':
  82             bitrate = int_or_none(xpath_text(relinker, './bitrate'))
  83             formats.append({
  84                 'url': media_url,
  85                 'tbr': bitrate if bitrate > 0 else None,
  86                 'format_id': join_nonempty('https', bitrate, delim='-'),
  87             })
  88         else:
  89             raise ExtractorError('Unrecognized media file found')
  90
  91         if (not formats and geoprotection is True) or '/video_no_available.mp4' in media_url:
  92             self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True)
  93
  94         if not audio_only and not is_live:
  95             formats.extend(self._create_http_urls(media_url, relinker_url, formats, video_id))
  96
  97         return filter_dict({
  98             'is_live': is_live,
  99             'duration': duration,
 100             'formats': formats,
 101         })
 102
 103     def _create_http_urls(self, manifest_url, relinker_url, fmts, video_id):
 104         _MANIFEST_REG = r'/(?P<id>\w+)(?:_(?P<quality>[\d\,]+))?(?:\.mp4)?(?:\.csmil)?/playlist\.m3u8'
 105         _MP4_TMPL = '%s&overrideUserAgentRule=mp4-%s'
 106         _QUALITY = {
 107             # tbr: w, h
 108             250: [352, 198],
 109             400: [512, 288],
 110             600: [512, 288],
 111             700: [512, 288],
 112             800: [700, 394],
 113             1200: [736, 414],
 114             1500: [920, 518],
 115             1800: [1024, 576],
 116             2400: [1280, 720],
 117             3200: [1440, 810],
 118             3600: [1440, 810],
 119             5000: [1920, 1080],
 120             10000: [1920, 1080],
 121         }
 122
 123         def percentage(number, target, pc=20, roof=125):
 124             '''check if the target is in the range of number +/- percent'''
 125             if not number or number < 0:
 126                 return False
 127             return abs(target - number) < min(float(number) * float(pc) / 100.0, roof)
 128
 129         def get_format_info(tbr):
 130             import math
 131             br = int_or_none(tbr)
 132             if len(fmts) == 1 and not br:
 133                 br = fmts[0].get('tbr')
 134             if br and br > 300:
 135                 tbr = math.floor(br / 100) * 100
 136             else:
 137                 tbr = 250
 138
 139             # try extracting info from available m3u8 formats
 140             format_copy = [None, None]
 141             for f in fmts:
 142                 if f.get('tbr'):
 143                     if percentage(tbr, f['tbr']):
 144                         format_copy[0] = f.copy()
 145                 if [f.get('width'), f.get('height')] == _QUALITY.get(tbr):
 146                     format_copy[1] = f.copy()
 147                     format_copy[1]['tbr'] = tbr
 148
 149             # prefer format with similar bitrate because there might be
 150             # multiple video with the same resolution but different bitrate
 151             format_copy = format_copy[0] or format_copy[1] or {}
 152             return {
 153                 'format_id': f'https-{tbr}',
 154                 'width': format_copy.get('width'),
 155                 'height': format_copy.get('height'),
 156                 'tbr': format_copy.get('tbr'),
 157                 'vcodec': format_copy.get('vcodec'),
 158                 'acodec': format_copy.get('acodec'),
 159                 'fps': format_copy.get('fps'),
 160             } if format_copy else {
 161                 'format_id': f'https-{tbr}',
 162                 'width': _QUALITY[tbr][0],
 163                 'height': _QUALITY[tbr][1],
 164                 'tbr': tbr,
 165                 'vcodec': 'avc1',
 166                 'acodec': 'mp4a',
 167                 'fps': 25,
 168             }
 169
 170         # Check if MP4 download is available
 171         try:
 172             self._request_webpage(
 173                 HEADRequest(_MP4_TMPL % (relinker_url, '*')), video_id, 'Checking MP4 availability')
 174         except ExtractorError as e:
 175             self.to_screen(f'{video_id}: MP4 direct download is not available: {e.cause}')
 176             return []
 177
 178         # filter out single-stream formats
 179         fmts = [f for f in fmts
 180                 if not f.get('vcodec') == 'none' and not f.get('acodec') == 'none']
 181
 182         mobj = re.search(_MANIFEST_REG, manifest_url)
 183         if not mobj:
 184             return []
 185         available_qualities = mobj.group('quality').split(',') if mobj.group('quality') else ['*']
 186
 187         formats = []
 188         for q in filter(None, available_qualities):
 189             self.write_debug(f'Creating https format for quality {q}')
 190             formats.append({
 191                 'url': _MP4_TMPL % (relinker_url, q),
 192                 'protocol': 'https',
 193                 'ext': 'mp4',
 194                 **get_format_info(q)
 195             })
 196         return formats
 197
 198     @staticmethod
 199     def _get_thumbnails_list(thumbs, url):
 200         return [{
 201             'url': urljoin(url, thumb_url),
 202         } for thumb_url in (thumbs or {}).values() if thumb_url]
 203
 204     @staticmethod
 205     def _extract_subtitles(url, video_data):
 206         STL_EXT = 'stl'
 207         SRT_EXT = 'srt'
 208         subtitles = {}
 209         subtitles_array = video_data.get('subtitlesArray') or video_data.get('subtitleList') or []
 210         for k in ('subtitles', 'subtitlesUrl'):
 211             subtitles_array.append({'url': video_data.get(k)})
 212         for subtitle in subtitles_array:
 213             sub_url = subtitle.get('url')
 214             if sub_url and isinstance(sub_url, str):
 215                 sub_lang = subtitle.get('language') or 'it'
 216                 sub_url = urljoin(url, sub_url)
 217                 sub_ext = determine_ext(sub_url, SRT_EXT)
 218                 subtitles.setdefault(sub_lang, []).append({
 219                     'ext': sub_ext,
 220                     'url': sub_url,
 221                 })
 222                 if STL_EXT == sub_ext:
 223                     subtitles[sub_lang].append({
 224                         'ext': SRT_EXT,
 225                         'url': sub_url[:-len(STL_EXT)] + SRT_EXT,
 226                     })
 227         return subtitles
 228
 229
 230 class RaiPlayIE(RaiBaseIE):
 231     _VALID_URL = rf'(?P<base>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>{RaiBaseIE._UUID_RE}))\.(?:html|json)'
 232     _TESTS = [{
 233         'url': 'https://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
 234         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 235         'info_dict': {
 236             'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
 237             'ext': 'mp4',
 238             'title': 'Report del 07/04/2014',
 239             'alt_title': 'St 2013/14 - Report - Espresso nel caffè - 07/04/2014',
 240             'description': 'md5:d730c168a58f4bb35600fc2f881ec04e',
 241             'thumbnail': r're:^https?://www\.raiplay\.it/.+\.jpg',
 242             'uploader': 'Rai 3',
 243             'creator': 'Rai 3',
 244             'duration': 6160,
 245             'series': 'Report',
 246             'season': '2013/14',
 247             'subtitles': {'it': 'count:4'},
 248             'release_year': 2022,
 249             'episode': 'Espresso nel caffè - 07/04/2014',
 250             'timestamp': 1396919880,
 251             'upload_date': '20140408',
 252             'formats': 'count:4',
 253         },
 254         'params': {'skip_download': True},
 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://www\.raiplay\.it/dl/img/.+\.jpg',
 266             'uploader': 'Rai Premium',
 267             'creator': 'Rai Fiction',
 268             'duration': 6493,
 269             'series': 'Blanca',
 270             'season': 'Season 1',
 271             'episode_number': 1,
 272             'release_year': 2021,
 273             'season_number': 1,
 274             'episode': 'Senza occhi',
 275             'timestamp': 1637318940,
 276             'upload_date': '20211119',
 277             'formats': 'count:12',
 278         },
 279         'params': {'skip_download': True},
 280         'expected_warnings': ['Video not available. Likely due to geo-restriction.']
 281     }, {
 282         # 1500 quality
 283         'url': 'https://www.raiplay.it/video/2012/09/S1E11---Tutto-cio-che-luccica-0cab3323-732e-45d6-8e86-7704acab6598.html',
 284         'md5': 'a634d20e8ab2d43724c273563f6bf87a',
 285         'info_dict': {
 286             'id': '0cab3323-732e-45d6-8e86-7704acab6598',
 287             'ext': 'mp4',
 288             'title': 'Mia and Me - S1E11 - Tutto ciò che luccica',
 289             'alt_title': 'St 1 Ep 11 - Mia and Me - Tutto ciò che luccica',
 290             'description': 'md5:4969e594184b1920c4c1f2b704da9dea',
 291             'thumbnail': r're:^https?://.*\.jpg$',
 292             'uploader': 'Rai Gulp',
 293             'series': 'Mia and Me',
 294             'season': 'Season 1',
 295             'episode_number': 11,
 296             'release_year': 2015,
 297             'season_number': 1,
 298             'episode': 'Tutto ciò che luccica',
 299             'timestamp': 1348495020,
 300             'upload_date': '20120924',
 301         },
 302     }, {
 303         'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
 304         'only_matching': True,
 305     }, {
 306         # subtitles at 'subtitlesArray' key (see #27698)
 307         'url': 'https://www.raiplay.it/video/2020/12/Report---04-01-2021-2e90f1de-8eee-4de4-ac0e-78d21db5b600.html',
 308         'only_matching': True,
 309     }, {
 310         # DRM protected
 311         'url': 'https://www.raiplay.it/video/2021/06/Lo-straordinario-mondo-di-Zoey-S2E1-Lo-straordinario-ritorno-di-Zoey-3ba992de-2332-41ad-9214-73e32ab209f4.html',
 312         'only_matching': True,
 313     }]
 314
 315     def _real_extract(self, url):
 316         base, video_id = self._match_valid_url(url).groups()
 317
 318         media = self._download_json(
 319             f'{base}.json', video_id, 'Downloading video JSON')
 320
 321         if not self.get_param('allow_unplayable_formats'):
 322             if traverse_obj(media, (('program_info', None), 'rights_management', 'rights', 'drm')):
 323                 self.report_drm(video_id)
 324
 325         video = media['video']
 326         relinker_info = self._extract_relinker_info(video['content_url'], video_id)
 327         date_published = join_nonempty(
 328             media.get('date_published'), media.get('time_published'), delim=' ')
 329         season = media.get('season')
 330         alt_title = join_nonempty(media.get('subtitle'), media.get('toptitle'), delim=' - ')
 331
 332         return {
 333             'id': remove_start(media.get('id'), 'ContentItem-') or video_id,
 334             'display_id': video_id,
 335             'title': media.get('name'),
 336             'alt_title': strip_or_none(alt_title or None),
 337             'description': media.get('description'),
 338             'uploader': strip_or_none(
 339                 traverse_obj(media, ('program_info', 'channel'))
 340                 or media.get('channel') or None),
 341             'creator': strip_or_none(
 342                 traverse_obj(media, ('program_info', 'editor'))
 343                 or media.get('editor') or None),
 344             'duration': parse_duration(video.get('duration')),
 345             'timestamp': unified_timestamp(date_published),
 346             'thumbnails': self._get_thumbnails_list(media.get('images'), url),
 347             'series': traverse_obj(media, ('program_info', 'name')),
 348             'season_number': int_or_none(season),
 349             'season': season if (season and not season.isdigit()) else None,
 350             'episode': media.get('episode_title'),
 351             'episode_number': int_or_none(media.get('episode')),
 352             'subtitles': self._extract_subtitles(url, video),
 353             'release_year': int_or_none(traverse_obj(media, ('track_info', 'edit_year'))),
 354             **relinker_info
 355         }
 356
 357
 358 class RaiPlayLiveIE(RaiPlayIE):  # XXX: Do not subclass from concrete IE
 359     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+))'
 360     _TESTS = [{
 361         'url': 'http://www.raiplay.it/dirette/rainews24',
 362         'info_dict': {
 363             'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
 364             'display_id': 'rainews24',
 365             'ext': 'mp4',
 366             'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 367             'description': 'md5:4d00bcf6dc98b27c6ec480de329d1497',
 368             'uploader': 'Rai News 24',
 369             'creator': 'Rai News 24',
 370             'is_live': True,
 371             'live_status': 'is_live',
 372             'upload_date': '20090502',
 373             'timestamp': 1241276220,
 374             'formats': 'count:3',
 375         },
 376         'params': {'skip_download': True},
 377     }]
 378
 379
 380 class RaiPlayPlaylistIE(InfoExtractor):
 381     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/programmi/(?P<id>[^/?#&]+))(?:/(?P<extra_id>[^?#&]+))?'
 382     _TESTS = [{
 383         # entire series episodes + extras...
 384         'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/',
 385         'info_dict': {
 386             'id': 'nondirloalmiocapo',
 387             'title': 'Non dirlo al mio capo',
 388             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 389         },
 390         'playlist_mincount': 30,
 391     }, {
 392         # single season
 393         'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/episodi/stagione-2/',
 394         'info_dict': {
 395             'id': 'nondirloalmiocapo',
 396             'title': 'Non dirlo al mio capo - Stagione 2',
 397             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 398         },
 399         'playlist_count': 12,
 400     }]
 401
 402     def _real_extract(self, url):
 403         base, playlist_id, extra_id = self._match_valid_url(url).groups()
 404
 405         program = self._download_json(
 406             f'{base}.json', playlist_id, 'Downloading program JSON')
 407
 408         if extra_id:
 409             extra_id = extra_id.upper().rstrip('/')
 410
 411         playlist_title = program.get('name')
 412         entries = []
 413         for b in (program.get('blocks') or []):
 414             for s in (b.get('sets') or []):
 415                 if extra_id:
 416                     if extra_id != join_nonempty(
 417                             b.get('name'), s.get('name'), delim='/').replace(' ', '-').upper():
 418                         continue
 419                     playlist_title = join_nonempty(playlist_title, s.get('name'), delim=' - ')
 420
 421                 s_id = s.get('id')
 422                 if not s_id:
 423                     continue
 424                 medias = self._download_json(
 425                     f'{base}/{s_id}.json', s_id,
 426                     'Downloading content set JSON', fatal=False)
 427                 if not medias:
 428                     continue
 429                 for m in (medias.get('items') or []):
 430                     path_id = m.get('path_id')
 431                     if not path_id:
 432                         continue
 433                     video_url = urljoin(url, path_id)
 434                     entries.append(self.url_result(
 435                         video_url, ie=RaiPlayIE.ie_key(),
 436                         video_id=RaiPlayIE._match_id(video_url)))
 437
 438         return self.playlist_result(
 439             entries, playlist_id, playlist_title,
 440             try_get(program, lambda x: x['program_info']['description']))
 441
 442
 443 class RaiPlaySoundIE(RaiBaseIE):
 444     _VALID_URL = rf'(?P<base>https?://(?:www\.)?raiplaysound\.it/.+?-(?P<id>{RaiBaseIE._UUID_RE}))\.(?:html|json)'
 445     _TESTS = [{
 446         'url': 'https://www.raiplaysound.it/audio/2021/12/IL-RUGGITO-DEL-CONIGLIO-1ebae2a7-7cdb-42bb-842e-fe0d193e9707.html',
 447         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 448         'info_dict': {
 449             'id': '1ebae2a7-7cdb-42bb-842e-fe0d193e9707',
 450             'ext': 'mp3',
 451             'title': 'Il Ruggito del Coniglio del 10/12/2021',
 452             'alt_title': 'md5:0e6476cd57858bb0f3fcc835d305b455',
 453             'description': 'md5:2a17d2107e59a4a8faa0e18334139ee2',
 454             'thumbnail': r're:^https?://.+\.jpg$',
 455             'uploader': 'rai radio 2',
 456             'duration': 5685,
 457             'series': 'Il Ruggito del Coniglio',
 458             'episode': 'Il Ruggito del Coniglio del 10/12/2021',
 459             'creator': 'rai radio 2',
 460             'timestamp': 1638346620,
 461             'upload_date': '20211201',
 462         },
 463         'params': {'skip_download': True},
 464     }]
 465
 466     def _real_extract(self, url):
 467         base, audio_id = self._match_valid_url(url).group('base', 'id')
 468         media = self._download_json(f'{base}.json', audio_id, 'Downloading audio JSON')
 469         uid = try_get(media, lambda x: remove_start(remove_start(x['uniquename'], 'ContentItem-'), 'Page-'))
 470
 471         info = {}
 472         formats = []
 473         relinkers = set(traverse_obj(media, (('downloadable_audio', 'audio', ('live', 'cards', 0, 'audio')), 'url')))
 474         for r in relinkers:
 475             info = self._extract_relinker_info(r, audio_id, True)
 476             formats.extend(info.get('formats'))
 477
 478         date_published = try_get(media, (lambda x: f'{x["create_date"]} {x.get("create_time") or ""}',
 479                                          lambda x: x['live']['create_date']))
 480
 481         podcast_info = traverse_obj(media, 'podcast_info', ('live', 'cards', 0)) or {}
 482
 483         return {
 484             **info,
 485             'id': uid or audio_id,
 486             'display_id': audio_id,
 487             'title': traverse_obj(media, 'title', 'episode_title'),
 488             'alt_title': traverse_obj(media, ('track_info', 'media_name'), expected_type=strip_or_none),
 489             'description': media.get('description'),
 490             'uploader': traverse_obj(media, ('track_info', 'channel'), expected_type=strip_or_none),
 491             'creator': traverse_obj(media, ('track_info', 'editor'), expected_type=strip_or_none),
 492             'timestamp': unified_timestamp(date_published),
 493             'thumbnails': self._get_thumbnails_list(podcast_info.get('images'), url),
 494             'series': podcast_info.get('title'),
 495             'season_number': int_or_none(media.get('season')),
 496             'episode': media.get('episode_title'),
 497             'episode_number': int_or_none(media.get('episode')),
 498             'formats': formats,
 499         }
 500
 501
 502 class RaiPlaySoundLiveIE(RaiPlaySoundIE):  # XXX: Do not subclass from concrete IE
 503     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplaysound\.it/(?P<id>[^/?#&]+)$)'
 504     _TESTS = [{
 505         'url': 'https://www.raiplaysound.it/radio2',
 506         'info_dict': {
 507             'id': 'b00a50e6-f404-4af6-8f8c-ff3b9af73a44',
 508             'display_id': 'radio2',
 509             'ext': 'mp4',
 510             'title': r're:Rai Radio 2 \d+-\d+-\d+ \d+:\d+',
 511             'thumbnail': r're:^https://www\.raiplaysound\.it/dl/img/.+\.png',
 512             'uploader': 'rai radio 2',
 513             'series': 'Rai Radio 2',
 514             'creator': 'raiplaysound',
 515             'is_live': True,
 516             'live_status': 'is_live',
 517         },
 518         'params': {'skip_download': True},
 519     }]
 520
 521
 522 class RaiPlaySoundPlaylistIE(InfoExtractor):
 523     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplaysound\.it/(?:programmi|playlist|audiolibri)/(?P<id>[^/?#&]+))(?:/(?P<extra_id>[^?#&]+))?'
 524     _TESTS = [{
 525         # entire show
 526         'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio',
 527         'info_dict': {
 528             'id': 'ilruggitodelconiglio',
 529             'title': 'Il Ruggito del Coniglio',
 530             'description': 'md5:48cff6972435964284614d70474132e6',
 531         },
 532         'playlist_mincount': 65,
 533     }, {
 534         # single season
 535         'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio/puntate/prima-stagione-1995',
 536         'info_dict': {
 537             'id': 'ilruggitodelconiglio_puntate_prima-stagione-1995',
 538             'title': 'Prima Stagione 1995',
 539         },
 540         'playlist_count': 1,
 541     }]
 542
 543     def _real_extract(self, url):
 544         base, playlist_id, extra_id = self._match_valid_url(url).group('base', 'id', 'extra_id')
 545         url = f'{base}.json'
 546         program = self._download_json(url, playlist_id, 'Downloading program JSON')
 547
 548         if extra_id:
 549             extra_id = extra_id.rstrip('/')
 550             playlist_id += '_' + extra_id.replace('/', '_')
 551             path = next(c['path_id'] for c in program.get('filters') or [] if extra_id in c.get('weblink'))
 552             program = self._download_json(
 553                 urljoin('https://www.raiplaysound.it', path), playlist_id, 'Downloading program secondary JSON')
 554
 555         entries = [
 556             self.url_result(urljoin(base, c['path_id']), ie=RaiPlaySoundIE.ie_key())
 557             for c in traverse_obj(program, 'cards', ('block', 'cards')) or []
 558             if c.get('path_id')]
 559
 560         return self.playlist_result(entries, playlist_id, program.get('title'),
 561                                     traverse_obj(program, ('podcast_info', 'description')))
 562
 563
 564 class RaiIE(RaiBaseIE):
 565     _VALID_URL = rf'https?://[^/]+\.(?:rai\.(?:it|tv))/.+?-(?P<id>{RaiBaseIE._UUID_RE})(?:-.+?)?\.html'
 566     _TESTS = [{
 567         'url': 'https://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 568         'info_dict': {
 569             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 570             'ext': 'mp4',
 571             'title': 'TG PRIMO TEMPO',
 572             'thumbnail': r're:^https?://.*\.jpg',
 573             'duration': 1758,
 574             'upload_date': '20140612',
 575         },
 576         'params': {'skip_download': True},
 577         'expected_warnings': ['Video not available. Likely due to geo-restriction.']
 578     }, {
 579         'url': 'https://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 580         'info_dict': {
 581             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 582             'ext': 'mp4',
 583             'title': 'TG1 ore 20:00 del 03/11/2016',
 584             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 585             'thumbnail': r're:^https?://.*\.jpg$',
 586             'duration': 2214,
 587             'upload_date': '20161103'
 588         },
 589         'params': {'skip_download': True},
 590     }, {
 591         # Direct MMS: Media URL no longer works.
 592         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
 593         'only_matching': True,
 594     }]
 595
 596     def _real_extract(self, url):
 597         content_id = self._match_id(url)
 598         media = self._download_json(
 599             f'https://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-{content_id}.html?json',
 600             content_id, 'Downloading video JSON', fatal=False, expected_status=404)
 601
 602         if media is None:
 603             return None
 604
 605         if 'Audio' in media['type']:
 606             relinker_info = {
 607                 'formats': [{
 608                     'format_id': join_nonempty('https', media.get('formatoAudio'), delim='-'),
 609                     'url': media['audioUrl'],
 610                     'ext': media.get('formatoAudio'),
 611                     'vcodec': 'none',
 612                     'acodec': media.get('formatoAudio'),
 613                 }]
 614             }
 615         elif 'Video' in media['type']:
 616             relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
 617         else:
 618             raise ExtractorError('not a media file')
 619
 620         thumbnails = self._get_thumbnails_list(
 621             {image_type: media.get(image_type) for image_type in (
 622                 'image', 'image_medium', 'image_300')}, url)
 623
 624         return {
 625             'id': content_id,
 626             'title': strip_or_none(media.get('name') or media.get('title')),
 627             'description': strip_or_none(media.get('desc')) or None,
 628             'thumbnails': thumbnails,
 629             'uploader': strip_or_none(media.get('author')) or None,
 630             'upload_date': unified_strdate(media.get('date')),
 631             'duration': parse_duration(media.get('length')),
 632             'subtitles': self._extract_subtitles(url, media),
 633             **relinker_info
 634         }
 635
 636
 637 class RaiNewsIE(RaiIE):  # XXX: Do not subclass from concrete IE
 638     _VALID_URL = rf'https?://(www\.)?rainews\.it/(?!articoli)[^?#]+-(?P<id>{RaiBaseIE._UUID_RE})(?:-[^/?#]+)?\.html'
 639     _EMBED_REGEX = [rf'<iframe[^>]+data-src="(?P<url>/iframe/[^?#]+?{RaiBaseIE._UUID_RE}\.html)']
 640     _TESTS = [{
 641         # new rainews player (#3911)
 642         'url': 'https://www.rainews.it/rubriche/24mm/video/2022/05/24mm-del-29052022-12cf645d-1ffd-4220-b27c-07c226dbdecf.html',
 643         'info_dict': {
 644             'id': '12cf645d-1ffd-4220-b27c-07c226dbdecf',
 645             'ext': 'mp4',
 646             'title': 'Puntata del 29/05/2022',
 647             'duration': 1589,
 648             'upload_date': '20220529',
 649             'uploader': 'rainews',
 650         },
 651         'params': {'skip_download': True},
 652     }, {
 653         # old content with fallback method to extract media urls
 654         'url': 'https://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',
 655         'info_dict': {
 656             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 657             'ext': 'mp4',
 658             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 659             'description': 'I film in uscita questa settimana.',
 660             'thumbnail': r're:^https?://.*\.png$',
 661             'duration': 833,
 662             'upload_date': '20161103'
 663         },
 664         'params': {'skip_download': True},
 665         'expected_warnings': ['unable to extract player_data'],
 666     }, {
 667         # iframe + drm
 668         'url': 'https://www.rainews.it/iframe/video/2022/07/euro2022-europei-calcio-femminile-italia-belgio-gol-0-1-video-4de06a69-de75-4e32-a657-02f0885f8118.html',
 669         'only_matching': True,
 670     }]
 671     _PLAYER_TAG = 'news'
 672
 673     def _real_extract(self, url):
 674         video_id = self._match_id(url)
 675
 676         webpage = self._download_webpage(url, video_id)
 677
 678         player_data = self._search_json(
 679             rf'<rai{self._PLAYER_TAG}-player\s*data=\'', webpage, 'player_data', video_id,
 680             transform_source=clean_html, default={})
 681         track_info = player_data.get('track_info')
 682         relinker_url = traverse_obj(player_data, 'mediapolis', 'content_url')
 683
 684         if not relinker_url:
 685             # fallback on old implementation for some old content
 686             try:
 687                 return self._extract_from_content_id(video_id, url)
 688             except GeoRestrictedError:
 689                 raise
 690             except ExtractorError as e:
 691                 raise ExtractorError('Relinker URL not found', cause=e)
 692
 693         relinker_info = self._extract_relinker_info(urljoin(url, relinker_url), video_id)
 694
 695         return {
 696             'id': video_id,
 697             'title': player_data.get('title') or track_info.get('title') or self._og_search_title(webpage),
 698             'upload_date': unified_strdate(track_info.get('date')),
 699             'uploader': strip_or_none(track_info.get('editor') or None),
 700             **relinker_info
 701         }
 702
 703
 704 class RaiCulturaIE(RaiNewsIE):  # XXX: Do not subclass from concrete IE
 705     _VALID_URL = rf'https?://(www\.)?raicultura\.it/(?!articoli)[^?#]+-(?P<id>{RaiBaseIE._UUID_RE})(?:-[^/?#]+)?\.html'
 706     _EMBED_REGEX = [rf'<iframe[^>]+data-src="(?P<url>/iframe/[^?#]+?{RaiBaseIE._UUID_RE}\.html)']
 707     _TESTS = [{
 708         'url': 'https://www.raicultura.it/letteratura/articoli/2018/12/Alberto-Asor-Rosa-Letteratura-e-potere-05ba8775-82b5-45c5-a89d-dd955fbde1fb.html',
 709         'info_dict': {
 710             'id': '05ba8775-82b5-45c5-a89d-dd955fbde1fb',
 711             'ext': 'mp4',
 712             'title': 'Alberto Asor Rosa: Letteratura e potere',
 713             'duration': 1756,
 714             'upload_date': '20181206',
 715             'uploader': 'raicultura',
 716             'formats': 'count:2',
 717         },
 718         'params': {'skip_download': True},
 719     }]
 720     _PLAYER_TAG = 'cultura'
 721
 722
 723 class RaiSudtirolIE(RaiBaseIE):
 724     _VALID_URL = r'https?://raisudtirol\.rai\.it/.+media=(?P<id>\w+)'
 725     _TESTS = [{
 726         # mp4 file
 727         'url': 'https://raisudtirol.rai.it/la/index.php?media=Ptv1619729460',
 728         'info_dict': {
 729             'id': 'Ptv1619729460',
 730             'ext': 'mp4',
 731             'title': 'Euro: trasmisciun d\'economia - 29-04-2021 20:51',
 732             'series': 'Euro: trasmisciun d\'economia',
 733             'upload_date': '20210429',
 734             'thumbnail': r're:https://raisudtirol\.rai\.it/img/.+\.jpg',
 735             'uploader': 'raisudtirol',
 736             'formats': 'count:1',
 737         },
 738         'params': {'skip_download': True},
 739     }, {
 740         # m3u manifest
 741         'url': 'https://raisudtirol.rai.it/it/kidsplayer.php?lang=it&media=GUGGUG_P1.smil',
 742         'info_dict': {
 743             'id': 'GUGGUG_P1',
 744             'ext': 'mp4',
 745             'title': 'GUGGUG! La Prospettiva - Die Perspektive',
 746             'uploader': 'raisudtirol',
 747             'formats': 'count:6',
 748         },
 749         'params': {'skip_download': True},
 750     }]
 751
 752     def _real_extract(self, url):
 753         video_id = self._match_id(url)
 754         webpage = self._download_webpage(url, video_id)
 755
 756         video_date = self._html_search_regex(
 757             r'<span class="med_data">(.+?)</span>', webpage, 'video_date', default=None)
 758         video_title = self._html_search_regex([
 759             r'<span class="med_title">(.+?)</span>', r'title: \'(.+?)\','],
 760             webpage, 'video_title', default=None)
 761         video_url = self._html_search_regex([
 762             r'sources:\s*\[\{file:\s*"(.+?)"\}\]',
 763             r'<source\s+src="(.+?)"\s+type="application/x-mpegURL"'],
 764             webpage, 'video_url', default=None)
 765
 766         ext = determine_ext(video_url)
 767         if ext == 'm3u8':
 768             formats = self._extract_m3u8_formats(video_url, video_id)
 769         elif ext == 'mp4':
 770             formats = [{
 771                 'format_id': 'https-mp4',
 772                 'url': self._proto_relative_url(video_url),
 773                 'width': 1024,
 774                 'height': 576,
 775                 'fps': 25,
 776                 'vcodec': 'avc1',
 777                 'acodec': 'mp4a',
 778             }]
 779         else:
 780             formats = []
 781             self.raise_no_formats(f'Unrecognized media file: {video_url}')
 782
 783         return {
 784             'id': video_id,
 785             'title': join_nonempty(video_title, video_date, delim=' - '),
 786             'series': video_title if video_date else None,
 787             'upload_date': unified_strdate(video_date),
 788             'thumbnail': urljoin('https://raisudtirol.rai.it/', self._html_search_regex(
 789                 r'image: \'(.+?)\'', webpage, 'video_thumb', default=None)),
 790             'uploader': 'raisudtirol',
 791             'formats': formats,
 792         }