yt_dlp/extractor/rai.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..networking import HEADRequest
   5 from ..utils import (
   6     ExtractorError,
   7     GeoRestrictedError,
   8     clean_html,
   9     determine_ext,
  10     filter_dict,
  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 _fix_m3u8_formats(self, media_url, video_id):
  32         fmts = self._extract_m3u8_formats(
  33             media_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
  34
  35         # Fix malformed m3u8 manifests by setting audio-only/video-only formats
  36         for f in fmts:
  37             if not f.get('acodec'):
  38                 f['acodec'] = 'mp4a'
  39             if not f.get('vcodec'):
  40                 f['vcodec'] = 'avc1'
  41             man_url = f['url']
  42             if re.search(r'chunklist(?:_b\d+)*_ao[_.]', man_url):  # audio only
  43                 f['vcodec'] = 'none'
  44             elif re.search(r'chunklist(?:_b\d+)*_vo[_.]', man_url):  # video only
  45                 f['acodec'] = 'none'
  46             else:  # video+audio
  47                 if f['acodec'] == 'none':
  48                     f['acodec'] = 'mp4a'
  49                 if f['vcodec'] == 'none':
  50                     f['vcodec'] = 'avc1'
  51
  52         return fmts
  53
  54     def _extract_relinker_info(self, relinker_url, video_id, audio_only=False):
  55         def fix_cdata(s):
  56             # remove \r\n\t before and after <![CDATA[ ]]> to avoid
  57             # polluted text with xpath_text
  58             s = re.sub(r'(\]\]>)[\r\n\t]+(</)', '\\1\\2', s)
  59             return re.sub(r'(>)[\r\n\t]+(<!\[CDATA\[)', '\\1\\2', s)
  60
  61         if not re.match(r'https?://', relinker_url):
  62             return {'formats': [{'url': relinker_url}]}
  63
  64         # set User-Agent to generic 'Rai' to avoid quality filtering from
  65         # the media server and get the maximum qualities available
  66         relinker = self._download_xml(
  67             relinker_url, video_id, note='Downloading XML metadata',
  68             transform_source=fix_cdata, query={'output': 64},
  69             headers={**self.geo_verification_headers(), 'User-Agent': 'Rai'})
  70
  71         if xpath_text(relinker, './license_url', default='{}') != '{}':
  72             self.report_drm(video_id)
  73
  74         is_live = xpath_text(relinker, './is_live', default='N') == 'Y'
  75         duration = parse_duration(xpath_text(relinker, './duration', default=None))
  76         media_url = xpath_text(relinker, './url[@type="content"]', default=None)
  77
  78         if not media_url:
  79             self.raise_no_formats('The relinker returned no media url')
  80
  81         # geo flag is a bit unreliable and not properly set all the time
  82         geoprotection = xpath_text(relinker, './geoprotection', default='N') == 'Y'
  83
  84         ext = determine_ext(media_url)
  85         formats = []
  86
  87         if ext == 'mp3':
  88             formats.append({
  89                 'url': media_url,
  90                 'vcodec': 'none',
  91                 'acodec': 'mp3',
  92                 'format_id': 'https-mp3',
  93             })
  94         elif ext == 'm3u8' or 'format=m3u8' in media_url:
  95             formats.extend(self._fix_m3u8_formats(media_url, video_id))
  96         elif ext == 'f4m':
  97             # very likely no longer needed. Cannot find any url that uses it.
  98             manifest_url = update_url_query(
  99                 media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
 100                 {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
 101             formats.extend(self._extract_f4m_formats(
 102                 manifest_url, video_id, f4m_id='hds', fatal=False))
 103         elif ext == 'mp4':
 104             bitrate = int_or_none(xpath_text(relinker, './bitrate'))
 105             formats.append({
 106                 'url': media_url,
 107                 'tbr': bitrate if bitrate > 0 else None,
 108                 'format_id': join_nonempty('https', bitrate, delim='-'),
 109             })
 110         else:
 111             raise ExtractorError('Unrecognized media file found')
 112
 113         if (not formats and geoprotection is True) or '/video_no_available.mp4' in media_url:
 114             self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True)
 115
 116         if not audio_only and not is_live:
 117             formats.extend(self._create_http_urls(media_url, relinker_url, formats, video_id))
 118
 119         return filter_dict({
 120             'is_live': is_live,
 121             'duration': duration,
 122             'formats': formats,
 123         })
 124
 125     def _create_http_urls(self, manifest_url, relinker_url, fmts, video_id):
 126         _MANIFEST_REG = r'/(?P<id>\w+)(?:_(?P<quality>[\d\,]+))?(?:\.mp4)?(?:\.csmil)?/playlist\.m3u8'
 127         _MP4_TMPL = '%s&overrideUserAgentRule=mp4-%s'
 128         _QUALITY = {
 129             # tbr: w, h
 130             250: [352, 198],
 131             400: [512, 288],
 132             600: [512, 288],
 133             700: [512, 288],
 134             800: [700, 394],
 135             1200: [736, 414],
 136             1500: [920, 518],
 137             1800: [1024, 576],
 138             2400: [1280, 720],
 139             3200: [1440, 810],
 140             3600: [1440, 810],
 141             5000: [1920, 1080],
 142             10000: [1920, 1080],
 143         }
 144
 145         def percentage(number, target, pc=20, roof=125):
 146             """check if the target is in the range of number +/- percent"""
 147             if not number or number < 0:
 148                 return False
 149             return abs(target - number) < min(float(number) * float(pc) / 100.0, roof)
 150
 151         def get_format_info(tbr):
 152             import math
 153             br = int_or_none(tbr)
 154             if len(fmts) == 1 and not br:
 155                 br = fmts[0].get('tbr')
 156             if br and br > 300:
 157                 tbr = math.floor(br / 100) * 100
 158             else:
 159                 tbr = 250
 160
 161             # try extracting info from available m3u8 formats
 162             format_copy = [None, None]
 163             for f in fmts:
 164                 if f.get('tbr'):
 165                     if percentage(tbr, f['tbr']):
 166                         format_copy[0] = f.copy()
 167                 if [f.get('width'), f.get('height')] == _QUALITY.get(tbr):
 168                     format_copy[1] = f.copy()
 169                     format_copy[1]['tbr'] = tbr
 170
 171             # prefer format with similar bitrate because there might be
 172             # multiple video with the same resolution but different bitrate
 173             format_copy = format_copy[0] or format_copy[1] or {}
 174             return {
 175                 'format_id': f'https-{tbr}',
 176                 'width': format_copy.get('width'),
 177                 'height': format_copy.get('height'),
 178                 'tbr': format_copy.get('tbr') or tbr,
 179                 'vcodec': format_copy.get('vcodec') or 'avc1',
 180                 'acodec': format_copy.get('acodec') or 'mp4a',
 181                 'fps': format_copy.get('fps') or 25,
 182             } if format_copy else {
 183                 'format_id': f'https-{tbr}',
 184                 'width': _QUALITY[tbr][0],
 185                 'height': _QUALITY[tbr][1],
 186                 'tbr': tbr,
 187                 'vcodec': 'avc1',
 188                 'acodec': 'mp4a',
 189                 'fps': 25,
 190             }
 191
 192         # Check if MP4 download is available
 193         try:
 194             self._request_webpage(
 195                 HEADRequest(_MP4_TMPL % (relinker_url, '*')), video_id, 'Checking MP4 availability')
 196         except ExtractorError as e:
 197             self.to_screen(f'{video_id}: MP4 direct download is not available: {e.cause}')
 198             return []
 199
 200         # filter out single-stream formats
 201         fmts = [f for f in fmts
 202                 if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
 203
 204         mobj = re.search(_MANIFEST_REG, manifest_url)
 205         if not mobj:
 206             return []
 207         available_qualities = mobj.group('quality').split(',') if mobj.group('quality') else ['*']
 208
 209         formats = []
 210         for q in filter(None, available_qualities):
 211             self.write_debug(f'Creating https format for quality {q}')
 212             formats.append({
 213                 'url': _MP4_TMPL % (relinker_url, q),
 214                 'protocol': 'https',
 215                 'ext': 'mp4',
 216                 **get_format_info(q),
 217             })
 218         return formats
 219
 220     @staticmethod
 221     def _get_thumbnails_list(thumbs, url):
 222         return [{
 223             'url': urljoin(url, thumb_url),
 224         } for thumb_url in (thumbs or {}).values() if thumb_url]
 225
 226     @staticmethod
 227     def _extract_subtitles(url, video_data):
 228         STL_EXT = 'stl'
 229         SRT_EXT = 'srt'
 230         subtitles = {}
 231         subtitles_array = video_data.get('subtitlesArray') or video_data.get('subtitleList') or []
 232         for k in ('subtitles', 'subtitlesUrl'):
 233             subtitles_array.append({'url': video_data.get(k)})
 234         for subtitle in subtitles_array:
 235             sub_url = subtitle.get('url')
 236             if sub_url and isinstance(sub_url, str):
 237                 sub_lang = subtitle.get('language') or 'it'
 238                 sub_url = urljoin(url, sub_url)
 239                 sub_ext = determine_ext(sub_url, SRT_EXT)
 240                 subtitles.setdefault(sub_lang, []).append({
 241                     'ext': sub_ext,
 242                     'url': sub_url,
 243                 })
 244                 if STL_EXT == sub_ext:
 245                     subtitles[sub_lang].append({
 246                         'ext': SRT_EXT,
 247                         'url': sub_url[:-len(STL_EXT)] + SRT_EXT,
 248                     })
 249         return subtitles
 250
 251
 252 class RaiPlayIE(RaiBaseIE):
 253     _VALID_URL = rf'(?P<base>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>{RaiBaseIE._UUID_RE}))\.(?:html|json)'
 254     _TESTS = [{
 255         'url': 'https://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
 256         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 257         'info_dict': {
 258             'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
 259             'ext': 'mp4',
 260             'title': 'Report del 07/04/2014',
 261             'alt_title': 'St 2013/14 - Report - Espresso nel caffè - 07/04/2014',
 262             'description': 'md5:d730c168a58f4bb35600fc2f881ec04e',
 263             'thumbnail': r're:^https?://www\.raiplay\.it/.+\.jpg',
 264             'uploader': 'Rai 3',
 265             'creator': 'Rai 3',
 266             'duration': 6160,
 267             'series': 'Report',
 268             'season': '2013/14',
 269             'subtitles': {'it': 'count:4'},
 270             'release_year': 2024,
 271             'episode': 'Espresso nel caffè - 07/04/2014',
 272             'timestamp': 1396919880,
 273             'upload_date': '20140408',
 274             'formats': 'count:4',
 275         },
 276         'params': {'skip_download': True},
 277     }, {
 278         # 1080p
 279         'url': 'https://www.raiplay.it/video/2021/11/Blanca-S1E1-Senza-occhi-b1255a4a-8e72-4a2f-b9f3-fc1308e00736.html',
 280         'md5': 'aeda7243115380b2dd5e881fd42d949a',
 281         'info_dict': {
 282             'id': 'b1255a4a-8e72-4a2f-b9f3-fc1308e00736',
 283             'ext': 'mp4',
 284             'title': 'Blanca - S1E1 - Senza occhi',
 285             'alt_title': 'St 1 Ep 1 - Blanca - Senza occhi',
 286             'description': 'md5:75f95d5c030ec8bac263b1212322e28c',
 287             'thumbnail': r're:^https://www\.raiplay\.it/dl/img/.+\.jpg',
 288             'uploader': 'Rai Premium',
 289             'creator': 'Rai Fiction',
 290             'duration': 6493,
 291             'series': 'Blanca',
 292             'season': 'Season 1',
 293             'episode_number': 1,
 294             'release_year': 2021,
 295             'season_number': 1,
 296             'episode': 'Senza occhi',
 297             'timestamp': 1637318940,
 298             'upload_date': '20211119',
 299             'formats': 'count:7',
 300         },
 301         'params': {'skip_download': True},
 302         'expected_warnings': ['Video not available. Likely due to geo-restriction.'],
 303     }, {
 304         # 1500 quality
 305         'url': 'https://www.raiplay.it/video/2012/09/S1E11---Tutto-cio-che-luccica-0cab3323-732e-45d6-8e86-7704acab6598.html',
 306         'md5': 'a634d20e8ab2d43724c273563f6bf87a',
 307         'info_dict': {
 308             'id': '0cab3323-732e-45d6-8e86-7704acab6598',
 309             'ext': 'mp4',
 310             'title': 'Mia and Me - S1E11 - Tutto ciò che luccica',
 311             'alt_title': 'St 1 Ep 11 - Mia and Me - Tutto ciò che luccica',
 312             'description': 'md5:4969e594184b1920c4c1f2b704da9dea',
 313             'thumbnail': r're:^https?://.*\.jpg$',
 314             'uploader': 'Rai Gulp',
 315             'series': 'Mia and Me',
 316             'season': 'Season 1',
 317             'episode_number': 11,
 318             'release_year': 2015,
 319             'season_number': 1,
 320             'episode': 'Tutto ciò che luccica',
 321             'timestamp': 1348495020,
 322             'upload_date': '20120924',
 323         },
 324     }, {
 325         'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
 326         'only_matching': True,
 327     }, {
 328         # subtitles at 'subtitlesArray' key (see #27698)
 329         'url': 'https://www.raiplay.it/video/2020/12/Report---04-01-2021-2e90f1de-8eee-4de4-ac0e-78d21db5b600.html',
 330         'only_matching': True,
 331     }, {
 332         # DRM protected
 333         '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',
 334         'only_matching': True,
 335     }]
 336
 337     def _real_extract(self, url):
 338         base, video_id = self._match_valid_url(url).groups()
 339
 340         media = self._download_json(
 341             f'{base}.json', video_id, 'Downloading video JSON')
 342
 343         if not self.get_param('allow_unplayable_formats'):
 344             if traverse_obj(media, (('program_info', None), 'rights_management', 'rights', 'drm')):
 345                 self.report_drm(video_id)
 346
 347         video = media['video']
 348         relinker_info = self._extract_relinker_info(video['content_url'], video_id)
 349         date_published = join_nonempty(
 350             media.get('date_published'), media.get('time_published'), delim=' ')
 351         season = media.get('season')
 352         alt_title = join_nonempty(media.get('subtitle'), media.get('toptitle'), delim=' - ')
 353
 354         return {
 355             'id': remove_start(media.get('id'), 'ContentItem-') or video_id,
 356             'display_id': video_id,
 357             'title': media.get('name'),
 358             'alt_title': strip_or_none(alt_title or None),
 359             'description': media.get('description'),
 360             'uploader': strip_or_none(
 361                 traverse_obj(media, ('program_info', 'channel'))
 362                 or media.get('channel') or None),
 363             'creator': strip_or_none(
 364                 traverse_obj(media, ('program_info', 'editor'))
 365                 or media.get('editor') or None),
 366             'duration': parse_duration(video.get('duration')),
 367             'timestamp': unified_timestamp(date_published),
 368             'thumbnails': self._get_thumbnails_list(media.get('images'), url),
 369             'series': traverse_obj(media, ('program_info', 'name')),
 370             'season_number': int_or_none(season),
 371             'season': season if (season and not season.isdigit()) else None,
 372             'episode': media.get('episode_title'),
 373             'episode_number': int_or_none(media.get('episode')),
 374             'subtitles': self._extract_subtitles(url, video),
 375             'release_year': int_or_none(traverse_obj(media, ('track_info', 'edit_year'))),
 376             **relinker_info,
 377         }
 378
 379
 380 class RaiPlayLiveIE(RaiPlayIE):  # XXX: Do not subclass from concrete IE
 381     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/dirette/(?P<id>[^/?#&]+))'
 382     _TESTS = [{
 383         'url': 'http://www.raiplay.it/dirette/rainews24',
 384         'info_dict': {
 385             'id': 'd784ad40-e0ae-4a69-aa76-37519d238a9c',
 386             'display_id': 'rainews24',
 387             'ext': 'mp4',
 388             'title': 're:^Diretta di Rai News 24 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 389             'description': 'md5:4d00bcf6dc98b27c6ec480de329d1497',
 390             'uploader': 'Rai News 24',
 391             'creator': 'Rai News 24',
 392             'is_live': True,
 393             'live_status': 'is_live',
 394             'upload_date': '20090502',
 395             'timestamp': 1241276220,
 396             'formats': 'count:3',
 397         },
 398         'params': {'skip_download': True},
 399     }]
 400
 401
 402 class RaiPlayPlaylistIE(InfoExtractor):
 403     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplay\.it/programmi/(?P<id>[^/?#&]+))(?:/(?P<extra_id>[^?#&]+))?'
 404     _TESTS = [{
 405         # entire series episodes + extras...
 406         'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/',
 407         'info_dict': {
 408             'id': 'nondirloalmiocapo',
 409             'title': 'Non dirlo al mio capo',
 410             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 411         },
 412         'playlist_mincount': 30,
 413     }, {
 414         # single season
 415         'url': 'https://www.raiplay.it/programmi/nondirloalmiocapo/episodi/stagione-2/',
 416         'info_dict': {
 417             'id': 'nondirloalmiocapo',
 418             'title': 'Non dirlo al mio capo - Stagione 2',
 419             'description': 'md5:98ab6b98f7f44c2843fd7d6f045f153b',
 420         },
 421         'playlist_count': 12,
 422     }]
 423
 424     def _real_extract(self, url):
 425         base, playlist_id, extra_id = self._match_valid_url(url).groups()
 426
 427         program = self._download_json(
 428             f'{base}.json', playlist_id, 'Downloading program JSON')
 429
 430         if extra_id:
 431             extra_id = extra_id.upper().rstrip('/')
 432
 433         playlist_title = program.get('name')
 434         entries = []
 435         for b in (program.get('blocks') or []):
 436             for s in (b.get('sets') or []):
 437                 if extra_id:
 438                     if extra_id != join_nonempty(
 439                             b.get('name'), s.get('name'), delim='/').replace(' ', '-').upper():
 440                         continue
 441                     playlist_title = join_nonempty(playlist_title, s.get('name'), delim=' - ')
 442
 443                 s_id = s.get('id')
 444                 if not s_id:
 445                     continue
 446                 medias = self._download_json(
 447                     f'{base}/{s_id}.json', s_id,
 448                     'Downloading content set JSON', fatal=False)
 449                 if not medias:
 450                     continue
 451                 for m in (medias.get('items') or []):
 452                     path_id = m.get('path_id')
 453                     if not path_id:
 454                         continue
 455                     video_url = urljoin(url, path_id)
 456                     entries.append(self.url_result(
 457                         video_url, ie=RaiPlayIE.ie_key(),
 458                         video_id=RaiPlayIE._match_id(video_url)))
 459
 460         return self.playlist_result(
 461             entries, playlist_id, playlist_title,
 462             try_get(program, lambda x: x['program_info']['description']))
 463
 464
 465 class RaiPlaySoundIE(RaiBaseIE):
 466     _VALID_URL = rf'(?P<base>https?://(?:www\.)?raiplaysound\.it/.+?-(?P<id>{RaiBaseIE._UUID_RE}))\.(?:html|json)'
 467     _TESTS = [{
 468         'url': 'https://www.raiplaysound.it/audio/2021/12/IL-RUGGITO-DEL-CONIGLIO-1ebae2a7-7cdb-42bb-842e-fe0d193e9707.html',
 469         'md5': '8970abf8caf8aef4696e7b1f2adfc696',
 470         'info_dict': {
 471             'id': '1ebae2a7-7cdb-42bb-842e-fe0d193e9707',
 472             'ext': 'mp3',
 473             'title': 'Il Ruggito del Coniglio del 10/12/2021',
 474             'alt_title': 'md5:0e6476cd57858bb0f3fcc835d305b455',
 475             'description': 'md5:2a17d2107e59a4a8faa0e18334139ee2',
 476             'thumbnail': r're:^https?://.+\.jpg$',
 477             'uploader': 'rai radio 2',
 478             'duration': 5685,
 479             'series': 'Il Ruggito del Coniglio',
 480             'episode': 'Il Ruggito del Coniglio del 10/12/2021',
 481             'creator': 'rai radio 2',
 482             'timestamp': 1638346620,
 483             'upload_date': '20211201',
 484         },
 485         'params': {'skip_download': True},
 486     }]
 487
 488     def _real_extract(self, url):
 489         base, audio_id = self._match_valid_url(url).group('base', 'id')
 490         media = self._download_json(f'{base}.json', audio_id, 'Downloading audio JSON')
 491         uid = try_get(media, lambda x: remove_start(remove_start(x['uniquename'], 'ContentItem-'), 'Page-'))
 492
 493         info = {}
 494         formats = []
 495         relinkers = set(traverse_obj(media, (('downloadable_audio', 'audio', ('live', 'cards', 0, 'audio')), 'url')))
 496         for r in relinkers:
 497             info = self._extract_relinker_info(r, audio_id, True)
 498             formats.extend(info.get('formats'))
 499
 500         date_published = try_get(media, (lambda x: f'{x["create_date"]} {x.get("create_time") or ""}',
 501                                          lambda x: x['live']['create_date']))
 502
 503         podcast_info = traverse_obj(media, 'podcast_info', ('live', 'cards', 0)) or {}
 504
 505         return {
 506             **info,
 507             'id': uid or audio_id,
 508             'display_id': audio_id,
 509             'title': traverse_obj(media, 'title', 'episode_title'),
 510             'alt_title': traverse_obj(media, ('track_info', 'media_name'), expected_type=strip_or_none),
 511             'description': media.get('description'),
 512             'uploader': traverse_obj(media, ('track_info', 'channel'), expected_type=strip_or_none),
 513             'creator': traverse_obj(media, ('track_info', 'editor'), expected_type=strip_or_none),
 514             'timestamp': unified_timestamp(date_published),
 515             'thumbnails': self._get_thumbnails_list(podcast_info.get('images'), url),
 516             'series': podcast_info.get('title'),
 517             'season_number': int_or_none(media.get('season')),
 518             'episode': media.get('episode_title'),
 519             'episode_number': int_or_none(media.get('episode')),
 520             'formats': formats,
 521         }
 522
 523
 524 class RaiPlaySoundLiveIE(RaiPlaySoundIE):  # XXX: Do not subclass from concrete IE
 525     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplaysound\.it/(?P<id>[^/?#&]+)$)'
 526     _TESTS = [{
 527         'url': 'https://www.raiplaysound.it/radio2',
 528         'info_dict': {
 529             'id': 'b00a50e6-f404-4af6-8f8c-ff3b9af73a44',
 530             'display_id': 'radio2',
 531             'ext': 'mp4',
 532             'title': r're:Rai Radio 2 \d+-\d+-\d+ \d+:\d+',
 533             'thumbnail': r're:^https://www\.raiplaysound\.it/dl/img/.+\.png',
 534             'uploader': 'rai radio 2',
 535             'series': 'Rai Radio 2',
 536             'creator': 'raiplaysound',
 537             'is_live': True,
 538             'live_status': 'is_live',
 539         },
 540         'params': {'skip_download': True},
 541     }]
 542
 543
 544 class RaiPlaySoundPlaylistIE(InfoExtractor):
 545     _VALID_URL = r'(?P<base>https?://(?:www\.)?raiplaysound\.it/(?:programmi|playlist|audiolibri)/(?P<id>[^/?#&]+))(?:/(?P<extra_id>[^?#&]+))?'
 546     _TESTS = [{
 547         # entire show
 548         'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio',
 549         'info_dict': {
 550             'id': 'ilruggitodelconiglio',
 551             'title': 'Il Ruggito del Coniglio',
 552             'description': 'md5:62a627b3a2d0635d08fa8b6e0a04f27e',
 553         },
 554         'playlist_mincount': 65,
 555     }, {
 556         # single season
 557         'url': 'https://www.raiplaysound.it/programmi/ilruggitodelconiglio/puntate/prima-stagione-1995',
 558         'info_dict': {
 559             'id': 'ilruggitodelconiglio_puntate_prima-stagione-1995',
 560             'title': 'Prima Stagione 1995',
 561         },
 562         'playlist_count': 1,
 563     }]
 564
 565     def _real_extract(self, url):
 566         base, playlist_id, extra_id = self._match_valid_url(url).group('base', 'id', 'extra_id')
 567         url = f'{base}.json'
 568         program = self._download_json(url, playlist_id, 'Downloading program JSON')
 569
 570         if extra_id:
 571             extra_id = extra_id.rstrip('/')
 572             playlist_id += '_' + extra_id.replace('/', '_')
 573             path = next(c['path_id'] for c in program.get('filters') or [] if extra_id in c.get('weblink'))
 574             program = self._download_json(
 575                 urljoin('https://www.raiplaysound.it', path), playlist_id, 'Downloading program secondary JSON')
 576
 577         entries = [
 578             self.url_result(urljoin(base, c['path_id']), ie=RaiPlaySoundIE.ie_key())
 579             for c in traverse_obj(program, 'cards', ('block', 'cards')) or []
 580             if c.get('path_id')]
 581
 582         return self.playlist_result(entries, playlist_id, program.get('title'),
 583                                     traverse_obj(program, ('podcast_info', 'description')))
 584
 585
 586 class RaiIE(RaiBaseIE):
 587     _VALID_URL = rf'https?://[^/]+\.(?:rai\.(?:it|tv))/.+?-(?P<id>{RaiBaseIE._UUID_RE})(?:-.+?)?\.html'
 588     _TESTS = [{
 589         'url': 'https://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
 590         'info_dict': {
 591             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
 592             'ext': 'mp4',
 593             'title': 'TG PRIMO TEMPO',
 594             'thumbnail': r're:^https?://.*\.jpg',
 595             'duration': 1758,
 596             'upload_date': '20140612',
 597         },
 598         'params': {'skip_download': True},
 599         'expected_warnings': ['Video not available. Likely due to geo-restriction.'],
 600     }, {
 601         'url': 'https://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
 602         'info_dict': {
 603             'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
 604             'ext': 'mp4',
 605             'title': 'TG1 ore 20:00 del 03/11/2016',
 606             'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
 607             'thumbnail': r're:^https?://.*\.jpg$',
 608             'duration': 2214,
 609             'upload_date': '20161103',
 610         },
 611         'params': {'skip_download': True},
 612     }, {
 613         # Direct MMS: Media URL no longer works.
 614         'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-b63a4089-ac28-48cf-bca5-9f5b5bc46df5.html',
 615         'only_matching': True,
 616     }]
 617
 618     def _real_extract(self, url):
 619         content_id = self._match_id(url)
 620         media = self._download_json(
 621             f'https://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-{content_id}.html?json',
 622             content_id, 'Downloading video JSON', fatal=False, expected_status=404)
 623
 624         if media is None:
 625             return None
 626
 627         if 'Audio' in media['type']:
 628             relinker_info = {
 629                 'formats': [{
 630                     'format_id': join_nonempty('https', media.get('formatoAudio'), delim='-'),
 631                     'url': media['audioUrl'],
 632                     'ext': media.get('formatoAudio'),
 633                     'vcodec': 'none',
 634                     'acodec': media.get('formatoAudio'),
 635                 }],
 636             }
 637         elif 'Video' in media['type']:
 638             relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
 639         else:
 640             raise ExtractorError('not a media file')
 641
 642         thumbnails = self._get_thumbnails_list(
 643             {image_type: media.get(image_type) for image_type in (
 644                 'image', 'image_medium', 'image_300')}, url)
 645
 646         return {
 647             'id': content_id,
 648             'title': strip_or_none(media.get('name') or media.get('title')),
 649             'description': strip_or_none(media.get('desc')) or None,
 650             'thumbnails': thumbnails,
 651             'uploader': strip_or_none(media.get('author')) or None,
 652             'upload_date': unified_strdate(media.get('date')),
 653             'duration': parse_duration(media.get('length')),
 654             'subtitles': self._extract_subtitles(url, media),
 655             **relinker_info,
 656         }
 657
 658
 659 class RaiNewsIE(RaiBaseIE):
 660     _VALID_URL = rf'https?://(www\.)?rainews\.it/(?!articoli)[^?#]+-(?P<id>{RaiBaseIE._UUID_RE})(?:-[^/?#]+)?\.html'
 661     _EMBED_REGEX = [rf'<iframe[^>]+data-src="(?P<url>/iframe/[^?#]+?{RaiBaseIE._UUID_RE}\.html)']
 662     _TESTS = [{
 663         # new rainews player (#3911)
 664         'url': 'https://www.rainews.it/video/2024/02/membri-della-croce-rossa-evacuano-gli-abitanti-di-un-villaggio-nella-regione-ucraina-di-kharkiv-il-filmato-dallucraina--31e8017c-845c-43f5-9c48-245b43c3a079.html',
 665         'info_dict': {
 666             'id': '31e8017c-845c-43f5-9c48-245b43c3a079',
 667             'ext': 'mp4',
 668             'title': 'md5:1e81364b09de4a149042bac3c7d36f0b',
 669             'duration': 196,
 670             'upload_date': '20240225',
 671             'uploader': 'rainews',
 672             'formats': 'count:2',
 673         },
 674         'params': {'skip_download': True},
 675     }, {
 676         # old content with fallback method to extract media urls
 677         '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',
 678         'info_dict': {
 679             'id': '1632c009-c843-4836-bb65-80c33084a64b',
 680             'ext': 'mp4',
 681             'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
 682             'description': 'I film in uscita questa settimana.',
 683             'thumbnail': r're:^https?://.*\.png$',
 684             'duration': 833,
 685             'upload_date': '20161103',
 686             'formats': 'count:8',
 687         },
 688         'params': {'skip_download': True},
 689         'expected_warnings': ['unable to extract player_data'],
 690     }, {
 691         # iframe + drm
 692         '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',
 693         'only_matching': True,
 694     }]
 695     _PLAYER_TAG = 'news'
 696
 697     def _real_extract(self, url):
 698         video_id = self._match_id(url)
 699
 700         webpage = self._download_webpage(url, video_id)
 701
 702         player_data = self._search_json(
 703             rf'<rai{self._PLAYER_TAG}-player\s*data=\'', webpage, 'player_data', video_id,
 704             transform_source=clean_html, default={})
 705         track_info = player_data.get('track_info')
 706         relinker_url = traverse_obj(player_data, 'mediapolis', 'content_url')
 707
 708         if not relinker_url:
 709             # fallback on old implementation for some old content
 710             try:
 711                 return RaiIE._real_extract(self, url)
 712             except GeoRestrictedError:
 713                 raise
 714             except ExtractorError as e:
 715                 raise ExtractorError('Relinker URL not found', cause=e)
 716
 717         relinker_info = self._extract_relinker_info(urljoin(url, relinker_url), video_id)
 718
 719         return {
 720             'id': video_id,
 721             'title': player_data.get('title') or track_info.get('title') or self._og_search_title(webpage),
 722             'upload_date': unified_strdate(track_info.get('date')),
 723             'uploader': strip_or_none(track_info.get('editor') or None),
 724             **relinker_info,
 725         }
 726
 727
 728 class RaiCulturaIE(RaiNewsIE):  # XXX: Do not subclass from concrete IE
 729     _VALID_URL = rf'https?://(www\.)?raicultura\.it/(?!articoli)[^?#]+-(?P<id>{RaiBaseIE._UUID_RE})(?:-[^/?#]+)?\.html'
 730     _EMBED_REGEX = [rf'<iframe[^>]+data-src="(?P<url>/iframe/[^?#]+?{RaiBaseIE._UUID_RE}\.html)']
 731     _TESTS = [{
 732         'url': 'https://www.raicultura.it/letteratura/articoli/2018/12/Alberto-Asor-Rosa-Letteratura-e-potere-05ba8775-82b5-45c5-a89d-dd955fbde1fb.html',
 733         'info_dict': {
 734             'id': '05ba8775-82b5-45c5-a89d-dd955fbde1fb',
 735             'ext': 'mp4',
 736             'title': 'Alberto Asor Rosa: Letteratura e potere',
 737             'duration': 1756,
 738             'upload_date': '20181206',
 739             'uploader': 'raicultura',
 740             'formats': 'count:2',
 741         },
 742         'params': {'skip_download': True},
 743     }]
 744     _PLAYER_TAG = 'cultura'
 745
 746
 747 class RaiSudtirolIE(RaiBaseIE):
 748     _VALID_URL = r'https?://raisudtirol\.rai\.it/.+media=(?P<id>\w+)'
 749     _TESTS = [{
 750         # mp4 file
 751         'url': 'https://raisudtirol.rai.it/la/index.php?media=Ptv1619729460',
 752         'info_dict': {
 753             'id': 'Ptv1619729460',
 754             'ext': 'mp4',
 755             'title': 'Euro: trasmisciun d\'economia - 29-04-2021 20:51',
 756             'series': 'Euro: trasmisciun d\'economia',
 757             'upload_date': '20210429',
 758             'thumbnail': r're:https://raisudtirol\.rai\.it/img/.+\.jpg',
 759             'uploader': 'raisudtirol',
 760             'formats': 'count:1',
 761         },
 762         'params': {'skip_download': True},
 763     }, {
 764         # m3u manifest
 765         'url': 'https://raisudtirol.rai.it/it/kidsplayer.php?lang=it&media=GUGGUG_P1.smil',
 766         'info_dict': {
 767             'id': 'GUGGUG_P1',
 768             'ext': 'mp4',
 769             'title': 'GUGGUG! La Prospettiva - Die Perspektive',
 770             'uploader': 'raisudtirol',
 771             'formats': 'count:6',
 772         },
 773         'params': {'skip_download': True},
 774     }]
 775
 776     def _real_extract(self, url):
 777         video_id = self._match_id(url)
 778         webpage = self._download_webpage(url, video_id)
 779
 780         video_date = self._html_search_regex(
 781             r'<span class="med_data">(.+?)</span>', webpage, 'video_date', default=None)
 782         video_title = self._html_search_regex([
 783             r'<span class="med_title">(.+?)</span>', r'title: \'(.+?)\','],
 784             webpage, 'video_title', default=None)
 785         video_url = self._html_search_regex([
 786             r'sources:\s*\[\{file:\s*"(.+?)"\}\]',
 787             r'<source\s+src="(.+?)"\s+type="application/x-mpegURL"'],
 788             webpage, 'video_url', default=None)
 789
 790         ext = determine_ext(video_url)
 791         if ext == 'm3u8':
 792             formats = self._extract_m3u8_formats(video_url, video_id)
 793         elif ext == 'mp4':
 794             formats = [{
 795                 'format_id': 'https-mp4',
 796                 'url': self._proto_relative_url(video_url),
 797                 'width': 1024,
 798                 'height': 576,
 799                 'fps': 25,
 800                 'vcodec': 'avc1',
 801                 'acodec': 'mp4a',
 802             }]
 803         else:
 804             formats = []
 805             self.raise_no_formats(f'Unrecognized media file: {video_url}')
 806
 807         return {
 808             'id': video_id,
 809             'title': join_nonempty(video_title, video_date, delim=' - '),
 810             'series': video_title if video_date else None,
 811             'upload_date': unified_strdate(video_date),
 812             'thumbnail': urljoin('https://raisudtirol.rai.it/', self._html_search_regex(
 813                 r'image: \'(.+?)\'', webpage, 'video_thumb', default=None)),
 814             'uploader': 'raisudtirol',
 815             'formats': formats,
 816         }