yt_dlp/extractor/orf.py

   1 import base64
   2 import functools
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..networking import HEADRequest
   7 from ..utils import (
   8     InAdvancePagedList,
   9     clean_html,
  10     determine_ext,
  11     float_or_none,
  12     int_or_none,
  13     join_nonempty,
  14     make_archive_id,
  15     mimetype2ext,
  16     orderedSet,
  17     parse_age_limit,
  18     remove_end,
  19     smuggle_url,
  20     strip_jsonp,
  21     try_call,
  22     unescapeHTML,
  23     unified_strdate,
  24     unsmuggle_url,
  25     url_or_none,
  26 )
  27 from ..utils.traversal import traverse_obj
  28
  29
  30 class ORFTVthekIE(InfoExtractor):
  31     IE_NAME = 'orf:tvthek'
  32     IE_DESC = 'ORF TVthek'
  33     _VALID_URL = r'(?P<url>https?://tvthek\.orf\.at/(?:(?:[^/]+/){2}){1,2}(?P<id>\d+))(/[^/]+/(?P<vid>\d+))?(?:$|[?#])'
  34
  35     _TESTS = [{
  36         'url': 'https://tvthek.orf.at/profile/ZIB-2/1211/ZIB-2/14121079',
  37         'info_dict': {
  38             'id': '14121079',
  39         },
  40         'playlist_count': 11,
  41         'params': {'noplaylist': True}
  42     }, {
  43         'url': 'https://tvthek.orf.at/profile/ZIB-2/1211/ZIB-2/14121079/Umfrage-Welches-Tier-ist-Sebastian-Kurz/15083150',
  44         'info_dict': {
  45             'id': '14121079',
  46         },
  47         'playlist_count': 1,
  48         'params': {'playlist_items': '5'}
  49     }, {
  50         'url': 'https://tvthek.orf.at/profile/ZIB-2/1211/ZIB-2/14121079/Umfrage-Welches-Tier-ist-Sebastian-Kurz/15083150',
  51         'info_dict': {
  52             'id': '14121079',
  53         },
  54         'playlist': [{
  55             'info_dict': {
  56                 'id': '15083150',
  57                 'ext': 'mp4',
  58                 'description': 'md5:7be1c485425f5f255a5e4e4815e77d04',
  59                 'thumbnail': 'https://api-tvthek.orf.at/uploads/media/segments/0130/59/824271ea35cd8931a0fb08ab316a5b0a1562342c.jpeg',
  60                 'title': 'Umfrage: Welches Tier ist Sebastian Kurz?',
  61             }
  62         }],
  63         'playlist_count': 1,
  64         'params': {'noplaylist': True, 'skip_download': 'm3u8'}
  65     }, {
  66         'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
  67         'playlist': [{
  68             'md5': '2942210346ed779588f428a92db88712',
  69             'info_dict': {
  70                 'id': '8896777',
  71                 'ext': 'mp4',
  72                 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
  73                 'description': 'md5:c1272f0245537812d4e36419c207b67d',
  74                 'duration': 2668,
  75                 'upload_date': '20141208',
  76             },
  77         }],
  78         'skip': 'Blocked outside of Austria / Germany',
  79     }, {
  80         'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
  81         'info_dict': {
  82             'id': '7982259',
  83             'ext': 'mp4',
  84             'title': 'Best of Ingrid Thurnher',
  85             'upload_date': '20140527',
  86             'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im Jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".',
  87         },
  88         'params': {
  89             'skip_download': True,  # rtsp downloads
  90         },
  91         'skip': 'Blocked outside of Austria / Germany',
  92     }, {
  93         'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
  94         'only_matching': True,
  95     }, {
  96         'url': 'http://tvthek.orf.at/profile/Universum/35429',
  97         'only_matching': True,
  98     }]
  99
 100     def _pagefunc(self, url, data_jsb, n, *, image=None):
 101         sd = data_jsb[n]
 102         video_id, title = str(sd['id']), sd['title']
 103         formats = []
 104         for fd in sd['sources']:
 105             src = url_or_none(fd.get('src'))
 106             if not src:
 107                 continue
 108             format_id = join_nonempty('delivery', 'quality', 'quality_string', from_dict=fd)
 109             ext = determine_ext(src)
 110             if ext == 'm3u8':
 111                 m3u8_formats = self._extract_m3u8_formats(
 112                     src, video_id, 'mp4', m3u8_id=format_id, fatal=False, note=f'Downloading {format_id} m3u8 manifest')
 113                 if any('/geoprotection' in f['url'] for f in m3u8_formats):
 114                     self.raise_geo_restricted()
 115                 formats.extend(m3u8_formats)
 116             elif ext == 'f4m':
 117                 formats.extend(self._extract_f4m_formats(
 118                     src, video_id, f4m_id=format_id, fatal=False))
 119             elif ext == 'mpd':
 120                 formats.extend(self._extract_mpd_formats(
 121                     src, video_id, mpd_id=format_id, fatal=False, note=f'Downloading {format_id} mpd manifest'))
 122             else:
 123                 formats.append({
 124                     'format_id': format_id,
 125                     'url': src,
 126                     'protocol': fd.get('protocol'),
 127                 })
 128
 129         # Check for geoblocking.
 130         # There is a property is_geoprotection, but that's always false
 131         geo_str = sd.get('geoprotection_string')
 132         http_url = next(
 133             (f['url'] for f in formats if re.match(r'^https?://.*\.mp4$', f['url'])),
 134             None) if geo_str else None
 135         if http_url:
 136             self._request_webpage(
 137                 HEADRequest(http_url), video_id, fatal=False, note='Testing for geoblocking',
 138                 errnote=f'This video seems to be blocked outside of {geo_str}. You may want to try the streaming-* formats')
 139
 140         subtitles = {}
 141         for sub in sd.get('subtitles', []):
 142             sub_src = sub.get('src')
 143             if not sub_src:
 144                 continue
 145             subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
 146                 'url': sub_src,
 147             })
 148
 149         upload_date = unified_strdate(sd.get('created_date'))
 150
 151         thumbnails = []
 152         preview = sd.get('preview_image_url')
 153         if preview:
 154             thumbnails.append({
 155                 'id': 'preview',
 156                 'url': preview,
 157                 'preference': 0,
 158             })
 159         image = sd.get('image_full_url') or image
 160         if image:
 161             thumbnails.append({
 162                 'id': 'full',
 163                 'url': image,
 164                 'preference': 1,
 165             })
 166
 167         yield {
 168             'id': video_id,
 169             'title': title,
 170             'webpage_url': smuggle_url(f'{url}/part/{video_id}', {'force_noplaylist': True}),
 171             'formats': formats,
 172             'subtitles': subtitles,
 173             'description': sd.get('description'),
 174             'duration': int_or_none(sd.get('duration_in_seconds')),
 175             'upload_date': upload_date,
 176             'thumbnails': thumbnails,
 177         }
 178
 179     def _real_extract(self, url):
 180         url, smuggled_data = unsmuggle_url(url)
 181         playlist_id, video_id, base_url = self._match_valid_url(url).group('id', 'vid', 'url')
 182         webpage = self._download_webpage(url, playlist_id)
 183
 184         data_jsb = self._parse_json(
 185             self._search_regex(
 186                 r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
 187                 webpage, 'playlist', group='json'),
 188             playlist_id, transform_source=unescapeHTML)['playlist']['videos']
 189
 190         if not self._yes_playlist(playlist_id, video_id, smuggled_data):
 191             data_jsb = [sd for sd in data_jsb if str(sd.get('id')) == video_id]
 192
 193         playlist_count = len(data_jsb)
 194         image = self._og_search_thumbnail(webpage) if playlist_count == 1 else None
 195
 196         page_func = functools.partial(self._pagefunc, base_url, data_jsb, image=image)
 197         return {
 198             '_type': 'playlist',
 199             'entries': InAdvancePagedList(page_func, playlist_count, 1),
 200             'id': playlist_id,
 201         }
 202
 203
 204 class ORFRadioIE(InfoExtractor):
 205     IE_NAME = 'orf:radio'
 206
 207     STATION_INFO = {
 208         'fm4': ('fm4', 'fm4', 'orffm4'),
 209         'noe': ('noe', 'oe2n', 'orfnoe'),
 210         'wien': ('wie', 'oe2w', 'orfwie'),
 211         'burgenland': ('bgl', 'oe2b', 'orfbgl'),
 212         'ooe': ('ooe', 'oe2o', 'orfooe'),
 213         'steiermark': ('stm', 'oe2st', 'orfstm'),
 214         'kaernten': ('ktn', 'oe2k', 'orfktn'),
 215         'salzburg': ('sbg', 'oe2s', 'orfsbg'),
 216         'tirol': ('tir', 'oe2t', 'orftir'),
 217         'vorarlberg': ('vbg', 'oe2v', 'orfvbg'),
 218         'oe3': ('oe3', 'oe3', 'orfoe3'),
 219         'oe1': ('oe1', 'oe1', 'orfoe1'),
 220     }
 221     _STATION_RE = '|'.join(map(re.escape, STATION_INFO.keys()))
 222
 223     _VALID_URL = rf'''(?x)
 224         https?://(?:
 225             (?P<station>{_STATION_RE})\.orf\.at/player|
 226             radiothek\.orf\.at/(?P<station2>{_STATION_RE})
 227         )/(?P<date>[0-9]+)/(?P<show>\w+)'''
 228
 229     _TESTS = [{
 230         'url': 'https://radiothek.orf.at/ooe/20220801/OGMO',
 231         'info_dict': {
 232             'id': 'OGMO',
 233             'title': 'Guten Morgen OÖ',
 234             'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 235         },
 236         'playlist': [{
 237             'md5': 'f33147d954a326e338ea52572c2810e8',
 238             'info_dict': {
 239                 'id': '2022-08-01_0459_tl_66_7DaysMon1_319062',
 240                 'ext': 'mp3',
 241                 'title': 'Guten Morgen OÖ',
 242                 'upload_date': '20220801',
 243                 'duration': 18000,
 244                 'timestamp': 1659322789,
 245                 'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 246             }
 247         }]
 248     }, {
 249         'url': 'https://ooe.orf.at/player/20220801/OGMO',
 250         'info_dict': {
 251             'id': 'OGMO',
 252             'title': 'Guten Morgen OÖ',
 253             'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 254         },
 255         'playlist': [{
 256             'md5': 'f33147d954a326e338ea52572c2810e8',
 257             'info_dict': {
 258                 'id': '2022-08-01_0459_tl_66_7DaysMon1_319062',
 259                 'ext': 'mp3',
 260                 'title': 'Guten Morgen OÖ',
 261                 'upload_date': '20220801',
 262                 'duration': 18000,
 263                 'timestamp': 1659322789,
 264                 'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 265             }
 266         }]
 267     }, {
 268         'url': 'http://fm4.orf.at/player/20170107/4CC',
 269         'only_matching': True,
 270     }, {
 271         'url': 'https://noe.orf.at/player/20200423/NGM',
 272         'only_matching': True,
 273     }, {
 274         'url': 'https://wien.orf.at/player/20200423/WGUM',
 275         'only_matching': True,
 276     }, {
 277         'url': 'https://burgenland.orf.at/player/20200423/BGM',
 278         'only_matching': True,
 279     }, {
 280         'url': 'https://steiermark.orf.at/player/20200423/STGMS',
 281         'only_matching': True,
 282     }, {
 283         'url': 'https://kaernten.orf.at/player/20200423/KGUMO',
 284         'only_matching': True,
 285     }, {
 286         'url': 'https://salzburg.orf.at/player/20200423/SGUM',
 287         'only_matching': True,
 288     }, {
 289         'url': 'https://tirol.orf.at/player/20200423/TGUMO',
 290         'only_matching': True,
 291     }, {
 292         'url': 'https://vorarlberg.orf.at/player/20200423/VGUM',
 293         'only_matching': True,
 294     }, {
 295         'url': 'https://oe3.orf.at/player/20200424/3WEK',
 296         'only_matching': True,
 297     }, {
 298         'url': 'http://oe1.orf.at/player/20170108/456544',
 299         'md5': '34d8a6e67ea888293741c86a099b745b',
 300         'info_dict': {
 301             'id': '2017-01-08_0759_tl_51_7DaysSun6_256141',
 302             'ext': 'mp3',
 303             'title': 'Morgenjournal',
 304             'duration': 609,
 305             'timestamp': 1483858796,
 306             'upload_date': '20170108',
 307         },
 308         'skip': 'Shows from ORF radios are only available for 7 days.'
 309     }]
 310
 311     def _entries(self, data, station):
 312         _, loop_station, old_ie = self.STATION_INFO[station]
 313         for info in data['streams']:
 314             item_id = info.get('loopStreamId')
 315             if not item_id:
 316                 continue
 317             video_id = item_id.replace('.mp3', '')
 318             yield {
 319                 'id': video_id,
 320                 'ext': 'mp3',
 321                 'url': f'https://loopstream01.apa.at/?channel={loop_station}&id={item_id}',
 322                 '_old_archive_ids': [make_archive_id(old_ie, video_id)],
 323                 'title': data.get('title'),
 324                 'description': clean_html(data.get('subtitle')),
 325                 'duration': try_call(lambda: (info['end'] - info['start']) / 1000),
 326                 'timestamp': int_or_none(info.get('start'), scale=1000),
 327                 'series': data.get('programTitle'),
 328             }
 329
 330     def _real_extract(self, url):
 331         station, station2, show_date, show_id = self._match_valid_url(url).group('station', 'station2', 'date', 'show')
 332         api_station, _, _ = self.STATION_INFO[station or station2]
 333         data = self._download_json(
 334             f'http://audioapi.orf.at/{api_station}/api/json/current/broadcast/{show_id}/{show_date}', show_id)
 335
 336         return self.playlist_result(
 337             self._entries(data, station or station2), show_id, data.get('title'), clean_html(data.get('subtitle')))
 338
 339
 340 class ORFPodcastIE(InfoExtractor):
 341     IE_NAME = 'orf:podcast'
 342     _STATION_RE = '|'.join(map(re.escape, (
 343         'bgl', 'fm4', 'ktn', 'noe', 'oe1', 'oe3',
 344         'ooe', 'sbg', 'stm', 'tir', 'tv', 'vbg', 'wie')))
 345     _VALID_URL = rf'https?://sound\.orf\.at/podcast/(?P<station>{_STATION_RE})/(?P<show>[\w-]+)/(?P<id>[\w-]+)'
 346     _TESTS = [{
 347         'url': 'https://sound.orf.at/podcast/oe3/fruehstueck-bei-mir/nicolas-stockhammer-15102023',
 348         'md5': '526a5700e03d271a1505386a8721ab9b',
 349         'info_dict': {
 350             'id': 'nicolas-stockhammer-15102023',
 351             'ext': 'mp3',
 352             'title': 'Nicolas Stockhammer (15.10.2023)',
 353             'duration': 3396.0,
 354             'series': 'Frühstück bei mir',
 355         },
 356         'skip': 'ORF podcasts are only available for a limited time'
 357     }]
 358
 359     def _real_extract(self, url):
 360         station, show, show_id = self._match_valid_url(url).group('station', 'show', 'id')
 361         data = self._download_json(
 362             f'https://audioapi.orf.at/radiothek/api/2.0/podcast/{station}/{show}/{show_id}', show_id)
 363
 364         return {
 365             'id': show_id,
 366             'ext': 'mp3',
 367             'vcodec': 'none',
 368             **traverse_obj(data, ('payload', {
 369                 'url': ('enclosures', 0, 'url'),
 370                 'ext': ('enclosures', 0, 'type', {mimetype2ext}),
 371                 'title': 'title',
 372                 'description': ('description', {clean_html}),
 373                 'duration': ('duration', {functools.partial(float_or_none, scale=1000)}),
 374                 'series': ('podcast', 'title'),
 375             })),
 376         }
 377
 378
 379 class ORFIPTVIE(InfoExtractor):
 380     IE_NAME = 'orf:iptv'
 381     IE_DESC = 'iptv.ORF.at'
 382     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 383
 384     _TEST = {
 385         'url': 'http://iptv.orf.at/stories/2275236/',
 386         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 387         'info_dict': {
 388             'id': '350612',
 389             'ext': 'flv',
 390             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 391             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 392             'duration': 68.197,
 393             'thumbnail': r're:^https?://.*\.jpg$',
 394             'upload_date': '20150425',
 395         },
 396     }
 397
 398     def _real_extract(self, url):
 399         story_id = self._match_id(url)
 400
 401         webpage = self._download_webpage(
 402             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 403
 404         video_id = self._search_regex(
 405             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 406
 407         data = self._download_json(
 408             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 409             video_id)[0]
 410
 411         duration = float_or_none(data['duration'], 1000)
 412
 413         video = data['sources']['default']
 414         load_balancer_url = video['loadBalancerUrl']
 415         abr = int_or_none(video.get('audioBitrate'))
 416         vbr = int_or_none(video.get('bitrate'))
 417         fps = int_or_none(video.get('videoFps'))
 418         width = int_or_none(video.get('videoWidth'))
 419         height = int_or_none(video.get('videoHeight'))
 420         thumbnail = video.get('preview')
 421
 422         rendition = self._download_json(
 423             load_balancer_url, video_id, transform_source=strip_jsonp)
 424
 425         f = {
 426             'abr': abr,
 427             'vbr': vbr,
 428             'fps': fps,
 429             'width': width,
 430             'height': height,
 431         }
 432
 433         formats = []
 434         for format_id, format_url in rendition['redirect'].items():
 435             if format_id == 'rtmp':
 436                 ff = f.copy()
 437                 ff.update({
 438                     'url': format_url,
 439                     'format_id': format_id,
 440                 })
 441                 formats.append(ff)
 442             elif determine_ext(format_url) == 'f4m':
 443                 formats.extend(self._extract_f4m_formats(
 444                     format_url, video_id, f4m_id=format_id))
 445             elif determine_ext(format_url) == 'm3u8':
 446                 formats.extend(self._extract_m3u8_formats(
 447                     format_url, video_id, 'mp4', m3u8_id=format_id))
 448             else:
 449                 continue
 450
 451         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 452         description = self._og_search_description(webpage)
 453         upload_date = unified_strdate(self._html_search_meta(
 454             'dc.date', webpage, 'upload date'))
 455
 456         return {
 457             'id': video_id,
 458             'title': title,
 459             'description': description,
 460             'duration': duration,
 461             'thumbnail': thumbnail,
 462             'upload_date': upload_date,
 463             'formats': formats,
 464         }
 465
 466
 467 class ORFFM4StoryIE(InfoExtractor):
 468     IE_NAME = 'orf:fm4:story'
 469     IE_DESC = 'fm4.orf.at stories'
 470     _VALID_URL = r'https?://fm4\.orf\.at/stories/(?P<id>\d+)'
 471
 472     _TEST = {
 473         'url': 'http://fm4.orf.at/stories/2865738/',
 474         'playlist': [{
 475             'md5': 'e1c2c706c45c7b34cf478bbf409907ca',
 476             'info_dict': {
 477                 'id': '547792',
 478                 'ext': 'flv',
 479                 'title': 'Manu Delago und Inner Tongue live',
 480                 'description': 'Manu Delago und Inner Tongue haben bei der FM4 Soundpark Session live alles gegeben. Hier gibt es Fotos und die gesamte Session als Video.',
 481                 'duration': 1748.52,
 482                 'thumbnail': r're:^https?://.*\.jpg$',
 483                 'upload_date': '20170913',
 484             },
 485         }, {
 486             'md5': 'c6dd2179731f86f4f55a7b49899d515f',
 487             'info_dict': {
 488                 'id': '547798',
 489                 'ext': 'flv',
 490                 'title': 'Manu Delago und Inner Tongue live (2)',
 491                 'duration': 1504.08,
 492                 'thumbnail': r're:^https?://.*\.jpg$',
 493                 'upload_date': '20170913',
 494                 'description': 'Manu Delago und Inner Tongue haben bei der FM4 Soundpark Session live alles gegeben. Hier gibt es Fotos und die gesamte Session als Video.',
 495             },
 496         }],
 497     }
 498
 499     def _real_extract(self, url):
 500         story_id = self._match_id(url)
 501         webpage = self._download_webpage(url, story_id)
 502
 503         entries = []
 504         all_ids = orderedSet(re.findall(r'data-video(?:id)?="(\d+)"', webpage))
 505         for idx, video_id in enumerate(all_ids):
 506             data = self._download_json(
 507                 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 508                 video_id)[0]
 509
 510             duration = float_or_none(data['duration'], 1000)
 511
 512             video = data['sources']['q8c']
 513             load_balancer_url = video['loadBalancerUrl']
 514             abr = int_or_none(video.get('audioBitrate'))
 515             vbr = int_or_none(video.get('bitrate'))
 516             fps = int_or_none(video.get('videoFps'))
 517             width = int_or_none(video.get('videoWidth'))
 518             height = int_or_none(video.get('videoHeight'))
 519             thumbnail = video.get('preview')
 520
 521             rendition = self._download_json(
 522                 load_balancer_url, video_id, transform_source=strip_jsonp)
 523
 524             f = {
 525                 'abr': abr,
 526                 'vbr': vbr,
 527                 'fps': fps,
 528                 'width': width,
 529                 'height': height,
 530             }
 531
 532             formats = []
 533             for format_id, format_url in rendition['redirect'].items():
 534                 if format_id == 'rtmp':
 535                     ff = f.copy()
 536                     ff.update({
 537                         'url': format_url,
 538                         'format_id': format_id,
 539                     })
 540                     formats.append(ff)
 541                 elif determine_ext(format_url) == 'f4m':
 542                     formats.extend(self._extract_f4m_formats(
 543                         format_url, video_id, f4m_id=format_id))
 544                 elif determine_ext(format_url) == 'm3u8':
 545                     formats.extend(self._extract_m3u8_formats(
 546                         format_url, video_id, 'mp4', m3u8_id=format_id))
 547                 else:
 548                     continue
 549
 550             title = remove_end(self._og_search_title(webpage), ' - fm4.ORF.at')
 551             if idx >= 1:
 552                 # Titles are duplicates, make them unique
 553                 title += ' (' + str(idx + 1) + ')'
 554             description = self._og_search_description(webpage)
 555             upload_date = unified_strdate(self._html_search_meta(
 556                 'dc.date', webpage, 'upload date'))
 557
 558             entries.append({
 559                 'id': video_id,
 560                 'title': title,
 561                 'description': description,
 562                 'duration': duration,
 563                 'thumbnail': thumbnail,
 564                 'upload_date': upload_date,
 565                 'formats': formats,
 566             })
 567
 568         return self.playlist_result(entries)
 569
 570
 571 class ORFONIE(InfoExtractor):
 572     IE_NAME = 'orf:on'
 573     _VALID_URL = r'https?://on\.orf\.at/video/(?P<id>\d+)'
 574     _TESTS = [{
 575         'url': 'https://on.orf.at/video/14210000/school-of-champions-48',
 576         'info_dict': {
 577             'id': '14210000',
 578             'ext': 'mp4',
 579             'duration': 2651.08,
 580             'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0167/98/thumb_16697671_segments_highlight_teaser.jpeg',
 581             'title': 'School of Champions (4/8)',
 582             'description': 'md5:d09ad279fc2e8502611e7648484b6afd',
 583             'media_type': 'episode',
 584             'timestamp': 1706472362,
 585             'upload_date': '20240128',
 586         }
 587     }, {
 588         'url': 'https://on.orf.at/video/3220355',
 589         'md5': 'f94d98e667cf9a3851317efb4e136662',
 590         'info_dict': {
 591             'id': '3220355',
 592             'ext': 'mp4',
 593             'duration': 445.04,
 594             'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0002/60/thumb_159573_segments_highlight_teaser.png',
 595             'title': '50 Jahre Burgenland: Der Festumzug',
 596             'description': 'md5:1560bf855119544ee8c4fa5376a2a6b0',
 597             'media_type': 'episode',
 598             'timestamp': 52916400,
 599             'upload_date': '19710905',
 600         }
 601     }]
 602
 603     def _extract_video(self, video_id):
 604         encrypted_id = base64.b64encode(f'3dSlfek03nsLKdj4Jsd{video_id}'.encode()).decode()
 605         api_json = self._download_json(
 606             f'https://api-tvthek.orf.at/api/v4.3/public/episode/encrypted/{encrypted_id}', video_id)
 607
 608         if traverse_obj(api_json, 'is_drm_protected'):
 609             self.report_drm(video_id)
 610
 611         formats, subtitles = [], {}
 612         for manifest_type in traverse_obj(api_json, ('sources', {dict.keys}, ...)):
 613             for manifest_url in traverse_obj(api_json, ('sources', manifest_type, ..., 'src', {url_or_none})):
 614                 if manifest_type == 'hls':
 615                     fmts, subs = self._extract_m3u8_formats_and_subtitles(
 616                         manifest_url, video_id, fatal=False, m3u8_id='hls')
 617                 elif manifest_type == 'dash':
 618                     fmts, subs = self._extract_mpd_formats_and_subtitles(
 619                         manifest_url, video_id, fatal=False, mpd_id='dash')
 620                 else:
 621                     continue
 622                 formats.extend(fmts)
 623                 self._merge_subtitles(subs, target=subtitles)
 624
 625         for sub_url in traverse_obj(api_json, (
 626                 '_embedded', 'subtitle',
 627                 ('xml_url', 'sami_url', 'stl_url', 'ttml_url', 'srt_url', 'vtt_url'), {url_or_none})):
 628             self._merge_subtitles({'de': [{'url': sub_url}]}, target=subtitles)
 629
 630         return {
 631             'id': video_id,
 632             'formats': formats,
 633             'subtitles': subtitles,
 634             **traverse_obj(api_json, {
 635                 'age_limit': ('age_classification', {parse_age_limit}),
 636                 'duration': ('duration_second', {float_or_none}),
 637                 'title': (('title', 'headline'), {str}),
 638                 'description': (('description', 'teaser_text'), {str}),
 639                 'media_type': ('video_type', {str}),
 640             }, get_all=False),
 641         }
 642
 643     def _real_extract(self, url):
 644         video_id = self._match_id(url)
 645         webpage = self._download_webpage(url, video_id)
 646
 647         return {
 648             'id': video_id,
 649             'title': self._html_search_meta(['og:title', 'twitter:title'], webpage, default=None),
 650             'description': self._html_search_meta(
 651                 ['description', 'og:description', 'twitter:description'], webpage, default=None),
 652             **self._search_json_ld(webpage, video_id, fatal=False),
 653             **self._extract_video(video_id),
 654         }