yt_dlp/extractor/vrt.py

   1 import json
   2 import time
   3 import urllib.parse
   4
   5 from .gigya import GigyaBaseIE
   6 from ..networking.exceptions import HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     clean_html,
  10     extract_attributes,
  11     float_or_none,
  12     get_element_by_class,
  13     get_element_html_by_class,
  14     int_or_none,
  15     join_nonempty,
  16     jwt_encode_hs256,
  17     make_archive_id,
  18     merge_dicts,
  19     parse_age_limit,
  20     parse_iso8601,
  21     str_or_none,
  22     strip_or_none,
  23     traverse_obj,
  24     url_or_none,
  25     urlencode_postdata,
  26 )
  27
  28
  29 class VRTBaseIE(GigyaBaseIE):
  30     _GEO_BYPASS = False
  31     _PLAYER_INFO = {
  32         'platform': 'desktop',
  33         'app': {
  34             'type': 'browser',
  35             'name': 'Chrome',
  36         },
  37         'device': 'undefined (undefined)',
  38         'os': {
  39             'name': 'Windows',
  40             'version': 'x86_64',
  41         },
  42         'player': {
  43             'name': 'VRT web player',
  44             'version': '2.7.4-prod-2023-04-19T06:05:45',
  45         },
  46     }
  47     # From https://player.vrt.be/vrtnws/js/main.js & https://player.vrt.be/ketnet/js/main.8cdb11341bcb79e4cd44.js
  48     _JWT_KEY_ID = '0-0Fp51UZykfaiCJrfTE3+oMI8zvDteYfPtR+2n1R+z8w='
  49     _JWT_SIGNING_KEY = 'b5f500d55cb44715107249ccd8a5c0136cfb2788dbb71b90a4f142423bacaf38'  # -dev
  50     # player-stag.vrt.be key:    d23987504521ae6fbf2716caca6700a24bb1579477b43c84e146b279de5ca595
  51     # player.vrt.be key:         2a9251d782700769fb856da5725daf38661874ca6f80ae7dc2b05ec1a81a24ae
  52
  53     def _extract_formats_and_subtitles(self, data, video_id):
  54         if traverse_obj(data, 'drm'):
  55             self.report_drm(video_id)
  56
  57         formats, subtitles = [], {}
  58         for target in traverse_obj(data, ('targetUrls', lambda _, v: url_or_none(v['url']) and v['type'])):
  59             format_type = target['type'].upper()
  60             format_url = target['url']
  61             if format_type in ('HLS', 'HLS_AES'):
  62                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
  63                     format_url, video_id, 'mp4', m3u8_id=format_type, fatal=False)
  64                 formats.extend(fmts)
  65                 self._merge_subtitles(subs, target=subtitles)
  66             elif format_type == 'HDS':
  67                 formats.extend(self._extract_f4m_formats(
  68                     format_url, video_id, f4m_id=format_type, fatal=False))
  69             elif format_type == 'MPEG_DASH':
  70                 fmts, subs = self._extract_mpd_formats_and_subtitles(
  71                     format_url, video_id, mpd_id=format_type, fatal=False)
  72                 formats.extend(fmts)
  73                 self._merge_subtitles(subs, target=subtitles)
  74             elif format_type == 'HSS':
  75                 fmts, subs = self._extract_ism_formats_and_subtitles(
  76                     format_url, video_id, ism_id='mss', fatal=False)
  77                 formats.extend(fmts)
  78                 self._merge_subtitles(subs, target=subtitles)
  79             else:
  80                 formats.append({
  81                     'format_id': format_type,
  82                     'url': format_url,
  83                 })
  84
  85         for sub in traverse_obj(data, ('subtitleUrls', lambda _, v: v['url'] and v['type'] == 'CLOSED')):
  86             subtitles.setdefault('nl', []).append({'url': sub['url']})
  87
  88         return formats, subtitles
  89
  90     def _call_api(self, video_id, client='null', id_token=None, version='v2'):
  91         player_info = {'exp': (round(time.time(), 3) + 900), **self._PLAYER_INFO}
  92         player_token = self._download_json(
  93             'https://media-services-public.vrt.be/vualto-video-aggregator-web/rest/external/v2/tokens',
  94             video_id, 'Downloading player token', headers={
  95                 **self.geo_verification_headers(),
  96                 'Content-Type': 'application/json',
  97             }, data=json.dumps({
  98                 'identityToken': id_token or {},
  99                 'playerInfo': jwt_encode_hs256(player_info, self._JWT_SIGNING_KEY, headers={
 100                     'kid': self._JWT_KEY_ID,
 101                 }).decode(),
 102             }, separators=(',', ':')).encode())['vrtPlayerToken']
 103
 104         return self._download_json(
 105             f'https://media-services-public.vrt.be/media-aggregator/{version}/media-items/{video_id}',
 106             video_id, 'Downloading API JSON', query={
 107                 'vrtPlayerToken': player_token,
 108                 'client': client,
 109             }, expected_status=400)
 110
 111
 112 class VRTIE(VRTBaseIE):
 113     IE_DESC = 'VRT NWS, Flanders News, Flandern Info and Sporza'
 114     _VALID_URL = r'https?://(?:www\.)?(?P<site>vrt\.be/vrtnws|sporza\.be)/[a-z]{2}/\d{4}/\d{2}/\d{2}/(?P<id>[^/?&#]+)'
 115     _TESTS = [{
 116         'url': 'https://www.vrt.be/vrtnws/nl/2019/05/15/beelden-van-binnenkant-notre-dame-een-maand-na-de-brand/',
 117         'info_dict': {
 118             'id': 'pbs-pub-7855fc7b-1448-49bc-b073-316cb60caa71$vid-2ca50305-c38a-4762-9890-65cbd098b7bd',
 119             'ext': 'mp4',
 120             'title': 'Beelden van binnenkant Notre-Dame, één maand na de brand',
 121             'description': 'md5:6fd85f999b2d1841aa5568f4bf02c3ff',
 122             'duration': 31.2,
 123             'thumbnail': 'https://images.vrt.be/orig/2019/05/15/2d914d61-7710-11e9-abcc-02b7b76bf47f.jpg',
 124         },
 125         'params': {'skip_download': 'm3u8'},
 126     }, {
 127         'url': 'https://sporza.be/nl/2019/05/15/de-belgian-cats-zijn-klaar-voor-het-ek/',
 128         'info_dict': {
 129             'id': 'pbs-pub-f2c86a46-8138-413a-a4b9-a0015a16ce2c$vid-1f112b31-e58e-4379-908d-aca6d80f8818',
 130             'ext': 'mp4',
 131             'title': 'De Belgian Cats zijn klaar voor het EK',
 132             'description': 'Video: De Belgian Cats zijn klaar voor het EK mét Ann Wauters | basketbal, sport in het journaal',
 133             'duration': 115.17,
 134             'thumbnail': 'https://images.vrt.be/orig/2019/05/15/11c0dba3-770e-11e9-abcc-02b7b76bf47f.jpg',
 135         },
 136         'params': {'skip_download': 'm3u8'},
 137     }]
 138     _CLIENT_MAP = {
 139         'vrt.be/vrtnws': 'vrtnieuws',
 140         'sporza.be': 'sporza',
 141     }
 142
 143     def _real_extract(self, url):
 144         site, display_id = self._match_valid_url(url).groups()
 145         webpage = self._download_webpage(url, display_id)
 146         attrs = extract_attributes(get_element_html_by_class('vrtvideo', webpage) or '')
 147
 148         asset_id = attrs.get('data-video-id') or attrs['data-videoid']
 149         publication_id = traverse_obj(attrs, 'data-publication-id', 'data-publicationid')
 150         if publication_id:
 151             asset_id = f'{publication_id}${asset_id}'
 152         client = traverse_obj(attrs, 'data-client-code', 'data-client') or self._CLIENT_MAP[site]
 153
 154         data = self._call_api(asset_id, client)
 155         formats, subtitles = self._extract_formats_and_subtitles(data, asset_id)
 156
 157         description = self._html_search_meta(
 158             ['og:description', 'twitter:description', 'description'], webpage)
 159         if description == '…':
 160             description = None
 161
 162         return {
 163             'id': asset_id,
 164             'formats': formats,
 165             'subtitles': subtitles,
 166             'description': description,
 167             'thumbnail': url_or_none(attrs.get('data-posterimage')),
 168             'duration': float_or_none(attrs.get('data-duration'), 1000),
 169             '_old_archive_ids': [make_archive_id('Canvas', asset_id)],
 170             **traverse_obj(data, {
 171                 'title': ('title', {str}),
 172                 'description': ('shortDescription', {str}),
 173                 'duration': ('duration', {float_or_none(scale=1000)}),
 174                 'thumbnail': ('posterImageUrl', {url_or_none}),
 175             }),
 176         }
 177
 178
 179 class VrtNUIE(VRTBaseIE):
 180     IE_DESC = 'VRT MAX'
 181     _VALID_URL = r'https?://(?:www\.)?vrt\.be/vrtnu/a-z/(?:[^/]+/){2}(?P<id>[^/?#&]+)'
 182     _TESTS = [{
 183         # CONTENT_IS_AGE_RESTRICTED
 184         'url': 'https://www.vrt.be/vrtnu/a-z/de-ideale-wereld/2023-vj/de-ideale-wereld-d20230116/',
 185         'info_dict': {
 186             'id': 'pbs-pub-855b00a8-6ce2-4032-ac4f-1fcf3ae78524$vid-d2243aa1-ec46-4e34-a55b-92568459906f',
 187             'ext': 'mp4',
 188             'title': 'Tom Waes',
 189             'description': 'Satirisch actualiteitenmagazine met Ella Leyers. Tom Waes is te gast.',
 190             'timestamp': 1673905125,
 191             'release_timestamp': 1673905125,
 192             'series': 'De ideale wereld',
 193             'season_id': '1672830988794',
 194             'episode': 'Aflevering 1',
 195             'episode_number': 1,
 196             'episode_id': '1672830988861',
 197             'display_id': 'de-ideale-wereld-d20230116',
 198             'channel': 'VRT',
 199             'duration': 1939.0,
 200             'thumbnail': 'https://images.vrt.be/orig/2023/01/10/1bb39cb3-9115-11ed-b07d-02b7b76bf47f.jpg',
 201             'release_date': '20230116',
 202             'upload_date': '20230116',
 203             'age_limit': 12,
 204         },
 205     }, {
 206         'url': 'https://www.vrt.be/vrtnu/a-z/buurman--wat-doet-u-nu-/6/buurman--wat-doet-u-nu--s6-trailer/',
 207         'info_dict': {
 208             'id': 'pbs-pub-ad4050eb-d9e5-48c2-9ec8-b6c355032361$vid-0465537a-34a8-4617-8352-4d8d983b4eee',
 209             'ext': 'mp4',
 210             'title': 'Trailer seizoen 6 \'Buurman, wat doet u nu?\'',
 211             'description': 'md5:197424726c61384b4e5c519f16c0cf02',
 212             'timestamp': 1652940000,
 213             'release_timestamp': 1652940000,
 214             'series': 'Buurman, wat doet u nu?',
 215             'season': 'Seizoen 6',
 216             'season_number': 6,
 217             'season_id': '1652344200907',
 218             'episode': 'Aflevering 0',
 219             'episode_number': 0,
 220             'episode_id': '1652951873524',
 221             'display_id': 'buurman--wat-doet-u-nu--s6-trailer',
 222             'channel': 'VRT',
 223             'duration': 33.13,
 224             'thumbnail': 'https://images.vrt.be/orig/2022/05/23/3c234d21-da83-11ec-b07d-02b7b76bf47f.jpg',
 225             'release_date': '20220519',
 226             'upload_date': '20220519',
 227         },
 228         'params': {'skip_download': 'm3u8'},
 229     }]
 230     _NETRC_MACHINE = 'vrtnu'
 231     _authenticated = False
 232
 233     def _perform_login(self, username, password):
 234         auth_info = self._gigya_login({
 235             'APIKey': '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy',
 236             'targetEnv': 'jssdk',
 237             'loginID': username,
 238             'password': password,
 239             'authMode': 'cookie',
 240         })
 241
 242         if auth_info.get('errorDetails'):
 243             raise ExtractorError(f'Unable to login. VrtNU said: {auth_info["errorDetails"]}', expected=True)
 244
 245         # Sometimes authentication fails for no good reason, retry
 246         for retry in self.RetryManager():
 247             if retry.attempt > 1:
 248                 self._sleep(1, None)
 249             try:
 250                 self._request_webpage(
 251                     'https://token.vrt.be/vrtnuinitlogin', None, note='Requesting XSRF Token',
 252                     errnote='Could not get XSRF Token', query={
 253                         'provider': 'site',
 254                         'destination': 'https://www.vrt.be/vrtnu/',
 255                     })
 256                 self._request_webpage(
 257                     'https://login.vrt.be/perform_login', None,
 258                     note='Performing login', errnote='Login failed',
 259                     query={'client_id': 'vrtnu-site'}, data=urlencode_postdata({
 260                         'UID': auth_info['UID'],
 261                         'UIDSignature': auth_info['UIDSignature'],
 262                         'signatureTimestamp': auth_info['signatureTimestamp'],
 263                         '_csrf': self._get_cookies('https://login.vrt.be').get('OIDCXSRF').value,
 264                     }))
 265             except ExtractorError as e:
 266                 if isinstance(e.cause, HTTPError) and e.cause.status == 401:
 267                     retry.error = e
 268                     continue
 269                 raise
 270
 271         self._authenticated = True
 272
 273     def _real_extract(self, url):
 274         display_id = self._match_id(url)
 275         parsed_url = urllib.parse.urlparse(url)
 276         details = self._download_json(
 277             f'{parsed_url.scheme}://{parsed_url.netloc}{parsed_url.path.rstrip("/")}.model.json',
 278             display_id, 'Downloading asset JSON', 'Unable to download asset JSON')['details']
 279
 280         watch_info = traverse_obj(details, (
 281             'actions', lambda _, v: v['type'] == 'watch-episode', {dict}), get_all=False) or {}
 282         video_id = join_nonempty(
 283             'episodePublicationId', 'episodeVideoId', delim='$', from_dict=watch_info)
 284         if '$' not in video_id:
 285             raise ExtractorError('Unable to extract video ID')
 286
 287         vrtnutoken = self._download_json(
 288             'https://token.vrt.be/refreshtoken', video_id, note='Retrieving vrtnutoken',
 289             errnote='Token refresh failed')['vrtnutoken'] if self._authenticated else None
 290
 291         video_info = self._call_api(video_id, 'vrtnu-web@PROD', vrtnutoken)
 292
 293         if 'title' not in video_info:
 294             code = video_info.get('code')
 295             if code in ('AUTHENTICATION_REQUIRED', 'CONTENT_IS_AGE_RESTRICTED'):
 296                 self.raise_login_required(code, method='password')
 297             elif code in ('INVALID_LOCATION', 'CONTENT_AVAILABLE_ONLY_IN_BE'):
 298                 self.raise_geo_restricted(countries=['BE'])
 299             elif code == 'CONTENT_AVAILABLE_ONLY_FOR_BE_RESIDENTS_AND_EXPATS':
 300                 if not self._authenticated:
 301                     self.raise_login_required(code, method='password')
 302                 self.raise_geo_restricted(countries=['BE'])
 303             raise ExtractorError(code, expected=True)
 304
 305         formats, subtitles = self._extract_formats_and_subtitles(video_info, video_id)
 306
 307         return {
 308             **traverse_obj(details, {
 309                 'title': 'title',
 310                 'description': ('description', {clean_html}),
 311                 'timestamp': ('data', 'episode', 'onTime', 'raw', {parse_iso8601}),
 312                 'release_timestamp': ('data', 'episode', 'onTime', 'raw', {parse_iso8601}),
 313                 'series': ('data', 'program', 'title'),
 314                 'season': ('data', 'season', 'title', 'value'),
 315                 'season_number': ('data', 'season', 'title', 'raw', {int_or_none}),
 316                 'season_id': ('data', 'season', 'id', {str_or_none}),
 317                 'episode': ('data', 'episode', 'number', 'value', {str_or_none}),
 318                 'episode_number': ('data', 'episode', 'number', 'raw', {int_or_none}),
 319                 'episode_id': ('data', 'episode', 'id', {str_or_none}),
 320                 'age_limit': ('data', 'episode', 'age', 'raw', {parse_age_limit}),
 321             }),
 322             'id': video_id,
 323             'display_id': display_id,
 324             'channel': 'VRT',
 325             'formats': formats,
 326             'duration': float_or_none(video_info.get('duration'), 1000),
 327             'thumbnail': url_or_none(video_info.get('posterImageUrl')),
 328             'subtitles': subtitles,
 329             '_old_archive_ids': [make_archive_id('Canvas', video_id)],
 330         }
 331
 332
 333 class KetnetIE(VRTBaseIE):
 334     _VALID_URL = r'https?://(?:www\.)?ketnet\.be/(?P<id>(?:[^/]+/)*[^/?#&]+)'
 335     _TESTS = [{
 336         'url': 'https://www.ketnet.be/kijken/m/meisjes/6/meisjes-s6a5',
 337         'info_dict': {
 338             'id': 'pbs-pub-39f8351c-a0a0-43e6-8394-205d597d6162$vid-5e306921-a9aa-4fa9-9f39-5b82c8f1028e',
 339             'ext': 'mp4',
 340             'title': 'Meisjes',
 341             'episode': 'Reeks 6: Week 5',
 342             'season': 'Reeks 6',
 343             'series': 'Meisjes',
 344             'timestamp': 1685251800,
 345             'upload_date': '20230528',
 346         },
 347         'params': {'skip_download': 'm3u8'},
 348     }]
 349
 350     def _real_extract(self, url):
 351         display_id = self._match_id(url)
 352
 353         video = self._download_json(
 354             'https://senior-bff.ketnet.be/graphql', display_id, query={
 355                 'query': '''{
 356   video(id: "content/ketnet/nl/%s.model.json") {
 357     description
 358     episodeNr
 359     imageUrl
 360     mediaReference
 361     programTitle
 362     publicationDate
 363     seasonTitle
 364     subtitleVideodetail
 365     titleVideodetail
 366   }
 367 }''' % display_id,  # noqa: UP031
 368             })['data']['video']
 369
 370         video_id = urllib.parse.unquote(video['mediaReference'])
 371         data = self._call_api(video_id, 'ketnet@PROD', version='v1')
 372         formats, subtitles = self._extract_formats_and_subtitles(data, video_id)
 373
 374         return {
 375             'id': video_id,
 376             'formats': formats,
 377             'subtitles': subtitles,
 378             '_old_archive_ids': [make_archive_id('Canvas', video_id)],
 379             **traverse_obj(video, {
 380                 'title': ('titleVideodetail', {str}),
 381                 'description': ('description', {str}),
 382                 'thumbnail': ('thumbnail', {url_or_none}),
 383                 'timestamp': ('publicationDate', {parse_iso8601}),
 384                 'series': ('programTitle', {str}),
 385                 'season': ('seasonTitle', {str}),
 386                 'episode': ('subtitleVideodetail', {str}),
 387                 'episode_number': ('episodeNr', {int_or_none}),
 388             }),
 389         }
 390
 391
 392 class DagelijkseKostIE(VRTBaseIE):
 393     IE_DESC = 'dagelijksekost.een.be'
 394     _VALID_URL = r'https?://dagelijksekost\.een\.be/gerechten/(?P<id>[^/?#&]+)'
 395     _TESTS = [{
 396         'url': 'https://dagelijksekost.een.be/gerechten/hachis-parmentier-met-witloof',
 397         'info_dict': {
 398             'id': 'md-ast-27a4d1ff-7d7b-425e-b84f-a4d227f592fa',
 399             'ext': 'mp4',
 400             'title': 'Hachis parmentier met witloof',
 401             'description': 'md5:9960478392d87f63567b5b117688cdc5',
 402             'display_id': 'hachis-parmentier-met-witloof',
 403         },
 404         'params': {'skip_download': 'm3u8'},
 405     }]
 406
 407     def _real_extract(self, url):
 408         display_id = self._match_id(url)
 409         webpage = self._download_webpage(url, display_id)
 410         video_id = self._html_search_regex(
 411             r'data-url=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id', group='id')
 412
 413         data = self._call_api(video_id, 'dako@prod', version='v1')
 414         formats, subtitles = self._extract_formats_and_subtitles(data, video_id)
 415
 416         return {
 417             'id': video_id,
 418             'formats': formats,
 419             'subtitles': subtitles,
 420             'display_id': display_id,
 421             'title': strip_or_none(get_element_by_class(
 422                 'dish-metadata__title', webpage) or self._html_search_meta('twitter:title', webpage)),
 423             'description': clean_html(get_element_by_class(
 424                 'dish-description', webpage)) or self._html_search_meta(
 425                 ['description', 'twitter:description', 'og:description'], webpage),
 426             '_old_archive_ids': [make_archive_id('Canvas', video_id)],
 427         }
 428
 429
 430 class Radio1BeIE(VRTBaseIE):
 431     _VALID_URL = r'https?://radio1\.be/(?:lees|luister/select)/(?P<id>[\w/-]+)'
 432     _TESTS = [{
 433         'url': 'https://radio1.be/luister/select/de-ochtend/komt-n-va-volgend-jaar-op-in-wallonie',
 434         'info_dict': {
 435             'id': 'eb6c22e9-544f-44f4-af39-cf8cccd29e22',
 436             'title': 'Komt N-VA volgend jaar op in Wallonië?',
 437             'display_id': 'de-ochtend/komt-n-va-volgend-jaar-op-in-wallonie',
 438             'description': 'md5:b374ea1c9302f38362df9dea1931468e',
 439             'thumbnail': r're:https?://cds\.vrt\.radio/[^/#\?&]+',
 440         },
 441         'playlist_mincount': 1,
 442     }, {
 443         'url': 'https://radio1.be/lees/europese-unie-wil-onmiddellijke-humanitaire-pauze-en-duurzaam-staakt-het-vuren-in-gaza?view=web',
 444         'info_dict': {
 445             'id': '5d47f102-dbdb-4fa0-832b-26c1870311f2',
 446             'title': 'Europese Unie wil "onmiddellijke humanitaire pauze" en "duurzaam staakt-het-vuren" in Gaza',
 447             'description': 'md5:1aad1fae7d39edeffde5d3e67d276b64',
 448             'thumbnail': r're:https?://cds\.vrt\.radio/[^/#\?&]+',
 449             'display_id': 'europese-unie-wil-onmiddellijke-humanitaire-pauze-en-duurzaam-staakt-het-vuren-in-gaza',
 450         },
 451         'playlist_mincount': 1,
 452     }]
 453
 454     def _extract_video_entries(self, next_js_data, display_id):
 455         video_data = traverse_obj(
 456             next_js_data, ((None, ('paragraphs', ...)), {lambda x: x if x['mediaReference'] else None}))
 457         for data in video_data:
 458             media_reference = data['mediaReference']
 459             formats, subtitles = self._extract_formats_and_subtitles(
 460                 self._call_api(media_reference), display_id)
 461
 462             yield {
 463                 'id': media_reference,
 464                 'formats': formats,
 465                 'subtitles': subtitles,
 466                 **traverse_obj(data, {
 467                     'title': ('title', {str}),
 468                     'description': ('body', {clean_html}),
 469                 }),
 470             }
 471
 472     def _real_extract(self, url):
 473         display_id = self._match_id(url)
 474         webpage = self._download_webpage(url, display_id)
 475         next_js_data = self._search_nextjs_data(webpage, display_id)['props']['pageProps']['item']
 476
 477         return self.playlist_result(
 478             self._extract_video_entries(next_js_data, display_id), **merge_dicts(traverse_obj(
 479                 next_js_data, ({
 480                     'id': ('id', {str}),
 481                     'title': ('title', {str}),
 482                     'description': (('description', 'content'), {clean_html}),
 483                 }), get_all=False), {
 484                     'display_id': display_id,
 485                     'title': self._html_search_meta(['name', 'og:title', 'twitter:title'], webpage),
 486                     'description': self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage),
 487                     'thumbnail': self._html_search_meta(['og:image', 'twitter:image'], webpage),
 488             }))