yt_dlp/extractor/cbc.py

   1 import base64
   2 import json
   3 import re
   4 import time
   5 import urllib.parse
   6 import xml.etree.ElementTree
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_str,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     int_or_none,
  15     join_nonempty,
  16     js_to_json,
  17     orderedSet,
  18     parse_iso8601,
  19     smuggle_url,
  20     strip_or_none,
  21     traverse_obj,
  22     try_get,
  23 )
  24
  25
  26 class CBCIE(InfoExtractor):
  27     IE_NAME = 'cbc.ca'
  28     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P<id>[^/?#]+)'
  29     _TESTS = [{
  30         # with mediaId
  31         'url': 'http://www.cbc.ca/22minutes/videos/clips-season-23/don-cherry-play-offs',
  32         'md5': '97e24d09672fc4cf56256d6faa6c25bc',
  33         'info_dict': {
  34             'id': '2682904050',
  35             'ext': 'mp4',
  36             'title': 'Don Cherry – All-Stars',
  37             'description': 'Don Cherry has a bee in his bonnet about AHL player John Scott because that guy’s got heart.',
  38             'timestamp': 1454463000,
  39             'upload_date': '20160203',
  40             'uploader': 'CBCC-NEW',
  41         },
  42         'skip': 'Geo-restricted to Canada',
  43     }, {
  44         # with clipId, feed available via tpfeed.cbc.ca and feed.theplatform.com
  45         'url': 'http://www.cbc.ca/22minutes/videos/22-minutes-update/22-minutes-update-episode-4',
  46         'md5': '162adfa070274b144f4fdc3c3b8207db',
  47         'info_dict': {
  48             'id': '2414435309',
  49             'ext': 'mp4',
  50             'title': '22 Minutes Update: What Not To Wear Quebec',
  51             'description': "This week's latest Canadian top political story is What Not To Wear Quebec.",
  52             'upload_date': '20131025',
  53             'uploader': 'CBCC-NEW',
  54             'timestamp': 1382717907,
  55         },
  56         'skip': 'No longer available',
  57     }, {
  58         # with clipId, feed only available via tpfeed.cbc.ca
  59         'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
  60         'md5': '0274a90b51a9b4971fe005c63f592f12',
  61         'info_dict': {
  62             'id': '2487345465',
  63             'ext': 'mp4',
  64             'title': 'Robin Williams freestyles on 90 Minutes Live',
  65             'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
  66             'upload_date': '19780210',
  67             'uploader': 'CBCC-NEW',
  68             'timestamp': 255977160,
  69         },
  70         'skip': '404 Not Found',
  71     }, {
  72         # multiple iframes
  73         'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot',
  74         'playlist': [{
  75             'md5': '377572d0b49c4ce0c9ad77470e0b96b4',
  76             'info_dict': {
  77                 'id': '2680832926',
  78                 'ext': 'mp4',
  79                 'title': 'An Eagle\'s-Eye View Off Burrard Bridge',
  80                 'description': 'Hercules the eagle flies from Vancouver\'s Burrard Bridge down to a nearby park with a mini-camera strapped to his back.',
  81                 'upload_date': '20160201',
  82                 'timestamp': 1454342820,
  83                 'uploader': 'CBCC-NEW',
  84             },
  85         }, {
  86             'md5': '415a0e3f586113894174dfb31aa5bb1a',
  87             'info_dict': {
  88                 'id': '2658915080',
  89                 'ext': 'mp4',
  90                 'title': 'Fly like an eagle!',
  91                 'description': 'Eagle equipped with a mini camera flies from the world\'s tallest tower',
  92                 'upload_date': '20150315',
  93                 'timestamp': 1426443984,
  94                 'uploader': 'CBCC-NEW',
  95             },
  96         }],
  97         'skip': 'Geo-restricted to Canada',
  98     }, {
  99         # multiple CBC.APP.Caffeine.initInstance(...)
 100         'url': 'http://www.cbc.ca/news/canada/calgary/dog-indoor-exercise-winter-1.3928238',
 101         'info_dict': {
 102             'title': 'Keep Rover active during the deep freeze with doggie pushups and other fun indoor tasks',  # FIXME
 103             'id': 'dog-indoor-exercise-winter-1.3928238',
 104             'description': 'md5:c18552e41726ee95bd75210d1ca9194c',
 105         },
 106         'playlist_mincount': 6,
 107     }]
 108
 109     @classmethod
 110     def suitable(cls, url):
 111         return False if CBCPlayerIE.suitable(url) else super(CBCIE, cls).suitable(url)
 112
 113     def _extract_player_init(self, player_init, display_id):
 114         player_info = self._parse_json(player_init, display_id, js_to_json)
 115         media_id = player_info.get('mediaId')
 116         if not media_id:
 117             clip_id = player_info['clipId']
 118             feed = self._download_json(
 119                 'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={:mpsReleases}{%s}' % clip_id,
 120                 clip_id, fatal=False)
 121             if feed:
 122                 media_id = try_get(feed, lambda x: x['entries'][0]['guid'], compat_str)
 123             if not media_id:
 124                 media_id = self._download_json(
 125                     'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
 126                     clip_id)['entries'][0]['id'].split('/')[-1]
 127         return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
 128
 129     def _real_extract(self, url):
 130         display_id = self._match_id(url)
 131         webpage = self._download_webpage(url, display_id)
 132         title = (self._og_search_title(webpage, default=None)
 133                  or self._html_search_meta('twitter:title', webpage, 'title', default=None)
 134                  or self._html_extract_title(webpage))
 135         entries = [
 136             self._extract_player_init(player_init, display_id)
 137             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
 138         media_ids = []
 139         for media_id_re in (
 140                 r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"',
 141                 r'<div[^>]+\bid=["\']player-(\d+)',
 142                 r'guid["\']\s*:\s*["\'](\d+)'):
 143             media_ids.extend(re.findall(media_id_re, webpage))
 144         entries.extend([
 145             self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
 146             for media_id in orderedSet(media_ids)])
 147         return self.playlist_result(
 148             entries, display_id, strip_or_none(title),
 149             self._og_search_description(webpage))
 150
 151
 152 class CBCPlayerIE(InfoExtractor):
 153     IE_NAME = 'cbc.ca:player'
 154     _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/(?:video/)?|i/caffeine/syndicate/\?mediaId=))(?P<id>(?:\d\.)?\d+)'
 155     _TESTS = [{
 156         'url': 'http://www.cbc.ca/player/play/2683190193',
 157         'md5': '64d25f841ddf4ddb28a235338af32e2c',
 158         'info_dict': {
 159             'id': '2683190193',
 160             'ext': 'mp4',
 161             'title': 'Gerry Runs a Sweat Shop',
 162             'description': 'md5:b457e1c01e8ff408d9d801c1c2cd29b0',
 163             'timestamp': 1455071400,
 164             'upload_date': '20160210',
 165             'uploader': 'CBCC-NEW',
 166         },
 167         'skip': 'Geo-restricted to Canada and no longer available',
 168     }, {
 169         'url': 'http://www.cbc.ca/i/caffeine/syndicate/?mediaId=2657631896',
 170         'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
 171         'info_dict': {
 172             'id': '2657631896',
 173             'ext': 'mp3',
 174             'title': 'CBC Montreal is organizing its first ever community hackathon!',
 175             'description': 'md5:dd3b692f0a139b0369943150bd1c46a9',
 176             'timestamp': 1425704400,
 177             'upload_date': '20150307',
 178             'uploader': 'CBCC-NEW',
 179             'thumbnail': 'http://thumbnails.cbc.ca/maven_legacy/thumbnails/sonali-karnick-220.jpg',
 180             'chapters': [],
 181             'duration': 494.811,
 182             'categories': ['AudioMobile/All in a Weekend Montreal'],
 183             'tags': 'count:8',
 184             'location': 'Quebec',
 185             'series': 'All in a Weekend Montreal',
 186             'season': 'Season 2015',
 187             'season_number': 2015,
 188             'media_type': 'Excerpt',
 189         },
 190     }, {
 191         'url': 'http://www.cbc.ca/i/caffeine/syndicate/?mediaId=2164402062',
 192         'md5': '33fcd8f6719b9dd60a5e73adcb83b9f6',
 193         'info_dict': {
 194             'id': '2164402062',
 195             'ext': 'mp4',
 196             'title': 'Cancer survivor four times over',
 197             'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
 198             'timestamp': 1320410746,
 199             'upload_date': '20111104',
 200             'uploader': 'CBCC-NEW',
 201             'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/277/67/cancer_852x480_2164412612.jpg',
 202             'chapters': [],
 203             'duration': 186.867,
 204             'series': 'CBC News: Windsor at 6:00',
 205             'categories': ['News/Canada/Windsor'],
 206             'location': 'Windsor',
 207             'tags': ['cancer'],
 208             'creators': ['Allison Johnson'],
 209             'media_type': 'Excerpt',
 210         },
 211     }, {
 212         # Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/
 213         'url': 'https://www.cbc.ca/player/play/1.2985700',
 214         'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
 215         'info_dict': {
 216             'id': '2657631896',
 217             'ext': 'mp3',
 218             'title': 'CBC Montreal is organizing its first ever community hackathon!',
 219             'description': 'The modern technology we tend to depend on so heavily, is never without it\'s share of hiccups and headaches. Next weekend - CBC Montreal will be getting members of the public for its first Hackathon.',
 220             'timestamp': 1425704400,
 221             'upload_date': '20150307',
 222             'uploader': 'CBCC-NEW',
 223             'thumbnail': 'http://thumbnails.cbc.ca/maven_legacy/thumbnails/sonali-karnick-220.jpg',
 224             'chapters': [],
 225             'duration': 494.811,
 226             'categories': ['AudioMobile/All in a Weekend Montreal'],
 227             'tags': 'count:8',
 228             'location': 'Quebec',
 229             'series': 'All in a Weekend Montreal',
 230             'season': 'Season 2015',
 231             'season_number': 2015,
 232             'media_type': 'Excerpt',
 233         },
 234     }, {
 235         'url': 'https://www.cbc.ca/player/play/1.1711287',
 236         'md5': '33fcd8f6719b9dd60a5e73adcb83b9f6',
 237         'info_dict': {
 238             'id': '2164402062',
 239             'ext': 'mp4',
 240             'title': 'Cancer survivor four times over',
 241             'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
 242             'timestamp': 1320410746,
 243             'upload_date': '20111104',
 244             'uploader': 'CBCC-NEW',
 245             'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/277/67/cancer_852x480_2164412612.jpg',
 246             'chapters': [],
 247             'duration': 186.867,
 248             'series': 'CBC News: Windsor at 6:00',
 249             'categories': ['News/Canada/Windsor'],
 250             'location': 'Windsor',
 251             'tags': ['cancer'],
 252             'creators': ['Allison Johnson'],
 253             'media_type': 'Excerpt',
 254         },
 255     }, {
 256         # Has subtitles
 257         # These broadcasts expire after ~1 month, can find new test URL here:
 258         # https://www.cbc.ca/player/news/TV%20Shows/The%20National/Latest%20Broadcast
 259         'url': 'https://www.cbc.ca/player/play/1.7159484',
 260         'md5': '6ed6cd0fc2ef568d2297ba68a763d455',
 261         'info_dict': {
 262             'id': '2324213316001',
 263             'ext': 'mp4',
 264             'title': 'The National | School boards sue social media giants',
 265             'description': 'md5:4b4db69322fa32186c3ce426da07402c',
 266             'timestamp': 1711681200,
 267             'duration': 2743.400,
 268             'subtitles': {'eng': [{'ext': 'vtt', 'protocol': 'm3u8_native'}]},
 269             'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/607/559/thumbnail.jpeg',
 270             'uploader': 'CBCC-NEW',
 271             'chapters': 'count:5',
 272             'upload_date': '20240329',
 273             'categories': 'count:4',
 274             'series': 'The National - Full Show',
 275             'tags': 'count:1',
 276             'creators': ['News'],
 277             'location': 'Canada',
 278             'media_type': 'Full Program',
 279         },
 280     }, {
 281         'url': 'https://www.cbc.ca/player/play/video/1.7194274',
 282         'md5': '188b96cf6bdcb2540e178a6caa957128',
 283         'info_dict': {
 284             'id': '2334524995812',
 285             'ext': 'mp4',
 286             'title': '#TheMoment a rare white spirit moose was spotted in Alberta',
 287             'description': 'md5:18ae269a2d0265c5b0bbe4b2e1ac61a3',
 288             'timestamp': 1714788791,
 289             'duration': 77.678,
 290             'subtitles': {'eng': [{'ext': 'vtt', 'protocol': 'm3u8_native'}]},
 291             'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/201/543/THE_MOMENT.jpg',
 292             'uploader': 'CBCC-NEW',
 293             'chapters': 'count:0',
 294             'upload_date': '20240504',
 295             'categories': 'count:3',
 296             'series': 'The National',
 297             'tags': 'count:15',
 298             'creators': ['encoder'],
 299             'location': 'Canada',
 300             'media_type': 'Excerpt',
 301         },
 302     }, {
 303         'url': 'cbcplayer:1.7159484',
 304         'only_matching': True,
 305     }, {
 306         'url': 'cbcplayer:2164402062',
 307         'only_matching': True,
 308     }, {
 309         'url': 'http://www.cbc.ca/player/play/2657631896',
 310         'only_matching': True,
 311     }]
 312
 313     def _real_extract(self, url):
 314         video_id = self._match_id(url)
 315         if '.' in video_id:
 316             webpage = self._download_webpage(f'https://www.cbc.ca/player/play/{video_id}', video_id)
 317             video_id = self._search_json(
 318                 r'window\.__INITIAL_STATE__\s*=', webpage,
 319                 'initial state', video_id)['video']['currentClip']['mediaId']
 320
 321         return {
 322             '_type': 'url_transparent',
 323             'ie_key': 'ThePlatform',
 324             'url': smuggle_url(
 325                 'http://link.theplatform.com/s/ExhSPC/media/guid/2655402169/%s?mbr=true&formats=MPEG4,FLV,MP3' % video_id, {
 326                     'force_smil_url': True
 327                 }),
 328             'id': video_id,
 329             '_format_sort_fields': ('res', 'proto')  # Prioritize direct http formats over HLS
 330         }
 331
 332
 333 class CBCPlayerPlaylistIE(InfoExtractor):
 334     IE_NAME = 'cbc.ca:player:playlist'
 335     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?:player/)(?!play/)(?P<id>[^?#]+)'
 336     _TESTS = [{
 337         'url': 'https://www.cbc.ca/player/news/TV%20Shows/The%20National/Latest%20Broadcast',
 338         'playlist_mincount': 25,
 339         'info_dict': {
 340             'id': 'news/tv shows/the national/latest broadcast',
 341         }
 342     }, {
 343         'url': 'https://www.cbc.ca/player/news/Canada/North',
 344         'playlist_mincount': 25,
 345         'info_dict': {
 346             'id': 'news/canada/north',
 347         }
 348     }]
 349
 350     def _real_extract(self, url):
 351         playlist_id = urllib.parse.unquote(self._match_id(url)).lower()
 352         webpage = self._download_webpage(url, playlist_id)
 353         json_content = self._search_json(
 354             r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', playlist_id)
 355
 356         def entries():
 357             for video_id in traverse_obj(json_content, (
 358                 'video', 'clipsByCategory', lambda k, _: k.lower() == playlist_id, 'items', ..., 'id'
 359             )):
 360                 yield self.url_result(f'https://www.cbc.ca/player/play/{video_id}', CBCPlayerIE)
 361
 362         return self.playlist_result(entries(), playlist_id)
 363
 364
 365 class CBCGemIE(InfoExtractor):
 366     IE_NAME = 'gem.cbc.ca'
 367     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>[0-9a-z-]+/s[0-9]+[a-z][0-9]+)'
 368     _TESTS = [{
 369         # This is a normal, public, TV show video
 370         'url': 'https://gem.cbc.ca/media/schitts-creek/s06e01',
 371         'md5': '93dbb31c74a8e45b378cf13bd3f6f11e',
 372         'info_dict': {
 373             'id': 'schitts-creek/s06e01',
 374             'ext': 'mp4',
 375             'title': 'Smoke Signals',
 376             'description': 'md5:929868d20021c924020641769eb3e7f1',
 377             'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/episode/perso/cbc_schitts_creek_season_06e01_thumbnail_v01.jpg?im=Resize=(Size)',
 378             'duration': 1314,
 379             'categories': ['comedy'],
 380             'series': 'Schitt\'s Creek',
 381             'season': 'Season 6',
 382             'season_number': 6,
 383             'episode': 'Smoke Signals',
 384             'episode_number': 1,
 385             'episode_id': 'schitts-creek/s06e01',
 386         },
 387         'params': {'format': 'bv'},
 388         'skip': 'Geo-restricted to Canada',
 389     }, {
 390         # This video requires an account in the browser, but works fine in yt-dlp
 391         'url': 'https://gem.cbc.ca/media/schitts-creek/s01e01',
 392         'md5': '297a9600f554f2258aed01514226a697',
 393         'info_dict': {
 394             'id': 'schitts-creek/s01e01',
 395             'ext': 'mp4',
 396             'title': 'The Cup Runneth Over',
 397             'description': 'md5:9bca14ea49ab808097530eb05a29e797',
 398             'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/episode/perso/cbc_schitts_creek_season_01e01_thumbnail_v01.jpg?im=Resize=(Size)',
 399             'series': 'Schitt\'s Creek',
 400             'season_number': 1,
 401             'season': 'Season 1',
 402             'episode_number': 1,
 403             'episode': 'The Cup Runneth Over',
 404             'episode_id': 'schitts-creek/s01e01',
 405             'duration': 1309,
 406             'categories': ['comedy'],
 407         },
 408         'params': {'format': 'bv'},
 409         'skip': 'Geo-restricted to Canada',
 410     }, {
 411         'url': 'https://gem.cbc.ca/nadiyas-family-favourites/s01e01',
 412         'only_matching': True,
 413     }]
 414
 415     _GEO_COUNTRIES = ['CA']
 416     _TOKEN_API_KEY = '3f4beddd-2061-49b0-ae80-6f1f2ed65b37'
 417     _NETRC_MACHINE = 'cbcgem'
 418     _claims_token = None
 419
 420     def _new_claims_token(self, email, password):
 421         data = json.dumps({
 422             'email': email,
 423             'password': password,
 424         }).encode()
 425         headers = {'content-type': 'application/json'}
 426         query = {'apikey': self._TOKEN_API_KEY}
 427         resp = self._download_json('https://api.loginradius.com/identity/v2/auth/login',
 428                                    None, data=data, headers=headers, query=query)
 429         access_token = resp['access_token']
 430
 431         query = {
 432             'access_token': access_token,
 433             'apikey': self._TOKEN_API_KEY,
 434             'jwtapp': 'jwt',
 435         }
 436         resp = self._download_json('https://cloud-api.loginradius.com/sso/jwt/api/token',
 437                                    None, headers=headers, query=query)
 438         sig = resp['signature']
 439
 440         data = json.dumps({'jwt': sig}).encode()
 441         headers = {'content-type': 'application/json', 'ott-device-type': 'web'}
 442         resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/token',
 443                                    None, data=data, headers=headers, expected_status=426)
 444         cbc_access_token = resp['accessToken']
 445
 446         headers = {'content-type': 'application/json', 'ott-device-type': 'web', 'ott-access-token': cbc_access_token}
 447         resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/profile',
 448                                    None, headers=headers, expected_status=426)
 449         return resp['claimsToken']
 450
 451     def _get_claims_token_expiry(self):
 452         # Token is a JWT
 453         # JWT is decoded here and 'exp' field is extracted
 454         # It is a Unix timestamp for when the token expires
 455         b64_data = self._claims_token.split('.')[1]
 456         data = base64.urlsafe_b64decode(b64_data + "==")
 457         return json.loads(data)['exp']
 458
 459     def claims_token_expired(self):
 460         exp = self._get_claims_token_expiry()
 461         if exp - time.time() < 10:
 462             # It will expire in less than 10 seconds, or has already expired
 463             return True
 464         return False
 465
 466     def claims_token_valid(self):
 467         return self._claims_token is not None and not self.claims_token_expired()
 468
 469     def _get_claims_token(self, email, password):
 470         if not self.claims_token_valid():
 471             self._claims_token = self._new_claims_token(email, password)
 472             self.cache.store(self._NETRC_MACHINE, 'claims_token', self._claims_token)
 473         return self._claims_token
 474
 475     def _real_initialize(self):
 476         if self.claims_token_valid():
 477             return
 478         self._claims_token = self.cache.load(self._NETRC_MACHINE, 'claims_token')
 479
 480     def _find_secret_formats(self, formats, video_id):
 481         """ Find a valid video url and convert it to the secret variant """
 482         base_format = next((f for f in formats if f.get('vcodec') != 'none'), None)
 483         if not base_format:
 484             return
 485
 486         base_url = re.sub(r'(Manifest\(.*?),filter=[\w-]+(.*?\))', r'\1\2', base_format['url'])
 487         url = re.sub(r'(Manifest\(.*?),format=[\w-]+(.*?\))', r'\1\2', base_url)
 488
 489         secret_xml = self._download_xml(url, video_id, note='Downloading secret XML', fatal=False)
 490         if not isinstance(secret_xml, xml.etree.ElementTree.Element):
 491             return
 492
 493         for child in secret_xml:
 494             if child.attrib.get('Type') != 'video':
 495                 continue
 496             for video_quality in child:
 497                 bitrate = int_or_none(video_quality.attrib.get('Bitrate'))
 498                 if not bitrate or 'Index' not in video_quality.attrib:
 499                     continue
 500                 height = int_or_none(video_quality.attrib.get('MaxHeight'))
 501
 502                 yield {
 503                     **base_format,
 504                     'format_id': join_nonempty('sec', height),
 505                     # Note: \g<1> is necessary instead of \1 since bitrate is a number
 506                     'url': re.sub(r'(QualityLevels\()\d+(\))', fr'\g<1>{bitrate}\2', base_url),
 507                     'width': int_or_none(video_quality.attrib.get('MaxWidth')),
 508                     'tbr': bitrate / 1000.0,
 509                     'height': height,
 510                 }
 511
 512     def _real_extract(self, url):
 513         video_id = self._match_id(url)
 514         video_info = self._download_json(
 515             f'https://services.radio-canada.ca/ott/cbc-api/v2/assets/{video_id}',
 516             video_id, expected_status=426)
 517
 518         email, password = self._get_login_info()
 519         if email and password:
 520             claims_token = self._get_claims_token(email, password)
 521             headers = {'x-claims-token': claims_token}
 522         else:
 523             headers = {}
 524         m3u8_info = self._download_json(video_info['playSession']['url'], video_id, headers=headers)
 525         m3u8_url = m3u8_info.get('url')
 526
 527         if m3u8_info.get('errorCode') == 1:
 528             self.raise_geo_restricted(countries=['CA'])
 529         elif m3u8_info.get('errorCode') == 35:
 530             self.raise_login_required(method='password')
 531         elif m3u8_info.get('errorCode') != 0:
 532             raise ExtractorError(f'{self.IE_NAME} said: {m3u8_info.get("errorCode")} - {m3u8_info.get("message")}')
 533
 534         formats = self._extract_m3u8_formats(m3u8_url, video_id, m3u8_id='hls')
 535         self._remove_duplicate_formats(formats)
 536         formats.extend(self._find_secret_formats(formats, video_id))
 537
 538         for format in formats:
 539             if format.get('vcodec') == 'none':
 540                 if format.get('ext') is None:
 541                     format['ext'] = 'm4a'
 542                 if format.get('acodec') is None:
 543                     format['acodec'] = 'mp4a.40.2'
 544
 545                 # Put described audio at the beginning of the list, so that it
 546                 # isn't chosen by default, as most people won't want it.
 547                 if 'descriptive' in format['format_id'].lower():
 548                     format['preference'] = -2
 549
 550         return {
 551             'id': video_id,
 552             'title': video_info['title'],
 553             'description': video_info.get('description'),
 554             'thumbnail': video_info.get('image'),
 555             'series': video_info.get('series'),
 556             'season_number': video_info.get('season'),
 557             'season': f'Season {video_info.get("season")}',
 558             'episode_number': video_info.get('episode'),
 559             'episode': video_info.get('title'),
 560             'episode_id': video_id,
 561             'duration': video_info.get('duration'),
 562             'categories': [video_info.get('category')],
 563             'formats': formats,
 564             'release_timestamp': video_info.get('airDate'),
 565             'timestamp': video_info.get('availableDate'),
 566         }
 567
 568
 569 class CBCGemPlaylistIE(InfoExtractor):
 570     IE_NAME = 'gem.cbc.ca:playlist'
 571     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>(?P<show>[0-9a-z-]+)/s(?P<season>[0-9]+))/?(?:[?#]|$)'
 572     _TESTS = [{
 573         # TV show playlist, all public videos
 574         'url': 'https://gem.cbc.ca/media/schitts-creek/s06',
 575         'playlist_count': 16,
 576         'info_dict': {
 577             'id': 'schitts-creek/s06',
 578             'title': 'Season 6',
 579             'description': 'md5:6a92104a56cbeb5818cc47884d4326a2',
 580             'series': 'Schitt\'s Creek',
 581             'season_number': 6,
 582             'season': 'Season 6',
 583             'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/season/perso/cbc_schitts_creek_season_06_carousel_v03.jpg?impolicy=ott&im=Resize=(_Size_)&quality=75',
 584         },
 585     }, {
 586         'url': 'https://gem.cbc.ca/schitts-creek/s06',
 587         'only_matching': True,
 588     }]
 589     _API_BASE = 'https://services.radio-canada.ca/ott/cbc-api/v2/shows/'
 590
 591     def _real_extract(self, url):
 592         match = self._match_valid_url(url)
 593         season_id = match.group('id')
 594         show = match.group('show')
 595         show_info = self._download_json(self._API_BASE + show, season_id, expected_status=426)
 596         season = int(match.group('season'))
 597
 598         season_info = next((s for s in show_info['seasons'] if s.get('season') == season), None)
 599
 600         if season_info is None:
 601             raise ExtractorError(f'Couldn\'t find season {season} of {show}')
 602
 603         episodes = []
 604         for episode in season_info['assets']:
 605             episodes.append({
 606                 '_type': 'url_transparent',
 607                 'ie_key': 'CBCGem',
 608                 'url': 'https://gem.cbc.ca/media/' + episode['id'],
 609                 'id': episode['id'],
 610                 'title': episode.get('title'),
 611                 'description': episode.get('description'),
 612                 'thumbnail': episode.get('image'),
 613                 'series': episode.get('series'),
 614                 'season_number': episode.get('season'),
 615                 'season': season_info['title'],
 616                 'season_id': season_info.get('id'),
 617                 'episode_number': episode.get('episode'),
 618                 'episode': episode.get('title'),
 619                 'episode_id': episode['id'],
 620                 'duration': episode.get('duration'),
 621                 'categories': [episode.get('category')],
 622             })
 623
 624         thumbnail = None
 625         tn_uri = season_info.get('image')
 626         # the-national was observed to use a "data:image/png;base64"
 627         # URI for their 'image' value. The image was 1x1, and is
 628         # probably just a placeholder, so it is ignored.
 629         if tn_uri is not None and not tn_uri.startswith('data:'):
 630             thumbnail = tn_uri
 631
 632         return {
 633             '_type': 'playlist',
 634             'entries': episodes,
 635             'id': season_id,
 636             'title': season_info['title'],
 637             'description': season_info.get('description'),
 638             'thumbnail': thumbnail,
 639             'series': show_info.get('title'),
 640             'season_number': season_info.get('season'),
 641             'season': season_info['title'],
 642         }
 643
 644
 645 class CBCGemLiveIE(InfoExtractor):
 646     IE_NAME = 'gem.cbc.ca:live'
 647     _VALID_URL = r'https?://gem\.cbc\.ca/live(?:-event)?/(?P<id>\d+)'
 648     _TESTS = [
 649         {
 650             'url': 'https://gem.cbc.ca/live/920604739687',
 651             'info_dict': {
 652                 'title': 'Ottawa',
 653                 'description': 'The live TV channel and local programming from Ottawa',
 654                 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/CBC_OTT_VMS/Live_Channel_Static_Images/Ottawa_2880x1620.jpg',
 655                 'is_live': True,
 656                 'id': 'AyqZwxRqh8EH',
 657                 'ext': 'mp4',
 658                 'timestamp': 1492106160,
 659                 'upload_date': '20170413',
 660                 'uploader': 'CBCC-NEW',
 661             },
 662             'skip': 'Live might have ended',
 663         },
 664         {
 665             'url': 'https://gem.cbc.ca/live/44',
 666             'info_dict': {
 667                 'id': '44',
 668                 'ext': 'mp4',
 669                 'is_live': True,
 670                 'title': r're:^Ottawa [0-9\-: ]+',
 671                 'description': 'The live TV channel and local programming from Ottawa',
 672                 'live_status': 'is_live',
 673                 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*'
 674             },
 675             'params': {'skip_download': True},
 676             'skip': 'Live might have ended',
 677         },
 678         {
 679             'url': 'https://gem.cbc.ca/live-event/10835',
 680             'info_dict': {
 681                 'id': '10835',
 682                 'ext': 'mp4',
 683                 'is_live': True,
 684                 'title': r're:^The National \| Biden’s trip wraps up, Paltrow testifies, Bird flu [0-9\-: ]+',
 685                 'description': 'March 24, 2023 | President Biden’s Ottawa visit ends with big pledges from both countries. Plus, Gwyneth Paltrow testifies in her ski collision trial.',
 686                 'live_status': 'is_live',
 687                 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*',
 688                 'timestamp': 1679706000,
 689                 'upload_date': '20230325',
 690             },
 691             'params': {'skip_download': True},
 692             'skip': 'Live might have ended',
 693         }
 694     ]
 695
 696     def _real_extract(self, url):
 697         video_id = self._match_id(url)
 698         webpage = self._download_webpage(url, video_id)
 699         video_info = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['data']
 700
 701         # Two types of metadata JSON
 702         if not video_info.get('formattedIdMedia'):
 703             video_info = traverse_obj(
 704                 video_info, (('freeTv', ('streams', ...)), 'items', lambda _, v: v['key'] == video_id, {dict}),
 705                 get_all=False, default={})
 706
 707         video_stream_id = video_info.get('formattedIdMedia')
 708         if not video_stream_id:
 709             raise ExtractorError('Couldn\'t find video metadata, maybe this livestream is now offline', expected=True)
 710
 711         stream_data = self._download_json(
 712             'https://services.radio-canada.ca/media/validation/v2/', video_id, query={
 713                 'appCode': 'mpx',
 714                 'connectionType': 'hd',
 715                 'deviceType': 'ipad',
 716                 'idMedia': video_stream_id,
 717                 'multibitrate': 'true',
 718                 'output': 'json',
 719                 'tech': 'hls',
 720                 'manifestType': 'desktop',
 721             })
 722
 723         return {
 724             'id': video_id,
 725             'formats': self._extract_m3u8_formats(stream_data['url'], video_id, 'mp4', live=True),
 726             'is_live': True,
 727             **traverse_obj(video_info, {
 728                 'title': 'title',
 729                 'description': 'description',
 730                 'thumbnail': ('images', 'card', 'url'),
 731                 'timestamp': ('airDate', {parse_iso8601}),
 732             })
 733         }