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