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/|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': 'cbcplayer:1.7159484',
 282         'only_matching': True,
 283     }, {
 284         'url': 'cbcplayer:2164402062',
 285         'only_matching': True,
 286     }, {
 287         'url': 'http://www.cbc.ca/player/play/2657631896',
 288         'only_matching': True,
 289     }]
 290
 291     def _real_extract(self, url):
 292         video_id = self._match_id(url)
 293         if '.' in video_id:
 294             webpage = self._download_webpage(f'https://www.cbc.ca/player/play/{video_id}', video_id)
 295             video_id = self._search_json(
 296                 r'window\.__INITIAL_STATE__\s*=', webpage,
 297                 'initial state', video_id)['video']['currentClip']['mediaId']
 298
 299         return {
 300             '_type': 'url_transparent',
 301             'ie_key': 'ThePlatform',
 302             'url': smuggle_url(
 303                 'http://link.theplatform.com/s/ExhSPC/media/guid/2655402169/%s?mbr=true&formats=MPEG4,FLV,MP3' % video_id, {
 304                     'force_smil_url': True
 305                 }),
 306             'id': video_id,
 307             '_format_sort_fields': ('res', 'proto')  # Prioritize direct http formats over HLS
 308         }
 309
 310
 311 class CBCPlayerPlaylistIE(InfoExtractor):
 312     IE_NAME = 'cbc.ca:player:playlist'
 313     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?:player/)(?!play/)(?P<id>[^?#]+)'
 314     _TESTS = [{
 315         'url': 'https://www.cbc.ca/player/news/TV%20Shows/The%20National/Latest%20Broadcast',
 316         'playlist_mincount': 25,
 317         'info_dict': {
 318             'id': 'news/tv shows/the national/latest broadcast',
 319         }
 320     }, {
 321         'url': 'https://www.cbc.ca/player/news/Canada/North',
 322         'playlist_mincount': 25,
 323         'info_dict': {
 324             'id': 'news/canada/north',
 325         }
 326     }]
 327
 328     def _real_extract(self, url):
 329         playlist_id = urllib.parse.unquote(self._match_id(url)).lower()
 330         webpage = self._download_webpage(url, playlist_id)
 331         json_content = self._search_json(
 332             r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', playlist_id)
 333
 334         def entries():
 335             for video_id in traverse_obj(json_content, (
 336                 'video', 'clipsByCategory', lambda k, _: k.lower() == playlist_id, 'items', ..., 'id'
 337             )):
 338                 yield self.url_result(f'https://www.cbc.ca/player/play/{video_id}', CBCPlayerIE)
 339
 340         return self.playlist_result(entries(), playlist_id)
 341
 342
 343 class CBCGemIE(InfoExtractor):
 344     IE_NAME = 'gem.cbc.ca'
 345     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>[0-9a-z-]+/s[0-9]+[a-z][0-9]+)'
 346     _TESTS = [{
 347         # This is a normal, public, TV show video
 348         'url': 'https://gem.cbc.ca/media/schitts-creek/s06e01',
 349         'md5': '93dbb31c74a8e45b378cf13bd3f6f11e',
 350         'info_dict': {
 351             'id': 'schitts-creek/s06e01',
 352             'ext': 'mp4',
 353             'title': 'Smoke Signals',
 354             'description': 'md5:929868d20021c924020641769eb3e7f1',
 355             'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/episode/perso/cbc_schitts_creek_season_06e01_thumbnail_v01.jpg?im=Resize=(Size)',
 356             'duration': 1314,
 357             'categories': ['comedy'],
 358             'series': 'Schitt\'s Creek',
 359             'season': 'Season 6',
 360             'season_number': 6,
 361             'episode': 'Smoke Signals',
 362             'episode_number': 1,
 363             'episode_id': 'schitts-creek/s06e01',
 364         },
 365         'params': {'format': 'bv'},
 366         'skip': 'Geo-restricted to Canada',
 367     }, {
 368         # This video requires an account in the browser, but works fine in yt-dlp
 369         'url': 'https://gem.cbc.ca/media/schitts-creek/s01e01',
 370         'md5': '297a9600f554f2258aed01514226a697',
 371         'info_dict': {
 372             'id': 'schitts-creek/s01e01',
 373             'ext': 'mp4',
 374             'title': 'The Cup Runneth Over',
 375             'description': 'md5:9bca14ea49ab808097530eb05a29e797',
 376             'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/episode/perso/cbc_schitts_creek_season_01e01_thumbnail_v01.jpg?im=Resize=(Size)',
 377             'series': 'Schitt\'s Creek',
 378             'season_number': 1,
 379             'season': 'Season 1',
 380             'episode_number': 1,
 381             'episode': 'The Cup Runneth Over',
 382             'episode_id': 'schitts-creek/s01e01',
 383             'duration': 1309,
 384             'categories': ['comedy'],
 385         },
 386         'params': {'format': 'bv'},
 387         'skip': 'Geo-restricted to Canada',
 388     }, {
 389         'url': 'https://gem.cbc.ca/nadiyas-family-favourites/s01e01',
 390         'only_matching': True,
 391     }]
 392
 393     _GEO_COUNTRIES = ['CA']
 394     _TOKEN_API_KEY = '3f4beddd-2061-49b0-ae80-6f1f2ed65b37'
 395     _NETRC_MACHINE = 'cbcgem'
 396     _claims_token = None
 397
 398     def _new_claims_token(self, email, password):
 399         data = json.dumps({
 400             'email': email,
 401             'password': password,
 402         }).encode()
 403         headers = {'content-type': 'application/json'}
 404         query = {'apikey': self._TOKEN_API_KEY}
 405         resp = self._download_json('https://api.loginradius.com/identity/v2/auth/login',
 406                                    None, data=data, headers=headers, query=query)
 407         access_token = resp['access_token']
 408
 409         query = {
 410             'access_token': access_token,
 411             'apikey': self._TOKEN_API_KEY,
 412             'jwtapp': 'jwt',
 413         }
 414         resp = self._download_json('https://cloud-api.loginradius.com/sso/jwt/api/token',
 415                                    None, headers=headers, query=query)
 416         sig = resp['signature']
 417
 418         data = json.dumps({'jwt': sig}).encode()
 419         headers = {'content-type': 'application/json', 'ott-device-type': 'web'}
 420         resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/token',
 421                                    None, data=data, headers=headers, expected_status=426)
 422         cbc_access_token = resp['accessToken']
 423
 424         headers = {'content-type': 'application/json', 'ott-device-type': 'web', 'ott-access-token': cbc_access_token}
 425         resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/profile',
 426                                    None, headers=headers, expected_status=426)
 427         return resp['claimsToken']
 428
 429     def _get_claims_token_expiry(self):
 430         # Token is a JWT
 431         # JWT is decoded here and 'exp' field is extracted
 432         # It is a Unix timestamp for when the token expires
 433         b64_data = self._claims_token.split('.')[1]
 434         data = base64.urlsafe_b64decode(b64_data + "==")
 435         return json.loads(data)['exp']
 436
 437     def claims_token_expired(self):
 438         exp = self._get_claims_token_expiry()
 439         if exp - time.time() < 10:
 440             # It will expire in less than 10 seconds, or has already expired
 441             return True
 442         return False
 443
 444     def claims_token_valid(self):
 445         return self._claims_token is not None and not self.claims_token_expired()
 446
 447     def _get_claims_token(self, email, password):
 448         if not self.claims_token_valid():
 449             self._claims_token = self._new_claims_token(email, password)
 450             self.cache.store(self._NETRC_MACHINE, 'claims_token', self._claims_token)
 451         return self._claims_token
 452
 453     def _real_initialize(self):
 454         if self.claims_token_valid():
 455             return
 456         self._claims_token = self.cache.load(self._NETRC_MACHINE, 'claims_token')
 457
 458     def _find_secret_formats(self, formats, video_id):
 459         """ Find a valid video url and convert it to the secret variant """
 460         base_format = next((f for f in formats if f.get('vcodec') != 'none'), None)
 461         if not base_format:
 462             return
 463
 464         base_url = re.sub(r'(Manifest\(.*?),filter=[\w-]+(.*?\))', r'\1\2', base_format['url'])
 465         url = re.sub(r'(Manifest\(.*?),format=[\w-]+(.*?\))', r'\1\2', base_url)
 466
 467         secret_xml = self._download_xml(url, video_id, note='Downloading secret XML', fatal=False)
 468         if not isinstance(secret_xml, xml.etree.ElementTree.Element):
 469             return
 470
 471         for child in secret_xml:
 472             if child.attrib.get('Type') != 'video':
 473                 continue
 474             for video_quality in child:
 475                 bitrate = int_or_none(video_quality.attrib.get('Bitrate'))
 476                 if not bitrate or 'Index' not in video_quality.attrib:
 477                     continue
 478                 height = int_or_none(video_quality.attrib.get('MaxHeight'))
 479
 480                 yield {
 481                     **base_format,
 482                     'format_id': join_nonempty('sec', height),
 483                     # Note: \g<1> is necessary instead of \1 since bitrate is a number
 484                     'url': re.sub(r'(QualityLevels\()\d+(\))', fr'\g<1>{bitrate}\2', base_url),
 485                     'width': int_or_none(video_quality.attrib.get('MaxWidth')),
 486                     'tbr': bitrate / 1000.0,
 487                     'height': height,
 488                 }
 489
 490     def _real_extract(self, url):
 491         video_id = self._match_id(url)
 492         video_info = self._download_json(
 493             f'https://services.radio-canada.ca/ott/cbc-api/v2/assets/{video_id}',
 494             video_id, expected_status=426)
 495
 496         email, password = self._get_login_info()
 497         if email and password:
 498             claims_token = self._get_claims_token(email, password)
 499             headers = {'x-claims-token': claims_token}
 500         else:
 501             headers = {}
 502         m3u8_info = self._download_json(video_info['playSession']['url'], video_id, headers=headers)
 503         m3u8_url = m3u8_info.get('url')
 504
 505         if m3u8_info.get('errorCode') == 1:
 506             self.raise_geo_restricted(countries=['CA'])
 507         elif m3u8_info.get('errorCode') == 35:
 508             self.raise_login_required(method='password')
 509         elif m3u8_info.get('errorCode') != 0:
 510             raise ExtractorError(f'{self.IE_NAME} said: {m3u8_info.get("errorCode")} - {m3u8_info.get("message")}')
 511
 512         formats = self._extract_m3u8_formats(m3u8_url, video_id, m3u8_id='hls')
 513         self._remove_duplicate_formats(formats)
 514         formats.extend(self._find_secret_formats(formats, video_id))
 515
 516         for format in formats:
 517             if format.get('vcodec') == 'none':
 518                 if format.get('ext') is None:
 519                     format['ext'] = 'm4a'
 520                 if format.get('acodec') is None:
 521                     format['acodec'] = 'mp4a.40.2'
 522
 523                 # Put described audio at the beginning of the list, so that it
 524                 # isn't chosen by default, as most people won't want it.
 525                 if 'descriptive' in format['format_id'].lower():
 526                     format['preference'] = -2
 527
 528         return {
 529             'id': video_id,
 530             'title': video_info['title'],
 531             'description': video_info.get('description'),
 532             'thumbnail': video_info.get('image'),
 533             'series': video_info.get('series'),
 534             'season_number': video_info.get('season'),
 535             'season': f'Season {video_info.get("season")}',
 536             'episode_number': video_info.get('episode'),
 537             'episode': video_info.get('title'),
 538             'episode_id': video_id,
 539             'duration': video_info.get('duration'),
 540             'categories': [video_info.get('category')],
 541             'formats': formats,
 542             'release_timestamp': video_info.get('airDate'),
 543             'timestamp': video_info.get('availableDate'),
 544         }
 545
 546
 547 class CBCGemPlaylistIE(InfoExtractor):
 548     IE_NAME = 'gem.cbc.ca:playlist'
 549     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>(?P<show>[0-9a-z-]+)/s(?P<season>[0-9]+))/?(?:[?#]|$)'
 550     _TESTS = [{
 551         # TV show playlist, all public videos
 552         'url': 'https://gem.cbc.ca/media/schitts-creek/s06',
 553         'playlist_count': 16,
 554         'info_dict': {
 555             'id': 'schitts-creek/s06',
 556             'title': 'Season 6',
 557             'description': 'md5:6a92104a56cbeb5818cc47884d4326a2',
 558             'series': 'Schitt\'s Creek',
 559             'season_number': 6,
 560             'season': 'Season 6',
 561             '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',
 562         },
 563     }, {
 564         'url': 'https://gem.cbc.ca/schitts-creek/s06',
 565         'only_matching': True,
 566     }]
 567     _API_BASE = 'https://services.radio-canada.ca/ott/cbc-api/v2/shows/'
 568
 569     def _real_extract(self, url):
 570         match = self._match_valid_url(url)
 571         season_id = match.group('id')
 572         show = match.group('show')
 573         show_info = self._download_json(self._API_BASE + show, season_id, expected_status=426)
 574         season = int(match.group('season'))
 575
 576         season_info = next((s for s in show_info['seasons'] if s.get('season') == season), None)
 577
 578         if season_info is None:
 579             raise ExtractorError(f'Couldn\'t find season {season} of {show}')
 580
 581         episodes = []
 582         for episode in season_info['assets']:
 583             episodes.append({
 584                 '_type': 'url_transparent',
 585                 'ie_key': 'CBCGem',
 586                 'url': 'https://gem.cbc.ca/media/' + episode['id'],
 587                 'id': episode['id'],
 588                 'title': episode.get('title'),
 589                 'description': episode.get('description'),
 590                 'thumbnail': episode.get('image'),
 591                 'series': episode.get('series'),
 592                 'season_number': episode.get('season'),
 593                 'season': season_info['title'],
 594                 'season_id': season_info.get('id'),
 595                 'episode_number': episode.get('episode'),
 596                 'episode': episode.get('title'),
 597                 'episode_id': episode['id'],
 598                 'duration': episode.get('duration'),
 599                 'categories': [episode.get('category')],
 600             })
 601
 602         thumbnail = None
 603         tn_uri = season_info.get('image')
 604         # the-national was observed to use a "data:image/png;base64"
 605         # URI for their 'image' value. The image was 1x1, and is
 606         # probably just a placeholder, so it is ignored.
 607         if tn_uri is not None and not tn_uri.startswith('data:'):
 608             thumbnail = tn_uri
 609
 610         return {
 611             '_type': 'playlist',
 612             'entries': episodes,
 613             'id': season_id,
 614             'title': season_info['title'],
 615             'description': season_info.get('description'),
 616             'thumbnail': thumbnail,
 617             'series': show_info.get('title'),
 618             'season_number': season_info.get('season'),
 619             'season': season_info['title'],
 620         }
 621
 622
 623 class CBCGemLiveIE(InfoExtractor):
 624     IE_NAME = 'gem.cbc.ca:live'
 625     _VALID_URL = r'https?://gem\.cbc\.ca/live(?:-event)?/(?P<id>\d+)'
 626     _TESTS = [
 627         {
 628             'url': 'https://gem.cbc.ca/live/920604739687',
 629             'info_dict': {
 630                 'title': 'Ottawa',
 631                 'description': 'The live TV channel and local programming from Ottawa',
 632                 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/CBC_OTT_VMS/Live_Channel_Static_Images/Ottawa_2880x1620.jpg',
 633                 'is_live': True,
 634                 'id': 'AyqZwxRqh8EH',
 635                 'ext': 'mp4',
 636                 'timestamp': 1492106160,
 637                 'upload_date': '20170413',
 638                 'uploader': 'CBCC-NEW',
 639             },
 640             'skip': 'Live might have ended',
 641         },
 642         {
 643             'url': 'https://gem.cbc.ca/live/44',
 644             'info_dict': {
 645                 'id': '44',
 646                 'ext': 'mp4',
 647                 'is_live': True,
 648                 'title': r're:^Ottawa [0-9\-: ]+',
 649                 'description': 'The live TV channel and local programming from Ottawa',
 650                 'live_status': 'is_live',
 651                 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*'
 652             },
 653             'params': {'skip_download': True},
 654             'skip': 'Live might have ended',
 655         },
 656         {
 657             'url': 'https://gem.cbc.ca/live-event/10835',
 658             'info_dict': {
 659                 'id': '10835',
 660                 'ext': 'mp4',
 661                 'is_live': True,
 662                 'title': r're:^The National \| Biden’s trip wraps up, Paltrow testifies, Bird flu [0-9\-: ]+',
 663                 '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.',
 664                 'live_status': 'is_live',
 665                 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*',
 666                 'timestamp': 1679706000,
 667                 'upload_date': '20230325',
 668             },
 669             'params': {'skip_download': True},
 670             'skip': 'Live might have ended',
 671         }
 672     ]
 673
 674     def _real_extract(self, url):
 675         video_id = self._match_id(url)
 676         webpage = self._download_webpage(url, video_id)
 677         video_info = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['data']
 678
 679         # Two types of metadata JSON
 680         if not video_info.get('formattedIdMedia'):
 681             video_info = traverse_obj(
 682                 video_info, (('freeTv', ('streams', ...)), 'items', lambda _, v: v['key'] == video_id, {dict}),
 683                 get_all=False, default={})
 684
 685         video_stream_id = video_info.get('formattedIdMedia')
 686         if not video_stream_id:
 687             raise ExtractorError('Couldn\'t find video metadata, maybe this livestream is now offline', expected=True)
 688
 689         stream_data = self._download_json(
 690             'https://services.radio-canada.ca/media/validation/v2/', video_id, query={
 691                 'appCode': 'mpx',
 692                 'connectionType': 'hd',
 693                 'deviceType': 'ipad',
 694                 'idMedia': video_stream_id,
 695                 'multibitrate': 'true',
 696                 'output': 'json',
 697                 'tech': 'hls',
 698                 'manifestType': 'desktop',
 699             })
 700
 701         return {
 702             'id': video_id,
 703             'formats': self._extract_m3u8_formats(stream_data['url'], video_id, 'mp4', live=True),
 704             'is_live': True,
 705             **traverse_obj(video_info, {
 706                 'title': 'title',
 707                 'description': 'description',
 708                 'thumbnail': ('images', 'card', 'url'),
 709                 'timestamp': ('airDate', {parse_iso8601}),
 710             })
 711         }