yt_dlp/extractor/cbc.py

   1 import base64
   2 import functools
   3 import json
   4 import re
   5 import time
   6 import urllib.parse
   7
   8 from .common import InfoExtractor
   9 from ..networking import HEADRequest
  10 from ..utils import (
  11     ExtractorError,
  12     float_or_none,
  13     int_or_none,
  14     js_to_json,
  15     mimetype2ext,
  16     orderedSet,
  17     parse_iso8601,
  18     replace_extension,
  19     smuggle_url,
  20     strip_or_none,
  21     traverse_obj,
  22     try_get,
  23     update_url,
  24     url_basename,
  25     url_or_none,
  26 )
  27
  28
  29 class CBCIE(InfoExtractor):
  30     IE_NAME = 'cbc.ca'
  31     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P<id>[^/?#]+)'
  32     _TESTS = [{
  33         # with mediaId
  34         'url': 'http://www.cbc.ca/22minutes/videos/clips-season-23/don-cherry-play-offs',
  35         'md5': '97e24d09672fc4cf56256d6faa6c25bc',
  36         'info_dict': {
  37             'id': '2682904050',
  38             'ext': 'mp4',
  39             'title': 'Don Cherry – All-Stars',
  40             'description': 'Don Cherry has a bee in his bonnet about AHL player John Scott because that guy’s got heart.',
  41             'timestamp': 1454463000,
  42             'upload_date': '20160203',
  43             'uploader': 'CBCC-NEW',
  44         },
  45         'skip': 'Geo-restricted to Canada',
  46     }, {
  47         # with clipId, feed available via tpfeed.cbc.ca and feed.theplatform.com
  48         'url': 'http://www.cbc.ca/22minutes/videos/22-minutes-update/22-minutes-update-episode-4',
  49         'md5': '162adfa070274b144f4fdc3c3b8207db',
  50         'info_dict': {
  51             'id': '2414435309',
  52             'ext': 'mp4',
  53             'title': '22 Minutes Update: What Not To Wear Quebec',
  54             'description': "This week's latest Canadian top political story is What Not To Wear Quebec.",
  55             'upload_date': '20131025',
  56             'uploader': 'CBCC-NEW',
  57             'timestamp': 1382717907,
  58         },
  59         'skip': 'No longer available',
  60     }, {
  61         # with clipId, feed only available via tpfeed.cbc.ca
  62         'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
  63         'md5': '0274a90b51a9b4971fe005c63f592f12',
  64         'info_dict': {
  65             'id': '2487345465',
  66             'ext': 'mp4',
  67             'title': 'Robin Williams freestyles on 90 Minutes Live',
  68             'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
  69             'upload_date': '19780210',
  70             'uploader': 'CBCC-NEW',
  71             'timestamp': 255977160,
  72         },
  73         'skip': '404 Not Found',
  74     }, {
  75         # multiple iframes
  76         'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot',
  77         'playlist': [{
  78             'md5': '377572d0b49c4ce0c9ad77470e0b96b4',
  79             'info_dict': {
  80                 'id': '2680832926',
  81                 'ext': 'mp4',
  82                 'title': 'An Eagle\'s-Eye View Off Burrard Bridge',
  83                 'description': 'Hercules the eagle flies from Vancouver\'s Burrard Bridge down to a nearby park with a mini-camera strapped to his back.',
  84                 'upload_date': '20160201',
  85                 'timestamp': 1454342820,
  86                 'uploader': 'CBCC-NEW',
  87             },
  88         }, {
  89             'md5': '415a0e3f586113894174dfb31aa5bb1a',
  90             'info_dict': {
  91                 'id': '2658915080',
  92                 'ext': 'mp4',
  93                 'title': 'Fly like an eagle!',
  94                 'description': 'Eagle equipped with a mini camera flies from the world\'s tallest tower',
  95                 'upload_date': '20150315',
  96                 'timestamp': 1426443984,
  97                 'uploader': 'CBCC-NEW',
  98             },
  99         }],
 100         'skip': 'Geo-restricted to Canada',
 101     }, {
 102         # multiple CBC.APP.Caffeine.initInstance(...)
 103         'url': 'http://www.cbc.ca/news/canada/calgary/dog-indoor-exercise-winter-1.3928238',
 104         'info_dict': {
 105             'title': 'Keep Rover active during the deep freeze with doggie pushups and other fun indoor tasks',  # FIXME: actual title includes " | CBC News"
 106             'id': 'dog-indoor-exercise-winter-1.3928238',
 107             'description': 'md5:c18552e41726ee95bd75210d1ca9194c',
 108         },
 109         'playlist_mincount': 6,
 110     }]
 111
 112     @classmethod
 113     def suitable(cls, url):
 114         return False if CBCPlayerIE.suitable(url) else super().suitable(url)
 115
 116     def _extract_player_init(self, player_init, display_id):
 117         player_info = self._parse_json(player_init, display_id, js_to_json)
 118         media_id = player_info.get('mediaId')
 119         if not media_id:
 120             clip_id = player_info['clipId']
 121             feed = self._download_json(
 122                 f'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={{:mpsReleases}}{{{clip_id}}}',
 123                 clip_id, fatal=False)
 124             if feed:
 125                 media_id = try_get(feed, lambda x: x['entries'][0]['guid'], str)
 126             if not media_id:
 127                 media_id = self._download_json(
 128                     'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
 129                     clip_id)['entries'][0]['id'].split('/')[-1]
 130         return self.url_result(f'cbcplayer:{media_id}', 'CBCPlayer', media_id)
 131
 132     def _real_extract(self, url):
 133         display_id = self._match_id(url)
 134         webpage = self._download_webpage(url, display_id)
 135         title = (self._og_search_title(webpage, default=None)
 136                  or self._html_search_meta('twitter:title', webpage, 'title', default=None)
 137                  or self._html_extract_title(webpage))
 138         entries = [
 139             self._extract_player_init(player_init, display_id)
 140             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
 141         media_ids = []
 142         for media_id_re in (
 143                 r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"',
 144                 r'<div[^>]+\bid=["\']player-(\d+)',
 145                 r'guid["\']\s*:\s*["\'](\d+)'):
 146             media_ids.extend(re.findall(media_id_re, webpage))
 147         entries.extend([
 148             self.url_result(f'cbcplayer:{media_id}', 'CBCPlayer', media_id)
 149             for media_id in orderedSet(media_ids)])
 150         return self.playlist_result(
 151             entries, display_id, strip_or_none(title),
 152             self._og_search_description(webpage))
 153
 154
 155 class CBCPlayerIE(InfoExtractor):
 156     IE_NAME = 'cbc.ca:player'
 157     _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/(?:video/)?|i/caffeine/syndicate/\?mediaId=))(?P<id>(?:\d\.)?\d+)'
 158     _GEO_COUNTRIES = ['CA']
 159     _TESTS = [{
 160         'url': 'http://www.cbc.ca/player/play/2683190193',
 161         'md5': '64d25f841ddf4ddb28a235338af32e2c',
 162         'info_dict': {
 163             'id': '2683190193',
 164             'ext': 'mp4',
 165             'title': 'Gerry Runs a Sweat Shop',
 166             'description': 'md5:b457e1c01e8ff408d9d801c1c2cd29b0',
 167             'timestamp': 1455071400,
 168             'upload_date': '20160210',
 169             'uploader': 'CBCC-NEW',
 170         },
 171         'skip': 'Geo-restricted to Canada and no longer available',
 172     }, {
 173         'url': 'http://www.cbc.ca/i/caffeine/syndicate/?mediaId=2657631896',
 174         'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
 175         'info_dict': {
 176             'id': '2657631896',
 177             'ext': 'mp3',
 178             'title': 'CBC Montreal is organizing its first ever community hackathon!',
 179             'description': 'md5:dd3b692f0a139b0369943150bd1c46a9',
 180             'timestamp': 1425704400,
 181             'upload_date': '20150307',
 182             'thumbnail': 'https://i.cbc.ca/ais/1.2985700,1717262248558/full/max/0/default.jpg',
 183             'chapters': [],
 184             'duration': 494.811,
 185             'categories': ['All in a Weekend Montreal'],
 186             'tags': 'count:11',
 187             'location': 'Quebec',
 188             'series': 'All in a Weekend Montreal',
 189             'season': 'Season 2015',
 190             'season_number': 2015,
 191             'media_type': 'Excerpt',
 192             'genres': ['Other'],
 193         },
 194     }, {
 195         'url': 'http://www.cbc.ca/i/caffeine/syndicate/?mediaId=2164402062',
 196         'info_dict': {
 197             'id': '2164402062',
 198             'ext': 'mp4',
 199             'title': 'Cancer survivor four times over',
 200             'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
 201             'timestamp': 1320410746,
 202             'upload_date': '20111104',
 203             'thumbnail': 'https://i.cbc.ca/ais/1.1711287,1717139372111/full/max/0/default.jpg',
 204             'chapters': [],
 205             'duration': 186.867,
 206             'series': 'CBC News: Windsor at 6:00',
 207             'categories': ['Windsor'],
 208             'location': 'Windsor',
 209             'tags': ['Cancer', 'News/Canada/Windsor', 'Windsor'],
 210             'media_type': 'Excerpt',
 211             'genres': ['News'],
 212         },
 213         'params': {'skip_download': 'm3u8'},
 214     }, {
 215         # Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/
 216         'url': 'https://www.cbc.ca/player/play/1.2985700',
 217         'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
 218         'info_dict': {
 219             'id': '1.2985700',
 220             'ext': 'mp3',
 221             'title': 'CBC Montreal is organizing its first ever community hackathon!',
 222             '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.',
 223             'timestamp': 1425704400,
 224             'upload_date': '20150307',
 225             'thumbnail': 'https://i.cbc.ca/ais/1.2985700,1717262248558/full/max/0/default.jpg',
 226             'chapters': [],
 227             'duration': 494.811,
 228             'categories': ['All in a Weekend Montreal'],
 229             'tags': 'count:11',
 230             'location': 'Quebec',
 231             'series': 'All in a Weekend Montreal',
 232             'season': 'Season 2015',
 233             'season_number': 2015,
 234             'media_type': 'Excerpt',
 235             'genres': ['Other'],
 236         },
 237     }, {
 238         'url': 'https://www.cbc.ca/player/play/1.1711287',
 239         'info_dict': {
 240             'id': '1.1711287',
 241             'ext': 'mp4',
 242             'title': 'Cancer survivor four times over',
 243             'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
 244             'timestamp': 1320410746,
 245             'upload_date': '20111104',
 246             'thumbnail': 'https://i.cbc.ca/ais/1.1711287,1717139372111/full/max/0/default.jpg',
 247             'chapters': [],
 248             'duration': 186.867,
 249             'series': 'CBC News: Windsor at 6:00',
 250             'categories': ['Windsor'],
 251             'location': 'Windsor',
 252             'tags': ['Cancer', 'News/Canada/Windsor', 'Windsor'],
 253             'media_type': 'Excerpt',
 254             'genres': ['News'],
 255         },
 256         'params': {'skip_download': 'm3u8'},
 257     }, {
 258         # Has subtitles
 259         # These broadcasts expire after ~1 month, can find new test URL here:
 260         # https://www.cbc.ca/player/news/TV%20Shows/The%20National/Latest%20Broadcast
 261         'url': 'https://www.cbc.ca/player/play/video/9.6424403',
 262         'md5': '8025909eaffcf0adf59922904def9a5e',
 263         'info_dict': {
 264             'id': '9.6424403',
 265             'ext': 'mp4',
 266             'title': 'The National | N.W.T. wildfire emergency',
 267             'description': 'md5:ada33d36d1df69347ed575905bfd496c',
 268             'timestamp': 1718589600,
 269             'duration': 2692.833,
 270             'subtitles': {
 271                 'en-US': [{
 272                     'name': 'English Captions',
 273                     'url': 'https://cbchls.akamaized.net/delivery/news-shows/2024/06/17/NAT_JUN16-00-55-00/NAT_JUN16_cc.vtt',
 274                 }],
 275             },
 276             'thumbnail': 'https://i.cbc.ca/ais/6272b5c6-5e78-4c05-915d-0e36672e33d1,1714756287822/full/max/0/default.jpg',
 277             'chapters': 'count:5',
 278             'upload_date': '20240617',
 279             'categories': ['News', 'The National', 'The National Latest Broadcasts'],
 280             'series': 'The National - Full Show',
 281             'tags': ['The National'],
 282             'location': 'Canada',
 283             'media_type': 'Full Program',
 284             'genres': ['News'],
 285         },
 286     }, {
 287         'url': 'https://www.cbc.ca/player/play/video/1.7194274',
 288         'md5': '188b96cf6bdcb2540e178a6caa957128',
 289         'info_dict': {
 290             'id': '1.7194274',
 291             'ext': 'mp4',
 292             'title': '#TheMoment a rare white spirit moose was spotted in Alberta',
 293             'description': 'md5:18ae269a2d0265c5b0bbe4b2e1ac61a3',
 294             'timestamp': 1714788791,
 295             'duration': 77.678,
 296             'subtitles': {'eng': [{'ext': 'vtt', 'protocol': 'm3u8_native'}]},
 297             'thumbnail': 'https://i.cbc.ca/ais/1.7194274,1717224990425/full/max/0/default.jpg',
 298             'chapters': [],
 299             'categories': 'count:3',
 300             'series': 'The National',
 301             'tags': 'count:17',
 302             'location': 'Canada',
 303             'media_type': 'Excerpt',
 304             'upload_date': '20240504',
 305             'genres': ['News'],
 306         },
 307     }, {
 308         'url': 'https://www.cbc.ca/player/play/video/9.6427282',
 309         'info_dict': {
 310             'id': '9.6427282',
 311             'ext': 'mp4',
 312             'title': 'Men\'s Soccer - Argentina vs Morocco',
 313             'description': 'Argentina faces Morocco on the football pitch at Saint Etienne Stadium.',
 314             'series': 'CBC Sports',
 315             'media_type': 'Event Coverage',
 316             'thumbnail': 'https://i.cbc.ca/ais/a4c5c0c2-99fa-4bd3-8061-5a63879c1b33,1718828053500/full/max/0/default.jpg',
 317             'timestamp': 1721825400.0,
 318             'upload_date': '20240724',
 319             'duration': 10568.0,
 320             'chapters': [],
 321             'genres': [],
 322             'tags': ['2024 Paris Olympic Games'],
 323             'categories': ['Olympics Summer Soccer', 'Summer Olympics Replays', 'Summer Olympics Soccer Replays'],
 324             'location': 'Canada',
 325         },
 326         'params': {'skip_download': 'm3u8'},
 327     }, {
 328         'url': 'https://www.cbc.ca/player/play/video/9.6459530',
 329         'md5': '6c1bb76693ab321a2e99c347a1d5ecbc',
 330         'info_dict': {
 331             'id': '9.6459530',
 332             'ext': 'mp4',
 333             'title': 'Parts of Jasper incinerated as wildfire rages',
 334             'description': 'md5:6f1caa8d128ad3f629257ef5fecf0962',
 335             'series': 'The National',
 336             'media_type': 'Excerpt',
 337             'thumbnail': 'https://i.cbc.ca/ais/507c0086-31a2-494d-96e4-bffb1048d045,1721953984375/full/max/0/default.jpg',
 338             'timestamp': 1721964091.012,
 339             'upload_date': '20240726',
 340             'duration': 952.285,
 341             'chapters': [],
 342             'genres': [],
 343             'tags': 'count:23',
 344             'categories': ['News (FAST)', 'News', 'The National', 'TV News Shows', 'The National '],
 345         },
 346     }, {
 347         'url': 'https://www.cbc.ca/player/play/video/9.6420651',
 348         'md5': '71a850c2c6ee5e912de169f5311bb533',
 349         'info_dict': {
 350             'id': '9.6420651',
 351             'ext': 'mp4',
 352             'title': 'Is it a breath of fresh air? Measuring air quality in Edmonton',
 353             'description': 'md5:3922b92cc8b69212d739bd9dd095b1c3',
 354             'series': 'CBC News Edmonton',
 355             'media_type': 'Excerpt',
 356             'thumbnail': 'https://i.cbc.ca/ais/73c4ab9c-7ad4-46ee-bb9b-020fdc01c745,1718214547576/full/max/0/default.jpg',
 357             'timestamp': 1718220065.768,
 358             'upload_date': '20240612',
 359             'duration': 286.086,
 360             'chapters': [],
 361             'genres': ['News'],
 362             'categories': ['News', 'Edmonton'],
 363             'tags': 'count:7',
 364             'location': 'Edmonton',
 365         },
 366     }, {
 367         'url': 'cbcplayer:1.7159484',
 368         'only_matching': True,
 369     }, {
 370         'url': 'cbcplayer:2164402062',
 371         'only_matching': True,
 372     }, {
 373         'url': 'http://www.cbc.ca/player/play/2657631896',
 374         'only_matching': True,
 375     }]
 376
 377     def _parse_param(self, asset_data, name):
 378         return traverse_obj(asset_data, ('params', lambda _, v: v['name'] == name, 'value', {str}, any))
 379
 380     def _real_extract(self, url):
 381         video_id = self._match_id(url)
 382         webpage = self._download_webpage(f'https://www.cbc.ca/player/play/{video_id}', video_id)
 383         data = self._search_json(
 384             r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', video_id)['video']['currentClip']
 385         assets = traverse_obj(
 386             data, ('media', 'assets', lambda _, v: url_or_none(v['key']) and v['type']))
 387
 388         if not assets and (media_id := traverse_obj(data, ('mediaId', {str}))):
 389             # XXX: Deprecated; CBC is migrating off of ThePlatform
 390             return {
 391                 '_type': 'url_transparent',
 392                 'ie_key': 'ThePlatform',
 393                 'url': smuggle_url(
 394                     f'http://link.theplatform.com/s/ExhSPC/media/guid/2655402169/{media_id}?mbr=true&formats=MPEG4,FLV,MP3', {
 395                         'force_smil_url': True,
 396                     }),
 397                 'id': media_id,
 398                 '_format_sort_fields': ('res', 'proto'),  # Prioritize direct http formats over HLS
 399             }
 400
 401         is_live = traverse_obj(data, ('media', 'streamType', {str})) == 'Live'
 402         formats, subtitles = [], {}
 403
 404         for sub in traverse_obj(data, ('media', 'textTracks', lambda _, v: url_or_none(v['src']))):
 405             subtitles.setdefault(sub.get('language') or 'und', []).append({
 406                 'url': sub['src'],
 407                 'name': sub.get('label'),
 408             })
 409
 410         for asset in assets:
 411             asset_key = asset['key']
 412             asset_type = asset['type']
 413             if asset_type != 'medianet':
 414                 self.report_warning(f'Skipping unsupported asset type "{asset_type}": {asset_key}')
 415                 continue
 416             asset_data = self._download_json(asset_key, video_id, f'Downloading {asset_type} JSON')
 417             ext = mimetype2ext(self._parse_param(asset_data, 'contentType'))
 418             if ext == 'm3u8':
 419                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
 420                     asset_data['url'], video_id, 'mp4', m3u8_id='hls', live=is_live)
 421                 formats.extend(fmts)
 422                 # Avoid slow/error-prone webvtt-over-m3u8 if direct https vtt is available
 423                 if not subtitles:
 424                     self._merge_subtitles(subs, target=subtitles)
 425                 if is_live or not fmts:
 426                     continue
 427                 # Check for direct https mp4 format
 428                 best_video_fmt = traverse_obj(fmts, (
 429                     lambda _, v: v.get('vcodec') != 'none' and v['tbr'], all,
 430                     {functools.partial(sorted, key=lambda x: x['tbr'])}, -1, {dict})) or {}
 431                 base_url = self._search_regex(
 432                     r'(https?://[^?#]+?/)hdntl=', best_video_fmt.get('url'), 'base url', default=None)
 433                 if not base_url or '/live/' in base_url:
 434                     continue
 435                 mp4_url = base_url + replace_extension(url_basename(best_video_fmt['url']), 'mp4')
 436                 if self._request_webpage(
 437                         HEADRequest(mp4_url), video_id, 'Checking for https format',
 438                         errnote=False, fatal=False):
 439                     formats.append({
 440                         **best_video_fmt,
 441                         'url': mp4_url,
 442                         'format_id': 'https-mp4',
 443                         'protocol': 'https',
 444                         'manifest_url': None,
 445                         'acodec': None,
 446                     })
 447             else:
 448                 formats.append({
 449                     'url': asset_data['url'],
 450                     'ext': ext,
 451                     'vcodec': 'none' if self._parse_param(asset_data, 'mediaType') == 'audio' else None,
 452                 })
 453
 454         chapters = traverse_obj(data, (
 455             'media', 'chapters', lambda _, v: float(v['startTime']) is not None, {
 456                 'start_time': ('startTime', {float_or_none(scale=1000)}),
 457                 'end_time': ('endTime', {float_or_none(scale=1000)}),
 458                 'title': ('name', {str}),
 459             }))
 460         # Filter out pointless single chapters with start_time==0 and no end_time
 461         if len(chapters) == 1 and not (chapters[0].get('start_time') or chapters[0].get('end_time')):
 462             chapters = []
 463
 464         return {
 465             **traverse_obj(data, {
 466                 'title': ('title', {str}),
 467                 'description': ('description', {str.strip}),
 468                 'thumbnail': ('image', 'url', {url_or_none}, {update_url(query=None)}),
 469                 'timestamp': ('publishedAt', {float_or_none(scale=1000)}),
 470                 'media_type': ('media', 'clipType', {str}),
 471                 'series': ('showName', {str}),
 472                 'season_number': ('media', 'season', {int_or_none}),
 473                 'duration': ('media', 'duration', {float_or_none}, {lambda x: None if is_live else x}),
 474                 'location': ('media', 'region', {str}),
 475                 'tags': ('tags', ..., 'name', {str}),
 476                 'genres': ('media', 'genre', all),
 477                 'categories': ('categories', ..., 'name', {str}),
 478             }),
 479             'id': video_id,
 480             'formats': formats,
 481             'subtitles': subtitles,
 482             'chapters': chapters,
 483             'is_live': is_live,
 484         }
 485
 486
 487 class CBCPlayerPlaylistIE(InfoExtractor):
 488     IE_NAME = 'cbc.ca:player:playlist'
 489     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?:player/)(?!play/)(?P<id>[^?#]+)'
 490     _TESTS = [{
 491         'url': 'https://www.cbc.ca/player/news/TV%20Shows/The%20National/Latest%20Broadcast',
 492         'playlist_mincount': 25,
 493         'info_dict': {
 494             'id': 'news/tv shows/the national/latest broadcast',
 495         },
 496     }, {
 497         'url': 'https://www.cbc.ca/player/news/Canada/North',
 498         'playlist_mincount': 25,
 499         'info_dict': {
 500             'id': 'news/canada/north',
 501         },
 502     }]
 503
 504     def _real_extract(self, url):
 505         playlist_id = urllib.parse.unquote(self._match_id(url)).lower()
 506         webpage = self._download_webpage(url, playlist_id)
 507         json_content = self._search_json(
 508             r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', playlist_id)
 509
 510         def entries():
 511             for video_id in traverse_obj(json_content, (
 512                 'video', 'clipsByCategory', lambda k, _: k.lower() == playlist_id, 'items', ..., 'id',
 513             )):
 514                 yield self.url_result(f'https://www.cbc.ca/player/play/{video_id}', CBCPlayerIE)
 515
 516         return self.playlist_result(entries(), playlist_id)
 517
 518
 519 class CBCGemIE(InfoExtractor):
 520     IE_NAME = 'gem.cbc.ca'
 521     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>[0-9a-z-]+/s[0-9]+[a-z][0-9]+)'
 522     _TESTS = [{
 523         # This is a normal, public, TV show video
 524         'url': 'https://gem.cbc.ca/media/schitts-creek/s06e01',
 525         'info_dict': {
 526             'id': 'schitts-creek/s06e01',
 527             'ext': 'mp4',
 528             'title': 'Smoke Signals',
 529             'description': 'md5:929868d20021c924020641769eb3e7f1',
 530             'thumbnail': r're:https://images\.radio-canada\.ca/[^#?]+/cbc_schitts_creek_season_06e01_thumbnail_v01\.jpg',
 531             'duration': 1324,
 532             'categories': ['comedy'],
 533             'series': 'Schitt\'s Creek',
 534             'season': 'Season 6',
 535             'season_number': 6,
 536             'episode': 'Smoke Signals',
 537             'episode_number': 1,
 538             'episode_id': 'schitts-creek/s06e01',
 539             'upload_date': '20210618',
 540             'timestamp': 1623988800,
 541             'release_date': '20200107',
 542             'release_timestamp': 1578427200,
 543         },
 544         'params': {'format': 'bv'},
 545     }, {
 546         # This video requires an account in the browser, but works fine in yt-dlp
 547         'url': 'https://gem.cbc.ca/media/schitts-creek/s01e01',
 548         'info_dict': {
 549             'id': 'schitts-creek/s01e01',
 550             'ext': 'mp4',
 551             'title': 'The Cup Runneth Over',
 552             'description': 'md5:9bca14ea49ab808097530eb05a29e797',
 553             'thumbnail': r're:https://images\.radio-canada\.ca/[^#?]+/cbc_schitts_creek_season_01e01_thumbnail_v01\.jpg',
 554             'series': 'Schitt\'s Creek',
 555             'season_number': 1,
 556             'season': 'Season 1',
 557             'episode_number': 1,
 558             'episode': 'The Cup Runneth Over',
 559             'episode_id': 'schitts-creek/s01e01',
 560             'duration': 1309,
 561             'categories': ['comedy'],
 562             'upload_date': '20210617',
 563             'timestamp': 1623902400,
 564             'release_date': '20151124',
 565             'release_timestamp': 1448323200,
 566         },
 567         'params': {'format': 'bv'},
 568     }, {
 569         'url': 'https://gem.cbc.ca/nadiyas-family-favourites/s01e01',
 570         'only_matching': True,
 571     }]
 572
 573     _GEO_COUNTRIES = ['CA']
 574     _TOKEN_API_KEY = '3f4beddd-2061-49b0-ae80-6f1f2ed65b37'
 575     _NETRC_MACHINE = 'cbcgem'
 576     _claims_token = None
 577
 578     def _new_claims_token(self, email, password):
 579         data = json.dumps({
 580             'email': email,
 581             'password': password,
 582         }).encode()
 583         headers = {'content-type': 'application/json'}
 584         query = {'apikey': self._TOKEN_API_KEY}
 585         resp = self._download_json('https://api.loginradius.com/identity/v2/auth/login',
 586                                    None, data=data, headers=headers, query=query)
 587         access_token = resp['access_token']
 588
 589         query = {
 590             'access_token': access_token,
 591             'apikey': self._TOKEN_API_KEY,
 592             'jwtapp': 'jwt',
 593         }
 594         resp = self._download_json('https://cloud-api.loginradius.com/sso/jwt/api/token',
 595                                    None, headers=headers, query=query)
 596         sig = resp['signature']
 597
 598         data = json.dumps({'jwt': sig}).encode()
 599         headers = {'content-type': 'application/json', 'ott-device-type': 'web'}
 600         resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/token',
 601                                    None, data=data, headers=headers, expected_status=426)
 602         cbc_access_token = resp['accessToken']
 603
 604         headers = {'content-type': 'application/json', 'ott-device-type': 'web', 'ott-access-token': cbc_access_token}
 605         resp = self._download_json('https://services.radio-canada.ca/ott/cbc-api/v2/profile',
 606                                    None, headers=headers, expected_status=426)
 607         return resp['claimsToken']
 608
 609     def _get_claims_token_expiry(self):
 610         # Token is a JWT
 611         # JWT is decoded here and 'exp' field is extracted
 612         # It is a Unix timestamp for when the token expires
 613         b64_data = self._claims_token.split('.')[1]
 614         data = base64.urlsafe_b64decode(b64_data + '==')
 615         return json.loads(data)['exp']
 616
 617     def claims_token_expired(self):
 618         exp = self._get_claims_token_expiry()
 619         # It will expire in less than 10 seconds, or has already expired
 620         return exp - time.time() < 10
 621
 622     def claims_token_valid(self):
 623         return self._claims_token is not None and not self.claims_token_expired()
 624
 625     def _get_claims_token(self, email, password):
 626         if not self.claims_token_valid():
 627             self._claims_token = self._new_claims_token(email, password)
 628             self.cache.store(self._NETRC_MACHINE, 'claims_token', self._claims_token)
 629         return self._claims_token
 630
 631     def _real_initialize(self):
 632         if self.claims_token_valid():
 633             return
 634         self._claims_token = self.cache.load(self._NETRC_MACHINE, 'claims_token')
 635
 636     def _real_extract(self, url):
 637         video_id = self._match_id(url)
 638         video_info = self._download_json(
 639             f'https://services.radio-canada.ca/ott/cbc-api/v2/assets/{video_id}',
 640             video_id, expected_status=426)
 641
 642         email, password = self._get_login_info()
 643         if email and password:
 644             claims_token = self._get_claims_token(email, password)
 645             headers = {'x-claims-token': claims_token}
 646         else:
 647             headers = {}
 648         m3u8_info = self._download_json(video_info['playSession']['url'], video_id, headers=headers)
 649
 650         if m3u8_info.get('errorCode') == 1:
 651             self.raise_geo_restricted(countries=['CA'])
 652         elif m3u8_info.get('errorCode') == 35:
 653             self.raise_login_required(method='password')
 654         elif m3u8_info.get('errorCode') != 0:
 655             raise ExtractorError(f'{self.IE_NAME} said: {m3u8_info.get("errorCode")} - {m3u8_info.get("message")}')
 656
 657         formats = self._extract_m3u8_formats(
 658             m3u8_info['url'], video_id, 'mp4', m3u8_id='hls', query={'manifestType': ''})
 659         self._remove_duplicate_formats(formats)
 660
 661         for fmt in formats:
 662             if fmt.get('vcodec') == 'none':
 663                 if fmt.get('ext') is None:
 664                     fmt['ext'] = 'm4a'
 665                 if fmt.get('acodec') is None:
 666                     fmt['acodec'] = 'mp4a.40.2'
 667
 668                 # Put described audio at the beginning of the list, so that it
 669                 # isn't chosen by default, as most people won't want it.
 670                 if 'descriptive' in fmt['format_id'].lower():
 671                     fmt['preference'] = -2
 672
 673         return {
 674             'id': video_id,
 675             'episode_id': video_id,
 676             'formats': formats,
 677             **traverse_obj(video_info, {
 678                 'title': ('title', {str}),
 679                 'episode': ('title', {str}),
 680                 'description': ('description', {str}),
 681                 'thumbnail': ('image', {url_or_none}),
 682                 'series': ('series', {str}),
 683                 'season_number': ('season', {int_or_none}),
 684                 'episode_number': ('episode', {int_or_none}),
 685                 'duration': ('duration', {int_or_none}),
 686                 'categories': ('category', {str}, all),
 687                 'release_timestamp': ('airDate', {int_or_none(scale=1000)}),
 688                 'timestamp': ('availableDate', {int_or_none(scale=1000)}),
 689             }),
 690         }
 691
 692
 693 class CBCGemPlaylistIE(InfoExtractor):
 694     IE_NAME = 'gem.cbc.ca:playlist'
 695     _VALID_URL = r'https?://gem\.cbc\.ca/(?:media/)?(?P<id>(?P<show>[0-9a-z-]+)/s(?P<season>[0-9]+))/?(?:[?#]|$)'
 696     _TESTS = [{
 697         # TV show playlist, all public videos
 698         'url': 'https://gem.cbc.ca/media/schitts-creek/s06',
 699         'playlist_count': 16,
 700         'info_dict': {
 701             'id': 'schitts-creek/s06',
 702             'title': 'Season 6',
 703             'description': 'md5:6a92104a56cbeb5818cc47884d4326a2',
 704             'series': 'Schitt\'s Creek',
 705             'season_number': 6,
 706             'season': 'Season 6',
 707             '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',
 708         },
 709     }, {
 710         'url': 'https://gem.cbc.ca/schitts-creek/s06',
 711         'only_matching': True,
 712     }]
 713     _API_BASE = 'https://services.radio-canada.ca/ott/cbc-api/v2/shows/'
 714
 715     def _real_extract(self, url):
 716         match = self._match_valid_url(url)
 717         season_id = match.group('id')
 718         show = match.group('show')
 719         show_info = self._download_json(self._API_BASE + show, season_id, expected_status=426)
 720         season = int(match.group('season'))
 721
 722         season_info = next((s for s in show_info['seasons'] if s.get('season') == season), None)
 723
 724         if season_info is None:
 725             raise ExtractorError(f'Couldn\'t find season {season} of {show}')
 726
 727         episodes = []
 728         for episode in season_info['assets']:
 729             episodes.append({
 730                 '_type': 'url_transparent',
 731                 'ie_key': 'CBCGem',
 732                 'url': 'https://gem.cbc.ca/media/' + episode['id'],
 733                 'id': episode['id'],
 734                 'title': episode.get('title'),
 735                 'description': episode.get('description'),
 736                 'thumbnail': episode.get('image'),
 737                 'series': episode.get('series'),
 738                 'season_number': episode.get('season'),
 739                 'season': season_info['title'],
 740                 'season_id': season_info.get('id'),
 741                 'episode_number': episode.get('episode'),
 742                 'episode': episode.get('title'),
 743                 'episode_id': episode['id'],
 744                 'duration': episode.get('duration'),
 745                 'categories': [episode.get('category')],
 746             })
 747
 748         thumbnail = None
 749         tn_uri = season_info.get('image')
 750         # the-national was observed to use a "data:image/png;base64"
 751         # URI for their 'image' value. The image was 1x1, and is
 752         # probably just a placeholder, so it is ignored.
 753         if tn_uri is not None and not tn_uri.startswith('data:'):
 754             thumbnail = tn_uri
 755
 756         return {
 757             '_type': 'playlist',
 758             'entries': episodes,
 759             'id': season_id,
 760             'title': season_info['title'],
 761             'description': season_info.get('description'),
 762             'thumbnail': thumbnail,
 763             'series': show_info.get('title'),
 764             'season_number': season_info.get('season'),
 765             'season': season_info['title'],
 766         }
 767
 768
 769 class CBCGemLiveIE(InfoExtractor):
 770     IE_NAME = 'gem.cbc.ca:live'
 771     _VALID_URL = r'https?://gem\.cbc\.ca/live(?:-event)?/(?P<id>\d+)'
 772     _TESTS = [
 773         {
 774             'url': 'https://gem.cbc.ca/live/920604739687',
 775             'info_dict': {
 776                 'title': 'Ottawa',
 777                 'description': 'The live TV channel and local programming from Ottawa',
 778                 'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/CBC_OTT_VMS/Live_Channel_Static_Images/Ottawa_2880x1620.jpg',
 779                 'live_status': 'is_live',
 780                 'id': 'AyqZwxRqh8EH',
 781                 'ext': 'mp4',
 782                 'release_timestamp': 1492106160,
 783                 'release_date': '20170413',
 784                 'uploader': 'CBCC-NEW',
 785             },
 786             'skip': 'Live might have ended',
 787         },
 788         {
 789             'url': 'https://gem.cbc.ca/live/44',
 790             'info_dict': {
 791                 'id': '44',
 792                 'ext': 'mp4',
 793                 'is_live': True,
 794                 'title': r're:^Ottawa [0-9\-: ]+',
 795                 'description': 'The live TV channel and local programming from Ottawa',
 796                 'live_status': 'is_live',
 797                 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*',
 798             },
 799             'params': {'skip_download': True},
 800             'skip': 'Live might have ended',
 801         },
 802         {
 803             'url': 'https://gem.cbc.ca/live-event/10835',
 804             'info_dict': {
 805                 'id': '10835',
 806                 'ext': 'mp4',
 807                 'is_live': True,
 808                 'title': r're:^The National \| Biden’s trip wraps up, Paltrow testifies, Bird flu [0-9\-: ]+',
 809                 '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.',
 810                 'live_status': 'is_live',
 811                 'thumbnail': r're:https://images.gem.cbc.ca/v1/cbc-gem/live/.*',
 812                 'release_timestamp': 1679706000,
 813                 'release_date': '20230325',
 814             },
 815             'params': {'skip_download': True},
 816             'skip': 'Live might have ended',
 817         },
 818         {   # event replay (medianetlive)
 819             'url': 'https://gem.cbc.ca/live-event/42314',
 820             'md5': '297a9600f554f2258aed01514226a697',
 821             'info_dict': {
 822                 'id': '42314',
 823                 'ext': 'mp4',
 824                 'live_status': 'was_live',
 825                 'title': 'Women\'s Soccer - Canada vs New Zealand',
 826                 'description': 'md5:36200e5f1a70982277b5a6ecea86155d',
 827                 'thumbnail': r're:https://.+default\.jpg',
 828                 'release_timestamp': 1721917200,
 829                 'release_date': '20240725',
 830             },
 831             'params': {'skip_download': True},
 832             'skip': 'Replay might no longer be available',
 833         },
 834         {   # event replay (medianetlive)
 835             'url': 'https://gem.cbc.ca/live-event/43273',
 836             'only_matching': True,
 837         },
 838     ]
 839     _GEO_COUNTRIES = ['CA']
 840
 841     def _real_extract(self, url):
 842         video_id = self._match_id(url)
 843         webpage = self._download_webpage(url, video_id)
 844         video_info = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['data']
 845
 846         # Three types of video_info JSON: info in root, freeTv stream/item, event replay
 847         if not video_info.get('formattedIdMedia'):
 848             if traverse_obj(video_info, ('event', 'key')) == video_id:
 849                 video_info = video_info['event']
 850             else:
 851                 video_info = traverse_obj(video_info, (
 852                     ('freeTv', ('streams', ...)), 'items',
 853                     lambda _, v: v['key'].partition('-')[0] == video_id, any)) or {}
 854
 855         video_stream_id = video_info.get('formattedIdMedia')
 856         if not video_stream_id:
 857             raise ExtractorError(
 858                 'Couldn\'t find video metadata, maybe this livestream is now offline', expected=True)
 859
 860         live_status = 'was_live' if video_info.get('isVodEnabled') else 'is_live'
 861         release_timestamp = traverse_obj(video_info, ('airDate', {parse_iso8601}))
 862
 863         if live_status == 'is_live' and release_timestamp and release_timestamp > time.time():
 864             formats = []
 865             live_status = 'is_upcoming'
 866             self.raise_no_formats('This livestream has not yet started', expected=True)
 867         else:
 868             stream_data = self._download_json(
 869                 'https://services.radio-canada.ca/media/validation/v2/', video_id, query={
 870                     'appCode': 'medianetlive',
 871                     'connectionType': 'hd',
 872                     'deviceType': 'ipad',
 873                     'idMedia': video_stream_id,
 874                     'multibitrate': 'true',
 875                     'output': 'json',
 876                     'tech': 'hls',
 877                     'manifestType': 'desktop',
 878                 })
 879             formats = self._extract_m3u8_formats(
 880                 stream_data['url'], video_id, 'mp4', live=live_status == 'is_live')
 881
 882         return {
 883             'id': video_id,
 884             'formats': formats,
 885             'live_status': live_status,
 886             'release_timestamp': release_timestamp,
 887             **traverse_obj(video_info, {
 888                 'title': ('title', {str}),
 889                 'description': ('description', {str}),
 890                 'thumbnail': ('images', 'card', 'url'),
 891             }),
 892         }