yt_dlp/extractor/youtube.py

   1 import base64
   2 import calendar
   3 import collections
   4 import copy
   5 import datetime
   6 import enum
   7 import hashlib
   8 import itertools
   9 import json
  10 import math
  11 import os.path
  12 import random
  13 import re
  14 import sys
  15 import threading
  16 import time
  17 import traceback
  18 import urllib.error
  19 import urllib.parse
  20
  21 from .common import InfoExtractor, SearchInfoExtractor
  22 from .openload import PhantomJSwrapper
  23 from ..compat import functools
  24 from ..jsinterp import JSInterpreter
  25 from ..utils import (
  26     NO_DEFAULT,
  27     ExtractorError,
  28     LazyList,
  29     UserNotLive,
  30     bug_reports_message,
  31     classproperty,
  32     clean_html,
  33     datetime_from_str,
  34     dict_get,
  35     filter_dict,
  36     float_or_none,
  37     format_field,
  38     get_first,
  39     int_or_none,
  40     is_html,
  41     join_nonempty,
  42     js_to_json,
  43     mimetype2ext,
  44     network_exceptions,
  45     orderedSet,
  46     parse_codecs,
  47     parse_count,
  48     parse_duration,
  49     parse_iso8601,
  50     parse_qs,
  51     qualities,
  52     remove_start,
  53     smuggle_url,
  54     str_or_none,
  55     str_to_int,
  56     strftime_or_none,
  57     traverse_obj,
  58     try_get,
  59     unescapeHTML,
  60     unified_strdate,
  61     unified_timestamp,
  62     unsmuggle_url,
  63     update_url_query,
  64     url_or_none,
  65     urljoin,
  66     variadic,
  67 )
  68
  69 # any clients starting with _ cannot be explicitly requested by the user
  70 INNERTUBE_CLIENTS = {
  71     'web': {
  72         'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
  73         'INNERTUBE_CONTEXT': {
  74             'client': {
  75                 'clientName': 'WEB',
  76                 'clientVersion': '2.20220801.00.00',
  77             }
  78         },
  79         'INNERTUBE_CONTEXT_CLIENT_NAME': 1
  80     },
  81     'web_embedded': {
  82         'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
  83         'INNERTUBE_CONTEXT': {
  84             'client': {
  85                 'clientName': 'WEB_EMBEDDED_PLAYER',
  86                 'clientVersion': '1.20220731.00.00',
  87             },
  88         },
  89         'INNERTUBE_CONTEXT_CLIENT_NAME': 56
  90     },
  91     'web_music': {
  92         'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
  93         'INNERTUBE_HOST': 'music.youtube.com',
  94         'INNERTUBE_CONTEXT': {
  95             'client': {
  96                 'clientName': 'WEB_REMIX',
  97                 'clientVersion': '1.20220727.01.00',
  98             }
  99         },
 100         'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
 101     },
 102     'web_creator': {
 103         'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
 104         'INNERTUBE_CONTEXT': {
 105             'client': {
 106                 'clientName': 'WEB_CREATOR',
 107                 'clientVersion': '1.20220726.00.00',
 108             }
 109         },
 110         'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
 111     },
 112     'android': {
 113         'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
 114         'INNERTUBE_CONTEXT': {
 115             'client': {
 116                 'clientName': 'ANDROID',
 117                 'clientVersion': '17.31.35',
 118                 'androidSdkVersion': 30,
 119                 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
 120             }
 121         },
 122         'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
 123         'REQUIRE_JS_PLAYER': False
 124     },
 125     'android_embedded': {
 126         'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
 127         'INNERTUBE_CONTEXT': {
 128             'client': {
 129                 'clientName': 'ANDROID_EMBEDDED_PLAYER',
 130                 'clientVersion': '17.31.35',
 131                 'androidSdkVersion': 30,
 132                 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
 133             },
 134         },
 135         'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
 136         'REQUIRE_JS_PLAYER': False
 137     },
 138     'android_music': {
 139         'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
 140         'INNERTUBE_CONTEXT': {
 141             'client': {
 142                 'clientName': 'ANDROID_MUSIC',
 143                 'clientVersion': '5.16.51',
 144                 'androidSdkVersion': 30,
 145                 'userAgent': 'com.google.android.apps.youtube.music/5.16.51 (Linux; U; Android 11) gzip'
 146             }
 147         },
 148         'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
 149         'REQUIRE_JS_PLAYER': False
 150     },
 151     'android_creator': {
 152         'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
 153         'INNERTUBE_CONTEXT': {
 154             'client': {
 155                 'clientName': 'ANDROID_CREATOR',
 156                 'clientVersion': '22.30.100',
 157                 'androidSdkVersion': 30,
 158                 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip'
 159             },
 160         },
 161         'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
 162         'REQUIRE_JS_PLAYER': False
 163     },
 164     # iOS clients have HLS live streams. Setting device model to get 60fps formats.
 165     # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
 166     'ios': {
 167         'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
 168         'INNERTUBE_CONTEXT': {
 169             'client': {
 170                 'clientName': 'IOS',
 171                 'clientVersion': '17.33.2',
 172                 'deviceModel': 'iPhone14,3',
 173                 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
 174             }
 175         },
 176         'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
 177         'REQUIRE_JS_PLAYER': False
 178     },
 179     'ios_embedded': {
 180         'INNERTUBE_CONTEXT': {
 181             'client': {
 182                 'clientName': 'IOS_MESSAGES_EXTENSION',
 183                 'clientVersion': '17.33.2',
 184                 'deviceModel': 'iPhone14,3',
 185                 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
 186             },
 187         },
 188         'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
 189         'REQUIRE_JS_PLAYER': False
 190     },
 191     'ios_music': {
 192         'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
 193         'INNERTUBE_CONTEXT': {
 194             'client': {
 195                 'clientName': 'IOS_MUSIC',
 196                 'clientVersion': '5.21',
 197                 'deviceModel': 'iPhone14,3',
 198                 'userAgent': 'com.google.ios.youtubemusic/5.21 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
 199             },
 200         },
 201         'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
 202         'REQUIRE_JS_PLAYER': False
 203     },
 204     'ios_creator': {
 205         'INNERTUBE_CONTEXT': {
 206             'client': {
 207                 'clientName': 'IOS_CREATOR',
 208                 'clientVersion': '22.33.101',
 209                 'deviceModel': 'iPhone14,3',
 210                 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
 211             },
 212         },
 213         'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
 214         'REQUIRE_JS_PLAYER': False
 215     },
 216     # mweb has 'ultralow' formats
 217     # See: https://github.com/yt-dlp/yt-dlp/pull/557
 218     'mweb': {
 219         'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
 220         'INNERTUBE_CONTEXT': {
 221             'client': {
 222                 'clientName': 'MWEB',
 223                 'clientVersion': '2.20220801.00.00',
 224             }
 225         },
 226         'INNERTUBE_CONTEXT_CLIENT_NAME': 2
 227     },
 228     # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
 229     # See: https://github.com/zerodytrash/YouTube-Internal-Clients
 230     'tv_embedded': {
 231         'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
 232         'INNERTUBE_CONTEXT': {
 233             'client': {
 234                 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
 235                 'clientVersion': '2.0',
 236             },
 237         },
 238         'INNERTUBE_CONTEXT_CLIENT_NAME': 85
 239     },
 240 }
 241
 242
 243 def _split_innertube_client(client_name):
 244     variant, *base = client_name.rsplit('.', 1)
 245     if base:
 246         return variant, base[0], variant
 247     base, *variant = client_name.split('_', 1)
 248     return client_name, base, variant[0] if variant else None
 249
 250
 251 def build_innertube_clients():
 252     THIRD_PARTY = {
 253         'embedUrl': 'https://www.youtube.com/',  # Can be any valid URL
 254     }
 255     BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
 256     priority = qualities(BASE_CLIENTS[::-1])
 257
 258     for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
 259         ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
 260         ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
 261         ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
 262         ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
 263
 264         _, base_client, variant = _split_innertube_client(client)
 265         ytcfg['priority'] = 10 * priority(base_client)
 266
 267         if not variant:
 268             INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
 269             embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
 270             embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
 271             embedscreen['priority'] -= 3
 272         elif variant == 'embedded':
 273             ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
 274             ytcfg['priority'] -= 2
 275         else:
 276             ytcfg['priority'] -= 3
 277
 278
 279 build_innertube_clients()
 280
 281
 282 class BadgeType(enum.Enum):
 283     AVAILABILITY_UNLISTED = enum.auto()
 284     AVAILABILITY_PRIVATE = enum.auto()
 285     AVAILABILITY_PUBLIC = enum.auto()
 286     AVAILABILITY_PREMIUM = enum.auto()
 287     AVAILABILITY_SUBSCRIPTION = enum.auto()
 288     LIVE_NOW = enum.auto()
 289
 290
 291 class YoutubeBaseInfoExtractor(InfoExtractor):
 292     """Provide base functions for Youtube extractors"""
 293
 294     _RESERVED_NAMES = (
 295         r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|'
 296         r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
 297         r'browse|oembed|get_video_info|iframe_api|s/player|source|'
 298         r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
 299
 300     _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
 301
 302     # _NETRC_MACHINE = 'youtube'
 303
 304     # If True it will raise an error if no login info is provided
 305     _LOGIN_REQUIRED = False
 306
 307     _INVIDIOUS_SITES = (
 308         # invidious-redirect websites
 309         r'(?:www\.)?redirect\.invidious\.io',
 310         r'(?:(?:www|dev)\.)?invidio\.us',
 311         # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
 312         r'(?:www\.)?invidious\.pussthecat\.org',
 313         r'(?:www\.)?invidious\.zee\.li',
 314         r'(?:www\.)?invidious\.ethibox\.fr',
 315         r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
 316         r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
 317         r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
 318         # youtube-dl invidious instances list
 319         r'(?:(?:www|no)\.)?invidiou\.sh',
 320         r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
 321         r'(?:www\.)?invidious\.kabi\.tk',
 322         r'(?:www\.)?invidious\.mastodon\.host',
 323         r'(?:www\.)?invidious\.zapashcanon\.fr',
 324         r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
 325         r'(?:www\.)?invidious\.tinfoil-hat\.net',
 326         r'(?:www\.)?invidious\.himiko\.cloud',
 327         r'(?:www\.)?invidious\.reallyancient\.tech',
 328         r'(?:www\.)?invidious\.tube',
 329         r'(?:www\.)?invidiou\.site',
 330         r'(?:www\.)?invidious\.site',
 331         r'(?:www\.)?invidious\.xyz',
 332         r'(?:www\.)?invidious\.nixnet\.xyz',
 333         r'(?:www\.)?invidious\.048596\.xyz',
 334         r'(?:www\.)?invidious\.drycat\.fr',
 335         r'(?:www\.)?inv\.skyn3t\.in',
 336         r'(?:www\.)?tube\.poal\.co',
 337         r'(?:www\.)?tube\.connect\.cafe',
 338         r'(?:www\.)?vid\.wxzm\.sx',
 339         r'(?:www\.)?vid\.mint\.lgbt',
 340         r'(?:www\.)?vid\.puffyan\.us',
 341         r'(?:www\.)?yewtu\.be',
 342         r'(?:www\.)?yt\.elukerio\.org',
 343         r'(?:www\.)?yt\.lelux\.fi',
 344         r'(?:www\.)?invidious\.ggc-project\.de',
 345         r'(?:www\.)?yt\.maisputain\.ovh',
 346         r'(?:www\.)?ytprivate\.com',
 347         r'(?:www\.)?invidious\.13ad\.de',
 348         r'(?:www\.)?invidious\.toot\.koeln',
 349         r'(?:www\.)?invidious\.fdn\.fr',
 350         r'(?:www\.)?watch\.nettohikari\.com',
 351         r'(?:www\.)?invidious\.namazso\.eu',
 352         r'(?:www\.)?invidious\.silkky\.cloud',
 353         r'(?:www\.)?invidious\.exonip\.de',
 354         r'(?:www\.)?invidious\.riverside\.rocks',
 355         r'(?:www\.)?invidious\.blamefran\.net',
 356         r'(?:www\.)?invidious\.moomoo\.de',
 357         r'(?:www\.)?ytb\.trom\.tf',
 358         r'(?:www\.)?yt\.cyberhost\.uk',
 359         r'(?:www\.)?kgg2m7yk5aybusll\.onion',
 360         r'(?:www\.)?qklhadlycap4cnod\.onion',
 361         r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
 362         r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
 363         r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
 364         r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
 365         r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
 366         r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
 367         r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
 368         r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
 369         r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
 370         r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
 371         # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
 372         r'(?:www\.)?piped\.kavin\.rocks',
 373         r'(?:www\.)?piped\.tokhmi\.xyz',
 374         r'(?:www\.)?piped\.syncpundit\.io',
 375         r'(?:www\.)?piped\.mha\.fi',
 376         r'(?:www\.)?watch\.whatever\.social',
 377         r'(?:www\.)?piped\.garudalinux\.org',
 378         r'(?:www\.)?piped\.rivo\.lol',
 379         r'(?:www\.)?piped-libre\.kavin\.rocks',
 380         r'(?:www\.)?yt\.jae\.fi',
 381         r'(?:www\.)?piped\.mint\.lgbt',
 382         r'(?:www\.)?il\.ax',
 383         r'(?:www\.)?piped\.esmailelbob\.xyz',
 384         r'(?:www\.)?piped\.projectsegfau\.lt',
 385         r'(?:www\.)?piped\.privacydev\.net',
 386         r'(?:www\.)?piped\.palveluntarjoaja\.eu',
 387         r'(?:www\.)?piped\.smnz\.de',
 388         r'(?:www\.)?piped\.adminforge\.de',
 389         r'(?:www\.)?watch\.whatevertinfoil\.de',
 390         r'(?:www\.)?piped\.qdi\.fi',
 391     )
 392
 393     # extracted from account/account_menu ep
 394     # XXX: These are the supported YouTube UI and API languages,
 395     # which is slightly different from languages supported for translation in YouTube studio
 396     _SUPPORTED_LANG_CODES = [
 397         'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
 398         'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
 399         'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
 400         'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
 401         'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
 402         'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
 403     ]
 404
 405     _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
 406
 407     @functools.cached_property
 408     def _preferred_lang(self):
 409         """
 410         Returns a language code supported by YouTube for the user preferred language.
 411         Returns None if no preferred language set.
 412         """
 413         preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
 414         if not preferred_lang:
 415             return
 416         if preferred_lang not in self._SUPPORTED_LANG_CODES:
 417             raise ExtractorError(
 418                 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
 419                 expected=True)
 420         elif preferred_lang != 'en':
 421             self.report_warning(
 422                 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
 423         return preferred_lang
 424
 425     def _initialize_consent(self):
 426         cookies = self._get_cookies('https://www.youtube.com/')
 427         if cookies.get('__Secure-3PSID'):
 428             return
 429         consent_id = None
 430         consent = cookies.get('CONSENT')
 431         if consent:
 432             if 'YES' in consent.value:
 433                 return
 434             consent_id = self._search_regex(
 435                 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
 436         if not consent_id:
 437             consent_id = random.randint(100, 999)
 438         self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
 439
 440     def _initialize_pref(self):
 441         cookies = self._get_cookies('https://www.youtube.com/')
 442         pref_cookie = cookies.get('PREF')
 443         pref = {}
 444         if pref_cookie:
 445             try:
 446                 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
 447             except ValueError:
 448                 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
 449         pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
 450         self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
 451
 452     def _real_initialize(self):
 453         self._initialize_pref()
 454         self._initialize_consent()
 455         self._check_login_required()
 456
 457     def _check_login_required(self):
 458         if self._LOGIN_REQUIRED and not self._cookies_passed:
 459             self.raise_login_required('Login details are needed to download this content', method='cookies')
 460
 461     _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
 462     _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
 463
 464     def _get_default_ytcfg(self, client='web'):
 465         return copy.deepcopy(INNERTUBE_CLIENTS[client])
 466
 467     def _get_innertube_host(self, client='web'):
 468         return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
 469
 470     def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
 471         # try_get but with fallback to default ytcfg client values when present
 472         _func = lambda y: try_get(y, getter, expected_type)
 473         return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
 474
 475     def _extract_client_name(self, ytcfg, default_client='web'):
 476         return self._ytcfg_get_safe(
 477             ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
 478                     lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
 479
 480     def _extract_client_version(self, ytcfg, default_client='web'):
 481         return self._ytcfg_get_safe(
 482             ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
 483                     lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
 484
 485     def _select_api_hostname(self, req_api_hostname, default_client=None):
 486         return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
 487                 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
 488
 489     def _extract_api_key(self, ytcfg=None, default_client='web'):
 490         return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
 491
 492     def _extract_context(self, ytcfg=None, default_client='web'):
 493         context = get_first(
 494             (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
 495         # Enforce language and tz for extraction
 496         client_context = traverse_obj(context, 'client', expected_type=dict, default={})
 497         client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
 498         return context
 499
 500     _SAPISID = None
 501
 502     def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
 503         time_now = round(time.time())
 504         if self._SAPISID is None:
 505             yt_cookies = self._get_cookies('https://www.youtube.com')
 506             # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
 507             # See: https://github.com/yt-dlp/yt-dlp/issues/393
 508             sapisid_cookie = dict_get(
 509                 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
 510             if sapisid_cookie and sapisid_cookie.value:
 511                 self._SAPISID = sapisid_cookie.value
 512                 self.write_debug('Extracted SAPISID cookie')
 513                 # SAPISID cookie is required if not already present
 514                 if not yt_cookies.get('SAPISID'):
 515                     self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
 516                     self._set_cookie(
 517                         '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
 518             else:
 519                 self._SAPISID = False
 520         if not self._SAPISID:
 521             return None
 522         # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
 523         sapisidhash = hashlib.sha1(
 524             f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
 525         return f'SAPISIDHASH {time_now}_{sapisidhash}'
 526
 527     def _call_api(self, ep, query, video_id, fatal=True, headers=None,
 528                   note='Downloading API JSON', errnote='Unable to download API page',
 529                   context=None, api_key=None, api_hostname=None, default_client='web'):
 530
 531         data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
 532         data.update(query)
 533         real_headers = self.generate_api_headers(default_client=default_client)
 534         real_headers.update({'content-type': 'application/json'})
 535         if headers:
 536             real_headers.update(headers)
 537         api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
 538                    or api_key or self._extract_api_key(default_client=default_client))
 539         return self._download_json(
 540             f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
 541             video_id=video_id, fatal=fatal, note=note, errnote=errnote,
 542             data=json.dumps(data).encode('utf8'), headers=real_headers,
 543             query={'key': api_key, 'prettyPrint': 'false'})
 544
 545     def extract_yt_initial_data(self, item_id, webpage, fatal=True):
 546         return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
 547
 548     @staticmethod
 549     def _extract_session_index(*data):
 550         """
 551         Index of current account in account list.
 552         See: https://github.com/yt-dlp/yt-dlp/pull/519
 553         """
 554         for ytcfg in data:
 555             session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
 556             if session_index is not None:
 557                 return session_index
 558
 559     # Deprecated?
 560     def _extract_identity_token(self, ytcfg=None, webpage=None):
 561         if ytcfg:
 562             token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
 563             if token:
 564                 return token
 565         if webpage:
 566             return self._search_regex(
 567                 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
 568                 'identity token', default=None, fatal=False)
 569
 570     @staticmethod
 571     def _extract_account_syncid(*args):
 572         """
 573         Extract syncId required to download private playlists of secondary channels
 574         @params response and/or ytcfg
 575         """
 576         for data in args:
 577             # ytcfg includes channel_syncid if on secondary channel
 578             delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
 579             if delegated_sid:
 580                 return delegated_sid
 581             sync_ids = (try_get(
 582                 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
 583                        lambda x: x['DATASYNC_ID']), str) or '').split('||')
 584             if len(sync_ids) >= 2 and sync_ids[1]:
 585                 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
 586                 # and just "user_syncid||" for primary channel. We only want the channel_syncid
 587                 return sync_ids[0]
 588
 589     @staticmethod
 590     def _extract_visitor_data(*args):
 591         """
 592         Extracts visitorData from an API response or ytcfg
 593         Appears to be used to track session state
 594         """
 595         return get_first(
 596             args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
 597             expected_type=str)
 598
 599     @functools.cached_property
 600     def is_authenticated(self):
 601         return bool(self._generate_sapisidhash_header())
 602
 603     def extract_ytcfg(self, video_id, webpage):
 604         if not webpage:
 605             return {}
 606         return self._parse_json(
 607             self._search_regex(
 608                 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
 609                 default='{}'), video_id, fatal=False) or {}
 610
 611     def generate_api_headers(
 612             self, *, ytcfg=None, account_syncid=None, session_index=None,
 613             visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
 614
 615         origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
 616         headers = {
 617             'X-YouTube-Client-Name': str(
 618                 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
 619             'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
 620             'Origin': origin,
 621             'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
 622             'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
 623             'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
 624             'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
 625         }
 626         if session_index is None:
 627             session_index = self._extract_session_index(ytcfg)
 628         if account_syncid or session_index is not None:
 629             headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
 630
 631         auth = self._generate_sapisidhash_header(origin)
 632         if auth is not None:
 633             headers['Authorization'] = auth
 634             headers['X-Origin'] = origin
 635         return filter_dict(headers)
 636
 637     def _download_ytcfg(self, client, video_id):
 638         url = {
 639             'web': 'https://www.youtube.com',
 640             'web_music': 'https://music.youtube.com',
 641             'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
 642         }.get(client)
 643         if not url:
 644             return {}
 645         webpage = self._download_webpage(
 646             url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
 647         return self.extract_ytcfg(video_id, webpage) or {}
 648
 649     @staticmethod
 650     def _build_api_continuation_query(continuation, ctp=None):
 651         query = {
 652             'continuation': continuation
 653         }
 654         # TODO: Inconsistency with clickTrackingParams.
 655         # Currently we have a fixed ctp contained within context (from ytcfg)
 656         # and a ctp in root query for continuation.
 657         if ctp:
 658             query['clickTracking'] = {'clickTrackingParams': ctp}
 659         return query
 660
 661     @classmethod
 662     def _extract_next_continuation_data(cls, renderer):
 663         next_continuation = try_get(
 664             renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
 665                        lambda x: x['continuation']['reloadContinuationData']), dict)
 666         if not next_continuation:
 667             return
 668         continuation = next_continuation.get('continuation')
 669         if not continuation:
 670             return
 671         ctp = next_continuation.get('clickTrackingParams')
 672         return cls._build_api_continuation_query(continuation, ctp)
 673
 674     @classmethod
 675     def _extract_continuation_ep_data(cls, continuation_ep: dict):
 676         if isinstance(continuation_ep, dict):
 677             continuation = try_get(
 678                 continuation_ep, lambda x: x['continuationCommand']['token'], str)
 679             if not continuation:
 680                 return
 681             ctp = continuation_ep.get('clickTrackingParams')
 682             return cls._build_api_continuation_query(continuation, ctp)
 683
 684     @classmethod
 685     def _extract_continuation(cls, renderer):
 686         next_continuation = cls._extract_next_continuation_data(renderer)
 687         if next_continuation:
 688             return next_continuation
 689
 690         return traverse_obj(renderer, (
 691             ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
 692             ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
 693         ), get_all=False, expected_type=cls._extract_continuation_ep_data)
 694
 695     @classmethod
 696     def _extract_alerts(cls, data):
 697         for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
 698             if not isinstance(alert_dict, dict):
 699                 continue
 700             for alert in alert_dict.values():
 701                 alert_type = alert.get('type')
 702                 if not alert_type:
 703                     continue
 704                 message = cls._get_text(alert, 'text')
 705                 if message:
 706                     yield alert_type, message
 707
 708     def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
 709         errors, warnings = [], []
 710         for alert_type, alert_message in alerts:
 711             if alert_type.lower() == 'error' and fatal:
 712                 errors.append([alert_type, alert_message])
 713             elif alert_message not in self._IGNORED_WARNINGS:
 714                 warnings.append([alert_type, alert_message])
 715
 716         for alert_type, alert_message in (warnings + errors[:-1]):
 717             self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
 718         if errors:
 719             raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
 720
 721     def _extract_and_report_alerts(self, data, *args, **kwargs):
 722         return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
 723
 724     def _extract_badges(self, renderer: dict):
 725         privacy_icon_map = {
 726             'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
 727             'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
 728             'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC
 729         }
 730
 731         badge_style_map = {
 732             'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
 733             'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
 734             'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW
 735         }
 736
 737         label_map = {
 738             'unlisted': BadgeType.AVAILABILITY_UNLISTED,
 739             'private': BadgeType.AVAILABILITY_PRIVATE,
 740             'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
 741             'live': BadgeType.LIVE_NOW,
 742             'premium': BadgeType.AVAILABILITY_PREMIUM
 743         }
 744
 745         badges = []
 746         for badge in traverse_obj(renderer, ('badges', ..., 'metadataBadgeRenderer'), default=[]):
 747             badge_type = (
 748                 privacy_icon_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
 749                 or badge_style_map.get(traverse_obj(badge, 'style'))
 750             )
 751             if badge_type:
 752                 badges.append({'type': badge_type})
 753                 continue
 754
 755             # fallback, won't work in some languages
 756             label = traverse_obj(badge, 'label', expected_type=str, default='')
 757             for match, label_badge_type in label_map.items():
 758                 if match in label.lower():
 759                     badges.append({'type': badge_type})
 760                     continue
 761
 762         return badges
 763
 764     @staticmethod
 765     def _has_badge(badges, badge_type):
 766         return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
 767
 768     @staticmethod
 769     def _get_text(data, *path_list, max_runs=None):
 770         for path in path_list or [None]:
 771             if path is None:
 772                 obj = [data]
 773             else:
 774                 obj = traverse_obj(data, path, default=[])
 775                 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
 776                     obj = [obj]
 777             for item in obj:
 778                 text = try_get(item, lambda x: x['simpleText'], str)
 779                 if text:
 780                     return text
 781                 runs = try_get(item, lambda x: x['runs'], list) or []
 782                 if not runs and isinstance(item, list):
 783                     runs = item
 784
 785                 runs = runs[:min(len(runs), max_runs or len(runs))]
 786                 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
 787                 if text:
 788                     return text
 789
 790     def _get_count(self, data, *path_list):
 791         count_text = self._get_text(data, *path_list) or ''
 792         count = parse_count(count_text)
 793         if count is None:
 794             count = str_to_int(
 795                 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
 796         return count
 797
 798     @staticmethod
 799     def _extract_thumbnails(data, *path_list):
 800         """
 801         Extract thumbnails from thumbnails dict
 802         @param path_list: path list to level that contains 'thumbnails' key
 803         """
 804         thumbnails = []
 805         for path in path_list or [()]:
 806             for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
 807                 thumbnail_url = url_or_none(thumbnail.get('url'))
 808                 if not thumbnail_url:
 809                     continue
 810                 # Sometimes youtube gives a wrong thumbnail URL. See:
 811                 # https://github.com/yt-dlp/yt-dlp/issues/233
 812                 # https://github.com/ytdl-org/youtube-dl/issues/28023
 813                 if 'maxresdefault' in thumbnail_url:
 814                     thumbnail_url = thumbnail_url.split('?')[0]
 815                 thumbnails.append({
 816                     'url': thumbnail_url,
 817                     'height': int_or_none(thumbnail.get('height')),
 818                     'width': int_or_none(thumbnail.get('width')),
 819                 })
 820         return thumbnails
 821
 822     @staticmethod
 823     def extract_relative_time(relative_time_text):
 824         """
 825         Extracts a relative time from string and converts to dt object
 826         e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
 827         """
 828         mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text)
 829         if mobj:
 830             start = mobj.group('start')
 831             if start:
 832                 return datetime_from_str(start)
 833             try:
 834                 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
 835             except ValueError:
 836                 return None
 837
 838     def _parse_time_text(self, text):
 839         if not text:
 840             return
 841         dt = self.extract_relative_time(text)
 842         timestamp = None
 843         if isinstance(dt, datetime.datetime):
 844             timestamp = calendar.timegm(dt.timetuple())
 845
 846         if timestamp is None:
 847             timestamp = (
 848                 unified_timestamp(text) or unified_timestamp(
 849                     self._search_regex(
 850                         (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
 851                         text.lower(), 'time text', default=None)))
 852
 853         if text and timestamp is None and self._preferred_lang in (None, 'en'):
 854             self.report_warning(
 855                 f'Cannot parse localized time text "{text}"', only_once=True)
 856         return timestamp
 857
 858     def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
 859                           ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
 860                           default_client='web'):
 861         for retry in self.RetryManager():
 862             try:
 863                 response = self._call_api(
 864                     ep=ep, fatal=True, headers=headers,
 865                     video_id=item_id, query=query, note=note,
 866                     context=self._extract_context(ytcfg, default_client),
 867                     api_key=self._extract_api_key(ytcfg, default_client),
 868                     api_hostname=api_hostname, default_client=default_client)
 869             except ExtractorError as e:
 870                 if not isinstance(e.cause, network_exceptions):
 871                     return self._error_or_warning(e, fatal=fatal)
 872                 elif not isinstance(e.cause, urllib.error.HTTPError):
 873                     retry.error = e
 874                     continue
 875
 876                 first_bytes = e.cause.read(512)
 877                 if not is_html(first_bytes):
 878                     yt_error = try_get(
 879                         self._parse_json(
 880                             self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
 881                         lambda x: x['error']['message'], str)
 882                     if yt_error:
 883                         self._report_alerts([('ERROR', yt_error)], fatal=False)
 884                 # Downloading page may result in intermittent 5xx HTTP error
 885                 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
 886                 # We also want to catch all other network exceptions since errors in later pages can be troublesome
 887                 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
 888                 if e.cause.code not in (403, 429):
 889                     retry.error = e
 890                     continue
 891                 return self._error_or_warning(e, fatal=fatal)
 892
 893             try:
 894                 self._extract_and_report_alerts(response, only_once=True)
 895             except ExtractorError as e:
 896                 # YouTube servers may return errors we want to retry on in a 200 OK response
 897                 # See: https://github.com/yt-dlp/yt-dlp/issues/839
 898                 if 'unknown error' in e.msg.lower():
 899                     retry.error = e
 900                     continue
 901                 return self._error_or_warning(e, fatal=fatal)
 902             # Youtube sometimes sends incomplete data
 903             # See: https://github.com/ytdl-org/youtube-dl/issues/28194
 904             if not traverse_obj(response, *variadic(check_get_keys)):
 905                 retry.error = ExtractorError('Incomplete data received', expected=True)
 906                 continue
 907
 908             return response
 909
 910     @staticmethod
 911     def is_music_url(url):
 912         return re.match(r'https?://music\.youtube\.com/', url) is not None
 913
 914     def _extract_video(self, renderer):
 915         video_id = renderer.get('videoId')
 916
 917         reel_header_renderer = traverse_obj(renderer, (
 918             'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
 919             'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
 920
 921         title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
 922         description = self._get_text(renderer, 'descriptionSnippet')
 923
 924         duration = int_or_none(renderer.get('lengthSeconds'))
 925         if duration is None:
 926             duration = parse_duration(self._get_text(
 927                 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
 928         if duration is None:
 929             # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
 930             duration = parse_duration(self._search_regex(
 931                 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
 932                 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
 933                 video_id, default=None, group='duration'))
 934
 935         channel_id = traverse_obj(
 936             renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
 937             expected_type=str, get_all=False)
 938         if not channel_id:
 939             channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
 940
 941         overlay_style = traverse_obj(
 942             renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
 943             get_all=False, expected_type=str)
 944         badges = self._extract_badges(renderer)
 945
 946         navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
 947             renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
 948             expected_type=str)) or ''
 949         url = f'https://www.youtube.com/watch?v={video_id}'
 950         if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
 951             url = f'https://www.youtube.com/shorts/{video_id}'
 952
 953         time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
 954                      or self._get_text(reel_header_renderer, 'timestampText') or '')
 955         scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
 956
 957         live_status = (
 958             'is_upcoming' if scheduled_timestamp is not None
 959             else 'was_live' if 'streamed' in time_text.lower()
 960             else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
 961             else None)
 962
 963         # videoInfo is a string like '50K views • 10 years ago'.
 964         view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
 965         view_count = (0 if 'no views' in view_count_text.lower()
 966                       else self._get_count({'simpleText': view_count_text}))
 967         view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
 968
 969         return {
 970             '_type': 'url',
 971             'ie_key': YoutubeIE.ie_key(),
 972             'id': video_id,
 973             'url': url,
 974             'title': title,
 975             'description': description,
 976             'duration': duration,
 977             'channel_id': channel_id,
 978             'channel': (self._get_text(renderer, 'ownerText', 'shortBylineText')
 979                         or self._get_text(reel_header_renderer, 'channelTitleText')),
 980             'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
 981             'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
 982             'timestamp': (self._parse_time_text(time_text)
 983                           if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
 984                           else None),
 985             'release_timestamp': scheduled_timestamp,
 986             'availability':
 987                 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
 988                 else self._availability(
 989                     is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
 990                     needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
 991                     needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
 992                     is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
 993             view_count_field: view_count,
 994             'live_status': live_status
 995         }
 996
 997
 998 class YoutubeIE(YoutubeBaseInfoExtractor):
 999     IE_DESC = 'YouTube'
1000     _VALID_URL = r"""(?x)^
1001                      (
1002                          (?:https?://|//)                                    # http(s):// or protocol-independent URL
1003                          (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1004                             (?:www\.)?deturl\.com/www\.youtube\.com|
1005                             (?:www\.)?pwnyoutube\.com|
1006                             (?:www\.)?hooktube\.com|
1007                             (?:www\.)?yourepeat\.com|
1008                             tube\.majestyc\.net|
1009                             %(invidious)s|
1010                             youtube\.googleapis\.com)/                        # the various hostnames, with wildcard subdomains
1011                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
1012                          (?:                                                  # the various things that can precede the ID:
1013                              (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream))  # v/ or embed/ or e/ or shorts/
1014                              |(?:                                             # or the v= param in all its forms
1015                                  (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)?  # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1016                                  (?:\?|\#!?)                                  # the params delimiter ? or # or #!
1017                                  (?:.*?[&;])??                                # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1018                                  v=
1019                              )
1020                          ))
1021                          |(?:
1022                             youtu\.be|                                        # just youtu.be/xxxx
1023                             vid\.plus|                                        # or vid.plus/xxxx
1024                             zwearz\.com/watch|                                # or zwearz.com/watch/xxxx
1025                             %(invidious)s
1026                          )/
1027                          |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1028                          )
1029                      )?                                                       # all until now is optional -> you can pass the naked ID
1030                      (?P<id>[0-9A-Za-z_-]{11})                                # here is it! the YouTube video ID
1031                      (?(1).+)?                                                # if we found the ID, everything can follow
1032                      (?:\#|$)""" % {
1033         'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1034     }
1035     _EMBED_REGEX = [
1036         r'''(?x)
1037             (?:
1038                 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1039                 data-video-url=|
1040                 <embed[^>]+?src=|
1041                 embedSWF\(?:\s*|
1042                 <object[^>]+data=|
1043                 new\s+SWFObject\(
1044             )
1045             (["\'])
1046                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1047                 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1048             \1''',
1049         # https://wordpress.org/plugins/lazy-load-for-videos/
1050         r'''(?xs)
1051             <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1052             \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1053     ]
1054     _RETURN_TYPE = 'video'  # XXX: How to handle multifeed?
1055
1056     _PLAYER_INFO_RE = (
1057         r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1058         r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
1059         r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1060     )
1061     _formats = {
1062         '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1063         '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1064         '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1065         '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1066         '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1067         '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1068         '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1069         '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1070         # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1071         '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1072         '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1073         '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1074         '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1075         '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1076         '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1077         '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1078         '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1079         '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1080
1081
1082         # 3D videos
1083         '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1084         '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1085         '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1086         '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1087         '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1088         '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1089         '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1090
1091         # Apple HTTP Live Streaming
1092         '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1093         '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1094         '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1095         '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1096         '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1097         '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1098         '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1099         '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1100
1101         # DASH mp4 video
1102         '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1103         '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1104         '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1105         '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1106         '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1107         '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'},  # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1108         '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1109         '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1110         '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1111         '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1112         '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1113         '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1114
1115         # Dash mp4 audio
1116         '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1117         '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1118         '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1119         '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1120         '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1121         '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1122         '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1123
1124         # Dash webm
1125         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1126         '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1127         '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1128         '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1129         '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1130         '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1131         '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1132         '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1133         '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1134         '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1135         '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1136         '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1137         '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1138         '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1139         '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1140         # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1141         '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1142         '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1143         '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1144         '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1145         '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1146         '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1147
1148         # Dash webm audio
1149         '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1150         '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1151
1152         # Dash webm audio with opus inside
1153         '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1154         '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1155         '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1156
1157         # RTMP (unnamed)
1158         '_rtmp': {'protocol': 'rtmp'},
1159
1160         # av01 video only formats sometimes served with "unknown" codecs
1161         '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1162         '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1163         '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1164         '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1165         '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1166         '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1167         '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1168         '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1169     }
1170     _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1171
1172     _GEO_BYPASS = False
1173
1174     IE_NAME = 'youtube'
1175     _TESTS = [
1176         {
1177             'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1178             'info_dict': {
1179                 'id': 'BaW_jenozKc',
1180                 'ext': 'mp4',
1181                 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1182                 'uploader': 'Philipp Hagemeister',
1183                 'uploader_id': 'phihag',
1184                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1185                 'channel': 'Philipp Hagemeister',
1186                 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1187                 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1188                 'upload_date': '20121002',
1189                 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1190                 'categories': ['Science & Technology'],
1191                 'tags': ['youtube-dl'],
1192                 'duration': 10,
1193                 'view_count': int,
1194                 'like_count': int,
1195                 'availability': 'public',
1196                 'playable_in_embed': True,
1197                 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1198                 'live_status': 'not_live',
1199                 'age_limit': 0,
1200                 'start_time': 1,
1201                 'end_time': 9,
1202                 'comment_count': int,
1203                 'channel_follower_count': int
1204             }
1205         },
1206         {
1207             'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1208             'note': 'Embed-only video (#1746)',
1209             'info_dict': {
1210                 'id': 'yZIXLfi8CZQ',
1211                 'ext': 'mp4',
1212                 'upload_date': '20120608',
1213                 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1214                 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1215                 'uploader': 'SET India',
1216                 'uploader_id': 'setindia',
1217                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1218                 'age_limit': 18,
1219             },
1220             'skip': 'Private video',
1221         },
1222         {
1223             'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1224             'note': 'Use the first video ID in the URL',
1225             'info_dict': {
1226                 'id': 'BaW_jenozKc',
1227                 'ext': 'mp4',
1228                 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1229                 'uploader': 'Philipp Hagemeister',
1230                 'uploader_id': 'phihag',
1231                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1232                 'channel': 'Philipp Hagemeister',
1233                 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1234                 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1235                 'upload_date': '20121002',
1236                 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1237                 'categories': ['Science & Technology'],
1238                 'tags': ['youtube-dl'],
1239                 'duration': 10,
1240                 'view_count': int,
1241                 'like_count': int,
1242                 'availability': 'public',
1243                 'playable_in_embed': True,
1244                 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1245                 'live_status': 'not_live',
1246                 'age_limit': 0,
1247                 'comment_count': int,
1248                 'channel_follower_count': int
1249             },
1250             'params': {
1251                 'skip_download': True,
1252             },
1253         },
1254         {
1255             'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1256             'note': '256k DASH audio (format 141) via DASH manifest',
1257             'info_dict': {
1258                 'id': 'a9LDPn-MO4I',
1259                 'ext': 'm4a',
1260                 'upload_date': '20121002',
1261                 'uploader_id': '8KVIDEO',
1262                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1263                 'description': '',
1264                 'uploader': '8KVIDEO',
1265                 'title': 'UHDTV TEST 8K VIDEO.mp4'
1266             },
1267             'params': {
1268                 'youtube_include_dash_manifest': True,
1269                 'format': '141',
1270             },
1271             'skip': 'format 141 not served anymore',
1272         },
1273         # DASH manifest with encrypted signature
1274         {
1275             'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1276             'info_dict': {
1277                 'id': 'IB3lcPjvWLA',
1278                 'ext': 'm4a',
1279                 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1280                 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1281                 'duration': 244,
1282                 'uploader': 'AfrojackVEVO',
1283                 'uploader_id': 'AfrojackVEVO',
1284                 'upload_date': '20131011',
1285                 'abr': 129.495,
1286                 'like_count': int,
1287                 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1288                 'playable_in_embed': True,
1289                 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1290                 'view_count': int,
1291                 'track': 'The Spark',
1292                 'live_status': 'not_live',
1293                 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1294                 'channel': 'Afrojack',
1295                 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1296                 'tags': 'count:19',
1297                 'availability': 'public',
1298                 'categories': ['Music'],
1299                 'age_limit': 0,
1300                 'alt_title': 'The Spark',
1301                 'channel_follower_count': int
1302             },
1303             'params': {
1304                 'youtube_include_dash_manifest': True,
1305                 'format': '141/bestaudio[ext=m4a]',
1306             },
1307         },
1308         # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1309         {
1310             'note': 'Embed allowed age-gate video',
1311             'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1312             'info_dict': {
1313                 'id': 'HtVdAasjOgU',
1314                 'ext': 'mp4',
1315                 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1316                 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1317                 'duration': 142,
1318                 'uploader': 'The Witcher',
1319                 'uploader_id': 'WitcherGame',
1320                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1321                 'upload_date': '20140605',
1322                 'age_limit': 18,
1323                 'categories': ['Gaming'],
1324                 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1325                 'availability': 'needs_auth',
1326                 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1327                 'like_count': int,
1328                 'channel': 'The Witcher',
1329                 'live_status': 'not_live',
1330                 'tags': 'count:17',
1331                 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1332                 'playable_in_embed': True,
1333                 'view_count': int,
1334                 'channel_follower_count': int
1335             },
1336         },
1337         {
1338             'note': 'Age-gate video with embed allowed in public site',
1339             'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1340             'info_dict': {
1341                 'id': 'HsUATh_Nc2U',
1342                 'ext': 'mp4',
1343                 'title': 'Godzilla 2 (Official Video)',
1344                 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1345                 'upload_date': '20200408',
1346                 'uploader_id': 'FlyingKitty900',
1347                 'uploader': 'FlyingKitty',
1348                 'age_limit': 18,
1349                 'availability': 'needs_auth',
1350                 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1351                 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1352                 'channel': 'FlyingKitty',
1353                 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1354                 'view_count': int,
1355                 'categories': ['Entertainment'],
1356                 'live_status': 'not_live',
1357                 'tags': ['Flyingkitty', 'godzilla 2'],
1358                 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1359                 'like_count': int,
1360                 'duration': 177,
1361                 'playable_in_embed': True,
1362                 'channel_follower_count': int
1363             },
1364         },
1365         {
1366             'note': 'Age-gate video embedable only with clientScreen=EMBED',
1367             'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1368             'info_dict': {
1369                 'id': 'Tq92D6wQ1mg',
1370                 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1371                 'ext': 'mp4',
1372                 'upload_date': '20191228',
1373                 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1374                 'uploader': 'Projekt Melody',
1375                 'description': 'md5:17eccca93a786d51bc67646756894066',
1376                 'age_limit': 18,
1377                 'like_count': int,
1378                 'availability': 'needs_auth',
1379                 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1380                 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1381                 'view_count': int,
1382                 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1383                 'channel': 'Projekt Melody',
1384                 'live_status': 'not_live',
1385                 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1386                 'playable_in_embed': True,
1387                 'categories': ['Entertainment'],
1388                 'duration': 106,
1389                 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1390                 'comment_count': int,
1391                 'channel_follower_count': int
1392             },
1393         },
1394         {
1395             'note': 'Non-Agegated non-embeddable video',
1396             'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1397             'info_dict': {
1398                 'id': 'MeJVWBSsPAY',
1399                 'ext': 'mp4',
1400                 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1401                 'uploader': 'Herr Lurik',
1402                 'uploader_id': 'st3in234',
1403                 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1404                 'upload_date': '20130730',
1405                 'track': 'Such mich find mich',
1406                 'age_limit': 0,
1407                 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1408                 'like_count': int,
1409                 'playable_in_embed': False,
1410                 'creator': 'OOMPH!',
1411                 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1412                 'view_count': int,
1413                 'alt_title': 'Such mich find mich',
1414                 'duration': 210,
1415                 'channel': 'Herr Lurik',
1416                 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1417                 'categories': ['Music'],
1418                 'availability': 'public',
1419                 'uploader_url': 'http://www.youtube.com/user/st3in234',
1420                 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1421                 'live_status': 'not_live',
1422                 'artist': 'OOMPH!',
1423                 'channel_follower_count': int
1424             },
1425         },
1426         {
1427             'note': 'Non-bypassable age-gated video',
1428             'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1429             'only_matching': True,
1430         },
1431         # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1432         # YouTube Red ad is not captured for creator
1433         {
1434             'url': '__2ABJjxzNo',
1435             'info_dict': {
1436                 'id': '__2ABJjxzNo',
1437                 'ext': 'mp4',
1438                 'duration': 266,
1439                 'upload_date': '20100430',
1440                 'uploader_id': 'deadmau5',
1441                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1442                 'creator': 'deadmau5',
1443                 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1444                 'uploader': 'deadmau5',
1445                 'title': 'Deadmau5 - Some Chords (HD)',
1446                 'alt_title': 'Some Chords',
1447                 'availability': 'public',
1448                 'tags': 'count:14',
1449                 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1450                 'view_count': int,
1451                 'live_status': 'not_live',
1452                 'channel': 'deadmau5',
1453                 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1454                 'like_count': int,
1455                 'track': 'Some Chords',
1456                 'artist': 'deadmau5',
1457                 'playable_in_embed': True,
1458                 'age_limit': 0,
1459                 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1460                 'categories': ['Music'],
1461                 'album': 'Some Chords',
1462                 'channel_follower_count': int
1463             },
1464             'expected_warnings': [
1465                 'DASH manifest missing',
1466             ]
1467         },
1468         # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1469         {
1470             'url': 'lqQg6PlCWgI',
1471             'info_dict': {
1472                 'id': 'lqQg6PlCWgI',
1473                 'ext': 'mp4',
1474                 'duration': 6085,
1475                 'upload_date': '20150827',
1476                 'uploader_id': 'olympic',
1477                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1478                 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1479                 'uploader': 'Olympics',
1480                 'title': 'Hockey - Women -  GER-AUS - London 2012 Olympic Games',
1481                 'like_count': int,
1482                 'release_timestamp': 1343767800,
1483                 'playable_in_embed': True,
1484                 'categories': ['Sports'],
1485                 'release_date': '20120731',
1486                 'channel': 'Olympics',
1487                 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1488                 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1489                 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1490                 'age_limit': 0,
1491                 'availability': 'public',
1492                 'live_status': 'was_live',
1493                 'view_count': int,
1494                 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1495                 'channel_follower_count': int
1496             },
1497             'params': {
1498                 'skip_download': 'requires avconv',
1499             }
1500         },
1501         # Non-square pixels
1502         {
1503             'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1504             'info_dict': {
1505                 'id': '_b-2C3KPAM0',
1506                 'ext': 'mp4',
1507                 'stretched_ratio': 16 / 9.,
1508                 'duration': 85,
1509                 'upload_date': '20110310',
1510                 'uploader_id': 'AllenMeow',
1511                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1512                 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
1513                 'uploader': '孫ᄋᄅ',
1514                 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1515                 'playable_in_embed': True,
1516                 'channel': '孫ᄋᄅ',
1517                 'age_limit': 0,
1518                 'tags': 'count:11',
1519                 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1520                 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1521                 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1522                 'view_count': int,
1523                 'categories': ['People & Blogs'],
1524                 'like_count': int,
1525                 'live_status': 'not_live',
1526                 'availability': 'unlisted',
1527                 'comment_count': int,
1528                 'channel_follower_count': int
1529             },
1530         },
1531         # url_encoded_fmt_stream_map is empty string
1532         {
1533             'url': 'qEJwOuvDf7I',
1534             'info_dict': {
1535                 'id': 'qEJwOuvDf7I',
1536                 'ext': 'webm',
1537                 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1538                 'description': '',
1539                 'upload_date': '20150404',
1540                 'uploader_id': 'spbelect',
1541                 'uploader': 'Наблюдатели Петербурга',
1542             },
1543             'params': {
1544                 'skip_download': 'requires avconv',
1545             },
1546             'skip': 'This live event has ended.',
1547         },
1548         # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1549         {
1550             'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1551             'info_dict': {
1552                 'id': 'FIl7x6_3R5Y',
1553                 'ext': 'webm',
1554                 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1555                 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1556                 'duration': 220,
1557                 'upload_date': '20150625',
1558                 'uploader_id': 'dorappi2000',
1559                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1560                 'uploader': 'dorappi2000',
1561                 'formats': 'mincount:31',
1562             },
1563             'skip': 'not actual anymore',
1564         },
1565         # DASH manifest with segment_list
1566         {
1567             'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1568             'md5': '8ce563a1d667b599d21064e982ab9e31',
1569             'info_dict': {
1570                 'id': 'CsmdDsKjzN8',
1571                 'ext': 'mp4',
1572                 'upload_date': '20150501',  # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1573                 'uploader': 'Airtek',
1574                 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1575                 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1576                 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1577             },
1578             'params': {
1579                 'youtube_include_dash_manifest': True,
1580                 'format': '135',  # bestvideo
1581             },
1582             'skip': 'This live event has ended.',
1583         },
1584         {
1585             # Multifeed videos (multiple cameras), URL can be of any Camera
1586             'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1587             'info_dict': {
1588                 'id': 'zaPI8MvL8pg',
1589                 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1590                 'description': 'md5:563ccbc698b39298481ca3c571169519',
1591             },
1592             'playlist': [{
1593                 'info_dict': {
1594                     'id': 'j5yGuxZ8lLU',
1595                     'ext': 'mp4',
1596                     'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1597                     'uploader': 'WiiLikeToPlay',
1598                     'description': 'md5:563ccbc698b39298481ca3c571169519',
1599                     'uploader_url': 'http://www.youtube.com/user/WiiRikeToPray',
1600                     'duration': 10120,
1601                     'channel_follower_count': int,
1602                     'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1603                     'availability': 'public',
1604                     'playable_in_embed': True,
1605                     'upload_date': '20131105',
1606                     'uploader_id': 'WiiRikeToPray',
1607                     'categories': ['Gaming'],
1608                     'live_status': 'was_live',
1609                     'tags': 'count:24',
1610                     'release_timestamp': 1383701910,
1611                     'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1612                     'comment_count': int,
1613                     'age_limit': 0,
1614                     'like_count': int,
1615                     'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1616                     'channel': 'WiiLikeToPlay',
1617                     'view_count': int,
1618                     'release_date': '20131106',
1619                 },
1620             }, {
1621                 'info_dict': {
1622                     'id': 'zaPI8MvL8pg',
1623                     'ext': 'mp4',
1624                     'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1625                     'uploader_id': 'WiiRikeToPray',
1626                     'availability': 'public',
1627                     'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1628                     'channel': 'WiiLikeToPlay',
1629                     'uploader_url': 'http://www.youtube.com/user/WiiRikeToPray',
1630                     'channel_follower_count': int,
1631                     'description': 'md5:563ccbc698b39298481ca3c571169519',
1632                     'duration': 10108,
1633                     'age_limit': 0,
1634                     'like_count': int,
1635                     'tags': 'count:24',
1636                     'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1637                     'uploader': 'WiiLikeToPlay',
1638                     'release_timestamp': 1383701915,
1639                     'comment_count': int,
1640                     'upload_date': '20131105',
1641                     'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1642                     'release_date': '20131106',
1643                     'playable_in_embed': True,
1644                     'live_status': 'was_live',
1645                     'categories': ['Gaming'],
1646                     'view_count': int,
1647                 },
1648             }, {
1649                 'info_dict': {
1650                     'id': 'R7r3vfO7Hao',
1651                     'ext': 'mp4',
1652                     'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1653                     'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1654                     'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1655                     'like_count': int,
1656                     'availability': 'public',
1657                     'playable_in_embed': True,
1658                     'upload_date': '20131105',
1659                     'description': 'md5:563ccbc698b39298481ca3c571169519',
1660                     'uploader_id': 'WiiRikeToPray',
1661                     'uploader_url': 'http://www.youtube.com/user/WiiRikeToPray',
1662                     'channel_follower_count': int,
1663                     'tags': 'count:24',
1664                     'release_date': '20131106',
1665                     'uploader': 'WiiLikeToPlay',
1666                     'comment_count': int,
1667                     'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1668                     'channel': 'WiiLikeToPlay',
1669                     'categories': ['Gaming'],
1670                     'release_timestamp': 1383701914,
1671                     'live_status': 'was_live',
1672                     'age_limit': 0,
1673                     'duration': 10128,
1674                     'view_count': int,
1675                 },
1676             }],
1677             'params': {'skip_download': True},
1678         },
1679         {
1680             # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1681             'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1682             'info_dict': {
1683                 'id': 'gVfLd0zydlo',
1684                 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1685             },
1686             'playlist_count': 2,
1687             'skip': 'Not multifeed anymore',
1688         },
1689         {
1690             'url': 'https://vid.plus/FlRa-iH7PGw',
1691             'only_matching': True,
1692         },
1693         {
1694             'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1695             'only_matching': True,
1696         },
1697         {
1698             # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1699             # Also tests cut-off URL expansion in video description (see
1700             # https://github.com/ytdl-org/youtube-dl/issues/1892,
1701             # https://github.com/ytdl-org/youtube-dl/issues/8164)
1702             'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1703             'info_dict': {
1704                 'id': 'lsguqyKfVQg',
1705                 'ext': 'mp4',
1706                 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1707                 'alt_title': 'Dark Walk',
1708                 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1709                 'duration': 133,
1710                 'upload_date': '20151119',
1711                 'uploader_id': 'IronSoulElf',
1712                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1713                 'uploader': 'IronSoulElf',
1714                 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1715                 'track': 'Dark Walk',
1716                 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1717                 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1718                 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1719                 'categories': ['Film & Animation'],
1720                 'view_count': int,
1721                 'live_status': 'not_live',
1722                 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1723                 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1724                 'tags': 'count:13',
1725                 'availability': 'public',
1726                 'channel': 'IronSoulElf',
1727                 'playable_in_embed': True,
1728                 'like_count': int,
1729                 'age_limit': 0,
1730                 'channel_follower_count': int
1731             },
1732             'params': {
1733                 'skip_download': True,
1734             },
1735         },
1736         {
1737             # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1738             'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1739             'only_matching': True,
1740         },
1741         {
1742             # Video with yt:stretch=17:0
1743             'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1744             'info_dict': {
1745                 'id': 'Q39EVAstoRM',
1746                 'ext': 'mp4',
1747                 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1748                 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1749                 'upload_date': '20151107',
1750                 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1751                 'uploader': 'CH GAMER DROID',
1752             },
1753             'params': {
1754                 'skip_download': True,
1755             },
1756             'skip': 'This video does not exist.',
1757         },
1758         {
1759             # Video with incomplete 'yt:stretch=16:'
1760             'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1761             'only_matching': True,
1762         },
1763         {
1764             # Video licensed under Creative Commons
1765             'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1766             'info_dict': {
1767                 'id': 'M4gD1WSo5mA',
1768                 'ext': 'mp4',
1769                 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1770                 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1771                 'duration': 721,
1772                 'upload_date': '20150128',
1773                 'uploader_id': 'BerkmanCenter',
1774                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1775                 'uploader': 'The Berkman Klein Center for Internet & Society',
1776                 'license': 'Creative Commons Attribution license (reuse allowed)',
1777                 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1778                 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1779                 'like_count': int,
1780                 'age_limit': 0,
1781                 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1782                 'channel': 'The Berkman Klein Center for Internet & Society',
1783                 'availability': 'public',
1784                 'view_count': int,
1785                 'categories': ['Education'],
1786                 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1787                 'live_status': 'not_live',
1788                 'playable_in_embed': True,
1789                 'comment_count': int,
1790                 'channel_follower_count': int,
1791                 'chapters': list,
1792             },
1793             'params': {
1794                 'skip_download': True,
1795             },
1796         },
1797         {
1798             # Channel-like uploader_url
1799             'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1800             'info_dict': {
1801                 'id': 'eQcmzGIKrzg',
1802                 'ext': 'mp4',
1803                 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1804                 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1805                 'duration': 4060,
1806                 'upload_date': '20151120',
1807                 'uploader': 'Bernie Sanders',
1808                 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1809                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1810                 'license': 'Creative Commons Attribution license (reuse allowed)',
1811                 'playable_in_embed': True,
1812                 'tags': 'count:12',
1813                 'like_count': int,
1814                 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1815                 'age_limit': 0,
1816                 'availability': 'public',
1817                 'categories': ['News & Politics'],
1818                 'channel': 'Bernie Sanders',
1819                 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1820                 'view_count': int,
1821                 'live_status': 'not_live',
1822                 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1823                 'comment_count': int,
1824                 'channel_follower_count': int,
1825                 'chapters': list,
1826             },
1827             'params': {
1828                 'skip_download': True,
1829             },
1830         },
1831         {
1832             'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1833             'only_matching': True,
1834         },
1835         {
1836             # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1837             'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1838             'only_matching': True,
1839         },
1840         {
1841             # Rental video preview
1842             'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1843             'info_dict': {
1844                 'id': 'uGpuVWrhIzE',
1845                 'ext': 'mp4',
1846                 'title': 'Piku - Trailer',
1847                 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1848                 'upload_date': '20150811',
1849                 'uploader': 'FlixMatrix',
1850                 'uploader_id': 'FlixMatrixKaravan',
1851                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1852                 'license': 'Standard YouTube License',
1853             },
1854             'params': {
1855                 'skip_download': True,
1856             },
1857             'skip': 'This video is not available.',
1858         },
1859         {
1860             # YouTube Red video with episode data
1861             'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1862             'info_dict': {
1863                 'id': 'iqKdEhx-dD4',
1864                 'ext': 'mp4',
1865                 'title': 'Isolation - Mind Field (Ep 1)',
1866                 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1867                 'duration': 2085,
1868                 'upload_date': '20170118',
1869                 'uploader': 'Vsauce',
1870                 'uploader_id': 'Vsauce',
1871                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1872                 'series': 'Mind Field',
1873                 'season_number': 1,
1874                 'episode_number': 1,
1875                 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1876                 'tags': 'count:12',
1877                 'view_count': int,
1878                 'availability': 'public',
1879                 'age_limit': 0,
1880                 'channel': 'Vsauce',
1881                 'episode': 'Episode 1',
1882                 'categories': ['Entertainment'],
1883                 'season': 'Season 1',
1884                 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1885                 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1886                 'like_count': int,
1887                 'playable_in_embed': True,
1888                 'live_status': 'not_live',
1889                 'channel_follower_count': int
1890             },
1891             'params': {
1892                 'skip_download': True,
1893             },
1894             'expected_warnings': [
1895                 'Skipping DASH manifest',
1896             ],
1897         },
1898         {
1899             # The following content has been identified by the YouTube community
1900             # as inappropriate or offensive to some audiences.
1901             'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1902             'info_dict': {
1903                 'id': '6SJNVb0GnPI',
1904                 'ext': 'mp4',
1905                 'title': 'Race Differences in Intelligence',
1906                 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1907                 'duration': 965,
1908                 'upload_date': '20140124',
1909                 'uploader': 'New Century Foundation',
1910                 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1911                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1912             },
1913             'params': {
1914                 'skip_download': True,
1915             },
1916             'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1917         },
1918         {
1919             # itag 212
1920             'url': '1t24XAntNCY',
1921             'only_matching': True,
1922         },
1923         {
1924             # geo restricted to JP
1925             'url': 'sJL6WA-aGkQ',
1926             'only_matching': True,
1927         },
1928         {
1929             'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1930             'only_matching': True,
1931         },
1932         {
1933             'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1934             'only_matching': True,
1935         },
1936         {
1937             # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1938             'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1939             'only_matching': True,
1940         },
1941         {
1942             # DRM protected
1943             'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1944             'only_matching': True,
1945         },
1946         {
1947             # Video with unsupported adaptive stream type formats
1948             'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1949             'info_dict': {
1950                 'id': 'Z4Vy8R84T1U',
1951                 'ext': 'mp4',
1952                 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1953                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1954                 'duration': 433,
1955                 'upload_date': '20130923',
1956                 'uploader': 'Amelia Putri Harwita',
1957                 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1958                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1959                 'formats': 'maxcount:10',
1960             },
1961             'params': {
1962                 'skip_download': True,
1963                 'youtube_include_dash_manifest': False,
1964             },
1965             'skip': 'not actual anymore',
1966         },
1967         {
1968             # Youtube Music Auto-generated description
1969             'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1970             'info_dict': {
1971                 'id': 'MgNrAu2pzNs',
1972                 'ext': 'mp4',
1973                 'title': 'Voyeur Girl',
1974                 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1975                 'upload_date': '20190312',
1976                 'uploader': 'Stephen - Topic',
1977                 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1978                 'artist': 'Stephen',
1979                 'track': 'Voyeur Girl',
1980                 'album': 'it\'s too much love to know my dear',
1981                 'release_date': '20190313',
1982                 'release_year': 2019,
1983                 'alt_title': 'Voyeur Girl',
1984                 'view_count': int,
1985                 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1986                 'playable_in_embed': True,
1987                 'like_count': int,
1988                 'categories': ['Music'],
1989                 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1990                 'channel': 'Stephen',
1991                 'availability': 'public',
1992                 'creator': 'Stephen',
1993                 'duration': 169,
1994                 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1995                 'age_limit': 0,
1996                 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1997                 'tags': 'count:11',
1998                 'live_status': 'not_live',
1999                 'channel_follower_count': int
2000             },
2001             'params': {
2002                 'skip_download': True,
2003             },
2004         },
2005         {
2006             'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2007             'only_matching': True,
2008         },
2009         {
2010             # invalid -> valid video id redirection
2011             'url': 'DJztXj2GPfl',
2012             'info_dict': {
2013                 'id': 'DJztXj2GPfk',
2014                 'ext': 'mp4',
2015                 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2016                 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2017                 'upload_date': '20090125',
2018                 'uploader': 'Prochorowka',
2019                 'uploader_id': 'Prochorowka',
2020                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
2021                 'artist': 'Panjabi MC',
2022                 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2023                 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2024             },
2025             'params': {
2026                 'skip_download': True,
2027             },
2028             'skip': 'Video unavailable',
2029         },
2030         {
2031             # empty description results in an empty string
2032             'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2033             'info_dict': {
2034                 'id': 'x41yOUIvK2k',
2035                 'ext': 'mp4',
2036                 'title': 'IMG 3456',
2037                 'description': '',
2038                 'upload_date': '20170613',
2039                 'uploader_id': 'ElevageOrVert',
2040                 'uploader': 'ElevageOrVert',
2041                 'view_count': int,
2042                 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2043                 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
2044                 'like_count': int,
2045                 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2046                 'tags': [],
2047                 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2048                 'availability': 'public',
2049                 'age_limit': 0,
2050                 'categories': ['Pets & Animals'],
2051                 'duration': 7,
2052                 'playable_in_embed': True,
2053                 'live_status': 'not_live',
2054                 'channel': 'ElevageOrVert',
2055                 'channel_follower_count': int
2056             },
2057             'params': {
2058                 'skip_download': True,
2059             },
2060         },
2061         {
2062             # with '};' inside yt initial data (see [1])
2063             # see [2] for an example with '};' inside ytInitialPlayerResponse
2064             # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2065             # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2066             'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2067             'info_dict': {
2068                 'id': 'CHqg6qOn4no',
2069                 'ext': 'mp4',
2070                 'title': 'Part 77   Sort a list of simple types in c#',
2071                 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2072                 'upload_date': '20130831',
2073                 'uploader_id': 'kudvenkat',
2074                 'uploader': 'kudvenkat',
2075                 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2076                 'like_count': int,
2077                 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
2078                 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2079                 'live_status': 'not_live',
2080                 'categories': ['Education'],
2081                 'availability': 'public',
2082                 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2083                 'tags': 'count:12',
2084                 'playable_in_embed': True,
2085                 'age_limit': 0,
2086                 'view_count': int,
2087                 'duration': 522,
2088                 'channel': 'kudvenkat',
2089                 'comment_count': int,
2090                 'channel_follower_count': int,
2091                 'chapters': list,
2092             },
2093             'params': {
2094                 'skip_download': True,
2095             },
2096         },
2097         {
2098             # another example of '};' in ytInitialData
2099             'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2100             'only_matching': True,
2101         },
2102         {
2103             'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2104             'only_matching': True,
2105         },
2106         {
2107             # https://github.com/ytdl-org/youtube-dl/pull/28094
2108             'url': 'OtqTfy26tG0',
2109             'info_dict': {
2110                 'id': 'OtqTfy26tG0',
2111                 'ext': 'mp4',
2112                 'title': 'Burn Out',
2113                 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2114                 'upload_date': '20141120',
2115                 'uploader': 'The Cinematic Orchestra - Topic',
2116                 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2117                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2118                 'artist': 'The Cinematic Orchestra',
2119                 'track': 'Burn Out',
2120                 'album': 'Every Day',
2121                 'like_count': int,
2122                 'live_status': 'not_live',
2123                 'alt_title': 'Burn Out',
2124                 'duration': 614,
2125                 'age_limit': 0,
2126                 'view_count': int,
2127                 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2128                 'creator': 'The Cinematic Orchestra',
2129                 'channel': 'The Cinematic Orchestra',
2130                 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2131                 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2132                 'availability': 'public',
2133                 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2134                 'categories': ['Music'],
2135                 'playable_in_embed': True,
2136                 'channel_follower_count': int
2137             },
2138             'params': {
2139                 'skip_download': True,
2140             },
2141         },
2142         {
2143             # controversial video, only works with bpctr when authenticated with cookies
2144             'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2145             'only_matching': True,
2146         },
2147         {
2148             # controversial video, requires bpctr/contentCheckOk
2149             'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2150             'info_dict': {
2151                 'id': 'SZJvDhaSDnc',
2152                 'ext': 'mp4',
2153                 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2154                 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2155                 'uploader': 'CBS Mornings',
2156                 'uploader_id': 'CBSThisMorning',
2157                 'upload_date': '20140716',
2158                 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2159                 'duration': 170,
2160                 'categories': ['News & Politics'],
2161                 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
2162                 'view_count': int,
2163                 'channel': 'CBS Mornings',
2164                 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2165                 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2166                 'age_limit': 18,
2167                 'availability': 'needs_auth',
2168                 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2169                 'like_count': int,
2170                 'live_status': 'not_live',
2171                 'playable_in_embed': True,
2172                 'channel_follower_count': int
2173             }
2174         },
2175         {
2176             # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2177             'url': 'cBvYw8_A0vQ',
2178             'info_dict': {
2179                 'id': 'cBvYw8_A0vQ',
2180                 'ext': 'mp4',
2181                 'title': '4K Ueno Okachimachi  Street  Scenes  上野御徒町歩き',
2182                 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2183                 'upload_date': '20201120',
2184                 'uploader': 'Walk around Japan',
2185                 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2186                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2187                 'duration': 1456,
2188                 'categories': ['Travel & Events'],
2189                 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2190                 'view_count': int,
2191                 'channel': 'Walk around Japan',
2192                 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2193                 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2194                 'age_limit': 0,
2195                 'availability': 'public',
2196                 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2197                 'live_status': 'not_live',
2198                 'playable_in_embed': True,
2199                 'channel_follower_count': int
2200             },
2201             'params': {
2202                 'skip_download': True,
2203             },
2204         }, {
2205             # Has multiple audio streams
2206             'url': 'WaOKSUlf4TM',
2207             'only_matching': True
2208         }, {
2209             # Requires Premium: has format 141 when requested using YTM url
2210             'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2211             'only_matching': True
2212         }, {
2213             # multiple subtitles with same lang_code
2214             'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2215             'only_matching': True,
2216         }, {
2217             # Force use android client fallback
2218             'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2219             'info_dict': {
2220                 'id': 'YOelRv7fMxY',
2221                 'title': 'DIGGING A SECRET TUNNEL Part 1',
2222                 'ext': '3gp',
2223                 'upload_date': '20210624',
2224                 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2225                 'uploader': 'colinfurze',
2226                 'uploader_id': 'colinfurze',
2227                 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2228                 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2229                 'duration': 596,
2230                 'categories': ['Entertainment'],
2231                 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2232                 'view_count': int,
2233                 'channel': 'colinfurze',
2234                 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2235                 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2236                 'age_limit': 0,
2237                 'availability': 'public',
2238                 'like_count': int,
2239                 'live_status': 'not_live',
2240                 'playable_in_embed': True,
2241                 'channel_follower_count': int,
2242                 'chapters': list,
2243             },
2244             'params': {
2245                 'format': '17',  # 3gp format available on android
2246                 'extractor_args': {'youtube': {'player_client': ['android']}},
2247             },
2248         },
2249         {
2250             # Skip download of additional client configs (remix client config in this case)
2251             'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2252             'only_matching': True,
2253             'params': {
2254                 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2255             },
2256         }, {
2257             # shorts
2258             'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2259             'only_matching': True,
2260         }, {
2261             'note': 'Storyboards',
2262             'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2263             'info_dict': {
2264                 'id': '5KLPxDtMqe8',
2265                 'ext': 'mhtml',
2266                 'format_id': 'sb0',
2267                 'title': 'Your Brain is Plastic',
2268                 'uploader_id': 'scishow',
2269                 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2270                 'upload_date': '20140324',
2271                 'uploader': 'SciShow',
2272                 'like_count': int,
2273                 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2274                 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2275                 'view_count': int,
2276                 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2277                 'playable_in_embed': True,
2278                 'tags': 'count:12',
2279                 'uploader_url': 'http://www.youtube.com/user/scishow',
2280                 'availability': 'public',
2281                 'channel': 'SciShow',
2282                 'live_status': 'not_live',
2283                 'duration': 248,
2284                 'categories': ['Education'],
2285                 'age_limit': 0,
2286                 'channel_follower_count': int,
2287                 'chapters': list,
2288             }, 'params': {'format': 'mhtml', 'skip_download': True}
2289         }, {
2290             # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2291             'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2292             'info_dict': {
2293                 'id': '2NUZ8W2llS4',
2294                 'ext': 'mp4',
2295                 'title': 'The NP that test your phone performance 🙂',
2296                 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2297                 'uploader': 'Leon Nguyen',
2298                 'uploader_id': 'VNSXIII',
2299                 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2300                 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2301                 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2302                 'duration': 21,
2303                 'view_count': int,
2304                 'age_limit': 0,
2305                 'categories': ['Gaming'],
2306                 'tags': 'count:23',
2307                 'playable_in_embed': True,
2308                 'live_status': 'not_live',
2309                 'upload_date': '20220103',
2310                 'like_count': int,
2311                 'availability': 'public',
2312                 'channel': 'Leon Nguyen',
2313                 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2314                 'comment_count': int,
2315                 'channel_follower_count': int
2316             }
2317         }, {
2318             # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2319             'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2320             'info_dict': {
2321                 'id': '2NUZ8W2llS4',
2322                 'ext': 'mp4',
2323                 'title': 'The NP that test your phone performance 🙂',
2324                 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2325                 'uploader': 'Leon Nguyen',
2326                 'uploader_id': 'VNSXIII',
2327                 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2328                 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2329                 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2330                 'duration': 21,
2331                 'view_count': int,
2332                 'age_limit': 0,
2333                 'categories': ['Gaming'],
2334                 'tags': 'count:23',
2335                 'playable_in_embed': True,
2336                 'live_status': 'not_live',
2337                 'upload_date': '20220102',
2338                 'like_count': int,
2339                 'availability': 'public',
2340                 'channel': 'Leon Nguyen',
2341                 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2342                 'comment_count': int,
2343                 'channel_follower_count': int
2344             },
2345             'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2346         }, {
2347             # date text is premiered video, ensure upload date in UTC (published 1641172509)
2348             'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2349             'info_dict': {
2350                 'id': 'mzZzzBU6lrM',
2351                 'ext': 'mp4',
2352                 'title': 'I Met GeorgeNotFound In Real Life...',
2353                 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2354                 'uploader': 'Quackity',
2355                 'uploader_id': 'QuackityHQ',
2356                 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2357                 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2358                 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2359                 'duration': 955,
2360                 'view_count': int,
2361                 'age_limit': 0,
2362                 'categories': ['Entertainment'],
2363                 'tags': 'count:26',
2364                 'playable_in_embed': True,
2365                 'live_status': 'not_live',
2366                 'release_timestamp': 1641172509,
2367                 'release_date': '20220103',
2368                 'upload_date': '20220103',
2369                 'like_count': int,
2370                 'availability': 'public',
2371                 'channel': 'Quackity',
2372                 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2373                 'channel_follower_count': int
2374             }
2375         },
2376         {   # continuous livestream. Microformat upload date should be preferred.
2377             # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2378             'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2379             'info_dict': {
2380                 'id': 'kgx4WGK0oNU',
2381                 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2382                 'ext': 'mp4',
2383                 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2384                 'availability': 'public',
2385                 'age_limit': 0,
2386                 'release_timestamp': 1637975704,
2387                 'upload_date': '20210619',
2388                 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2389                 'live_status': 'is_live',
2390                 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2391                 'uploader': '阿鲍Abao',
2392                 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2393                 'channel': 'Abao in Tokyo',
2394                 'channel_follower_count': int,
2395                 'release_date': '20211127',
2396                 'tags': 'count:39',
2397                 'categories': ['People & Blogs'],
2398                 'like_count': int,
2399                 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2400                 'view_count': int,
2401                 'playable_in_embed': True,
2402                 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2403                 'concurrent_view_count': int,
2404             },
2405             'params': {'skip_download': True}
2406         }, {
2407             # Story. Requires specific player params to work.
2408             'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2409             'info_dict': {
2410                 'id': 'vv8qTUWmulI',
2411                 'ext': 'mp4',
2412                 'availability': 'unlisted',
2413                 'view_count': int,
2414                 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2415                 'upload_date': '20220526',
2416                 'categories': ['Education'],
2417                 'title': 'Story',
2418                 'channel': 'IT\'S HISTORY',
2419                 'description': '',
2420                 'uploader_id': 'BlastfromthePast',
2421                 'duration': 12,
2422                 'uploader': 'IT\'S HISTORY',
2423                 'playable_in_embed': True,
2424                 'age_limit': 0,
2425                 'live_status': 'not_live',
2426                 'tags': [],
2427                 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2428                 'uploader_url': 'http://www.youtube.com/user/BlastfromthePast',
2429                 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2430             },
2431             'skip': 'stories get removed after some period of time',
2432         }, {
2433             'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2434             'info_dict': {
2435                 'id': 'tjjjtzRLHvA',
2436                 'ext': 'mp4',
2437                 'title': 'ハッシュタグ無し };if window.ytcsi',
2438                 'upload_date': '20220323',
2439                 'like_count': int,
2440                 'availability': 'unlisted',
2441                 'channel': 'nao20010128nao',
2442                 'thumbnail': 'https://i.ytimg.com/vi_webp/tjjjtzRLHvA/maxresdefault.webp',
2443                 'age_limit': 0,
2444                 'uploader': 'nao20010128nao',
2445                 'uploader_id': 'nao20010128nao',
2446                 'categories': ['Music'],
2447                 'view_count': int,
2448                 'description': '',
2449                 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2450                 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2451                 'live_status': 'not_live',
2452                 'playable_in_embed': True,
2453                 'channel_follower_count': int,
2454                 'duration': 6,
2455                 'tags': [],
2456                 'uploader_url': 'http://www.youtube.com/user/nao20010128nao',
2457             }
2458         }, {
2459             # Prefer primary title+description language metadata by default
2460             # Do not prefer translated description if primary is empty
2461             'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2462             'info_dict': {
2463                 'id': 'el3E4MbxRqQ',
2464                 'ext': 'mp4',
2465                 'title': 'dlp test video 2 - primary sv no desc',
2466                 'description': '',
2467                 'channel': 'cole-dlp-test-acc',
2468                 'tags': [],
2469                 'view_count': int,
2470                 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2471                 'like_count': int,
2472                 'playable_in_embed': True,
2473                 'availability': 'unlisted',
2474                 'thumbnail': 'https://i.ytimg.com/vi_webp/el3E4MbxRqQ/maxresdefault.webp',
2475                 'age_limit': 0,
2476                 'duration': 5,
2477                 'uploader_id': 'UCiu-3thuViMebBjw_5nWYrA',
2478                 'uploader_url': 'http://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2479                 'live_status': 'not_live',
2480                 'upload_date': '20220908',
2481                 'categories': ['People & Blogs'],
2482                 'uploader': 'cole-dlp-test-acc',
2483                 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2484             },
2485             'params': {'skip_download': True}
2486         }, {
2487             # Extractor argument: prefer translated title+description
2488             'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2489             'info_dict': {
2490                 'id': 'gHKT4uU8Zng',
2491                 'ext': 'mp4',
2492                 'channel': 'cole-dlp-test-acc',
2493                 'tags': [],
2494                 'duration': 5,
2495                 'live_status': 'not_live',
2496                 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2497                 'upload_date': '20220728',
2498                 'uploader_id': 'UCiu-3thuViMebBjw_5nWYrA',
2499                 'view_count': int,
2500                 'categories': ['People & Blogs'],
2501                 'thumbnail': 'https://i.ytimg.com/vi_webp/gHKT4uU8Zng/maxresdefault.webp',
2502                 'title': 'dlp test video title translated (fr)',
2503                 'availability': 'public',
2504                 'uploader': 'cole-dlp-test-acc',
2505                 'age_limit': 0,
2506                 'description': 'dlp test video description translated (fr)',
2507                 'playable_in_embed': True,
2508                 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2509                 'uploader_url': 'http://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2510             },
2511             'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2512             'expected_warnings': [r'Preferring "fr" translated fields'],
2513         }, {
2514             'note': '6 channel audio',
2515             'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2516             'only_matching': True,
2517         }, {
2518             'note': 'Multiple HLS formats with same itag',
2519             'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2520             'info_dict': {
2521                 'id': 'kX3nB4PpJko',
2522                 'ext': 'mp4',
2523                 'categories': ['Entertainment'],
2524                 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2525                 'uploader_url': 'http://www.youtube.com/user/MrBeast6000',
2526                 'live_status': 'not_live',
2527                 'duration': 937,
2528                 'channel_follower_count': int,
2529                 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2530                 'title': 'Last To Take Hand Off Jet, Keeps It!',
2531                 'channel': 'MrBeast',
2532                 'playable_in_embed': True,
2533                 'view_count': int,
2534                 'upload_date': '20221112',
2535                 'uploader': 'MrBeast',
2536                 'uploader_id': 'MrBeast6000',
2537                 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2538                 'age_limit': 0,
2539                 'availability': 'public',
2540                 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2541                 'like_count': int,
2542                 'tags': [],
2543             },
2544             'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2545         }
2546     ]
2547
2548     _WEBPAGE_TESTS = [
2549         # YouTube <object> embed
2550         {
2551             'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2552             'md5': '873c81d308b979f0e23ee7e620b312a3',
2553             'info_dict': {
2554                 'id': 'msN87y-iEx0',
2555                 'ext': 'mp4',
2556                 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2557                 'upload_date': '20080526',
2558                 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2559                 'uploader': 'Christopher Sykes',
2560                 'uploader_id': 'ChristopherJSykes',
2561                 'age_limit': 0,
2562                 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2563                 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2564                 'playable_in_embed': True,
2565                 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2566                 'like_count': int,
2567                 'comment_count': int,
2568                 'channel': 'Christopher Sykes',
2569                 'live_status': 'not_live',
2570                 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2571                 'availability': 'public',
2572                 'duration': 195,
2573                 'view_count': int,
2574                 'categories': ['Science & Technology'],
2575                 'channel_follower_count': int,
2576                 'uploader_url': 'http://www.youtube.com/user/ChristopherJSykes',
2577             },
2578             'params': {
2579                 'skip_download': True,
2580             }
2581         },
2582     ]
2583
2584     @classmethod
2585     def suitable(cls, url):
2586         from ..utils import parse_qs
2587
2588         qs = parse_qs(url)
2589         if qs.get('list', [None])[0]:
2590             return False
2591         return super().suitable(url)
2592
2593     def __init__(self, *args, **kwargs):
2594         super().__init__(*args, **kwargs)
2595         self._code_cache = {}
2596         self._player_cache = {}
2597
2598     def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2599         lock = threading.Lock()
2600         start_time = time.time()
2601         formats = [f for f in formats if f.get('is_from_start')]
2602
2603         def refetch_manifest(format_id, delay):
2604             nonlocal formats, start_time, is_live
2605             if time.time() <= start_time + delay:
2606                 return
2607
2608             _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2609             video_details = traverse_obj(
2610                 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2611             microformats = traverse_obj(
2612                 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2613                 expected_type=dict, default=[])
2614             _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2615             is_live = live_status == 'is_live'
2616             start_time = time.time()
2617
2618         def mpd_feed(format_id, delay):
2619             """
2620             @returns (manifest_url, manifest_stream_number, is_live) or None
2621             """
2622             with lock:
2623                 refetch_manifest(format_id, delay)
2624
2625             f = next((f for f in formats if f['format_id'] == format_id), None)
2626             if not f:
2627                 if not is_live:
2628                     self.to_screen(f'{video_id}: Video is no longer live')
2629                 else:
2630                     self.report_warning(
2631                         f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2632                 return None
2633             return f['manifest_url'], f['manifest_stream_number'], is_live
2634
2635         for f in formats:
2636             f['is_live'] = is_live
2637             gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2638                                     live_start_time, mpd_feed, not is_live and f.copy())
2639             if is_live:
2640                 f['fragments'] = gen
2641                 f['protocol'] = 'http_dash_segments_generator'
2642             else:
2643                 f['fragments'] = LazyList(gen({}))
2644                 del f['is_from_start']
2645
2646     def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2647         FETCH_SPAN, MAX_DURATION = 5, 432000
2648
2649         mpd_url, stream_number, is_live = None, None, True
2650
2651         begin_index = 0
2652         download_start_time = ctx.get('start') or time.time()
2653
2654         lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2655         if lack_early_segments:
2656             self.report_warning(bug_reports_message(
2657                 'Starting download from the last 120 hours of the live stream since '
2658                 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2659             lack_early_segments = True
2660
2661         known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2662         fragments, fragment_base_url = None, None
2663
2664         def _extract_sequence_from_mpd(refresh_sequence, immediate):
2665             nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2666             # Obtain from MPD's maximum seq value
2667             old_mpd_url = mpd_url
2668             last_error = ctx.pop('last_error', None)
2669             expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
2670             mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2671                                                or (mpd_url, stream_number, False))
2672             if not refresh_sequence:
2673                 if expire_fast and not is_live:
2674                     return False, last_seq
2675                 elif old_mpd_url == mpd_url:
2676                     return True, last_seq
2677             if manifestless_orig_fmt:
2678                 fmt_info = manifestless_orig_fmt
2679             else:
2680                 try:
2681                     fmts, _ = self._extract_mpd_formats_and_subtitles(
2682                         mpd_url, None, note=False, errnote=False, fatal=False)
2683                 except ExtractorError:
2684                     fmts = None
2685                 if not fmts:
2686                     no_fragment_score += 2
2687                     return False, last_seq
2688                 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2689             fragments = fmt_info['fragments']
2690             fragment_base_url = fmt_info['fragment_base_url']
2691             assert fragment_base_url
2692
2693             _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2694             return True, _last_seq
2695
2696         self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2697         while is_live:
2698             fetch_time = time.time()
2699             if no_fragment_score > 30:
2700                 return
2701             if last_segment_url:
2702                 # Obtain from "X-Head-Seqnum" header value from each segment
2703                 try:
2704                     urlh = self._request_webpage(
2705                         last_segment_url, None, note=False, errnote=False, fatal=False)
2706                 except ExtractorError:
2707                     urlh = None
2708                 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2709                 if last_seq is None:
2710                     no_fragment_score += 2
2711                     last_segment_url = None
2712                     continue
2713             else:
2714                 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2715                 no_fragment_score += 2
2716                 if not should_continue:
2717                     continue
2718
2719             if known_idx > last_seq:
2720                 last_segment_url = None
2721                 continue
2722
2723             last_seq += 1
2724
2725             if begin_index < 0 and known_idx < 0:
2726                 # skip from the start when it's negative value
2727                 known_idx = last_seq + begin_index
2728             if lack_early_segments:
2729                 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2730             try:
2731                 for idx in range(known_idx, last_seq):
2732                     # do not update sequence here or you'll get skipped some part of it
2733                     should_continue, _ = _extract_sequence_from_mpd(False, False)
2734                     if not should_continue:
2735                         known_idx = idx - 1
2736                         raise ExtractorError('breaking out of outer loop')
2737                     last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2738                     yield {
2739                         'url': last_segment_url,
2740                         'fragment_count': last_seq,
2741                     }
2742                 if known_idx == last_seq:
2743                     no_fragment_score += 5
2744                 else:
2745                     no_fragment_score = 0
2746                 known_idx = last_seq
2747             except ExtractorError:
2748                 continue
2749
2750             if manifestless_orig_fmt:
2751                 # Stop at the first iteration if running for post-live manifestless;
2752                 # fragment count no longer increase since it starts
2753                 break
2754
2755             time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2756
2757     def _extract_player_url(self, *ytcfgs, webpage=None):
2758         player_url = traverse_obj(
2759             ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2760             get_all=False, expected_type=str)
2761         if not player_url:
2762             return
2763         return urljoin('https://www.youtube.com', player_url)
2764
2765     def _download_player_url(self, video_id, fatal=False):
2766         res = self._download_webpage(
2767             'https://www.youtube.com/iframe_api',
2768             note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2769         if res:
2770             player_version = self._search_regex(
2771                 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2772             if player_version:
2773                 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2774
2775     def _signature_cache_id(self, example_sig):
2776         """ Return a string representation of a signature """
2777         return '.'.join(str(len(part)) for part in example_sig.split('.'))
2778
2779     @classmethod
2780     def _extract_player_info(cls, player_url):
2781         for player_re in cls._PLAYER_INFO_RE:
2782             id_m = re.search(player_re, player_url)
2783             if id_m:
2784                 break
2785         else:
2786             raise ExtractorError('Cannot identify player %r' % player_url)
2787         return id_m.group('id')
2788
2789     def _load_player(self, video_id, player_url, fatal=True):
2790         player_id = self._extract_player_info(player_url)
2791         if player_id not in self._code_cache:
2792             code = self._download_webpage(
2793                 player_url, video_id, fatal=fatal,
2794                 note='Downloading player ' + player_id,
2795                 errnote='Download of %s failed' % player_url)
2796             if code:
2797                 self._code_cache[player_id] = code
2798         return self._code_cache.get(player_id)
2799
2800     def _extract_signature_function(self, video_id, player_url, example_sig):
2801         player_id = self._extract_player_info(player_url)
2802
2803         # Read from filesystem cache
2804         func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2805         assert os.path.basename(func_id) == func_id
2806
2807         self.write_debug(f'Extracting signature function {func_id}')
2808         cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2809
2810         if not cache_spec:
2811             code = self._load_player(video_id, player_url)
2812         if code:
2813             res = self._parse_sig_js(code)
2814             test_string = ''.join(map(chr, range(len(example_sig))))
2815             cache_spec = [ord(c) for c in res(test_string)]
2816             self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2817
2818         return lambda s: ''.join(s[i] for i in cache_spec)
2819
2820     def _print_sig_code(self, func, example_sig):
2821         if not self.get_param('youtube_print_sig_code'):
2822             return
2823
2824         def gen_sig_code(idxs):
2825             def _genslice(start, end, step):
2826                 starts = '' if start == 0 else str(start)
2827                 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2828                 steps = '' if step == 1 else (':%d' % step)
2829                 return f's[{starts}{ends}{steps}]'
2830
2831             step = None
2832             # Quelch pyflakes warnings - start will be set when step is set
2833             start = '(Never used)'
2834             for i, prev in zip(idxs[1:], idxs[:-1]):
2835                 if step is not None:
2836                     if i - prev == step:
2837                         continue
2838                     yield _genslice(start, prev, step)
2839                     step = None
2840                     continue
2841                 if i - prev in [-1, 1]:
2842                     step = i - prev
2843                     start = prev
2844                     continue
2845                 else:
2846                     yield 's[%d]' % prev
2847             if step is None:
2848                 yield 's[%d]' % i
2849             else:
2850                 yield _genslice(start, i, step)
2851
2852         test_string = ''.join(map(chr, range(len(example_sig))))
2853         cache_res = func(test_string)
2854         cache_spec = [ord(c) for c in cache_res]
2855         expr_code = ' + '.join(gen_sig_code(cache_spec))
2856         signature_id_tuple = '(%s)' % (
2857             ', '.join(str(len(p)) for p in example_sig.split('.')))
2858         code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2859                 '    return %s\n') % (signature_id_tuple, expr_code)
2860         self.to_screen('Extracted signature function:\n' + code)
2861
2862     def _parse_sig_js(self, jscode):
2863         funcname = self._search_regex(
2864             (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2865              r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2866              r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2867              r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2868              r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\);[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\)',
2869              r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2870              r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2871              # Obsolete patterns
2872              r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2873              r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2874              r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2875              r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2876              r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2877              r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2878              r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2879              r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2880             jscode, 'Initial JS player signature function name', group='sig')
2881
2882         jsi = JSInterpreter(jscode)
2883         initial_function = jsi.extract_function(funcname)
2884         return lambda s: initial_function([s])
2885
2886     def _cached(self, func, *cache_id):
2887         def inner(*args, **kwargs):
2888             if cache_id not in self._player_cache:
2889                 try:
2890                     self._player_cache[cache_id] = func(*args, **kwargs)
2891                 except ExtractorError as e:
2892                     self._player_cache[cache_id] = e
2893                 except Exception as e:
2894                     self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
2895
2896             ret = self._player_cache[cache_id]
2897             if isinstance(ret, Exception):
2898                 raise ret
2899             return ret
2900         return inner
2901
2902     def _decrypt_signature(self, s, video_id, player_url):
2903         """Turn the encrypted s field into a working signature"""
2904         extract_sig = self._cached(
2905             self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
2906         func = extract_sig(video_id, player_url, s)
2907         self._print_sig_code(func, s)
2908         return func(s)
2909
2910     def _decrypt_nsig(self, s, video_id, player_url):
2911         """Turn the encrypted n field into a working signature"""
2912         if player_url is None:
2913             raise ExtractorError('Cannot decrypt nsig without player_url')
2914         player_url = urljoin('https://www.youtube.com', player_url)
2915
2916         try:
2917             jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
2918         except ExtractorError as e:
2919             raise ExtractorError('Unable to extract nsig function code', cause=e)
2920         if self.get_param('youtube_print_sig_code'):
2921             self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2922
2923         try:
2924             extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
2925             ret = extract_nsig(jsi, func_code)(s)
2926         except JSInterpreter.Exception as e:
2927             try:
2928                 jsi = PhantomJSwrapper(self, timeout=5000)
2929             except ExtractorError:
2930                 raise e
2931             self.report_warning(
2932                 f'Native nsig extraction failed: Trying with PhantomJS\n'
2933                 f'         n = {s} ; player = {player_url}', video_id)
2934             self.write_debug(e, only_once=True)
2935
2936             args, func_body = func_code
2937             ret = jsi.execute(
2938                 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
2939                 video_id=video_id, note='Executing signature code').strip()
2940
2941         self.write_debug(f'Decrypted nsig {s} => {ret}')
2942         return ret
2943
2944     def _extract_n_function_name(self, jscode):
2945         funcname, idx = self._search_regex(
2946             r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2947             jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2948         if not idx:
2949             return funcname
2950
2951         return json.loads(js_to_json(self._search_regex(
2952             rf'var {re.escape(funcname)}\s*=\s*(\[.+?\]);', jscode,
2953             f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
2954
2955     def _extract_n_function_code(self, video_id, player_url):
2956         player_id = self._extract_player_info(player_url)
2957         func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
2958         jscode = func_code or self._load_player(video_id, player_url)
2959         jsi = JSInterpreter(jscode)
2960
2961         if func_code:
2962             return jsi, player_id, func_code
2963
2964         func_name = self._extract_n_function_name(jscode)
2965
2966         # For redundancy
2967         func_code = self._search_regex(
2968             r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
2969                      # NB: The end of the regex is intentionally kept strict
2970                      {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
2971             jscode, 'nsig function', group=('var', 'code'), default=None)
2972         if func_code:
2973             func_code = ([func_code[0]], func_code[1])
2974         else:
2975             self.write_debug('Extracting nsig function with jsinterp')
2976             func_code = jsi.extract_function_code(func_name)
2977
2978         self.cache.store('youtube-nsig', player_id, func_code)
2979         return jsi, player_id, func_code
2980
2981     def _extract_n_function_from_code(self, jsi, func_code):
2982         func = jsi.extract_function_from_code(*func_code)
2983
2984         def extract_nsig(s):
2985             try:
2986                 ret = func([s])
2987             except JSInterpreter.Exception:
2988                 raise
2989             except Exception as e:
2990                 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
2991
2992             if ret.startswith('enhanced_except_'):
2993                 raise JSInterpreter.Exception('Signature function returned an exception')
2994             return ret
2995
2996         return extract_nsig
2997
2998     def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
2999         """
3000         Extract signatureTimestamp (sts)
3001         Required to tell API what sig/player version is in use.
3002         """
3003         sts = None
3004         if isinstance(ytcfg, dict):
3005             sts = int_or_none(ytcfg.get('STS'))
3006
3007         if not sts:
3008             # Attempt to extract from player
3009             if player_url is None:
3010                 error_msg = 'Cannot extract signature timestamp without player_url.'
3011                 if fatal:
3012                     raise ExtractorError(error_msg)
3013                 self.report_warning(error_msg)
3014                 return
3015             code = self._load_player(video_id, player_url, fatal=fatal)
3016             if code:
3017                 sts = int_or_none(self._search_regex(
3018                     r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3019                     'JS player signature timestamp', group='sts', fatal=fatal))
3020         return sts
3021
3022     def _mark_watched(self, video_id, player_responses):
3023         for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3024             label = 'fully ' if is_full else ''
3025             url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3026                             expected_type=url_or_none)
3027             if not url:
3028                 self.report_warning(f'Unable to mark {label}watched')
3029                 return
3030             parsed_url = urllib.parse.urlparse(url)
3031             qs = urllib.parse.parse_qs(parsed_url.query)
3032
3033             # cpn generation algorithm is reverse engineered from base.js.
3034             # In fact it works even with dummy cpn.
3035             CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3036             cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3037
3038             # # more consistent results setting it to right before the end
3039             video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3040
3041             qs.update({
3042                 'ver': ['2'],
3043                 'cpn': [cpn],
3044                 'cmt': video_length,
3045                 'el': 'detailpage',  # otherwise defaults to "shorts"
3046             })
3047
3048             if is_full:
3049                 # these seem to mark watchtime "history" in the real world
3050                 # they're required, so send in a single value
3051                 qs.update({
3052                     'st': 0,
3053                     'et': video_length,
3054                 })
3055
3056             url = urllib.parse.urlunparse(
3057                 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3058
3059             self._download_webpage(
3060                 url, video_id, f'Marking {label}watched',
3061                 'Unable to mark watched', fatal=False)
3062
3063     @classmethod
3064     def _extract_from_webpage(cls, url, webpage):
3065         # Invidious Instances
3066         # https://github.com/yt-dlp/yt-dlp/issues/195
3067         # https://github.com/iv-org/invidious/pull/1730
3068         mobj = re.search(
3069             r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3070             webpage)
3071         if mobj:
3072             yield cls.url_result(mobj.group('url'), cls)
3073             raise cls.StopExtraction()
3074
3075         yield from super()._extract_from_webpage(url, webpage)
3076
3077         # lazyYT YouTube embed
3078         for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3079             yield cls.url_result(unescapeHTML(id_), cls, id_)
3080
3081         # Wordpress "YouTube Video Importer" plugin
3082         for m in re.findall(r'''(?x)<div[^>]+
3083                 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3084                 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3085             yield cls.url_result(m[-1], cls, m[-1])
3086
3087     @classmethod
3088     def extract_id(cls, url):
3089         video_id = cls.get_temp_id(url)
3090         if not video_id:
3091             raise ExtractorError(f'Invalid URL: {url}')
3092         return video_id
3093
3094     def _extract_chapters_from_json(self, data, duration):
3095         chapter_list = traverse_obj(
3096             data, (
3097                 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3098                 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3099             ), expected_type=list)
3100
3101         return self._extract_chapters(
3102             chapter_list,
3103             chapter_time=lambda chapter: float_or_none(
3104                 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3105             chapter_title=lambda chapter: traverse_obj(
3106                 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3107             duration=duration)
3108
3109     def _extract_chapters_from_engagement_panel(self, data, duration):
3110         content_list = traverse_obj(
3111             data,
3112             ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3113             expected_type=list, default=[])
3114         chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3115         chapter_title = lambda chapter: self._get_text(chapter, 'title')
3116
3117         return next(filter(None, (
3118             self._extract_chapters(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3119                                    chapter_time, chapter_title, duration)
3120             for contents in content_list)), [])
3121
3122     def _extract_chapters_from_description(self, description, duration):
3123         duration_re = r'(?:\d+:)?\d{1,2}:\d{2}'
3124         sep_re = r'(?m)^\s*(%s)\b\W*\s(%s)\s*$'
3125         return self._extract_chapters(
3126             re.findall(sep_re % (duration_re, r'.+?'), description or ''),
3127             chapter_time=lambda x: parse_duration(x[0]), chapter_title=lambda x: x[1],
3128             duration=duration, strict=False) or self._extract_chapters(
3129             re.findall(sep_re % (r'.+?', duration_re), description or ''),
3130             chapter_time=lambda x: parse_duration(x[1]), chapter_title=lambda x: x[0],
3131             duration=duration, strict=False)
3132
3133     def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration, strict=True):
3134         if not duration:
3135             return
3136         chapter_list = [{
3137             'start_time': chapter_time(chapter),
3138             'title': chapter_title(chapter),
3139         } for chapter in chapter_list or []]
3140         if not strict:
3141             chapter_list.sort(key=lambda c: c['start_time'] or 0)
3142
3143         chapters = [{'start_time': 0}]
3144         for idx, chapter in enumerate(chapter_list):
3145             if chapter['start_time'] is None:
3146                 self.report_warning(f'Incomplete chapter {idx}')
3147             elif chapters[-1]['start_time'] <= chapter['start_time'] <= duration:
3148                 chapters.append(chapter)
3149             elif chapter not in chapters:
3150                 self.report_warning(
3151                     f'Invalid start time ({chapter["start_time"]} < {chapters[-1]["start_time"]}) for chapter "{chapter["title"]}"')
3152         return chapters[1:]
3153
3154     def _extract_comment(self, comment_renderer, parent=None):
3155         comment_id = comment_renderer.get('commentId')
3156         if not comment_id:
3157             return
3158
3159         text = self._get_text(comment_renderer, 'contentText')
3160
3161         # Timestamp is an estimate calculated from the current time and time_text
3162         time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3163         timestamp = self._parse_time_text(time_text)
3164
3165         author = self._get_text(comment_renderer, 'authorText')
3166         author_id = try_get(comment_renderer,
3167                             lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], str)
3168
3169         votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
3170                                                        lambda x: x['likeCount']), str)) or 0
3171         author_thumbnail = try_get(comment_renderer,
3172                                    lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], str)
3173
3174         author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
3175         is_favorited = 'creatorHeart' in (try_get(
3176             comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
3177         return {
3178             'id': comment_id,
3179             'text': text,
3180             'timestamp': timestamp,
3181             'time_text': time_text,
3182             'like_count': votes,
3183             'is_favorited': is_favorited,
3184             'author': author,
3185             'author_id': author_id,
3186             'author_thumbnail': author_thumbnail,
3187             'author_is_uploader': author_is_uploader,
3188             'parent': parent or 'root'
3189         }
3190
3191     def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3192
3193         get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3194
3195         def extract_header(contents):
3196             _continuation = None
3197             for content in contents:
3198                 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3199                 expected_comment_count = self._get_count(
3200                     comments_header_renderer, 'countText', 'commentsCount')
3201
3202                 if expected_comment_count:
3203                     tracker['est_total'] = expected_comment_count
3204                     self.to_screen(f'Downloading ~{expected_comment_count} comments')
3205                 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top')  # 1 = new, 0 = top
3206
3207                 sort_menu_item = try_get(
3208                     comments_header_renderer,
3209                     lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3210                 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3211
3212                 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3213                 if not _continuation:
3214                     continue
3215
3216                 sort_text = str_or_none(sort_menu_item.get('title'))
3217                 if not sort_text:
3218                     sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3219                 self.to_screen('Sorting comments by %s' % sort_text.lower())
3220                 break
3221             return _continuation
3222
3223         def extract_thread(contents):
3224             if not parent:
3225                 tracker['current_page_thread'] = 0
3226             for content in contents:
3227                 if not parent and tracker['total_parent_comments'] >= max_parents:
3228                     yield
3229                 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3230                 comment_renderer = get_first(
3231                     (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3232                     expected_type=dict, default={})
3233
3234                 comment = self._extract_comment(comment_renderer, parent)
3235                 if not comment:
3236                     continue
3237
3238                 tracker['running_total'] += 1
3239                 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3240                 yield comment
3241
3242                 # Attempt to get the replies
3243                 comment_replies_renderer = try_get(
3244                     comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3245
3246                 if comment_replies_renderer:
3247                     tracker['current_page_thread'] += 1
3248                     comment_entries_iter = self._comment_entries(
3249                         comment_replies_renderer, ytcfg, video_id,
3250                         parent=comment.get('id'), tracker=tracker)
3251                     yield from itertools.islice(comment_entries_iter, min(
3252                         max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3253
3254         # Keeps track of counts across recursive calls
3255         if not tracker:
3256             tracker = dict(
3257                 running_total=0,
3258                 est_total=0,
3259                 current_page_thread=0,
3260                 total_parent_comments=0,
3261                 total_reply_comments=0)
3262
3263         # TODO: Deprecated
3264         # YouTube comments have a max depth of 2
3265         max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3266         if max_depth:
3267             self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3268                                                 'Set max replies in the max-comments extractor argument instead')
3269         if max_depth == 1 and parent:
3270             return
3271
3272         max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3273             lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3274
3275         continuation = self._extract_continuation(root_continuation_data)
3276
3277         response = None
3278         is_forced_continuation = False
3279         is_first_continuation = parent is None
3280         if is_first_continuation and not continuation:
3281             # Sometimes you can get comments by generating the continuation yourself,
3282             # even if YouTube initially reports them being disabled - e.g. stories comments.
3283             # Note: if the comment section is actually disabled, YouTube may return a response with
3284             # required check_get_keys missing. So we will disable that check initially in this case.
3285             continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3286             is_forced_continuation = True
3287
3288         for page_num in itertools.count(0):
3289             if not continuation:
3290                 break
3291             headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3292             comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
3293             if page_num == 0:
3294                 if is_first_continuation:
3295                     note_prefix = 'Downloading comment section API JSON'
3296                 else:
3297                     note_prefix = '    Downloading comment API JSON reply thread %d %s' % (
3298                         tracker['current_page_thread'], comment_prog_str)
3299             else:
3300                 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3301                     '       ' if parent else '', ' replies' if parent else '',
3302                     page_num, comment_prog_str)
3303             try:
3304                 response = self._extract_response(
3305                     item_id=None, query=continuation,
3306                     ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3307                     check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
3308             except ExtractorError as e:
3309                 # Ignore incomplete data error for replies if retries didn't work.
3310                 # This is to allow any other parent comments and comment threads to be downloaded.
3311                 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3312                 if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True:
3313                     self.report_warning(
3314                         'Received incomplete data for a comment reply thread and retrying did not help. '
3315                         'Ignoring to let other comments be downloaded.')
3316                 else:
3317                     raise
3318             is_forced_continuation = False
3319             continuation_contents = traverse_obj(
3320                 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
3321
3322             continuation = None
3323             for continuation_section in continuation_contents:
3324                 continuation_items = traverse_obj(
3325                     continuation_section,
3326                     (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
3327                     get_all=False, expected_type=list) or []
3328                 if is_first_continuation:
3329                     continuation = extract_header(continuation_items)
3330                     is_first_continuation = False
3331                     if continuation:
3332                         break
3333                     continue
3334
3335                 for entry in extract_thread(continuation_items):
3336                     if not entry:
3337                         return
3338                     yield entry
3339                 continuation = self._extract_continuation({'contents': continuation_items})
3340                 if continuation:
3341                     break
3342
3343         message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3344         if message and not parent and tracker['running_total'] == 0:
3345             self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3346             raise self.CommentsDisabled
3347
3348     @staticmethod
3349     def _generate_comment_continuation(video_id):
3350         """
3351         Generates initial comment section continuation token from given video id
3352         """
3353         token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3354         return base64.b64encode(token.encode()).decode()
3355
3356     def _get_comments(self, ytcfg, video_id, contents, webpage):
3357         """Entry for comment extraction"""
3358         def _real_comment_extract(contents):
3359             renderer = next((
3360                 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3361                 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3362             yield from self._comment_entries(renderer, ytcfg, video_id)
3363
3364         max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3365         return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3366
3367     @staticmethod
3368     def _get_checkok_params():
3369         return {'contentCheckOk': True, 'racyCheckOk': True}
3370
3371     @classmethod
3372     def _generate_player_context(cls, sts=None):
3373         context = {
3374             'html5Preference': 'HTML5_PREF_WANTS',
3375         }
3376         if sts is not None:
3377             context['signatureTimestamp'] = sts
3378         return {
3379             'playbackContext': {
3380                 'contentPlaybackContext': context
3381             },
3382             **cls._get_checkok_params()
3383         }
3384
3385     @staticmethod
3386     def _is_agegated(player_response):
3387         if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3388             return True
3389
3390         reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
3391         AGE_GATE_REASONS = (
3392             'confirm your age', 'age-restricted', 'inappropriate',  # reason
3393             'age_verification_required', 'age_check_required',  # status
3394         )
3395         return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3396
3397     @staticmethod
3398     def _is_unplayable(player_response):
3399         return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3400
3401     _STORY_PLAYER_PARAMS = '8AEB'
3402
3403     def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3404
3405         session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3406         syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3407         sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3408         headers = self.generate_api_headers(
3409             ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3410
3411         yt_query = {
3412             'videoId': video_id,
3413         }
3414         if smuggled_data.get('is_story') or _split_innertube_client(client)[0] == 'android':
3415             yt_query['params'] = self._STORY_PLAYER_PARAMS
3416
3417         yt_query.update(self._generate_player_context(sts))
3418         return self._extract_response(
3419             item_id=video_id, ep='player', query=yt_query,
3420             ytcfg=player_ytcfg, headers=headers, fatal=True,
3421             default_client=client,
3422             note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3423         ) or None
3424
3425     def _get_requested_clients(self, url, smuggled_data):
3426         requested_clients = []
3427         default = ['android', 'web']
3428         allowed_clients = sorted(
3429             (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3430             key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3431         for client in self._configuration_arg('player_client'):
3432             if client in allowed_clients:
3433                 requested_clients.append(client)
3434             elif client == 'default':
3435                 requested_clients.extend(default)
3436             elif client == 'all':
3437                 requested_clients.extend(allowed_clients)
3438             else:
3439                 self.report_warning(f'Skipping unsupported client {client}')
3440         if not requested_clients:
3441             requested_clients = default
3442
3443         if smuggled_data.get('is_music_url') or self.is_music_url(url):
3444             requested_clients.extend(
3445                 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3446
3447         return orderedSet(requested_clients)
3448
3449     def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3450         initial_pr = None
3451         if webpage:
3452             initial_pr = self._search_json(
3453                 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3454
3455         all_clients = set(clients)
3456         clients = clients[::-1]
3457         prs = []
3458
3459         def append_client(*client_names):
3460             """ Append the first client name that exists but not already used """
3461             for client_name in client_names:
3462                 actual_client = _split_innertube_client(client_name)[0]
3463                 if actual_client in INNERTUBE_CLIENTS:
3464                     if actual_client not in all_clients:
3465                         clients.append(client_name)
3466                         all_clients.add(actual_client)
3467                         return
3468
3469         # Android player_response does not have microFormats which are needed for
3470         # extraction of some data. So we return the initial_pr with formats
3471         # stripped out even if not requested by the user
3472         # See: https://github.com/yt-dlp/yt-dlp/issues/501
3473         if initial_pr:
3474             pr = dict(initial_pr)
3475             pr['streamingData'] = None
3476             prs.append(pr)
3477
3478         last_error = None
3479         tried_iframe_fallback = False
3480         player_url = None
3481         while clients:
3482             client, base_client, variant = _split_innertube_client(clients.pop())
3483             player_ytcfg = master_ytcfg if client == 'web' else {}
3484             if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3485                 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3486
3487             player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3488             require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3489             if 'js' in self._configuration_arg('player_skip'):
3490                 require_js_player = False
3491                 player_url = None
3492
3493             if not player_url and not tried_iframe_fallback and require_js_player:
3494                 player_url = self._download_player_url(video_id)
3495                 tried_iframe_fallback = True
3496
3497             try:
3498                 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3499                     client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3500             except ExtractorError as e:
3501                 if last_error:
3502                     self.report_warning(last_error)
3503                 last_error = e
3504                 continue
3505
3506             if pr:
3507                 # YouTube may return a different video player response than expected.
3508                 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3509                 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3510                 if pr_video_id and pr_video_id != video_id:
3511                     self.report_warning(
3512                         f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3513                 else:
3514                     prs.append(pr)
3515
3516             # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3517             if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3518                 append_client(f'{base_client}_creator')
3519             elif self._is_agegated(pr):
3520                 if variant == 'tv_embedded':
3521                     append_client(f'{base_client}_embedded')
3522                 elif not variant:
3523                     append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3524
3525         if last_error:
3526             if not len(prs):
3527                 raise last_error
3528             self.report_warning(last_error)
3529         return prs, player_url
3530
3531     def _needs_live_processing(self, live_status, duration):
3532         if (live_status == 'is_live' and self.get_param('live_from_start')
3533                 or live_status == 'post_live' and (duration or 0) > 4 * 3600):
3534             return live_status
3535
3536     def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3537         itags, stream_ids = collections.defaultdict(set), []
3538         itag_qualities, res_qualities = {}, {0: None}
3539         q = qualities([
3540             # Normally tiny is the smallest video-only formats. But
3541             # audio-only formats with unknown quality may get tagged as tiny
3542             'tiny',
3543             'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high',  # Audio only formats
3544             'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3545         ])
3546         streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3547
3548         for fmt in streaming_formats:
3549             if fmt.get('targetDurationSec'):
3550                 continue
3551
3552             itag = str_or_none(fmt.get('itag'))
3553             audio_track = fmt.get('audioTrack') or {}
3554             stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3555             if stream_id in stream_ids:
3556                 continue
3557
3558             quality = fmt.get('quality')
3559             height = int_or_none(fmt.get('height'))
3560             if quality == 'tiny' or not quality:
3561                 quality = fmt.get('audioQuality', '').lower() or quality
3562             # The 3gp format (17) in android client has a quality of "small",
3563             # but is actually worse than other formats
3564             if itag == '17':
3565                 quality = 'tiny'
3566             if quality:
3567                 if itag:
3568                     itag_qualities[itag] = quality
3569                 if height:
3570                     res_qualities[height] = quality
3571             # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3572             # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3573             # number of fragment that would subsequently requested with (`&sq=N`)
3574             if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3575                 continue
3576
3577             fmt_url = fmt.get('url')
3578             if not fmt_url:
3579                 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3580                 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3581                 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3582                 if not all((sc, fmt_url, player_url, encrypted_sig)):
3583                     continue
3584                 try:
3585                     fmt_url += '&%s=%s' % (
3586                         traverse_obj(sc, ('sp', -1)) or 'signature',
3587                         self._decrypt_signature(encrypted_sig, video_id, player_url)
3588                     )
3589                 except ExtractorError as e:
3590                     self.report_warning('Signature extraction failed: Some formats may be missing',
3591                                         video_id=video_id, only_once=True)
3592                     self.write_debug(e, only_once=True)
3593                     continue
3594
3595             query = parse_qs(fmt_url)
3596             throttled = False
3597             if query.get('n'):
3598                 try:
3599                     decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3600                     fmt_url = update_url_query(fmt_url, {
3601                         'n': decrypt_nsig(query['n'][0], video_id, player_url)
3602                     })
3603                 except ExtractorError as e:
3604                     phantomjs_hint = ''
3605                     if isinstance(e, JSInterpreter.Exception):
3606                         phantomjs_hint = (f'         Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3607                                           f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3608                     if player_url:
3609                         self.report_warning(
3610                             f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3611                             f'         n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3612                         self.write_debug(e, only_once=True)
3613                     else:
3614                         self.report_warning(
3615                             'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3616                             video_id=video_id, only_once=True)
3617                     throttled = True
3618
3619             tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3620             language_preference = (
3621                 10 if audio_track.get('audioIsDefault') and 10
3622                 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3623                 else -1)
3624             # Some formats may have much smaller duration than others (possibly damaged during encoding)
3625             # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3626             # Make sure to avoid false positives with small duration differences.
3627             # E.g. __2ABJjxzNo, ySuUZEjARPY
3628             is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3629             if is_damaged:
3630                 self.report_warning(
3631                     f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3632             dct = {
3633                 'asr': int_or_none(fmt.get('audioSampleRate')),
3634                 'filesize': int_or_none(fmt.get('contentLength')),
3635                 'format_id': itag,
3636                 'format_note': join_nonempty(
3637                     '%s%s' % (audio_track.get('displayName') or '',
3638                               ' (default)' if language_preference > 0 else ''),
3639                     fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3640                     try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3641                     try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3642                     throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3643                 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3644                 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3645                 'fps': int_or_none(fmt.get('fps')) or None,
3646                 'audio_channels': fmt.get('audioChannels'),
3647                 'height': height,
3648                 'quality': q(quality),
3649                 'has_drm': bool(fmt.get('drmFamilies')),
3650                 'tbr': tbr,
3651                 'url': fmt_url,
3652                 'width': int_or_none(fmt.get('width')),
3653                 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3654                                           'desc' if language_preference < -1 else ''),
3655                 'language_preference': language_preference,
3656                 # Strictly de-prioritize damaged and 3gp formats
3657                 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3658             }
3659             mime_mobj = re.match(
3660                 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3661             if mime_mobj:
3662                 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3663                 dct.update(parse_codecs(mime_mobj.group(2)))
3664             no_audio = dct.get('acodec') == 'none'
3665             no_video = dct.get('vcodec') == 'none'
3666             if no_audio:
3667                 dct['vbr'] = tbr
3668             if no_video:
3669                 dct['abr'] = tbr
3670             if no_audio or no_video:
3671                 dct['downloader_options'] = {
3672                     # Youtube throttles chunks >~10M
3673                     'http_chunk_size': 10485760,
3674                 }
3675                 if dct.get('ext'):
3676                     dct['container'] = dct['ext'] + '_dash'
3677
3678             if itag:
3679                 itags[itag].add(('https', dct.get('language')))
3680                 stream_ids.append(stream_id)
3681             yield dct
3682
3683         needs_live_processing = self._needs_live_processing(live_status, duration)
3684         skip_bad_formats = not self._configuration_arg('include_incomplete_formats')
3685
3686         skip_manifests = set(self._configuration_arg('skip'))
3687         if (not self.get_param('youtube_include_hls_manifest', True)
3688                 or needs_live_processing == 'is_live'  # These will be filtered out by YoutubeDL anyway
3689                 or needs_live_processing and skip_bad_formats):
3690             skip_manifests.add('hls')
3691
3692         if not self.get_param('youtube_include_dash_manifest', True):
3693             skip_manifests.add('dash')
3694         if self._configuration_arg('include_live_dash'):
3695             self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3696                                                 'Use include_incomplete_formats extractor argument instead')
3697         elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3698             skip_manifests.add('dash')
3699
3700         def process_manifest_format(f, proto, itag):
3701             key = (proto, f.get('language'))
3702             if key in itags[itag]:
3703                 return False
3704             itags[itag].add(key)
3705
3706             if any(p != proto for p, _ in itags[itag]):
3707                 f['format_id'] = f'{itag}-{proto}'
3708             elif itag:
3709                 f['format_id'] = itag
3710
3711             f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3712             if f['quality'] == -1 and f.get('height'):
3713                 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3714             return True
3715
3716         subtitles = {}
3717         for sd in streaming_data:
3718             hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3719             if hls_manifest_url:
3720                 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3721                     hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3722                 subtitles = self._merge_subtitles(subs, subtitles)
3723                 for f in fmts:
3724                     if process_manifest_format(f, 'hls', self._search_regex(
3725                             r'/itag/(\d+)', f['url'], 'itag', default=None)):
3726                         yield f
3727
3728             dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3729             if dash_manifest_url:
3730                 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3731                 subtitles = self._merge_subtitles(subs, subtitles)  # Prioritize HLS subs over DASH
3732                 for f in formats:
3733                     if process_manifest_format(f, 'dash', f['format_id']):
3734                         f['filesize'] = int_or_none(self._search_regex(
3735                             r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3736                         if needs_live_processing:
3737                             f['is_from_start'] = True
3738
3739                         yield f
3740         yield subtitles
3741
3742     def _extract_storyboard(self, player_responses, duration):
3743         spec = get_first(
3744             player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3745         base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3746         if not base_url:
3747             return
3748         L = len(spec) - 1
3749         for i, args in enumerate(spec):
3750             args = args.split('#')
3751             counts = list(map(int_or_none, args[:5]))
3752             if len(args) != 8 or not all(counts):
3753                 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3754                 continue
3755             width, height, frame_count, cols, rows = counts
3756             N, sigh = args[6:]
3757
3758             url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3759             fragment_count = frame_count / (cols * rows)
3760             fragment_duration = duration / fragment_count
3761             yield {
3762                 'format_id': f'sb{i}',
3763                 'format_note': 'storyboard',
3764                 'ext': 'mhtml',
3765                 'protocol': 'mhtml',
3766                 'acodec': 'none',
3767                 'vcodec': 'none',
3768                 'url': url,
3769                 'width': width,
3770                 'height': height,
3771                 'fps': frame_count / duration,
3772                 'rows': rows,
3773                 'columns': cols,
3774                 'fragments': [{
3775                     'url': url.replace('$M', str(j)),
3776                     'duration': min(fragment_duration, duration - (j * fragment_duration)),
3777                 } for j in range(math.ceil(fragment_count))],
3778             }
3779
3780     def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3781         webpage = None
3782         if 'webpage' not in self._configuration_arg('player_skip'):
3783             query = {'bpctr': '9999999999', 'has_verified': '1'}
3784             if smuggled_data.get('is_story'):
3785                 query['pp'] = self._STORY_PLAYER_PARAMS
3786             webpage = self._download_webpage(
3787                 webpage_url, video_id, fatal=False, query=query)
3788
3789         master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3790
3791         player_responses, player_url = self._extract_player_responses(
3792             self._get_requested_clients(url, smuggled_data),
3793             video_id, webpage, master_ytcfg, smuggled_data)
3794
3795         return webpage, master_ytcfg, player_responses, player_url
3796
3797     def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3798         live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3799         is_live = get_first(video_details, 'isLive')
3800         if is_live is None:
3801             is_live = get_first(live_broadcast_details, 'isLiveNow')
3802         live_content = get_first(video_details, 'isLiveContent')
3803         is_upcoming = get_first(video_details, 'isUpcoming')
3804         post_live = get_first(video_details, 'isPostLiveDvr')
3805         live_status = ('post_live' if post_live
3806                        else 'is_live' if is_live
3807                        else 'is_upcoming' if is_upcoming
3808                        else 'was_live' if live_content
3809                        else 'not_live' if False in (is_live, live_content)
3810                        else None)
3811         streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3812         *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
3813
3814         return live_broadcast_details, live_status, streaming_data, formats, subtitles
3815
3816     def _real_extract(self, url):
3817         url, smuggled_data = unsmuggle_url(url, {})
3818         video_id = self._match_id(url)
3819
3820         base_url = self.http_scheme() + '//www.youtube.com/'
3821         webpage_url = base_url + 'watch?v=' + video_id
3822
3823         webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3824
3825         playability_statuses = traverse_obj(
3826             player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3827
3828         trailer_video_id = get_first(
3829             playability_statuses,
3830             ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3831             expected_type=str)
3832         if trailer_video_id:
3833             return self.url_result(
3834                 trailer_video_id, self.ie_key(), trailer_video_id)
3835
3836         search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3837                        if webpage else (lambda x: None))
3838
3839         video_details = traverse_obj(
3840             player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3841         microformats = traverse_obj(
3842             player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3843             expected_type=dict, default=[])
3844
3845         translated_title = self._get_text(microformats, (..., 'title'))
3846         video_title = (self._preferred_lang and translated_title
3847                        or get_first(video_details, 'title')  # primary
3848                        or translated_title
3849                        or search_meta(['og:title', 'twitter:title', 'title']))
3850         translated_description = self._get_text(microformats, (..., 'description'))
3851         original_description = get_first(video_details, 'shortDescription')
3852         video_description = (
3853             self._preferred_lang and translated_description
3854             # If original description is blank, it will be an empty string.
3855             # Do not prefer translated description in this case.
3856             or original_description if original_description is not None else translated_description)
3857
3858         multifeed_metadata_list = get_first(
3859             player_responses,
3860             ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3861             expected_type=str)
3862         if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3863             if self.get_param('noplaylist'):
3864                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3865             else:
3866                 entries = []
3867                 feed_ids = []
3868                 for feed in multifeed_metadata_list.split(','):
3869                     # Unquote should take place before split on comma (,) since textual
3870                     # fields may contain comma as well (see
3871                     # https://github.com/ytdl-org/youtube-dl/issues/8536)
3872                     feed_data = urllib.parse.parse_qs(
3873                         urllib.parse.unquote_plus(feed))
3874
3875                     def feed_entry(name):
3876                         return try_get(
3877                             feed_data, lambda x: x[name][0], str)
3878
3879                     feed_id = feed_entry('id')
3880                     if not feed_id:
3881                         continue
3882                     feed_title = feed_entry('title')
3883                     title = video_title
3884                     if feed_title:
3885                         title += ' (%s)' % feed_title
3886                     entries.append({
3887                         '_type': 'url_transparent',
3888                         'ie_key': 'Youtube',
3889                         'url': smuggle_url(
3890                             '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3891                             {'force_singlefeed': True}),
3892                         'title': title,
3893                     })
3894                     feed_ids.append(feed_id)
3895                 self.to_screen(
3896                     'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3897                     % (', '.join(feed_ids), video_id))
3898                 return self.playlist_result(
3899                     entries, video_id, video_title, video_description)
3900
3901         duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
3902                     or int_or_none(get_first(microformats, 'lengthSeconds'))
3903                     or parse_duration(search_meta('duration')) or None)
3904
3905         live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
3906             self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
3907         if live_status == 'post_live':
3908             self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
3909
3910         if not formats:
3911             if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3912                 self.report_drm(video_id)
3913             pemr = get_first(
3914                 playability_statuses,
3915                 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3916             reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3917             subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3918             if subreason:
3919                 if subreason == 'The uploader has not made this video available in your country.':
3920                     countries = get_first(microformats, 'availableCountries')
3921                     if not countries:
3922                         regions_allowed = search_meta('regionsAllowed')
3923                         countries = regions_allowed.split(',') if regions_allowed else None
3924                     self.raise_geo_restricted(subreason, countries, metadata_available=True)
3925                 reason += f'. {subreason}'
3926             if reason:
3927                 self.raise_no_formats(reason, expected=True)
3928
3929         keywords = get_first(video_details, 'keywords', expected_type=list) or []
3930         if not keywords and webpage:
3931             keywords = [
3932                 unescapeHTML(m.group('content'))
3933                 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3934         for keyword in keywords:
3935             if keyword.startswith('yt:stretch='):
3936                 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3937                 if mobj:
3938                     # NB: float is intentional for forcing float division
3939                     w, h = (float(v) for v in mobj.groups())
3940                     if w > 0 and h > 0:
3941                         ratio = w / h
3942                         for f in formats:
3943                             if f.get('vcodec') != 'none':
3944                                 f['stretched_ratio'] = ratio
3945                         break
3946         thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3947         thumbnail_url = search_meta(['og:image', 'twitter:image'])
3948         if thumbnail_url:
3949             thumbnails.append({
3950                 'url': thumbnail_url,
3951             })
3952         original_thumbnails = thumbnails.copy()
3953
3954         # The best resolution thumbnails sometimes does not appear in the webpage
3955         # See: https://github.com/yt-dlp/yt-dlp/issues/340
3956         # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3957         thumbnail_names = [
3958             # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
3959             # in resolution, these are not the custom thumbnail. So de-prioritize them
3960             'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3961             'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3962         ]
3963         n_thumbnail_names = len(thumbnail_names)
3964         thumbnails.extend({
3965             'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3966                 video_id=video_id, name=name, ext=ext,
3967                 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
3968         } for name in thumbnail_names for ext in ('webp', 'jpg'))
3969         for thumb in thumbnails:
3970             i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3971             thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3972         self._remove_duplicate_formats(thumbnails)
3973         self._downloader._sort_thumbnails(original_thumbnails)
3974
3975         category = get_first(microformats, 'category') or search_meta('genre')
3976         channel_id = str_or_none(
3977             get_first(video_details, 'channelId')
3978             or get_first(microformats, 'externalChannelId')
3979             or search_meta('channelId'))
3980         owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3981
3982         live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3983         live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3984         if not duration and live_end_time and live_start_time:
3985             duration = live_end_time - live_start_time
3986
3987         needs_live_processing = self._needs_live_processing(live_status, duration)
3988
3989         def is_bad_format(fmt):
3990             if needs_live_processing and not fmt.get('is_from_start'):
3991                 return True
3992             elif (live_status == 'is_live' and needs_live_processing != 'is_live'
3993                     and fmt.get('protocol') == 'http_dash_segments'):
3994                 return True
3995
3996         for fmt in filter(is_bad_format, formats):
3997             fmt['preference'] = (fmt.get('preference') or -1) - 10
3998             fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 4 hours)', delim=' ')
3999
4000         if needs_live_processing:
4001             self._prepare_live_from_start_formats(
4002                 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4003
4004         formats.extend(self._extract_storyboard(player_responses, duration))
4005
4006         # source_preference is lower for throttled/potentially damaged formats
4007         self._sort_formats(formats, (
4008             'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto'))
4009
4010         info = {
4011             'id': video_id,
4012             'title': video_title,
4013             'formats': formats,
4014             'thumbnails': thumbnails,
4015             # The best thumbnail that we are sure exists. Prevents unnecessary
4016             # URL checking if user don't care about getting the best possible thumbnail
4017             'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4018             'description': video_description,
4019             'uploader': get_first(video_details, 'author'),
4020             'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
4021             'uploader_url': owner_profile_url,
4022             'channel_id': channel_id,
4023             'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s'),
4024             'duration': duration,
4025             'view_count': int_or_none(
4026                 get_first((video_details, microformats), (..., 'viewCount'))
4027                 or search_meta('interactionCount')),
4028             'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4029             'age_limit': 18 if (
4030                 get_first(microformats, 'isFamilySafe') is False
4031                 or search_meta('isFamilyFriendly') == 'false'
4032                 or search_meta('og:restrictions:age') == '18+') else 0,
4033             'webpage_url': webpage_url,
4034             'categories': [category] if category else None,
4035             'tags': keywords,
4036             'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4037             'live_status': live_status,
4038             'release_timestamp': live_start_time,
4039         }
4040
4041         subtitles = {}
4042         pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4043         if pctr:
4044             def get_lang_code(track):
4045                 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4046                         or track.get('languageCode'))
4047
4048             # Converted into dicts to remove duplicates
4049             captions = {
4050                 get_lang_code(sub): sub
4051                 for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])}
4052             translation_languages = {
4053                 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4054                 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])}
4055
4056             def process_language(container, base_url, lang_code, sub_name, query):
4057                 lang_subs = container.setdefault(lang_code, [])
4058                 for fmt in self._SUBTITLE_FORMATS:
4059                     query.update({
4060                         'fmt': fmt,
4061                     })
4062                     lang_subs.append({
4063                         'ext': fmt,
4064                         'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4065                         'name': sub_name,
4066                     })
4067
4068             # NB: Constructing the full subtitle dictionary is slow
4069             get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4070                 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4071             for lang_code, caption_track in captions.items():
4072                 base_url = caption_track.get('baseUrl')
4073                 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4074                 if not base_url:
4075                     continue
4076                 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4077                 if caption_track.get('kind') != 'asr':
4078                     if not lang_code:
4079                         continue
4080                     process_language(
4081                         subtitles, base_url, lang_code, lang_name, {})
4082                     if not caption_track.get('isTranslatable'):
4083                         continue
4084                 for trans_code, trans_name in translation_languages.items():
4085                     if not trans_code:
4086                         continue
4087                     orig_trans_code = trans_code
4088                     if caption_track.get('kind') != 'asr':
4089                         if not get_translated_subs:
4090                             continue
4091                         trans_code += f'-{lang_code}'
4092                         trans_name += format_field(lang_name, None, ' from %s')
4093                     # Add an "-orig" label to the original language so that it can be distinguished.
4094                     # The subs are returned without "-orig" as well for compatibility
4095                     if lang_code == f'a-{orig_trans_code}':
4096                         process_language(
4097                             automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4098                     # Setting tlang=lang returns damaged subtitles.
4099                     process_language(automatic_captions, base_url, trans_code, trans_name,
4100                                      {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4101
4102         info['automatic_captions'] = automatic_captions
4103         info['subtitles'] = subtitles
4104
4105         parsed_url = urllib.parse.urlparse(url)
4106         for component in [parsed_url.fragment, parsed_url.query]:
4107             query = urllib.parse.parse_qs(component)
4108             for k, v in query.items():
4109                 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4110                     d_k += '_time'
4111                     if d_k not in info and k in s_ks:
4112                         info[d_k] = parse_duration(query[k][0])
4113
4114         # Youtube Music Auto-generated description
4115         if video_description:
4116             mobj = re.search(
4117                 r'''(?xs)
4118                     (?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+
4119                     (?P<album>[^\n]+)
4120                     (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4121                     (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4122                     (.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?
4123                     .+\nAuto-generated\ by\ YouTube\.\s*$
4124                 ''', video_description)
4125             if mobj:
4126                 release_year = mobj.group('release_year')
4127                 release_date = mobj.group('release_date')
4128                 if release_date:
4129                     release_date = release_date.replace('-', '')
4130                     if not release_year:
4131                         release_year = release_date[:4]
4132                 info.update({
4133                     'album': mobj.group('album'.strip()),
4134                     'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4135                     'track': mobj.group('track').strip(),
4136                     'release_date': release_date,
4137                     'release_year': int_or_none(release_year),
4138                 })
4139
4140         initial_data = None
4141         if webpage:
4142             initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4143         if not initial_data:
4144             query = {'videoId': video_id}
4145             query.update(self._get_checkok_params())
4146             initial_data = self._extract_response(
4147                 item_id=video_id, ep='next', fatal=False,
4148                 ytcfg=master_ytcfg, query=query,
4149                 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4150                 note='Downloading initial data API JSON')
4151
4152         info['comment_count'] = traverse_obj(initial_data, (
4153             'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4154             'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount', 'simpleText'
4155         ), (
4156             'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4157             'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo', 'runs', ..., 'text'
4158         ), expected_type=int_or_none, get_all=False)
4159
4160         try:  # This will error if there is no livechat
4161             initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4162         except (KeyError, IndexError, TypeError):
4163             pass
4164         else:
4165             info.setdefault('subtitles', {})['live_chat'] = [{
4166                 # url is needed to set cookies
4167                 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4168                 'video_id': video_id,
4169                 'ext': 'json',
4170                 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4171                              else 'youtube_live_chat_replay'),
4172             }]
4173
4174         if initial_data:
4175             info['chapters'] = (
4176                 self._extract_chapters_from_json(initial_data, duration)
4177                 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4178                 or self._extract_chapters_from_description(video_description, duration)
4179                 or None)
4180
4181         contents = traverse_obj(
4182             initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4183             expected_type=list, default=[])
4184
4185         vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4186         if vpir:
4187             stl = vpir.get('superTitleLink')
4188             if stl:
4189                 stl = self._get_text(stl)
4190                 if try_get(
4191                         vpir,
4192                         lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4193                     info['location'] = stl
4194                 else:
4195                     mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4196                     if mobj:
4197                         info.update({
4198                             'series': mobj.group(1),
4199                             'season_number': int(mobj.group(2)),
4200                             'episode_number': int(mobj.group(3)),
4201                         })
4202             for tlb in (try_get(
4203                     vpir,
4204                     lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4205                     list) or []):
4206                 tbrs = variadic(
4207                     traverse_obj(
4208                         tlb, 'toggleButtonRenderer',
4209                         ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer'),
4210                         default=[]))
4211                 for tbr in tbrs:
4212                     for getter, regex in [(
4213                             lambda x: x['defaultText']['accessibility']['accessibilityData'],
4214                             r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4215                                 lambda x: x['accessibility'],
4216                                 lambda x: x['accessibilityData']['accessibilityData'],
4217                             ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4218                         label = (try_get(tbr, getter, dict) or {}).get('label')
4219                         if label:
4220                             mobj = re.match(regex, label)
4221                             if mobj:
4222                                 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4223                                 break
4224             sbr_tooltip = try_get(
4225                 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4226             if sbr_tooltip:
4227                 like_count, dislike_count = sbr_tooltip.split(' / ')
4228                 info.update({
4229                     'like_count': str_to_int(like_count),
4230                     'dislike_count': str_to_int(dislike_count),
4231                 })
4232             vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4233             if vcr:
4234                 vc = self._get_count(vcr, 'viewCount')
4235                 # Upcoming premieres with waiting count are treated as live here
4236                 if vcr.get('isLive'):
4237                     info['concurrent_view_count'] = vc
4238                 elif info.get('view_count') is None:
4239                     info['view_count'] = vc
4240
4241         vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4242         if vsir:
4243             vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4244             info.update({
4245                 'channel': self._get_text(vor, 'title'),
4246                 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4247
4248             rows = try_get(
4249                 vsir,
4250                 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4251                 list) or []
4252             multiple_songs = False
4253             for row in rows:
4254                 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4255                     multiple_songs = True
4256                     break
4257             for row in rows:
4258                 mrr = row.get('metadataRowRenderer') or {}
4259                 mrr_title = mrr.get('title')
4260                 if not mrr_title:
4261                     continue
4262                 mrr_title = self._get_text(mrr, 'title')
4263                 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4264                 if mrr_title == 'License':
4265                     info['license'] = mrr_contents_text
4266                 elif not multiple_songs:
4267                     if mrr_title == 'Album':
4268                         info['album'] = mrr_contents_text
4269                     elif mrr_title == 'Artist':
4270                         info['artist'] = mrr_contents_text
4271                     elif mrr_title == 'Song':
4272                         info['track'] = mrr_contents_text
4273
4274         fallbacks = {
4275             'channel': 'uploader',
4276             'channel_id': 'uploader_id',
4277             'channel_url': 'uploader_url',
4278         }
4279
4280         # The upload date for scheduled, live and past live streams / premieres in microformats
4281         # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4282         # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4283         upload_date = (
4284             unified_strdate(get_first(microformats, 'uploadDate'))
4285             or unified_strdate(search_meta('uploadDate')))
4286         if not upload_date or (
4287             live_status in ('not_live', None)
4288             and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4289         ):
4290             upload_date = strftime_or_none(
4291                 self._parse_time_text(self._get_text(vpir, 'dateText')), '%Y%m%d') or upload_date
4292         info['upload_date'] = upload_date
4293
4294         for to, frm in fallbacks.items():
4295             if not info.get(to):
4296                 info[to] = info.get(frm)
4297
4298         for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4299             v = info.get(s_k)
4300             if v:
4301                 info[d_k] = v
4302
4303         badges = self._extract_badges(traverse_obj(contents, (..., 'videoPrimaryInfoRenderer'), get_all=False))
4304
4305         is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4306                       or get_first(video_details, 'isPrivate', expected_type=bool))
4307
4308         info['availability'] = (
4309             'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4310             else self._availability(
4311                 is_private=is_private,
4312                 needs_premium=(
4313                     self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4314                     or False if initial_data and is_private is not None else None),
4315                 needs_subscription=(
4316                     self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4317                     or False if initial_data and is_private is not None else None),
4318                 needs_auth=info['age_limit'] >= 18,
4319                 is_unlisted=None if is_private is None else (
4320                     self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4321                     or get_first(microformats, 'isUnlisted', expected_type=bool))))
4322
4323         info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4324
4325         self.mark_watched(video_id, player_responses)
4326
4327         return info
4328
4329
4330 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4331     @staticmethod
4332     def passthrough_smuggled_data(func):
4333         def _smuggle(info, smuggled_data):
4334             if info.get('_type') not in ('url', 'url_transparent'):
4335                 return info
4336             if smuggled_data.get('is_music_url'):
4337                 parsed_url = urllib.parse.urlparse(info['url'])
4338                 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4339                     smuggled_data.pop('is_music_url')
4340                     info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4341             if smuggled_data:
4342                 info['url'] = smuggle_url(info['url'], smuggled_data)
4343             return info
4344
4345         @functools.wraps(func)
4346         def wrapper(self, url):
4347             url, smuggled_data = unsmuggle_url(url, {})
4348             if self.is_music_url(url):
4349                 smuggled_data['is_music_url'] = True
4350             info_dict = func(self, url, smuggled_data)
4351             if smuggled_data:
4352                 _smuggle(info_dict, smuggled_data)
4353                 if info_dict.get('entries'):
4354                     info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4355             return info_dict
4356         return wrapper
4357
4358     def _extract_channel_id(self, webpage):
4359         channel_id = self._html_search_meta(
4360             'channelId', webpage, 'channel id', default=None)
4361         if channel_id:
4362             return channel_id
4363         channel_url = self._html_search_meta(
4364             ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
4365              'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
4366              'twitter:app:url:googleplay'), webpage, 'channel url')
4367         return self._search_regex(
4368             r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
4369             channel_url, 'channel id')
4370
4371     @staticmethod
4372     def _extract_basic_item_renderer(item):
4373         # Modified from _extract_grid_item_renderer
4374         known_basic_renderers = (
4375             'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4376         )
4377         for key, renderer in item.items():
4378             if not isinstance(renderer, dict):
4379                 continue
4380             elif key in known_basic_renderers:
4381                 return renderer
4382             elif key.startswith('grid') and key.endswith('Renderer'):
4383                 return renderer
4384
4385     def _grid_entries(self, grid_renderer):
4386         for item in grid_renderer['items']:
4387             if not isinstance(item, dict):
4388                 continue
4389             renderer = self._extract_basic_item_renderer(item)
4390             if not isinstance(renderer, dict):
4391                 continue
4392             title = self._get_text(renderer, 'title')
4393
4394             # playlist
4395             playlist_id = renderer.get('playlistId')
4396             if playlist_id:
4397                 yield self.url_result(
4398                     'https://www.youtube.com/playlist?list=%s' % playlist_id,
4399                     ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4400                     video_title=title)
4401                 continue
4402             # video
4403             video_id = renderer.get('videoId')
4404             if video_id:
4405                 yield self._extract_video(renderer)
4406                 continue
4407             # channel
4408             channel_id = renderer.get('channelId')
4409             if channel_id:
4410                 yield self.url_result(
4411                     'https://www.youtube.com/channel/%s' % channel_id,
4412                     ie=YoutubeTabIE.ie_key(), video_title=title)
4413                 continue
4414             # generic endpoint URL support
4415             ep_url = urljoin('https://www.youtube.com/', try_get(
4416                 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4417                 str))
4418             if ep_url:
4419                 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4420                     if ie.suitable(ep_url):
4421                         yield self.url_result(
4422                             ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4423                         break
4424
4425     def _music_reponsive_list_entry(self, renderer):
4426         video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4427         if video_id:
4428             return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4429                                    ie=YoutubeIE.ie_key(), video_id=video_id)
4430         playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4431         if playlist_id:
4432             video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4433             if video_id:
4434                 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4435                                        ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4436             return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4437                                    ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4438         browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4439         if browse_id:
4440             return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4441                                    ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4442
4443     def _shelf_entries_from_content(self, shelf_renderer):
4444         content = shelf_renderer.get('content')
4445         if not isinstance(content, dict):
4446             return
4447         renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4448         if renderer:
4449             # TODO: add support for nested playlists so each shelf is processed
4450             # as separate playlist
4451             # TODO: this includes only first N items
4452             yield from self._grid_entries(renderer)
4453         renderer = content.get('horizontalListRenderer')
4454         if renderer:
4455             # TODO
4456             pass
4457
4458     def _shelf_entries(self, shelf_renderer, skip_channels=False):
4459         ep = try_get(
4460             shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4461             str)
4462         shelf_url = urljoin('https://www.youtube.com', ep)
4463         if shelf_url:
4464             # Skipping links to another channels, note that checking for
4465             # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4466             # will not work
4467             if skip_channels and '/channels?' in shelf_url:
4468                 return
4469             title = self._get_text(shelf_renderer, 'title')
4470             yield self.url_result(shelf_url, video_title=title)
4471         # Shelf may not contain shelf URL, fallback to extraction from content
4472         yield from self._shelf_entries_from_content(shelf_renderer)
4473
4474     def _playlist_entries(self, video_list_renderer):
4475         for content in video_list_renderer['contents']:
4476             if not isinstance(content, dict):
4477                 continue
4478             renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4479             if not isinstance(renderer, dict):
4480                 continue
4481             video_id = renderer.get('videoId')
4482             if not video_id:
4483                 continue
4484             yield self._extract_video(renderer)
4485
4486     def _rich_entries(self, rich_grid_renderer):
4487         renderer = traverse_obj(
4488             rich_grid_renderer, ('content', ('videoRenderer', 'reelItemRenderer')), get_all=False) or {}
4489         video_id = renderer.get('videoId')
4490         if not video_id:
4491             return
4492         yield self._extract_video(renderer)
4493
4494     def _video_entry(self, video_renderer):
4495         video_id = video_renderer.get('videoId')
4496         if video_id:
4497             return self._extract_video(video_renderer)
4498
4499     def _hashtag_tile_entry(self, hashtag_tile_renderer):
4500         url = urljoin('https://youtube.com', traverse_obj(
4501             hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4502         if url:
4503             return self.url_result(
4504                 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4505
4506     def _post_thread_entries(self, post_thread_renderer):
4507         post_renderer = try_get(
4508             post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4509         if not post_renderer:
4510             return
4511         # video attachment
4512         video_renderer = try_get(
4513             post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4514         video_id = video_renderer.get('videoId')
4515         if video_id:
4516             entry = self._extract_video(video_renderer)
4517             if entry:
4518                 yield entry
4519         # playlist attachment
4520         playlist_id = try_get(
4521             post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4522         if playlist_id:
4523             yield self.url_result(
4524                 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4525                 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4526         # inline video links
4527         runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4528         for run in runs:
4529             if not isinstance(run, dict):
4530                 continue
4531             ep_url = try_get(
4532                 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4533             if not ep_url:
4534                 continue
4535             if not YoutubeIE.suitable(ep_url):
4536                 continue
4537             ep_video_id = YoutubeIE._match_id(ep_url)
4538             if video_id == ep_video_id:
4539                 continue
4540             yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4541
4542     def _post_thread_continuation_entries(self, post_thread_continuation):
4543         contents = post_thread_continuation.get('contents')
4544         if not isinstance(contents, list):
4545             return
4546         for content in contents:
4547             renderer = content.get('backstagePostThreadRenderer')
4548             if isinstance(renderer, dict):
4549                 yield from self._post_thread_entries(renderer)
4550                 continue
4551             renderer = content.get('videoRenderer')
4552             if isinstance(renderer, dict):
4553                 yield self._video_entry(renderer)
4554
4555     r''' # unused
4556     def _rich_grid_entries(self, contents):
4557         for content in contents:
4558             video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4559             if video_renderer:
4560                 entry = self._video_entry(video_renderer)
4561                 if entry:
4562                     yield entry
4563     '''
4564
4565     def _report_history_entries(self, renderer):
4566         for url in traverse_obj(renderer, (
4567                 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4568                 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4569                 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4570             yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4571
4572     def _extract_entries(self, parent_renderer, continuation_list):
4573         # continuation_list is modified in-place with continuation_list = [continuation_token]
4574         continuation_list[:] = [None]
4575         contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4576         for content in contents:
4577             if not isinstance(content, dict):
4578                 continue
4579             is_renderer = traverse_obj(
4580                 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4581                 expected_type=dict)
4582             if not is_renderer:
4583                 if content.get('richItemRenderer'):
4584                     for entry in self._rich_entries(content['richItemRenderer']):
4585                         yield entry
4586                     continuation_list[0] = self._extract_continuation(parent_renderer)
4587                 elif content.get('reportHistorySectionRenderer'):  # https://www.youtube.com/reporthistory
4588                     table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4589                     yield from self._report_history_entries(table)
4590                     continuation_list[0] = self._extract_continuation(table)
4591                 continue
4592
4593             isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4594             for isr_content in isr_contents:
4595                 if not isinstance(isr_content, dict):
4596                     continue
4597
4598                 known_renderers = {
4599                     'playlistVideoListRenderer': self._playlist_entries,
4600                     'gridRenderer': self._grid_entries,
4601                     'reelShelfRenderer': self._grid_entries,
4602                     'shelfRenderer': self._shelf_entries,
4603                     'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4604                     'backstagePostThreadRenderer': self._post_thread_entries,
4605                     'videoRenderer': lambda x: [self._video_entry(x)],
4606                     'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4607                     'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4608                     'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4609                 }
4610                 for key, renderer in isr_content.items():
4611                     if key not in known_renderers:
4612                         continue
4613                     for entry in known_renderers[key](renderer):
4614                         if entry:
4615                             yield entry
4616                     continuation_list[0] = self._extract_continuation(renderer)
4617                     break
4618
4619             if not continuation_list[0]:
4620                 continuation_list[0] = self._extract_continuation(is_renderer)
4621
4622         if not continuation_list[0]:
4623             continuation_list[0] = self._extract_continuation(parent_renderer)
4624
4625     def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4626         continuation_list = [None]
4627         extract_entries = lambda x: self._extract_entries(x, continuation_list)
4628         tab_content = try_get(tab, lambda x: x['content'], dict)
4629         if not tab_content:
4630             return
4631         parent_renderer = (
4632             try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4633             or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4634         yield from extract_entries(parent_renderer)
4635         continuation = continuation_list[0]
4636
4637         for page_num in itertools.count(1):
4638             if not continuation:
4639                 break
4640             headers = self.generate_api_headers(
4641                 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4642             response = self._extract_response(
4643                 item_id=f'{item_id} page {page_num}',
4644                 query=continuation, headers=headers, ytcfg=ytcfg,
4645                 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4646
4647             if not response:
4648                 break
4649             # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4650             # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4651             visitor_data = self._extract_visitor_data(response) or visitor_data
4652
4653             known_renderers = {
4654                 'videoRenderer': (self._grid_entries, 'items'),  # for membership tab
4655                 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4656                 'gridVideoRenderer': (self._grid_entries, 'items'),
4657                 'gridChannelRenderer': (self._grid_entries, 'items'),
4658                 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4659                 'itemSectionRenderer': (extract_entries, 'contents'),  # for feeds
4660                 'richItemRenderer': (extract_entries, 'contents'),  # for hashtag
4661                 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4662                 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4663                 'playlistVideoListContinuation': (self._playlist_entries, None),
4664                 'gridContinuation': (self._grid_entries, None),
4665                 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4666                 'sectionListContinuation': (extract_entries, None),  # for feeds
4667             }
4668
4669             continuation_items = traverse_obj(response, (
4670                 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4671                 'appendContinuationItemsAction', 'continuationItems'
4672             ), 'continuationContents', get_all=False)
4673             continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4674
4675             video_items_renderer = None
4676             for key in continuation_item.keys():
4677                 if key not in known_renderers:
4678                     continue
4679                 func, parent_key = known_renderers[key]
4680                 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4681                 continuation_list = [None]
4682                 yield from func(video_items_renderer)
4683                 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4684
4685             if not video_items_renderer:
4686                 break
4687
4688     @staticmethod
4689     def _extract_selected_tab(tabs, fatal=True):
4690         for tab_renderer in tabs:
4691             if tab_renderer.get('selected'):
4692                 return tab_renderer
4693         if fatal:
4694             raise ExtractorError('Unable to find selected tab')
4695
4696     @staticmethod
4697     def _extract_tab_renderers(response):
4698         return traverse_obj(
4699             response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
4700
4701     def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4702         metadata = self._extract_metadata_from_tabs(item_id, data)
4703
4704         selected_tab = self._extract_selected_tab(tabs)
4705         metadata['title'] += format_field(selected_tab, 'title', ' - %s')
4706         metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
4707
4708         return self.playlist_result(
4709             self._entries(
4710                 selected_tab, metadata['id'], ytcfg,
4711                 self._extract_account_syncid(ytcfg, data),
4712                 self._extract_visitor_data(data, ytcfg)),
4713             **metadata)
4714
4715     def _extract_metadata_from_tabs(self, item_id, data):
4716         info = {'id': item_id}
4717
4718         metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
4719         if metadata_renderer:
4720             info.update({
4721                 'uploader': metadata_renderer.get('title'),
4722                 'uploader_id': metadata_renderer.get('externalId'),
4723                 'uploader_url': metadata_renderer.get('channelUrl'),
4724             })
4725             if info['uploader_id']:
4726                 info['id'] = info['uploader_id']
4727         else:
4728             metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
4729
4730         # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4731         # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4732         def _get_uncropped(url):
4733             return url_or_none((url or '').split('=')[0] + '=s0')
4734
4735         avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
4736         if avatar_thumbnails:
4737             uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4738             if uncropped_avatar:
4739                 avatar_thumbnails.append({
4740                     'url': uncropped_avatar,
4741                     'id': 'avatar_uncropped',
4742                     'preference': 1
4743                 })
4744
4745         channel_banners = self._extract_thumbnails(
4746             data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
4747         for banner in channel_banners:
4748             banner['preference'] = -10
4749
4750         if channel_banners:
4751             uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4752             if uncropped_banner:
4753                 channel_banners.append({
4754                     'url': uncropped_banner,
4755                     'id': 'banner_uncropped',
4756                     'preference': -5
4757                 })
4758
4759         # Deprecated - remove primary_sidebar_renderer when layout discontinued
4760         primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4761         playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
4762
4763         primary_thumbnails = self._extract_thumbnails(
4764             primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4765         playlist_thumbnails = self._extract_thumbnails(
4766             playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
4767
4768         info.update({
4769             'title': (traverse_obj(metadata_renderer, 'title')
4770                       or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
4771                       or info['id']),
4772             'availability': self._extract_availability(data),
4773             'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4774             'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
4775             'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
4776             'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
4777         })
4778
4779         # Playlist stats is a text runs array containing [video count, view count, last updated].
4780         # last updated or (view count and last updated) may be missing.
4781         playlist_stats = get_first(
4782             (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
4783
4784         last_updated_unix = self._parse_time_text(
4785             self._get_text(playlist_stats, 2)  # deprecated, remove when old layout discontinued
4786             or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
4787         info['modified_date'] = strftime_or_none(last_updated_unix, '%Y%m%d')
4788
4789         info['view_count'] = self._get_count(playlist_stats, 1)
4790         if info['view_count'] is None:  # 0 is allowed
4791             info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
4792
4793         info['playlist_count'] = self._get_count(playlist_stats, 0)
4794         if info['playlist_count'] is None:  # 0 is allowed
4795             info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
4796
4797         if not info.get('uploader_id'):
4798             owner = traverse_obj(playlist_header_renderer, 'ownerText')
4799             if not owner:  # Deprecated
4800                 owner = traverse_obj(
4801                     self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
4802                     ('videoOwner', 'videoOwnerRenderer', 'title'))
4803             owner_text = self._get_text(owner)
4804             browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
4805             info.update({
4806                 'uploader': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
4807                 'uploader_id': browse_ep.get('browseId'),
4808                 'uploader_url': urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl'))
4809             })
4810
4811         info.update({
4812             'channel': info['uploader'],
4813             'channel_id': info['uploader_id'],
4814             'channel_url': info['uploader_url']
4815         })
4816         return info
4817
4818     def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4819         first_id = last_id = response = None
4820         for page_num in itertools.count(1):
4821             videos = list(self._playlist_entries(playlist))
4822             if not videos:
4823                 return
4824             start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
4825             if start >= len(videos):
4826                 return
4827             yield from videos[start:]
4828             first_id = first_id or videos[0]['id']
4829             last_id = videos[-1]['id']
4830             watch_endpoint = try_get(
4831                 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
4832             headers = self.generate_api_headers(
4833                 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4834                 visitor_data=self._extract_visitor_data(response, data, ytcfg))
4835             query = {
4836                 'playlistId': playlist_id,
4837                 'videoId': watch_endpoint.get('videoId') or last_id,
4838                 'index': watch_endpoint.get('index') or len(videos),
4839                 'params': watch_endpoint.get('params') or 'OAE%3D'
4840             }
4841             response = self._extract_response(
4842                 item_id='%s page %d' % (playlist_id, page_num),
4843                 query=query, ep='next', headers=headers, ytcfg=ytcfg,
4844                 check_get_keys='contents'
4845             )
4846             playlist = try_get(
4847                 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
4848
4849     def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
4850         title = playlist.get('title') or try_get(
4851             data, lambda x: x['titleText']['simpleText'], str)
4852         playlist_id = playlist.get('playlistId') or item_id
4853
4854         # Delegating everything except mix playlists to regular tab-based playlist URL
4855         playlist_url = urljoin(url, try_get(
4856             playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4857             str))
4858
4859         # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
4860         # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
4861         is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
4862
4863         if playlist_url and playlist_url != url and not is_known_unviewable:
4864             return self.url_result(
4865                 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4866                 video_title=title)
4867
4868         return self.playlist_result(
4869             self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
4870             playlist_id=playlist_id, playlist_title=title)
4871
4872     def _extract_availability(self, data):
4873         """
4874         Gets the availability of a given playlist/tab.
4875         Note: Unless YouTube tells us explicitly, we do not assume it is public
4876         @param data: response
4877         """
4878         sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
4879         playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
4880         player_header_privacy = playlist_header_renderer.get('privacy')
4881
4882         badges = self._extract_badges(sidebar_renderer)
4883
4884         # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
4885         privacy_setting_icon = get_first(
4886             (playlist_header_renderer, sidebar_renderer),
4887             ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
4888              lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
4889             expected_type=str)
4890
4891         microformats_is_unlisted = traverse_obj(
4892             data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
4893
4894         return (
4895             'public' if (
4896                 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4897                 or player_header_privacy == 'PUBLIC'
4898                 or privacy_setting_icon == 'PRIVACY_PUBLIC')
4899             else self._availability(
4900                 is_private=(
4901                     self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4902                     or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
4903                     else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
4904                 is_unlisted=(
4905                     self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4906                     or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
4907                     else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
4908                     else microformats_is_unlisted if microformats_is_unlisted is not None else None),
4909                 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
4910                 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
4911                 needs_auth=False))
4912
4913     @staticmethod
4914     def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
4915         sidebar_renderer = try_get(
4916             data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
4917         for item in sidebar_renderer:
4918             renderer = try_get(item, lambda x: x[info_renderer], expected_type)
4919             if renderer:
4920                 return renderer
4921
4922     def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
4923         """
4924         Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
4925         """
4926         is_playlist = bool(traverse_obj(
4927             data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
4928         if not is_playlist:
4929             return
4930         headers = self.generate_api_headers(
4931             ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
4932             visitor_data=self._extract_visitor_data(data, ytcfg))
4933         query = {
4934             'params': 'wgYCCAA=',
4935             'browseId': f'VL{item_id}'
4936         }
4937         return self._extract_response(
4938             item_id=item_id, headers=headers, query=query,
4939             check_get_keys='contents', fatal=False, ytcfg=ytcfg,
4940             note='Redownloading playlist API JSON with unavailable videos')
4941
4942     @functools.cached_property
4943     def skip_webpage(self):
4944         return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
4945
4946     def _extract_webpage(self, url, item_id, fatal=True):
4947         webpage, data = None, None
4948         for retry in self.RetryManager(fatal=fatal):
4949             try:
4950                 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
4951                 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
4952             except ExtractorError as e:
4953                 if isinstance(e.cause, network_exceptions):
4954                     if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
4955                         retry.error = e
4956                         continue
4957                 self._error_or_warning(e, fatal=fatal)
4958                 break
4959
4960             try:
4961                 self._extract_and_report_alerts(data)
4962             except ExtractorError as e:
4963                 self._error_or_warning(e, fatal=fatal)
4964                 break
4965
4966             # Sometimes youtube returns a webpage with incomplete ytInitialData
4967             # See: https://github.com/yt-dlp/yt-dlp/issues/116
4968             if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
4969                 retry.error = ExtractorError('Incomplete yt initial data received')
4970                 continue
4971
4972         return webpage, data
4973
4974     def _report_playlist_authcheck(self, ytcfg, fatal=True):
4975         """Use if failed to extract ytcfg (and data) from initial webpage"""
4976         if not ytcfg and self.is_authenticated:
4977             msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
4978             if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
4979                 raise ExtractorError(
4980                     f'{msg}. If you are not downloading private content, or '
4981                     'your cookies are only for the first account and channel,'
4982                     ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
4983                     expected=True)
4984             self.report_warning(msg, only_once=True)
4985
4986     def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
4987         data = None
4988         if not self.skip_webpage:
4989             webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
4990             ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
4991             # Reject webpage data if redirected to home page without explicitly requesting
4992             selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
4993             if (url != 'https://www.youtube.com/feed/recommended'
4994                     and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch'  # Home page
4995                     and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
4996                 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
4997                 if fatal:
4998                     raise ExtractorError(msg, expected=True)
4999                 self.report_warning(msg, only_once=True)
5000         if not data:
5001             self._report_playlist_authcheck(ytcfg, fatal=fatal)
5002             data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5003         return data, ytcfg
5004
5005     def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5006         headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5007         resolve_response = self._extract_response(
5008             item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5009             ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5010         endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5011         for ep_key, ep in endpoints.items():
5012             params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5013             if params:
5014                 return self._extract_response(
5015                     item_id=item_id, query=params, ep=ep, headers=headers,
5016                     ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5017                     check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5018         err_note = 'Failed to resolve url (does the playlist exist?)'
5019         if fatal:
5020             raise ExtractorError(err_note, expected=True)
5021         self.report_warning(err_note, item_id)
5022
5023     _SEARCH_PARAMS = None
5024
5025     def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5026         data = {'query': query}
5027         if params is NO_DEFAULT:
5028             params = self._SEARCH_PARAMS
5029         if params:
5030             data['params'] = params
5031
5032         content_keys = (
5033             ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5034             ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5035             # ytmusic search
5036             ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5037             ('continuationContents', ),
5038         )
5039         display_id = f'query "{query}"'
5040         check_get_keys = tuple({keys[0] for keys in content_keys})
5041         ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5042         self._report_playlist_authcheck(ytcfg, fatal=False)
5043
5044         continuation_list = [None]
5045         search = None
5046         for page_num in itertools.count(1):
5047             data.update(continuation_list[0] or {})
5048             headers = self.generate_api_headers(
5049                 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5050             search = self._extract_response(
5051                 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5052                 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5053             slr_contents = traverse_obj(search, *content_keys)
5054             yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5055             if not continuation_list[0]:
5056                 break
5057
5058
5059 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5060     IE_DESC = 'YouTube Tabs'
5061     _VALID_URL = r'''(?x:
5062         https?://
5063             (?:\w+\.)?
5064             (?:
5065                 youtube(?:kids)?\.com|
5066                 %(invidious)s
5067             )/
5068             (?:
5069                 (?P<channel_type>channel|c|user|browse)/|
5070                 (?P<not_channel>
5071                     feed/|hashtag/|
5072                     (?:playlist|watch)\?.*?\blist=
5073                 )|
5074                 (?!(?:%(reserved_names)s)\b)  # Direct URLs
5075             )
5076             (?P<id>[^/?\#&]+)
5077     )''' % {
5078         'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5079         'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5080     }
5081     IE_NAME = 'youtube:tab'
5082
5083     _TESTS = [{
5084         'note': 'playlists, multipage',
5085         'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5086         'playlist_mincount': 94,
5087         'info_dict': {
5088             'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5089             'title': 'Igor Kleiner - Playlists',
5090             'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5091             'uploader': 'Igor Kleiner',
5092             'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5093             'channel': 'Igor Kleiner',
5094             'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5095             'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5096             'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5097             'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5098             'channel_follower_count': int
5099         },
5100     }, {
5101         'note': 'playlists, multipage, different order',
5102         'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5103         'playlist_mincount': 94,
5104         'info_dict': {
5105             'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5106             'title': 'Igor Kleiner - Playlists',
5107             'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5108             'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5109             'uploader': 'Igor Kleiner',
5110             'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5111             'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5112             'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5113             'channel': 'Igor Kleiner',
5114             'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5115             'channel_follower_count': int
5116         },
5117     }, {
5118         'note': 'playlists, series',
5119         'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5120         'playlist_mincount': 5,
5121         'info_dict': {
5122             'id': 'UCYO_jab_esuFRV4b17AJtAw',
5123             'title': '3Blue1Brown - Playlists',
5124             'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5125             'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
5126             'uploader': '3Blue1Brown',
5127             'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5128             'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5129             'channel': '3Blue1Brown',
5130             'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5131             'tags': ['Mathematics'],
5132             'channel_follower_count': int
5133         },
5134     }, {
5135         'note': 'playlists, singlepage',
5136         'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5137         'playlist_mincount': 4,
5138         'info_dict': {
5139             'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5140             'title': 'ThirstForScience - Playlists',
5141             'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5142             'uploader': 'ThirstForScience',
5143             'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5144             'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5145             'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5146             'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5147             'tags': 'count:13',
5148             'channel': 'ThirstForScience',
5149             'channel_follower_count': int
5150         }
5151     }, {
5152         'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5153         'only_matching': True,
5154     }, {
5155         'note': 'basic, single video playlist',
5156         'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5157         'info_dict': {
5158             'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5159             'uploader': 'Sergey M.',
5160             'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5161             'title': 'youtube-dl public playlist',
5162             'description': '',
5163             'tags': [],
5164             'view_count': int,
5165             'modified_date': '20201130',
5166             'channel': 'Sergey M.',
5167             'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5168             'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5169             'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5170             'availability': 'public',
5171         },
5172         'playlist_count': 1,
5173     }, {
5174         'note': 'empty playlist',
5175         'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5176         'info_dict': {
5177             'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5178             'uploader': 'Sergey M.',
5179             'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5180             'title': 'youtube-dl empty playlist',
5181             'tags': [],
5182             'channel': 'Sergey M.',
5183             'description': '',
5184             'modified_date': '20160902',
5185             'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5186             'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5187             'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5188             'availability': 'public',
5189         },
5190         'playlist_count': 0,
5191     }, {
5192         'note': 'Home tab',
5193         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5194         'info_dict': {
5195             'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5196             'title': 'lex will - Home',
5197             'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5198             'uploader': 'lex will',
5199             'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5200             'channel': 'lex will',
5201             'tags': ['bible', 'history', 'prophesy'],
5202             'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5203             'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5204             'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5205             'channel_follower_count': int
5206         },
5207         'playlist_mincount': 2,
5208     }, {
5209         'note': 'Videos tab',
5210         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5211         'info_dict': {
5212             'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5213             'title': 'lex will - Videos',
5214             'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5215             'uploader': 'lex will',
5216             'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5217             'tags': ['bible', 'history', 'prophesy'],
5218             'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5219             'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5220             'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5221             'channel': 'lex will',
5222             'channel_follower_count': int
5223         },
5224         'playlist_mincount': 975,
5225     }, {
5226         'note': 'Videos tab, sorted by popular',
5227         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5228         'info_dict': {
5229             'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5230             'title': 'lex will - Videos',
5231             'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5232             'uploader': 'lex will',
5233             'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5234             'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5235             'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5236             'channel': 'lex will',
5237             'tags': ['bible', 'history', 'prophesy'],
5238             'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5239             'channel_follower_count': int
5240         },
5241         'playlist_mincount': 199,
5242     }, {
5243         'note': 'Playlists tab',
5244         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5245         'info_dict': {
5246             'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5247             'title': 'lex will - Playlists',
5248             'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5249             'uploader': 'lex will',
5250             'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5251             'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5252             'channel': 'lex will',
5253             'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5254             'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5255             'tags': ['bible', 'history', 'prophesy'],
5256             'channel_follower_count': int
5257         },
5258         'playlist_mincount': 17,
5259     }, {
5260         'note': 'Community tab',
5261         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5262         'info_dict': {
5263             'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5264             'title': 'lex will - Community',
5265             'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5266             'uploader': 'lex will',
5267             'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5268             'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5269             'channel': 'lex will',
5270             'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5271             'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5272             'tags': ['bible', 'history', 'prophesy'],
5273             'channel_follower_count': int
5274         },
5275         'playlist_mincount': 18,
5276     }, {
5277         'note': 'Channels tab',
5278         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5279         'info_dict': {
5280             'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5281             'title': 'lex will - Channels',
5282             'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5283             'uploader': 'lex will',
5284             'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5285             'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5286             'channel': 'lex will',
5287             'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5288             'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5289             'tags': ['bible', 'history', 'prophesy'],
5290             'channel_follower_count': int
5291         },
5292         'playlist_mincount': 12,
5293     }, {
5294         'note': 'Search tab',
5295         'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5296         'playlist_mincount': 40,
5297         'info_dict': {
5298             'id': 'UCYO_jab_esuFRV4b17AJtAw',
5299             'title': '3Blue1Brown - Search - linear algebra',
5300             'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5301             'uploader': '3Blue1Brown',
5302             'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw',
5303             'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5304             'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5305             'tags': ['Mathematics'],
5306             'channel': '3Blue1Brown',
5307             'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5308             'channel_follower_count': int
5309         },
5310     }, {
5311         'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5312         'only_matching': True,
5313     }, {
5314         'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5315         'only_matching': True,
5316     }, {
5317         'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5318         'only_matching': True,
5319     }, {
5320         'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5321         'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5322         'info_dict': {
5323             'title': '29C3: Not my department',
5324             'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5325             'uploader': 'Christiaan008',
5326             'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5327             'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5328             'tags': [],
5329             'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008',
5330             'view_count': int,
5331             'modified_date': '20150605',
5332             'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5333             'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008',
5334             'channel': 'Christiaan008',
5335             'availability': 'public',
5336         },
5337         'playlist_count': 96,
5338     }, {
5339         'note': 'Large playlist',
5340         'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5341         'info_dict': {
5342             'title': 'Uploads from Cauchemar',
5343             'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5344             'uploader': 'Cauchemar',
5345             'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5346             'channel_url': 'https://www.youtube.com/c/Cauchemar89',
5347             'tags': [],
5348             'modified_date': r're:\d{8}',
5349             'channel': 'Cauchemar',
5350             'uploader_url': 'https://www.youtube.com/c/Cauchemar89',
5351             'view_count': int,
5352             'description': '',
5353             'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5354             'availability': 'public',
5355         },
5356         'playlist_mincount': 1123,
5357         'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5358     }, {
5359         'note': 'even larger playlist, 8832 videos',
5360         'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5361         'only_matching': True,
5362     }, {
5363         'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5364         'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5365         'info_dict': {
5366             'title': 'Uploads from Interstellar Movie',
5367             'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5368             'uploader': 'Interstellar Movie',
5369             'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5370             'uploader_url': 'https://www.youtube.com/c/InterstellarMovie',
5371             'tags': [],
5372             'view_count': int,
5373             'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5374             'channel_url': 'https://www.youtube.com/c/InterstellarMovie',
5375             'channel': 'Interstellar Movie',
5376             'description': '',
5377             'modified_date': r're:\d{8}',
5378             'availability': 'public',
5379         },
5380         'playlist_mincount': 21,
5381     }, {
5382         'note': 'Playlist with "show unavailable videos" button',
5383         'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5384         'info_dict': {
5385             'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5386             'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5387             'uploader': 'Phim Siêu Nhân Nhật Bản',
5388             'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5389             'view_count': int,
5390             'channel': 'Phim Siêu Nhân Nhật Bản',
5391             'tags': [],
5392             'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5393             'description': '',
5394             'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5395             'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5396             'modified_date': r're:\d{8}',
5397             'availability': 'public',
5398         },
5399         'playlist_mincount': 200,
5400         'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5401     }, {
5402         'note': 'Playlist with unavailable videos in page 7',
5403         'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5404         'info_dict': {
5405             'title': 'Uploads from BlankTV',
5406             'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5407             'uploader': 'BlankTV',
5408             'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5409             'channel': 'BlankTV',
5410             'channel_url': 'https://www.youtube.com/c/blanktv',
5411             'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5412             'view_count': int,
5413             'tags': [],
5414             'uploader_url': 'https://www.youtube.com/c/blanktv',
5415             'modified_date': r're:\d{8}',
5416             'description': '',
5417             'availability': 'public',
5418         },
5419         'playlist_mincount': 1000,
5420         'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5421     }, {
5422         'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5423         'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5424         'info_dict': {
5425             'title': 'Data Analysis with Dr Mike Pound',
5426             'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5427             'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5428             'uploader': 'Computerphile',
5429             'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5430             'uploader_url': 'https://www.youtube.com/user/Computerphile',
5431             'tags': [],
5432             'view_count': int,
5433             'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5434             'channel_url': 'https://www.youtube.com/user/Computerphile',
5435             'channel': 'Computerphile',
5436             'availability': 'public',
5437             'modified_date': '20190712',
5438         },
5439         'playlist_mincount': 11,
5440     }, {
5441         'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5442         'only_matching': True,
5443     }, {
5444         'note': 'Playlist URL that does not actually serve a playlist',
5445         'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5446         'info_dict': {
5447             'id': 'FqZTN594JQw',
5448             'ext': 'webm',
5449             'title': "Smiley's People 01 detective, Adventure Series, Action",
5450             'uploader': 'STREEM',
5451             'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
5452             'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
5453             'upload_date': '20150526',
5454             'license': 'Standard YouTube License',
5455             'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5456             'categories': ['People & Blogs'],
5457             'tags': list,
5458             'view_count': int,
5459             'like_count': int,
5460         },
5461         'params': {
5462             'skip_download': True,
5463         },
5464         'skip': 'This video is not available.',
5465         'add_ie': [YoutubeIE.ie_key()],
5466     }, {
5467         'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5468         'only_matching': True,
5469     }, {
5470         'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5471         'only_matching': True,
5472     }, {
5473         'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5474         'info_dict': {
5475             'id': 'Wq15eF5vCbI',  # This will keep changing
5476             'ext': 'mp4',
5477             'title': str,
5478             'uploader': 'Sky News',
5479             'uploader_id': 'skynews',
5480             'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
5481             'upload_date': r're:\d{8}',
5482             'description': str,
5483             'categories': ['News & Politics'],
5484             'tags': list,
5485             'like_count': int,
5486             'release_timestamp': int,
5487             'channel': 'Sky News',
5488             'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5489             'age_limit': 0,
5490             'view_count': int,
5491             'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5492             'playable_in_embed': True,
5493             'release_date': r're:\d+',
5494             'availability': 'public',
5495             'live_status': 'is_live',
5496             'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5497             'channel_follower_count': int,
5498             'concurrent_view_count': int,
5499         },
5500         'params': {
5501             'skip_download': True,
5502         },
5503         'expected_warnings': ['Ignoring subtitle tracks found in '],
5504     }, {
5505         'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5506         'info_dict': {
5507             'id': 'a48o2S1cPoo',
5508             'ext': 'mp4',
5509             'title': 'The Young Turks - Live Main Show',
5510             'uploader': 'The Young Turks',
5511             'uploader_id': 'TheYoungTurks',
5512             'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
5513             'upload_date': '20150715',
5514             'license': 'Standard YouTube License',
5515             'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5516             'categories': ['News & Politics'],
5517             'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5518             'like_count': int,
5519         },
5520         'params': {
5521             'skip_download': True,
5522         },
5523         'only_matching': True,
5524     }, {
5525         'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5526         'only_matching': True,
5527     }, {
5528         'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5529         'only_matching': True,
5530     }, {
5531         'note': 'A channel that is not live. Should raise error',
5532         'url': 'https://www.youtube.com/user/numberphile/live',
5533         'only_matching': True,
5534     }, {
5535         'url': 'https://www.youtube.com/feed/trending',
5536         'only_matching': True,
5537     }, {
5538         'url': 'https://www.youtube.com/feed/library',
5539         'only_matching': True,
5540     }, {
5541         'url': 'https://www.youtube.com/feed/history',
5542         'only_matching': True,
5543     }, {
5544         'url': 'https://www.youtube.com/feed/subscriptions',
5545         'only_matching': True,
5546     }, {
5547         'url': 'https://www.youtube.com/feed/watch_later',
5548         'only_matching': True,
5549     }, {
5550         'note': 'Recommended - redirects to home page.',
5551         'url': 'https://www.youtube.com/feed/recommended',
5552         'only_matching': True,
5553     }, {
5554         'note': 'inline playlist with not always working continuations',
5555         'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5556         'only_matching': True,
5557     }, {
5558         'url': 'https://www.youtube.com/course',
5559         'only_matching': True,
5560     }, {
5561         'url': 'https://www.youtube.com/zsecurity',
5562         'only_matching': True,
5563     }, {
5564         'url': 'http://www.youtube.com/NASAgovVideo/videos',
5565         'only_matching': True,
5566     }, {
5567         'url': 'https://www.youtube.com/TheYoungTurks/live',
5568         'only_matching': True,
5569     }, {
5570         'url': 'https://www.youtube.com/hashtag/cctv9',
5571         'info_dict': {
5572             'id': 'cctv9',
5573             'title': '#cctv9',
5574             'tags': [],
5575         },
5576         'playlist_mincount': 300,  # not consistent but should be over 300
5577     }, {
5578         'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5579         'only_matching': True,
5580     }, {
5581         'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5582         'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5583         'only_matching': True
5584     }, {
5585         'note': '/browse/ should redirect to /channel/',
5586         'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5587         'only_matching': True
5588     }, {
5589         'note': 'VLPL, should redirect to playlist?list=PL...',
5590         'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5591         'info_dict': {
5592             'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5593             'uploader': 'NoCopyrightSounds',
5594             'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5595             'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5596             'title': 'NCS : All Releases 💿',
5597             'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5598             'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds',
5599             'modified_date': r're:\d{8}',
5600             'view_count': int,
5601             'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5602             'tags': [],
5603             'channel': 'NoCopyrightSounds',
5604             'availability': 'public',
5605         },
5606         'playlist_mincount': 166,
5607         'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5608     }, {
5609         'note': 'Topic, should redirect to playlist?list=UU...',
5610         'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5611         'info_dict': {
5612             'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5613             'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5614             'title': 'Uploads from Royalty Free Music - Topic',
5615             'uploader': 'Royalty Free Music - Topic',
5616             'tags': [],
5617             'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5618             'channel': 'Royalty Free Music - Topic',
5619             'view_count': int,
5620             'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5621             'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5622             'modified_date': r're:\d{8}',
5623             'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5624             'description': '',
5625             'availability': 'public',
5626         },
5627         'playlist_mincount': 101,
5628     }, {
5629         # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5630         # Treat as a general feed
5631         'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5632         'info_dict': {
5633             'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5634             'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5635             'tags': [],
5636         },
5637         'playlist_mincount': 9,
5638     }, {
5639         'note': 'Youtube music Album',
5640         'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5641         'info_dict': {
5642             'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5643             'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5644             'tags': [],
5645             'view_count': int,
5646             'description': '',
5647             'availability': 'unlisted',
5648             'modified_date': r're:\d{8}',
5649         },
5650         'playlist_count': 50,
5651     }, {
5652         'note': 'unlisted single video playlist',
5653         'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5654         'info_dict': {
5655             'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5656             'uploader': 'colethedj',
5657             'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5658             'title': 'yt-dlp unlisted playlist test',
5659             'availability': 'unlisted',
5660             'tags': [],
5661             'modified_date': '20220418',
5662             'channel': 'colethedj',
5663             'view_count': int,
5664             'description': '',
5665             'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5666             'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5667             'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5668         },
5669         'playlist_count': 1,
5670     }, {
5671         'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5672         'url': 'https://www.youtube.com/feed/recommended',
5673         'info_dict': {
5674             'id': 'recommended',
5675             'title': 'recommended',
5676             'tags': [],
5677         },
5678         'playlist_mincount': 50,
5679         'params': {
5680             'skip_download': True,
5681             'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5682         },
5683     }, {
5684         'note': 'API Fallback: /videos tab, sorted by oldest first',
5685         'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5686         'info_dict': {
5687             'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5688             'title': 'Cody\'sLab - Videos',
5689             'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5690             'uploader': 'Cody\'sLab',
5691             'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5692             'channel': 'Cody\'sLab',
5693             'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5694             'tags': [],
5695             'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5696             'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5697             'channel_follower_count': int
5698         },
5699         'playlist_mincount': 650,
5700         'params': {
5701             'skip_download': True,
5702             'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5703         },
5704         'skip': 'Query for sorting no longer works',
5705     }, {
5706         'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5707         'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5708         'info_dict': {
5709             'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5710             'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5711             'title': 'Uploads from Royalty Free Music - Topic',
5712             'uploader': 'Royalty Free Music - Topic',
5713             'modified_date': r're:\d{8}',
5714             'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5715             'description': '',
5716             'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5717             'tags': [],
5718             'channel': 'Royalty Free Music - Topic',
5719             'view_count': int,
5720             'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5721             'availability': 'public',
5722         },
5723         'playlist_mincount': 101,
5724         'params': {
5725             'skip_download': True,
5726             'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5727         },
5728     }, {
5729         'note': 'non-standard redirect to regional channel',
5730         'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5731         'only_matching': True
5732     }, {
5733         'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5734         'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5735         'info_dict': {
5736             'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5737             'modified_date': '20220407',
5738             'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5739             'tags': [],
5740             'uploader_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5741             'uploader': 'pukkandan',
5742             'availability': 'unlisted',
5743             'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5744             'channel': 'pukkandan',
5745             'description': 'Test for collaborative playlist',
5746             'title': 'yt-dlp test - collaborative playlist',
5747             'view_count': int,
5748             'uploader_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5749         },
5750         'playlist_mincount': 2
5751     }, {
5752         'note': 'translated tab name',
5753         'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
5754         'info_dict': {
5755             'id': 'UCiu-3thuViMebBjw_5nWYrA',
5756             'tags': [],
5757             'uploader_id': 'UCiu-3thuViMebBjw_5nWYrA',
5758             'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5759             'description': 'test description',
5760             'title': 'cole-dlp-test-acc - 再生リスト',
5761             'uploader_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5762             'uploader': 'cole-dlp-test-acc',
5763             'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5764             'channel': 'cole-dlp-test-acc',
5765             'channel_follower_count': int,
5766         },
5767         'playlist_mincount': 1,
5768         'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
5769         'expected_warnings': ['Preferring "ja"'],
5770     }, {
5771         # XXX: this should really check flat playlist entries, but the test suite doesn't support that
5772         'note': 'preferred lang set with playlist with translated video titles',
5773         'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
5774         'info_dict': {
5775             'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
5776             'tags': [],
5777             'view_count': int,
5778             'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5779             'uploader': 'cole-dlp-test-acc',
5780             'uploader_id': 'UCiu-3thuViMebBjw_5nWYrA',
5781             'channel': 'cole-dlp-test-acc',
5782             'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5783             'description': 'test',
5784             'uploader_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5785             'title': 'dlp test playlist',
5786             'availability': 'public',
5787         },
5788         'playlist_mincount': 1,
5789         'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
5790         'expected_warnings': ['Preferring "ja"'],
5791     }, {
5792         # shorts audio pivot for 2GtVksBMYFM.
5793         'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
5794         'info_dict': {
5795             'id': 'sfv_audio_pivot',
5796             'title': 'sfv_audio_pivot',
5797             'tags': [],
5798         },
5799         'playlist_mincount': 50,
5800
5801     }, {
5802         # Channel with a real live tab (not to be mistaken with streams tab)
5803         # Do not treat like it should redirect to live stream
5804         'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
5805         'info_dict': {
5806             'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
5807             'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
5808             'tags': [],
5809         },
5810         'playlist_mincount': 20,
5811     }, {
5812         # Tab name is not the same as tab id
5813         'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
5814         'info_dict': {
5815             'id': 'UCQvWX73GQygcwXOTSf_VDVg',
5816             'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
5817             'tags': [],
5818         },
5819         'playlist_mincount': 8,
5820     }, {
5821         # Home tab id is literally home. Not to get mistaken with featured
5822         'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
5823         'info_dict': {
5824             'id': 'UCQvWX73GQygcwXOTSf_VDVg',
5825             'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
5826             'tags': [],
5827         },
5828         'playlist_mincount': 8,
5829     }, {
5830         # Should get three playlists for videos, shorts and streams tabs
5831         'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
5832         'info_dict': {
5833             'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
5834             'title': 'Polka Ch. 尾丸ポルカ',
5835             'channel_follower_count': int,
5836             'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
5837             'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
5838             'uploader': 'Polka Ch. 尾丸ポルカ',
5839             'description': 'md5:3b8df1ac5af337aa206e37ee3d181ec9',
5840             'channel': 'Polka Ch. 尾丸ポルカ',
5841             'tags': 'count:35',
5842             'uploader_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
5843             'uploader_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
5844         },
5845         'playlist_count': 3,
5846     }, {
5847         # Shorts tab with channel with handle
5848         'url': 'https://www.youtube.com/@NotJustBikes/shorts',
5849         'info_dict': {
5850             'id': 'UC0intLFzLaudFG-xAvUEO-A',
5851             'title': 'Not Just Bikes - Shorts',
5852             'tags': 'count:12',
5853             'uploader': 'Not Just Bikes',
5854             'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
5855             'description': 'md5:7513148b1f02b924783157d84c4ea555',
5856             'channel_follower_count': int,
5857             'uploader_id': 'UC0intLFzLaudFG-xAvUEO-A',
5858             'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
5859             'uploader_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
5860             'channel': 'Not Just Bikes',
5861         },
5862         'playlist_mincount': 10,
5863     }, {
5864         # Streams tab
5865         'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
5866         'info_dict': {
5867             'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
5868             'title': '中村悠一 - Live',
5869             'tags': 'count:7',
5870             'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
5871             'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
5872             'uploader_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
5873             'channel': '中村悠一',
5874             'uploader_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
5875             'channel_follower_count': int,
5876             'uploader': '中村悠一',
5877             'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
5878         },
5879         'playlist_mincount': 60,
5880     }, {
5881         # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
5882         # See test_youtube_lists
5883         'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
5884         'only_matching': True,
5885     }, {
5886         # No uploads and no UCID given. Should fail with no uploads error
5887         # See test_youtube_lists
5888         'url': 'https://www.youtube.com/news',
5889         'only_matching': True
5890     }, {
5891         # No videos tab but has a shorts tab
5892         'url': 'https://www.youtube.com/c/TKFShorts',
5893         'info_dict': {
5894             'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
5895             'title': 'Shorts Break - Shorts',
5896             'tags': 'count:32',
5897             'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
5898             'channel': 'Shorts Break',
5899             'description': 'md5:a6c234cf3d50d878ef8721e34457cd11',
5900             'uploader': 'Shorts Break',
5901             'channel_follower_count': int,
5902             'uploader_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
5903             'uploader_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
5904             'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
5905         },
5906         'playlist_mincount': 30,
5907     }, {
5908         # Trending Now Tab. tab id is empty
5909         'url': 'https://www.youtube.com/feed/trending',
5910         'info_dict': {
5911             'id': 'trending',
5912             'title': 'trending - Now',
5913             'tags': [],
5914         },
5915         'playlist_mincount': 30,
5916     }, {
5917         # Trending Gaming Tab. tab id is empty
5918         'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
5919         'info_dict': {
5920             'id': 'trending',
5921             'title': 'trending - Gaming',
5922             'tags': [],
5923         },
5924         'playlist_mincount': 30,
5925     }, {
5926         # Shorts url result in shorts tab
5927         'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
5928         'info_dict': {
5929             'id': 'UCiu-3thuViMebBjw_5nWYrA',
5930             'title': 'cole-dlp-test-acc - Shorts',
5931             'uploader_id': 'UCiu-3thuViMebBjw_5nWYrA',
5932             'channel': 'cole-dlp-test-acc',
5933             'channel_follower_count': int,
5934             'description': 'test description',
5935             'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5936             'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5937             'tags': [],
5938             'uploader': 'cole-dlp-test-acc',
5939             'uploader_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5940
5941         },
5942         'playlist': [{
5943             'info_dict': {
5944                 '_type': 'url',
5945                 'ie_key': 'Youtube',
5946                 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
5947                 'id': 'sSM9J5YH_60',
5948                 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5949                 'title': 'SHORT short',
5950                 'channel': 'cole-dlp-test-acc',
5951                 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5952                 'view_count': int,
5953                 'thumbnails': list,
5954             }
5955         }],
5956         'params': {'extract_flat': True},
5957     }, {
5958         # Live video status should be extracted
5959         'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
5960         'info_dict': {
5961             'id': 'UCQvWX73GQygcwXOTSf_VDVg',
5962             'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live',  # TODO, should be Minecraft - Live or Minecraft - Topic - Live
5963             'tags': []
5964         },
5965         'playlist': [{
5966             'info_dict': {
5967                 '_type': 'url',
5968                 'ie_key': 'Youtube',
5969                 'url': 'startswith:https://www.youtube.com/watch?v=',
5970                 'id': str,
5971                 'title': str,
5972                 'live_status': 'is_live',
5973                 'channel_id': str,
5974                 'channel_url': str,
5975                 'concurrent_view_count': int,
5976                 'channel': str,
5977             }
5978         }],
5979         'params': {'extract_flat': True},
5980         'playlist_mincount': 1
5981     }]
5982
5983     @classmethod
5984     def suitable(cls, url):
5985         return False if YoutubeIE.suitable(url) else super().suitable(url)
5986
5987     _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
5988
5989     def _get_url_mobj(self, url):
5990         mobj = self._URL_RE.match(url).groupdict()
5991         mobj.update((k, '') for k, v in mobj.items() if v is None)
5992         return mobj
5993
5994     def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
5995         tab_name = (tab.get('title') or '').lower()
5996         tab_url = urljoin(base_url, traverse_obj(
5997             tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
5998
5999         tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6000                   or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6001         if tab_id:
6002             return {
6003                 'TAB_ID_SPONSORSHIPS': 'membership',
6004             }.get(tab_id, tab_id), tab_name
6005
6006         # Fallback to tab name if we cannot get the tab id.
6007         # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6008         # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6009         if tab_name:
6010             self.write_debug(f'Falling back to selected tab name: {tab_name}')
6011         return {
6012             'home': 'featured',
6013             'live': 'streams',
6014         }.get(tab_name, tab_name), tab_name
6015
6016     def _has_tab(self, tabs, tab_id):
6017         return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6018
6019     @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6020     def _real_extract(self, url, smuggled_data):
6021         item_id = self._match_id(url)
6022         url = urllib.parse.urlunparse(
6023             urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6024         compat_opts = self.get_param('compat_opts', [])
6025
6026         mobj = self._get_url_mobj(url)
6027         pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6028         if is_channel and smuggled_data.get('is_music_url'):
6029             if item_id[:2] == 'VL':  # Youtube music VL channels have an equivalent playlist
6030                 return self.url_result(
6031                     f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6032             elif item_id[:2] == 'MP':  # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6033                 mdata = self._extract_tab_endpoint(
6034                     f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6035                 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6036                                     get_all=False, expected_type=str)
6037                 if not murl:
6038                     raise ExtractorError('Failed to resolve album to playlist')
6039                 return self.url_result(murl, YoutubeTabIE)
6040             elif mobj['channel_type'] == 'browse':  # Youtube music /browse/ should be changed to /channel/
6041                 return self.url_result(
6042                     f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6043
6044         original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6045         if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6046             url = f'{pre}/videos{post}'
6047
6048         # Handle both video/playlist URLs
6049         qs = parse_qs(url)
6050         video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6051         if not video_id and mobj['not_channel'].startswith('watch'):
6052             if not playlist_id:
6053                 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6054                 raise ExtractorError('A video URL was given without video ID', expected=True)
6055             # Common mistake: https://www.youtube.com/watch?list=playlist_id
6056             self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6057             return self.url_result(
6058                 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6059
6060         if not self._yes_playlist(playlist_id, video_id):
6061             return self.url_result(
6062                 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6063
6064         data, ytcfg = self._extract_data(url, display_id)
6065
6066         # YouTube may provide a non-standard redirect to the regional channel
6067         # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6068         # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6069         redirect_url = traverse_obj(
6070             data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6071         if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6072             redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6073             self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6074             return self.url_result(redirect_url, YoutubeTabIE)
6075
6076         tabs, extra_tabs = self._extract_tab_renderers(data), []
6077         if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6078             selected_tab = self._extract_selected_tab(tabs)
6079             selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url)  # NB: Name may be translated
6080             self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6081
6082             if not original_tab_id and selected_tab_name:
6083                 self.to_screen('Downloading all uploads of the channel. '
6084                                'To download only the videos in a specific tab, pass the tab\'s URL')
6085                 if self._has_tab(tabs, 'streams'):
6086                     extra_tabs.append(''.join((pre, '/streams', post)))
6087                 if self._has_tab(tabs, 'shorts'):
6088                     extra_tabs.append(''.join((pre, '/shorts', post)))
6089                 # XXX: Members-only tab should also be extracted
6090
6091                 if not extra_tabs and selected_tab_id != 'videos':
6092                     # Channel does not have streams, shorts or videos tabs
6093                     if item_id[:2] != 'UC':
6094                         raise ExtractorError('This channel has no uploads', expected=True)
6095
6096                     # Topic channels don't have /videos. Use the equivalent playlist instead
6097                     pl_id = f'UU{item_id[2:]}'
6098                     pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6099                     try:
6100                         data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6101                     except ExtractorError:
6102                         raise ExtractorError('This channel has no uploads', expected=True)
6103                     else:
6104                         item_id, url = pl_id, pl_url
6105                         self.to_screen(
6106                             f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6107
6108                 elif extra_tabs and selected_tab_id != 'videos':
6109                     # When there are shorts/live tabs but not videos tab
6110                     url, data = f'{pre}{post}', None
6111
6112             elif (original_tab_id or 'videos') != selected_tab_id:
6113                 if original_tab_id == 'live':
6114                     # Live tab should have redirected to the video
6115                     # Except in the case the channel has an actual live tab
6116                     # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6117                     raise UserNotLive(video_id=item_id)
6118                 elif selected_tab_name:
6119                     raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6120
6121                 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6122                 url = f'{pre}{post}'
6123
6124         # YouTube sometimes provides a button to reload playlist with unavailable videos.
6125         if 'no-youtube-unavailable-videos' not in compat_opts:
6126             data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6127         self._extract_and_report_alerts(data, only_once=True)
6128
6129         tabs, entries = self._extract_tab_renderers(data), []
6130         if tabs:
6131             entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6132             entries[0].update({
6133                 'extractor_key': YoutubeTabIE.ie_key(),
6134                 'extractor': YoutubeTabIE.IE_NAME,
6135                 'webpage_url': url,
6136             })
6137         if self.get_param('playlist_items') == '0':
6138             entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6139         else:  # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6140             entries.extend(map(self._real_extract, extra_tabs))
6141
6142         if len(entries) == 1:
6143             return entries[0]
6144         elif entries:
6145             metadata = self._extract_metadata_from_tabs(item_id, data)
6146             uploads_url = 'the Uploads (UU) playlist URL'
6147             if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6148                 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6149             self.to_screen(
6150                 'Downloading as multiple playlists, separated by tabs. '
6151                 f'To download as a single playlist instead, pass {uploads_url}')
6152             return self.playlist_result(entries, item_id, **metadata)
6153
6154         # Inline playlist
6155         playlist = traverse_obj(
6156             data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6157         if playlist:
6158             return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6159
6160         video_id = traverse_obj(
6161             data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6162         if video_id:
6163             if tab != '/live':  # live tab is expected to redirect to video
6164                 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6165             return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6166
6167         raise ExtractorError('Unable to recognize tab page')
6168
6169
6170 class YoutubePlaylistIE(InfoExtractor):
6171     IE_DESC = 'YouTube playlists'
6172     _VALID_URL = r'''(?x)(?:
6173                         (?:https?://)?
6174                         (?:\w+\.)?
6175                         (?:
6176                             (?:
6177                                 youtube(?:kids)?\.com|
6178                                 %(invidious)s
6179                             )
6180                             /.*?\?.*?\blist=
6181                         )?
6182                         (?P<id>%(playlist_id)s)
6183                      )''' % {
6184         'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6185         'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6186     }
6187     IE_NAME = 'youtube:playlist'
6188     _TESTS = [{
6189         'note': 'issue #673',
6190         'url': 'PLBB231211A4F62143',
6191         'info_dict': {
6192             'title': '[OLD]Team Fortress 2 (Class-based LP)',
6193             'id': 'PLBB231211A4F62143',
6194             'uploader': 'Wickman',
6195             'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6196             'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6197             'view_count': int,
6198             'uploader_url': 'https://www.youtube.com/c/WickmanVT',
6199             'modified_date': r're:\d{8}',
6200             'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6201             'channel': 'Wickman',
6202             'tags': [],
6203             'channel_url': 'https://www.youtube.com/c/WickmanVT',
6204             'availability': 'public',
6205         },
6206         'playlist_mincount': 29,
6207     }, {
6208         'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6209         'info_dict': {
6210             'title': 'YDL_safe_search',
6211             'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6212         },
6213         'playlist_count': 2,
6214         'skip': 'This playlist is private',
6215     }, {
6216         'note': 'embedded',
6217         'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6218         'playlist_count': 4,
6219         'info_dict': {
6220             'title': 'JODA15',
6221             'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6222             'uploader': 'milan',
6223             'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6224             'description': '',
6225             'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6226             'tags': [],
6227             'modified_date': '20140919',
6228             'view_count': int,
6229             'channel': 'milan',
6230             'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6231             'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6232             'availability': 'public',
6233         },
6234         'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6235     }, {
6236         'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6237         'playlist_mincount': 455,
6238         'info_dict': {
6239             'title': '2018 Chinese New Singles (11/6 updated)',
6240             'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6241             'uploader': 'LBK',
6242             'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6243             'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6244             'channel': 'LBK',
6245             'view_count': int,
6246             'channel_url': 'https://www.youtube.com/c/愛低音的國王',
6247             'tags': [],
6248             'uploader_url': 'https://www.youtube.com/c/愛低音的國王',
6249             'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6250             'modified_date': r're:\d{8}',
6251             'availability': 'public',
6252         },
6253         'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6254     }, {
6255         'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6256         'only_matching': True,
6257     }, {
6258         # music album playlist
6259         'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6260         'only_matching': True,
6261     }]
6262
6263     @classmethod
6264     def suitable(cls, url):
6265         if YoutubeTabIE.suitable(url):
6266             return False
6267         from ..utils import parse_qs
6268         qs = parse_qs(url)
6269         if qs.get('v', [None])[0]:
6270             return False
6271         return super().suitable(url)
6272
6273     def _real_extract(self, url):
6274         playlist_id = self._match_id(url)
6275         is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6276         url = update_url_query(
6277             'https://www.youtube.com/playlist',
6278             parse_qs(url) or {'list': playlist_id})
6279         if is_music_url:
6280             url = smuggle_url(url, {'is_music_url': True})
6281         return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6282
6283
6284 class YoutubeYtBeIE(InfoExtractor):
6285     IE_DESC = 'youtu.be'
6286     _VALID_URL = r'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{11})/*?.*?\blist=(?P<playlist_id>%(playlist_id)s)' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
6287     _TESTS = [{
6288         'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6289         'info_dict': {
6290             'id': 'yeWKywCrFtk',
6291             'ext': 'mp4',
6292             'title': 'Small Scale Baler and Braiding Rugs',
6293             'uploader': 'Backus-Page House Museum',
6294             'uploader_id': 'backuspagemuseum',
6295             'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
6296             'upload_date': '20161008',
6297             'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6298             'categories': ['Nonprofits & Activism'],
6299             'tags': list,
6300             'like_count': int,
6301             'age_limit': 0,
6302             'playable_in_embed': True,
6303             'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp',
6304             'channel': 'Backus-Page House Museum',
6305             'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6306             'live_status': 'not_live',
6307             'view_count': int,
6308             'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6309             'availability': 'public',
6310             'duration': 59,
6311             'comment_count': int,
6312             'channel_follower_count': int
6313         },
6314         'params': {
6315             'noplaylist': True,
6316             'skip_download': True,
6317         },
6318     }, {
6319         'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6320         'only_matching': True,
6321     }]
6322
6323     def _real_extract(self, url):
6324         mobj = self._match_valid_url(url)
6325         video_id = mobj.group('id')
6326         playlist_id = mobj.group('playlist_id')
6327         return self.url_result(
6328             update_url_query('https://www.youtube.com/watch', {
6329                 'v': video_id,
6330                 'list': playlist_id,
6331                 'feature': 'youtu.be',
6332             }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6333
6334
6335 class YoutubeLivestreamEmbedIE(InfoExtractor):
6336     IE_DESC = 'YouTube livestream embeds'
6337     _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6338     _TESTS = [{
6339         'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6340         'only_matching': True,
6341     }]
6342
6343     def _real_extract(self, url):
6344         channel_id = self._match_id(url)
6345         return self.url_result(
6346             f'https://www.youtube.com/channel/{channel_id}/live',
6347             ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6348
6349
6350 class YoutubeYtUserIE(InfoExtractor):
6351     IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6352     IE_NAME = 'youtube:user'
6353     _VALID_URL = r'ytuser:(?P<id>.+)'
6354     _TESTS = [{
6355         'url': 'ytuser:phihag',
6356         'only_matching': True,
6357     }]
6358
6359     def _real_extract(self, url):
6360         user_id = self._match_id(url)
6361         return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6362
6363
6364 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6365     IE_NAME = 'youtube:favorites'
6366     IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6367     _VALID_URL = r':ytfav(?:ou?rite)?s?'
6368     _LOGIN_REQUIRED = True
6369     _TESTS = [{
6370         'url': ':ytfav',
6371         'only_matching': True,
6372     }, {
6373         'url': ':ytfavorites',
6374         'only_matching': True,
6375     }]
6376
6377     def _real_extract(self, url):
6378         return self.url_result(
6379             'https://www.youtube.com/playlist?list=LL',
6380             ie=YoutubeTabIE.ie_key())
6381
6382
6383 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6384     IE_NAME = 'youtube:notif'
6385     IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6386     _VALID_URL = r':ytnotif(?:ication)?s?'
6387     _LOGIN_REQUIRED = True
6388     _TESTS = [{
6389         'url': ':ytnotif',
6390         'only_matching': True,
6391     }, {
6392         'url': ':ytnotifications',
6393         'only_matching': True,
6394     }]
6395
6396     def _extract_notification_menu(self, response, continuation_list):
6397         notification_list = traverse_obj(
6398             response,
6399             ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6400             ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6401             expected_type=list) or []
6402         continuation_list[0] = None
6403         for item in notification_list:
6404             entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6405             if entry:
6406                 yield entry
6407             continuation = item.get('continuationItemRenderer')
6408             if continuation:
6409                 continuation_list[0] = continuation
6410
6411     def _extract_notification_renderer(self, notification):
6412         video_id = traverse_obj(
6413             notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6414         url = f'https://www.youtube.com/watch?v={video_id}'
6415         channel_id = None
6416         if not video_id:
6417             browse_ep = traverse_obj(
6418                 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6419             channel_id = traverse_obj(browse_ep, 'browseId', expected_type=str)
6420             post_id = self._search_regex(
6421                 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6422                 'post id', default=None)
6423             if not channel_id or not post_id:
6424                 return
6425             # The direct /post url redirects to this in the browser
6426             url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6427
6428         channel = traverse_obj(
6429             notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6430             expected_type=str)
6431         notification_title = self._get_text(notification, 'shortMessage')
6432         if notification_title:
6433             notification_title = notification_title.replace('\xad', '')  # remove soft hyphens
6434         # TODO: handle recommended videos
6435         title = self._search_regex(
6436             rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6437             'video title', default=None)
6438         timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6439                      if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6440                      else None)
6441         return {
6442             '_type': 'url',
6443             'url': url,
6444             'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6445             'video_id': video_id,
6446             'title': title,
6447             'channel_id': channel_id,
6448             'channel': channel,
6449             'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6450             'timestamp': timestamp,
6451         }
6452
6453     def _notification_menu_entries(self, ytcfg):
6454         continuation_list = [None]
6455         response = None
6456         for page in itertools.count(1):
6457             ctoken = traverse_obj(
6458                 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6459             response = self._extract_response(
6460                 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6461                 ep='notification/get_notification_menu', check_get_keys='actions',
6462                 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6463             yield from self._extract_notification_menu(response, continuation_list)
6464             if not continuation_list[0]:
6465                 break
6466
6467     def _real_extract(self, url):
6468         display_id = 'notifications'
6469         ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6470         self._report_playlist_authcheck(ytcfg)
6471         return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6472
6473
6474 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6475     IE_DESC = 'YouTube search'
6476     IE_NAME = 'youtube:search'
6477     _SEARCH_KEY = 'ytsearch'
6478     _SEARCH_PARAMS = 'EgIQAQ%3D%3D'  # Videos only
6479     _TESTS = [{
6480         'url': 'ytsearch5:youtube-dl test video',
6481         'playlist_count': 5,
6482         'info_dict': {
6483             'id': 'youtube-dl test video',
6484             'title': 'youtube-dl test video',
6485         }
6486     }]
6487
6488
6489 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6490     IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6491     _SEARCH_KEY = 'ytsearchdate'
6492     IE_DESC = 'YouTube search, newest videos first'
6493     _SEARCH_PARAMS = 'CAISAhAB'  # Videos only, sorted by date
6494     _TESTS = [{
6495         'url': 'ytsearchdate5:youtube-dl test video',
6496         'playlist_count': 5,
6497         'info_dict': {
6498             'id': 'youtube-dl test video',
6499             'title': 'youtube-dl test video',
6500         }
6501     }]
6502
6503
6504 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6505     IE_DESC = 'YouTube search URLs with sorting and filter support'
6506     IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6507     _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6508     _TESTS = [{
6509         'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6510         'playlist_mincount': 5,
6511         'info_dict': {
6512             'id': 'youtube-dl test video',
6513             'title': 'youtube-dl test video',
6514         }
6515     }, {
6516         'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6517         'playlist_mincount': 5,
6518         'info_dict': {
6519             'id': 'python',
6520             'title': 'python',
6521         }
6522     }, {
6523         'url': 'https://www.youtube.com/results?search_query=%23cats',
6524         'playlist_mincount': 1,
6525         'info_dict': {
6526             'id': '#cats',
6527             'title': '#cats',
6528             # The test suite does not have support for nested playlists
6529             # 'entries': [{
6530             #     'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6531             #     'title': '#cats',
6532             # }],
6533         },
6534     }, {
6535         'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
6536         'only_matching': True,
6537     }]
6538
6539     def _real_extract(self, url):
6540         qs = parse_qs(url)
6541         query = (qs.get('search_query') or qs.get('q'))[0]
6542         return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
6543
6544
6545 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
6546     IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
6547     IE_NAME = 'youtube:music:search_url'
6548     _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6549     _TESTS = [{
6550         'url': 'https://music.youtube.com/search?q=royalty+free+music',
6551         'playlist_count': 16,
6552         'info_dict': {
6553             'id': 'royalty free music',
6554             'title': 'royalty free music',
6555         }
6556     }, {
6557         'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
6558         'playlist_mincount': 30,
6559         'info_dict': {
6560             'id': 'royalty free music - songs',
6561             'title': 'royalty free music - songs',
6562         },
6563         'params': {'extract_flat': 'in_playlist'}
6564     }, {
6565         'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
6566         'playlist_mincount': 30,
6567         'info_dict': {
6568             'id': 'royalty free music - community playlists',
6569             'title': 'royalty free music - community playlists',
6570         },
6571         'params': {'extract_flat': 'in_playlist'}
6572     }]
6573
6574     _SECTIONS = {
6575         'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
6576         'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
6577         'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
6578         'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
6579         'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
6580         'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
6581     }
6582
6583     def _real_extract(self, url):
6584         qs = parse_qs(url)
6585         query = (qs.get('search_query') or qs.get('q'))[0]
6586         params = qs.get('sp', (None,))[0]
6587         if params:
6588             section = next((k for k, v in self._SECTIONS.items() if v == params), params)
6589         else:
6590             section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
6591             params = self._SECTIONS.get(section)
6592             if not params:
6593                 section = None
6594         title = join_nonempty(query, section, delim=' - ')
6595         return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
6596
6597
6598 class YoutubeFeedsInfoExtractor(InfoExtractor):
6599     """
6600     Base class for feed extractors
6601     Subclasses must re-define the _FEED_NAME property.
6602     """
6603     _LOGIN_REQUIRED = True
6604     _FEED_NAME = 'feeds'
6605
6606     def _real_initialize(self):
6607         YoutubeBaseInfoExtractor._check_login_required(self)
6608
6609     @classproperty
6610     def IE_NAME(self):
6611         return f'youtube:{self._FEED_NAME}'
6612
6613     def _real_extract(self, url):
6614         return self.url_result(
6615             f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
6616
6617
6618 class YoutubeWatchLaterIE(InfoExtractor):
6619     IE_NAME = 'youtube:watchlater'
6620     IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
6621     _VALID_URL = r':ytwatchlater'
6622     _TESTS = [{
6623         'url': ':ytwatchlater',
6624         'only_matching': True,
6625     }]
6626
6627     def _real_extract(self, url):
6628         return self.url_result(
6629             'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
6630
6631
6632 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
6633     IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
6634     _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
6635     _FEED_NAME = 'recommended'
6636     _LOGIN_REQUIRED = False
6637     _TESTS = [{
6638         'url': ':ytrec',
6639         'only_matching': True,
6640     }, {
6641         'url': ':ytrecommended',
6642         'only_matching': True,
6643     }, {
6644         'url': 'https://youtube.com',
6645         'only_matching': True,
6646     }]
6647
6648
6649 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
6650     IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
6651     _VALID_URL = r':ytsub(?:scription)?s?'
6652     _FEED_NAME = 'subscriptions'
6653     _TESTS = [{
6654         'url': ':ytsubs',
6655         'only_matching': True,
6656     }, {
6657         'url': ':ytsubscriptions',
6658         'only_matching': True,
6659     }]
6660
6661
6662 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
6663     IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
6664     _VALID_URL = r':ythis(?:tory)?'
6665     _FEED_NAME = 'history'
6666     _TESTS = [{
6667         'url': ':ythistory',
6668         'only_matching': True,
6669     }]
6670
6671
6672 class YoutubeStoriesIE(InfoExtractor):
6673     IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
6674     IE_NAME = 'youtube:stories'
6675     _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
6676     _TESTS = [{
6677         'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
6678         'only_matching': True,
6679     }]
6680
6681     def _real_extract(self, url):
6682         playlist_id = f'RLTD{self._match_id(url)}'
6683         return self.url_result(
6684             smuggle_url(f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1', {'is_story': True}),
6685             ie=YoutubeTabIE, video_id=playlist_id)
6686
6687
6688 class YoutubeShortsAudioPivotIE(InfoExtractor):
6689     IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
6690     IE_NAME = 'youtube:shorts:pivot:audio'
6691     _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
6692     _TESTS = [{
6693         'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
6694         'only_matching': True,
6695     }]
6696
6697     @staticmethod
6698     def _generate_audio_pivot_params(video_id):
6699         """
6700         Generates sfv_audio_pivot browse params for this video id
6701         """
6702         pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
6703         return urllib.parse.quote(base64.b64encode(pb_params).decode())
6704
6705     def _real_extract(self, url):
6706         video_id = self._match_id(url)
6707         return self.url_result(
6708             f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
6709             ie=YoutubeTabIE)
6710
6711
6712 class YoutubeTruncatedURLIE(InfoExtractor):
6713     IE_NAME = 'youtube:truncated_url'
6714     IE_DESC = False  # Do not list
6715     _VALID_URL = r'''(?x)
6716         (?:https?://)?
6717         (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
6718         (?:watch\?(?:
6719             feature=[a-z_]+|
6720             annotation_id=annotation_[^&]+|
6721             x-yt-cl=[0-9]+|
6722             hl=[^&]*|
6723             t=[0-9]+
6724         )?
6725         |
6726             attribution_link\?a=[^&]+
6727         )
6728         $
6729     '''
6730
6731     _TESTS = [{
6732         'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
6733         'only_matching': True,
6734     }, {
6735         'url': 'https://www.youtube.com/watch?',
6736         'only_matching': True,
6737     }, {
6738         'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
6739         'only_matching': True,
6740     }, {
6741         'url': 'https://www.youtube.com/watch?feature=foo',
6742         'only_matching': True,
6743     }, {
6744         'url': 'https://www.youtube.com/watch?hl=en-GB',
6745         'only_matching': True,
6746     }, {
6747         'url': 'https://www.youtube.com/watch?t=2372',
6748         'only_matching': True,
6749     }]
6750
6751     def _real_extract(self, url):
6752         raise ExtractorError(
6753             'Did you forget to quote the URL? Remember that & is a meta '
6754             'character in most shells, so you want to put the URL in quotes, '
6755             'like  youtube-dl '
6756             '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
6757             ' or simply  youtube-dl BaW_jenozKc  .',
6758             expected=True)
6759
6760
6761 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
6762     IE_NAME = 'youtube:clip'
6763     _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
6764     _TESTS = [{
6765         # FIXME: Other metadata should be extracted from the clip, not from the base video
6766         'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
6767         'info_dict': {
6768             'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
6769             'ext': 'mp4',
6770             'section_start': 29.0,
6771             'section_end': 39.7,
6772             'duration': 10.7,
6773             'age_limit': 0,
6774             'availability': 'public',
6775             'categories': ['Gaming'],
6776             'channel': 'Scott The Woz',
6777             'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
6778             'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
6779             'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
6780             'like_count': int,
6781             'playable_in_embed': True,
6782             'tags': 'count:17',
6783             'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
6784             'title': 'Mobile Games on Console - Scott The Woz',
6785             'upload_date': '20210920',
6786             'uploader': 'Scott The Woz',
6787             'uploader_id': 'scottthewoz',
6788             'uploader_url': 'http://www.youtube.com/user/scottthewoz',
6789             'view_count': int,
6790             'live_status': 'not_live',
6791             'channel_follower_count': int
6792         }
6793     }]
6794
6795     def _real_extract(self, url):
6796         clip_id = self._match_id(url)
6797         _, data = self._extract_webpage(url, clip_id)
6798
6799         video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
6800         if not video_id:
6801             raise ExtractorError('Unable to find video ID')
6802
6803         clip_data = traverse_obj(data, (
6804             'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
6805             'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
6806             'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
6807             'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
6808
6809         return {
6810             '_type': 'url_transparent',
6811             'url': f'https://www.youtube.com/watch?v={video_id}',
6812             'ie_key': YoutubeIE.ie_key(),
6813             'id': clip_id,
6814             'section_start': int(clip_data['startTimeMs']) / 1000,
6815             'section_end': int(clip_data['endTimeMs']) / 1000,
6816         }
6817
6818
6819 class YoutubeTruncatedIDIE(InfoExtractor):
6820     IE_NAME = 'youtube:truncated_id'
6821     IE_DESC = False  # Do not list
6822     _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
6823
6824     _TESTS = [{
6825         'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
6826         'only_matching': True,
6827     }]
6828
6829     def _real_extract(self, url):
6830         video_id = self._match_id(url)
6831         raise ExtractorError(
6832             f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
6833             expected=True)