]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[extractor/youtube] Subtitles cannot be translated to `und`
[yt-dlp.git] / 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 r'(?:www\.)?piped\.video',
392 r'(?:www\.)?piped\.aeong\.one',
393 )
394
395 # extracted from account/account_menu ep
396 # XXX: These are the supported YouTube UI and API languages,
397 # which is slightly different from languages supported for translation in YouTube studio
398 _SUPPORTED_LANG_CODES = [
399 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
400 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
401 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
402 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
403 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
404 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
405 ]
406
407 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
408
409 @functools.cached_property
410 def _preferred_lang(self):
411 """
412 Returns a language code supported by YouTube for the user preferred language.
413 Returns None if no preferred language set.
414 """
415 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
416 if not preferred_lang:
417 return
418 if preferred_lang not in self._SUPPORTED_LANG_CODES:
419 raise ExtractorError(
420 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
421 expected=True)
422 elif preferred_lang != 'en':
423 self.report_warning(
424 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
425 return preferred_lang
426
427 def _initialize_consent(self):
428 cookies = self._get_cookies('https://www.youtube.com/')
429 if cookies.get('__Secure-3PSID'):
430 return
431 consent_id = None
432 consent = cookies.get('CONSENT')
433 if consent:
434 if 'YES' in consent.value:
435 return
436 consent_id = self._search_regex(
437 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
438 if not consent_id:
439 consent_id = random.randint(100, 999)
440 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
441
442 def _initialize_pref(self):
443 cookies = self._get_cookies('https://www.youtube.com/')
444 pref_cookie = cookies.get('PREF')
445 pref = {}
446 if pref_cookie:
447 try:
448 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
449 except ValueError:
450 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
451 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
452 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
453
454 def _real_initialize(self):
455 self._initialize_pref()
456 self._initialize_consent()
457 self._check_login_required()
458
459 def _check_login_required(self):
460 if self._LOGIN_REQUIRED and not self._cookies_passed:
461 self.raise_login_required('Login details are needed to download this content', method='cookies')
462
463 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
464 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
465
466 def _get_default_ytcfg(self, client='web'):
467 return copy.deepcopy(INNERTUBE_CLIENTS[client])
468
469 def _get_innertube_host(self, client='web'):
470 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
471
472 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
473 # try_get but with fallback to default ytcfg client values when present
474 _func = lambda y: try_get(y, getter, expected_type)
475 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
476
477 def _extract_client_name(self, ytcfg, default_client='web'):
478 return self._ytcfg_get_safe(
479 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
480 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
481
482 def _extract_client_version(self, ytcfg, default_client='web'):
483 return self._ytcfg_get_safe(
484 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
485 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
486
487 def _select_api_hostname(self, req_api_hostname, default_client=None):
488 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
489 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
490
491 def _extract_api_key(self, ytcfg=None, default_client='web'):
492 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
493
494 def _extract_context(self, ytcfg=None, default_client='web'):
495 context = get_first(
496 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
497 # Enforce language and tz for extraction
498 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
499 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
500 return context
501
502 _SAPISID = None
503
504 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
505 time_now = round(time.time())
506 if self._SAPISID is None:
507 yt_cookies = self._get_cookies('https://www.youtube.com')
508 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
509 # See: https://github.com/yt-dlp/yt-dlp/issues/393
510 sapisid_cookie = dict_get(
511 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
512 if sapisid_cookie and sapisid_cookie.value:
513 self._SAPISID = sapisid_cookie.value
514 self.write_debug('Extracted SAPISID cookie')
515 # SAPISID cookie is required if not already present
516 if not yt_cookies.get('SAPISID'):
517 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
518 self._set_cookie(
519 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
520 else:
521 self._SAPISID = False
522 if not self._SAPISID:
523 return None
524 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
525 sapisidhash = hashlib.sha1(
526 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
527 return f'SAPISIDHASH {time_now}_{sapisidhash}'
528
529 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
530 note='Downloading API JSON', errnote='Unable to download API page',
531 context=None, api_key=None, api_hostname=None, default_client='web'):
532
533 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
534 data.update(query)
535 real_headers = self.generate_api_headers(default_client=default_client)
536 real_headers.update({'content-type': 'application/json'})
537 if headers:
538 real_headers.update(headers)
539 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
540 or api_key or self._extract_api_key(default_client=default_client))
541 return self._download_json(
542 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
543 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
544 data=json.dumps(data).encode('utf8'), headers=real_headers,
545 query={'key': api_key, 'prettyPrint': 'false'})
546
547 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
548 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
549
550 @staticmethod
551 def _extract_session_index(*data):
552 """
553 Index of current account in account list.
554 See: https://github.com/yt-dlp/yt-dlp/pull/519
555 """
556 for ytcfg in data:
557 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
558 if session_index is not None:
559 return session_index
560
561 # Deprecated?
562 def _extract_identity_token(self, ytcfg=None, webpage=None):
563 if ytcfg:
564 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
565 if token:
566 return token
567 if webpage:
568 return self._search_regex(
569 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
570 'identity token', default=None, fatal=False)
571
572 @staticmethod
573 def _extract_account_syncid(*args):
574 """
575 Extract syncId required to download private playlists of secondary channels
576 @params response and/or ytcfg
577 """
578 for data in args:
579 # ytcfg includes channel_syncid if on secondary channel
580 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
581 if delegated_sid:
582 return delegated_sid
583 sync_ids = (try_get(
584 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
585 lambda x: x['DATASYNC_ID']), str) or '').split('||')
586 if len(sync_ids) >= 2 and sync_ids[1]:
587 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
588 # and just "user_syncid||" for primary channel. We only want the channel_syncid
589 return sync_ids[0]
590
591 @staticmethod
592 def _extract_visitor_data(*args):
593 """
594 Extracts visitorData from an API response or ytcfg
595 Appears to be used to track session state
596 """
597 return get_first(
598 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
599 expected_type=str)
600
601 @functools.cached_property
602 def is_authenticated(self):
603 return bool(self._generate_sapisidhash_header())
604
605 def extract_ytcfg(self, video_id, webpage):
606 if not webpage:
607 return {}
608 return self._parse_json(
609 self._search_regex(
610 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
611 default='{}'), video_id, fatal=False) or {}
612
613 def generate_api_headers(
614 self, *, ytcfg=None, account_syncid=None, session_index=None,
615 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
616
617 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
618 headers = {
619 'X-YouTube-Client-Name': str(
620 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
621 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
622 'Origin': origin,
623 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
624 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
625 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
626 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
627 }
628 if session_index is None:
629 session_index = self._extract_session_index(ytcfg)
630 if account_syncid or session_index is not None:
631 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
632
633 auth = self._generate_sapisidhash_header(origin)
634 if auth is not None:
635 headers['Authorization'] = auth
636 headers['X-Origin'] = origin
637 return filter_dict(headers)
638
639 def _download_ytcfg(self, client, video_id):
640 url = {
641 'web': 'https://www.youtube.com',
642 'web_music': 'https://music.youtube.com',
643 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
644 }.get(client)
645 if not url:
646 return {}
647 webpage = self._download_webpage(
648 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
649 return self.extract_ytcfg(video_id, webpage) or {}
650
651 @staticmethod
652 def _build_api_continuation_query(continuation, ctp=None):
653 query = {
654 'continuation': continuation
655 }
656 # TODO: Inconsistency with clickTrackingParams.
657 # Currently we have a fixed ctp contained within context (from ytcfg)
658 # and a ctp in root query for continuation.
659 if ctp:
660 query['clickTracking'] = {'clickTrackingParams': ctp}
661 return query
662
663 @classmethod
664 def _extract_next_continuation_data(cls, renderer):
665 next_continuation = try_get(
666 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
667 lambda x: x['continuation']['reloadContinuationData']), dict)
668 if not next_continuation:
669 return
670 continuation = next_continuation.get('continuation')
671 if not continuation:
672 return
673 ctp = next_continuation.get('clickTrackingParams')
674 return cls._build_api_continuation_query(continuation, ctp)
675
676 @classmethod
677 def _extract_continuation_ep_data(cls, continuation_ep: dict):
678 if isinstance(continuation_ep, dict):
679 continuation = try_get(
680 continuation_ep, lambda x: x['continuationCommand']['token'], str)
681 if not continuation:
682 return
683 ctp = continuation_ep.get('clickTrackingParams')
684 return cls._build_api_continuation_query(continuation, ctp)
685
686 @classmethod
687 def _extract_continuation(cls, renderer):
688 next_continuation = cls._extract_next_continuation_data(renderer)
689 if next_continuation:
690 return next_continuation
691
692 return traverse_obj(renderer, (
693 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
694 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
695 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
696
697 @classmethod
698 def _extract_alerts(cls, data):
699 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
700 if not isinstance(alert_dict, dict):
701 continue
702 for alert in alert_dict.values():
703 alert_type = alert.get('type')
704 if not alert_type:
705 continue
706 message = cls._get_text(alert, 'text')
707 if message:
708 yield alert_type, message
709
710 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
711 errors, warnings = [], []
712 for alert_type, alert_message in alerts:
713 if alert_type.lower() == 'error' and fatal:
714 errors.append([alert_type, alert_message])
715 elif alert_message not in self._IGNORED_WARNINGS:
716 warnings.append([alert_type, alert_message])
717
718 for alert_type, alert_message in (warnings + errors[:-1]):
719 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
720 if errors:
721 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
722
723 def _extract_and_report_alerts(self, data, *args, **kwargs):
724 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
725
726 def _extract_badges(self, renderer: dict):
727 privacy_icon_map = {
728 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
729 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
730 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC
731 }
732
733 badge_style_map = {
734 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
735 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
736 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW
737 }
738
739 label_map = {
740 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
741 'private': BadgeType.AVAILABILITY_PRIVATE,
742 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
743 'live': BadgeType.LIVE_NOW,
744 'premium': BadgeType.AVAILABILITY_PREMIUM
745 }
746
747 badges = []
748 for badge in traverse_obj(renderer, ('badges', ..., 'metadataBadgeRenderer'), default=[]):
749 badge_type = (
750 privacy_icon_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
751 or badge_style_map.get(traverse_obj(badge, 'style'))
752 )
753 if badge_type:
754 badges.append({'type': badge_type})
755 continue
756
757 # fallback, won't work in some languages
758 label = traverse_obj(badge, 'label', expected_type=str, default='')
759 for match, label_badge_type in label_map.items():
760 if match in label.lower():
761 badges.append({'type': badge_type})
762 continue
763
764 return badges
765
766 @staticmethod
767 def _has_badge(badges, badge_type):
768 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
769
770 @staticmethod
771 def _get_text(data, *path_list, max_runs=None):
772 for path in path_list or [None]:
773 if path is None:
774 obj = [data]
775 else:
776 obj = traverse_obj(data, path, default=[])
777 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
778 obj = [obj]
779 for item in obj:
780 text = try_get(item, lambda x: x['simpleText'], str)
781 if text:
782 return text
783 runs = try_get(item, lambda x: x['runs'], list) or []
784 if not runs and isinstance(item, list):
785 runs = item
786
787 runs = runs[:min(len(runs), max_runs or len(runs))]
788 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[]))
789 if text:
790 return text
791
792 def _get_count(self, data, *path_list):
793 count_text = self._get_text(data, *path_list) or ''
794 count = parse_count(count_text)
795 if count is None:
796 count = str_to_int(
797 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
798 return count
799
800 @staticmethod
801 def _extract_thumbnails(data, *path_list):
802 """
803 Extract thumbnails from thumbnails dict
804 @param path_list: path list to level that contains 'thumbnails' key
805 """
806 thumbnails = []
807 for path in path_list or [()]:
808 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]):
809 thumbnail_url = url_or_none(thumbnail.get('url'))
810 if not thumbnail_url:
811 continue
812 # Sometimes youtube gives a wrong thumbnail URL. See:
813 # https://github.com/yt-dlp/yt-dlp/issues/233
814 # https://github.com/ytdl-org/youtube-dl/issues/28023
815 if 'maxresdefault' in thumbnail_url:
816 thumbnail_url = thumbnail_url.split('?')[0]
817 thumbnails.append({
818 'url': thumbnail_url,
819 'height': int_or_none(thumbnail.get('height')),
820 'width': int_or_none(thumbnail.get('width')),
821 })
822 return thumbnails
823
824 @staticmethod
825 def extract_relative_time(relative_time_text):
826 """
827 Extracts a relative time from string and converts to dt object
828 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
829 """
830 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)
831 if mobj:
832 start = mobj.group('start')
833 if start:
834 return datetime_from_str(start)
835 try:
836 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
837 except ValueError:
838 return None
839
840 def _parse_time_text(self, text):
841 if not text:
842 return
843 dt = self.extract_relative_time(text)
844 timestamp = None
845 if isinstance(dt, datetime.datetime):
846 timestamp = calendar.timegm(dt.timetuple())
847
848 if timestamp is None:
849 timestamp = (
850 unified_timestamp(text) or unified_timestamp(
851 self._search_regex(
852 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
853 text.lower(), 'time text', default=None)))
854
855 if text and timestamp is None and self._preferred_lang in (None, 'en'):
856 self.report_warning(
857 f'Cannot parse localized time text "{text}"', only_once=True)
858 return timestamp
859
860 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
861 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
862 default_client='web'):
863 for retry in self.RetryManager():
864 try:
865 response = self._call_api(
866 ep=ep, fatal=True, headers=headers,
867 video_id=item_id, query=query, note=note,
868 context=self._extract_context(ytcfg, default_client),
869 api_key=self._extract_api_key(ytcfg, default_client),
870 api_hostname=api_hostname, default_client=default_client)
871 except ExtractorError as e:
872 if not isinstance(e.cause, network_exceptions):
873 return self._error_or_warning(e, fatal=fatal)
874 elif not isinstance(e.cause, urllib.error.HTTPError):
875 retry.error = e
876 continue
877
878 first_bytes = e.cause.read(512)
879 if not is_html(first_bytes):
880 yt_error = try_get(
881 self._parse_json(
882 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
883 lambda x: x['error']['message'], str)
884 if yt_error:
885 self._report_alerts([('ERROR', yt_error)], fatal=False)
886 # Downloading page may result in intermittent 5xx HTTP error
887 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
888 # We also want to catch all other network exceptions since errors in later pages can be troublesome
889 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
890 if e.cause.code not in (403, 429):
891 retry.error = e
892 continue
893 return self._error_or_warning(e, fatal=fatal)
894
895 try:
896 self._extract_and_report_alerts(response, only_once=True)
897 except ExtractorError as e:
898 # YouTube servers may return errors we want to retry on in a 200 OK response
899 # See: https://github.com/yt-dlp/yt-dlp/issues/839
900 if 'unknown error' in e.msg.lower():
901 retry.error = e
902 continue
903 return self._error_or_warning(e, fatal=fatal)
904 # Youtube sometimes sends incomplete data
905 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
906 if not traverse_obj(response, *variadic(check_get_keys)):
907 retry.error = ExtractorError('Incomplete data received', expected=True)
908 continue
909
910 return response
911
912 @staticmethod
913 def is_music_url(url):
914 return re.match(r'https?://music\.youtube\.com/', url) is not None
915
916 def _extract_video(self, renderer):
917 video_id = renderer.get('videoId')
918
919 reel_header_renderer = traverse_obj(renderer, (
920 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
921 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
922
923 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
924 description = self._get_text(renderer, 'descriptionSnippet')
925
926 duration = int_or_none(renderer.get('lengthSeconds'))
927 if duration is None:
928 duration = parse_duration(self._get_text(
929 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
930 if duration is None:
931 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
932 duration = parse_duration(self._search_regex(
933 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
934 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
935 video_id, default=None, group='duration'))
936
937 channel_id = traverse_obj(
938 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
939 expected_type=str, get_all=False)
940 if not channel_id:
941 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
942
943 overlay_style = traverse_obj(
944 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
945 get_all=False, expected_type=str)
946 badges = self._extract_badges(renderer)
947
948 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
949 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
950 expected_type=str)) or ''
951 url = f'https://www.youtube.com/watch?v={video_id}'
952 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
953 url = f'https://www.youtube.com/shorts/{video_id}'
954
955 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
956 or self._get_text(reel_header_renderer, 'timestampText') or '')
957 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
958
959 live_status = (
960 'is_upcoming' if scheduled_timestamp is not None
961 else 'was_live' if 'streamed' in time_text.lower()
962 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
963 else None)
964
965 # videoInfo is a string like '50K views • 10 years ago'.
966 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
967 view_count = (0 if 'no views' in view_count_text.lower()
968 else self._get_count({'simpleText': view_count_text}))
969 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
970
971 return {
972 '_type': 'url',
973 'ie_key': YoutubeIE.ie_key(),
974 'id': video_id,
975 'url': url,
976 'title': title,
977 'description': description,
978 'duration': duration,
979 'channel_id': channel_id,
980 'channel': (self._get_text(renderer, 'ownerText', 'shortBylineText')
981 or self._get_text(reel_header_renderer, 'channelTitleText')),
982 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
983 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
984 'timestamp': (self._parse_time_text(time_text)
985 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
986 else None),
987 'release_timestamp': scheduled_timestamp,
988 'availability':
989 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
990 else self._availability(
991 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
992 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
993 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
994 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
995 view_count_field: view_count,
996 'live_status': live_status
997 }
998
999
1000 class YoutubeIE(YoutubeBaseInfoExtractor):
1001 IE_DESC = 'YouTube'
1002 _VALID_URL = r"""(?x)^
1003 (
1004 (?:https?://|//) # http(s):// or protocol-independent URL
1005 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1006 (?:www\.)?deturl\.com/www\.youtube\.com|
1007 (?:www\.)?pwnyoutube\.com|
1008 (?:www\.)?hooktube\.com|
1009 (?:www\.)?yourepeat\.com|
1010 tube\.majestyc\.net|
1011 %(invidious)s|
1012 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1013 (?:.*?\#/)? # handle anchor (#/) redirect urls
1014 (?: # the various things that can precede the ID:
1015 (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1016 |(?: # or the v= param in all its forms
1017 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1018 (?:\?|\#!?) # the params delimiter ? or # or #!
1019 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1020 v=
1021 )
1022 ))
1023 |(?:
1024 youtu\.be| # just youtu.be/xxxx
1025 vid\.plus| # or vid.plus/xxxx
1026 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1027 %(invidious)s
1028 )/
1029 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1030 )
1031 )? # all until now is optional -> you can pass the naked ID
1032 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1033 (?(1).+)? # if we found the ID, everything can follow
1034 (?:\#|$)""" % {
1035 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1036 }
1037 _EMBED_REGEX = [
1038 r'''(?x)
1039 (?:
1040 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1041 data-video-url=|
1042 <embed[^>]+?src=|
1043 embedSWF\(?:\s*|
1044 <object[^>]+data=|
1045 new\s+SWFObject\(
1046 )
1047 (["\'])
1048 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1049 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1050 \1''',
1051 # https://wordpress.org/plugins/lazy-load-for-videos/
1052 r'''(?xs)
1053 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1054 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1055 ]
1056 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1057
1058 _PLAYER_INFO_RE = (
1059 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1060 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$',
1061 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1062 )
1063 _formats = {
1064 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1065 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1066 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1067 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1068 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1069 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1070 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1071 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1072 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1073 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1074 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1075 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1076 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1077 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1078 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1079 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1080 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1081 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1082
1083
1084 # 3D videos
1085 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1086 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1087 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1088 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1089 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1090 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1091 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1092
1093 # Apple HTTP Live Streaming
1094 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1095 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1096 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1097 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1098 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1099 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1100 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1101 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1102
1103 # DASH mp4 video
1104 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1105 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1106 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1107 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1108 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1109 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1110 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1111 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1112 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1113 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1114 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1115 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1116
1117 # Dash mp4 audio
1118 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1119 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1120 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1121 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1122 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1123 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1124 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1125
1126 # Dash webm
1127 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1128 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1129 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1130 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1131 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1132 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1133 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1134 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1135 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1136 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1137 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1138 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1139 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1140 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1141 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1142 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1143 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1144 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1145 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1146 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1147 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1148 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1149
1150 # Dash webm audio
1151 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1152 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1153
1154 # Dash webm audio with opus inside
1155 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1156 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1157 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1158
1159 # RTMP (unnamed)
1160 '_rtmp': {'protocol': 'rtmp'},
1161
1162 # av01 video only formats sometimes served with "unknown" codecs
1163 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1164 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1165 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1166 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1167 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1168 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1169 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1170 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1171 }
1172 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1173
1174 _GEO_BYPASS = False
1175
1176 IE_NAME = 'youtube'
1177 _TESTS = [
1178 {
1179 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1180 'info_dict': {
1181 'id': 'BaW_jenozKc',
1182 'ext': 'mp4',
1183 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1184 'uploader': 'Philipp Hagemeister',
1185 'uploader_id': 'phihag',
1186 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1187 'channel': 'Philipp Hagemeister',
1188 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1189 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1190 'upload_date': '20121002',
1191 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1192 'categories': ['Science & Technology'],
1193 'tags': ['youtube-dl'],
1194 'duration': 10,
1195 'view_count': int,
1196 'like_count': int,
1197 'availability': 'public',
1198 'playable_in_embed': True,
1199 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1200 'live_status': 'not_live',
1201 'age_limit': 0,
1202 'start_time': 1,
1203 'end_time': 9,
1204 'comment_count': int,
1205 'channel_follower_count': int
1206 }
1207 },
1208 {
1209 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1210 'note': 'Embed-only video (#1746)',
1211 'info_dict': {
1212 'id': 'yZIXLfi8CZQ',
1213 'ext': 'mp4',
1214 'upload_date': '20120608',
1215 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1216 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1217 'uploader': 'SET India',
1218 'uploader_id': 'setindia',
1219 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
1220 'age_limit': 18,
1221 },
1222 'skip': 'Private video',
1223 },
1224 {
1225 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1226 'note': 'Use the first video ID in the URL',
1227 'info_dict': {
1228 'id': 'BaW_jenozKc',
1229 'ext': 'mp4',
1230 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1231 'uploader': 'Philipp Hagemeister',
1232 'uploader_id': 'phihag',
1233 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
1234 'channel': 'Philipp Hagemeister',
1235 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1236 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1237 'upload_date': '20121002',
1238 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1239 'categories': ['Science & Technology'],
1240 'tags': ['youtube-dl'],
1241 'duration': 10,
1242 'view_count': int,
1243 'like_count': int,
1244 'availability': 'public',
1245 'playable_in_embed': True,
1246 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1247 'live_status': 'not_live',
1248 'age_limit': 0,
1249 'comment_count': int,
1250 'channel_follower_count': int
1251 },
1252 'params': {
1253 'skip_download': True,
1254 },
1255 },
1256 {
1257 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1258 'note': '256k DASH audio (format 141) via DASH manifest',
1259 'info_dict': {
1260 'id': 'a9LDPn-MO4I',
1261 'ext': 'm4a',
1262 'upload_date': '20121002',
1263 'uploader_id': '8KVIDEO',
1264 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
1265 'description': '',
1266 'uploader': '8KVIDEO',
1267 'title': 'UHDTV TEST 8K VIDEO.mp4'
1268 },
1269 'params': {
1270 'youtube_include_dash_manifest': True,
1271 'format': '141',
1272 },
1273 'skip': 'format 141 not served anymore',
1274 },
1275 # DASH manifest with encrypted signature
1276 {
1277 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1278 'info_dict': {
1279 'id': 'IB3lcPjvWLA',
1280 'ext': 'm4a',
1281 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1282 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1283 'duration': 244,
1284 'uploader': 'AfrojackVEVO',
1285 'uploader_id': 'AfrojackVEVO',
1286 'upload_date': '20131011',
1287 'abr': 129.495,
1288 'like_count': int,
1289 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1290 'playable_in_embed': True,
1291 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1292 'view_count': int,
1293 'track': 'The Spark',
1294 'live_status': 'not_live',
1295 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1296 'channel': 'Afrojack',
1297 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO',
1298 'tags': 'count:19',
1299 'availability': 'public',
1300 'categories': ['Music'],
1301 'age_limit': 0,
1302 'alt_title': 'The Spark',
1303 'channel_follower_count': int
1304 },
1305 'params': {
1306 'youtube_include_dash_manifest': True,
1307 'format': '141/bestaudio[ext=m4a]',
1308 },
1309 },
1310 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1311 {
1312 'note': 'Embed allowed age-gate video',
1313 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1314 'info_dict': {
1315 'id': 'HtVdAasjOgU',
1316 'ext': 'mp4',
1317 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1318 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1319 'duration': 142,
1320 'uploader': 'The Witcher',
1321 'uploader_id': 'WitcherGame',
1322 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
1323 'upload_date': '20140605',
1324 'age_limit': 18,
1325 'categories': ['Gaming'],
1326 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1327 'availability': 'needs_auth',
1328 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1329 'like_count': int,
1330 'channel': 'The Witcher',
1331 'live_status': 'not_live',
1332 'tags': 'count:17',
1333 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1334 'playable_in_embed': True,
1335 'view_count': int,
1336 'channel_follower_count': int
1337 },
1338 },
1339 {
1340 'note': 'Age-gate video with embed allowed in public site',
1341 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1342 'info_dict': {
1343 'id': 'HsUATh_Nc2U',
1344 'ext': 'mp4',
1345 'title': 'Godzilla 2 (Official Video)',
1346 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1347 'upload_date': '20200408',
1348 'uploader_id': 'FlyingKitty900',
1349 'uploader': 'FlyingKitty',
1350 'age_limit': 18,
1351 'availability': 'needs_auth',
1352 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1353 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900',
1354 'channel': 'FlyingKitty',
1355 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1356 'view_count': int,
1357 'categories': ['Entertainment'],
1358 'live_status': 'not_live',
1359 'tags': ['Flyingkitty', 'godzilla 2'],
1360 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1361 'like_count': int,
1362 'duration': 177,
1363 'playable_in_embed': True,
1364 'channel_follower_count': int
1365 },
1366 },
1367 {
1368 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1369 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1370 'info_dict': {
1371 'id': 'Tq92D6wQ1mg',
1372 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1373 'ext': 'mp4',
1374 'upload_date': '20191228',
1375 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1376 'uploader': 'Projekt Melody',
1377 'description': 'md5:17eccca93a786d51bc67646756894066',
1378 'age_limit': 18,
1379 'like_count': int,
1380 'availability': 'needs_auth',
1381 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1382 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1383 'view_count': int,
1384 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1385 'channel': 'Projekt Melody',
1386 'live_status': 'not_live',
1387 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1388 'playable_in_embed': True,
1389 'categories': ['Entertainment'],
1390 'duration': 106,
1391 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1392 'comment_count': int,
1393 'channel_follower_count': int
1394 },
1395 },
1396 {
1397 'note': 'Non-Agegated non-embeddable video',
1398 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1399 'info_dict': {
1400 'id': 'MeJVWBSsPAY',
1401 'ext': 'mp4',
1402 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1403 'uploader': 'Herr Lurik',
1404 'uploader_id': 'st3in234',
1405 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1406 'upload_date': '20130730',
1407 'track': 'Such mich find mich',
1408 'age_limit': 0,
1409 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1410 'like_count': int,
1411 'playable_in_embed': False,
1412 'creator': 'OOMPH!',
1413 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1414 'view_count': int,
1415 'alt_title': 'Such mich find mich',
1416 'duration': 210,
1417 'channel': 'Herr Lurik',
1418 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1419 'categories': ['Music'],
1420 'availability': 'public',
1421 'uploader_url': 'http://www.youtube.com/user/st3in234',
1422 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1423 'live_status': 'not_live',
1424 'artist': 'OOMPH!',
1425 'channel_follower_count': int
1426 },
1427 },
1428 {
1429 'note': 'Non-bypassable age-gated video',
1430 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1431 'only_matching': True,
1432 },
1433 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1434 # YouTube Red ad is not captured for creator
1435 {
1436 'url': '__2ABJjxzNo',
1437 'info_dict': {
1438 'id': '__2ABJjxzNo',
1439 'ext': 'mp4',
1440 'duration': 266,
1441 'upload_date': '20100430',
1442 'uploader_id': 'deadmau5',
1443 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
1444 'creator': 'deadmau5',
1445 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1446 'uploader': 'deadmau5',
1447 'title': 'Deadmau5 - Some Chords (HD)',
1448 'alt_title': 'Some Chords',
1449 'availability': 'public',
1450 'tags': 'count:14',
1451 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1452 'view_count': int,
1453 'live_status': 'not_live',
1454 'channel': 'deadmau5',
1455 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1456 'like_count': int,
1457 'track': 'Some Chords',
1458 'artist': 'deadmau5',
1459 'playable_in_embed': True,
1460 'age_limit': 0,
1461 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1462 'categories': ['Music'],
1463 'album': 'Some Chords',
1464 'channel_follower_count': int
1465 },
1466 'expected_warnings': [
1467 'DASH manifest missing',
1468 ]
1469 },
1470 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1471 {
1472 'url': 'lqQg6PlCWgI',
1473 'info_dict': {
1474 'id': 'lqQg6PlCWgI',
1475 'ext': 'mp4',
1476 'duration': 6085,
1477 'upload_date': '20150827',
1478 'uploader_id': 'olympic',
1479 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
1480 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1481 'uploader': 'Olympics',
1482 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1483 'like_count': int,
1484 'release_timestamp': 1343767800,
1485 'playable_in_embed': True,
1486 'categories': ['Sports'],
1487 'release_date': '20120731',
1488 'channel': 'Olympics',
1489 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1490 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1491 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1492 'age_limit': 0,
1493 'availability': 'public',
1494 'live_status': 'was_live',
1495 'view_count': int,
1496 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1497 'channel_follower_count': int
1498 },
1499 'params': {
1500 'skip_download': 'requires avconv',
1501 }
1502 },
1503 # Non-square pixels
1504 {
1505 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1506 'info_dict': {
1507 'id': '_b-2C3KPAM0',
1508 'ext': 'mp4',
1509 'stretched_ratio': 16 / 9.,
1510 'duration': 85,
1511 'upload_date': '20110310',
1512 'uploader_id': 'AllenMeow',
1513 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
1514 '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',
1515 'uploader': '孫ᄋᄅ',
1516 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1517 'playable_in_embed': True,
1518 'channel': '孫ᄋᄅ',
1519 'age_limit': 0,
1520 'tags': 'count:11',
1521 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1522 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1523 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1524 'view_count': int,
1525 'categories': ['People & Blogs'],
1526 'like_count': int,
1527 'live_status': 'not_live',
1528 'availability': 'unlisted',
1529 'comment_count': int,
1530 'channel_follower_count': int
1531 },
1532 },
1533 # url_encoded_fmt_stream_map is empty string
1534 {
1535 'url': 'qEJwOuvDf7I',
1536 'info_dict': {
1537 'id': 'qEJwOuvDf7I',
1538 'ext': 'webm',
1539 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1540 'description': '',
1541 'upload_date': '20150404',
1542 'uploader_id': 'spbelect',
1543 'uploader': 'Наблюдатели Петербурга',
1544 },
1545 'params': {
1546 'skip_download': 'requires avconv',
1547 },
1548 'skip': 'This live event has ended.',
1549 },
1550 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1551 {
1552 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1553 'info_dict': {
1554 'id': 'FIl7x6_3R5Y',
1555 'ext': 'webm',
1556 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1557 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1558 'duration': 220,
1559 'upload_date': '20150625',
1560 'uploader_id': 'dorappi2000',
1561 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
1562 'uploader': 'dorappi2000',
1563 'formats': 'mincount:31',
1564 },
1565 'skip': 'not actual anymore',
1566 },
1567 # DASH manifest with segment_list
1568 {
1569 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1570 'md5': '8ce563a1d667b599d21064e982ab9e31',
1571 'info_dict': {
1572 'id': 'CsmdDsKjzN8',
1573 'ext': 'mp4',
1574 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1575 'uploader': 'Airtek',
1576 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1577 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
1578 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1579 },
1580 'params': {
1581 'youtube_include_dash_manifest': True,
1582 'format': '135', # bestvideo
1583 },
1584 'skip': 'This live event has ended.',
1585 },
1586 {
1587 # Multifeed videos (multiple cameras), URL can be of any Camera
1588 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1589 'info_dict': {
1590 'id': 'zaPI8MvL8pg',
1591 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1592 'description': 'md5:563ccbc698b39298481ca3c571169519',
1593 },
1594 'playlist': [{
1595 'info_dict': {
1596 'id': 'j5yGuxZ8lLU',
1597 'ext': 'mp4',
1598 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1599 'uploader': 'WiiLikeToPlay',
1600 'description': 'md5:563ccbc698b39298481ca3c571169519',
1601 'uploader_url': 'http://www.youtube.com/user/WiiRikeToPray',
1602 'duration': 10120,
1603 'channel_follower_count': int,
1604 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1605 'availability': 'public',
1606 'playable_in_embed': True,
1607 'upload_date': '20131105',
1608 'uploader_id': 'WiiRikeToPray',
1609 'categories': ['Gaming'],
1610 'live_status': 'was_live',
1611 'tags': 'count:24',
1612 'release_timestamp': 1383701910,
1613 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1614 'comment_count': int,
1615 'age_limit': 0,
1616 'like_count': int,
1617 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1618 'channel': 'WiiLikeToPlay',
1619 'view_count': int,
1620 'release_date': '20131106',
1621 },
1622 }, {
1623 'info_dict': {
1624 'id': 'zaPI8MvL8pg',
1625 'ext': 'mp4',
1626 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1627 'uploader_id': 'WiiRikeToPray',
1628 'availability': 'public',
1629 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1630 'channel': 'WiiLikeToPlay',
1631 'uploader_url': 'http://www.youtube.com/user/WiiRikeToPray',
1632 'channel_follower_count': int,
1633 'description': 'md5:563ccbc698b39298481ca3c571169519',
1634 'duration': 10108,
1635 'age_limit': 0,
1636 'like_count': int,
1637 'tags': 'count:24',
1638 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1639 'uploader': 'WiiLikeToPlay',
1640 'release_timestamp': 1383701915,
1641 'comment_count': int,
1642 'upload_date': '20131105',
1643 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1644 'release_date': '20131106',
1645 'playable_in_embed': True,
1646 'live_status': 'was_live',
1647 'categories': ['Gaming'],
1648 'view_count': int,
1649 },
1650 }, {
1651 'info_dict': {
1652 'id': 'R7r3vfO7Hao',
1653 'ext': 'mp4',
1654 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1655 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1656 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1657 'like_count': int,
1658 'availability': 'public',
1659 'playable_in_embed': True,
1660 'upload_date': '20131105',
1661 'description': 'md5:563ccbc698b39298481ca3c571169519',
1662 'uploader_id': 'WiiRikeToPray',
1663 'uploader_url': 'http://www.youtube.com/user/WiiRikeToPray',
1664 'channel_follower_count': int,
1665 'tags': 'count:24',
1666 'release_date': '20131106',
1667 'uploader': 'WiiLikeToPlay',
1668 'comment_count': int,
1669 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1670 'channel': 'WiiLikeToPlay',
1671 'categories': ['Gaming'],
1672 'release_timestamp': 1383701914,
1673 'live_status': 'was_live',
1674 'age_limit': 0,
1675 'duration': 10128,
1676 'view_count': int,
1677 },
1678 }],
1679 'params': {'skip_download': True},
1680 },
1681 {
1682 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1683 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1684 'info_dict': {
1685 'id': 'gVfLd0zydlo',
1686 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1687 },
1688 'playlist_count': 2,
1689 'skip': 'Not multifeed anymore',
1690 },
1691 {
1692 'url': 'https://vid.plus/FlRa-iH7PGw',
1693 'only_matching': True,
1694 },
1695 {
1696 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1697 'only_matching': True,
1698 },
1699 {
1700 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1701 # Also tests cut-off URL expansion in video description (see
1702 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1703 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1704 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1705 'info_dict': {
1706 'id': 'lsguqyKfVQg',
1707 'ext': 'mp4',
1708 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1709 'alt_title': 'Dark Walk',
1710 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1711 'duration': 133,
1712 'upload_date': '20151119',
1713 'uploader_id': 'IronSoulElf',
1714 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
1715 'uploader': 'IronSoulElf',
1716 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1717 'track': 'Dark Walk',
1718 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1719 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1720 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1721 'categories': ['Film & Animation'],
1722 'view_count': int,
1723 'live_status': 'not_live',
1724 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1725 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1726 'tags': 'count:13',
1727 'availability': 'public',
1728 'channel': 'IronSoulElf',
1729 'playable_in_embed': True,
1730 'like_count': int,
1731 'age_limit': 0,
1732 'channel_follower_count': int
1733 },
1734 'params': {
1735 'skip_download': True,
1736 },
1737 },
1738 {
1739 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1740 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1741 'only_matching': True,
1742 },
1743 {
1744 # Video with yt:stretch=17:0
1745 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1746 'info_dict': {
1747 'id': 'Q39EVAstoRM',
1748 'ext': 'mp4',
1749 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1750 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1751 'upload_date': '20151107',
1752 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
1753 'uploader': 'CH GAMER DROID',
1754 },
1755 'params': {
1756 'skip_download': True,
1757 },
1758 'skip': 'This video does not exist.',
1759 },
1760 {
1761 # Video with incomplete 'yt:stretch=16:'
1762 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1763 'only_matching': True,
1764 },
1765 {
1766 # Video licensed under Creative Commons
1767 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1768 'info_dict': {
1769 'id': 'M4gD1WSo5mA',
1770 'ext': 'mp4',
1771 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1772 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1773 'duration': 721,
1774 'upload_date': '20150128',
1775 'uploader_id': 'BerkmanCenter',
1776 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1777 'uploader': 'The Berkman Klein Center for Internet & Society',
1778 'license': 'Creative Commons Attribution license (reuse allowed)',
1779 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1780 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1781 'like_count': int,
1782 'age_limit': 0,
1783 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1784 'channel': 'The Berkman Klein Center for Internet & Society',
1785 'availability': 'public',
1786 'view_count': int,
1787 'categories': ['Education'],
1788 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1789 'live_status': 'not_live',
1790 'playable_in_embed': True,
1791 'comment_count': int,
1792 'channel_follower_count': int,
1793 'chapters': list,
1794 },
1795 'params': {
1796 'skip_download': True,
1797 },
1798 },
1799 {
1800 # Channel-like uploader_url
1801 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1802 'info_dict': {
1803 'id': 'eQcmzGIKrzg',
1804 'ext': 'mp4',
1805 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1806 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1807 'duration': 4060,
1808 'upload_date': '20151120',
1809 'uploader': 'Bernie Sanders',
1810 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1811 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1812 'license': 'Creative Commons Attribution license (reuse allowed)',
1813 'playable_in_embed': True,
1814 'tags': 'count:12',
1815 'like_count': int,
1816 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1817 'age_limit': 0,
1818 'availability': 'public',
1819 'categories': ['News & Politics'],
1820 'channel': 'Bernie Sanders',
1821 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1822 'view_count': int,
1823 'live_status': 'not_live',
1824 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1825 'comment_count': int,
1826 'channel_follower_count': int,
1827 'chapters': list,
1828 },
1829 'params': {
1830 'skip_download': True,
1831 },
1832 },
1833 {
1834 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1835 'only_matching': True,
1836 },
1837 {
1838 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1839 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1840 'only_matching': True,
1841 },
1842 {
1843 # Rental video preview
1844 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1845 'info_dict': {
1846 'id': 'uGpuVWrhIzE',
1847 'ext': 'mp4',
1848 'title': 'Piku - Trailer',
1849 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1850 'upload_date': '20150811',
1851 'uploader': 'FlixMatrix',
1852 'uploader_id': 'FlixMatrixKaravan',
1853 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1854 'license': 'Standard YouTube License',
1855 },
1856 'params': {
1857 'skip_download': True,
1858 },
1859 'skip': 'This video is not available.',
1860 },
1861 {
1862 # YouTube Red video with episode data
1863 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1864 'info_dict': {
1865 'id': 'iqKdEhx-dD4',
1866 'ext': 'mp4',
1867 'title': 'Isolation - Mind Field (Ep 1)',
1868 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1869 'duration': 2085,
1870 'upload_date': '20170118',
1871 'uploader': 'Vsauce',
1872 'uploader_id': 'Vsauce',
1873 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1874 'series': 'Mind Field',
1875 'season_number': 1,
1876 'episode_number': 1,
1877 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1878 'tags': 'count:12',
1879 'view_count': int,
1880 'availability': 'public',
1881 'age_limit': 0,
1882 'channel': 'Vsauce',
1883 'episode': 'Episode 1',
1884 'categories': ['Entertainment'],
1885 'season': 'Season 1',
1886 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1887 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1888 'like_count': int,
1889 'playable_in_embed': True,
1890 'live_status': 'not_live',
1891 'channel_follower_count': int
1892 },
1893 'params': {
1894 'skip_download': True,
1895 },
1896 'expected_warnings': [
1897 'Skipping DASH manifest',
1898 ],
1899 },
1900 {
1901 # The following content has been identified by the YouTube community
1902 # as inappropriate or offensive to some audiences.
1903 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1904 'info_dict': {
1905 'id': '6SJNVb0GnPI',
1906 'ext': 'mp4',
1907 'title': 'Race Differences in Intelligence',
1908 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1909 'duration': 965,
1910 'upload_date': '20140124',
1911 'uploader': 'New Century Foundation',
1912 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1913 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1914 },
1915 'params': {
1916 'skip_download': True,
1917 },
1918 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1919 },
1920 {
1921 # itag 212
1922 'url': '1t24XAntNCY',
1923 'only_matching': True,
1924 },
1925 {
1926 # geo restricted to JP
1927 'url': 'sJL6WA-aGkQ',
1928 'only_matching': True,
1929 },
1930 {
1931 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1932 'only_matching': True,
1933 },
1934 {
1935 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1936 'only_matching': True,
1937 },
1938 {
1939 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1940 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1941 'only_matching': True,
1942 },
1943 {
1944 # DRM protected
1945 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1946 'only_matching': True,
1947 },
1948 {
1949 # Video with unsupported adaptive stream type formats
1950 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1951 'info_dict': {
1952 'id': 'Z4Vy8R84T1U',
1953 'ext': 'mp4',
1954 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1955 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1956 'duration': 433,
1957 'upload_date': '20130923',
1958 'uploader': 'Amelia Putri Harwita',
1959 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1960 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1961 'formats': 'maxcount:10',
1962 },
1963 'params': {
1964 'skip_download': True,
1965 'youtube_include_dash_manifest': False,
1966 },
1967 'skip': 'not actual anymore',
1968 },
1969 {
1970 # Youtube Music Auto-generated description
1971 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1972 'info_dict': {
1973 'id': 'MgNrAu2pzNs',
1974 'ext': 'mp4',
1975 'title': 'Voyeur Girl',
1976 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1977 'upload_date': '20190312',
1978 'uploader': 'Stephen - Topic',
1979 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1980 'artist': 'Stephen',
1981 'track': 'Voyeur Girl',
1982 'album': 'it\'s too much love to know my dear',
1983 'release_date': '20190313',
1984 'release_year': 2019,
1985 'alt_title': 'Voyeur Girl',
1986 'view_count': int,
1987 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1988 'playable_in_embed': True,
1989 'like_count': int,
1990 'categories': ['Music'],
1991 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
1992 'channel': 'Stephen',
1993 'availability': 'public',
1994 'creator': 'Stephen',
1995 'duration': 169,
1996 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
1997 'age_limit': 0,
1998 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1999 'tags': 'count:11',
2000 'live_status': 'not_live',
2001 'channel_follower_count': int
2002 },
2003 'params': {
2004 'skip_download': True,
2005 },
2006 },
2007 {
2008 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2009 'only_matching': True,
2010 },
2011 {
2012 # invalid -> valid video id redirection
2013 'url': 'DJztXj2GPfl',
2014 'info_dict': {
2015 'id': 'DJztXj2GPfk',
2016 'ext': 'mp4',
2017 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2018 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2019 'upload_date': '20090125',
2020 'uploader': 'Prochorowka',
2021 'uploader_id': 'Prochorowka',
2022 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
2023 'artist': 'Panjabi MC',
2024 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2025 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2026 },
2027 'params': {
2028 'skip_download': True,
2029 },
2030 'skip': 'Video unavailable',
2031 },
2032 {
2033 # empty description results in an empty string
2034 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2035 'info_dict': {
2036 'id': 'x41yOUIvK2k',
2037 'ext': 'mp4',
2038 'title': 'IMG 3456',
2039 'description': '',
2040 'upload_date': '20170613',
2041 'uploader_id': 'ElevageOrVert',
2042 'uploader': 'ElevageOrVert',
2043 'view_count': int,
2044 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2045 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert',
2046 'like_count': int,
2047 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2048 'tags': [],
2049 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2050 'availability': 'public',
2051 'age_limit': 0,
2052 'categories': ['Pets & Animals'],
2053 'duration': 7,
2054 'playable_in_embed': True,
2055 'live_status': 'not_live',
2056 'channel': 'ElevageOrVert',
2057 'channel_follower_count': int
2058 },
2059 'params': {
2060 'skip_download': True,
2061 },
2062 },
2063 {
2064 # with '};' inside yt initial data (see [1])
2065 # see [2] for an example with '};' inside ytInitialPlayerResponse
2066 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2067 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2068 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2069 'info_dict': {
2070 'id': 'CHqg6qOn4no',
2071 'ext': 'mp4',
2072 'title': 'Part 77 Sort a list of simple types in c#',
2073 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2074 'upload_date': '20130831',
2075 'uploader_id': 'kudvenkat',
2076 'uploader': 'kudvenkat',
2077 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2078 'like_count': int,
2079 'uploader_url': 'http://www.youtube.com/user/kudvenkat',
2080 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2081 'live_status': 'not_live',
2082 'categories': ['Education'],
2083 'availability': 'public',
2084 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2085 'tags': 'count:12',
2086 'playable_in_embed': True,
2087 'age_limit': 0,
2088 'view_count': int,
2089 'duration': 522,
2090 'channel': 'kudvenkat',
2091 'comment_count': int,
2092 'channel_follower_count': int,
2093 'chapters': list,
2094 },
2095 'params': {
2096 'skip_download': True,
2097 },
2098 },
2099 {
2100 # another example of '};' in ytInitialData
2101 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2102 'only_matching': True,
2103 },
2104 {
2105 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2106 'only_matching': True,
2107 },
2108 {
2109 # https://github.com/ytdl-org/youtube-dl/pull/28094
2110 'url': 'OtqTfy26tG0',
2111 'info_dict': {
2112 'id': 'OtqTfy26tG0',
2113 'ext': 'mp4',
2114 'title': 'Burn Out',
2115 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2116 'upload_date': '20141120',
2117 'uploader': 'The Cinematic Orchestra - Topic',
2118 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2119 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2120 'artist': 'The Cinematic Orchestra',
2121 'track': 'Burn Out',
2122 'album': 'Every Day',
2123 'like_count': int,
2124 'live_status': 'not_live',
2125 'alt_title': 'Burn Out',
2126 'duration': 614,
2127 'age_limit': 0,
2128 'view_count': int,
2129 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2130 'creator': 'The Cinematic Orchestra',
2131 'channel': 'The Cinematic Orchestra',
2132 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2133 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2134 'availability': 'public',
2135 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2136 'categories': ['Music'],
2137 'playable_in_embed': True,
2138 'channel_follower_count': int
2139 },
2140 'params': {
2141 'skip_download': True,
2142 },
2143 },
2144 {
2145 # controversial video, only works with bpctr when authenticated with cookies
2146 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2147 'only_matching': True,
2148 },
2149 {
2150 # controversial video, requires bpctr/contentCheckOk
2151 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2152 'info_dict': {
2153 'id': 'SZJvDhaSDnc',
2154 'ext': 'mp4',
2155 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2156 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2157 'uploader': 'CBS Mornings',
2158 'uploader_id': 'CBSThisMorning',
2159 'upload_date': '20140716',
2160 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2161 'duration': 170,
2162 'categories': ['News & Politics'],
2163 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning',
2164 'view_count': int,
2165 'channel': 'CBS Mornings',
2166 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2167 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2168 'age_limit': 18,
2169 'availability': 'needs_auth',
2170 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2171 'like_count': int,
2172 'live_status': 'not_live',
2173 'playable_in_embed': True,
2174 'channel_follower_count': int
2175 }
2176 },
2177 {
2178 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2179 'url': 'cBvYw8_A0vQ',
2180 'info_dict': {
2181 'id': 'cBvYw8_A0vQ',
2182 'ext': 'mp4',
2183 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2184 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2185 'upload_date': '20201120',
2186 'uploader': 'Walk around Japan',
2187 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2188 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2189 'duration': 1456,
2190 'categories': ['Travel & Events'],
2191 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2192 'view_count': int,
2193 'channel': 'Walk around Japan',
2194 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2195 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2196 'age_limit': 0,
2197 'availability': 'public',
2198 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2199 'live_status': 'not_live',
2200 'playable_in_embed': True,
2201 'channel_follower_count': int
2202 },
2203 'params': {
2204 'skip_download': True,
2205 },
2206 }, {
2207 # Has multiple audio streams
2208 'url': 'WaOKSUlf4TM',
2209 'only_matching': True
2210 }, {
2211 # Requires Premium: has format 141 when requested using YTM url
2212 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2213 'only_matching': True
2214 }, {
2215 # multiple subtitles with same lang_code
2216 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2217 'only_matching': True,
2218 }, {
2219 # Force use android client fallback
2220 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2221 'info_dict': {
2222 'id': 'YOelRv7fMxY',
2223 'title': 'DIGGING A SECRET TUNNEL Part 1',
2224 'ext': '3gp',
2225 'upload_date': '20210624',
2226 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2227 'uploader': 'colinfurze',
2228 'uploader_id': 'colinfurze',
2229 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2230 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2231 'duration': 596,
2232 'categories': ['Entertainment'],
2233 'uploader_url': 'http://www.youtube.com/user/colinfurze',
2234 'view_count': int,
2235 'channel': 'colinfurze',
2236 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2237 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2238 'age_limit': 0,
2239 'availability': 'public',
2240 'like_count': int,
2241 'live_status': 'not_live',
2242 'playable_in_embed': True,
2243 'channel_follower_count': int,
2244 'chapters': list,
2245 },
2246 'params': {
2247 'format': '17', # 3gp format available on android
2248 'extractor_args': {'youtube': {'player_client': ['android']}},
2249 },
2250 },
2251 {
2252 # Skip download of additional client configs (remix client config in this case)
2253 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2254 'only_matching': True,
2255 'params': {
2256 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2257 },
2258 }, {
2259 # shorts
2260 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2261 'only_matching': True,
2262 }, {
2263 'note': 'Storyboards',
2264 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2265 'info_dict': {
2266 'id': '5KLPxDtMqe8',
2267 'ext': 'mhtml',
2268 'format_id': 'sb0',
2269 'title': 'Your Brain is Plastic',
2270 'uploader_id': 'scishow',
2271 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2272 'upload_date': '20140324',
2273 'uploader': 'SciShow',
2274 'like_count': int,
2275 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2276 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2277 'view_count': int,
2278 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2279 'playable_in_embed': True,
2280 'tags': 'count:12',
2281 'uploader_url': 'http://www.youtube.com/user/scishow',
2282 'availability': 'public',
2283 'channel': 'SciShow',
2284 'live_status': 'not_live',
2285 'duration': 248,
2286 'categories': ['Education'],
2287 'age_limit': 0,
2288 'channel_follower_count': int,
2289 'chapters': list,
2290 }, 'params': {'format': 'mhtml', 'skip_download': True}
2291 }, {
2292 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2293 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2294 'info_dict': {
2295 'id': '2NUZ8W2llS4',
2296 'ext': 'mp4',
2297 'title': 'The NP that test your phone performance 🙂',
2298 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2299 'uploader': 'Leon Nguyen',
2300 'uploader_id': 'VNSXIII',
2301 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2302 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2303 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2304 'duration': 21,
2305 'view_count': int,
2306 'age_limit': 0,
2307 'categories': ['Gaming'],
2308 'tags': 'count:23',
2309 'playable_in_embed': True,
2310 'live_status': 'not_live',
2311 'upload_date': '20220103',
2312 'like_count': int,
2313 'availability': 'public',
2314 'channel': 'Leon Nguyen',
2315 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2316 'comment_count': int,
2317 'channel_follower_count': int
2318 }
2319 }, {
2320 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2321 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2322 'info_dict': {
2323 'id': '2NUZ8W2llS4',
2324 'ext': 'mp4',
2325 'title': 'The NP that test your phone performance 🙂',
2326 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2327 'uploader': 'Leon Nguyen',
2328 'uploader_id': 'VNSXIII',
2329 'uploader_url': 'http://www.youtube.com/user/VNSXIII',
2330 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2331 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2332 'duration': 21,
2333 'view_count': int,
2334 'age_limit': 0,
2335 'categories': ['Gaming'],
2336 'tags': 'count:23',
2337 'playable_in_embed': True,
2338 'live_status': 'not_live',
2339 'upload_date': '20220102',
2340 'like_count': int,
2341 'availability': 'public',
2342 'channel': 'Leon Nguyen',
2343 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2344 'comment_count': int,
2345 'channel_follower_count': int
2346 },
2347 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2348 }, {
2349 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2350 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2351 'info_dict': {
2352 'id': 'mzZzzBU6lrM',
2353 'ext': 'mp4',
2354 'title': 'I Met GeorgeNotFound In Real Life...',
2355 'description': 'md5:cca98a355c7184e750f711f3a1b22c84',
2356 'uploader': 'Quackity',
2357 'uploader_id': 'QuackityHQ',
2358 'uploader_url': 'http://www.youtube.com/user/QuackityHQ',
2359 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2360 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2361 'duration': 955,
2362 'view_count': int,
2363 'age_limit': 0,
2364 'categories': ['Entertainment'],
2365 'tags': 'count:26',
2366 'playable_in_embed': True,
2367 'live_status': 'not_live',
2368 'release_timestamp': 1641172509,
2369 'release_date': '20220103',
2370 'upload_date': '20220103',
2371 'like_count': int,
2372 'availability': 'public',
2373 'channel': 'Quackity',
2374 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2375 'channel_follower_count': int
2376 }
2377 },
2378 { # continuous livestream. Microformat upload date should be preferred.
2379 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2380 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2381 'info_dict': {
2382 'id': 'kgx4WGK0oNU',
2383 '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}',
2384 'ext': 'mp4',
2385 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2386 'availability': 'public',
2387 'age_limit': 0,
2388 'release_timestamp': 1637975704,
2389 'upload_date': '20210619',
2390 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2391 'live_status': 'is_live',
2392 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2393 'uploader': '阿鲍Abao',
2394 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2395 'channel': 'Abao in Tokyo',
2396 'channel_follower_count': int,
2397 'release_date': '20211127',
2398 'tags': 'count:39',
2399 'categories': ['People & Blogs'],
2400 'like_count': int,
2401 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2402 'view_count': int,
2403 'playable_in_embed': True,
2404 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2405 'concurrent_view_count': int,
2406 },
2407 'params': {'skip_download': True}
2408 }, {
2409 # Story. Requires specific player params to work.
2410 'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2411 'info_dict': {
2412 'id': 'vv8qTUWmulI',
2413 'ext': 'mp4',
2414 'availability': 'unlisted',
2415 'view_count': int,
2416 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2417 'upload_date': '20220526',
2418 'categories': ['Education'],
2419 'title': 'Story',
2420 'channel': 'IT\'S HISTORY',
2421 'description': '',
2422 'uploader_id': 'BlastfromthePast',
2423 'duration': 12,
2424 'uploader': 'IT\'S HISTORY',
2425 'playable_in_embed': True,
2426 'age_limit': 0,
2427 'live_status': 'not_live',
2428 'tags': [],
2429 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2430 'uploader_url': 'http://www.youtube.com/user/BlastfromthePast',
2431 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2432 },
2433 'skip': 'stories get removed after some period of time',
2434 }, {
2435 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2436 'info_dict': {
2437 'id': 'tjjjtzRLHvA',
2438 'ext': 'mp4',
2439 'title': 'ハッシュタグ無し };if window.ytcsi',
2440 'upload_date': '20220323',
2441 'like_count': int,
2442 'availability': 'unlisted',
2443 'channel': 'nao20010128nao',
2444 'thumbnail': 'https://i.ytimg.com/vi_webp/tjjjtzRLHvA/maxresdefault.webp',
2445 'age_limit': 0,
2446 'uploader': 'nao20010128nao',
2447 'uploader_id': 'nao20010128nao',
2448 'categories': ['Music'],
2449 'view_count': int,
2450 'description': '',
2451 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2452 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2453 'live_status': 'not_live',
2454 'playable_in_embed': True,
2455 'channel_follower_count': int,
2456 'duration': 6,
2457 'tags': [],
2458 'uploader_url': 'http://www.youtube.com/user/nao20010128nao',
2459 }
2460 }, {
2461 # Prefer primary title+description language metadata by default
2462 # Do not prefer translated description if primary is empty
2463 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2464 'info_dict': {
2465 'id': 'el3E4MbxRqQ',
2466 'ext': 'mp4',
2467 'title': 'dlp test video 2 - primary sv no desc',
2468 'description': '',
2469 'channel': 'cole-dlp-test-acc',
2470 'tags': [],
2471 'view_count': int,
2472 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2473 'like_count': int,
2474 'playable_in_embed': True,
2475 'availability': 'unlisted',
2476 'thumbnail': 'https://i.ytimg.com/vi_webp/el3E4MbxRqQ/maxresdefault.webp',
2477 'age_limit': 0,
2478 'duration': 5,
2479 'uploader_id': 'UCiu-3thuViMebBjw_5nWYrA',
2480 'uploader_url': 'http://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2481 'live_status': 'not_live',
2482 'upload_date': '20220908',
2483 'categories': ['People & Blogs'],
2484 'uploader': 'cole-dlp-test-acc',
2485 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2486 },
2487 'params': {'skip_download': True}
2488 }, {
2489 # Extractor argument: prefer translated title+description
2490 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2491 'info_dict': {
2492 'id': 'gHKT4uU8Zng',
2493 'ext': 'mp4',
2494 'channel': 'cole-dlp-test-acc',
2495 'tags': [],
2496 'duration': 5,
2497 'live_status': 'not_live',
2498 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2499 'upload_date': '20220728',
2500 'uploader_id': 'UCiu-3thuViMebBjw_5nWYrA',
2501 'view_count': int,
2502 'categories': ['People & Blogs'],
2503 'thumbnail': 'https://i.ytimg.com/vi_webp/gHKT4uU8Zng/maxresdefault.webp',
2504 'title': 'dlp test video title translated (fr)',
2505 'availability': 'public',
2506 'uploader': 'cole-dlp-test-acc',
2507 'age_limit': 0,
2508 'description': 'dlp test video description translated (fr)',
2509 'playable_in_embed': True,
2510 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2511 'uploader_url': 'http://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2512 },
2513 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2514 'expected_warnings': [r'Preferring "fr" translated fields'],
2515 }, {
2516 'note': '6 channel audio',
2517 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2518 'only_matching': True,
2519 }, {
2520 'note': 'Multiple HLS formats with same itag',
2521 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2522 'info_dict': {
2523 'id': 'kX3nB4PpJko',
2524 'ext': 'mp4',
2525 'categories': ['Entertainment'],
2526 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2527 'uploader_url': 'http://www.youtube.com/user/MrBeast6000',
2528 'live_status': 'not_live',
2529 'duration': 937,
2530 'channel_follower_count': int,
2531 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2532 'title': 'Last To Take Hand Off Jet, Keeps It!',
2533 'channel': 'MrBeast',
2534 'playable_in_embed': True,
2535 'view_count': int,
2536 'upload_date': '20221112',
2537 'uploader': 'MrBeast',
2538 'uploader_id': 'MrBeast6000',
2539 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2540 'age_limit': 0,
2541 'availability': 'public',
2542 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2543 'like_count': int,
2544 'tags': [],
2545 },
2546 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2547 }
2548 ]
2549
2550 _WEBPAGE_TESTS = [
2551 # YouTube <object> embed
2552 {
2553 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2554 'md5': '873c81d308b979f0e23ee7e620b312a3',
2555 'info_dict': {
2556 'id': 'msN87y-iEx0',
2557 'ext': 'mp4',
2558 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2559 'upload_date': '20080526',
2560 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2561 'uploader': 'Christopher Sykes',
2562 'uploader_id': 'ChristopherJSykes',
2563 'age_limit': 0,
2564 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2565 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2566 'playable_in_embed': True,
2567 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2568 'like_count': int,
2569 'comment_count': int,
2570 'channel': 'Christopher Sykes',
2571 'live_status': 'not_live',
2572 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2573 'availability': 'public',
2574 'duration': 195,
2575 'view_count': int,
2576 'categories': ['Science & Technology'],
2577 'channel_follower_count': int,
2578 'uploader_url': 'http://www.youtube.com/user/ChristopherJSykes',
2579 },
2580 'params': {
2581 'skip_download': True,
2582 }
2583 },
2584 ]
2585
2586 @classmethod
2587 def suitable(cls, url):
2588 from ..utils import parse_qs
2589
2590 qs = parse_qs(url)
2591 if qs.get('list', [None])[0]:
2592 return False
2593 return super().suitable(url)
2594
2595 def __init__(self, *args, **kwargs):
2596 super().__init__(*args, **kwargs)
2597 self._code_cache = {}
2598 self._player_cache = {}
2599
2600 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2601 lock = threading.Lock()
2602 start_time = time.time()
2603 formats = [f for f in formats if f.get('is_from_start')]
2604
2605 def refetch_manifest(format_id, delay):
2606 nonlocal formats, start_time, is_live
2607 if time.time() <= start_time + delay:
2608 return
2609
2610 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2611 video_details = traverse_obj(
2612 prs, (..., 'videoDetails'), expected_type=dict, default=[])
2613 microformats = traverse_obj(
2614 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2615 expected_type=dict, default=[])
2616 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2617 is_live = live_status == 'is_live'
2618 start_time = time.time()
2619
2620 def mpd_feed(format_id, delay):
2621 """
2622 @returns (manifest_url, manifest_stream_number, is_live) or None
2623 """
2624 with lock:
2625 refetch_manifest(format_id, delay)
2626
2627 f = next((f for f in formats if f['format_id'] == format_id), None)
2628 if not f:
2629 if not is_live:
2630 self.to_screen(f'{video_id}: Video is no longer live')
2631 else:
2632 self.report_warning(
2633 f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}')
2634 return None
2635 return f['manifest_url'], f['manifest_stream_number'], is_live
2636
2637 for f in formats:
2638 f['is_live'] = is_live
2639 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2640 live_start_time, mpd_feed, not is_live and f.copy())
2641 if is_live:
2642 f['fragments'] = gen
2643 f['protocol'] = 'http_dash_segments_generator'
2644 else:
2645 f['fragments'] = LazyList(gen({}))
2646 del f['is_from_start']
2647
2648 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2649 FETCH_SPAN, MAX_DURATION = 5, 432000
2650
2651 mpd_url, stream_number, is_live = None, None, True
2652
2653 begin_index = 0
2654 download_start_time = ctx.get('start') or time.time()
2655
2656 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2657 if lack_early_segments:
2658 self.report_warning(bug_reports_message(
2659 'Starting download from the last 120 hours of the live stream since '
2660 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2661 lack_early_segments = True
2662
2663 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2664 fragments, fragment_base_url = None, None
2665
2666 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2667 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2668 # Obtain from MPD's maximum seq value
2669 old_mpd_url = mpd_url
2670 last_error = ctx.pop('last_error', None)
2671 expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
2672 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2673 or (mpd_url, stream_number, False))
2674 if not refresh_sequence:
2675 if expire_fast and not is_live:
2676 return False, last_seq
2677 elif old_mpd_url == mpd_url:
2678 return True, last_seq
2679 if manifestless_orig_fmt:
2680 fmt_info = manifestless_orig_fmt
2681 else:
2682 try:
2683 fmts, _ = self._extract_mpd_formats_and_subtitles(
2684 mpd_url, None, note=False, errnote=False, fatal=False)
2685 except ExtractorError:
2686 fmts = None
2687 if not fmts:
2688 no_fragment_score += 2
2689 return False, last_seq
2690 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2691 fragments = fmt_info['fragments']
2692 fragment_base_url = fmt_info['fragment_base_url']
2693 assert fragment_base_url
2694
2695 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2696 return True, _last_seq
2697
2698 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2699 while is_live:
2700 fetch_time = time.time()
2701 if no_fragment_score > 30:
2702 return
2703 if last_segment_url:
2704 # Obtain from "X-Head-Seqnum" header value from each segment
2705 try:
2706 urlh = self._request_webpage(
2707 last_segment_url, None, note=False, errnote=False, fatal=False)
2708 except ExtractorError:
2709 urlh = None
2710 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2711 if last_seq is None:
2712 no_fragment_score += 2
2713 last_segment_url = None
2714 continue
2715 else:
2716 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2717 no_fragment_score += 2
2718 if not should_continue:
2719 continue
2720
2721 if known_idx > last_seq:
2722 last_segment_url = None
2723 continue
2724
2725 last_seq += 1
2726
2727 if begin_index < 0 and known_idx < 0:
2728 # skip from the start when it's negative value
2729 known_idx = last_seq + begin_index
2730 if lack_early_segments:
2731 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2732 try:
2733 for idx in range(known_idx, last_seq):
2734 # do not update sequence here or you'll get skipped some part of it
2735 should_continue, _ = _extract_sequence_from_mpd(False, False)
2736 if not should_continue:
2737 known_idx = idx - 1
2738 raise ExtractorError('breaking out of outer loop')
2739 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2740 yield {
2741 'url': last_segment_url,
2742 'fragment_count': last_seq,
2743 }
2744 if known_idx == last_seq:
2745 no_fragment_score += 5
2746 else:
2747 no_fragment_score = 0
2748 known_idx = last_seq
2749 except ExtractorError:
2750 continue
2751
2752 if manifestless_orig_fmt:
2753 # Stop at the first iteration if running for post-live manifestless;
2754 # fragment count no longer increase since it starts
2755 break
2756
2757 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2758
2759 def _extract_player_url(self, *ytcfgs, webpage=None):
2760 player_url = traverse_obj(
2761 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2762 get_all=False, expected_type=str)
2763 if not player_url:
2764 return
2765 return urljoin('https://www.youtube.com', player_url)
2766
2767 def _download_player_url(self, video_id, fatal=False):
2768 res = self._download_webpage(
2769 'https://www.youtube.com/iframe_api',
2770 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2771 if res:
2772 player_version = self._search_regex(
2773 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2774 if player_version:
2775 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2776
2777 def _signature_cache_id(self, example_sig):
2778 """ Return a string representation of a signature """
2779 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2780
2781 @classmethod
2782 def _extract_player_info(cls, player_url):
2783 for player_re in cls._PLAYER_INFO_RE:
2784 id_m = re.search(player_re, player_url)
2785 if id_m:
2786 break
2787 else:
2788 raise ExtractorError('Cannot identify player %r' % player_url)
2789 return id_m.group('id')
2790
2791 def _load_player(self, video_id, player_url, fatal=True):
2792 player_id = self._extract_player_info(player_url)
2793 if player_id not in self._code_cache:
2794 code = self._download_webpage(
2795 player_url, video_id, fatal=fatal,
2796 note='Downloading player ' + player_id,
2797 errnote='Download of %s failed' % player_url)
2798 if code:
2799 self._code_cache[player_id] = code
2800 return self._code_cache.get(player_id)
2801
2802 def _extract_signature_function(self, video_id, player_url, example_sig):
2803 player_id = self._extract_player_info(player_url)
2804
2805 # Read from filesystem cache
2806 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2807 assert os.path.basename(func_id) == func_id
2808
2809 self.write_debug(f'Extracting signature function {func_id}')
2810 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2811
2812 if not cache_spec:
2813 code = self._load_player(video_id, player_url)
2814 if code:
2815 res = self._parse_sig_js(code)
2816 test_string = ''.join(map(chr, range(len(example_sig))))
2817 cache_spec = [ord(c) for c in res(test_string)]
2818 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2819
2820 return lambda s: ''.join(s[i] for i in cache_spec)
2821
2822 def _print_sig_code(self, func, example_sig):
2823 if not self.get_param('youtube_print_sig_code'):
2824 return
2825
2826 def gen_sig_code(idxs):
2827 def _genslice(start, end, step):
2828 starts = '' if start == 0 else str(start)
2829 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2830 steps = '' if step == 1 else (':%d' % step)
2831 return f's[{starts}{ends}{steps}]'
2832
2833 step = None
2834 # Quelch pyflakes warnings - start will be set when step is set
2835 start = '(Never used)'
2836 for i, prev in zip(idxs[1:], idxs[:-1]):
2837 if step is not None:
2838 if i - prev == step:
2839 continue
2840 yield _genslice(start, prev, step)
2841 step = None
2842 continue
2843 if i - prev in [-1, 1]:
2844 step = i - prev
2845 start = prev
2846 continue
2847 else:
2848 yield 's[%d]' % prev
2849 if step is None:
2850 yield 's[%d]' % i
2851 else:
2852 yield _genslice(start, i, step)
2853
2854 test_string = ''.join(map(chr, range(len(example_sig))))
2855 cache_res = func(test_string)
2856 cache_spec = [ord(c) for c in cache_res]
2857 expr_code = ' + '.join(gen_sig_code(cache_spec))
2858 signature_id_tuple = '(%s)' % (
2859 ', '.join(str(len(p)) for p in example_sig.split('.')))
2860 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2861 ' return %s\n') % (signature_id_tuple, expr_code)
2862 self.to_screen('Extracted signature function:\n' + code)
2863
2864 def _parse_sig_js(self, jscode):
2865 funcname = self._search_regex(
2866 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2867 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2868 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2869 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2870 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}\(a,\d+\)',
2871 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\(\s*""\s*\)',
2872 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
2873 # Obsolete patterns
2874 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2875 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
2876 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2877 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2878 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2879 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2880 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2881 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
2882 jscode, 'Initial JS player signature function name', group='sig')
2883
2884 jsi = JSInterpreter(jscode)
2885 initial_function = jsi.extract_function(funcname)
2886 return lambda s: initial_function([s])
2887
2888 def _cached(self, func, *cache_id):
2889 def inner(*args, **kwargs):
2890 if cache_id not in self._player_cache:
2891 try:
2892 self._player_cache[cache_id] = func(*args, **kwargs)
2893 except ExtractorError as e:
2894 self._player_cache[cache_id] = e
2895 except Exception as e:
2896 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
2897
2898 ret = self._player_cache[cache_id]
2899 if isinstance(ret, Exception):
2900 raise ret
2901 return ret
2902 return inner
2903
2904 def _decrypt_signature(self, s, video_id, player_url):
2905 """Turn the encrypted s field into a working signature"""
2906 extract_sig = self._cached(
2907 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
2908 func = extract_sig(video_id, player_url, s)
2909 self._print_sig_code(func, s)
2910 return func(s)
2911
2912 def _decrypt_nsig(self, s, video_id, player_url):
2913 """Turn the encrypted n field into a working signature"""
2914 if player_url is None:
2915 raise ExtractorError('Cannot decrypt nsig without player_url')
2916 player_url = urljoin('https://www.youtube.com', player_url)
2917
2918 try:
2919 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
2920 except ExtractorError as e:
2921 raise ExtractorError('Unable to extract nsig function code', cause=e)
2922 if self.get_param('youtube_print_sig_code'):
2923 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
2924
2925 try:
2926 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
2927 ret = extract_nsig(jsi, func_code)(s)
2928 except JSInterpreter.Exception as e:
2929 try:
2930 jsi = PhantomJSwrapper(self, timeout=5000)
2931 except ExtractorError:
2932 raise e
2933 self.report_warning(
2934 f'Native nsig extraction failed: Trying with PhantomJS\n'
2935 f' n = {s} ; player = {player_url}', video_id)
2936 self.write_debug(e, only_once=True)
2937
2938 args, func_body = func_code
2939 ret = jsi.execute(
2940 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
2941 video_id=video_id, note='Executing signature code').strip()
2942
2943 self.write_debug(f'Decrypted nsig {s} => {ret}')
2944 return ret
2945
2946 def _extract_n_function_name(self, jscode):
2947 funcname, idx = self._search_regex(
2948 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
2949 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
2950 if not idx:
2951 return funcname
2952
2953 return json.loads(js_to_json(self._search_regex(
2954 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\]);', jscode,
2955 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
2956
2957 def _extract_n_function_code(self, video_id, player_url):
2958 player_id = self._extract_player_info(player_url)
2959 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
2960 jscode = func_code or self._load_player(video_id, player_url)
2961 jsi = JSInterpreter(jscode)
2962
2963 if func_code:
2964 return jsi, player_id, func_code
2965
2966 func_name = self._extract_n_function_name(jscode)
2967
2968 # For redundancy
2969 func_code = self._search_regex(
2970 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
2971 # NB: The end of the regex is intentionally kept strict
2972 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
2973 jscode, 'nsig function', group=('var', 'code'), default=None)
2974 if func_code:
2975 func_code = ([func_code[0]], func_code[1])
2976 else:
2977 self.write_debug('Extracting nsig function with jsinterp')
2978 func_code = jsi.extract_function_code(func_name)
2979
2980 self.cache.store('youtube-nsig', player_id, func_code)
2981 return jsi, player_id, func_code
2982
2983 def _extract_n_function_from_code(self, jsi, func_code):
2984 func = jsi.extract_function_from_code(*func_code)
2985
2986 def extract_nsig(s):
2987 try:
2988 ret = func([s])
2989 except JSInterpreter.Exception:
2990 raise
2991 except Exception as e:
2992 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
2993
2994 if ret.startswith('enhanced_except_'):
2995 raise JSInterpreter.Exception('Signature function returned an exception')
2996 return ret
2997
2998 return extract_nsig
2999
3000 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3001 """
3002 Extract signatureTimestamp (sts)
3003 Required to tell API what sig/player version is in use.
3004 """
3005 sts = None
3006 if isinstance(ytcfg, dict):
3007 sts = int_or_none(ytcfg.get('STS'))
3008
3009 if not sts:
3010 # Attempt to extract from player
3011 if player_url is None:
3012 error_msg = 'Cannot extract signature timestamp without player_url.'
3013 if fatal:
3014 raise ExtractorError(error_msg)
3015 self.report_warning(error_msg)
3016 return
3017 code = self._load_player(video_id, player_url, fatal=fatal)
3018 if code:
3019 sts = int_or_none(self._search_regex(
3020 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3021 'JS player signature timestamp', group='sts', fatal=fatal))
3022 return sts
3023
3024 def _mark_watched(self, video_id, player_responses):
3025 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3026 label = 'fully ' if is_full else ''
3027 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3028 expected_type=url_or_none)
3029 if not url:
3030 self.report_warning(f'Unable to mark {label}watched')
3031 return
3032 parsed_url = urllib.parse.urlparse(url)
3033 qs = urllib.parse.parse_qs(parsed_url.query)
3034
3035 # cpn generation algorithm is reverse engineered from base.js.
3036 # In fact it works even with dummy cpn.
3037 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3038 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3039
3040 # # more consistent results setting it to right before the end
3041 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3042
3043 qs.update({
3044 'ver': ['2'],
3045 'cpn': [cpn],
3046 'cmt': video_length,
3047 'el': 'detailpage', # otherwise defaults to "shorts"
3048 })
3049
3050 if is_full:
3051 # these seem to mark watchtime "history" in the real world
3052 # they're required, so send in a single value
3053 qs.update({
3054 'st': 0,
3055 'et': video_length,
3056 })
3057
3058 url = urllib.parse.urlunparse(
3059 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3060
3061 self._download_webpage(
3062 url, video_id, f'Marking {label}watched',
3063 'Unable to mark watched', fatal=False)
3064
3065 @classmethod
3066 def _extract_from_webpage(cls, url, webpage):
3067 # Invidious Instances
3068 # https://github.com/yt-dlp/yt-dlp/issues/195
3069 # https://github.com/iv-org/invidious/pull/1730
3070 mobj = re.search(
3071 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3072 webpage)
3073 if mobj:
3074 yield cls.url_result(mobj.group('url'), cls)
3075 raise cls.StopExtraction()
3076
3077 yield from super()._extract_from_webpage(url, webpage)
3078
3079 # lazyYT YouTube embed
3080 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3081 yield cls.url_result(unescapeHTML(id_), cls, id_)
3082
3083 # Wordpress "YouTube Video Importer" plugin
3084 for m in re.findall(r'''(?x)<div[^>]+
3085 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3086 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3087 yield cls.url_result(m[-1], cls, m[-1])
3088
3089 @classmethod
3090 def extract_id(cls, url):
3091 video_id = cls.get_temp_id(url)
3092 if not video_id:
3093 raise ExtractorError(f'Invalid URL: {url}')
3094 return video_id
3095
3096 def _extract_chapters_from_json(self, data, duration):
3097 chapter_list = traverse_obj(
3098 data, (
3099 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3100 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3101 ), expected_type=list)
3102
3103 return self._extract_chapters(
3104 chapter_list,
3105 chapter_time=lambda chapter: float_or_none(
3106 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3107 chapter_title=lambda chapter: traverse_obj(
3108 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3109 duration=duration)
3110
3111 def _extract_chapters_from_engagement_panel(self, data, duration):
3112 content_list = traverse_obj(
3113 data,
3114 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3115 expected_type=list, default=[])
3116 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3117 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3118
3119 return next(filter(None, (
3120 self._extract_chapters(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3121 chapter_time, chapter_title, duration)
3122 for contents in content_list)), [])
3123
3124 def _extract_chapters_from_description(self, description, duration):
3125 duration_re = r'(?:\d+:)?\d{1,2}:\d{2}'
3126 sep_re = r'(?m)^\s*(%s)\b\W*\s(%s)\s*$'
3127 return self._extract_chapters(
3128 re.findall(sep_re % (duration_re, r'.+?'), description or ''),
3129 chapter_time=lambda x: parse_duration(x[0]), chapter_title=lambda x: x[1],
3130 duration=duration, strict=False) or self._extract_chapters(
3131 re.findall(sep_re % (r'.+?', duration_re), description or ''),
3132 chapter_time=lambda x: parse_duration(x[1]), chapter_title=lambda x: x[0],
3133 duration=duration, strict=False)
3134
3135 def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration, strict=True):
3136 if not duration:
3137 return
3138 chapter_list = [{
3139 'start_time': chapter_time(chapter),
3140 'title': chapter_title(chapter),
3141 } for chapter in chapter_list or []]
3142 if not strict:
3143 chapter_list.sort(key=lambda c: c['start_time'] or 0)
3144
3145 chapters = [{'start_time': 0}]
3146 for idx, chapter in enumerate(chapter_list):
3147 if chapter['start_time'] is None:
3148 self.report_warning(f'Incomplete chapter {idx}')
3149 elif chapters[-1]['start_time'] <= chapter['start_time'] <= duration:
3150 chapters.append(chapter)
3151 elif chapter not in chapters:
3152 self.report_warning(
3153 f'Invalid start time ({chapter["start_time"]} < {chapters[-1]["start_time"]}) for chapter "{chapter["title"]}"')
3154 return chapters[1:]
3155
3156 def _extract_comment(self, comment_renderer, parent=None):
3157 comment_id = comment_renderer.get('commentId')
3158 if not comment_id:
3159 return
3160
3161 text = self._get_text(comment_renderer, 'contentText')
3162
3163 # Timestamp is an estimate calculated from the current time and time_text
3164 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3165 timestamp = self._parse_time_text(time_text)
3166
3167 author = self._get_text(comment_renderer, 'authorText')
3168 author_id = try_get(comment_renderer,
3169 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], str)
3170
3171 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
3172 lambda x: x['likeCount']), str)) or 0
3173 author_thumbnail = try_get(comment_renderer,
3174 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], str)
3175
3176 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
3177 is_favorited = 'creatorHeart' in (try_get(
3178 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
3179 return {
3180 'id': comment_id,
3181 'text': text,
3182 'timestamp': timestamp,
3183 'time_text': time_text,
3184 'like_count': votes,
3185 'is_favorited': is_favorited,
3186 'author': author,
3187 'author_id': author_id,
3188 'author_thumbnail': author_thumbnail,
3189 'author_is_uploader': author_is_uploader,
3190 'parent': parent or 'root'
3191 }
3192
3193 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3194
3195 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3196
3197 def extract_header(contents):
3198 _continuation = None
3199 for content in contents:
3200 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3201 expected_comment_count = self._get_count(
3202 comments_header_renderer, 'countText', 'commentsCount')
3203
3204 if expected_comment_count:
3205 tracker['est_total'] = expected_comment_count
3206 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3207 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3208
3209 sort_menu_item = try_get(
3210 comments_header_renderer,
3211 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3212 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3213
3214 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3215 if not _continuation:
3216 continue
3217
3218 sort_text = str_or_none(sort_menu_item.get('title'))
3219 if not sort_text:
3220 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3221 self.to_screen('Sorting comments by %s' % sort_text.lower())
3222 break
3223 return _continuation
3224
3225 def extract_thread(contents):
3226 if not parent:
3227 tracker['current_page_thread'] = 0
3228 for content in contents:
3229 if not parent and tracker['total_parent_comments'] >= max_parents:
3230 yield
3231 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3232 comment_renderer = get_first(
3233 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3234 expected_type=dict, default={})
3235
3236 comment = self._extract_comment(comment_renderer, parent)
3237 if not comment:
3238 continue
3239
3240 tracker['running_total'] += 1
3241 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3242 yield comment
3243
3244 # Attempt to get the replies
3245 comment_replies_renderer = try_get(
3246 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3247
3248 if comment_replies_renderer:
3249 tracker['current_page_thread'] += 1
3250 comment_entries_iter = self._comment_entries(
3251 comment_replies_renderer, ytcfg, video_id,
3252 parent=comment.get('id'), tracker=tracker)
3253 yield from itertools.islice(comment_entries_iter, min(
3254 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3255
3256 # Keeps track of counts across recursive calls
3257 if not tracker:
3258 tracker = dict(
3259 running_total=0,
3260 est_total=0,
3261 current_page_thread=0,
3262 total_parent_comments=0,
3263 total_reply_comments=0)
3264
3265 # TODO: Deprecated
3266 # YouTube comments have a max depth of 2
3267 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3268 if max_depth:
3269 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3270 'Set max replies in the max-comments extractor argument instead')
3271 if max_depth == 1 and parent:
3272 return
3273
3274 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3275 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3276
3277 continuation = self._extract_continuation(root_continuation_data)
3278
3279 response = None
3280 is_forced_continuation = False
3281 is_first_continuation = parent is None
3282 if is_first_continuation and not continuation:
3283 # Sometimes you can get comments by generating the continuation yourself,
3284 # even if YouTube initially reports them being disabled - e.g. stories comments.
3285 # Note: if the comment section is actually disabled, YouTube may return a response with
3286 # required check_get_keys missing. So we will disable that check initially in this case.
3287 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3288 is_forced_continuation = True
3289
3290 for page_num in itertools.count(0):
3291 if not continuation:
3292 break
3293 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3294 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
3295 if page_num == 0:
3296 if is_first_continuation:
3297 note_prefix = 'Downloading comment section API JSON'
3298 else:
3299 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3300 tracker['current_page_thread'], comment_prog_str)
3301 else:
3302 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3303 ' ' if parent else '', ' replies' if parent else '',
3304 page_num, comment_prog_str)
3305 try:
3306 response = self._extract_response(
3307 item_id=None, query=continuation,
3308 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3309 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
3310 except ExtractorError as e:
3311 # Ignore incomplete data error for replies if retries didn't work.
3312 # This is to allow any other parent comments and comment threads to be downloaded.
3313 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3314 if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True:
3315 self.report_warning(
3316 'Received incomplete data for a comment reply thread and retrying did not help. '
3317 'Ignoring to let other comments be downloaded.')
3318 else:
3319 raise
3320 is_forced_continuation = False
3321 continuation_contents = traverse_obj(
3322 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
3323
3324 continuation = None
3325 for continuation_section in continuation_contents:
3326 continuation_items = traverse_obj(
3327 continuation_section,
3328 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
3329 get_all=False, expected_type=list) or []
3330 if is_first_continuation:
3331 continuation = extract_header(continuation_items)
3332 is_first_continuation = False
3333 if continuation:
3334 break
3335 continue
3336
3337 for entry in extract_thread(continuation_items):
3338 if not entry:
3339 return
3340 yield entry
3341 continuation = self._extract_continuation({'contents': continuation_items})
3342 if continuation:
3343 break
3344
3345 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3346 if message and not parent and tracker['running_total'] == 0:
3347 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3348 raise self.CommentsDisabled
3349
3350 @staticmethod
3351 def _generate_comment_continuation(video_id):
3352 """
3353 Generates initial comment section continuation token from given video id
3354 """
3355 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3356 return base64.b64encode(token.encode()).decode()
3357
3358 def _get_comments(self, ytcfg, video_id, contents, webpage):
3359 """Entry for comment extraction"""
3360 def _real_comment_extract(contents):
3361 renderer = next((
3362 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3363 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3364 yield from self._comment_entries(renderer, ytcfg, video_id)
3365
3366 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3367 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3368
3369 @staticmethod
3370 def _get_checkok_params():
3371 return {'contentCheckOk': True, 'racyCheckOk': True}
3372
3373 @classmethod
3374 def _generate_player_context(cls, sts=None):
3375 context = {
3376 'html5Preference': 'HTML5_PREF_WANTS',
3377 }
3378 if sts is not None:
3379 context['signatureTimestamp'] = sts
3380 return {
3381 'playbackContext': {
3382 'contentPlaybackContext': context
3383 },
3384 **cls._get_checkok_params()
3385 }
3386
3387 @staticmethod
3388 def _is_agegated(player_response):
3389 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3390 return True
3391
3392 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[])
3393 AGE_GATE_REASONS = (
3394 'confirm your age', 'age-restricted', 'inappropriate', # reason
3395 'age_verification_required', 'age_check_required', # status
3396 )
3397 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3398
3399 @staticmethod
3400 def _is_unplayable(player_response):
3401 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3402
3403 _STORY_PLAYER_PARAMS = '8AEB'
3404
3405 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3406
3407 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3408 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3409 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3410 headers = self.generate_api_headers(
3411 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3412
3413 yt_query = {
3414 'videoId': video_id,
3415 }
3416 if smuggled_data.get('is_story') or _split_innertube_client(client)[0] == 'android':
3417 yt_query['params'] = self._STORY_PLAYER_PARAMS
3418
3419 yt_query.update(self._generate_player_context(sts))
3420 return self._extract_response(
3421 item_id=video_id, ep='player', query=yt_query,
3422 ytcfg=player_ytcfg, headers=headers, fatal=True,
3423 default_client=client,
3424 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3425 ) or None
3426
3427 def _get_requested_clients(self, url, smuggled_data):
3428 requested_clients = []
3429 default = ['android', 'web']
3430 allowed_clients = sorted(
3431 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3432 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3433 for client in self._configuration_arg('player_client'):
3434 if client in allowed_clients:
3435 requested_clients.append(client)
3436 elif client == 'default':
3437 requested_clients.extend(default)
3438 elif client == 'all':
3439 requested_clients.extend(allowed_clients)
3440 else:
3441 self.report_warning(f'Skipping unsupported client {client}')
3442 if not requested_clients:
3443 requested_clients = default
3444
3445 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3446 requested_clients.extend(
3447 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3448
3449 return orderedSet(requested_clients)
3450
3451 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3452 initial_pr = None
3453 if webpage:
3454 initial_pr = self._search_json(
3455 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3456
3457 all_clients = set(clients)
3458 clients = clients[::-1]
3459 prs = []
3460
3461 def append_client(*client_names):
3462 """ Append the first client name that exists but not already used """
3463 for client_name in client_names:
3464 actual_client = _split_innertube_client(client_name)[0]
3465 if actual_client in INNERTUBE_CLIENTS:
3466 if actual_client not in all_clients:
3467 clients.append(client_name)
3468 all_clients.add(actual_client)
3469 return
3470
3471 # Android player_response does not have microFormats which are needed for
3472 # extraction of some data. So we return the initial_pr with formats
3473 # stripped out even if not requested by the user
3474 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3475 if initial_pr:
3476 pr = dict(initial_pr)
3477 pr['streamingData'] = None
3478 prs.append(pr)
3479
3480 last_error = None
3481 tried_iframe_fallback = False
3482 player_url = None
3483 while clients:
3484 client, base_client, variant = _split_innertube_client(clients.pop())
3485 player_ytcfg = master_ytcfg if client == 'web' else {}
3486 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3487 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3488
3489 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3490 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3491 if 'js' in self._configuration_arg('player_skip'):
3492 require_js_player = False
3493 player_url = None
3494
3495 if not player_url and not tried_iframe_fallback and require_js_player:
3496 player_url = self._download_player_url(video_id)
3497 tried_iframe_fallback = True
3498
3499 try:
3500 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3501 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3502 except ExtractorError as e:
3503 if last_error:
3504 self.report_warning(last_error)
3505 last_error = e
3506 continue
3507
3508 if pr:
3509 # YouTube may return a different video player response than expected.
3510 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3511 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3512 if pr_video_id and pr_video_id != video_id:
3513 self.report_warning(
3514 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3515 else:
3516 prs.append(pr)
3517
3518 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3519 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3520 append_client(f'{base_client}_creator')
3521 elif self._is_agegated(pr):
3522 if variant == 'tv_embedded':
3523 append_client(f'{base_client}_embedded')
3524 elif not variant:
3525 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3526
3527 if last_error:
3528 if not len(prs):
3529 raise last_error
3530 self.report_warning(last_error)
3531 return prs, player_url
3532
3533 def _needs_live_processing(self, live_status, duration):
3534 if (live_status == 'is_live' and self.get_param('live_from_start')
3535 or live_status == 'post_live' and (duration or 0) > 4 * 3600):
3536 return live_status
3537
3538 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3539 itags, stream_ids = collections.defaultdict(set), []
3540 itag_qualities, res_qualities = {}, {0: None}
3541 q = qualities([
3542 # Normally tiny is the smallest video-only formats. But
3543 # audio-only formats with unknown quality may get tagged as tiny
3544 'tiny',
3545 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3546 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3547 ])
3548 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[])
3549
3550 for fmt in streaming_formats:
3551 if fmt.get('targetDurationSec'):
3552 continue
3553
3554 itag = str_or_none(fmt.get('itag'))
3555 audio_track = fmt.get('audioTrack') or {}
3556 stream_id = '%s.%s' % (itag or '', audio_track.get('id', ''))
3557 if stream_id in stream_ids:
3558 continue
3559
3560 quality = fmt.get('quality')
3561 height = int_or_none(fmt.get('height'))
3562 if quality == 'tiny' or not quality:
3563 quality = fmt.get('audioQuality', '').lower() or quality
3564 # The 3gp format (17) in android client has a quality of "small",
3565 # but is actually worse than other formats
3566 if itag == '17':
3567 quality = 'tiny'
3568 if quality:
3569 if itag:
3570 itag_qualities[itag] = quality
3571 if height:
3572 res_qualities[height] = quality
3573 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3574 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3575 # number of fragment that would subsequently requested with (`&sq=N`)
3576 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3577 continue
3578
3579 fmt_url = fmt.get('url')
3580 if not fmt_url:
3581 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3582 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3583 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3584 if not all((sc, fmt_url, player_url, encrypted_sig)):
3585 continue
3586 try:
3587 fmt_url += '&%s=%s' % (
3588 traverse_obj(sc, ('sp', -1)) or 'signature',
3589 self._decrypt_signature(encrypted_sig, video_id, player_url)
3590 )
3591 except ExtractorError as e:
3592 self.report_warning('Signature extraction failed: Some formats may be missing',
3593 video_id=video_id, only_once=True)
3594 self.write_debug(e, only_once=True)
3595 continue
3596
3597 query = parse_qs(fmt_url)
3598 throttled = False
3599 if query.get('n'):
3600 try:
3601 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3602 fmt_url = update_url_query(fmt_url, {
3603 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3604 })
3605 except ExtractorError as e:
3606 phantomjs_hint = ''
3607 if isinstance(e, JSInterpreter.Exception):
3608 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3609 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3610 if player_url:
3611 self.report_warning(
3612 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3613 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3614 self.write_debug(e, only_once=True)
3615 else:
3616 self.report_warning(
3617 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3618 video_id=video_id, only_once=True)
3619 throttled = True
3620
3621 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3622 language_preference = (
3623 10 if audio_track.get('audioIsDefault') and 10
3624 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3625 else -1)
3626 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3627 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3628 # Make sure to avoid false positives with small duration differences.
3629 # E.g. __2ABJjxzNo, ySuUZEjARPY
3630 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3631 if is_damaged:
3632 self.report_warning(
3633 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3634 dct = {
3635 'asr': int_or_none(fmt.get('audioSampleRate')),
3636 'filesize': int_or_none(fmt.get('contentLength')),
3637 'format_id': itag,
3638 'format_note': join_nonempty(
3639 '%s%s' % (audio_track.get('displayName') or '',
3640 ' (default)' if language_preference > 0 else ''),
3641 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3642 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3643 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3644 throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '),
3645 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3646 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3647 'fps': int_or_none(fmt.get('fps')) or None,
3648 'audio_channels': fmt.get('audioChannels'),
3649 'height': height,
3650 'quality': q(quality),
3651 'has_drm': bool(fmt.get('drmFamilies')),
3652 'tbr': tbr,
3653 'url': fmt_url,
3654 'width': int_or_none(fmt.get('width')),
3655 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3656 'desc' if language_preference < -1 else ''),
3657 'language_preference': language_preference,
3658 # Strictly de-prioritize damaged and 3gp formats
3659 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3660 }
3661 mime_mobj = re.match(
3662 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3663 if mime_mobj:
3664 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3665 dct.update(parse_codecs(mime_mobj.group(2)))
3666 no_audio = dct.get('acodec') == 'none'
3667 no_video = dct.get('vcodec') == 'none'
3668 if no_audio:
3669 dct['vbr'] = tbr
3670 if no_video:
3671 dct['abr'] = tbr
3672 if no_audio or no_video:
3673 dct['downloader_options'] = {
3674 # Youtube throttles chunks >~10M
3675 'http_chunk_size': 10485760,
3676 }
3677 if dct.get('ext'):
3678 dct['container'] = dct['ext'] + '_dash'
3679
3680 if itag:
3681 itags[itag].add(('https', dct.get('language')))
3682 stream_ids.append(stream_id)
3683 yield dct
3684
3685 needs_live_processing = self._needs_live_processing(live_status, duration)
3686 skip_bad_formats = not self._configuration_arg('include_incomplete_formats')
3687
3688 skip_manifests = set(self._configuration_arg('skip'))
3689 if (not self.get_param('youtube_include_hls_manifest', True)
3690 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3691 or needs_live_processing and skip_bad_formats):
3692 skip_manifests.add('hls')
3693
3694 if not self.get_param('youtube_include_dash_manifest', True):
3695 skip_manifests.add('dash')
3696 if self._configuration_arg('include_live_dash'):
3697 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3698 'Use include_incomplete_formats extractor argument instead')
3699 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3700 skip_manifests.add('dash')
3701
3702 def process_manifest_format(f, proto, itag):
3703 key = (proto, f.get('language'))
3704 if key in itags[itag]:
3705 return False
3706 itags[itag].add(key)
3707
3708 if any(p != proto for p, _ in itags[itag]):
3709 f['format_id'] = f'{itag}-{proto}'
3710 elif itag:
3711 f['format_id'] = itag
3712
3713 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3714 if f['quality'] == -1 and f.get('height'):
3715 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3716 return True
3717
3718 subtitles = {}
3719 for sd in streaming_data:
3720 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3721 if hls_manifest_url:
3722 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3723 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3724 subtitles = self._merge_subtitles(subs, subtitles)
3725 for f in fmts:
3726 if process_manifest_format(f, 'hls', self._search_regex(
3727 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3728 yield f
3729
3730 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3731 if dash_manifest_url:
3732 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3733 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3734 for f in formats:
3735 if process_manifest_format(f, 'dash', f['format_id']):
3736 f['filesize'] = int_or_none(self._search_regex(
3737 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3738 if needs_live_processing:
3739 f['is_from_start'] = True
3740
3741 yield f
3742 yield subtitles
3743
3744 def _extract_storyboard(self, player_responses, duration):
3745 spec = get_first(
3746 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3747 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3748 if not base_url:
3749 return
3750 L = len(spec) - 1
3751 for i, args in enumerate(spec):
3752 args = args.split('#')
3753 counts = list(map(int_or_none, args[:5]))
3754 if len(args) != 8 or not all(counts):
3755 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3756 continue
3757 width, height, frame_count, cols, rows = counts
3758 N, sigh = args[6:]
3759
3760 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3761 fragment_count = frame_count / (cols * rows)
3762 fragment_duration = duration / fragment_count
3763 yield {
3764 'format_id': f'sb{i}',
3765 'format_note': 'storyboard',
3766 'ext': 'mhtml',
3767 'protocol': 'mhtml',
3768 'acodec': 'none',
3769 'vcodec': 'none',
3770 'url': url,
3771 'width': width,
3772 'height': height,
3773 'fps': frame_count / duration,
3774 'rows': rows,
3775 'columns': cols,
3776 'fragments': [{
3777 'url': url.replace('$M', str(j)),
3778 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3779 } for j in range(math.ceil(fragment_count))],
3780 }
3781
3782 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3783 webpage = None
3784 if 'webpage' not in self._configuration_arg('player_skip'):
3785 query = {'bpctr': '9999999999', 'has_verified': '1'}
3786 if smuggled_data.get('is_story'):
3787 query['pp'] = self._STORY_PLAYER_PARAMS
3788 webpage = self._download_webpage(
3789 webpage_url, video_id, fatal=False, query=query)
3790
3791 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3792
3793 player_responses, player_url = self._extract_player_responses(
3794 self._get_requested_clients(url, smuggled_data),
3795 video_id, webpage, master_ytcfg, smuggled_data)
3796
3797 return webpage, master_ytcfg, player_responses, player_url
3798
3799 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3800 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3801 is_live = get_first(video_details, 'isLive')
3802 if is_live is None:
3803 is_live = get_first(live_broadcast_details, 'isLiveNow')
3804 live_content = get_first(video_details, 'isLiveContent')
3805 is_upcoming = get_first(video_details, 'isUpcoming')
3806 post_live = get_first(video_details, 'isPostLiveDvr')
3807 live_status = ('post_live' if post_live
3808 else 'is_live' if is_live
3809 else 'is_upcoming' if is_upcoming
3810 else 'was_live' if live_content
3811 else 'not_live' if False in (is_live, live_content)
3812 else None)
3813 streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[])
3814 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
3815
3816 return live_broadcast_details, live_status, streaming_data, formats, subtitles
3817
3818 def _real_extract(self, url):
3819 url, smuggled_data = unsmuggle_url(url, {})
3820 video_id = self._match_id(url)
3821
3822 base_url = self.http_scheme() + '//www.youtube.com/'
3823 webpage_url = base_url + 'watch?v=' + video_id
3824
3825 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3826
3827 playability_statuses = traverse_obj(
3828 player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[])
3829
3830 trailer_video_id = get_first(
3831 playability_statuses,
3832 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3833 expected_type=str)
3834 if trailer_video_id:
3835 return self.url_result(
3836 trailer_video_id, self.ie_key(), trailer_video_id)
3837
3838 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3839 if webpage else (lambda x: None))
3840
3841 video_details = traverse_obj(
3842 player_responses, (..., 'videoDetails'), expected_type=dict, default=[])
3843 microformats = traverse_obj(
3844 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3845 expected_type=dict, default=[])
3846
3847 translated_title = self._get_text(microformats, (..., 'title'))
3848 video_title = (self._preferred_lang and translated_title
3849 or get_first(video_details, 'title') # primary
3850 or translated_title
3851 or search_meta(['og:title', 'twitter:title', 'title']))
3852 translated_description = self._get_text(microformats, (..., 'description'))
3853 original_description = get_first(video_details, 'shortDescription')
3854 video_description = (
3855 self._preferred_lang and translated_description
3856 # If original description is blank, it will be an empty string.
3857 # Do not prefer translated description in this case.
3858 or original_description if original_description is not None else translated_description)
3859
3860 multifeed_metadata_list = get_first(
3861 player_responses,
3862 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
3863 expected_type=str)
3864 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
3865 if self.get_param('noplaylist'):
3866 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3867 else:
3868 entries = []
3869 feed_ids = []
3870 for feed in multifeed_metadata_list.split(','):
3871 # Unquote should take place before split on comma (,) since textual
3872 # fields may contain comma as well (see
3873 # https://github.com/ytdl-org/youtube-dl/issues/8536)
3874 feed_data = urllib.parse.parse_qs(
3875 urllib.parse.unquote_plus(feed))
3876
3877 def feed_entry(name):
3878 return try_get(
3879 feed_data, lambda x: x[name][0], str)
3880
3881 feed_id = feed_entry('id')
3882 if not feed_id:
3883 continue
3884 feed_title = feed_entry('title')
3885 title = video_title
3886 if feed_title:
3887 title += ' (%s)' % feed_title
3888 entries.append({
3889 '_type': 'url_transparent',
3890 'ie_key': 'Youtube',
3891 'url': smuggle_url(
3892 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
3893 {'force_singlefeed': True}),
3894 'title': title,
3895 })
3896 feed_ids.append(feed_id)
3897 self.to_screen(
3898 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
3899 % (', '.join(feed_ids), video_id))
3900 return self.playlist_result(
3901 entries, video_id, video_title, video_description)
3902
3903 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
3904 or int_or_none(get_first(microformats, 'lengthSeconds'))
3905 or parse_duration(search_meta('duration')) or None)
3906
3907 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
3908 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
3909 if live_status == 'post_live':
3910 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
3911
3912 if not formats:
3913 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
3914 self.report_drm(video_id)
3915 pemr = get_first(
3916 playability_statuses,
3917 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
3918 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
3919 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
3920 if subreason:
3921 if subreason == 'The uploader has not made this video available in your country.':
3922 countries = get_first(microformats, 'availableCountries')
3923 if not countries:
3924 regions_allowed = search_meta('regionsAllowed')
3925 countries = regions_allowed.split(',') if regions_allowed else None
3926 self.raise_geo_restricted(subreason, countries, metadata_available=True)
3927 reason += f'. {subreason}'
3928 if reason:
3929 self.raise_no_formats(reason, expected=True)
3930
3931 keywords = get_first(video_details, 'keywords', expected_type=list) or []
3932 if not keywords and webpage:
3933 keywords = [
3934 unescapeHTML(m.group('content'))
3935 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
3936 for keyword in keywords:
3937 if keyword.startswith('yt:stretch='):
3938 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
3939 if mobj:
3940 # NB: float is intentional for forcing float division
3941 w, h = (float(v) for v in mobj.groups())
3942 if w > 0 and h > 0:
3943 ratio = w / h
3944 for f in formats:
3945 if f.get('vcodec') != 'none':
3946 f['stretched_ratio'] = ratio
3947 break
3948 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
3949 thumbnail_url = search_meta(['og:image', 'twitter:image'])
3950 if thumbnail_url:
3951 thumbnails.append({
3952 'url': thumbnail_url,
3953 })
3954 original_thumbnails = thumbnails.copy()
3955
3956 # The best resolution thumbnails sometimes does not appear in the webpage
3957 # See: https://github.com/yt-dlp/yt-dlp/issues/340
3958 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
3959 thumbnail_names = [
3960 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
3961 # in resolution, these are not the custom thumbnail. So de-prioritize them
3962 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
3963 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
3964 ]
3965 n_thumbnail_names = len(thumbnail_names)
3966 thumbnails.extend({
3967 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
3968 video_id=video_id, name=name, ext=ext,
3969 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
3970 } for name in thumbnail_names for ext in ('webp', 'jpg'))
3971 for thumb in thumbnails:
3972 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
3973 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
3974 self._remove_duplicate_formats(thumbnails)
3975 self._downloader._sort_thumbnails(original_thumbnails)
3976
3977 category = get_first(microformats, 'category') or search_meta('genre')
3978 channel_id = str_or_none(
3979 get_first(video_details, 'channelId')
3980 or get_first(microformats, 'externalChannelId')
3981 or search_meta('channelId'))
3982 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
3983
3984 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
3985 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
3986 if not duration and live_end_time and live_start_time:
3987 duration = live_end_time - live_start_time
3988
3989 needs_live_processing = self._needs_live_processing(live_status, duration)
3990
3991 def is_bad_format(fmt):
3992 if needs_live_processing and not fmt.get('is_from_start'):
3993 return True
3994 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
3995 and fmt.get('protocol') == 'http_dash_segments'):
3996 return True
3997
3998 for fmt in filter(is_bad_format, formats):
3999 fmt['preference'] = (fmt.get('preference') or -1) - 10
4000 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 4 hours)', delim=' ')
4001
4002 if needs_live_processing:
4003 self._prepare_live_from_start_formats(
4004 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4005
4006 formats.extend(self._extract_storyboard(player_responses, duration))
4007
4008 info = {
4009 'id': video_id,
4010 'title': video_title,
4011 'formats': formats,
4012 'thumbnails': thumbnails,
4013 # The best thumbnail that we are sure exists. Prevents unnecessary
4014 # URL checking if user don't care about getting the best possible thumbnail
4015 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4016 'description': video_description,
4017 'uploader': get_first(video_details, 'author'),
4018 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
4019 'uploader_url': owner_profile_url,
4020 'channel_id': channel_id,
4021 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s'),
4022 'duration': duration,
4023 'view_count': int_or_none(
4024 get_first((video_details, microformats), (..., 'viewCount'))
4025 or search_meta('interactionCount')),
4026 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4027 'age_limit': 18 if (
4028 get_first(microformats, 'isFamilySafe') is False
4029 or search_meta('isFamilyFriendly') == 'false'
4030 or search_meta('og:restrictions:age') == '18+') else 0,
4031 'webpage_url': webpage_url,
4032 'categories': [category] if category else None,
4033 'tags': keywords,
4034 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4035 'live_status': live_status,
4036 'release_timestamp': live_start_time,
4037 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4038 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
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' and trans_code != 'und':
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)