[ie/youtube] Add age-gate workaround for some embeddable videos (#11821)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
blob0d3963116ed5ee375596032a685a4130b712f39a
1 import base64
2 import calendar
3 import collections
4 import copy
5 import datetime as dt
6 import enum
7 import functools
8 import hashlib
9 import itertools
10 import json
11 import math
12 import os.path
13 import random
14 import re
15 import shlex
16 import sys
17 import threading
18 import time
19 import traceback
20 import urllib.parse
22 from .common import InfoExtractor, SearchInfoExtractor
23 from .openload import PhantomJSwrapper
24 from ..jsinterp import JSInterpreter
25 from ..networking.exceptions import HTTPError, network_exceptions
26 from ..utils import (
27 NO_DEFAULT,
28 ExtractorError,
29 LazyList,
30 UserNotLive,
31 bug_reports_message,
32 classproperty,
33 clean_html,
34 datetime_from_str,
35 dict_get,
36 filesize_from_tbr,
37 filter_dict,
38 float_or_none,
39 format_field,
40 get_first,
41 int_or_none,
42 is_html,
43 join_nonempty,
44 js_to_json,
45 mimetype2ext,
46 orderedSet,
47 parse_codecs,
48 parse_count,
49 parse_duration,
50 parse_iso8601,
51 parse_qs,
52 qualities,
53 remove_end,
54 remove_start,
55 smuggle_url,
56 str_or_none,
57 str_to_int,
58 strftime_or_none,
59 traverse_obj,
60 try_call,
61 try_get,
62 unescapeHTML,
63 unified_strdate,
64 unified_timestamp,
65 unsmuggle_url,
66 update_url_query,
67 url_or_none,
68 urljoin,
69 variadic,
72 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
73 STREAMING_DATA_PO_TOKEN = '__yt_dlp_po_token'
75 # any clients starting with _ cannot be explicitly requested by the user
76 INNERTUBE_CLIENTS = {
77 'web': {
78 'INNERTUBE_CONTEXT': {
79 'client': {
80 'clientName': 'WEB',
81 'clientVersion': '2.20241126.01.00',
84 'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
85 'REQUIRE_PO_TOKEN': True,
86 'SUPPORTS_COOKIES': True,
88 # Safari UA returns pre-merged video+audio 144p/240p/360p/720p/1080p HLS formats
89 'web_safari': {
90 'INNERTUBE_CONTEXT': {
91 'client': {
92 'clientName': 'WEB',
93 'clientVersion': '2.20241126.01.00',
94 'userAgent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.5 Safari/605.1.15,gzip(gfe)',
97 'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
98 'REQUIRE_PO_TOKEN': True,
99 'SUPPORTS_COOKIES': True,
101 'web_embedded': {
102 'INNERTUBE_CONTEXT': {
103 'client': {
104 'clientName': 'WEB_EMBEDDED_PLAYER',
105 'clientVersion': '1.20241201.00.00',
108 'INNERTUBE_CONTEXT_CLIENT_NAME': 56,
109 'SUPPORTS_COOKIES': True,
111 'web_music': {
112 'INNERTUBE_HOST': 'music.youtube.com',
113 'INNERTUBE_CONTEXT': {
114 'client': {
115 'clientName': 'WEB_REMIX',
116 'clientVersion': '1.20241127.01.00',
119 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
120 'SUPPORTS_COOKIES': True,
122 # This client now requires sign-in for every video
123 'web_creator': {
124 'INNERTUBE_CONTEXT': {
125 'client': {
126 'clientName': 'WEB_CREATOR',
127 'clientVersion': '1.20241203.01.00',
130 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
131 'REQUIRE_AUTH': True,
132 'SUPPORTS_COOKIES': True,
134 'android': {
135 'INNERTUBE_CONTEXT': {
136 'client': {
137 'clientName': 'ANDROID',
138 'clientVersion': '19.44.38',
139 'androidSdkVersion': 30,
140 'userAgent': 'com.google.android.youtube/19.44.38 (Linux; U; Android 11) gzip',
141 'osName': 'Android',
142 'osVersion': '11',
145 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
146 'REQUIRE_JS_PLAYER': False,
147 'REQUIRE_PO_TOKEN': True,
149 # This client now requires sign-in for every video
150 'android_music': {
151 'INNERTUBE_CONTEXT': {
152 'client': {
153 'clientName': 'ANDROID_MUSIC',
154 'clientVersion': '7.27.52',
155 'androidSdkVersion': 30,
156 'userAgent': 'com.google.android.apps.youtube.music/7.27.52 (Linux; U; Android 11) gzip',
157 'osName': 'Android',
158 'osVersion': '11',
161 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
162 'REQUIRE_JS_PLAYER': False,
163 'REQUIRE_PO_TOKEN': True,
164 'REQUIRE_AUTH': True,
165 'SUPPORTS_COOKIES': True,
167 # This client now requires sign-in for every video
168 'android_creator': {
169 'INNERTUBE_CONTEXT': {
170 'client': {
171 'clientName': 'ANDROID_CREATOR',
172 'clientVersion': '24.45.100',
173 'androidSdkVersion': 30,
174 'userAgent': 'com.google.android.apps.youtube.creator/24.45.100 (Linux; U; Android 11) gzip',
175 'osName': 'Android',
176 'osVersion': '11',
179 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
180 'REQUIRE_JS_PLAYER': False,
181 'REQUIRE_PO_TOKEN': True,
182 'REQUIRE_AUTH': True,
184 # YouTube Kids videos aren't returned on this client for some reason
185 'android_vr': {
186 'INNERTUBE_CONTEXT': {
187 'client': {
188 'clientName': 'ANDROID_VR',
189 'clientVersion': '1.60.19',
190 'deviceMake': 'Oculus',
191 'deviceModel': 'Quest 3',
192 'androidSdkVersion': 32,
193 'userAgent': 'com.google.android.apps.youtube.vr.oculus/1.60.19 (Linux; U; Android 12L; eureka-user Build/SQ3A.220605.009.A1) gzip',
194 'osName': 'Android',
195 'osVersion': '12L',
198 'INNERTUBE_CONTEXT_CLIENT_NAME': 28,
199 'REQUIRE_JS_PLAYER': False,
200 'SUPPORTS_COOKIES': True,
202 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
203 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
204 'ios': {
205 'INNERTUBE_CONTEXT': {
206 'client': {
207 'clientName': 'IOS',
208 'clientVersion': '19.45.4',
209 'deviceMake': 'Apple',
210 'deviceModel': 'iPhone16,2',
211 'userAgent': 'com.google.ios.youtube/19.45.4 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
212 'osName': 'iPhone',
213 'osVersion': '18.1.0.22B83',
216 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
217 'REQUIRE_JS_PLAYER': False,
219 # This client now requires sign-in for every video
220 'ios_music': {
221 'INNERTUBE_CONTEXT': {
222 'client': {
223 'clientName': 'IOS_MUSIC',
224 'clientVersion': '7.27.0',
225 'deviceMake': 'Apple',
226 'deviceModel': 'iPhone16,2',
227 'userAgent': 'com.google.ios.youtubemusic/7.27.0 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
228 'osName': 'iPhone',
229 'osVersion': '18.1.0.22B83',
232 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
233 'REQUIRE_JS_PLAYER': False,
234 'REQUIRE_AUTH': True,
235 'SUPPORTS_COOKIES': True,
237 # This client now requires sign-in for every video
238 'ios_creator': {
239 'INNERTUBE_CONTEXT': {
240 'client': {
241 'clientName': 'IOS_CREATOR',
242 'clientVersion': '24.45.100',
243 'deviceMake': 'Apple',
244 'deviceModel': 'iPhone16,2',
245 'userAgent': 'com.google.ios.ytcreator/24.45.100 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
246 'osName': 'iPhone',
247 'osVersion': '18.1.0.22B83',
250 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
251 'REQUIRE_JS_PLAYER': False,
252 'REQUIRE_AUTH': True,
254 # mweb has 'ultralow' formats
255 # See: https://github.com/yt-dlp/yt-dlp/pull/557
256 'mweb': {
257 'INNERTUBE_CONTEXT': {
258 'client': {
259 'clientName': 'MWEB',
260 'clientVersion': '2.20241202.07.00',
261 'userAgent': 'Mozilla/5.0 (iPad; CPU OS 16_7_10 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Mobile/15E148 Safari/604.1,gzip(gfe)',
264 'INNERTUBE_CONTEXT_CLIENT_NAME': 2,
265 'SUPPORTS_COOKIES': True,
267 'tv': {
268 'INNERTUBE_CONTEXT': {
269 'client': {
270 'clientName': 'TVHTML5',
271 'clientVersion': '7.20241201.18.00',
274 'INNERTUBE_CONTEXT_CLIENT_NAME': 7,
275 'SUPPORTS_COOKIES': True,
277 # This client now requires sign-in for every video
278 # It was previously an age-gate workaround for videos that were `playable_in_embed`
279 # It may still be useful if signed into an EU account that is not age-verified
280 'tv_embedded': {
281 'INNERTUBE_CONTEXT': {
282 'client': {
283 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
284 'clientVersion': '2.0',
287 'INNERTUBE_CONTEXT_CLIENT_NAME': 85,
288 'REQUIRE_AUTH': True,
289 'SUPPORTS_COOKIES': True,
294 def _split_innertube_client(client_name):
295 variant, *base = client_name.rsplit('.', 1)
296 if base:
297 return variant, base[0], variant
298 base, *variant = client_name.split('_', 1)
299 return client_name, base, variant[0] if variant else None
302 def short_client_name(client_name):
303 main, *parts = _split_innertube_client(client_name)[0].split('_')
304 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
307 def build_innertube_clients():
308 THIRD_PARTY = {
309 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
311 BASE_CLIENTS = ('ios', 'web', 'tv', 'mweb', 'android')
312 priority = qualities(BASE_CLIENTS[::-1])
314 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
315 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
316 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
317 ytcfg.setdefault('REQUIRE_PO_TOKEN', False)
318 ytcfg.setdefault('REQUIRE_AUTH', False)
319 ytcfg.setdefault('SUPPORTS_COOKIES', False)
320 ytcfg.setdefault('PLAYER_PARAMS', None)
321 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
323 _, base_client, variant = _split_innertube_client(client)
324 ytcfg['priority'] = 10 * priority(base_client)
326 if variant == 'embedded':
327 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
328 ytcfg['priority'] -= 2
329 elif variant:
330 ytcfg['priority'] -= 3
333 build_innertube_clients()
336 class BadgeType(enum.Enum):
337 AVAILABILITY_UNLISTED = enum.auto()
338 AVAILABILITY_PRIVATE = enum.auto()
339 AVAILABILITY_PUBLIC = enum.auto()
340 AVAILABILITY_PREMIUM = enum.auto()
341 AVAILABILITY_SUBSCRIPTION = enum.auto()
342 LIVE_NOW = enum.auto()
343 VERIFIED = enum.auto()
346 class YoutubeBaseInfoExtractor(InfoExtractor):
347 """Provide base functions for Youtube extractors"""
349 _RESERVED_NAMES = (
350 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
351 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
352 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
353 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
355 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
357 # _NETRC_MACHINE = 'youtube'
359 # If True it will raise an error if no login info is provided
360 _LOGIN_REQUIRED = False
362 _INVIDIOUS_SITES = (
363 # invidious-redirect websites
364 r'(?:www\.)?redirect\.invidious\.io',
365 r'(?:(?:www|dev)\.)?invidio\.us',
366 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
367 r'(?:www\.)?invidious\.pussthecat\.org',
368 r'(?:www\.)?invidious\.zee\.li',
369 r'(?:www\.)?invidious\.ethibox\.fr',
370 r'(?:www\.)?iv\.ggtyler\.dev',
371 r'(?:www\.)?inv\.vern\.i2p',
372 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
373 r'(?:www\.)?inv\.riverside\.rocks',
374 r'(?:www\.)?invidious\.silur\.me',
375 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
376 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
377 r'(?:www\.)?invidious\.slipfox\.xyz',
378 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
379 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
380 r'(?:www\.)?invidious\.tiekoetter\.com',
381 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
382 r'(?:www\.)?invidious\.nerdvpn\.de',
383 r'(?:www\.)?invidious\.weblibre\.org',
384 r'(?:www\.)?inv\.odyssey346\.dev',
385 r'(?:www\.)?invidious\.dhusch\.de',
386 r'(?:www\.)?iv\.melmac\.space',
387 r'(?:www\.)?watch\.thekitty\.zone',
388 r'(?:www\.)?invidious\.privacydev\.net',
389 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
390 r'(?:www\.)?invidious\.drivet\.xyz',
391 r'(?:www\.)?vid\.priv\.au',
392 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
393 r'(?:www\.)?inv\.vern\.cc',
394 r'(?:www\.)?invidious\.esmailelbob\.xyz',
395 r'(?:www\.)?invidious\.sethforprivacy\.com',
396 r'(?:www\.)?yt\.oelrichsgarcia\.de',
397 r'(?:www\.)?yt\.artemislena\.eu',
398 r'(?:www\.)?invidious\.flokinet\.to',
399 r'(?:www\.)?invidious\.baczek\.me',
400 r'(?:www\.)?y\.com\.sb',
401 r'(?:www\.)?invidious\.epicsite\.xyz',
402 r'(?:www\.)?invidious\.lidarshield\.cloud',
403 r'(?:www\.)?yt\.funami\.tech',
404 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
405 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
406 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
407 # youtube-dl invidious instances list
408 r'(?:(?:www|no)\.)?invidiou\.sh',
409 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
410 r'(?:www\.)?invidious\.kabi\.tk',
411 r'(?:www\.)?invidious\.mastodon\.host',
412 r'(?:www\.)?invidious\.zapashcanon\.fr',
413 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
414 r'(?:www\.)?invidious\.tinfoil-hat\.net',
415 r'(?:www\.)?invidious\.himiko\.cloud',
416 r'(?:www\.)?invidious\.reallyancient\.tech',
417 r'(?:www\.)?invidious\.tube',
418 r'(?:www\.)?invidiou\.site',
419 r'(?:www\.)?invidious\.site',
420 r'(?:www\.)?invidious\.xyz',
421 r'(?:www\.)?invidious\.nixnet\.xyz',
422 r'(?:www\.)?invidious\.048596\.xyz',
423 r'(?:www\.)?invidious\.drycat\.fr',
424 r'(?:www\.)?inv\.skyn3t\.in',
425 r'(?:www\.)?tube\.poal\.co',
426 r'(?:www\.)?tube\.connect\.cafe',
427 r'(?:www\.)?vid\.wxzm\.sx',
428 r'(?:www\.)?vid\.mint\.lgbt',
429 r'(?:www\.)?vid\.puffyan\.us',
430 r'(?:www\.)?yewtu\.be',
431 r'(?:www\.)?yt\.elukerio\.org',
432 r'(?:www\.)?yt\.lelux\.fi',
433 r'(?:www\.)?invidious\.ggc-project\.de',
434 r'(?:www\.)?yt\.maisputain\.ovh',
435 r'(?:www\.)?ytprivate\.com',
436 r'(?:www\.)?invidious\.13ad\.de',
437 r'(?:www\.)?invidious\.toot\.koeln',
438 r'(?:www\.)?invidious\.fdn\.fr',
439 r'(?:www\.)?watch\.nettohikari\.com',
440 r'(?:www\.)?invidious\.namazso\.eu',
441 r'(?:www\.)?invidious\.silkky\.cloud',
442 r'(?:www\.)?invidious\.exonip\.de',
443 r'(?:www\.)?invidious\.riverside\.rocks',
444 r'(?:www\.)?invidious\.blamefran\.net',
445 r'(?:www\.)?invidious\.moomoo\.de',
446 r'(?:www\.)?ytb\.trom\.tf',
447 r'(?:www\.)?yt\.cyberhost\.uk',
448 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
449 r'(?:www\.)?qklhadlycap4cnod\.onion',
450 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
451 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
452 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
453 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
454 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
455 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
456 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
457 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
458 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
459 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
460 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
461 r'(?:www\.)?piped\.kavin\.rocks',
462 r'(?:www\.)?piped\.tokhmi\.xyz',
463 r'(?:www\.)?piped\.syncpundit\.io',
464 r'(?:www\.)?piped\.mha\.fi',
465 r'(?:www\.)?watch\.whatever\.social',
466 r'(?:www\.)?piped\.garudalinux\.org',
467 r'(?:www\.)?piped\.rivo\.lol',
468 r'(?:www\.)?piped-libre\.kavin\.rocks',
469 r'(?:www\.)?yt\.jae\.fi',
470 r'(?:www\.)?piped\.mint\.lgbt',
471 r'(?:www\.)?il\.ax',
472 r'(?:www\.)?piped\.esmailelbob\.xyz',
473 r'(?:www\.)?piped\.projectsegfau\.lt',
474 r'(?:www\.)?piped\.privacydev\.net',
475 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
476 r'(?:www\.)?piped\.smnz\.de',
477 r'(?:www\.)?piped\.adminforge\.de',
478 r'(?:www\.)?watch\.whatevertinfoil\.de',
479 r'(?:www\.)?piped\.qdi\.fi',
480 r'(?:(?:www|cf)\.)?piped\.video',
481 r'(?:www\.)?piped\.aeong\.one',
482 r'(?:www\.)?piped\.moomoo\.me',
483 r'(?:www\.)?piped\.chauvet\.pro',
484 r'(?:www\.)?watch\.leptons\.xyz',
485 r'(?:www\.)?pd\.vern\.cc',
486 r'(?:www\.)?piped\.hostux\.net',
487 r'(?:www\.)?piped\.lunar\.icu',
488 # Hyperpipe instances from https://hyperpipe.codeberg.page/
489 r'(?:www\.)?hyperpipe\.surge\.sh',
490 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
491 r'(?:www\.)?listen\.whatever\.social',
492 r'(?:www\.)?music\.adminforge\.de',
495 # extracted from account/account_menu ep
496 # XXX: These are the supported YouTube UI and API languages,
497 # which is slightly different from languages supported for translation in YouTube studio
498 _SUPPORTED_LANG_CODES = [
499 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
500 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
501 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
502 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
503 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
504 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko',
507 _IGNORED_WARNINGS = {
508 'Unavailable videos will be hidden during playback',
509 'Unavailable videos are hidden',
512 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
513 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
515 _NETRC_MACHINE = 'youtube'
517 def ucid_or_none(self, ucid):
518 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
520 def handle_or_none(self, handle):
521 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', urllib.parse.unquote(handle or ''),
522 '@-handle', default=None)
524 def handle_from_url(self, url):
525 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
526 urllib.parse.unquote(url or ''), 'channel handle', default=None)
528 def ucid_from_url(self, url):
529 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
530 url, 'channel id', default=None)
532 @functools.cached_property
533 def _preferred_lang(self):
535 Returns a language code supported by YouTube for the user preferred language.
536 Returns None if no preferred language set.
538 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
539 if not preferred_lang:
540 return
541 if preferred_lang not in self._SUPPORTED_LANG_CODES:
542 raise ExtractorError(
543 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
544 expected=True)
545 elif preferred_lang != 'en':
546 self.report_warning(
547 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
548 return preferred_lang
550 def _initialize_consent(self):
551 cookies = self._get_cookies('https://www.youtube.com/')
552 if cookies.get('__Secure-3PSID'):
553 return
554 socs = cookies.get('SOCS')
555 if socs and not socs.value.startswith('CAA'): # not consented
556 return
557 self._set_cookie('.youtube.com', 'SOCS', 'CAI', secure=True) # accept all (required for mixes)
559 def _initialize_pref(self):
560 cookies = self._get_cookies('https://www.youtube.com/')
561 pref_cookie = cookies.get('PREF')
562 pref = {}
563 if pref_cookie:
564 try:
565 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
566 except ValueError:
567 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
568 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
569 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
571 def _real_initialize(self):
572 self._initialize_pref()
573 self._initialize_consent()
574 self._check_login_required()
576 def _perform_login(self, username, password):
577 if username.startswith('oauth'):
578 raise ExtractorError(
579 f'Login with OAuth is no longer supported. {self._youtube_login_hint}', expected=True)
581 self.report_warning(
582 f'Login with password is not supported for YouTube. {self._youtube_login_hint}')
584 @property
585 def _youtube_login_hint(self):
586 return (f'{self._login_hint(method="cookies")}. Also see '
587 'https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies '
588 'for tips on effectively exporting YouTube cookies')
590 def _check_login_required(self):
591 if self._LOGIN_REQUIRED and not self.is_authenticated:
592 self.raise_login_required(
593 f'Login details are needed to download this content. {self._youtube_login_hint}', method=None)
595 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
596 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
598 def _get_default_ytcfg(self, client='web'):
599 return copy.deepcopy(INNERTUBE_CLIENTS[client])
601 def _get_innertube_host(self, client='web'):
602 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
604 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
605 # try_get but with fallback to default ytcfg client values when present
606 _func = lambda y: try_get(y, getter, expected_type)
607 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
609 def _extract_client_name(self, ytcfg, default_client='web'):
610 return self._ytcfg_get_safe(
611 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
612 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
614 def _extract_client_version(self, ytcfg, default_client='web'):
615 return self._ytcfg_get_safe(
616 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
617 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
619 def _select_api_hostname(self, req_api_hostname, default_client=None):
620 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
621 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
623 def _extract_context(self, ytcfg=None, default_client='web'):
624 context = get_first(
625 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
626 # Enforce language and tz for extraction
627 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
628 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
629 return context
631 _SAPISID = None
633 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
634 time_now = round(time.time())
635 if self._SAPISID is None:
636 yt_cookies = self._get_cookies('https://www.youtube.com')
637 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
638 # See: https://github.com/yt-dlp/yt-dlp/issues/393
639 sapisid_cookie = dict_get(
640 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
641 if sapisid_cookie and sapisid_cookie.value:
642 self._SAPISID = sapisid_cookie.value
643 self.write_debug('Extracted SAPISID cookie')
644 # SAPISID cookie is required if not already present
645 if not yt_cookies.get('SAPISID'):
646 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
647 self._set_cookie(
648 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
649 else:
650 self._SAPISID = False
651 if not self._SAPISID:
652 return None
653 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
654 sapisidhash = hashlib.sha1(
655 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
656 return f'SAPISIDHASH {time_now}_{sapisidhash}'
658 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
659 note='Downloading API JSON', errnote='Unable to download API page',
660 context=None, api_key=None, api_hostname=None, default_client='web'):
662 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
663 data.update(query)
664 real_headers = self.generate_api_headers(default_client=default_client)
665 real_headers.update({'content-type': 'application/json'})
666 if headers:
667 real_headers.update(headers)
668 return self._download_json(
669 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
670 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
671 data=json.dumps(data).encode('utf8'), headers=real_headers,
672 query=filter_dict({
673 'key': self._configuration_arg(
674 'innertube_key', [api_key], ie_key=YoutubeIE.ie_key(), casesense=True)[0],
675 'prettyPrint': 'false',
676 }, cndn=lambda _, v: v))
678 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
679 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
681 @staticmethod
682 def _extract_session_index(*data):
684 Index of current account in account list.
685 See: https://github.com/yt-dlp/yt-dlp/pull/519
687 for ytcfg in data:
688 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
689 if session_index is not None:
690 return session_index
692 def _data_sync_id_to_delegated_session_id(self, data_sync_id):
693 if not data_sync_id:
694 return
695 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
696 # and just "user_syncid||" for primary channel. We only want the channel_syncid
697 channel_syncid, _, user_syncid = data_sync_id.partition('||')
698 if user_syncid:
699 return channel_syncid
701 def _extract_account_syncid(self, *args):
703 Extract current session ID required to download private playlists of secondary channels
704 @params response and/or ytcfg
706 # ytcfg includes channel_syncid if on secondary channel
707 if delegated_sid := traverse_obj(args, (..., 'DELEGATED_SESSION_ID', {str}, any)):
708 return delegated_sid
710 data_sync_id = self._extract_data_sync_id(*args)
711 return self._data_sync_id_to_delegated_session_id(data_sync_id)
713 def _extract_data_sync_id(self, *args):
715 Extract current account dataSyncId.
716 In the format DELEGATED_SESSION_ID||USER_SESSION_ID or USER_SESSION_ID||
717 @params response and/or ytcfg
719 if data_sync_id := self._configuration_arg('data_sync_id', [None], ie_key=YoutubeIE, casesense=True)[0]:
720 return data_sync_id
722 return traverse_obj(
723 args, (..., ('DATASYNC_ID', ('responseContext', 'mainAppWebResponseContext', 'datasyncId')), {str}, any))
725 def _extract_visitor_data(self, *args):
727 Extracts visitorData from an API response or ytcfg
728 Appears to be used to track session state
730 if visitor_data := self._configuration_arg('visitor_data', [None], ie_key=YoutubeIE, casesense=True)[0]:
731 return visitor_data
732 return get_first(
733 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
734 expected_type=str)
736 @functools.cached_property
737 def is_authenticated(self):
738 return bool(self._generate_sapisidhash_header())
740 def extract_ytcfg(self, video_id, webpage):
741 if not webpage:
742 return {}
743 return self._parse_json(
744 self._search_regex(
745 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
746 default='{}'), video_id, fatal=False) or {}
748 def _generate_cookie_auth_headers(self, *, ytcfg=None, account_syncid=None, session_index=None, origin=None, **kwargs):
749 headers = {}
750 account_syncid = account_syncid or self._extract_account_syncid(ytcfg)
751 if account_syncid:
752 headers['X-Goog-PageId'] = account_syncid
753 if session_index is None:
754 session_index = self._extract_session_index(ytcfg)
755 if account_syncid or session_index is not None:
756 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
758 auth = self._generate_sapisidhash_header(origin)
759 if auth is not None:
760 headers['Authorization'] = auth
761 headers['X-Origin'] = origin
763 return headers
765 def generate_api_headers(
766 self, *, ytcfg=None, account_syncid=None, session_index=None,
767 visitor_data=None, api_hostname=None, default_client='web', **kwargs):
769 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
770 headers = {
771 'X-YouTube-Client-Name': str(
772 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
773 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
774 'Origin': origin,
775 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
776 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client),
777 **self._generate_cookie_auth_headers(ytcfg=ytcfg, account_syncid=account_syncid, session_index=session_index, origin=origin),
779 return filter_dict(headers)
781 def _download_ytcfg(self, client, video_id):
782 url = {
783 'web': 'https://www.youtube.com',
784 'web_music': 'https://music.youtube.com',
785 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1',
786 }.get(client)
787 if not url:
788 return {}
789 webpage = self._download_webpage(
790 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
791 return self.extract_ytcfg(video_id, webpage) or {}
793 @staticmethod
794 def _build_api_continuation_query(continuation, ctp=None):
795 query = {
796 'continuation': continuation,
798 # TODO: Inconsistency with clickTrackingParams.
799 # Currently we have a fixed ctp contained within context (from ytcfg)
800 # and a ctp in root query for continuation.
801 if ctp:
802 query['clickTracking'] = {'clickTrackingParams': ctp}
803 return query
805 @classmethod
806 def _extract_next_continuation_data(cls, renderer):
807 next_continuation = try_get(
808 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
809 lambda x: x['continuation']['reloadContinuationData']), dict)
810 if not next_continuation:
811 return
812 continuation = next_continuation.get('continuation')
813 if not continuation:
814 return
815 ctp = next_continuation.get('clickTrackingParams')
816 return cls._build_api_continuation_query(continuation, ctp)
818 @classmethod
819 def _extract_continuation_ep_data(cls, continuation_ep: dict):
820 if isinstance(continuation_ep, dict):
821 continuation = try_get(
822 continuation_ep, lambda x: x['continuationCommand']['token'], str)
823 if not continuation:
824 return
825 ctp = continuation_ep.get('clickTrackingParams')
826 return cls._build_api_continuation_query(continuation, ctp)
828 @classmethod
829 def _extract_continuation(cls, renderer):
830 next_continuation = cls._extract_next_continuation_data(renderer)
831 if next_continuation:
832 return next_continuation
834 return traverse_obj(renderer, (
835 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
836 ('continuationEndpoint', ('button', 'buttonRenderer', 'command')),
837 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
839 @classmethod
840 def _extract_alerts(cls, data):
841 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
842 if not isinstance(alert_dict, dict):
843 continue
844 for alert in alert_dict.values():
845 alert_type = alert.get('type')
846 if not alert_type:
847 continue
848 message = cls._get_text(alert, 'text')
849 if message:
850 yield alert_type, message
852 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
853 errors, warnings = [], []
854 for alert_type, alert_message in alerts:
855 if alert_type.lower() == 'error' and fatal:
856 errors.append([alert_type, alert_message])
857 elif alert_message not in self._IGNORED_WARNINGS:
858 warnings.append([alert_type, alert_message])
860 for alert_type, alert_message in (warnings + errors[:-1]):
861 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
862 if errors:
863 raise ExtractorError(f'YouTube said: {errors[-1][1]}', expected=expected)
865 def _extract_and_report_alerts(self, data, *args, **kwargs):
866 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
868 def _extract_badges(self, badge_list: list):
870 Extract known BadgeType's from a list of badge renderers.
871 @returns [{'type': BadgeType}]
873 icon_type_map = {
874 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
875 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
876 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC,
877 'CHECK_CIRCLE_THICK': BadgeType.VERIFIED,
878 'OFFICIAL_ARTIST_BADGE': BadgeType.VERIFIED,
879 'CHECK': BadgeType.VERIFIED,
882 badge_style_map = {
883 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
884 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
885 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
886 'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
887 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED,
890 label_map = {
891 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
892 'private': BadgeType.AVAILABILITY_PRIVATE,
893 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
894 'live': BadgeType.LIVE_NOW,
895 'premium': BadgeType.AVAILABILITY_PREMIUM,
896 'verified': BadgeType.VERIFIED,
897 'official artist channel': BadgeType.VERIFIED,
900 badges = []
901 for badge in traverse_obj(badge_list, (..., lambda key, _: re.search(r'[bB]adgeRenderer$', key))):
902 badge_type = (
903 icon_type_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
904 or badge_style_map.get(traverse_obj(badge, 'style'))
906 if badge_type:
907 badges.append({'type': badge_type})
908 continue
910 # fallback, won't work in some languages
911 label = traverse_obj(
912 badge, 'label', ('accessibilityData', 'label'), 'tooltip', 'iconTooltip', get_all=False, expected_type=str, default='')
913 for match, label_badge_type in label_map.items():
914 if match in label.lower():
915 badges.append({'type': label_badge_type})
916 break
918 return badges
920 @staticmethod
921 def _has_badge(badges, badge_type):
922 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
924 @staticmethod
925 def _get_text(data, *path_list, max_runs=None):
926 for path in path_list or [None]:
927 if path is None:
928 obj = [data]
929 else:
930 obj = traverse_obj(data, path, default=[])
931 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
932 obj = [obj]
933 for item in obj:
934 text = try_get(item, lambda x: x['simpleText'], str)
935 if text:
936 return text
937 runs = try_get(item, lambda x: x['runs'], list) or []
938 if not runs and isinstance(item, list):
939 runs = item
941 runs = runs[:min(len(runs), max_runs or len(runs))]
942 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
943 if text:
944 return text
946 def _get_count(self, data, *path_list):
947 count_text = self._get_text(data, *path_list) or ''
948 count = parse_count(count_text)
949 if count is None:
950 count = str_to_int(
951 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
952 return count
954 @staticmethod
955 def _extract_thumbnails(data, *path_list, final_key='thumbnails'):
957 Extract thumbnails from thumbnails dict
958 @param path_list: path list to level that contains 'thumbnails' key
960 thumbnails = []
961 for path in path_list or [()]:
962 for thumbnail in traverse_obj(data, (*variadic(path), final_key, ...)):
963 thumbnail_url = url_or_none(thumbnail.get('url'))
964 if not thumbnail_url:
965 continue
966 # Sometimes youtube gives a wrong thumbnail URL. See:
967 # https://github.com/yt-dlp/yt-dlp/issues/233
968 # https://github.com/ytdl-org/youtube-dl/issues/28023
969 if 'maxresdefault' in thumbnail_url:
970 thumbnail_url = thumbnail_url.split('?')[0]
971 thumbnails.append({
972 'url': thumbnail_url,
973 'height': int_or_none(thumbnail.get('height')),
974 'width': int_or_none(thumbnail.get('width')),
976 return thumbnails
978 @staticmethod
979 def extract_relative_time(relative_time_text):
981 Extracts a relative time from string and converts to dt object
982 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today', '8 yr ago'
985 # XXX: this could be moved to a general function in utils/_utils.py
986 # The relative time text strings are roughly the same as what
987 # Javascript's Intl.RelativeTimeFormat function generates.
988 # See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat
989 mobj = re.search(
990 r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>sec(?:ond)?|s|min(?:ute)?|h(?:our|r)?|d(?:ay)?|w(?:eek|k)?|mo(?:nth)?|y(?:ear|r)?)s?\s*ago',
991 relative_time_text)
992 if mobj:
993 start = mobj.group('start')
994 if start:
995 return datetime_from_str(start)
996 try:
997 return datetime_from_str('now-{}{}'.format(mobj.group('time'), mobj.group('unit')))
998 except ValueError:
999 return None
1001 def _parse_time_text(self, text):
1002 if not text:
1003 return
1004 dt_ = self.extract_relative_time(text)
1005 timestamp = None
1006 if isinstance(dt_, dt.datetime):
1007 timestamp = calendar.timegm(dt_.timetuple())
1009 if timestamp is None:
1010 timestamp = (
1011 unified_timestamp(text) or unified_timestamp(
1012 self._search_regex(
1013 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
1014 text.lower(), 'time text', default=None)))
1016 if text and timestamp is None and self._preferred_lang in (None, 'en'):
1017 self.report_warning(
1018 f'Cannot parse localized time text "{text}"', only_once=True)
1019 return timestamp
1021 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
1022 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
1023 default_client='web'):
1024 raise_for_incomplete = bool(self._configuration_arg('raise_incomplete_data', ie_key=YoutubeIE))
1025 # Incomplete Data should be a warning by default when retries are exhausted, while other errors should be fatal.
1026 icd_retries = iter(self.RetryManager(fatal=raise_for_incomplete))
1027 icd_rm = next(icd_retries)
1028 main_retries = iter(self.RetryManager())
1029 main_rm = next(main_retries)
1030 # Manual retry loop for multiple RetryManagers
1031 # The proper RetryManager MUST be advanced after an error
1032 # and its result MUST be checked if the manager is non fatal
1033 while True:
1034 try:
1035 response = self._call_api(
1036 ep=ep, fatal=True, headers=headers,
1037 video_id=item_id, query=query, note=note,
1038 context=self._extract_context(ytcfg, default_client),
1039 api_hostname=api_hostname, default_client=default_client)
1040 except ExtractorError as e:
1041 if not isinstance(e.cause, network_exceptions):
1042 return self._error_or_warning(e, fatal=fatal)
1043 elif not isinstance(e.cause, HTTPError):
1044 main_rm.error = e
1045 next(main_retries)
1046 continue
1048 first_bytes = e.cause.response.read(512)
1049 if not is_html(first_bytes):
1050 yt_error = try_get(
1051 self._parse_json(
1052 self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
1053 lambda x: x['error']['message'], str)
1054 if yt_error:
1055 self._report_alerts([('ERROR', yt_error)], fatal=False)
1056 # Downloading page may result in intermittent 5xx HTTP error
1057 # Sometimes a 404 is also received. See: https://github.com/ytdl-org/youtube-dl/issues/28289
1058 # We also want to catch all other network exceptions since errors in later pages can be troublesome
1059 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
1060 if e.cause.status not in (403, 429):
1061 main_rm.error = e
1062 next(main_retries)
1063 continue
1064 return self._error_or_warning(e, fatal=fatal)
1066 try:
1067 self._extract_and_report_alerts(response, only_once=True)
1068 except ExtractorError as e:
1069 # YouTube's servers may return errors we want to retry on in a 200 OK response
1070 # See: https://github.com/yt-dlp/yt-dlp/issues/839
1071 if 'unknown error' in e.msg.lower():
1072 main_rm.error = e
1073 next(main_retries)
1074 continue
1075 return self._error_or_warning(e, fatal=fatal)
1076 # Youtube sometimes sends incomplete data
1077 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1078 if not traverse_obj(response, *variadic(check_get_keys)):
1079 icd_rm.error = ExtractorError('Incomplete data received', expected=True)
1080 should_retry = next(icd_retries, None)
1081 if not should_retry:
1082 return None
1083 continue
1085 return response
1087 @staticmethod
1088 def is_music_url(url):
1089 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
1091 def _extract_video(self, renderer):
1092 video_id = renderer.get('videoId')
1094 reel_header_renderer = traverse_obj(renderer, (
1095 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1096 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1098 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1099 description = self._get_text(renderer, 'descriptionSnippet')
1101 duration = int_or_none(renderer.get('lengthSeconds'))
1102 if duration is None:
1103 duration = parse_duration(self._get_text(
1104 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1105 if duration is None:
1106 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1107 duration = parse_duration(self._search_regex(
1108 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1109 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1110 video_id, default=None, group='duration'))
1112 channel_id = traverse_obj(
1113 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1114 expected_type=str, get_all=False)
1115 if not channel_id:
1116 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1118 channel_id = self.ucid_or_none(channel_id)
1120 overlay_style = traverse_obj(
1121 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1122 get_all=False, expected_type=str)
1123 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1124 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1125 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1126 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1127 expected_type=str)) or ''
1128 url = f'https://www.youtube.com/watch?v={video_id}'
1129 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1130 url = f'https://www.youtube.com/shorts/{video_id}'
1132 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1133 or self._get_text(reel_header_renderer, 'timestampText') or '')
1134 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1136 live_status = (
1137 'is_upcoming' if scheduled_timestamp is not None
1138 else 'was_live' if 'streamed' in time_text.lower()
1139 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1140 else None)
1142 # videoInfo is a string like '50K views • 10 years ago'.
1143 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1144 view_count = (0 if 'no views' in view_count_text.lower()
1145 else self._get_count({'simpleText': view_count_text}))
1146 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1148 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1149 or self._get_text(reel_header_renderer, 'channelTitleText'))
1151 channel_handle = traverse_obj(renderer, (
1152 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1153 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1154 expected_type=self.handle_from_url, get_all=False)
1155 return {
1156 '_type': 'url',
1157 'ie_key': YoutubeIE.ie_key(),
1158 'id': video_id,
1159 'url': url,
1160 'title': title,
1161 'description': description,
1162 'duration': duration,
1163 'channel_id': channel_id,
1164 'channel': channel,
1165 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1166 'uploader': channel,
1167 'uploader_id': channel_handle,
1168 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1169 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1170 'timestamp': (self._parse_time_text(time_text)
1171 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1172 else None),
1173 'release_timestamp': scheduled_timestamp,
1174 'availability':
1175 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1176 else self._availability(
1177 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1178 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1179 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1180 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1181 view_count_field: view_count,
1182 'live_status': live_status,
1183 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None,
1187 class YoutubeIE(YoutubeBaseInfoExtractor):
1188 IE_DESC = 'YouTube'
1189 _VALID_URL = r'''(?x)^
1191 (?:https?://|//) # http(s):// or protocol-independent URL
1192 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1193 (?:www\.)?deturl\.com/www\.youtube\.com|
1194 (?:www\.)?pwnyoutube\.com|
1195 (?:www\.)?hooktube\.com|
1196 (?:www\.)?yourepeat\.com|
1197 tube\.majestyc\.net|
1198 {invidious}|
1199 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1200 (?:.*?\#/)? # handle anchor (#/) redirect urls
1201 (?: # the various things that can precede the ID:
1202 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1203 |(?: # or the v= param in all its forms
1204 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1205 (?:\?|\#!?) # the params delimiter ? or # or #!
1206 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1210 |(?:
1211 youtu\.be| # just youtu.be/xxxx
1212 vid\.plus| # or vid.plus/xxxx
1213 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1214 {invidious}
1216 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1218 )? # all until now is optional -> you can pass the naked ID
1219 (?P<id>[0-9A-Za-z_-]{{11}}) # here is it! the YouTube video ID
1220 (?(1).+)? # if we found the ID, everything can follow
1221 (?:\#|$)'''.format(
1222 invidious='|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1224 _EMBED_REGEX = [
1225 r'''(?x)
1227 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1228 data-video-url=|
1229 <embed[^>]+?src=|
1230 embedSWF\(?:\s*|
1231 <object[^>]+data=|
1232 new\s+SWFObject\(
1234 (["\'])
1235 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1236 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1237 \1''',
1238 # https://wordpress.org/plugins/lazy-load-for-videos/
1239 r'''(?xs)
1240 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1241 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1243 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1245 _PLAYER_INFO_RE = (
1246 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1247 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$',
1248 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1250 _formats = { # NB: Used in YoutubeWebArchiveIE and GoogleDriveIE
1251 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1252 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1253 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1254 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1255 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1256 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1257 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1258 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1259 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1260 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1261 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1262 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1263 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1264 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1265 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1266 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1267 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1268 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1271 # 3D videos
1272 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1273 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1274 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1275 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1276 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1277 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1278 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1280 # Apple HTTP Live Streaming
1281 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1282 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1283 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1284 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1285 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1286 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1287 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1288 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1290 # DASH mp4 video
1291 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1292 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1293 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1294 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1295 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1296 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1297 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1298 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1299 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1300 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1301 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1302 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1304 # Dash mp4 audio
1305 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1306 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1307 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1308 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1309 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1310 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1311 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1313 # Dash webm
1314 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1315 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1316 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1317 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1318 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1319 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1320 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1321 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1322 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1323 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1324 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1325 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1326 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1327 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1328 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1329 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1330 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1331 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1332 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1333 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1334 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1335 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1337 # Dash webm audio
1338 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1339 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1341 # Dash webm audio with opus inside
1342 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1343 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1344 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1346 # RTMP (unnamed)
1347 '_rtmp': {'protocol': 'rtmp'},
1349 # av01 video only formats sometimes served with "unknown" codecs
1350 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1351 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1352 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1353 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1354 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1355 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1356 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1357 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1359 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1360 _DEFAULT_CLIENTS = ('ios', 'mweb')
1361 _DEFAULT_AUTHED_CLIENTS = ('web_creator', 'mweb')
1363 _GEO_BYPASS = False
1365 IE_NAME = 'youtube'
1366 _TESTS = [
1368 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1369 'info_dict': {
1370 'id': 'BaW_jenozKc',
1371 'ext': 'mp4',
1372 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1373 'channel': 'Philipp Hagemeister',
1374 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1375 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1376 'upload_date': '20121002',
1377 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1378 'categories': ['Science & Technology'],
1379 'tags': ['youtube-dl'],
1380 'duration': 10,
1381 'view_count': int,
1382 'like_count': int,
1383 'availability': 'public',
1384 'playable_in_embed': True,
1385 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1386 'live_status': 'not_live',
1387 'age_limit': 0,
1388 'start_time': 1,
1389 'end_time': 9,
1390 'comment_count': int,
1391 'channel_follower_count': int,
1392 'uploader': 'Philipp Hagemeister',
1393 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1394 'uploader_id': '@PhilippHagemeister',
1395 'heatmap': 'count:100',
1396 'timestamp': 1349198244,
1400 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1401 'note': 'Embed-only video (#1746)',
1402 'info_dict': {
1403 'id': 'yZIXLfi8CZQ',
1404 'ext': 'mp4',
1405 'upload_date': '20120608',
1406 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1407 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1408 'age_limit': 18,
1410 'skip': 'Private video',
1413 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1414 'note': 'Use the first video ID in the URL',
1415 'info_dict': {
1416 'id': 'BaW_jenozKc',
1417 'ext': 'mp4',
1418 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1419 'channel': 'Philipp Hagemeister',
1420 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1421 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1422 'upload_date': '20121002',
1423 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1424 'categories': ['Science & Technology'],
1425 'tags': ['youtube-dl'],
1426 'duration': 10,
1427 'view_count': int,
1428 'like_count': int,
1429 'availability': 'public',
1430 'playable_in_embed': True,
1431 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1432 'live_status': 'not_live',
1433 'age_limit': 0,
1434 'comment_count': int,
1435 'channel_follower_count': int,
1436 'uploader': 'Philipp Hagemeister',
1437 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1438 'uploader_id': '@PhilippHagemeister',
1439 'heatmap': 'count:100',
1440 'timestamp': 1349198244,
1442 'params': {
1443 'skip_download': True,
1447 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1448 'note': '256k DASH audio (format 141) via DASH manifest',
1449 'info_dict': {
1450 'id': 'a9LDPn-MO4I',
1451 'ext': 'm4a',
1452 'upload_date': '20121002',
1453 'description': '',
1454 'title': 'UHDTV TEST 8K VIDEO.mp4',
1456 'params': {
1457 'youtube_include_dash_manifest': True,
1458 'format': '141',
1460 'skip': 'format 141 not served anymore',
1462 # DASH manifest with encrypted signature
1464 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1465 'info_dict': {
1466 'id': 'IB3lcPjvWLA',
1467 'ext': 'm4a',
1468 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1469 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1470 'duration': 244,
1471 'upload_date': '20131011',
1472 'abr': 129.495,
1473 'like_count': int,
1474 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1475 'playable_in_embed': True,
1476 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1477 'view_count': int,
1478 'track': 'The Spark',
1479 'live_status': 'not_live',
1480 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1481 'channel': 'Afrojack',
1482 'tags': 'count:19',
1483 'availability': 'public',
1484 'categories': ['Music'],
1485 'age_limit': 0,
1486 'alt_title': 'The Spark',
1487 'channel_follower_count': int,
1488 'uploader': 'Afrojack',
1489 'uploader_url': 'https://www.youtube.com/@Afrojack',
1490 'uploader_id': '@Afrojack',
1492 'params': {
1493 'youtube_include_dash_manifest': True,
1494 'format': '141/bestaudio[ext=m4a]',
1497 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1499 'note': 'Embed allowed age-gate video; works with web_embedded',
1500 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1501 'info_dict': {
1502 'id': 'HtVdAasjOgU',
1503 'ext': 'mp4',
1504 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1505 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1506 'duration': 142,
1507 'upload_date': '20140605',
1508 'age_limit': 18,
1509 'categories': ['Gaming'],
1510 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1511 'availability': 'needs_auth',
1512 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1513 'like_count': int,
1514 'channel': 'The Witcher',
1515 'live_status': 'not_live',
1516 'tags': 'count:17',
1517 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1518 'playable_in_embed': True,
1519 'view_count': int,
1520 'channel_follower_count': int,
1521 'uploader': 'The Witcher',
1522 'uploader_url': 'https://www.youtube.com/@thewitcher',
1523 'uploader_id': '@thewitcher',
1524 'comment_count': int,
1525 'channel_is_verified': True,
1526 'heatmap': 'count:100',
1527 'timestamp': 1401991663,
1531 'note': 'Age-gate video with embed allowed in public site',
1532 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1533 'info_dict': {
1534 'id': 'HsUATh_Nc2U',
1535 'ext': 'mp4',
1536 'title': 'Godzilla 2 (Official Video)',
1537 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1538 'upload_date': '20200408',
1539 'age_limit': 18,
1540 'availability': 'needs_auth',
1541 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1542 'channel': 'FlyingKitty',
1543 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1544 'view_count': int,
1545 'categories': ['Entertainment'],
1546 'live_status': 'not_live',
1547 'tags': ['Flyingkitty', 'godzilla 2'],
1548 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1549 'like_count': int,
1550 'duration': 177,
1551 'playable_in_embed': True,
1552 'channel_follower_count': int,
1553 'uploader': 'FlyingKitty',
1554 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1555 'uploader_id': '@FlyingKitty900',
1556 'comment_count': int,
1557 'channel_is_verified': True,
1559 'skip': 'Age-restricted; requires authentication',
1562 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1563 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1564 'info_dict': {
1565 'id': 'Tq92D6wQ1mg',
1566 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1567 'ext': 'mp4',
1568 'upload_date': '20191228',
1569 'description': 'md5:17eccca93a786d51bc67646756894066',
1570 'age_limit': 18,
1571 'like_count': int,
1572 'availability': 'needs_auth',
1573 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1574 'view_count': int,
1575 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1576 'channel': 'Projekt Melody',
1577 'live_status': 'not_live',
1578 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1579 'playable_in_embed': True,
1580 'categories': ['Entertainment'],
1581 'duration': 106,
1582 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1583 'comment_count': int,
1584 'channel_follower_count': int,
1585 'uploader': 'Projekt Melody',
1586 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1587 'uploader_id': '@ProjektMelody',
1588 'timestamp': 1577508724,
1590 'skip': 'Age-restricted; requires authentication',
1593 'note': 'Non-Agegated non-embeddable video',
1594 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1595 'info_dict': {
1596 'id': 'MeJVWBSsPAY',
1597 'ext': 'mp4',
1598 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1599 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1600 'upload_date': '20130730',
1601 'track': 'Such mich find mich',
1602 'age_limit': 0,
1603 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1604 'like_count': int,
1605 'playable_in_embed': False,
1606 'creator': 'OOMPH!',
1607 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1608 'view_count': int,
1609 'alt_title': 'Such mich find mich',
1610 'duration': 210,
1611 'channel': 'Herr Lurik',
1612 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1613 'categories': ['Music'],
1614 'availability': 'public',
1615 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1616 'live_status': 'not_live',
1617 'artist': 'OOMPH!',
1618 'channel_follower_count': int,
1619 'uploader': 'Herr Lurik',
1620 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1621 'uploader_id': '@HerrLurik',
1625 'note': 'Non-bypassable age-gated video',
1626 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1627 'only_matching': True,
1629 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1630 # YouTube Red ad is not captured for creator
1632 'url': '__2ABJjxzNo',
1633 'info_dict': {
1634 'id': '__2ABJjxzNo',
1635 'ext': 'mp4',
1636 'duration': 266,
1637 'upload_date': '20100430',
1638 'creator': 'deadmau5',
1639 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1640 'title': 'Deadmau5 - Some Chords (HD)',
1641 'alt_title': 'Some Chords',
1642 'availability': 'public',
1643 'tags': 'count:14',
1644 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1645 'view_count': int,
1646 'live_status': 'not_live',
1647 'channel': 'deadmau5',
1648 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1649 'like_count': int,
1650 'track': 'Some Chords',
1651 'artist': 'deadmau5',
1652 'playable_in_embed': True,
1653 'age_limit': 0,
1654 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1655 'categories': ['Music'],
1656 'album': 'Some Chords',
1657 'channel_follower_count': int,
1658 'uploader': 'deadmau5',
1659 'uploader_url': 'https://www.youtube.com/@deadmau5',
1660 'uploader_id': '@deadmau5',
1662 'expected_warnings': [
1663 'DASH manifest missing',
1666 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1668 'url': 'lqQg6PlCWgI',
1669 'info_dict': {
1670 'id': 'lqQg6PlCWgI',
1671 'ext': 'mp4',
1672 'duration': 6085,
1673 'upload_date': '20150827',
1674 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1675 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1676 'like_count': int,
1677 'release_timestamp': 1343767800,
1678 'playable_in_embed': True,
1679 'categories': ['Sports'],
1680 'release_date': '20120731',
1681 'channel': 'Olympics',
1682 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1683 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1684 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1685 'age_limit': 0,
1686 'availability': 'public',
1687 'live_status': 'was_live',
1688 'view_count': int,
1689 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1690 'channel_follower_count': int,
1691 'uploader': 'Olympics',
1692 'uploader_url': 'https://www.youtube.com/@Olympics',
1693 'uploader_id': '@Olympics',
1694 'channel_is_verified': True,
1695 'timestamp': 1440707674,
1697 'params': {
1698 'skip_download': 'requires avconv',
1701 # Non-square pixels
1703 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1704 'info_dict': {
1705 'id': '_b-2C3KPAM0',
1706 'ext': 'mp4',
1707 'stretched_ratio': 16 / 9.,
1708 'duration': 85,
1709 'upload_date': '20110310',
1710 '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',
1711 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1712 'playable_in_embed': True,
1713 'channel': '孫ᄋᄅ',
1714 'age_limit': 0,
1715 'tags': 'count:11',
1716 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1717 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1718 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1719 'view_count': int,
1720 'categories': ['People & Blogs'],
1721 'like_count': int,
1722 'live_status': 'not_live',
1723 'availability': 'unlisted',
1724 'comment_count': int,
1725 'channel_follower_count': int,
1726 'uploader': '孫ᄋᄅ',
1727 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1728 'uploader_id': '@AllenMeow',
1729 'timestamp': 1299776999,
1732 # url_encoded_fmt_stream_map is empty string
1734 'url': 'qEJwOuvDf7I',
1735 'info_dict': {
1736 'id': 'qEJwOuvDf7I',
1737 'ext': 'webm',
1738 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1739 'description': '',
1740 'upload_date': '20150404',
1742 'params': {
1743 'skip_download': 'requires avconv',
1745 'skip': 'This live event has ended.',
1747 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1749 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1750 'info_dict': {
1751 'id': 'FIl7x6_3R5Y',
1752 'ext': 'webm',
1753 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1754 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1755 'duration': 220,
1756 'upload_date': '20150625',
1757 'formats': 'mincount:31',
1759 'skip': 'not actual anymore',
1761 # DASH manifest with segment_list
1763 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1764 'md5': '8ce563a1d667b599d21064e982ab9e31',
1765 'info_dict': {
1766 'id': 'CsmdDsKjzN8',
1767 'ext': 'mp4',
1768 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1769 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1770 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1772 'params': {
1773 'youtube_include_dash_manifest': True,
1774 'format': '135', # bestvideo
1776 'skip': 'This live event has ended.',
1779 # Multifeed videos (multiple cameras), URL can be of any Camera
1780 # TODO: fix multifeed titles
1781 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1782 'info_dict': {
1783 'id': 'zaPI8MvL8pg',
1784 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1785 'description': 'md5:563ccbc698b39298481ca3c571169519',
1787 'playlist': [{
1788 'info_dict': {
1789 'id': 'j5yGuxZ8lLU',
1790 'ext': 'mp4',
1791 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1792 'description': 'md5:563ccbc698b39298481ca3c571169519',
1793 'duration': 10120,
1794 'channel_follower_count': int,
1795 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1796 'availability': 'public',
1797 'playable_in_embed': True,
1798 'upload_date': '20131105',
1799 'categories': ['Gaming'],
1800 'live_status': 'was_live',
1801 'tags': 'count:24',
1802 'release_timestamp': 1383701910,
1803 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1804 'comment_count': int,
1805 'age_limit': 0,
1806 'like_count': int,
1807 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1808 'channel': 'WiiLikeToPlay',
1809 'view_count': int,
1810 'release_date': '20131106',
1811 'uploader': 'WiiLikeToPlay',
1812 'uploader_id': '@WLTP',
1813 'uploader_url': 'https://www.youtube.com/@WLTP',
1815 }, {
1816 'info_dict': {
1817 'id': 'zaPI8MvL8pg',
1818 'ext': 'mp4',
1819 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1820 'availability': 'public',
1821 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1822 'channel': 'WiiLikeToPlay',
1823 'channel_follower_count': int,
1824 'description': 'md5:563ccbc698b39298481ca3c571169519',
1825 'duration': 10108,
1826 'age_limit': 0,
1827 'like_count': int,
1828 'tags': 'count:24',
1829 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1830 'release_timestamp': 1383701915,
1831 'comment_count': int,
1832 'upload_date': '20131105',
1833 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1834 'release_date': '20131106',
1835 'playable_in_embed': True,
1836 'live_status': 'was_live',
1837 'categories': ['Gaming'],
1838 'view_count': int,
1839 'uploader': 'WiiLikeToPlay',
1840 'uploader_id': '@WLTP',
1841 'uploader_url': 'https://www.youtube.com/@WLTP',
1843 }, {
1844 'info_dict': {
1845 'id': 'R7r3vfO7Hao',
1846 'ext': 'mp4',
1847 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1848 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1849 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1850 'like_count': int,
1851 'availability': 'public',
1852 'playable_in_embed': True,
1853 'upload_date': '20131105',
1854 'description': 'md5:563ccbc698b39298481ca3c571169519',
1855 'channel_follower_count': int,
1856 'tags': 'count:24',
1857 'release_date': '20131106',
1858 'comment_count': int,
1859 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1860 'channel': 'WiiLikeToPlay',
1861 'categories': ['Gaming'],
1862 'release_timestamp': 1383701914,
1863 'live_status': 'was_live',
1864 'age_limit': 0,
1865 'duration': 10128,
1866 'view_count': int,
1867 'uploader': 'WiiLikeToPlay',
1868 'uploader_id': '@WLTP',
1869 'uploader_url': 'https://www.youtube.com/@WLTP',
1872 'params': {'skip_download': True},
1873 'skip': 'Not multifeed anymore',
1876 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1877 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1878 'info_dict': {
1879 'id': 'gVfLd0zydlo',
1880 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1882 'playlist_count': 2,
1883 'skip': 'Not multifeed anymore',
1886 'url': 'https://vid.plus/FlRa-iH7PGw',
1887 'only_matching': True,
1890 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1891 'only_matching': True,
1894 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1895 # Also tests cut-off URL expansion in video description (see
1896 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1897 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1898 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1899 'info_dict': {
1900 'id': 'lsguqyKfVQg',
1901 'ext': 'mp4',
1902 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1903 'alt_title': 'Dark Walk',
1904 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1905 'duration': 133,
1906 'upload_date': '20151119',
1907 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1908 'track': 'Dark Walk',
1909 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1910 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1911 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1912 'categories': ['Film & Animation'],
1913 'view_count': int,
1914 'live_status': 'not_live',
1915 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1916 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1917 'tags': 'count:13',
1918 'availability': 'public',
1919 'channel': 'IronSoulElf',
1920 'playable_in_embed': True,
1921 'like_count': int,
1922 'age_limit': 0,
1923 'channel_follower_count': int,
1925 'params': {
1926 'skip_download': True,
1930 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1931 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1932 'only_matching': True,
1935 # Video with yt:stretch=17:0
1936 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1937 'info_dict': {
1938 'id': 'Q39EVAstoRM',
1939 'ext': 'mp4',
1940 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1941 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1942 'upload_date': '20151107',
1944 'params': {
1945 'skip_download': True,
1947 'skip': 'This video does not exist.',
1950 # Video with incomplete 'yt:stretch=16:'
1951 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1952 'only_matching': True,
1955 # Video licensed under Creative Commons
1956 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1957 'info_dict': {
1958 'id': 'M4gD1WSo5mA',
1959 'ext': 'mp4',
1960 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1961 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1962 'duration': 721,
1963 'upload_date': '20150128',
1964 'license': 'Creative Commons Attribution license (reuse allowed)',
1965 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1966 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1967 'like_count': int,
1968 'age_limit': 0,
1969 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1970 'channel': 'The Berkman Klein Center for Internet & Society',
1971 'availability': 'public',
1972 'view_count': int,
1973 'categories': ['Education'],
1974 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1975 'live_status': 'not_live',
1976 'playable_in_embed': True,
1977 'channel_follower_count': int,
1978 'chapters': list,
1979 'uploader': 'The Berkman Klein Center for Internet & Society',
1980 'uploader_id': '@BKCHarvard',
1981 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1982 'timestamp': 1422422076,
1984 'params': {
1985 'skip_download': True,
1989 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1990 'info_dict': {
1991 'id': 'eQcmzGIKrzg',
1992 'ext': 'mp4',
1993 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1994 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1995 'duration': 4060,
1996 'upload_date': '20151120',
1997 'license': 'Creative Commons Attribution license (reuse allowed)',
1998 'playable_in_embed': True,
1999 'tags': 'count:12',
2000 'like_count': int,
2001 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
2002 'age_limit': 0,
2003 'availability': 'public',
2004 'categories': ['News & Politics'],
2005 'channel': 'Bernie Sanders',
2006 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
2007 'view_count': int,
2008 'live_status': 'not_live',
2009 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
2010 'comment_count': int,
2011 'channel_follower_count': int,
2012 'chapters': list,
2013 'uploader': 'Bernie Sanders',
2014 'uploader_url': 'https://www.youtube.com/@BernieSanders',
2015 'uploader_id': '@BernieSanders',
2016 'channel_is_verified': True,
2017 'heatmap': 'count:100',
2018 'timestamp': 1447987198,
2020 'params': {
2021 'skip_download': True,
2025 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
2026 'only_matching': True,
2029 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
2030 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
2031 'only_matching': True,
2034 # Rental video preview
2035 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
2036 'info_dict': {
2037 'id': 'uGpuVWrhIzE',
2038 'ext': 'mp4',
2039 'title': 'Piku - Trailer',
2040 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
2041 'upload_date': '20150811',
2042 'license': 'Standard YouTube License',
2044 'params': {
2045 'skip_download': True,
2047 'skip': 'This video is not available.',
2050 # YouTube Red video with episode data
2051 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
2052 'info_dict': {
2053 'id': 'iqKdEhx-dD4',
2054 'ext': 'mp4',
2055 'title': 'Isolation - Mind Field (Ep 1)',
2056 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
2057 'duration': 2085,
2058 'upload_date': '20170118',
2059 'series': 'Mind Field',
2060 'season_number': 1,
2061 'episode_number': 1,
2062 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
2063 'tags': 'count:12',
2064 'view_count': int,
2065 'availability': 'public',
2066 'age_limit': 0,
2067 'channel': 'Vsauce',
2068 'episode': 'Episode 1',
2069 'categories': ['Entertainment'],
2070 'season': 'Season 1',
2071 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
2072 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
2073 'like_count': int,
2074 'playable_in_embed': True,
2075 'live_status': 'not_live',
2076 'channel_follower_count': int,
2077 'uploader': 'Vsauce',
2078 'uploader_url': 'https://www.youtube.com/@Vsauce',
2079 'uploader_id': '@Vsauce',
2080 'comment_count': int,
2081 'channel_is_verified': True,
2082 'timestamp': 1484761047,
2084 'params': {
2085 'skip_download': True,
2087 'expected_warnings': [
2088 'Skipping DASH manifest',
2092 # The following content has been identified by the YouTube community
2093 # as inappropriate or offensive to some audiences.
2094 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
2095 'info_dict': {
2096 'id': '6SJNVb0GnPI',
2097 'ext': 'mp4',
2098 'title': 'Race Differences in Intelligence',
2099 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
2100 'duration': 965,
2101 'upload_date': '20140124',
2103 'params': {
2104 'skip_download': True,
2106 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
2109 # itag 212
2110 'url': '1t24XAntNCY',
2111 'only_matching': True,
2114 # geo restricted to JP
2115 'url': 'sJL6WA-aGkQ',
2116 'only_matching': True,
2119 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2120 'only_matching': True,
2123 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2124 'only_matching': True,
2127 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2128 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2129 'only_matching': True,
2132 # DRM protected
2133 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2134 'only_matching': True,
2137 # Video with unsupported adaptive stream type formats
2138 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2139 'info_dict': {
2140 'id': 'Z4Vy8R84T1U',
2141 'ext': 'mp4',
2142 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2143 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2144 'duration': 433,
2145 'upload_date': '20130923',
2146 'formats': 'maxcount:10',
2148 'params': {
2149 'skip_download': True,
2150 'youtube_include_dash_manifest': False,
2152 'skip': 'not actual anymore',
2155 # Youtube Music Auto-generated description
2156 # TODO: fix metadata extraction
2157 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2158 'info_dict': {
2159 'id': 'MgNrAu2pzNs',
2160 'ext': 'mp4',
2161 'title': 'Voyeur Girl',
2162 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2163 'upload_date': '20190312',
2164 'artists': ['Stephen'],
2165 'creators': ['Stephen'],
2166 'track': 'Voyeur Girl',
2167 'album': 'it\'s too much love to know my dear',
2168 'release_date': '20190313',
2169 'alt_title': 'Voyeur Girl',
2170 'view_count': int,
2171 'playable_in_embed': True,
2172 'like_count': int,
2173 'categories': ['Music'],
2174 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2175 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2176 'uploader': 'Stephen',
2177 'availability': 'public',
2178 'duration': 169,
2179 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2180 'age_limit': 0,
2181 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2182 'tags': 'count:11',
2183 'live_status': 'not_live',
2184 'channel_follower_count': int,
2186 'params': {
2187 'skip_download': True,
2191 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2192 'only_matching': True,
2195 # invalid -> valid video id redirection
2196 'url': 'DJztXj2GPfl',
2197 'info_dict': {
2198 'id': 'DJztXj2GPfk',
2199 'ext': 'mp4',
2200 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2201 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2202 'upload_date': '20090125',
2203 'artist': 'Panjabi MC',
2204 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2205 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2207 'params': {
2208 'skip_download': True,
2210 'skip': 'Video unavailable',
2213 # empty description results in an empty string
2214 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2215 'info_dict': {
2216 'id': 'x41yOUIvK2k',
2217 'ext': 'mp4',
2218 'title': 'IMG 3456',
2219 'description': '',
2220 'upload_date': '20170613',
2221 'view_count': int,
2222 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2223 'like_count': int,
2224 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2225 'tags': [],
2226 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2227 'availability': 'public',
2228 'age_limit': 0,
2229 'categories': ['Pets & Animals'],
2230 'duration': 7,
2231 'playable_in_embed': True,
2232 'live_status': 'not_live',
2233 'channel': 'l\'Or Vert asbl',
2234 'channel_follower_count': int,
2235 'uploader': 'l\'Or Vert asbl',
2236 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2237 'uploader_id': '@ElevageOrVert',
2238 'timestamp': 1497343210,
2240 'params': {
2241 'skip_download': True,
2245 # with '};' inside yt initial data (see [1])
2246 # see [2] for an example with '};' inside ytInitialPlayerResponse
2247 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2248 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2249 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2250 'info_dict': {
2251 'id': 'CHqg6qOn4no',
2252 'ext': 'mp4',
2253 'title': 'Part 77 Sort a list of simple types in c#',
2254 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2255 'upload_date': '20130831',
2256 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2257 'like_count': int,
2258 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2259 'live_status': 'not_live',
2260 'categories': ['Education'],
2261 'availability': 'public',
2262 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2263 'tags': 'count:12',
2264 'playable_in_embed': True,
2265 'age_limit': 0,
2266 'view_count': int,
2267 'duration': 522,
2268 'channel': 'kudvenkat',
2269 'comment_count': int,
2270 'channel_follower_count': int,
2271 'chapters': list,
2272 'uploader': 'kudvenkat',
2273 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2274 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2275 'channel_is_verified': True,
2276 'heatmap': 'count:100',
2277 'timestamp': 1377976349,
2279 'params': {
2280 'skip_download': True,
2284 # another example of '};' in ytInitialData
2285 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2286 'only_matching': True,
2289 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2290 'only_matching': True,
2293 # https://github.com/ytdl-org/youtube-dl/pull/28094
2294 'url': 'OtqTfy26tG0',
2295 'info_dict': {
2296 'id': 'OtqTfy26tG0',
2297 'ext': 'mp4',
2298 'title': 'Burn Out',
2299 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2300 'upload_date': '20141120',
2301 'artist': 'The Cinematic Orchestra',
2302 'track': 'Burn Out',
2303 'album': 'Every Day',
2304 'like_count': int,
2305 'live_status': 'not_live',
2306 'alt_title': 'Burn Out',
2307 'duration': 614,
2308 'age_limit': 0,
2309 'view_count': int,
2310 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2311 'creator': 'The Cinematic Orchestra',
2312 'channel': 'The Cinematic Orchestra',
2313 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2314 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2315 'availability': 'public',
2316 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2317 'categories': ['Music'],
2318 'playable_in_embed': True,
2319 'channel_follower_count': int,
2320 'uploader': 'The Cinematic Orchestra',
2321 'comment_count': int,
2323 'params': {
2324 'skip_download': True,
2328 # controversial video, only works with bpctr when authenticated with cookies
2329 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2330 'only_matching': True,
2333 # controversial video, requires bpctr/contentCheckOk
2334 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2335 'info_dict': {
2336 'id': 'SZJvDhaSDnc',
2337 'ext': 'mp4',
2338 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2339 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2340 'upload_date': '20140716',
2341 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2342 'duration': 170,
2343 'categories': ['News & Politics'],
2344 'view_count': int,
2345 'channel': 'CBS Mornings',
2346 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2347 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2348 'age_limit': 18,
2349 'availability': 'needs_auth',
2350 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2351 'like_count': int,
2352 'live_status': 'not_live',
2353 'playable_in_embed': True,
2354 'channel_follower_count': int,
2355 'uploader': 'CBS Mornings',
2356 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2357 'uploader_id': '@CBSMornings',
2358 'comment_count': int,
2359 'channel_is_verified': True,
2360 'timestamp': 1405513526,
2362 'skip': 'Age-restricted; requires authentication',
2365 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2366 'url': 'cBvYw8_A0vQ',
2367 'info_dict': {
2368 'id': 'cBvYw8_A0vQ',
2369 'ext': 'mp4',
2370 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2371 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2372 'upload_date': '20201120',
2373 'duration': 1456,
2374 'categories': ['Travel & Events'],
2375 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2376 'view_count': int,
2377 'channel': 'Walk around Japan',
2378 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2379 'thumbnail': 'https://i.ytimg.com/vi/cBvYw8_A0vQ/hqdefault.jpg',
2380 'age_limit': 0,
2381 'availability': 'public',
2382 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2383 'live_status': 'not_live',
2384 'playable_in_embed': True,
2385 'channel_follower_count': int,
2386 'uploader': 'Walk around Japan',
2387 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2388 'uploader_id': '@walkaroundjapan7124',
2389 'timestamp': 1605884416,
2391 'params': {
2392 'skip_download': True,
2394 }, {
2395 # Has multiple audio streams
2396 'url': 'WaOKSUlf4TM',
2397 'only_matching': True,
2398 }, {
2399 # Requires Premium: has format 141 when requested using YTM url
2400 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2401 'only_matching': True,
2402 }, {
2403 # multiple subtitles with same lang_code
2404 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2405 'only_matching': True,
2406 }, {
2407 # Force use android client fallback
2408 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2409 'info_dict': {
2410 'id': 'YOelRv7fMxY',
2411 'title': 'DIGGING A SECRET TUNNEL Part 1',
2412 'ext': '3gp',
2413 'upload_date': '20210624',
2414 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2415 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2416 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2417 'duration': 596,
2418 'categories': ['Entertainment'],
2419 'view_count': int,
2420 'channel': 'colinfurze',
2421 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2422 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2423 'age_limit': 0,
2424 'availability': 'public',
2425 'like_count': int,
2426 'live_status': 'not_live',
2427 'playable_in_embed': True,
2428 'channel_follower_count': int,
2429 'chapters': list,
2430 'uploader': 'colinfurze',
2431 'uploader_url': 'https://www.youtube.com/@colinfurze',
2432 'uploader_id': '@colinfurze',
2433 'comment_count': int,
2434 'channel_is_verified': True,
2435 'heatmap': 'count:100',
2437 'params': {
2438 'format': '17', # 3gp format available on android
2439 'extractor_args': {'youtube': {'player_client': ['android']}},
2441 'skip': 'android client broken',
2444 # Skip download of additional client configs (remix client config in this case)
2445 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2446 'only_matching': True,
2447 'params': {
2448 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2450 }, {
2451 # shorts
2452 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2453 'only_matching': True,
2454 }, {
2455 'note': 'Storyboards',
2456 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2457 'info_dict': {
2458 'id': '5KLPxDtMqe8',
2459 'ext': 'mhtml',
2460 'format_id': 'sb0',
2461 'title': 'Your Brain is Plastic',
2462 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2463 'upload_date': '20140324',
2464 'like_count': int,
2465 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2466 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2467 'view_count': int,
2468 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2469 'playable_in_embed': True,
2470 'tags': 'count:12',
2471 'availability': 'public',
2472 'channel': 'SciShow',
2473 'live_status': 'not_live',
2474 'duration': 248,
2475 'categories': ['Education'],
2476 'age_limit': 0,
2477 'channel_follower_count': int,
2478 'chapters': list,
2479 'uploader': 'SciShow',
2480 'uploader_url': 'https://www.youtube.com/@SciShow',
2481 'uploader_id': '@SciShow',
2482 'comment_count': int,
2483 'channel_is_verified': True,
2484 'heatmap': 'count:100',
2485 'timestamp': 1395685455,
2486 }, 'params': {'format': 'mhtml', 'skip_download': True},
2487 }, {
2488 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2489 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2490 'info_dict': {
2491 'id': '2NUZ8W2llS4',
2492 'ext': 'mp4',
2493 'title': 'The NP that test your phone performance 🙂',
2494 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2495 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2496 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2497 'duration': 21,
2498 'view_count': int,
2499 'age_limit': 0,
2500 'categories': ['Gaming'],
2501 'tags': 'count:23',
2502 'playable_in_embed': True,
2503 'live_status': 'not_live',
2504 'upload_date': '20220103',
2505 'like_count': int,
2506 'availability': 'public',
2507 'channel': 'Leon Nguyen',
2508 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2509 'comment_count': int,
2510 'channel_follower_count': int,
2511 'uploader': 'Leon Nguyen',
2512 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2513 'uploader_id': '@LeonNguyen',
2514 'heatmap': 'count:100',
2515 'timestamp': 1641170939,
2517 }, {
2518 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2519 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2520 'info_dict': {
2521 'id': 'mzZzzBU6lrM',
2522 'ext': 'mp4',
2523 'title': 'I Met GeorgeNotFound In Real Life...',
2524 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2525 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2526 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2527 'duration': 955,
2528 'view_count': int,
2529 'age_limit': 0,
2530 'categories': ['Entertainment'],
2531 'tags': 'count:26',
2532 'playable_in_embed': True,
2533 'live_status': 'not_live',
2534 'release_timestamp': 1641172509,
2535 'release_date': '20220103',
2536 'upload_date': '20220103',
2537 'like_count': int,
2538 'availability': 'public',
2539 'channel': 'Quackity',
2540 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2541 'channel_follower_count': int,
2542 'uploader': 'Quackity',
2543 'uploader_id': '@Quackity',
2544 'uploader_url': 'https://www.youtube.com/@Quackity',
2545 'comment_count': int,
2546 'channel_is_verified': True,
2547 'heatmap': 'count:100',
2548 'timestamp': 1641172509,
2551 { # continuous livestream.
2552 # Upload date was 2022-07-12T05:12:29-07:00, while stream start is 2022-07-12T15:59:30+00:00
2553 'url': 'https://www.youtube.com/watch?v=jfKfPfyJRdk',
2554 'info_dict': {
2555 'id': 'jfKfPfyJRdk',
2556 'ext': 'mp4',
2557 'channel_id': 'UCSJ4gkVC6NrvII8umztf0Ow',
2558 'like_count': int,
2559 'uploader': 'Lofi Girl',
2560 'categories': ['Music'],
2561 'concurrent_view_count': int,
2562 'playable_in_embed': True,
2563 'timestamp': 1657627949,
2564 'release_date': '20220712',
2565 'channel_url': 'https://www.youtube.com/channel/UCSJ4gkVC6NrvII8umztf0Ow',
2566 'description': 'md5:13a6f76df898f5674f9127139f3df6f7',
2567 'age_limit': 0,
2568 'thumbnail': 'https://i.ytimg.com/vi/jfKfPfyJRdk/maxresdefault.jpg',
2569 'release_timestamp': 1657641570,
2570 'uploader_url': 'https://www.youtube.com/@LofiGirl',
2571 'channel_follower_count': int,
2572 'channel_is_verified': True,
2573 'title': r're:^lofi hip hop radio 📚 - beats to relax/study to',
2574 'view_count': int,
2575 'live_status': 'is_live',
2576 'tags': 'count:32',
2577 'channel': 'Lofi Girl',
2578 'availability': 'public',
2579 'upload_date': '20220712',
2580 'uploader_id': '@LofiGirl',
2582 'params': {'skip_download': True},
2583 }, {
2584 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2585 'info_dict': {
2586 'id': 'tjjjtzRLHvA',
2587 'ext': 'mp4',
2588 'title': 'ハッシュタグ無し };if window.ytcsi',
2589 'upload_date': '20220323',
2590 'like_count': int,
2591 'availability': 'unlisted',
2592 'channel': 'Lesmiscore',
2593 'thumbnail': r're:^https?://.*\.jpg',
2594 'age_limit': 0,
2595 'categories': ['Music'],
2596 'view_count': int,
2597 'description': '',
2598 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2599 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2600 'live_status': 'not_live',
2601 'playable_in_embed': True,
2602 'channel_follower_count': int,
2603 'duration': 6,
2604 'tags': [],
2605 'uploader_id': '@lesmiscore',
2606 'uploader': 'Lesmiscore',
2607 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2608 'timestamp': 1648005313,
2610 }, {
2611 # Prefer primary title+description language metadata by default
2612 # Do not prefer translated description if primary is empty
2613 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2614 'info_dict': {
2615 'id': 'el3E4MbxRqQ',
2616 'ext': 'mp4',
2617 'title': 'dlp test video 2 - primary sv no desc',
2618 'description': '',
2619 'channel': 'cole-dlp-test-acc',
2620 'tags': [],
2621 'view_count': int,
2622 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2623 'like_count': int,
2624 'playable_in_embed': True,
2625 'availability': 'unlisted',
2626 'thumbnail': r're:^https?://.*\.jpg',
2627 'age_limit': 0,
2628 'duration': 5,
2629 'live_status': 'not_live',
2630 'upload_date': '20220908',
2631 'categories': ['People & Blogs'],
2632 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2633 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2634 'uploader_id': '@coletdjnz',
2635 'uploader': 'cole-dlp-test-acc',
2636 'timestamp': 1662677394,
2638 'params': {'skip_download': True},
2639 }, {
2640 # Extractor argument: prefer translated title+description
2641 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2642 'info_dict': {
2643 'id': 'gHKT4uU8Zng',
2644 'ext': 'mp4',
2645 'channel': 'cole-dlp-test-acc',
2646 'tags': [],
2647 'duration': 5,
2648 'live_status': 'not_live',
2649 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2650 'upload_date': '20220729',
2651 'view_count': int,
2652 'categories': ['People & Blogs'],
2653 'thumbnail': r're:^https?://.*\.jpg',
2654 'title': 'dlp test video title translated (fr)',
2655 'availability': 'public',
2656 'age_limit': 0,
2657 'description': 'dlp test video description translated (fr)',
2658 'playable_in_embed': True,
2659 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2660 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2661 'uploader_id': '@coletdjnz',
2662 'uploader': 'cole-dlp-test-acc',
2663 'timestamp': 1659073275,
2664 'like_count': int,
2666 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2667 'expected_warnings': [r'Preferring "fr" translated fields'],
2668 }, {
2669 'note': '6 channel audio',
2670 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2671 'only_matching': True,
2672 }, {
2673 'note': 'Multiple HLS formats with same itag',
2674 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2675 'info_dict': {
2676 'id': 'kX3nB4PpJko',
2677 'ext': 'mp4',
2678 'categories': ['Entertainment'],
2679 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2680 'live_status': 'not_live',
2681 'duration': 937,
2682 'channel_follower_count': int,
2683 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2684 'title': 'Last To Take Hand Off Jet, Keeps It!',
2685 'channel': 'MrBeast',
2686 'playable_in_embed': True,
2687 'view_count': int,
2688 'upload_date': '20221112',
2689 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2690 'age_limit': 0,
2691 'availability': 'public',
2692 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2693 'like_count': int,
2694 'tags': [],
2695 'uploader': 'MrBeast',
2696 'uploader_url': 'https://www.youtube.com/@MrBeast',
2697 'uploader_id': '@MrBeast',
2698 'comment_count': int,
2699 'channel_is_verified': True,
2700 'heatmap': 'count:100',
2702 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2703 }, {
2704 'note': 'Audio formats with Dynamic Range Compression',
2705 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2706 'info_dict': {
2707 'id': 'Tq92D6wQ1mg',
2708 'ext': 'webm',
2709 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2710 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2711 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2712 'channel_follower_count': int,
2713 'description': 'md5:17eccca93a786d51bc67646756894066',
2714 'upload_date': '20191228',
2715 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2716 'playable_in_embed': True,
2717 'like_count': int,
2718 'categories': ['Entertainment'],
2719 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2720 'age_limit': 18,
2721 'channel': 'Projekt Melody',
2722 'view_count': int,
2723 'availability': 'needs_auth',
2724 'comment_count': int,
2725 'live_status': 'not_live',
2726 'duration': 106,
2727 'uploader': 'Projekt Melody',
2728 'uploader_id': '@ProjektMelody',
2729 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2730 'timestamp': 1577508724,
2732 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2733 'skip': 'Age-restricted; requires authentication',
2736 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2737 'info_dict': {
2738 'id': 'qVv6vCqciTM',
2739 'ext': 'mp4',
2740 'age_limit': 0,
2741 'comment_count': int,
2742 'chapters': 'count:13',
2743 'upload_date': '20221223',
2744 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2745 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2746 'like_count': int,
2747 'release_date': '20221223',
2748 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2749 'title': '【 #インターネット女クリスマス 】3Dで歌ってはしゃぐインターネットの女たち【月ノ美兎/名取さな】',
2750 'view_count': int,
2751 'playable_in_embed': True,
2752 'duration': 4438,
2753 'availability': 'public',
2754 'channel_follower_count': int,
2755 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2756 'categories': ['Entertainment'],
2757 'live_status': 'was_live',
2758 'release_timestamp': 1671793345,
2759 'channel': 'さなちゃんねる',
2760 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2761 'uploader': 'さなちゃんねる',
2762 'uploader_url': 'https://www.youtube.com/@sana_natori',
2763 'uploader_id': '@sana_natori',
2764 'channel_is_verified': True,
2765 'heatmap': 'count:100',
2766 'timestamp': 1671798112,
2770 # Fallbacks when webpage and web client is unavailable
2771 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2772 'info_dict': {
2773 'id': 'wSSmNUl9Snw',
2774 'ext': 'mp4',
2775 # 'categories': ['Science & Technology'],
2776 'view_count': int,
2777 'chapters': 'count:2',
2778 'channel': 'Scott Manley',
2779 'like_count': int,
2780 'age_limit': 0,
2781 # 'availability': 'public',
2782 'channel_follower_count': int,
2783 'live_status': 'not_live',
2784 'upload_date': '20170831',
2785 'duration': 682,
2786 'tags': 'count:8',
2787 'uploader_url': 'https://www.youtube.com/@scottmanley',
2788 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2789 'uploader': 'Scott Manley',
2790 'uploader_id': '@scottmanley',
2791 'title': 'The Computer Hack That Saved Apollo 14',
2792 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2793 'thumbnail': r're:^https?://.*\.webp',
2794 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2795 'playable_in_embed': True,
2796 'comment_count': int,
2797 'channel_is_verified': True,
2798 'heatmap': 'count:100',
2800 'params': {
2801 'extractor_args': {'youtube': {'player_client': ['ios'], 'player_skip': ['webpage']}},
2805 # uploader_id has non-ASCII characters that are percent-encoded in YT's JSON
2806 'url': 'https://www.youtube.com/shorts/18NGQq7p3LY',
2807 'info_dict': {
2808 'id': '18NGQq7p3LY',
2809 'ext': 'mp4',
2810 'title': '아이브 이서 장원영 리즈 삐끼삐끼 챌린지',
2811 'description': '',
2812 'uploader': 'ㅇㅇ',
2813 'uploader_id': '@으아-v1k',
2814 'uploader_url': 'https://www.youtube.com/@으아-v1k',
2815 'channel': 'ㅇㅇ',
2816 'channel_id': 'UCC25oTm2J7ZVoi5TngOHg9g',
2817 'channel_url': 'https://www.youtube.com/channel/UCC25oTm2J7ZVoi5TngOHg9g',
2818 'thumbnail': r're:https?://.+/.+\.jpg',
2819 'playable_in_embed': True,
2820 'age_limit': 0,
2821 'duration': 3,
2822 'timestamp': 1724306170,
2823 'upload_date': '20240822',
2824 'availability': 'public',
2825 'live_status': 'not_live',
2826 'view_count': int,
2827 'like_count': int,
2828 'channel_follower_count': int,
2829 'categories': ['People & Blogs'],
2830 'tags': [],
2835 _WEBPAGE_TESTS = [
2836 # YouTube <object> embed
2838 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2839 'md5': '873c81d308b979f0e23ee7e620b312a3',
2840 'info_dict': {
2841 'id': 'msN87y-iEx0',
2842 'ext': 'mp4',
2843 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2844 'upload_date': '20080526',
2845 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2846 'age_limit': 0,
2847 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2848 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2849 'playable_in_embed': True,
2850 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2851 'like_count': int,
2852 'comment_count': int,
2853 'channel': 'Christopher Sykes',
2854 'live_status': 'not_live',
2855 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2856 'availability': 'public',
2857 'duration': 195,
2858 'view_count': int,
2859 'categories': ['Science & Technology'],
2860 'channel_follower_count': int,
2861 'uploader': 'Christopher Sykes',
2862 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2863 'uploader_id': '@ChristopherSykesDocumentaries',
2864 'heatmap': 'count:100',
2865 'timestamp': 1211825920,
2867 'params': {
2868 'skip_download': True,
2873 @classmethod
2874 def suitable(cls, url):
2875 from ..utils import parse_qs
2877 qs = parse_qs(url)
2878 if qs.get('list', [None])[0]:
2879 return False
2880 return super().suitable(url)
2882 def __init__(self, *args, **kwargs):
2883 super().__init__(*args, **kwargs)
2884 self._code_cache = {}
2885 self._player_cache = {}
2887 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2888 lock = threading.Lock()
2889 start_time = time.time()
2890 formats = [f for f in formats if f.get('is_from_start')]
2892 def refetch_manifest(format_id, delay):
2893 nonlocal formats, start_time, is_live
2894 if time.time() <= start_time + delay:
2895 return
2897 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2898 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2899 microformats = traverse_obj(
2900 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2901 expected_type=dict)
2902 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2903 is_live = live_status == 'is_live'
2904 start_time = time.time()
2906 def mpd_feed(format_id, delay):
2908 @returns (manifest_url, manifest_stream_number, is_live) or None
2910 for retry in self.RetryManager(fatal=False):
2911 with lock:
2912 refetch_manifest(format_id, delay)
2914 f = next((f for f in formats if f['format_id'] == format_id), None)
2915 if not f:
2916 if not is_live:
2917 retry.error = f'{video_id}: Video is no longer live'
2918 else:
2919 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2920 continue
2921 return f['manifest_url'], f['manifest_stream_number'], is_live
2922 return None
2924 for f in formats:
2925 f['is_live'] = is_live
2926 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2927 live_start_time, mpd_feed, not is_live and f.copy())
2928 if is_live:
2929 f['fragments'] = gen
2930 f['protocol'] = 'http_dash_segments_generator'
2931 else:
2932 f['fragments'] = LazyList(gen({}))
2933 del f['is_from_start']
2935 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2936 FETCH_SPAN, MAX_DURATION = 5, 432000
2938 mpd_url, stream_number, is_live = None, None, True
2940 begin_index = 0
2941 download_start_time = ctx.get('start') or time.time()
2943 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2944 if lack_early_segments:
2945 self.report_warning(bug_reports_message(
2946 'Starting download from the last 120 hours of the live stream since '
2947 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2948 lack_early_segments = True
2950 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2951 fragments, fragment_base_url = None, None
2953 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2954 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2955 # Obtain from MPD's maximum seq value
2956 old_mpd_url = mpd_url
2957 last_error = ctx.pop('last_error', None)
2958 expire_fast = immediate or (last_error and isinstance(last_error, HTTPError) and last_error.status == 403)
2959 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2960 or (mpd_url, stream_number, False))
2961 if not refresh_sequence:
2962 if expire_fast and not is_live:
2963 return False, last_seq
2964 elif old_mpd_url == mpd_url:
2965 return True, last_seq
2966 if manifestless_orig_fmt:
2967 fmt_info = manifestless_orig_fmt
2968 else:
2969 try:
2970 fmts, _ = self._extract_mpd_formats_and_subtitles(
2971 mpd_url, None, note=False, errnote=False, fatal=False)
2972 except ExtractorError:
2973 fmts = None
2974 if not fmts:
2975 no_fragment_score += 2
2976 return False, last_seq
2977 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2978 fragments = fmt_info['fragments']
2979 fragment_base_url = fmt_info['fragment_base_url']
2980 assert fragment_base_url
2982 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2983 return True, _last_seq
2985 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2986 while is_live:
2987 fetch_time = time.time()
2988 if no_fragment_score > 30:
2989 return
2990 if last_segment_url:
2991 # Obtain from "X-Head-Seqnum" header value from each segment
2992 try:
2993 urlh = self._request_webpage(
2994 last_segment_url, None, note=False, errnote=False, fatal=False)
2995 except ExtractorError:
2996 urlh = None
2997 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2998 if last_seq is None:
2999 no_fragment_score += 2
3000 last_segment_url = None
3001 continue
3002 else:
3003 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
3004 no_fragment_score += 2
3005 if not should_continue:
3006 continue
3008 if known_idx > last_seq:
3009 last_segment_url = None
3010 continue
3012 last_seq += 1
3014 if begin_index < 0 and known_idx < 0:
3015 # skip from the start when it's negative value
3016 known_idx = last_seq + begin_index
3017 if lack_early_segments:
3018 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
3019 try:
3020 for idx in range(known_idx, last_seq):
3021 # do not update sequence here or you'll get skipped some part of it
3022 should_continue, _ = _extract_sequence_from_mpd(False, False)
3023 if not should_continue:
3024 known_idx = idx - 1
3025 raise ExtractorError('breaking out of outer loop')
3026 last_segment_url = urljoin(fragment_base_url, f'sq/{idx}')
3027 yield {
3028 'url': last_segment_url,
3029 'fragment_count': last_seq,
3031 if known_idx == last_seq:
3032 no_fragment_score += 5
3033 else:
3034 no_fragment_score = 0
3035 known_idx = last_seq
3036 except ExtractorError:
3037 continue
3039 if manifestless_orig_fmt:
3040 # Stop at the first iteration if running for post-live manifestless;
3041 # fragment count no longer increase since it starts
3042 break
3044 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
3046 def _extract_player_url(self, *ytcfgs, webpage=None):
3047 player_url = traverse_obj(
3048 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
3049 get_all=False, expected_type=str)
3050 if not player_url:
3051 return
3052 return urljoin('https://www.youtube.com', player_url)
3054 def _download_player_url(self, video_id, fatal=False):
3055 res = self._download_webpage(
3056 'https://www.youtube.com/iframe_api',
3057 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
3058 if res:
3059 player_version = self._search_regex(
3060 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
3061 if player_version:
3062 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
3064 def _signature_cache_id(self, example_sig):
3065 """ Return a string representation of a signature """
3066 return '.'.join(str(len(part)) for part in example_sig.split('.'))
3068 @classmethod
3069 def _extract_player_info(cls, player_url):
3070 for player_re in cls._PLAYER_INFO_RE:
3071 id_m = re.search(player_re, player_url)
3072 if id_m:
3073 break
3074 else:
3075 raise ExtractorError(f'Cannot identify player {player_url!r}')
3076 return id_m.group('id')
3078 def _load_player(self, video_id, player_url, fatal=True):
3079 player_id = self._extract_player_info(player_url)
3080 if player_id not in self._code_cache:
3081 code = self._download_webpage(
3082 player_url, video_id, fatal=fatal,
3083 note='Downloading player ' + player_id,
3084 errnote=f'Download of {player_url} failed')
3085 if code:
3086 self._code_cache[player_id] = code
3087 return self._code_cache.get(player_id)
3089 def _extract_signature_function(self, video_id, player_url, example_sig):
3090 player_id = self._extract_player_info(player_url)
3092 # Read from filesystem cache
3093 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
3094 assert os.path.basename(func_id) == func_id
3096 self.write_debug(f'Extracting signature function {func_id}')
3097 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
3099 if not cache_spec:
3100 code = self._load_player(video_id, player_url)
3101 if code:
3102 res = self._parse_sig_js(code)
3103 test_string = ''.join(map(chr, range(len(example_sig))))
3104 cache_spec = [ord(c) for c in res(test_string)]
3105 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
3107 return lambda s: ''.join(s[i] for i in cache_spec)
3109 def _print_sig_code(self, func, example_sig):
3110 if not self.get_param('youtube_print_sig_code'):
3111 return
3113 def gen_sig_code(idxs):
3114 def _genslice(start, end, step):
3115 starts = '' if start == 0 else str(start)
3116 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
3117 steps = '' if step == 1 else (':%d' % step)
3118 return f's[{starts}{ends}{steps}]'
3120 step = None
3121 # Quelch pyflakes warnings - start will be set when step is set
3122 start = '(Never used)'
3123 for i, prev in zip(idxs[1:], idxs[:-1]):
3124 if step is not None:
3125 if i - prev == step:
3126 continue
3127 yield _genslice(start, prev, step)
3128 step = None
3129 continue
3130 if i - prev in [-1, 1]:
3131 step = i - prev
3132 start = prev
3133 continue
3134 else:
3135 yield 's[%d]' % prev
3136 if step is None:
3137 yield 's[%d]' % i
3138 else:
3139 yield _genslice(start, i, step)
3141 test_string = ''.join(map(chr, range(len(example_sig))))
3142 cache_res = func(test_string)
3143 cache_spec = [ord(c) for c in cache_res]
3144 expr_code = ' + '.join(gen_sig_code(cache_spec))
3145 signature_id_tuple = '({})'.format(', '.join(str(len(p)) for p in example_sig.split('.')))
3146 code = (f'if tuple(len(p) for p in s.split(\'.\')) == {signature_id_tuple}:\n'
3147 f' return {expr_code}\n')
3148 self.to_screen('Extracted signature function:\n' + code)
3150 def _parse_sig_js(self, jscode):
3151 # Examples where `sig` is funcname:
3152 # sig=function(a){a=a.split(""); ... ;return a.join("")};
3153 # ;c&&(c=sig(decodeURIComponent(c)),a.set(b,encodeURIComponent(c)));return a};
3154 # {var l=f,m=h.sp,n=sig(decodeURIComponent(h.s));l.set(m,encodeURIComponent(n))}
3155 # sig=function(J){J=J.split(""); ... ;return J.join("")};
3156 # ;N&&(N=sig(decodeURIComponent(N)),J.set(R,encodeURIComponent(N)));return J};
3157 # {var H=u,k=f.sp,v=sig(decodeURIComponent(f.s));H.set(k,encodeURIComponent(v))}
3158 funcname = self._search_regex(
3159 (r'\b(?P<var>[a-zA-Z0-9_$]+)&&\((?P=var)=(?P<sig>[a-zA-Z0-9_$]{2,})\(decodeURIComponent\((?P=var)\)\)',
3160 r'(?P<sig>[a-zA-Z0-9_$]+)\s*=\s*function\(\s*(?P<arg>[a-zA-Z0-9_$]+)\s*\)\s*{\s*(?P=arg)\s*=\s*(?P=arg)\.split\(\s*""\s*\)\s*;\s*[^}]+;\s*return\s+(?P=arg)\.join\(\s*""\s*\)',
3161 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}\(a,\d+\))?',
3162 # Old patterns
3163 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3164 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3165 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3166 # Obsolete patterns
3167 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3168 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3169 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3170 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3171 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3172 jscode, 'Initial JS player signature function name', group='sig')
3174 jsi = JSInterpreter(jscode)
3175 initial_function = jsi.extract_function(funcname)
3176 return lambda s: initial_function([s])
3178 def _cached(self, func, *cache_id):
3179 def inner(*args, **kwargs):
3180 if cache_id not in self._player_cache:
3181 try:
3182 self._player_cache[cache_id] = func(*args, **kwargs)
3183 except ExtractorError as e:
3184 self._player_cache[cache_id] = e
3185 except Exception as e:
3186 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3188 ret = self._player_cache[cache_id]
3189 if isinstance(ret, Exception):
3190 raise ret
3191 return ret
3192 return inner
3194 def _decrypt_signature(self, s, video_id, player_url):
3195 """Turn the encrypted s field into a working signature"""
3196 extract_sig = self._cached(
3197 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3198 func = extract_sig(video_id, player_url, s)
3199 self._print_sig_code(func, s)
3200 return func(s)
3202 def _decrypt_nsig(self, s, video_id, player_url):
3203 """Turn the encrypted n field into a working signature"""
3204 if player_url is None:
3205 raise ExtractorError('Cannot decrypt nsig without player_url')
3206 player_url = urljoin('https://www.youtube.com', player_url)
3208 try:
3209 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3210 except ExtractorError as e:
3211 raise ExtractorError('Unable to extract nsig function code', cause=e)
3212 if self.get_param('youtube_print_sig_code'):
3213 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3215 try:
3216 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3217 ret = extract_nsig(jsi, func_code)(s)
3218 except JSInterpreter.Exception as e:
3219 try:
3220 jsi = PhantomJSwrapper(self, timeout=5000)
3221 except ExtractorError:
3222 raise e
3223 self.report_warning(
3224 f'Native nsig extraction failed: Trying with PhantomJS\n'
3225 f' n = {s} ; player = {player_url}', video_id)
3226 self.write_debug(e, only_once=True)
3228 args, func_body = func_code
3229 ret = jsi.execute(
3230 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3231 video_id=video_id, note='Executing signature code').strip()
3233 self.write_debug(f'Decrypted nsig {s} => {ret}')
3234 return ret
3236 def _extract_n_function_name(self, jscode, player_url=None):
3237 # Examples (with placeholders nfunc, narray, idx):
3238 # * .get("n"))&&(b=nfunc(b)
3239 # * .get("n"))&&(b=narray[idx](b)
3240 # * b=String.fromCharCode(110),c=a.get(b))&&c=narray[idx](c)
3241 # * a.D&&(b="nn"[+a.D],c=a.get(b))&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("")
3242 # * a.D&&(PL(a),b=a.j.n||null)&&(b=narray[0](b),a.set("n",b),narray.length||nfunc("")
3243 # * a.D&&(b="nn"[+a.D],vL(a),c=a.j[b]||null)&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("")
3244 # * J.J="";J.url="";J.Z&&(R="nn"[+J.Z],mW(J),N=J.K[R]||null)&&(N=narray[idx](N),J.set(R,N))}};
3245 funcname, idx = self._search_regex(
3246 r'''(?x)
3248 \.get\("n"\)\)&&\(b=|
3250 b=String\.fromCharCode\(110\)|
3251 (?P<str_idx>[a-zA-Z0-9_$.]+)&&\(b="nn"\[\+(?P=str_idx)\]
3254 ,[a-zA-Z0-9_$]+\(a\))?,c=a\.
3256 get\(b\)|
3257 [a-zA-Z0-9_$]+\[b\]\|\|null
3258 )\)&&\(c=|
3259 \b(?P<var>[a-zA-Z0-9_$]+)=
3260 )(?P<nfunc>[a-zA-Z0-9_$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z]\)
3261 (?(var),[a-zA-Z0-9_$]+\.set\((?:"n+"|[a-zA-Z0-9_$]+)\,(?P=var)\))''',
3262 jscode, 'n function name', group=('nfunc', 'idx'), default=(None, None))
3263 if not funcname:
3264 self.report_warning(join_nonempty(
3265 'Falling back to generic n function search',
3266 player_url and f' player = {player_url}', delim='\n'))
3267 return self._search_regex(
3268 r'''(?xs)
3269 ;\s*(?P<name>[a-zA-Z0-9_$]+)\s*=\s*function\([a-zA-Z0-9_$]+\)
3270 \s*\{(?:(?!};).)+?return\s*(?P<q>["'])[\w-]+_w8_(?P=q)\s*\+\s*[a-zA-Z0-9_$]+''',
3271 jscode, 'Initial JS player n function name', group='name')
3272 elif not idx:
3273 return funcname
3275 return json.loads(js_to_json(self._search_regex(
3276 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
3277 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3279 def _fixup_n_function_code(self, argnames, code):
3280 return argnames, re.sub(
3281 rf';\s*if\s*\(\s*typeof\s+[a-zA-Z0-9_$]+\s*===?\s*(["\'])undefined\1\s*\)\s*return\s+{argnames[0]};',
3282 ';', code)
3284 def _extract_n_function_code(self, video_id, player_url):
3285 player_id = self._extract_player_info(player_url)
3286 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2024.07.09')
3287 jscode = func_code or self._load_player(video_id, player_url)
3288 jsi = JSInterpreter(jscode)
3290 if func_code:
3291 return jsi, player_id, func_code
3293 func_name = self._extract_n_function_name(jscode, player_url=player_url)
3295 # XXX: Workaround for the `typeof` gotcha
3296 func_code = self._fixup_n_function_code(*jsi.extract_function_code(func_name))
3298 self.cache.store('youtube-nsig', player_id, func_code)
3299 return jsi, player_id, func_code
3301 def _extract_n_function_from_code(self, jsi, func_code):
3302 func = jsi.extract_function_from_code(*func_code)
3304 def extract_nsig(s):
3305 try:
3306 ret = func([s])
3307 except JSInterpreter.Exception:
3308 raise
3309 except Exception as e:
3310 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3312 if ret.startswith('enhanced_except_') or ret.endswith(s):
3313 raise JSInterpreter.Exception('Signature function returned an exception')
3314 return ret
3316 return extract_nsig
3318 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3320 Extract signatureTimestamp (sts)
3321 Required to tell API what sig/player version is in use.
3323 sts = None
3324 if isinstance(ytcfg, dict):
3325 sts = int_or_none(ytcfg.get('STS'))
3327 if not sts:
3328 # Attempt to extract from player
3329 if player_url is None:
3330 error_msg = 'Cannot extract signature timestamp without player_url.'
3331 if fatal:
3332 raise ExtractorError(error_msg)
3333 self.report_warning(error_msg)
3334 return
3335 code = self._load_player(video_id, player_url, fatal=fatal)
3336 if code:
3337 sts = int_or_none(self._search_regex(
3338 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3339 'JS player signature timestamp', group='sts', fatal=fatal))
3340 return sts
3342 def _mark_watched(self, video_id, player_responses):
3343 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3344 label = 'fully ' if is_full else ''
3345 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3346 expected_type=url_or_none)
3347 if not url:
3348 self.report_warning(f'Unable to mark {label}watched')
3349 return
3350 parsed_url = urllib.parse.urlparse(url)
3351 qs = urllib.parse.parse_qs(parsed_url.query)
3353 # cpn generation algorithm is reverse engineered from base.js.
3354 # In fact it works even with dummy cpn.
3355 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3356 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(16))
3358 # # more consistent results setting it to right before the end
3359 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3361 qs.update({
3362 'ver': ['2'],
3363 'cpn': [cpn],
3364 'cmt': video_length,
3365 'el': 'detailpage', # otherwise defaults to "shorts"
3368 if is_full:
3369 # these seem to mark watchtime "history" in the real world
3370 # they're required, so send in a single value
3371 qs.update({
3372 'st': 0,
3373 'et': video_length,
3376 url = urllib.parse.urlunparse(
3377 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3379 self._download_webpage(
3380 url, video_id, f'Marking {label}watched',
3381 'Unable to mark watched', fatal=False)
3383 @classmethod
3384 def _extract_from_webpage(cls, url, webpage):
3385 # Invidious Instances
3386 # https://github.com/yt-dlp/yt-dlp/issues/195
3387 # https://github.com/iv-org/invidious/pull/1730
3388 mobj = re.search(
3389 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3390 webpage)
3391 if mobj:
3392 yield cls.url_result(mobj.group('url'), cls)
3393 raise cls.StopExtraction
3395 yield from super()._extract_from_webpage(url, webpage)
3397 # lazyYT YouTube embed
3398 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3399 yield cls.url_result(unescapeHTML(id_), cls, id_)
3401 # Wordpress "YouTube Video Importer" plugin
3402 for m in re.findall(r'''(?x)<div[^>]+
3403 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3404 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3405 yield cls.url_result(m[-1], cls, m[-1])
3407 @classmethod
3408 def extract_id(cls, url):
3409 video_id = cls.get_temp_id(url)
3410 if not video_id:
3411 raise ExtractorError(f'Invalid URL: {url}')
3412 return video_id
3414 def _extract_chapters_from_json(self, data, duration):
3415 chapter_list = traverse_obj(
3416 data, (
3417 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3418 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters',
3419 ), expected_type=list)
3421 return self._extract_chapters_helper(
3422 chapter_list,
3423 start_function=lambda chapter: float_or_none(
3424 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3425 title_function=lambda chapter: traverse_obj(
3426 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3427 duration=duration)
3429 def _extract_chapters_from_engagement_panel(self, data, duration):
3430 content_list = traverse_obj(
3431 data,
3432 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3433 expected_type=list)
3434 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3435 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3437 return next(filter(None, (
3438 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3439 chapter_time, chapter_title, duration)
3440 for contents in content_list)), [])
3442 def _extract_heatmap(self, data):
3443 return traverse_obj(data, (
3444 'frameworkUpdates', 'entityBatchUpdate', 'mutations',
3445 lambda _, v: v['payload']['macroMarkersListEntity']['markersList']['markerType'] == 'MARKER_TYPE_HEATMAP',
3446 'payload', 'macroMarkersListEntity', 'markersList', 'markers', ..., {
3447 'start_time': ('startMillis', {float_or_none(scale=1000)}),
3448 'end_time': {lambda x: (int(x['startMillis']) + int(x['durationMillis'])) / 1000},
3449 'value': ('intensityScoreNormalized', {float_or_none}),
3450 })) or None
3452 def _extract_comment(self, entities, parent=None):
3453 comment_entity_payload = get_first(entities, ('payload', 'commentEntityPayload', {dict}))
3454 if not (comment_id := traverse_obj(comment_entity_payload, ('properties', 'commentId', {str}))):
3455 return
3457 toolbar_entity_payload = get_first(entities, ('payload', 'engagementToolbarStateEntityPayload', {dict}))
3458 time_text = traverse_obj(comment_entity_payload, ('properties', 'publishedTime', {str})) or ''
3460 return {
3461 'id': comment_id,
3462 'parent': parent or 'root',
3463 **traverse_obj(comment_entity_payload, {
3464 'text': ('properties', 'content', 'content', {str}),
3465 'like_count': ('toolbar', 'likeCountA11y', {parse_count}),
3466 'author_id': ('author', 'channelId', {self.ucid_or_none}),
3467 'author': ('author', 'displayName', {str}),
3468 'author_thumbnail': ('author', 'avatarThumbnailUrl', {url_or_none}),
3469 'author_is_uploader': ('author', 'isCreator', {bool}),
3470 'author_is_verified': ('author', 'isVerified', {bool}),
3471 'author_url': ('author', 'channelCommand', 'innertubeCommand', (
3472 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'),
3473 ), {urljoin('https://www.youtube.com')}),
3474 }, get_all=False),
3475 'is_favorited': (None if toolbar_entity_payload is None else
3476 toolbar_entity_payload.get('heartState') == 'TOOLBAR_HEART_STATE_HEARTED'),
3477 '_time_text': time_text, # FIXME: non-standard, but we need a way of showing that it is an estimate.
3478 'timestamp': self._parse_time_text(time_text),
3481 def _extract_comment_old(self, comment_renderer, parent=None):
3482 comment_id = comment_renderer.get('commentId')
3483 if not comment_id:
3484 return
3486 info = {
3487 'id': comment_id,
3488 'text': self._get_text(comment_renderer, 'contentText'),
3489 'like_count': self._get_count(comment_renderer, 'voteCount'),
3490 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3491 'author': self._get_text(comment_renderer, 'authorText'),
3492 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3493 'parent': parent or 'root',
3496 # Timestamp is an estimate calculated from the current time and time_text
3497 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3498 timestamp = self._parse_time_text(time_text)
3500 info.update({
3501 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3502 '_time_text': time_text,
3503 'timestamp': timestamp,
3506 info['author_url'] = urljoin(
3507 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3508 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3509 expected_type=str, get_all=False))
3511 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3512 if author_is_uploader is not None:
3513 info['author_is_uploader'] = author_is_uploader
3515 comment_abr = traverse_obj(
3516 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3517 if comment_abr is not None:
3518 info['is_favorited'] = 'creatorHeart' in comment_abr
3520 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3521 if self._has_badge(badges, BadgeType.VERIFIED):
3522 info['author_is_verified'] = True
3524 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3525 if is_pinned:
3526 info['is_pinned'] = True
3528 return info
3530 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3532 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3534 def extract_header(contents):
3535 _continuation = None
3536 for content in contents:
3537 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3538 expected_comment_count = self._get_count(
3539 comments_header_renderer, 'countText', 'commentsCount')
3541 if expected_comment_count is not None:
3542 tracker['est_total'] = expected_comment_count
3543 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3544 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3546 sort_menu_item = try_get(
3547 comments_header_renderer,
3548 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3549 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3551 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3552 if not _continuation:
3553 continue
3555 sort_text = str_or_none(sort_menu_item.get('title'))
3556 if not sort_text:
3557 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3558 self.to_screen(f'Sorting comments by {sort_text.lower()}')
3559 break
3560 return _continuation
3562 def extract_thread(contents, entity_payloads):
3563 if not parent:
3564 tracker['current_page_thread'] = 0
3565 for content in contents:
3566 if not parent and tracker['total_parent_comments'] >= max_parents:
3567 yield
3568 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3570 # old comment format
3571 if not entity_payloads:
3572 comment_renderer = get_first(
3573 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3574 expected_type=dict, default={})
3576 comment = self._extract_comment_old(comment_renderer, parent)
3578 # new comment format
3579 else:
3580 view_model = (
3581 traverse_obj(comment_thread_renderer, ('commentViewModel', 'commentViewModel', {dict}))
3582 or traverse_obj(content, ('commentViewModel', {dict})))
3583 comment_keys = traverse_obj(view_model, (('commentKey', 'toolbarStateKey'), {str}))
3584 if not comment_keys:
3585 continue
3586 entities = traverse_obj(entity_payloads, lambda _, v: v['entityKey'] in comment_keys)
3587 comment = self._extract_comment(entities, parent)
3588 if comment:
3589 comment['is_pinned'] = traverse_obj(view_model, ('pinnedText', {str})) is not None
3591 if not comment:
3592 continue
3593 comment_id = comment['id']
3595 if comment.get('is_pinned'):
3596 tracker['pinned_comment_ids'].add(comment_id)
3597 # Sometimes YouTube may break and give us infinite looping comments.
3598 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3599 if comment_id in tracker['seen_comment_ids']:
3600 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3601 # Pinned comments may appear a second time in newest first sort
3602 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3603 continue
3604 self.report_warning(
3605 'Detected YouTube comments looping. Stopping comment extraction '
3606 f'{"for this thread" if parent else ""} as we probably cannot get any more.')
3607 yield
3608 else:
3609 tracker['seen_comment_ids'].add(comment['id'])
3611 tracker['running_total'] += 1
3612 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3613 yield comment
3615 # Attempt to get the replies
3616 comment_replies_renderer = try_get(
3617 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3619 if comment_replies_renderer:
3620 tracker['current_page_thread'] += 1
3621 comment_entries_iter = self._comment_entries(
3622 comment_replies_renderer, ytcfg, video_id,
3623 parent=comment.get('id'), tracker=tracker)
3624 yield from itertools.islice(comment_entries_iter, min(
3625 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3627 # Keeps track of counts across recursive calls
3628 if not tracker:
3629 tracker = {
3630 'running_total': 0,
3631 'est_total': None,
3632 'current_page_thread': 0,
3633 'total_parent_comments': 0,
3634 'total_reply_comments': 0,
3635 'seen_comment_ids': set(),
3636 'pinned_comment_ids': set(),
3639 # TODO: Deprecated
3640 # YouTube comments have a max depth of 2
3641 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3642 if max_depth:
3643 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3644 'Set max replies in the max-comments extractor argument instead')
3645 if max_depth == 1 and parent:
3646 return
3648 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = (
3649 int_or_none(p, default=sys.maxsize) for p in self._configuration_arg('max_comments') + [''] * 4)
3651 continuation = self._extract_continuation(root_continuation_data)
3653 response = None
3654 is_forced_continuation = False
3655 is_first_continuation = parent is None
3656 if is_first_continuation and not continuation:
3657 # Sometimes you can get comments by generating the continuation yourself,
3658 # even if YouTube initially reports them being disabled - e.g. stories comments.
3659 # Note: if the comment section is actually disabled, YouTube may return a response with
3660 # required check_get_keys missing. So we will disable that check initially in this case.
3661 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3662 is_forced_continuation = True
3664 continuation_items_path = (
3665 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3666 for page_num in itertools.count(0):
3667 if not continuation:
3668 break
3669 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3670 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3671 if page_num == 0:
3672 if is_first_continuation:
3673 note_prefix = 'Downloading comment section API JSON'
3674 else:
3675 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3676 tracker['current_page_thread'], comment_prog_str)
3677 else:
3678 note_prefix = '{}Downloading comment{} API JSON page {} {}'.format(
3679 ' ' if parent else '', ' replies' if parent else '',
3680 page_num, comment_prog_str)
3682 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3683 # Ignore check if YouTube says the comment count is 0.
3684 check_get_keys = None
3685 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3686 check_get_keys = [[*continuation_items_path, ..., (
3687 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentViewModel', 'commentRenderer'))]]
3688 try:
3689 response = self._extract_response(
3690 item_id=None, query=continuation,
3691 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3692 check_get_keys=check_get_keys)
3693 except ExtractorError as e:
3694 # Ignore incomplete data error for replies if retries didn't work.
3695 # This is to allow any other parent comments and comment threads to be downloaded.
3696 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3697 if 'incomplete data' in str(e).lower() and parent:
3698 if self.get_param('ignoreerrors') in (True, 'only_download'):
3699 self.report_warning(
3700 'Received incomplete data for a comment reply thread and retrying did not help. '
3701 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
3702 return
3703 else:
3704 raise ExtractorError(
3705 'Incomplete data received for comment reply thread. '
3706 'Pass --ignore-errors to ignore and allow rest of comments to download.',
3707 expected=True)
3708 raise
3709 is_forced_continuation = False
3710 continuation = None
3711 mutations = traverse_obj(response, ('frameworkUpdates', 'entityBatchUpdate', 'mutations', ..., {dict}))
3712 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3713 if is_first_continuation:
3714 continuation = extract_header(continuation_items)
3715 is_first_continuation = False
3716 if continuation:
3717 break
3718 continue
3720 for entry in extract_thread(continuation_items, mutations):
3721 if not entry:
3722 return
3723 yield entry
3724 continuation = self._extract_continuation({'contents': continuation_items})
3725 if continuation:
3726 break
3728 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3729 if message and not parent and tracker['running_total'] == 0:
3730 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3731 raise self.CommentsDisabled
3733 @staticmethod
3734 def _generate_comment_continuation(video_id):
3736 Generates initial comment section continuation token from given video id
3738 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3739 return base64.b64encode(token.encode()).decode()
3741 def _get_comments(self, ytcfg, video_id, contents, webpage):
3742 """Entry for comment extraction"""
3743 def _real_comment_extract(contents):
3744 renderer = next((
3745 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3746 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3747 yield from self._comment_entries(renderer, ytcfg, video_id)
3749 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3750 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3752 @staticmethod
3753 def _get_checkok_params():
3754 return {'contentCheckOk': True, 'racyCheckOk': True}
3756 @classmethod
3757 def _generate_player_context(cls, sts=None):
3758 context = {
3759 'html5Preference': 'HTML5_PREF_WANTS',
3761 if sts is not None:
3762 context['signatureTimestamp'] = sts
3763 return {
3764 'playbackContext': {
3765 'contentPlaybackContext': context,
3767 **cls._get_checkok_params(),
3770 def _get_config_po_token(self, client):
3771 po_token_strs = self._configuration_arg('po_token', [], ie_key=YoutubeIE, casesense=True)
3772 for token_str in po_token_strs:
3773 po_token_client, sep, po_token = token_str.partition('+')
3774 if not sep:
3775 self.report_warning(
3776 f'Invalid po_token configuration format. Expected "client+po_token", got "{token_str}"', only_once=True)
3777 continue
3778 if po_token_client == client:
3779 return po_token
3781 def fetch_po_token(self, client='web', visitor_data=None, data_sync_id=None, player_url=None, **kwargs):
3782 # PO Token is bound to visitor_data / Visitor ID when logged out. Must have visitor_data for it to function.
3783 if not visitor_data and not self.is_authenticated and player_url:
3784 self.report_warning(
3785 f'Unable to fetch PO Token for {client} client: Missing required Visitor Data. '
3786 f'You may need to pass Visitor Data with --extractor-args "youtube:visitor_data=XXX"')
3787 return
3789 config_po_token = self._get_config_po_token(client)
3790 if config_po_token:
3791 # PO token is bound to data_sync_id / account Session ID when logged in. However, for the config po_token,
3792 # if using first channel in an account then we don't need the data_sync_id anymore...
3793 if not data_sync_id and self.is_authenticated and player_url:
3794 self.report_warning(
3795 f'Got a PO Token for {client} client, but missing Data Sync ID for account. Formats may not work.'
3796 f'You may need to pass a Data Sync ID with --extractor-args "youtube:data_sync_id=XXX"')
3798 return config_po_token
3800 # Require PO Token if logged in for external fetching
3801 if not data_sync_id and self.is_authenticated and player_url:
3802 self.report_warning(
3803 f'Unable to fetch PO Token for {client} client: Missing required Data Sync ID for account. '
3804 f'You may need to pass a Data Sync ID with --extractor-args "youtube:data_sync_id=XXX"')
3805 return
3807 return self._fetch_po_token(
3808 client=client,
3809 visitor_data=visitor_data,
3810 data_sync_id=data_sync_id,
3811 player_url=player_url,
3812 **kwargs,
3815 def _fetch_po_token(self, client, visitor_data=None, data_sync_id=None, player_url=None, **kwargs):
3816 """External PO Token fetch stub"""
3818 @staticmethod
3819 def _is_agegated(player_response):
3820 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3821 return True
3823 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3824 AGE_GATE_REASONS = (
3825 'confirm your age', 'age-restricted', 'inappropriate', # reason
3826 'age_verification_required', 'age_check_required', # status
3828 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3830 @staticmethod
3831 def _is_unplayable(player_response):
3832 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3834 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, visitor_data, data_sync_id, po_token):
3835 headers = self.generate_api_headers(
3836 ytcfg=player_ytcfg,
3837 default_client=client,
3838 visitor_data=visitor_data,
3839 session_index=self._extract_session_index(master_ytcfg, player_ytcfg),
3840 account_syncid=(
3841 self._data_sync_id_to_delegated_session_id(data_sync_id)
3842 or self._extract_account_syncid(master_ytcfg, initial_pr, player_ytcfg)
3846 yt_query = {
3847 'videoId': video_id,
3850 default_pp = traverse_obj(
3851 INNERTUBE_CLIENTS, (_split_innertube_client(client)[0], 'PLAYER_PARAMS', {str}))
3852 if player_params := self._configuration_arg('player_params', [default_pp], casesense=True)[0]:
3853 yt_query['params'] = player_params
3855 if po_token:
3856 yt_query['serviceIntegrityDimensions'] = {'poToken': po_token}
3858 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3859 yt_query.update(self._generate_player_context(sts))
3860 return self._extract_response(
3861 item_id=video_id, ep='player', query=yt_query,
3862 ytcfg=player_ytcfg, headers=headers, fatal=True,
3863 default_client=client,
3864 note='Downloading {} player API JSON'.format(client.replace('_', ' ').strip()),
3865 ) or None
3867 def _get_requested_clients(self, url, smuggled_data):
3868 requested_clients = []
3869 excluded_clients = []
3870 default_clients = self._DEFAULT_AUTHED_CLIENTS if self.is_authenticated else self._DEFAULT_CLIENTS
3871 allowed_clients = sorted(
3872 (client for client in INNERTUBE_CLIENTS if client[:1] != '_'),
3873 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3874 for client in self._configuration_arg('player_client'):
3875 if client == 'default':
3876 requested_clients.extend(default_clients)
3877 elif client == 'all':
3878 requested_clients.extend(allowed_clients)
3879 elif client.startswith('-'):
3880 excluded_clients.append(client[1:])
3881 elif client not in allowed_clients:
3882 self.report_warning(f'Skipping unsupported client "{client}"')
3883 else:
3884 requested_clients.append(client)
3885 if not requested_clients:
3886 requested_clients.extend(default_clients)
3887 for excluded_client in excluded_clients:
3888 if excluded_client in requested_clients:
3889 requested_clients.remove(excluded_client)
3890 if not requested_clients:
3891 raise ExtractorError('No player clients have been requested', expected=True)
3893 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3894 for requested_client in requested_clients:
3895 _, base_client, variant = _split_innertube_client(requested_client)
3896 music_client = f'{base_client}_music' if base_client != 'mweb' else 'web_music'
3897 if variant != 'music' and music_client in INNERTUBE_CLIENTS:
3898 client_info = INNERTUBE_CLIENTS[music_client]
3899 if not client_info['REQUIRE_AUTH'] or (self.is_authenticated and client_info['SUPPORTS_COOKIES']):
3900 requested_clients.append(music_client)
3902 if self.is_authenticated:
3903 unsupported_clients = [
3904 client for client in requested_clients if not INNERTUBE_CLIENTS[client]['SUPPORTS_COOKIES']
3906 for client in unsupported_clients:
3907 self.report_warning(f'Skipping client "{client}" since it does not support cookies', only_once=True)
3908 requested_clients.remove(client)
3910 return orderedSet(requested_clients)
3912 def _invalid_player_response(self, pr, video_id):
3913 # YouTube may return a different video player response than expected.
3914 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3915 if (pr_id := traverse_obj(pr, ('videoDetails', 'videoId'))) != video_id:
3916 return pr_id
3918 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3919 initial_pr = None
3920 if webpage:
3921 initial_pr = self._search_json(
3922 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3924 prs = []
3925 deprioritized_prs = []
3927 if initial_pr and not self._invalid_player_response(initial_pr, video_id):
3928 # Android player_response does not have microFormats which are needed for
3929 # extraction of some data. So we return the initial_pr with formats
3930 # stripped out even if not requested by the user
3931 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3932 prs.append({**initial_pr, 'streamingData': None})
3934 all_clients = set(clients)
3935 clients = clients[::-1]
3937 def append_client(*client_names):
3938 """ Append the first client name that exists but not already used """
3939 for client_name in client_names:
3940 actual_client = _split_innertube_client(client_name)[0]
3941 if actual_client in INNERTUBE_CLIENTS:
3942 if actual_client not in all_clients:
3943 clients.append(client_name)
3944 all_clients.add(actual_client)
3945 return
3947 tried_iframe_fallback = False
3948 player_url = visitor_data = data_sync_id = None
3949 skipped_clients = {}
3950 while clients:
3951 deprioritize_pr = False
3952 client, base_client, variant = _split_innertube_client(clients.pop())
3953 player_ytcfg = master_ytcfg if client == 'web' else {}
3954 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3955 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3957 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3958 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3959 if 'js' in self._configuration_arg('player_skip'):
3960 require_js_player = False
3961 player_url = None
3963 if not player_url and not tried_iframe_fallback and require_js_player:
3964 player_url = self._download_player_url(video_id)
3965 tried_iframe_fallback = True
3967 visitor_data = visitor_data or self._extract_visitor_data(master_ytcfg, initial_pr, player_ytcfg)
3968 data_sync_id = data_sync_id or self._extract_data_sync_id(master_ytcfg, initial_pr, player_ytcfg)
3969 po_token = self.fetch_po_token(
3970 client=client, visitor_data=visitor_data,
3971 data_sync_id=data_sync_id if self.is_authenticated else None,
3972 player_url=player_url if require_js_player else None,
3975 require_po_token = self._get_default_ytcfg(client).get('REQUIRE_PO_TOKEN')
3976 if not po_token and require_po_token:
3977 self.report_warning(
3978 f'No PO Token provided for {client} client, '
3979 f'which is required for working {client} formats. '
3980 f'You can manually pass a PO Token for this client with '
3981 f'--extractor-args "youtube:po_token={client}+XXX"',
3982 only_once=True)
3983 deprioritize_pr = True
3985 pr = initial_pr if client == 'web' else None
3986 try:
3987 pr = pr or self._extract_player_response(
3988 client, video_id,
3989 master_ytcfg=player_ytcfg or master_ytcfg,
3990 player_ytcfg=player_ytcfg,
3991 player_url=player_url,
3992 initial_pr=initial_pr,
3993 visitor_data=visitor_data,
3994 data_sync_id=data_sync_id,
3995 po_token=po_token)
3996 except ExtractorError as e:
3997 self.report_warning(e)
3998 continue
4000 if pr_id := self._invalid_player_response(pr, video_id):
4001 skipped_clients[client] = pr_id
4002 elif pr:
4003 # Save client name for introspection later
4004 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
4005 sd[STREAMING_DATA_CLIENT_NAME] = client
4006 sd[STREAMING_DATA_PO_TOKEN] = po_token
4007 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
4008 f[STREAMING_DATA_CLIENT_NAME] = client
4009 f[STREAMING_DATA_PO_TOKEN] = po_token
4010 if deprioritize_pr:
4011 deprioritized_prs.append(pr)
4012 else:
4013 prs.append(pr)
4015 # web_embedded can work around age-gate and age-verification for some embeddable videos
4016 if self._is_agegated(pr) and variant != 'web_embedded':
4017 append_client(f'web_embedded.{base_client}')
4018 # Unauthenticated users will only get web_embedded client formats if age-gated
4019 if self._is_agegated(pr) and not self.is_authenticated:
4020 self.to_screen(
4021 f'{video_id}: This video is age-restricted; some formats may be missing '
4022 f'without authentication. {self._login_hint()}', only_once=True)
4024 ''' This code is pointless while web_creator is in _DEFAULT_AUTHED_CLIENTS
4025 # EU countries require age-verification for accounts to access age-restricted videos
4026 # If account is not age-verified, _is_agegated() will be truthy for non-embedded clients
4027 embedding_is_disabled = variant == 'web_embedded' and self._is_unplayable(pr)
4028 if self.is_authenticated and (self._is_agegated(pr) or embedding_is_disabled):
4029 self.to_screen(
4030 f'{video_id}: This video is age-restricted and YouTube is requiring '
4031 'account age-verification; some formats may be missing', only_once=True)
4032 # web_creator can work around the age-verification requirement
4033 # android_vr may also be able to work around age-verification
4034 # tv_embedded may(?) still work around age-verification if the video is embeddable
4035 append_client('web_creator')
4038 prs.extend(deprioritized_prs)
4040 if skipped_clients:
4041 self.report_warning(
4042 f'Skipping player responses from {"/".join(skipped_clients)} clients '
4043 f'(got player responses for video "{"/".join(set(skipped_clients.values()))}" instead of "{video_id}")')
4044 if not prs:
4045 raise ExtractorError(
4046 'All player responses are invalid. Your IP is likely being blocked by Youtube', expected=True)
4047 elif not prs:
4048 raise ExtractorError('Failed to extract any player response')
4049 return prs, player_url
4051 def _needs_live_processing(self, live_status, duration):
4052 if ((live_status == 'is_live' and self.get_param('live_from_start'))
4053 or (live_status == 'post_live' and (duration or 0) > 2 * 3600)):
4054 return live_status
4056 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
4057 CHUNK_SIZE = 10 << 20
4058 PREFERRED_LANG_VALUE = 10
4059 original_language = None
4060 itags, stream_ids = collections.defaultdict(set), []
4061 itag_qualities, res_qualities = {}, {0: None}
4062 q = qualities([
4063 # Normally tiny is the smallest video-only formats. But
4064 # audio-only formats with unknown quality may get tagged as tiny
4065 'tiny',
4066 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
4067 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres',
4069 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
4070 format_types = self._configuration_arg('formats')
4071 all_formats = 'duplicate' in format_types
4072 if self._configuration_arg('include_duplicate_formats'):
4073 all_formats = True
4074 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
4075 'Use formats=duplicate extractor argument instead')
4077 def build_fragments(f):
4078 return LazyList({
4079 'url': update_url_query(f['url'], {
4080 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}',
4082 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
4084 for fmt in streaming_formats:
4085 if fmt.get('targetDurationSec'):
4086 continue
4088 itag = str_or_none(fmt.get('itag'))
4089 audio_track = fmt.get('audioTrack') or {}
4090 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
4091 if not all_formats:
4092 if stream_id in stream_ids:
4093 continue
4095 quality = fmt.get('quality')
4096 height = int_or_none(fmt.get('height'))
4097 if quality == 'tiny' or not quality:
4098 quality = fmt.get('audioQuality', '').lower() or quality
4099 # The 3gp format (17) in android client has a quality of "small",
4100 # but is actually worse than other formats
4101 if itag == '17':
4102 quality = 'tiny'
4103 if quality:
4104 if itag:
4105 itag_qualities[itag] = quality
4106 if height:
4107 res_qualities[height] = quality
4109 display_name = audio_track.get('displayName') or ''
4110 is_original = 'original' in display_name.lower()
4111 is_descriptive = 'descriptive' in display_name.lower()
4112 is_default = audio_track.get('audioIsDefault')
4113 language_code = audio_track.get('id', '').split('.')[0]
4114 if language_code and (is_original or (is_default and not original_language)):
4115 original_language = language_code
4117 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
4118 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
4119 # number of fragment that would subsequently requested with (`&sq=N`)
4120 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
4121 continue
4123 fmt_url = fmt.get('url')
4124 if not fmt_url:
4125 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
4126 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
4127 encrypted_sig = try_get(sc, lambda x: x['s'][0])
4128 if not all((sc, fmt_url, player_url, encrypted_sig)):
4129 continue
4130 try:
4131 fmt_url += '&{}={}'.format(
4132 traverse_obj(sc, ('sp', -1)) or 'signature',
4133 self._decrypt_signature(encrypted_sig, video_id, player_url),
4135 except ExtractorError as e:
4136 self.report_warning('Signature extraction failed: Some formats may be missing',
4137 video_id=video_id, only_once=True)
4138 self.write_debug(e, only_once=True)
4139 continue
4141 query = parse_qs(fmt_url)
4142 if query.get('n'):
4143 try:
4144 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
4145 fmt_url = update_url_query(fmt_url, {
4146 'n': decrypt_nsig(query['n'][0], video_id, player_url),
4148 except ExtractorError as e:
4149 phantomjs_hint = ''
4150 if isinstance(e, JSInterpreter.Exception):
4151 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
4152 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
4153 if player_url:
4154 self.report_warning(
4155 f'nsig extraction failed: Some formats may be missing\n{phantomjs_hint}'
4156 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
4157 self.write_debug(e, only_once=True)
4158 else:
4159 self.report_warning(
4160 'Cannot decrypt nsig without player_url: Some formats may be missing',
4161 video_id=video_id, only_once=True)
4162 continue
4164 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
4165 format_duration = traverse_obj(fmt, ('approxDurationMs', {float_or_none(scale=1000)}))
4166 # Some formats may have much smaller duration than others (possibly damaged during encoding)
4167 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
4168 # Make sure to avoid false positives with small duration differences.
4169 # E.g. __2ABJjxzNo, ySuUZEjARPY
4170 is_damaged = try_call(lambda: format_duration < duration // 2)
4171 if is_damaged:
4172 self.report_warning(
4173 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
4175 client_name = fmt[STREAMING_DATA_CLIENT_NAME]
4176 po_token = fmt.get(STREAMING_DATA_PO_TOKEN)
4178 if po_token:
4179 fmt_url = update_url_query(fmt_url, {'pot': po_token})
4181 # Clients that require PO Token return videoplayback URLs that may return 403
4182 is_broken = (not po_token and self._get_default_ytcfg(client_name).get('REQUIRE_PO_TOKEN'))
4183 if is_broken:
4184 self.report_warning(
4185 f'{video_id}: {client_name} client formats require a PO Token which was not provided. '
4186 'They will be deprioritized as they may yield HTTP Error 403', only_once=True)
4188 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
4189 fps = int_or_none(fmt.get('fps')) or 0
4190 dct = {
4191 'asr': int_or_none(fmt.get('audioSampleRate')),
4192 'filesize': int_or_none(fmt.get('contentLength')),
4193 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
4194 'format_note': join_nonempty(
4195 join_nonempty(display_name, is_default and ' (default)', delim=''),
4196 name, fmt.get('isDrc') and 'DRC',
4197 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
4198 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
4199 is_damaged and 'DAMAGED', is_broken and 'BROKEN',
4200 (self.get_param('verbose') or all_formats) and short_client_name(client_name),
4201 delim=', '),
4202 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
4203 'source_preference': (-5 if itag == '22' else -1) + (100 if 'Premium' in name else 0),
4204 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
4205 'audio_channels': fmt.get('audioChannels'),
4206 'height': height,
4207 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
4208 'has_drm': bool(fmt.get('drmFamilies')),
4209 'tbr': tbr,
4210 'filesize_approx': filesize_from_tbr(tbr, format_duration),
4211 'url': fmt_url,
4212 'width': int_or_none(fmt.get('width')),
4213 'language': join_nonempty(language_code, 'desc' if is_descriptive else '') or None,
4214 'language_preference': PREFERRED_LANG_VALUE if is_original else 5 if is_default else -10 if is_descriptive else -1,
4215 # Strictly de-prioritize broken, damaged and 3gp formats
4216 'preference': -20 if is_broken else -10 if is_damaged else -2 if itag == '17' else None,
4218 mime_mobj = re.match(
4219 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
4220 if mime_mobj:
4221 dct['ext'] = mimetype2ext(mime_mobj.group(1))
4222 dct.update(parse_codecs(mime_mobj.group(2)))
4223 if itag:
4224 itags[itag].add(('https', dct.get('language')))
4225 stream_ids.append(stream_id)
4226 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
4227 if single_stream and dct.get('ext'):
4228 dct['container'] = dct['ext'] + '_dash'
4230 if (all_formats or 'dashy' in format_types) and dct['filesize']:
4231 yield {
4232 **dct,
4233 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
4234 'protocol': 'http_dash_segments',
4235 'fragments': build_fragments(dct),
4237 if all_formats or 'dashy' not in format_types:
4238 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
4239 yield dct
4241 needs_live_processing = self._needs_live_processing(live_status, duration)
4242 skip_bad_formats = 'incomplete' not in format_types
4243 if self._configuration_arg('include_incomplete_formats'):
4244 skip_bad_formats = False
4245 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
4246 'Use formats=incomplete extractor argument instead')
4248 skip_manifests = set(self._configuration_arg('skip'))
4249 if (not self.get_param('youtube_include_hls_manifest', True)
4250 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
4251 or (needs_live_processing and skip_bad_formats)):
4252 skip_manifests.add('hls')
4254 if not self.get_param('youtube_include_dash_manifest', True):
4255 skip_manifests.add('dash')
4256 if self._configuration_arg('include_live_dash'):
4257 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
4258 'Use formats=incomplete extractor argument instead')
4259 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
4260 skip_manifests.add('dash')
4262 def process_manifest_format(f, proto, client_name, itag, po_token):
4263 key = (proto, f.get('language'))
4264 if not all_formats and key in itags[itag]:
4265 return False
4266 itags[itag].add(key)
4268 if f.get('source_preference') is None:
4269 f['source_preference'] = -1
4271 # Clients that require PO Token return videoplayback URLs that may return 403
4272 # hls does not currently require PO Token
4273 if (not po_token and self._get_default_ytcfg(client_name).get('REQUIRE_PO_TOKEN')) and proto != 'hls':
4274 self.report_warning(
4275 f'{video_id}: {client_name} client {proto} formats require a PO Token which was not provided. '
4276 'They will be deprioritized as they may yield HTTP Error 403', only_once=True)
4277 f['format_note'] = join_nonempty(f.get('format_note'), 'BROKEN', delim=' ')
4278 f['source_preference'] -= 20
4280 if itag and all_formats:
4281 f['format_id'] = f'{itag}-{proto}'
4282 elif any(p != proto for p, _ in itags[itag]):
4283 f['format_id'] = f'{itag}-{proto}'
4284 elif itag:
4285 f['format_id'] = itag
4287 if original_language and f.get('language') == original_language:
4288 f['format_note'] = join_nonempty(f.get('format_note'), '(default)', delim=' ')
4289 f['language_preference'] = PREFERRED_LANG_VALUE
4291 if itag in ('616', '235'):
4292 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
4293 f['source_preference'] += 100
4295 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
4296 if f['quality'] == -1 and f.get('height'):
4297 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
4298 if self.get_param('verbose') or all_formats:
4299 f['format_note'] = join_nonempty(
4300 f.get('format_note'), short_client_name(client_name), delim=', ')
4301 if f.get('fps') and f['fps'] <= 1:
4302 del f['fps']
4304 if proto == 'hls' and f.get('has_drm'):
4305 f['has_drm'] = 'maybe'
4306 f['source_preference'] -= 5
4307 return True
4309 subtitles = {}
4310 for sd in streaming_data:
4311 client_name = sd[STREAMING_DATA_CLIENT_NAME]
4312 po_token = sd.get(STREAMING_DATA_PO_TOKEN)
4313 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
4314 if hls_manifest_url:
4315 if po_token:
4316 hls_manifest_url = hls_manifest_url.rstrip('/') + f'/pot/{po_token}'
4317 fmts, subs = self._extract_m3u8_formats_and_subtitles(
4318 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
4319 subtitles = self._merge_subtitles(subs, subtitles)
4320 for f in fmts:
4321 if process_manifest_format(f, 'hls', client_name, self._search_regex(
4322 r'/itag/(\d+)', f['url'], 'itag', default=None), po_token):
4323 yield f
4325 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
4326 if dash_manifest_url:
4327 if po_token:
4328 dash_manifest_url = dash_manifest_url.rstrip('/') + f'/pot/{po_token}'
4329 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
4330 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
4331 for f in formats:
4332 if process_manifest_format(f, 'dash', client_name, f['format_id'], po_token):
4333 f['filesize'] = int_or_none(self._search_regex(
4334 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
4335 if needs_live_processing:
4336 f['is_from_start'] = True
4338 yield f
4339 yield subtitles
4341 def _extract_storyboard(self, player_responses, duration):
4342 spec = get_first(
4343 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
4344 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
4345 if not base_url:
4346 return
4347 L = len(spec) - 1
4348 for i, args in enumerate(spec):
4349 args = args.split('#')
4350 counts = list(map(int_or_none, args[:5]))
4351 if len(args) != 8 or not all(counts):
4352 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
4353 continue
4354 width, height, frame_count, cols, rows = counts
4355 N, sigh = args[6:]
4357 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
4358 fragment_count = frame_count / (cols * rows)
4359 fragment_duration = duration / fragment_count
4360 yield {
4361 'format_id': f'sb{i}',
4362 'format_note': 'storyboard',
4363 'ext': 'mhtml',
4364 'protocol': 'mhtml',
4365 'acodec': 'none',
4366 'vcodec': 'none',
4367 'url': url,
4368 'width': width,
4369 'height': height,
4370 'fps': frame_count / duration,
4371 'rows': rows,
4372 'columns': cols,
4373 'fragments': [{
4374 'url': url.replace('$M', str(j)),
4375 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4376 } for j in range(math.ceil(fragment_count))],
4379 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4380 webpage = None
4381 if 'webpage' not in self._configuration_arg('player_skip'):
4382 query = {'bpctr': '9999999999', 'has_verified': '1'}
4383 pp = self._configuration_arg('player_params', [None], casesense=True)[0]
4384 if pp:
4385 query['pp'] = pp
4386 webpage = self._download_webpage(
4387 webpage_url, video_id, fatal=False, query=query)
4389 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4391 player_responses, player_url = self._extract_player_responses(
4392 self._get_requested_clients(url, smuggled_data),
4393 video_id, webpage, master_ytcfg, smuggled_data)
4395 return webpage, master_ytcfg, player_responses, player_url
4397 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4398 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4399 is_live = get_first(video_details, 'isLive')
4400 if is_live is None:
4401 is_live = get_first(live_broadcast_details, 'isLiveNow')
4402 live_content = get_first(video_details, 'isLiveContent')
4403 is_upcoming = get_first(video_details, 'isUpcoming')
4404 post_live = get_first(video_details, 'isPostLiveDvr')
4405 live_status = ('post_live' if post_live
4406 else 'is_live' if is_live
4407 else 'is_upcoming' if is_upcoming
4408 else 'was_live' if live_content
4409 else 'not_live' if False in (is_live, live_content)
4410 else None)
4411 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4412 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4413 if all(f.get('has_drm') for f in formats):
4414 # If there are no formats that definitely don't have DRM, all have DRM
4415 for f in formats:
4416 f['has_drm'] = True
4418 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4420 def _real_extract(self, url):
4421 url, smuggled_data = unsmuggle_url(url, {})
4422 video_id = self._match_id(url)
4424 base_url = self.http_scheme() + '//www.youtube.com/'
4425 webpage_url = base_url + 'watch?v=' + video_id
4427 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4429 playability_statuses = traverse_obj(
4430 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4432 trailer_video_id = get_first(
4433 playability_statuses,
4434 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4435 expected_type=str)
4436 if trailer_video_id:
4437 return self.url_result(
4438 trailer_video_id, self.ie_key(), trailer_video_id)
4440 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4441 if webpage else (lambda x: None))
4443 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4444 microformats = traverse_obj(
4445 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4446 expected_type=dict)
4448 translated_title = self._get_text(microformats, (..., 'title'))
4449 video_title = ((self._preferred_lang and translated_title)
4450 or get_first(video_details, 'title') # primary
4451 or translated_title
4452 or search_meta(['og:title', 'twitter:title', 'title']))
4453 translated_description = self._get_text(microformats, (..., 'description'))
4454 original_description = get_first(video_details, 'shortDescription')
4455 video_description = (
4456 (self._preferred_lang and translated_description)
4457 # If original description is blank, it will be an empty string.
4458 # Do not prefer translated description in this case.
4459 or original_description if original_description is not None else translated_description)
4461 multifeed_metadata_list = get_first(
4462 player_responses,
4463 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4464 expected_type=str)
4465 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4466 if self.get_param('noplaylist'):
4467 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
4468 else:
4469 entries = []
4470 feed_ids = []
4471 for feed in multifeed_metadata_list.split(','):
4472 # Unquote should take place before split on comma (,) since textual
4473 # fields may contain comma as well (see
4474 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4475 feed_data = urllib.parse.parse_qs(
4476 urllib.parse.unquote_plus(feed))
4478 def feed_entry(name):
4479 return try_get(
4480 feed_data, lambda x: x[name][0], str)
4482 feed_id = feed_entry('id')
4483 if not feed_id:
4484 continue
4485 feed_title = feed_entry('title')
4486 title = video_title
4487 if feed_title:
4488 title += f' ({feed_title})'
4489 entries.append({
4490 '_type': 'url_transparent',
4491 'ie_key': 'Youtube',
4492 'url': smuggle_url(
4493 '{}watch?v={}'.format(base_url, feed_data['id'][0]),
4494 {'force_singlefeed': True}),
4495 'title': title,
4497 feed_ids.append(feed_id)
4498 self.to_screen(
4499 'Downloading multifeed video ({}) - add --no-playlist to just download video {}'.format(
4500 ', '.join(feed_ids), video_id))
4501 return self.playlist_result(
4502 entries, video_id, video_title, video_description)
4504 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4505 or int_or_none(get_first(microformats, 'lengthSeconds'))
4506 or parse_duration(search_meta('duration')) or None)
4508 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4509 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4510 if live_status == 'post_live':
4511 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4513 if not formats:
4514 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4515 self.report_drm(video_id)
4516 pemr = get_first(
4517 playability_statuses,
4518 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4519 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4520 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4521 if subreason:
4522 if subreason == 'The uploader has not made this video available in your country.':
4523 countries = get_first(microformats, 'availableCountries')
4524 if not countries:
4525 regions_allowed = search_meta('regionsAllowed')
4526 countries = regions_allowed.split(',') if regions_allowed else None
4527 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4528 reason += f'. {subreason}'
4529 if reason:
4530 if 'sign in' in reason.lower():
4531 reason = remove_end(reason, 'This helps protect our community. Learn more')
4532 reason = f'{remove_end(reason.strip(), ".")}. {self._youtube_login_hint}'
4533 self.raise_no_formats(reason, expected=True)
4535 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4536 if not keywords and webpage:
4537 keywords = [
4538 unescapeHTML(m.group('content'))
4539 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4540 for keyword in keywords:
4541 if keyword.startswith('yt:stretch='):
4542 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4543 if mobj:
4544 # NB: float is intentional for forcing float division
4545 w, h = (float(v) for v in mobj.groups())
4546 if w > 0 and h > 0:
4547 ratio = w / h
4548 for f in formats:
4549 if f.get('vcodec') != 'none':
4550 f['stretched_ratio'] = ratio
4551 break
4552 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4553 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4554 if thumbnail_url:
4555 thumbnails.append({
4556 'url': thumbnail_url,
4558 original_thumbnails = thumbnails.copy()
4560 # The best resolution thumbnails sometimes does not appear in the webpage
4561 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4562 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4563 thumbnail_names = [
4564 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4565 # in resolution, these are not the custom thumbnail. So de-prioritize them
4566 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4567 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3',
4569 n_thumbnail_names = len(thumbnail_names)
4570 thumbnails.extend({
4571 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4572 video_id=video_id, name=name, ext=ext,
4573 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4574 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4575 for thumb in thumbnails:
4576 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4577 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4578 self._remove_duplicate_formats(thumbnails)
4579 self._downloader._sort_thumbnails(original_thumbnails)
4581 category = get_first(microformats, 'category') or search_meta('genre')
4582 channel_id = self.ucid_or_none(str_or_none(
4583 get_first(video_details, 'channelId')
4584 or get_first(microformats, 'externalChannelId')
4585 or search_meta('channelId')))
4586 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4588 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4589 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4590 if not duration and live_end_time and live_start_time:
4591 duration = live_end_time - live_start_time
4593 needs_live_processing = self._needs_live_processing(live_status, duration)
4595 def is_bad_format(fmt):
4596 if needs_live_processing and not fmt.get('is_from_start'):
4597 return True
4598 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4599 and fmt.get('protocol') == 'http_dash_segments'):
4600 return True
4602 for fmt in filter(is_bad_format, formats):
4603 fmt['preference'] = (fmt.get('preference') or -1) - 10
4604 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4606 if needs_live_processing:
4607 self._prepare_live_from_start_formats(
4608 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4610 formats.extend(self._extract_storyboard(player_responses, duration))
4612 channel_handle = self.handle_from_url(owner_profile_url)
4614 info = {
4615 'id': video_id,
4616 'title': video_title,
4617 'formats': formats,
4618 'thumbnails': thumbnails,
4619 # The best thumbnail that we are sure exists. Prevents unnecessary
4620 # URL checking if user don't care about getting the best possible thumbnail
4621 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4622 'description': video_description,
4623 'channel_id': channel_id,
4624 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4625 'duration': duration,
4626 'view_count': int_or_none(
4627 get_first((video_details, microformats), (..., 'viewCount'))
4628 or search_meta('interactionCount')),
4629 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4630 'age_limit': 18 if (
4631 get_first(microformats, 'isFamilySafe') is False
4632 or search_meta('isFamilyFriendly') == 'false'
4633 or search_meta('og:restrictions:age') == '18+') else 0,
4634 'webpage_url': webpage_url,
4635 'categories': [category] if category else None,
4636 'tags': keywords,
4637 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4638 'live_status': live_status,
4639 'release_timestamp': live_start_time,
4640 '_format_sort_fields': ( # source_preference is lower for potentially damaged formats
4641 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec', 'channels', 'acodec', 'lang', 'proto'),
4644 subtitles = {}
4645 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4646 if pctr:
4647 def get_lang_code(track):
4648 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4649 or track.get('languageCode'))
4651 # Converted into dicts to remove duplicates
4652 captions = {
4653 get_lang_code(sub): sub
4654 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4655 translation_languages = {
4656 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4657 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4659 def process_language(container, base_url, lang_code, sub_name, query):
4660 lang_subs = container.setdefault(lang_code, [])
4661 for fmt in self._SUBTITLE_FORMATS:
4662 query.update({
4663 'fmt': fmt,
4665 lang_subs.append({
4666 'ext': fmt,
4667 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4668 'name': sub_name,
4671 # NB: Constructing the full subtitle dictionary is slow
4672 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4673 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4674 for lang_code, caption_track in captions.items():
4675 base_url = caption_track.get('baseUrl')
4676 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4677 if not base_url:
4678 continue
4679 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4680 if caption_track.get('kind') != 'asr':
4681 if not lang_code:
4682 continue
4683 process_language(
4684 subtitles, base_url, lang_code, lang_name, {})
4685 if not caption_track.get('isTranslatable'):
4686 continue
4687 for trans_code, trans_name in translation_languages.items():
4688 if not trans_code:
4689 continue
4690 orig_trans_code = trans_code
4691 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4692 if not get_translated_subs:
4693 continue
4694 trans_code += f'-{lang_code}'
4695 trans_name += format_field(lang_name, None, ' from %s')
4696 if lang_code == f'a-{orig_trans_code}':
4697 # Set audio language based on original subtitles
4698 for f in formats:
4699 if f.get('acodec') != 'none' and not f.get('language'):
4700 f['language'] = orig_trans_code
4701 # Add an "-orig" label to the original language so that it can be distinguished.
4702 # The subs are returned without "-orig" as well for compatibility
4703 process_language(
4704 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4705 # Setting tlang=lang returns damaged subtitles.
4706 process_language(automatic_captions, base_url, trans_code, trans_name,
4707 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4709 info['automatic_captions'] = automatic_captions
4710 info['subtitles'] = subtitles
4712 parsed_url = urllib.parse.urlparse(url)
4713 for component in [parsed_url.fragment, parsed_url.query]:
4714 query = urllib.parse.parse_qs(component)
4715 for k, v in query.items():
4716 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4717 d_k += '_time'
4718 if d_k not in info and k in s_ks:
4719 info[d_k] = parse_duration(v[0])
4721 # Youtube Music Auto-generated description
4722 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4723 # XXX: Causes catastrophic backtracking if description has "·"
4724 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4725 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4726 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4727 mobj = re.search(
4728 r'''(?xs)
4729 (?=(?P<track>[^\n·]+))(?P=track)·
4730 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4731 (?=(?P<album>[^\n]+))(?P=album)\n
4732 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4733 (?:.+?Released\ on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4734 (.+?\nArtist\s*:\s*
4735 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4736 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4737 ''', video_description)
4738 if mobj:
4739 release_year = mobj.group('release_year')
4740 release_date = mobj.group('release_date')
4741 if release_date:
4742 release_date = release_date.replace('-', '')
4743 if not release_year:
4744 release_year = release_date[:4]
4745 info.update({
4746 'album': mobj.group('album'.strip()),
4747 'artists': ([a] if (a := mobj.group('clean_artist'))
4748 else [a.strip() for a in mobj.group('artist').split('·')]),
4749 'track': mobj.group('track').strip(),
4750 'release_date': release_date,
4751 'release_year': int_or_none(release_year),
4754 initial_data = None
4755 if webpage:
4756 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4757 if not traverse_obj(initial_data, 'contents'):
4758 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4759 initial_data = None
4760 if not initial_data:
4761 query = {'videoId': video_id}
4762 query.update(self._get_checkok_params())
4763 initial_data = self._extract_response(
4764 item_id=video_id, ep='next', fatal=False,
4765 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4766 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4767 note='Downloading initial data API JSON')
4769 COMMENTS_SECTION_IDS = ('comment-item-section', 'engagement-panel-comments-section')
4770 info['comment_count'] = traverse_obj(initial_data, (
4771 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4772 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount',
4773 ), (
4774 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] in COMMENTS_SECTION_IDS,
4775 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo',
4776 ), expected_type=self._get_count, get_all=False)
4778 try: # This will error if there is no livechat
4779 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4780 except (KeyError, IndexError, TypeError):
4781 pass
4782 else:
4783 info.setdefault('subtitles', {})['live_chat'] = [{
4784 # url is needed to set cookies
4785 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4786 'video_id': video_id,
4787 'ext': 'json',
4788 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4789 else 'youtube_live_chat_replay'),
4792 if initial_data:
4793 info['chapters'] = (
4794 self._extract_chapters_from_json(initial_data, duration)
4795 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4796 or self._extract_chapters_from_description(video_description, duration)
4797 or None)
4799 info['heatmap'] = self._extract_heatmap(initial_data)
4801 contents = traverse_obj(
4802 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4803 expected_type=list, default=[])
4805 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4806 if vpir:
4807 stl = vpir.get('superTitleLink')
4808 if stl:
4809 stl = self._get_text(stl)
4810 if try_get(
4811 vpir,
4812 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4813 info['location'] = stl
4814 else:
4815 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4816 if mobj:
4817 info.update({
4818 'series': mobj.group(1),
4819 'season_number': int(mobj.group(2)),
4820 'episode_number': int(mobj.group(3)),
4822 for tlb in (try_get(
4823 vpir,
4824 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4825 list) or []):
4826 tbrs = variadic(
4827 traverse_obj(
4828 tlb, ('toggleButtonRenderer', ...),
4829 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4830 for tbr in tbrs:
4831 for getter, regex in [(
4832 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4833 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4834 lambda x: x['accessibility'],
4835 lambda x: x['accessibilityData']['accessibilityData'],
4836 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4837 label = (try_get(tbr, getter, dict) or {}).get('label')
4838 if label:
4839 mobj = re.match(regex, label)
4840 if mobj:
4841 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4842 break
4844 info['like_count'] = traverse_obj(vpir, (
4845 'videoActions', 'menuRenderer', 'topLevelButtons', ...,
4846 'segmentedLikeDislikeButtonViewModel', 'likeButtonViewModel', 'likeButtonViewModel',
4847 'toggleButtonViewModel', 'toggleButtonViewModel', 'defaultButtonViewModel',
4848 'buttonViewModel', 'accessibilityText', {parse_count}), get_all=False)
4850 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4851 if vcr:
4852 vc = self._get_count(vcr, 'viewCount')
4853 # Upcoming premieres with waiting count are treated as live here
4854 if vcr.get('isLive'):
4855 info['concurrent_view_count'] = vc
4856 elif info.get('view_count') is None:
4857 info['view_count'] = vc
4859 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4860 if vsir:
4861 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4862 info.update({
4863 'channel': self._get_text(vor, 'title'),
4864 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4866 if not channel_handle:
4867 channel_handle = self.handle_from_url(
4868 traverse_obj(vor, (
4869 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4870 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4871 {str}), get_all=False))
4873 rows = try_get(
4874 vsir,
4875 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4876 list) or []
4877 multiple_songs = False
4878 for row in rows:
4879 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4880 multiple_songs = True
4881 break
4882 for row in rows:
4883 mrr = row.get('metadataRowRenderer') or {}
4884 mrr_title = mrr.get('title')
4885 if not mrr_title:
4886 continue
4887 mrr_title = self._get_text(mrr, 'title')
4888 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4889 if mrr_title == 'License':
4890 info['license'] = mrr_contents_text
4891 elif not multiple_songs:
4892 if mrr_title == 'Album':
4893 info['album'] = mrr_contents_text
4894 elif mrr_title == 'Artist':
4895 info['artists'] = [mrr_contents_text] if mrr_contents_text else None
4896 elif mrr_title == 'Song':
4897 info['track'] = mrr_contents_text
4898 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4899 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4900 info['channel_is_verified'] = True
4902 info.update({
4903 'uploader': info.get('channel'),
4904 'uploader_id': channel_handle,
4905 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4908 # We only want timestamp IF it has time precision AND a timezone
4909 # Currently the uploadDate in microformats appears to be in US/Pacific timezone.
4910 timestamp = (
4911 parse_iso8601(get_first(microformats, 'uploadDate'), timezone=NO_DEFAULT)
4912 or parse_iso8601(search_meta('uploadDate'), timezone=NO_DEFAULT)
4914 upload_date = (
4915 dt.datetime.fromtimestamp(timestamp, dt.timezone.utc).strftime('%Y%m%d') if timestamp else
4917 unified_strdate(get_first(microformats, 'uploadDate'))
4918 or unified_strdate(search_meta('uploadDate'))
4921 # In the case we cannot get the timestamp:
4922 # The upload date for scheduled, live and past live streams / premieres in microformats
4923 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4924 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4925 if not upload_date or (not timestamp and live_status in ('not_live', None)):
4926 # this should be in UTC, as configured in the cookie/client context
4927 upload_date = strftime_or_none(
4928 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4930 info['upload_date'] = upload_date
4931 info['timestamp'] = timestamp
4933 if upload_date and live_status not in ('is_live', 'post_live', 'is_upcoming'):
4934 # Newly uploaded videos' HLS formats are potentially problematic and need to be checked
4935 upload_datetime = datetime_from_str(upload_date).replace(tzinfo=dt.timezone.utc)
4936 if upload_datetime >= datetime_from_str('today-2days'):
4937 for fmt in info['formats']:
4938 if fmt.get('protocol') == 'm3u8_native':
4939 fmt['__needs_testing'] = True
4941 for s_k, d_k in [('artists', 'creators'), ('track', 'alt_title')]:
4942 v = info.get(s_k)
4943 if v:
4944 info[d_k] = v
4946 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4948 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4949 or get_first(video_details, 'isPrivate', expected_type=bool))
4951 info['availability'] = (
4952 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4953 else self._availability(
4954 is_private=is_private,
4955 needs_premium=(
4956 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4957 or False if initial_data and is_private is not None else None),
4958 needs_subscription=(
4959 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4960 or False if initial_data and is_private is not None else None),
4961 needs_auth=info['age_limit'] >= 18,
4962 is_unlisted=None if is_private is None else (
4963 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4964 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4966 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4968 self.mark_watched(video_id, player_responses)
4970 return info
4973 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4974 @staticmethod
4975 def passthrough_smuggled_data(func):
4976 def _smuggle(info, smuggled_data):
4977 if info.get('_type') not in ('url', 'url_transparent'):
4978 return info
4979 if smuggled_data.get('is_music_url'):
4980 parsed_url = urllib.parse.urlparse(info['url'])
4981 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4982 smuggled_data.pop('is_music_url')
4983 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4984 if smuggled_data:
4985 info['url'] = smuggle_url(info['url'], smuggled_data)
4986 return info
4988 @functools.wraps(func)
4989 def wrapper(self, url):
4990 url, smuggled_data = unsmuggle_url(url, {})
4991 if self.is_music_url(url):
4992 smuggled_data['is_music_url'] = True
4993 info_dict = func(self, url, smuggled_data)
4994 if smuggled_data:
4995 _smuggle(info_dict, smuggled_data)
4996 if info_dict.get('entries'):
4997 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4998 return info_dict
4999 return wrapper
5001 @staticmethod
5002 def _extract_basic_item_renderer(item):
5003 # Modified from _extract_grid_item_renderer
5004 known_basic_renderers = (
5005 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer',
5007 for key, renderer in item.items():
5008 if not isinstance(renderer, dict):
5009 continue
5010 elif key in known_basic_renderers:
5011 return renderer
5012 elif key.startswith('grid') and key.endswith('Renderer'):
5013 return renderer
5015 def _extract_channel_renderer(self, renderer):
5016 channel_id = self.ucid_or_none(renderer['channelId'])
5017 title = self._get_text(renderer, 'title')
5018 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
5019 channel_handle = self.handle_from_url(
5020 traverse_obj(renderer, (
5021 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
5022 ('browseEndpoint', 'canonicalBaseUrl')),
5023 {str}), get_all=False))
5024 if not channel_handle:
5025 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
5026 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
5027 return {
5028 '_type': 'url',
5029 'url': channel_url,
5030 'id': channel_id,
5031 'ie_key': YoutubeTabIE.ie_key(),
5032 'channel': title,
5033 'uploader': title,
5034 'channel_id': channel_id,
5035 'channel_url': channel_url,
5036 'title': title,
5037 'uploader_id': channel_handle,
5038 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5039 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
5040 # However, in feed/channels this is set correctly to the subscriber count
5041 'channel_follower_count': traverse_obj(
5042 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
5043 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
5044 'playlist_count': (
5045 # videoCountText may be the subscriber count
5046 self._get_count(renderer, 'videoCountText')
5047 if self._get_count(renderer, 'subscriberCountText') is not None else None),
5048 'description': self._get_text(renderer, 'descriptionSnippet'),
5049 'channel_is_verified': True if self._has_badge(
5050 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
5053 def _grid_entries(self, grid_renderer):
5054 for item in grid_renderer['items']:
5055 if not isinstance(item, dict):
5056 continue
5057 if lockup_view_model := traverse_obj(item, ('lockupViewModel', {dict})):
5058 if entry := self._extract_lockup_view_model(lockup_view_model):
5059 yield entry
5060 continue
5061 renderer = self._extract_basic_item_renderer(item)
5062 if not isinstance(renderer, dict):
5063 continue
5064 title = self._get_text(renderer, 'title')
5066 # playlist
5067 playlist_id = renderer.get('playlistId')
5068 if playlist_id:
5069 yield self.url_result(
5070 f'https://www.youtube.com/playlist?list={playlist_id}',
5071 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5072 video_title=title)
5073 continue
5074 # video
5075 video_id = renderer.get('videoId')
5076 if video_id:
5077 yield self._extract_video(renderer)
5078 continue
5079 # channel
5080 channel_id = renderer.get('channelId')
5081 if channel_id:
5082 yield self._extract_channel_renderer(renderer)
5083 continue
5084 # generic endpoint URL support
5085 ep_url = urljoin('https://www.youtube.com/', try_get(
5086 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
5087 str))
5088 if ep_url:
5089 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
5090 if ie.suitable(ep_url):
5091 yield self.url_result(
5092 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
5093 break
5095 def _music_reponsive_list_entry(self, renderer):
5096 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
5097 if video_id:
5098 title = traverse_obj(renderer, (
5099 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
5100 'text', 'runs', 0, 'text'))
5101 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
5102 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
5103 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
5104 if playlist_id:
5105 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
5106 if video_id:
5107 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
5108 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5109 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
5110 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5111 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
5112 if browse_id:
5113 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
5114 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
5116 def _shelf_entries_from_content(self, shelf_renderer):
5117 content = shelf_renderer.get('content')
5118 if not isinstance(content, dict):
5119 return
5120 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
5121 if renderer:
5122 # TODO: add support for nested playlists so each shelf is processed
5123 # as separate playlist
5124 # TODO: this includes only first N items
5125 yield from self._grid_entries(renderer)
5126 renderer = content.get('horizontalListRenderer')
5127 if renderer:
5128 # TODO: handle case
5129 pass
5131 def _shelf_entries(self, shelf_renderer, skip_channels=False):
5132 ep = try_get(
5133 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5134 str)
5135 shelf_url = urljoin('https://www.youtube.com', ep)
5136 if shelf_url:
5137 # Skipping links to another channels, note that checking for
5138 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
5139 # will not work
5140 if skip_channels and '/channels?' in shelf_url:
5141 return
5142 title = self._get_text(shelf_renderer, 'title')
5143 yield self.url_result(shelf_url, video_title=title)
5144 # Shelf may not contain shelf URL, fallback to extraction from content
5145 yield from self._shelf_entries_from_content(shelf_renderer)
5147 def _playlist_entries(self, video_list_renderer):
5148 for content in video_list_renderer['contents']:
5149 if not isinstance(content, dict):
5150 continue
5151 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
5152 if not isinstance(renderer, dict):
5153 continue
5154 video_id = renderer.get('videoId')
5155 if not video_id:
5156 continue
5157 yield self._extract_video(renderer)
5159 def _extract_lockup_view_model(self, view_model):
5160 content_id = view_model.get('contentId')
5161 if not content_id:
5162 return
5163 content_type = view_model.get('contentType')
5164 if content_type not in ('LOCKUP_CONTENT_TYPE_PLAYLIST', 'LOCKUP_CONTENT_TYPE_PODCAST'):
5165 self.report_warning(
5166 f'Unsupported lockup view model content type "{content_type}"{bug_reports_message()}', only_once=True)
5167 return
5168 return self.url_result(
5169 f'https://www.youtube.com/playlist?list={content_id}', ie=YoutubeTabIE, video_id=content_id,
5170 title=traverse_obj(view_model, (
5171 'metadata', 'lockupMetadataViewModel', 'title', 'content', {str})),
5172 thumbnails=self._extract_thumbnails(view_model, (
5173 'contentImage', 'collectionThumbnailViewModel', 'primaryThumbnail', 'thumbnailViewModel', 'image'), final_key='sources'))
5175 def _rich_entries(self, rich_grid_renderer):
5176 if lockup_view_model := traverse_obj(rich_grid_renderer, ('content', 'lockupViewModel', {dict})):
5177 if entry := self._extract_lockup_view_model(lockup_view_model):
5178 yield entry
5179 return
5180 renderer = traverse_obj(
5181 rich_grid_renderer,
5182 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer', 'shortsLockupViewModel'), any)) or {}
5183 video_id = renderer.get('videoId')
5184 if video_id:
5185 yield self._extract_video(renderer)
5186 return
5187 playlist_id = renderer.get('playlistId')
5188 if playlist_id:
5189 yield self.url_result(
5190 f'https://www.youtube.com/playlist?list={playlist_id}',
5191 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5192 video_title=self._get_text(renderer, 'title'))
5193 return
5194 # shortsLockupViewModel extraction
5195 entity_id = renderer.get('entityId')
5196 if entity_id:
5197 video_id = traverse_obj(renderer, ('onTap', 'innertubeCommand', 'reelWatchEndpoint', 'videoId', {str}))
5198 if not video_id:
5199 return
5200 yield self.url_result(
5201 f'https://www.youtube.com/shorts/{video_id}',
5202 ie=YoutubeIE, video_id=video_id,
5203 **traverse_obj(renderer, ('overlayMetadata', {
5204 'title': ('primaryText', 'content', {str}),
5205 'view_count': ('secondaryText', 'content', {parse_count}),
5206 })),
5207 thumbnails=self._extract_thumbnails(renderer, 'thumbnail', final_key='sources'))
5208 return
5210 def _video_entry(self, video_renderer):
5211 video_id = video_renderer.get('videoId')
5212 if video_id:
5213 return self._extract_video(video_renderer)
5215 def _hashtag_tile_entry(self, hashtag_tile_renderer):
5216 url = urljoin('https://youtube.com', traverse_obj(
5217 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
5218 if url:
5219 return self.url_result(
5220 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
5222 def _post_thread_entries(self, post_thread_renderer):
5223 post_renderer = try_get(
5224 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
5225 if not post_renderer:
5226 return
5227 # video attachment
5228 video_renderer = try_get(
5229 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
5230 video_id = video_renderer.get('videoId')
5231 if video_id:
5232 entry = self._extract_video(video_renderer)
5233 if entry:
5234 yield entry
5235 # playlist attachment
5236 playlist_id = try_get(
5237 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
5238 if playlist_id:
5239 yield self.url_result(
5240 f'https://www.youtube.com/playlist?list={playlist_id}',
5241 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
5242 # inline video links
5243 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
5244 for run in runs:
5245 if not isinstance(run, dict):
5246 continue
5247 ep_url = try_get(
5248 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
5249 if not ep_url:
5250 continue
5251 if not YoutubeIE.suitable(ep_url):
5252 continue
5253 ep_video_id = YoutubeIE._match_id(ep_url)
5254 if video_id == ep_video_id:
5255 continue
5256 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
5258 def _post_thread_continuation_entries(self, post_thread_continuation):
5259 contents = post_thread_continuation.get('contents')
5260 if not isinstance(contents, list):
5261 return
5262 for content in contents:
5263 renderer = content.get('backstagePostThreadRenderer')
5264 if isinstance(renderer, dict):
5265 yield from self._post_thread_entries(renderer)
5266 continue
5267 renderer = content.get('videoRenderer')
5268 if isinstance(renderer, dict):
5269 yield self._video_entry(renderer)
5271 r''' # unused
5272 def _rich_grid_entries(self, contents):
5273 for content in contents:
5274 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
5275 if video_renderer:
5276 entry = self._video_entry(video_renderer)
5277 if entry:
5278 yield entry
5281 def _report_history_entries(self, renderer):
5282 for url in traverse_obj(renderer, (
5283 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
5284 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
5285 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
5286 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
5288 def _extract_entries(self, parent_renderer, continuation_list):
5289 # continuation_list is modified in-place with continuation_list = [continuation_token]
5290 continuation_list[:] = [None]
5291 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
5292 for content in contents:
5293 if not isinstance(content, dict):
5294 continue
5295 is_renderer = traverse_obj(
5296 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
5297 expected_type=dict)
5298 if not is_renderer:
5299 if content.get('richItemRenderer'):
5300 for entry in self._rich_entries(content['richItemRenderer']):
5301 yield entry
5302 continuation_list[0] = self._extract_continuation(parent_renderer)
5303 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
5304 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
5305 yield from self._report_history_entries(table)
5306 continuation_list[0] = self._extract_continuation(table)
5307 continue
5309 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
5310 for isr_content in isr_contents:
5311 if not isinstance(isr_content, dict):
5312 continue
5314 known_renderers = {
5315 'playlistVideoListRenderer': self._playlist_entries,
5316 'gridRenderer': self._grid_entries,
5317 'reelShelfRenderer': self._grid_entries,
5318 'shelfRenderer': self._shelf_entries,
5319 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
5320 'backstagePostThreadRenderer': self._post_thread_entries,
5321 'videoRenderer': lambda x: [self._video_entry(x)],
5322 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
5323 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
5324 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
5325 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
5326 'lockupViewModel': lambda x: [self._extract_lockup_view_model(x)],
5328 for key, renderer in isr_content.items():
5329 if key not in known_renderers:
5330 continue
5331 for entry in known_renderers[key](renderer):
5332 if entry:
5333 yield entry
5334 continuation_list[0] = self._extract_continuation(renderer)
5335 break
5337 if not continuation_list[0]:
5338 continuation_list[0] = self._extract_continuation(is_renderer)
5340 if not continuation_list[0]:
5341 continuation_list[0] = self._extract_continuation(parent_renderer)
5343 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
5344 continuation_list = [None]
5345 extract_entries = lambda x: self._extract_entries(x, continuation_list)
5346 tab_content = try_get(tab, lambda x: x['content'], dict)
5347 if not tab_content:
5348 return
5349 parent_renderer = (
5350 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
5351 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
5352 yield from extract_entries(parent_renderer)
5353 continuation = continuation_list[0]
5354 seen_continuations = set()
5355 for page_num in itertools.count(1):
5356 if not continuation:
5357 break
5358 continuation_token = continuation.get('continuation')
5359 if continuation_token is not None and continuation_token in seen_continuations:
5360 self.write_debug('Detected YouTube feed looping - assuming end of feed.')
5361 break
5362 seen_continuations.add(continuation_token)
5363 headers = self.generate_api_headers(
5364 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
5365 response = self._extract_response(
5366 item_id=f'{item_id} page {page_num}',
5367 query=continuation, headers=headers, ytcfg=ytcfg,
5368 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
5370 if not response:
5371 break
5372 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
5373 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
5374 visitor_data = self._extract_visitor_data(response) or visitor_data
5376 known_renderers = {
5377 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
5378 'gridPlaylistRenderer': (self._grid_entries, 'items'),
5379 'gridVideoRenderer': (self._grid_entries, 'items'),
5380 'gridChannelRenderer': (self._grid_entries, 'items'),
5381 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
5382 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
5383 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
5384 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
5385 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
5386 'playlistVideoListContinuation': (self._playlist_entries, None),
5387 'gridContinuation': (self._grid_entries, None),
5388 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
5389 'sectionListContinuation': (extract_entries, None), # for feeds
5392 continuation_items = traverse_obj(response, (
5393 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
5394 'appendContinuationItemsAction', 'continuationItems',
5395 ), 'continuationContents', get_all=False)
5396 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
5398 video_items_renderer = None
5399 for key in continuation_item:
5400 if key not in known_renderers:
5401 continue
5402 func, parent_key = known_renderers[key]
5403 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
5404 continuation_list = [None]
5405 yield from func(video_items_renderer)
5406 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
5408 if not video_items_renderer:
5409 break
5411 @staticmethod
5412 def _extract_selected_tab(tabs, fatal=True):
5413 for tab_renderer in tabs:
5414 if tab_renderer.get('selected'):
5415 return tab_renderer
5416 if fatal:
5417 raise ExtractorError('Unable to find selected tab')
5419 @staticmethod
5420 def _extract_tab_renderers(response):
5421 return traverse_obj(
5422 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
5424 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
5425 metadata = self._extract_metadata_from_tabs(item_id, data)
5427 selected_tab = self._extract_selected_tab(tabs)
5428 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
5429 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
5431 return self.playlist_result(
5432 self._entries(
5433 selected_tab, metadata['id'], ytcfg,
5434 self._extract_account_syncid(ytcfg, data),
5435 self._extract_visitor_data(data, ytcfg)),
5436 **metadata)
5438 def _extract_metadata_from_tabs(self, item_id, data):
5439 info = {'id': item_id}
5441 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
5442 if metadata_renderer:
5443 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
5444 ('channelUrl', {self.ucid_from_url}))
5445 info.update({
5446 'channel': metadata_renderer.get('title'),
5447 'channel_id': channel_id,
5449 if info['channel_id']:
5450 info['id'] = info['channel_id']
5451 else:
5452 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5454 # pageHeaderViewModel slow rollout began April 2024
5455 page_header_view_model = traverse_obj(data, (
5456 'header', 'pageHeaderRenderer', 'content', 'pageHeaderViewModel', {dict}))
5458 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5459 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5460 def _get_uncropped(url):
5461 return url_or_none((url or '').split('=')[0] + '=s0')
5463 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5464 if avatar_thumbnails:
5465 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5466 if uncropped_avatar:
5467 avatar_thumbnails.append({
5468 'url': uncropped_avatar,
5469 'id': 'avatar_uncropped',
5470 'preference': 1,
5473 channel_banners = (
5474 self._extract_thumbnails(data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5475 or self._extract_thumbnails(
5476 page_header_view_model, ('banner', 'imageBannerViewModel', 'image'), final_key='sources'))
5477 for banner in channel_banners:
5478 banner['preference'] = -10
5480 if channel_banners:
5481 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5482 if uncropped_banner:
5483 channel_banners.append({
5484 'url': uncropped_banner,
5485 'id': 'banner_uncropped',
5486 'preference': -5,
5489 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5490 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5491 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5493 primary_thumbnails = self._extract_thumbnails(
5494 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5495 playlist_thumbnails = self._extract_thumbnails(
5496 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5498 info.update({
5499 'title': (traverse_obj(metadata_renderer, 'title')
5500 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5501 or info['id']),
5502 'availability': self._extract_availability(data),
5503 'channel_follower_count': (
5504 self._get_count(data, ('header', ..., 'subscriberCountText'))
5505 or traverse_obj(page_header_view_model, (
5506 'metadata', 'contentMetadataViewModel', 'metadataRows', ..., 'metadataParts',
5507 lambda _, v: 'subscribers' in v['text']['content'], 'text', 'content', {parse_count}, any))),
5508 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5509 'tags': (traverse_obj(data, ('microformat', 'microformatDataRenderer', 'tags', ..., {str}))
5510 or traverse_obj(metadata_renderer, ('keywords', {lambda x: x and shlex.split(x)}, ...))),
5511 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5514 channel_handle = (
5515 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5516 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5518 if channel_handle:
5519 info.update({
5520 'uploader_id': channel_handle,
5521 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5524 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5525 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5526 info['channel_is_verified'] = True
5527 # Playlist stats is a text runs array containing [video count, view count, last updated].
5528 # last updated or (view count and last updated) may be missing.
5529 playlist_stats = get_first(
5530 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5532 last_updated_unix = self._parse_time_text(
5533 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5534 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5535 info['modified_date'] = strftime_or_none(last_updated_unix)
5537 info['view_count'] = self._get_count(playlist_stats, 1)
5538 if info['view_count'] is None: # 0 is allowed
5539 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5540 if info['view_count'] is None:
5541 info['view_count'] = self._get_count(data, (
5542 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5543 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5545 info['playlist_count'] = self._get_count(playlist_stats, 0)
5546 if info['playlist_count'] is None: # 0 is allowed
5547 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5549 if not info.get('channel_id'):
5550 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5551 if not owner: # Deprecated
5552 owner = traverse_obj(
5553 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5554 ('videoOwner', 'videoOwnerRenderer', 'title'))
5555 owner_text = self._get_text(owner)
5556 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5557 info.update({
5558 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5559 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5560 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl'))),
5563 info.update({
5564 'uploader': info['channel'],
5565 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5566 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5569 return info
5571 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5572 first_id = last_id = response = None
5573 for page_num in itertools.count(1):
5574 videos = list(self._playlist_entries(playlist))
5575 if not videos:
5576 return
5577 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5578 if start >= len(videos):
5579 return
5580 yield from videos[start:]
5581 first_id = first_id or videos[0]['id']
5582 last_id = videos[-1]['id']
5583 watch_endpoint = try_get(
5584 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5585 headers = self.generate_api_headers(
5586 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5587 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5588 query = {
5589 'playlistId': playlist_id,
5590 'videoId': watch_endpoint.get('videoId') or last_id,
5591 'index': watch_endpoint.get('index') or len(videos),
5592 'params': watch_endpoint.get('params') or 'OAE%3D',
5594 response = self._extract_response(
5595 item_id=f'{playlist_id} page {page_num}',
5596 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5597 check_get_keys='contents',
5599 playlist = try_get(
5600 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5602 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5603 title = playlist.get('title') or try_get(
5604 data, lambda x: x['titleText']['simpleText'], str)
5605 playlist_id = playlist.get('playlistId') or item_id
5607 # Delegating everything except mix playlists to regular tab-based playlist URL
5608 playlist_url = urljoin(url, try_get(
5609 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5610 str))
5612 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5613 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5614 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5616 if playlist_url and playlist_url != url and not is_known_unviewable:
5617 return self.url_result(
5618 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5619 video_title=title)
5621 return self.playlist_result(
5622 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5623 playlist_id=playlist_id, playlist_title=title)
5625 def _extract_availability(self, data):
5627 Gets the availability of a given playlist/tab.
5628 Note: Unless YouTube tells us explicitly, we do not assume it is public
5629 @param data: response
5631 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5632 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5633 player_header_privacy = playlist_header_renderer.get('privacy')
5635 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5637 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5638 privacy_setting_icon = get_first(
5639 (playlist_header_renderer, sidebar_renderer),
5640 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5641 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5642 expected_type=str)
5644 microformats_is_unlisted = traverse_obj(
5645 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5647 return (
5648 'public' if (
5649 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5650 or player_header_privacy == 'PUBLIC'
5651 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5652 else self._availability(
5653 is_private=(
5654 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5655 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5656 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5657 is_unlisted=(
5658 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5659 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5660 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5661 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5662 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5663 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5664 needs_auth=False))
5666 @staticmethod
5667 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5668 sidebar_renderer = try_get(
5669 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5670 for item in sidebar_renderer:
5671 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5672 if renderer:
5673 return renderer
5675 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5677 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5679 is_playlist = bool(traverse_obj(
5680 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5681 if not is_playlist:
5682 return
5683 headers = self.generate_api_headers(
5684 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5685 visitor_data=self._extract_visitor_data(data, ytcfg))
5686 query = {
5687 'params': 'wgYCCAA=',
5688 'browseId': f'VL{item_id}',
5690 return self._extract_response(
5691 item_id=item_id, headers=headers, query=query,
5692 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5693 note='Redownloading playlist API JSON with unavailable videos')
5695 @functools.cached_property
5696 def skip_webpage(self):
5697 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5699 def _extract_webpage(self, url, item_id, fatal=True):
5700 webpage, data = None, None
5701 for retry in self.RetryManager(fatal=fatal):
5702 try:
5703 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5704 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5705 except ExtractorError as e:
5706 if isinstance(e.cause, network_exceptions):
5707 if not isinstance(e.cause, HTTPError) or e.cause.status not in (403, 429):
5708 retry.error = e
5709 continue
5710 self._error_or_warning(e, fatal=fatal)
5711 break
5713 try:
5714 self._extract_and_report_alerts(data)
5715 except ExtractorError as e:
5716 self._error_or_warning(e, fatal=fatal)
5717 break
5719 # Sometimes youtube returns a webpage with incomplete ytInitialData
5720 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5721 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5722 retry.error = ExtractorError('Incomplete yt initial data received')
5723 data = None
5724 continue
5726 return webpage, data
5728 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5729 """Use if failed to extract ytcfg (and data) from initial webpage"""
5730 if not ytcfg and self.is_authenticated:
5731 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5732 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5733 raise ExtractorError(
5734 f'{msg}. If you are not downloading private content, or '
5735 'your cookies are only for the first account and channel,'
5736 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5737 expected=True)
5738 self.report_warning(msg, only_once=True)
5740 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5741 data = None
5742 if not self.skip_webpage:
5743 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5744 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5745 # Reject webpage data if redirected to home page without explicitly requesting
5746 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5747 if (url != 'https://www.youtube.com/feed/recommended'
5748 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5749 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5750 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5751 if fatal:
5752 raise ExtractorError(msg, expected=True)
5753 self.report_warning(msg, only_once=True)
5754 if not data:
5755 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5756 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5757 return data, ytcfg
5759 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5760 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5761 resolve_response = self._extract_response(
5762 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5763 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5764 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5765 for ep_key, ep in endpoints.items():
5766 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5767 if params:
5768 return self._extract_response(
5769 item_id=item_id, query=params, ep=ep, headers=headers,
5770 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5771 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5772 err_note = 'Failed to resolve url (does the playlist exist?)'
5773 if fatal:
5774 raise ExtractorError(err_note, expected=True)
5775 self.report_warning(err_note, item_id)
5777 _SEARCH_PARAMS = None
5779 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5780 data = {'query': query}
5781 if params is NO_DEFAULT:
5782 params = self._SEARCH_PARAMS
5783 if params:
5784 data['params'] = params
5786 content_keys = (
5787 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5788 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5789 # ytmusic search
5790 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5791 ('continuationContents', ),
5793 display_id = f'query "{query}"'
5794 check_get_keys = tuple({keys[0] for keys in content_keys})
5795 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5796 self._report_playlist_authcheck(ytcfg, fatal=False)
5798 continuation_list = [None]
5799 search = None
5800 for page_num in itertools.count(1):
5801 data.update(continuation_list[0] or {})
5802 headers = self.generate_api_headers(
5803 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5804 search = self._extract_response(
5805 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5806 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5807 slr_contents = traverse_obj(search, *content_keys)
5808 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5809 if not continuation_list[0]:
5810 break
5813 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5814 IE_DESC = 'YouTube Tabs'
5815 _VALID_URL = r'''(?x:
5816 https?://
5817 (?!consent\.)(?:\w+\.)?
5819 youtube(?:kids)?\.com|
5820 {invidious}
5823 (?P<channel_type>channel|c|user|browse)/|
5824 (?P<not_channel>
5825 feed/|hashtag/|
5826 (?:playlist|watch)\?.*?\blist=
5828 (?!(?:{reserved_names})\b) # Direct URLs
5830 (?P<id>[^/?\#&]+)
5831 )'''.format(
5832 reserved_names=YoutubeBaseInfoExtractor._RESERVED_NAMES,
5833 invidious='|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5835 IE_NAME = 'youtube:tab'
5837 _TESTS = [{
5838 'note': 'playlists, multipage',
5839 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5840 'playlist_mincount': 94,
5841 'info_dict': {
5842 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5843 'title': 'Igor Kleiner Ph.D. - Playlists',
5844 'description': 'md5:15d7dd9e333cb987907fcb0d604b233a',
5845 'uploader': 'Igor Kleiner Ph.D.',
5846 'uploader_id': '@IgorDataScience',
5847 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5848 'channel': 'Igor Kleiner Ph.D.',
5849 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5850 'tags': ['критическое мышление', 'наука просто', 'математика', 'анализ данных'],
5851 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5852 'channel_follower_count': int,
5854 }, {
5855 'note': 'playlists, multipage, different order',
5856 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5857 'playlist_mincount': 94,
5858 'info_dict': {
5859 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5860 'title': 'Igor Kleiner Ph.D. - Playlists',
5861 'description': 'md5:15d7dd9e333cb987907fcb0d604b233a',
5862 'uploader': 'Igor Kleiner Ph.D.',
5863 'uploader_id': '@IgorDataScience',
5864 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5865 'tags': ['критическое мышление', 'наука просто', 'математика', 'анализ данных'],
5866 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5867 'channel': 'Igor Kleiner Ph.D.',
5868 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5869 'channel_follower_count': int,
5871 }, {
5872 'note': 'playlists, series',
5873 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5874 'playlist_mincount': 5,
5875 'info_dict': {
5876 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5877 'title': '3Blue1Brown - Playlists',
5878 'description': 'md5:602e3789e6a0cb7d9d352186b720e395',
5879 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5880 'channel': '3Blue1Brown',
5881 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5882 'uploader_id': '@3blue1brown',
5883 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5884 'uploader': '3Blue1Brown',
5885 'tags': ['Mathematics'],
5886 'channel_follower_count': int,
5887 'channel_is_verified': True,
5889 }, {
5890 'note': 'playlists, singlepage',
5891 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5892 'playlist_mincount': 4,
5893 'info_dict': {
5894 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5895 'title': 'ThirstForScience - Playlists',
5896 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5897 'uploader': 'ThirstForScience',
5898 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5899 'uploader_id': '@ThirstForScience',
5900 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5901 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5902 'tags': 'count:12',
5903 'channel': 'ThirstForScience',
5904 'channel_follower_count': int,
5906 }, {
5907 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5908 'only_matching': True,
5909 }, {
5910 'note': 'basic, single video playlist',
5911 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5912 'info_dict': {
5913 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5914 'title': 'youtube-dl public playlist',
5915 'description': '',
5916 'tags': [],
5917 'view_count': int,
5918 'modified_date': '20201130',
5919 'channel': 'Sergey M.',
5920 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5921 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5922 'availability': 'public',
5923 'uploader': 'Sergey M.',
5924 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5925 'uploader_id': '@sergeym.6173',
5927 'playlist_count': 1,
5928 }, {
5929 'note': 'empty playlist',
5930 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5931 'info_dict': {
5932 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5933 'title': 'youtube-dl empty playlist',
5934 'tags': [],
5935 'channel': 'Sergey M.',
5936 'description': '',
5937 'modified_date': '20230921',
5938 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5939 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5940 'availability': 'unlisted',
5941 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5942 'uploader_id': '@sergeym.6173',
5943 'uploader': 'Sergey M.',
5945 'playlist_count': 0,
5946 }, {
5947 'note': 'Home tab',
5948 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5949 'info_dict': {
5950 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5951 'title': 'lex will - Home',
5952 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5953 'uploader': 'lex will',
5954 'uploader_id': '@lexwill718',
5955 'channel': 'lex will',
5956 'tags': ['bible', 'history', 'prophesy'],
5957 'uploader_url': 'https://www.youtube.com/@lexwill718',
5958 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5959 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5960 'channel_follower_count': int,
5962 'playlist_mincount': 2,
5963 }, {
5964 'note': 'Videos tab',
5965 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5966 'info_dict': {
5967 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5968 'title': 'lex will - Videos',
5969 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5970 'uploader': 'lex will',
5971 'uploader_id': '@lexwill718',
5972 'tags': ['bible', 'history', 'prophesy'],
5973 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5974 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5975 'uploader_url': 'https://www.youtube.com/@lexwill718',
5976 'channel': 'lex will',
5977 'channel_follower_count': int,
5979 'playlist_mincount': 975,
5980 }, {
5981 'note': 'Videos tab, sorted by popular',
5982 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5983 'info_dict': {
5984 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5985 'title': 'lex will - Videos',
5986 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5987 'uploader': 'lex will',
5988 'uploader_id': '@lexwill718',
5989 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5990 'uploader_url': 'https://www.youtube.com/@lexwill718',
5991 'channel': 'lex will',
5992 'tags': ['bible', 'history', 'prophesy'],
5993 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5994 'channel_follower_count': int,
5996 'playlist_mincount': 199,
5997 }, {
5998 'note': 'Playlists tab',
5999 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
6000 'info_dict': {
6001 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
6002 'title': 'lex will - Playlists',
6003 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
6004 'uploader': 'lex will',
6005 'uploader_id': '@lexwill718',
6006 'uploader_url': 'https://www.youtube.com/@lexwill718',
6007 'channel': 'lex will',
6008 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
6009 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
6010 'tags': ['bible', 'history', 'prophesy'],
6011 'channel_follower_count': int,
6013 'playlist_mincount': 17,
6014 }, {
6015 'note': 'Community tab',
6016 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
6017 'info_dict': {
6018 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
6019 'title': 'lex will - Community',
6020 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
6021 'channel': 'lex will',
6022 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
6023 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
6024 'tags': ['bible', 'history', 'prophesy'],
6025 'channel_follower_count': int,
6026 'uploader_url': 'https://www.youtube.com/@lexwill718',
6027 'uploader_id': '@lexwill718',
6028 'uploader': 'lex will',
6030 'playlist_mincount': 18,
6031 }, {
6032 'note': 'Channels tab',
6033 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
6034 'info_dict': {
6035 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
6036 'title': 'lex will - Channels',
6037 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
6038 'channel': 'lex will',
6039 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
6040 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
6041 'tags': ['bible', 'history', 'prophesy'],
6042 'channel_follower_count': int,
6043 'uploader_url': 'https://www.youtube.com/@lexwill718',
6044 'uploader_id': '@lexwill718',
6045 'uploader': 'lex will',
6047 'playlist_mincount': 12,
6048 }, {
6049 'note': 'Search tab',
6050 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
6051 'playlist_mincount': 40,
6052 'info_dict': {
6053 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6054 'title': '3Blue1Brown - Search - linear algebra',
6055 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
6056 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6057 'tags': ['Mathematics'],
6058 'channel': '3Blue1Brown',
6059 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6060 'channel_follower_count': int,
6061 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6062 'uploader_id': '@3blue1brown',
6063 'uploader': '3Blue1Brown',
6064 'channel_is_verified': True,
6066 }, {
6067 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
6068 'only_matching': True,
6069 }, {
6070 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
6071 'only_matching': True,
6072 }, {
6073 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
6074 'only_matching': True,
6075 }, {
6076 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
6077 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
6078 'info_dict': {
6079 'title': '29C3: Not my department',
6080 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
6081 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
6082 'tags': [],
6083 'view_count': int,
6084 'modified_date': '20150605',
6085 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
6086 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
6087 'channel': 'Christiaan008',
6088 'availability': 'public',
6089 'uploader_id': '@ChRiStIaAn008',
6090 'uploader': 'Christiaan008',
6091 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
6093 'playlist_count': 96,
6094 }, {
6095 'note': 'Large playlist',
6096 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
6097 'info_dict': {
6098 'title': 'Uploads from Cauchemar',
6099 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
6100 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
6101 'tags': [],
6102 'modified_date': r're:\d{8}',
6103 'channel': 'Cauchemar',
6104 'view_count': int,
6105 'description': '',
6106 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
6107 'availability': 'public',
6108 'uploader_id': '@Cauchemar89',
6109 'uploader': 'Cauchemar',
6110 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
6112 'playlist_mincount': 1123,
6113 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6114 }, {
6115 'note': 'even larger playlist, 8832 videos',
6116 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
6117 'only_matching': True,
6118 }, {
6119 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
6120 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
6121 'info_dict': {
6122 'title': 'Uploads from Interstellar Movie',
6123 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
6124 'tags': [],
6125 'view_count': int,
6126 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
6127 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
6128 'channel': 'Interstellar Movie',
6129 'description': '',
6130 'modified_date': r're:\d{8}',
6131 'availability': 'public',
6132 'uploader_id': '@InterstellarMovie',
6133 'uploader': 'Interstellar Movie',
6134 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
6136 'playlist_mincount': 21,
6137 }, {
6138 'note': 'Playlist with "show unavailable videos" button',
6139 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
6140 'info_dict': {
6141 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
6142 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
6143 'view_count': int,
6144 'channel': 'Phim Siêu Nhân Nhật Bản',
6145 'tags': [],
6146 'description': '',
6147 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
6148 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
6149 'modified_date': r're:\d{8}',
6150 'availability': 'public',
6151 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
6152 'uploader_id': '@phimsieunhannhatban',
6153 'uploader': 'Phim Siêu Nhân Nhật Bản',
6155 'playlist_mincount': 200,
6156 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6157 }, {
6158 'note': 'Playlist with unavailable videos in page 7',
6159 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
6160 'info_dict': {
6161 'title': 'Uploads from BlankTV',
6162 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
6163 'channel': 'BlankTV',
6164 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
6165 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
6166 'view_count': int,
6167 'tags': [],
6168 'modified_date': r're:\d{8}',
6169 'description': '',
6170 'availability': 'public',
6171 'uploader_id': '@blanktv',
6172 'uploader': 'BlankTV',
6173 'uploader_url': 'https://www.youtube.com/@blanktv',
6175 'playlist_mincount': 1000,
6176 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6177 }, {
6178 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
6179 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
6180 'info_dict': {
6181 'title': 'Data Analysis with Dr Mike Pound',
6182 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
6183 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
6184 'tags': [],
6185 'view_count': int,
6186 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
6187 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
6188 'channel': 'Computerphile',
6189 'availability': 'public',
6190 'modified_date': '20190712',
6191 'uploader_id': '@Computerphile',
6192 'uploader': 'Computerphile',
6193 'uploader_url': 'https://www.youtube.com/@Computerphile',
6195 'playlist_mincount': 11,
6196 }, {
6197 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
6198 'only_matching': True,
6199 }, {
6200 'note': 'Playlist URL that does not actually serve a playlist',
6201 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
6202 'info_dict': {
6203 'id': 'FqZTN594JQw',
6204 'ext': 'webm',
6205 'title': "Smiley's People 01 detective, Adventure Series, Action",
6206 'upload_date': '20150526',
6207 'license': 'Standard YouTube License',
6208 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
6209 'categories': ['People & Blogs'],
6210 'tags': list,
6211 'view_count': int,
6212 'like_count': int,
6214 'params': {
6215 'skip_download': True,
6217 'skip': 'This video is not available.',
6218 'add_ie': [YoutubeIE.ie_key()],
6219 }, {
6220 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
6221 'only_matching': True,
6222 }, {
6223 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
6224 'only_matching': True,
6225 }, {
6226 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
6227 'info_dict': {
6228 'id': 'hGkQjiJLjWQ', # This will keep changing
6229 'ext': 'mp4',
6230 'title': str,
6231 'upload_date': r're:\d{8}',
6232 'description': str,
6233 'categories': ['News & Politics'],
6234 'tags': list,
6235 'like_count': int,
6236 'release_timestamp': int,
6237 'channel': 'Sky News',
6238 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
6239 'age_limit': 0,
6240 'view_count': int,
6241 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
6242 'playable_in_embed': True,
6243 'release_date': r're:\d+',
6244 'availability': 'public',
6245 'live_status': 'is_live',
6246 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
6247 'channel_follower_count': int,
6248 'concurrent_view_count': int,
6249 'uploader_url': 'https://www.youtube.com/@SkyNews',
6250 'uploader_id': '@SkyNews',
6251 'uploader': 'Sky News',
6252 'channel_is_verified': True,
6254 'params': {
6255 'skip_download': True,
6257 'expected_warnings': ['Ignoring subtitle tracks found in '],
6258 }, {
6259 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
6260 'info_dict': {
6261 'id': 'a48o2S1cPoo',
6262 'ext': 'mp4',
6263 'title': 'The Young Turks - Live Main Show',
6264 'upload_date': '20150715',
6265 'license': 'Standard YouTube License',
6266 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
6267 'categories': ['News & Politics'],
6268 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
6269 'like_count': int,
6271 'params': {
6272 'skip_download': True,
6274 'only_matching': True,
6275 }, {
6276 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
6277 'only_matching': True,
6278 }, {
6279 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
6280 'only_matching': True,
6281 }, {
6282 'note': 'A channel that is not live. Should raise error',
6283 'url': 'https://www.youtube.com/user/numberphile/live',
6284 'only_matching': True,
6285 }, {
6286 'url': 'https://www.youtube.com/feed/trending',
6287 'only_matching': True,
6288 }, {
6289 'url': 'https://www.youtube.com/feed/library',
6290 'only_matching': True,
6291 }, {
6292 'url': 'https://www.youtube.com/feed/history',
6293 'only_matching': True,
6294 }, {
6295 'url': 'https://www.youtube.com/feed/subscriptions',
6296 'only_matching': True,
6297 }, {
6298 'url': 'https://www.youtube.com/feed/watch_later',
6299 'only_matching': True,
6300 }, {
6301 'note': 'Recommended - redirects to home page.',
6302 'url': 'https://www.youtube.com/feed/recommended',
6303 'only_matching': True,
6304 }, {
6305 'note': 'inline playlist with not always working continuations',
6306 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
6307 'only_matching': True,
6308 }, {
6309 'url': 'https://www.youtube.com/course',
6310 'only_matching': True,
6311 }, {
6312 'url': 'https://www.youtube.com/zsecurity',
6313 'only_matching': True,
6314 }, {
6315 'url': 'http://www.youtube.com/NASAgovVideo/videos',
6316 'only_matching': True,
6317 }, {
6318 'url': 'https://www.youtube.com/TheYoungTurks/live',
6319 'only_matching': True,
6320 }, {
6321 'url': 'https://www.youtube.com/hashtag/cctv9',
6322 'info_dict': {
6323 'id': 'cctv9',
6324 'title': 'cctv9 - All',
6325 'tags': [],
6327 'playlist_mincount': 300, # not consistent but should be over 300
6328 }, {
6329 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
6330 'only_matching': True,
6331 }, {
6332 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
6333 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
6334 'only_matching': True,
6335 }, {
6336 'note': '/browse/ should redirect to /channel/',
6337 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
6338 'only_matching': True,
6339 }, {
6340 'note': 'VLPL, should redirect to playlist?list=PL...',
6341 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
6342 'info_dict': {
6343 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
6344 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
6345 'title': 'NCS : All Releases 💿',
6346 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
6347 'modified_date': r're:\d{8}',
6348 'view_count': int,
6349 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
6350 'tags': [],
6351 'channel': 'NoCopyrightSounds',
6352 'availability': 'public',
6353 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
6354 'uploader': 'NoCopyrightSounds',
6355 'uploader_id': '@NoCopyrightSounds',
6357 'playlist_mincount': 166,
6358 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
6359 }, {
6360 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
6361 'note': 'Topic, should redirect to playlist?list=UU...',
6362 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6363 'info_dict': {
6364 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6365 'title': 'Uploads from Royalty Free Music - Topic',
6366 'tags': [],
6367 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6368 'channel': 'Royalty Free Music - Topic',
6369 'view_count': int,
6370 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6371 'modified_date': r're:\d{8}',
6372 'description': '',
6373 'availability': 'public',
6374 'uploader': 'Royalty Free Music - Topic',
6376 'playlist_mincount': 101,
6377 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6378 }, {
6379 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
6380 # Treat as a general feed
6381 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
6382 'info_dict': {
6383 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
6384 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
6385 'tags': [],
6387 'playlist_mincount': 9,
6388 }, {
6389 'note': 'Youtube music Album',
6390 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
6391 'info_dict': {
6392 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
6393 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
6394 'tags': [],
6395 'view_count': int,
6396 'description': '',
6397 'availability': 'unlisted',
6398 'modified_date': r're:\d{8}',
6400 'playlist_count': 50,
6401 'expected_warnings': ['YouTube Music is not directly supported'],
6402 }, {
6403 'note': 'unlisted single video playlist',
6404 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
6405 'info_dict': {
6406 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
6407 'title': 'yt-dlp unlisted playlist test',
6408 'availability': 'unlisted',
6409 'tags': [],
6410 'modified_date': '20220418',
6411 'channel': 'colethedj',
6412 'view_count': int,
6413 'description': '',
6414 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
6415 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
6416 'uploader_url': 'https://www.youtube.com/@colethedj1894',
6417 'uploader_id': '@colethedj1894',
6418 'uploader': 'colethedj',
6420 'playlist': [{
6421 'info_dict': {
6422 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
6423 'id': 'BaW_jenozKc',
6424 '_type': 'url',
6425 'ie_key': 'Youtube',
6426 'duration': 10,
6427 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
6428 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
6429 'view_count': int,
6430 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
6431 'channel': 'Philipp Hagemeister',
6432 'uploader_id': '@PhilippHagemeister',
6433 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
6434 'uploader': 'Philipp Hagemeister',
6437 'playlist_count': 1,
6438 'params': {'extract_flat': True},
6439 }, {
6440 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
6441 'url': 'https://www.youtube.com/feed/recommended',
6442 'info_dict': {
6443 'id': 'recommended',
6444 'title': 'recommended',
6445 'tags': [],
6447 'playlist_mincount': 50,
6448 'params': {
6449 'skip_download': True,
6450 'extractor_args': {'youtubetab': {'skip': ['webpage']}},
6452 }, {
6453 'note': 'API Fallback: /videos tab, sorted by oldest first',
6454 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
6455 'info_dict': {
6456 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6457 'title': 'Cody\'sLab - Videos',
6458 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6459 'channel': 'Cody\'sLab',
6460 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6461 'tags': [],
6462 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6463 'channel_follower_count': int,
6465 'playlist_mincount': 650,
6466 'params': {
6467 'skip_download': True,
6468 'extractor_args': {'youtubetab': {'skip': ['webpage']}},
6470 'skip': 'Query for sorting no longer works',
6471 }, {
6472 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6473 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6474 'info_dict': {
6475 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6476 'title': 'Uploads from Royalty Free Music - Topic',
6477 'modified_date': r're:\d{8}',
6478 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6479 'description': '',
6480 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6481 'tags': [],
6482 'channel': 'Royalty Free Music - Topic',
6483 'view_count': int,
6484 'availability': 'public',
6485 'uploader': 'Royalty Free Music - Topic',
6487 'playlist_mincount': 101,
6488 'params': {
6489 'skip_download': True,
6490 'extractor_args': {'youtubetab': {'skip': ['webpage']}},
6492 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6493 }, {
6494 'note': 'non-standard redirect to regional channel',
6495 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6496 'only_matching': True,
6497 }, {
6498 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6499 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6500 'info_dict': {
6501 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6502 'modified_date': '20220407',
6503 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6504 'tags': [],
6505 'availability': 'unlisted',
6506 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6507 'channel': 'pukkandan',
6508 'description': 'Test for collaborative playlist',
6509 'title': 'yt-dlp test - collaborative playlist',
6510 'view_count': int,
6511 'uploader_url': 'https://www.youtube.com/@pukkandan',
6512 'uploader_id': '@pukkandan',
6513 'uploader': 'pukkandan',
6515 'playlist_mincount': 2,
6516 }, {
6517 'note': 'translated tab name',
6518 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6519 'info_dict': {
6520 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6521 'tags': [],
6522 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6523 'description': 'test description',
6524 'title': 'cole-dlp-test-acc - 再生リスト',
6525 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6526 'channel': 'cole-dlp-test-acc',
6527 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6528 'uploader_id': '@coletdjnz',
6529 'uploader': 'cole-dlp-test-acc',
6531 'playlist_mincount': 1,
6532 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6533 'expected_warnings': ['Preferring "ja"'],
6534 }, {
6535 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6536 'note': 'preferred lang set with playlist with translated video titles',
6537 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6538 'info_dict': {
6539 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6540 'tags': [],
6541 'view_count': int,
6542 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6543 'channel': 'cole-dlp-test-acc',
6544 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6545 'description': 'test',
6546 'title': 'dlp test playlist',
6547 'availability': 'public',
6548 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6549 'uploader_id': '@coletdjnz',
6550 'uploader': 'cole-dlp-test-acc',
6552 'playlist_mincount': 1,
6553 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6554 'expected_warnings': ['Preferring "ja"'],
6555 }, {
6556 # shorts audio pivot for 2GtVksBMYFM.
6557 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6558 'info_dict': {
6559 'id': 'sfv_audio_pivot',
6560 'title': 'sfv_audio_pivot',
6561 'tags': [],
6563 'playlist_mincount': 50,
6565 }, {
6566 # Channel with a real live tab (not to be mistaken with streams tab)
6567 # Do not treat like it should redirect to live stream
6568 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6569 'info_dict': {
6570 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6571 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6572 'tags': [],
6574 'playlist_mincount': 20,
6575 }, {
6576 # Tab name is not the same as tab id
6577 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6578 'info_dict': {
6579 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6580 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6581 'tags': [],
6583 'playlist_mincount': 8,
6584 }, {
6585 # Home tab id is literally home. Not to get mistaken with featured
6586 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6587 'info_dict': {
6588 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6589 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6590 'tags': [],
6592 'playlist_mincount': 8,
6593 }, {
6594 # Should get three playlists for videos, shorts and streams tabs
6595 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6596 'info_dict': {
6597 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6598 'title': 'Polka Ch. 尾丸ポルカ',
6599 'channel_follower_count': int,
6600 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6601 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6602 'description': 'md5:49809d8bf9da539bc48ed5d1f83c33f2',
6603 'channel': 'Polka Ch. 尾丸ポルカ',
6604 'tags': 'count:35',
6605 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6606 'uploader': 'Polka Ch. 尾丸ポルカ',
6607 'uploader_id': '@OmaruPolka',
6608 'channel_is_verified': True,
6610 'playlist_count': 3,
6611 }, {
6612 # Shorts tab with channel with handle
6613 # TODO: fix channel description
6614 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6615 'info_dict': {
6616 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6617 'title': 'Not Just Bikes - Shorts',
6618 'tags': 'count:10',
6619 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6620 'description': 'md5:5e82545b3a041345927a92d0585df247',
6621 'channel_follower_count': int,
6622 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6623 'channel': 'Not Just Bikes',
6624 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6625 'uploader': 'Not Just Bikes',
6626 'uploader_id': '@NotJustBikes',
6627 'channel_is_verified': True,
6629 'playlist_mincount': 10,
6630 }, {
6631 # Streams tab
6632 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6633 'info_dict': {
6634 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6635 'title': '中村悠一 - Live',
6636 'tags': 'count:7',
6637 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6638 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6639 'channel': '中村悠一',
6640 'channel_follower_count': int,
6641 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6642 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6643 'uploader_id': '@Yuichi-Nakamura',
6644 'uploader': '中村悠一',
6646 'playlist_mincount': 60,
6647 }, {
6648 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6649 # See test_youtube_lists
6650 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6651 'only_matching': True,
6652 }, {
6653 # No uploads and no UCID given. Should fail with no uploads error
6654 # See test_youtube_lists
6655 'url': 'https://www.youtube.com/news',
6656 'only_matching': True,
6657 }, {
6658 # No videos tab but has a shorts tab
6659 'url': 'https://www.youtube.com/c/TKFShorts',
6660 'info_dict': {
6661 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6662 'title': 'Shorts Break - Shorts',
6663 'tags': 'count:48',
6664 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6665 'channel': 'Shorts Break',
6666 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6667 'channel_follower_count': int,
6668 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6669 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6670 'uploader': 'Shorts Break',
6671 'uploader_id': '@ShortsBreak_Official',
6673 'playlist_mincount': 30,
6674 }, {
6675 # Trending Now Tab. tab id is empty
6676 'url': 'https://www.youtube.com/feed/trending',
6677 'info_dict': {
6678 'id': 'trending',
6679 'title': 'trending - Now',
6680 'tags': [],
6682 'playlist_mincount': 30,
6683 }, {
6684 # Trending Gaming Tab. tab id is empty
6685 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6686 'info_dict': {
6687 'id': 'trending',
6688 'title': 'trending - Gaming',
6689 'tags': [],
6691 'playlist_mincount': 30,
6692 }, {
6693 # Shorts url result in shorts tab
6694 # TODO: Fix channel id extraction
6695 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6696 'info_dict': {
6697 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6698 'title': 'cole-dlp-test-acc - Shorts',
6699 'channel': 'cole-dlp-test-acc',
6700 'description': 'test description',
6701 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6702 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6703 'tags': [],
6704 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6705 'uploader_id': '@coletdjnz',
6706 'uploader': 'cole-dlp-test-acc',
6708 'playlist': [{
6709 'info_dict': {
6710 # Channel data is not currently available for short renderers (as of 2023-03-01)
6711 '_type': 'url',
6712 'ie_key': 'Youtube',
6713 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6714 'id': 'sSM9J5YH_60',
6715 'title': 'SHORT short',
6716 'view_count': int,
6717 'thumbnails': list,
6720 'params': {'extract_flat': True},
6721 }, {
6722 # Live video status should be extracted
6723 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6724 'info_dict': {
6725 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6726 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO: should be Minecraft - Live or Minecraft - Topic - Live
6727 'tags': [],
6729 'playlist': [{
6730 'info_dict': {
6731 '_type': 'url',
6732 'ie_key': 'Youtube',
6733 'url': 'startswith:https://www.youtube.com/watch?v=',
6734 'id': str,
6735 'title': str,
6736 'live_status': 'is_live',
6737 'channel_id': str,
6738 'channel_url': str,
6739 'concurrent_view_count': int,
6740 'channel': str,
6741 'uploader': str,
6742 'uploader_url': str,
6743 'uploader_id': str,
6744 'channel_is_verified': bool, # this will keep changing
6747 'params': {'extract_flat': True, 'playlist_items': '1'},
6748 'playlist_mincount': 1,
6749 }, {
6750 # Channel renderer metadata. Contains number of videos on the channel
6751 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6752 'info_dict': {
6753 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6754 'title': 'cole-dlp-test-acc - Channels',
6755 'channel': 'cole-dlp-test-acc',
6756 'description': 'test description',
6757 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6758 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6759 'tags': [],
6760 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6761 'uploader_id': '@coletdjnz',
6762 'uploader': 'cole-dlp-test-acc',
6764 'playlist': [{
6765 'info_dict': {
6766 '_type': 'url',
6767 'ie_key': 'YoutubeTab',
6768 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6769 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6770 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6771 'title': 'PewDiePie',
6772 'channel': 'PewDiePie',
6773 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6774 'thumbnails': list,
6775 'channel_follower_count': int,
6776 'playlist_count': int,
6777 'uploader': 'PewDiePie',
6778 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6779 'uploader_id': '@PewDiePie',
6780 'channel_is_verified': True,
6783 'params': {'extract_flat': True},
6784 }, {
6785 'url': 'https://www.youtube.com/@3blue1brown/about',
6786 'info_dict': {
6787 'id': '@3blue1brown',
6788 'tags': ['Mathematics'],
6789 'title': '3Blue1Brown',
6790 'channel_follower_count': int,
6791 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6792 'channel': '3Blue1Brown',
6793 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6794 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
6795 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6796 'uploader_id': '@3blue1brown',
6797 'uploader': '3Blue1Brown',
6798 'channel_is_verified': True,
6800 'playlist_count': 0,
6801 }, {
6802 # Podcasts tab, with rich entry lockupViewModel
6803 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6804 'info_dict': {
6805 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6806 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6807 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6808 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6809 'title': '99% Invisible - Podcasts',
6810 'uploader': '99% Invisible',
6811 'channel_follower_count': int,
6812 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6813 'tags': [],
6814 'channel': '99% Invisible',
6815 'uploader_id': '@99percentinvisiblepodcast',
6817 'playlist_count': 5,
6818 }, {
6819 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6820 'url': 'https://www.youtube.com/@AHimitsu/releases',
6821 'info_dict': {
6822 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6823 'channel': 'A Himitsu',
6824 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6825 'title': 'A Himitsu - Releases',
6826 'uploader_id': '@AHimitsu',
6827 'uploader': 'A Himitsu',
6828 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6829 'tags': 'count:12',
6830 'description': 'I make music',
6831 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6832 'channel_follower_count': int,
6833 'channel_is_verified': True,
6835 'playlist_mincount': 10,
6836 }, {
6837 # Playlist with only shorts, shown as reel renderers
6838 # FIXME: future: YouTube currently doesn't give continuation for this,
6839 # may do in future.
6840 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
6841 'info_dict': {
6842 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
6843 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
6844 'view_count': int,
6845 'uploader_id': '@BangyShorts',
6846 'description': '',
6847 'uploader_url': 'https://www.youtube.com/@BangyShorts',
6848 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
6849 'channel': 'Bangy Shorts',
6850 'uploader': 'Bangy Shorts',
6851 'tags': [],
6852 'availability': 'public',
6853 'modified_date': r're:\d{8}',
6854 'title': 'Uploads from Bangy Shorts',
6856 'playlist_mincount': 100,
6857 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6858 }, {
6859 'note': 'Tags containing spaces',
6860 'url': 'https://www.youtube.com/channel/UC7_YxT-KID8kRbqZo7MyscQ',
6861 'playlist_count': 3,
6862 'info_dict': {
6863 'id': 'UC7_YxT-KID8kRbqZo7MyscQ',
6864 'channel': 'Markiplier',
6865 'channel_id': 'UC7_YxT-KID8kRbqZo7MyscQ',
6866 'title': 'Markiplier',
6867 'channel_follower_count': int,
6868 'description': 'md5:0c010910558658824402809750dc5d97',
6869 'uploader_id': '@markiplier',
6870 'uploader_url': 'https://www.youtube.com/@markiplier',
6871 'uploader': 'Markiplier',
6872 'channel_url': 'https://www.youtube.com/channel/UC7_YxT-KID8kRbqZo7MyscQ',
6873 'channel_is_verified': True,
6874 'tags': ['markiplier', 'comedy', 'gaming', 'funny videos', 'funny moments',
6875 'sketch comedy', 'laughing', 'lets play', 'challenge videos', 'hilarious',
6876 'challenges', 'sketches', 'scary games', 'funny games', 'rage games',
6877 'mark fischbach'],
6881 @classmethod
6882 def suitable(cls, url):
6883 return False if YoutubeIE.suitable(url) else super().suitable(url)
6885 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6887 def _get_url_mobj(self, url):
6888 mobj = self._URL_RE.match(url).groupdict()
6889 mobj.update((k, '') for k, v in mobj.items() if v is None)
6890 return mobj
6892 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6893 tab_name = (tab.get('title') or '').lower()
6894 tab_url = urljoin(base_url, traverse_obj(
6895 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6897 tab_id = ((tab_url and self._get_url_mobj(tab_url)['tab'][1:])
6898 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6899 if tab_id:
6900 return {
6901 'TAB_ID_SPONSORSHIPS': 'membership',
6902 }.get(tab_id, tab_id), tab_name
6904 # Fallback to tab name if we cannot get the tab id.
6905 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6906 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6907 if tab_name:
6908 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6909 return {
6910 'home': 'featured',
6911 'live': 'streams',
6912 }.get(tab_name, tab_name), tab_name
6914 def _has_tab(self, tabs, tab_id):
6915 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6917 def _empty_playlist(self, item_id, data):
6918 return self.playlist_result([], item_id, **self._extract_metadata_from_tabs(item_id, data))
6920 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6921 def _real_extract(self, url, smuggled_data):
6922 item_id = self._match_id(url)
6923 url = urllib.parse.urlunparse(
6924 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6925 compat_opts = self.get_param('compat_opts', [])
6927 mobj = self._get_url_mobj(url)
6928 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6929 if is_channel and smuggled_data.get('is_music_url'):
6930 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6931 return self.url_result(
6932 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6933 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6934 mdata = self._extract_tab_endpoint(
6935 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6936 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6937 get_all=False, expected_type=str)
6938 if not murl:
6939 raise ExtractorError('Failed to resolve album to playlist')
6940 return self.url_result(murl, YoutubeTabIE)
6941 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6942 return self.url_result(
6943 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6945 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6946 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6947 url = f'{pre}/videos{post}'
6948 if smuggled_data.get('is_music_url'):
6949 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6951 # Handle both video/playlist URLs
6952 qs = parse_qs(url)
6953 video_id, playlist_id = (traverse_obj(qs, (key, 0)) for key in ('v', 'list'))
6954 if not video_id and mobj['not_channel'].startswith('watch'):
6955 if not playlist_id:
6956 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6957 raise ExtractorError('A video URL was given without video ID', expected=True)
6958 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6959 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6960 return self.url_result(
6961 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6963 if not self._yes_playlist(playlist_id, video_id):
6964 return self.url_result(
6965 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6967 data, ytcfg = self._extract_data(url, display_id)
6969 # YouTube may provide a non-standard redirect to the regional channel
6970 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6971 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6972 redirect_url = traverse_obj(
6973 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6974 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6975 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6976 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6977 return self.url_result(redirect_url, YoutubeTabIE)
6979 tabs, extra_tabs = self._extract_tab_renderers(data), []
6980 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6981 selected_tab = self._extract_selected_tab(tabs)
6982 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6983 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6985 # /about is no longer a tab
6986 if original_tab_id == 'about':
6987 return self._empty_playlist(item_id, data)
6989 if not original_tab_id and selected_tab_name:
6990 self.to_screen('Downloading all uploads of the channel. '
6991 'To download only the videos in a specific tab, pass the tab\'s URL')
6992 if self._has_tab(tabs, 'streams'):
6993 extra_tabs.append(''.join((pre, '/streams', post)))
6994 if self._has_tab(tabs, 'shorts'):
6995 extra_tabs.append(''.join((pre, '/shorts', post)))
6996 # XXX: Members-only tab should also be extracted
6998 if not extra_tabs and selected_tab_id != 'videos':
6999 # Channel does not have streams, shorts or videos tabs
7000 if item_id[:2] != 'UC':
7001 return self._empty_playlist(item_id, data)
7003 # Topic channels don't have /videos. Use the equivalent playlist instead
7004 pl_id = f'UU{item_id[2:]}'
7005 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
7006 try:
7007 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
7008 except ExtractorError:
7009 return self._empty_playlist(item_id, data)
7010 else:
7011 item_id, url = pl_id, pl_url
7012 self.to_screen(
7013 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
7015 elif extra_tabs and selected_tab_id != 'videos':
7016 # When there are shorts/live tabs but not videos tab
7017 url, data = f'{pre}{post}', None
7019 elif (original_tab_id or 'videos') != selected_tab_id:
7020 if original_tab_id == 'live':
7021 # Live tab should have redirected to the video
7022 # Except in the case the channel has an actual live tab
7023 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
7024 raise UserNotLive(video_id=item_id)
7025 elif selected_tab_name:
7026 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
7028 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
7029 url = f'{pre}{post}'
7031 # YouTube sometimes provides a button to reload playlist with unavailable videos.
7032 if 'no-youtube-unavailable-videos' not in compat_opts:
7033 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
7034 self._extract_and_report_alerts(data, only_once=True)
7036 tabs, entries = self._extract_tab_renderers(data), []
7037 if tabs:
7038 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
7039 entries[0].update({
7040 'extractor_key': YoutubeTabIE.ie_key(),
7041 'extractor': YoutubeTabIE.IE_NAME,
7042 'webpage_url': url,
7044 if self.get_param('playlist_items') == '0':
7045 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
7046 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
7047 entries.extend(map(self._real_extract, extra_tabs))
7049 if len(entries) == 1:
7050 return entries[0]
7051 elif entries:
7052 metadata = self._extract_metadata_from_tabs(item_id, data)
7053 uploads_url = 'the Uploads (UU) playlist URL'
7054 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
7055 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
7056 self.to_screen(
7057 'Downloading as multiple playlists, separated by tabs. '
7058 f'To download as a single playlist instead, pass {uploads_url}')
7059 return self.playlist_result(entries, item_id, **metadata)
7061 # Inline playlist
7062 playlist = traverse_obj(
7063 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
7064 if playlist:
7065 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
7067 video_id = traverse_obj(
7068 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
7069 if video_id:
7070 if tab != '/live': # live tab is expected to redirect to video
7071 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
7072 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
7074 raise ExtractorError('Unable to recognize tab page')
7077 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
7078 IE_DESC = 'YouTube playlists'
7079 _VALID_URL = r'''(?x)(?:
7080 (?:https?://)?
7081 (?:\w+\.)?
7084 youtube(?:kids)?\.com|
7085 {invidious}
7087 /.*?\?.*?\blist=
7089 (?P<id>{playlist_id})
7090 )'''.format(
7091 playlist_id=YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
7092 invidious='|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
7094 IE_NAME = 'youtube:playlist'
7095 _TESTS = [{
7096 'note': 'issue #673',
7097 'url': 'PLBB231211A4F62143',
7098 'info_dict': {
7099 'title': '[OLD]Team Fortress 2 (Class-based LP)',
7100 'id': 'PLBB231211A4F62143',
7101 'uploader': 'Wickman',
7102 'uploader_id': '@WickmanVT',
7103 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
7104 'view_count': int,
7105 'uploader_url': 'https://www.youtube.com/@WickmanVT',
7106 'modified_date': r're:\d{8}',
7107 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
7108 'channel': 'Wickman',
7109 'tags': [],
7110 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
7111 'availability': 'public',
7113 'playlist_mincount': 29,
7114 }, {
7115 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
7116 'info_dict': {
7117 'title': 'YDL_safe_search',
7118 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
7120 'playlist_count': 2,
7121 'skip': 'This playlist is private',
7122 }, {
7123 'note': 'embedded',
7124 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
7125 'playlist_count': 4,
7126 'info_dict': {
7127 'title': 'JODA15',
7128 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
7129 'uploader': 'milan',
7130 'uploader_id': '@milan5503',
7131 'description': '',
7132 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
7133 'tags': [],
7134 'modified_date': '20140919',
7135 'view_count': int,
7136 'channel': 'milan',
7137 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
7138 'uploader_url': 'https://www.youtube.com/@milan5503',
7139 'availability': 'public',
7141 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden', 'Retrying', 'Giving up'],
7142 }, {
7143 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
7144 'playlist_mincount': 455,
7145 'info_dict': {
7146 'title': '2018 Chinese New Singles (11/6 updated)',
7147 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
7148 'uploader': 'LBK',
7149 'uploader_id': '@music_king',
7150 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
7151 'channel': 'LBK',
7152 'view_count': int,
7153 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
7154 'tags': [],
7155 'uploader_url': 'https://www.youtube.com/@music_king',
7156 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
7157 'modified_date': r're:\d{8}',
7158 'availability': 'public',
7160 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
7161 }, {
7162 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
7163 'only_matching': True,
7164 }, {
7165 # music album playlist
7166 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
7167 'only_matching': True,
7170 @classmethod
7171 def suitable(cls, url):
7172 if YoutubeTabIE.suitable(url):
7173 return False
7174 from ..utils import parse_qs
7175 qs = parse_qs(url)
7176 if qs.get('v', [None])[0]:
7177 return False
7178 return super().suitable(url)
7180 def _real_extract(self, url):
7181 playlist_id = self._match_id(url)
7182 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
7183 url = update_url_query(
7184 'https://www.youtube.com/playlist',
7185 parse_qs(url) or {'list': playlist_id})
7186 if is_music_url:
7187 url = smuggle_url(url, {'is_music_url': True})
7188 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
7191 class YoutubeYtBeIE(YoutubeBaseInfoExtractor):
7192 IE_DESC = 'youtu.be'
7193 _VALID_URL = rf'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{{11}})/*?.*?\blist=(?P<playlist_id>{YoutubeBaseInfoExtractor._PLAYLIST_ID_RE})'
7194 _TESTS = [{
7195 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
7196 'info_dict': {
7197 'id': 'yeWKywCrFtk',
7198 'ext': 'mp4',
7199 'title': 'Small Scale Baler and Braiding Rugs',
7200 'uploader': 'Backus-Page House Museum',
7201 'uploader_id': '@backuspagemuseum',
7202 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
7203 'upload_date': '20161008',
7204 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
7205 'categories': ['Nonprofits & Activism'],
7206 'tags': list,
7207 'like_count': int,
7208 'age_limit': 0,
7209 'playable_in_embed': True,
7210 'thumbnail': r're:^https?://.*\.webp',
7211 'channel': 'Backus-Page House Museum',
7212 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
7213 'live_status': 'not_live',
7214 'view_count': int,
7215 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
7216 'availability': 'public',
7217 'duration': 59,
7218 'comment_count': int,
7219 'channel_follower_count': int,
7221 'params': {
7222 'noplaylist': True,
7223 'skip_download': True,
7225 }, {
7226 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
7227 'only_matching': True,
7230 def _real_extract(self, url):
7231 mobj = self._match_valid_url(url)
7232 video_id = mobj.group('id')
7233 playlist_id = mobj.group('playlist_id')
7234 return self.url_result(
7235 update_url_query('https://www.youtube.com/watch', {
7236 'v': video_id,
7237 'list': playlist_id,
7238 'feature': 'youtu.be',
7239 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
7242 class YoutubeLivestreamEmbedIE(YoutubeBaseInfoExtractor):
7243 IE_DESC = 'YouTube livestream embeds'
7244 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
7245 _TESTS = [{
7246 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
7247 'only_matching': True,
7250 def _real_extract(self, url):
7251 channel_id = self._match_id(url)
7252 return self.url_result(
7253 f'https://www.youtube.com/channel/{channel_id}/live',
7254 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
7257 class YoutubeYtUserIE(YoutubeBaseInfoExtractor):
7258 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
7259 IE_NAME = 'youtube:user'
7260 _VALID_URL = r'ytuser:(?P<id>.+)'
7261 _TESTS = [{
7262 'url': 'ytuser:phihag',
7263 'only_matching': True,
7266 def _real_extract(self, url):
7267 user_id = self._match_id(url)
7268 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
7271 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
7272 IE_NAME = 'youtube:favorites'
7273 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
7274 _VALID_URL = r':ytfav(?:ou?rite)?s?'
7275 _LOGIN_REQUIRED = True
7276 _TESTS = [{
7277 'url': ':ytfav',
7278 'only_matching': True,
7279 }, {
7280 'url': ':ytfavorites',
7281 'only_matching': True,
7284 def _real_extract(self, url):
7285 return self.url_result(
7286 'https://www.youtube.com/playlist?list=LL',
7287 ie=YoutubeTabIE.ie_key())
7290 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
7291 IE_NAME = 'youtube:notif'
7292 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
7293 _VALID_URL = r':ytnotif(?:ication)?s?'
7294 _LOGIN_REQUIRED = True
7295 _TESTS = [{
7296 'url': ':ytnotif',
7297 'only_matching': True,
7298 }, {
7299 'url': ':ytnotifications',
7300 'only_matching': True,
7303 def _extract_notification_menu(self, response, continuation_list):
7304 notification_list = traverse_obj(
7305 response,
7306 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
7307 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
7308 expected_type=list) or []
7309 continuation_list[0] = None
7310 for item in notification_list:
7311 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
7312 if entry:
7313 yield entry
7314 continuation = item.get('continuationItemRenderer')
7315 if continuation:
7316 continuation_list[0] = continuation
7318 def _extract_notification_renderer(self, notification):
7319 video_id = traverse_obj(
7320 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
7321 url = f'https://www.youtube.com/watch?v={video_id}'
7322 channel_id = None
7323 if not video_id:
7324 browse_ep = traverse_obj(
7325 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
7326 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
7327 post_id = self._search_regex(
7328 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
7329 'post id', default=None)
7330 if not channel_id or not post_id:
7331 return
7332 # The direct /post url redirects to this in the browser
7333 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
7335 channel = traverse_obj(
7336 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
7337 expected_type=str)
7338 notification_title = self._get_text(notification, 'shortMessage')
7339 if notification_title:
7340 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
7341 # TODO: handle recommended videos
7342 title = self._search_regex(
7343 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
7344 'video title', default=None)
7345 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
7346 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
7347 else None)
7348 return {
7349 '_type': 'url',
7350 'url': url,
7351 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
7352 'video_id': video_id,
7353 'title': title,
7354 'channel_id': channel_id,
7355 'channel': channel,
7356 'uploader': channel,
7357 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
7358 'timestamp': timestamp,
7361 def _notification_menu_entries(self, ytcfg):
7362 continuation_list = [None]
7363 response = None
7364 for page in itertools.count(1):
7365 ctoken = traverse_obj(
7366 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
7367 response = self._extract_response(
7368 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
7369 ep='notification/get_notification_menu', check_get_keys='actions',
7370 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
7371 yield from self._extract_notification_menu(response, continuation_list)
7372 if not continuation_list[0]:
7373 break
7375 def _real_extract(self, url):
7376 display_id = 'notifications'
7377 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
7378 self._report_playlist_authcheck(ytcfg)
7379 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
7382 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
7383 IE_DESC = 'YouTube search'
7384 IE_NAME = 'youtube:search'
7385 _SEARCH_KEY = 'ytsearch'
7386 _SEARCH_PARAMS = 'EgIQAfABAQ==' # Videos only
7387 _TESTS = [{
7388 'url': 'ytsearch5:youtube-dl test video',
7389 'playlist_count': 5,
7390 'info_dict': {
7391 'id': 'youtube-dl test video',
7392 'title': 'youtube-dl test video',
7394 }, {
7395 'note': 'Suicide/self-harm search warning',
7396 'url': 'ytsearch1:i hate myself and i wanna die',
7397 'playlist_count': 1,
7398 'info_dict': {
7399 'id': 'i hate myself and i wanna die',
7400 'title': 'i hate myself and i wanna die',
7405 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
7406 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
7407 _SEARCH_KEY = 'ytsearchdate'
7408 IE_DESC = 'YouTube search, newest videos first'
7409 _SEARCH_PARAMS = 'CAISAhAB8AEB' # Videos only, sorted by date
7410 _TESTS = [{
7411 'url': 'ytsearchdate5:youtube-dl test video',
7412 'playlist_count': 5,
7413 'info_dict': {
7414 'id': 'youtube-dl test video',
7415 'title': 'youtube-dl test video',
7420 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
7421 IE_DESC = 'YouTube search URLs with sorting and filter support'
7422 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
7423 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7424 _TESTS = [{
7425 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
7426 'playlist_mincount': 5,
7427 'info_dict': {
7428 'id': 'youtube-dl test video',
7429 'title': 'youtube-dl test video',
7431 }, {
7432 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
7433 'playlist_mincount': 5,
7434 'info_dict': {
7435 'id': 'python',
7436 'title': 'python',
7438 }, {
7439 'url': 'https://www.youtube.com/results?search_query=%23cats',
7440 'playlist_mincount': 1,
7441 'info_dict': {
7442 'id': '#cats',
7443 'title': '#cats',
7444 # The test suite does not have support for nested playlists
7445 # 'entries': [{
7446 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
7447 # 'title': '#cats',
7448 # }],
7450 }, {
7451 # Channel results
7452 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
7453 'info_dict': {
7454 'id': 'kurzgesagt',
7455 'title': 'kurzgesagt',
7457 'playlist': [{
7458 'info_dict': {
7459 '_type': 'url',
7460 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7461 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7462 'ie_key': 'YoutubeTab',
7463 'channel': 'Kurzgesagt – In a Nutshell',
7464 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
7465 'title': 'Kurzgesagt – In a Nutshell',
7466 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7467 # No longer available for search as it is set to the handle.
7468 # 'playlist_count': int,
7469 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7470 'thumbnails': list,
7471 'uploader_id': '@kurzgesagt',
7472 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
7473 'uploader': 'Kurzgesagt – In a Nutshell',
7474 'channel_is_verified': True,
7475 'channel_follower_count': int,
7478 'params': {'extract_flat': True, 'playlist_items': '1'},
7479 'playlist_mincount': 1,
7480 }, {
7481 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
7482 'only_matching': True,
7485 def _real_extract(self, url):
7486 qs = parse_qs(url)
7487 query = (qs.get('search_query') or qs.get('q'))[0]
7488 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
7491 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
7492 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
7493 IE_NAME = 'youtube:music:search_url'
7494 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7495 _TESTS = [{
7496 'url': 'https://music.youtube.com/search?q=royalty+free+music',
7497 'playlist_count': 16,
7498 'info_dict': {
7499 'id': 'royalty free music',
7500 'title': 'royalty free music',
7502 }, {
7503 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7504 'playlist_mincount': 30,
7505 'info_dict': {
7506 'id': 'royalty free music - songs',
7507 'title': 'royalty free music - songs',
7509 'params': {'extract_flat': 'in_playlist'},
7510 }, {
7511 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7512 'playlist_mincount': 30,
7513 'info_dict': {
7514 'id': 'royalty free music - community playlists',
7515 'title': 'royalty free music - community playlists',
7517 'params': {'extract_flat': 'in_playlist'},
7520 _SECTIONS = {
7521 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7522 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7523 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7524 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7525 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7526 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7529 def _real_extract(self, url):
7530 qs = parse_qs(url)
7531 query = (qs.get('search_query') or qs.get('q'))[0]
7532 params = qs.get('sp', (None,))[0]
7533 if params:
7534 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7535 else:
7536 section = urllib.parse.unquote_plus(([*url.split('#'), ''])[1]).lower()
7537 params = self._SECTIONS.get(section)
7538 if not params:
7539 section = None
7540 title = join_nonempty(query, section, delim=' - ')
7541 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7544 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
7546 Base class for feed extractors
7547 Subclasses must re-define the _FEED_NAME property.
7549 _LOGIN_REQUIRED = True
7550 _FEED_NAME = 'feeds'
7552 @classproperty
7553 def IE_NAME(cls):
7554 return f'youtube:{cls._FEED_NAME}'
7556 def _real_extract(self, url):
7557 return self.url_result(
7558 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7561 class YoutubeWatchLaterIE(YoutubeBaseInfoExtractor):
7562 IE_NAME = 'youtube:watchlater'
7563 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7564 _VALID_URL = r':ytwatchlater'
7565 _TESTS = [{
7566 'url': ':ytwatchlater',
7567 'only_matching': True,
7570 def _real_extract(self, url):
7571 return self.url_result(
7572 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7575 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7576 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7577 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7578 _FEED_NAME = 'recommended'
7579 _LOGIN_REQUIRED = False
7580 _TESTS = [{
7581 'url': ':ytrec',
7582 'only_matching': True,
7583 }, {
7584 'url': ':ytrecommended',
7585 'only_matching': True,
7586 }, {
7587 'url': 'https://youtube.com',
7588 'only_matching': True,
7592 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7593 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7594 _VALID_URL = r':ytsub(?:scription)?s?'
7595 _FEED_NAME = 'subscriptions'
7596 _TESTS = [{
7597 'url': ':ytsubs',
7598 'only_matching': True,
7599 }, {
7600 'url': ':ytsubscriptions',
7601 'only_matching': True,
7605 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7606 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7607 _VALID_URL = r':ythis(?:tory)?'
7608 _FEED_NAME = 'history'
7609 _TESTS = [{
7610 'url': ':ythistory',
7611 'only_matching': True,
7615 class YoutubeShortsAudioPivotIE(YoutubeBaseInfoExtractor):
7616 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7617 IE_NAME = 'youtube:shorts:pivot:audio'
7618 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7619 _TESTS = [{
7620 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7621 'only_matching': True,
7624 @staticmethod
7625 def _generate_audio_pivot_params(video_id):
7627 Generates sfv_audio_pivot browse params for this video id
7629 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7630 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7632 def _real_extract(self, url):
7633 video_id = self._match_id(url)
7634 return self.url_result(
7635 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7636 ie=YoutubeTabIE)
7639 class YoutubeTruncatedURLIE(YoutubeBaseInfoExtractor):
7640 IE_NAME = 'youtube:truncated_url'
7641 IE_DESC = False # Do not list
7642 _VALID_URL = r'''(?x)
7643 (?:https?://)?
7644 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7645 (?:watch\?(?:
7646 feature=[a-z_]+|
7647 annotation_id=annotation_[^&]+|
7648 x-yt-cl=[0-9]+|
7649 hl=[^&]*|
7650 t=[0-9]+
7653 attribution_link\?a=[^&]+
7658 _TESTS = [{
7659 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7660 'only_matching': True,
7661 }, {
7662 'url': 'https://www.youtube.com/watch?',
7663 'only_matching': True,
7664 }, {
7665 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7666 'only_matching': True,
7667 }, {
7668 'url': 'https://www.youtube.com/watch?feature=foo',
7669 'only_matching': True,
7670 }, {
7671 'url': 'https://www.youtube.com/watch?hl=en-GB',
7672 'only_matching': True,
7673 }, {
7674 'url': 'https://www.youtube.com/watch?t=2372',
7675 'only_matching': True,
7678 def _real_extract(self, url):
7679 raise ExtractorError(
7680 'Did you forget to quote the URL? Remember that & is a meta '
7681 'character in most shells, so you want to put the URL in quotes, '
7682 'like yt-dlp '
7683 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7684 ' or simply yt-dlp BaW_jenozKc .',
7685 expected=True)
7688 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7689 IE_NAME = 'youtube:clip'
7690 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7691 _TESTS = [{
7692 # FIXME: Other metadata should be extracted from the clip, not from the base video
7693 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7694 'info_dict': {
7695 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7696 'ext': 'mp4',
7697 'section_start': 29.0,
7698 'section_end': 39.7,
7699 'duration': 10.7,
7700 'age_limit': 0,
7701 'availability': 'public',
7702 'categories': ['Gaming'],
7703 'channel': 'Scott The Woz',
7704 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7705 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7706 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7707 'like_count': int,
7708 'playable_in_embed': True,
7709 'tags': 'count:17',
7710 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7711 'title': 'Mobile Games on Console - Scott The Woz',
7712 'upload_date': '20210920',
7713 'uploader': 'Scott The Woz',
7714 'uploader_id': '@ScottTheWoz',
7715 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7716 'view_count': int,
7717 'live_status': 'not_live',
7718 'channel_follower_count': int,
7719 'chapters': 'count:20',
7720 'comment_count': int,
7721 'heatmap': 'count:100',
7725 def _real_extract(self, url):
7726 clip_id = self._match_id(url)
7727 _, data = self._extract_webpage(url, clip_id)
7729 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7730 if not video_id:
7731 raise ExtractorError('Unable to find video ID')
7733 clip_data = traverse_obj(data, (
7734 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7735 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7736 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7737 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7739 return {
7740 '_type': 'url_transparent',
7741 'url': f'https://www.youtube.com/watch?v={video_id}',
7742 'ie_key': YoutubeIE.ie_key(),
7743 'id': clip_id,
7744 'section_start': int(clip_data['startTimeMs']) / 1000,
7745 'section_end': int(clip_data['endTimeMs']) / 1000,
7746 '_format_sort_fields': ( # https protocol is prioritized for ffmpeg compatibility
7747 'proto:https', 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec', 'channels', 'acodec', 'lang'),
7751 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7752 IE_NAME = 'youtube:consent'
7753 IE_DESC = False # Do not list
7754 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7755 _TESTS = [{
7756 'url': 'https://consent.youtube.com/m?continue=https%3A%2F%2Fwww.youtube.com%2Flive%2FqVv6vCqciTM%3Fcbrd%3D1&gl=NL&m=0&pc=yt&hl=en&src=1',
7757 'info_dict': {
7758 'id': 'qVv6vCqciTM',
7759 'ext': 'mp4',
7760 'age_limit': 0,
7761 'uploader_id': '@sana_natori',
7762 'comment_count': int,
7763 'chapters': 'count:13',
7764 'upload_date': '20221223',
7765 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7766 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7767 'uploader_url': 'https://www.youtube.com/@sana_natori',
7768 'like_count': int,
7769 'release_date': '20221223',
7770 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7771 'title': '【 #インターネット女クリスマス 】3Dで歌ってはしゃぐインターネットの女たち【月ノ美兎/名取さな】',
7772 'view_count': int,
7773 'playable_in_embed': True,
7774 'duration': 4438,
7775 'availability': 'public',
7776 'channel_follower_count': int,
7777 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7778 'categories': ['Entertainment'],
7779 'live_status': 'was_live',
7780 'release_timestamp': 1671793345,
7781 'channel': 'さなちゃんねる',
7782 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7783 'uploader': 'さなちゃんねる',
7784 'channel_is_verified': True,
7785 'heatmap': 'count:100',
7787 'add_ie': ['Youtube'],
7788 'params': {'skip_download': 'Youtube'},
7791 def _real_extract(self, url):
7792 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7793 if not redirect_url:
7794 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7795 return self.url_result(redirect_url)
7798 class YoutubeTruncatedIDIE(YoutubeBaseInfoExtractor):
7799 IE_NAME = 'youtube:truncated_id'
7800 IE_DESC = False # Do not list
7801 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7803 _TESTS = [{
7804 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7805 'only_matching': True,
7808 def _real_extract(self, url):
7809 video_id = self._match_id(url)
7810 raise ExtractorError(
7811 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7812 expected=True)