]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[ie/youtube] Fix tbr calculation (#9489)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import collections
4 import copy
5 import datetime
6 import enum
7 import hashlib
8 import itertools
9 import json
10 import math
11 import os.path
12 import random
13 import re
14 import shlex
15 import sys
16 import threading
17 import time
18 import traceback
19 import urllib.parse
20
21 from .common import InfoExtractor, SearchInfoExtractor
22 from .openload import PhantomJSwrapper
23 from ..compat import functools
24 from ..jsinterp import JSInterpreter
25 from ..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 filter_dict,
37 float_or_none,
38 format_field,
39 get_first,
40 int_or_none,
41 is_html,
42 join_nonempty,
43 js_to_json,
44 mimetype2ext,
45 orderedSet,
46 parse_codecs,
47 parse_count,
48 parse_duration,
49 parse_iso8601,
50 parse_qs,
51 qualities,
52 remove_start,
53 smuggle_url,
54 str_or_none,
55 str_to_int,
56 strftime_or_none,
57 traverse_obj,
58 try_get,
59 unescapeHTML,
60 unified_strdate,
61 unified_timestamp,
62 unsmuggle_url,
63 update_url_query,
64 url_or_none,
65 urljoin,
66 variadic,
67 )
68
69 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
70 # any clients starting with _ cannot be explicitly requested by the user
71 INNERTUBE_CLIENTS = {
72 'web': {
73 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
74 'INNERTUBE_CONTEXT': {
75 'client': {
76 'clientName': 'WEB',
77 'clientVersion': '2.20220801.00.00',
78 }
79 },
80 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
81 },
82 'web_embedded': {
83 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
84 'INNERTUBE_CONTEXT': {
85 'client': {
86 'clientName': 'WEB_EMBEDDED_PLAYER',
87 'clientVersion': '1.20220731.00.00',
88 },
89 },
90 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
91 },
92 'web_music': {
93 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
94 'INNERTUBE_HOST': 'music.youtube.com',
95 'INNERTUBE_CONTEXT': {
96 'client': {
97 'clientName': 'WEB_REMIX',
98 'clientVersion': '1.20220727.01.00',
99 }
100 },
101 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
102 },
103 'web_creator': {
104 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
105 'INNERTUBE_CONTEXT': {
106 'client': {
107 'clientName': 'WEB_CREATOR',
108 'clientVersion': '1.20220726.00.00',
109 }
110 },
111 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
112 },
113 'android': {
114 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
115 'INNERTUBE_CONTEXT': {
116 'client': {
117 'clientName': 'ANDROID',
118 'clientVersion': '19.09.37',
119 'androidSdkVersion': 30,
120 'userAgent': 'com.google.android.youtube/19.09.37 (Linux; U; Android 11) gzip'
121 }
122 },
123 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
124 'REQUIRE_JS_PLAYER': False
125 },
126 'android_embedded': {
127 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
128 'INNERTUBE_CONTEXT': {
129 'client': {
130 'clientName': 'ANDROID_EMBEDDED_PLAYER',
131 'clientVersion': '19.09.37',
132 'androidSdkVersion': 30,
133 'userAgent': 'com.google.android.youtube/19.09.37 (Linux; U; Android 11) gzip'
134 },
135 },
136 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
137 'REQUIRE_JS_PLAYER': False
138 },
139 'android_music': {
140 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
141 'INNERTUBE_CONTEXT': {
142 'client': {
143 'clientName': 'ANDROID_MUSIC',
144 'clientVersion': '6.42.52',
145 'androidSdkVersion': 30,
146 'userAgent': 'com.google.android.apps.youtube.music/6.42.52 (Linux; U; Android 11) gzip'
147 }
148 },
149 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
150 'REQUIRE_JS_PLAYER': False
151 },
152 'android_creator': {
153 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
154 'INNERTUBE_CONTEXT': {
155 'client': {
156 'clientName': 'ANDROID_CREATOR',
157 'clientVersion': '22.30.100',
158 'androidSdkVersion': 30,
159 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip'
160 },
161 },
162 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
163 'REQUIRE_JS_PLAYER': False
164 },
165 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
166 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
167 'ios': {
168 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
169 'INNERTUBE_CONTEXT': {
170 'client': {
171 'clientName': 'IOS',
172 'clientVersion': '19.09.3',
173 'deviceModel': 'iPhone14,3',
174 'userAgent': 'com.google.ios.youtube/19.09.3 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
175 }
176 },
177 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
178 'REQUIRE_JS_PLAYER': False
179 },
180 'ios_embedded': {
181 'INNERTUBE_CONTEXT': {
182 'client': {
183 'clientName': 'IOS_MESSAGES_EXTENSION',
184 'clientVersion': '19.09.3',
185 'deviceModel': 'iPhone14,3',
186 'userAgent': 'com.google.ios.youtube/19.09.3 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
187 },
188 },
189 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
190 'REQUIRE_JS_PLAYER': False
191 },
192 'ios_music': {
193 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
194 'INNERTUBE_CONTEXT': {
195 'client': {
196 'clientName': 'IOS_MUSIC',
197 'clientVersion': '6.33.3',
198 'deviceModel': 'iPhone14,3',
199 'userAgent': 'com.google.ios.youtubemusic/6.33.3 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
200 },
201 },
202 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
203 'REQUIRE_JS_PLAYER': False
204 },
205 'ios_creator': {
206 'INNERTUBE_CONTEXT': {
207 'client': {
208 'clientName': 'IOS_CREATOR',
209 'clientVersion': '22.33.101',
210 'deviceModel': 'iPhone14,3',
211 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
212 },
213 },
214 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
215 'REQUIRE_JS_PLAYER': False
216 },
217 # mweb has 'ultralow' formats
218 # See: https://github.com/yt-dlp/yt-dlp/pull/557
219 'mweb': {
220 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
221 'INNERTUBE_CONTEXT': {
222 'client': {
223 'clientName': 'MWEB',
224 'clientVersion': '2.20220801.00.00',
225 }
226 },
227 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
228 },
229 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
230 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
231 'tv_embedded': {
232 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
233 'INNERTUBE_CONTEXT': {
234 'client': {
235 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
236 'clientVersion': '2.0',
237 },
238 },
239 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
240 },
241 }
242
243
244 def _split_innertube_client(client_name):
245 variant, *base = client_name.rsplit('.', 1)
246 if base:
247 return variant, base[0], variant
248 base, *variant = client_name.split('_', 1)
249 return client_name, base, variant[0] if variant else None
250
251
252 def short_client_name(client_name):
253 main, *parts = _split_innertube_client(client_name)[0].replace('embedscreen', 'e_s').split('_')
254 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
255
256
257 def build_innertube_clients():
258 THIRD_PARTY = {
259 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
260 }
261 BASE_CLIENTS = ('ios', 'android', 'web', 'tv', 'mweb')
262 priority = qualities(BASE_CLIENTS[::-1])
263
264 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
265 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
266 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
267 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
268 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
269
270 _, base_client, variant = _split_innertube_client(client)
271 ytcfg['priority'] = 10 * priority(base_client)
272
273 if not variant:
274 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
275 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
276 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
277 embedscreen['priority'] -= 3
278 elif variant == 'embedded':
279 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
280 ytcfg['priority'] -= 2
281 else:
282 ytcfg['priority'] -= 3
283
284
285 build_innertube_clients()
286
287
288 class BadgeType(enum.Enum):
289 AVAILABILITY_UNLISTED = enum.auto()
290 AVAILABILITY_PRIVATE = enum.auto()
291 AVAILABILITY_PUBLIC = enum.auto()
292 AVAILABILITY_PREMIUM = enum.auto()
293 AVAILABILITY_SUBSCRIPTION = enum.auto()
294 LIVE_NOW = enum.auto()
295 VERIFIED = enum.auto()
296
297
298 class YoutubeBaseInfoExtractor(InfoExtractor):
299 """Provide base functions for Youtube extractors"""
300
301 _RESERVED_NAMES = (
302 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
303 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
304 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
305 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
306
307 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
308
309 # _NETRC_MACHINE = 'youtube'
310
311 # If True it will raise an error if no login info is provided
312 _LOGIN_REQUIRED = False
313
314 _INVIDIOUS_SITES = (
315 # invidious-redirect websites
316 r'(?:www\.)?redirect\.invidious\.io',
317 r'(?:(?:www|dev)\.)?invidio\.us',
318 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
319 r'(?:www\.)?invidious\.pussthecat\.org',
320 r'(?:www\.)?invidious\.zee\.li',
321 r'(?:www\.)?invidious\.ethibox\.fr',
322 r'(?:www\.)?iv\.ggtyler\.dev',
323 r'(?:www\.)?inv\.vern\.i2p',
324 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
325 r'(?:www\.)?inv\.riverside\.rocks',
326 r'(?:www\.)?invidious\.silur\.me',
327 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
328 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
329 r'(?:www\.)?invidious\.slipfox\.xyz',
330 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
331 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
332 r'(?:www\.)?invidious\.tiekoetter\.com',
333 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
334 r'(?:www\.)?invidious\.nerdvpn\.de',
335 r'(?:www\.)?invidious\.weblibre\.org',
336 r'(?:www\.)?inv\.odyssey346\.dev',
337 r'(?:www\.)?invidious\.dhusch\.de',
338 r'(?:www\.)?iv\.melmac\.space',
339 r'(?:www\.)?watch\.thekitty\.zone',
340 r'(?:www\.)?invidious\.privacydev\.net',
341 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
342 r'(?:www\.)?invidious\.drivet\.xyz',
343 r'(?:www\.)?vid\.priv\.au',
344 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
345 r'(?:www\.)?inv\.vern\.cc',
346 r'(?:www\.)?invidious\.esmailelbob\.xyz',
347 r'(?:www\.)?invidious\.sethforprivacy\.com',
348 r'(?:www\.)?yt\.oelrichsgarcia\.de',
349 r'(?:www\.)?yt\.artemislena\.eu',
350 r'(?:www\.)?invidious\.flokinet\.to',
351 r'(?:www\.)?invidious\.baczek\.me',
352 r'(?:www\.)?y\.com\.sb',
353 r'(?:www\.)?invidious\.epicsite\.xyz',
354 r'(?:www\.)?invidious\.lidarshield\.cloud',
355 r'(?:www\.)?yt\.funami\.tech',
356 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
357 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
358 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
359 # youtube-dl invidious instances list
360 r'(?:(?:www|no)\.)?invidiou\.sh',
361 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
362 r'(?:www\.)?invidious\.kabi\.tk',
363 r'(?:www\.)?invidious\.mastodon\.host',
364 r'(?:www\.)?invidious\.zapashcanon\.fr',
365 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
366 r'(?:www\.)?invidious\.tinfoil-hat\.net',
367 r'(?:www\.)?invidious\.himiko\.cloud',
368 r'(?:www\.)?invidious\.reallyancient\.tech',
369 r'(?:www\.)?invidious\.tube',
370 r'(?:www\.)?invidiou\.site',
371 r'(?:www\.)?invidious\.site',
372 r'(?:www\.)?invidious\.xyz',
373 r'(?:www\.)?invidious\.nixnet\.xyz',
374 r'(?:www\.)?invidious\.048596\.xyz',
375 r'(?:www\.)?invidious\.drycat\.fr',
376 r'(?:www\.)?inv\.skyn3t\.in',
377 r'(?:www\.)?tube\.poal\.co',
378 r'(?:www\.)?tube\.connect\.cafe',
379 r'(?:www\.)?vid\.wxzm\.sx',
380 r'(?:www\.)?vid\.mint\.lgbt',
381 r'(?:www\.)?vid\.puffyan\.us',
382 r'(?:www\.)?yewtu\.be',
383 r'(?:www\.)?yt\.elukerio\.org',
384 r'(?:www\.)?yt\.lelux\.fi',
385 r'(?:www\.)?invidious\.ggc-project\.de',
386 r'(?:www\.)?yt\.maisputain\.ovh',
387 r'(?:www\.)?ytprivate\.com',
388 r'(?:www\.)?invidious\.13ad\.de',
389 r'(?:www\.)?invidious\.toot\.koeln',
390 r'(?:www\.)?invidious\.fdn\.fr',
391 r'(?:www\.)?watch\.nettohikari\.com',
392 r'(?:www\.)?invidious\.namazso\.eu',
393 r'(?:www\.)?invidious\.silkky\.cloud',
394 r'(?:www\.)?invidious\.exonip\.de',
395 r'(?:www\.)?invidious\.riverside\.rocks',
396 r'(?:www\.)?invidious\.blamefran\.net',
397 r'(?:www\.)?invidious\.moomoo\.de',
398 r'(?:www\.)?ytb\.trom\.tf',
399 r'(?:www\.)?yt\.cyberhost\.uk',
400 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
401 r'(?:www\.)?qklhadlycap4cnod\.onion',
402 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
403 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
404 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
405 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
406 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
407 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
408 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
409 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
410 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
411 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
412 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
413 r'(?:www\.)?piped\.kavin\.rocks',
414 r'(?:www\.)?piped\.tokhmi\.xyz',
415 r'(?:www\.)?piped\.syncpundit\.io',
416 r'(?:www\.)?piped\.mha\.fi',
417 r'(?:www\.)?watch\.whatever\.social',
418 r'(?:www\.)?piped\.garudalinux\.org',
419 r'(?:www\.)?piped\.rivo\.lol',
420 r'(?:www\.)?piped-libre\.kavin\.rocks',
421 r'(?:www\.)?yt\.jae\.fi',
422 r'(?:www\.)?piped\.mint\.lgbt',
423 r'(?:www\.)?il\.ax',
424 r'(?:www\.)?piped\.esmailelbob\.xyz',
425 r'(?:www\.)?piped\.projectsegfau\.lt',
426 r'(?:www\.)?piped\.privacydev\.net',
427 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
428 r'(?:www\.)?piped\.smnz\.de',
429 r'(?:www\.)?piped\.adminforge\.de',
430 r'(?:www\.)?watch\.whatevertinfoil\.de',
431 r'(?:www\.)?piped\.qdi\.fi',
432 r'(?:(?:www|cf)\.)?piped\.video',
433 r'(?:www\.)?piped\.aeong\.one',
434 r'(?:www\.)?piped\.moomoo\.me',
435 r'(?:www\.)?piped\.chauvet\.pro',
436 r'(?:www\.)?watch\.leptons\.xyz',
437 r'(?:www\.)?pd\.vern\.cc',
438 r'(?:www\.)?piped\.hostux\.net',
439 r'(?:www\.)?piped\.lunar\.icu',
440 # Hyperpipe instances from https://hyperpipe.codeberg.page/
441 r'(?:www\.)?hyperpipe\.surge\.sh',
442 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
443 r'(?:www\.)?listen\.whatever\.social',
444 r'(?:www\.)?music\.adminforge\.de',
445 )
446
447 # extracted from account/account_menu ep
448 # XXX: These are the supported YouTube UI and API languages,
449 # which is slightly different from languages supported for translation in YouTube studio
450 _SUPPORTED_LANG_CODES = [
451 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
452 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
453 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
454 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
455 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
456 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
457 ]
458
459 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
460
461 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
462 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
463
464 def ucid_or_none(self, ucid):
465 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
466
467 def handle_or_none(self, handle):
468 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
469
470 def handle_from_url(self, url):
471 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
472 url, 'channel handle', default=None)
473
474 def ucid_from_url(self, url):
475 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
476 url, 'channel id', default=None)
477
478 @functools.cached_property
479 def _preferred_lang(self):
480 """
481 Returns a language code supported by YouTube for the user preferred language.
482 Returns None if no preferred language set.
483 """
484 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
485 if not preferred_lang:
486 return
487 if preferred_lang not in self._SUPPORTED_LANG_CODES:
488 raise ExtractorError(
489 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
490 expected=True)
491 elif preferred_lang != 'en':
492 self.report_warning(
493 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
494 return preferred_lang
495
496 def _initialize_consent(self):
497 cookies = self._get_cookies('https://www.youtube.com/')
498 if cookies.get('__Secure-3PSID'):
499 return
500 socs = cookies.get('SOCS')
501 if socs and not socs.value.startswith('CAA'): # not consented
502 return
503 self._set_cookie('.youtube.com', 'SOCS', 'CAI', secure=True) # accept all (required for mixes)
504
505 def _initialize_pref(self):
506 cookies = self._get_cookies('https://www.youtube.com/')
507 pref_cookie = cookies.get('PREF')
508 pref = {}
509 if pref_cookie:
510 try:
511 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
512 except ValueError:
513 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
514 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
515 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
516
517 def _real_initialize(self):
518 self._initialize_pref()
519 self._initialize_consent()
520 self._check_login_required()
521
522 def _check_login_required(self):
523 if self._LOGIN_REQUIRED and not self._cookies_passed:
524 self.raise_login_required('Login details are needed to download this content', method='cookies')
525
526 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
527 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
528
529 def _get_default_ytcfg(self, client='web'):
530 return copy.deepcopy(INNERTUBE_CLIENTS[client])
531
532 def _get_innertube_host(self, client='web'):
533 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
534
535 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
536 # try_get but with fallback to default ytcfg client values when present
537 _func = lambda y: try_get(y, getter, expected_type)
538 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
539
540 def _extract_client_name(self, ytcfg, default_client='web'):
541 return self._ytcfg_get_safe(
542 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
543 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
544
545 def _extract_client_version(self, ytcfg, default_client='web'):
546 return self._ytcfg_get_safe(
547 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
548 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
549
550 def _select_api_hostname(self, req_api_hostname, default_client=None):
551 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
552 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
553
554 def _extract_api_key(self, ytcfg=None, default_client='web'):
555 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
556
557 def _extract_context(self, ytcfg=None, default_client='web'):
558 context = get_first(
559 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
560 # Enforce language and tz for extraction
561 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
562 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
563 return context
564
565 _SAPISID = None
566
567 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
568 time_now = round(time.time())
569 if self._SAPISID is None:
570 yt_cookies = self._get_cookies('https://www.youtube.com')
571 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
572 # See: https://github.com/yt-dlp/yt-dlp/issues/393
573 sapisid_cookie = dict_get(
574 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
575 if sapisid_cookie and sapisid_cookie.value:
576 self._SAPISID = sapisid_cookie.value
577 self.write_debug('Extracted SAPISID cookie')
578 # SAPISID cookie is required if not already present
579 if not yt_cookies.get('SAPISID'):
580 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
581 self._set_cookie(
582 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
583 else:
584 self._SAPISID = False
585 if not self._SAPISID:
586 return None
587 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
588 sapisidhash = hashlib.sha1(
589 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
590 return f'SAPISIDHASH {time_now}_{sapisidhash}'
591
592 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
593 note='Downloading API JSON', errnote='Unable to download API page',
594 context=None, api_key=None, api_hostname=None, default_client='web'):
595
596 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
597 data.update(query)
598 real_headers = self.generate_api_headers(default_client=default_client)
599 real_headers.update({'content-type': 'application/json'})
600 if headers:
601 real_headers.update(headers)
602 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
603 or api_key or self._extract_api_key(default_client=default_client))
604 return self._download_json(
605 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
606 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
607 data=json.dumps(data).encode('utf8'), headers=real_headers,
608 query={'key': api_key, 'prettyPrint': 'false'})
609
610 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
611 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
612
613 @staticmethod
614 def _extract_session_index(*data):
615 """
616 Index of current account in account list.
617 See: https://github.com/yt-dlp/yt-dlp/pull/519
618 """
619 for ytcfg in data:
620 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
621 if session_index is not None:
622 return session_index
623
624 # Deprecated?
625 def _extract_identity_token(self, ytcfg=None, webpage=None):
626 if ytcfg:
627 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
628 if token:
629 return token
630 if webpage:
631 return self._search_regex(
632 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
633 'identity token', default=None, fatal=False)
634
635 @staticmethod
636 def _extract_account_syncid(*args):
637 """
638 Extract syncId required to download private playlists of secondary channels
639 @params response and/or ytcfg
640 """
641 for data in args:
642 # ytcfg includes channel_syncid if on secondary channel
643 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
644 if delegated_sid:
645 return delegated_sid
646 sync_ids = (try_get(
647 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
648 lambda x: x['DATASYNC_ID']), str) or '').split('||')
649 if len(sync_ids) >= 2 and sync_ids[1]:
650 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
651 # and just "user_syncid||" for primary channel. We only want the channel_syncid
652 return sync_ids[0]
653
654 @staticmethod
655 def _extract_visitor_data(*args):
656 """
657 Extracts visitorData from an API response or ytcfg
658 Appears to be used to track session state
659 """
660 return get_first(
661 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
662 expected_type=str)
663
664 @functools.cached_property
665 def is_authenticated(self):
666 return bool(self._generate_sapisidhash_header())
667
668 def extract_ytcfg(self, video_id, webpage):
669 if not webpage:
670 return {}
671 return self._parse_json(
672 self._search_regex(
673 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
674 default='{}'), video_id, fatal=False) or {}
675
676 def generate_api_headers(
677 self, *, ytcfg=None, account_syncid=None, session_index=None,
678 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
679
680 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
681 headers = {
682 'X-YouTube-Client-Name': str(
683 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
684 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
685 'Origin': origin,
686 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
687 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
688 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
689 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
690 }
691 if session_index is None:
692 session_index = self._extract_session_index(ytcfg)
693 if account_syncid or session_index is not None:
694 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
695
696 auth = self._generate_sapisidhash_header(origin)
697 if auth is not None:
698 headers['Authorization'] = auth
699 headers['X-Origin'] = origin
700 return filter_dict(headers)
701
702 def _download_ytcfg(self, client, video_id):
703 url = {
704 'web': 'https://www.youtube.com',
705 'web_music': 'https://music.youtube.com',
706 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
707 }.get(client)
708 if not url:
709 return {}
710 webpage = self._download_webpage(
711 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
712 return self.extract_ytcfg(video_id, webpage) or {}
713
714 @staticmethod
715 def _build_api_continuation_query(continuation, ctp=None):
716 query = {
717 'continuation': continuation
718 }
719 # TODO: Inconsistency with clickTrackingParams.
720 # Currently we have a fixed ctp contained within context (from ytcfg)
721 # and a ctp in root query for continuation.
722 if ctp:
723 query['clickTracking'] = {'clickTrackingParams': ctp}
724 return query
725
726 @classmethod
727 def _extract_next_continuation_data(cls, renderer):
728 next_continuation = try_get(
729 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
730 lambda x: x['continuation']['reloadContinuationData']), dict)
731 if not next_continuation:
732 return
733 continuation = next_continuation.get('continuation')
734 if not continuation:
735 return
736 ctp = next_continuation.get('clickTrackingParams')
737 return cls._build_api_continuation_query(continuation, ctp)
738
739 @classmethod
740 def _extract_continuation_ep_data(cls, continuation_ep: dict):
741 if isinstance(continuation_ep, dict):
742 continuation = try_get(
743 continuation_ep, lambda x: x['continuationCommand']['token'], str)
744 if not continuation:
745 return
746 ctp = continuation_ep.get('clickTrackingParams')
747 return cls._build_api_continuation_query(continuation, ctp)
748
749 @classmethod
750 def _extract_continuation(cls, renderer):
751 next_continuation = cls._extract_next_continuation_data(renderer)
752 if next_continuation:
753 return next_continuation
754
755 return traverse_obj(renderer, (
756 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
757 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
758 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
759
760 @classmethod
761 def _extract_alerts(cls, data):
762 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
763 if not isinstance(alert_dict, dict):
764 continue
765 for alert in alert_dict.values():
766 alert_type = alert.get('type')
767 if not alert_type:
768 continue
769 message = cls._get_text(alert, 'text')
770 if message:
771 yield alert_type, message
772
773 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
774 errors, warnings = [], []
775 for alert_type, alert_message in alerts:
776 if alert_type.lower() == 'error' and fatal:
777 errors.append([alert_type, alert_message])
778 elif alert_message not in self._IGNORED_WARNINGS:
779 warnings.append([alert_type, alert_message])
780
781 for alert_type, alert_message in (warnings + errors[:-1]):
782 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
783 if errors:
784 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
785
786 def _extract_and_report_alerts(self, data, *args, **kwargs):
787 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
788
789 def _extract_badges(self, badge_list: list):
790 """
791 Extract known BadgeType's from a list of badge renderers.
792 @returns [{'type': BadgeType}]
793 """
794 icon_type_map = {
795 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
796 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
797 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC,
798 'CHECK_CIRCLE_THICK': BadgeType.VERIFIED,
799 'OFFICIAL_ARTIST_BADGE': BadgeType.VERIFIED,
800 'CHECK': BadgeType.VERIFIED,
801 }
802
803 badge_style_map = {
804 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
805 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
806 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
807 'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
808 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED,
809 }
810
811 label_map = {
812 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
813 'private': BadgeType.AVAILABILITY_PRIVATE,
814 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
815 'live': BadgeType.LIVE_NOW,
816 'premium': BadgeType.AVAILABILITY_PREMIUM,
817 'verified': BadgeType.VERIFIED,
818 'official artist channel': BadgeType.VERIFIED,
819 }
820
821 badges = []
822 for badge in traverse_obj(badge_list, (..., lambda key, _: re.search(r'[bB]adgeRenderer$', key))):
823 badge_type = (
824 icon_type_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
825 or badge_style_map.get(traverse_obj(badge, 'style'))
826 )
827 if badge_type:
828 badges.append({'type': badge_type})
829 continue
830
831 # fallback, won't work in some languages
832 label = traverse_obj(
833 badge, 'label', ('accessibilityData', 'label'), 'tooltip', 'iconTooltip', get_all=False, expected_type=str, default='')
834 for match, label_badge_type in label_map.items():
835 if match in label.lower():
836 badges.append({'type': label_badge_type})
837 break
838
839 return badges
840
841 @staticmethod
842 def _has_badge(badges, badge_type):
843 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
844
845 @staticmethod
846 def _get_text(data, *path_list, max_runs=None):
847 for path in path_list or [None]:
848 if path is None:
849 obj = [data]
850 else:
851 obj = traverse_obj(data, path, default=[])
852 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
853 obj = [obj]
854 for item in obj:
855 text = try_get(item, lambda x: x['simpleText'], str)
856 if text:
857 return text
858 runs = try_get(item, lambda x: x['runs'], list) or []
859 if not runs and isinstance(item, list):
860 runs = item
861
862 runs = runs[:min(len(runs), max_runs or len(runs))]
863 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
864 if text:
865 return text
866
867 def _get_count(self, data, *path_list):
868 count_text = self._get_text(data, *path_list) or ''
869 count = parse_count(count_text)
870 if count is None:
871 count = str_to_int(
872 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
873 return count
874
875 @staticmethod
876 def _extract_thumbnails(data, *path_list):
877 """
878 Extract thumbnails from thumbnails dict
879 @param path_list: path list to level that contains 'thumbnails' key
880 """
881 thumbnails = []
882 for path in path_list or [()]:
883 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...)):
884 thumbnail_url = url_or_none(thumbnail.get('url'))
885 if not thumbnail_url:
886 continue
887 # Sometimes youtube gives a wrong thumbnail URL. See:
888 # https://github.com/yt-dlp/yt-dlp/issues/233
889 # https://github.com/ytdl-org/youtube-dl/issues/28023
890 if 'maxresdefault' in thumbnail_url:
891 thumbnail_url = thumbnail_url.split('?')[0]
892 thumbnails.append({
893 'url': thumbnail_url,
894 'height': int_or_none(thumbnail.get('height')),
895 'width': int_or_none(thumbnail.get('width')),
896 })
897 return thumbnails
898
899 @staticmethod
900 def extract_relative_time(relative_time_text):
901 """
902 Extracts a relative time from string and converts to dt object
903 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today', '8 yr ago'
904 """
905
906 # XXX: this could be moved to a general function in utils/_utils.py
907 # The relative time text strings are roughly the same as what
908 # Javascript's Intl.RelativeTimeFormat function generates.
909 # See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat
910 mobj = re.search(
911 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',
912 relative_time_text)
913 if mobj:
914 start = mobj.group('start')
915 if start:
916 return datetime_from_str(start)
917 try:
918 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
919 except ValueError:
920 return None
921
922 def _parse_time_text(self, text):
923 if not text:
924 return
925 dt = self.extract_relative_time(text)
926 timestamp = None
927 if isinstance(dt, datetime.datetime):
928 timestamp = calendar.timegm(dt.timetuple())
929
930 if timestamp is None:
931 timestamp = (
932 unified_timestamp(text) or unified_timestamp(
933 self._search_regex(
934 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
935 text.lower(), 'time text', default=None)))
936
937 if text and timestamp is None and self._preferred_lang in (None, 'en'):
938 self.report_warning(
939 f'Cannot parse localized time text "{text}"', only_once=True)
940 return timestamp
941
942 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
943 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
944 default_client='web'):
945 raise_for_incomplete = bool(self._configuration_arg('raise_incomplete_data', ie_key=YoutubeIE))
946 # Incomplete Data should be a warning by default when retries are exhausted, while other errors should be fatal.
947 icd_retries = iter(self.RetryManager(fatal=raise_for_incomplete))
948 icd_rm = next(icd_retries)
949 main_retries = iter(self.RetryManager())
950 main_rm = next(main_retries)
951 # Manual retry loop for multiple RetryManagers
952 # The proper RetryManager MUST be advanced after an error
953 # and its result MUST be checked if the manager is non fatal
954 while True:
955 try:
956 response = self._call_api(
957 ep=ep, fatal=True, headers=headers,
958 video_id=item_id, query=query, note=note,
959 context=self._extract_context(ytcfg, default_client),
960 api_key=self._extract_api_key(ytcfg, default_client),
961 api_hostname=api_hostname, default_client=default_client)
962 except ExtractorError as e:
963 if not isinstance(e.cause, network_exceptions):
964 return self._error_or_warning(e, fatal=fatal)
965 elif not isinstance(e.cause, HTTPError):
966 main_rm.error = e
967 next(main_retries)
968 continue
969
970 first_bytes = e.cause.response.read(512)
971 if not is_html(first_bytes):
972 yt_error = try_get(
973 self._parse_json(
974 self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
975 lambda x: x['error']['message'], str)
976 if yt_error:
977 self._report_alerts([('ERROR', yt_error)], fatal=False)
978 # Downloading page may result in intermittent 5xx HTTP error
979 # Sometimes a 404 is also received. See: https://github.com/ytdl-org/youtube-dl/issues/28289
980 # We also want to catch all other network exceptions since errors in later pages can be troublesome
981 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
982 if e.cause.status not in (403, 429):
983 main_rm.error = e
984 next(main_retries)
985 continue
986 return self._error_or_warning(e, fatal=fatal)
987
988 try:
989 self._extract_and_report_alerts(response, only_once=True)
990 except ExtractorError as e:
991 # YouTube's servers may return errors we want to retry on in a 200 OK response
992 # See: https://github.com/yt-dlp/yt-dlp/issues/839
993 if 'unknown error' in e.msg.lower():
994 main_rm.error = e
995 next(main_retries)
996 continue
997 return self._error_or_warning(e, fatal=fatal)
998 # Youtube sometimes sends incomplete data
999 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1000 if not traverse_obj(response, *variadic(check_get_keys)):
1001 icd_rm.error = ExtractorError('Incomplete data received', expected=True)
1002 should_retry = next(icd_retries, None)
1003 if not should_retry:
1004 return None
1005 continue
1006
1007 return response
1008
1009 @staticmethod
1010 def is_music_url(url):
1011 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
1012
1013 def _extract_video(self, renderer):
1014 video_id = renderer.get('videoId')
1015
1016 reel_header_renderer = traverse_obj(renderer, (
1017 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1018 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1019
1020 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1021 description = self._get_text(renderer, 'descriptionSnippet')
1022
1023 duration = int_or_none(renderer.get('lengthSeconds'))
1024 if duration is None:
1025 duration = parse_duration(self._get_text(
1026 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1027 if duration is None:
1028 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1029 duration = parse_duration(self._search_regex(
1030 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1031 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1032 video_id, default=None, group='duration'))
1033
1034 channel_id = traverse_obj(
1035 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1036 expected_type=str, get_all=False)
1037 if not channel_id:
1038 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1039
1040 channel_id = self.ucid_or_none(channel_id)
1041
1042 overlay_style = traverse_obj(
1043 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1044 get_all=False, expected_type=str)
1045 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1046 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1047 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1048 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1049 expected_type=str)) or ''
1050 url = f'https://www.youtube.com/watch?v={video_id}'
1051 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1052 url = f'https://www.youtube.com/shorts/{video_id}'
1053
1054 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1055 or self._get_text(reel_header_renderer, 'timestampText') or '')
1056 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1057
1058 live_status = (
1059 'is_upcoming' if scheduled_timestamp is not None
1060 else 'was_live' if 'streamed' in time_text.lower()
1061 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1062 else None)
1063
1064 # videoInfo is a string like '50K views • 10 years ago'.
1065 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1066 view_count = (0 if 'no views' in view_count_text.lower()
1067 else self._get_count({'simpleText': view_count_text}))
1068 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1069
1070 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1071 or self._get_text(reel_header_renderer, 'channelTitleText'))
1072
1073 channel_handle = traverse_obj(renderer, (
1074 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1075 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1076 expected_type=self.handle_from_url, get_all=False)
1077 return {
1078 '_type': 'url',
1079 'ie_key': YoutubeIE.ie_key(),
1080 'id': video_id,
1081 'url': url,
1082 'title': title,
1083 'description': description,
1084 'duration': duration,
1085 'channel_id': channel_id,
1086 'channel': channel,
1087 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1088 'uploader': channel,
1089 'uploader_id': channel_handle,
1090 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1091 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1092 'timestamp': (self._parse_time_text(time_text)
1093 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1094 else None),
1095 'release_timestamp': scheduled_timestamp,
1096 'availability':
1097 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1098 else self._availability(
1099 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1100 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1101 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1102 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1103 view_count_field: view_count,
1104 'live_status': live_status,
1105 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None
1106 }
1107
1108
1109 class YoutubeIE(YoutubeBaseInfoExtractor):
1110 IE_DESC = 'YouTube'
1111 _VALID_URL = r"""(?x)^
1112 (
1113 (?:https?://|//) # http(s):// or protocol-independent URL
1114 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1115 (?:www\.)?deturl\.com/www\.youtube\.com|
1116 (?:www\.)?pwnyoutube\.com|
1117 (?:www\.)?hooktube\.com|
1118 (?:www\.)?yourepeat\.com|
1119 tube\.majestyc\.net|
1120 %(invidious)s|
1121 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1122 (?:.*?\#/)? # handle anchor (#/) redirect urls
1123 (?: # the various things that can precede the ID:
1124 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1125 |(?: # or the v= param in all its forms
1126 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1127 (?:\?|\#!?) # the params delimiter ? or # or #!
1128 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1129 v=
1130 )
1131 ))
1132 |(?:
1133 youtu\.be| # just youtu.be/xxxx
1134 vid\.plus| # or vid.plus/xxxx
1135 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1136 %(invidious)s
1137 )/
1138 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1139 )
1140 )? # all until now is optional -> you can pass the naked ID
1141 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1142 (?(1).+)? # if we found the ID, everything can follow
1143 (?:\#|$)""" % {
1144 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1145 }
1146 _EMBED_REGEX = [
1147 r'''(?x)
1148 (?:
1149 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1150 data-video-url=|
1151 <embed[^>]+?src=|
1152 embedSWF\(?:\s*|
1153 <object[^>]+data=|
1154 new\s+SWFObject\(
1155 )
1156 (["\'])
1157 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1158 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1159 \1''',
1160 # https://wordpress.org/plugins/lazy-load-for-videos/
1161 r'''(?xs)
1162 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1163 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1164 ]
1165 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1166
1167 _PLAYER_INFO_RE = (
1168 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1169 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$',
1170 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1171 )
1172 _formats = {
1173 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1174 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1175 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1176 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1177 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1178 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1179 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1180 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1181 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1182 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1183 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1184 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1185 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1186 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1187 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1188 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1189 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1190 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1191
1192
1193 # 3D videos
1194 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1195 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1196 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1197 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1198 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1199 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1200 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1201
1202 # Apple HTTP Live Streaming
1203 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1204 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1205 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1206 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1207 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1208 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1209 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1210 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1211
1212 # DASH mp4 video
1213 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1214 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1215 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1216 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1217 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1218 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1219 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1220 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1221 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1222 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1223 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1224 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1225
1226 # Dash mp4 audio
1227 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1228 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1229 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1230 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1231 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1232 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1233 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1234
1235 # Dash webm
1236 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1237 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1238 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1239 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1240 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1241 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1242 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1243 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1244 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1245 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1246 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1247 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1248 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1249 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1250 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1251 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1252 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1253 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1254 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1255 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1256 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1257 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1258
1259 # Dash webm audio
1260 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1261 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1262
1263 # Dash webm audio with opus inside
1264 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1265 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1266 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1267
1268 # RTMP (unnamed)
1269 '_rtmp': {'protocol': 'rtmp'},
1270
1271 # av01 video only formats sometimes served with "unknown" codecs
1272 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1273 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1274 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1275 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1276 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1277 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1278 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1279 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1280 }
1281 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1282
1283 _GEO_BYPASS = False
1284
1285 IE_NAME = 'youtube'
1286 _TESTS = [
1287 {
1288 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1289 'info_dict': {
1290 'id': 'BaW_jenozKc',
1291 'ext': 'mp4',
1292 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1293 'channel': 'Philipp Hagemeister',
1294 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1295 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1296 'upload_date': '20121002',
1297 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1298 'categories': ['Science & Technology'],
1299 'tags': ['youtube-dl'],
1300 'duration': 10,
1301 'view_count': int,
1302 'like_count': int,
1303 'availability': 'public',
1304 'playable_in_embed': True,
1305 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1306 'live_status': 'not_live',
1307 'age_limit': 0,
1308 'start_time': 1,
1309 'end_time': 9,
1310 'comment_count': int,
1311 'channel_follower_count': int,
1312 'uploader': 'Philipp Hagemeister',
1313 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1314 'uploader_id': '@PhilippHagemeister',
1315 'heatmap': 'count:100',
1316 }
1317 },
1318 {
1319 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1320 'note': 'Embed-only video (#1746)',
1321 'info_dict': {
1322 'id': 'yZIXLfi8CZQ',
1323 'ext': 'mp4',
1324 'upload_date': '20120608',
1325 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1326 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1327 'age_limit': 18,
1328 },
1329 'skip': 'Private video',
1330 },
1331 {
1332 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1333 'note': 'Use the first video ID in the URL',
1334 'info_dict': {
1335 'id': 'BaW_jenozKc',
1336 'ext': 'mp4',
1337 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1338 'channel': 'Philipp Hagemeister',
1339 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1340 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1341 'upload_date': '20121002',
1342 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1343 'categories': ['Science & Technology'],
1344 'tags': ['youtube-dl'],
1345 'duration': 10,
1346 'view_count': int,
1347 'like_count': int,
1348 'availability': 'public',
1349 'playable_in_embed': True,
1350 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1351 'live_status': 'not_live',
1352 'age_limit': 0,
1353 'comment_count': int,
1354 'channel_follower_count': int,
1355 'uploader': 'Philipp Hagemeister',
1356 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1357 'uploader_id': '@PhilippHagemeister',
1358 'heatmap': 'count:100',
1359 },
1360 'params': {
1361 'skip_download': True,
1362 },
1363 },
1364 {
1365 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1366 'note': '256k DASH audio (format 141) via DASH manifest',
1367 'info_dict': {
1368 'id': 'a9LDPn-MO4I',
1369 'ext': 'm4a',
1370 'upload_date': '20121002',
1371 'description': '',
1372 'title': 'UHDTV TEST 8K VIDEO.mp4'
1373 },
1374 'params': {
1375 'youtube_include_dash_manifest': True,
1376 'format': '141',
1377 },
1378 'skip': 'format 141 not served anymore',
1379 },
1380 # DASH manifest with encrypted signature
1381 {
1382 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1383 'info_dict': {
1384 'id': 'IB3lcPjvWLA',
1385 'ext': 'm4a',
1386 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1387 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1388 'duration': 244,
1389 'upload_date': '20131011',
1390 'abr': 129.495,
1391 'like_count': int,
1392 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1393 'playable_in_embed': True,
1394 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1395 'view_count': int,
1396 'track': 'The Spark',
1397 'live_status': 'not_live',
1398 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1399 'channel': 'Afrojack',
1400 'tags': 'count:19',
1401 'availability': 'public',
1402 'categories': ['Music'],
1403 'age_limit': 0,
1404 'alt_title': 'The Spark',
1405 'channel_follower_count': int,
1406 'uploader': 'Afrojack',
1407 'uploader_url': 'https://www.youtube.com/@Afrojack',
1408 'uploader_id': '@Afrojack',
1409 },
1410 'params': {
1411 'youtube_include_dash_manifest': True,
1412 'format': '141/bestaudio[ext=m4a]',
1413 },
1414 },
1415 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1416 {
1417 'note': 'Embed allowed age-gate video',
1418 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1419 'info_dict': {
1420 'id': 'HtVdAasjOgU',
1421 'ext': 'mp4',
1422 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1423 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1424 'duration': 142,
1425 'upload_date': '20140605',
1426 'age_limit': 18,
1427 'categories': ['Gaming'],
1428 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1429 'availability': 'needs_auth',
1430 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1431 'like_count': int,
1432 'channel': 'The Witcher',
1433 'live_status': 'not_live',
1434 'tags': 'count:17',
1435 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1436 'playable_in_embed': True,
1437 'view_count': int,
1438 'channel_follower_count': int,
1439 'uploader': 'The Witcher',
1440 'uploader_url': 'https://www.youtube.com/@thewitcher',
1441 'uploader_id': '@thewitcher',
1442 'comment_count': int,
1443 'channel_is_verified': True,
1444 'heatmap': 'count:100',
1445 },
1446 },
1447 {
1448 'note': 'Age-gate video with embed allowed in public site',
1449 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1450 'info_dict': {
1451 'id': 'HsUATh_Nc2U',
1452 'ext': 'mp4',
1453 'title': 'Godzilla 2 (Official Video)',
1454 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1455 'upload_date': '20200408',
1456 'age_limit': 18,
1457 'availability': 'needs_auth',
1458 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1459 'channel': 'FlyingKitty',
1460 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1461 'view_count': int,
1462 'categories': ['Entertainment'],
1463 'live_status': 'not_live',
1464 'tags': ['Flyingkitty', 'godzilla 2'],
1465 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1466 'like_count': int,
1467 'duration': 177,
1468 'playable_in_embed': True,
1469 'channel_follower_count': int,
1470 'uploader': 'FlyingKitty',
1471 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1472 'uploader_id': '@FlyingKitty900',
1473 'comment_count': int,
1474 'channel_is_verified': True,
1475 },
1476 },
1477 {
1478 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1479 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1480 'info_dict': {
1481 'id': 'Tq92D6wQ1mg',
1482 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1483 'ext': 'mp4',
1484 'upload_date': '20191228',
1485 'description': 'md5:17eccca93a786d51bc67646756894066',
1486 'age_limit': 18,
1487 'like_count': int,
1488 'availability': 'needs_auth',
1489 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1490 'view_count': int,
1491 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1492 'channel': 'Projekt Melody',
1493 'live_status': 'not_live',
1494 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1495 'playable_in_embed': True,
1496 'categories': ['Entertainment'],
1497 'duration': 106,
1498 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1499 'comment_count': int,
1500 'channel_follower_count': int,
1501 'uploader': 'Projekt Melody',
1502 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1503 'uploader_id': '@ProjektMelody',
1504 },
1505 },
1506 {
1507 'note': 'Non-Agegated non-embeddable video',
1508 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1509 'info_dict': {
1510 'id': 'MeJVWBSsPAY',
1511 'ext': 'mp4',
1512 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1513 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1514 'upload_date': '20130730',
1515 'track': 'Such mich find mich',
1516 'age_limit': 0,
1517 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1518 'like_count': int,
1519 'playable_in_embed': False,
1520 'creator': 'OOMPH!',
1521 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1522 'view_count': int,
1523 'alt_title': 'Such mich find mich',
1524 'duration': 210,
1525 'channel': 'Herr Lurik',
1526 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1527 'categories': ['Music'],
1528 'availability': 'public',
1529 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1530 'live_status': 'not_live',
1531 'artist': 'OOMPH!',
1532 'channel_follower_count': int,
1533 'uploader': 'Herr Lurik',
1534 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1535 'uploader_id': '@HerrLurik',
1536 },
1537 },
1538 {
1539 'note': 'Non-bypassable age-gated video',
1540 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1541 'only_matching': True,
1542 },
1543 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1544 # YouTube Red ad is not captured for creator
1545 {
1546 'url': '__2ABJjxzNo',
1547 'info_dict': {
1548 'id': '__2ABJjxzNo',
1549 'ext': 'mp4',
1550 'duration': 266,
1551 'upload_date': '20100430',
1552 'creator': 'deadmau5',
1553 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1554 'title': 'Deadmau5 - Some Chords (HD)',
1555 'alt_title': 'Some Chords',
1556 'availability': 'public',
1557 'tags': 'count:14',
1558 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1559 'view_count': int,
1560 'live_status': 'not_live',
1561 'channel': 'deadmau5',
1562 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1563 'like_count': int,
1564 'track': 'Some Chords',
1565 'artist': 'deadmau5',
1566 'playable_in_embed': True,
1567 'age_limit': 0,
1568 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1569 'categories': ['Music'],
1570 'album': 'Some Chords',
1571 'channel_follower_count': int,
1572 'uploader': 'deadmau5',
1573 'uploader_url': 'https://www.youtube.com/@deadmau5',
1574 'uploader_id': '@deadmau5',
1575 },
1576 'expected_warnings': [
1577 'DASH manifest missing',
1578 ]
1579 },
1580 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1581 {
1582 'url': 'lqQg6PlCWgI',
1583 'info_dict': {
1584 'id': 'lqQg6PlCWgI',
1585 'ext': 'mp4',
1586 'duration': 6085,
1587 'upload_date': '20150827',
1588 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1589 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1590 'like_count': int,
1591 'release_timestamp': 1343767800,
1592 'playable_in_embed': True,
1593 'categories': ['Sports'],
1594 'release_date': '20120731',
1595 'channel': 'Olympics',
1596 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1597 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1598 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1599 'age_limit': 0,
1600 'availability': 'public',
1601 'live_status': 'was_live',
1602 'view_count': int,
1603 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1604 'channel_follower_count': int,
1605 'uploader': 'Olympics',
1606 'uploader_url': 'https://www.youtube.com/@Olympics',
1607 'uploader_id': '@Olympics',
1608 'channel_is_verified': True,
1609 },
1610 'params': {
1611 'skip_download': 'requires avconv',
1612 }
1613 },
1614 # Non-square pixels
1615 {
1616 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1617 'info_dict': {
1618 'id': '_b-2C3KPAM0',
1619 'ext': 'mp4',
1620 'stretched_ratio': 16 / 9.,
1621 'duration': 85,
1622 'upload_date': '20110310',
1623 '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',
1624 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1625 'playable_in_embed': True,
1626 'channel': '孫ᄋᄅ',
1627 'age_limit': 0,
1628 'tags': 'count:11',
1629 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1630 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1631 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1632 'view_count': int,
1633 'categories': ['People & Blogs'],
1634 'like_count': int,
1635 'live_status': 'not_live',
1636 'availability': 'unlisted',
1637 'comment_count': int,
1638 'channel_follower_count': int,
1639 'uploader': '孫ᄋᄅ',
1640 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1641 'uploader_id': '@AllenMeow',
1642 },
1643 },
1644 # url_encoded_fmt_stream_map is empty string
1645 {
1646 'url': 'qEJwOuvDf7I',
1647 'info_dict': {
1648 'id': 'qEJwOuvDf7I',
1649 'ext': 'webm',
1650 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1651 'description': '',
1652 'upload_date': '20150404',
1653 },
1654 'params': {
1655 'skip_download': 'requires avconv',
1656 },
1657 'skip': 'This live event has ended.',
1658 },
1659 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1660 {
1661 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1662 'info_dict': {
1663 'id': 'FIl7x6_3R5Y',
1664 'ext': 'webm',
1665 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1666 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1667 'duration': 220,
1668 'upload_date': '20150625',
1669 'formats': 'mincount:31',
1670 },
1671 'skip': 'not actual anymore',
1672 },
1673 # DASH manifest with segment_list
1674 {
1675 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1676 'md5': '8ce563a1d667b599d21064e982ab9e31',
1677 'info_dict': {
1678 'id': 'CsmdDsKjzN8',
1679 'ext': 'mp4',
1680 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1681 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1682 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1683 },
1684 'params': {
1685 'youtube_include_dash_manifest': True,
1686 'format': '135', # bestvideo
1687 },
1688 'skip': 'This live event has ended.',
1689 },
1690 {
1691 # Multifeed videos (multiple cameras), URL can be of any Camera
1692 # TODO: fix multifeed titles
1693 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1694 'info_dict': {
1695 'id': 'zaPI8MvL8pg',
1696 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1697 'description': 'md5:563ccbc698b39298481ca3c571169519',
1698 },
1699 'playlist': [{
1700 'info_dict': {
1701 'id': 'j5yGuxZ8lLU',
1702 'ext': 'mp4',
1703 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1704 'description': 'md5:563ccbc698b39298481ca3c571169519',
1705 'duration': 10120,
1706 'channel_follower_count': int,
1707 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1708 'availability': 'public',
1709 'playable_in_embed': True,
1710 'upload_date': '20131105',
1711 'categories': ['Gaming'],
1712 'live_status': 'was_live',
1713 'tags': 'count:24',
1714 'release_timestamp': 1383701910,
1715 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1716 'comment_count': int,
1717 'age_limit': 0,
1718 'like_count': int,
1719 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1720 'channel': 'WiiLikeToPlay',
1721 'view_count': int,
1722 'release_date': '20131106',
1723 'uploader': 'WiiLikeToPlay',
1724 'uploader_id': '@WLTP',
1725 'uploader_url': 'https://www.youtube.com/@WLTP',
1726 },
1727 }, {
1728 'info_dict': {
1729 'id': 'zaPI8MvL8pg',
1730 'ext': 'mp4',
1731 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1732 'availability': 'public',
1733 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1734 'channel': 'WiiLikeToPlay',
1735 'channel_follower_count': int,
1736 'description': 'md5:563ccbc698b39298481ca3c571169519',
1737 'duration': 10108,
1738 'age_limit': 0,
1739 'like_count': int,
1740 'tags': 'count:24',
1741 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1742 'release_timestamp': 1383701915,
1743 'comment_count': int,
1744 'upload_date': '20131105',
1745 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1746 'release_date': '20131106',
1747 'playable_in_embed': True,
1748 'live_status': 'was_live',
1749 'categories': ['Gaming'],
1750 'view_count': int,
1751 'uploader': 'WiiLikeToPlay',
1752 'uploader_id': '@WLTP',
1753 'uploader_url': 'https://www.youtube.com/@WLTP',
1754 },
1755 }, {
1756 'info_dict': {
1757 'id': 'R7r3vfO7Hao',
1758 'ext': 'mp4',
1759 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1760 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1761 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1762 'like_count': int,
1763 'availability': 'public',
1764 'playable_in_embed': True,
1765 'upload_date': '20131105',
1766 'description': 'md5:563ccbc698b39298481ca3c571169519',
1767 'channel_follower_count': int,
1768 'tags': 'count:24',
1769 'release_date': '20131106',
1770 'comment_count': int,
1771 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1772 'channel': 'WiiLikeToPlay',
1773 'categories': ['Gaming'],
1774 'release_timestamp': 1383701914,
1775 'live_status': 'was_live',
1776 'age_limit': 0,
1777 'duration': 10128,
1778 'view_count': int,
1779 'uploader': 'WiiLikeToPlay',
1780 'uploader_id': '@WLTP',
1781 'uploader_url': 'https://www.youtube.com/@WLTP',
1782 },
1783 }],
1784 'params': {'skip_download': True},
1785 },
1786 {
1787 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1788 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1789 'info_dict': {
1790 'id': 'gVfLd0zydlo',
1791 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1792 },
1793 'playlist_count': 2,
1794 'skip': 'Not multifeed anymore',
1795 },
1796 {
1797 'url': 'https://vid.plus/FlRa-iH7PGw',
1798 'only_matching': True,
1799 },
1800 {
1801 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1802 'only_matching': True,
1803 },
1804 {
1805 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1806 # Also tests cut-off URL expansion in video description (see
1807 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1808 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1809 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1810 'info_dict': {
1811 'id': 'lsguqyKfVQg',
1812 'ext': 'mp4',
1813 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1814 'alt_title': 'Dark Walk',
1815 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1816 'duration': 133,
1817 'upload_date': '20151119',
1818 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1819 'track': 'Dark Walk',
1820 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1821 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1822 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1823 'categories': ['Film & Animation'],
1824 'view_count': int,
1825 'live_status': 'not_live',
1826 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1827 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1828 'tags': 'count:13',
1829 'availability': 'public',
1830 'channel': 'IronSoulElf',
1831 'playable_in_embed': True,
1832 'like_count': int,
1833 'age_limit': 0,
1834 'channel_follower_count': int
1835 },
1836 'params': {
1837 'skip_download': True,
1838 },
1839 },
1840 {
1841 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1842 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1843 'only_matching': True,
1844 },
1845 {
1846 # Video with yt:stretch=17:0
1847 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1848 'info_dict': {
1849 'id': 'Q39EVAstoRM',
1850 'ext': 'mp4',
1851 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1852 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1853 'upload_date': '20151107',
1854 },
1855 'params': {
1856 'skip_download': True,
1857 },
1858 'skip': 'This video does not exist.',
1859 },
1860 {
1861 # Video with incomplete 'yt:stretch=16:'
1862 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1863 'only_matching': True,
1864 },
1865 {
1866 # Video licensed under Creative Commons
1867 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1868 'info_dict': {
1869 'id': 'M4gD1WSo5mA',
1870 'ext': 'mp4',
1871 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1872 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1873 'duration': 721,
1874 'upload_date': '20150128',
1875 'license': 'Creative Commons Attribution license (reuse allowed)',
1876 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1877 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1878 'like_count': int,
1879 'age_limit': 0,
1880 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1881 'channel': 'The Berkman Klein Center for Internet & Society',
1882 'availability': 'public',
1883 'view_count': int,
1884 'categories': ['Education'],
1885 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1886 'live_status': 'not_live',
1887 'playable_in_embed': True,
1888 'channel_follower_count': int,
1889 'chapters': list,
1890 'uploader': 'The Berkman Klein Center for Internet & Society',
1891 'uploader_id': '@BKCHarvard',
1892 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1893 },
1894 'params': {
1895 'skip_download': True,
1896 },
1897 },
1898 {
1899 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1900 'info_dict': {
1901 'id': 'eQcmzGIKrzg',
1902 'ext': 'mp4',
1903 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1904 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1905 'duration': 4060,
1906 'upload_date': '20151120',
1907 'license': 'Creative Commons Attribution license (reuse allowed)',
1908 'playable_in_embed': True,
1909 'tags': 'count:12',
1910 'like_count': int,
1911 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1912 'age_limit': 0,
1913 'availability': 'public',
1914 'categories': ['News & Politics'],
1915 'channel': 'Bernie Sanders',
1916 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1917 'view_count': int,
1918 'live_status': 'not_live',
1919 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1920 'comment_count': int,
1921 'channel_follower_count': int,
1922 'chapters': list,
1923 'uploader': 'Bernie Sanders',
1924 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1925 'uploader_id': '@BernieSanders',
1926 'channel_is_verified': True,
1927 'heatmap': 'count:100',
1928 },
1929 'params': {
1930 'skip_download': True,
1931 },
1932 },
1933 {
1934 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1935 'only_matching': True,
1936 },
1937 {
1938 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1939 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1940 'only_matching': True,
1941 },
1942 {
1943 # Rental video preview
1944 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1945 'info_dict': {
1946 'id': 'uGpuVWrhIzE',
1947 'ext': 'mp4',
1948 'title': 'Piku - Trailer',
1949 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1950 'upload_date': '20150811',
1951 'license': 'Standard YouTube License',
1952 },
1953 'params': {
1954 'skip_download': True,
1955 },
1956 'skip': 'This video is not available.',
1957 },
1958 {
1959 # YouTube Red video with episode data
1960 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1961 'info_dict': {
1962 'id': 'iqKdEhx-dD4',
1963 'ext': 'mp4',
1964 'title': 'Isolation - Mind Field (Ep 1)',
1965 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1966 'duration': 2085,
1967 'upload_date': '20170118',
1968 'series': 'Mind Field',
1969 'season_number': 1,
1970 'episode_number': 1,
1971 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1972 'tags': 'count:12',
1973 'view_count': int,
1974 'availability': 'public',
1975 'age_limit': 0,
1976 'channel': 'Vsauce',
1977 'episode': 'Episode 1',
1978 'categories': ['Entertainment'],
1979 'season': 'Season 1',
1980 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1981 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1982 'like_count': int,
1983 'playable_in_embed': True,
1984 'live_status': 'not_live',
1985 'channel_follower_count': int,
1986 'uploader': 'Vsauce',
1987 'uploader_url': 'https://www.youtube.com/@Vsauce',
1988 'uploader_id': '@Vsauce',
1989 'comment_count': int,
1990 'channel_is_verified': True,
1991 },
1992 'params': {
1993 'skip_download': True,
1994 },
1995 'expected_warnings': [
1996 'Skipping DASH manifest',
1997 ],
1998 },
1999 {
2000 # The following content has been identified by the YouTube community
2001 # as inappropriate or offensive to some audiences.
2002 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
2003 'info_dict': {
2004 'id': '6SJNVb0GnPI',
2005 'ext': 'mp4',
2006 'title': 'Race Differences in Intelligence',
2007 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
2008 'duration': 965,
2009 'upload_date': '20140124',
2010 },
2011 'params': {
2012 'skip_download': True,
2013 },
2014 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
2015 },
2016 {
2017 # itag 212
2018 'url': '1t24XAntNCY',
2019 'only_matching': True,
2020 },
2021 {
2022 # geo restricted to JP
2023 'url': 'sJL6WA-aGkQ',
2024 'only_matching': True,
2025 },
2026 {
2027 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2028 'only_matching': True,
2029 },
2030 {
2031 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2032 'only_matching': True,
2033 },
2034 {
2035 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2036 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2037 'only_matching': True,
2038 },
2039 {
2040 # DRM protected
2041 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2042 'only_matching': True,
2043 },
2044 {
2045 # Video with unsupported adaptive stream type formats
2046 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2047 'info_dict': {
2048 'id': 'Z4Vy8R84T1U',
2049 'ext': 'mp4',
2050 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2051 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2052 'duration': 433,
2053 'upload_date': '20130923',
2054 'formats': 'maxcount:10',
2055 },
2056 'params': {
2057 'skip_download': True,
2058 'youtube_include_dash_manifest': False,
2059 },
2060 'skip': 'not actual anymore',
2061 },
2062 {
2063 # Youtube Music Auto-generated description
2064 # TODO: fix metadata extraction
2065 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2066 'info_dict': {
2067 'id': 'MgNrAu2pzNs',
2068 'ext': 'mp4',
2069 'title': 'Voyeur Girl',
2070 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2071 'upload_date': '20190312',
2072 'artists': ['Stephen'],
2073 'creators': ['Stephen'],
2074 'track': 'Voyeur Girl',
2075 'album': 'it\'s too much love to know my dear',
2076 'release_date': '20190313',
2077 'alt_title': 'Voyeur Girl',
2078 'view_count': int,
2079 'playable_in_embed': True,
2080 'like_count': int,
2081 'categories': ['Music'],
2082 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2083 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2084 'uploader': 'Stephen',
2085 'availability': 'public',
2086 'duration': 169,
2087 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2088 'age_limit': 0,
2089 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2090 'tags': 'count:11',
2091 'live_status': 'not_live',
2092 'channel_follower_count': int
2093 },
2094 'params': {
2095 'skip_download': True,
2096 },
2097 },
2098 {
2099 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2100 'only_matching': True,
2101 },
2102 {
2103 # invalid -> valid video id redirection
2104 'url': 'DJztXj2GPfl',
2105 'info_dict': {
2106 'id': 'DJztXj2GPfk',
2107 'ext': 'mp4',
2108 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2109 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2110 'upload_date': '20090125',
2111 'artist': 'Panjabi MC',
2112 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2113 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2114 },
2115 'params': {
2116 'skip_download': True,
2117 },
2118 'skip': 'Video unavailable',
2119 },
2120 {
2121 # empty description results in an empty string
2122 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2123 'info_dict': {
2124 'id': 'x41yOUIvK2k',
2125 'ext': 'mp4',
2126 'title': 'IMG 3456',
2127 'description': '',
2128 'upload_date': '20170613',
2129 'view_count': int,
2130 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2131 'like_count': int,
2132 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2133 'tags': [],
2134 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2135 'availability': 'public',
2136 'age_limit': 0,
2137 'categories': ['Pets & Animals'],
2138 'duration': 7,
2139 'playable_in_embed': True,
2140 'live_status': 'not_live',
2141 'channel': 'l\'Or Vert asbl',
2142 'channel_follower_count': int,
2143 'uploader': 'l\'Or Vert asbl',
2144 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2145 'uploader_id': '@ElevageOrVert',
2146 },
2147 'params': {
2148 'skip_download': True,
2149 },
2150 },
2151 {
2152 # with '};' inside yt initial data (see [1])
2153 # see [2] for an example with '};' inside ytInitialPlayerResponse
2154 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2155 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2156 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2157 'info_dict': {
2158 'id': 'CHqg6qOn4no',
2159 'ext': 'mp4',
2160 'title': 'Part 77 Sort a list of simple types in c#',
2161 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2162 'upload_date': '20130831',
2163 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2164 'like_count': int,
2165 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2166 'live_status': 'not_live',
2167 'categories': ['Education'],
2168 'availability': 'public',
2169 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2170 'tags': 'count:12',
2171 'playable_in_embed': True,
2172 'age_limit': 0,
2173 'view_count': int,
2174 'duration': 522,
2175 'channel': 'kudvenkat',
2176 'comment_count': int,
2177 'channel_follower_count': int,
2178 'chapters': list,
2179 'uploader': 'kudvenkat',
2180 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2181 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2182 'channel_is_verified': True,
2183 'heatmap': 'count:100',
2184 },
2185 'params': {
2186 'skip_download': True,
2187 },
2188 },
2189 {
2190 # another example of '};' in ytInitialData
2191 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2192 'only_matching': True,
2193 },
2194 {
2195 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2196 'only_matching': True,
2197 },
2198 {
2199 # https://github.com/ytdl-org/youtube-dl/pull/28094
2200 'url': 'OtqTfy26tG0',
2201 'info_dict': {
2202 'id': 'OtqTfy26tG0',
2203 'ext': 'mp4',
2204 'title': 'Burn Out',
2205 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2206 'upload_date': '20141120',
2207 'artist': 'The Cinematic Orchestra',
2208 'track': 'Burn Out',
2209 'album': 'Every Day',
2210 'like_count': int,
2211 'live_status': 'not_live',
2212 'alt_title': 'Burn Out',
2213 'duration': 614,
2214 'age_limit': 0,
2215 'view_count': int,
2216 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2217 'creator': 'The Cinematic Orchestra',
2218 'channel': 'The Cinematic Orchestra',
2219 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2220 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2221 'availability': 'public',
2222 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2223 'categories': ['Music'],
2224 'playable_in_embed': True,
2225 'channel_follower_count': int,
2226 'uploader': 'The Cinematic Orchestra',
2227 'comment_count': int,
2228 },
2229 'params': {
2230 'skip_download': True,
2231 },
2232 },
2233 {
2234 # controversial video, only works with bpctr when authenticated with cookies
2235 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2236 'only_matching': True,
2237 },
2238 {
2239 # controversial video, requires bpctr/contentCheckOk
2240 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2241 'info_dict': {
2242 'id': 'SZJvDhaSDnc',
2243 'ext': 'mp4',
2244 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2245 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2246 'upload_date': '20140716',
2247 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2248 'duration': 170,
2249 'categories': ['News & Politics'],
2250 'view_count': int,
2251 'channel': 'CBS Mornings',
2252 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2253 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2254 'age_limit': 18,
2255 'availability': 'needs_auth',
2256 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2257 'like_count': int,
2258 'live_status': 'not_live',
2259 'playable_in_embed': True,
2260 'channel_follower_count': int,
2261 'uploader': 'CBS Mornings',
2262 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2263 'uploader_id': '@CBSMornings',
2264 'comment_count': int,
2265 'channel_is_verified': True,
2266 }
2267 },
2268 {
2269 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2270 'url': 'cBvYw8_A0vQ',
2271 'info_dict': {
2272 'id': 'cBvYw8_A0vQ',
2273 'ext': 'mp4',
2274 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2275 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2276 'upload_date': '20201120',
2277 'duration': 1456,
2278 'categories': ['Travel & Events'],
2279 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2280 'view_count': int,
2281 'channel': 'Walk around Japan',
2282 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2283 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2284 'age_limit': 0,
2285 'availability': 'public',
2286 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2287 'live_status': 'not_live',
2288 'playable_in_embed': True,
2289 'channel_follower_count': int,
2290 'uploader': 'Walk around Japan',
2291 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2292 'uploader_id': '@walkaroundjapan7124',
2293 },
2294 'params': {
2295 'skip_download': True,
2296 },
2297 }, {
2298 # Has multiple audio streams
2299 'url': 'WaOKSUlf4TM',
2300 'only_matching': True
2301 }, {
2302 # Requires Premium: has format 141 when requested using YTM url
2303 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2304 'only_matching': True
2305 }, {
2306 # multiple subtitles with same lang_code
2307 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2308 'only_matching': True,
2309 }, {
2310 # Force use android client fallback
2311 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2312 'info_dict': {
2313 'id': 'YOelRv7fMxY',
2314 'title': 'DIGGING A SECRET TUNNEL Part 1',
2315 'ext': '3gp',
2316 'upload_date': '20210624',
2317 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2318 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2319 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2320 'duration': 596,
2321 'categories': ['Entertainment'],
2322 'view_count': int,
2323 'channel': 'colinfurze',
2324 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2325 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2326 'age_limit': 0,
2327 'availability': 'public',
2328 'like_count': int,
2329 'live_status': 'not_live',
2330 'playable_in_embed': True,
2331 'channel_follower_count': int,
2332 'chapters': list,
2333 'uploader': 'colinfurze',
2334 'uploader_url': 'https://www.youtube.com/@colinfurze',
2335 'uploader_id': '@colinfurze',
2336 'comment_count': int,
2337 'channel_is_verified': True,
2338 'heatmap': 'count:100',
2339 },
2340 'params': {
2341 'format': '17', # 3gp format available on android
2342 'extractor_args': {'youtube': {'player_client': ['android']}},
2343 },
2344 },
2345 {
2346 # Skip download of additional client configs (remix client config in this case)
2347 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2348 'only_matching': True,
2349 'params': {
2350 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2351 },
2352 }, {
2353 # shorts
2354 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2355 'only_matching': True,
2356 }, {
2357 'note': 'Storyboards',
2358 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2359 'info_dict': {
2360 'id': '5KLPxDtMqe8',
2361 'ext': 'mhtml',
2362 'format_id': 'sb0',
2363 'title': 'Your Brain is Plastic',
2364 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2365 'upload_date': '20140324',
2366 'like_count': int,
2367 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2368 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2369 'view_count': int,
2370 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2371 'playable_in_embed': True,
2372 'tags': 'count:12',
2373 'availability': 'public',
2374 'channel': 'SciShow',
2375 'live_status': 'not_live',
2376 'duration': 248,
2377 'categories': ['Education'],
2378 'age_limit': 0,
2379 'channel_follower_count': int,
2380 'chapters': list,
2381 'uploader': 'SciShow',
2382 'uploader_url': 'https://www.youtube.com/@SciShow',
2383 'uploader_id': '@SciShow',
2384 'comment_count': int,
2385 'channel_is_verified': True,
2386 'heatmap': 'count:100',
2387 }, 'params': {'format': 'mhtml', 'skip_download': True}
2388 }, {
2389 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2390 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2391 'info_dict': {
2392 'id': '2NUZ8W2llS4',
2393 'ext': 'mp4',
2394 'title': 'The NP that test your phone performance 🙂',
2395 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2396 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2397 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2398 'duration': 21,
2399 'view_count': int,
2400 'age_limit': 0,
2401 'categories': ['Gaming'],
2402 'tags': 'count:23',
2403 'playable_in_embed': True,
2404 'live_status': 'not_live',
2405 'upload_date': '20220103',
2406 'like_count': int,
2407 'availability': 'public',
2408 'channel': 'Leon Nguyen',
2409 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2410 'comment_count': int,
2411 'channel_follower_count': int,
2412 'uploader': 'Leon Nguyen',
2413 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2414 'uploader_id': '@LeonNguyen',
2415 'heatmap': 'count:100',
2416 }
2417 }, {
2418 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2419 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2420 'info_dict': {
2421 'id': '2NUZ8W2llS4',
2422 'ext': 'mp4',
2423 'title': 'The NP that test your phone performance 🙂',
2424 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2425 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2426 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2427 'duration': 21,
2428 'view_count': int,
2429 'age_limit': 0,
2430 'categories': ['Gaming'],
2431 'tags': 'count:23',
2432 'playable_in_embed': True,
2433 'live_status': 'not_live',
2434 'upload_date': '20220102',
2435 'like_count': int,
2436 'availability': 'public',
2437 'channel': 'Leon Nguyen',
2438 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2439 'comment_count': int,
2440 'channel_follower_count': int,
2441 'uploader': 'Leon Nguyen',
2442 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2443 'uploader_id': '@LeonNguyen',
2444 'heatmap': 'count:100',
2445 },
2446 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2447 }, {
2448 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2449 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2450 'info_dict': {
2451 'id': 'mzZzzBU6lrM',
2452 'ext': 'mp4',
2453 'title': 'I Met GeorgeNotFound In Real Life...',
2454 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2455 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2456 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2457 'duration': 955,
2458 'view_count': int,
2459 'age_limit': 0,
2460 'categories': ['Entertainment'],
2461 'tags': 'count:26',
2462 'playable_in_embed': True,
2463 'live_status': 'not_live',
2464 'release_timestamp': 1641172509,
2465 'release_date': '20220103',
2466 'upload_date': '20220103',
2467 'like_count': int,
2468 'availability': 'public',
2469 'channel': 'Quackity',
2470 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2471 'channel_follower_count': int,
2472 'uploader': 'Quackity',
2473 'uploader_id': '@Quackity',
2474 'uploader_url': 'https://www.youtube.com/@Quackity',
2475 'comment_count': int,
2476 'channel_is_verified': True,
2477 'heatmap': 'count:100',
2478 }
2479 },
2480 { # continuous livestream. Microformat upload date should be preferred.
2481 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2482 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2483 'info_dict': {
2484 'id': 'kgx4WGK0oNU',
2485 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2486 'ext': 'mp4',
2487 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2488 'availability': 'public',
2489 'age_limit': 0,
2490 'release_timestamp': 1637975704,
2491 'upload_date': '20210619',
2492 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2493 'live_status': 'is_live',
2494 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2495 'channel': 'Abao in Tokyo',
2496 'channel_follower_count': int,
2497 'release_date': '20211127',
2498 'tags': 'count:39',
2499 'categories': ['People & Blogs'],
2500 'like_count': int,
2501 'view_count': int,
2502 'playable_in_embed': True,
2503 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2504 'concurrent_view_count': int,
2505 'uploader': 'Abao in Tokyo',
2506 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2507 'uploader_id': '@abaointokyo',
2508 },
2509 'params': {'skip_download': True}
2510 }, {
2511 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2512 'info_dict': {
2513 'id': 'tjjjtzRLHvA',
2514 'ext': 'mp4',
2515 'title': 'ハッシュタグ無し };if window.ytcsi',
2516 'upload_date': '20220323',
2517 'like_count': int,
2518 'availability': 'unlisted',
2519 'channel': 'Lesmiscore',
2520 'thumbnail': r're:^https?://.*\.jpg',
2521 'age_limit': 0,
2522 'categories': ['Music'],
2523 'view_count': int,
2524 'description': '',
2525 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2526 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2527 'live_status': 'not_live',
2528 'playable_in_embed': True,
2529 'channel_follower_count': int,
2530 'duration': 6,
2531 'tags': [],
2532 'uploader_id': '@lesmiscore',
2533 'uploader': 'Lesmiscore',
2534 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2535 }
2536 }, {
2537 # Prefer primary title+description language metadata by default
2538 # Do not prefer translated description if primary is empty
2539 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2540 'info_dict': {
2541 'id': 'el3E4MbxRqQ',
2542 'ext': 'mp4',
2543 'title': 'dlp test video 2 - primary sv no desc',
2544 'description': '',
2545 'channel': 'cole-dlp-test-acc',
2546 'tags': [],
2547 'view_count': int,
2548 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2549 'like_count': int,
2550 'playable_in_embed': True,
2551 'availability': 'unlisted',
2552 'thumbnail': r're:^https?://.*\.jpg',
2553 'age_limit': 0,
2554 'duration': 5,
2555 'live_status': 'not_live',
2556 'upload_date': '20220908',
2557 'categories': ['People & Blogs'],
2558 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2559 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2560 'uploader_id': '@coletdjnz',
2561 'uploader': 'cole-dlp-test-acc',
2562 },
2563 'params': {'skip_download': True}
2564 }, {
2565 # Extractor argument: prefer translated title+description
2566 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2567 'info_dict': {
2568 'id': 'gHKT4uU8Zng',
2569 'ext': 'mp4',
2570 'channel': 'cole-dlp-test-acc',
2571 'tags': [],
2572 'duration': 5,
2573 'live_status': 'not_live',
2574 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2575 'upload_date': '20220728',
2576 'view_count': int,
2577 'categories': ['People & Blogs'],
2578 'thumbnail': r're:^https?://.*\.jpg',
2579 'title': 'dlp test video title translated (fr)',
2580 'availability': 'public',
2581 'age_limit': 0,
2582 'description': 'dlp test video description translated (fr)',
2583 'playable_in_embed': True,
2584 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2585 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2586 'uploader_id': '@coletdjnz',
2587 'uploader': 'cole-dlp-test-acc',
2588 },
2589 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2590 'expected_warnings': [r'Preferring "fr" translated fields'],
2591 }, {
2592 'note': '6 channel audio',
2593 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2594 'only_matching': True,
2595 }, {
2596 'note': 'Multiple HLS formats with same itag',
2597 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2598 'info_dict': {
2599 'id': 'kX3nB4PpJko',
2600 'ext': 'mp4',
2601 'categories': ['Entertainment'],
2602 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2603 'live_status': 'not_live',
2604 'duration': 937,
2605 'channel_follower_count': int,
2606 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2607 'title': 'Last To Take Hand Off Jet, Keeps It!',
2608 'channel': 'MrBeast',
2609 'playable_in_embed': True,
2610 'view_count': int,
2611 'upload_date': '20221112',
2612 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2613 'age_limit': 0,
2614 'availability': 'public',
2615 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2616 'like_count': int,
2617 'tags': [],
2618 'uploader': 'MrBeast',
2619 'uploader_url': 'https://www.youtube.com/@MrBeast',
2620 'uploader_id': '@MrBeast',
2621 'comment_count': int,
2622 'channel_is_verified': True,
2623 'heatmap': 'count:100',
2624 },
2625 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2626 }, {
2627 'note': 'Audio formats with Dynamic Range Compression',
2628 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2629 'info_dict': {
2630 'id': 'Tq92D6wQ1mg',
2631 'ext': 'webm',
2632 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2633 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2634 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2635 'channel_follower_count': int,
2636 'description': 'md5:17eccca93a786d51bc67646756894066',
2637 'upload_date': '20191228',
2638 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2639 'playable_in_embed': True,
2640 'like_count': int,
2641 'categories': ['Entertainment'],
2642 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2643 'age_limit': 18,
2644 'channel': 'Projekt Melody',
2645 'view_count': int,
2646 'availability': 'needs_auth',
2647 'comment_count': int,
2648 'live_status': 'not_live',
2649 'duration': 106,
2650 'uploader': 'Projekt Melody',
2651 'uploader_id': '@ProjektMelody',
2652 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2653 },
2654 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2655 },
2656 {
2657 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2658 'info_dict': {
2659 'id': 'qVv6vCqciTM',
2660 'ext': 'mp4',
2661 'age_limit': 0,
2662 'comment_count': int,
2663 'chapters': 'count:13',
2664 'upload_date': '20221223',
2665 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2666 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2667 'like_count': int,
2668 'release_date': '20221223',
2669 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2670 '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】',
2671 'view_count': int,
2672 'playable_in_embed': True,
2673 'duration': 4438,
2674 'availability': 'public',
2675 'channel_follower_count': int,
2676 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2677 'categories': ['Entertainment'],
2678 'live_status': 'was_live',
2679 'release_timestamp': 1671793345,
2680 'channel': 'さなちゃんねる',
2681 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2682 'uploader': 'さなちゃんねる',
2683 'uploader_url': 'https://www.youtube.com/@sana_natori',
2684 'uploader_id': '@sana_natori',
2685 'channel_is_verified': True,
2686 'heatmap': 'count:100',
2687 },
2688 },
2689 {
2690 # Fallbacks when webpage and web client is unavailable
2691 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2692 'info_dict': {
2693 'id': 'wSSmNUl9Snw',
2694 'ext': 'mp4',
2695 # 'categories': ['Science & Technology'],
2696 'view_count': int,
2697 'chapters': 'count:2',
2698 'channel': 'Scott Manley',
2699 'like_count': int,
2700 'age_limit': 0,
2701 # 'availability': 'public',
2702 'channel_follower_count': int,
2703 'live_status': 'not_live',
2704 'upload_date': '20170831',
2705 'duration': 682,
2706 'tags': 'count:8',
2707 'uploader_url': 'https://www.youtube.com/@scottmanley',
2708 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2709 'uploader': 'Scott Manley',
2710 'uploader_id': '@scottmanley',
2711 'title': 'The Computer Hack That Saved Apollo 14',
2712 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2713 'thumbnail': r're:^https?://.*\.webp',
2714 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2715 'playable_in_embed': True,
2716 'comment_count': int,
2717 'channel_is_verified': True,
2718 'heatmap': 'count:100',
2719 },
2720 'params': {
2721 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2722 },
2723 },
2724 ]
2725
2726 _WEBPAGE_TESTS = [
2727 # YouTube <object> embed
2728 {
2729 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2730 'md5': '873c81d308b979f0e23ee7e620b312a3',
2731 'info_dict': {
2732 'id': 'msN87y-iEx0',
2733 'ext': 'mp4',
2734 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2735 'upload_date': '20080526',
2736 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2737 'age_limit': 0,
2738 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2739 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2740 'playable_in_embed': True,
2741 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2742 'like_count': int,
2743 'comment_count': int,
2744 'channel': 'Christopher Sykes',
2745 'live_status': 'not_live',
2746 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2747 'availability': 'public',
2748 'duration': 195,
2749 'view_count': int,
2750 'categories': ['Science & Technology'],
2751 'channel_follower_count': int,
2752 'uploader': 'Christopher Sykes',
2753 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2754 'uploader_id': '@ChristopherSykesDocumentaries',
2755 'heatmap': 'count:100',
2756 },
2757 'params': {
2758 'skip_download': True,
2759 }
2760 },
2761 ]
2762
2763 @classmethod
2764 def suitable(cls, url):
2765 from ..utils import parse_qs
2766
2767 qs = parse_qs(url)
2768 if qs.get('list', [None])[0]:
2769 return False
2770 return super().suitable(url)
2771
2772 def __init__(self, *args, **kwargs):
2773 super().__init__(*args, **kwargs)
2774 self._code_cache = {}
2775 self._player_cache = {}
2776
2777 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2778 lock = threading.Lock()
2779 start_time = time.time()
2780 formats = [f for f in formats if f.get('is_from_start')]
2781
2782 def refetch_manifest(format_id, delay):
2783 nonlocal formats, start_time, is_live
2784 if time.time() <= start_time + delay:
2785 return
2786
2787 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2788 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2789 microformats = traverse_obj(
2790 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2791 expected_type=dict)
2792 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2793 is_live = live_status == 'is_live'
2794 start_time = time.time()
2795
2796 def mpd_feed(format_id, delay):
2797 """
2798 @returns (manifest_url, manifest_stream_number, is_live) or None
2799 """
2800 for retry in self.RetryManager(fatal=False):
2801 with lock:
2802 refetch_manifest(format_id, delay)
2803
2804 f = next((f for f in formats if f['format_id'] == format_id), None)
2805 if not f:
2806 if not is_live:
2807 retry.error = f'{video_id}: Video is no longer live'
2808 else:
2809 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2810 continue
2811 return f['manifest_url'], f['manifest_stream_number'], is_live
2812 return None
2813
2814 for f in formats:
2815 f['is_live'] = is_live
2816 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2817 live_start_time, mpd_feed, not is_live and f.copy())
2818 if is_live:
2819 f['fragments'] = gen
2820 f['protocol'] = 'http_dash_segments_generator'
2821 else:
2822 f['fragments'] = LazyList(gen({}))
2823 del f['is_from_start']
2824
2825 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2826 FETCH_SPAN, MAX_DURATION = 5, 432000
2827
2828 mpd_url, stream_number, is_live = None, None, True
2829
2830 begin_index = 0
2831 download_start_time = ctx.get('start') or time.time()
2832
2833 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2834 if lack_early_segments:
2835 self.report_warning(bug_reports_message(
2836 'Starting download from the last 120 hours of the live stream since '
2837 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2838 lack_early_segments = True
2839
2840 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2841 fragments, fragment_base_url = None, None
2842
2843 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2844 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2845 # Obtain from MPD's maximum seq value
2846 old_mpd_url = mpd_url
2847 last_error = ctx.pop('last_error', None)
2848 expire_fast = immediate or last_error and isinstance(last_error, HTTPError) and last_error.status == 403
2849 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2850 or (mpd_url, stream_number, False))
2851 if not refresh_sequence:
2852 if expire_fast and not is_live:
2853 return False, last_seq
2854 elif old_mpd_url == mpd_url:
2855 return True, last_seq
2856 if manifestless_orig_fmt:
2857 fmt_info = manifestless_orig_fmt
2858 else:
2859 try:
2860 fmts, _ = self._extract_mpd_formats_and_subtitles(
2861 mpd_url, None, note=False, errnote=False, fatal=False)
2862 except ExtractorError:
2863 fmts = None
2864 if not fmts:
2865 no_fragment_score += 2
2866 return False, last_seq
2867 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2868 fragments = fmt_info['fragments']
2869 fragment_base_url = fmt_info['fragment_base_url']
2870 assert fragment_base_url
2871
2872 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2873 return True, _last_seq
2874
2875 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2876 while is_live:
2877 fetch_time = time.time()
2878 if no_fragment_score > 30:
2879 return
2880 if last_segment_url:
2881 # Obtain from "X-Head-Seqnum" header value from each segment
2882 try:
2883 urlh = self._request_webpage(
2884 last_segment_url, None, note=False, errnote=False, fatal=False)
2885 except ExtractorError:
2886 urlh = None
2887 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2888 if last_seq is None:
2889 no_fragment_score += 2
2890 last_segment_url = None
2891 continue
2892 else:
2893 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2894 no_fragment_score += 2
2895 if not should_continue:
2896 continue
2897
2898 if known_idx > last_seq:
2899 last_segment_url = None
2900 continue
2901
2902 last_seq += 1
2903
2904 if begin_index < 0 and known_idx < 0:
2905 # skip from the start when it's negative value
2906 known_idx = last_seq + begin_index
2907 if lack_early_segments:
2908 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2909 try:
2910 for idx in range(known_idx, last_seq):
2911 # do not update sequence here or you'll get skipped some part of it
2912 should_continue, _ = _extract_sequence_from_mpd(False, False)
2913 if not should_continue:
2914 known_idx = idx - 1
2915 raise ExtractorError('breaking out of outer loop')
2916 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2917 yield {
2918 'url': last_segment_url,
2919 'fragment_count': last_seq,
2920 }
2921 if known_idx == last_seq:
2922 no_fragment_score += 5
2923 else:
2924 no_fragment_score = 0
2925 known_idx = last_seq
2926 except ExtractorError:
2927 continue
2928
2929 if manifestless_orig_fmt:
2930 # Stop at the first iteration if running for post-live manifestless;
2931 # fragment count no longer increase since it starts
2932 break
2933
2934 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2935
2936 def _extract_player_url(self, *ytcfgs, webpage=None):
2937 player_url = traverse_obj(
2938 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2939 get_all=False, expected_type=str)
2940 if not player_url:
2941 return
2942 return urljoin('https://www.youtube.com', player_url)
2943
2944 def _download_player_url(self, video_id, fatal=False):
2945 res = self._download_webpage(
2946 'https://www.youtube.com/iframe_api',
2947 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2948 if res:
2949 player_version = self._search_regex(
2950 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2951 if player_version:
2952 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2953
2954 def _signature_cache_id(self, example_sig):
2955 """ Return a string representation of a signature """
2956 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2957
2958 @classmethod
2959 def _extract_player_info(cls, player_url):
2960 for player_re in cls._PLAYER_INFO_RE:
2961 id_m = re.search(player_re, player_url)
2962 if id_m:
2963 break
2964 else:
2965 raise ExtractorError('Cannot identify player %r' % player_url)
2966 return id_m.group('id')
2967
2968 def _load_player(self, video_id, player_url, fatal=True):
2969 player_id = self._extract_player_info(player_url)
2970 if player_id not in self._code_cache:
2971 code = self._download_webpage(
2972 player_url, video_id, fatal=fatal,
2973 note='Downloading player ' + player_id,
2974 errnote='Download of %s failed' % player_url)
2975 if code:
2976 self._code_cache[player_id] = code
2977 return self._code_cache.get(player_id)
2978
2979 def _extract_signature_function(self, video_id, player_url, example_sig):
2980 player_id = self._extract_player_info(player_url)
2981
2982 # Read from filesystem cache
2983 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2984 assert os.path.basename(func_id) == func_id
2985
2986 self.write_debug(f'Extracting signature function {func_id}')
2987 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2988
2989 if not cache_spec:
2990 code = self._load_player(video_id, player_url)
2991 if code:
2992 res = self._parse_sig_js(code)
2993 test_string = ''.join(map(chr, range(len(example_sig))))
2994 cache_spec = [ord(c) for c in res(test_string)]
2995 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2996
2997 return lambda s: ''.join(s[i] for i in cache_spec)
2998
2999 def _print_sig_code(self, func, example_sig):
3000 if not self.get_param('youtube_print_sig_code'):
3001 return
3002
3003 def gen_sig_code(idxs):
3004 def _genslice(start, end, step):
3005 starts = '' if start == 0 else str(start)
3006 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
3007 steps = '' if step == 1 else (':%d' % step)
3008 return f's[{starts}{ends}{steps}]'
3009
3010 step = None
3011 # Quelch pyflakes warnings - start will be set when step is set
3012 start = '(Never used)'
3013 for i, prev in zip(idxs[1:], idxs[:-1]):
3014 if step is not None:
3015 if i - prev == step:
3016 continue
3017 yield _genslice(start, prev, step)
3018 step = None
3019 continue
3020 if i - prev in [-1, 1]:
3021 step = i - prev
3022 start = prev
3023 continue
3024 else:
3025 yield 's[%d]' % prev
3026 if step is None:
3027 yield 's[%d]' % i
3028 else:
3029 yield _genslice(start, i, step)
3030
3031 test_string = ''.join(map(chr, range(len(example_sig))))
3032 cache_res = func(test_string)
3033 cache_spec = [ord(c) for c in cache_res]
3034 expr_code = ' + '.join(gen_sig_code(cache_spec))
3035 signature_id_tuple = '(%s)' % (
3036 ', '.join(str(len(p)) for p in example_sig.split('.')))
3037 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
3038 ' return %s\n') % (signature_id_tuple, expr_code)
3039 self.to_screen('Extracted signature function:\n' + code)
3040
3041 def _parse_sig_js(self, jscode):
3042 funcname = self._search_regex(
3043 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3044 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3045 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3046 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3047 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}\(a,\d+\))?',
3048 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3049 # Obsolete patterns
3050 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3051 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3052 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3053 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3054 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3055 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3056 jscode, 'Initial JS player signature function name', group='sig')
3057
3058 jsi = JSInterpreter(jscode)
3059 initial_function = jsi.extract_function(funcname)
3060 return lambda s: initial_function([s])
3061
3062 def _cached(self, func, *cache_id):
3063 def inner(*args, **kwargs):
3064 if cache_id not in self._player_cache:
3065 try:
3066 self._player_cache[cache_id] = func(*args, **kwargs)
3067 except ExtractorError as e:
3068 self._player_cache[cache_id] = e
3069 except Exception as e:
3070 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3071
3072 ret = self._player_cache[cache_id]
3073 if isinstance(ret, Exception):
3074 raise ret
3075 return ret
3076 return inner
3077
3078 def _decrypt_signature(self, s, video_id, player_url):
3079 """Turn the encrypted s field into a working signature"""
3080 extract_sig = self._cached(
3081 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3082 func = extract_sig(video_id, player_url, s)
3083 self._print_sig_code(func, s)
3084 return func(s)
3085
3086 def _decrypt_nsig(self, s, video_id, player_url):
3087 """Turn the encrypted n field into a working signature"""
3088 if player_url is None:
3089 raise ExtractorError('Cannot decrypt nsig without player_url')
3090 player_url = urljoin('https://www.youtube.com', player_url)
3091
3092 try:
3093 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3094 except ExtractorError as e:
3095 raise ExtractorError('Unable to extract nsig function code', cause=e)
3096 if self.get_param('youtube_print_sig_code'):
3097 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3098
3099 try:
3100 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3101 ret = extract_nsig(jsi, func_code)(s)
3102 except JSInterpreter.Exception as e:
3103 try:
3104 jsi = PhantomJSwrapper(self, timeout=5000)
3105 except ExtractorError:
3106 raise e
3107 self.report_warning(
3108 f'Native nsig extraction failed: Trying with PhantomJS\n'
3109 f' n = {s} ; player = {player_url}', video_id)
3110 self.write_debug(e, only_once=True)
3111
3112 args, func_body = func_code
3113 ret = jsi.execute(
3114 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3115 video_id=video_id, note='Executing signature code').strip()
3116
3117 self.write_debug(f'Decrypted nsig {s} => {ret}')
3118 return ret
3119
3120 def _extract_n_function_name(self, jscode):
3121 funcname, idx = self._search_regex(
3122 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3123 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3124 if not idx:
3125 return funcname
3126
3127 return json.loads(js_to_json(self._search_regex(
3128 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
3129 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3130
3131 def _extract_n_function_code(self, video_id, player_url):
3132 player_id = self._extract_player_info(player_url)
3133 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3134 jscode = func_code or self._load_player(video_id, player_url)
3135 jsi = JSInterpreter(jscode)
3136
3137 if func_code:
3138 return jsi, player_id, func_code
3139
3140 func_name = self._extract_n_function_name(jscode)
3141
3142 # For redundancy
3143 func_code = self._search_regex(
3144 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3145 # NB: The end of the regex is intentionally kept strict
3146 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3147 jscode, 'nsig function', group=('var', 'code'), default=None)
3148 if func_code:
3149 func_code = ([func_code[0]], func_code[1])
3150 else:
3151 self.write_debug('Extracting nsig function with jsinterp')
3152 func_code = jsi.extract_function_code(func_name)
3153
3154 self.cache.store('youtube-nsig', player_id, func_code)
3155 return jsi, player_id, func_code
3156
3157 def _extract_n_function_from_code(self, jsi, func_code):
3158 func = jsi.extract_function_from_code(*func_code)
3159
3160 def extract_nsig(s):
3161 try:
3162 ret = func([s])
3163 except JSInterpreter.Exception:
3164 raise
3165 except Exception as e:
3166 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3167
3168 if ret.startswith('enhanced_except_'):
3169 raise JSInterpreter.Exception('Signature function returned an exception')
3170 return ret
3171
3172 return extract_nsig
3173
3174 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3175 """
3176 Extract signatureTimestamp (sts)
3177 Required to tell API what sig/player version is in use.
3178 """
3179 sts = None
3180 if isinstance(ytcfg, dict):
3181 sts = int_or_none(ytcfg.get('STS'))
3182
3183 if not sts:
3184 # Attempt to extract from player
3185 if player_url is None:
3186 error_msg = 'Cannot extract signature timestamp without player_url.'
3187 if fatal:
3188 raise ExtractorError(error_msg)
3189 self.report_warning(error_msg)
3190 return
3191 code = self._load_player(video_id, player_url, fatal=fatal)
3192 if code:
3193 sts = int_or_none(self._search_regex(
3194 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3195 'JS player signature timestamp', group='sts', fatal=fatal))
3196 return sts
3197
3198 def _mark_watched(self, video_id, player_responses):
3199 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3200 label = 'fully ' if is_full else ''
3201 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3202 expected_type=url_or_none)
3203 if not url:
3204 self.report_warning(f'Unable to mark {label}watched')
3205 return
3206 parsed_url = urllib.parse.urlparse(url)
3207 qs = urllib.parse.parse_qs(parsed_url.query)
3208
3209 # cpn generation algorithm is reverse engineered from base.js.
3210 # In fact it works even with dummy cpn.
3211 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3212 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3213
3214 # # more consistent results setting it to right before the end
3215 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3216
3217 qs.update({
3218 'ver': ['2'],
3219 'cpn': [cpn],
3220 'cmt': video_length,
3221 'el': 'detailpage', # otherwise defaults to "shorts"
3222 })
3223
3224 if is_full:
3225 # these seem to mark watchtime "history" in the real world
3226 # they're required, so send in a single value
3227 qs.update({
3228 'st': 0,
3229 'et': video_length,
3230 })
3231
3232 url = urllib.parse.urlunparse(
3233 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3234
3235 self._download_webpage(
3236 url, video_id, f'Marking {label}watched',
3237 'Unable to mark watched', fatal=False)
3238
3239 @classmethod
3240 def _extract_from_webpage(cls, url, webpage):
3241 # Invidious Instances
3242 # https://github.com/yt-dlp/yt-dlp/issues/195
3243 # https://github.com/iv-org/invidious/pull/1730
3244 mobj = re.search(
3245 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3246 webpage)
3247 if mobj:
3248 yield cls.url_result(mobj.group('url'), cls)
3249 raise cls.StopExtraction()
3250
3251 yield from super()._extract_from_webpage(url, webpage)
3252
3253 # lazyYT YouTube embed
3254 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3255 yield cls.url_result(unescapeHTML(id_), cls, id_)
3256
3257 # Wordpress "YouTube Video Importer" plugin
3258 for m in re.findall(r'''(?x)<div[^>]+
3259 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3260 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3261 yield cls.url_result(m[-1], cls, m[-1])
3262
3263 @classmethod
3264 def extract_id(cls, url):
3265 video_id = cls.get_temp_id(url)
3266 if not video_id:
3267 raise ExtractorError(f'Invalid URL: {url}')
3268 return video_id
3269
3270 def _extract_chapters_from_json(self, data, duration):
3271 chapter_list = traverse_obj(
3272 data, (
3273 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3274 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3275 ), expected_type=list)
3276
3277 return self._extract_chapters_helper(
3278 chapter_list,
3279 start_function=lambda chapter: float_or_none(
3280 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3281 title_function=lambda chapter: traverse_obj(
3282 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3283 duration=duration)
3284
3285 def _extract_chapters_from_engagement_panel(self, data, duration):
3286 content_list = traverse_obj(
3287 data,
3288 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3289 expected_type=list)
3290 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3291 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3292
3293 return next(filter(None, (
3294 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3295 chapter_time, chapter_title, duration)
3296 for contents in content_list)), [])
3297
3298 def _extract_heatmap(self, data):
3299 return traverse_obj(data, (
3300 'frameworkUpdates', 'entityBatchUpdate', 'mutations',
3301 lambda _, v: v['payload']['macroMarkersListEntity']['markersList']['markerType'] == 'MARKER_TYPE_HEATMAP',
3302 'payload', 'macroMarkersListEntity', 'markersList', 'markers', ..., {
3303 'start_time': ('startMillis', {functools.partial(float_or_none, scale=1000)}),
3304 'end_time': {lambda x: (int(x['startMillis']) + int(x['durationMillis'])) / 1000},
3305 'value': ('intensityScoreNormalized', {float_or_none}),
3306 })) or None
3307
3308 def _extract_comment(self, comment_renderer, parent=None):
3309 comment_id = comment_renderer.get('commentId')
3310 if not comment_id:
3311 return
3312
3313 info = {
3314 'id': comment_id,
3315 'text': self._get_text(comment_renderer, 'contentText'),
3316 'like_count': self._get_count(comment_renderer, 'voteCount'),
3317 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3318 'author': self._get_text(comment_renderer, 'authorText'),
3319 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3320 'parent': parent or 'root',
3321 }
3322
3323 # Timestamp is an estimate calculated from the current time and time_text
3324 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3325 timestamp = self._parse_time_text(time_text)
3326
3327 info.update({
3328 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3329 '_time_text': time_text,
3330 'timestamp': timestamp,
3331 })
3332
3333 info['author_url'] = urljoin(
3334 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3335 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3336 expected_type=str, get_all=False))
3337
3338 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3339 if author_is_uploader is not None:
3340 info['author_is_uploader'] = author_is_uploader
3341
3342 comment_abr = traverse_obj(
3343 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3344 if comment_abr is not None:
3345 info['is_favorited'] = 'creatorHeart' in comment_abr
3346
3347 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3348 if self._has_badge(badges, BadgeType.VERIFIED):
3349 info['author_is_verified'] = True
3350
3351 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3352 if is_pinned:
3353 info['is_pinned'] = True
3354
3355 return info
3356
3357 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3358
3359 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3360
3361 def extract_header(contents):
3362 _continuation = None
3363 for content in contents:
3364 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3365 expected_comment_count = self._get_count(
3366 comments_header_renderer, 'countText', 'commentsCount')
3367
3368 if expected_comment_count is not None:
3369 tracker['est_total'] = expected_comment_count
3370 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3371 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3372
3373 sort_menu_item = try_get(
3374 comments_header_renderer,
3375 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3376 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3377
3378 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3379 if not _continuation:
3380 continue
3381
3382 sort_text = str_or_none(sort_menu_item.get('title'))
3383 if not sort_text:
3384 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3385 self.to_screen('Sorting comments by %s' % sort_text.lower())
3386 break
3387 return _continuation
3388
3389 def extract_thread(contents):
3390 if not parent:
3391 tracker['current_page_thread'] = 0
3392 for content in contents:
3393 if not parent and tracker['total_parent_comments'] >= max_parents:
3394 yield
3395 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3396 comment_renderer = get_first(
3397 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3398 expected_type=dict, default={})
3399
3400 comment = self._extract_comment(comment_renderer, parent)
3401 if not comment:
3402 continue
3403 comment_id = comment['id']
3404 if comment.get('is_pinned'):
3405 tracker['pinned_comment_ids'].add(comment_id)
3406 # Sometimes YouTube may break and give us infinite looping comments.
3407 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3408 if comment_id in tracker['seen_comment_ids']:
3409 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3410 # Pinned comments may appear a second time in newest first sort
3411 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3412 continue
3413 self.report_warning(
3414 'Detected YouTube comments looping. Stopping comment extraction '
3415 f'{"for this thread" if parent else ""} as we probably cannot get any more.')
3416 yield
3417 else:
3418 tracker['seen_comment_ids'].add(comment['id'])
3419
3420 tracker['running_total'] += 1
3421 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3422 yield comment
3423
3424 # Attempt to get the replies
3425 comment_replies_renderer = try_get(
3426 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3427
3428 if comment_replies_renderer:
3429 tracker['current_page_thread'] += 1
3430 comment_entries_iter = self._comment_entries(
3431 comment_replies_renderer, ytcfg, video_id,
3432 parent=comment.get('id'), tracker=tracker)
3433 yield from itertools.islice(comment_entries_iter, min(
3434 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3435
3436 # Keeps track of counts across recursive calls
3437 if not tracker:
3438 tracker = dict(
3439 running_total=0,
3440 est_total=None,
3441 current_page_thread=0,
3442 total_parent_comments=0,
3443 total_reply_comments=0,
3444 seen_comment_ids=set(),
3445 pinned_comment_ids=set()
3446 )
3447
3448 # TODO: Deprecated
3449 # YouTube comments have a max depth of 2
3450 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3451 if max_depth:
3452 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3453 'Set max replies in the max-comments extractor argument instead')
3454 if max_depth == 1 and parent:
3455 return
3456
3457 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3458 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3459
3460 continuation = self._extract_continuation(root_continuation_data)
3461
3462 response = None
3463 is_forced_continuation = False
3464 is_first_continuation = parent is None
3465 if is_first_continuation and not continuation:
3466 # Sometimes you can get comments by generating the continuation yourself,
3467 # even if YouTube initially reports them being disabled - e.g. stories comments.
3468 # Note: if the comment section is actually disabled, YouTube may return a response with
3469 # required check_get_keys missing. So we will disable that check initially in this case.
3470 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3471 is_forced_continuation = True
3472
3473 continuation_items_path = (
3474 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3475 for page_num in itertools.count(0):
3476 if not continuation:
3477 break
3478 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3479 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3480 if page_num == 0:
3481 if is_first_continuation:
3482 note_prefix = 'Downloading comment section API JSON'
3483 else:
3484 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3485 tracker['current_page_thread'], comment_prog_str)
3486 else:
3487 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3488 ' ' if parent else '', ' replies' if parent else '',
3489 page_num, comment_prog_str)
3490
3491 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3492 # Ignore check if YouTube says the comment count is 0.
3493 check_get_keys = None
3494 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3495 check_get_keys = [[*continuation_items_path, ..., (
3496 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
3497 try:
3498 response = self._extract_response(
3499 item_id=None, query=continuation,
3500 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3501 check_get_keys=check_get_keys)
3502 except ExtractorError as e:
3503 # Ignore incomplete data error for replies if retries didn't work.
3504 # This is to allow any other parent comments and comment threads to be downloaded.
3505 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3506 if 'incomplete data' in str(e).lower() and parent:
3507 if self.get_param('ignoreerrors') in (True, 'only_download'):
3508 self.report_warning(
3509 'Received incomplete data for a comment reply thread and retrying did not help. '
3510 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
3511 return
3512 else:
3513 raise ExtractorError(
3514 'Incomplete data received for comment reply thread. '
3515 'Pass --ignore-errors to ignore and allow rest of comments to download.',
3516 expected=True)
3517 raise
3518 is_forced_continuation = False
3519 continuation = None
3520 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3521 if is_first_continuation:
3522 continuation = extract_header(continuation_items)
3523 is_first_continuation = False
3524 if continuation:
3525 break
3526 continue
3527
3528 for entry in extract_thread(continuation_items):
3529 if not entry:
3530 return
3531 yield entry
3532 continuation = self._extract_continuation({'contents': continuation_items})
3533 if continuation:
3534 break
3535
3536 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3537 if message and not parent and tracker['running_total'] == 0:
3538 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3539 raise self.CommentsDisabled
3540
3541 @staticmethod
3542 def _generate_comment_continuation(video_id):
3543 """
3544 Generates initial comment section continuation token from given video id
3545 """
3546 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3547 return base64.b64encode(token.encode()).decode()
3548
3549 def _get_comments(self, ytcfg, video_id, contents, webpage):
3550 """Entry for comment extraction"""
3551 def _real_comment_extract(contents):
3552 renderer = next((
3553 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3554 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3555 yield from self._comment_entries(renderer, ytcfg, video_id)
3556
3557 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3558 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3559
3560 @staticmethod
3561 def _get_checkok_params():
3562 return {'contentCheckOk': True, 'racyCheckOk': True}
3563
3564 @classmethod
3565 def _generate_player_context(cls, sts=None):
3566 context = {
3567 'html5Preference': 'HTML5_PREF_WANTS',
3568 }
3569 if sts is not None:
3570 context['signatureTimestamp'] = sts
3571 return {
3572 'playbackContext': {
3573 'contentPlaybackContext': context
3574 },
3575 **cls._get_checkok_params()
3576 }
3577
3578 @staticmethod
3579 def _is_agegated(player_response):
3580 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3581 return True
3582
3583 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3584 AGE_GATE_REASONS = (
3585 'confirm your age', 'age-restricted', 'inappropriate', # reason
3586 'age_verification_required', 'age_check_required', # status
3587 )
3588 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3589
3590 @staticmethod
3591 def _is_unplayable(player_response):
3592 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3593
3594 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3595
3596 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3597 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3598 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3599 headers = self.generate_api_headers(
3600 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3601
3602 yt_query = {
3603 'videoId': video_id,
3604 }
3605 if _split_innertube_client(client)[0] == 'android':
3606 yt_query['params'] = 'CgIQBg=='
3607
3608 pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0]
3609 if pp_arg:
3610 yt_query['params'] = pp_arg
3611
3612 yt_query.update(self._generate_player_context(sts))
3613 return self._extract_response(
3614 item_id=video_id, ep='player', query=yt_query,
3615 ytcfg=player_ytcfg, headers=headers, fatal=True,
3616 default_client=client,
3617 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3618 ) or None
3619
3620 def _get_requested_clients(self, url, smuggled_data):
3621 requested_clients = []
3622 default = ['ios', 'android', 'web']
3623 allowed_clients = sorted(
3624 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3625 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3626 for client in self._configuration_arg('player_client'):
3627 if client in allowed_clients:
3628 requested_clients.append(client)
3629 elif client == 'default':
3630 requested_clients.extend(default)
3631 elif client == 'all':
3632 requested_clients.extend(allowed_clients)
3633 else:
3634 self.report_warning(f'Skipping unsupported client {client}')
3635 if not requested_clients:
3636 requested_clients = default
3637
3638 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3639 requested_clients.extend(
3640 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3641
3642 return orderedSet(requested_clients)
3643
3644 def _invalid_player_response(self, pr, video_id):
3645 # YouTube may return a different video player response than expected.
3646 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3647 if (pr_id := traverse_obj(pr, ('videoDetails', 'videoId'))) != video_id:
3648 return pr_id
3649
3650 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3651 initial_pr = None
3652 if webpage:
3653 initial_pr = self._search_json(
3654 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3655
3656 prs = []
3657 if initial_pr and not self._invalid_player_response(initial_pr, video_id):
3658 # Android player_response does not have microFormats which are needed for
3659 # extraction of some data. So we return the initial_pr with formats
3660 # stripped out even if not requested by the user
3661 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3662 prs.append({**initial_pr, 'streamingData': None})
3663
3664 all_clients = set(clients)
3665 clients = clients[::-1]
3666
3667 def append_client(*client_names):
3668 """ Append the first client name that exists but not already used """
3669 for client_name in client_names:
3670 actual_client = _split_innertube_client(client_name)[0]
3671 if actual_client in INNERTUBE_CLIENTS:
3672 if actual_client not in all_clients:
3673 clients.append(client_name)
3674 all_clients.add(actual_client)
3675 return
3676
3677 tried_iframe_fallback = False
3678 player_url = None
3679 skipped_clients = {}
3680 while clients:
3681 client, base_client, variant = _split_innertube_client(clients.pop())
3682 player_ytcfg = master_ytcfg if client == 'web' else {}
3683 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3684 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3685
3686 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3687 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3688 if 'js' in self._configuration_arg('player_skip'):
3689 require_js_player = False
3690 player_url = None
3691
3692 if not player_url and not tried_iframe_fallback and require_js_player:
3693 player_url = self._download_player_url(video_id)
3694 tried_iframe_fallback = True
3695
3696 try:
3697 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3698 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3699 except ExtractorError as e:
3700 self.report_warning(e)
3701 continue
3702
3703 if pr_id := self._invalid_player_response(pr, video_id):
3704 skipped_clients[client] = pr_id
3705 elif pr:
3706 # Save client name for introspection later
3707 name = short_client_name(client)
3708 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3709 sd[STREAMING_DATA_CLIENT_NAME] = name
3710 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3711 f[STREAMING_DATA_CLIENT_NAME] = name
3712 prs.append(pr)
3713
3714 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3715 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3716 append_client(f'{base_client}_creator')
3717 elif self._is_agegated(pr):
3718 if variant == 'tv_embedded':
3719 append_client(f'{base_client}_embedded')
3720 elif not variant:
3721 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3722
3723 if skipped_clients:
3724 self.report_warning(
3725 f'Skipping player responses from {"/".join(skipped_clients)} clients '
3726 f'(got player responses for video "{"/".join(set(skipped_clients.values()))}" instead of "{video_id}")')
3727 if not prs:
3728 raise ExtractorError(
3729 'All player responses are invalid. Your IP is likely being blocked by Youtube', expected=True)
3730 elif not prs:
3731 raise ExtractorError('Failed to extract any player response')
3732 return prs, player_url
3733
3734 def _needs_live_processing(self, live_status, duration):
3735 if (live_status == 'is_live' and self.get_param('live_from_start')
3736 or live_status == 'post_live' and (duration or 0) > 2 * 3600):
3737 return live_status
3738
3739 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3740 CHUNK_SIZE = 10 << 20
3741 itags, stream_ids = collections.defaultdict(set), []
3742 itag_qualities, res_qualities = {}, {0: None}
3743 q = qualities([
3744 # Normally tiny is the smallest video-only formats. But
3745 # audio-only formats with unknown quality may get tagged as tiny
3746 'tiny',
3747 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3748 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3749 ])
3750 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3751 format_types = self._configuration_arg('formats')
3752 all_formats = 'duplicate' in format_types
3753 if self._configuration_arg('include_duplicate_formats'):
3754 all_formats = True
3755 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
3756 'Use formats=duplicate extractor argument instead')
3757
3758 def build_fragments(f):
3759 return LazyList({
3760 'url': update_url_query(f['url'], {
3761 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3762 })
3763 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3764
3765 for fmt in streaming_formats:
3766 if fmt.get('targetDurationSec'):
3767 continue
3768
3769 itag = str_or_none(fmt.get('itag'))
3770 audio_track = fmt.get('audioTrack') or {}
3771 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3772 if not all_formats:
3773 if stream_id in stream_ids:
3774 continue
3775
3776 quality = fmt.get('quality')
3777 height = int_or_none(fmt.get('height'))
3778 if quality == 'tiny' or not quality:
3779 quality = fmt.get('audioQuality', '').lower() or quality
3780 # The 3gp format (17) in android client has a quality of "small",
3781 # but is actually worse than other formats
3782 if itag == '17':
3783 quality = 'tiny'
3784 if quality:
3785 if itag:
3786 itag_qualities[itag] = quality
3787 if height:
3788 res_qualities[height] = quality
3789 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3790 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3791 # number of fragment that would subsequently requested with (`&sq=N`)
3792 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3793 continue
3794
3795 fmt_url = fmt.get('url')
3796 if not fmt_url:
3797 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3798 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3799 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3800 if not all((sc, fmt_url, player_url, encrypted_sig)):
3801 continue
3802 try:
3803 fmt_url += '&%s=%s' % (
3804 traverse_obj(sc, ('sp', -1)) or 'signature',
3805 self._decrypt_signature(encrypted_sig, video_id, player_url)
3806 )
3807 except ExtractorError as e:
3808 self.report_warning('Signature extraction failed: Some formats may be missing',
3809 video_id=video_id, only_once=True)
3810 self.write_debug(e, only_once=True)
3811 continue
3812
3813 query = parse_qs(fmt_url)
3814 throttled = False
3815 if query.get('n'):
3816 try:
3817 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3818 fmt_url = update_url_query(fmt_url, {
3819 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3820 })
3821 except ExtractorError as e:
3822 phantomjs_hint = ''
3823 if isinstance(e, JSInterpreter.Exception):
3824 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3825 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3826 if player_url:
3827 self.report_warning(
3828 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3829 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3830 self.write_debug(e, only_once=True)
3831 else:
3832 self.report_warning(
3833 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3834 video_id=video_id, only_once=True)
3835 throttled = True
3836
3837 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1024)
3838 language_preference = (
3839 10 if audio_track.get('audioIsDefault') and 10
3840 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3841 else -1)
3842 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3843 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3844 # Make sure to avoid false positives with small duration differences.
3845 # E.g. __2ABJjxzNo, ySuUZEjARPY
3846 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3847 if is_damaged:
3848 self.report_warning(
3849 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3850
3851 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3852 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
3853 fps = int_or_none(fmt.get('fps')) or 0
3854 dct = {
3855 'asr': int_or_none(fmt.get('audioSampleRate')),
3856 'filesize': int_or_none(fmt.get('contentLength')),
3857 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3858 'format_note': join_nonempty(
3859 join_nonempty(audio_track.get('displayName'),
3860 language_preference > 0 and ' (default)', delim=''),
3861 name, fmt.get('isDrc') and 'DRC',
3862 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3863 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3864 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3865 (self.get_param('verbose') or all_formats) and client_name,
3866 delim=', '),
3867 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3868 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
3869 + (100 if 'Premium' in name else 0)),
3870 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
3871 'audio_channels': fmt.get('audioChannels'),
3872 'height': height,
3873 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3874 'has_drm': bool(fmt.get('drmFamilies')),
3875 'tbr': tbr,
3876 'url': fmt_url,
3877 'width': int_or_none(fmt.get('width')),
3878 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3879 'desc' if language_preference < -1 else '') or None,
3880 'language_preference': language_preference,
3881 # Strictly de-prioritize damaged and 3gp formats
3882 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3883 }
3884 mime_mobj = re.match(
3885 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3886 if mime_mobj:
3887 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3888 dct.update(parse_codecs(mime_mobj.group(2)))
3889 if itag:
3890 itags[itag].add(('https', dct.get('language')))
3891 stream_ids.append(stream_id)
3892 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3893 if single_stream and dct.get('ext'):
3894 dct['container'] = dct['ext'] + '_dash'
3895
3896 if (all_formats or 'dashy' in format_types) and dct['filesize']:
3897 yield {
3898 **dct,
3899 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3900 'protocol': 'http_dash_segments',
3901 'fragments': build_fragments(dct),
3902 }
3903 if all_formats or 'dashy' not in format_types:
3904 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3905 yield dct
3906
3907 needs_live_processing = self._needs_live_processing(live_status, duration)
3908 skip_bad_formats = 'incomplete' not in format_types
3909 if self._configuration_arg('include_incomplete_formats'):
3910 skip_bad_formats = False
3911 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
3912 'Use formats=incomplete extractor argument instead')
3913
3914 skip_manifests = set(self._configuration_arg('skip'))
3915 if (not self.get_param('youtube_include_hls_manifest', True)
3916 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3917 or needs_live_processing and skip_bad_formats):
3918 skip_manifests.add('hls')
3919
3920 if not self.get_param('youtube_include_dash_manifest', True):
3921 skip_manifests.add('dash')
3922 if self._configuration_arg('include_live_dash'):
3923 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3924 'Use formats=incomplete extractor argument instead')
3925 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3926 skip_manifests.add('dash')
3927
3928 def process_manifest_format(f, proto, client_name, itag):
3929 key = (proto, f.get('language'))
3930 if not all_formats and key in itags[itag]:
3931 return False
3932 itags[itag].add(key)
3933
3934 if itag and all_formats:
3935 f['format_id'] = f'{itag}-{proto}'
3936 elif any(p != proto for p, _ in itags[itag]):
3937 f['format_id'] = f'{itag}-{proto}'
3938 elif itag:
3939 f['format_id'] = itag
3940
3941 if f.get('source_preference') is None:
3942 f['source_preference'] = -1
3943
3944 if itag in ('616', '235'):
3945 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
3946 f['source_preference'] += 100
3947
3948 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3949 if f['quality'] == -1 and f.get('height'):
3950 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3951 if self.get_param('verbose') or all_formats:
3952 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3953 if f.get('fps') and f['fps'] <= 1:
3954 del f['fps']
3955
3956 if proto == 'hls' and f.get('has_drm'):
3957 f['has_drm'] = 'maybe'
3958 f['source_preference'] -= 5
3959 return True
3960
3961 subtitles = {}
3962 for sd in streaming_data:
3963 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3964
3965 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3966 if hls_manifest_url:
3967 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3968 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3969 subtitles = self._merge_subtitles(subs, subtitles)
3970 for f in fmts:
3971 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3972 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3973 yield f
3974
3975 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3976 if dash_manifest_url:
3977 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3978 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3979 for f in formats:
3980 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3981 f['filesize'] = int_or_none(self._search_regex(
3982 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3983 if needs_live_processing:
3984 f['is_from_start'] = True
3985
3986 yield f
3987 yield subtitles
3988
3989 def _extract_storyboard(self, player_responses, duration):
3990 spec = get_first(
3991 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3992 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3993 if not base_url:
3994 return
3995 L = len(spec) - 1
3996 for i, args in enumerate(spec):
3997 args = args.split('#')
3998 counts = list(map(int_or_none, args[:5]))
3999 if len(args) != 8 or not all(counts):
4000 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
4001 continue
4002 width, height, frame_count, cols, rows = counts
4003 N, sigh = args[6:]
4004
4005 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
4006 fragment_count = frame_count / (cols * rows)
4007 fragment_duration = duration / fragment_count
4008 yield {
4009 'format_id': f'sb{i}',
4010 'format_note': 'storyboard',
4011 'ext': 'mhtml',
4012 'protocol': 'mhtml',
4013 'acodec': 'none',
4014 'vcodec': 'none',
4015 'url': url,
4016 'width': width,
4017 'height': height,
4018 'fps': frame_count / duration,
4019 'rows': rows,
4020 'columns': cols,
4021 'fragments': [{
4022 'url': url.replace('$M', str(j)),
4023 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4024 } for j in range(math.ceil(fragment_count))],
4025 }
4026
4027 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4028 webpage = None
4029 if 'webpage' not in self._configuration_arg('player_skip'):
4030 query = {'bpctr': '9999999999', 'has_verified': '1'}
4031 pp = self._configuration_arg('player_params', [None], casesense=True)[0]
4032 if pp:
4033 query['pp'] = pp
4034 webpage = self._download_webpage(
4035 webpage_url, video_id, fatal=False, query=query)
4036
4037 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4038
4039 player_responses, player_url = self._extract_player_responses(
4040 self._get_requested_clients(url, smuggled_data),
4041 video_id, webpage, master_ytcfg, smuggled_data)
4042
4043 return webpage, master_ytcfg, player_responses, player_url
4044
4045 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4046 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4047 is_live = get_first(video_details, 'isLive')
4048 if is_live is None:
4049 is_live = get_first(live_broadcast_details, 'isLiveNow')
4050 live_content = get_first(video_details, 'isLiveContent')
4051 is_upcoming = get_first(video_details, 'isUpcoming')
4052 post_live = get_first(video_details, 'isPostLiveDvr')
4053 live_status = ('post_live' if post_live
4054 else 'is_live' if is_live
4055 else 'is_upcoming' if is_upcoming
4056 else 'was_live' if live_content
4057 else 'not_live' if False in (is_live, live_content)
4058 else None)
4059 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4060 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4061 if all(f.get('has_drm') for f in formats):
4062 # If there are no formats that definitely don't have DRM, all have DRM
4063 for f in formats:
4064 f['has_drm'] = True
4065
4066 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4067
4068 def _real_extract(self, url):
4069 url, smuggled_data = unsmuggle_url(url, {})
4070 video_id = self._match_id(url)
4071
4072 base_url = self.http_scheme() + '//www.youtube.com/'
4073 webpage_url = base_url + 'watch?v=' + video_id
4074
4075 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4076
4077 playability_statuses = traverse_obj(
4078 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4079
4080 trailer_video_id = get_first(
4081 playability_statuses,
4082 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4083 expected_type=str)
4084 if trailer_video_id:
4085 return self.url_result(
4086 trailer_video_id, self.ie_key(), trailer_video_id)
4087
4088 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4089 if webpage else (lambda x: None))
4090
4091 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4092 microformats = traverse_obj(
4093 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4094 expected_type=dict)
4095
4096 translated_title = self._get_text(microformats, (..., 'title'))
4097 video_title = (self._preferred_lang and translated_title
4098 or get_first(video_details, 'title') # primary
4099 or translated_title
4100 or search_meta(['og:title', 'twitter:title', 'title']))
4101 translated_description = self._get_text(microformats, (..., 'description'))
4102 original_description = get_first(video_details, 'shortDescription')
4103 video_description = (
4104 self._preferred_lang and translated_description
4105 # If original description is blank, it will be an empty string.
4106 # Do not prefer translated description in this case.
4107 or original_description if original_description is not None else translated_description)
4108
4109 multifeed_metadata_list = get_first(
4110 player_responses,
4111 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4112 expected_type=str)
4113 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4114 if self.get_param('noplaylist'):
4115 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4116 else:
4117 entries = []
4118 feed_ids = []
4119 for feed in multifeed_metadata_list.split(','):
4120 # Unquote should take place before split on comma (,) since textual
4121 # fields may contain comma as well (see
4122 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4123 feed_data = urllib.parse.parse_qs(
4124 urllib.parse.unquote_plus(feed))
4125
4126 def feed_entry(name):
4127 return try_get(
4128 feed_data, lambda x: x[name][0], str)
4129
4130 feed_id = feed_entry('id')
4131 if not feed_id:
4132 continue
4133 feed_title = feed_entry('title')
4134 title = video_title
4135 if feed_title:
4136 title += ' (%s)' % feed_title
4137 entries.append({
4138 '_type': 'url_transparent',
4139 'ie_key': 'Youtube',
4140 'url': smuggle_url(
4141 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4142 {'force_singlefeed': True}),
4143 'title': title,
4144 })
4145 feed_ids.append(feed_id)
4146 self.to_screen(
4147 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4148 % (', '.join(feed_ids), video_id))
4149 return self.playlist_result(
4150 entries, video_id, video_title, video_description)
4151
4152 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4153 or int_or_none(get_first(microformats, 'lengthSeconds'))
4154 or parse_duration(search_meta('duration')) or None)
4155
4156 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4157 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4158 if live_status == 'post_live':
4159 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4160
4161 if not formats:
4162 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4163 self.report_drm(video_id)
4164 pemr = get_first(
4165 playability_statuses,
4166 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4167 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4168 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4169 if subreason:
4170 if subreason == 'The uploader has not made this video available in your country.':
4171 countries = get_first(microformats, 'availableCountries')
4172 if not countries:
4173 regions_allowed = search_meta('regionsAllowed')
4174 countries = regions_allowed.split(',') if regions_allowed else None
4175 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4176 reason += f'. {subreason}'
4177 if reason:
4178 self.raise_no_formats(reason, expected=True)
4179
4180 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4181 if not keywords and webpage:
4182 keywords = [
4183 unescapeHTML(m.group('content'))
4184 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4185 for keyword in keywords:
4186 if keyword.startswith('yt:stretch='):
4187 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4188 if mobj:
4189 # NB: float is intentional for forcing float division
4190 w, h = (float(v) for v in mobj.groups())
4191 if w > 0 and h > 0:
4192 ratio = w / h
4193 for f in formats:
4194 if f.get('vcodec') != 'none':
4195 f['stretched_ratio'] = ratio
4196 break
4197 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4198 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4199 if thumbnail_url:
4200 thumbnails.append({
4201 'url': thumbnail_url,
4202 })
4203 original_thumbnails = thumbnails.copy()
4204
4205 # The best resolution thumbnails sometimes does not appear in the webpage
4206 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4207 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4208 thumbnail_names = [
4209 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4210 # in resolution, these are not the custom thumbnail. So de-prioritize them
4211 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4212 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4213 ]
4214 n_thumbnail_names = len(thumbnail_names)
4215 thumbnails.extend({
4216 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4217 video_id=video_id, name=name, ext=ext,
4218 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4219 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4220 for thumb in thumbnails:
4221 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4222 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4223 self._remove_duplicate_formats(thumbnails)
4224 self._downloader._sort_thumbnails(original_thumbnails)
4225
4226 category = get_first(microformats, 'category') or search_meta('genre')
4227 channel_id = self.ucid_or_none(str_or_none(
4228 get_first(video_details, 'channelId')
4229 or get_first(microformats, 'externalChannelId')
4230 or search_meta('channelId')))
4231 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4232
4233 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4234 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4235 if not duration and live_end_time and live_start_time:
4236 duration = live_end_time - live_start_time
4237
4238 needs_live_processing = self._needs_live_processing(live_status, duration)
4239
4240 def is_bad_format(fmt):
4241 if needs_live_processing and not fmt.get('is_from_start'):
4242 return True
4243 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4244 and fmt.get('protocol') == 'http_dash_segments'):
4245 return True
4246
4247 for fmt in filter(is_bad_format, formats):
4248 fmt['preference'] = (fmt.get('preference') or -1) - 10
4249 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4250
4251 if needs_live_processing:
4252 self._prepare_live_from_start_formats(
4253 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4254
4255 formats.extend(self._extract_storyboard(player_responses, duration))
4256
4257 channel_handle = self.handle_from_url(owner_profile_url)
4258
4259 info = {
4260 'id': video_id,
4261 'title': video_title,
4262 'formats': formats,
4263 'thumbnails': thumbnails,
4264 # The best thumbnail that we are sure exists. Prevents unnecessary
4265 # URL checking if user don't care about getting the best possible thumbnail
4266 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4267 'description': video_description,
4268 'channel_id': channel_id,
4269 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4270 'duration': duration,
4271 'view_count': int_or_none(
4272 get_first((video_details, microformats), (..., 'viewCount'))
4273 or search_meta('interactionCount')),
4274 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4275 'age_limit': 18 if (
4276 get_first(microformats, 'isFamilySafe') is False
4277 or search_meta('isFamilyFriendly') == 'false'
4278 or search_meta('og:restrictions:age') == '18+') else 0,
4279 'webpage_url': webpage_url,
4280 'categories': [category] if category else None,
4281 'tags': keywords,
4282 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4283 'live_status': live_status,
4284 'release_timestamp': live_start_time,
4285 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4286 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4287 }
4288
4289 subtitles = {}
4290 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4291 if pctr:
4292 def get_lang_code(track):
4293 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4294 or track.get('languageCode'))
4295
4296 # Converted into dicts to remove duplicates
4297 captions = {
4298 get_lang_code(sub): sub
4299 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4300 translation_languages = {
4301 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4302 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4303
4304 def process_language(container, base_url, lang_code, sub_name, query):
4305 lang_subs = container.setdefault(lang_code, [])
4306 for fmt in self._SUBTITLE_FORMATS:
4307 query.update({
4308 'fmt': fmt,
4309 })
4310 lang_subs.append({
4311 'ext': fmt,
4312 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4313 'name': sub_name,
4314 })
4315
4316 # NB: Constructing the full subtitle dictionary is slow
4317 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4318 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4319 for lang_code, caption_track in captions.items():
4320 base_url = caption_track.get('baseUrl')
4321 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4322 if not base_url:
4323 continue
4324 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4325 if caption_track.get('kind') != 'asr':
4326 if not lang_code:
4327 continue
4328 process_language(
4329 subtitles, base_url, lang_code, lang_name, {})
4330 if not caption_track.get('isTranslatable'):
4331 continue
4332 for trans_code, trans_name in translation_languages.items():
4333 if not trans_code:
4334 continue
4335 orig_trans_code = trans_code
4336 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4337 if not get_translated_subs:
4338 continue
4339 trans_code += f'-{lang_code}'
4340 trans_name += format_field(lang_name, None, ' from %s')
4341 if lang_code == f'a-{orig_trans_code}':
4342 # Set audio language based on original subtitles
4343 for f in formats:
4344 if f.get('acodec') != 'none' and not f.get('language'):
4345 f['language'] = orig_trans_code
4346 # Add an "-orig" label to the original language so that it can be distinguished.
4347 # The subs are returned without "-orig" as well for compatibility
4348 process_language(
4349 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4350 # Setting tlang=lang returns damaged subtitles.
4351 process_language(automatic_captions, base_url, trans_code, trans_name,
4352 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4353
4354 info['automatic_captions'] = automatic_captions
4355 info['subtitles'] = subtitles
4356
4357 parsed_url = urllib.parse.urlparse(url)
4358 for component in [parsed_url.fragment, parsed_url.query]:
4359 query = urllib.parse.parse_qs(component)
4360 for k, v in query.items():
4361 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4362 d_k += '_time'
4363 if d_k not in info and k in s_ks:
4364 info[d_k] = parse_duration(query[k][0])
4365
4366 # Youtube Music Auto-generated description
4367 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4368 # XXX: Causes catastrophic backtracking if description has "·"
4369 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4370 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4371 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4372 mobj = re.search(
4373 r'''(?xs)
4374 (?=(?P<track>[^\n·]+))(?P=track)·
4375 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4376 (?=(?P<album>[^\n]+))(?P=album)\n
4377 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4378 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4379 (.+?\nArtist\s*:\s*
4380 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4381 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4382 ''', video_description)
4383 if mobj:
4384 release_year = mobj.group('release_year')
4385 release_date = mobj.group('release_date')
4386 if release_date:
4387 release_date = release_date.replace('-', '')
4388 if not release_year:
4389 release_year = release_date[:4]
4390 info.update({
4391 'album': mobj.group('album'.strip()),
4392 'artists': ([a] if (a := mobj.group('clean_artist'))
4393 else [a.strip() for a in mobj.group('artist').split('·')]),
4394 'track': mobj.group('track').strip(),
4395 'release_date': release_date,
4396 'release_year': int_or_none(release_year),
4397 })
4398
4399 initial_data = None
4400 if webpage:
4401 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4402 if not traverse_obj(initial_data, 'contents'):
4403 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4404 initial_data = None
4405 if not initial_data:
4406 query = {'videoId': video_id}
4407 query.update(self._get_checkok_params())
4408 initial_data = self._extract_response(
4409 item_id=video_id, ep='next', fatal=False,
4410 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4411 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4412 note='Downloading initial data API JSON')
4413
4414 info['comment_count'] = traverse_obj(initial_data, (
4415 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4416 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4417 ), (
4418 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4419 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4420 ), expected_type=self._get_count, get_all=False)
4421
4422 try: # This will error if there is no livechat
4423 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4424 except (KeyError, IndexError, TypeError):
4425 pass
4426 else:
4427 info.setdefault('subtitles', {})['live_chat'] = [{
4428 # url is needed to set cookies
4429 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4430 'video_id': video_id,
4431 'ext': 'json',
4432 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4433 else 'youtube_live_chat_replay'),
4434 }]
4435
4436 if initial_data:
4437 info['chapters'] = (
4438 self._extract_chapters_from_json(initial_data, duration)
4439 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4440 or self._extract_chapters_from_description(video_description, duration)
4441 or None)
4442
4443 info['heatmap'] = self._extract_heatmap(initial_data)
4444
4445 contents = traverse_obj(
4446 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4447 expected_type=list, default=[])
4448
4449 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4450 if vpir:
4451 stl = vpir.get('superTitleLink')
4452 if stl:
4453 stl = self._get_text(stl)
4454 if try_get(
4455 vpir,
4456 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4457 info['location'] = stl
4458 else:
4459 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4460 if mobj:
4461 info.update({
4462 'series': mobj.group(1),
4463 'season_number': int(mobj.group(2)),
4464 'episode_number': int(mobj.group(3)),
4465 })
4466 for tlb in (try_get(
4467 vpir,
4468 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4469 list) or []):
4470 tbrs = variadic(
4471 traverse_obj(
4472 tlb, ('toggleButtonRenderer', ...),
4473 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4474 for tbr in tbrs:
4475 for getter, regex in [(
4476 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4477 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4478 lambda x: x['accessibility'],
4479 lambda x: x['accessibilityData']['accessibilityData'],
4480 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4481 label = (try_get(tbr, getter, dict) or {}).get('label')
4482 if label:
4483 mobj = re.match(regex, label)
4484 if mobj:
4485 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4486 break
4487
4488 info['like_count'] = traverse_obj(vpir, (
4489 'videoActions', 'menuRenderer', 'topLevelButtons', ...,
4490 'segmentedLikeDislikeButtonViewModel', 'likeButtonViewModel', 'likeButtonViewModel',
4491 'toggleButtonViewModel', 'toggleButtonViewModel', 'defaultButtonViewModel',
4492 'buttonViewModel', 'accessibilityText', {parse_count}), get_all=False)
4493
4494 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4495 if vcr:
4496 vc = self._get_count(vcr, 'viewCount')
4497 # Upcoming premieres with waiting count are treated as live here
4498 if vcr.get('isLive'):
4499 info['concurrent_view_count'] = vc
4500 elif info.get('view_count') is None:
4501 info['view_count'] = vc
4502
4503 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4504 if vsir:
4505 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4506 info.update({
4507 'channel': self._get_text(vor, 'title'),
4508 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4509
4510 if not channel_handle:
4511 channel_handle = self.handle_from_url(
4512 traverse_obj(vor, (
4513 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4514 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4515 {str}), get_all=False))
4516
4517 rows = try_get(
4518 vsir,
4519 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4520 list) or []
4521 multiple_songs = False
4522 for row in rows:
4523 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4524 multiple_songs = True
4525 break
4526 for row in rows:
4527 mrr = row.get('metadataRowRenderer') or {}
4528 mrr_title = mrr.get('title')
4529 if not mrr_title:
4530 continue
4531 mrr_title = self._get_text(mrr, 'title')
4532 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4533 if mrr_title == 'License':
4534 info['license'] = mrr_contents_text
4535 elif not multiple_songs:
4536 if mrr_title == 'Album':
4537 info['album'] = mrr_contents_text
4538 elif mrr_title == 'Artist':
4539 info['artists'] = [mrr_contents_text] if mrr_contents_text else None
4540 elif mrr_title == 'Song':
4541 info['track'] = mrr_contents_text
4542 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4543 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4544 info['channel_is_verified'] = True
4545
4546 info.update({
4547 'uploader': info.get('channel'),
4548 'uploader_id': channel_handle,
4549 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4550 })
4551 # The upload date for scheduled, live and past live streams / premieres in microformats
4552 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4553 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4554 upload_date = (
4555 unified_strdate(get_first(microformats, 'uploadDate'))
4556 or unified_strdate(search_meta('uploadDate')))
4557 if not upload_date or (
4558 live_status in ('not_live', None)
4559 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4560 ):
4561 upload_date = strftime_or_none(
4562 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4563 info['upload_date'] = upload_date
4564
4565 if upload_date and live_status not in ('is_live', 'post_live', 'is_upcoming'):
4566 # Newly uploaded videos' HLS formats are potentially problematic and need to be checked
4567 upload_datetime = datetime_from_str(upload_date).replace(tzinfo=datetime.timezone.utc)
4568 if upload_datetime >= datetime_from_str('today-2days'):
4569 for fmt in info['formats']:
4570 if fmt.get('protocol') == 'm3u8_native':
4571 fmt['__needs_testing'] = True
4572
4573 for s_k, d_k in [('artists', 'creators'), ('track', 'alt_title')]:
4574 v = info.get(s_k)
4575 if v:
4576 info[d_k] = v
4577
4578 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4579
4580 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4581 or get_first(video_details, 'isPrivate', expected_type=bool))
4582
4583 info['availability'] = (
4584 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4585 else self._availability(
4586 is_private=is_private,
4587 needs_premium=(
4588 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4589 or False if initial_data and is_private is not None else None),
4590 needs_subscription=(
4591 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4592 or False if initial_data and is_private is not None else None),
4593 needs_auth=info['age_limit'] >= 18,
4594 is_unlisted=None if is_private is None else (
4595 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4596 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4597
4598 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4599
4600 self.mark_watched(video_id, player_responses)
4601
4602 return info
4603
4604
4605 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4606 @staticmethod
4607 def passthrough_smuggled_data(func):
4608 def _smuggle(info, smuggled_data):
4609 if info.get('_type') not in ('url', 'url_transparent'):
4610 return info
4611 if smuggled_data.get('is_music_url'):
4612 parsed_url = urllib.parse.urlparse(info['url'])
4613 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4614 smuggled_data.pop('is_music_url')
4615 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4616 if smuggled_data:
4617 info['url'] = smuggle_url(info['url'], smuggled_data)
4618 return info
4619
4620 @functools.wraps(func)
4621 def wrapper(self, url):
4622 url, smuggled_data = unsmuggle_url(url, {})
4623 if self.is_music_url(url):
4624 smuggled_data['is_music_url'] = True
4625 info_dict = func(self, url, smuggled_data)
4626 if smuggled_data:
4627 _smuggle(info_dict, smuggled_data)
4628 if info_dict.get('entries'):
4629 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4630 return info_dict
4631 return wrapper
4632
4633 @staticmethod
4634 def _extract_basic_item_renderer(item):
4635 # Modified from _extract_grid_item_renderer
4636 known_basic_renderers = (
4637 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4638 )
4639 for key, renderer in item.items():
4640 if not isinstance(renderer, dict):
4641 continue
4642 elif key in known_basic_renderers:
4643 return renderer
4644 elif key.startswith('grid') and key.endswith('Renderer'):
4645 return renderer
4646
4647 def _extract_channel_renderer(self, renderer):
4648 channel_id = self.ucid_or_none(renderer['channelId'])
4649 title = self._get_text(renderer, 'title')
4650 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4651 channel_handle = self.handle_from_url(
4652 traverse_obj(renderer, (
4653 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4654 ('browseEndpoint', 'canonicalBaseUrl')),
4655 {str}), get_all=False))
4656 if not channel_handle:
4657 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
4658 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
4659 return {
4660 '_type': 'url',
4661 'url': channel_url,
4662 'id': channel_id,
4663 'ie_key': YoutubeTabIE.ie_key(),
4664 'channel': title,
4665 'uploader': title,
4666 'channel_id': channel_id,
4667 'channel_url': channel_url,
4668 'title': title,
4669 'uploader_id': channel_handle,
4670 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4671 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
4672 # However, in feed/channels this is set correctly to the subscriber count
4673 'channel_follower_count': traverse_obj(
4674 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
4675 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4676 'playlist_count': (
4677 # videoCountText may be the subscriber count
4678 self._get_count(renderer, 'videoCountText')
4679 if self._get_count(renderer, 'subscriberCountText') is not None else None),
4680 'description': self._get_text(renderer, 'descriptionSnippet'),
4681 'channel_is_verified': True if self._has_badge(
4682 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
4683 }
4684
4685 def _grid_entries(self, grid_renderer):
4686 for item in grid_renderer['items']:
4687 if not isinstance(item, dict):
4688 continue
4689 renderer = self._extract_basic_item_renderer(item)
4690 if not isinstance(renderer, dict):
4691 continue
4692 title = self._get_text(renderer, 'title')
4693
4694 # playlist
4695 playlist_id = renderer.get('playlistId')
4696 if playlist_id:
4697 yield self.url_result(
4698 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4699 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4700 video_title=title)
4701 continue
4702 # video
4703 video_id = renderer.get('videoId')
4704 if video_id:
4705 yield self._extract_video(renderer)
4706 continue
4707 # channel
4708 channel_id = renderer.get('channelId')
4709 if channel_id:
4710 yield self._extract_channel_renderer(renderer)
4711 continue
4712 # generic endpoint URL support
4713 ep_url = urljoin('https://www.youtube.com/', try_get(
4714 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4715 str))
4716 if ep_url:
4717 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4718 if ie.suitable(ep_url):
4719 yield self.url_result(
4720 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4721 break
4722
4723 def _music_reponsive_list_entry(self, renderer):
4724 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4725 if video_id:
4726 title = traverse_obj(renderer, (
4727 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4728 'text', 'runs', 0, 'text'))
4729 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4730 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4731 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4732 if playlist_id:
4733 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4734 if video_id:
4735 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4736 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4737 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4738 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4739 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4740 if browse_id:
4741 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4742 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4743
4744 def _shelf_entries_from_content(self, shelf_renderer):
4745 content = shelf_renderer.get('content')
4746 if not isinstance(content, dict):
4747 return
4748 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4749 if renderer:
4750 # TODO: add support for nested playlists so each shelf is processed
4751 # as separate playlist
4752 # TODO: this includes only first N items
4753 yield from self._grid_entries(renderer)
4754 renderer = content.get('horizontalListRenderer')
4755 if renderer:
4756 # TODO
4757 pass
4758
4759 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4760 ep = try_get(
4761 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4762 str)
4763 shelf_url = urljoin('https://www.youtube.com', ep)
4764 if shelf_url:
4765 # Skipping links to another channels, note that checking for
4766 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4767 # will not work
4768 if skip_channels and '/channels?' in shelf_url:
4769 return
4770 title = self._get_text(shelf_renderer, 'title')
4771 yield self.url_result(shelf_url, video_title=title)
4772 # Shelf may not contain shelf URL, fallback to extraction from content
4773 yield from self._shelf_entries_from_content(shelf_renderer)
4774
4775 def _playlist_entries(self, video_list_renderer):
4776 for content in video_list_renderer['contents']:
4777 if not isinstance(content, dict):
4778 continue
4779 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4780 if not isinstance(renderer, dict):
4781 continue
4782 video_id = renderer.get('videoId')
4783 if not video_id:
4784 continue
4785 yield self._extract_video(renderer)
4786
4787 def _rich_entries(self, rich_grid_renderer):
4788 renderer = traverse_obj(
4789 rich_grid_renderer,
4790 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4791 video_id = renderer.get('videoId')
4792 if video_id:
4793 yield self._extract_video(renderer)
4794 return
4795 playlist_id = renderer.get('playlistId')
4796 if playlist_id:
4797 yield self.url_result(
4798 f'https://www.youtube.com/playlist?list={playlist_id}',
4799 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4800 video_title=self._get_text(renderer, 'title'))
4801 return
4802
4803 def _video_entry(self, video_renderer):
4804 video_id = video_renderer.get('videoId')
4805 if video_id:
4806 return self._extract_video(video_renderer)
4807
4808 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4809 url = urljoin('https://youtube.com', traverse_obj(
4810 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4811 if url:
4812 return self.url_result(
4813 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4814
4815 def _post_thread_entries(self, post_thread_renderer):
4816 post_renderer = try_get(
4817 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4818 if not post_renderer:
4819 return
4820 # video attachment
4821 video_renderer = try_get(
4822 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4823 video_id = video_renderer.get('videoId')
4824 if video_id:
4825 entry = self._extract_video(video_renderer)
4826 if entry:
4827 yield entry
4828 # playlist attachment
4829 playlist_id = try_get(
4830 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4831 if playlist_id:
4832 yield self.url_result(
4833 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4834 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4835 # inline video links
4836 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4837 for run in runs:
4838 if not isinstance(run, dict):
4839 continue
4840 ep_url = try_get(
4841 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4842 if not ep_url:
4843 continue
4844 if not YoutubeIE.suitable(ep_url):
4845 continue
4846 ep_video_id = YoutubeIE._match_id(ep_url)
4847 if video_id == ep_video_id:
4848 continue
4849 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4850
4851 def _post_thread_continuation_entries(self, post_thread_continuation):
4852 contents = post_thread_continuation.get('contents')
4853 if not isinstance(contents, list):
4854 return
4855 for content in contents:
4856 renderer = content.get('backstagePostThreadRenderer')
4857 if isinstance(renderer, dict):
4858 yield from self._post_thread_entries(renderer)
4859 continue
4860 renderer = content.get('videoRenderer')
4861 if isinstance(renderer, dict):
4862 yield self._video_entry(renderer)
4863
4864 r''' # unused
4865 def _rich_grid_entries(self, contents):
4866 for content in contents:
4867 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4868 if video_renderer:
4869 entry = self._video_entry(video_renderer)
4870 if entry:
4871 yield entry
4872 '''
4873
4874 def _report_history_entries(self, renderer):
4875 for url in traverse_obj(renderer, (
4876 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4877 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4878 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4879 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4880
4881 def _extract_entries(self, parent_renderer, continuation_list):
4882 # continuation_list is modified in-place with continuation_list = [continuation_token]
4883 continuation_list[:] = [None]
4884 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4885 for content in contents:
4886 if not isinstance(content, dict):
4887 continue
4888 is_renderer = traverse_obj(
4889 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4890 expected_type=dict)
4891 if not is_renderer:
4892 if content.get('richItemRenderer'):
4893 for entry in self._rich_entries(content['richItemRenderer']):
4894 yield entry
4895 continuation_list[0] = self._extract_continuation(parent_renderer)
4896 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4897 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4898 yield from self._report_history_entries(table)
4899 continuation_list[0] = self._extract_continuation(table)
4900 continue
4901
4902 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4903 for isr_content in isr_contents:
4904 if not isinstance(isr_content, dict):
4905 continue
4906
4907 known_renderers = {
4908 'playlistVideoListRenderer': self._playlist_entries,
4909 'gridRenderer': self._grid_entries,
4910 'reelShelfRenderer': self._grid_entries,
4911 'shelfRenderer': self._shelf_entries,
4912 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4913 'backstagePostThreadRenderer': self._post_thread_entries,
4914 'videoRenderer': lambda x: [self._video_entry(x)],
4915 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4916 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4917 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
4918 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
4919 }
4920 for key, renderer in isr_content.items():
4921 if key not in known_renderers:
4922 continue
4923 for entry in known_renderers[key](renderer):
4924 if entry:
4925 yield entry
4926 continuation_list[0] = self._extract_continuation(renderer)
4927 break
4928
4929 if not continuation_list[0]:
4930 continuation_list[0] = self._extract_continuation(is_renderer)
4931
4932 if not continuation_list[0]:
4933 continuation_list[0] = self._extract_continuation(parent_renderer)
4934
4935 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4936 continuation_list = [None]
4937 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4938 tab_content = try_get(tab, lambda x: x['content'], dict)
4939 if not tab_content:
4940 return
4941 parent_renderer = (
4942 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4943 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4944 yield from extract_entries(parent_renderer)
4945 continuation = continuation_list[0]
4946 seen_continuations = set()
4947 for page_num in itertools.count(1):
4948 if not continuation:
4949 break
4950 continuation_token = continuation.get('continuation')
4951 if continuation_token is not None and continuation_token in seen_continuations:
4952 self.write_debug('Detected YouTube feed looping - assuming end of feed.')
4953 break
4954 seen_continuations.add(continuation_token)
4955 headers = self.generate_api_headers(
4956 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4957 response = self._extract_response(
4958 item_id=f'{item_id} page {page_num}',
4959 query=continuation, headers=headers, ytcfg=ytcfg,
4960 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4961
4962 if not response:
4963 break
4964 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4965 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4966 visitor_data = self._extract_visitor_data(response) or visitor_data
4967
4968 known_renderers = {
4969 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4970 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4971 'gridVideoRenderer': (self._grid_entries, 'items'),
4972 'gridChannelRenderer': (self._grid_entries, 'items'),
4973 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4974 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4975 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4976 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4977 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4978 'playlistVideoListContinuation': (self._playlist_entries, None),
4979 'gridContinuation': (self._grid_entries, None),
4980 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4981 'sectionListContinuation': (extract_entries, None), # for feeds
4982 }
4983
4984 continuation_items = traverse_obj(response, (
4985 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4986 'appendContinuationItemsAction', 'continuationItems'
4987 ), 'continuationContents', get_all=False)
4988 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4989
4990 video_items_renderer = None
4991 for key in continuation_item.keys():
4992 if key not in known_renderers:
4993 continue
4994 func, parent_key = known_renderers[key]
4995 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4996 continuation_list = [None]
4997 yield from func(video_items_renderer)
4998 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4999
5000 if not video_items_renderer:
5001 break
5002
5003 @staticmethod
5004 def _extract_selected_tab(tabs, fatal=True):
5005 for tab_renderer in tabs:
5006 if tab_renderer.get('selected'):
5007 return tab_renderer
5008 if fatal:
5009 raise ExtractorError('Unable to find selected tab')
5010
5011 @staticmethod
5012 def _extract_tab_renderers(response):
5013 return traverse_obj(
5014 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
5015
5016 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
5017 metadata = self._extract_metadata_from_tabs(item_id, data)
5018
5019 selected_tab = self._extract_selected_tab(tabs)
5020 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
5021 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
5022
5023 return self.playlist_result(
5024 self._entries(
5025 selected_tab, metadata['id'], ytcfg,
5026 self._extract_account_syncid(ytcfg, data),
5027 self._extract_visitor_data(data, ytcfg)),
5028 **metadata)
5029
5030 def _extract_metadata_from_tabs(self, item_id, data):
5031 info = {'id': item_id}
5032
5033 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
5034 if metadata_renderer:
5035 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
5036 ('channelUrl', {self.ucid_from_url}))
5037 info.update({
5038 'channel': metadata_renderer.get('title'),
5039 'channel_id': channel_id,
5040 })
5041 if info['channel_id']:
5042 info['id'] = info['channel_id']
5043 else:
5044 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5045
5046 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5047 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5048 def _get_uncropped(url):
5049 return url_or_none((url or '').split('=')[0] + '=s0')
5050
5051 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5052 if avatar_thumbnails:
5053 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5054 if uncropped_avatar:
5055 avatar_thumbnails.append({
5056 'url': uncropped_avatar,
5057 'id': 'avatar_uncropped',
5058 'preference': 1
5059 })
5060
5061 channel_banners = self._extract_thumbnails(
5062 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5063 for banner in channel_banners:
5064 banner['preference'] = -10
5065
5066 if channel_banners:
5067 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5068 if uncropped_banner:
5069 channel_banners.append({
5070 'url': uncropped_banner,
5071 'id': 'banner_uncropped',
5072 'preference': -5
5073 })
5074
5075 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5076 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5077 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5078
5079 primary_thumbnails = self._extract_thumbnails(
5080 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5081 playlist_thumbnails = self._extract_thumbnails(
5082 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5083
5084 info.update({
5085 'title': (traverse_obj(metadata_renderer, 'title')
5086 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5087 or info['id']),
5088 'availability': self._extract_availability(data),
5089 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
5090 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5091 'tags': (traverse_obj(data, ('microformat', 'microformatDataRenderer', 'tags', ..., {str}))
5092 or traverse_obj(metadata_renderer, ('keywords', {lambda x: x and shlex.split(x)}, ...))),
5093 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5094 })
5095
5096 channel_handle = (
5097 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5098 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5099
5100 if channel_handle:
5101 info.update({
5102 'uploader_id': channel_handle,
5103 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5104 })
5105
5106 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5107 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5108 info['channel_is_verified'] = True
5109 # Playlist stats is a text runs array containing [video count, view count, last updated].
5110 # last updated or (view count and last updated) may be missing.
5111 playlist_stats = get_first(
5112 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5113
5114 last_updated_unix = self._parse_time_text(
5115 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5116 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5117 info['modified_date'] = strftime_or_none(last_updated_unix)
5118
5119 info['view_count'] = self._get_count(playlist_stats, 1)
5120 if info['view_count'] is None: # 0 is allowed
5121 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5122 if info['view_count'] is None:
5123 info['view_count'] = self._get_count(data, (
5124 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5125 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5126
5127 info['playlist_count'] = self._get_count(playlist_stats, 0)
5128 if info['playlist_count'] is None: # 0 is allowed
5129 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5130
5131 if not info.get('channel_id'):
5132 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5133 if not owner: # Deprecated
5134 owner = traverse_obj(
5135 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5136 ('videoOwner', 'videoOwnerRenderer', 'title'))
5137 owner_text = self._get_text(owner)
5138 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5139 info.update({
5140 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5141 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5142 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
5143 })
5144
5145 info.update({
5146 'uploader': info['channel'],
5147 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5148 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5149 })
5150
5151 return info
5152
5153 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5154 first_id = last_id = response = None
5155 for page_num in itertools.count(1):
5156 videos = list(self._playlist_entries(playlist))
5157 if not videos:
5158 return
5159 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5160 if start >= len(videos):
5161 return
5162 yield from videos[start:]
5163 first_id = first_id or videos[0]['id']
5164 last_id = videos[-1]['id']
5165 watch_endpoint = try_get(
5166 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5167 headers = self.generate_api_headers(
5168 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5169 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5170 query = {
5171 'playlistId': playlist_id,
5172 'videoId': watch_endpoint.get('videoId') or last_id,
5173 'index': watch_endpoint.get('index') or len(videos),
5174 'params': watch_endpoint.get('params') or 'OAE%3D'
5175 }
5176 response = self._extract_response(
5177 item_id='%s page %d' % (playlist_id, page_num),
5178 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5179 check_get_keys='contents'
5180 )
5181 playlist = try_get(
5182 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5183
5184 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5185 title = playlist.get('title') or try_get(
5186 data, lambda x: x['titleText']['simpleText'], str)
5187 playlist_id = playlist.get('playlistId') or item_id
5188
5189 # Delegating everything except mix playlists to regular tab-based playlist URL
5190 playlist_url = urljoin(url, try_get(
5191 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5192 str))
5193
5194 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5195 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5196 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5197
5198 if playlist_url and playlist_url != url and not is_known_unviewable:
5199 return self.url_result(
5200 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5201 video_title=title)
5202
5203 return self.playlist_result(
5204 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5205 playlist_id=playlist_id, playlist_title=title)
5206
5207 def _extract_availability(self, data):
5208 """
5209 Gets the availability of a given playlist/tab.
5210 Note: Unless YouTube tells us explicitly, we do not assume it is public
5211 @param data: response
5212 """
5213 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5214 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5215 player_header_privacy = playlist_header_renderer.get('privacy')
5216
5217 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5218
5219 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5220 privacy_setting_icon = get_first(
5221 (playlist_header_renderer, sidebar_renderer),
5222 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5223 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5224 expected_type=str)
5225
5226 microformats_is_unlisted = traverse_obj(
5227 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5228
5229 return (
5230 'public' if (
5231 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5232 or player_header_privacy == 'PUBLIC'
5233 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5234 else self._availability(
5235 is_private=(
5236 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5237 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5238 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5239 is_unlisted=(
5240 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5241 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5242 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5243 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5244 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5245 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5246 needs_auth=False))
5247
5248 @staticmethod
5249 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5250 sidebar_renderer = try_get(
5251 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5252 for item in sidebar_renderer:
5253 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5254 if renderer:
5255 return renderer
5256
5257 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5258 """
5259 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5260 """
5261 is_playlist = bool(traverse_obj(
5262 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5263 if not is_playlist:
5264 return
5265 headers = self.generate_api_headers(
5266 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5267 visitor_data=self._extract_visitor_data(data, ytcfg))
5268 query = {
5269 'params': 'wgYCCAA=',
5270 'browseId': f'VL{item_id}'
5271 }
5272 return self._extract_response(
5273 item_id=item_id, headers=headers, query=query,
5274 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5275 note='Redownloading playlist API JSON with unavailable videos')
5276
5277 @functools.cached_property
5278 def skip_webpage(self):
5279 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5280
5281 def _extract_webpage(self, url, item_id, fatal=True):
5282 webpage, data = None, None
5283 for retry in self.RetryManager(fatal=fatal):
5284 try:
5285 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5286 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5287 except ExtractorError as e:
5288 if isinstance(e.cause, network_exceptions):
5289 if not isinstance(e.cause, HTTPError) or e.cause.status not in (403, 429):
5290 retry.error = e
5291 continue
5292 self._error_or_warning(e, fatal=fatal)
5293 break
5294
5295 try:
5296 self._extract_and_report_alerts(data)
5297 except ExtractorError as e:
5298 self._error_or_warning(e, fatal=fatal)
5299 break
5300
5301 # Sometimes youtube returns a webpage with incomplete ytInitialData
5302 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5303 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5304 retry.error = ExtractorError('Incomplete yt initial data received')
5305 data = None
5306 continue
5307
5308 return webpage, data
5309
5310 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5311 """Use if failed to extract ytcfg (and data) from initial webpage"""
5312 if not ytcfg and self.is_authenticated:
5313 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5314 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5315 raise ExtractorError(
5316 f'{msg}. If you are not downloading private content, or '
5317 'your cookies are only for the first account and channel,'
5318 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5319 expected=True)
5320 self.report_warning(msg, only_once=True)
5321
5322 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5323 data = None
5324 if not self.skip_webpage:
5325 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5326 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5327 # Reject webpage data if redirected to home page without explicitly requesting
5328 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5329 if (url != 'https://www.youtube.com/feed/recommended'
5330 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5331 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5332 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5333 if fatal:
5334 raise ExtractorError(msg, expected=True)
5335 self.report_warning(msg, only_once=True)
5336 if not data:
5337 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5338 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5339 return data, ytcfg
5340
5341 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5342 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5343 resolve_response = self._extract_response(
5344 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5345 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5346 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5347 for ep_key, ep in endpoints.items():
5348 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5349 if params:
5350 return self._extract_response(
5351 item_id=item_id, query=params, ep=ep, headers=headers,
5352 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5353 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5354 err_note = 'Failed to resolve url (does the playlist exist?)'
5355 if fatal:
5356 raise ExtractorError(err_note, expected=True)
5357 self.report_warning(err_note, item_id)
5358
5359 _SEARCH_PARAMS = None
5360
5361 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5362 data = {'query': query}
5363 if params is NO_DEFAULT:
5364 params = self._SEARCH_PARAMS
5365 if params:
5366 data['params'] = params
5367
5368 content_keys = (
5369 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5370 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5371 # ytmusic search
5372 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5373 ('continuationContents', ),
5374 )
5375 display_id = f'query "{query}"'
5376 check_get_keys = tuple({keys[0] for keys in content_keys})
5377 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5378 self._report_playlist_authcheck(ytcfg, fatal=False)
5379
5380 continuation_list = [None]
5381 search = None
5382 for page_num in itertools.count(1):
5383 data.update(continuation_list[0] or {})
5384 headers = self.generate_api_headers(
5385 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5386 search = self._extract_response(
5387 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5388 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5389 slr_contents = traverse_obj(search, *content_keys)
5390 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5391 if not continuation_list[0]:
5392 break
5393
5394
5395 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5396 IE_DESC = 'YouTube Tabs'
5397 _VALID_URL = r'''(?x:
5398 https?://
5399 (?!consent\.)(?:\w+\.)?
5400 (?:
5401 youtube(?:kids)?\.com|
5402 %(invidious)s
5403 )/
5404 (?:
5405 (?P<channel_type>channel|c|user|browse)/|
5406 (?P<not_channel>
5407 feed/|hashtag/|
5408 (?:playlist|watch)\?.*?\blist=
5409 )|
5410 (?!(?:%(reserved_names)s)\b) # Direct URLs
5411 )
5412 (?P<id>[^/?\#&]+)
5413 )''' % {
5414 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5415 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5416 }
5417 IE_NAME = 'youtube:tab'
5418
5419 _TESTS = [{
5420 'note': 'playlists, multipage',
5421 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5422 'playlist_mincount': 94,
5423 'info_dict': {
5424 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5425 'title': 'Igor Kleiner Ph.D. - Playlists',
5426 'description': 'md5:15d7dd9e333cb987907fcb0d604b233a',
5427 'uploader': 'Igor Kleiner Ph.D.',
5428 'uploader_id': '@IgorDataScience',
5429 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5430 'channel': 'Igor Kleiner Ph.D.',
5431 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5432 'tags': ['критическое мышление', 'наука просто', 'математика', 'анализ данных'],
5433 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5434 'channel_follower_count': int
5435 },
5436 }, {
5437 'note': 'playlists, multipage, different order',
5438 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5439 'playlist_mincount': 94,
5440 'info_dict': {
5441 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5442 'title': 'Igor Kleiner Ph.D. - Playlists',
5443 'description': 'md5:15d7dd9e333cb987907fcb0d604b233a',
5444 'uploader': 'Igor Kleiner Ph.D.',
5445 'uploader_id': '@IgorDataScience',
5446 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5447 'tags': ['критическое мышление', 'наука просто', 'математика', 'анализ данных'],
5448 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5449 'channel': 'Igor Kleiner Ph.D.',
5450 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5451 'channel_follower_count': int
5452 },
5453 }, {
5454 'note': 'playlists, series',
5455 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5456 'playlist_mincount': 5,
5457 'info_dict': {
5458 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5459 'title': '3Blue1Brown - Playlists',
5460 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
5461 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5462 'channel': '3Blue1Brown',
5463 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5464 'uploader_id': '@3blue1brown',
5465 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5466 'uploader': '3Blue1Brown',
5467 'tags': ['Mathematics'],
5468 'channel_follower_count': int,
5469 'channel_is_verified': True,
5470 },
5471 }, {
5472 'note': 'playlists, singlepage',
5473 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5474 'playlist_mincount': 4,
5475 'info_dict': {
5476 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5477 'title': 'ThirstForScience - Playlists',
5478 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5479 'uploader': 'ThirstForScience',
5480 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5481 'uploader_id': '@ThirstForScience',
5482 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5483 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5484 'tags': 'count:12',
5485 'channel': 'ThirstForScience',
5486 'channel_follower_count': int
5487 }
5488 }, {
5489 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5490 'only_matching': True,
5491 }, {
5492 'note': 'basic, single video playlist',
5493 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5494 'info_dict': {
5495 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5496 'title': 'youtube-dl public playlist',
5497 'description': '',
5498 'tags': [],
5499 'view_count': int,
5500 'modified_date': '20201130',
5501 'channel': 'Sergey M.',
5502 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5503 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5504 'availability': 'public',
5505 'uploader': 'Sergey M.',
5506 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5507 'uploader_id': '@sergeym.6173',
5508 },
5509 'playlist_count': 1,
5510 }, {
5511 'note': 'empty playlist',
5512 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5513 'info_dict': {
5514 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5515 'title': 'youtube-dl empty playlist',
5516 'tags': [],
5517 'channel': 'Sergey M.',
5518 'description': '',
5519 'modified_date': '20230921',
5520 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5521 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5522 'availability': 'unlisted',
5523 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5524 'uploader_id': '@sergeym.6173',
5525 'uploader': 'Sergey M.',
5526 },
5527 'playlist_count': 0,
5528 }, {
5529 'note': 'Home tab',
5530 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5531 'info_dict': {
5532 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5533 'title': 'lex will - Home',
5534 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5535 'uploader': 'lex will',
5536 'uploader_id': '@lexwill718',
5537 'channel': 'lex will',
5538 'tags': ['bible', 'history', 'prophesy'],
5539 'uploader_url': 'https://www.youtube.com/@lexwill718',
5540 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5541 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5542 'channel_follower_count': int
5543 },
5544 'playlist_mincount': 2,
5545 }, {
5546 'note': 'Videos tab',
5547 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5548 'info_dict': {
5549 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5550 'title': 'lex will - Videos',
5551 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5552 'uploader': 'lex will',
5553 'uploader_id': '@lexwill718',
5554 'tags': ['bible', 'history', 'prophesy'],
5555 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5556 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5557 'uploader_url': 'https://www.youtube.com/@lexwill718',
5558 'channel': 'lex will',
5559 'channel_follower_count': int
5560 },
5561 'playlist_mincount': 975,
5562 }, {
5563 'note': 'Videos tab, sorted by popular',
5564 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5565 'info_dict': {
5566 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5567 'title': 'lex will - Videos',
5568 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5569 'uploader': 'lex will',
5570 'uploader_id': '@lexwill718',
5571 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5572 'uploader_url': 'https://www.youtube.com/@lexwill718',
5573 'channel': 'lex will',
5574 'tags': ['bible', 'history', 'prophesy'],
5575 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5576 'channel_follower_count': int
5577 },
5578 'playlist_mincount': 199,
5579 }, {
5580 'note': 'Playlists tab',
5581 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5582 'info_dict': {
5583 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5584 'title': 'lex will - Playlists',
5585 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5586 'uploader': 'lex will',
5587 'uploader_id': '@lexwill718',
5588 'uploader_url': 'https://www.youtube.com/@lexwill718',
5589 'channel': 'lex will',
5590 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5591 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5592 'tags': ['bible', 'history', 'prophesy'],
5593 'channel_follower_count': int
5594 },
5595 'playlist_mincount': 17,
5596 }, {
5597 'note': 'Community tab',
5598 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5599 'info_dict': {
5600 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5601 'title': 'lex will - Community',
5602 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5603 'channel': 'lex will',
5604 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5605 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5606 'tags': ['bible', 'history', 'prophesy'],
5607 'channel_follower_count': int,
5608 'uploader_url': 'https://www.youtube.com/@lexwill718',
5609 'uploader_id': '@lexwill718',
5610 'uploader': 'lex will',
5611 },
5612 'playlist_mincount': 18,
5613 }, {
5614 'note': 'Channels tab',
5615 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5616 'info_dict': {
5617 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5618 'title': 'lex will - Channels',
5619 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5620 'channel': 'lex will',
5621 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5622 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5623 'tags': ['bible', 'history', 'prophesy'],
5624 'channel_follower_count': int,
5625 'uploader_url': 'https://www.youtube.com/@lexwill718',
5626 'uploader_id': '@lexwill718',
5627 'uploader': 'lex will',
5628 },
5629 'playlist_mincount': 12,
5630 }, {
5631 'note': 'Search tab',
5632 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5633 'playlist_mincount': 40,
5634 'info_dict': {
5635 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5636 'title': '3Blue1Brown - Search - linear algebra',
5637 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
5638 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5639 'tags': ['Mathematics'],
5640 'channel': '3Blue1Brown',
5641 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5642 'channel_follower_count': int,
5643 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5644 'uploader_id': '@3blue1brown',
5645 'uploader': '3Blue1Brown',
5646 'channel_is_verified': True,
5647 },
5648 }, {
5649 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5650 'only_matching': True,
5651 }, {
5652 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5653 'only_matching': True,
5654 }, {
5655 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5656 'only_matching': True,
5657 }, {
5658 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5659 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5660 'info_dict': {
5661 'title': '29C3: Not my department',
5662 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5663 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5664 'tags': [],
5665 'view_count': int,
5666 'modified_date': '20150605',
5667 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5668 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5669 'channel': 'Christiaan008',
5670 'availability': 'public',
5671 'uploader_id': '@ChRiStIaAn008',
5672 'uploader': 'Christiaan008',
5673 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5674 },
5675 'playlist_count': 96,
5676 }, {
5677 'note': 'Large playlist',
5678 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5679 'info_dict': {
5680 'title': 'Uploads from Cauchemar',
5681 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5682 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5683 'tags': [],
5684 'modified_date': r're:\d{8}',
5685 'channel': 'Cauchemar',
5686 'view_count': int,
5687 'description': '',
5688 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5689 'availability': 'public',
5690 'uploader_id': '@Cauchemar89',
5691 'uploader': 'Cauchemar',
5692 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5693 },
5694 'playlist_mincount': 1123,
5695 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5696 }, {
5697 'note': 'even larger playlist, 8832 videos',
5698 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5699 'only_matching': True,
5700 }, {
5701 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5702 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5703 'info_dict': {
5704 'title': 'Uploads from Interstellar Movie',
5705 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5706 'tags': [],
5707 'view_count': int,
5708 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5709 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5710 'channel': 'Interstellar Movie',
5711 'description': '',
5712 'modified_date': r're:\d{8}',
5713 'availability': 'public',
5714 'uploader_id': '@InterstellarMovie',
5715 'uploader': 'Interstellar Movie',
5716 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5717 },
5718 'playlist_mincount': 21,
5719 }, {
5720 'note': 'Playlist with "show unavailable videos" button',
5721 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5722 'info_dict': {
5723 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5724 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5725 'view_count': int,
5726 'channel': 'Phim Siêu Nhân Nhật Bản',
5727 'tags': [],
5728 'description': '',
5729 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5730 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5731 'modified_date': r're:\d{8}',
5732 'availability': 'public',
5733 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5734 'uploader_id': '@phimsieunhannhatban',
5735 'uploader': 'Phim Siêu Nhân Nhật Bản',
5736 },
5737 'playlist_mincount': 200,
5738 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5739 }, {
5740 'note': 'Playlist with unavailable videos in page 7',
5741 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5742 'info_dict': {
5743 'title': 'Uploads from BlankTV',
5744 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5745 'channel': 'BlankTV',
5746 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5747 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5748 'view_count': int,
5749 'tags': [],
5750 'modified_date': r're:\d{8}',
5751 'description': '',
5752 'availability': 'public',
5753 'uploader_id': '@blanktv',
5754 'uploader': 'BlankTV',
5755 'uploader_url': 'https://www.youtube.com/@blanktv',
5756 },
5757 'playlist_mincount': 1000,
5758 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5759 }, {
5760 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5761 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5762 'info_dict': {
5763 'title': 'Data Analysis with Dr Mike Pound',
5764 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5765 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5766 'tags': [],
5767 'view_count': int,
5768 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5769 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5770 'channel': 'Computerphile',
5771 'availability': 'public',
5772 'modified_date': '20190712',
5773 'uploader_id': '@Computerphile',
5774 'uploader': 'Computerphile',
5775 'uploader_url': 'https://www.youtube.com/@Computerphile',
5776 },
5777 'playlist_mincount': 11,
5778 }, {
5779 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5780 'only_matching': True,
5781 }, {
5782 'note': 'Playlist URL that does not actually serve a playlist',
5783 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5784 'info_dict': {
5785 'id': 'FqZTN594JQw',
5786 'ext': 'webm',
5787 'title': "Smiley's People 01 detective, Adventure Series, Action",
5788 'upload_date': '20150526',
5789 'license': 'Standard YouTube License',
5790 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5791 'categories': ['People & Blogs'],
5792 'tags': list,
5793 'view_count': int,
5794 'like_count': int,
5795 },
5796 'params': {
5797 'skip_download': True,
5798 },
5799 'skip': 'This video is not available.',
5800 'add_ie': [YoutubeIE.ie_key()],
5801 }, {
5802 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5803 'only_matching': True,
5804 }, {
5805 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5806 'only_matching': True,
5807 }, {
5808 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5809 'info_dict': {
5810 'id': 'hGkQjiJLjWQ', # This will keep changing
5811 'ext': 'mp4',
5812 'title': str,
5813 'upload_date': r're:\d{8}',
5814 'description': str,
5815 'categories': ['News & Politics'],
5816 'tags': list,
5817 'like_count': int,
5818 'release_timestamp': int,
5819 'channel': 'Sky News',
5820 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5821 'age_limit': 0,
5822 'view_count': int,
5823 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5824 'playable_in_embed': True,
5825 'release_date': r're:\d+',
5826 'availability': 'public',
5827 'live_status': 'is_live',
5828 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5829 'channel_follower_count': int,
5830 'concurrent_view_count': int,
5831 'uploader_url': 'https://www.youtube.com/@SkyNews',
5832 'uploader_id': '@SkyNews',
5833 'uploader': 'Sky News',
5834 'channel_is_verified': True,
5835 },
5836 'params': {
5837 'skip_download': True,
5838 },
5839 'expected_warnings': ['Ignoring subtitle tracks found in '],
5840 }, {
5841 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5842 'info_dict': {
5843 'id': 'a48o2S1cPoo',
5844 'ext': 'mp4',
5845 'title': 'The Young Turks - Live Main Show',
5846 'upload_date': '20150715',
5847 'license': 'Standard YouTube License',
5848 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5849 'categories': ['News & Politics'],
5850 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5851 'like_count': int,
5852 },
5853 'params': {
5854 'skip_download': True,
5855 },
5856 'only_matching': True,
5857 }, {
5858 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5859 'only_matching': True,
5860 }, {
5861 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5862 'only_matching': True,
5863 }, {
5864 'note': 'A channel that is not live. Should raise error',
5865 'url': 'https://www.youtube.com/user/numberphile/live',
5866 'only_matching': True,
5867 }, {
5868 'url': 'https://www.youtube.com/feed/trending',
5869 'only_matching': True,
5870 }, {
5871 'url': 'https://www.youtube.com/feed/library',
5872 'only_matching': True,
5873 }, {
5874 'url': 'https://www.youtube.com/feed/history',
5875 'only_matching': True,
5876 }, {
5877 'url': 'https://www.youtube.com/feed/subscriptions',
5878 'only_matching': True,
5879 }, {
5880 'url': 'https://www.youtube.com/feed/watch_later',
5881 'only_matching': True,
5882 }, {
5883 'note': 'Recommended - redirects to home page.',
5884 'url': 'https://www.youtube.com/feed/recommended',
5885 'only_matching': True,
5886 }, {
5887 'note': 'inline playlist with not always working continuations',
5888 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5889 'only_matching': True,
5890 }, {
5891 'url': 'https://www.youtube.com/course',
5892 'only_matching': True,
5893 }, {
5894 'url': 'https://www.youtube.com/zsecurity',
5895 'only_matching': True,
5896 }, {
5897 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5898 'only_matching': True,
5899 }, {
5900 'url': 'https://www.youtube.com/TheYoungTurks/live',
5901 'only_matching': True,
5902 }, {
5903 'url': 'https://www.youtube.com/hashtag/cctv9',
5904 'info_dict': {
5905 'id': 'cctv9',
5906 'title': 'cctv9 - All',
5907 'tags': [],
5908 },
5909 'playlist_mincount': 300, # not consistent but should be over 300
5910 }, {
5911 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5912 'only_matching': True,
5913 }, {
5914 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5915 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5916 'only_matching': True
5917 }, {
5918 'note': '/browse/ should redirect to /channel/',
5919 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5920 'only_matching': True
5921 }, {
5922 'note': 'VLPL, should redirect to playlist?list=PL...',
5923 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5924 'info_dict': {
5925 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5926 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5927 'title': 'NCS : All Releases 💿',
5928 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5929 'modified_date': r're:\d{8}',
5930 'view_count': int,
5931 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5932 'tags': [],
5933 'channel': 'NoCopyrightSounds',
5934 'availability': 'public',
5935 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5936 'uploader': 'NoCopyrightSounds',
5937 'uploader_id': '@NoCopyrightSounds',
5938 },
5939 'playlist_mincount': 166,
5940 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5941 }, {
5942 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5943 'note': 'Topic, should redirect to playlist?list=UU...',
5944 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5945 'info_dict': {
5946 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5947 'title': 'Uploads from Royalty Free Music - Topic',
5948 'tags': [],
5949 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5950 'channel': 'Royalty Free Music - Topic',
5951 'view_count': int,
5952 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5953 'modified_date': r're:\d{8}',
5954 'description': '',
5955 'availability': 'public',
5956 'uploader': 'Royalty Free Music - Topic',
5957 },
5958 'playlist_mincount': 101,
5959 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5960 }, {
5961 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5962 # Treat as a general feed
5963 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5964 'info_dict': {
5965 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5966 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5967 'tags': [],
5968 },
5969 'playlist_mincount': 9,
5970 }, {
5971 'note': 'Youtube music Album',
5972 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5973 'info_dict': {
5974 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5975 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5976 'tags': [],
5977 'view_count': int,
5978 'description': '',
5979 'availability': 'unlisted',
5980 'modified_date': r're:\d{8}',
5981 },
5982 'playlist_count': 50,
5983 'expected_warnings': ['YouTube Music is not directly supported'],
5984 }, {
5985 'note': 'unlisted single video playlist',
5986 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5987 'info_dict': {
5988 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5989 'title': 'yt-dlp unlisted playlist test',
5990 'availability': 'unlisted',
5991 'tags': [],
5992 'modified_date': '20220418',
5993 'channel': 'colethedj',
5994 'view_count': int,
5995 'description': '',
5996 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5997 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5998 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5999 'uploader_id': '@colethedj1894',
6000 'uploader': 'colethedj',
6001 },
6002 'playlist': [{
6003 'info_dict': {
6004 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
6005 'id': 'BaW_jenozKc',
6006 '_type': 'url',
6007 'ie_key': 'Youtube',
6008 'duration': 10,
6009 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
6010 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
6011 'view_count': int,
6012 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
6013 'channel': 'Philipp Hagemeister',
6014 'uploader_id': '@PhilippHagemeister',
6015 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
6016 'uploader': 'Philipp Hagemeister',
6017 }
6018 }],
6019 'playlist_count': 1,
6020 'params': {'extract_flat': True},
6021 }, {
6022 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
6023 'url': 'https://www.youtube.com/feed/recommended',
6024 'info_dict': {
6025 'id': 'recommended',
6026 'title': 'recommended',
6027 'tags': [],
6028 },
6029 'playlist_mincount': 50,
6030 'params': {
6031 'skip_download': True,
6032 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6033 },
6034 }, {
6035 'note': 'API Fallback: /videos tab, sorted by oldest first',
6036 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
6037 'info_dict': {
6038 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6039 'title': 'Cody\'sLab - Videos',
6040 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6041 'channel': 'Cody\'sLab',
6042 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6043 'tags': [],
6044 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6045 'channel_follower_count': int
6046 },
6047 'playlist_mincount': 650,
6048 'params': {
6049 'skip_download': True,
6050 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6051 },
6052 'skip': 'Query for sorting no longer works',
6053 }, {
6054 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6055 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6056 'info_dict': {
6057 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6058 'title': 'Uploads from Royalty Free Music - Topic',
6059 'modified_date': r're:\d{8}',
6060 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6061 'description': '',
6062 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6063 'tags': [],
6064 'channel': 'Royalty Free Music - Topic',
6065 'view_count': int,
6066 'availability': 'public',
6067 'uploader': 'Royalty Free Music - Topic',
6068 },
6069 'playlist_mincount': 101,
6070 'params': {
6071 'skip_download': True,
6072 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6073 },
6074 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6075 }, {
6076 'note': 'non-standard redirect to regional channel',
6077 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6078 'only_matching': True
6079 }, {
6080 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6081 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6082 'info_dict': {
6083 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6084 'modified_date': '20220407',
6085 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6086 'tags': [],
6087 'availability': 'unlisted',
6088 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6089 'channel': 'pukkandan',
6090 'description': 'Test for collaborative playlist',
6091 'title': 'yt-dlp test - collaborative playlist',
6092 'view_count': int,
6093 'uploader_url': 'https://www.youtube.com/@pukkandan',
6094 'uploader_id': '@pukkandan',
6095 'uploader': 'pukkandan',
6096 },
6097 'playlist_mincount': 2
6098 }, {
6099 'note': 'translated tab name',
6100 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6101 'info_dict': {
6102 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6103 'tags': [],
6104 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6105 'description': 'test description',
6106 'title': 'cole-dlp-test-acc - 再生リスト',
6107 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6108 'channel': 'cole-dlp-test-acc',
6109 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6110 'uploader_id': '@coletdjnz',
6111 'uploader': 'cole-dlp-test-acc',
6112 },
6113 'playlist_mincount': 1,
6114 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6115 'expected_warnings': ['Preferring "ja"'],
6116 }, {
6117 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6118 'note': 'preferred lang set with playlist with translated video titles',
6119 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6120 'info_dict': {
6121 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6122 'tags': [],
6123 'view_count': int,
6124 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6125 'channel': 'cole-dlp-test-acc',
6126 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6127 'description': 'test',
6128 'title': 'dlp test playlist',
6129 'availability': 'public',
6130 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6131 'uploader_id': '@coletdjnz',
6132 'uploader': 'cole-dlp-test-acc',
6133 },
6134 'playlist_mincount': 1,
6135 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6136 'expected_warnings': ['Preferring "ja"'],
6137 }, {
6138 # shorts audio pivot for 2GtVksBMYFM.
6139 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6140 'info_dict': {
6141 'id': 'sfv_audio_pivot',
6142 'title': 'sfv_audio_pivot',
6143 'tags': [],
6144 },
6145 'playlist_mincount': 50,
6146
6147 }, {
6148 # Channel with a real live tab (not to be mistaken with streams tab)
6149 # Do not treat like it should redirect to live stream
6150 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6151 'info_dict': {
6152 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6153 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6154 'tags': [],
6155 },
6156 'playlist_mincount': 20,
6157 }, {
6158 # Tab name is not the same as tab id
6159 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6160 'info_dict': {
6161 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6162 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6163 'tags': [],
6164 },
6165 'playlist_mincount': 8,
6166 }, {
6167 # Home tab id is literally home. Not to get mistaken with featured
6168 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6169 'info_dict': {
6170 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6171 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6172 'tags': [],
6173 },
6174 'playlist_mincount': 8,
6175 }, {
6176 # Should get three playlists for videos, shorts and streams tabs
6177 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6178 'info_dict': {
6179 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6180 'title': 'Polka Ch. 尾丸ポルカ',
6181 'channel_follower_count': int,
6182 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6183 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6184 'description': 'md5:49809d8bf9da539bc48ed5d1f83c33f2',
6185 'channel': 'Polka Ch. 尾丸ポルカ',
6186 'tags': 'count:35',
6187 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6188 'uploader': 'Polka Ch. 尾丸ポルカ',
6189 'uploader_id': '@OmaruPolka',
6190 'channel_is_verified': True,
6191 },
6192 'playlist_count': 3,
6193 }, {
6194 # Shorts tab with channel with handle
6195 # TODO: fix channel description
6196 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6197 'info_dict': {
6198 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6199 'title': 'Not Just Bikes - Shorts',
6200 'tags': 'count:10',
6201 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6202 'description': 'md5:5e82545b3a041345927a92d0585df247',
6203 'channel_follower_count': int,
6204 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6205 'channel': 'Not Just Bikes',
6206 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6207 'uploader': 'Not Just Bikes',
6208 'uploader_id': '@NotJustBikes',
6209 'channel_is_verified': True,
6210 },
6211 'playlist_mincount': 10,
6212 }, {
6213 # Streams tab
6214 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6215 'info_dict': {
6216 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6217 'title': '中村悠一 - Live',
6218 'tags': 'count:7',
6219 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6220 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6221 'channel': '中村悠一',
6222 'channel_follower_count': int,
6223 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6224 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6225 'uploader_id': '@Yuichi-Nakamura',
6226 'uploader': '中村悠一',
6227 },
6228 'playlist_mincount': 60,
6229 }, {
6230 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6231 # See test_youtube_lists
6232 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6233 'only_matching': True,
6234 }, {
6235 # No uploads and no UCID given. Should fail with no uploads error
6236 # See test_youtube_lists
6237 'url': 'https://www.youtube.com/news',
6238 'only_matching': True
6239 }, {
6240 # No videos tab but has a shorts tab
6241 'url': 'https://www.youtube.com/c/TKFShorts',
6242 'info_dict': {
6243 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6244 'title': 'Shorts Break - Shorts',
6245 'tags': 'count:48',
6246 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6247 'channel': 'Shorts Break',
6248 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6249 'channel_follower_count': int,
6250 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6251 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6252 'uploader': 'Shorts Break',
6253 'uploader_id': '@ShortsBreak_Official',
6254 },
6255 'playlist_mincount': 30,
6256 }, {
6257 # Trending Now Tab. tab id is empty
6258 'url': 'https://www.youtube.com/feed/trending',
6259 'info_dict': {
6260 'id': 'trending',
6261 'title': 'trending - Now',
6262 'tags': [],
6263 },
6264 'playlist_mincount': 30,
6265 }, {
6266 # Trending Gaming Tab. tab id is empty
6267 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6268 'info_dict': {
6269 'id': 'trending',
6270 'title': 'trending - Gaming',
6271 'tags': [],
6272 },
6273 'playlist_mincount': 30,
6274 }, {
6275 # Shorts url result in shorts tab
6276 # TODO: Fix channel id extraction
6277 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6278 'info_dict': {
6279 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6280 'title': 'cole-dlp-test-acc - Shorts',
6281 'channel': 'cole-dlp-test-acc',
6282 'description': 'test description',
6283 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6284 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6285 'tags': [],
6286 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6287 'uploader_id': '@coletdjnz',
6288 'uploader': 'cole-dlp-test-acc',
6289 },
6290 'playlist': [{
6291 'info_dict': {
6292 # Channel data is not currently available for short renderers (as of 2023-03-01)
6293 '_type': 'url',
6294 'ie_key': 'Youtube',
6295 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6296 'id': 'sSM9J5YH_60',
6297 'title': 'SHORT short',
6298 'view_count': int,
6299 'thumbnails': list,
6300 }
6301 }],
6302 'params': {'extract_flat': True},
6303 }, {
6304 # Live video status should be extracted
6305 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6306 'info_dict': {
6307 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6308 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6309 'tags': []
6310 },
6311 'playlist': [{
6312 'info_dict': {
6313 '_type': 'url',
6314 'ie_key': 'Youtube',
6315 'url': 'startswith:https://www.youtube.com/watch?v=',
6316 'id': str,
6317 'title': str,
6318 'live_status': 'is_live',
6319 'channel_id': str,
6320 'channel_url': str,
6321 'concurrent_view_count': int,
6322 'channel': str,
6323 'uploader': str,
6324 'uploader_url': str,
6325 'uploader_id': str,
6326 'channel_is_verified': bool, # this will keep changing
6327 }
6328 }],
6329 'params': {'extract_flat': True, 'playlist_items': '1'},
6330 'playlist_mincount': 1
6331 }, {
6332 # Channel renderer metadata. Contains number of videos on the channel
6333 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6334 'info_dict': {
6335 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6336 'title': 'cole-dlp-test-acc - Channels',
6337 'channel': 'cole-dlp-test-acc',
6338 'description': 'test description',
6339 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6340 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6341 'tags': [],
6342 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6343 'uploader_id': '@coletdjnz',
6344 'uploader': 'cole-dlp-test-acc',
6345 },
6346 'playlist': [{
6347 'info_dict': {
6348 '_type': 'url',
6349 'ie_key': 'YoutubeTab',
6350 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6351 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6352 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6353 'title': 'PewDiePie',
6354 'channel': 'PewDiePie',
6355 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6356 'thumbnails': list,
6357 'channel_follower_count': int,
6358 'playlist_count': int,
6359 'uploader': 'PewDiePie',
6360 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6361 'uploader_id': '@PewDiePie',
6362 'channel_is_verified': True,
6363 }
6364 }],
6365 'params': {'extract_flat': True},
6366 }, {
6367 'url': 'https://www.youtube.com/@3blue1brown/about',
6368 'info_dict': {
6369 'id': '@3blue1brown',
6370 'tags': ['Mathematics'],
6371 'title': '3Blue1Brown',
6372 'channel_follower_count': int,
6373 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6374 'channel': '3Blue1Brown',
6375 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6376 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
6377 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6378 'uploader_id': '@3blue1brown',
6379 'uploader': '3Blue1Brown',
6380 'channel_is_verified': True,
6381 },
6382 'playlist_count': 0,
6383 }, {
6384 # Podcasts tab, with rich entry playlistRenderers
6385 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6386 'info_dict': {
6387 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6388 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6389 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6390 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6391 'title': '99 Percent Invisible - Podcasts',
6392 'uploader': '99 Percent Invisible',
6393 'channel_follower_count': int,
6394 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6395 'tags': [],
6396 'channel': '99 Percent Invisible',
6397 'uploader_id': '@99percentinvisiblepodcast',
6398 },
6399 'playlist_count': 0,
6400 }, {
6401 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6402 'url': 'https://www.youtube.com/@AHimitsu/releases',
6403 'info_dict': {
6404 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6405 'channel': 'A Himitsu',
6406 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6407 'title': 'A Himitsu - Releases',
6408 'uploader_id': '@AHimitsu',
6409 'uploader': 'A Himitsu',
6410 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6411 'tags': 'count:12',
6412 'description': 'I make music',
6413 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6414 'channel_follower_count': int,
6415 'channel_is_verified': True,
6416 },
6417 'playlist_mincount': 10,
6418 }, {
6419 # Playlist with only shorts, shown as reel renderers
6420 # FIXME: future: YouTube currently doesn't give continuation for this,
6421 # may do in future.
6422 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
6423 'info_dict': {
6424 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
6425 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
6426 'view_count': int,
6427 'uploader_id': '@BangyShorts',
6428 'description': '',
6429 'uploader_url': 'https://www.youtube.com/@BangyShorts',
6430 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
6431 'channel': 'Bangy Shorts',
6432 'uploader': 'Bangy Shorts',
6433 'tags': [],
6434 'availability': 'public',
6435 'modified_date': r're:\d{8}',
6436 'title': 'Uploads from Bangy Shorts',
6437 },
6438 'playlist_mincount': 100,
6439 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6440 }, {
6441 'note': 'Tags containing spaces',
6442 'url': 'https://www.youtube.com/channel/UC7_YxT-KID8kRbqZo7MyscQ',
6443 'playlist_count': 3,
6444 'info_dict': {
6445 'id': 'UC7_YxT-KID8kRbqZo7MyscQ',
6446 'channel': 'Markiplier',
6447 'channel_id': 'UC7_YxT-KID8kRbqZo7MyscQ',
6448 'title': 'Markiplier',
6449 'channel_follower_count': int,
6450 'description': 'md5:0c010910558658824402809750dc5d97',
6451 'uploader_id': '@markiplier',
6452 'uploader_url': 'https://www.youtube.com/@markiplier',
6453 'uploader': 'Markiplier',
6454 'channel_url': 'https://www.youtube.com/channel/UC7_YxT-KID8kRbqZo7MyscQ',
6455 'channel_is_verified': True,
6456 'tags': ['markiplier', 'comedy', 'gaming', 'funny videos', 'funny moments',
6457 'sketch comedy', 'laughing', 'lets play', 'challenge videos', 'hilarious',
6458 'challenges', 'sketches', 'scary games', 'funny games', 'rage games',
6459 'mark fischbach'],
6460 },
6461 }]
6462
6463 @classmethod
6464 def suitable(cls, url):
6465 return False if YoutubeIE.suitable(url) else super().suitable(url)
6466
6467 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6468
6469 def _get_url_mobj(self, url):
6470 mobj = self._URL_RE.match(url).groupdict()
6471 mobj.update((k, '') for k, v in mobj.items() if v is None)
6472 return mobj
6473
6474 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6475 tab_name = (tab.get('title') or '').lower()
6476 tab_url = urljoin(base_url, traverse_obj(
6477 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6478
6479 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6480 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6481 if tab_id:
6482 return {
6483 'TAB_ID_SPONSORSHIPS': 'membership',
6484 }.get(tab_id, tab_id), tab_name
6485
6486 # Fallback to tab name if we cannot get the tab id.
6487 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6488 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6489 if tab_name:
6490 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6491 return {
6492 'home': 'featured',
6493 'live': 'streams',
6494 }.get(tab_name, tab_name), tab_name
6495
6496 def _has_tab(self, tabs, tab_id):
6497 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6498
6499 def _empty_playlist(self, item_id, data):
6500 return self.playlist_result([], item_id, **self._extract_metadata_from_tabs(item_id, data))
6501
6502 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6503 def _real_extract(self, url, smuggled_data):
6504 item_id = self._match_id(url)
6505 url = urllib.parse.urlunparse(
6506 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6507 compat_opts = self.get_param('compat_opts', [])
6508
6509 mobj = self._get_url_mobj(url)
6510 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6511 if is_channel and smuggled_data.get('is_music_url'):
6512 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6513 return self.url_result(
6514 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6515 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6516 mdata = self._extract_tab_endpoint(
6517 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6518 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6519 get_all=False, expected_type=str)
6520 if not murl:
6521 raise ExtractorError('Failed to resolve album to playlist')
6522 return self.url_result(murl, YoutubeTabIE)
6523 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6524 return self.url_result(
6525 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6526
6527 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6528 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6529 url = f'{pre}/videos{post}'
6530 if smuggled_data.get('is_music_url'):
6531 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6532
6533 # Handle both video/playlist URLs
6534 qs = parse_qs(url)
6535 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6536 if not video_id and mobj['not_channel'].startswith('watch'):
6537 if not playlist_id:
6538 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6539 raise ExtractorError('A video URL was given without video ID', expected=True)
6540 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6541 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6542 return self.url_result(
6543 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6544
6545 if not self._yes_playlist(playlist_id, video_id):
6546 return self.url_result(
6547 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6548
6549 data, ytcfg = self._extract_data(url, display_id)
6550
6551 # YouTube may provide a non-standard redirect to the regional channel
6552 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6553 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6554 redirect_url = traverse_obj(
6555 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6556 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6557 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6558 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6559 return self.url_result(redirect_url, YoutubeTabIE)
6560
6561 tabs, extra_tabs = self._extract_tab_renderers(data), []
6562 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6563 selected_tab = self._extract_selected_tab(tabs)
6564 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6565 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6566
6567 # /about is no longer a tab
6568 if original_tab_id == 'about':
6569 return self._empty_playlist(item_id, data)
6570
6571 if not original_tab_id and selected_tab_name:
6572 self.to_screen('Downloading all uploads of the channel. '
6573 'To download only the videos in a specific tab, pass the tab\'s URL')
6574 if self._has_tab(tabs, 'streams'):
6575 extra_tabs.append(''.join((pre, '/streams', post)))
6576 if self._has_tab(tabs, 'shorts'):
6577 extra_tabs.append(''.join((pre, '/shorts', post)))
6578 # XXX: Members-only tab should also be extracted
6579
6580 if not extra_tabs and selected_tab_id != 'videos':
6581 # Channel does not have streams, shorts or videos tabs
6582 if item_id[:2] != 'UC':
6583 return self._empty_playlist(item_id, data)
6584
6585 # Topic channels don't have /videos. Use the equivalent playlist instead
6586 pl_id = f'UU{item_id[2:]}'
6587 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6588 try:
6589 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6590 except ExtractorError:
6591 return self._empty_playlist(item_id, data)
6592 else:
6593 item_id, url = pl_id, pl_url
6594 self.to_screen(
6595 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6596
6597 elif extra_tabs and selected_tab_id != 'videos':
6598 # When there are shorts/live tabs but not videos tab
6599 url, data = f'{pre}{post}', None
6600
6601 elif (original_tab_id or 'videos') != selected_tab_id:
6602 if original_tab_id == 'live':
6603 # Live tab should have redirected to the video
6604 # Except in the case the channel has an actual live tab
6605 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6606 raise UserNotLive(video_id=item_id)
6607 elif selected_tab_name:
6608 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6609
6610 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6611 url = f'{pre}{post}'
6612
6613 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6614 if 'no-youtube-unavailable-videos' not in compat_opts:
6615 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6616 self._extract_and_report_alerts(data, only_once=True)
6617
6618 tabs, entries = self._extract_tab_renderers(data), []
6619 if tabs:
6620 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6621 entries[0].update({
6622 'extractor_key': YoutubeTabIE.ie_key(),
6623 'extractor': YoutubeTabIE.IE_NAME,
6624 'webpage_url': url,
6625 })
6626 if self.get_param('playlist_items') == '0':
6627 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6628 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6629 entries.extend(map(self._real_extract, extra_tabs))
6630
6631 if len(entries) == 1:
6632 return entries[0]
6633 elif entries:
6634 metadata = self._extract_metadata_from_tabs(item_id, data)
6635 uploads_url = 'the Uploads (UU) playlist URL'
6636 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6637 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6638 self.to_screen(
6639 'Downloading as multiple playlists, separated by tabs. '
6640 f'To download as a single playlist instead, pass {uploads_url}')
6641 return self.playlist_result(entries, item_id, **metadata)
6642
6643 # Inline playlist
6644 playlist = traverse_obj(
6645 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6646 if playlist:
6647 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6648
6649 video_id = traverse_obj(
6650 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6651 if video_id:
6652 if tab != '/live': # live tab is expected to redirect to video
6653 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6654 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6655
6656 raise ExtractorError('Unable to recognize tab page')
6657
6658
6659 class YoutubePlaylistIE(InfoExtractor):
6660 IE_DESC = 'YouTube playlists'
6661 _VALID_URL = r'''(?x)(?:
6662 (?:https?://)?
6663 (?:\w+\.)?
6664 (?:
6665 (?:
6666 youtube(?:kids)?\.com|
6667 %(invidious)s
6668 )
6669 /.*?\?.*?\blist=
6670 )?
6671 (?P<id>%(playlist_id)s)
6672 )''' % {
6673 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6674 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6675 }
6676 IE_NAME = 'youtube:playlist'
6677 _TESTS = [{
6678 'note': 'issue #673',
6679 'url': 'PLBB231211A4F62143',
6680 'info_dict': {
6681 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6682 'id': 'PLBB231211A4F62143',
6683 'uploader': 'Wickman',
6684 'uploader_id': '@WickmanVT',
6685 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6686 'view_count': int,
6687 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6688 'modified_date': r're:\d{8}',
6689 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6690 'channel': 'Wickman',
6691 'tags': [],
6692 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6693 'availability': 'public',
6694 },
6695 'playlist_mincount': 29,
6696 }, {
6697 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6698 'info_dict': {
6699 'title': 'YDL_safe_search',
6700 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6701 },
6702 'playlist_count': 2,
6703 'skip': 'This playlist is private',
6704 }, {
6705 'note': 'embedded',
6706 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6707 'playlist_count': 4,
6708 'info_dict': {
6709 'title': 'JODA15',
6710 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6711 'uploader': 'milan',
6712 'uploader_id': '@milan5503',
6713 'description': '',
6714 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6715 'tags': [],
6716 'modified_date': '20140919',
6717 'view_count': int,
6718 'channel': 'milan',
6719 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6720 'uploader_url': 'https://www.youtube.com/@milan5503',
6721 'availability': 'public',
6722 },
6723 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden', 'Retrying', 'Giving up'],
6724 }, {
6725 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6726 'playlist_mincount': 455,
6727 'info_dict': {
6728 'title': '2018 Chinese New Singles (11/6 updated)',
6729 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6730 'uploader': 'LBK',
6731 'uploader_id': '@music_king',
6732 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6733 'channel': 'LBK',
6734 'view_count': int,
6735 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6736 'tags': [],
6737 'uploader_url': 'https://www.youtube.com/@music_king',
6738 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6739 'modified_date': r're:\d{8}',
6740 'availability': 'public',
6741 },
6742 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6743 }, {
6744 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6745 'only_matching': True,
6746 }, {
6747 # music album playlist
6748 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6749 'only_matching': True,
6750 }]
6751
6752 @classmethod
6753 def suitable(cls, url):
6754 if YoutubeTabIE.suitable(url):
6755 return False
6756 from ..utils import parse_qs
6757 qs = parse_qs(url)
6758 if qs.get('v', [None])[0]:
6759 return False
6760 return super().suitable(url)
6761
6762 def _real_extract(self, url):
6763 playlist_id = self._match_id(url)
6764 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6765 url = update_url_query(
6766 'https://www.youtube.com/playlist',
6767 parse_qs(url) or {'list': playlist_id})
6768 if is_music_url:
6769 url = smuggle_url(url, {'is_music_url': True})
6770 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6771
6772
6773 class YoutubeYtBeIE(InfoExtractor):
6774 IE_DESC = 'youtu.be'
6775 _VALID_URL = r'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{11})/*?.*?\blist=(?P<playlist_id>%(playlist_id)s)' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
6776 _TESTS = [{
6777 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6778 'info_dict': {
6779 'id': 'yeWKywCrFtk',
6780 'ext': 'mp4',
6781 'title': 'Small Scale Baler and Braiding Rugs',
6782 'uploader': 'Backus-Page House Museum',
6783 'uploader_id': '@backuspagemuseum',
6784 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6785 'upload_date': '20161008',
6786 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6787 'categories': ['Nonprofits & Activism'],
6788 'tags': list,
6789 'like_count': int,
6790 'age_limit': 0,
6791 'playable_in_embed': True,
6792 'thumbnail': r're:^https?://.*\.webp',
6793 'channel': 'Backus-Page House Museum',
6794 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6795 'live_status': 'not_live',
6796 'view_count': int,
6797 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6798 'availability': 'public',
6799 'duration': 59,
6800 'comment_count': int,
6801 'channel_follower_count': int
6802 },
6803 'params': {
6804 'noplaylist': True,
6805 'skip_download': True,
6806 },
6807 }, {
6808 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6809 'only_matching': True,
6810 }]
6811
6812 def _real_extract(self, url):
6813 mobj = self._match_valid_url(url)
6814 video_id = mobj.group('id')
6815 playlist_id = mobj.group('playlist_id')
6816 return self.url_result(
6817 update_url_query('https://www.youtube.com/watch', {
6818 'v': video_id,
6819 'list': playlist_id,
6820 'feature': 'youtu.be',
6821 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6822
6823
6824 class YoutubeLivestreamEmbedIE(InfoExtractor):
6825 IE_DESC = 'YouTube livestream embeds'
6826 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6827 _TESTS = [{
6828 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6829 'only_matching': True,
6830 }]
6831
6832 def _real_extract(self, url):
6833 channel_id = self._match_id(url)
6834 return self.url_result(
6835 f'https://www.youtube.com/channel/{channel_id}/live',
6836 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6837
6838
6839 class YoutubeYtUserIE(InfoExtractor):
6840 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6841 IE_NAME = 'youtube:user'
6842 _VALID_URL = r'ytuser:(?P<id>.+)'
6843 _TESTS = [{
6844 'url': 'ytuser:phihag',
6845 'only_matching': True,
6846 }]
6847
6848 def _real_extract(self, url):
6849 user_id = self._match_id(url)
6850 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6851
6852
6853 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6854 IE_NAME = 'youtube:favorites'
6855 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6856 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6857 _LOGIN_REQUIRED = True
6858 _TESTS = [{
6859 'url': ':ytfav',
6860 'only_matching': True,
6861 }, {
6862 'url': ':ytfavorites',
6863 'only_matching': True,
6864 }]
6865
6866 def _real_extract(self, url):
6867 return self.url_result(
6868 'https://www.youtube.com/playlist?list=LL',
6869 ie=YoutubeTabIE.ie_key())
6870
6871
6872 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6873 IE_NAME = 'youtube:notif'
6874 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6875 _VALID_URL = r':ytnotif(?:ication)?s?'
6876 _LOGIN_REQUIRED = True
6877 _TESTS = [{
6878 'url': ':ytnotif',
6879 'only_matching': True,
6880 }, {
6881 'url': ':ytnotifications',
6882 'only_matching': True,
6883 }]
6884
6885 def _extract_notification_menu(self, response, continuation_list):
6886 notification_list = traverse_obj(
6887 response,
6888 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6889 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6890 expected_type=list) or []
6891 continuation_list[0] = None
6892 for item in notification_list:
6893 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6894 if entry:
6895 yield entry
6896 continuation = item.get('continuationItemRenderer')
6897 if continuation:
6898 continuation_list[0] = continuation
6899
6900 def _extract_notification_renderer(self, notification):
6901 video_id = traverse_obj(
6902 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6903 url = f'https://www.youtube.com/watch?v={video_id}'
6904 channel_id = None
6905 if not video_id:
6906 browse_ep = traverse_obj(
6907 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6908 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6909 post_id = self._search_regex(
6910 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6911 'post id', default=None)
6912 if not channel_id or not post_id:
6913 return
6914 # The direct /post url redirects to this in the browser
6915 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6916
6917 channel = traverse_obj(
6918 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6919 expected_type=str)
6920 notification_title = self._get_text(notification, 'shortMessage')
6921 if notification_title:
6922 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6923 # TODO: handle recommended videos
6924 title = self._search_regex(
6925 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6926 'video title', default=None)
6927 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6928 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6929 else None)
6930 return {
6931 '_type': 'url',
6932 'url': url,
6933 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6934 'video_id': video_id,
6935 'title': title,
6936 'channel_id': channel_id,
6937 'channel': channel,
6938 'uploader': channel,
6939 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6940 'timestamp': timestamp,
6941 }
6942
6943 def _notification_menu_entries(self, ytcfg):
6944 continuation_list = [None]
6945 response = None
6946 for page in itertools.count(1):
6947 ctoken = traverse_obj(
6948 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6949 response = self._extract_response(
6950 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6951 ep='notification/get_notification_menu', check_get_keys='actions',
6952 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6953 yield from self._extract_notification_menu(response, continuation_list)
6954 if not continuation_list[0]:
6955 break
6956
6957 def _real_extract(self, url):
6958 display_id = 'notifications'
6959 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6960 self._report_playlist_authcheck(ytcfg)
6961 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6962
6963
6964 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6965 IE_DESC = 'YouTube search'
6966 IE_NAME = 'youtube:search'
6967 _SEARCH_KEY = 'ytsearch'
6968 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6969 _TESTS = [{
6970 'url': 'ytsearch5:youtube-dl test video',
6971 'playlist_count': 5,
6972 'info_dict': {
6973 'id': 'youtube-dl test video',
6974 'title': 'youtube-dl test video',
6975 }
6976 }]
6977
6978
6979 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6980 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6981 _SEARCH_KEY = 'ytsearchdate'
6982 IE_DESC = 'YouTube search, newest videos first'
6983 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6984 _TESTS = [{
6985 'url': 'ytsearchdate5:youtube-dl test video',
6986 'playlist_count': 5,
6987 'info_dict': {
6988 'id': 'youtube-dl test video',
6989 'title': 'youtube-dl test video',
6990 }
6991 }]
6992
6993
6994 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6995 IE_DESC = 'YouTube search URLs with sorting and filter support'
6996 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6997 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6998 _TESTS = [{
6999 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
7000 'playlist_mincount': 5,
7001 'info_dict': {
7002 'id': 'youtube-dl test video',
7003 'title': 'youtube-dl test video',
7004 }
7005 }, {
7006 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
7007 'playlist_mincount': 5,
7008 'info_dict': {
7009 'id': 'python',
7010 'title': 'python',
7011 }
7012 }, {
7013 'url': 'https://www.youtube.com/results?search_query=%23cats',
7014 'playlist_mincount': 1,
7015 'info_dict': {
7016 'id': '#cats',
7017 'title': '#cats',
7018 # The test suite does not have support for nested playlists
7019 # 'entries': [{
7020 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
7021 # 'title': '#cats',
7022 # }],
7023 },
7024 }, {
7025 # Channel results
7026 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
7027 'info_dict': {
7028 'id': 'kurzgesagt',
7029 'title': 'kurzgesagt',
7030 },
7031 'playlist': [{
7032 'info_dict': {
7033 '_type': 'url',
7034 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7035 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7036 'ie_key': 'YoutubeTab',
7037 'channel': 'Kurzgesagt – In a Nutshell',
7038 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
7039 'title': 'Kurzgesagt – In a Nutshell',
7040 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7041 # No longer available for search as it is set to the handle.
7042 # 'playlist_count': int,
7043 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7044 'thumbnails': list,
7045 'uploader_id': '@kurzgesagt',
7046 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
7047 'uploader': 'Kurzgesagt – In a Nutshell',
7048 'channel_is_verified': True,
7049 'channel_follower_count': int,
7050 }
7051 }],
7052 'params': {'extract_flat': True, 'playlist_items': '1'},
7053 'playlist_mincount': 1,
7054 }, {
7055 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
7056 'only_matching': True,
7057 }]
7058
7059 def _real_extract(self, url):
7060 qs = parse_qs(url)
7061 query = (qs.get('search_query') or qs.get('q'))[0]
7062 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
7063
7064
7065 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
7066 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
7067 IE_NAME = 'youtube:music:search_url'
7068 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7069 _TESTS = [{
7070 'url': 'https://music.youtube.com/search?q=royalty+free+music',
7071 'playlist_count': 16,
7072 'info_dict': {
7073 'id': 'royalty free music',
7074 'title': 'royalty free music',
7075 }
7076 }, {
7077 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7078 'playlist_mincount': 30,
7079 'info_dict': {
7080 'id': 'royalty free music - songs',
7081 'title': 'royalty free music - songs',
7082 },
7083 'params': {'extract_flat': 'in_playlist'}
7084 }, {
7085 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7086 'playlist_mincount': 30,
7087 'info_dict': {
7088 'id': 'royalty free music - community playlists',
7089 'title': 'royalty free music - community playlists',
7090 },
7091 'params': {'extract_flat': 'in_playlist'}
7092 }]
7093
7094 _SECTIONS = {
7095 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7096 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7097 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7098 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7099 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7100 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7101 }
7102
7103 def _real_extract(self, url):
7104 qs = parse_qs(url)
7105 query = (qs.get('search_query') or qs.get('q'))[0]
7106 params = qs.get('sp', (None,))[0]
7107 if params:
7108 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7109 else:
7110 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
7111 params = self._SECTIONS.get(section)
7112 if not params:
7113 section = None
7114 title = join_nonempty(query, section, delim=' - ')
7115 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7116
7117
7118 class YoutubeFeedsInfoExtractor(InfoExtractor):
7119 """
7120 Base class for feed extractors
7121 Subclasses must re-define the _FEED_NAME property.
7122 """
7123 _LOGIN_REQUIRED = True
7124 _FEED_NAME = 'feeds'
7125
7126 def _real_initialize(self):
7127 YoutubeBaseInfoExtractor._check_login_required(self)
7128
7129 @classproperty
7130 def IE_NAME(self):
7131 return f'youtube:{self._FEED_NAME}'
7132
7133 def _real_extract(self, url):
7134 return self.url_result(
7135 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7136
7137
7138 class YoutubeWatchLaterIE(InfoExtractor):
7139 IE_NAME = 'youtube:watchlater'
7140 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7141 _VALID_URL = r':ytwatchlater'
7142 _TESTS = [{
7143 'url': ':ytwatchlater',
7144 'only_matching': True,
7145 }]
7146
7147 def _real_extract(self, url):
7148 return self.url_result(
7149 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7150
7151
7152 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7153 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7154 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7155 _FEED_NAME = 'recommended'
7156 _LOGIN_REQUIRED = False
7157 _TESTS = [{
7158 'url': ':ytrec',
7159 'only_matching': True,
7160 }, {
7161 'url': ':ytrecommended',
7162 'only_matching': True,
7163 }, {
7164 'url': 'https://youtube.com',
7165 'only_matching': True,
7166 }]
7167
7168
7169 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7170 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7171 _VALID_URL = r':ytsub(?:scription)?s?'
7172 _FEED_NAME = 'subscriptions'
7173 _TESTS = [{
7174 'url': ':ytsubs',
7175 'only_matching': True,
7176 }, {
7177 'url': ':ytsubscriptions',
7178 'only_matching': True,
7179 }]
7180
7181
7182 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7183 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7184 _VALID_URL = r':ythis(?:tory)?'
7185 _FEED_NAME = 'history'
7186 _TESTS = [{
7187 'url': ':ythistory',
7188 'only_matching': True,
7189 }]
7190
7191
7192 class YoutubeShortsAudioPivotIE(InfoExtractor):
7193 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7194 IE_NAME = 'youtube:shorts:pivot:audio'
7195 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7196 _TESTS = [{
7197 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7198 'only_matching': True,
7199 }]
7200
7201 @staticmethod
7202 def _generate_audio_pivot_params(video_id):
7203 """
7204 Generates sfv_audio_pivot browse params for this video id
7205 """
7206 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7207 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7208
7209 def _real_extract(self, url):
7210 video_id = self._match_id(url)
7211 return self.url_result(
7212 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7213 ie=YoutubeTabIE)
7214
7215
7216 class YoutubeTruncatedURLIE(InfoExtractor):
7217 IE_NAME = 'youtube:truncated_url'
7218 IE_DESC = False # Do not list
7219 _VALID_URL = r'''(?x)
7220 (?:https?://)?
7221 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7222 (?:watch\?(?:
7223 feature=[a-z_]+|
7224 annotation_id=annotation_[^&]+|
7225 x-yt-cl=[0-9]+|
7226 hl=[^&]*|
7227 t=[0-9]+
7228 )?
7229 |
7230 attribution_link\?a=[^&]+
7231 )
7232 $
7233 '''
7234
7235 _TESTS = [{
7236 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7237 'only_matching': True,
7238 }, {
7239 'url': 'https://www.youtube.com/watch?',
7240 'only_matching': True,
7241 }, {
7242 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7243 'only_matching': True,
7244 }, {
7245 'url': 'https://www.youtube.com/watch?feature=foo',
7246 'only_matching': True,
7247 }, {
7248 'url': 'https://www.youtube.com/watch?hl=en-GB',
7249 'only_matching': True,
7250 }, {
7251 'url': 'https://www.youtube.com/watch?t=2372',
7252 'only_matching': True,
7253 }]
7254
7255 def _real_extract(self, url):
7256 raise ExtractorError(
7257 'Did you forget to quote the URL? Remember that & is a meta '
7258 'character in most shells, so you want to put the URL in quotes, '
7259 'like youtube-dl '
7260 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7261 ' or simply youtube-dl BaW_jenozKc .',
7262 expected=True)
7263
7264
7265 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7266 IE_NAME = 'youtube:clip'
7267 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7268 _TESTS = [{
7269 # FIXME: Other metadata should be extracted from the clip, not from the base video
7270 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7271 'info_dict': {
7272 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7273 'ext': 'mp4',
7274 'section_start': 29.0,
7275 'section_end': 39.7,
7276 'duration': 10.7,
7277 'age_limit': 0,
7278 'availability': 'public',
7279 'categories': ['Gaming'],
7280 'channel': 'Scott The Woz',
7281 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7282 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7283 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7284 'like_count': int,
7285 'playable_in_embed': True,
7286 'tags': 'count:17',
7287 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7288 'title': 'Mobile Games on Console - Scott The Woz',
7289 'upload_date': '20210920',
7290 'uploader': 'Scott The Woz',
7291 'uploader_id': '@ScottTheWoz',
7292 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7293 'view_count': int,
7294 'live_status': 'not_live',
7295 'channel_follower_count': int,
7296 'chapters': 'count:20',
7297 'comment_count': int,
7298 'heatmap': 'count:100',
7299 }
7300 }]
7301
7302 def _real_extract(self, url):
7303 clip_id = self._match_id(url)
7304 _, data = self._extract_webpage(url, clip_id)
7305
7306 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7307 if not video_id:
7308 raise ExtractorError('Unable to find video ID')
7309
7310 clip_data = traverse_obj(data, (
7311 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7312 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7313 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7314 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7315
7316 return {
7317 '_type': 'url_transparent',
7318 'url': f'https://www.youtube.com/watch?v={video_id}',
7319 'ie_key': YoutubeIE.ie_key(),
7320 'id': clip_id,
7321 'section_start': int(clip_data['startTimeMs']) / 1000,
7322 'section_end': int(clip_data['endTimeMs']) / 1000,
7323 }
7324
7325
7326 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7327 IE_NAME = 'youtube:consent'
7328 IE_DESC = False # Do not list
7329 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7330 _TESTS = [{
7331 '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',
7332 'info_dict': {
7333 'id': 'qVv6vCqciTM',
7334 'ext': 'mp4',
7335 'age_limit': 0,
7336 'uploader_id': '@sana_natori',
7337 'comment_count': int,
7338 'chapters': 'count:13',
7339 'upload_date': '20221223',
7340 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7341 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7342 'uploader_url': 'https://www.youtube.com/@sana_natori',
7343 'like_count': int,
7344 'release_date': '20221223',
7345 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7346 '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】',
7347 'view_count': int,
7348 'playable_in_embed': True,
7349 'duration': 4438,
7350 'availability': 'public',
7351 'channel_follower_count': int,
7352 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7353 'categories': ['Entertainment'],
7354 'live_status': 'was_live',
7355 'release_timestamp': 1671793345,
7356 'channel': 'さなちゃんねる',
7357 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7358 'uploader': 'さなちゃんねる',
7359 'channel_is_verified': True,
7360 'heatmap': 'count:100',
7361 },
7362 'add_ie': ['Youtube'],
7363 'params': {'skip_download': 'Youtube'},
7364 }]
7365
7366 def _real_extract(self, url):
7367 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7368 if not redirect_url:
7369 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7370 return self.url_result(redirect_url)
7371
7372
7373 class YoutubeTruncatedIDIE(InfoExtractor):
7374 IE_NAME = 'youtube:truncated_id'
7375 IE_DESC = False # Do not list
7376 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7377
7378 _TESTS = [{
7379 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7380 'only_matching': True,
7381 }]
7382
7383 def _real_extract(self, url):
7384 video_id = self._match_id(url)
7385 raise ExtractorError(
7386 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7387 expected=True)