]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[cleanup] Misc (#8598)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import collections
4 import copy
5 import datetime
6 import enum
7 import hashlib
8 import itertools
9 import json
10 import math
11 import os.path
12 import random
13 import re
14 import sys
15 import threading
16 import time
17 import traceback
18 import urllib.parse
19
20 from .common import InfoExtractor, SearchInfoExtractor
21 from .openload import PhantomJSwrapper
22 from ..compat import functools
23 from ..jsinterp import JSInterpreter
24 from ..networking.exceptions import HTTPError, network_exceptions
25 from ..utils import (
26 NO_DEFAULT,
27 ExtractorError,
28 LazyList,
29 UserNotLive,
30 bug_reports_message,
31 classproperty,
32 clean_html,
33 datetime_from_str,
34 dict_get,
35 filter_dict,
36 float_or_none,
37 format_field,
38 get_first,
39 int_or_none,
40 is_html,
41 join_nonempty,
42 js_to_json,
43 mimetype2ext,
44 orderedSet,
45 parse_codecs,
46 parse_count,
47 parse_duration,
48 parse_iso8601,
49 parse_qs,
50 qualities,
51 remove_start,
52 smuggle_url,
53 str_or_none,
54 str_to_int,
55 strftime_or_none,
56 traverse_obj,
57 try_get,
58 unescapeHTML,
59 unified_strdate,
60 unified_timestamp,
61 unsmuggle_url,
62 update_url_query,
63 url_or_none,
64 urljoin,
65 variadic,
66 )
67
68 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
69 # any clients starting with _ cannot be explicitly requested by the user
70 INNERTUBE_CLIENTS = {
71 'web': {
72 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
73 'INNERTUBE_CONTEXT': {
74 'client': {
75 'clientName': 'WEB',
76 'clientVersion': '2.20220801.00.00',
77 }
78 },
79 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
80 },
81 'web_embedded': {
82 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
83 'INNERTUBE_CONTEXT': {
84 'client': {
85 'clientName': 'WEB_EMBEDDED_PLAYER',
86 'clientVersion': '1.20220731.00.00',
87 },
88 },
89 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
90 },
91 'web_music': {
92 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
93 'INNERTUBE_HOST': 'music.youtube.com',
94 'INNERTUBE_CONTEXT': {
95 'client': {
96 'clientName': 'WEB_REMIX',
97 'clientVersion': '1.20220727.01.00',
98 }
99 },
100 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
101 },
102 'web_creator': {
103 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
104 'INNERTUBE_CONTEXT': {
105 'client': {
106 'clientName': 'WEB_CREATOR',
107 'clientVersion': '1.20220726.00.00',
108 }
109 },
110 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
111 },
112 'android': {
113 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
114 'INNERTUBE_CONTEXT': {
115 'client': {
116 'clientName': 'ANDROID',
117 'clientVersion': '17.31.35',
118 'androidSdkVersion': 30,
119 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
120 }
121 },
122 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
123 'REQUIRE_JS_PLAYER': False
124 },
125 'android_embedded': {
126 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
127 'INNERTUBE_CONTEXT': {
128 'client': {
129 'clientName': 'ANDROID_EMBEDDED_PLAYER',
130 'clientVersion': '17.31.35',
131 'androidSdkVersion': 30,
132 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
133 },
134 },
135 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
136 'REQUIRE_JS_PLAYER': False
137 },
138 'android_music': {
139 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
140 'INNERTUBE_CONTEXT': {
141 'client': {
142 'clientName': 'ANDROID_MUSIC',
143 'clientVersion': '5.16.51',
144 'androidSdkVersion': 30,
145 'userAgent': 'com.google.android.apps.youtube.music/5.16.51 (Linux; U; Android 11) gzip'
146 }
147 },
148 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
149 'REQUIRE_JS_PLAYER': False
150 },
151 'android_creator': {
152 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
153 'INNERTUBE_CONTEXT': {
154 'client': {
155 'clientName': 'ANDROID_CREATOR',
156 'clientVersion': '22.30.100',
157 'androidSdkVersion': 30,
158 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip'
159 },
160 },
161 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
162 'REQUIRE_JS_PLAYER': False
163 },
164 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
165 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
166 'ios': {
167 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
168 'INNERTUBE_CONTEXT': {
169 'client': {
170 'clientName': 'IOS',
171 'clientVersion': '17.33.2',
172 'deviceModel': 'iPhone14,3',
173 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
174 }
175 },
176 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
177 'REQUIRE_JS_PLAYER': False
178 },
179 'ios_embedded': {
180 'INNERTUBE_CONTEXT': {
181 'client': {
182 'clientName': 'IOS_MESSAGES_EXTENSION',
183 'clientVersion': '17.33.2',
184 'deviceModel': 'iPhone14,3',
185 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
186 },
187 },
188 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
189 'REQUIRE_JS_PLAYER': False
190 },
191 'ios_music': {
192 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
193 'INNERTUBE_CONTEXT': {
194 'client': {
195 'clientName': 'IOS_MUSIC',
196 'clientVersion': '5.21',
197 'deviceModel': 'iPhone14,3',
198 'userAgent': 'com.google.ios.youtubemusic/5.21 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
199 },
200 },
201 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
202 'REQUIRE_JS_PLAYER': False
203 },
204 'ios_creator': {
205 'INNERTUBE_CONTEXT': {
206 'client': {
207 'clientName': 'IOS_CREATOR',
208 'clientVersion': '22.33.101',
209 'deviceModel': 'iPhone14,3',
210 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
211 },
212 },
213 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
214 'REQUIRE_JS_PLAYER': False
215 },
216 # mweb has 'ultralow' formats
217 # See: https://github.com/yt-dlp/yt-dlp/pull/557
218 'mweb': {
219 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
220 'INNERTUBE_CONTEXT': {
221 'client': {
222 'clientName': 'MWEB',
223 'clientVersion': '2.20220801.00.00',
224 }
225 },
226 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
227 },
228 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
229 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
230 'tv_embedded': {
231 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
232 'INNERTUBE_CONTEXT': {
233 'client': {
234 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
235 'clientVersion': '2.0',
236 },
237 },
238 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
239 },
240 }
241
242
243 def _split_innertube_client(client_name):
244 variant, *base = client_name.rsplit('.', 1)
245 if base:
246 return variant, base[0], variant
247 base, *variant = client_name.split('_', 1)
248 return client_name, base, variant[0] if variant else None
249
250
251 def short_client_name(client_name):
252 main, *parts = _split_innertube_client(client_name)[0].replace('embedscreen', 'e_s').split('_')
253 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
254
255
256 def build_innertube_clients():
257 THIRD_PARTY = {
258 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
259 }
260 BASE_CLIENTS = ('ios', 'android', 'web', 'tv', 'mweb')
261 priority = qualities(BASE_CLIENTS[::-1])
262
263 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
264 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
265 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
266 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
267 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
268
269 _, base_client, variant = _split_innertube_client(client)
270 ytcfg['priority'] = 10 * priority(base_client)
271
272 if not variant:
273 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
274 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
275 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
276 embedscreen['priority'] -= 3
277 elif variant == 'embedded':
278 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
279 ytcfg['priority'] -= 2
280 else:
281 ytcfg['priority'] -= 3
282
283
284 build_innertube_clients()
285
286
287 class BadgeType(enum.Enum):
288 AVAILABILITY_UNLISTED = enum.auto()
289 AVAILABILITY_PRIVATE = enum.auto()
290 AVAILABILITY_PUBLIC = enum.auto()
291 AVAILABILITY_PREMIUM = enum.auto()
292 AVAILABILITY_SUBSCRIPTION = enum.auto()
293 LIVE_NOW = enum.auto()
294 VERIFIED = enum.auto()
295
296
297 class YoutubeBaseInfoExtractor(InfoExtractor):
298 """Provide base functions for Youtube extractors"""
299
300 _RESERVED_NAMES = (
301 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
302 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
303 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
304 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
305
306 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
307
308 # _NETRC_MACHINE = 'youtube'
309
310 # If True it will raise an error if no login info is provided
311 _LOGIN_REQUIRED = False
312
313 _INVIDIOUS_SITES = (
314 # invidious-redirect websites
315 r'(?:www\.)?redirect\.invidious\.io',
316 r'(?:(?:www|dev)\.)?invidio\.us',
317 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
318 r'(?:www\.)?invidious\.pussthecat\.org',
319 r'(?:www\.)?invidious\.zee\.li',
320 r'(?:www\.)?invidious\.ethibox\.fr',
321 r'(?:www\.)?iv\.ggtyler\.dev',
322 r'(?:www\.)?inv\.vern\.i2p',
323 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
324 r'(?:www\.)?inv\.riverside\.rocks',
325 r'(?:www\.)?invidious\.silur\.me',
326 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
327 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
328 r'(?:www\.)?invidious\.slipfox\.xyz',
329 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
330 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
331 r'(?:www\.)?invidious\.tiekoetter\.com',
332 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
333 r'(?:www\.)?invidious\.nerdvpn\.de',
334 r'(?:www\.)?invidious\.weblibre\.org',
335 r'(?:www\.)?inv\.odyssey346\.dev',
336 r'(?:www\.)?invidious\.dhusch\.de',
337 r'(?:www\.)?iv\.melmac\.space',
338 r'(?:www\.)?watch\.thekitty\.zone',
339 r'(?:www\.)?invidious\.privacydev\.net',
340 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
341 r'(?:www\.)?invidious\.drivet\.xyz',
342 r'(?:www\.)?vid\.priv\.au',
343 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
344 r'(?:www\.)?inv\.vern\.cc',
345 r'(?:www\.)?invidious\.esmailelbob\.xyz',
346 r'(?:www\.)?invidious\.sethforprivacy\.com',
347 r'(?:www\.)?yt\.oelrichsgarcia\.de',
348 r'(?:www\.)?yt\.artemislena\.eu',
349 r'(?:www\.)?invidious\.flokinet\.to',
350 r'(?:www\.)?invidious\.baczek\.me',
351 r'(?:www\.)?y\.com\.sb',
352 r'(?:www\.)?invidious\.epicsite\.xyz',
353 r'(?:www\.)?invidious\.lidarshield\.cloud',
354 r'(?:www\.)?yt\.funami\.tech',
355 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
356 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
357 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
358 # youtube-dl invidious instances list
359 r'(?:(?:www|no)\.)?invidiou\.sh',
360 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
361 r'(?:www\.)?invidious\.kabi\.tk',
362 r'(?:www\.)?invidious\.mastodon\.host',
363 r'(?:www\.)?invidious\.zapashcanon\.fr',
364 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
365 r'(?:www\.)?invidious\.tinfoil-hat\.net',
366 r'(?:www\.)?invidious\.himiko\.cloud',
367 r'(?:www\.)?invidious\.reallyancient\.tech',
368 r'(?:www\.)?invidious\.tube',
369 r'(?:www\.)?invidiou\.site',
370 r'(?:www\.)?invidious\.site',
371 r'(?:www\.)?invidious\.xyz',
372 r'(?:www\.)?invidious\.nixnet\.xyz',
373 r'(?:www\.)?invidious\.048596\.xyz',
374 r'(?:www\.)?invidious\.drycat\.fr',
375 r'(?:www\.)?inv\.skyn3t\.in',
376 r'(?:www\.)?tube\.poal\.co',
377 r'(?:www\.)?tube\.connect\.cafe',
378 r'(?:www\.)?vid\.wxzm\.sx',
379 r'(?:www\.)?vid\.mint\.lgbt',
380 r'(?:www\.)?vid\.puffyan\.us',
381 r'(?:www\.)?yewtu\.be',
382 r'(?:www\.)?yt\.elukerio\.org',
383 r'(?:www\.)?yt\.lelux\.fi',
384 r'(?:www\.)?invidious\.ggc-project\.de',
385 r'(?:www\.)?yt\.maisputain\.ovh',
386 r'(?:www\.)?ytprivate\.com',
387 r'(?:www\.)?invidious\.13ad\.de',
388 r'(?:www\.)?invidious\.toot\.koeln',
389 r'(?:www\.)?invidious\.fdn\.fr',
390 r'(?:www\.)?watch\.nettohikari\.com',
391 r'(?:www\.)?invidious\.namazso\.eu',
392 r'(?:www\.)?invidious\.silkky\.cloud',
393 r'(?:www\.)?invidious\.exonip\.de',
394 r'(?:www\.)?invidious\.riverside\.rocks',
395 r'(?:www\.)?invidious\.blamefran\.net',
396 r'(?:www\.)?invidious\.moomoo\.de',
397 r'(?:www\.)?ytb\.trom\.tf',
398 r'(?:www\.)?yt\.cyberhost\.uk',
399 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
400 r'(?:www\.)?qklhadlycap4cnod\.onion',
401 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
402 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
403 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
404 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
405 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
406 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
407 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
408 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
409 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
410 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
411 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
412 r'(?:www\.)?piped\.kavin\.rocks',
413 r'(?:www\.)?piped\.tokhmi\.xyz',
414 r'(?:www\.)?piped\.syncpundit\.io',
415 r'(?:www\.)?piped\.mha\.fi',
416 r'(?:www\.)?watch\.whatever\.social',
417 r'(?:www\.)?piped\.garudalinux\.org',
418 r'(?:www\.)?piped\.rivo\.lol',
419 r'(?:www\.)?piped-libre\.kavin\.rocks',
420 r'(?:www\.)?yt\.jae\.fi',
421 r'(?:www\.)?piped\.mint\.lgbt',
422 r'(?:www\.)?il\.ax',
423 r'(?:www\.)?piped\.esmailelbob\.xyz',
424 r'(?:www\.)?piped\.projectsegfau\.lt',
425 r'(?:www\.)?piped\.privacydev\.net',
426 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
427 r'(?:www\.)?piped\.smnz\.de',
428 r'(?:www\.)?piped\.adminforge\.de',
429 r'(?:www\.)?watch\.whatevertinfoil\.de',
430 r'(?:www\.)?piped\.qdi\.fi',
431 r'(?:(?:www|cf)\.)?piped\.video',
432 r'(?:www\.)?piped\.aeong\.one',
433 r'(?:www\.)?piped\.moomoo\.me',
434 r'(?:www\.)?piped\.chauvet\.pro',
435 r'(?:www\.)?watch\.leptons\.xyz',
436 r'(?:www\.)?pd\.vern\.cc',
437 r'(?:www\.)?piped\.hostux\.net',
438 r'(?:www\.)?piped\.lunar\.icu',
439 # Hyperpipe instances from https://hyperpipe.codeberg.page/
440 r'(?:www\.)?hyperpipe\.surge\.sh',
441 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
442 r'(?:www\.)?listen\.whatever\.social',
443 r'(?:www\.)?music\.adminforge\.de',
444 )
445
446 # extracted from account/account_menu ep
447 # XXX: These are the supported YouTube UI and API languages,
448 # which is slightly different from languages supported for translation in YouTube studio
449 _SUPPORTED_LANG_CODES = [
450 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
451 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
452 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
453 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
454 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
455 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
456 ]
457
458 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
459
460 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
461 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
462
463 def ucid_or_none(self, ucid):
464 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
465
466 def handle_or_none(self, handle):
467 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
468
469 def handle_from_url(self, url):
470 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
471 url, 'channel handle', default=None)
472
473 def ucid_from_url(self, url):
474 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
475 url, 'channel id', default=None)
476
477 @functools.cached_property
478 def _preferred_lang(self):
479 """
480 Returns a language code supported by YouTube for the user preferred language.
481 Returns None if no preferred language set.
482 """
483 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
484 if not preferred_lang:
485 return
486 if preferred_lang not in self._SUPPORTED_LANG_CODES:
487 raise ExtractorError(
488 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
489 expected=True)
490 elif preferred_lang != 'en':
491 self.report_warning(
492 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
493 return preferred_lang
494
495 def _initialize_consent(self):
496 cookies = self._get_cookies('https://www.youtube.com/')
497 if cookies.get('__Secure-3PSID'):
498 return
499 socs = cookies.get('SOCS')
500 if socs and not socs.value.startswith('CAA'): # not consented
501 return
502 self._set_cookie('.youtube.com', 'SOCS', 'CAI', secure=True) # accept all (required for mixes)
503
504 def _initialize_pref(self):
505 cookies = self._get_cookies('https://www.youtube.com/')
506 pref_cookie = cookies.get('PREF')
507 pref = {}
508 if pref_cookie:
509 try:
510 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
511 except ValueError:
512 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
513 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
514 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
515
516 def _real_initialize(self):
517 self._initialize_pref()
518 self._initialize_consent()
519 self._check_login_required()
520
521 def _check_login_required(self):
522 if self._LOGIN_REQUIRED and not self._cookies_passed:
523 self.raise_login_required('Login details are needed to download this content', method='cookies')
524
525 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
526 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
527
528 def _get_default_ytcfg(self, client='web'):
529 return copy.deepcopy(INNERTUBE_CLIENTS[client])
530
531 def _get_innertube_host(self, client='web'):
532 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
533
534 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
535 # try_get but with fallback to default ytcfg client values when present
536 _func = lambda y: try_get(y, getter, expected_type)
537 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
538
539 def _extract_client_name(self, ytcfg, default_client='web'):
540 return self._ytcfg_get_safe(
541 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
542 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
543
544 def _extract_client_version(self, ytcfg, default_client='web'):
545 return self._ytcfg_get_safe(
546 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
547 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
548
549 def _select_api_hostname(self, req_api_hostname, default_client=None):
550 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
551 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
552
553 def _extract_api_key(self, ytcfg=None, default_client='web'):
554 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
555
556 def _extract_context(self, ytcfg=None, default_client='web'):
557 context = get_first(
558 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
559 # Enforce language and tz for extraction
560 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
561 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
562 return context
563
564 _SAPISID = None
565
566 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
567 time_now = round(time.time())
568 if self._SAPISID is None:
569 yt_cookies = self._get_cookies('https://www.youtube.com')
570 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
571 # See: https://github.com/yt-dlp/yt-dlp/issues/393
572 sapisid_cookie = dict_get(
573 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
574 if sapisid_cookie and sapisid_cookie.value:
575 self._SAPISID = sapisid_cookie.value
576 self.write_debug('Extracted SAPISID cookie')
577 # SAPISID cookie is required if not already present
578 if not yt_cookies.get('SAPISID'):
579 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
580 self._set_cookie(
581 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
582 else:
583 self._SAPISID = False
584 if not self._SAPISID:
585 return None
586 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
587 sapisidhash = hashlib.sha1(
588 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
589 return f'SAPISIDHASH {time_now}_{sapisidhash}'
590
591 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
592 note='Downloading API JSON', errnote='Unable to download API page',
593 context=None, api_key=None, api_hostname=None, default_client='web'):
594
595 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
596 data.update(query)
597 real_headers = self.generate_api_headers(default_client=default_client)
598 real_headers.update({'content-type': 'application/json'})
599 if headers:
600 real_headers.update(headers)
601 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
602 or api_key or self._extract_api_key(default_client=default_client))
603 return self._download_json(
604 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
605 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
606 data=json.dumps(data).encode('utf8'), headers=real_headers,
607 query={'key': api_key, 'prettyPrint': 'false'})
608
609 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
610 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
611
612 @staticmethod
613 def _extract_session_index(*data):
614 """
615 Index of current account in account list.
616 See: https://github.com/yt-dlp/yt-dlp/pull/519
617 """
618 for ytcfg in data:
619 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
620 if session_index is not None:
621 return session_index
622
623 # Deprecated?
624 def _extract_identity_token(self, ytcfg=None, webpage=None):
625 if ytcfg:
626 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
627 if token:
628 return token
629 if webpage:
630 return self._search_regex(
631 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
632 'identity token', default=None, fatal=False)
633
634 @staticmethod
635 def _extract_account_syncid(*args):
636 """
637 Extract syncId required to download private playlists of secondary channels
638 @params response and/or ytcfg
639 """
640 for data in args:
641 # ytcfg includes channel_syncid if on secondary channel
642 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
643 if delegated_sid:
644 return delegated_sid
645 sync_ids = (try_get(
646 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
647 lambda x: x['DATASYNC_ID']), str) or '').split('||')
648 if len(sync_ids) >= 2 and sync_ids[1]:
649 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
650 # and just "user_syncid||" for primary channel. We only want the channel_syncid
651 return sync_ids[0]
652
653 @staticmethod
654 def _extract_visitor_data(*args):
655 """
656 Extracts visitorData from an API response or ytcfg
657 Appears to be used to track session state
658 """
659 return get_first(
660 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
661 expected_type=str)
662
663 @functools.cached_property
664 def is_authenticated(self):
665 return bool(self._generate_sapisidhash_header())
666
667 def extract_ytcfg(self, video_id, webpage):
668 if not webpage:
669 return {}
670 return self._parse_json(
671 self._search_regex(
672 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
673 default='{}'), video_id, fatal=False) or {}
674
675 def generate_api_headers(
676 self, *, ytcfg=None, account_syncid=None, session_index=None,
677 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
678
679 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
680 headers = {
681 'X-YouTube-Client-Name': str(
682 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
683 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
684 'Origin': origin,
685 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
686 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
687 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
688 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
689 }
690 if session_index is None:
691 session_index = self._extract_session_index(ytcfg)
692 if account_syncid or session_index is not None:
693 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
694
695 auth = self._generate_sapisidhash_header(origin)
696 if auth is not None:
697 headers['Authorization'] = auth
698 headers['X-Origin'] = origin
699 return filter_dict(headers)
700
701 def _download_ytcfg(self, client, video_id):
702 url = {
703 'web': 'https://www.youtube.com',
704 'web_music': 'https://music.youtube.com',
705 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
706 }.get(client)
707 if not url:
708 return {}
709 webpage = self._download_webpage(
710 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
711 return self.extract_ytcfg(video_id, webpage) or {}
712
713 @staticmethod
714 def _build_api_continuation_query(continuation, ctp=None):
715 query = {
716 'continuation': continuation
717 }
718 # TODO: Inconsistency with clickTrackingParams.
719 # Currently we have a fixed ctp contained within context (from ytcfg)
720 # and a ctp in root query for continuation.
721 if ctp:
722 query['clickTracking'] = {'clickTrackingParams': ctp}
723 return query
724
725 @classmethod
726 def _extract_next_continuation_data(cls, renderer):
727 next_continuation = try_get(
728 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
729 lambda x: x['continuation']['reloadContinuationData']), dict)
730 if not next_continuation:
731 return
732 continuation = next_continuation.get('continuation')
733 if not continuation:
734 return
735 ctp = next_continuation.get('clickTrackingParams')
736 return cls._build_api_continuation_query(continuation, ctp)
737
738 @classmethod
739 def _extract_continuation_ep_data(cls, continuation_ep: dict):
740 if isinstance(continuation_ep, dict):
741 continuation = try_get(
742 continuation_ep, lambda x: x['continuationCommand']['token'], str)
743 if not continuation:
744 return
745 ctp = continuation_ep.get('clickTrackingParams')
746 return cls._build_api_continuation_query(continuation, ctp)
747
748 @classmethod
749 def _extract_continuation(cls, renderer):
750 next_continuation = cls._extract_next_continuation_data(renderer)
751 if next_continuation:
752 return next_continuation
753
754 return traverse_obj(renderer, (
755 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
756 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
757 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
758
759 @classmethod
760 def _extract_alerts(cls, data):
761 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
762 if not isinstance(alert_dict, dict):
763 continue
764 for alert in alert_dict.values():
765 alert_type = alert.get('type')
766 if not alert_type:
767 continue
768 message = cls._get_text(alert, 'text')
769 if message:
770 yield alert_type, message
771
772 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
773 errors, warnings = [], []
774 for alert_type, alert_message in alerts:
775 if alert_type.lower() == 'error' and fatal:
776 errors.append([alert_type, alert_message])
777 elif alert_message not in self._IGNORED_WARNINGS:
778 warnings.append([alert_type, alert_message])
779
780 for alert_type, alert_message in (warnings + errors[:-1]):
781 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
782 if errors:
783 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
784
785 def _extract_and_report_alerts(self, data, *args, **kwargs):
786 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
787
788 def _extract_badges(self, badge_list: list):
789 """
790 Extract known BadgeType's from a list of badge renderers.
791 @returns [{'type': BadgeType}]
792 """
793 icon_type_map = {
794 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
795 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
796 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC,
797 'CHECK_CIRCLE_THICK': BadgeType.VERIFIED,
798 'OFFICIAL_ARTIST_BADGE': BadgeType.VERIFIED,
799 'CHECK': BadgeType.VERIFIED,
800 }
801
802 badge_style_map = {
803 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
804 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
805 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
806 'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
807 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED,
808 }
809
810 label_map = {
811 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
812 'private': BadgeType.AVAILABILITY_PRIVATE,
813 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
814 'live': BadgeType.LIVE_NOW,
815 'premium': BadgeType.AVAILABILITY_PREMIUM,
816 'verified': BadgeType.VERIFIED,
817 'official artist channel': BadgeType.VERIFIED,
818 }
819
820 badges = []
821 for badge in traverse_obj(badge_list, (..., lambda key, _: re.search(r'[bB]adgeRenderer$', key))):
822 badge_type = (
823 icon_type_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
824 or badge_style_map.get(traverse_obj(badge, 'style'))
825 )
826 if badge_type:
827 badges.append({'type': badge_type})
828 continue
829
830 # fallback, won't work in some languages
831 label = traverse_obj(
832 badge, 'label', ('accessibilityData', 'label'), 'tooltip', 'iconTooltip', get_all=False, expected_type=str, default='')
833 for match, label_badge_type in label_map.items():
834 if match in label.lower():
835 badges.append({'type': label_badge_type})
836 break
837
838 return badges
839
840 @staticmethod
841 def _has_badge(badges, badge_type):
842 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
843
844 @staticmethod
845 def _get_text(data, *path_list, max_runs=None):
846 for path in path_list or [None]:
847 if path is None:
848 obj = [data]
849 else:
850 obj = traverse_obj(data, path, default=[])
851 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
852 obj = [obj]
853 for item in obj:
854 text = try_get(item, lambda x: x['simpleText'], str)
855 if text:
856 return text
857 runs = try_get(item, lambda x: x['runs'], list) or []
858 if not runs and isinstance(item, list):
859 runs = item
860
861 runs = runs[:min(len(runs), max_runs or len(runs))]
862 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
863 if text:
864 return text
865
866 def _get_count(self, data, *path_list):
867 count_text = self._get_text(data, *path_list) or ''
868 count = parse_count(count_text)
869 if count is None:
870 count = str_to_int(
871 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
872 return count
873
874 @staticmethod
875 def _extract_thumbnails(data, *path_list):
876 """
877 Extract thumbnails from thumbnails dict
878 @param path_list: path list to level that contains 'thumbnails' key
879 """
880 thumbnails = []
881 for path in path_list or [()]:
882 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...)):
883 thumbnail_url = url_or_none(thumbnail.get('url'))
884 if not thumbnail_url:
885 continue
886 # Sometimes youtube gives a wrong thumbnail URL. See:
887 # https://github.com/yt-dlp/yt-dlp/issues/233
888 # https://github.com/ytdl-org/youtube-dl/issues/28023
889 if 'maxresdefault' in thumbnail_url:
890 thumbnail_url = thumbnail_url.split('?')[0]
891 thumbnails.append({
892 'url': thumbnail_url,
893 'height': int_or_none(thumbnail.get('height')),
894 'width': int_or_none(thumbnail.get('width')),
895 })
896 return thumbnails
897
898 @staticmethod
899 def extract_relative_time(relative_time_text):
900 """
901 Extracts a relative time from string and converts to dt object
902 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today', '8 yr ago'
903 """
904
905 # XXX: this could be moved to a general function in utils/_utils.py
906 # The relative time text strings are roughly the same as what
907 # Javascript's Intl.RelativeTimeFormat function generates.
908 # See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat
909 mobj = re.search(
910 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',
911 relative_time_text)
912 if mobj:
913 start = mobj.group('start')
914 if start:
915 return datetime_from_str(start)
916 try:
917 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
918 except ValueError:
919 return None
920
921 def _parse_time_text(self, text):
922 if not text:
923 return
924 dt = self.extract_relative_time(text)
925 timestamp = None
926 if isinstance(dt, datetime.datetime):
927 timestamp = calendar.timegm(dt.timetuple())
928
929 if timestamp is None:
930 timestamp = (
931 unified_timestamp(text) or unified_timestamp(
932 self._search_regex(
933 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
934 text.lower(), 'time text', default=None)))
935
936 if text and timestamp is None and self._preferred_lang in (None, 'en'):
937 self.report_warning(
938 f'Cannot parse localized time text "{text}"', only_once=True)
939 return timestamp
940
941 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
942 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
943 default_client='web'):
944 raise_for_incomplete = bool(self._configuration_arg('raise_incomplete_data', ie_key=YoutubeIE))
945 # Incomplete Data should be a warning by default when retries are exhausted, while other errors should be fatal.
946 icd_retries = iter(self.RetryManager(fatal=raise_for_incomplete))
947 icd_rm = next(icd_retries)
948 main_retries = iter(self.RetryManager())
949 main_rm = next(main_retries)
950 # Manual retry loop for multiple RetryManagers
951 # The proper RetryManager MUST be advanced after an error
952 # and its result MUST be checked if the manager is non fatal
953 while True:
954 try:
955 response = self._call_api(
956 ep=ep, fatal=True, headers=headers,
957 video_id=item_id, query=query, note=note,
958 context=self._extract_context(ytcfg, default_client),
959 api_key=self._extract_api_key(ytcfg, default_client),
960 api_hostname=api_hostname, default_client=default_client)
961 except ExtractorError as e:
962 if not isinstance(e.cause, network_exceptions):
963 return self._error_or_warning(e, fatal=fatal)
964 elif not isinstance(e.cause, HTTPError):
965 main_rm.error = e
966 next(main_retries)
967 continue
968
969 first_bytes = e.cause.response.read(512)
970 if not is_html(first_bytes):
971 yt_error = try_get(
972 self._parse_json(
973 self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
974 lambda x: x['error']['message'], str)
975 if yt_error:
976 self._report_alerts([('ERROR', yt_error)], fatal=False)
977 # Downloading page may result in intermittent 5xx HTTP error
978 # Sometimes a 404 is also received. See: https://github.com/ytdl-org/youtube-dl/issues/28289
979 # We also want to catch all other network exceptions since errors in later pages can be troublesome
980 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
981 if e.cause.status not in (403, 429):
982 main_rm.error = e
983 next(main_retries)
984 continue
985 return self._error_or_warning(e, fatal=fatal)
986
987 try:
988 self._extract_and_report_alerts(response, only_once=True)
989 except ExtractorError as e:
990 # YouTube's servers may return errors we want to retry on in a 200 OK response
991 # See: https://github.com/yt-dlp/yt-dlp/issues/839
992 if 'unknown error' in e.msg.lower():
993 main_rm.error = e
994 next(main_retries)
995 continue
996 return self._error_or_warning(e, fatal=fatal)
997 # Youtube sometimes sends incomplete data
998 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
999 if not traverse_obj(response, *variadic(check_get_keys)):
1000 icd_rm.error = ExtractorError('Incomplete data received', expected=True)
1001 should_retry = next(icd_retries, None)
1002 if not should_retry:
1003 return None
1004 continue
1005
1006 return response
1007
1008 @staticmethod
1009 def is_music_url(url):
1010 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
1011
1012 def _extract_video(self, renderer):
1013 video_id = renderer.get('videoId')
1014
1015 reel_header_renderer = traverse_obj(renderer, (
1016 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1017 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1018
1019 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1020 description = self._get_text(renderer, 'descriptionSnippet')
1021
1022 duration = int_or_none(renderer.get('lengthSeconds'))
1023 if duration is None:
1024 duration = parse_duration(self._get_text(
1025 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1026 if duration is None:
1027 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1028 duration = parse_duration(self._search_regex(
1029 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1030 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1031 video_id, default=None, group='duration'))
1032
1033 channel_id = traverse_obj(
1034 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1035 expected_type=str, get_all=False)
1036 if not channel_id:
1037 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1038
1039 channel_id = self.ucid_or_none(channel_id)
1040
1041 overlay_style = traverse_obj(
1042 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1043 get_all=False, expected_type=str)
1044 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1045 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1046 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1047 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1048 expected_type=str)) or ''
1049 url = f'https://www.youtube.com/watch?v={video_id}'
1050 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1051 url = f'https://www.youtube.com/shorts/{video_id}'
1052
1053 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1054 or self._get_text(reel_header_renderer, 'timestampText') or '')
1055 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1056
1057 live_status = (
1058 'is_upcoming' if scheduled_timestamp is not None
1059 else 'was_live' if 'streamed' in time_text.lower()
1060 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1061 else None)
1062
1063 # videoInfo is a string like '50K views • 10 years ago'.
1064 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1065 view_count = (0 if 'no views' in view_count_text.lower()
1066 else self._get_count({'simpleText': view_count_text}))
1067 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1068
1069 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1070 or self._get_text(reel_header_renderer, 'channelTitleText'))
1071
1072 channel_handle = traverse_obj(renderer, (
1073 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1074 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1075 expected_type=self.handle_from_url, get_all=False)
1076 return {
1077 '_type': 'url',
1078 'ie_key': YoutubeIE.ie_key(),
1079 'id': video_id,
1080 'url': url,
1081 'title': title,
1082 'description': description,
1083 'duration': duration,
1084 'channel_id': channel_id,
1085 'channel': channel,
1086 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1087 'uploader': channel,
1088 'uploader_id': channel_handle,
1089 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1090 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1091 'timestamp': (self._parse_time_text(time_text)
1092 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1093 else None),
1094 'release_timestamp': scheduled_timestamp,
1095 'availability':
1096 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1097 else self._availability(
1098 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1099 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1100 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1101 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1102 view_count_field: view_count,
1103 'live_status': live_status,
1104 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None
1105 }
1106
1107
1108 class YoutubeIE(YoutubeBaseInfoExtractor):
1109 IE_DESC = 'YouTube'
1110 _VALID_URL = r"""(?x)^
1111 (
1112 (?:https?://|//) # http(s):// or protocol-independent URL
1113 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1114 (?:www\.)?deturl\.com/www\.youtube\.com|
1115 (?:www\.)?pwnyoutube\.com|
1116 (?:www\.)?hooktube\.com|
1117 (?:www\.)?yourepeat\.com|
1118 tube\.majestyc\.net|
1119 %(invidious)s|
1120 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1121 (?:.*?\#/)? # handle anchor (#/) redirect urls
1122 (?: # the various things that can precede the ID:
1123 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1124 |(?: # or the v= param in all its forms
1125 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1126 (?:\?|\#!?) # the params delimiter ? or # or #!
1127 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1128 v=
1129 )
1130 ))
1131 |(?:
1132 youtu\.be| # just youtu.be/xxxx
1133 vid\.plus| # or vid.plus/xxxx
1134 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1135 %(invidious)s
1136 )/
1137 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1138 )
1139 )? # all until now is optional -> you can pass the naked ID
1140 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1141 (?(1).+)? # if we found the ID, everything can follow
1142 (?:\#|$)""" % {
1143 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1144 }
1145 _EMBED_REGEX = [
1146 r'''(?x)
1147 (?:
1148 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1149 data-video-url=|
1150 <embed[^>]+?src=|
1151 embedSWF\(?:\s*|
1152 <object[^>]+data=|
1153 new\s+SWFObject\(
1154 )
1155 (["\'])
1156 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1157 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1158 \1''',
1159 # https://wordpress.org/plugins/lazy-load-for-videos/
1160 r'''(?xs)
1161 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1162 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1163 ]
1164 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1165
1166 _PLAYER_INFO_RE = (
1167 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1168 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$',
1169 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1170 )
1171 _formats = {
1172 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1173 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1174 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1175 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1176 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1177 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1178 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1179 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1180 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1181 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1182 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1183 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1184 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1185 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1186 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1187 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1188 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1189 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1190
1191
1192 # 3D videos
1193 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1194 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1195 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1196 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1197 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1198 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1199 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1200
1201 # Apple HTTP Live Streaming
1202 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1203 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1204 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1205 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1206 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1207 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1208 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1209 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1210
1211 # DASH mp4 video
1212 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1213 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1214 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1215 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1216 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1217 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1218 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1219 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1220 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1221 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1222 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1223 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1224
1225 # Dash mp4 audio
1226 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1227 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1228 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1229 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1230 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1231 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1232 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1233
1234 # Dash webm
1235 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1236 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1237 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1238 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1239 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1240 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1241 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1242 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1243 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1244 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1245 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1246 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1247 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1248 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1249 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1250 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1251 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1252 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1253 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1254 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1255 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1256 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1257
1258 # Dash webm audio
1259 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1260 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1261
1262 # Dash webm audio with opus inside
1263 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1264 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1265 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1266
1267 # RTMP (unnamed)
1268 '_rtmp': {'protocol': 'rtmp'},
1269
1270 # av01 video only formats sometimes served with "unknown" codecs
1271 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1272 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1273 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1274 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1275 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1276 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1277 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1278 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1279 }
1280 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1281
1282 _GEO_BYPASS = False
1283
1284 IE_NAME = 'youtube'
1285 _TESTS = [
1286 {
1287 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1288 'info_dict': {
1289 'id': 'BaW_jenozKc',
1290 'ext': 'mp4',
1291 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1292 'channel': 'Philipp Hagemeister',
1293 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1294 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1295 'upload_date': '20121002',
1296 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1297 'categories': ['Science & Technology'],
1298 'tags': ['youtube-dl'],
1299 'duration': 10,
1300 'view_count': int,
1301 'like_count': int,
1302 'availability': 'public',
1303 'playable_in_embed': True,
1304 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1305 'live_status': 'not_live',
1306 'age_limit': 0,
1307 'start_time': 1,
1308 'end_time': 9,
1309 'comment_count': int,
1310 'channel_follower_count': int,
1311 'uploader': 'Philipp Hagemeister',
1312 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1313 'uploader_id': '@PhilippHagemeister',
1314 'heatmap': 'count:100',
1315 }
1316 },
1317 {
1318 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1319 'note': 'Embed-only video (#1746)',
1320 'info_dict': {
1321 'id': 'yZIXLfi8CZQ',
1322 'ext': 'mp4',
1323 'upload_date': '20120608',
1324 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1325 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1326 'age_limit': 18,
1327 },
1328 'skip': 'Private video',
1329 },
1330 {
1331 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1332 'note': 'Use the first video ID in the URL',
1333 'info_dict': {
1334 'id': 'BaW_jenozKc',
1335 'ext': 'mp4',
1336 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1337 'channel': 'Philipp Hagemeister',
1338 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1339 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1340 'upload_date': '20121002',
1341 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1342 'categories': ['Science & Technology'],
1343 'tags': ['youtube-dl'],
1344 'duration': 10,
1345 'view_count': int,
1346 'like_count': int,
1347 'availability': 'public',
1348 'playable_in_embed': True,
1349 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1350 'live_status': 'not_live',
1351 'age_limit': 0,
1352 'comment_count': int,
1353 'channel_follower_count': int,
1354 'uploader': 'Philipp Hagemeister',
1355 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1356 'uploader_id': '@PhilippHagemeister',
1357 'heatmap': 'count:100',
1358 },
1359 'params': {
1360 'skip_download': True,
1361 },
1362 },
1363 {
1364 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1365 'note': '256k DASH audio (format 141) via DASH manifest',
1366 'info_dict': {
1367 'id': 'a9LDPn-MO4I',
1368 'ext': 'm4a',
1369 'upload_date': '20121002',
1370 'description': '',
1371 'title': 'UHDTV TEST 8K VIDEO.mp4'
1372 },
1373 'params': {
1374 'youtube_include_dash_manifest': True,
1375 'format': '141',
1376 },
1377 'skip': 'format 141 not served anymore',
1378 },
1379 # DASH manifest with encrypted signature
1380 {
1381 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1382 'info_dict': {
1383 'id': 'IB3lcPjvWLA',
1384 'ext': 'm4a',
1385 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1386 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1387 'duration': 244,
1388 'upload_date': '20131011',
1389 'abr': 129.495,
1390 'like_count': int,
1391 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1392 'playable_in_embed': True,
1393 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1394 'view_count': int,
1395 'track': 'The Spark',
1396 'live_status': 'not_live',
1397 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1398 'channel': 'Afrojack',
1399 'tags': 'count:19',
1400 'availability': 'public',
1401 'categories': ['Music'],
1402 'age_limit': 0,
1403 'alt_title': 'The Spark',
1404 'channel_follower_count': int,
1405 'uploader': 'Afrojack',
1406 'uploader_url': 'https://www.youtube.com/@Afrojack',
1407 'uploader_id': '@Afrojack',
1408 },
1409 'params': {
1410 'youtube_include_dash_manifest': True,
1411 'format': '141/bestaudio[ext=m4a]',
1412 },
1413 },
1414 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1415 {
1416 'note': 'Embed allowed age-gate video',
1417 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1418 'info_dict': {
1419 'id': 'HtVdAasjOgU',
1420 'ext': 'mp4',
1421 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1422 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1423 'duration': 142,
1424 'upload_date': '20140605',
1425 'age_limit': 18,
1426 'categories': ['Gaming'],
1427 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1428 'availability': 'needs_auth',
1429 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1430 'like_count': int,
1431 'channel': 'The Witcher',
1432 'live_status': 'not_live',
1433 'tags': 'count:17',
1434 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1435 'playable_in_embed': True,
1436 'view_count': int,
1437 'channel_follower_count': int,
1438 'uploader': 'The Witcher',
1439 'uploader_url': 'https://www.youtube.com/@thewitcher',
1440 'uploader_id': '@thewitcher',
1441 'comment_count': int,
1442 'channel_is_verified': True,
1443 'heatmap': 'count:100',
1444 },
1445 },
1446 {
1447 'note': 'Age-gate video with embed allowed in public site',
1448 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1449 'info_dict': {
1450 'id': 'HsUATh_Nc2U',
1451 'ext': 'mp4',
1452 'title': 'Godzilla 2 (Official Video)',
1453 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1454 'upload_date': '20200408',
1455 'age_limit': 18,
1456 'availability': 'needs_auth',
1457 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1458 'channel': 'FlyingKitty',
1459 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1460 'view_count': int,
1461 'categories': ['Entertainment'],
1462 'live_status': 'not_live',
1463 'tags': ['Flyingkitty', 'godzilla 2'],
1464 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1465 'like_count': int,
1466 'duration': 177,
1467 'playable_in_embed': True,
1468 'channel_follower_count': int,
1469 'uploader': 'FlyingKitty',
1470 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1471 'uploader_id': '@FlyingKitty900',
1472 'comment_count': int,
1473 'channel_is_verified': True,
1474 },
1475 },
1476 {
1477 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1478 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1479 'info_dict': {
1480 'id': 'Tq92D6wQ1mg',
1481 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1482 'ext': 'mp4',
1483 'upload_date': '20191228',
1484 'description': 'md5:17eccca93a786d51bc67646756894066',
1485 'age_limit': 18,
1486 'like_count': int,
1487 'availability': 'needs_auth',
1488 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1489 'view_count': int,
1490 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1491 'channel': 'Projekt Melody',
1492 'live_status': 'not_live',
1493 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1494 'playable_in_embed': True,
1495 'categories': ['Entertainment'],
1496 'duration': 106,
1497 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1498 'comment_count': int,
1499 'channel_follower_count': int,
1500 'uploader': 'Projekt Melody',
1501 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1502 'uploader_id': '@ProjektMelody',
1503 },
1504 },
1505 {
1506 'note': 'Non-Agegated non-embeddable video',
1507 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1508 'info_dict': {
1509 'id': 'MeJVWBSsPAY',
1510 'ext': 'mp4',
1511 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1512 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1513 'upload_date': '20130730',
1514 'track': 'Such mich find mich',
1515 'age_limit': 0,
1516 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1517 'like_count': int,
1518 'playable_in_embed': False,
1519 'creator': 'OOMPH!',
1520 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1521 'view_count': int,
1522 'alt_title': 'Such mich find mich',
1523 'duration': 210,
1524 'channel': 'Herr Lurik',
1525 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1526 'categories': ['Music'],
1527 'availability': 'public',
1528 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1529 'live_status': 'not_live',
1530 'artist': 'OOMPH!',
1531 'channel_follower_count': int,
1532 'uploader': 'Herr Lurik',
1533 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1534 'uploader_id': '@HerrLurik',
1535 },
1536 },
1537 {
1538 'note': 'Non-bypassable age-gated video',
1539 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1540 'only_matching': True,
1541 },
1542 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1543 # YouTube Red ad is not captured for creator
1544 {
1545 'url': '__2ABJjxzNo',
1546 'info_dict': {
1547 'id': '__2ABJjxzNo',
1548 'ext': 'mp4',
1549 'duration': 266,
1550 'upload_date': '20100430',
1551 'creator': 'deadmau5',
1552 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1553 'title': 'Deadmau5 - Some Chords (HD)',
1554 'alt_title': 'Some Chords',
1555 'availability': 'public',
1556 'tags': 'count:14',
1557 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1558 'view_count': int,
1559 'live_status': 'not_live',
1560 'channel': 'deadmau5',
1561 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1562 'like_count': int,
1563 'track': 'Some Chords',
1564 'artist': 'deadmau5',
1565 'playable_in_embed': True,
1566 'age_limit': 0,
1567 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1568 'categories': ['Music'],
1569 'album': 'Some Chords',
1570 'channel_follower_count': int,
1571 'uploader': 'deadmau5',
1572 'uploader_url': 'https://www.youtube.com/@deadmau5',
1573 'uploader_id': '@deadmau5',
1574 },
1575 'expected_warnings': [
1576 'DASH manifest missing',
1577 ]
1578 },
1579 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1580 {
1581 'url': 'lqQg6PlCWgI',
1582 'info_dict': {
1583 'id': 'lqQg6PlCWgI',
1584 'ext': 'mp4',
1585 'duration': 6085,
1586 'upload_date': '20150827',
1587 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1588 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1589 'like_count': int,
1590 'release_timestamp': 1343767800,
1591 'playable_in_embed': True,
1592 'categories': ['Sports'],
1593 'release_date': '20120731',
1594 'channel': 'Olympics',
1595 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1596 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1597 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1598 'age_limit': 0,
1599 'availability': 'public',
1600 'live_status': 'was_live',
1601 'view_count': int,
1602 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1603 'channel_follower_count': int,
1604 'uploader': 'Olympics',
1605 'uploader_url': 'https://www.youtube.com/@Olympics',
1606 'uploader_id': '@Olympics',
1607 'channel_is_verified': True,
1608 },
1609 'params': {
1610 'skip_download': 'requires avconv',
1611 }
1612 },
1613 # Non-square pixels
1614 {
1615 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1616 'info_dict': {
1617 'id': '_b-2C3KPAM0',
1618 'ext': 'mp4',
1619 'stretched_ratio': 16 / 9.,
1620 'duration': 85,
1621 'upload_date': '20110310',
1622 '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',
1623 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1624 'playable_in_embed': True,
1625 'channel': '孫ᄋᄅ',
1626 'age_limit': 0,
1627 'tags': 'count:11',
1628 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1629 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1630 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1631 'view_count': int,
1632 'categories': ['People & Blogs'],
1633 'like_count': int,
1634 'live_status': 'not_live',
1635 'availability': 'unlisted',
1636 'comment_count': int,
1637 'channel_follower_count': int,
1638 'uploader': '孫ᄋᄅ',
1639 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1640 'uploader_id': '@AllenMeow',
1641 },
1642 },
1643 # url_encoded_fmt_stream_map is empty string
1644 {
1645 'url': 'qEJwOuvDf7I',
1646 'info_dict': {
1647 'id': 'qEJwOuvDf7I',
1648 'ext': 'webm',
1649 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1650 'description': '',
1651 'upload_date': '20150404',
1652 },
1653 'params': {
1654 'skip_download': 'requires avconv',
1655 },
1656 'skip': 'This live event has ended.',
1657 },
1658 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1659 {
1660 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1661 'info_dict': {
1662 'id': 'FIl7x6_3R5Y',
1663 'ext': 'webm',
1664 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1665 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1666 'duration': 220,
1667 'upload_date': '20150625',
1668 'formats': 'mincount:31',
1669 },
1670 'skip': 'not actual anymore',
1671 },
1672 # DASH manifest with segment_list
1673 {
1674 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1675 'md5': '8ce563a1d667b599d21064e982ab9e31',
1676 'info_dict': {
1677 'id': 'CsmdDsKjzN8',
1678 'ext': 'mp4',
1679 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1680 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1681 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1682 },
1683 'params': {
1684 'youtube_include_dash_manifest': True,
1685 'format': '135', # bestvideo
1686 },
1687 'skip': 'This live event has ended.',
1688 },
1689 {
1690 # Multifeed videos (multiple cameras), URL can be of any Camera
1691 # TODO: fix multifeed titles
1692 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1693 'info_dict': {
1694 'id': 'zaPI8MvL8pg',
1695 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1696 'description': 'md5:563ccbc698b39298481ca3c571169519',
1697 },
1698 'playlist': [{
1699 'info_dict': {
1700 'id': 'j5yGuxZ8lLU',
1701 'ext': 'mp4',
1702 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1703 'description': 'md5:563ccbc698b39298481ca3c571169519',
1704 'duration': 10120,
1705 'channel_follower_count': int,
1706 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1707 'availability': 'public',
1708 'playable_in_embed': True,
1709 'upload_date': '20131105',
1710 'categories': ['Gaming'],
1711 'live_status': 'was_live',
1712 'tags': 'count:24',
1713 'release_timestamp': 1383701910,
1714 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1715 'comment_count': int,
1716 'age_limit': 0,
1717 'like_count': int,
1718 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1719 'channel': 'WiiLikeToPlay',
1720 'view_count': int,
1721 'release_date': '20131106',
1722 'uploader': 'WiiLikeToPlay',
1723 'uploader_id': '@WLTP',
1724 'uploader_url': 'https://www.youtube.com/@WLTP',
1725 },
1726 }, {
1727 'info_dict': {
1728 'id': 'zaPI8MvL8pg',
1729 'ext': 'mp4',
1730 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1731 'availability': 'public',
1732 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1733 'channel': 'WiiLikeToPlay',
1734 'channel_follower_count': int,
1735 'description': 'md5:563ccbc698b39298481ca3c571169519',
1736 'duration': 10108,
1737 'age_limit': 0,
1738 'like_count': int,
1739 'tags': 'count:24',
1740 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1741 'release_timestamp': 1383701915,
1742 'comment_count': int,
1743 'upload_date': '20131105',
1744 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1745 'release_date': '20131106',
1746 'playable_in_embed': True,
1747 'live_status': 'was_live',
1748 'categories': ['Gaming'],
1749 'view_count': int,
1750 'uploader': 'WiiLikeToPlay',
1751 'uploader_id': '@WLTP',
1752 'uploader_url': 'https://www.youtube.com/@WLTP',
1753 },
1754 }, {
1755 'info_dict': {
1756 'id': 'R7r3vfO7Hao',
1757 'ext': 'mp4',
1758 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1759 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1760 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1761 'like_count': int,
1762 'availability': 'public',
1763 'playable_in_embed': True,
1764 'upload_date': '20131105',
1765 'description': 'md5:563ccbc698b39298481ca3c571169519',
1766 'channel_follower_count': int,
1767 'tags': 'count:24',
1768 'release_date': '20131106',
1769 'comment_count': int,
1770 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1771 'channel': 'WiiLikeToPlay',
1772 'categories': ['Gaming'],
1773 'release_timestamp': 1383701914,
1774 'live_status': 'was_live',
1775 'age_limit': 0,
1776 'duration': 10128,
1777 'view_count': int,
1778 'uploader': 'WiiLikeToPlay',
1779 'uploader_id': '@WLTP',
1780 'uploader_url': 'https://www.youtube.com/@WLTP',
1781 },
1782 }],
1783 'params': {'skip_download': True},
1784 },
1785 {
1786 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1787 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1788 'info_dict': {
1789 'id': 'gVfLd0zydlo',
1790 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1791 },
1792 'playlist_count': 2,
1793 'skip': 'Not multifeed anymore',
1794 },
1795 {
1796 'url': 'https://vid.plus/FlRa-iH7PGw',
1797 'only_matching': True,
1798 },
1799 {
1800 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1801 'only_matching': True,
1802 },
1803 {
1804 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1805 # Also tests cut-off URL expansion in video description (see
1806 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1807 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1808 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1809 'info_dict': {
1810 'id': 'lsguqyKfVQg',
1811 'ext': 'mp4',
1812 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1813 'alt_title': 'Dark Walk',
1814 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1815 'duration': 133,
1816 'upload_date': '20151119',
1817 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1818 'track': 'Dark Walk',
1819 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1820 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1821 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1822 'categories': ['Film & Animation'],
1823 'view_count': int,
1824 'live_status': 'not_live',
1825 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1826 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1827 'tags': 'count:13',
1828 'availability': 'public',
1829 'channel': 'IronSoulElf',
1830 'playable_in_embed': True,
1831 'like_count': int,
1832 'age_limit': 0,
1833 'channel_follower_count': int
1834 },
1835 'params': {
1836 'skip_download': True,
1837 },
1838 },
1839 {
1840 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1841 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1842 'only_matching': True,
1843 },
1844 {
1845 # Video with yt:stretch=17:0
1846 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1847 'info_dict': {
1848 'id': 'Q39EVAstoRM',
1849 'ext': 'mp4',
1850 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1851 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1852 'upload_date': '20151107',
1853 },
1854 'params': {
1855 'skip_download': True,
1856 },
1857 'skip': 'This video does not exist.',
1858 },
1859 {
1860 # Video with incomplete 'yt:stretch=16:'
1861 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1862 'only_matching': True,
1863 },
1864 {
1865 # Video licensed under Creative Commons
1866 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1867 'info_dict': {
1868 'id': 'M4gD1WSo5mA',
1869 'ext': 'mp4',
1870 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1871 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1872 'duration': 721,
1873 'upload_date': '20150128',
1874 'license': 'Creative Commons Attribution license (reuse allowed)',
1875 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1876 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1877 'like_count': int,
1878 'age_limit': 0,
1879 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1880 'channel': 'The Berkman Klein Center for Internet & Society',
1881 'availability': 'public',
1882 'view_count': int,
1883 'categories': ['Education'],
1884 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1885 'live_status': 'not_live',
1886 'playable_in_embed': True,
1887 'channel_follower_count': int,
1888 'chapters': list,
1889 'uploader': 'The Berkman Klein Center for Internet & Society',
1890 'uploader_id': '@BKCHarvard',
1891 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1892 },
1893 'params': {
1894 'skip_download': True,
1895 },
1896 },
1897 {
1898 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1899 'info_dict': {
1900 'id': 'eQcmzGIKrzg',
1901 'ext': 'mp4',
1902 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1903 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1904 'duration': 4060,
1905 'upload_date': '20151120',
1906 'license': 'Creative Commons Attribution license (reuse allowed)',
1907 'playable_in_embed': True,
1908 'tags': 'count:12',
1909 'like_count': int,
1910 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1911 'age_limit': 0,
1912 'availability': 'public',
1913 'categories': ['News & Politics'],
1914 'channel': 'Bernie Sanders',
1915 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1916 'view_count': int,
1917 'live_status': 'not_live',
1918 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1919 'comment_count': int,
1920 'channel_follower_count': int,
1921 'chapters': list,
1922 'uploader': 'Bernie Sanders',
1923 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1924 'uploader_id': '@BernieSanders',
1925 'channel_is_verified': True,
1926 'heatmap': 'count:100',
1927 },
1928 'params': {
1929 'skip_download': True,
1930 },
1931 },
1932 {
1933 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1934 'only_matching': True,
1935 },
1936 {
1937 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1938 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1939 'only_matching': True,
1940 },
1941 {
1942 # Rental video preview
1943 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1944 'info_dict': {
1945 'id': 'uGpuVWrhIzE',
1946 'ext': 'mp4',
1947 'title': 'Piku - Trailer',
1948 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1949 'upload_date': '20150811',
1950 'license': 'Standard YouTube License',
1951 },
1952 'params': {
1953 'skip_download': True,
1954 },
1955 'skip': 'This video is not available.',
1956 },
1957 {
1958 # YouTube Red video with episode data
1959 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1960 'info_dict': {
1961 'id': 'iqKdEhx-dD4',
1962 'ext': 'mp4',
1963 'title': 'Isolation - Mind Field (Ep 1)',
1964 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1965 'duration': 2085,
1966 'upload_date': '20170118',
1967 'series': 'Mind Field',
1968 'season_number': 1,
1969 'episode_number': 1,
1970 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1971 'tags': 'count:12',
1972 'view_count': int,
1973 'availability': 'public',
1974 'age_limit': 0,
1975 'channel': 'Vsauce',
1976 'episode': 'Episode 1',
1977 'categories': ['Entertainment'],
1978 'season': 'Season 1',
1979 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1980 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1981 'like_count': int,
1982 'playable_in_embed': True,
1983 'live_status': 'not_live',
1984 'channel_follower_count': int,
1985 'uploader': 'Vsauce',
1986 'uploader_url': 'https://www.youtube.com/@Vsauce',
1987 'uploader_id': '@Vsauce',
1988 'comment_count': int,
1989 'channel_is_verified': True,
1990 },
1991 'params': {
1992 'skip_download': True,
1993 },
1994 'expected_warnings': [
1995 'Skipping DASH manifest',
1996 ],
1997 },
1998 {
1999 # The following content has been identified by the YouTube community
2000 # as inappropriate or offensive to some audiences.
2001 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
2002 'info_dict': {
2003 'id': '6SJNVb0GnPI',
2004 'ext': 'mp4',
2005 'title': 'Race Differences in Intelligence',
2006 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
2007 'duration': 965,
2008 'upload_date': '20140124',
2009 },
2010 'params': {
2011 'skip_download': True,
2012 },
2013 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
2014 },
2015 {
2016 # itag 212
2017 'url': '1t24XAntNCY',
2018 'only_matching': True,
2019 },
2020 {
2021 # geo restricted to JP
2022 'url': 'sJL6WA-aGkQ',
2023 'only_matching': True,
2024 },
2025 {
2026 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2027 'only_matching': True,
2028 },
2029 {
2030 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2031 'only_matching': True,
2032 },
2033 {
2034 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2035 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2036 'only_matching': True,
2037 },
2038 {
2039 # DRM protected
2040 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2041 'only_matching': True,
2042 },
2043 {
2044 # Video with unsupported adaptive stream type formats
2045 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2046 'info_dict': {
2047 'id': 'Z4Vy8R84T1U',
2048 'ext': 'mp4',
2049 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2050 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2051 'duration': 433,
2052 'upload_date': '20130923',
2053 'formats': 'maxcount:10',
2054 },
2055 'params': {
2056 'skip_download': True,
2057 'youtube_include_dash_manifest': False,
2058 },
2059 'skip': 'not actual anymore',
2060 },
2061 {
2062 # Youtube Music Auto-generated description
2063 # TODO: fix metadata extraction
2064 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2065 'info_dict': {
2066 'id': 'MgNrAu2pzNs',
2067 'ext': 'mp4',
2068 'title': 'Voyeur Girl',
2069 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2070 'upload_date': '20190312',
2071 'artist': 'Stephen',
2072 'track': 'Voyeur Girl',
2073 'album': 'it\'s too much love to know my dear',
2074 'release_date': '20190313',
2075 'alt_title': 'Voyeur Girl',
2076 'view_count': int,
2077 'playable_in_embed': True,
2078 'like_count': int,
2079 'categories': ['Music'],
2080 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2081 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2082 'uploader': 'Stephen',
2083 'availability': 'public',
2084 'creator': 'Stephen',
2085 'duration': 169,
2086 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2087 'age_limit': 0,
2088 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2089 'tags': 'count:11',
2090 'live_status': 'not_live',
2091 'channel_follower_count': int
2092 },
2093 'params': {
2094 'skip_download': True,
2095 },
2096 },
2097 {
2098 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2099 'only_matching': True,
2100 },
2101 {
2102 # invalid -> valid video id redirection
2103 'url': 'DJztXj2GPfl',
2104 'info_dict': {
2105 'id': 'DJztXj2GPfk',
2106 'ext': 'mp4',
2107 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2108 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2109 'upload_date': '20090125',
2110 'artist': 'Panjabi MC',
2111 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2112 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2113 },
2114 'params': {
2115 'skip_download': True,
2116 },
2117 'skip': 'Video unavailable',
2118 },
2119 {
2120 # empty description results in an empty string
2121 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2122 'info_dict': {
2123 'id': 'x41yOUIvK2k',
2124 'ext': 'mp4',
2125 'title': 'IMG 3456',
2126 'description': '',
2127 'upload_date': '20170613',
2128 'view_count': int,
2129 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2130 'like_count': int,
2131 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2132 'tags': [],
2133 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2134 'availability': 'public',
2135 'age_limit': 0,
2136 'categories': ['Pets & Animals'],
2137 'duration': 7,
2138 'playable_in_embed': True,
2139 'live_status': 'not_live',
2140 'channel': 'l\'Or Vert asbl',
2141 'channel_follower_count': int,
2142 'uploader': 'l\'Or Vert asbl',
2143 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2144 'uploader_id': '@ElevageOrVert',
2145 },
2146 'params': {
2147 'skip_download': True,
2148 },
2149 },
2150 {
2151 # with '};' inside yt initial data (see [1])
2152 # see [2] for an example with '};' inside ytInitialPlayerResponse
2153 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2154 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2155 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2156 'info_dict': {
2157 'id': 'CHqg6qOn4no',
2158 'ext': 'mp4',
2159 'title': 'Part 77 Sort a list of simple types in c#',
2160 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2161 'upload_date': '20130831',
2162 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2163 'like_count': int,
2164 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2165 'live_status': 'not_live',
2166 'categories': ['Education'],
2167 'availability': 'public',
2168 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2169 'tags': 'count:12',
2170 'playable_in_embed': True,
2171 'age_limit': 0,
2172 'view_count': int,
2173 'duration': 522,
2174 'channel': 'kudvenkat',
2175 'comment_count': int,
2176 'channel_follower_count': int,
2177 'chapters': list,
2178 'uploader': 'kudvenkat',
2179 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2180 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2181 'channel_is_verified': True,
2182 'heatmap': 'count:100',
2183 },
2184 'params': {
2185 'skip_download': True,
2186 },
2187 },
2188 {
2189 # another example of '};' in ytInitialData
2190 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2191 'only_matching': True,
2192 },
2193 {
2194 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2195 'only_matching': True,
2196 },
2197 {
2198 # https://github.com/ytdl-org/youtube-dl/pull/28094
2199 'url': 'OtqTfy26tG0',
2200 'info_dict': {
2201 'id': 'OtqTfy26tG0',
2202 'ext': 'mp4',
2203 'title': 'Burn Out',
2204 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2205 'upload_date': '20141120',
2206 'artist': 'The Cinematic Orchestra',
2207 'track': 'Burn Out',
2208 'album': 'Every Day',
2209 'like_count': int,
2210 'live_status': 'not_live',
2211 'alt_title': 'Burn Out',
2212 'duration': 614,
2213 'age_limit': 0,
2214 'view_count': int,
2215 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2216 'creator': 'The Cinematic Orchestra',
2217 'channel': 'The Cinematic Orchestra',
2218 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2219 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2220 'availability': 'public',
2221 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2222 'categories': ['Music'],
2223 'playable_in_embed': True,
2224 'channel_follower_count': int,
2225 'uploader': 'The Cinematic Orchestra',
2226 'comment_count': int,
2227 },
2228 'params': {
2229 'skip_download': True,
2230 },
2231 },
2232 {
2233 # controversial video, only works with bpctr when authenticated with cookies
2234 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2235 'only_matching': True,
2236 },
2237 {
2238 # controversial video, requires bpctr/contentCheckOk
2239 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2240 'info_dict': {
2241 'id': 'SZJvDhaSDnc',
2242 'ext': 'mp4',
2243 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2244 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2245 'upload_date': '20140716',
2246 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2247 'duration': 170,
2248 'categories': ['News & Politics'],
2249 'view_count': int,
2250 'channel': 'CBS Mornings',
2251 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2252 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2253 'age_limit': 18,
2254 'availability': 'needs_auth',
2255 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2256 'like_count': int,
2257 'live_status': 'not_live',
2258 'playable_in_embed': True,
2259 'channel_follower_count': int,
2260 'uploader': 'CBS Mornings',
2261 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2262 'uploader_id': '@CBSMornings',
2263 'comment_count': int,
2264 'channel_is_verified': True,
2265 }
2266 },
2267 {
2268 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2269 'url': 'cBvYw8_A0vQ',
2270 'info_dict': {
2271 'id': 'cBvYw8_A0vQ',
2272 'ext': 'mp4',
2273 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2274 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2275 'upload_date': '20201120',
2276 'duration': 1456,
2277 'categories': ['Travel & Events'],
2278 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2279 'view_count': int,
2280 'channel': 'Walk around Japan',
2281 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2282 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2283 'age_limit': 0,
2284 'availability': 'public',
2285 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2286 'live_status': 'not_live',
2287 'playable_in_embed': True,
2288 'channel_follower_count': int,
2289 'uploader': 'Walk around Japan',
2290 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2291 'uploader_id': '@walkaroundjapan7124',
2292 },
2293 'params': {
2294 'skip_download': True,
2295 },
2296 }, {
2297 # Has multiple audio streams
2298 'url': 'WaOKSUlf4TM',
2299 'only_matching': True
2300 }, {
2301 # Requires Premium: has format 141 when requested using YTM url
2302 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2303 'only_matching': True
2304 }, {
2305 # multiple subtitles with same lang_code
2306 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2307 'only_matching': True,
2308 }, {
2309 # Force use android client fallback
2310 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2311 'info_dict': {
2312 'id': 'YOelRv7fMxY',
2313 'title': 'DIGGING A SECRET TUNNEL Part 1',
2314 'ext': '3gp',
2315 'upload_date': '20210624',
2316 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2317 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2318 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2319 'duration': 596,
2320 'categories': ['Entertainment'],
2321 'view_count': int,
2322 'channel': 'colinfurze',
2323 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2324 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2325 'age_limit': 0,
2326 'availability': 'public',
2327 'like_count': int,
2328 'live_status': 'not_live',
2329 'playable_in_embed': True,
2330 'channel_follower_count': int,
2331 'chapters': list,
2332 'uploader': 'colinfurze',
2333 'uploader_url': 'https://www.youtube.com/@colinfurze',
2334 'uploader_id': '@colinfurze',
2335 'comment_count': int,
2336 'channel_is_verified': True,
2337 'heatmap': 'count:100',
2338 },
2339 'params': {
2340 'format': '17', # 3gp format available on android
2341 'extractor_args': {'youtube': {'player_client': ['android']}},
2342 },
2343 },
2344 {
2345 # Skip download of additional client configs (remix client config in this case)
2346 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2347 'only_matching': True,
2348 'params': {
2349 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2350 },
2351 }, {
2352 # shorts
2353 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2354 'only_matching': True,
2355 }, {
2356 'note': 'Storyboards',
2357 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2358 'info_dict': {
2359 'id': '5KLPxDtMqe8',
2360 'ext': 'mhtml',
2361 'format_id': 'sb0',
2362 'title': 'Your Brain is Plastic',
2363 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2364 'upload_date': '20140324',
2365 'like_count': int,
2366 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2367 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2368 'view_count': int,
2369 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2370 'playable_in_embed': True,
2371 'tags': 'count:12',
2372 'availability': 'public',
2373 'channel': 'SciShow',
2374 'live_status': 'not_live',
2375 'duration': 248,
2376 'categories': ['Education'],
2377 'age_limit': 0,
2378 'channel_follower_count': int,
2379 'chapters': list,
2380 'uploader': 'SciShow',
2381 'uploader_url': 'https://www.youtube.com/@SciShow',
2382 'uploader_id': '@SciShow',
2383 'comment_count': int,
2384 'channel_is_verified': True,
2385 'heatmap': 'count:100',
2386 }, 'params': {'format': 'mhtml', 'skip_download': True}
2387 }, {
2388 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2389 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2390 'info_dict': {
2391 'id': '2NUZ8W2llS4',
2392 'ext': 'mp4',
2393 'title': 'The NP that test your phone performance 🙂',
2394 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2395 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2396 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2397 'duration': 21,
2398 'view_count': int,
2399 'age_limit': 0,
2400 'categories': ['Gaming'],
2401 'tags': 'count:23',
2402 'playable_in_embed': True,
2403 'live_status': 'not_live',
2404 'upload_date': '20220103',
2405 'like_count': int,
2406 'availability': 'public',
2407 'channel': 'Leon Nguyen',
2408 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2409 'comment_count': int,
2410 'channel_follower_count': int,
2411 'uploader': 'Leon Nguyen',
2412 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2413 'uploader_id': '@LeonNguyen',
2414 'heatmap': 'count:100',
2415 }
2416 }, {
2417 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2418 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2419 'info_dict': {
2420 'id': '2NUZ8W2llS4',
2421 'ext': 'mp4',
2422 'title': 'The NP that test your phone performance 🙂',
2423 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2424 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2425 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2426 'duration': 21,
2427 'view_count': int,
2428 'age_limit': 0,
2429 'categories': ['Gaming'],
2430 'tags': 'count:23',
2431 'playable_in_embed': True,
2432 'live_status': 'not_live',
2433 'upload_date': '20220102',
2434 'like_count': int,
2435 'availability': 'public',
2436 'channel': 'Leon Nguyen',
2437 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2438 'comment_count': int,
2439 'channel_follower_count': int,
2440 'uploader': 'Leon Nguyen',
2441 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2442 'uploader_id': '@LeonNguyen',
2443 'heatmap': 'count:100',
2444 },
2445 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2446 }, {
2447 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2448 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2449 'info_dict': {
2450 'id': 'mzZzzBU6lrM',
2451 'ext': 'mp4',
2452 'title': 'I Met GeorgeNotFound In Real Life...',
2453 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2454 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2455 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2456 'duration': 955,
2457 'view_count': int,
2458 'age_limit': 0,
2459 'categories': ['Entertainment'],
2460 'tags': 'count:26',
2461 'playable_in_embed': True,
2462 'live_status': 'not_live',
2463 'release_timestamp': 1641172509,
2464 'release_date': '20220103',
2465 'upload_date': '20220103',
2466 'like_count': int,
2467 'availability': 'public',
2468 'channel': 'Quackity',
2469 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2470 'channel_follower_count': int,
2471 'uploader': 'Quackity',
2472 'uploader_id': '@Quackity',
2473 'uploader_url': 'https://www.youtube.com/@Quackity',
2474 'comment_count': int,
2475 'channel_is_verified': True,
2476 'heatmap': 'count:100',
2477 }
2478 },
2479 { # continuous livestream. Microformat upload date should be preferred.
2480 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2481 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2482 'info_dict': {
2483 'id': 'kgx4WGK0oNU',
2484 '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}',
2485 'ext': 'mp4',
2486 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2487 'availability': 'public',
2488 'age_limit': 0,
2489 'release_timestamp': 1637975704,
2490 'upload_date': '20210619',
2491 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2492 'live_status': 'is_live',
2493 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2494 'channel': 'Abao in Tokyo',
2495 'channel_follower_count': int,
2496 'release_date': '20211127',
2497 'tags': 'count:39',
2498 'categories': ['People & Blogs'],
2499 'like_count': int,
2500 'view_count': int,
2501 'playable_in_embed': True,
2502 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2503 'concurrent_view_count': int,
2504 'uploader': 'Abao in Tokyo',
2505 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2506 'uploader_id': '@abaointokyo',
2507 },
2508 'params': {'skip_download': True}
2509 }, {
2510 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2511 'info_dict': {
2512 'id': 'tjjjtzRLHvA',
2513 'ext': 'mp4',
2514 'title': 'ハッシュタグ無し };if window.ytcsi',
2515 'upload_date': '20220323',
2516 'like_count': int,
2517 'availability': 'unlisted',
2518 'channel': 'Lesmiscore',
2519 'thumbnail': r're:^https?://.*\.jpg',
2520 'age_limit': 0,
2521 'categories': ['Music'],
2522 'view_count': int,
2523 'description': '',
2524 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2525 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2526 'live_status': 'not_live',
2527 'playable_in_embed': True,
2528 'channel_follower_count': int,
2529 'duration': 6,
2530 'tags': [],
2531 'uploader_id': '@lesmiscore',
2532 'uploader': 'Lesmiscore',
2533 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2534 }
2535 }, {
2536 # Prefer primary title+description language metadata by default
2537 # Do not prefer translated description if primary is empty
2538 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2539 'info_dict': {
2540 'id': 'el3E4MbxRqQ',
2541 'ext': 'mp4',
2542 'title': 'dlp test video 2 - primary sv no desc',
2543 'description': '',
2544 'channel': 'cole-dlp-test-acc',
2545 'tags': [],
2546 'view_count': int,
2547 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2548 'like_count': int,
2549 'playable_in_embed': True,
2550 'availability': 'unlisted',
2551 'thumbnail': r're:^https?://.*\.jpg',
2552 'age_limit': 0,
2553 'duration': 5,
2554 'live_status': 'not_live',
2555 'upload_date': '20220908',
2556 'categories': ['People & Blogs'],
2557 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2558 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2559 'uploader_id': '@coletdjnz',
2560 'uploader': 'cole-dlp-test-acc',
2561 },
2562 'params': {'skip_download': True}
2563 }, {
2564 # Extractor argument: prefer translated title+description
2565 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2566 'info_dict': {
2567 'id': 'gHKT4uU8Zng',
2568 'ext': 'mp4',
2569 'channel': 'cole-dlp-test-acc',
2570 'tags': [],
2571 'duration': 5,
2572 'live_status': 'not_live',
2573 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2574 'upload_date': '20220728',
2575 'view_count': int,
2576 'categories': ['People & Blogs'],
2577 'thumbnail': r're:^https?://.*\.jpg',
2578 'title': 'dlp test video title translated (fr)',
2579 'availability': 'public',
2580 'age_limit': 0,
2581 'description': 'dlp test video description translated (fr)',
2582 'playable_in_embed': True,
2583 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2584 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2585 'uploader_id': '@coletdjnz',
2586 'uploader': 'cole-dlp-test-acc',
2587 },
2588 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2589 'expected_warnings': [r'Preferring "fr" translated fields'],
2590 }, {
2591 'note': '6 channel audio',
2592 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2593 'only_matching': True,
2594 }, {
2595 'note': 'Multiple HLS formats with same itag',
2596 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2597 'info_dict': {
2598 'id': 'kX3nB4PpJko',
2599 'ext': 'mp4',
2600 'categories': ['Entertainment'],
2601 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2602 'live_status': 'not_live',
2603 'duration': 937,
2604 'channel_follower_count': int,
2605 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2606 'title': 'Last To Take Hand Off Jet, Keeps It!',
2607 'channel': 'MrBeast',
2608 'playable_in_embed': True,
2609 'view_count': int,
2610 'upload_date': '20221112',
2611 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2612 'age_limit': 0,
2613 'availability': 'public',
2614 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2615 'like_count': int,
2616 'tags': [],
2617 'uploader': 'MrBeast',
2618 'uploader_url': 'https://www.youtube.com/@MrBeast',
2619 'uploader_id': '@MrBeast',
2620 'comment_count': int,
2621 'channel_is_verified': True,
2622 'heatmap': 'count:100',
2623 },
2624 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2625 }, {
2626 'note': 'Audio formats with Dynamic Range Compression',
2627 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2628 'info_dict': {
2629 'id': 'Tq92D6wQ1mg',
2630 'ext': 'webm',
2631 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2632 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2633 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2634 'channel_follower_count': int,
2635 'description': 'md5:17eccca93a786d51bc67646756894066',
2636 'upload_date': '20191228',
2637 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2638 'playable_in_embed': True,
2639 'like_count': int,
2640 'categories': ['Entertainment'],
2641 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2642 'age_limit': 18,
2643 'channel': 'Projekt Melody',
2644 'view_count': int,
2645 'availability': 'needs_auth',
2646 'comment_count': int,
2647 'live_status': 'not_live',
2648 'duration': 106,
2649 'uploader': 'Projekt Melody',
2650 'uploader_id': '@ProjektMelody',
2651 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2652 },
2653 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2654 },
2655 {
2656 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2657 'info_dict': {
2658 'id': 'qVv6vCqciTM',
2659 'ext': 'mp4',
2660 'age_limit': 0,
2661 'comment_count': int,
2662 'chapters': 'count:13',
2663 'upload_date': '20221223',
2664 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2665 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2666 'like_count': int,
2667 'release_date': '20221223',
2668 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2669 '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】',
2670 'view_count': int,
2671 'playable_in_embed': True,
2672 'duration': 4438,
2673 'availability': 'public',
2674 'channel_follower_count': int,
2675 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2676 'categories': ['Entertainment'],
2677 'live_status': 'was_live',
2678 'release_timestamp': 1671793345,
2679 'channel': 'さなちゃんねる',
2680 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2681 'uploader': 'さなちゃんねる',
2682 'uploader_url': 'https://www.youtube.com/@sana_natori',
2683 'uploader_id': '@sana_natori',
2684 'channel_is_verified': True,
2685 'heatmap': 'count:100',
2686 },
2687 },
2688 {
2689 # Fallbacks when webpage and web client is unavailable
2690 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2691 'info_dict': {
2692 'id': 'wSSmNUl9Snw',
2693 'ext': 'mp4',
2694 # 'categories': ['Science & Technology'],
2695 'view_count': int,
2696 'chapters': 'count:2',
2697 'channel': 'Scott Manley',
2698 'like_count': int,
2699 'age_limit': 0,
2700 # 'availability': 'public',
2701 'channel_follower_count': int,
2702 'live_status': 'not_live',
2703 'upload_date': '20170831',
2704 'duration': 682,
2705 'tags': 'count:8',
2706 'uploader_url': 'https://www.youtube.com/@scottmanley',
2707 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2708 'uploader': 'Scott Manley',
2709 'uploader_id': '@scottmanley',
2710 'title': 'The Computer Hack That Saved Apollo 14',
2711 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2712 'thumbnail': r're:^https?://.*\.webp',
2713 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2714 'playable_in_embed': True,
2715 'comment_count': int,
2716 'channel_is_verified': True,
2717 'heatmap': 'count:100',
2718 },
2719 'params': {
2720 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2721 },
2722 },
2723 ]
2724
2725 _WEBPAGE_TESTS = [
2726 # YouTube <object> embed
2727 {
2728 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2729 'md5': '873c81d308b979f0e23ee7e620b312a3',
2730 'info_dict': {
2731 'id': 'msN87y-iEx0',
2732 'ext': 'mp4',
2733 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2734 'upload_date': '20080526',
2735 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2736 'age_limit': 0,
2737 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2738 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2739 'playable_in_embed': True,
2740 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2741 'like_count': int,
2742 'comment_count': int,
2743 'channel': 'Christopher Sykes',
2744 'live_status': 'not_live',
2745 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2746 'availability': 'public',
2747 'duration': 195,
2748 'view_count': int,
2749 'categories': ['Science & Technology'],
2750 'channel_follower_count': int,
2751 'uploader': 'Christopher Sykes',
2752 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2753 'uploader_id': '@ChristopherSykesDocumentaries',
2754 'heatmap': 'count:100',
2755 },
2756 'params': {
2757 'skip_download': True,
2758 }
2759 },
2760 ]
2761
2762 @classmethod
2763 def suitable(cls, url):
2764 from ..utils import parse_qs
2765
2766 qs = parse_qs(url)
2767 if qs.get('list', [None])[0]:
2768 return False
2769 return super().suitable(url)
2770
2771 def __init__(self, *args, **kwargs):
2772 super().__init__(*args, **kwargs)
2773 self._code_cache = {}
2774 self._player_cache = {}
2775
2776 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2777 lock = threading.Lock()
2778 start_time = time.time()
2779 formats = [f for f in formats if f.get('is_from_start')]
2780
2781 def refetch_manifest(format_id, delay):
2782 nonlocal formats, start_time, is_live
2783 if time.time() <= start_time + delay:
2784 return
2785
2786 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2787 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2788 microformats = traverse_obj(
2789 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2790 expected_type=dict)
2791 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2792 is_live = live_status == 'is_live'
2793 start_time = time.time()
2794
2795 def mpd_feed(format_id, delay):
2796 """
2797 @returns (manifest_url, manifest_stream_number, is_live) or None
2798 """
2799 for retry in self.RetryManager(fatal=False):
2800 with lock:
2801 refetch_manifest(format_id, delay)
2802
2803 f = next((f for f in formats if f['format_id'] == format_id), None)
2804 if not f:
2805 if not is_live:
2806 retry.error = f'{video_id}: Video is no longer live'
2807 else:
2808 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2809 continue
2810 return f['manifest_url'], f['manifest_stream_number'], is_live
2811 return None
2812
2813 for f in formats:
2814 f['is_live'] = is_live
2815 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2816 live_start_time, mpd_feed, not is_live and f.copy())
2817 if is_live:
2818 f['fragments'] = gen
2819 f['protocol'] = 'http_dash_segments_generator'
2820 else:
2821 f['fragments'] = LazyList(gen({}))
2822 del f['is_from_start']
2823
2824 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2825 FETCH_SPAN, MAX_DURATION = 5, 432000
2826
2827 mpd_url, stream_number, is_live = None, None, True
2828
2829 begin_index = 0
2830 download_start_time = ctx.get('start') or time.time()
2831
2832 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2833 if lack_early_segments:
2834 self.report_warning(bug_reports_message(
2835 'Starting download from the last 120 hours of the live stream since '
2836 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2837 lack_early_segments = True
2838
2839 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2840 fragments, fragment_base_url = None, None
2841
2842 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2843 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2844 # Obtain from MPD's maximum seq value
2845 old_mpd_url = mpd_url
2846 last_error = ctx.pop('last_error', None)
2847 expire_fast = immediate or last_error and isinstance(last_error, HTTPError) and last_error.status == 403
2848 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2849 or (mpd_url, stream_number, False))
2850 if not refresh_sequence:
2851 if expire_fast and not is_live:
2852 return False, last_seq
2853 elif old_mpd_url == mpd_url:
2854 return True, last_seq
2855 if manifestless_orig_fmt:
2856 fmt_info = manifestless_orig_fmt
2857 else:
2858 try:
2859 fmts, _ = self._extract_mpd_formats_and_subtitles(
2860 mpd_url, None, note=False, errnote=False, fatal=False)
2861 except ExtractorError:
2862 fmts = None
2863 if not fmts:
2864 no_fragment_score += 2
2865 return False, last_seq
2866 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2867 fragments = fmt_info['fragments']
2868 fragment_base_url = fmt_info['fragment_base_url']
2869 assert fragment_base_url
2870
2871 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2872 return True, _last_seq
2873
2874 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2875 while is_live:
2876 fetch_time = time.time()
2877 if no_fragment_score > 30:
2878 return
2879 if last_segment_url:
2880 # Obtain from "X-Head-Seqnum" header value from each segment
2881 try:
2882 urlh = self._request_webpage(
2883 last_segment_url, None, note=False, errnote=False, fatal=False)
2884 except ExtractorError:
2885 urlh = None
2886 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2887 if last_seq is None:
2888 no_fragment_score += 2
2889 last_segment_url = None
2890 continue
2891 else:
2892 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2893 no_fragment_score += 2
2894 if not should_continue:
2895 continue
2896
2897 if known_idx > last_seq:
2898 last_segment_url = None
2899 continue
2900
2901 last_seq += 1
2902
2903 if begin_index < 0 and known_idx < 0:
2904 # skip from the start when it's negative value
2905 known_idx = last_seq + begin_index
2906 if lack_early_segments:
2907 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2908 try:
2909 for idx in range(known_idx, last_seq):
2910 # do not update sequence here or you'll get skipped some part of it
2911 should_continue, _ = _extract_sequence_from_mpd(False, False)
2912 if not should_continue:
2913 known_idx = idx - 1
2914 raise ExtractorError('breaking out of outer loop')
2915 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2916 yield {
2917 'url': last_segment_url,
2918 'fragment_count': last_seq,
2919 }
2920 if known_idx == last_seq:
2921 no_fragment_score += 5
2922 else:
2923 no_fragment_score = 0
2924 known_idx = last_seq
2925 except ExtractorError:
2926 continue
2927
2928 if manifestless_orig_fmt:
2929 # Stop at the first iteration if running for post-live manifestless;
2930 # fragment count no longer increase since it starts
2931 break
2932
2933 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2934
2935 def _extract_player_url(self, *ytcfgs, webpage=None):
2936 player_url = traverse_obj(
2937 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2938 get_all=False, expected_type=str)
2939 if not player_url:
2940 return
2941 return urljoin('https://www.youtube.com', player_url)
2942
2943 def _download_player_url(self, video_id, fatal=False):
2944 res = self._download_webpage(
2945 'https://www.youtube.com/iframe_api',
2946 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2947 if res:
2948 player_version = self._search_regex(
2949 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2950 if player_version:
2951 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2952
2953 def _signature_cache_id(self, example_sig):
2954 """ Return a string representation of a signature """
2955 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2956
2957 @classmethod
2958 def _extract_player_info(cls, player_url):
2959 for player_re in cls._PLAYER_INFO_RE:
2960 id_m = re.search(player_re, player_url)
2961 if id_m:
2962 break
2963 else:
2964 raise ExtractorError('Cannot identify player %r' % player_url)
2965 return id_m.group('id')
2966
2967 def _load_player(self, video_id, player_url, fatal=True):
2968 player_id = self._extract_player_info(player_url)
2969 if player_id not in self._code_cache:
2970 code = self._download_webpage(
2971 player_url, video_id, fatal=fatal,
2972 note='Downloading player ' + player_id,
2973 errnote='Download of %s failed' % player_url)
2974 if code:
2975 self._code_cache[player_id] = code
2976 return self._code_cache.get(player_id)
2977
2978 def _extract_signature_function(self, video_id, player_url, example_sig):
2979 player_id = self._extract_player_info(player_url)
2980
2981 # Read from filesystem cache
2982 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2983 assert os.path.basename(func_id) == func_id
2984
2985 self.write_debug(f'Extracting signature function {func_id}')
2986 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2987
2988 if not cache_spec:
2989 code = self._load_player(video_id, player_url)
2990 if code:
2991 res = self._parse_sig_js(code)
2992 test_string = ''.join(map(chr, range(len(example_sig))))
2993 cache_spec = [ord(c) for c in res(test_string)]
2994 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2995
2996 return lambda s: ''.join(s[i] for i in cache_spec)
2997
2998 def _print_sig_code(self, func, example_sig):
2999 if not self.get_param('youtube_print_sig_code'):
3000 return
3001
3002 def gen_sig_code(idxs):
3003 def _genslice(start, end, step):
3004 starts = '' if start == 0 else str(start)
3005 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
3006 steps = '' if step == 1 else (':%d' % step)
3007 return f's[{starts}{ends}{steps}]'
3008
3009 step = None
3010 # Quelch pyflakes warnings - start will be set when step is set
3011 start = '(Never used)'
3012 for i, prev in zip(idxs[1:], idxs[:-1]):
3013 if step is not None:
3014 if i - prev == step:
3015 continue
3016 yield _genslice(start, prev, step)
3017 step = None
3018 continue
3019 if i - prev in [-1, 1]:
3020 step = i - prev
3021 start = prev
3022 continue
3023 else:
3024 yield 's[%d]' % prev
3025 if step is None:
3026 yield 's[%d]' % i
3027 else:
3028 yield _genslice(start, i, step)
3029
3030 test_string = ''.join(map(chr, range(len(example_sig))))
3031 cache_res = func(test_string)
3032 cache_spec = [ord(c) for c in cache_res]
3033 expr_code = ' + '.join(gen_sig_code(cache_spec))
3034 signature_id_tuple = '(%s)' % (
3035 ', '.join(str(len(p)) for p in example_sig.split('.')))
3036 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
3037 ' return %s\n') % (signature_id_tuple, expr_code)
3038 self.to_screen('Extracted signature function:\n' + code)
3039
3040 def _parse_sig_js(self, jscode):
3041 funcname = self._search_regex(
3042 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3043 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3044 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3045 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3046 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}\(a,\d+\))?',
3047 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3048 # Obsolete patterns
3049 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3050 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3051 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3052 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3053 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3054 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3055 jscode, 'Initial JS player signature function name', group='sig')
3056
3057 jsi = JSInterpreter(jscode)
3058 initial_function = jsi.extract_function(funcname)
3059 return lambda s: initial_function([s])
3060
3061 def _cached(self, func, *cache_id):
3062 def inner(*args, **kwargs):
3063 if cache_id not in self._player_cache:
3064 try:
3065 self._player_cache[cache_id] = func(*args, **kwargs)
3066 except ExtractorError as e:
3067 self._player_cache[cache_id] = e
3068 except Exception as e:
3069 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3070
3071 ret = self._player_cache[cache_id]
3072 if isinstance(ret, Exception):
3073 raise ret
3074 return ret
3075 return inner
3076
3077 def _decrypt_signature(self, s, video_id, player_url):
3078 """Turn the encrypted s field into a working signature"""
3079 extract_sig = self._cached(
3080 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3081 func = extract_sig(video_id, player_url, s)
3082 self._print_sig_code(func, s)
3083 return func(s)
3084
3085 def _decrypt_nsig(self, s, video_id, player_url):
3086 """Turn the encrypted n field into a working signature"""
3087 if player_url is None:
3088 raise ExtractorError('Cannot decrypt nsig without player_url')
3089 player_url = urljoin('https://www.youtube.com', player_url)
3090
3091 try:
3092 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3093 except ExtractorError as e:
3094 raise ExtractorError('Unable to extract nsig function code', cause=e)
3095 if self.get_param('youtube_print_sig_code'):
3096 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3097
3098 try:
3099 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3100 ret = extract_nsig(jsi, func_code)(s)
3101 except JSInterpreter.Exception as e:
3102 try:
3103 jsi = PhantomJSwrapper(self, timeout=5000)
3104 except ExtractorError:
3105 raise e
3106 self.report_warning(
3107 f'Native nsig extraction failed: Trying with PhantomJS\n'
3108 f' n = {s} ; player = {player_url}', video_id)
3109 self.write_debug(e, only_once=True)
3110
3111 args, func_body = func_code
3112 ret = jsi.execute(
3113 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3114 video_id=video_id, note='Executing signature code').strip()
3115
3116 self.write_debug(f'Decrypted nsig {s} => {ret}')
3117 return ret
3118
3119 def _extract_n_function_name(self, jscode):
3120 funcname, idx = self._search_regex(
3121 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3122 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3123 if not idx:
3124 return funcname
3125
3126 return json.loads(js_to_json(self._search_regex(
3127 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
3128 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3129
3130 def _extract_n_function_code(self, video_id, player_url):
3131 player_id = self._extract_player_info(player_url)
3132 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3133 jscode = func_code or self._load_player(video_id, player_url)
3134 jsi = JSInterpreter(jscode)
3135
3136 if func_code:
3137 return jsi, player_id, func_code
3138
3139 func_name = self._extract_n_function_name(jscode)
3140
3141 # For redundancy
3142 func_code = self._search_regex(
3143 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3144 # NB: The end of the regex is intentionally kept strict
3145 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3146 jscode, 'nsig function', group=('var', 'code'), default=None)
3147 if func_code:
3148 func_code = ([func_code[0]], func_code[1])
3149 else:
3150 self.write_debug('Extracting nsig function with jsinterp')
3151 func_code = jsi.extract_function_code(func_name)
3152
3153 self.cache.store('youtube-nsig', player_id, func_code)
3154 return jsi, player_id, func_code
3155
3156 def _extract_n_function_from_code(self, jsi, func_code):
3157 func = jsi.extract_function_from_code(*func_code)
3158
3159 def extract_nsig(s):
3160 try:
3161 ret = func([s])
3162 except JSInterpreter.Exception:
3163 raise
3164 except Exception as e:
3165 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3166
3167 if ret.startswith('enhanced_except_'):
3168 raise JSInterpreter.Exception('Signature function returned an exception')
3169 return ret
3170
3171 return extract_nsig
3172
3173 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3174 """
3175 Extract signatureTimestamp (sts)
3176 Required to tell API what sig/player version is in use.
3177 """
3178 sts = None
3179 if isinstance(ytcfg, dict):
3180 sts = int_or_none(ytcfg.get('STS'))
3181
3182 if not sts:
3183 # Attempt to extract from player
3184 if player_url is None:
3185 error_msg = 'Cannot extract signature timestamp without player_url.'
3186 if fatal:
3187 raise ExtractorError(error_msg)
3188 self.report_warning(error_msg)
3189 return
3190 code = self._load_player(video_id, player_url, fatal=fatal)
3191 if code:
3192 sts = int_or_none(self._search_regex(
3193 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3194 'JS player signature timestamp', group='sts', fatal=fatal))
3195 return sts
3196
3197 def _mark_watched(self, video_id, player_responses):
3198 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3199 label = 'fully ' if is_full else ''
3200 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3201 expected_type=url_or_none)
3202 if not url:
3203 self.report_warning(f'Unable to mark {label}watched')
3204 return
3205 parsed_url = urllib.parse.urlparse(url)
3206 qs = urllib.parse.parse_qs(parsed_url.query)
3207
3208 # cpn generation algorithm is reverse engineered from base.js.
3209 # In fact it works even with dummy cpn.
3210 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3211 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3212
3213 # # more consistent results setting it to right before the end
3214 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3215
3216 qs.update({
3217 'ver': ['2'],
3218 'cpn': [cpn],
3219 'cmt': video_length,
3220 'el': 'detailpage', # otherwise defaults to "shorts"
3221 })
3222
3223 if is_full:
3224 # these seem to mark watchtime "history" in the real world
3225 # they're required, so send in a single value
3226 qs.update({
3227 'st': 0,
3228 'et': video_length,
3229 })
3230
3231 url = urllib.parse.urlunparse(
3232 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3233
3234 self._download_webpage(
3235 url, video_id, f'Marking {label}watched',
3236 'Unable to mark watched', fatal=False)
3237
3238 @classmethod
3239 def _extract_from_webpage(cls, url, webpage):
3240 # Invidious Instances
3241 # https://github.com/yt-dlp/yt-dlp/issues/195
3242 # https://github.com/iv-org/invidious/pull/1730
3243 mobj = re.search(
3244 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3245 webpage)
3246 if mobj:
3247 yield cls.url_result(mobj.group('url'), cls)
3248 raise cls.StopExtraction()
3249
3250 yield from super()._extract_from_webpage(url, webpage)
3251
3252 # lazyYT YouTube embed
3253 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3254 yield cls.url_result(unescapeHTML(id_), cls, id_)
3255
3256 # Wordpress "YouTube Video Importer" plugin
3257 for m in re.findall(r'''(?x)<div[^>]+
3258 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3259 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3260 yield cls.url_result(m[-1], cls, m[-1])
3261
3262 @classmethod
3263 def extract_id(cls, url):
3264 video_id = cls.get_temp_id(url)
3265 if not video_id:
3266 raise ExtractorError(f'Invalid URL: {url}')
3267 return video_id
3268
3269 def _extract_chapters_from_json(self, data, duration):
3270 chapter_list = traverse_obj(
3271 data, (
3272 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3273 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3274 ), expected_type=list)
3275
3276 return self._extract_chapters_helper(
3277 chapter_list,
3278 start_function=lambda chapter: float_or_none(
3279 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3280 title_function=lambda chapter: traverse_obj(
3281 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3282 duration=duration)
3283
3284 def _extract_chapters_from_engagement_panel(self, data, duration):
3285 content_list = traverse_obj(
3286 data,
3287 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3288 expected_type=list)
3289 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3290 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3291
3292 return next(filter(None, (
3293 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3294 chapter_time, chapter_title, duration)
3295 for contents in content_list)), [])
3296
3297 def _extract_heatmap(self, data):
3298 return traverse_obj(data, (
3299 'frameworkUpdates', 'entityBatchUpdate', 'mutations',
3300 lambda _, v: v['payload']['macroMarkersListEntity']['markersList']['markerType'] == 'MARKER_TYPE_HEATMAP',
3301 'payload', 'macroMarkersListEntity', 'markersList', 'markers', ..., {
3302 'start_time': ('startMillis', {functools.partial(float_or_none, scale=1000)}),
3303 'end_time': {lambda x: (int(x['startMillis']) + int(x['durationMillis'])) / 1000},
3304 'value': ('intensityScoreNormalized', {float_or_none}),
3305 })) or None
3306
3307 def _extract_comment(self, comment_renderer, parent=None):
3308 comment_id = comment_renderer.get('commentId')
3309 if not comment_id:
3310 return
3311
3312 info = {
3313 'id': comment_id,
3314 'text': self._get_text(comment_renderer, 'contentText'),
3315 'like_count': self._get_count(comment_renderer, 'voteCount'),
3316 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3317 'author': self._get_text(comment_renderer, 'authorText'),
3318 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3319 'parent': parent or 'root',
3320 }
3321
3322 # Timestamp is an estimate calculated from the current time and time_text
3323 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3324 timestamp = self._parse_time_text(time_text)
3325
3326 info.update({
3327 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3328 '_time_text': time_text,
3329 'timestamp': timestamp,
3330 })
3331
3332 info['author_url'] = urljoin(
3333 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3334 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3335 expected_type=str, get_all=False))
3336
3337 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3338 if author_is_uploader is not None:
3339 info['author_is_uploader'] = author_is_uploader
3340
3341 comment_abr = traverse_obj(
3342 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3343 if comment_abr is not None:
3344 info['is_favorited'] = 'creatorHeart' in comment_abr
3345
3346 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3347 if self._has_badge(badges, BadgeType.VERIFIED):
3348 info['author_is_verified'] = True
3349
3350 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3351 if is_pinned:
3352 info['is_pinned'] = True
3353
3354 return info
3355
3356 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3357
3358 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3359
3360 def extract_header(contents):
3361 _continuation = None
3362 for content in contents:
3363 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3364 expected_comment_count = self._get_count(
3365 comments_header_renderer, 'countText', 'commentsCount')
3366
3367 if expected_comment_count is not None:
3368 tracker['est_total'] = expected_comment_count
3369 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3370 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3371
3372 sort_menu_item = try_get(
3373 comments_header_renderer,
3374 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3375 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3376
3377 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3378 if not _continuation:
3379 continue
3380
3381 sort_text = str_or_none(sort_menu_item.get('title'))
3382 if not sort_text:
3383 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3384 self.to_screen('Sorting comments by %s' % sort_text.lower())
3385 break
3386 return _continuation
3387
3388 def extract_thread(contents):
3389 if not parent:
3390 tracker['current_page_thread'] = 0
3391 for content in contents:
3392 if not parent and tracker['total_parent_comments'] >= max_parents:
3393 yield
3394 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3395 comment_renderer = get_first(
3396 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3397 expected_type=dict, default={})
3398
3399 comment = self._extract_comment(comment_renderer, parent)
3400 if not comment:
3401 continue
3402 comment_id = comment['id']
3403 if comment.get('is_pinned'):
3404 tracker['pinned_comment_ids'].add(comment_id)
3405 # Sometimes YouTube may break and give us infinite looping comments.
3406 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3407 if comment_id in tracker['seen_comment_ids']:
3408 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3409 # Pinned comments may appear a second time in newest first sort
3410 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3411 continue
3412 self.report_warning(
3413 'Detected YouTube comments looping. Stopping comment extraction '
3414 f'{"for this thread" if parent else ""} as we probably cannot get any more.')
3415 yield
3416 else:
3417 tracker['seen_comment_ids'].add(comment['id'])
3418
3419 tracker['running_total'] += 1
3420 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3421 yield comment
3422
3423 # Attempt to get the replies
3424 comment_replies_renderer = try_get(
3425 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3426
3427 if comment_replies_renderer:
3428 tracker['current_page_thread'] += 1
3429 comment_entries_iter = self._comment_entries(
3430 comment_replies_renderer, ytcfg, video_id,
3431 parent=comment.get('id'), tracker=tracker)
3432 yield from itertools.islice(comment_entries_iter, min(
3433 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3434
3435 # Keeps track of counts across recursive calls
3436 if not tracker:
3437 tracker = dict(
3438 running_total=0,
3439 est_total=None,
3440 current_page_thread=0,
3441 total_parent_comments=0,
3442 total_reply_comments=0,
3443 seen_comment_ids=set(),
3444 pinned_comment_ids=set()
3445 )
3446
3447 # TODO: Deprecated
3448 # YouTube comments have a max depth of 2
3449 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3450 if max_depth:
3451 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3452 'Set max replies in the max-comments extractor argument instead')
3453 if max_depth == 1 and parent:
3454 return
3455
3456 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3457 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3458
3459 continuation = self._extract_continuation(root_continuation_data)
3460
3461 response = None
3462 is_forced_continuation = False
3463 is_first_continuation = parent is None
3464 if is_first_continuation and not continuation:
3465 # Sometimes you can get comments by generating the continuation yourself,
3466 # even if YouTube initially reports them being disabled - e.g. stories comments.
3467 # Note: if the comment section is actually disabled, YouTube may return a response with
3468 # required check_get_keys missing. So we will disable that check initially in this case.
3469 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3470 is_forced_continuation = True
3471
3472 continuation_items_path = (
3473 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3474 for page_num in itertools.count(0):
3475 if not continuation:
3476 break
3477 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3478 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3479 if page_num == 0:
3480 if is_first_continuation:
3481 note_prefix = 'Downloading comment section API JSON'
3482 else:
3483 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3484 tracker['current_page_thread'], comment_prog_str)
3485 else:
3486 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3487 ' ' if parent else '', ' replies' if parent else '',
3488 page_num, comment_prog_str)
3489
3490 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3491 # Ignore check if YouTube says the comment count is 0.
3492 check_get_keys = None
3493 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3494 check_get_keys = [[*continuation_items_path, ..., (
3495 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
3496 try:
3497 response = self._extract_response(
3498 item_id=None, query=continuation,
3499 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3500 check_get_keys=check_get_keys)
3501 except ExtractorError as e:
3502 # Ignore incomplete data error for replies if retries didn't work.
3503 # This is to allow any other parent comments and comment threads to be downloaded.
3504 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3505 if 'incomplete data' in str(e).lower() and parent:
3506 if self.get_param('ignoreerrors') in (True, 'only_download'):
3507 self.report_warning(
3508 'Received incomplete data for a comment reply thread and retrying did not help. '
3509 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
3510 return
3511 else:
3512 raise ExtractorError(
3513 'Incomplete data received for comment reply thread. '
3514 'Pass --ignore-errors to ignore and allow rest of comments to download.',
3515 expected=True)
3516 raise
3517 is_forced_continuation = False
3518 continuation = None
3519 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3520 if is_first_continuation:
3521 continuation = extract_header(continuation_items)
3522 is_first_continuation = False
3523 if continuation:
3524 break
3525 continue
3526
3527 for entry in extract_thread(continuation_items):
3528 if not entry:
3529 return
3530 yield entry
3531 continuation = self._extract_continuation({'contents': continuation_items})
3532 if continuation:
3533 break
3534
3535 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3536 if message and not parent and tracker['running_total'] == 0:
3537 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3538 raise self.CommentsDisabled
3539
3540 @staticmethod
3541 def _generate_comment_continuation(video_id):
3542 """
3543 Generates initial comment section continuation token from given video id
3544 """
3545 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3546 return base64.b64encode(token.encode()).decode()
3547
3548 def _get_comments(self, ytcfg, video_id, contents, webpage):
3549 """Entry for comment extraction"""
3550 def _real_comment_extract(contents):
3551 renderer = next((
3552 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3553 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3554 yield from self._comment_entries(renderer, ytcfg, video_id)
3555
3556 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3557 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3558
3559 @staticmethod
3560 def _get_checkok_params():
3561 return {'contentCheckOk': True, 'racyCheckOk': True}
3562
3563 @classmethod
3564 def _generate_player_context(cls, sts=None):
3565 context = {
3566 'html5Preference': 'HTML5_PREF_WANTS',
3567 }
3568 if sts is not None:
3569 context['signatureTimestamp'] = sts
3570 return {
3571 'playbackContext': {
3572 'contentPlaybackContext': context
3573 },
3574 **cls._get_checkok_params()
3575 }
3576
3577 @staticmethod
3578 def _is_agegated(player_response):
3579 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3580 return True
3581
3582 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3583 AGE_GATE_REASONS = (
3584 'confirm your age', 'age-restricted', 'inappropriate', # reason
3585 'age_verification_required', 'age_check_required', # status
3586 )
3587 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3588
3589 @staticmethod
3590 def _is_unplayable(player_response):
3591 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3592
3593 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3594
3595 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3596 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3597 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3598 headers = self.generate_api_headers(
3599 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3600
3601 yt_query = {
3602 'videoId': video_id,
3603 }
3604 if _split_innertube_client(client)[0] == 'android':
3605 yt_query['params'] = 'CgIQBg=='
3606
3607 pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0]
3608 if pp_arg:
3609 yt_query['params'] = pp_arg
3610
3611 yt_query.update(self._generate_player_context(sts))
3612 return self._extract_response(
3613 item_id=video_id, ep='player', query=yt_query,
3614 ytcfg=player_ytcfg, headers=headers, fatal=True,
3615 default_client=client,
3616 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3617 ) or None
3618
3619 def _get_requested_clients(self, url, smuggled_data):
3620 requested_clients = []
3621 default = ['ios', 'android', 'web']
3622 allowed_clients = sorted(
3623 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3624 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3625 for client in self._configuration_arg('player_client'):
3626 if client in allowed_clients:
3627 requested_clients.append(client)
3628 elif client == 'default':
3629 requested_clients.extend(default)
3630 elif client == 'all':
3631 requested_clients.extend(allowed_clients)
3632 else:
3633 self.report_warning(f'Skipping unsupported client {client}')
3634 if not requested_clients:
3635 requested_clients = default
3636
3637 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3638 requested_clients.extend(
3639 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3640
3641 return orderedSet(requested_clients)
3642
3643 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3644 initial_pr = None
3645 if webpage:
3646 initial_pr = self._search_json(
3647 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3648
3649 all_clients = set(clients)
3650 clients = clients[::-1]
3651 prs = []
3652
3653 def append_client(*client_names):
3654 """ Append the first client name that exists but not already used """
3655 for client_name in client_names:
3656 actual_client = _split_innertube_client(client_name)[0]
3657 if actual_client in INNERTUBE_CLIENTS:
3658 if actual_client not in all_clients:
3659 clients.append(client_name)
3660 all_clients.add(actual_client)
3661 return
3662
3663 # Android player_response does not have microFormats which are needed for
3664 # extraction of some data. So we return the initial_pr with formats
3665 # stripped out even if not requested by the user
3666 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3667 if initial_pr:
3668 pr = dict(initial_pr)
3669 pr['streamingData'] = None
3670 prs.append(pr)
3671
3672 last_error = None
3673 tried_iframe_fallback = False
3674 player_url = None
3675 while clients:
3676 client, base_client, variant = _split_innertube_client(clients.pop())
3677 player_ytcfg = master_ytcfg if client == 'web' else {}
3678 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3679 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3680
3681 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3682 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3683 if 'js' in self._configuration_arg('player_skip'):
3684 require_js_player = False
3685 player_url = None
3686
3687 if not player_url and not tried_iframe_fallback and require_js_player:
3688 player_url = self._download_player_url(video_id)
3689 tried_iframe_fallback = True
3690
3691 try:
3692 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3693 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3694 except ExtractorError as e:
3695 if last_error:
3696 self.report_warning(last_error)
3697 last_error = e
3698 continue
3699
3700 if pr:
3701 # YouTube may return a different video player response than expected.
3702 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3703 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3704 if pr_video_id and pr_video_id != video_id:
3705 self.report_warning(
3706 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3707 else:
3708 # Save client name for introspection later
3709 name = short_client_name(client)
3710 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3711 sd[STREAMING_DATA_CLIENT_NAME] = name
3712 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3713 f[STREAMING_DATA_CLIENT_NAME] = name
3714 prs.append(pr)
3715
3716 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3717 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3718 append_client(f'{base_client}_creator')
3719 elif self._is_agegated(pr):
3720 if variant == 'tv_embedded':
3721 append_client(f'{base_client}_embedded')
3722 elif not variant:
3723 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3724
3725 if last_error:
3726 if not len(prs):
3727 raise last_error
3728 self.report_warning(last_error)
3729 return prs, player_url
3730
3731 def _needs_live_processing(self, live_status, duration):
3732 if (live_status == 'is_live' and self.get_param('live_from_start')
3733 or live_status == 'post_live' and (duration or 0) > 2 * 3600):
3734 return live_status
3735
3736 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3737 CHUNK_SIZE = 10 << 20
3738 itags, stream_ids = collections.defaultdict(set), []
3739 itag_qualities, res_qualities = {}, {0: None}
3740 q = qualities([
3741 # Normally tiny is the smallest video-only formats. But
3742 # audio-only formats with unknown quality may get tagged as tiny
3743 'tiny',
3744 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3745 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3746 ])
3747 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3748 format_types = self._configuration_arg('formats')
3749 all_formats = 'duplicate' in format_types
3750 if self._configuration_arg('include_duplicate_formats'):
3751 all_formats = True
3752 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
3753 'Use formats=duplicate extractor argument instead')
3754
3755 def build_fragments(f):
3756 return LazyList({
3757 'url': update_url_query(f['url'], {
3758 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3759 })
3760 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3761
3762 for fmt in streaming_formats:
3763 if fmt.get('targetDurationSec'):
3764 continue
3765
3766 itag = str_or_none(fmt.get('itag'))
3767 audio_track = fmt.get('audioTrack') or {}
3768 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3769 if not all_formats:
3770 if stream_id in stream_ids:
3771 continue
3772
3773 quality = fmt.get('quality')
3774 height = int_or_none(fmt.get('height'))
3775 if quality == 'tiny' or not quality:
3776 quality = fmt.get('audioQuality', '').lower() or quality
3777 # The 3gp format (17) in android client has a quality of "small",
3778 # but is actually worse than other formats
3779 if itag == '17':
3780 quality = 'tiny'
3781 if quality:
3782 if itag:
3783 itag_qualities[itag] = quality
3784 if height:
3785 res_qualities[height] = quality
3786 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3787 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3788 # number of fragment that would subsequently requested with (`&sq=N`)
3789 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3790 continue
3791
3792 fmt_url = fmt.get('url')
3793 if not fmt_url:
3794 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3795 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3796 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3797 if not all((sc, fmt_url, player_url, encrypted_sig)):
3798 continue
3799 try:
3800 fmt_url += '&%s=%s' % (
3801 traverse_obj(sc, ('sp', -1)) or 'signature',
3802 self._decrypt_signature(encrypted_sig, video_id, player_url)
3803 )
3804 except ExtractorError as e:
3805 self.report_warning('Signature extraction failed: Some formats may be missing',
3806 video_id=video_id, only_once=True)
3807 self.write_debug(e, only_once=True)
3808 continue
3809
3810 query = parse_qs(fmt_url)
3811 throttled = False
3812 if query.get('n'):
3813 try:
3814 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3815 fmt_url = update_url_query(fmt_url, {
3816 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3817 })
3818 except ExtractorError as e:
3819 phantomjs_hint = ''
3820 if isinstance(e, JSInterpreter.Exception):
3821 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3822 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3823 if player_url:
3824 self.report_warning(
3825 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3826 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3827 self.write_debug(e, only_once=True)
3828 else:
3829 self.report_warning(
3830 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3831 video_id=video_id, only_once=True)
3832 throttled = True
3833
3834 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3835 language_preference = (
3836 10 if audio_track.get('audioIsDefault') and 10
3837 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3838 else -1)
3839 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3840 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3841 # Make sure to avoid false positives with small duration differences.
3842 # E.g. __2ABJjxzNo, ySuUZEjARPY
3843 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3844 if is_damaged:
3845 self.report_warning(
3846 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3847
3848 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3849 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
3850 fps = int_or_none(fmt.get('fps')) or 0
3851 dct = {
3852 'asr': int_or_none(fmt.get('audioSampleRate')),
3853 'filesize': int_or_none(fmt.get('contentLength')),
3854 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3855 'format_note': join_nonempty(
3856 join_nonempty(audio_track.get('displayName'),
3857 language_preference > 0 and ' (default)', delim=''),
3858 name, fmt.get('isDrc') and 'DRC',
3859 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3860 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3861 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3862 (self.get_param('verbose') or all_formats) and client_name,
3863 delim=', '),
3864 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3865 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
3866 + (100 if 'Premium' in name else 0)),
3867 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
3868 'audio_channels': fmt.get('audioChannels'),
3869 'height': height,
3870 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3871 'has_drm': bool(fmt.get('drmFamilies')),
3872 'tbr': tbr,
3873 'url': fmt_url,
3874 'width': int_or_none(fmt.get('width')),
3875 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3876 'desc' if language_preference < -1 else '') or None,
3877 'language_preference': language_preference,
3878 # Strictly de-prioritize damaged and 3gp formats
3879 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3880 }
3881 mime_mobj = re.match(
3882 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3883 if mime_mobj:
3884 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3885 dct.update(parse_codecs(mime_mobj.group(2)))
3886 if itag:
3887 itags[itag].add(('https', dct.get('language')))
3888 stream_ids.append(stream_id)
3889 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3890 if single_stream and dct.get('ext'):
3891 dct['container'] = dct['ext'] + '_dash'
3892
3893 if (all_formats or 'dashy' in format_types) and dct['filesize']:
3894 yield {
3895 **dct,
3896 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3897 'protocol': 'http_dash_segments',
3898 'fragments': build_fragments(dct),
3899 }
3900 if all_formats or 'dashy' not in format_types:
3901 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3902 yield dct
3903
3904 needs_live_processing = self._needs_live_processing(live_status, duration)
3905 skip_bad_formats = 'incomplete' not in format_types
3906 if self._configuration_arg('include_incomplete_formats'):
3907 skip_bad_formats = False
3908 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
3909 'Use formats=incomplete extractor argument instead')
3910
3911 skip_manifests = set(self._configuration_arg('skip'))
3912 if (not self.get_param('youtube_include_hls_manifest', True)
3913 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3914 or needs_live_processing and skip_bad_formats):
3915 skip_manifests.add('hls')
3916
3917 if not self.get_param('youtube_include_dash_manifest', True):
3918 skip_manifests.add('dash')
3919 if self._configuration_arg('include_live_dash'):
3920 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3921 'Use formats=incomplete extractor argument instead')
3922 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3923 skip_manifests.add('dash')
3924
3925 def process_manifest_format(f, proto, client_name, itag):
3926 key = (proto, f.get('language'))
3927 if not all_formats and key in itags[itag]:
3928 return False
3929 itags[itag].add(key)
3930
3931 if itag and all_formats:
3932 f['format_id'] = f'{itag}-{proto}'
3933 elif any(p != proto for p, _ in itags[itag]):
3934 f['format_id'] = f'{itag}-{proto}'
3935 elif itag:
3936 f['format_id'] = itag
3937
3938 if f.get('source_preference') is None:
3939 f['source_preference'] = -1
3940
3941 if itag in ('616', '235'):
3942 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
3943 f['source_preference'] += 100
3944
3945 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3946 if f['quality'] == -1 and f.get('height'):
3947 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3948 if self.get_param('verbose') or all_formats:
3949 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3950 if f.get('fps') and f['fps'] <= 1:
3951 del f['fps']
3952
3953 if proto == 'hls' and f.get('has_drm'):
3954 f['has_drm'] = 'maybe'
3955 f['source_preference'] -= 5
3956 return True
3957
3958 subtitles = {}
3959 for sd in streaming_data:
3960 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3961
3962 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3963 if hls_manifest_url:
3964 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3965 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3966 subtitles = self._merge_subtitles(subs, subtitles)
3967 for f in fmts:
3968 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3969 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3970 yield f
3971
3972 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3973 if dash_manifest_url:
3974 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3975 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3976 for f in formats:
3977 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3978 f['filesize'] = int_or_none(self._search_regex(
3979 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3980 if needs_live_processing:
3981 f['is_from_start'] = True
3982
3983 yield f
3984 yield subtitles
3985
3986 def _extract_storyboard(self, player_responses, duration):
3987 spec = get_first(
3988 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3989 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3990 if not base_url:
3991 return
3992 L = len(spec) - 1
3993 for i, args in enumerate(spec):
3994 args = args.split('#')
3995 counts = list(map(int_or_none, args[:5]))
3996 if len(args) != 8 or not all(counts):
3997 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3998 continue
3999 width, height, frame_count, cols, rows = counts
4000 N, sigh = args[6:]
4001
4002 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
4003 fragment_count = frame_count / (cols * rows)
4004 fragment_duration = duration / fragment_count
4005 yield {
4006 'format_id': f'sb{i}',
4007 'format_note': 'storyboard',
4008 'ext': 'mhtml',
4009 'protocol': 'mhtml',
4010 'acodec': 'none',
4011 'vcodec': 'none',
4012 'url': url,
4013 'width': width,
4014 'height': height,
4015 'fps': frame_count / duration,
4016 'rows': rows,
4017 'columns': cols,
4018 'fragments': [{
4019 'url': url.replace('$M', str(j)),
4020 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4021 } for j in range(math.ceil(fragment_count))],
4022 }
4023
4024 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4025 webpage = None
4026 if 'webpage' not in self._configuration_arg('player_skip'):
4027 query = {'bpctr': '9999999999', 'has_verified': '1'}
4028 pp = self._configuration_arg('player_params', [None], casesense=True)[0]
4029 if pp:
4030 query['pp'] = pp
4031 webpage = self._download_webpage(
4032 webpage_url, video_id, fatal=False, query=query)
4033
4034 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4035
4036 player_responses, player_url = self._extract_player_responses(
4037 self._get_requested_clients(url, smuggled_data),
4038 video_id, webpage, master_ytcfg, smuggled_data)
4039
4040 return webpage, master_ytcfg, player_responses, player_url
4041
4042 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4043 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4044 is_live = get_first(video_details, 'isLive')
4045 if is_live is None:
4046 is_live = get_first(live_broadcast_details, 'isLiveNow')
4047 live_content = get_first(video_details, 'isLiveContent')
4048 is_upcoming = get_first(video_details, 'isUpcoming')
4049 post_live = get_first(video_details, 'isPostLiveDvr')
4050 live_status = ('post_live' if post_live
4051 else 'is_live' if is_live
4052 else 'is_upcoming' if is_upcoming
4053 else 'was_live' if live_content
4054 else 'not_live' if False in (is_live, live_content)
4055 else None)
4056 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4057 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4058 if all(f.get('has_drm') for f in formats):
4059 # If there are no formats that definitely don't have DRM, all have DRM
4060 for f in formats:
4061 f['has_drm'] = True
4062
4063 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4064
4065 def _real_extract(self, url):
4066 url, smuggled_data = unsmuggle_url(url, {})
4067 video_id = self._match_id(url)
4068
4069 base_url = self.http_scheme() + '//www.youtube.com/'
4070 webpage_url = base_url + 'watch?v=' + video_id
4071
4072 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4073
4074 playability_statuses = traverse_obj(
4075 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4076
4077 trailer_video_id = get_first(
4078 playability_statuses,
4079 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4080 expected_type=str)
4081 if trailer_video_id:
4082 return self.url_result(
4083 trailer_video_id, self.ie_key(), trailer_video_id)
4084
4085 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4086 if webpage else (lambda x: None))
4087
4088 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4089 microformats = traverse_obj(
4090 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4091 expected_type=dict)
4092
4093 translated_title = self._get_text(microformats, (..., 'title'))
4094 video_title = (self._preferred_lang and translated_title
4095 or get_first(video_details, 'title') # primary
4096 or translated_title
4097 or search_meta(['og:title', 'twitter:title', 'title']))
4098 translated_description = self._get_text(microformats, (..., 'description'))
4099 original_description = get_first(video_details, 'shortDescription')
4100 video_description = (
4101 self._preferred_lang and translated_description
4102 # If original description is blank, it will be an empty string.
4103 # Do not prefer translated description in this case.
4104 or original_description if original_description is not None else translated_description)
4105
4106 multifeed_metadata_list = get_first(
4107 player_responses,
4108 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4109 expected_type=str)
4110 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4111 if self.get_param('noplaylist'):
4112 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4113 else:
4114 entries = []
4115 feed_ids = []
4116 for feed in multifeed_metadata_list.split(','):
4117 # Unquote should take place before split on comma (,) since textual
4118 # fields may contain comma as well (see
4119 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4120 feed_data = urllib.parse.parse_qs(
4121 urllib.parse.unquote_plus(feed))
4122
4123 def feed_entry(name):
4124 return try_get(
4125 feed_data, lambda x: x[name][0], str)
4126
4127 feed_id = feed_entry('id')
4128 if not feed_id:
4129 continue
4130 feed_title = feed_entry('title')
4131 title = video_title
4132 if feed_title:
4133 title += ' (%s)' % feed_title
4134 entries.append({
4135 '_type': 'url_transparent',
4136 'ie_key': 'Youtube',
4137 'url': smuggle_url(
4138 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4139 {'force_singlefeed': True}),
4140 'title': title,
4141 })
4142 feed_ids.append(feed_id)
4143 self.to_screen(
4144 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4145 % (', '.join(feed_ids), video_id))
4146 return self.playlist_result(
4147 entries, video_id, video_title, video_description)
4148
4149 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4150 or int_or_none(get_first(microformats, 'lengthSeconds'))
4151 or parse_duration(search_meta('duration')) or None)
4152
4153 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4154 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4155 if live_status == 'post_live':
4156 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4157
4158 if not formats:
4159 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4160 self.report_drm(video_id)
4161 pemr = get_first(
4162 playability_statuses,
4163 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4164 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4165 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4166 if subreason:
4167 if subreason == 'The uploader has not made this video available in your country.':
4168 countries = get_first(microformats, 'availableCountries')
4169 if not countries:
4170 regions_allowed = search_meta('regionsAllowed')
4171 countries = regions_allowed.split(',') if regions_allowed else None
4172 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4173 reason += f'. {subreason}'
4174 if reason:
4175 self.raise_no_formats(reason, expected=True)
4176
4177 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4178 if not keywords and webpage:
4179 keywords = [
4180 unescapeHTML(m.group('content'))
4181 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4182 for keyword in keywords:
4183 if keyword.startswith('yt:stretch='):
4184 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4185 if mobj:
4186 # NB: float is intentional for forcing float division
4187 w, h = (float(v) for v in mobj.groups())
4188 if w > 0 and h > 0:
4189 ratio = w / h
4190 for f in formats:
4191 if f.get('vcodec') != 'none':
4192 f['stretched_ratio'] = ratio
4193 break
4194 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4195 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4196 if thumbnail_url:
4197 thumbnails.append({
4198 'url': thumbnail_url,
4199 })
4200 original_thumbnails = thumbnails.copy()
4201
4202 # The best resolution thumbnails sometimes does not appear in the webpage
4203 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4204 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4205 thumbnail_names = [
4206 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4207 # in resolution, these are not the custom thumbnail. So de-prioritize them
4208 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4209 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4210 ]
4211 n_thumbnail_names = len(thumbnail_names)
4212 thumbnails.extend({
4213 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4214 video_id=video_id, name=name, ext=ext,
4215 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4216 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4217 for thumb in thumbnails:
4218 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4219 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4220 self._remove_duplicate_formats(thumbnails)
4221 self._downloader._sort_thumbnails(original_thumbnails)
4222
4223 category = get_first(microformats, 'category') or search_meta('genre')
4224 channel_id = self.ucid_or_none(str_or_none(
4225 get_first(video_details, 'channelId')
4226 or get_first(microformats, 'externalChannelId')
4227 or search_meta('channelId')))
4228 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4229
4230 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4231 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4232 if not duration and live_end_time and live_start_time:
4233 duration = live_end_time - live_start_time
4234
4235 needs_live_processing = self._needs_live_processing(live_status, duration)
4236
4237 def is_bad_format(fmt):
4238 if needs_live_processing and not fmt.get('is_from_start'):
4239 return True
4240 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4241 and fmt.get('protocol') == 'http_dash_segments'):
4242 return True
4243
4244 for fmt in filter(is_bad_format, formats):
4245 fmt['preference'] = (fmt.get('preference') or -1) - 10
4246 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4247
4248 if needs_live_processing:
4249 self._prepare_live_from_start_formats(
4250 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4251
4252 formats.extend(self._extract_storyboard(player_responses, duration))
4253
4254 channel_handle = self.handle_from_url(owner_profile_url)
4255
4256 info = {
4257 'id': video_id,
4258 'title': video_title,
4259 'formats': formats,
4260 'thumbnails': thumbnails,
4261 # The best thumbnail that we are sure exists. Prevents unnecessary
4262 # URL checking if user don't care about getting the best possible thumbnail
4263 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4264 'description': video_description,
4265 'channel_id': channel_id,
4266 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4267 'duration': duration,
4268 'view_count': int_or_none(
4269 get_first((video_details, microformats), (..., 'viewCount'))
4270 or search_meta('interactionCount')),
4271 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4272 'age_limit': 18 if (
4273 get_first(microformats, 'isFamilySafe') is False
4274 or search_meta('isFamilyFriendly') == 'false'
4275 or search_meta('og:restrictions:age') == '18+') else 0,
4276 'webpage_url': webpage_url,
4277 'categories': [category] if category else None,
4278 'tags': keywords,
4279 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4280 'live_status': live_status,
4281 'release_timestamp': live_start_time,
4282 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4283 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4284 }
4285
4286 subtitles = {}
4287 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4288 if pctr:
4289 def get_lang_code(track):
4290 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4291 or track.get('languageCode'))
4292
4293 # Converted into dicts to remove duplicates
4294 captions = {
4295 get_lang_code(sub): sub
4296 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4297 translation_languages = {
4298 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4299 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4300
4301 def process_language(container, base_url, lang_code, sub_name, query):
4302 lang_subs = container.setdefault(lang_code, [])
4303 for fmt in self._SUBTITLE_FORMATS:
4304 query.update({
4305 'fmt': fmt,
4306 })
4307 lang_subs.append({
4308 'ext': fmt,
4309 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4310 'name': sub_name,
4311 })
4312
4313 # NB: Constructing the full subtitle dictionary is slow
4314 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4315 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4316 for lang_code, caption_track in captions.items():
4317 base_url = caption_track.get('baseUrl')
4318 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4319 if not base_url:
4320 continue
4321 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4322 if caption_track.get('kind') != 'asr':
4323 if not lang_code:
4324 continue
4325 process_language(
4326 subtitles, base_url, lang_code, lang_name, {})
4327 if not caption_track.get('isTranslatable'):
4328 continue
4329 for trans_code, trans_name in translation_languages.items():
4330 if not trans_code:
4331 continue
4332 orig_trans_code = trans_code
4333 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4334 if not get_translated_subs:
4335 continue
4336 trans_code += f'-{lang_code}'
4337 trans_name += format_field(lang_name, None, ' from %s')
4338 if lang_code == f'a-{orig_trans_code}':
4339 # Set audio language based on original subtitles
4340 for f in formats:
4341 if f.get('acodec') != 'none' and not f.get('language'):
4342 f['language'] = orig_trans_code
4343 # Add an "-orig" label to the original language so that it can be distinguished.
4344 # The subs are returned without "-orig" as well for compatibility
4345 process_language(
4346 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4347 # Setting tlang=lang returns damaged subtitles.
4348 process_language(automatic_captions, base_url, trans_code, trans_name,
4349 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4350
4351 info['automatic_captions'] = automatic_captions
4352 info['subtitles'] = subtitles
4353
4354 parsed_url = urllib.parse.urlparse(url)
4355 for component in [parsed_url.fragment, parsed_url.query]:
4356 query = urllib.parse.parse_qs(component)
4357 for k, v in query.items():
4358 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4359 d_k += '_time'
4360 if d_k not in info and k in s_ks:
4361 info[d_k] = parse_duration(query[k][0])
4362
4363 # Youtube Music Auto-generated description
4364 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4365 # XXX: Causes catastrophic backtracking if description has "·"
4366 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4367 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4368 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4369 mobj = re.search(
4370 r'''(?xs)
4371 (?=(?P<track>[^\n·]+))(?P=track)·
4372 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4373 (?=(?P<album>[^\n]+))(?P=album)\n
4374 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4375 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4376 (.+?\nArtist\s*:\s*
4377 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4378 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4379 ''', video_description)
4380 if mobj:
4381 release_year = mobj.group('release_year')
4382 release_date = mobj.group('release_date')
4383 if release_date:
4384 release_date = release_date.replace('-', '')
4385 if not release_year:
4386 release_year = release_date[:4]
4387 info.update({
4388 'album': mobj.group('album'.strip()),
4389 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4390 'track': mobj.group('track').strip(),
4391 'release_date': release_date,
4392 'release_year': int_or_none(release_year),
4393 })
4394
4395 initial_data = None
4396 if webpage:
4397 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4398 if not traverse_obj(initial_data, 'contents'):
4399 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4400 initial_data = None
4401 if not initial_data:
4402 query = {'videoId': video_id}
4403 query.update(self._get_checkok_params())
4404 initial_data = self._extract_response(
4405 item_id=video_id, ep='next', fatal=False,
4406 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4407 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4408 note='Downloading initial data API JSON')
4409
4410 info['comment_count'] = traverse_obj(initial_data, (
4411 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4412 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4413 ), (
4414 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4415 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4416 ), expected_type=self._get_count, get_all=False)
4417
4418 try: # This will error if there is no livechat
4419 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4420 except (KeyError, IndexError, TypeError):
4421 pass
4422 else:
4423 info.setdefault('subtitles', {})['live_chat'] = [{
4424 # url is needed to set cookies
4425 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4426 'video_id': video_id,
4427 'ext': 'json',
4428 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4429 else 'youtube_live_chat_replay'),
4430 }]
4431
4432 if initial_data:
4433 info['chapters'] = (
4434 self._extract_chapters_from_json(initial_data, duration)
4435 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4436 or self._extract_chapters_from_description(video_description, duration)
4437 or None)
4438
4439 info['heatmap'] = self._extract_heatmap(initial_data)
4440
4441 contents = traverse_obj(
4442 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4443 expected_type=list, default=[])
4444
4445 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4446 if vpir:
4447 stl = vpir.get('superTitleLink')
4448 if stl:
4449 stl = self._get_text(stl)
4450 if try_get(
4451 vpir,
4452 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4453 info['location'] = stl
4454 else:
4455 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4456 if mobj:
4457 info.update({
4458 'series': mobj.group(1),
4459 'season_number': int(mobj.group(2)),
4460 'episode_number': int(mobj.group(3)),
4461 })
4462 for tlb in (try_get(
4463 vpir,
4464 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4465 list) or []):
4466 tbrs = variadic(
4467 traverse_obj(
4468 tlb, ('toggleButtonRenderer', ...),
4469 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4470 for tbr in tbrs:
4471 for getter, regex in [(
4472 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4473 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4474 lambda x: x['accessibility'],
4475 lambda x: x['accessibilityData']['accessibilityData'],
4476 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4477 label = (try_get(tbr, getter, dict) or {}).get('label')
4478 if label:
4479 mobj = re.match(regex, label)
4480 if mobj:
4481 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4482 break
4483
4484 info['like_count'] = traverse_obj(vpir, (
4485 'videoActions', 'menuRenderer', 'topLevelButtons', ...,
4486 'segmentedLikeDislikeButtonViewModel', 'likeButtonViewModel', 'likeButtonViewModel',
4487 'toggleButtonViewModel', 'toggleButtonViewModel', 'defaultButtonViewModel',
4488 'buttonViewModel', 'accessibilityText', {parse_count}), get_all=False)
4489
4490 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4491 if vcr:
4492 vc = self._get_count(vcr, 'viewCount')
4493 # Upcoming premieres with waiting count are treated as live here
4494 if vcr.get('isLive'):
4495 info['concurrent_view_count'] = vc
4496 elif info.get('view_count') is None:
4497 info['view_count'] = vc
4498
4499 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4500 if vsir:
4501 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4502 info.update({
4503 'channel': self._get_text(vor, 'title'),
4504 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4505
4506 if not channel_handle:
4507 channel_handle = self.handle_from_url(
4508 traverse_obj(vor, (
4509 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4510 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4511 {str}), get_all=False))
4512
4513 rows = try_get(
4514 vsir,
4515 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4516 list) or []
4517 multiple_songs = False
4518 for row in rows:
4519 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4520 multiple_songs = True
4521 break
4522 for row in rows:
4523 mrr = row.get('metadataRowRenderer') or {}
4524 mrr_title = mrr.get('title')
4525 if not mrr_title:
4526 continue
4527 mrr_title = self._get_text(mrr, 'title')
4528 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4529 if mrr_title == 'License':
4530 info['license'] = mrr_contents_text
4531 elif not multiple_songs:
4532 if mrr_title == 'Album':
4533 info['album'] = mrr_contents_text
4534 elif mrr_title == 'Artist':
4535 info['artist'] = mrr_contents_text
4536 elif mrr_title == 'Song':
4537 info['track'] = mrr_contents_text
4538 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4539 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4540 info['channel_is_verified'] = True
4541
4542 info.update({
4543 'uploader': info.get('channel'),
4544 'uploader_id': channel_handle,
4545 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4546 })
4547 # The upload date for scheduled, live and past live streams / premieres in microformats
4548 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4549 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4550 upload_date = (
4551 unified_strdate(get_first(microformats, 'uploadDate'))
4552 or unified_strdate(search_meta('uploadDate')))
4553 if not upload_date or (
4554 live_status in ('not_live', None)
4555 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4556 ):
4557 upload_date = strftime_or_none(
4558 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4559 info['upload_date'] = upload_date
4560
4561 if upload_date and live_status not in ('is_live', 'post_live', 'is_upcoming'):
4562 # Newly uploaded videos' HLS formats are potentially problematic and need to be checked
4563 upload_datetime = datetime_from_str(upload_date).replace(tzinfo=datetime.timezone.utc)
4564 if upload_datetime >= datetime_from_str('today-2days'):
4565 for fmt in info['formats']:
4566 if fmt.get('protocol') == 'm3u8_native':
4567 fmt['__needs_testing'] = True
4568
4569 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4570 v = info.get(s_k)
4571 if v:
4572 info[d_k] = v
4573
4574 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4575
4576 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4577 or get_first(video_details, 'isPrivate', expected_type=bool))
4578
4579 info['availability'] = (
4580 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4581 else self._availability(
4582 is_private=is_private,
4583 needs_premium=(
4584 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4585 or False if initial_data and is_private is not None else None),
4586 needs_subscription=(
4587 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4588 or False if initial_data and is_private is not None else None),
4589 needs_auth=info['age_limit'] >= 18,
4590 is_unlisted=None if is_private is None else (
4591 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4592 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4593
4594 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4595
4596 self.mark_watched(video_id, player_responses)
4597
4598 return info
4599
4600
4601 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4602 @staticmethod
4603 def passthrough_smuggled_data(func):
4604 def _smuggle(info, smuggled_data):
4605 if info.get('_type') not in ('url', 'url_transparent'):
4606 return info
4607 if smuggled_data.get('is_music_url'):
4608 parsed_url = urllib.parse.urlparse(info['url'])
4609 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4610 smuggled_data.pop('is_music_url')
4611 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4612 if smuggled_data:
4613 info['url'] = smuggle_url(info['url'], smuggled_data)
4614 return info
4615
4616 @functools.wraps(func)
4617 def wrapper(self, url):
4618 url, smuggled_data = unsmuggle_url(url, {})
4619 if self.is_music_url(url):
4620 smuggled_data['is_music_url'] = True
4621 info_dict = func(self, url, smuggled_data)
4622 if smuggled_data:
4623 _smuggle(info_dict, smuggled_data)
4624 if info_dict.get('entries'):
4625 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4626 return info_dict
4627 return wrapper
4628
4629 @staticmethod
4630 def _extract_basic_item_renderer(item):
4631 # Modified from _extract_grid_item_renderer
4632 known_basic_renderers = (
4633 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4634 )
4635 for key, renderer in item.items():
4636 if not isinstance(renderer, dict):
4637 continue
4638 elif key in known_basic_renderers:
4639 return renderer
4640 elif key.startswith('grid') and key.endswith('Renderer'):
4641 return renderer
4642
4643 def _extract_channel_renderer(self, renderer):
4644 channel_id = self.ucid_or_none(renderer['channelId'])
4645 title = self._get_text(renderer, 'title')
4646 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4647 channel_handle = self.handle_from_url(
4648 traverse_obj(renderer, (
4649 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4650 ('browseEndpoint', 'canonicalBaseUrl')),
4651 {str}), get_all=False))
4652 if not channel_handle:
4653 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
4654 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
4655 return {
4656 '_type': 'url',
4657 'url': channel_url,
4658 'id': channel_id,
4659 'ie_key': YoutubeTabIE.ie_key(),
4660 'channel': title,
4661 'uploader': title,
4662 'channel_id': channel_id,
4663 'channel_url': channel_url,
4664 'title': title,
4665 'uploader_id': channel_handle,
4666 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4667 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
4668 # However, in feed/channels this is set correctly to the subscriber count
4669 'channel_follower_count': traverse_obj(
4670 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
4671 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4672 'playlist_count': (
4673 # videoCountText may be the subscriber count
4674 self._get_count(renderer, 'videoCountText')
4675 if self._get_count(renderer, 'subscriberCountText') is not None else None),
4676 'description': self._get_text(renderer, 'descriptionSnippet'),
4677 'channel_is_verified': True if self._has_badge(
4678 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
4679 }
4680
4681 def _grid_entries(self, grid_renderer):
4682 for item in grid_renderer['items']:
4683 if not isinstance(item, dict):
4684 continue
4685 renderer = self._extract_basic_item_renderer(item)
4686 if not isinstance(renderer, dict):
4687 continue
4688 title = self._get_text(renderer, 'title')
4689
4690 # playlist
4691 playlist_id = renderer.get('playlistId')
4692 if playlist_id:
4693 yield self.url_result(
4694 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4695 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4696 video_title=title)
4697 continue
4698 # video
4699 video_id = renderer.get('videoId')
4700 if video_id:
4701 yield self._extract_video(renderer)
4702 continue
4703 # channel
4704 channel_id = renderer.get('channelId')
4705 if channel_id:
4706 yield self._extract_channel_renderer(renderer)
4707 continue
4708 # generic endpoint URL support
4709 ep_url = urljoin('https://www.youtube.com/', try_get(
4710 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4711 str))
4712 if ep_url:
4713 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4714 if ie.suitable(ep_url):
4715 yield self.url_result(
4716 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4717 break
4718
4719 def _music_reponsive_list_entry(self, renderer):
4720 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4721 if video_id:
4722 title = traverse_obj(renderer, (
4723 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4724 'text', 'runs', 0, 'text'))
4725 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4726 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4727 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4728 if playlist_id:
4729 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4730 if video_id:
4731 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4732 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4733 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4734 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4735 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4736 if browse_id:
4737 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4738 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4739
4740 def _shelf_entries_from_content(self, shelf_renderer):
4741 content = shelf_renderer.get('content')
4742 if not isinstance(content, dict):
4743 return
4744 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4745 if renderer:
4746 # TODO: add support for nested playlists so each shelf is processed
4747 # as separate playlist
4748 # TODO: this includes only first N items
4749 yield from self._grid_entries(renderer)
4750 renderer = content.get('horizontalListRenderer')
4751 if renderer:
4752 # TODO
4753 pass
4754
4755 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4756 ep = try_get(
4757 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4758 str)
4759 shelf_url = urljoin('https://www.youtube.com', ep)
4760 if shelf_url:
4761 # Skipping links to another channels, note that checking for
4762 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4763 # will not work
4764 if skip_channels and '/channels?' in shelf_url:
4765 return
4766 title = self._get_text(shelf_renderer, 'title')
4767 yield self.url_result(shelf_url, video_title=title)
4768 # Shelf may not contain shelf URL, fallback to extraction from content
4769 yield from self._shelf_entries_from_content(shelf_renderer)
4770
4771 def _playlist_entries(self, video_list_renderer):
4772 for content in video_list_renderer['contents']:
4773 if not isinstance(content, dict):
4774 continue
4775 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4776 if not isinstance(renderer, dict):
4777 continue
4778 video_id = renderer.get('videoId')
4779 if not video_id:
4780 continue
4781 yield self._extract_video(renderer)
4782
4783 def _rich_entries(self, rich_grid_renderer):
4784 renderer = traverse_obj(
4785 rich_grid_renderer,
4786 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4787 video_id = renderer.get('videoId')
4788 if video_id:
4789 yield self._extract_video(renderer)
4790 return
4791 playlist_id = renderer.get('playlistId')
4792 if playlist_id:
4793 yield self.url_result(
4794 f'https://www.youtube.com/playlist?list={playlist_id}',
4795 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4796 video_title=self._get_text(renderer, 'title'))
4797 return
4798
4799 def _video_entry(self, video_renderer):
4800 video_id = video_renderer.get('videoId')
4801 if video_id:
4802 return self._extract_video(video_renderer)
4803
4804 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4805 url = urljoin('https://youtube.com', traverse_obj(
4806 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4807 if url:
4808 return self.url_result(
4809 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4810
4811 def _post_thread_entries(self, post_thread_renderer):
4812 post_renderer = try_get(
4813 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4814 if not post_renderer:
4815 return
4816 # video attachment
4817 video_renderer = try_get(
4818 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4819 video_id = video_renderer.get('videoId')
4820 if video_id:
4821 entry = self._extract_video(video_renderer)
4822 if entry:
4823 yield entry
4824 # playlist attachment
4825 playlist_id = try_get(
4826 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4827 if playlist_id:
4828 yield self.url_result(
4829 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4830 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4831 # inline video links
4832 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4833 for run in runs:
4834 if not isinstance(run, dict):
4835 continue
4836 ep_url = try_get(
4837 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4838 if not ep_url:
4839 continue
4840 if not YoutubeIE.suitable(ep_url):
4841 continue
4842 ep_video_id = YoutubeIE._match_id(ep_url)
4843 if video_id == ep_video_id:
4844 continue
4845 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4846
4847 def _post_thread_continuation_entries(self, post_thread_continuation):
4848 contents = post_thread_continuation.get('contents')
4849 if not isinstance(contents, list):
4850 return
4851 for content in contents:
4852 renderer = content.get('backstagePostThreadRenderer')
4853 if isinstance(renderer, dict):
4854 yield from self._post_thread_entries(renderer)
4855 continue
4856 renderer = content.get('videoRenderer')
4857 if isinstance(renderer, dict):
4858 yield self._video_entry(renderer)
4859
4860 r''' # unused
4861 def _rich_grid_entries(self, contents):
4862 for content in contents:
4863 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4864 if video_renderer:
4865 entry = self._video_entry(video_renderer)
4866 if entry:
4867 yield entry
4868 '''
4869
4870 def _report_history_entries(self, renderer):
4871 for url in traverse_obj(renderer, (
4872 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4873 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4874 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4875 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4876
4877 def _extract_entries(self, parent_renderer, continuation_list):
4878 # continuation_list is modified in-place with continuation_list = [continuation_token]
4879 continuation_list[:] = [None]
4880 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4881 for content in contents:
4882 if not isinstance(content, dict):
4883 continue
4884 is_renderer = traverse_obj(
4885 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4886 expected_type=dict)
4887 if not is_renderer:
4888 if content.get('richItemRenderer'):
4889 for entry in self._rich_entries(content['richItemRenderer']):
4890 yield entry
4891 continuation_list[0] = self._extract_continuation(parent_renderer)
4892 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4893 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4894 yield from self._report_history_entries(table)
4895 continuation_list[0] = self._extract_continuation(table)
4896 continue
4897
4898 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4899 for isr_content in isr_contents:
4900 if not isinstance(isr_content, dict):
4901 continue
4902
4903 known_renderers = {
4904 'playlistVideoListRenderer': self._playlist_entries,
4905 'gridRenderer': self._grid_entries,
4906 'reelShelfRenderer': self._grid_entries,
4907 'shelfRenderer': self._shelf_entries,
4908 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4909 'backstagePostThreadRenderer': self._post_thread_entries,
4910 'videoRenderer': lambda x: [self._video_entry(x)],
4911 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4912 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4913 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
4914 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
4915 }
4916 for key, renderer in isr_content.items():
4917 if key not in known_renderers:
4918 continue
4919 for entry in known_renderers[key](renderer):
4920 if entry:
4921 yield entry
4922 continuation_list[0] = self._extract_continuation(renderer)
4923 break
4924
4925 if not continuation_list[0]:
4926 continuation_list[0] = self._extract_continuation(is_renderer)
4927
4928 if not continuation_list[0]:
4929 continuation_list[0] = self._extract_continuation(parent_renderer)
4930
4931 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4932 continuation_list = [None]
4933 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4934 tab_content = try_get(tab, lambda x: x['content'], dict)
4935 if not tab_content:
4936 return
4937 parent_renderer = (
4938 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4939 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4940 yield from extract_entries(parent_renderer)
4941 continuation = continuation_list[0]
4942 seen_continuations = set()
4943 for page_num in itertools.count(1):
4944 if not continuation:
4945 break
4946 continuation_token = continuation.get('continuation')
4947 if continuation_token is not None and continuation_token in seen_continuations:
4948 self.write_debug('Detected YouTube feed looping - assuming end of feed.')
4949 break
4950 seen_continuations.add(continuation_token)
4951 headers = self.generate_api_headers(
4952 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4953 response = self._extract_response(
4954 item_id=f'{item_id} page {page_num}',
4955 query=continuation, headers=headers, ytcfg=ytcfg,
4956 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4957
4958 if not response:
4959 break
4960 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4961 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4962 visitor_data = self._extract_visitor_data(response) or visitor_data
4963
4964 known_renderers = {
4965 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4966 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4967 'gridVideoRenderer': (self._grid_entries, 'items'),
4968 'gridChannelRenderer': (self._grid_entries, 'items'),
4969 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4970 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4971 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4972 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4973 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4974 'playlistVideoListContinuation': (self._playlist_entries, None),
4975 'gridContinuation': (self._grid_entries, None),
4976 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4977 'sectionListContinuation': (extract_entries, None), # for feeds
4978 }
4979
4980 continuation_items = traverse_obj(response, (
4981 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4982 'appendContinuationItemsAction', 'continuationItems'
4983 ), 'continuationContents', get_all=False)
4984 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4985
4986 video_items_renderer = None
4987 for key in continuation_item.keys():
4988 if key not in known_renderers:
4989 continue
4990 func, parent_key = known_renderers[key]
4991 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4992 continuation_list = [None]
4993 yield from func(video_items_renderer)
4994 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4995
4996 if not video_items_renderer:
4997 break
4998
4999 @staticmethod
5000 def _extract_selected_tab(tabs, fatal=True):
5001 for tab_renderer in tabs:
5002 if tab_renderer.get('selected'):
5003 return tab_renderer
5004 if fatal:
5005 raise ExtractorError('Unable to find selected tab')
5006
5007 @staticmethod
5008 def _extract_tab_renderers(response):
5009 return traverse_obj(
5010 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
5011
5012 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
5013 metadata = self._extract_metadata_from_tabs(item_id, data)
5014
5015 selected_tab = self._extract_selected_tab(tabs)
5016 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
5017 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
5018
5019 return self.playlist_result(
5020 self._entries(
5021 selected_tab, metadata['id'], ytcfg,
5022 self._extract_account_syncid(ytcfg, data),
5023 self._extract_visitor_data(data, ytcfg)),
5024 **metadata)
5025
5026 def _extract_metadata_from_tabs(self, item_id, data):
5027 info = {'id': item_id}
5028
5029 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
5030 if metadata_renderer:
5031 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
5032 ('channelUrl', {self.ucid_from_url}))
5033 info.update({
5034 'channel': metadata_renderer.get('title'),
5035 'channel_id': channel_id,
5036 })
5037 if info['channel_id']:
5038 info['id'] = info['channel_id']
5039 else:
5040 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5041
5042 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5043 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5044 def _get_uncropped(url):
5045 return url_or_none((url or '').split('=')[0] + '=s0')
5046
5047 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5048 if avatar_thumbnails:
5049 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5050 if uncropped_avatar:
5051 avatar_thumbnails.append({
5052 'url': uncropped_avatar,
5053 'id': 'avatar_uncropped',
5054 'preference': 1
5055 })
5056
5057 channel_banners = self._extract_thumbnails(
5058 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5059 for banner in channel_banners:
5060 banner['preference'] = -10
5061
5062 if channel_banners:
5063 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5064 if uncropped_banner:
5065 channel_banners.append({
5066 'url': uncropped_banner,
5067 'id': 'banner_uncropped',
5068 'preference': -5
5069 })
5070
5071 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5072 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5073 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5074
5075 primary_thumbnails = self._extract_thumbnails(
5076 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5077 playlist_thumbnails = self._extract_thumbnails(
5078 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5079
5080 info.update({
5081 'title': (traverse_obj(metadata_renderer, 'title')
5082 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5083 or info['id']),
5084 'availability': self._extract_availability(data),
5085 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
5086 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5087 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
5088 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5089 })
5090
5091 channel_handle = (
5092 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5093 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5094
5095 if channel_handle:
5096 info.update({
5097 'uploader_id': channel_handle,
5098 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5099 })
5100
5101 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5102 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5103 info['channel_is_verified'] = True
5104 # Playlist stats is a text runs array containing [video count, view count, last updated].
5105 # last updated or (view count and last updated) may be missing.
5106 playlist_stats = get_first(
5107 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5108
5109 last_updated_unix = self._parse_time_text(
5110 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5111 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5112 info['modified_date'] = strftime_or_none(last_updated_unix)
5113
5114 info['view_count'] = self._get_count(playlist_stats, 1)
5115 if info['view_count'] is None: # 0 is allowed
5116 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5117 if info['view_count'] is None:
5118 info['view_count'] = self._get_count(data, (
5119 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5120 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5121
5122 info['playlist_count'] = self._get_count(playlist_stats, 0)
5123 if info['playlist_count'] is None: # 0 is allowed
5124 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5125
5126 if not info.get('channel_id'):
5127 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5128 if not owner: # Deprecated
5129 owner = traverse_obj(
5130 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5131 ('videoOwner', 'videoOwnerRenderer', 'title'))
5132 owner_text = self._get_text(owner)
5133 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5134 info.update({
5135 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5136 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5137 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
5138 })
5139
5140 info.update({
5141 'uploader': info['channel'],
5142 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5143 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5144 })
5145
5146 return info
5147
5148 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5149 first_id = last_id = response = None
5150 for page_num in itertools.count(1):
5151 videos = list(self._playlist_entries(playlist))
5152 if not videos:
5153 return
5154 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5155 if start >= len(videos):
5156 return
5157 yield from videos[start:]
5158 first_id = first_id or videos[0]['id']
5159 last_id = videos[-1]['id']
5160 watch_endpoint = try_get(
5161 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5162 headers = self.generate_api_headers(
5163 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5164 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5165 query = {
5166 'playlistId': playlist_id,
5167 'videoId': watch_endpoint.get('videoId') or last_id,
5168 'index': watch_endpoint.get('index') or len(videos),
5169 'params': watch_endpoint.get('params') or 'OAE%3D'
5170 }
5171 response = self._extract_response(
5172 item_id='%s page %d' % (playlist_id, page_num),
5173 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5174 check_get_keys='contents'
5175 )
5176 playlist = try_get(
5177 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5178
5179 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5180 title = playlist.get('title') or try_get(
5181 data, lambda x: x['titleText']['simpleText'], str)
5182 playlist_id = playlist.get('playlistId') or item_id
5183
5184 # Delegating everything except mix playlists to regular tab-based playlist URL
5185 playlist_url = urljoin(url, try_get(
5186 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5187 str))
5188
5189 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5190 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5191 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5192
5193 if playlist_url and playlist_url != url and not is_known_unviewable:
5194 return self.url_result(
5195 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5196 video_title=title)
5197
5198 return self.playlist_result(
5199 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5200 playlist_id=playlist_id, playlist_title=title)
5201
5202 def _extract_availability(self, data):
5203 """
5204 Gets the availability of a given playlist/tab.
5205 Note: Unless YouTube tells us explicitly, we do not assume it is public
5206 @param data: response
5207 """
5208 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5209 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5210 player_header_privacy = playlist_header_renderer.get('privacy')
5211
5212 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5213
5214 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5215 privacy_setting_icon = get_first(
5216 (playlist_header_renderer, sidebar_renderer),
5217 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5218 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5219 expected_type=str)
5220
5221 microformats_is_unlisted = traverse_obj(
5222 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5223
5224 return (
5225 'public' if (
5226 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5227 or player_header_privacy == 'PUBLIC'
5228 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5229 else self._availability(
5230 is_private=(
5231 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5232 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5233 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5234 is_unlisted=(
5235 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5236 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5237 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5238 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5239 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5240 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5241 needs_auth=False))
5242
5243 @staticmethod
5244 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5245 sidebar_renderer = try_get(
5246 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5247 for item in sidebar_renderer:
5248 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5249 if renderer:
5250 return renderer
5251
5252 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5253 """
5254 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5255 """
5256 is_playlist = bool(traverse_obj(
5257 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5258 if not is_playlist:
5259 return
5260 headers = self.generate_api_headers(
5261 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5262 visitor_data=self._extract_visitor_data(data, ytcfg))
5263 query = {
5264 'params': 'wgYCCAA=',
5265 'browseId': f'VL{item_id}'
5266 }
5267 return self._extract_response(
5268 item_id=item_id, headers=headers, query=query,
5269 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5270 note='Redownloading playlist API JSON with unavailable videos')
5271
5272 @functools.cached_property
5273 def skip_webpage(self):
5274 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5275
5276 def _extract_webpage(self, url, item_id, fatal=True):
5277 webpage, data = None, None
5278 for retry in self.RetryManager(fatal=fatal):
5279 try:
5280 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5281 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5282 except ExtractorError as e:
5283 if isinstance(e.cause, network_exceptions):
5284 if not isinstance(e.cause, HTTPError) or e.cause.status not in (403, 429):
5285 retry.error = e
5286 continue
5287 self._error_or_warning(e, fatal=fatal)
5288 break
5289
5290 try:
5291 self._extract_and_report_alerts(data)
5292 except ExtractorError as e:
5293 self._error_or_warning(e, fatal=fatal)
5294 break
5295
5296 # Sometimes youtube returns a webpage with incomplete ytInitialData
5297 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5298 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5299 retry.error = ExtractorError('Incomplete yt initial data received')
5300 data = None
5301 continue
5302
5303 return webpage, data
5304
5305 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5306 """Use if failed to extract ytcfg (and data) from initial webpage"""
5307 if not ytcfg and self.is_authenticated:
5308 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5309 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5310 raise ExtractorError(
5311 f'{msg}. If you are not downloading private content, or '
5312 'your cookies are only for the first account and channel,'
5313 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5314 expected=True)
5315 self.report_warning(msg, only_once=True)
5316
5317 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5318 data = None
5319 if not self.skip_webpage:
5320 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5321 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5322 # Reject webpage data if redirected to home page without explicitly requesting
5323 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5324 if (url != 'https://www.youtube.com/feed/recommended'
5325 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5326 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5327 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5328 if fatal:
5329 raise ExtractorError(msg, expected=True)
5330 self.report_warning(msg, only_once=True)
5331 if not data:
5332 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5333 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5334 return data, ytcfg
5335
5336 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5337 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5338 resolve_response = self._extract_response(
5339 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5340 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5341 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5342 for ep_key, ep in endpoints.items():
5343 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5344 if params:
5345 return self._extract_response(
5346 item_id=item_id, query=params, ep=ep, headers=headers,
5347 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5348 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5349 err_note = 'Failed to resolve url (does the playlist exist?)'
5350 if fatal:
5351 raise ExtractorError(err_note, expected=True)
5352 self.report_warning(err_note, item_id)
5353
5354 _SEARCH_PARAMS = None
5355
5356 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5357 data = {'query': query}
5358 if params is NO_DEFAULT:
5359 params = self._SEARCH_PARAMS
5360 if params:
5361 data['params'] = params
5362
5363 content_keys = (
5364 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5365 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5366 # ytmusic search
5367 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5368 ('continuationContents', ),
5369 )
5370 display_id = f'query "{query}"'
5371 check_get_keys = tuple({keys[0] for keys in content_keys})
5372 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5373 self._report_playlist_authcheck(ytcfg, fatal=False)
5374
5375 continuation_list = [None]
5376 search = None
5377 for page_num in itertools.count(1):
5378 data.update(continuation_list[0] or {})
5379 headers = self.generate_api_headers(
5380 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5381 search = self._extract_response(
5382 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5383 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5384 slr_contents = traverse_obj(search, *content_keys)
5385 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5386 if not continuation_list[0]:
5387 break
5388
5389
5390 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5391 IE_DESC = 'YouTube Tabs'
5392 _VALID_URL = r'''(?x:
5393 https?://
5394 (?!consent\.)(?:\w+\.)?
5395 (?:
5396 youtube(?:kids)?\.com|
5397 %(invidious)s
5398 )/
5399 (?:
5400 (?P<channel_type>channel|c|user|browse)/|
5401 (?P<not_channel>
5402 feed/|hashtag/|
5403 (?:playlist|watch)\?.*?\blist=
5404 )|
5405 (?!(?:%(reserved_names)s)\b) # Direct URLs
5406 )
5407 (?P<id>[^/?\#&]+)
5408 )''' % {
5409 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5410 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5411 }
5412 IE_NAME = 'youtube:tab'
5413
5414 _TESTS = [{
5415 'note': 'playlists, multipage',
5416 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5417 'playlist_mincount': 94,
5418 'info_dict': {
5419 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5420 'title': 'Igor Kleiner - Playlists',
5421 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5422 'uploader': 'Igor Kleiner',
5423 'uploader_id': '@IgorDataScience',
5424 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5425 'channel': 'Igor Kleiner',
5426 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5427 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5428 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5429 'channel_follower_count': int
5430 },
5431 }, {
5432 'note': 'playlists, multipage, different order',
5433 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5434 'playlist_mincount': 94,
5435 'info_dict': {
5436 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5437 'title': 'Igor Kleiner - Playlists',
5438 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5439 'uploader': 'Igor Kleiner',
5440 'uploader_id': '@IgorDataScience',
5441 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5442 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5443 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5444 'channel': 'Igor Kleiner',
5445 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5446 'channel_follower_count': int
5447 },
5448 }, {
5449 'note': 'playlists, series',
5450 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5451 'playlist_mincount': 5,
5452 'info_dict': {
5453 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5454 'title': '3Blue1Brown - Playlists',
5455 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5456 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5457 'channel': '3Blue1Brown',
5458 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5459 'uploader_id': '@3blue1brown',
5460 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5461 'uploader': '3Blue1Brown',
5462 'tags': ['Mathematics'],
5463 'channel_follower_count': int,
5464 'channel_is_verified': True,
5465 },
5466 }, {
5467 'note': 'playlists, singlepage',
5468 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5469 'playlist_mincount': 4,
5470 'info_dict': {
5471 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5472 'title': 'ThirstForScience - Playlists',
5473 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5474 'uploader': 'ThirstForScience',
5475 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5476 'uploader_id': '@ThirstForScience',
5477 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5478 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5479 'tags': 'count:13',
5480 'channel': 'ThirstForScience',
5481 'channel_follower_count': int
5482 }
5483 }, {
5484 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5485 'only_matching': True,
5486 }, {
5487 'note': 'basic, single video playlist',
5488 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5489 'info_dict': {
5490 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5491 'title': 'youtube-dl public playlist',
5492 'description': '',
5493 'tags': [],
5494 'view_count': int,
5495 'modified_date': '20201130',
5496 'channel': 'Sergey M.',
5497 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5498 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5499 'availability': 'public',
5500 'uploader': 'Sergey M.',
5501 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5502 'uploader_id': '@sergeym.6173',
5503 },
5504 'playlist_count': 1,
5505 }, {
5506 'note': 'empty playlist',
5507 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5508 'info_dict': {
5509 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5510 'title': 'youtube-dl empty playlist',
5511 'tags': [],
5512 'channel': 'Sergey M.',
5513 'description': '',
5514 'modified_date': '20160902',
5515 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5516 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5517 'availability': 'public',
5518 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5519 'uploader_id': '@sergeym.6173',
5520 'uploader': 'Sergey M.',
5521 },
5522 'playlist_count': 0,
5523 }, {
5524 'note': 'Home tab',
5525 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5526 'info_dict': {
5527 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5528 'title': 'lex will - Home',
5529 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5530 'uploader': 'lex will',
5531 'uploader_id': '@lexwill718',
5532 'channel': 'lex will',
5533 'tags': ['bible', 'history', 'prophesy'],
5534 'uploader_url': 'https://www.youtube.com/@lexwill718',
5535 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5536 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5537 'channel_follower_count': int
5538 },
5539 'playlist_mincount': 2,
5540 }, {
5541 'note': 'Videos tab',
5542 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5543 'info_dict': {
5544 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5545 'title': 'lex will - Videos',
5546 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5547 'uploader': 'lex will',
5548 'uploader_id': '@lexwill718',
5549 'tags': ['bible', 'history', 'prophesy'],
5550 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5551 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5552 'uploader_url': 'https://www.youtube.com/@lexwill718',
5553 'channel': 'lex will',
5554 'channel_follower_count': int
5555 },
5556 'playlist_mincount': 975,
5557 }, {
5558 'note': 'Videos tab, sorted by popular',
5559 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5560 'info_dict': {
5561 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5562 'title': 'lex will - Videos',
5563 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5564 'uploader': 'lex will',
5565 'uploader_id': '@lexwill718',
5566 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5567 'uploader_url': 'https://www.youtube.com/@lexwill718',
5568 'channel': 'lex will',
5569 'tags': ['bible', 'history', 'prophesy'],
5570 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5571 'channel_follower_count': int
5572 },
5573 'playlist_mincount': 199,
5574 }, {
5575 'note': 'Playlists tab',
5576 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5577 'info_dict': {
5578 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5579 'title': 'lex will - Playlists',
5580 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5581 'uploader': 'lex will',
5582 'uploader_id': '@lexwill718',
5583 'uploader_url': 'https://www.youtube.com/@lexwill718',
5584 'channel': 'lex will',
5585 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5586 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5587 'tags': ['bible', 'history', 'prophesy'],
5588 'channel_follower_count': int
5589 },
5590 'playlist_mincount': 17,
5591 }, {
5592 'note': 'Community tab',
5593 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5594 'info_dict': {
5595 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5596 'title': 'lex will - Community',
5597 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5598 'channel': 'lex will',
5599 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5600 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5601 'tags': ['bible', 'history', 'prophesy'],
5602 'channel_follower_count': int,
5603 'uploader_url': 'https://www.youtube.com/@lexwill718',
5604 'uploader_id': '@lexwill718',
5605 'uploader': 'lex will',
5606 },
5607 'playlist_mincount': 18,
5608 }, {
5609 'note': 'Channels tab',
5610 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5611 'info_dict': {
5612 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5613 'title': 'lex will - Channels',
5614 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5615 'channel': 'lex will',
5616 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5617 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5618 'tags': ['bible', 'history', 'prophesy'],
5619 'channel_follower_count': int,
5620 'uploader_url': 'https://www.youtube.com/@lexwill718',
5621 'uploader_id': '@lexwill718',
5622 'uploader': 'lex will',
5623 },
5624 'playlist_mincount': 12,
5625 }, {
5626 'note': 'Search tab',
5627 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5628 'playlist_mincount': 40,
5629 'info_dict': {
5630 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5631 'title': '3Blue1Brown - Search - linear algebra',
5632 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5633 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5634 'tags': ['Mathematics'],
5635 'channel': '3Blue1Brown',
5636 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5637 'channel_follower_count': int,
5638 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5639 'uploader_id': '@3blue1brown',
5640 'uploader': '3Blue1Brown',
5641 'channel_is_verified': True,
5642 },
5643 }, {
5644 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5645 'only_matching': True,
5646 }, {
5647 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5648 'only_matching': True,
5649 }, {
5650 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5651 'only_matching': True,
5652 }, {
5653 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5654 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5655 'info_dict': {
5656 'title': '29C3: Not my department',
5657 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5658 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5659 'tags': [],
5660 'view_count': int,
5661 'modified_date': '20150605',
5662 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5663 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5664 'channel': 'Christiaan008',
5665 'availability': 'public',
5666 'uploader_id': '@ChRiStIaAn008',
5667 'uploader': 'Christiaan008',
5668 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5669 },
5670 'playlist_count': 96,
5671 }, {
5672 'note': 'Large playlist',
5673 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5674 'info_dict': {
5675 'title': 'Uploads from Cauchemar',
5676 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5677 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5678 'tags': [],
5679 'modified_date': r're:\d{8}',
5680 'channel': 'Cauchemar',
5681 'view_count': int,
5682 'description': '',
5683 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5684 'availability': 'public',
5685 'uploader_id': '@Cauchemar89',
5686 'uploader': 'Cauchemar',
5687 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5688 },
5689 'playlist_mincount': 1123,
5690 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5691 }, {
5692 'note': 'even larger playlist, 8832 videos',
5693 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5694 'only_matching': True,
5695 }, {
5696 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5697 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5698 'info_dict': {
5699 'title': 'Uploads from Interstellar Movie',
5700 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5701 'tags': [],
5702 'view_count': int,
5703 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5704 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5705 'channel': 'Interstellar Movie',
5706 'description': '',
5707 'modified_date': r're:\d{8}',
5708 'availability': 'public',
5709 'uploader_id': '@InterstellarMovie',
5710 'uploader': 'Interstellar Movie',
5711 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5712 },
5713 'playlist_mincount': 21,
5714 }, {
5715 'note': 'Playlist with "show unavailable videos" button',
5716 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5717 'info_dict': {
5718 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5719 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5720 'view_count': int,
5721 'channel': 'Phim Siêu Nhân Nhật Bản',
5722 'tags': [],
5723 'description': '',
5724 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5725 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5726 'modified_date': r're:\d{8}',
5727 'availability': 'public',
5728 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5729 'uploader_id': '@phimsieunhannhatban',
5730 'uploader': 'Phim Siêu Nhân Nhật Bản',
5731 },
5732 'playlist_mincount': 200,
5733 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5734 }, {
5735 'note': 'Playlist with unavailable videos in page 7',
5736 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5737 'info_dict': {
5738 'title': 'Uploads from BlankTV',
5739 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5740 'channel': 'BlankTV',
5741 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5742 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5743 'view_count': int,
5744 'tags': [],
5745 'modified_date': r're:\d{8}',
5746 'description': '',
5747 'availability': 'public',
5748 'uploader_id': '@blanktv',
5749 'uploader': 'BlankTV',
5750 'uploader_url': 'https://www.youtube.com/@blanktv',
5751 },
5752 'playlist_mincount': 1000,
5753 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5754 }, {
5755 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5756 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5757 'info_dict': {
5758 'title': 'Data Analysis with Dr Mike Pound',
5759 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5760 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5761 'tags': [],
5762 'view_count': int,
5763 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5764 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5765 'channel': 'Computerphile',
5766 'availability': 'public',
5767 'modified_date': '20190712',
5768 'uploader_id': '@Computerphile',
5769 'uploader': 'Computerphile',
5770 'uploader_url': 'https://www.youtube.com/@Computerphile',
5771 },
5772 'playlist_mincount': 11,
5773 }, {
5774 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5775 'only_matching': True,
5776 }, {
5777 'note': 'Playlist URL that does not actually serve a playlist',
5778 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5779 'info_dict': {
5780 'id': 'FqZTN594JQw',
5781 'ext': 'webm',
5782 'title': "Smiley's People 01 detective, Adventure Series, Action",
5783 'upload_date': '20150526',
5784 'license': 'Standard YouTube License',
5785 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5786 'categories': ['People & Blogs'],
5787 'tags': list,
5788 'view_count': int,
5789 'like_count': int,
5790 },
5791 'params': {
5792 'skip_download': True,
5793 },
5794 'skip': 'This video is not available.',
5795 'add_ie': [YoutubeIE.ie_key()],
5796 }, {
5797 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5798 'only_matching': True,
5799 }, {
5800 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5801 'only_matching': True,
5802 }, {
5803 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5804 'info_dict': {
5805 'id': 'hGkQjiJLjWQ', # This will keep changing
5806 'ext': 'mp4',
5807 'title': str,
5808 'upload_date': r're:\d{8}',
5809 'description': str,
5810 'categories': ['News & Politics'],
5811 'tags': list,
5812 'like_count': int,
5813 'release_timestamp': int,
5814 'channel': 'Sky News',
5815 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5816 'age_limit': 0,
5817 'view_count': int,
5818 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5819 'playable_in_embed': True,
5820 'release_date': r're:\d+',
5821 'availability': 'public',
5822 'live_status': 'is_live',
5823 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5824 'channel_follower_count': int,
5825 'concurrent_view_count': int,
5826 'uploader_url': 'https://www.youtube.com/@SkyNews',
5827 'uploader_id': '@SkyNews',
5828 'uploader': 'Sky News',
5829 'channel_is_verified': True,
5830 },
5831 'params': {
5832 'skip_download': True,
5833 },
5834 'expected_warnings': ['Ignoring subtitle tracks found in '],
5835 }, {
5836 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5837 'info_dict': {
5838 'id': 'a48o2S1cPoo',
5839 'ext': 'mp4',
5840 'title': 'The Young Turks - Live Main Show',
5841 'upload_date': '20150715',
5842 'license': 'Standard YouTube License',
5843 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5844 'categories': ['News & Politics'],
5845 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5846 'like_count': int,
5847 },
5848 'params': {
5849 'skip_download': True,
5850 },
5851 'only_matching': True,
5852 }, {
5853 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5854 'only_matching': True,
5855 }, {
5856 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5857 'only_matching': True,
5858 }, {
5859 'note': 'A channel that is not live. Should raise error',
5860 'url': 'https://www.youtube.com/user/numberphile/live',
5861 'only_matching': True,
5862 }, {
5863 'url': 'https://www.youtube.com/feed/trending',
5864 'only_matching': True,
5865 }, {
5866 'url': 'https://www.youtube.com/feed/library',
5867 'only_matching': True,
5868 }, {
5869 'url': 'https://www.youtube.com/feed/history',
5870 'only_matching': True,
5871 }, {
5872 'url': 'https://www.youtube.com/feed/subscriptions',
5873 'only_matching': True,
5874 }, {
5875 'url': 'https://www.youtube.com/feed/watch_later',
5876 'only_matching': True,
5877 }, {
5878 'note': 'Recommended - redirects to home page.',
5879 'url': 'https://www.youtube.com/feed/recommended',
5880 'only_matching': True,
5881 }, {
5882 'note': 'inline playlist with not always working continuations',
5883 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5884 'only_matching': True,
5885 }, {
5886 'url': 'https://www.youtube.com/course',
5887 'only_matching': True,
5888 }, {
5889 'url': 'https://www.youtube.com/zsecurity',
5890 'only_matching': True,
5891 }, {
5892 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5893 'only_matching': True,
5894 }, {
5895 'url': 'https://www.youtube.com/TheYoungTurks/live',
5896 'only_matching': True,
5897 }, {
5898 'url': 'https://www.youtube.com/hashtag/cctv9',
5899 'info_dict': {
5900 'id': 'cctv9',
5901 'title': '#cctv9',
5902 'tags': [],
5903 },
5904 'playlist_mincount': 300, # not consistent but should be over 300
5905 }, {
5906 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5907 'only_matching': True,
5908 }, {
5909 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5910 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5911 'only_matching': True
5912 }, {
5913 'note': '/browse/ should redirect to /channel/',
5914 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5915 'only_matching': True
5916 }, {
5917 'note': 'VLPL, should redirect to playlist?list=PL...',
5918 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5919 'info_dict': {
5920 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5921 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5922 'title': 'NCS : All Releases 💿',
5923 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5924 'modified_date': r're:\d{8}',
5925 'view_count': int,
5926 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5927 'tags': [],
5928 'channel': 'NoCopyrightSounds',
5929 'availability': 'public',
5930 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5931 'uploader': 'NoCopyrightSounds',
5932 'uploader_id': '@NoCopyrightSounds',
5933 },
5934 'playlist_mincount': 166,
5935 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5936 }, {
5937 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5938 'note': 'Topic, should redirect to playlist?list=UU...',
5939 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5940 'info_dict': {
5941 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5942 'title': 'Uploads from Royalty Free Music - Topic',
5943 'tags': [],
5944 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5945 'channel': 'Royalty Free Music - Topic',
5946 'view_count': int,
5947 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5948 'modified_date': r're:\d{8}',
5949 'description': '',
5950 'availability': 'public',
5951 'uploader': 'Royalty Free Music - Topic',
5952 },
5953 'playlist_mincount': 101,
5954 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5955 }, {
5956 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5957 # Treat as a general feed
5958 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5959 'info_dict': {
5960 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5961 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5962 'tags': [],
5963 },
5964 'playlist_mincount': 9,
5965 }, {
5966 'note': 'Youtube music Album',
5967 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5968 'info_dict': {
5969 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5970 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5971 'tags': [],
5972 'view_count': int,
5973 'description': '',
5974 'availability': 'unlisted',
5975 'modified_date': r're:\d{8}',
5976 },
5977 'playlist_count': 50,
5978 'expected_warnings': ['YouTube Music is not directly supported'],
5979 }, {
5980 'note': 'unlisted single video playlist',
5981 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5982 'info_dict': {
5983 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5984 'title': 'yt-dlp unlisted playlist test',
5985 'availability': 'unlisted',
5986 'tags': [],
5987 'modified_date': '20220418',
5988 'channel': 'colethedj',
5989 'view_count': int,
5990 'description': '',
5991 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5992 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5993 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5994 'uploader_id': '@colethedj1894',
5995 'uploader': 'colethedj',
5996 },
5997 'playlist': [{
5998 'info_dict': {
5999 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
6000 'id': 'BaW_jenozKc',
6001 '_type': 'url',
6002 'ie_key': 'Youtube',
6003 'duration': 10,
6004 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
6005 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
6006 'view_count': int,
6007 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
6008 'channel': 'Philipp Hagemeister',
6009 'uploader_id': '@PhilippHagemeister',
6010 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
6011 'uploader': 'Philipp Hagemeister',
6012 }
6013 }],
6014 'playlist_count': 1,
6015 'params': {'extract_flat': True},
6016 }, {
6017 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
6018 'url': 'https://www.youtube.com/feed/recommended',
6019 'info_dict': {
6020 'id': 'recommended',
6021 'title': 'recommended',
6022 'tags': [],
6023 },
6024 'playlist_mincount': 50,
6025 'params': {
6026 'skip_download': True,
6027 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6028 },
6029 }, {
6030 'note': 'API Fallback: /videos tab, sorted by oldest first',
6031 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
6032 'info_dict': {
6033 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6034 'title': 'Cody\'sLab - Videos',
6035 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6036 'channel': 'Cody\'sLab',
6037 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6038 'tags': [],
6039 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6040 'channel_follower_count': int
6041 },
6042 'playlist_mincount': 650,
6043 'params': {
6044 'skip_download': True,
6045 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6046 },
6047 'skip': 'Query for sorting no longer works',
6048 }, {
6049 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6050 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6051 'info_dict': {
6052 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6053 'title': 'Uploads from Royalty Free Music - Topic',
6054 'modified_date': r're:\d{8}',
6055 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6056 'description': '',
6057 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6058 'tags': [],
6059 'channel': 'Royalty Free Music - Topic',
6060 'view_count': int,
6061 'availability': 'public',
6062 'uploader': 'Royalty Free Music - Topic',
6063 },
6064 'playlist_mincount': 101,
6065 'params': {
6066 'skip_download': True,
6067 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6068 },
6069 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6070 }, {
6071 'note': 'non-standard redirect to regional channel',
6072 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6073 'only_matching': True
6074 }, {
6075 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6076 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6077 'info_dict': {
6078 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6079 'modified_date': '20220407',
6080 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6081 'tags': [],
6082 'availability': 'unlisted',
6083 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6084 'channel': 'pukkandan',
6085 'description': 'Test for collaborative playlist',
6086 'title': 'yt-dlp test - collaborative playlist',
6087 'view_count': int,
6088 'uploader_url': 'https://www.youtube.com/@pukkandan',
6089 'uploader_id': '@pukkandan',
6090 'uploader': 'pukkandan',
6091 },
6092 'playlist_mincount': 2
6093 }, {
6094 'note': 'translated tab name',
6095 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6096 'info_dict': {
6097 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6098 'tags': [],
6099 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6100 'description': 'test description',
6101 'title': 'cole-dlp-test-acc - 再生リスト',
6102 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6103 'channel': 'cole-dlp-test-acc',
6104 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6105 'uploader_id': '@coletdjnz',
6106 'uploader': 'cole-dlp-test-acc',
6107 },
6108 'playlist_mincount': 1,
6109 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6110 'expected_warnings': ['Preferring "ja"'],
6111 }, {
6112 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6113 'note': 'preferred lang set with playlist with translated video titles',
6114 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6115 'info_dict': {
6116 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6117 'tags': [],
6118 'view_count': int,
6119 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6120 'channel': 'cole-dlp-test-acc',
6121 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6122 'description': 'test',
6123 'title': 'dlp test playlist',
6124 'availability': 'public',
6125 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6126 'uploader_id': '@coletdjnz',
6127 'uploader': 'cole-dlp-test-acc',
6128 },
6129 'playlist_mincount': 1,
6130 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6131 'expected_warnings': ['Preferring "ja"'],
6132 }, {
6133 # shorts audio pivot for 2GtVksBMYFM.
6134 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6135 'info_dict': {
6136 'id': 'sfv_audio_pivot',
6137 'title': 'sfv_audio_pivot',
6138 'tags': [],
6139 },
6140 'playlist_mincount': 50,
6141
6142 }, {
6143 # Channel with a real live tab (not to be mistaken with streams tab)
6144 # Do not treat like it should redirect to live stream
6145 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6146 'info_dict': {
6147 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6148 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6149 'tags': [],
6150 },
6151 'playlist_mincount': 20,
6152 }, {
6153 # Tab name is not the same as tab id
6154 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6155 'info_dict': {
6156 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6157 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6158 'tags': [],
6159 },
6160 'playlist_mincount': 8,
6161 }, {
6162 # Home tab id is literally home. Not to get mistaken with featured
6163 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6164 'info_dict': {
6165 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6166 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6167 'tags': [],
6168 },
6169 'playlist_mincount': 8,
6170 }, {
6171 # Should get three playlists for videos, shorts and streams tabs
6172 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6173 'info_dict': {
6174 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6175 'title': 'Polka Ch. 尾丸ポルカ',
6176 'channel_follower_count': int,
6177 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6178 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6179 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6180 'channel': 'Polka Ch. 尾丸ポルカ',
6181 'tags': 'count:35',
6182 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6183 'uploader': 'Polka Ch. 尾丸ポルカ',
6184 'uploader_id': '@OmaruPolka',
6185 },
6186 'playlist_count': 3,
6187 }, {
6188 # Shorts tab with channel with handle
6189 # TODO: fix channel description
6190 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6191 'info_dict': {
6192 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6193 'title': 'Not Just Bikes - Shorts',
6194 'tags': 'count:12',
6195 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6196 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6197 'channel_follower_count': int,
6198 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6199 'channel': 'Not Just Bikes',
6200 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6201 'uploader': 'Not Just Bikes',
6202 'uploader_id': '@NotJustBikes',
6203 },
6204 'playlist_mincount': 10,
6205 }, {
6206 # Streams tab
6207 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6208 'info_dict': {
6209 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6210 'title': '中村悠一 - Live',
6211 'tags': 'count:7',
6212 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6213 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6214 'channel': '中村悠一',
6215 'channel_follower_count': int,
6216 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6217 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6218 'uploader_id': '@Yuichi-Nakamura',
6219 'uploader': '中村悠一',
6220 },
6221 'playlist_mincount': 60,
6222 }, {
6223 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6224 # See test_youtube_lists
6225 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6226 'only_matching': True,
6227 }, {
6228 # No uploads and no UCID given. Should fail with no uploads error
6229 # See test_youtube_lists
6230 'url': 'https://www.youtube.com/news',
6231 'only_matching': True
6232 }, {
6233 # No videos tab but has a shorts tab
6234 'url': 'https://www.youtube.com/c/TKFShorts',
6235 'info_dict': {
6236 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6237 'title': 'Shorts Break - Shorts',
6238 'tags': 'count:48',
6239 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6240 'channel': 'Shorts Break',
6241 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6242 'channel_follower_count': int,
6243 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6244 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6245 'uploader': 'Shorts Break',
6246 'uploader_id': '@ShortsBreak_Official',
6247 },
6248 'playlist_mincount': 30,
6249 }, {
6250 # Trending Now Tab. tab id is empty
6251 'url': 'https://www.youtube.com/feed/trending',
6252 'info_dict': {
6253 'id': 'trending',
6254 'title': 'trending - Now',
6255 'tags': [],
6256 },
6257 'playlist_mincount': 30,
6258 }, {
6259 # Trending Gaming Tab. tab id is empty
6260 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6261 'info_dict': {
6262 'id': 'trending',
6263 'title': 'trending - Gaming',
6264 'tags': [],
6265 },
6266 'playlist_mincount': 30,
6267 }, {
6268 # Shorts url result in shorts tab
6269 # TODO: Fix channel id extraction
6270 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6271 'info_dict': {
6272 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6273 'title': 'cole-dlp-test-acc - Shorts',
6274 'channel': 'cole-dlp-test-acc',
6275 'description': 'test description',
6276 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6277 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6278 'tags': [],
6279 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6280 'uploader_id': '@coletdjnz',
6281 'uploader': 'cole-dlp-test-acc',
6282 },
6283 'playlist': [{
6284 'info_dict': {
6285 # Channel data is not currently available for short renderers (as of 2023-03-01)
6286 '_type': 'url',
6287 'ie_key': 'Youtube',
6288 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6289 'id': 'sSM9J5YH_60',
6290 'title': 'SHORT short',
6291 'view_count': int,
6292 'thumbnails': list,
6293 }
6294 }],
6295 'params': {'extract_flat': True},
6296 }, {
6297 # Live video status should be extracted
6298 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6299 'info_dict': {
6300 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6301 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6302 'tags': []
6303 },
6304 'playlist': [{
6305 'info_dict': {
6306 '_type': 'url',
6307 'ie_key': 'Youtube',
6308 'url': 'startswith:https://www.youtube.com/watch?v=',
6309 'id': str,
6310 'title': str,
6311 'live_status': 'is_live',
6312 'channel_id': str,
6313 'channel_url': str,
6314 'concurrent_view_count': int,
6315 'channel': str,
6316 'uploader': str,
6317 'uploader_url': str,
6318 'uploader_id': str,
6319 'channel_is_verified': bool, # this will keep changing
6320 }
6321 }],
6322 'params': {'extract_flat': True, 'playlist_items': '1'},
6323 'playlist_mincount': 1
6324 }, {
6325 # Channel renderer metadata. Contains number of videos on the channel
6326 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6327 'info_dict': {
6328 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6329 'title': 'cole-dlp-test-acc - Channels',
6330 'channel': 'cole-dlp-test-acc',
6331 'description': 'test description',
6332 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6333 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6334 'tags': [],
6335 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6336 'uploader_id': '@coletdjnz',
6337 'uploader': 'cole-dlp-test-acc',
6338 },
6339 'playlist': [{
6340 'info_dict': {
6341 '_type': 'url',
6342 'ie_key': 'YoutubeTab',
6343 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6344 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6345 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6346 'title': 'PewDiePie',
6347 'channel': 'PewDiePie',
6348 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6349 'thumbnails': list,
6350 'channel_follower_count': int,
6351 'playlist_count': int,
6352 'uploader': 'PewDiePie',
6353 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6354 'uploader_id': '@PewDiePie',
6355 'channel_is_verified': True,
6356 }
6357 }],
6358 'params': {'extract_flat': True},
6359 }, {
6360 'url': 'https://www.youtube.com/@3blue1brown/about',
6361 'info_dict': {
6362 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6363 'tags': ['Mathematics'],
6364 'title': '3Blue1Brown - About',
6365 'channel_follower_count': int,
6366 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6367 'channel': '3Blue1Brown',
6368 'view_count': int,
6369 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6370 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6371 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6372 'uploader_id': '@3blue1brown',
6373 'uploader': '3Blue1Brown',
6374 'channel_is_verified': True,
6375 },
6376 'playlist_count': 0,
6377 }, {
6378 # Podcasts tab, with rich entry playlistRenderers
6379 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6380 'info_dict': {
6381 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6382 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6383 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6384 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6385 'title': '99 Percent Invisible - Podcasts',
6386 'uploader': '99 Percent Invisible',
6387 'channel_follower_count': int,
6388 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6389 'tags': [],
6390 'channel': '99 Percent Invisible',
6391 'uploader_id': '@99percentinvisiblepodcast',
6392 },
6393 'playlist_count': 1,
6394 }, {
6395 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6396 'url': 'https://www.youtube.com/@AHimitsu/releases',
6397 'info_dict': {
6398 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6399 'channel': 'A Himitsu',
6400 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6401 'title': 'A Himitsu - Releases',
6402 'uploader_id': '@AHimitsu',
6403 'uploader': 'A Himitsu',
6404 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6405 'tags': 'count:16',
6406 'description': 'I make music',
6407 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6408 'channel_follower_count': int,
6409 'channel_is_verified': True,
6410 },
6411 'playlist_mincount': 10,
6412 }, {
6413 # Playlist with only shorts, shown as reel renderers
6414 # FIXME: future: YouTube currently doesn't give continuation for this,
6415 # may do in future.
6416 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
6417 'info_dict': {
6418 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
6419 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
6420 'view_count': int,
6421 'uploader_id': '@BangyShorts',
6422 'description': '',
6423 'uploader_url': 'https://www.youtube.com/@BangyShorts',
6424 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
6425 'channel': 'Bangy Shorts',
6426 'uploader': 'Bangy Shorts',
6427 'tags': [],
6428 'availability': 'public',
6429 'modified_date': '20230626',
6430 'title': 'Uploads from Bangy Shorts',
6431 },
6432 'playlist_mincount': 100,
6433 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6434 }]
6435
6436 @classmethod
6437 def suitable(cls, url):
6438 return False if YoutubeIE.suitable(url) else super().suitable(url)
6439
6440 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6441
6442 def _get_url_mobj(self, url):
6443 mobj = self._URL_RE.match(url).groupdict()
6444 mobj.update((k, '') for k, v in mobj.items() if v is None)
6445 return mobj
6446
6447 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6448 tab_name = (tab.get('title') or '').lower()
6449 tab_url = urljoin(base_url, traverse_obj(
6450 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6451
6452 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6453 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6454 if tab_id:
6455 return {
6456 'TAB_ID_SPONSORSHIPS': 'membership',
6457 }.get(tab_id, tab_id), tab_name
6458
6459 # Fallback to tab name if we cannot get the tab id.
6460 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6461 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6462 if tab_name:
6463 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6464 return {
6465 'home': 'featured',
6466 'live': 'streams',
6467 }.get(tab_name, tab_name), tab_name
6468
6469 def _has_tab(self, tabs, tab_id):
6470 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6471
6472 def _empty_playlist(self, item_id, data):
6473 return self.playlist_result([], item_id, **self._extract_metadata_from_tabs(item_id, data))
6474
6475 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6476 def _real_extract(self, url, smuggled_data):
6477 item_id = self._match_id(url)
6478 url = urllib.parse.urlunparse(
6479 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6480 compat_opts = self.get_param('compat_opts', [])
6481
6482 mobj = self._get_url_mobj(url)
6483 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6484 if is_channel and smuggled_data.get('is_music_url'):
6485 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6486 return self.url_result(
6487 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6488 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6489 mdata = self._extract_tab_endpoint(
6490 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6491 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6492 get_all=False, expected_type=str)
6493 if not murl:
6494 raise ExtractorError('Failed to resolve album to playlist')
6495 return self.url_result(murl, YoutubeTabIE)
6496 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6497 return self.url_result(
6498 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6499
6500 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6501 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6502 url = f'{pre}/videos{post}'
6503 if smuggled_data.get('is_music_url'):
6504 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6505
6506 # Handle both video/playlist URLs
6507 qs = parse_qs(url)
6508 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6509 if not video_id and mobj['not_channel'].startswith('watch'):
6510 if not playlist_id:
6511 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6512 raise ExtractorError('A video URL was given without video ID', expected=True)
6513 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6514 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6515 return self.url_result(
6516 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6517
6518 if not self._yes_playlist(playlist_id, video_id):
6519 return self.url_result(
6520 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6521
6522 data, ytcfg = self._extract_data(url, display_id)
6523
6524 # YouTube may provide a non-standard redirect to the regional channel
6525 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6526 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6527 redirect_url = traverse_obj(
6528 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6529 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6530 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6531 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6532 return self.url_result(redirect_url, YoutubeTabIE)
6533
6534 tabs, extra_tabs = self._extract_tab_renderers(data), []
6535 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6536 selected_tab = self._extract_selected_tab(tabs)
6537 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6538 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6539
6540 # /about is no longer a tab
6541 if original_tab_id == 'about':
6542 return self._empty_playlist(item_id, data)
6543
6544 if not original_tab_id and selected_tab_name:
6545 self.to_screen('Downloading all uploads of the channel. '
6546 'To download only the videos in a specific tab, pass the tab\'s URL')
6547 if self._has_tab(tabs, 'streams'):
6548 extra_tabs.append(''.join((pre, '/streams', post)))
6549 if self._has_tab(tabs, 'shorts'):
6550 extra_tabs.append(''.join((pre, '/shorts', post)))
6551 # XXX: Members-only tab should also be extracted
6552
6553 if not extra_tabs and selected_tab_id != 'videos':
6554 # Channel does not have streams, shorts or videos tabs
6555 if item_id[:2] != 'UC':
6556 return self._empty_playlist(item_id, data)
6557
6558 # Topic channels don't have /videos. Use the equivalent playlist instead
6559 pl_id = f'UU{item_id[2:]}'
6560 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6561 try:
6562 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6563 except ExtractorError:
6564 return self._empty_playlist(item_id, data)
6565 else:
6566 item_id, url = pl_id, pl_url
6567 self.to_screen(
6568 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6569
6570 elif extra_tabs and selected_tab_id != 'videos':
6571 # When there are shorts/live tabs but not videos tab
6572 url, data = f'{pre}{post}', None
6573
6574 elif (original_tab_id or 'videos') != selected_tab_id:
6575 if original_tab_id == 'live':
6576 # Live tab should have redirected to the video
6577 # Except in the case the channel has an actual live tab
6578 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6579 raise UserNotLive(video_id=item_id)
6580 elif selected_tab_name:
6581 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6582
6583 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6584 url = f'{pre}{post}'
6585
6586 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6587 if 'no-youtube-unavailable-videos' not in compat_opts:
6588 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6589 self._extract_and_report_alerts(data, only_once=True)
6590
6591 tabs, entries = self._extract_tab_renderers(data), []
6592 if tabs:
6593 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6594 entries[0].update({
6595 'extractor_key': YoutubeTabIE.ie_key(),
6596 'extractor': YoutubeTabIE.IE_NAME,
6597 'webpage_url': url,
6598 })
6599 if self.get_param('playlist_items') == '0':
6600 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6601 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6602 entries.extend(map(self._real_extract, extra_tabs))
6603
6604 if len(entries) == 1:
6605 return entries[0]
6606 elif entries:
6607 metadata = self._extract_metadata_from_tabs(item_id, data)
6608 uploads_url = 'the Uploads (UU) playlist URL'
6609 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6610 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6611 self.to_screen(
6612 'Downloading as multiple playlists, separated by tabs. '
6613 f'To download as a single playlist instead, pass {uploads_url}')
6614 return self.playlist_result(entries, item_id, **metadata)
6615
6616 # Inline playlist
6617 playlist = traverse_obj(
6618 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6619 if playlist:
6620 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6621
6622 video_id = traverse_obj(
6623 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6624 if video_id:
6625 if tab != '/live': # live tab is expected to redirect to video
6626 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6627 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6628
6629 raise ExtractorError('Unable to recognize tab page')
6630
6631
6632 class YoutubePlaylistIE(InfoExtractor):
6633 IE_DESC = 'YouTube playlists'
6634 _VALID_URL = r'''(?x)(?:
6635 (?:https?://)?
6636 (?:\w+\.)?
6637 (?:
6638 (?:
6639 youtube(?:kids)?\.com|
6640 %(invidious)s
6641 )
6642 /.*?\?.*?\blist=
6643 )?
6644 (?P<id>%(playlist_id)s)
6645 )''' % {
6646 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6647 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6648 }
6649 IE_NAME = 'youtube:playlist'
6650 _TESTS = [{
6651 'note': 'issue #673',
6652 'url': 'PLBB231211A4F62143',
6653 'info_dict': {
6654 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6655 'id': 'PLBB231211A4F62143',
6656 'uploader': 'Wickman',
6657 'uploader_id': '@WickmanVT',
6658 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6659 'view_count': int,
6660 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6661 'modified_date': r're:\d{8}',
6662 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6663 'channel': 'Wickman',
6664 'tags': [],
6665 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6666 'availability': 'public',
6667 },
6668 'playlist_mincount': 29,
6669 }, {
6670 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6671 'info_dict': {
6672 'title': 'YDL_safe_search',
6673 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6674 },
6675 'playlist_count': 2,
6676 'skip': 'This playlist is private',
6677 }, {
6678 'note': 'embedded',
6679 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6680 'playlist_count': 4,
6681 'info_dict': {
6682 'title': 'JODA15',
6683 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6684 'uploader': 'milan',
6685 'uploader_id': '@milan5503',
6686 'description': '',
6687 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6688 'tags': [],
6689 'modified_date': '20140919',
6690 'view_count': int,
6691 'channel': 'milan',
6692 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6693 'uploader_url': 'https://www.youtube.com/@milan5503',
6694 'availability': 'public',
6695 },
6696 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden', 'Retrying', 'Giving up'],
6697 }, {
6698 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6699 'playlist_mincount': 455,
6700 'info_dict': {
6701 'title': '2018 Chinese New Singles (11/6 updated)',
6702 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6703 'uploader': 'LBK',
6704 'uploader_id': '@music_king',
6705 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6706 'channel': 'LBK',
6707 'view_count': int,
6708 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6709 'tags': [],
6710 'uploader_url': 'https://www.youtube.com/@music_king',
6711 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6712 'modified_date': r're:\d{8}',
6713 'availability': 'public',
6714 },
6715 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6716 }, {
6717 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6718 'only_matching': True,
6719 }, {
6720 # music album playlist
6721 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6722 'only_matching': True,
6723 }]
6724
6725 @classmethod
6726 def suitable(cls, url):
6727 if YoutubeTabIE.suitable(url):
6728 return False
6729 from ..utils import parse_qs
6730 qs = parse_qs(url)
6731 if qs.get('v', [None])[0]:
6732 return False
6733 return super().suitable(url)
6734
6735 def _real_extract(self, url):
6736 playlist_id = self._match_id(url)
6737 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6738 url = update_url_query(
6739 'https://www.youtube.com/playlist',
6740 parse_qs(url) or {'list': playlist_id})
6741 if is_music_url:
6742 url = smuggle_url(url, {'is_music_url': True})
6743 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6744
6745
6746 class YoutubeYtBeIE(InfoExtractor):
6747 IE_DESC = 'youtu.be'
6748 _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}
6749 _TESTS = [{
6750 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6751 'info_dict': {
6752 'id': 'yeWKywCrFtk',
6753 'ext': 'mp4',
6754 'title': 'Small Scale Baler and Braiding Rugs',
6755 'uploader': 'Backus-Page House Museum',
6756 'uploader_id': '@backuspagemuseum',
6757 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6758 'upload_date': '20161008',
6759 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6760 'categories': ['Nonprofits & Activism'],
6761 'tags': list,
6762 'like_count': int,
6763 'age_limit': 0,
6764 'playable_in_embed': True,
6765 'thumbnail': r're:^https?://.*\.webp',
6766 'channel': 'Backus-Page House Museum',
6767 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6768 'live_status': 'not_live',
6769 'view_count': int,
6770 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6771 'availability': 'public',
6772 'duration': 59,
6773 'comment_count': int,
6774 'channel_follower_count': int
6775 },
6776 'params': {
6777 'noplaylist': True,
6778 'skip_download': True,
6779 },
6780 }, {
6781 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6782 'only_matching': True,
6783 }]
6784
6785 def _real_extract(self, url):
6786 mobj = self._match_valid_url(url)
6787 video_id = mobj.group('id')
6788 playlist_id = mobj.group('playlist_id')
6789 return self.url_result(
6790 update_url_query('https://www.youtube.com/watch', {
6791 'v': video_id,
6792 'list': playlist_id,
6793 'feature': 'youtu.be',
6794 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6795
6796
6797 class YoutubeLivestreamEmbedIE(InfoExtractor):
6798 IE_DESC = 'YouTube livestream embeds'
6799 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6800 _TESTS = [{
6801 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6802 'only_matching': True,
6803 }]
6804
6805 def _real_extract(self, url):
6806 channel_id = self._match_id(url)
6807 return self.url_result(
6808 f'https://www.youtube.com/channel/{channel_id}/live',
6809 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6810
6811
6812 class YoutubeYtUserIE(InfoExtractor):
6813 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6814 IE_NAME = 'youtube:user'
6815 _VALID_URL = r'ytuser:(?P<id>.+)'
6816 _TESTS = [{
6817 'url': 'ytuser:phihag',
6818 'only_matching': True,
6819 }]
6820
6821 def _real_extract(self, url):
6822 user_id = self._match_id(url)
6823 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6824
6825
6826 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6827 IE_NAME = 'youtube:favorites'
6828 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6829 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6830 _LOGIN_REQUIRED = True
6831 _TESTS = [{
6832 'url': ':ytfav',
6833 'only_matching': True,
6834 }, {
6835 'url': ':ytfavorites',
6836 'only_matching': True,
6837 }]
6838
6839 def _real_extract(self, url):
6840 return self.url_result(
6841 'https://www.youtube.com/playlist?list=LL',
6842 ie=YoutubeTabIE.ie_key())
6843
6844
6845 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6846 IE_NAME = 'youtube:notif'
6847 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6848 _VALID_URL = r':ytnotif(?:ication)?s?'
6849 _LOGIN_REQUIRED = True
6850 _TESTS = [{
6851 'url': ':ytnotif',
6852 'only_matching': True,
6853 }, {
6854 'url': ':ytnotifications',
6855 'only_matching': True,
6856 }]
6857
6858 def _extract_notification_menu(self, response, continuation_list):
6859 notification_list = traverse_obj(
6860 response,
6861 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6862 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6863 expected_type=list) or []
6864 continuation_list[0] = None
6865 for item in notification_list:
6866 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6867 if entry:
6868 yield entry
6869 continuation = item.get('continuationItemRenderer')
6870 if continuation:
6871 continuation_list[0] = continuation
6872
6873 def _extract_notification_renderer(self, notification):
6874 video_id = traverse_obj(
6875 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6876 url = f'https://www.youtube.com/watch?v={video_id}'
6877 channel_id = None
6878 if not video_id:
6879 browse_ep = traverse_obj(
6880 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6881 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6882 post_id = self._search_regex(
6883 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6884 'post id', default=None)
6885 if not channel_id or not post_id:
6886 return
6887 # The direct /post url redirects to this in the browser
6888 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6889
6890 channel = traverse_obj(
6891 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6892 expected_type=str)
6893 notification_title = self._get_text(notification, 'shortMessage')
6894 if notification_title:
6895 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6896 # TODO: handle recommended videos
6897 title = self._search_regex(
6898 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6899 'video title', default=None)
6900 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6901 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6902 else None)
6903 return {
6904 '_type': 'url',
6905 'url': url,
6906 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6907 'video_id': video_id,
6908 'title': title,
6909 'channel_id': channel_id,
6910 'channel': channel,
6911 'uploader': channel,
6912 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6913 'timestamp': timestamp,
6914 }
6915
6916 def _notification_menu_entries(self, ytcfg):
6917 continuation_list = [None]
6918 response = None
6919 for page in itertools.count(1):
6920 ctoken = traverse_obj(
6921 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6922 response = self._extract_response(
6923 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6924 ep='notification/get_notification_menu', check_get_keys='actions',
6925 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6926 yield from self._extract_notification_menu(response, continuation_list)
6927 if not continuation_list[0]:
6928 break
6929
6930 def _real_extract(self, url):
6931 display_id = 'notifications'
6932 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6933 self._report_playlist_authcheck(ytcfg)
6934 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6935
6936
6937 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6938 IE_DESC = 'YouTube search'
6939 IE_NAME = 'youtube:search'
6940 _SEARCH_KEY = 'ytsearch'
6941 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6942 _TESTS = [{
6943 'url': 'ytsearch5:youtube-dl test video',
6944 'playlist_count': 5,
6945 'info_dict': {
6946 'id': 'youtube-dl test video',
6947 'title': 'youtube-dl test video',
6948 }
6949 }]
6950
6951
6952 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6953 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6954 _SEARCH_KEY = 'ytsearchdate'
6955 IE_DESC = 'YouTube search, newest videos first'
6956 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6957 _TESTS = [{
6958 'url': 'ytsearchdate5:youtube-dl test video',
6959 'playlist_count': 5,
6960 'info_dict': {
6961 'id': 'youtube-dl test video',
6962 'title': 'youtube-dl test video',
6963 }
6964 }]
6965
6966
6967 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6968 IE_DESC = 'YouTube search URLs with sorting and filter support'
6969 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6970 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6971 _TESTS = [{
6972 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6973 'playlist_mincount': 5,
6974 'info_dict': {
6975 'id': 'youtube-dl test video',
6976 'title': 'youtube-dl test video',
6977 }
6978 }, {
6979 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6980 'playlist_mincount': 5,
6981 'info_dict': {
6982 'id': 'python',
6983 'title': 'python',
6984 }
6985 }, {
6986 'url': 'https://www.youtube.com/results?search_query=%23cats',
6987 'playlist_mincount': 1,
6988 'info_dict': {
6989 'id': '#cats',
6990 'title': '#cats',
6991 # The test suite does not have support for nested playlists
6992 # 'entries': [{
6993 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6994 # 'title': '#cats',
6995 # }],
6996 },
6997 }, {
6998 # Channel results
6999 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
7000 'info_dict': {
7001 'id': 'kurzgesagt',
7002 'title': 'kurzgesagt',
7003 },
7004 'playlist': [{
7005 'info_dict': {
7006 '_type': 'url',
7007 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7008 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7009 'ie_key': 'YoutubeTab',
7010 'channel': 'Kurzgesagt – In a Nutshell',
7011 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
7012 'title': 'Kurzgesagt – In a Nutshell',
7013 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7014 # No longer available for search as it is set to the handle.
7015 # 'playlist_count': int,
7016 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7017 'thumbnails': list,
7018 'uploader_id': '@kurzgesagt',
7019 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
7020 'uploader': 'Kurzgesagt – In a Nutshell',
7021 'channel_is_verified': True,
7022 'channel_follower_count': int,
7023 }
7024 }],
7025 'params': {'extract_flat': True, 'playlist_items': '1'},
7026 'playlist_mincount': 1,
7027 }, {
7028 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
7029 'only_matching': True,
7030 }]
7031
7032 def _real_extract(self, url):
7033 qs = parse_qs(url)
7034 query = (qs.get('search_query') or qs.get('q'))[0]
7035 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
7036
7037
7038 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
7039 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
7040 IE_NAME = 'youtube:music:search_url'
7041 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7042 _TESTS = [{
7043 'url': 'https://music.youtube.com/search?q=royalty+free+music',
7044 'playlist_count': 16,
7045 'info_dict': {
7046 'id': 'royalty free music',
7047 'title': 'royalty free music',
7048 }
7049 }, {
7050 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7051 'playlist_mincount': 30,
7052 'info_dict': {
7053 'id': 'royalty free music - songs',
7054 'title': 'royalty free music - songs',
7055 },
7056 'params': {'extract_flat': 'in_playlist'}
7057 }, {
7058 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7059 'playlist_mincount': 30,
7060 'info_dict': {
7061 'id': 'royalty free music - community playlists',
7062 'title': 'royalty free music - community playlists',
7063 },
7064 'params': {'extract_flat': 'in_playlist'}
7065 }]
7066
7067 _SECTIONS = {
7068 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7069 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7070 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7071 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7072 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7073 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7074 }
7075
7076 def _real_extract(self, url):
7077 qs = parse_qs(url)
7078 query = (qs.get('search_query') or qs.get('q'))[0]
7079 params = qs.get('sp', (None,))[0]
7080 if params:
7081 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7082 else:
7083 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
7084 params = self._SECTIONS.get(section)
7085 if not params:
7086 section = None
7087 title = join_nonempty(query, section, delim=' - ')
7088 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7089
7090
7091 class YoutubeFeedsInfoExtractor(InfoExtractor):
7092 """
7093 Base class for feed extractors
7094 Subclasses must re-define the _FEED_NAME property.
7095 """
7096 _LOGIN_REQUIRED = True
7097 _FEED_NAME = 'feeds'
7098
7099 def _real_initialize(self):
7100 YoutubeBaseInfoExtractor._check_login_required(self)
7101
7102 @classproperty
7103 def IE_NAME(self):
7104 return f'youtube:{self._FEED_NAME}'
7105
7106 def _real_extract(self, url):
7107 return self.url_result(
7108 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7109
7110
7111 class YoutubeWatchLaterIE(InfoExtractor):
7112 IE_NAME = 'youtube:watchlater'
7113 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7114 _VALID_URL = r':ytwatchlater'
7115 _TESTS = [{
7116 'url': ':ytwatchlater',
7117 'only_matching': True,
7118 }]
7119
7120 def _real_extract(self, url):
7121 return self.url_result(
7122 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7123
7124
7125 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7126 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7127 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7128 _FEED_NAME = 'recommended'
7129 _LOGIN_REQUIRED = False
7130 _TESTS = [{
7131 'url': ':ytrec',
7132 'only_matching': True,
7133 }, {
7134 'url': ':ytrecommended',
7135 'only_matching': True,
7136 }, {
7137 'url': 'https://youtube.com',
7138 'only_matching': True,
7139 }]
7140
7141
7142 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7143 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7144 _VALID_URL = r':ytsub(?:scription)?s?'
7145 _FEED_NAME = 'subscriptions'
7146 _TESTS = [{
7147 'url': ':ytsubs',
7148 'only_matching': True,
7149 }, {
7150 'url': ':ytsubscriptions',
7151 'only_matching': True,
7152 }]
7153
7154
7155 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7156 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7157 _VALID_URL = r':ythis(?:tory)?'
7158 _FEED_NAME = 'history'
7159 _TESTS = [{
7160 'url': ':ythistory',
7161 'only_matching': True,
7162 }]
7163
7164
7165 class YoutubeShortsAudioPivotIE(InfoExtractor):
7166 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7167 IE_NAME = 'youtube:shorts:pivot:audio'
7168 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7169 _TESTS = [{
7170 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7171 'only_matching': True,
7172 }]
7173
7174 @staticmethod
7175 def _generate_audio_pivot_params(video_id):
7176 """
7177 Generates sfv_audio_pivot browse params for this video id
7178 """
7179 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7180 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7181
7182 def _real_extract(self, url):
7183 video_id = self._match_id(url)
7184 return self.url_result(
7185 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7186 ie=YoutubeTabIE)
7187
7188
7189 class YoutubeTruncatedURLIE(InfoExtractor):
7190 IE_NAME = 'youtube:truncated_url'
7191 IE_DESC = False # Do not list
7192 _VALID_URL = r'''(?x)
7193 (?:https?://)?
7194 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7195 (?:watch\?(?:
7196 feature=[a-z_]+|
7197 annotation_id=annotation_[^&]+|
7198 x-yt-cl=[0-9]+|
7199 hl=[^&]*|
7200 t=[0-9]+
7201 )?
7202 |
7203 attribution_link\?a=[^&]+
7204 )
7205 $
7206 '''
7207
7208 _TESTS = [{
7209 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7210 'only_matching': True,
7211 }, {
7212 'url': 'https://www.youtube.com/watch?',
7213 'only_matching': True,
7214 }, {
7215 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7216 'only_matching': True,
7217 }, {
7218 'url': 'https://www.youtube.com/watch?feature=foo',
7219 'only_matching': True,
7220 }, {
7221 'url': 'https://www.youtube.com/watch?hl=en-GB',
7222 'only_matching': True,
7223 }, {
7224 'url': 'https://www.youtube.com/watch?t=2372',
7225 'only_matching': True,
7226 }]
7227
7228 def _real_extract(self, url):
7229 raise ExtractorError(
7230 'Did you forget to quote the URL? Remember that & is a meta '
7231 'character in most shells, so you want to put the URL in quotes, '
7232 'like youtube-dl '
7233 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7234 ' or simply youtube-dl BaW_jenozKc .',
7235 expected=True)
7236
7237
7238 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7239 IE_NAME = 'youtube:clip'
7240 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7241 _TESTS = [{
7242 # FIXME: Other metadata should be extracted from the clip, not from the base video
7243 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7244 'info_dict': {
7245 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7246 'ext': 'mp4',
7247 'section_start': 29.0,
7248 'section_end': 39.7,
7249 'duration': 10.7,
7250 'age_limit': 0,
7251 'availability': 'public',
7252 'categories': ['Gaming'],
7253 'channel': 'Scott The Woz',
7254 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7255 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7256 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7257 'like_count': int,
7258 'playable_in_embed': True,
7259 'tags': 'count:17',
7260 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7261 'title': 'Mobile Games on Console - Scott The Woz',
7262 'upload_date': '20210920',
7263 'uploader': 'Scott The Woz',
7264 'uploader_id': '@ScottTheWoz',
7265 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7266 'view_count': int,
7267 'live_status': 'not_live',
7268 'channel_follower_count': int,
7269 'chapters': 'count:20',
7270 'comment_count': int,
7271 'heatmap': 'count:100',
7272 }
7273 }]
7274
7275 def _real_extract(self, url):
7276 clip_id = self._match_id(url)
7277 _, data = self._extract_webpage(url, clip_id)
7278
7279 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7280 if not video_id:
7281 raise ExtractorError('Unable to find video ID')
7282
7283 clip_data = traverse_obj(data, (
7284 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7285 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7286 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7287 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7288
7289 return {
7290 '_type': 'url_transparent',
7291 'url': f'https://www.youtube.com/watch?v={video_id}',
7292 'ie_key': YoutubeIE.ie_key(),
7293 'id': clip_id,
7294 'section_start': int(clip_data['startTimeMs']) / 1000,
7295 'section_end': int(clip_data['endTimeMs']) / 1000,
7296 }
7297
7298
7299 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7300 IE_NAME = 'youtube:consent'
7301 IE_DESC = False # Do not list
7302 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7303 _TESTS = [{
7304 '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',
7305 'info_dict': {
7306 'id': 'qVv6vCqciTM',
7307 'ext': 'mp4',
7308 'age_limit': 0,
7309 'uploader_id': '@sana_natori',
7310 'comment_count': int,
7311 'chapters': 'count:13',
7312 'upload_date': '20221223',
7313 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7314 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7315 'uploader_url': 'https://www.youtube.com/@sana_natori',
7316 'like_count': int,
7317 'release_date': '20221223',
7318 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7319 '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】',
7320 'view_count': int,
7321 'playable_in_embed': True,
7322 'duration': 4438,
7323 'availability': 'public',
7324 'channel_follower_count': int,
7325 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7326 'categories': ['Entertainment'],
7327 'live_status': 'was_live',
7328 'release_timestamp': 1671793345,
7329 'channel': 'さなちゃんねる',
7330 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7331 'uploader': 'さなちゃんねる',
7332 'channel_is_verified': True,
7333 'heatmap': 'count:100',
7334 },
7335 'add_ie': ['Youtube'],
7336 'params': {'skip_download': 'Youtube'},
7337 }]
7338
7339 def _real_extract(self, url):
7340 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7341 if not redirect_url:
7342 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7343 return self.url_result(redirect_url)
7344
7345
7346 class YoutubeTruncatedIDIE(InfoExtractor):
7347 IE_NAME = 'youtube:truncated_id'
7348 IE_DESC = False # Do not list
7349 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7350
7351 _TESTS = [{
7352 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7353 'only_matching': True,
7354 }]
7355
7356 def _real_extract(self, url):
7357 video_id = self._match_id(url)
7358 raise ExtractorError(
7359 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7360 expected=True)