]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[cleanup] Misc (#8182)
[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\.)?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 for retry in self.RetryManager():
945 try:
946 response = self._call_api(
947 ep=ep, fatal=True, headers=headers,
948 video_id=item_id, query=query, note=note,
949 context=self._extract_context(ytcfg, default_client),
950 api_key=self._extract_api_key(ytcfg, default_client),
951 api_hostname=api_hostname, default_client=default_client)
952 except ExtractorError as e:
953 if not isinstance(e.cause, network_exceptions):
954 return self._error_or_warning(e, fatal=fatal)
955 elif not isinstance(e.cause, HTTPError):
956 retry.error = e
957 continue
958
959 first_bytes = e.cause.response.read(512)
960 if not is_html(first_bytes):
961 yt_error = try_get(
962 self._parse_json(
963 self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
964 lambda x: x['error']['message'], str)
965 if yt_error:
966 self._report_alerts([('ERROR', yt_error)], fatal=False)
967 # Downloading page may result in intermittent 5xx HTTP error
968 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
969 # We also want to catch all other network exceptions since errors in later pages can be troublesome
970 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
971 if e.cause.status not in (403, 429):
972 retry.error = e
973 continue
974 return self._error_or_warning(e, fatal=fatal)
975
976 try:
977 self._extract_and_report_alerts(response, only_once=True)
978 except ExtractorError as e:
979 # YouTube servers may return errors we want to retry on in a 200 OK response
980 # See: https://github.com/yt-dlp/yt-dlp/issues/839
981 if 'unknown error' in e.msg.lower():
982 retry.error = e
983 continue
984 return self._error_or_warning(e, fatal=fatal)
985 # Youtube sometimes sends incomplete data
986 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
987 if not traverse_obj(response, *variadic(check_get_keys)):
988 retry.error = ExtractorError('Incomplete data received', expected=True)
989 continue
990
991 return response
992
993 @staticmethod
994 def is_music_url(url):
995 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
996
997 def _extract_video(self, renderer):
998 video_id = renderer.get('videoId')
999
1000 reel_header_renderer = traverse_obj(renderer, (
1001 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1002 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1003
1004 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1005 description = self._get_text(renderer, 'descriptionSnippet')
1006
1007 duration = int_or_none(renderer.get('lengthSeconds'))
1008 if duration is None:
1009 duration = parse_duration(self._get_text(
1010 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1011 if duration is None:
1012 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1013 duration = parse_duration(self._search_regex(
1014 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1015 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1016 video_id, default=None, group='duration'))
1017
1018 channel_id = traverse_obj(
1019 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1020 expected_type=str, get_all=False)
1021 if not channel_id:
1022 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1023
1024 channel_id = self.ucid_or_none(channel_id)
1025
1026 overlay_style = traverse_obj(
1027 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1028 get_all=False, expected_type=str)
1029 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1030 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1031 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1032 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1033 expected_type=str)) or ''
1034 url = f'https://www.youtube.com/watch?v={video_id}'
1035 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1036 url = f'https://www.youtube.com/shorts/{video_id}'
1037
1038 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1039 or self._get_text(reel_header_renderer, 'timestampText') or '')
1040 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1041
1042 live_status = (
1043 'is_upcoming' if scheduled_timestamp is not None
1044 else 'was_live' if 'streamed' in time_text.lower()
1045 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1046 else None)
1047
1048 # videoInfo is a string like '50K views • 10 years ago'.
1049 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1050 view_count = (0 if 'no views' in view_count_text.lower()
1051 else self._get_count({'simpleText': view_count_text}))
1052 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1053
1054 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1055 or self._get_text(reel_header_renderer, 'channelTitleText'))
1056
1057 channel_handle = traverse_obj(renderer, (
1058 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1059 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1060 expected_type=self.handle_from_url, get_all=False)
1061 return {
1062 '_type': 'url',
1063 'ie_key': YoutubeIE.ie_key(),
1064 'id': video_id,
1065 'url': url,
1066 'title': title,
1067 'description': description,
1068 'duration': duration,
1069 'channel_id': channel_id,
1070 'channel': channel,
1071 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1072 'uploader': channel,
1073 'uploader_id': channel_handle,
1074 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1075 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1076 'timestamp': (self._parse_time_text(time_text)
1077 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1078 else None),
1079 'release_timestamp': scheduled_timestamp,
1080 'availability':
1081 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1082 else self._availability(
1083 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1084 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1085 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1086 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1087 view_count_field: view_count,
1088 'live_status': live_status,
1089 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None
1090 }
1091
1092
1093 class YoutubeIE(YoutubeBaseInfoExtractor):
1094 IE_DESC = 'YouTube'
1095 _VALID_URL = r"""(?x)^
1096 (
1097 (?:https?://|//) # http(s):// or protocol-independent URL
1098 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1099 (?:www\.)?deturl\.com/www\.youtube\.com|
1100 (?:www\.)?pwnyoutube\.com|
1101 (?:www\.)?hooktube\.com|
1102 (?:www\.)?yourepeat\.com|
1103 tube\.majestyc\.net|
1104 %(invidious)s|
1105 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1106 (?:.*?\#/)? # handle anchor (#/) redirect urls
1107 (?: # the various things that can precede the ID:
1108 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1109 |(?: # or the v= param in all its forms
1110 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1111 (?:\?|\#!?) # the params delimiter ? or # or #!
1112 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1113 v=
1114 )
1115 ))
1116 |(?:
1117 youtu\.be| # just youtu.be/xxxx
1118 vid\.plus| # or vid.plus/xxxx
1119 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1120 %(invidious)s
1121 )/
1122 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1123 )
1124 )? # all until now is optional -> you can pass the naked ID
1125 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1126 (?(1).+)? # if we found the ID, everything can follow
1127 (?:\#|$)""" % {
1128 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1129 }
1130 _EMBED_REGEX = [
1131 r'''(?x)
1132 (?:
1133 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1134 data-video-url=|
1135 <embed[^>]+?src=|
1136 embedSWF\(?:\s*|
1137 <object[^>]+data=|
1138 new\s+SWFObject\(
1139 )
1140 (["\'])
1141 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1142 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1143 \1''',
1144 # https://wordpress.org/plugins/lazy-load-for-videos/
1145 r'''(?xs)
1146 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1147 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1148 ]
1149 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1150
1151 _PLAYER_INFO_RE = (
1152 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1153 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$',
1154 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1155 )
1156 _formats = {
1157 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1158 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1159 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1160 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1161 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1162 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1163 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1164 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1165 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1166 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1167 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1168 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1169 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1170 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1171 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1172 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1173 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1174 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1175
1176
1177 # 3D videos
1178 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1179 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1180 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1181 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1182 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1183 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1184 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1185
1186 # Apple HTTP Live Streaming
1187 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1188 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1189 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1190 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1191 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1192 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1193 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1194 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1195
1196 # DASH mp4 video
1197 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1198 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1199 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1200 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1201 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1202 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1203 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1204 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1205 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1206 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1207 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1208 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1209
1210 # Dash mp4 audio
1211 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1212 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1213 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1214 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1215 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1216 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1217 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1218
1219 # Dash webm
1220 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1221 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1222 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1223 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1224 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1225 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1226 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1227 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1228 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1229 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1230 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1231 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1232 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1233 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1234 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1235 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1236 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1237 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1238 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1239 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1240 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1241 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1242
1243 # Dash webm audio
1244 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1245 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1246
1247 # Dash webm audio with opus inside
1248 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1249 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1250 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1251
1252 # RTMP (unnamed)
1253 '_rtmp': {'protocol': 'rtmp'},
1254
1255 # av01 video only formats sometimes served with "unknown" codecs
1256 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1257 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1258 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1259 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1260 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1261 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1262 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1263 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1264 }
1265 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1266
1267 _GEO_BYPASS = False
1268
1269 IE_NAME = 'youtube'
1270 _TESTS = [
1271 {
1272 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1273 'info_dict': {
1274 'id': 'BaW_jenozKc',
1275 'ext': 'mp4',
1276 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1277 'channel': 'Philipp Hagemeister',
1278 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1279 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1280 'upload_date': '20121002',
1281 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1282 'categories': ['Science & Technology'],
1283 'tags': ['youtube-dl'],
1284 'duration': 10,
1285 'view_count': int,
1286 'like_count': int,
1287 'availability': 'public',
1288 'playable_in_embed': True,
1289 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1290 'live_status': 'not_live',
1291 'age_limit': 0,
1292 'start_time': 1,
1293 'end_time': 9,
1294 'comment_count': int,
1295 'channel_follower_count': int,
1296 'uploader': 'Philipp Hagemeister',
1297 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1298 'uploader_id': '@PhilippHagemeister',
1299 'heatmap': 'count:100',
1300 }
1301 },
1302 {
1303 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1304 'note': 'Embed-only video (#1746)',
1305 'info_dict': {
1306 'id': 'yZIXLfi8CZQ',
1307 'ext': 'mp4',
1308 'upload_date': '20120608',
1309 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1310 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1311 'age_limit': 18,
1312 },
1313 'skip': 'Private video',
1314 },
1315 {
1316 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1317 'note': 'Use the first video ID in the URL',
1318 'info_dict': {
1319 'id': 'BaW_jenozKc',
1320 'ext': 'mp4',
1321 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1322 'channel': 'Philipp Hagemeister',
1323 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1324 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1325 'upload_date': '20121002',
1326 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1327 'categories': ['Science & Technology'],
1328 'tags': ['youtube-dl'],
1329 'duration': 10,
1330 'view_count': int,
1331 'like_count': int,
1332 'availability': 'public',
1333 'playable_in_embed': True,
1334 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1335 'live_status': 'not_live',
1336 'age_limit': 0,
1337 'comment_count': int,
1338 'channel_follower_count': int,
1339 'uploader': 'Philipp Hagemeister',
1340 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1341 'uploader_id': '@PhilippHagemeister',
1342 'heatmap': 'count:100',
1343 },
1344 'params': {
1345 'skip_download': True,
1346 },
1347 },
1348 {
1349 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1350 'note': '256k DASH audio (format 141) via DASH manifest',
1351 'info_dict': {
1352 'id': 'a9LDPn-MO4I',
1353 'ext': 'm4a',
1354 'upload_date': '20121002',
1355 'description': '',
1356 'title': 'UHDTV TEST 8K VIDEO.mp4'
1357 },
1358 'params': {
1359 'youtube_include_dash_manifest': True,
1360 'format': '141',
1361 },
1362 'skip': 'format 141 not served anymore',
1363 },
1364 # DASH manifest with encrypted signature
1365 {
1366 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1367 'info_dict': {
1368 'id': 'IB3lcPjvWLA',
1369 'ext': 'm4a',
1370 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1371 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1372 'duration': 244,
1373 'upload_date': '20131011',
1374 'abr': 129.495,
1375 'like_count': int,
1376 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1377 'playable_in_embed': True,
1378 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1379 'view_count': int,
1380 'track': 'The Spark',
1381 'live_status': 'not_live',
1382 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1383 'channel': 'Afrojack',
1384 'tags': 'count:19',
1385 'availability': 'public',
1386 'categories': ['Music'],
1387 'age_limit': 0,
1388 'alt_title': 'The Spark',
1389 'channel_follower_count': int,
1390 'uploader': 'Afrojack',
1391 'uploader_url': 'https://www.youtube.com/@Afrojack',
1392 'uploader_id': '@Afrojack',
1393 },
1394 'params': {
1395 'youtube_include_dash_manifest': True,
1396 'format': '141/bestaudio[ext=m4a]',
1397 },
1398 },
1399 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1400 {
1401 'note': 'Embed allowed age-gate video',
1402 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1403 'info_dict': {
1404 'id': 'HtVdAasjOgU',
1405 'ext': 'mp4',
1406 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1407 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1408 'duration': 142,
1409 'upload_date': '20140605',
1410 'age_limit': 18,
1411 'categories': ['Gaming'],
1412 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1413 'availability': 'needs_auth',
1414 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1415 'like_count': int,
1416 'channel': 'The Witcher',
1417 'live_status': 'not_live',
1418 'tags': 'count:17',
1419 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1420 'playable_in_embed': True,
1421 'view_count': int,
1422 'channel_follower_count': int,
1423 'uploader': 'The Witcher',
1424 'uploader_url': 'https://www.youtube.com/@thewitcher',
1425 'uploader_id': '@thewitcher',
1426 'comment_count': int,
1427 'channel_is_verified': True,
1428 'heatmap': 'count:100',
1429 },
1430 },
1431 {
1432 'note': 'Age-gate video with embed allowed in public site',
1433 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1434 'info_dict': {
1435 'id': 'HsUATh_Nc2U',
1436 'ext': 'mp4',
1437 'title': 'Godzilla 2 (Official Video)',
1438 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1439 'upload_date': '20200408',
1440 'age_limit': 18,
1441 'availability': 'needs_auth',
1442 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1443 'channel': 'FlyingKitty',
1444 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1445 'view_count': int,
1446 'categories': ['Entertainment'],
1447 'live_status': 'not_live',
1448 'tags': ['Flyingkitty', 'godzilla 2'],
1449 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1450 'like_count': int,
1451 'duration': 177,
1452 'playable_in_embed': True,
1453 'channel_follower_count': int,
1454 'uploader': 'FlyingKitty',
1455 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1456 'uploader_id': '@FlyingKitty900',
1457 'comment_count': int,
1458 'channel_is_verified': True,
1459 },
1460 },
1461 {
1462 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1463 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1464 'info_dict': {
1465 'id': 'Tq92D6wQ1mg',
1466 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1467 'ext': 'mp4',
1468 'upload_date': '20191228',
1469 'description': 'md5:17eccca93a786d51bc67646756894066',
1470 'age_limit': 18,
1471 'like_count': int,
1472 'availability': 'needs_auth',
1473 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1474 'view_count': int,
1475 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1476 'channel': 'Projekt Melody',
1477 'live_status': 'not_live',
1478 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1479 'playable_in_embed': True,
1480 'categories': ['Entertainment'],
1481 'duration': 106,
1482 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1483 'comment_count': int,
1484 'channel_follower_count': int,
1485 'uploader': 'Projekt Melody',
1486 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1487 'uploader_id': '@ProjektMelody',
1488 },
1489 },
1490 {
1491 'note': 'Non-Agegated non-embeddable video',
1492 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1493 'info_dict': {
1494 'id': 'MeJVWBSsPAY',
1495 'ext': 'mp4',
1496 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1497 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1498 'upload_date': '20130730',
1499 'track': 'Such mich find mich',
1500 'age_limit': 0,
1501 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1502 'like_count': int,
1503 'playable_in_embed': False,
1504 'creator': 'OOMPH!',
1505 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1506 'view_count': int,
1507 'alt_title': 'Such mich find mich',
1508 'duration': 210,
1509 'channel': 'Herr Lurik',
1510 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1511 'categories': ['Music'],
1512 'availability': 'public',
1513 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1514 'live_status': 'not_live',
1515 'artist': 'OOMPH!',
1516 'channel_follower_count': int,
1517 'uploader': 'Herr Lurik',
1518 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1519 'uploader_id': '@HerrLurik',
1520 },
1521 },
1522 {
1523 'note': 'Non-bypassable age-gated video',
1524 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1525 'only_matching': True,
1526 },
1527 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1528 # YouTube Red ad is not captured for creator
1529 {
1530 'url': '__2ABJjxzNo',
1531 'info_dict': {
1532 'id': '__2ABJjxzNo',
1533 'ext': 'mp4',
1534 'duration': 266,
1535 'upload_date': '20100430',
1536 'creator': 'deadmau5',
1537 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1538 'title': 'Deadmau5 - Some Chords (HD)',
1539 'alt_title': 'Some Chords',
1540 'availability': 'public',
1541 'tags': 'count:14',
1542 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1543 'view_count': int,
1544 'live_status': 'not_live',
1545 'channel': 'deadmau5',
1546 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1547 'like_count': int,
1548 'track': 'Some Chords',
1549 'artist': 'deadmau5',
1550 'playable_in_embed': True,
1551 'age_limit': 0,
1552 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1553 'categories': ['Music'],
1554 'album': 'Some Chords',
1555 'channel_follower_count': int,
1556 'uploader': 'deadmau5',
1557 'uploader_url': 'https://www.youtube.com/@deadmau5',
1558 'uploader_id': '@deadmau5',
1559 },
1560 'expected_warnings': [
1561 'DASH manifest missing',
1562 ]
1563 },
1564 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1565 {
1566 'url': 'lqQg6PlCWgI',
1567 'info_dict': {
1568 'id': 'lqQg6PlCWgI',
1569 'ext': 'mp4',
1570 'duration': 6085,
1571 'upload_date': '20150827',
1572 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1573 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1574 'like_count': int,
1575 'release_timestamp': 1343767800,
1576 'playable_in_embed': True,
1577 'categories': ['Sports'],
1578 'release_date': '20120731',
1579 'channel': 'Olympics',
1580 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1581 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1582 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1583 'age_limit': 0,
1584 'availability': 'public',
1585 'live_status': 'was_live',
1586 'view_count': int,
1587 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1588 'channel_follower_count': int,
1589 'uploader': 'Olympics',
1590 'uploader_url': 'https://www.youtube.com/@Olympics',
1591 'uploader_id': '@Olympics',
1592 'channel_is_verified': True,
1593 },
1594 'params': {
1595 'skip_download': 'requires avconv',
1596 }
1597 },
1598 # Non-square pixels
1599 {
1600 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1601 'info_dict': {
1602 'id': '_b-2C3KPAM0',
1603 'ext': 'mp4',
1604 'stretched_ratio': 16 / 9.,
1605 'duration': 85,
1606 'upload_date': '20110310',
1607 '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',
1608 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1609 'playable_in_embed': True,
1610 'channel': '孫ᄋᄅ',
1611 'age_limit': 0,
1612 'tags': 'count:11',
1613 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1614 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1615 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1616 'view_count': int,
1617 'categories': ['People & Blogs'],
1618 'like_count': int,
1619 'live_status': 'not_live',
1620 'availability': 'unlisted',
1621 'comment_count': int,
1622 'channel_follower_count': int,
1623 'uploader': '孫ᄋᄅ',
1624 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1625 'uploader_id': '@AllenMeow',
1626 },
1627 },
1628 # url_encoded_fmt_stream_map is empty string
1629 {
1630 'url': 'qEJwOuvDf7I',
1631 'info_dict': {
1632 'id': 'qEJwOuvDf7I',
1633 'ext': 'webm',
1634 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1635 'description': '',
1636 'upload_date': '20150404',
1637 },
1638 'params': {
1639 'skip_download': 'requires avconv',
1640 },
1641 'skip': 'This live event has ended.',
1642 },
1643 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1644 {
1645 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1646 'info_dict': {
1647 'id': 'FIl7x6_3R5Y',
1648 'ext': 'webm',
1649 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1650 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1651 'duration': 220,
1652 'upload_date': '20150625',
1653 'formats': 'mincount:31',
1654 },
1655 'skip': 'not actual anymore',
1656 },
1657 # DASH manifest with segment_list
1658 {
1659 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1660 'md5': '8ce563a1d667b599d21064e982ab9e31',
1661 'info_dict': {
1662 'id': 'CsmdDsKjzN8',
1663 'ext': 'mp4',
1664 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1665 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1666 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1667 },
1668 'params': {
1669 'youtube_include_dash_manifest': True,
1670 'format': '135', # bestvideo
1671 },
1672 'skip': 'This live event has ended.',
1673 },
1674 {
1675 # Multifeed videos (multiple cameras), URL can be of any Camera
1676 # TODO: fix multifeed titles
1677 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1678 'info_dict': {
1679 'id': 'zaPI8MvL8pg',
1680 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1681 'description': 'md5:563ccbc698b39298481ca3c571169519',
1682 },
1683 'playlist': [{
1684 'info_dict': {
1685 'id': 'j5yGuxZ8lLU',
1686 'ext': 'mp4',
1687 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1688 'description': 'md5:563ccbc698b39298481ca3c571169519',
1689 'duration': 10120,
1690 'channel_follower_count': int,
1691 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1692 'availability': 'public',
1693 'playable_in_embed': True,
1694 'upload_date': '20131105',
1695 'categories': ['Gaming'],
1696 'live_status': 'was_live',
1697 'tags': 'count:24',
1698 'release_timestamp': 1383701910,
1699 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1700 'comment_count': int,
1701 'age_limit': 0,
1702 'like_count': int,
1703 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1704 'channel': 'WiiLikeToPlay',
1705 'view_count': int,
1706 'release_date': '20131106',
1707 'uploader': 'WiiLikeToPlay',
1708 'uploader_id': '@WLTP',
1709 'uploader_url': 'https://www.youtube.com/@WLTP',
1710 },
1711 }, {
1712 'info_dict': {
1713 'id': 'zaPI8MvL8pg',
1714 'ext': 'mp4',
1715 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1716 'availability': 'public',
1717 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1718 'channel': 'WiiLikeToPlay',
1719 'channel_follower_count': int,
1720 'description': 'md5:563ccbc698b39298481ca3c571169519',
1721 'duration': 10108,
1722 'age_limit': 0,
1723 'like_count': int,
1724 'tags': 'count:24',
1725 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1726 'release_timestamp': 1383701915,
1727 'comment_count': int,
1728 'upload_date': '20131105',
1729 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1730 'release_date': '20131106',
1731 'playable_in_embed': True,
1732 'live_status': 'was_live',
1733 'categories': ['Gaming'],
1734 'view_count': int,
1735 'uploader': 'WiiLikeToPlay',
1736 'uploader_id': '@WLTP',
1737 'uploader_url': 'https://www.youtube.com/@WLTP',
1738 },
1739 }, {
1740 'info_dict': {
1741 'id': 'R7r3vfO7Hao',
1742 'ext': 'mp4',
1743 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1744 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1745 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1746 'like_count': int,
1747 'availability': 'public',
1748 'playable_in_embed': True,
1749 'upload_date': '20131105',
1750 'description': 'md5:563ccbc698b39298481ca3c571169519',
1751 'channel_follower_count': int,
1752 'tags': 'count:24',
1753 'release_date': '20131106',
1754 'comment_count': int,
1755 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1756 'channel': 'WiiLikeToPlay',
1757 'categories': ['Gaming'],
1758 'release_timestamp': 1383701914,
1759 'live_status': 'was_live',
1760 'age_limit': 0,
1761 'duration': 10128,
1762 'view_count': int,
1763 'uploader': 'WiiLikeToPlay',
1764 'uploader_id': '@WLTP',
1765 'uploader_url': 'https://www.youtube.com/@WLTP',
1766 },
1767 }],
1768 'params': {'skip_download': True},
1769 },
1770 {
1771 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1772 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1773 'info_dict': {
1774 'id': 'gVfLd0zydlo',
1775 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1776 },
1777 'playlist_count': 2,
1778 'skip': 'Not multifeed anymore',
1779 },
1780 {
1781 'url': 'https://vid.plus/FlRa-iH7PGw',
1782 'only_matching': True,
1783 },
1784 {
1785 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1786 'only_matching': True,
1787 },
1788 {
1789 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1790 # Also tests cut-off URL expansion in video description (see
1791 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1792 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1793 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1794 'info_dict': {
1795 'id': 'lsguqyKfVQg',
1796 'ext': 'mp4',
1797 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1798 'alt_title': 'Dark Walk',
1799 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1800 'duration': 133,
1801 'upload_date': '20151119',
1802 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1803 'track': 'Dark Walk',
1804 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1805 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1806 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1807 'categories': ['Film & Animation'],
1808 'view_count': int,
1809 'live_status': 'not_live',
1810 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1811 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1812 'tags': 'count:13',
1813 'availability': 'public',
1814 'channel': 'IronSoulElf',
1815 'playable_in_embed': True,
1816 'like_count': int,
1817 'age_limit': 0,
1818 'channel_follower_count': int
1819 },
1820 'params': {
1821 'skip_download': True,
1822 },
1823 },
1824 {
1825 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1826 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1827 'only_matching': True,
1828 },
1829 {
1830 # Video with yt:stretch=17:0
1831 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1832 'info_dict': {
1833 'id': 'Q39EVAstoRM',
1834 'ext': 'mp4',
1835 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1836 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1837 'upload_date': '20151107',
1838 },
1839 'params': {
1840 'skip_download': True,
1841 },
1842 'skip': 'This video does not exist.',
1843 },
1844 {
1845 # Video with incomplete 'yt:stretch=16:'
1846 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1847 'only_matching': True,
1848 },
1849 {
1850 # Video licensed under Creative Commons
1851 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1852 'info_dict': {
1853 'id': 'M4gD1WSo5mA',
1854 'ext': 'mp4',
1855 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1856 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1857 'duration': 721,
1858 'upload_date': '20150128',
1859 'license': 'Creative Commons Attribution license (reuse allowed)',
1860 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1861 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1862 'like_count': int,
1863 'age_limit': 0,
1864 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1865 'channel': 'The Berkman Klein Center for Internet & Society',
1866 'availability': 'public',
1867 'view_count': int,
1868 'categories': ['Education'],
1869 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1870 'live_status': 'not_live',
1871 'playable_in_embed': True,
1872 'channel_follower_count': int,
1873 'chapters': list,
1874 'uploader': 'The Berkman Klein Center for Internet & Society',
1875 'uploader_id': '@BKCHarvard',
1876 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1877 },
1878 'params': {
1879 'skip_download': True,
1880 },
1881 },
1882 {
1883 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1884 'info_dict': {
1885 'id': 'eQcmzGIKrzg',
1886 'ext': 'mp4',
1887 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1888 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1889 'duration': 4060,
1890 'upload_date': '20151120',
1891 'license': 'Creative Commons Attribution license (reuse allowed)',
1892 'playable_in_embed': True,
1893 'tags': 'count:12',
1894 'like_count': int,
1895 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1896 'age_limit': 0,
1897 'availability': 'public',
1898 'categories': ['News & Politics'],
1899 'channel': 'Bernie Sanders',
1900 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1901 'view_count': int,
1902 'live_status': 'not_live',
1903 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1904 'comment_count': int,
1905 'channel_follower_count': int,
1906 'chapters': list,
1907 'uploader': 'Bernie Sanders',
1908 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1909 'uploader_id': '@BernieSanders',
1910 'channel_is_verified': True,
1911 'heatmap': 'count:100',
1912 },
1913 'params': {
1914 'skip_download': True,
1915 },
1916 },
1917 {
1918 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1919 'only_matching': True,
1920 },
1921 {
1922 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1923 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1924 'only_matching': True,
1925 },
1926 {
1927 # Rental video preview
1928 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1929 'info_dict': {
1930 'id': 'uGpuVWrhIzE',
1931 'ext': 'mp4',
1932 'title': 'Piku - Trailer',
1933 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1934 'upload_date': '20150811',
1935 'license': 'Standard YouTube License',
1936 },
1937 'params': {
1938 'skip_download': True,
1939 },
1940 'skip': 'This video is not available.',
1941 },
1942 {
1943 # YouTube Red video with episode data
1944 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1945 'info_dict': {
1946 'id': 'iqKdEhx-dD4',
1947 'ext': 'mp4',
1948 'title': 'Isolation - Mind Field (Ep 1)',
1949 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1950 'duration': 2085,
1951 'upload_date': '20170118',
1952 'series': 'Mind Field',
1953 'season_number': 1,
1954 'episode_number': 1,
1955 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1956 'tags': 'count:12',
1957 'view_count': int,
1958 'availability': 'public',
1959 'age_limit': 0,
1960 'channel': 'Vsauce',
1961 'episode': 'Episode 1',
1962 'categories': ['Entertainment'],
1963 'season': 'Season 1',
1964 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1965 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1966 'like_count': int,
1967 'playable_in_embed': True,
1968 'live_status': 'not_live',
1969 'channel_follower_count': int,
1970 'uploader': 'Vsauce',
1971 'uploader_url': 'https://www.youtube.com/@Vsauce',
1972 'uploader_id': '@Vsauce',
1973 'comment_count': int,
1974 'channel_is_verified': True,
1975 },
1976 'params': {
1977 'skip_download': True,
1978 },
1979 'expected_warnings': [
1980 'Skipping DASH manifest',
1981 ],
1982 },
1983 {
1984 # The following content has been identified by the YouTube community
1985 # as inappropriate or offensive to some audiences.
1986 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1987 'info_dict': {
1988 'id': '6SJNVb0GnPI',
1989 'ext': 'mp4',
1990 'title': 'Race Differences in Intelligence',
1991 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1992 'duration': 965,
1993 'upload_date': '20140124',
1994 },
1995 'params': {
1996 'skip_download': True,
1997 },
1998 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1999 },
2000 {
2001 # itag 212
2002 'url': '1t24XAntNCY',
2003 'only_matching': True,
2004 },
2005 {
2006 # geo restricted to JP
2007 'url': 'sJL6WA-aGkQ',
2008 'only_matching': True,
2009 },
2010 {
2011 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2012 'only_matching': True,
2013 },
2014 {
2015 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2016 'only_matching': True,
2017 },
2018 {
2019 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2020 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2021 'only_matching': True,
2022 },
2023 {
2024 # DRM protected
2025 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2026 'only_matching': True,
2027 },
2028 {
2029 # Video with unsupported adaptive stream type formats
2030 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2031 'info_dict': {
2032 'id': 'Z4Vy8R84T1U',
2033 'ext': 'mp4',
2034 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2035 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2036 'duration': 433,
2037 'upload_date': '20130923',
2038 'formats': 'maxcount:10',
2039 },
2040 'params': {
2041 'skip_download': True,
2042 'youtube_include_dash_manifest': False,
2043 },
2044 'skip': 'not actual anymore',
2045 },
2046 {
2047 # Youtube Music Auto-generated description
2048 # TODO: fix metadata extraction
2049 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2050 'info_dict': {
2051 'id': 'MgNrAu2pzNs',
2052 'ext': 'mp4',
2053 'title': 'Voyeur Girl',
2054 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2055 'upload_date': '20190312',
2056 'artist': 'Stephen',
2057 'track': 'Voyeur Girl',
2058 'album': 'it\'s too much love to know my dear',
2059 'release_date': '20190313',
2060 'release_year': 2019,
2061 'alt_title': 'Voyeur Girl',
2062 'view_count': int,
2063 'playable_in_embed': True,
2064 'like_count': int,
2065 'categories': ['Music'],
2066 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2067 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2068 'uploader': 'Stephen',
2069 'availability': 'public',
2070 'creator': 'Stephen',
2071 'duration': 169,
2072 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2073 'age_limit': 0,
2074 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2075 'tags': 'count:11',
2076 'live_status': 'not_live',
2077 'channel_follower_count': int
2078 },
2079 'params': {
2080 'skip_download': True,
2081 },
2082 },
2083 {
2084 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2085 'only_matching': True,
2086 },
2087 {
2088 # invalid -> valid video id redirection
2089 'url': 'DJztXj2GPfl',
2090 'info_dict': {
2091 'id': 'DJztXj2GPfk',
2092 'ext': 'mp4',
2093 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2094 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2095 'upload_date': '20090125',
2096 'artist': 'Panjabi MC',
2097 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2098 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2099 },
2100 'params': {
2101 'skip_download': True,
2102 },
2103 'skip': 'Video unavailable',
2104 },
2105 {
2106 # empty description results in an empty string
2107 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2108 'info_dict': {
2109 'id': 'x41yOUIvK2k',
2110 'ext': 'mp4',
2111 'title': 'IMG 3456',
2112 'description': '',
2113 'upload_date': '20170613',
2114 'view_count': int,
2115 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2116 'like_count': int,
2117 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2118 'tags': [],
2119 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2120 'availability': 'public',
2121 'age_limit': 0,
2122 'categories': ['Pets & Animals'],
2123 'duration': 7,
2124 'playable_in_embed': True,
2125 'live_status': 'not_live',
2126 'channel': 'l\'Or Vert asbl',
2127 'channel_follower_count': int,
2128 'uploader': 'l\'Or Vert asbl',
2129 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2130 'uploader_id': '@ElevageOrVert',
2131 },
2132 'params': {
2133 'skip_download': True,
2134 },
2135 },
2136 {
2137 # with '};' inside yt initial data (see [1])
2138 # see [2] for an example with '};' inside ytInitialPlayerResponse
2139 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2140 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2141 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2142 'info_dict': {
2143 'id': 'CHqg6qOn4no',
2144 'ext': 'mp4',
2145 'title': 'Part 77 Sort a list of simple types in c#',
2146 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2147 'upload_date': '20130831',
2148 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2149 'like_count': int,
2150 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2151 'live_status': 'not_live',
2152 'categories': ['Education'],
2153 'availability': 'public',
2154 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2155 'tags': 'count:12',
2156 'playable_in_embed': True,
2157 'age_limit': 0,
2158 'view_count': int,
2159 'duration': 522,
2160 'channel': 'kudvenkat',
2161 'comment_count': int,
2162 'channel_follower_count': int,
2163 'chapters': list,
2164 'uploader': 'kudvenkat',
2165 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2166 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2167 'channel_is_verified': True,
2168 'heatmap': 'count:100',
2169 },
2170 'params': {
2171 'skip_download': True,
2172 },
2173 },
2174 {
2175 # another example of '};' in ytInitialData
2176 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2177 'only_matching': True,
2178 },
2179 {
2180 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2181 'only_matching': True,
2182 },
2183 {
2184 # https://github.com/ytdl-org/youtube-dl/pull/28094
2185 'url': 'OtqTfy26tG0',
2186 'info_dict': {
2187 'id': 'OtqTfy26tG0',
2188 'ext': 'mp4',
2189 'title': 'Burn Out',
2190 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2191 'upload_date': '20141120',
2192 'artist': 'The Cinematic Orchestra',
2193 'track': 'Burn Out',
2194 'album': 'Every Day',
2195 'like_count': int,
2196 'live_status': 'not_live',
2197 'alt_title': 'Burn Out',
2198 'duration': 614,
2199 'age_limit': 0,
2200 'view_count': int,
2201 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2202 'creator': 'The Cinematic Orchestra',
2203 'channel': 'The Cinematic Orchestra',
2204 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2205 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2206 'availability': 'public',
2207 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2208 'categories': ['Music'],
2209 'playable_in_embed': True,
2210 'channel_follower_count': int,
2211 'uploader': 'The Cinematic Orchestra',
2212 'comment_count': int,
2213 },
2214 'params': {
2215 'skip_download': True,
2216 },
2217 },
2218 {
2219 # controversial video, only works with bpctr when authenticated with cookies
2220 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2221 'only_matching': True,
2222 },
2223 {
2224 # controversial video, requires bpctr/contentCheckOk
2225 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2226 'info_dict': {
2227 'id': 'SZJvDhaSDnc',
2228 'ext': 'mp4',
2229 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2230 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2231 'upload_date': '20140716',
2232 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2233 'duration': 170,
2234 'categories': ['News & Politics'],
2235 'view_count': int,
2236 'channel': 'CBS Mornings',
2237 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2238 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2239 'age_limit': 18,
2240 'availability': 'needs_auth',
2241 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2242 'like_count': int,
2243 'live_status': 'not_live',
2244 'playable_in_embed': True,
2245 'channel_follower_count': int,
2246 'uploader': 'CBS Mornings',
2247 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2248 'uploader_id': '@CBSMornings',
2249 'comment_count': int,
2250 'channel_is_verified': True,
2251 }
2252 },
2253 {
2254 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2255 'url': 'cBvYw8_A0vQ',
2256 'info_dict': {
2257 'id': 'cBvYw8_A0vQ',
2258 'ext': 'mp4',
2259 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2260 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2261 'upload_date': '20201120',
2262 'duration': 1456,
2263 'categories': ['Travel & Events'],
2264 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2265 'view_count': int,
2266 'channel': 'Walk around Japan',
2267 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2268 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2269 'age_limit': 0,
2270 'availability': 'public',
2271 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2272 'live_status': 'not_live',
2273 'playable_in_embed': True,
2274 'channel_follower_count': int,
2275 'uploader': 'Walk around Japan',
2276 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2277 'uploader_id': '@walkaroundjapan7124',
2278 },
2279 'params': {
2280 'skip_download': True,
2281 },
2282 }, {
2283 # Has multiple audio streams
2284 'url': 'WaOKSUlf4TM',
2285 'only_matching': True
2286 }, {
2287 # Requires Premium: has format 141 when requested using YTM url
2288 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2289 'only_matching': True
2290 }, {
2291 # multiple subtitles with same lang_code
2292 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2293 'only_matching': True,
2294 }, {
2295 # Force use android client fallback
2296 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2297 'info_dict': {
2298 'id': 'YOelRv7fMxY',
2299 'title': 'DIGGING A SECRET TUNNEL Part 1',
2300 'ext': '3gp',
2301 'upload_date': '20210624',
2302 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2303 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2304 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2305 'duration': 596,
2306 'categories': ['Entertainment'],
2307 'view_count': int,
2308 'channel': 'colinfurze',
2309 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2310 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2311 'age_limit': 0,
2312 'availability': 'public',
2313 'like_count': int,
2314 'live_status': 'not_live',
2315 'playable_in_embed': True,
2316 'channel_follower_count': int,
2317 'chapters': list,
2318 'uploader': 'colinfurze',
2319 'uploader_url': 'https://www.youtube.com/@colinfurze',
2320 'uploader_id': '@colinfurze',
2321 'comment_count': int,
2322 'channel_is_verified': True,
2323 'heatmap': 'count:100',
2324 },
2325 'params': {
2326 'format': '17', # 3gp format available on android
2327 'extractor_args': {'youtube': {'player_client': ['android']}},
2328 },
2329 },
2330 {
2331 # Skip download of additional client configs (remix client config in this case)
2332 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2333 'only_matching': True,
2334 'params': {
2335 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2336 },
2337 }, {
2338 # shorts
2339 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2340 'only_matching': True,
2341 }, {
2342 'note': 'Storyboards',
2343 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2344 'info_dict': {
2345 'id': '5KLPxDtMqe8',
2346 'ext': 'mhtml',
2347 'format_id': 'sb0',
2348 'title': 'Your Brain is Plastic',
2349 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2350 'upload_date': '20140324',
2351 'like_count': int,
2352 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2353 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2354 'view_count': int,
2355 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2356 'playable_in_embed': True,
2357 'tags': 'count:12',
2358 'availability': 'public',
2359 'channel': 'SciShow',
2360 'live_status': 'not_live',
2361 'duration': 248,
2362 'categories': ['Education'],
2363 'age_limit': 0,
2364 'channel_follower_count': int,
2365 'chapters': list,
2366 'uploader': 'SciShow',
2367 'uploader_url': 'https://www.youtube.com/@SciShow',
2368 'uploader_id': '@SciShow',
2369 'comment_count': int,
2370 'channel_is_verified': True,
2371 'heatmap': 'count:100',
2372 }, 'params': {'format': 'mhtml', 'skip_download': True}
2373 }, {
2374 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2375 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2376 'info_dict': {
2377 'id': '2NUZ8W2llS4',
2378 'ext': 'mp4',
2379 'title': 'The NP that test your phone performance 🙂',
2380 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2381 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2382 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2383 'duration': 21,
2384 'view_count': int,
2385 'age_limit': 0,
2386 'categories': ['Gaming'],
2387 'tags': 'count:23',
2388 'playable_in_embed': True,
2389 'live_status': 'not_live',
2390 'upload_date': '20220103',
2391 'like_count': int,
2392 'availability': 'public',
2393 'channel': 'Leon Nguyen',
2394 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2395 'comment_count': int,
2396 'channel_follower_count': int,
2397 'uploader': 'Leon Nguyen',
2398 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2399 'uploader_id': '@LeonNguyen',
2400 'heatmap': 'count:100',
2401 }
2402 }, {
2403 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2404 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2405 'info_dict': {
2406 'id': '2NUZ8W2llS4',
2407 'ext': 'mp4',
2408 'title': 'The NP that test your phone performance 🙂',
2409 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2410 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2411 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2412 'duration': 21,
2413 'view_count': int,
2414 'age_limit': 0,
2415 'categories': ['Gaming'],
2416 'tags': 'count:23',
2417 'playable_in_embed': True,
2418 'live_status': 'not_live',
2419 'upload_date': '20220102',
2420 'like_count': int,
2421 'availability': 'public',
2422 'channel': 'Leon Nguyen',
2423 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2424 'comment_count': int,
2425 'channel_follower_count': int,
2426 'uploader': 'Leon Nguyen',
2427 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2428 'uploader_id': '@LeonNguyen',
2429 'heatmap': 'count:100',
2430 },
2431 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2432 }, {
2433 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2434 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2435 'info_dict': {
2436 'id': 'mzZzzBU6lrM',
2437 'ext': 'mp4',
2438 'title': 'I Met GeorgeNotFound In Real Life...',
2439 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2440 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2441 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2442 'duration': 955,
2443 'view_count': int,
2444 'age_limit': 0,
2445 'categories': ['Entertainment'],
2446 'tags': 'count:26',
2447 'playable_in_embed': True,
2448 'live_status': 'not_live',
2449 'release_timestamp': 1641172509,
2450 'release_date': '20220103',
2451 'upload_date': '20220103',
2452 'like_count': int,
2453 'availability': 'public',
2454 'channel': 'Quackity',
2455 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2456 'channel_follower_count': int,
2457 'uploader': 'Quackity',
2458 'uploader_id': '@Quackity',
2459 'uploader_url': 'https://www.youtube.com/@Quackity',
2460 'comment_count': int,
2461 'channel_is_verified': True,
2462 'heatmap': 'count:100',
2463 }
2464 },
2465 { # continuous livestream. Microformat upload date should be preferred.
2466 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2467 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2468 'info_dict': {
2469 'id': 'kgx4WGK0oNU',
2470 '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}',
2471 'ext': 'mp4',
2472 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2473 'availability': 'public',
2474 'age_limit': 0,
2475 'release_timestamp': 1637975704,
2476 'upload_date': '20210619',
2477 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2478 'live_status': 'is_live',
2479 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2480 'channel': 'Abao in Tokyo',
2481 'channel_follower_count': int,
2482 'release_date': '20211127',
2483 'tags': 'count:39',
2484 'categories': ['People & Blogs'],
2485 'like_count': int,
2486 'view_count': int,
2487 'playable_in_embed': True,
2488 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2489 'concurrent_view_count': int,
2490 'uploader': 'Abao in Tokyo',
2491 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2492 'uploader_id': '@abaointokyo',
2493 },
2494 'params': {'skip_download': True}
2495 }, {
2496 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2497 'info_dict': {
2498 'id': 'tjjjtzRLHvA',
2499 'ext': 'mp4',
2500 'title': 'ハッシュタグ無し };if window.ytcsi',
2501 'upload_date': '20220323',
2502 'like_count': int,
2503 'availability': 'unlisted',
2504 'channel': 'Lesmiscore',
2505 'thumbnail': r're:^https?://.*\.jpg',
2506 'age_limit': 0,
2507 'categories': ['Music'],
2508 'view_count': int,
2509 'description': '',
2510 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2511 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2512 'live_status': 'not_live',
2513 'playable_in_embed': True,
2514 'channel_follower_count': int,
2515 'duration': 6,
2516 'tags': [],
2517 'uploader_id': '@lesmiscore',
2518 'uploader': 'Lesmiscore',
2519 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2520 }
2521 }, {
2522 # Prefer primary title+description language metadata by default
2523 # Do not prefer translated description if primary is empty
2524 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2525 'info_dict': {
2526 'id': 'el3E4MbxRqQ',
2527 'ext': 'mp4',
2528 'title': 'dlp test video 2 - primary sv no desc',
2529 'description': '',
2530 'channel': 'cole-dlp-test-acc',
2531 'tags': [],
2532 'view_count': int,
2533 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2534 'like_count': int,
2535 'playable_in_embed': True,
2536 'availability': 'unlisted',
2537 'thumbnail': r're:^https?://.*\.jpg',
2538 'age_limit': 0,
2539 'duration': 5,
2540 'live_status': 'not_live',
2541 'upload_date': '20220908',
2542 'categories': ['People & Blogs'],
2543 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2544 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2545 'uploader_id': '@coletdjnz',
2546 'uploader': 'cole-dlp-test-acc',
2547 },
2548 'params': {'skip_download': True}
2549 }, {
2550 # Extractor argument: prefer translated title+description
2551 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2552 'info_dict': {
2553 'id': 'gHKT4uU8Zng',
2554 'ext': 'mp4',
2555 'channel': 'cole-dlp-test-acc',
2556 'tags': [],
2557 'duration': 5,
2558 'live_status': 'not_live',
2559 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2560 'upload_date': '20220728',
2561 'view_count': int,
2562 'categories': ['People & Blogs'],
2563 'thumbnail': r're:^https?://.*\.jpg',
2564 'title': 'dlp test video title translated (fr)',
2565 'availability': 'public',
2566 'age_limit': 0,
2567 'description': 'dlp test video description translated (fr)',
2568 'playable_in_embed': True,
2569 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2570 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2571 'uploader_id': '@coletdjnz',
2572 'uploader': 'cole-dlp-test-acc',
2573 },
2574 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2575 'expected_warnings': [r'Preferring "fr" translated fields'],
2576 }, {
2577 'note': '6 channel audio',
2578 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2579 'only_matching': True,
2580 }, {
2581 'note': 'Multiple HLS formats with same itag',
2582 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2583 'info_dict': {
2584 'id': 'kX3nB4PpJko',
2585 'ext': 'mp4',
2586 'categories': ['Entertainment'],
2587 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2588 'live_status': 'not_live',
2589 'duration': 937,
2590 'channel_follower_count': int,
2591 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2592 'title': 'Last To Take Hand Off Jet, Keeps It!',
2593 'channel': 'MrBeast',
2594 'playable_in_embed': True,
2595 'view_count': int,
2596 'upload_date': '20221112',
2597 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2598 'age_limit': 0,
2599 'availability': 'public',
2600 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2601 'like_count': int,
2602 'tags': [],
2603 'uploader': 'MrBeast',
2604 'uploader_url': 'https://www.youtube.com/@MrBeast',
2605 'uploader_id': '@MrBeast',
2606 'comment_count': int,
2607 'channel_is_verified': True,
2608 'heatmap': 'count:100',
2609 },
2610 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2611 }, {
2612 'note': 'Audio formats with Dynamic Range Compression',
2613 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2614 'info_dict': {
2615 'id': 'Tq92D6wQ1mg',
2616 'ext': 'webm',
2617 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2618 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2619 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2620 'channel_follower_count': int,
2621 'description': 'md5:17eccca93a786d51bc67646756894066',
2622 'upload_date': '20191228',
2623 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2624 'playable_in_embed': True,
2625 'like_count': int,
2626 'categories': ['Entertainment'],
2627 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2628 'age_limit': 18,
2629 'channel': 'Projekt Melody',
2630 'view_count': int,
2631 'availability': 'needs_auth',
2632 'comment_count': int,
2633 'live_status': 'not_live',
2634 'duration': 106,
2635 'uploader': 'Projekt Melody',
2636 'uploader_id': '@ProjektMelody',
2637 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2638 },
2639 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2640 },
2641 {
2642 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2643 'info_dict': {
2644 'id': 'qVv6vCqciTM',
2645 'ext': 'mp4',
2646 'age_limit': 0,
2647 'comment_count': int,
2648 'chapters': 'count:13',
2649 'upload_date': '20221223',
2650 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2651 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2652 'like_count': int,
2653 'release_date': '20221223',
2654 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2655 '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】',
2656 'view_count': int,
2657 'playable_in_embed': True,
2658 'duration': 4438,
2659 'availability': 'public',
2660 'channel_follower_count': int,
2661 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2662 'categories': ['Entertainment'],
2663 'live_status': 'was_live',
2664 'release_timestamp': 1671793345,
2665 'channel': 'さなちゃんねる',
2666 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2667 'uploader': 'さなちゃんねる',
2668 'uploader_url': 'https://www.youtube.com/@sana_natori',
2669 'uploader_id': '@sana_natori',
2670 'channel_is_verified': True,
2671 'heatmap': 'count:100',
2672 },
2673 },
2674 {
2675 # Fallbacks when webpage and web client is unavailable
2676 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2677 'info_dict': {
2678 'id': 'wSSmNUl9Snw',
2679 'ext': 'mp4',
2680 # 'categories': ['Science & Technology'],
2681 'view_count': int,
2682 'chapters': 'count:2',
2683 'channel': 'Scott Manley',
2684 'like_count': int,
2685 'age_limit': 0,
2686 # 'availability': 'public',
2687 'channel_follower_count': int,
2688 'live_status': 'not_live',
2689 'upload_date': '20170831',
2690 'duration': 682,
2691 'tags': 'count:8',
2692 'uploader_url': 'https://www.youtube.com/@scottmanley',
2693 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2694 'uploader': 'Scott Manley',
2695 'uploader_id': '@scottmanley',
2696 'title': 'The Computer Hack That Saved Apollo 14',
2697 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2698 'thumbnail': r're:^https?://.*\.webp',
2699 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2700 'playable_in_embed': True,
2701 'comment_count': int,
2702 'channel_is_verified': True,
2703 'heatmap': 'count:100',
2704 },
2705 'params': {
2706 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2707 },
2708 },
2709 ]
2710
2711 _WEBPAGE_TESTS = [
2712 # YouTube <object> embed
2713 {
2714 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2715 'md5': '873c81d308b979f0e23ee7e620b312a3',
2716 'info_dict': {
2717 'id': 'msN87y-iEx0',
2718 'ext': 'mp4',
2719 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2720 'upload_date': '20080526',
2721 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2722 'age_limit': 0,
2723 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2724 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2725 'playable_in_embed': True,
2726 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2727 'like_count': int,
2728 'comment_count': int,
2729 'channel': 'Christopher Sykes',
2730 'live_status': 'not_live',
2731 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2732 'availability': 'public',
2733 'duration': 195,
2734 'view_count': int,
2735 'categories': ['Science & Technology'],
2736 'channel_follower_count': int,
2737 'uploader': 'Christopher Sykes',
2738 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2739 'uploader_id': '@ChristopherSykesDocumentaries',
2740 'heatmap': 'count:100',
2741 },
2742 'params': {
2743 'skip_download': True,
2744 }
2745 },
2746 ]
2747
2748 @classmethod
2749 def suitable(cls, url):
2750 from ..utils import parse_qs
2751
2752 qs = parse_qs(url)
2753 if qs.get('list', [None])[0]:
2754 return False
2755 return super().suitable(url)
2756
2757 def __init__(self, *args, **kwargs):
2758 super().__init__(*args, **kwargs)
2759 self._code_cache = {}
2760 self._player_cache = {}
2761
2762 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2763 lock = threading.Lock()
2764 start_time = time.time()
2765 formats = [f for f in formats if f.get('is_from_start')]
2766
2767 def refetch_manifest(format_id, delay):
2768 nonlocal formats, start_time, is_live
2769 if time.time() <= start_time + delay:
2770 return
2771
2772 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2773 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2774 microformats = traverse_obj(
2775 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2776 expected_type=dict)
2777 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2778 is_live = live_status == 'is_live'
2779 start_time = time.time()
2780
2781 def mpd_feed(format_id, delay):
2782 """
2783 @returns (manifest_url, manifest_stream_number, is_live) or None
2784 """
2785 for retry in self.RetryManager(fatal=False):
2786 with lock:
2787 refetch_manifest(format_id, delay)
2788
2789 f = next((f for f in formats if f['format_id'] == format_id), None)
2790 if not f:
2791 if not is_live:
2792 retry.error = f'{video_id}: Video is no longer live'
2793 else:
2794 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2795 continue
2796 return f['manifest_url'], f['manifest_stream_number'], is_live
2797 return None
2798
2799 for f in formats:
2800 f['is_live'] = is_live
2801 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2802 live_start_time, mpd_feed, not is_live and f.copy())
2803 if is_live:
2804 f['fragments'] = gen
2805 f['protocol'] = 'http_dash_segments_generator'
2806 else:
2807 f['fragments'] = LazyList(gen({}))
2808 del f['is_from_start']
2809
2810 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2811 FETCH_SPAN, MAX_DURATION = 5, 432000
2812
2813 mpd_url, stream_number, is_live = None, None, True
2814
2815 begin_index = 0
2816 download_start_time = ctx.get('start') or time.time()
2817
2818 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2819 if lack_early_segments:
2820 self.report_warning(bug_reports_message(
2821 'Starting download from the last 120 hours of the live stream since '
2822 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2823 lack_early_segments = True
2824
2825 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2826 fragments, fragment_base_url = None, None
2827
2828 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2829 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2830 # Obtain from MPD's maximum seq value
2831 old_mpd_url = mpd_url
2832 last_error = ctx.pop('last_error', None)
2833 expire_fast = immediate or last_error and isinstance(last_error, HTTPError) and last_error.status == 403
2834 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2835 or (mpd_url, stream_number, False))
2836 if not refresh_sequence:
2837 if expire_fast and not is_live:
2838 return False, last_seq
2839 elif old_mpd_url == mpd_url:
2840 return True, last_seq
2841 if manifestless_orig_fmt:
2842 fmt_info = manifestless_orig_fmt
2843 else:
2844 try:
2845 fmts, _ = self._extract_mpd_formats_and_subtitles(
2846 mpd_url, None, note=False, errnote=False, fatal=False)
2847 except ExtractorError:
2848 fmts = None
2849 if not fmts:
2850 no_fragment_score += 2
2851 return False, last_seq
2852 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2853 fragments = fmt_info['fragments']
2854 fragment_base_url = fmt_info['fragment_base_url']
2855 assert fragment_base_url
2856
2857 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2858 return True, _last_seq
2859
2860 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2861 while is_live:
2862 fetch_time = time.time()
2863 if no_fragment_score > 30:
2864 return
2865 if last_segment_url:
2866 # Obtain from "X-Head-Seqnum" header value from each segment
2867 try:
2868 urlh = self._request_webpage(
2869 last_segment_url, None, note=False, errnote=False, fatal=False)
2870 except ExtractorError:
2871 urlh = None
2872 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2873 if last_seq is None:
2874 no_fragment_score += 2
2875 last_segment_url = None
2876 continue
2877 else:
2878 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2879 no_fragment_score += 2
2880 if not should_continue:
2881 continue
2882
2883 if known_idx > last_seq:
2884 last_segment_url = None
2885 continue
2886
2887 last_seq += 1
2888
2889 if begin_index < 0 and known_idx < 0:
2890 # skip from the start when it's negative value
2891 known_idx = last_seq + begin_index
2892 if lack_early_segments:
2893 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2894 try:
2895 for idx in range(known_idx, last_seq):
2896 # do not update sequence here or you'll get skipped some part of it
2897 should_continue, _ = _extract_sequence_from_mpd(False, False)
2898 if not should_continue:
2899 known_idx = idx - 1
2900 raise ExtractorError('breaking out of outer loop')
2901 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2902 yield {
2903 'url': last_segment_url,
2904 'fragment_count': last_seq,
2905 }
2906 if known_idx == last_seq:
2907 no_fragment_score += 5
2908 else:
2909 no_fragment_score = 0
2910 known_idx = last_seq
2911 except ExtractorError:
2912 continue
2913
2914 if manifestless_orig_fmt:
2915 # Stop at the first iteration if running for post-live manifestless;
2916 # fragment count no longer increase since it starts
2917 break
2918
2919 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2920
2921 def _extract_player_url(self, *ytcfgs, webpage=None):
2922 player_url = traverse_obj(
2923 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2924 get_all=False, expected_type=str)
2925 if not player_url:
2926 return
2927 return urljoin('https://www.youtube.com', player_url)
2928
2929 def _download_player_url(self, video_id, fatal=False):
2930 res = self._download_webpage(
2931 'https://www.youtube.com/iframe_api',
2932 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2933 if res:
2934 player_version = self._search_regex(
2935 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2936 if player_version:
2937 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2938
2939 def _signature_cache_id(self, example_sig):
2940 """ Return a string representation of a signature """
2941 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2942
2943 @classmethod
2944 def _extract_player_info(cls, player_url):
2945 for player_re in cls._PLAYER_INFO_RE:
2946 id_m = re.search(player_re, player_url)
2947 if id_m:
2948 break
2949 else:
2950 raise ExtractorError('Cannot identify player %r' % player_url)
2951 return id_m.group('id')
2952
2953 def _load_player(self, video_id, player_url, fatal=True):
2954 player_id = self._extract_player_info(player_url)
2955 if player_id not in self._code_cache:
2956 code = self._download_webpage(
2957 player_url, video_id, fatal=fatal,
2958 note='Downloading player ' + player_id,
2959 errnote='Download of %s failed' % player_url)
2960 if code:
2961 self._code_cache[player_id] = code
2962 return self._code_cache.get(player_id)
2963
2964 def _extract_signature_function(self, video_id, player_url, example_sig):
2965 player_id = self._extract_player_info(player_url)
2966
2967 # Read from filesystem cache
2968 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2969 assert os.path.basename(func_id) == func_id
2970
2971 self.write_debug(f'Extracting signature function {func_id}')
2972 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2973
2974 if not cache_spec:
2975 code = self._load_player(video_id, player_url)
2976 if code:
2977 res = self._parse_sig_js(code)
2978 test_string = ''.join(map(chr, range(len(example_sig))))
2979 cache_spec = [ord(c) for c in res(test_string)]
2980 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2981
2982 return lambda s: ''.join(s[i] for i in cache_spec)
2983
2984 def _print_sig_code(self, func, example_sig):
2985 if not self.get_param('youtube_print_sig_code'):
2986 return
2987
2988 def gen_sig_code(idxs):
2989 def _genslice(start, end, step):
2990 starts = '' if start == 0 else str(start)
2991 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2992 steps = '' if step == 1 else (':%d' % step)
2993 return f's[{starts}{ends}{steps}]'
2994
2995 step = None
2996 # Quelch pyflakes warnings - start will be set when step is set
2997 start = '(Never used)'
2998 for i, prev in zip(idxs[1:], idxs[:-1]):
2999 if step is not None:
3000 if i - prev == step:
3001 continue
3002 yield _genslice(start, prev, step)
3003 step = None
3004 continue
3005 if i - prev in [-1, 1]:
3006 step = i - prev
3007 start = prev
3008 continue
3009 else:
3010 yield 's[%d]' % prev
3011 if step is None:
3012 yield 's[%d]' % i
3013 else:
3014 yield _genslice(start, i, step)
3015
3016 test_string = ''.join(map(chr, range(len(example_sig))))
3017 cache_res = func(test_string)
3018 cache_spec = [ord(c) for c in cache_res]
3019 expr_code = ' + '.join(gen_sig_code(cache_spec))
3020 signature_id_tuple = '(%s)' % (
3021 ', '.join(str(len(p)) for p in example_sig.split('.')))
3022 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
3023 ' return %s\n') % (signature_id_tuple, expr_code)
3024 self.to_screen('Extracted signature function:\n' + code)
3025
3026 def _parse_sig_js(self, jscode):
3027 funcname = self._search_regex(
3028 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3029 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3030 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3031 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3032 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}\(a,\d+\))?',
3033 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3034 # Obsolete patterns
3035 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3036 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3037 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3038 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3039 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3040 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3041 jscode, 'Initial JS player signature function name', group='sig')
3042
3043 jsi = JSInterpreter(jscode)
3044 initial_function = jsi.extract_function(funcname)
3045 return lambda s: initial_function([s])
3046
3047 def _cached(self, func, *cache_id):
3048 def inner(*args, **kwargs):
3049 if cache_id not in self._player_cache:
3050 try:
3051 self._player_cache[cache_id] = func(*args, **kwargs)
3052 except ExtractorError as e:
3053 self._player_cache[cache_id] = e
3054 except Exception as e:
3055 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3056
3057 ret = self._player_cache[cache_id]
3058 if isinstance(ret, Exception):
3059 raise ret
3060 return ret
3061 return inner
3062
3063 def _decrypt_signature(self, s, video_id, player_url):
3064 """Turn the encrypted s field into a working signature"""
3065 extract_sig = self._cached(
3066 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3067 func = extract_sig(video_id, player_url, s)
3068 self._print_sig_code(func, s)
3069 return func(s)
3070
3071 def _decrypt_nsig(self, s, video_id, player_url):
3072 """Turn the encrypted n field into a working signature"""
3073 if player_url is None:
3074 raise ExtractorError('Cannot decrypt nsig without player_url')
3075 player_url = urljoin('https://www.youtube.com', player_url)
3076
3077 try:
3078 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3079 except ExtractorError as e:
3080 raise ExtractorError('Unable to extract nsig function code', cause=e)
3081 if self.get_param('youtube_print_sig_code'):
3082 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3083
3084 try:
3085 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3086 ret = extract_nsig(jsi, func_code)(s)
3087 except JSInterpreter.Exception as e:
3088 try:
3089 jsi = PhantomJSwrapper(self, timeout=5000)
3090 except ExtractorError:
3091 raise e
3092 self.report_warning(
3093 f'Native nsig extraction failed: Trying with PhantomJS\n'
3094 f' n = {s} ; player = {player_url}', video_id)
3095 self.write_debug(e, only_once=True)
3096
3097 args, func_body = func_code
3098 ret = jsi.execute(
3099 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3100 video_id=video_id, note='Executing signature code').strip()
3101
3102 self.write_debug(f'Decrypted nsig {s} => {ret}')
3103 return ret
3104
3105 def _extract_n_function_name(self, jscode):
3106 funcname, idx = self._search_regex(
3107 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3108 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3109 if not idx:
3110 return funcname
3111
3112 return json.loads(js_to_json(self._search_regex(
3113 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
3114 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3115
3116 def _extract_n_function_code(self, video_id, player_url):
3117 player_id = self._extract_player_info(player_url)
3118 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3119 jscode = func_code or self._load_player(video_id, player_url)
3120 jsi = JSInterpreter(jscode)
3121
3122 if func_code:
3123 return jsi, player_id, func_code
3124
3125 func_name = self._extract_n_function_name(jscode)
3126
3127 # For redundancy
3128 func_code = self._search_regex(
3129 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3130 # NB: The end of the regex is intentionally kept strict
3131 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3132 jscode, 'nsig function', group=('var', 'code'), default=None)
3133 if func_code:
3134 func_code = ([func_code[0]], func_code[1])
3135 else:
3136 self.write_debug('Extracting nsig function with jsinterp')
3137 func_code = jsi.extract_function_code(func_name)
3138
3139 self.cache.store('youtube-nsig', player_id, func_code)
3140 return jsi, player_id, func_code
3141
3142 def _extract_n_function_from_code(self, jsi, func_code):
3143 func = jsi.extract_function_from_code(*func_code)
3144
3145 def extract_nsig(s):
3146 try:
3147 ret = func([s])
3148 except JSInterpreter.Exception:
3149 raise
3150 except Exception as e:
3151 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3152
3153 if ret.startswith('enhanced_except_'):
3154 raise JSInterpreter.Exception('Signature function returned an exception')
3155 return ret
3156
3157 return extract_nsig
3158
3159 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3160 """
3161 Extract signatureTimestamp (sts)
3162 Required to tell API what sig/player version is in use.
3163 """
3164 sts = None
3165 if isinstance(ytcfg, dict):
3166 sts = int_or_none(ytcfg.get('STS'))
3167
3168 if not sts:
3169 # Attempt to extract from player
3170 if player_url is None:
3171 error_msg = 'Cannot extract signature timestamp without player_url.'
3172 if fatal:
3173 raise ExtractorError(error_msg)
3174 self.report_warning(error_msg)
3175 return
3176 code = self._load_player(video_id, player_url, fatal=fatal)
3177 if code:
3178 sts = int_or_none(self._search_regex(
3179 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3180 'JS player signature timestamp', group='sts', fatal=fatal))
3181 return sts
3182
3183 def _mark_watched(self, video_id, player_responses):
3184 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3185 label = 'fully ' if is_full else ''
3186 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3187 expected_type=url_or_none)
3188 if not url:
3189 self.report_warning(f'Unable to mark {label}watched')
3190 return
3191 parsed_url = urllib.parse.urlparse(url)
3192 qs = urllib.parse.parse_qs(parsed_url.query)
3193
3194 # cpn generation algorithm is reverse engineered from base.js.
3195 # In fact it works even with dummy cpn.
3196 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3197 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3198
3199 # # more consistent results setting it to right before the end
3200 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3201
3202 qs.update({
3203 'ver': ['2'],
3204 'cpn': [cpn],
3205 'cmt': video_length,
3206 'el': 'detailpage', # otherwise defaults to "shorts"
3207 })
3208
3209 if is_full:
3210 # these seem to mark watchtime "history" in the real world
3211 # they're required, so send in a single value
3212 qs.update({
3213 'st': 0,
3214 'et': video_length,
3215 })
3216
3217 url = urllib.parse.urlunparse(
3218 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3219
3220 self._download_webpage(
3221 url, video_id, f'Marking {label}watched',
3222 'Unable to mark watched', fatal=False)
3223
3224 @classmethod
3225 def _extract_from_webpage(cls, url, webpage):
3226 # Invidious Instances
3227 # https://github.com/yt-dlp/yt-dlp/issues/195
3228 # https://github.com/iv-org/invidious/pull/1730
3229 mobj = re.search(
3230 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3231 webpage)
3232 if mobj:
3233 yield cls.url_result(mobj.group('url'), cls)
3234 raise cls.StopExtraction()
3235
3236 yield from super()._extract_from_webpage(url, webpage)
3237
3238 # lazyYT YouTube embed
3239 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3240 yield cls.url_result(unescapeHTML(id_), cls, id_)
3241
3242 # Wordpress "YouTube Video Importer" plugin
3243 for m in re.findall(r'''(?x)<div[^>]+
3244 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3245 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3246 yield cls.url_result(m[-1], cls, m[-1])
3247
3248 @classmethod
3249 def extract_id(cls, url):
3250 video_id = cls.get_temp_id(url)
3251 if not video_id:
3252 raise ExtractorError(f'Invalid URL: {url}')
3253 return video_id
3254
3255 def _extract_chapters_from_json(self, data, duration):
3256 chapter_list = traverse_obj(
3257 data, (
3258 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3259 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3260 ), expected_type=list)
3261
3262 return self._extract_chapters_helper(
3263 chapter_list,
3264 start_function=lambda chapter: float_or_none(
3265 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3266 title_function=lambda chapter: traverse_obj(
3267 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3268 duration=duration)
3269
3270 def _extract_chapters_from_engagement_panel(self, data, duration):
3271 content_list = traverse_obj(
3272 data,
3273 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3274 expected_type=list)
3275 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3276 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3277
3278 return next(filter(None, (
3279 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3280 chapter_time, chapter_title, duration)
3281 for contents in content_list)), [])
3282
3283 def _extract_heatmap_from_player_overlay(self, data):
3284 content_list = traverse_obj(data, (
3285 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer', 'decoratedPlayerBarRenderer', 'playerBar',
3286 'multiMarkersPlayerBarRenderer', 'markersMap', ..., 'value', 'heatmap', 'heatmapRenderer', 'heatMarkers', {list}))
3287 return next(filter(None, (
3288 traverse_obj(contents, (..., 'heatMarkerRenderer', {
3289 'start_time': ('timeRangeStartMillis', {functools.partial(float_or_none, scale=1000)}),
3290 'end_time': {lambda x: (x['timeRangeStartMillis'] + x['markerDurationMillis']) / 1000},
3291 'value': ('heatMarkerIntensityScoreNormalized', {float_or_none}),
3292 })) for contents in content_list)), None)
3293
3294 def _extract_comment(self, comment_renderer, parent=None):
3295 comment_id = comment_renderer.get('commentId')
3296 if not comment_id:
3297 return
3298
3299 info = {
3300 'id': comment_id,
3301 'text': self._get_text(comment_renderer, 'contentText'),
3302 'like_count': self._get_count(comment_renderer, 'voteCount'),
3303 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3304 'author': self._get_text(comment_renderer, 'authorText'),
3305 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3306 'parent': parent or 'root',
3307 }
3308
3309 # Timestamp is an estimate calculated from the current time and time_text
3310 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3311 timestamp = self._parse_time_text(time_text)
3312
3313 info.update({
3314 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3315 '_time_text': time_text,
3316 'timestamp': timestamp,
3317 })
3318
3319 info['author_url'] = urljoin(
3320 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3321 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3322 expected_type=str, get_all=False))
3323
3324 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3325 if author_is_uploader is not None:
3326 info['author_is_uploader'] = author_is_uploader
3327
3328 comment_abr = traverse_obj(
3329 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3330 if comment_abr is not None:
3331 info['is_favorited'] = 'creatorHeart' in comment_abr
3332
3333 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3334 if self._has_badge(badges, BadgeType.VERIFIED):
3335 info['author_is_verified'] = True
3336
3337 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3338 if is_pinned:
3339 info['is_pinned'] = True
3340
3341 return info
3342
3343 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3344
3345 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3346
3347 def extract_header(contents):
3348 _continuation = None
3349 for content in contents:
3350 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3351 expected_comment_count = self._get_count(
3352 comments_header_renderer, 'countText', 'commentsCount')
3353
3354 if expected_comment_count is not None:
3355 tracker['est_total'] = expected_comment_count
3356 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3357 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3358
3359 sort_menu_item = try_get(
3360 comments_header_renderer,
3361 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3362 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3363
3364 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3365 if not _continuation:
3366 continue
3367
3368 sort_text = str_or_none(sort_menu_item.get('title'))
3369 if not sort_text:
3370 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3371 self.to_screen('Sorting comments by %s' % sort_text.lower())
3372 break
3373 return _continuation
3374
3375 def extract_thread(contents):
3376 if not parent:
3377 tracker['current_page_thread'] = 0
3378 for content in contents:
3379 if not parent and tracker['total_parent_comments'] >= max_parents:
3380 yield
3381 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3382 comment_renderer = get_first(
3383 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3384 expected_type=dict, default={})
3385
3386 comment = self._extract_comment(comment_renderer, parent)
3387 if not comment:
3388 continue
3389 comment_id = comment['id']
3390 if comment.get('is_pinned'):
3391 tracker['pinned_comment_ids'].add(comment_id)
3392 # Sometimes YouTube may break and give us infinite looping comments.
3393 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3394 if comment_id in tracker['seen_comment_ids']:
3395 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3396 # Pinned comments may appear a second time in newest first sort
3397 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3398 continue
3399 self.report_warning(
3400 'Detected YouTube comments looping. Stopping comment extraction '
3401 f'{"for this thread" if parent else ""} as we probably cannot get any more.')
3402 yield
3403 else:
3404 tracker['seen_comment_ids'].add(comment['id'])
3405
3406 tracker['running_total'] += 1
3407 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3408 yield comment
3409
3410 # Attempt to get the replies
3411 comment_replies_renderer = try_get(
3412 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3413
3414 if comment_replies_renderer:
3415 tracker['current_page_thread'] += 1
3416 comment_entries_iter = self._comment_entries(
3417 comment_replies_renderer, ytcfg, video_id,
3418 parent=comment.get('id'), tracker=tracker)
3419 yield from itertools.islice(comment_entries_iter, min(
3420 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3421
3422 # Keeps track of counts across recursive calls
3423 if not tracker:
3424 tracker = dict(
3425 running_total=0,
3426 est_total=None,
3427 current_page_thread=0,
3428 total_parent_comments=0,
3429 total_reply_comments=0,
3430 seen_comment_ids=set(),
3431 pinned_comment_ids=set()
3432 )
3433
3434 # TODO: Deprecated
3435 # YouTube comments have a max depth of 2
3436 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3437 if max_depth:
3438 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3439 'Set max replies in the max-comments extractor argument instead')
3440 if max_depth == 1 and parent:
3441 return
3442
3443 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3444 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3445
3446 continuation = self._extract_continuation(root_continuation_data)
3447
3448 response = None
3449 is_forced_continuation = False
3450 is_first_continuation = parent is None
3451 if is_first_continuation and not continuation:
3452 # Sometimes you can get comments by generating the continuation yourself,
3453 # even if YouTube initially reports them being disabled - e.g. stories comments.
3454 # Note: if the comment section is actually disabled, YouTube may return a response with
3455 # required check_get_keys missing. So we will disable that check initially in this case.
3456 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3457 is_forced_continuation = True
3458
3459 continuation_items_path = (
3460 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3461 for page_num in itertools.count(0):
3462 if not continuation:
3463 break
3464 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3465 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3466 if page_num == 0:
3467 if is_first_continuation:
3468 note_prefix = 'Downloading comment section API JSON'
3469 else:
3470 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3471 tracker['current_page_thread'], comment_prog_str)
3472 else:
3473 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3474 ' ' if parent else '', ' replies' if parent else '',
3475 page_num, comment_prog_str)
3476
3477 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3478 # Ignore check if YouTube says the comment count is 0.
3479 check_get_keys = None
3480 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3481 check_get_keys = [[*continuation_items_path, ..., (
3482 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
3483 try:
3484 response = self._extract_response(
3485 item_id=None, query=continuation,
3486 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3487 check_get_keys=check_get_keys)
3488 except ExtractorError as e:
3489 # Ignore incomplete data error for replies if retries didn't work.
3490 # This is to allow any other parent comments and comment threads to be downloaded.
3491 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3492 if 'incomplete data' in str(e).lower() and parent:
3493 if self.get_param('ignoreerrors') in (True, 'only_download'):
3494 self.report_warning(
3495 'Received incomplete data for a comment reply thread and retrying did not help. '
3496 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
3497 return
3498 else:
3499 raise ExtractorError(
3500 'Incomplete data received for comment reply thread. '
3501 'Pass --ignore-errors to ignore and allow rest of comments to download.',
3502 expected=True)
3503 raise
3504 is_forced_continuation = False
3505 continuation = None
3506 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3507 if is_first_continuation:
3508 continuation = extract_header(continuation_items)
3509 is_first_continuation = False
3510 if continuation:
3511 break
3512 continue
3513
3514 for entry in extract_thread(continuation_items):
3515 if not entry:
3516 return
3517 yield entry
3518 continuation = self._extract_continuation({'contents': continuation_items})
3519 if continuation:
3520 break
3521
3522 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3523 if message and not parent and tracker['running_total'] == 0:
3524 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3525 raise self.CommentsDisabled
3526
3527 @staticmethod
3528 def _generate_comment_continuation(video_id):
3529 """
3530 Generates initial comment section continuation token from given video id
3531 """
3532 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3533 return base64.b64encode(token.encode()).decode()
3534
3535 def _get_comments(self, ytcfg, video_id, contents, webpage):
3536 """Entry for comment extraction"""
3537 def _real_comment_extract(contents):
3538 renderer = next((
3539 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3540 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3541 yield from self._comment_entries(renderer, ytcfg, video_id)
3542
3543 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3544 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3545
3546 @staticmethod
3547 def _get_checkok_params():
3548 return {'contentCheckOk': True, 'racyCheckOk': True}
3549
3550 @classmethod
3551 def _generate_player_context(cls, sts=None):
3552 context = {
3553 'html5Preference': 'HTML5_PREF_WANTS',
3554 }
3555 if sts is not None:
3556 context['signatureTimestamp'] = sts
3557 return {
3558 'playbackContext': {
3559 'contentPlaybackContext': context
3560 },
3561 **cls._get_checkok_params()
3562 }
3563
3564 @staticmethod
3565 def _is_agegated(player_response):
3566 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3567 return True
3568
3569 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3570 AGE_GATE_REASONS = (
3571 'confirm your age', 'age-restricted', 'inappropriate', # reason
3572 'age_verification_required', 'age_check_required', # status
3573 )
3574 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3575
3576 @staticmethod
3577 def _is_unplayable(player_response):
3578 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3579
3580 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3581
3582 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3583 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3584 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3585 headers = self.generate_api_headers(
3586 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3587
3588 yt_query = {
3589 'videoId': video_id,
3590 }
3591 if _split_innertube_client(client)[0] == 'android':
3592 yt_query['params'] = 'CgIQBg=='
3593
3594 pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0]
3595 if pp_arg:
3596 yt_query['params'] = pp_arg
3597
3598 yt_query.update(self._generate_player_context(sts))
3599 return self._extract_response(
3600 item_id=video_id, ep='player', query=yt_query,
3601 ytcfg=player_ytcfg, headers=headers, fatal=True,
3602 default_client=client,
3603 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3604 ) or None
3605
3606 def _get_requested_clients(self, url, smuggled_data):
3607 requested_clients = []
3608 default = ['ios', 'android', 'web']
3609 allowed_clients = sorted(
3610 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3611 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3612 for client in self._configuration_arg('player_client'):
3613 if client in allowed_clients:
3614 requested_clients.append(client)
3615 elif client == 'default':
3616 requested_clients.extend(default)
3617 elif client == 'all':
3618 requested_clients.extend(allowed_clients)
3619 else:
3620 self.report_warning(f'Skipping unsupported client {client}')
3621 if not requested_clients:
3622 requested_clients = default
3623
3624 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3625 requested_clients.extend(
3626 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3627
3628 return orderedSet(requested_clients)
3629
3630 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3631 initial_pr = None
3632 if webpage:
3633 initial_pr = self._search_json(
3634 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3635
3636 all_clients = set(clients)
3637 clients = clients[::-1]
3638 prs = []
3639
3640 def append_client(*client_names):
3641 """ Append the first client name that exists but not already used """
3642 for client_name in client_names:
3643 actual_client = _split_innertube_client(client_name)[0]
3644 if actual_client in INNERTUBE_CLIENTS:
3645 if actual_client not in all_clients:
3646 clients.append(client_name)
3647 all_clients.add(actual_client)
3648 return
3649
3650 # Android player_response does not have microFormats which are needed for
3651 # extraction of some data. So we return the initial_pr with formats
3652 # stripped out even if not requested by the user
3653 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3654 if initial_pr:
3655 pr = dict(initial_pr)
3656 pr['streamingData'] = None
3657 prs.append(pr)
3658
3659 last_error = None
3660 tried_iframe_fallback = False
3661 player_url = None
3662 while clients:
3663 client, base_client, variant = _split_innertube_client(clients.pop())
3664 player_ytcfg = master_ytcfg if client == 'web' else {}
3665 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3666 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3667
3668 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3669 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3670 if 'js' in self._configuration_arg('player_skip'):
3671 require_js_player = False
3672 player_url = None
3673
3674 if not player_url and not tried_iframe_fallback and require_js_player:
3675 player_url = self._download_player_url(video_id)
3676 tried_iframe_fallback = True
3677
3678 try:
3679 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3680 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3681 except ExtractorError as e:
3682 if last_error:
3683 self.report_warning(last_error)
3684 last_error = e
3685 continue
3686
3687 if pr:
3688 # YouTube may return a different video player response than expected.
3689 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3690 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3691 if pr_video_id and pr_video_id != video_id:
3692 self.report_warning(
3693 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3694 else:
3695 # Save client name for introspection later
3696 name = short_client_name(client)
3697 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3698 sd[STREAMING_DATA_CLIENT_NAME] = name
3699 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3700 f[STREAMING_DATA_CLIENT_NAME] = name
3701 prs.append(pr)
3702
3703 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3704 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3705 append_client(f'{base_client}_creator')
3706 elif self._is_agegated(pr):
3707 if variant == 'tv_embedded':
3708 append_client(f'{base_client}_embedded')
3709 elif not variant:
3710 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3711
3712 if last_error:
3713 if not len(prs):
3714 raise last_error
3715 self.report_warning(last_error)
3716 return prs, player_url
3717
3718 def _needs_live_processing(self, live_status, duration):
3719 if (live_status == 'is_live' and self.get_param('live_from_start')
3720 or live_status == 'post_live' and (duration or 0) > 2 * 3600):
3721 return live_status
3722
3723 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3724 CHUNK_SIZE = 10 << 20
3725 itags, stream_ids = collections.defaultdict(set), []
3726 itag_qualities, res_qualities = {}, {0: None}
3727 q = qualities([
3728 # Normally tiny is the smallest video-only formats. But
3729 # audio-only formats with unknown quality may get tagged as tiny
3730 'tiny',
3731 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3732 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3733 ])
3734 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3735 format_types = self._configuration_arg('formats')
3736 all_formats = 'duplicate' in format_types
3737 if self._configuration_arg('include_duplicate_formats'):
3738 all_formats = True
3739 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
3740 'Use formats=duplicate extractor argument instead')
3741
3742 def build_fragments(f):
3743 return LazyList({
3744 'url': update_url_query(f['url'], {
3745 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3746 })
3747 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3748
3749 for fmt in streaming_formats:
3750 if fmt.get('targetDurationSec'):
3751 continue
3752
3753 itag = str_or_none(fmt.get('itag'))
3754 audio_track = fmt.get('audioTrack') or {}
3755 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3756 if not all_formats:
3757 if stream_id in stream_ids:
3758 continue
3759
3760 quality = fmt.get('quality')
3761 height = int_or_none(fmt.get('height'))
3762 if quality == 'tiny' or not quality:
3763 quality = fmt.get('audioQuality', '').lower() or quality
3764 # The 3gp format (17) in android client has a quality of "small",
3765 # but is actually worse than other formats
3766 if itag == '17':
3767 quality = 'tiny'
3768 if quality:
3769 if itag:
3770 itag_qualities[itag] = quality
3771 if height:
3772 res_qualities[height] = quality
3773 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3774 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3775 # number of fragment that would subsequently requested with (`&sq=N`)
3776 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3777 continue
3778
3779 fmt_url = fmt.get('url')
3780 if not fmt_url:
3781 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3782 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3783 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3784 if not all((sc, fmt_url, player_url, encrypted_sig)):
3785 continue
3786 try:
3787 fmt_url += '&%s=%s' % (
3788 traverse_obj(sc, ('sp', -1)) or 'signature',
3789 self._decrypt_signature(encrypted_sig, video_id, player_url)
3790 )
3791 except ExtractorError as e:
3792 self.report_warning('Signature extraction failed: Some formats may be missing',
3793 video_id=video_id, only_once=True)
3794 self.write_debug(e, only_once=True)
3795 continue
3796
3797 query = parse_qs(fmt_url)
3798 throttled = False
3799 if query.get('n'):
3800 try:
3801 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3802 fmt_url = update_url_query(fmt_url, {
3803 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3804 })
3805 except ExtractorError as e:
3806 phantomjs_hint = ''
3807 if isinstance(e, JSInterpreter.Exception):
3808 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3809 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3810 if player_url:
3811 self.report_warning(
3812 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3813 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3814 self.write_debug(e, only_once=True)
3815 else:
3816 self.report_warning(
3817 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3818 video_id=video_id, only_once=True)
3819 throttled = True
3820
3821 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3822 language_preference = (
3823 10 if audio_track.get('audioIsDefault') and 10
3824 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3825 else -1)
3826 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3827 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3828 # Make sure to avoid false positives with small duration differences.
3829 # E.g. __2ABJjxzNo, ySuUZEjARPY
3830 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3831 if is_damaged:
3832 self.report_warning(
3833 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3834
3835 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3836 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
3837 fps = int_or_none(fmt.get('fps')) or 0
3838 dct = {
3839 'asr': int_or_none(fmt.get('audioSampleRate')),
3840 'filesize': int_or_none(fmt.get('contentLength')),
3841 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3842 'format_note': join_nonempty(
3843 join_nonempty(audio_track.get('displayName'),
3844 language_preference > 0 and ' (default)', delim=''),
3845 name, fmt.get('isDrc') and 'DRC',
3846 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3847 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3848 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3849 (self.get_param('verbose') or all_formats) and client_name,
3850 delim=', '),
3851 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3852 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
3853 + (100 if 'Premium' in name else 0)),
3854 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
3855 'audio_channels': fmt.get('audioChannels'),
3856 'height': height,
3857 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3858 'has_drm': bool(fmt.get('drmFamilies')),
3859 'tbr': tbr,
3860 'url': fmt_url,
3861 'width': int_or_none(fmt.get('width')),
3862 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3863 'desc' if language_preference < -1 else '') or None,
3864 'language_preference': language_preference,
3865 # Strictly de-prioritize damaged and 3gp formats
3866 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3867 }
3868 mime_mobj = re.match(
3869 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3870 if mime_mobj:
3871 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3872 dct.update(parse_codecs(mime_mobj.group(2)))
3873 if itag:
3874 itags[itag].add(('https', dct.get('language')))
3875 stream_ids.append(stream_id)
3876 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3877 if single_stream and dct.get('ext'):
3878 dct['container'] = dct['ext'] + '_dash'
3879
3880 if (all_formats or 'dashy' in format_types) and dct['filesize']:
3881 yield {
3882 **dct,
3883 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3884 'protocol': 'http_dash_segments',
3885 'fragments': build_fragments(dct),
3886 }
3887 if all_formats or 'dashy' not in format_types:
3888 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3889 yield dct
3890
3891 needs_live_processing = self._needs_live_processing(live_status, duration)
3892 skip_bad_formats = 'incomplete' not in format_types
3893 if self._configuration_arg('include_incomplete_formats'):
3894 skip_bad_formats = False
3895 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
3896 'Use formats=incomplete extractor argument instead')
3897
3898 skip_manifests = set(self._configuration_arg('skip'))
3899 if (not self.get_param('youtube_include_hls_manifest', True)
3900 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3901 or needs_live_processing and skip_bad_formats):
3902 skip_manifests.add('hls')
3903
3904 if not self.get_param('youtube_include_dash_manifest', True):
3905 skip_manifests.add('dash')
3906 if self._configuration_arg('include_live_dash'):
3907 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3908 'Use formats=incomplete extractor argument instead')
3909 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3910 skip_manifests.add('dash')
3911
3912 def process_manifest_format(f, proto, client_name, itag):
3913 key = (proto, f.get('language'))
3914 if not all_formats and key in itags[itag]:
3915 return False
3916 itags[itag].add(key)
3917
3918 if itag and all_formats:
3919 f['format_id'] = f'{itag}-{proto}'
3920 elif any(p != proto for p, _ in itags[itag]):
3921 f['format_id'] = f'{itag}-{proto}'
3922 elif itag:
3923 f['format_id'] = itag
3924
3925 if f.get('source_preference') is None:
3926 f['source_preference'] = -1
3927
3928 if itag in ('616', '235'):
3929 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
3930 f['source_preference'] += 100
3931
3932 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3933 if f['quality'] == -1 and f.get('height'):
3934 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3935 if self.get_param('verbose') or all_formats:
3936 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3937 if f.get('fps') and f['fps'] <= 1:
3938 del f['fps']
3939
3940 if proto == 'hls' and f.get('has_drm'):
3941 f['has_drm'] = 'maybe'
3942 f['source_preference'] -= 5
3943 return True
3944
3945 subtitles = {}
3946 for sd in streaming_data:
3947 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3948
3949 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3950 if hls_manifest_url:
3951 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3952 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3953 subtitles = self._merge_subtitles(subs, subtitles)
3954 for f in fmts:
3955 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3956 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3957 yield f
3958
3959 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3960 if dash_manifest_url:
3961 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3962 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3963 for f in formats:
3964 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3965 f['filesize'] = int_or_none(self._search_regex(
3966 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3967 if needs_live_processing:
3968 f['is_from_start'] = True
3969
3970 yield f
3971 yield subtitles
3972
3973 def _extract_storyboard(self, player_responses, duration):
3974 spec = get_first(
3975 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3976 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3977 if not base_url:
3978 return
3979 L = len(spec) - 1
3980 for i, args in enumerate(spec):
3981 args = args.split('#')
3982 counts = list(map(int_or_none, args[:5]))
3983 if len(args) != 8 or not all(counts):
3984 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3985 continue
3986 width, height, frame_count, cols, rows = counts
3987 N, sigh = args[6:]
3988
3989 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3990 fragment_count = frame_count / (cols * rows)
3991 fragment_duration = duration / fragment_count
3992 yield {
3993 'format_id': f'sb{i}',
3994 'format_note': 'storyboard',
3995 'ext': 'mhtml',
3996 'protocol': 'mhtml',
3997 'acodec': 'none',
3998 'vcodec': 'none',
3999 'url': url,
4000 'width': width,
4001 'height': height,
4002 'fps': frame_count / duration,
4003 'rows': rows,
4004 'columns': cols,
4005 'fragments': [{
4006 'url': url.replace('$M', str(j)),
4007 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4008 } for j in range(math.ceil(fragment_count))],
4009 }
4010
4011 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4012 webpage = None
4013 if 'webpage' not in self._configuration_arg('player_skip'):
4014 query = {'bpctr': '9999999999', 'has_verified': '1'}
4015 pp = self._configuration_arg('player_params', [None], casesense=True)[0]
4016 if pp:
4017 query['pp'] = pp
4018 webpage = self._download_webpage(
4019 webpage_url, video_id, fatal=False, query=query)
4020
4021 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4022
4023 player_responses, player_url = self._extract_player_responses(
4024 self._get_requested_clients(url, smuggled_data),
4025 video_id, webpage, master_ytcfg, smuggled_data)
4026
4027 return webpage, master_ytcfg, player_responses, player_url
4028
4029 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4030 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4031 is_live = get_first(video_details, 'isLive')
4032 if is_live is None:
4033 is_live = get_first(live_broadcast_details, 'isLiveNow')
4034 live_content = get_first(video_details, 'isLiveContent')
4035 is_upcoming = get_first(video_details, 'isUpcoming')
4036 post_live = get_first(video_details, 'isPostLiveDvr')
4037 live_status = ('post_live' if post_live
4038 else 'is_live' if is_live
4039 else 'is_upcoming' if is_upcoming
4040 else 'was_live' if live_content
4041 else 'not_live' if False in (is_live, live_content)
4042 else None)
4043 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4044 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4045 if all(f.get('has_drm') for f in formats):
4046 # If there are no formats that definitely don't have DRM, all have DRM
4047 for f in formats:
4048 f['has_drm'] = True
4049
4050 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4051
4052 def _real_extract(self, url):
4053 url, smuggled_data = unsmuggle_url(url, {})
4054 video_id = self._match_id(url)
4055
4056 base_url = self.http_scheme() + '//www.youtube.com/'
4057 webpage_url = base_url + 'watch?v=' + video_id
4058
4059 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4060
4061 playability_statuses = traverse_obj(
4062 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4063
4064 trailer_video_id = get_first(
4065 playability_statuses,
4066 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4067 expected_type=str)
4068 if trailer_video_id:
4069 return self.url_result(
4070 trailer_video_id, self.ie_key(), trailer_video_id)
4071
4072 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4073 if webpage else (lambda x: None))
4074
4075 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4076 microformats = traverse_obj(
4077 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4078 expected_type=dict)
4079
4080 translated_title = self._get_text(microformats, (..., 'title'))
4081 video_title = (self._preferred_lang and translated_title
4082 or get_first(video_details, 'title') # primary
4083 or translated_title
4084 or search_meta(['og:title', 'twitter:title', 'title']))
4085 translated_description = self._get_text(microformats, (..., 'description'))
4086 original_description = get_first(video_details, 'shortDescription')
4087 video_description = (
4088 self._preferred_lang and translated_description
4089 # If original description is blank, it will be an empty string.
4090 # Do not prefer translated description in this case.
4091 or original_description if original_description is not None else translated_description)
4092
4093 multifeed_metadata_list = get_first(
4094 player_responses,
4095 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4096 expected_type=str)
4097 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4098 if self.get_param('noplaylist'):
4099 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4100 else:
4101 entries = []
4102 feed_ids = []
4103 for feed in multifeed_metadata_list.split(','):
4104 # Unquote should take place before split on comma (,) since textual
4105 # fields may contain comma as well (see
4106 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4107 feed_data = urllib.parse.parse_qs(
4108 urllib.parse.unquote_plus(feed))
4109
4110 def feed_entry(name):
4111 return try_get(
4112 feed_data, lambda x: x[name][0], str)
4113
4114 feed_id = feed_entry('id')
4115 if not feed_id:
4116 continue
4117 feed_title = feed_entry('title')
4118 title = video_title
4119 if feed_title:
4120 title += ' (%s)' % feed_title
4121 entries.append({
4122 '_type': 'url_transparent',
4123 'ie_key': 'Youtube',
4124 'url': smuggle_url(
4125 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4126 {'force_singlefeed': True}),
4127 'title': title,
4128 })
4129 feed_ids.append(feed_id)
4130 self.to_screen(
4131 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4132 % (', '.join(feed_ids), video_id))
4133 return self.playlist_result(
4134 entries, video_id, video_title, video_description)
4135
4136 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4137 or int_or_none(get_first(microformats, 'lengthSeconds'))
4138 or parse_duration(search_meta('duration')) or None)
4139
4140 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4141 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4142 if live_status == 'post_live':
4143 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4144
4145 if not formats:
4146 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4147 self.report_drm(video_id)
4148 pemr = get_first(
4149 playability_statuses,
4150 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4151 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4152 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4153 if subreason:
4154 if subreason == 'The uploader has not made this video available in your country.':
4155 countries = get_first(microformats, 'availableCountries')
4156 if not countries:
4157 regions_allowed = search_meta('regionsAllowed')
4158 countries = regions_allowed.split(',') if regions_allowed else None
4159 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4160 reason += f'. {subreason}'
4161 if reason:
4162 self.raise_no_formats(reason, expected=True)
4163
4164 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4165 if not keywords and webpage:
4166 keywords = [
4167 unescapeHTML(m.group('content'))
4168 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4169 for keyword in keywords:
4170 if keyword.startswith('yt:stretch='):
4171 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4172 if mobj:
4173 # NB: float is intentional for forcing float division
4174 w, h = (float(v) for v in mobj.groups())
4175 if w > 0 and h > 0:
4176 ratio = w / h
4177 for f in formats:
4178 if f.get('vcodec') != 'none':
4179 f['stretched_ratio'] = ratio
4180 break
4181 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4182 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4183 if thumbnail_url:
4184 thumbnails.append({
4185 'url': thumbnail_url,
4186 })
4187 original_thumbnails = thumbnails.copy()
4188
4189 # The best resolution thumbnails sometimes does not appear in the webpage
4190 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4191 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4192 thumbnail_names = [
4193 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4194 # in resolution, these are not the custom thumbnail. So de-prioritize them
4195 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4196 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4197 ]
4198 n_thumbnail_names = len(thumbnail_names)
4199 thumbnails.extend({
4200 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4201 video_id=video_id, name=name, ext=ext,
4202 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4203 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4204 for thumb in thumbnails:
4205 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4206 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4207 self._remove_duplicate_formats(thumbnails)
4208 self._downloader._sort_thumbnails(original_thumbnails)
4209
4210 category = get_first(microformats, 'category') or search_meta('genre')
4211 channel_id = self.ucid_or_none(str_or_none(
4212 get_first(video_details, 'channelId')
4213 or get_first(microformats, 'externalChannelId')
4214 or search_meta('channelId')))
4215 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4216
4217 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4218 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4219 if not duration and live_end_time and live_start_time:
4220 duration = live_end_time - live_start_time
4221
4222 needs_live_processing = self._needs_live_processing(live_status, duration)
4223
4224 def is_bad_format(fmt):
4225 if needs_live_processing and not fmt.get('is_from_start'):
4226 return True
4227 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4228 and fmt.get('protocol') == 'http_dash_segments'):
4229 return True
4230
4231 for fmt in filter(is_bad_format, formats):
4232 fmt['preference'] = (fmt.get('preference') or -1) - 10
4233 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4234
4235 if needs_live_processing:
4236 self._prepare_live_from_start_formats(
4237 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4238
4239 formats.extend(self._extract_storyboard(player_responses, duration))
4240
4241 channel_handle = self.handle_from_url(owner_profile_url)
4242
4243 info = {
4244 'id': video_id,
4245 'title': video_title,
4246 'formats': formats,
4247 'thumbnails': thumbnails,
4248 # The best thumbnail that we are sure exists. Prevents unnecessary
4249 # URL checking if user don't care about getting the best possible thumbnail
4250 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4251 'description': video_description,
4252 'channel_id': channel_id,
4253 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4254 'duration': duration,
4255 'view_count': int_or_none(
4256 get_first((video_details, microformats), (..., 'viewCount'))
4257 or search_meta('interactionCount')),
4258 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4259 'age_limit': 18 if (
4260 get_first(microformats, 'isFamilySafe') is False
4261 or search_meta('isFamilyFriendly') == 'false'
4262 or search_meta('og:restrictions:age') == '18+') else 0,
4263 'webpage_url': webpage_url,
4264 'categories': [category] if category else None,
4265 'tags': keywords,
4266 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4267 'live_status': live_status,
4268 'release_timestamp': live_start_time,
4269 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4270 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4271 }
4272
4273 subtitles = {}
4274 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4275 if pctr:
4276 def get_lang_code(track):
4277 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4278 or track.get('languageCode'))
4279
4280 # Converted into dicts to remove duplicates
4281 captions = {
4282 get_lang_code(sub): sub
4283 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4284 translation_languages = {
4285 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4286 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4287
4288 def process_language(container, base_url, lang_code, sub_name, query):
4289 lang_subs = container.setdefault(lang_code, [])
4290 for fmt in self._SUBTITLE_FORMATS:
4291 query.update({
4292 'fmt': fmt,
4293 })
4294 lang_subs.append({
4295 'ext': fmt,
4296 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4297 'name': sub_name,
4298 })
4299
4300 # NB: Constructing the full subtitle dictionary is slow
4301 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4302 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4303 for lang_code, caption_track in captions.items():
4304 base_url = caption_track.get('baseUrl')
4305 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4306 if not base_url:
4307 continue
4308 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4309 if caption_track.get('kind') != 'asr':
4310 if not lang_code:
4311 continue
4312 process_language(
4313 subtitles, base_url, lang_code, lang_name, {})
4314 if not caption_track.get('isTranslatable'):
4315 continue
4316 for trans_code, trans_name in translation_languages.items():
4317 if not trans_code:
4318 continue
4319 orig_trans_code = trans_code
4320 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4321 if not get_translated_subs:
4322 continue
4323 trans_code += f'-{lang_code}'
4324 trans_name += format_field(lang_name, None, ' from %s')
4325 if lang_code == f'a-{orig_trans_code}':
4326 # Set audio language based on original subtitles
4327 for f in formats:
4328 if f.get('acodec') != 'none' and not f.get('language'):
4329 f['language'] = orig_trans_code
4330 # Add an "-orig" label to the original language so that it can be distinguished.
4331 # The subs are returned without "-orig" as well for compatibility
4332 process_language(
4333 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4334 # Setting tlang=lang returns damaged subtitles.
4335 process_language(automatic_captions, base_url, trans_code, trans_name,
4336 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4337
4338 info['automatic_captions'] = automatic_captions
4339 info['subtitles'] = subtitles
4340
4341 parsed_url = urllib.parse.urlparse(url)
4342 for component in [parsed_url.fragment, parsed_url.query]:
4343 query = urllib.parse.parse_qs(component)
4344 for k, v in query.items():
4345 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4346 d_k += '_time'
4347 if d_k not in info and k in s_ks:
4348 info[d_k] = parse_duration(query[k][0])
4349
4350 # Youtube Music Auto-generated description
4351 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4352 # XXX: Causes catastrophic backtracking if description has "·"
4353 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4354 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4355 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4356 mobj = re.search(
4357 r'''(?xs)
4358 (?=(?P<track>[^\n·]+))(?P=track)·
4359 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4360 (?=(?P<album>[^\n]+))(?P=album)\n
4361 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4362 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4363 (.+?\nArtist\s*:\s*
4364 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4365 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4366 ''', video_description)
4367 if mobj:
4368 release_year = mobj.group('release_year')
4369 release_date = mobj.group('release_date')
4370 if release_date:
4371 release_date = release_date.replace('-', '')
4372 if not release_year:
4373 release_year = release_date[:4]
4374 info.update({
4375 'album': mobj.group('album'.strip()),
4376 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4377 'track': mobj.group('track').strip(),
4378 'release_date': release_date,
4379 'release_year': int_or_none(release_year),
4380 })
4381
4382 initial_data = None
4383 if webpage:
4384 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4385 if not traverse_obj(initial_data, 'contents'):
4386 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4387 initial_data = None
4388 if not initial_data:
4389 query = {'videoId': video_id}
4390 query.update(self._get_checkok_params())
4391 initial_data = self._extract_response(
4392 item_id=video_id, ep='next', fatal=False,
4393 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4394 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4395 note='Downloading initial data API JSON')
4396
4397 info['comment_count'] = traverse_obj(initial_data, (
4398 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4399 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4400 ), (
4401 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4402 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4403 ), expected_type=self._get_count, get_all=False)
4404
4405 try: # This will error if there is no livechat
4406 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4407 except (KeyError, IndexError, TypeError):
4408 pass
4409 else:
4410 info.setdefault('subtitles', {})['live_chat'] = [{
4411 # url is needed to set cookies
4412 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4413 'video_id': video_id,
4414 'ext': 'json',
4415 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4416 else 'youtube_live_chat_replay'),
4417 }]
4418
4419 if initial_data:
4420 info['chapters'] = (
4421 self._extract_chapters_from_json(initial_data, duration)
4422 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4423 or self._extract_chapters_from_description(video_description, duration)
4424 or None)
4425
4426 info['heatmap'] = self._extract_heatmap_from_player_overlay(initial_data)
4427
4428 contents = traverse_obj(
4429 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4430 expected_type=list, default=[])
4431
4432 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4433 if vpir:
4434 stl = vpir.get('superTitleLink')
4435 if stl:
4436 stl = self._get_text(stl)
4437 if try_get(
4438 vpir,
4439 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4440 info['location'] = stl
4441 else:
4442 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4443 if mobj:
4444 info.update({
4445 'series': mobj.group(1),
4446 'season_number': int(mobj.group(2)),
4447 'episode_number': int(mobj.group(3)),
4448 })
4449 for tlb in (try_get(
4450 vpir,
4451 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4452 list) or []):
4453 tbrs = variadic(
4454 traverse_obj(
4455 tlb, ('toggleButtonRenderer', ...),
4456 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4457 for tbr in tbrs:
4458 for getter, regex in [(
4459 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4460 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4461 lambda x: x['accessibility'],
4462 lambda x: x['accessibilityData']['accessibilityData'],
4463 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4464 label = (try_get(tbr, getter, dict) or {}).get('label')
4465 if label:
4466 mobj = re.match(regex, label)
4467 if mobj:
4468 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4469 break
4470 sbr_tooltip = try_get(
4471 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4472 if sbr_tooltip:
4473 like_count, dislike_count = sbr_tooltip.split(' / ')
4474 info.update({
4475 'like_count': str_to_int(like_count),
4476 'dislike_count': str_to_int(dislike_count),
4477 })
4478 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4479 if vcr:
4480 vc = self._get_count(vcr, 'viewCount')
4481 # Upcoming premieres with waiting count are treated as live here
4482 if vcr.get('isLive'):
4483 info['concurrent_view_count'] = vc
4484 elif info.get('view_count') is None:
4485 info['view_count'] = vc
4486
4487 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4488 if vsir:
4489 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4490 info.update({
4491 'channel': self._get_text(vor, 'title'),
4492 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4493
4494 if not channel_handle:
4495 channel_handle = self.handle_from_url(
4496 traverse_obj(vor, (
4497 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4498 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4499 {str}), get_all=False))
4500
4501 rows = try_get(
4502 vsir,
4503 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4504 list) or []
4505 multiple_songs = False
4506 for row in rows:
4507 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4508 multiple_songs = True
4509 break
4510 for row in rows:
4511 mrr = row.get('metadataRowRenderer') or {}
4512 mrr_title = mrr.get('title')
4513 if not mrr_title:
4514 continue
4515 mrr_title = self._get_text(mrr, 'title')
4516 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4517 if mrr_title == 'License':
4518 info['license'] = mrr_contents_text
4519 elif not multiple_songs:
4520 if mrr_title == 'Album':
4521 info['album'] = mrr_contents_text
4522 elif mrr_title == 'Artist':
4523 info['artist'] = mrr_contents_text
4524 elif mrr_title == 'Song':
4525 info['track'] = mrr_contents_text
4526 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4527 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4528 info['channel_is_verified'] = True
4529
4530 info.update({
4531 'uploader': info.get('channel'),
4532 'uploader_id': channel_handle,
4533 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4534 })
4535 # The upload date for scheduled, live and past live streams / premieres in microformats
4536 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4537 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4538 upload_date = (
4539 unified_strdate(get_first(microformats, 'uploadDate'))
4540 or unified_strdate(search_meta('uploadDate')))
4541 if not upload_date or (
4542 live_status in ('not_live', None)
4543 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4544 ):
4545 upload_date = strftime_or_none(
4546 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4547 info['upload_date'] = upload_date
4548
4549 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4550 v = info.get(s_k)
4551 if v:
4552 info[d_k] = v
4553
4554 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4555
4556 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4557 or get_first(video_details, 'isPrivate', expected_type=bool))
4558
4559 info['availability'] = (
4560 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4561 else self._availability(
4562 is_private=is_private,
4563 needs_premium=(
4564 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4565 or False if initial_data and is_private is not None else None),
4566 needs_subscription=(
4567 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4568 or False if initial_data and is_private is not None else None),
4569 needs_auth=info['age_limit'] >= 18,
4570 is_unlisted=None if is_private is None else (
4571 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4572 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4573
4574 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4575
4576 self.mark_watched(video_id, player_responses)
4577
4578 return info
4579
4580
4581 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4582 @staticmethod
4583 def passthrough_smuggled_data(func):
4584 def _smuggle(info, smuggled_data):
4585 if info.get('_type') not in ('url', 'url_transparent'):
4586 return info
4587 if smuggled_data.get('is_music_url'):
4588 parsed_url = urllib.parse.urlparse(info['url'])
4589 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4590 smuggled_data.pop('is_music_url')
4591 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4592 if smuggled_data:
4593 info['url'] = smuggle_url(info['url'], smuggled_data)
4594 return info
4595
4596 @functools.wraps(func)
4597 def wrapper(self, url):
4598 url, smuggled_data = unsmuggle_url(url, {})
4599 if self.is_music_url(url):
4600 smuggled_data['is_music_url'] = True
4601 info_dict = func(self, url, smuggled_data)
4602 if smuggled_data:
4603 _smuggle(info_dict, smuggled_data)
4604 if info_dict.get('entries'):
4605 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4606 return info_dict
4607 return wrapper
4608
4609 @staticmethod
4610 def _extract_basic_item_renderer(item):
4611 # Modified from _extract_grid_item_renderer
4612 known_basic_renderers = (
4613 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4614 )
4615 for key, renderer in item.items():
4616 if not isinstance(renderer, dict):
4617 continue
4618 elif key in known_basic_renderers:
4619 return renderer
4620 elif key.startswith('grid') and key.endswith('Renderer'):
4621 return renderer
4622
4623 def _extract_channel_renderer(self, renderer):
4624 channel_id = self.ucid_or_none(renderer['channelId'])
4625 title = self._get_text(renderer, 'title')
4626 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4627 channel_handle = self.handle_from_url(
4628 traverse_obj(renderer, (
4629 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4630 ('browseEndpoint', 'canonicalBaseUrl')),
4631 {str}), get_all=False))
4632 if not channel_handle:
4633 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
4634 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
4635 return {
4636 '_type': 'url',
4637 'url': channel_url,
4638 'id': channel_id,
4639 'ie_key': YoutubeTabIE.ie_key(),
4640 'channel': title,
4641 'uploader': title,
4642 'channel_id': channel_id,
4643 'channel_url': channel_url,
4644 'title': title,
4645 'uploader_id': channel_handle,
4646 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4647 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
4648 # However, in feed/channels this is set correctly to the subscriber count
4649 'channel_follower_count': traverse_obj(
4650 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
4651 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4652 'playlist_count': (
4653 # videoCountText may be the subscriber count
4654 self._get_count(renderer, 'videoCountText')
4655 if self._get_count(renderer, 'subscriberCountText') is not None else None),
4656 'description': self._get_text(renderer, 'descriptionSnippet'),
4657 'channel_is_verified': True if self._has_badge(
4658 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
4659 }
4660
4661 def _grid_entries(self, grid_renderer):
4662 for item in grid_renderer['items']:
4663 if not isinstance(item, dict):
4664 continue
4665 renderer = self._extract_basic_item_renderer(item)
4666 if not isinstance(renderer, dict):
4667 continue
4668 title = self._get_text(renderer, 'title')
4669
4670 # playlist
4671 playlist_id = renderer.get('playlistId')
4672 if playlist_id:
4673 yield self.url_result(
4674 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4675 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4676 video_title=title)
4677 continue
4678 # video
4679 video_id = renderer.get('videoId')
4680 if video_id:
4681 yield self._extract_video(renderer)
4682 continue
4683 # channel
4684 channel_id = renderer.get('channelId')
4685 if channel_id:
4686 yield self._extract_channel_renderer(renderer)
4687 continue
4688 # generic endpoint URL support
4689 ep_url = urljoin('https://www.youtube.com/', try_get(
4690 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4691 str))
4692 if ep_url:
4693 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4694 if ie.suitable(ep_url):
4695 yield self.url_result(
4696 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4697 break
4698
4699 def _music_reponsive_list_entry(self, renderer):
4700 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4701 if video_id:
4702 title = traverse_obj(renderer, (
4703 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4704 'text', 'runs', 0, 'text'))
4705 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4706 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4707 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4708 if playlist_id:
4709 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4710 if video_id:
4711 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4712 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4713 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4714 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4715 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4716 if browse_id:
4717 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4718 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4719
4720 def _shelf_entries_from_content(self, shelf_renderer):
4721 content = shelf_renderer.get('content')
4722 if not isinstance(content, dict):
4723 return
4724 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4725 if renderer:
4726 # TODO: add support for nested playlists so each shelf is processed
4727 # as separate playlist
4728 # TODO: this includes only first N items
4729 yield from self._grid_entries(renderer)
4730 renderer = content.get('horizontalListRenderer')
4731 if renderer:
4732 # TODO
4733 pass
4734
4735 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4736 ep = try_get(
4737 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4738 str)
4739 shelf_url = urljoin('https://www.youtube.com', ep)
4740 if shelf_url:
4741 # Skipping links to another channels, note that checking for
4742 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4743 # will not work
4744 if skip_channels and '/channels?' in shelf_url:
4745 return
4746 title = self._get_text(shelf_renderer, 'title')
4747 yield self.url_result(shelf_url, video_title=title)
4748 # Shelf may not contain shelf URL, fallback to extraction from content
4749 yield from self._shelf_entries_from_content(shelf_renderer)
4750
4751 def _playlist_entries(self, video_list_renderer):
4752 for content in video_list_renderer['contents']:
4753 if not isinstance(content, dict):
4754 continue
4755 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4756 if not isinstance(renderer, dict):
4757 continue
4758 video_id = renderer.get('videoId')
4759 if not video_id:
4760 continue
4761 yield self._extract_video(renderer)
4762
4763 def _rich_entries(self, rich_grid_renderer):
4764 renderer = traverse_obj(
4765 rich_grid_renderer,
4766 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4767 video_id = renderer.get('videoId')
4768 if video_id:
4769 yield self._extract_video(renderer)
4770 return
4771 playlist_id = renderer.get('playlistId')
4772 if playlist_id:
4773 yield self.url_result(
4774 f'https://www.youtube.com/playlist?list={playlist_id}',
4775 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4776 video_title=self._get_text(renderer, 'title'))
4777 return
4778
4779 def _video_entry(self, video_renderer):
4780 video_id = video_renderer.get('videoId')
4781 if video_id:
4782 return self._extract_video(video_renderer)
4783
4784 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4785 url = urljoin('https://youtube.com', traverse_obj(
4786 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4787 if url:
4788 return self.url_result(
4789 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4790
4791 def _post_thread_entries(self, post_thread_renderer):
4792 post_renderer = try_get(
4793 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4794 if not post_renderer:
4795 return
4796 # video attachment
4797 video_renderer = try_get(
4798 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4799 video_id = video_renderer.get('videoId')
4800 if video_id:
4801 entry = self._extract_video(video_renderer)
4802 if entry:
4803 yield entry
4804 # playlist attachment
4805 playlist_id = try_get(
4806 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4807 if playlist_id:
4808 yield self.url_result(
4809 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4810 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4811 # inline video links
4812 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4813 for run in runs:
4814 if not isinstance(run, dict):
4815 continue
4816 ep_url = try_get(
4817 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4818 if not ep_url:
4819 continue
4820 if not YoutubeIE.suitable(ep_url):
4821 continue
4822 ep_video_id = YoutubeIE._match_id(ep_url)
4823 if video_id == ep_video_id:
4824 continue
4825 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4826
4827 def _post_thread_continuation_entries(self, post_thread_continuation):
4828 contents = post_thread_continuation.get('contents')
4829 if not isinstance(contents, list):
4830 return
4831 for content in contents:
4832 renderer = content.get('backstagePostThreadRenderer')
4833 if isinstance(renderer, dict):
4834 yield from self._post_thread_entries(renderer)
4835 continue
4836 renderer = content.get('videoRenderer')
4837 if isinstance(renderer, dict):
4838 yield self._video_entry(renderer)
4839
4840 r''' # unused
4841 def _rich_grid_entries(self, contents):
4842 for content in contents:
4843 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4844 if video_renderer:
4845 entry = self._video_entry(video_renderer)
4846 if entry:
4847 yield entry
4848 '''
4849
4850 def _report_history_entries(self, renderer):
4851 for url in traverse_obj(renderer, (
4852 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4853 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4854 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4855 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4856
4857 def _extract_entries(self, parent_renderer, continuation_list):
4858 # continuation_list is modified in-place with continuation_list = [continuation_token]
4859 continuation_list[:] = [None]
4860 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4861 for content in contents:
4862 if not isinstance(content, dict):
4863 continue
4864 is_renderer = traverse_obj(
4865 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4866 expected_type=dict)
4867 if not is_renderer:
4868 if content.get('richItemRenderer'):
4869 for entry in self._rich_entries(content['richItemRenderer']):
4870 yield entry
4871 continuation_list[0] = self._extract_continuation(parent_renderer)
4872 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4873 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4874 yield from self._report_history_entries(table)
4875 continuation_list[0] = self._extract_continuation(table)
4876 continue
4877
4878 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4879 for isr_content in isr_contents:
4880 if not isinstance(isr_content, dict):
4881 continue
4882
4883 known_renderers = {
4884 'playlistVideoListRenderer': self._playlist_entries,
4885 'gridRenderer': self._grid_entries,
4886 'reelShelfRenderer': self._grid_entries,
4887 'shelfRenderer': self._shelf_entries,
4888 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4889 'backstagePostThreadRenderer': self._post_thread_entries,
4890 'videoRenderer': lambda x: [self._video_entry(x)],
4891 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4892 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4893 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
4894 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
4895 }
4896 for key, renderer in isr_content.items():
4897 if key not in known_renderers:
4898 continue
4899 for entry in known_renderers[key](renderer):
4900 if entry:
4901 yield entry
4902 continuation_list[0] = self._extract_continuation(renderer)
4903 break
4904
4905 if not continuation_list[0]:
4906 continuation_list[0] = self._extract_continuation(is_renderer)
4907
4908 if not continuation_list[0]:
4909 continuation_list[0] = self._extract_continuation(parent_renderer)
4910
4911 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4912 continuation_list = [None]
4913 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4914 tab_content = try_get(tab, lambda x: x['content'], dict)
4915 if not tab_content:
4916 return
4917 parent_renderer = (
4918 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4919 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4920 yield from extract_entries(parent_renderer)
4921 continuation = continuation_list[0]
4922 seen_continuations = set()
4923 for page_num in itertools.count(1):
4924 if not continuation:
4925 break
4926 continuation_token = continuation.get('continuation')
4927 if continuation_token is not None and continuation_token in seen_continuations:
4928 self.write_debug('Detected YouTube feed looping - assuming end of feed.')
4929 break
4930 seen_continuations.add(continuation_token)
4931 headers = self.generate_api_headers(
4932 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4933 response = self._extract_response(
4934 item_id=f'{item_id} page {page_num}',
4935 query=continuation, headers=headers, ytcfg=ytcfg,
4936 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4937
4938 if not response:
4939 break
4940 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4941 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4942 visitor_data = self._extract_visitor_data(response) or visitor_data
4943
4944 known_renderers = {
4945 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4946 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4947 'gridVideoRenderer': (self._grid_entries, 'items'),
4948 'gridChannelRenderer': (self._grid_entries, 'items'),
4949 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4950 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4951 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4952 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4953 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4954 'playlistVideoListContinuation': (self._playlist_entries, None),
4955 'gridContinuation': (self._grid_entries, None),
4956 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4957 'sectionListContinuation': (extract_entries, None), # for feeds
4958 }
4959
4960 continuation_items = traverse_obj(response, (
4961 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4962 'appendContinuationItemsAction', 'continuationItems'
4963 ), 'continuationContents', get_all=False)
4964 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4965
4966 video_items_renderer = None
4967 for key in continuation_item.keys():
4968 if key not in known_renderers:
4969 continue
4970 func, parent_key = known_renderers[key]
4971 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4972 continuation_list = [None]
4973 yield from func(video_items_renderer)
4974 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4975
4976 if not video_items_renderer:
4977 break
4978
4979 @staticmethod
4980 def _extract_selected_tab(tabs, fatal=True):
4981 for tab_renderer in tabs:
4982 if tab_renderer.get('selected'):
4983 return tab_renderer
4984 if fatal:
4985 raise ExtractorError('Unable to find selected tab')
4986
4987 @staticmethod
4988 def _extract_tab_renderers(response):
4989 return traverse_obj(
4990 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
4991
4992 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4993 metadata = self._extract_metadata_from_tabs(item_id, data)
4994
4995 selected_tab = self._extract_selected_tab(tabs)
4996 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
4997 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
4998
4999 return self.playlist_result(
5000 self._entries(
5001 selected_tab, metadata['id'], ytcfg,
5002 self._extract_account_syncid(ytcfg, data),
5003 self._extract_visitor_data(data, ytcfg)),
5004 **metadata)
5005
5006 def _extract_metadata_from_tabs(self, item_id, data):
5007 info = {'id': item_id}
5008
5009 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
5010 if metadata_renderer:
5011 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
5012 ('channelUrl', {self.ucid_from_url}))
5013 info.update({
5014 'channel': metadata_renderer.get('title'),
5015 'channel_id': channel_id,
5016 })
5017 if info['channel_id']:
5018 info['id'] = info['channel_id']
5019 else:
5020 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5021
5022 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5023 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5024 def _get_uncropped(url):
5025 return url_or_none((url or '').split('=')[0] + '=s0')
5026
5027 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5028 if avatar_thumbnails:
5029 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5030 if uncropped_avatar:
5031 avatar_thumbnails.append({
5032 'url': uncropped_avatar,
5033 'id': 'avatar_uncropped',
5034 'preference': 1
5035 })
5036
5037 channel_banners = self._extract_thumbnails(
5038 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5039 for banner in channel_banners:
5040 banner['preference'] = -10
5041
5042 if channel_banners:
5043 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5044 if uncropped_banner:
5045 channel_banners.append({
5046 'url': uncropped_banner,
5047 'id': 'banner_uncropped',
5048 'preference': -5
5049 })
5050
5051 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5052 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5053 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5054
5055 primary_thumbnails = self._extract_thumbnails(
5056 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5057 playlist_thumbnails = self._extract_thumbnails(
5058 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5059
5060 info.update({
5061 'title': (traverse_obj(metadata_renderer, 'title')
5062 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5063 or info['id']),
5064 'availability': self._extract_availability(data),
5065 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
5066 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5067 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
5068 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5069 })
5070
5071 channel_handle = (
5072 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5073 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5074
5075 if channel_handle:
5076 info.update({
5077 'uploader_id': channel_handle,
5078 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5079 })
5080
5081 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5082 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5083 info['channel_is_verified'] = True
5084 # Playlist stats is a text runs array containing [video count, view count, last updated].
5085 # last updated or (view count and last updated) may be missing.
5086 playlist_stats = get_first(
5087 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5088
5089 last_updated_unix = self._parse_time_text(
5090 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5091 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5092 info['modified_date'] = strftime_or_none(last_updated_unix)
5093
5094 info['view_count'] = self._get_count(playlist_stats, 1)
5095 if info['view_count'] is None: # 0 is allowed
5096 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5097 if info['view_count'] is None:
5098 info['view_count'] = self._get_count(data, (
5099 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5100 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5101
5102 info['playlist_count'] = self._get_count(playlist_stats, 0)
5103 if info['playlist_count'] is None: # 0 is allowed
5104 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5105
5106 if not info.get('channel_id'):
5107 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5108 if not owner: # Deprecated
5109 owner = traverse_obj(
5110 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5111 ('videoOwner', 'videoOwnerRenderer', 'title'))
5112 owner_text = self._get_text(owner)
5113 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5114 info.update({
5115 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5116 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5117 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
5118 })
5119
5120 info.update({
5121 'uploader': info['channel'],
5122 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5123 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5124 })
5125
5126 return info
5127
5128 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5129 first_id = last_id = response = None
5130 for page_num in itertools.count(1):
5131 videos = list(self._playlist_entries(playlist))
5132 if not videos:
5133 return
5134 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5135 if start >= len(videos):
5136 return
5137 yield from videos[start:]
5138 first_id = first_id or videos[0]['id']
5139 last_id = videos[-1]['id']
5140 watch_endpoint = try_get(
5141 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5142 headers = self.generate_api_headers(
5143 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5144 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5145 query = {
5146 'playlistId': playlist_id,
5147 'videoId': watch_endpoint.get('videoId') or last_id,
5148 'index': watch_endpoint.get('index') or len(videos),
5149 'params': watch_endpoint.get('params') or 'OAE%3D'
5150 }
5151 response = self._extract_response(
5152 item_id='%s page %d' % (playlist_id, page_num),
5153 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5154 check_get_keys='contents'
5155 )
5156 playlist = try_get(
5157 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5158
5159 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5160 title = playlist.get('title') or try_get(
5161 data, lambda x: x['titleText']['simpleText'], str)
5162 playlist_id = playlist.get('playlistId') or item_id
5163
5164 # Delegating everything except mix playlists to regular tab-based playlist URL
5165 playlist_url = urljoin(url, try_get(
5166 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5167 str))
5168
5169 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5170 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5171 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5172
5173 if playlist_url and playlist_url != url and not is_known_unviewable:
5174 return self.url_result(
5175 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5176 video_title=title)
5177
5178 return self.playlist_result(
5179 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5180 playlist_id=playlist_id, playlist_title=title)
5181
5182 def _extract_availability(self, data):
5183 """
5184 Gets the availability of a given playlist/tab.
5185 Note: Unless YouTube tells us explicitly, we do not assume it is public
5186 @param data: response
5187 """
5188 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5189 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5190 player_header_privacy = playlist_header_renderer.get('privacy')
5191
5192 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5193
5194 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5195 privacy_setting_icon = get_first(
5196 (playlist_header_renderer, sidebar_renderer),
5197 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5198 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5199 expected_type=str)
5200
5201 microformats_is_unlisted = traverse_obj(
5202 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5203
5204 return (
5205 'public' if (
5206 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5207 or player_header_privacy == 'PUBLIC'
5208 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5209 else self._availability(
5210 is_private=(
5211 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5212 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5213 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5214 is_unlisted=(
5215 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5216 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5217 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5218 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5219 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5220 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5221 needs_auth=False))
5222
5223 @staticmethod
5224 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5225 sidebar_renderer = try_get(
5226 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5227 for item in sidebar_renderer:
5228 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5229 if renderer:
5230 return renderer
5231
5232 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5233 """
5234 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5235 """
5236 is_playlist = bool(traverse_obj(
5237 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5238 if not is_playlist:
5239 return
5240 headers = self.generate_api_headers(
5241 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5242 visitor_data=self._extract_visitor_data(data, ytcfg))
5243 query = {
5244 'params': 'wgYCCAA=',
5245 'browseId': f'VL{item_id}'
5246 }
5247 return self._extract_response(
5248 item_id=item_id, headers=headers, query=query,
5249 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5250 note='Redownloading playlist API JSON with unavailable videos')
5251
5252 @functools.cached_property
5253 def skip_webpage(self):
5254 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5255
5256 def _extract_webpage(self, url, item_id, fatal=True):
5257 webpage, data = None, None
5258 for retry in self.RetryManager(fatal=fatal):
5259 try:
5260 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5261 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5262 except ExtractorError as e:
5263 if isinstance(e.cause, network_exceptions):
5264 if not isinstance(e.cause, HTTPError) or e.cause.status not in (403, 429):
5265 retry.error = e
5266 continue
5267 self._error_or_warning(e, fatal=fatal)
5268 break
5269
5270 try:
5271 self._extract_and_report_alerts(data)
5272 except ExtractorError as e:
5273 self._error_or_warning(e, fatal=fatal)
5274 break
5275
5276 # Sometimes youtube returns a webpage with incomplete ytInitialData
5277 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5278 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5279 retry.error = ExtractorError('Incomplete yt initial data received')
5280 continue
5281
5282 return webpage, data
5283
5284 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5285 """Use if failed to extract ytcfg (and data) from initial webpage"""
5286 if not ytcfg and self.is_authenticated:
5287 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5288 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5289 raise ExtractorError(
5290 f'{msg}. If you are not downloading private content, or '
5291 'your cookies are only for the first account and channel,'
5292 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5293 expected=True)
5294 self.report_warning(msg, only_once=True)
5295
5296 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5297 data = None
5298 if not self.skip_webpage:
5299 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5300 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5301 # Reject webpage data if redirected to home page without explicitly requesting
5302 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5303 if (url != 'https://www.youtube.com/feed/recommended'
5304 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5305 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5306 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5307 if fatal:
5308 raise ExtractorError(msg, expected=True)
5309 self.report_warning(msg, only_once=True)
5310 if not data:
5311 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5312 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5313 return data, ytcfg
5314
5315 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5316 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5317 resolve_response = self._extract_response(
5318 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5319 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5320 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5321 for ep_key, ep in endpoints.items():
5322 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5323 if params:
5324 return self._extract_response(
5325 item_id=item_id, query=params, ep=ep, headers=headers,
5326 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5327 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5328 err_note = 'Failed to resolve url (does the playlist exist?)'
5329 if fatal:
5330 raise ExtractorError(err_note, expected=True)
5331 self.report_warning(err_note, item_id)
5332
5333 _SEARCH_PARAMS = None
5334
5335 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5336 data = {'query': query}
5337 if params is NO_DEFAULT:
5338 params = self._SEARCH_PARAMS
5339 if params:
5340 data['params'] = params
5341
5342 content_keys = (
5343 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5344 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5345 # ytmusic search
5346 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5347 ('continuationContents', ),
5348 )
5349 display_id = f'query "{query}"'
5350 check_get_keys = tuple({keys[0] for keys in content_keys})
5351 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5352 self._report_playlist_authcheck(ytcfg, fatal=False)
5353
5354 continuation_list = [None]
5355 search = None
5356 for page_num in itertools.count(1):
5357 data.update(continuation_list[0] or {})
5358 headers = self.generate_api_headers(
5359 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5360 search = self._extract_response(
5361 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5362 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5363 slr_contents = traverse_obj(search, *content_keys)
5364 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5365 if not continuation_list[0]:
5366 break
5367
5368
5369 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5370 IE_DESC = 'YouTube Tabs'
5371 _VALID_URL = r'''(?x:
5372 https?://
5373 (?!consent\.)(?:\w+\.)?
5374 (?:
5375 youtube(?:kids)?\.com|
5376 %(invidious)s
5377 )/
5378 (?:
5379 (?P<channel_type>channel|c|user|browse)/|
5380 (?P<not_channel>
5381 feed/|hashtag/|
5382 (?:playlist|watch)\?.*?\blist=
5383 )|
5384 (?!(?:%(reserved_names)s)\b) # Direct URLs
5385 )
5386 (?P<id>[^/?\#&]+)
5387 )''' % {
5388 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5389 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5390 }
5391 IE_NAME = 'youtube:tab'
5392
5393 _TESTS = [{
5394 'note': 'playlists, multipage',
5395 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5396 'playlist_mincount': 94,
5397 'info_dict': {
5398 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5399 'title': 'Igor Kleiner - Playlists',
5400 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5401 'uploader': 'Igor Kleiner',
5402 'uploader_id': '@IgorDataScience',
5403 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5404 'channel': 'Igor Kleiner',
5405 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5406 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5407 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5408 'channel_follower_count': int
5409 },
5410 }, {
5411 'note': 'playlists, multipage, different order',
5412 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5413 'playlist_mincount': 94,
5414 'info_dict': {
5415 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5416 'title': 'Igor Kleiner - Playlists',
5417 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5418 'uploader': 'Igor Kleiner',
5419 'uploader_id': '@IgorDataScience',
5420 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5421 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5422 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5423 'channel': 'Igor Kleiner',
5424 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5425 'channel_follower_count': int
5426 },
5427 }, {
5428 'note': 'playlists, series',
5429 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5430 'playlist_mincount': 5,
5431 'info_dict': {
5432 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5433 'title': '3Blue1Brown - Playlists',
5434 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5435 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5436 'channel': '3Blue1Brown',
5437 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5438 'uploader_id': '@3blue1brown',
5439 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5440 'uploader': '3Blue1Brown',
5441 'tags': ['Mathematics'],
5442 'channel_follower_count': int,
5443 'channel_is_verified': True,
5444 },
5445 }, {
5446 'note': 'playlists, singlepage',
5447 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5448 'playlist_mincount': 4,
5449 'info_dict': {
5450 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5451 'title': 'ThirstForScience - Playlists',
5452 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5453 'uploader': 'ThirstForScience',
5454 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5455 'uploader_id': '@ThirstForScience',
5456 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5457 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5458 'tags': 'count:13',
5459 'channel': 'ThirstForScience',
5460 'channel_follower_count': int
5461 }
5462 }, {
5463 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5464 'only_matching': True,
5465 }, {
5466 'note': 'basic, single video playlist',
5467 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5468 'info_dict': {
5469 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5470 'title': 'youtube-dl public playlist',
5471 'description': '',
5472 'tags': [],
5473 'view_count': int,
5474 'modified_date': '20201130',
5475 'channel': 'Sergey M.',
5476 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5477 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5478 'availability': 'public',
5479 'uploader': 'Sergey M.',
5480 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5481 'uploader_id': '@sergeym.6173',
5482 },
5483 'playlist_count': 1,
5484 }, {
5485 'note': 'empty playlist',
5486 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5487 'info_dict': {
5488 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5489 'title': 'youtube-dl empty playlist',
5490 'tags': [],
5491 'channel': 'Sergey M.',
5492 'description': '',
5493 'modified_date': '20160902',
5494 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5495 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5496 'availability': 'public',
5497 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5498 'uploader_id': '@sergeym.6173',
5499 'uploader': 'Sergey M.',
5500 },
5501 'playlist_count': 0,
5502 }, {
5503 'note': 'Home tab',
5504 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5505 'info_dict': {
5506 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5507 'title': 'lex will - Home',
5508 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5509 'uploader': 'lex will',
5510 'uploader_id': '@lexwill718',
5511 'channel': 'lex will',
5512 'tags': ['bible', 'history', 'prophesy'],
5513 'uploader_url': 'https://www.youtube.com/@lexwill718',
5514 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5515 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5516 'channel_follower_count': int
5517 },
5518 'playlist_mincount': 2,
5519 }, {
5520 'note': 'Videos tab',
5521 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5522 'info_dict': {
5523 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5524 'title': 'lex will - Videos',
5525 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5526 'uploader': 'lex will',
5527 'uploader_id': '@lexwill718',
5528 'tags': ['bible', 'history', 'prophesy'],
5529 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5530 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5531 'uploader_url': 'https://www.youtube.com/@lexwill718',
5532 'channel': 'lex will',
5533 'channel_follower_count': int
5534 },
5535 'playlist_mincount': 975,
5536 }, {
5537 'note': 'Videos tab, sorted by popular',
5538 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5539 'info_dict': {
5540 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5541 'title': 'lex will - Videos',
5542 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5543 'uploader': 'lex will',
5544 'uploader_id': '@lexwill718',
5545 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5546 'uploader_url': 'https://www.youtube.com/@lexwill718',
5547 'channel': 'lex will',
5548 'tags': ['bible', 'history', 'prophesy'],
5549 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5550 'channel_follower_count': int
5551 },
5552 'playlist_mincount': 199,
5553 }, {
5554 'note': 'Playlists tab',
5555 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5556 'info_dict': {
5557 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5558 'title': 'lex will - Playlists',
5559 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5560 'uploader': 'lex will',
5561 'uploader_id': '@lexwill718',
5562 'uploader_url': 'https://www.youtube.com/@lexwill718',
5563 'channel': 'lex will',
5564 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5565 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5566 'tags': ['bible', 'history', 'prophesy'],
5567 'channel_follower_count': int
5568 },
5569 'playlist_mincount': 17,
5570 }, {
5571 'note': 'Community tab',
5572 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5573 'info_dict': {
5574 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5575 'title': 'lex will - Community',
5576 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5577 'channel': 'lex will',
5578 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5579 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5580 'tags': ['bible', 'history', 'prophesy'],
5581 'channel_follower_count': int,
5582 'uploader_url': 'https://www.youtube.com/@lexwill718',
5583 'uploader_id': '@lexwill718',
5584 'uploader': 'lex will',
5585 },
5586 'playlist_mincount': 18,
5587 }, {
5588 'note': 'Channels tab',
5589 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5590 'info_dict': {
5591 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5592 'title': 'lex will - Channels',
5593 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5594 'channel': 'lex will',
5595 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5596 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5597 'tags': ['bible', 'history', 'prophesy'],
5598 'channel_follower_count': int,
5599 'uploader_url': 'https://www.youtube.com/@lexwill718',
5600 'uploader_id': '@lexwill718',
5601 'uploader': 'lex will',
5602 },
5603 'playlist_mincount': 12,
5604 }, {
5605 'note': 'Search tab',
5606 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5607 'playlist_mincount': 40,
5608 'info_dict': {
5609 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5610 'title': '3Blue1Brown - Search - linear algebra',
5611 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5612 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5613 'tags': ['Mathematics'],
5614 'channel': '3Blue1Brown',
5615 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5616 'channel_follower_count': int,
5617 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5618 'uploader_id': '@3blue1brown',
5619 'uploader': '3Blue1Brown',
5620 'channel_is_verified': True,
5621 },
5622 }, {
5623 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5624 'only_matching': True,
5625 }, {
5626 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5627 'only_matching': True,
5628 }, {
5629 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5630 'only_matching': True,
5631 }, {
5632 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5633 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5634 'info_dict': {
5635 'title': '29C3: Not my department',
5636 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5637 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5638 'tags': [],
5639 'view_count': int,
5640 'modified_date': '20150605',
5641 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5642 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5643 'channel': 'Christiaan008',
5644 'availability': 'public',
5645 'uploader_id': '@ChRiStIaAn008',
5646 'uploader': 'Christiaan008',
5647 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5648 },
5649 'playlist_count': 96,
5650 }, {
5651 'note': 'Large playlist',
5652 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5653 'info_dict': {
5654 'title': 'Uploads from Cauchemar',
5655 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5656 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5657 'tags': [],
5658 'modified_date': r're:\d{8}',
5659 'channel': 'Cauchemar',
5660 'view_count': int,
5661 'description': '',
5662 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5663 'availability': 'public',
5664 'uploader_id': '@Cauchemar89',
5665 'uploader': 'Cauchemar',
5666 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5667 },
5668 'playlist_mincount': 1123,
5669 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5670 }, {
5671 'note': 'even larger playlist, 8832 videos',
5672 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5673 'only_matching': True,
5674 }, {
5675 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5676 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5677 'info_dict': {
5678 'title': 'Uploads from Interstellar Movie',
5679 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5680 'tags': [],
5681 'view_count': int,
5682 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5683 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5684 'channel': 'Interstellar Movie',
5685 'description': '',
5686 'modified_date': r're:\d{8}',
5687 'availability': 'public',
5688 'uploader_id': '@InterstellarMovie',
5689 'uploader': 'Interstellar Movie',
5690 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5691 },
5692 'playlist_mincount': 21,
5693 }, {
5694 'note': 'Playlist with "show unavailable videos" button',
5695 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5696 'info_dict': {
5697 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5698 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5699 'view_count': int,
5700 'channel': 'Phim Siêu Nhân Nhật Bản',
5701 'tags': [],
5702 'description': '',
5703 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5704 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5705 'modified_date': r're:\d{8}',
5706 'availability': 'public',
5707 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5708 'uploader_id': '@phimsieunhannhatban',
5709 'uploader': 'Phim Siêu Nhân Nhật Bản',
5710 },
5711 'playlist_mincount': 200,
5712 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5713 }, {
5714 'note': 'Playlist with unavailable videos in page 7',
5715 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5716 'info_dict': {
5717 'title': 'Uploads from BlankTV',
5718 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5719 'channel': 'BlankTV',
5720 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5721 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5722 'view_count': int,
5723 'tags': [],
5724 'modified_date': r're:\d{8}',
5725 'description': '',
5726 'availability': 'public',
5727 'uploader_id': '@blanktv',
5728 'uploader': 'BlankTV',
5729 'uploader_url': 'https://www.youtube.com/@blanktv',
5730 },
5731 'playlist_mincount': 1000,
5732 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5733 }, {
5734 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5735 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5736 'info_dict': {
5737 'title': 'Data Analysis with Dr Mike Pound',
5738 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5739 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5740 'tags': [],
5741 'view_count': int,
5742 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5743 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5744 'channel': 'Computerphile',
5745 'availability': 'public',
5746 'modified_date': '20190712',
5747 'uploader_id': '@Computerphile',
5748 'uploader': 'Computerphile',
5749 'uploader_url': 'https://www.youtube.com/@Computerphile',
5750 },
5751 'playlist_mincount': 11,
5752 }, {
5753 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5754 'only_matching': True,
5755 }, {
5756 'note': 'Playlist URL that does not actually serve a playlist',
5757 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5758 'info_dict': {
5759 'id': 'FqZTN594JQw',
5760 'ext': 'webm',
5761 'title': "Smiley's People 01 detective, Adventure Series, Action",
5762 'upload_date': '20150526',
5763 'license': 'Standard YouTube License',
5764 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5765 'categories': ['People & Blogs'],
5766 'tags': list,
5767 'view_count': int,
5768 'like_count': int,
5769 },
5770 'params': {
5771 'skip_download': True,
5772 },
5773 'skip': 'This video is not available.',
5774 'add_ie': [YoutubeIE.ie_key()],
5775 }, {
5776 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5777 'only_matching': True,
5778 }, {
5779 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5780 'only_matching': True,
5781 }, {
5782 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5783 'info_dict': {
5784 'id': 'hGkQjiJLjWQ', # This will keep changing
5785 'ext': 'mp4',
5786 'title': str,
5787 'upload_date': r're:\d{8}',
5788 'description': str,
5789 'categories': ['News & Politics'],
5790 'tags': list,
5791 'like_count': int,
5792 'release_timestamp': int,
5793 'channel': 'Sky News',
5794 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5795 'age_limit': 0,
5796 'view_count': int,
5797 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5798 'playable_in_embed': True,
5799 'release_date': r're:\d+',
5800 'availability': 'public',
5801 'live_status': 'is_live',
5802 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5803 'channel_follower_count': int,
5804 'concurrent_view_count': int,
5805 'uploader_url': 'https://www.youtube.com/@SkyNews',
5806 'uploader_id': '@SkyNews',
5807 'uploader': 'Sky News',
5808 'channel_is_verified': True,
5809 },
5810 'params': {
5811 'skip_download': True,
5812 },
5813 'expected_warnings': ['Ignoring subtitle tracks found in '],
5814 }, {
5815 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5816 'info_dict': {
5817 'id': 'a48o2S1cPoo',
5818 'ext': 'mp4',
5819 'title': 'The Young Turks - Live Main Show',
5820 'upload_date': '20150715',
5821 'license': 'Standard YouTube License',
5822 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5823 'categories': ['News & Politics'],
5824 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5825 'like_count': int,
5826 },
5827 'params': {
5828 'skip_download': True,
5829 },
5830 'only_matching': True,
5831 }, {
5832 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5833 'only_matching': True,
5834 }, {
5835 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5836 'only_matching': True,
5837 }, {
5838 'note': 'A channel that is not live. Should raise error',
5839 'url': 'https://www.youtube.com/user/numberphile/live',
5840 'only_matching': True,
5841 }, {
5842 'url': 'https://www.youtube.com/feed/trending',
5843 'only_matching': True,
5844 }, {
5845 'url': 'https://www.youtube.com/feed/library',
5846 'only_matching': True,
5847 }, {
5848 'url': 'https://www.youtube.com/feed/history',
5849 'only_matching': True,
5850 }, {
5851 'url': 'https://www.youtube.com/feed/subscriptions',
5852 'only_matching': True,
5853 }, {
5854 'url': 'https://www.youtube.com/feed/watch_later',
5855 'only_matching': True,
5856 }, {
5857 'note': 'Recommended - redirects to home page.',
5858 'url': 'https://www.youtube.com/feed/recommended',
5859 'only_matching': True,
5860 }, {
5861 'note': 'inline playlist with not always working continuations',
5862 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5863 'only_matching': True,
5864 }, {
5865 'url': 'https://www.youtube.com/course',
5866 'only_matching': True,
5867 }, {
5868 'url': 'https://www.youtube.com/zsecurity',
5869 'only_matching': True,
5870 }, {
5871 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5872 'only_matching': True,
5873 }, {
5874 'url': 'https://www.youtube.com/TheYoungTurks/live',
5875 'only_matching': True,
5876 }, {
5877 'url': 'https://www.youtube.com/hashtag/cctv9',
5878 'info_dict': {
5879 'id': 'cctv9',
5880 'title': '#cctv9',
5881 'tags': [],
5882 },
5883 'playlist_mincount': 300, # not consistent but should be over 300
5884 }, {
5885 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5886 'only_matching': True,
5887 }, {
5888 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5889 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5890 'only_matching': True
5891 }, {
5892 'note': '/browse/ should redirect to /channel/',
5893 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5894 'only_matching': True
5895 }, {
5896 'note': 'VLPL, should redirect to playlist?list=PL...',
5897 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5898 'info_dict': {
5899 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5900 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5901 'title': 'NCS : All Releases 💿',
5902 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5903 'modified_date': r're:\d{8}',
5904 'view_count': int,
5905 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5906 'tags': [],
5907 'channel': 'NoCopyrightSounds',
5908 'availability': 'public',
5909 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5910 'uploader': 'NoCopyrightSounds',
5911 'uploader_id': '@NoCopyrightSounds',
5912 },
5913 'playlist_mincount': 166,
5914 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5915 }, {
5916 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5917 'note': 'Topic, should redirect to playlist?list=UU...',
5918 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5919 'info_dict': {
5920 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5921 'title': 'Uploads from Royalty Free Music - Topic',
5922 'tags': [],
5923 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5924 'channel': 'Royalty Free Music - Topic',
5925 'view_count': int,
5926 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5927 'modified_date': r're:\d{8}',
5928 'description': '',
5929 'availability': 'public',
5930 'uploader': 'Royalty Free Music - Topic',
5931 },
5932 'playlist_mincount': 101,
5933 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5934 }, {
5935 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5936 # Treat as a general feed
5937 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5938 'info_dict': {
5939 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5940 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5941 'tags': [],
5942 },
5943 'playlist_mincount': 9,
5944 }, {
5945 'note': 'Youtube music Album',
5946 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5947 'info_dict': {
5948 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5949 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5950 'tags': [],
5951 'view_count': int,
5952 'description': '',
5953 'availability': 'unlisted',
5954 'modified_date': r're:\d{8}',
5955 },
5956 'playlist_count': 50,
5957 'expected_warnings': ['YouTube Music is not directly supported'],
5958 }, {
5959 'note': 'unlisted single video playlist',
5960 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5961 'info_dict': {
5962 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5963 'title': 'yt-dlp unlisted playlist test',
5964 'availability': 'unlisted',
5965 'tags': [],
5966 'modified_date': '20220418',
5967 'channel': 'colethedj',
5968 'view_count': int,
5969 'description': '',
5970 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5971 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5972 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5973 'uploader_id': '@colethedj1894',
5974 'uploader': 'colethedj',
5975 },
5976 'playlist': [{
5977 'info_dict': {
5978 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
5979 'id': 'BaW_jenozKc',
5980 '_type': 'url',
5981 'ie_key': 'Youtube',
5982 'duration': 10,
5983 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
5984 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
5985 'view_count': int,
5986 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
5987 'channel': 'Philipp Hagemeister',
5988 'uploader_id': '@PhilippHagemeister',
5989 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
5990 'uploader': 'Philipp Hagemeister',
5991 }
5992 }],
5993 'playlist_count': 1,
5994 'params': {'extract_flat': True},
5995 }, {
5996 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5997 'url': 'https://www.youtube.com/feed/recommended',
5998 'info_dict': {
5999 'id': 'recommended',
6000 'title': 'recommended',
6001 'tags': [],
6002 },
6003 'playlist_mincount': 50,
6004 'params': {
6005 'skip_download': True,
6006 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6007 },
6008 }, {
6009 'note': 'API Fallback: /videos tab, sorted by oldest first',
6010 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
6011 'info_dict': {
6012 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6013 'title': 'Cody\'sLab - Videos',
6014 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6015 'channel': 'Cody\'sLab',
6016 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6017 'tags': [],
6018 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6019 'channel_follower_count': int
6020 },
6021 'playlist_mincount': 650,
6022 'params': {
6023 'skip_download': True,
6024 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6025 },
6026 'skip': 'Query for sorting no longer works',
6027 }, {
6028 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6029 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6030 'info_dict': {
6031 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6032 'title': 'Uploads from Royalty Free Music - Topic',
6033 'modified_date': r're:\d{8}',
6034 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6035 'description': '',
6036 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6037 'tags': [],
6038 'channel': 'Royalty Free Music - Topic',
6039 'view_count': int,
6040 'availability': 'public',
6041 'uploader': 'Royalty Free Music - Topic',
6042 },
6043 'playlist_mincount': 101,
6044 'params': {
6045 'skip_download': True,
6046 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6047 },
6048 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6049 }, {
6050 'note': 'non-standard redirect to regional channel',
6051 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6052 'only_matching': True
6053 }, {
6054 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6055 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6056 'info_dict': {
6057 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6058 'modified_date': '20220407',
6059 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6060 'tags': [],
6061 'availability': 'unlisted',
6062 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6063 'channel': 'pukkandan',
6064 'description': 'Test for collaborative playlist',
6065 'title': 'yt-dlp test - collaborative playlist',
6066 'view_count': int,
6067 'uploader_url': 'https://www.youtube.com/@pukkandan',
6068 'uploader_id': '@pukkandan',
6069 'uploader': 'pukkandan',
6070 },
6071 'playlist_mincount': 2
6072 }, {
6073 'note': 'translated tab name',
6074 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6075 'info_dict': {
6076 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6077 'tags': [],
6078 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6079 'description': 'test description',
6080 'title': 'cole-dlp-test-acc - 再生リスト',
6081 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6082 'channel': 'cole-dlp-test-acc',
6083 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6084 'uploader_id': '@coletdjnz',
6085 'uploader': 'cole-dlp-test-acc',
6086 },
6087 'playlist_mincount': 1,
6088 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6089 'expected_warnings': ['Preferring "ja"'],
6090 }, {
6091 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6092 'note': 'preferred lang set with playlist with translated video titles',
6093 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6094 'info_dict': {
6095 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6096 'tags': [],
6097 'view_count': int,
6098 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6099 'channel': 'cole-dlp-test-acc',
6100 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6101 'description': 'test',
6102 'title': 'dlp test playlist',
6103 'availability': 'public',
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 # shorts audio pivot for 2GtVksBMYFM.
6113 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6114 'info_dict': {
6115 'id': 'sfv_audio_pivot',
6116 'title': 'sfv_audio_pivot',
6117 'tags': [],
6118 },
6119 'playlist_mincount': 50,
6120
6121 }, {
6122 # Channel with a real live tab (not to be mistaken with streams tab)
6123 # Do not treat like it should redirect to live stream
6124 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6125 'info_dict': {
6126 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6127 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6128 'tags': [],
6129 },
6130 'playlist_mincount': 20,
6131 }, {
6132 # Tab name is not the same as tab id
6133 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6134 'info_dict': {
6135 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6136 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6137 'tags': [],
6138 },
6139 'playlist_mincount': 8,
6140 }, {
6141 # Home tab id is literally home. Not to get mistaken with featured
6142 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6143 'info_dict': {
6144 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6145 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6146 'tags': [],
6147 },
6148 'playlist_mincount': 8,
6149 }, {
6150 # Should get three playlists for videos, shorts and streams tabs
6151 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6152 'info_dict': {
6153 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6154 'title': 'Polka Ch. 尾丸ポルカ',
6155 'channel_follower_count': int,
6156 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6157 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6158 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6159 'channel': 'Polka Ch. 尾丸ポルカ',
6160 'tags': 'count:35',
6161 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6162 'uploader': 'Polka Ch. 尾丸ポルカ',
6163 'uploader_id': '@OmaruPolka',
6164 },
6165 'playlist_count': 3,
6166 }, {
6167 # Shorts tab with channel with handle
6168 # TODO: fix channel description
6169 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6170 'info_dict': {
6171 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6172 'title': 'Not Just Bikes - Shorts',
6173 'tags': 'count:12',
6174 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6175 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6176 'channel_follower_count': int,
6177 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6178 'channel': 'Not Just Bikes',
6179 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6180 'uploader': 'Not Just Bikes',
6181 'uploader_id': '@NotJustBikes',
6182 },
6183 'playlist_mincount': 10,
6184 }, {
6185 # Streams tab
6186 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6187 'info_dict': {
6188 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6189 'title': '中村悠一 - Live',
6190 'tags': 'count:7',
6191 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6192 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6193 'channel': '中村悠一',
6194 'channel_follower_count': int,
6195 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6196 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6197 'uploader_id': '@Yuichi-Nakamura',
6198 'uploader': '中村悠一',
6199 },
6200 'playlist_mincount': 60,
6201 }, {
6202 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6203 # See test_youtube_lists
6204 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6205 'only_matching': True,
6206 }, {
6207 # No uploads and no UCID given. Should fail with no uploads error
6208 # See test_youtube_lists
6209 'url': 'https://www.youtube.com/news',
6210 'only_matching': True
6211 }, {
6212 # No videos tab but has a shorts tab
6213 'url': 'https://www.youtube.com/c/TKFShorts',
6214 'info_dict': {
6215 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6216 'title': 'Shorts Break - Shorts',
6217 'tags': 'count:48',
6218 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6219 'channel': 'Shorts Break',
6220 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6221 'channel_follower_count': int,
6222 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6223 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6224 'uploader': 'Shorts Break',
6225 'uploader_id': '@ShortsBreak_Official',
6226 },
6227 'playlist_mincount': 30,
6228 }, {
6229 # Trending Now Tab. tab id is empty
6230 'url': 'https://www.youtube.com/feed/trending',
6231 'info_dict': {
6232 'id': 'trending',
6233 'title': 'trending - Now',
6234 'tags': [],
6235 },
6236 'playlist_mincount': 30,
6237 }, {
6238 # Trending Gaming Tab. tab id is empty
6239 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6240 'info_dict': {
6241 'id': 'trending',
6242 'title': 'trending - Gaming',
6243 'tags': [],
6244 },
6245 'playlist_mincount': 30,
6246 }, {
6247 # Shorts url result in shorts tab
6248 # TODO: Fix channel id extraction
6249 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6250 'info_dict': {
6251 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6252 'title': 'cole-dlp-test-acc - Shorts',
6253 'channel': 'cole-dlp-test-acc',
6254 'description': 'test description',
6255 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6256 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6257 'tags': [],
6258 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6259 'uploader_id': '@coletdjnz',
6260 'uploader': 'cole-dlp-test-acc',
6261 },
6262 'playlist': [{
6263 'info_dict': {
6264 # Channel data is not currently available for short renderers (as of 2023-03-01)
6265 '_type': 'url',
6266 'ie_key': 'Youtube',
6267 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6268 'id': 'sSM9J5YH_60',
6269 'title': 'SHORT short',
6270 'view_count': int,
6271 'thumbnails': list,
6272 }
6273 }],
6274 'params': {'extract_flat': True},
6275 }, {
6276 # Live video status should be extracted
6277 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6278 'info_dict': {
6279 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6280 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6281 'tags': []
6282 },
6283 'playlist': [{
6284 'info_dict': {
6285 '_type': 'url',
6286 'ie_key': 'Youtube',
6287 'url': 'startswith:https://www.youtube.com/watch?v=',
6288 'id': str,
6289 'title': str,
6290 'live_status': 'is_live',
6291 'channel_id': str,
6292 'channel_url': str,
6293 'concurrent_view_count': int,
6294 'channel': str,
6295 'uploader': str,
6296 'uploader_url': str,
6297 'uploader_id': str,
6298 'channel_is_verified': bool, # this will keep changing
6299 }
6300 }],
6301 'params': {'extract_flat': True, 'playlist_items': '1'},
6302 'playlist_mincount': 1
6303 }, {
6304 # Channel renderer metadata. Contains number of videos on the channel
6305 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6306 'info_dict': {
6307 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6308 'title': 'cole-dlp-test-acc - Channels',
6309 'channel': 'cole-dlp-test-acc',
6310 'description': 'test description',
6311 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6312 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6313 'tags': [],
6314 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6315 'uploader_id': '@coletdjnz',
6316 'uploader': 'cole-dlp-test-acc',
6317 },
6318 'playlist': [{
6319 'info_dict': {
6320 '_type': 'url',
6321 'ie_key': 'YoutubeTab',
6322 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6323 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6324 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6325 'title': 'PewDiePie',
6326 'channel': 'PewDiePie',
6327 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6328 'thumbnails': list,
6329 'channel_follower_count': int,
6330 'playlist_count': int,
6331 'uploader': 'PewDiePie',
6332 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6333 'uploader_id': '@PewDiePie',
6334 'channel_is_verified': True,
6335 }
6336 }],
6337 'params': {'extract_flat': True},
6338 }, {
6339 'url': 'https://www.youtube.com/@3blue1brown/about',
6340 'info_dict': {
6341 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6342 'tags': ['Mathematics'],
6343 'title': '3Blue1Brown - About',
6344 'channel_follower_count': int,
6345 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6346 'channel': '3Blue1Brown',
6347 'view_count': int,
6348 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6349 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6350 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6351 'uploader_id': '@3blue1brown',
6352 'uploader': '3Blue1Brown',
6353 'channel_is_verified': True,
6354 },
6355 'playlist_count': 0,
6356 }, {
6357 # Podcasts tab, with rich entry playlistRenderers
6358 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6359 'info_dict': {
6360 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6361 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6362 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6363 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6364 'title': '99 Percent Invisible - Podcasts',
6365 'uploader': '99 Percent Invisible',
6366 'channel_follower_count': int,
6367 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6368 'tags': [],
6369 'channel': '99 Percent Invisible',
6370 'uploader_id': '@99percentinvisiblepodcast',
6371 },
6372 'playlist_count': 1,
6373 }, {
6374 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6375 'url': 'https://www.youtube.com/@AHimitsu/releases',
6376 'info_dict': {
6377 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6378 'channel': 'A Himitsu',
6379 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6380 'title': 'A Himitsu - Releases',
6381 'uploader_id': '@AHimitsu',
6382 'uploader': 'A Himitsu',
6383 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6384 'tags': 'count:16',
6385 'description': 'I make music',
6386 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6387 'channel_follower_count': int,
6388 'channel_is_verified': True,
6389 },
6390 'playlist_mincount': 10,
6391 }, {
6392 # Playlist with only shorts, shown as reel renderers
6393 # FIXME: future: YouTube currently doesn't give continuation for this,
6394 # may do in future.
6395 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
6396 'info_dict': {
6397 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
6398 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
6399 'view_count': int,
6400 'uploader_id': '@BangyShorts',
6401 'description': '',
6402 'uploader_url': 'https://www.youtube.com/@BangyShorts',
6403 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
6404 'channel': 'Bangy Shorts',
6405 'uploader': 'Bangy Shorts',
6406 'tags': [],
6407 'availability': 'public',
6408 'modified_date': '20230626',
6409 'title': 'Uploads from Bangy Shorts',
6410 },
6411 'playlist_mincount': 100,
6412 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6413 }]
6414
6415 @classmethod
6416 def suitable(cls, url):
6417 return False if YoutubeIE.suitable(url) else super().suitable(url)
6418
6419 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6420
6421 def _get_url_mobj(self, url):
6422 mobj = self._URL_RE.match(url).groupdict()
6423 mobj.update((k, '') for k, v in mobj.items() if v is None)
6424 return mobj
6425
6426 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6427 tab_name = (tab.get('title') or '').lower()
6428 tab_url = urljoin(base_url, traverse_obj(
6429 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6430
6431 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6432 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6433 if tab_id:
6434 return {
6435 'TAB_ID_SPONSORSHIPS': 'membership',
6436 }.get(tab_id, tab_id), tab_name
6437
6438 # Fallback to tab name if we cannot get the tab id.
6439 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6440 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6441 if tab_name:
6442 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6443 return {
6444 'home': 'featured',
6445 'live': 'streams',
6446 }.get(tab_name, tab_name), tab_name
6447
6448 def _has_tab(self, tabs, tab_id):
6449 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6450
6451 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6452 def _real_extract(self, url, smuggled_data):
6453 item_id = self._match_id(url)
6454 url = urllib.parse.urlunparse(
6455 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6456 compat_opts = self.get_param('compat_opts', [])
6457
6458 mobj = self._get_url_mobj(url)
6459 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6460 if is_channel and smuggled_data.get('is_music_url'):
6461 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6462 return self.url_result(
6463 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6464 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6465 mdata = self._extract_tab_endpoint(
6466 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6467 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6468 get_all=False, expected_type=str)
6469 if not murl:
6470 raise ExtractorError('Failed to resolve album to playlist')
6471 return self.url_result(murl, YoutubeTabIE)
6472 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6473 return self.url_result(
6474 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6475
6476 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6477 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6478 url = f'{pre}/videos{post}'
6479 if smuggled_data.get('is_music_url'):
6480 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6481
6482 # Handle both video/playlist URLs
6483 qs = parse_qs(url)
6484 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6485 if not video_id and mobj['not_channel'].startswith('watch'):
6486 if not playlist_id:
6487 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6488 raise ExtractorError('A video URL was given without video ID', expected=True)
6489 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6490 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6491 return self.url_result(
6492 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6493
6494 if not self._yes_playlist(playlist_id, video_id):
6495 return self.url_result(
6496 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6497
6498 data, ytcfg = self._extract_data(url, display_id)
6499
6500 # YouTube may provide a non-standard redirect to the regional channel
6501 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6502 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6503 redirect_url = traverse_obj(
6504 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6505 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6506 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6507 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6508 return self.url_result(redirect_url, YoutubeTabIE)
6509
6510 tabs, extra_tabs = self._extract_tab_renderers(data), []
6511 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6512 selected_tab = self._extract_selected_tab(tabs)
6513 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6514 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6515
6516 if not original_tab_id and selected_tab_name:
6517 self.to_screen('Downloading all uploads of the channel. '
6518 'To download only the videos in a specific tab, pass the tab\'s URL')
6519 if self._has_tab(tabs, 'streams'):
6520 extra_tabs.append(''.join((pre, '/streams', post)))
6521 if self._has_tab(tabs, 'shorts'):
6522 extra_tabs.append(''.join((pre, '/shorts', post)))
6523 # XXX: Members-only tab should also be extracted
6524
6525 if not extra_tabs and selected_tab_id != 'videos':
6526 # Channel does not have streams, shorts or videos tabs
6527 if item_id[:2] != 'UC':
6528 raise ExtractorError('This channel has no uploads', expected=True)
6529
6530 # Topic channels don't have /videos. Use the equivalent playlist instead
6531 pl_id = f'UU{item_id[2:]}'
6532 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6533 try:
6534 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6535 except ExtractorError:
6536 raise ExtractorError('This channel has no uploads', expected=True)
6537 else:
6538 item_id, url = pl_id, pl_url
6539 self.to_screen(
6540 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6541
6542 elif extra_tabs and selected_tab_id != 'videos':
6543 # When there are shorts/live tabs but not videos tab
6544 url, data = f'{pre}{post}', None
6545
6546 elif (original_tab_id or 'videos') != selected_tab_id:
6547 if original_tab_id == 'live':
6548 # Live tab should have redirected to the video
6549 # Except in the case the channel has an actual live tab
6550 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6551 raise UserNotLive(video_id=item_id)
6552 elif selected_tab_name:
6553 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6554
6555 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6556 url = f'{pre}{post}'
6557
6558 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6559 if 'no-youtube-unavailable-videos' not in compat_opts:
6560 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6561 self._extract_and_report_alerts(data, only_once=True)
6562
6563 tabs, entries = self._extract_tab_renderers(data), []
6564 if tabs:
6565 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6566 entries[0].update({
6567 'extractor_key': YoutubeTabIE.ie_key(),
6568 'extractor': YoutubeTabIE.IE_NAME,
6569 'webpage_url': url,
6570 })
6571 if self.get_param('playlist_items') == '0':
6572 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6573 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6574 entries.extend(map(self._real_extract, extra_tabs))
6575
6576 if len(entries) == 1:
6577 return entries[0]
6578 elif entries:
6579 metadata = self._extract_metadata_from_tabs(item_id, data)
6580 uploads_url = 'the Uploads (UU) playlist URL'
6581 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6582 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6583 self.to_screen(
6584 'Downloading as multiple playlists, separated by tabs. '
6585 f'To download as a single playlist instead, pass {uploads_url}')
6586 return self.playlist_result(entries, item_id, **metadata)
6587
6588 # Inline playlist
6589 playlist = traverse_obj(
6590 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6591 if playlist:
6592 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6593
6594 video_id = traverse_obj(
6595 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6596 if video_id:
6597 if tab != '/live': # live tab is expected to redirect to video
6598 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6599 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6600
6601 raise ExtractorError('Unable to recognize tab page')
6602
6603
6604 class YoutubePlaylistIE(InfoExtractor):
6605 IE_DESC = 'YouTube playlists'
6606 _VALID_URL = r'''(?x)(?:
6607 (?:https?://)?
6608 (?:\w+\.)?
6609 (?:
6610 (?:
6611 youtube(?:kids)?\.com|
6612 %(invidious)s
6613 )
6614 /.*?\?.*?\blist=
6615 )?
6616 (?P<id>%(playlist_id)s)
6617 )''' % {
6618 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6619 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6620 }
6621 IE_NAME = 'youtube:playlist'
6622 _TESTS = [{
6623 'note': 'issue #673',
6624 'url': 'PLBB231211A4F62143',
6625 'info_dict': {
6626 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6627 'id': 'PLBB231211A4F62143',
6628 'uploader': 'Wickman',
6629 'uploader_id': '@WickmanVT',
6630 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6631 'view_count': int,
6632 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6633 'modified_date': r're:\d{8}',
6634 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6635 'channel': 'Wickman',
6636 'tags': [],
6637 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6638 'availability': 'public',
6639 },
6640 'playlist_mincount': 29,
6641 }, {
6642 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6643 'info_dict': {
6644 'title': 'YDL_safe_search',
6645 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6646 },
6647 'playlist_count': 2,
6648 'skip': 'This playlist is private',
6649 }, {
6650 'note': 'embedded',
6651 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6652 'playlist_count': 4,
6653 'info_dict': {
6654 'title': 'JODA15',
6655 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6656 'uploader': 'milan',
6657 'uploader_id': '@milan5503',
6658 'description': '',
6659 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6660 'tags': [],
6661 'modified_date': '20140919',
6662 'view_count': int,
6663 'channel': 'milan',
6664 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6665 'uploader_url': 'https://www.youtube.com/@milan5503',
6666 'availability': 'public',
6667 },
6668 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6669 }, {
6670 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6671 'playlist_mincount': 455,
6672 'info_dict': {
6673 'title': '2018 Chinese New Singles (11/6 updated)',
6674 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6675 'uploader': 'LBK',
6676 'uploader_id': '@music_king',
6677 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6678 'channel': 'LBK',
6679 'view_count': int,
6680 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6681 'tags': [],
6682 'uploader_url': 'https://www.youtube.com/@music_king',
6683 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6684 'modified_date': r're:\d{8}',
6685 'availability': 'public',
6686 },
6687 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6688 }, {
6689 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6690 'only_matching': True,
6691 }, {
6692 # music album playlist
6693 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6694 'only_matching': True,
6695 }]
6696
6697 @classmethod
6698 def suitable(cls, url):
6699 if YoutubeTabIE.suitable(url):
6700 return False
6701 from ..utils import parse_qs
6702 qs = parse_qs(url)
6703 if qs.get('v', [None])[0]:
6704 return False
6705 return super().suitable(url)
6706
6707 def _real_extract(self, url):
6708 playlist_id = self._match_id(url)
6709 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6710 url = update_url_query(
6711 'https://www.youtube.com/playlist',
6712 parse_qs(url) or {'list': playlist_id})
6713 if is_music_url:
6714 url = smuggle_url(url, {'is_music_url': True})
6715 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6716
6717
6718 class YoutubeYtBeIE(InfoExtractor):
6719 IE_DESC = 'youtu.be'
6720 _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}
6721 _TESTS = [{
6722 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6723 'info_dict': {
6724 'id': 'yeWKywCrFtk',
6725 'ext': 'mp4',
6726 'title': 'Small Scale Baler and Braiding Rugs',
6727 'uploader': 'Backus-Page House Museum',
6728 'uploader_id': '@backuspagemuseum',
6729 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6730 'upload_date': '20161008',
6731 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6732 'categories': ['Nonprofits & Activism'],
6733 'tags': list,
6734 'like_count': int,
6735 'age_limit': 0,
6736 'playable_in_embed': True,
6737 'thumbnail': r're:^https?://.*\.webp',
6738 'channel': 'Backus-Page House Museum',
6739 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6740 'live_status': 'not_live',
6741 'view_count': int,
6742 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6743 'availability': 'public',
6744 'duration': 59,
6745 'comment_count': int,
6746 'channel_follower_count': int
6747 },
6748 'params': {
6749 'noplaylist': True,
6750 'skip_download': True,
6751 },
6752 }, {
6753 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6754 'only_matching': True,
6755 }]
6756
6757 def _real_extract(self, url):
6758 mobj = self._match_valid_url(url)
6759 video_id = mobj.group('id')
6760 playlist_id = mobj.group('playlist_id')
6761 return self.url_result(
6762 update_url_query('https://www.youtube.com/watch', {
6763 'v': video_id,
6764 'list': playlist_id,
6765 'feature': 'youtu.be',
6766 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6767
6768
6769 class YoutubeLivestreamEmbedIE(InfoExtractor):
6770 IE_DESC = 'YouTube livestream embeds'
6771 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6772 _TESTS = [{
6773 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6774 'only_matching': True,
6775 }]
6776
6777 def _real_extract(self, url):
6778 channel_id = self._match_id(url)
6779 return self.url_result(
6780 f'https://www.youtube.com/channel/{channel_id}/live',
6781 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6782
6783
6784 class YoutubeYtUserIE(InfoExtractor):
6785 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6786 IE_NAME = 'youtube:user'
6787 _VALID_URL = r'ytuser:(?P<id>.+)'
6788 _TESTS = [{
6789 'url': 'ytuser:phihag',
6790 'only_matching': True,
6791 }]
6792
6793 def _real_extract(self, url):
6794 user_id = self._match_id(url)
6795 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6796
6797
6798 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6799 IE_NAME = 'youtube:favorites'
6800 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6801 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6802 _LOGIN_REQUIRED = True
6803 _TESTS = [{
6804 'url': ':ytfav',
6805 'only_matching': True,
6806 }, {
6807 'url': ':ytfavorites',
6808 'only_matching': True,
6809 }]
6810
6811 def _real_extract(self, url):
6812 return self.url_result(
6813 'https://www.youtube.com/playlist?list=LL',
6814 ie=YoutubeTabIE.ie_key())
6815
6816
6817 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6818 IE_NAME = 'youtube:notif'
6819 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6820 _VALID_URL = r':ytnotif(?:ication)?s?'
6821 _LOGIN_REQUIRED = True
6822 _TESTS = [{
6823 'url': ':ytnotif',
6824 'only_matching': True,
6825 }, {
6826 'url': ':ytnotifications',
6827 'only_matching': True,
6828 }]
6829
6830 def _extract_notification_menu(self, response, continuation_list):
6831 notification_list = traverse_obj(
6832 response,
6833 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6834 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6835 expected_type=list) or []
6836 continuation_list[0] = None
6837 for item in notification_list:
6838 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6839 if entry:
6840 yield entry
6841 continuation = item.get('continuationItemRenderer')
6842 if continuation:
6843 continuation_list[0] = continuation
6844
6845 def _extract_notification_renderer(self, notification):
6846 video_id = traverse_obj(
6847 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6848 url = f'https://www.youtube.com/watch?v={video_id}'
6849 channel_id = None
6850 if not video_id:
6851 browse_ep = traverse_obj(
6852 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6853 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6854 post_id = self._search_regex(
6855 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6856 'post id', default=None)
6857 if not channel_id or not post_id:
6858 return
6859 # The direct /post url redirects to this in the browser
6860 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6861
6862 channel = traverse_obj(
6863 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6864 expected_type=str)
6865 notification_title = self._get_text(notification, 'shortMessage')
6866 if notification_title:
6867 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6868 # TODO: handle recommended videos
6869 title = self._search_regex(
6870 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6871 'video title', default=None)
6872 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6873 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6874 else None)
6875 return {
6876 '_type': 'url',
6877 'url': url,
6878 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6879 'video_id': video_id,
6880 'title': title,
6881 'channel_id': channel_id,
6882 'channel': channel,
6883 'uploader': channel,
6884 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6885 'timestamp': timestamp,
6886 }
6887
6888 def _notification_menu_entries(self, ytcfg):
6889 continuation_list = [None]
6890 response = None
6891 for page in itertools.count(1):
6892 ctoken = traverse_obj(
6893 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6894 response = self._extract_response(
6895 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6896 ep='notification/get_notification_menu', check_get_keys='actions',
6897 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6898 yield from self._extract_notification_menu(response, continuation_list)
6899 if not continuation_list[0]:
6900 break
6901
6902 def _real_extract(self, url):
6903 display_id = 'notifications'
6904 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6905 self._report_playlist_authcheck(ytcfg)
6906 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6907
6908
6909 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6910 IE_DESC = 'YouTube search'
6911 IE_NAME = 'youtube:search'
6912 _SEARCH_KEY = 'ytsearch'
6913 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6914 _TESTS = [{
6915 'url': 'ytsearch5:youtube-dl test video',
6916 'playlist_count': 5,
6917 'info_dict': {
6918 'id': 'youtube-dl test video',
6919 'title': 'youtube-dl test video',
6920 }
6921 }]
6922
6923
6924 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6925 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6926 _SEARCH_KEY = 'ytsearchdate'
6927 IE_DESC = 'YouTube search, newest videos first'
6928 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6929 _TESTS = [{
6930 'url': 'ytsearchdate5:youtube-dl test video',
6931 'playlist_count': 5,
6932 'info_dict': {
6933 'id': 'youtube-dl test video',
6934 'title': 'youtube-dl test video',
6935 }
6936 }]
6937
6938
6939 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6940 IE_DESC = 'YouTube search URLs with sorting and filter support'
6941 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6942 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6943 _TESTS = [{
6944 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6945 'playlist_mincount': 5,
6946 'info_dict': {
6947 'id': 'youtube-dl test video',
6948 'title': 'youtube-dl test video',
6949 }
6950 }, {
6951 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6952 'playlist_mincount': 5,
6953 'info_dict': {
6954 'id': 'python',
6955 'title': 'python',
6956 }
6957 }, {
6958 'url': 'https://www.youtube.com/results?search_query=%23cats',
6959 'playlist_mincount': 1,
6960 'info_dict': {
6961 'id': '#cats',
6962 'title': '#cats',
6963 # The test suite does not have support for nested playlists
6964 # 'entries': [{
6965 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6966 # 'title': '#cats',
6967 # }],
6968 },
6969 }, {
6970 # Channel results
6971 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
6972 'info_dict': {
6973 'id': 'kurzgesagt',
6974 'title': 'kurzgesagt',
6975 },
6976 'playlist': [{
6977 'info_dict': {
6978 '_type': 'url',
6979 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6980 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6981 'ie_key': 'YoutubeTab',
6982 'channel': 'Kurzgesagt – In a Nutshell',
6983 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
6984 'title': 'Kurzgesagt – In a Nutshell',
6985 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6986 # No longer available for search as it is set to the handle.
6987 # 'playlist_count': int,
6988 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6989 'thumbnails': list,
6990 'uploader_id': '@kurzgesagt',
6991 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
6992 'uploader': 'Kurzgesagt – In a Nutshell',
6993 'channel_is_verified': True,
6994 'channel_follower_count': int,
6995 }
6996 }],
6997 'params': {'extract_flat': True, 'playlist_items': '1'},
6998 'playlist_mincount': 1,
6999 }, {
7000 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
7001 'only_matching': True,
7002 }]
7003
7004 def _real_extract(self, url):
7005 qs = parse_qs(url)
7006 query = (qs.get('search_query') or qs.get('q'))[0]
7007 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
7008
7009
7010 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
7011 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
7012 IE_NAME = 'youtube:music:search_url'
7013 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7014 _TESTS = [{
7015 'url': 'https://music.youtube.com/search?q=royalty+free+music',
7016 'playlist_count': 16,
7017 'info_dict': {
7018 'id': 'royalty free music',
7019 'title': 'royalty free music',
7020 }
7021 }, {
7022 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7023 'playlist_mincount': 30,
7024 'info_dict': {
7025 'id': 'royalty free music - songs',
7026 'title': 'royalty free music - songs',
7027 },
7028 'params': {'extract_flat': 'in_playlist'}
7029 }, {
7030 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7031 'playlist_mincount': 30,
7032 'info_dict': {
7033 'id': 'royalty free music - community playlists',
7034 'title': 'royalty free music - community playlists',
7035 },
7036 'params': {'extract_flat': 'in_playlist'}
7037 }]
7038
7039 _SECTIONS = {
7040 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7041 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7042 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7043 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7044 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7045 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7046 }
7047
7048 def _real_extract(self, url):
7049 qs = parse_qs(url)
7050 query = (qs.get('search_query') or qs.get('q'))[0]
7051 params = qs.get('sp', (None,))[0]
7052 if params:
7053 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7054 else:
7055 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
7056 params = self._SECTIONS.get(section)
7057 if not params:
7058 section = None
7059 title = join_nonempty(query, section, delim=' - ')
7060 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7061
7062
7063 class YoutubeFeedsInfoExtractor(InfoExtractor):
7064 """
7065 Base class for feed extractors
7066 Subclasses must re-define the _FEED_NAME property.
7067 """
7068 _LOGIN_REQUIRED = True
7069 _FEED_NAME = 'feeds'
7070
7071 def _real_initialize(self):
7072 YoutubeBaseInfoExtractor._check_login_required(self)
7073
7074 @classproperty
7075 def IE_NAME(self):
7076 return f'youtube:{self._FEED_NAME}'
7077
7078 def _real_extract(self, url):
7079 return self.url_result(
7080 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7081
7082
7083 class YoutubeWatchLaterIE(InfoExtractor):
7084 IE_NAME = 'youtube:watchlater'
7085 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7086 _VALID_URL = r':ytwatchlater'
7087 _TESTS = [{
7088 'url': ':ytwatchlater',
7089 'only_matching': True,
7090 }]
7091
7092 def _real_extract(self, url):
7093 return self.url_result(
7094 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7095
7096
7097 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7098 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7099 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7100 _FEED_NAME = 'recommended'
7101 _LOGIN_REQUIRED = False
7102 _TESTS = [{
7103 'url': ':ytrec',
7104 'only_matching': True,
7105 }, {
7106 'url': ':ytrecommended',
7107 'only_matching': True,
7108 }, {
7109 'url': 'https://youtube.com',
7110 'only_matching': True,
7111 }]
7112
7113
7114 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7115 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7116 _VALID_URL = r':ytsub(?:scription)?s?'
7117 _FEED_NAME = 'subscriptions'
7118 _TESTS = [{
7119 'url': ':ytsubs',
7120 'only_matching': True,
7121 }, {
7122 'url': ':ytsubscriptions',
7123 'only_matching': True,
7124 }]
7125
7126
7127 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7128 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7129 _VALID_URL = r':ythis(?:tory)?'
7130 _FEED_NAME = 'history'
7131 _TESTS = [{
7132 'url': ':ythistory',
7133 'only_matching': True,
7134 }]
7135
7136
7137 class YoutubeShortsAudioPivotIE(InfoExtractor):
7138 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7139 IE_NAME = 'youtube:shorts:pivot:audio'
7140 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7141 _TESTS = [{
7142 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7143 'only_matching': True,
7144 }]
7145
7146 @staticmethod
7147 def _generate_audio_pivot_params(video_id):
7148 """
7149 Generates sfv_audio_pivot browse params for this video id
7150 """
7151 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7152 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7153
7154 def _real_extract(self, url):
7155 video_id = self._match_id(url)
7156 return self.url_result(
7157 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7158 ie=YoutubeTabIE)
7159
7160
7161 class YoutubeTruncatedURLIE(InfoExtractor):
7162 IE_NAME = 'youtube:truncated_url'
7163 IE_DESC = False # Do not list
7164 _VALID_URL = r'''(?x)
7165 (?:https?://)?
7166 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7167 (?:watch\?(?:
7168 feature=[a-z_]+|
7169 annotation_id=annotation_[^&]+|
7170 x-yt-cl=[0-9]+|
7171 hl=[^&]*|
7172 t=[0-9]+
7173 )?
7174 |
7175 attribution_link\?a=[^&]+
7176 )
7177 $
7178 '''
7179
7180 _TESTS = [{
7181 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7182 'only_matching': True,
7183 }, {
7184 'url': 'https://www.youtube.com/watch?',
7185 'only_matching': True,
7186 }, {
7187 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7188 'only_matching': True,
7189 }, {
7190 'url': 'https://www.youtube.com/watch?feature=foo',
7191 'only_matching': True,
7192 }, {
7193 'url': 'https://www.youtube.com/watch?hl=en-GB',
7194 'only_matching': True,
7195 }, {
7196 'url': 'https://www.youtube.com/watch?t=2372',
7197 'only_matching': True,
7198 }]
7199
7200 def _real_extract(self, url):
7201 raise ExtractorError(
7202 'Did you forget to quote the URL? Remember that & is a meta '
7203 'character in most shells, so you want to put the URL in quotes, '
7204 'like youtube-dl '
7205 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7206 ' or simply youtube-dl BaW_jenozKc .',
7207 expected=True)
7208
7209
7210 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7211 IE_NAME = 'youtube:clip'
7212 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7213 _TESTS = [{
7214 # FIXME: Other metadata should be extracted from the clip, not from the base video
7215 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7216 'info_dict': {
7217 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7218 'ext': 'mp4',
7219 'section_start': 29.0,
7220 'section_end': 39.7,
7221 'duration': 10.7,
7222 'age_limit': 0,
7223 'availability': 'public',
7224 'categories': ['Gaming'],
7225 'channel': 'Scott The Woz',
7226 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7227 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7228 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7229 'like_count': int,
7230 'playable_in_embed': True,
7231 'tags': 'count:17',
7232 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7233 'title': 'Mobile Games on Console - Scott The Woz',
7234 'upload_date': '20210920',
7235 'uploader': 'Scott The Woz',
7236 'uploader_id': '@ScottTheWoz',
7237 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7238 'view_count': int,
7239 'live_status': 'not_live',
7240 'channel_follower_count': int,
7241 'chapters': 'count:20',
7242 'comment_count': int,
7243 'heatmap': 'count:100',
7244 }
7245 }]
7246
7247 def _real_extract(self, url):
7248 clip_id = self._match_id(url)
7249 _, data = self._extract_webpage(url, clip_id)
7250
7251 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7252 if not video_id:
7253 raise ExtractorError('Unable to find video ID')
7254
7255 clip_data = traverse_obj(data, (
7256 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7257 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7258 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7259 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7260
7261 return {
7262 '_type': 'url_transparent',
7263 'url': f'https://www.youtube.com/watch?v={video_id}',
7264 'ie_key': YoutubeIE.ie_key(),
7265 'id': clip_id,
7266 'section_start': int(clip_data['startTimeMs']) / 1000,
7267 'section_end': int(clip_data['endTimeMs']) / 1000,
7268 }
7269
7270
7271 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7272 IE_NAME = 'youtube:consent'
7273 IE_DESC = False # Do not list
7274 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7275 _TESTS = [{
7276 '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',
7277 'info_dict': {
7278 'id': 'qVv6vCqciTM',
7279 'ext': 'mp4',
7280 'age_limit': 0,
7281 'uploader_id': '@sana_natori',
7282 'comment_count': int,
7283 'chapters': 'count:13',
7284 'upload_date': '20221223',
7285 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7286 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7287 'uploader_url': 'https://www.youtube.com/@sana_natori',
7288 'like_count': int,
7289 'release_date': '20221223',
7290 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7291 '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】',
7292 'view_count': int,
7293 'playable_in_embed': True,
7294 'duration': 4438,
7295 'availability': 'public',
7296 'channel_follower_count': int,
7297 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7298 'categories': ['Entertainment'],
7299 'live_status': 'was_live',
7300 'release_timestamp': 1671793345,
7301 'channel': 'さなちゃんねる',
7302 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7303 'uploader': 'さなちゃんねる',
7304 'channel_is_verified': True,
7305 'heatmap': 'count:100',
7306 },
7307 'add_ie': ['Youtube'],
7308 'params': {'skip_download': 'Youtube'},
7309 }]
7310
7311 def _real_extract(self, url):
7312 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7313 if not redirect_url:
7314 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7315 return self.url_result(redirect_url)
7316
7317
7318 class YoutubeTruncatedIDIE(InfoExtractor):
7319 IE_NAME = 'youtube:truncated_id'
7320 IE_DESC = False # Do not list
7321 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7322
7323 _TESTS = [{
7324 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7325 'only_matching': True,
7326 }]
7327
7328 def _real_extract(self, url):
7329 video_id = self._match_id(url)
7330 raise ExtractorError(
7331 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7332 expected=True)