]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[cleanup] Misc (#8338)
[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 raise_for_incomplete = bool(self._configuration_arg('raise_incomplete_data', ie_key=YoutubeIE))
945 # Incomplete Data should be a warning by default when retries are exhausted, while other errors should be fatal.
946 icd_retries = iter(self.RetryManager(fatal=raise_for_incomplete))
947 icd_rm = next(icd_retries)
948 main_retries = iter(self.RetryManager())
949 main_rm = next(main_retries)
950 # Manual retry loop for multiple RetryManagers
951 # The proper RetryManager MUST be advanced after an error
952 # and its result MUST be checked if the manager is non fatal
953 while True:
954 try:
955 response = self._call_api(
956 ep=ep, fatal=True, headers=headers,
957 video_id=item_id, query=query, note=note,
958 context=self._extract_context(ytcfg, default_client),
959 api_key=self._extract_api_key(ytcfg, default_client),
960 api_hostname=api_hostname, default_client=default_client)
961 except ExtractorError as e:
962 if not isinstance(e.cause, network_exceptions):
963 return self._error_or_warning(e, fatal=fatal)
964 elif not isinstance(e.cause, HTTPError):
965 main_rm.error = e
966 next(main_retries)
967 continue
968
969 first_bytes = e.cause.response.read(512)
970 if not is_html(first_bytes):
971 yt_error = try_get(
972 self._parse_json(
973 self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
974 lambda x: x['error']['message'], str)
975 if yt_error:
976 self._report_alerts([('ERROR', yt_error)], fatal=False)
977 # Downloading page may result in intermittent 5xx HTTP error
978 # Sometimes a 404 is also received. See: https://github.com/ytdl-org/youtube-dl/issues/28289
979 # We also want to catch all other network exceptions since errors in later pages can be troublesome
980 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
981 if e.cause.status not in (403, 429):
982 main_rm.error = e
983 next(main_retries)
984 continue
985 return self._error_or_warning(e, fatal=fatal)
986
987 try:
988 self._extract_and_report_alerts(response, only_once=True)
989 except ExtractorError as e:
990 # YouTube's servers may return errors we want to retry on in a 200 OK response
991 # See: https://github.com/yt-dlp/yt-dlp/issues/839
992 if 'unknown error' in e.msg.lower():
993 main_rm.error = e
994 next(main_retries)
995 continue
996 return self._error_or_warning(e, fatal=fatal)
997 # Youtube sometimes sends incomplete data
998 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
999 if not traverse_obj(response, *variadic(check_get_keys)):
1000 icd_rm.error = ExtractorError('Incomplete data received', expected=True)
1001 should_retry = next(icd_retries, None)
1002 if not should_retry:
1003 return None
1004 continue
1005
1006 return response
1007
1008 @staticmethod
1009 def is_music_url(url):
1010 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
1011
1012 def _extract_video(self, renderer):
1013 video_id = renderer.get('videoId')
1014
1015 reel_header_renderer = traverse_obj(renderer, (
1016 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1017 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1018
1019 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1020 description = self._get_text(renderer, 'descriptionSnippet')
1021
1022 duration = int_or_none(renderer.get('lengthSeconds'))
1023 if duration is None:
1024 duration = parse_duration(self._get_text(
1025 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1026 if duration is None:
1027 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1028 duration = parse_duration(self._search_regex(
1029 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1030 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1031 video_id, default=None, group='duration'))
1032
1033 channel_id = traverse_obj(
1034 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1035 expected_type=str, get_all=False)
1036 if not channel_id:
1037 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1038
1039 channel_id = self.ucid_or_none(channel_id)
1040
1041 overlay_style = traverse_obj(
1042 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1043 get_all=False, expected_type=str)
1044 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1045 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1046 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1047 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1048 expected_type=str)) or ''
1049 url = f'https://www.youtube.com/watch?v={video_id}'
1050 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1051 url = f'https://www.youtube.com/shorts/{video_id}'
1052
1053 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1054 or self._get_text(reel_header_renderer, 'timestampText') or '')
1055 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1056
1057 live_status = (
1058 'is_upcoming' if scheduled_timestamp is not None
1059 else 'was_live' if 'streamed' in time_text.lower()
1060 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1061 else None)
1062
1063 # videoInfo is a string like '50K views • 10 years ago'.
1064 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1065 view_count = (0 if 'no views' in view_count_text.lower()
1066 else self._get_count({'simpleText': view_count_text}))
1067 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1068
1069 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1070 or self._get_text(reel_header_renderer, 'channelTitleText'))
1071
1072 channel_handle = traverse_obj(renderer, (
1073 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1074 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1075 expected_type=self.handle_from_url, get_all=False)
1076 return {
1077 '_type': 'url',
1078 'ie_key': YoutubeIE.ie_key(),
1079 'id': video_id,
1080 'url': url,
1081 'title': title,
1082 'description': description,
1083 'duration': duration,
1084 'channel_id': channel_id,
1085 'channel': channel,
1086 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1087 'uploader': channel,
1088 'uploader_id': channel_handle,
1089 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1090 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1091 'timestamp': (self._parse_time_text(time_text)
1092 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1093 else None),
1094 'release_timestamp': scheduled_timestamp,
1095 'availability':
1096 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1097 else self._availability(
1098 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1099 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1100 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1101 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1102 view_count_field: view_count,
1103 'live_status': live_status,
1104 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None
1105 }
1106
1107
1108 class YoutubeIE(YoutubeBaseInfoExtractor):
1109 IE_DESC = 'YouTube'
1110 _VALID_URL = r"""(?x)^
1111 (
1112 (?:https?://|//) # http(s):// or protocol-independent URL
1113 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1114 (?:www\.)?deturl\.com/www\.youtube\.com|
1115 (?:www\.)?pwnyoutube\.com|
1116 (?:www\.)?hooktube\.com|
1117 (?:www\.)?yourepeat\.com|
1118 tube\.majestyc\.net|
1119 %(invidious)s|
1120 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1121 (?:.*?\#/)? # handle anchor (#/) redirect urls
1122 (?: # the various things that can precede the ID:
1123 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1124 |(?: # or the v= param in all its forms
1125 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1126 (?:\?|\#!?) # the params delimiter ? or # or #!
1127 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1128 v=
1129 )
1130 ))
1131 |(?:
1132 youtu\.be| # just youtu.be/xxxx
1133 vid\.plus| # or vid.plus/xxxx
1134 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1135 %(invidious)s
1136 )/
1137 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1138 )
1139 )? # all until now is optional -> you can pass the naked ID
1140 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1141 (?(1).+)? # if we found the ID, everything can follow
1142 (?:\#|$)""" % {
1143 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1144 }
1145 _EMBED_REGEX = [
1146 r'''(?x)
1147 (?:
1148 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1149 data-video-url=|
1150 <embed[^>]+?src=|
1151 embedSWF\(?:\s*|
1152 <object[^>]+data=|
1153 new\s+SWFObject\(
1154 )
1155 (["\'])
1156 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1157 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1158 \1''',
1159 # https://wordpress.org/plugins/lazy-load-for-videos/
1160 r'''(?xs)
1161 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1162 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1163 ]
1164 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1165
1166 _PLAYER_INFO_RE = (
1167 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1168 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
1169 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1170 )
1171 _formats = {
1172 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1173 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1174 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1175 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1176 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1177 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1178 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1179 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1180 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1181 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1182 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1183 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1184 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1185 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1186 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1187 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1188 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1189 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1190
1191
1192 # 3D videos
1193 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1194 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1195 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1196 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1197 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1198 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1199 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1200
1201 # Apple HTTP Live Streaming
1202 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1203 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1204 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1205 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1206 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1207 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1208 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1209 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1210
1211 # DASH mp4 video
1212 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1213 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1214 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1215 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1216 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1217 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1218 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1219 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1220 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1221 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1222 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1223 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1224
1225 # Dash mp4 audio
1226 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1227 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1228 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1229 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1230 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1231 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1232 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1233
1234 # Dash webm
1235 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1236 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1237 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1238 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1239 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1240 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1241 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1242 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1243 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1244 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1245 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1246 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1247 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1248 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1249 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1250 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1251 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1252 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1253 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1254 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1255 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1256 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1257
1258 # Dash webm audio
1259 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1260 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1261
1262 # Dash webm audio with opus inside
1263 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1264 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1265 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1266
1267 # RTMP (unnamed)
1268 '_rtmp': {'protocol': 'rtmp'},
1269
1270 # av01 video only formats sometimes served with "unknown" codecs
1271 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1272 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1273 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1274 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1275 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1276 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1277 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1278 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1279 }
1280 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1281
1282 _GEO_BYPASS = False
1283
1284 IE_NAME = 'youtube'
1285 _TESTS = [
1286 {
1287 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1288 'info_dict': {
1289 'id': 'BaW_jenozKc',
1290 'ext': 'mp4',
1291 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1292 'channel': 'Philipp Hagemeister',
1293 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1294 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1295 'upload_date': '20121002',
1296 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1297 'categories': ['Science & Technology'],
1298 'tags': ['youtube-dl'],
1299 'duration': 10,
1300 'view_count': int,
1301 'like_count': int,
1302 'availability': 'public',
1303 'playable_in_embed': True,
1304 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1305 'live_status': 'not_live',
1306 'age_limit': 0,
1307 'start_time': 1,
1308 'end_time': 9,
1309 'comment_count': int,
1310 'channel_follower_count': int,
1311 'uploader': 'Philipp Hagemeister',
1312 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1313 'uploader_id': '@PhilippHagemeister',
1314 'heatmap': 'count:100',
1315 }
1316 },
1317 {
1318 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1319 'note': 'Embed-only video (#1746)',
1320 'info_dict': {
1321 'id': 'yZIXLfi8CZQ',
1322 'ext': 'mp4',
1323 'upload_date': '20120608',
1324 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1325 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1326 'age_limit': 18,
1327 },
1328 'skip': 'Private video',
1329 },
1330 {
1331 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1332 'note': 'Use the first video ID in the URL',
1333 'info_dict': {
1334 'id': 'BaW_jenozKc',
1335 'ext': 'mp4',
1336 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1337 'channel': 'Philipp Hagemeister',
1338 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1339 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1340 'upload_date': '20121002',
1341 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1342 'categories': ['Science & Technology'],
1343 'tags': ['youtube-dl'],
1344 'duration': 10,
1345 'view_count': int,
1346 'like_count': int,
1347 'availability': 'public',
1348 'playable_in_embed': True,
1349 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1350 'live_status': 'not_live',
1351 'age_limit': 0,
1352 'comment_count': int,
1353 'channel_follower_count': int,
1354 'uploader': 'Philipp Hagemeister',
1355 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1356 'uploader_id': '@PhilippHagemeister',
1357 'heatmap': 'count:100',
1358 },
1359 'params': {
1360 'skip_download': True,
1361 },
1362 },
1363 {
1364 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1365 'note': '256k DASH audio (format 141) via DASH manifest',
1366 'info_dict': {
1367 'id': 'a9LDPn-MO4I',
1368 'ext': 'm4a',
1369 'upload_date': '20121002',
1370 'description': '',
1371 'title': 'UHDTV TEST 8K VIDEO.mp4'
1372 },
1373 'params': {
1374 'youtube_include_dash_manifest': True,
1375 'format': '141',
1376 },
1377 'skip': 'format 141 not served anymore',
1378 },
1379 # DASH manifest with encrypted signature
1380 {
1381 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1382 'info_dict': {
1383 'id': 'IB3lcPjvWLA',
1384 'ext': 'm4a',
1385 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1386 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1387 'duration': 244,
1388 'upload_date': '20131011',
1389 'abr': 129.495,
1390 'like_count': int,
1391 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1392 'playable_in_embed': True,
1393 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1394 'view_count': int,
1395 'track': 'The Spark',
1396 'live_status': 'not_live',
1397 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1398 'channel': 'Afrojack',
1399 'tags': 'count:19',
1400 'availability': 'public',
1401 'categories': ['Music'],
1402 'age_limit': 0,
1403 'alt_title': 'The Spark',
1404 'channel_follower_count': int,
1405 'uploader': 'Afrojack',
1406 'uploader_url': 'https://www.youtube.com/@Afrojack',
1407 'uploader_id': '@Afrojack',
1408 },
1409 'params': {
1410 'youtube_include_dash_manifest': True,
1411 'format': '141/bestaudio[ext=m4a]',
1412 },
1413 },
1414 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1415 {
1416 'note': 'Embed allowed age-gate video',
1417 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1418 'info_dict': {
1419 'id': 'HtVdAasjOgU',
1420 'ext': 'mp4',
1421 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1422 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1423 'duration': 142,
1424 'upload_date': '20140605',
1425 'age_limit': 18,
1426 'categories': ['Gaming'],
1427 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1428 'availability': 'needs_auth',
1429 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1430 'like_count': int,
1431 'channel': 'The Witcher',
1432 'live_status': 'not_live',
1433 'tags': 'count:17',
1434 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1435 'playable_in_embed': True,
1436 'view_count': int,
1437 'channel_follower_count': int,
1438 'uploader': 'The Witcher',
1439 'uploader_url': 'https://www.youtube.com/@thewitcher',
1440 'uploader_id': '@thewitcher',
1441 'comment_count': int,
1442 'channel_is_verified': True,
1443 'heatmap': 'count:100',
1444 },
1445 },
1446 {
1447 'note': 'Age-gate video with embed allowed in public site',
1448 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1449 'info_dict': {
1450 'id': 'HsUATh_Nc2U',
1451 'ext': 'mp4',
1452 'title': 'Godzilla 2 (Official Video)',
1453 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1454 'upload_date': '20200408',
1455 'age_limit': 18,
1456 'availability': 'needs_auth',
1457 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1458 'channel': 'FlyingKitty',
1459 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1460 'view_count': int,
1461 'categories': ['Entertainment'],
1462 'live_status': 'not_live',
1463 'tags': ['Flyingkitty', 'godzilla 2'],
1464 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1465 'like_count': int,
1466 'duration': 177,
1467 'playable_in_embed': True,
1468 'channel_follower_count': int,
1469 'uploader': 'FlyingKitty',
1470 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1471 'uploader_id': '@FlyingKitty900',
1472 'comment_count': int,
1473 'channel_is_verified': True,
1474 },
1475 },
1476 {
1477 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1478 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1479 'info_dict': {
1480 'id': 'Tq92D6wQ1mg',
1481 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1482 'ext': 'mp4',
1483 'upload_date': '20191228',
1484 'description': 'md5:17eccca93a786d51bc67646756894066',
1485 'age_limit': 18,
1486 'like_count': int,
1487 'availability': 'needs_auth',
1488 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1489 'view_count': int,
1490 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1491 'channel': 'Projekt Melody',
1492 'live_status': 'not_live',
1493 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1494 'playable_in_embed': True,
1495 'categories': ['Entertainment'],
1496 'duration': 106,
1497 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1498 'comment_count': int,
1499 'channel_follower_count': int,
1500 'uploader': 'Projekt Melody',
1501 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1502 'uploader_id': '@ProjektMelody',
1503 },
1504 },
1505 {
1506 'note': 'Non-Agegated non-embeddable video',
1507 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1508 'info_dict': {
1509 'id': 'MeJVWBSsPAY',
1510 'ext': 'mp4',
1511 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1512 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1513 'upload_date': '20130730',
1514 'track': 'Such mich find mich',
1515 'age_limit': 0,
1516 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1517 'like_count': int,
1518 'playable_in_embed': False,
1519 'creator': 'OOMPH!',
1520 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1521 'view_count': int,
1522 'alt_title': 'Such mich find mich',
1523 'duration': 210,
1524 'channel': 'Herr Lurik',
1525 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1526 'categories': ['Music'],
1527 'availability': 'public',
1528 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1529 'live_status': 'not_live',
1530 'artist': 'OOMPH!',
1531 'channel_follower_count': int,
1532 'uploader': 'Herr Lurik',
1533 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1534 'uploader_id': '@HerrLurik',
1535 },
1536 },
1537 {
1538 'note': 'Non-bypassable age-gated video',
1539 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1540 'only_matching': True,
1541 },
1542 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1543 # YouTube Red ad is not captured for creator
1544 {
1545 'url': '__2ABJjxzNo',
1546 'info_dict': {
1547 'id': '__2ABJjxzNo',
1548 'ext': 'mp4',
1549 'duration': 266,
1550 'upload_date': '20100430',
1551 'creator': 'deadmau5',
1552 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1553 'title': 'Deadmau5 - Some Chords (HD)',
1554 'alt_title': 'Some Chords',
1555 'availability': 'public',
1556 'tags': 'count:14',
1557 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1558 'view_count': int,
1559 'live_status': 'not_live',
1560 'channel': 'deadmau5',
1561 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1562 'like_count': int,
1563 'track': 'Some Chords',
1564 'artist': 'deadmau5',
1565 'playable_in_embed': True,
1566 'age_limit': 0,
1567 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1568 'categories': ['Music'],
1569 'album': 'Some Chords',
1570 'channel_follower_count': int,
1571 'uploader': 'deadmau5',
1572 'uploader_url': 'https://www.youtube.com/@deadmau5',
1573 'uploader_id': '@deadmau5',
1574 },
1575 'expected_warnings': [
1576 'DASH manifest missing',
1577 ]
1578 },
1579 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1580 {
1581 'url': 'lqQg6PlCWgI',
1582 'info_dict': {
1583 'id': 'lqQg6PlCWgI',
1584 'ext': 'mp4',
1585 'duration': 6085,
1586 'upload_date': '20150827',
1587 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1588 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1589 'like_count': int,
1590 'release_timestamp': 1343767800,
1591 'playable_in_embed': True,
1592 'categories': ['Sports'],
1593 'release_date': '20120731',
1594 'channel': 'Olympics',
1595 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1596 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1597 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1598 'age_limit': 0,
1599 'availability': 'public',
1600 'live_status': 'was_live',
1601 'view_count': int,
1602 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1603 'channel_follower_count': int,
1604 'uploader': 'Olympics',
1605 'uploader_url': 'https://www.youtube.com/@Olympics',
1606 'uploader_id': '@Olympics',
1607 'channel_is_verified': True,
1608 },
1609 'params': {
1610 'skip_download': 'requires avconv',
1611 }
1612 },
1613 # Non-square pixels
1614 {
1615 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1616 'info_dict': {
1617 'id': '_b-2C3KPAM0',
1618 'ext': 'mp4',
1619 'stretched_ratio': 16 / 9.,
1620 'duration': 85,
1621 'upload_date': '20110310',
1622 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
1623 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1624 'playable_in_embed': True,
1625 'channel': '孫ᄋᄅ',
1626 'age_limit': 0,
1627 'tags': 'count:11',
1628 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1629 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1630 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1631 'view_count': int,
1632 'categories': ['People & Blogs'],
1633 'like_count': int,
1634 'live_status': 'not_live',
1635 'availability': 'unlisted',
1636 'comment_count': int,
1637 'channel_follower_count': int,
1638 'uploader': '孫ᄋᄅ',
1639 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1640 'uploader_id': '@AllenMeow',
1641 },
1642 },
1643 # url_encoded_fmt_stream_map is empty string
1644 {
1645 'url': 'qEJwOuvDf7I',
1646 'info_dict': {
1647 'id': 'qEJwOuvDf7I',
1648 'ext': 'webm',
1649 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1650 'description': '',
1651 'upload_date': '20150404',
1652 },
1653 'params': {
1654 'skip_download': 'requires avconv',
1655 },
1656 'skip': 'This live event has ended.',
1657 },
1658 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1659 {
1660 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1661 'info_dict': {
1662 'id': 'FIl7x6_3R5Y',
1663 'ext': 'webm',
1664 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1665 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1666 'duration': 220,
1667 'upload_date': '20150625',
1668 'formats': 'mincount:31',
1669 },
1670 'skip': 'not actual anymore',
1671 },
1672 # DASH manifest with segment_list
1673 {
1674 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1675 'md5': '8ce563a1d667b599d21064e982ab9e31',
1676 'info_dict': {
1677 'id': 'CsmdDsKjzN8',
1678 'ext': 'mp4',
1679 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1680 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1681 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1682 },
1683 'params': {
1684 'youtube_include_dash_manifest': True,
1685 'format': '135', # bestvideo
1686 },
1687 'skip': 'This live event has ended.',
1688 },
1689 {
1690 # Multifeed videos (multiple cameras), URL can be of any Camera
1691 # TODO: fix multifeed titles
1692 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1693 'info_dict': {
1694 'id': 'zaPI8MvL8pg',
1695 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1696 'description': 'md5:563ccbc698b39298481ca3c571169519',
1697 },
1698 'playlist': [{
1699 'info_dict': {
1700 'id': 'j5yGuxZ8lLU',
1701 'ext': 'mp4',
1702 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1703 'description': 'md5:563ccbc698b39298481ca3c571169519',
1704 'duration': 10120,
1705 'channel_follower_count': int,
1706 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1707 'availability': 'public',
1708 'playable_in_embed': True,
1709 'upload_date': '20131105',
1710 'categories': ['Gaming'],
1711 'live_status': 'was_live',
1712 'tags': 'count:24',
1713 'release_timestamp': 1383701910,
1714 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1715 'comment_count': int,
1716 'age_limit': 0,
1717 'like_count': int,
1718 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1719 'channel': 'WiiLikeToPlay',
1720 'view_count': int,
1721 'release_date': '20131106',
1722 'uploader': 'WiiLikeToPlay',
1723 'uploader_id': '@WLTP',
1724 'uploader_url': 'https://www.youtube.com/@WLTP',
1725 },
1726 }, {
1727 'info_dict': {
1728 'id': 'zaPI8MvL8pg',
1729 'ext': 'mp4',
1730 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1731 'availability': 'public',
1732 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1733 'channel': 'WiiLikeToPlay',
1734 'channel_follower_count': int,
1735 'description': 'md5:563ccbc698b39298481ca3c571169519',
1736 'duration': 10108,
1737 'age_limit': 0,
1738 'like_count': int,
1739 'tags': 'count:24',
1740 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1741 'release_timestamp': 1383701915,
1742 'comment_count': int,
1743 'upload_date': '20131105',
1744 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1745 'release_date': '20131106',
1746 'playable_in_embed': True,
1747 'live_status': 'was_live',
1748 'categories': ['Gaming'],
1749 'view_count': int,
1750 'uploader': 'WiiLikeToPlay',
1751 'uploader_id': '@WLTP',
1752 'uploader_url': 'https://www.youtube.com/@WLTP',
1753 },
1754 }, {
1755 'info_dict': {
1756 'id': 'R7r3vfO7Hao',
1757 'ext': 'mp4',
1758 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1759 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1760 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1761 'like_count': int,
1762 'availability': 'public',
1763 'playable_in_embed': True,
1764 'upload_date': '20131105',
1765 'description': 'md5:563ccbc698b39298481ca3c571169519',
1766 'channel_follower_count': int,
1767 'tags': 'count:24',
1768 'release_date': '20131106',
1769 'comment_count': int,
1770 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1771 'channel': 'WiiLikeToPlay',
1772 'categories': ['Gaming'],
1773 'release_timestamp': 1383701914,
1774 'live_status': 'was_live',
1775 'age_limit': 0,
1776 'duration': 10128,
1777 'view_count': int,
1778 'uploader': 'WiiLikeToPlay',
1779 'uploader_id': '@WLTP',
1780 'uploader_url': 'https://www.youtube.com/@WLTP',
1781 },
1782 }],
1783 'params': {'skip_download': True},
1784 },
1785 {
1786 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1787 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1788 'info_dict': {
1789 'id': 'gVfLd0zydlo',
1790 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1791 },
1792 'playlist_count': 2,
1793 'skip': 'Not multifeed anymore',
1794 },
1795 {
1796 'url': 'https://vid.plus/FlRa-iH7PGw',
1797 'only_matching': True,
1798 },
1799 {
1800 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1801 'only_matching': True,
1802 },
1803 {
1804 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1805 # Also tests cut-off URL expansion in video description (see
1806 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1807 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1808 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1809 'info_dict': {
1810 'id': 'lsguqyKfVQg',
1811 'ext': 'mp4',
1812 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1813 'alt_title': 'Dark Walk',
1814 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1815 'duration': 133,
1816 'upload_date': '20151119',
1817 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1818 'track': 'Dark Walk',
1819 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1820 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1821 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1822 'categories': ['Film & Animation'],
1823 'view_count': int,
1824 'live_status': 'not_live',
1825 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1826 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1827 'tags': 'count:13',
1828 'availability': 'public',
1829 'channel': 'IronSoulElf',
1830 'playable_in_embed': True,
1831 'like_count': int,
1832 'age_limit': 0,
1833 'channel_follower_count': int
1834 },
1835 'params': {
1836 'skip_download': True,
1837 },
1838 },
1839 {
1840 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1841 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1842 'only_matching': True,
1843 },
1844 {
1845 # Video with yt:stretch=17:0
1846 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1847 'info_dict': {
1848 'id': 'Q39EVAstoRM',
1849 'ext': 'mp4',
1850 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1851 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1852 'upload_date': '20151107',
1853 },
1854 'params': {
1855 'skip_download': True,
1856 },
1857 'skip': 'This video does not exist.',
1858 },
1859 {
1860 # Video with incomplete 'yt:stretch=16:'
1861 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1862 'only_matching': True,
1863 },
1864 {
1865 # Video licensed under Creative Commons
1866 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1867 'info_dict': {
1868 'id': 'M4gD1WSo5mA',
1869 'ext': 'mp4',
1870 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1871 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1872 'duration': 721,
1873 'upload_date': '20150128',
1874 'license': 'Creative Commons Attribution license (reuse allowed)',
1875 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1876 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1877 'like_count': int,
1878 'age_limit': 0,
1879 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1880 'channel': 'The Berkman Klein Center for Internet & Society',
1881 'availability': 'public',
1882 'view_count': int,
1883 'categories': ['Education'],
1884 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1885 'live_status': 'not_live',
1886 'playable_in_embed': True,
1887 'channel_follower_count': int,
1888 'chapters': list,
1889 'uploader': 'The Berkman Klein Center for Internet & Society',
1890 'uploader_id': '@BKCHarvard',
1891 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1892 },
1893 'params': {
1894 'skip_download': True,
1895 },
1896 },
1897 {
1898 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1899 'info_dict': {
1900 'id': 'eQcmzGIKrzg',
1901 'ext': 'mp4',
1902 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1903 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1904 'duration': 4060,
1905 'upload_date': '20151120',
1906 'license': 'Creative Commons Attribution license (reuse allowed)',
1907 'playable_in_embed': True,
1908 'tags': 'count:12',
1909 'like_count': int,
1910 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1911 'age_limit': 0,
1912 'availability': 'public',
1913 'categories': ['News & Politics'],
1914 'channel': 'Bernie Sanders',
1915 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1916 'view_count': int,
1917 'live_status': 'not_live',
1918 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1919 'comment_count': int,
1920 'channel_follower_count': int,
1921 'chapters': list,
1922 'uploader': 'Bernie Sanders',
1923 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1924 'uploader_id': '@BernieSanders',
1925 'channel_is_verified': True,
1926 'heatmap': 'count:100',
1927 },
1928 'params': {
1929 'skip_download': True,
1930 },
1931 },
1932 {
1933 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1934 'only_matching': True,
1935 },
1936 {
1937 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1938 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1939 'only_matching': True,
1940 },
1941 {
1942 # Rental video preview
1943 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1944 'info_dict': {
1945 'id': 'uGpuVWrhIzE',
1946 'ext': 'mp4',
1947 'title': 'Piku - Trailer',
1948 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1949 'upload_date': '20150811',
1950 'license': 'Standard YouTube License',
1951 },
1952 'params': {
1953 'skip_download': True,
1954 },
1955 'skip': 'This video is not available.',
1956 },
1957 {
1958 # YouTube Red video with episode data
1959 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1960 'info_dict': {
1961 'id': 'iqKdEhx-dD4',
1962 'ext': 'mp4',
1963 'title': 'Isolation - Mind Field (Ep 1)',
1964 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1965 'duration': 2085,
1966 'upload_date': '20170118',
1967 'series': 'Mind Field',
1968 'season_number': 1,
1969 'episode_number': 1,
1970 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1971 'tags': 'count:12',
1972 'view_count': int,
1973 'availability': 'public',
1974 'age_limit': 0,
1975 'channel': 'Vsauce',
1976 'episode': 'Episode 1',
1977 'categories': ['Entertainment'],
1978 'season': 'Season 1',
1979 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1980 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1981 'like_count': int,
1982 'playable_in_embed': True,
1983 'live_status': 'not_live',
1984 'channel_follower_count': int,
1985 'uploader': 'Vsauce',
1986 'uploader_url': 'https://www.youtube.com/@Vsauce',
1987 'uploader_id': '@Vsauce',
1988 'comment_count': int,
1989 'channel_is_verified': True,
1990 },
1991 'params': {
1992 'skip_download': True,
1993 },
1994 'expected_warnings': [
1995 'Skipping DASH manifest',
1996 ],
1997 },
1998 {
1999 # The following content has been identified by the YouTube community
2000 # as inappropriate or offensive to some audiences.
2001 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
2002 'info_dict': {
2003 'id': '6SJNVb0GnPI',
2004 'ext': 'mp4',
2005 'title': 'Race Differences in Intelligence',
2006 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
2007 'duration': 965,
2008 'upload_date': '20140124',
2009 },
2010 'params': {
2011 'skip_download': True,
2012 },
2013 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
2014 },
2015 {
2016 # itag 212
2017 'url': '1t24XAntNCY',
2018 'only_matching': True,
2019 },
2020 {
2021 # geo restricted to JP
2022 'url': 'sJL6WA-aGkQ',
2023 'only_matching': True,
2024 },
2025 {
2026 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2027 'only_matching': True,
2028 },
2029 {
2030 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2031 'only_matching': True,
2032 },
2033 {
2034 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2035 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2036 'only_matching': True,
2037 },
2038 {
2039 # DRM protected
2040 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2041 'only_matching': True,
2042 },
2043 {
2044 # Video with unsupported adaptive stream type formats
2045 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2046 'info_dict': {
2047 'id': 'Z4Vy8R84T1U',
2048 'ext': 'mp4',
2049 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2050 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2051 'duration': 433,
2052 'upload_date': '20130923',
2053 'formats': 'maxcount:10',
2054 },
2055 'params': {
2056 'skip_download': True,
2057 'youtube_include_dash_manifest': False,
2058 },
2059 'skip': 'not actual anymore',
2060 },
2061 {
2062 # Youtube Music Auto-generated description
2063 # TODO: fix metadata extraction
2064 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2065 'info_dict': {
2066 'id': 'MgNrAu2pzNs',
2067 'ext': 'mp4',
2068 'title': 'Voyeur Girl',
2069 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2070 'upload_date': '20190312',
2071 'artist': 'Stephen',
2072 'track': 'Voyeur Girl',
2073 'album': 'it\'s too much love to know my dear',
2074 'release_date': '20190313',
2075 'release_year': 2019,
2076 'alt_title': 'Voyeur Girl',
2077 'view_count': int,
2078 'playable_in_embed': True,
2079 'like_count': int,
2080 'categories': ['Music'],
2081 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2082 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2083 'uploader': 'Stephen',
2084 'availability': 'public',
2085 'creator': 'Stephen',
2086 'duration': 169,
2087 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2088 'age_limit': 0,
2089 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2090 'tags': 'count:11',
2091 'live_status': 'not_live',
2092 'channel_follower_count': int
2093 },
2094 'params': {
2095 'skip_download': True,
2096 },
2097 },
2098 {
2099 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2100 'only_matching': True,
2101 },
2102 {
2103 # invalid -> valid video id redirection
2104 'url': 'DJztXj2GPfl',
2105 'info_dict': {
2106 'id': 'DJztXj2GPfk',
2107 'ext': 'mp4',
2108 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2109 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2110 'upload_date': '20090125',
2111 'artist': 'Panjabi MC',
2112 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2113 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2114 },
2115 'params': {
2116 'skip_download': True,
2117 },
2118 'skip': 'Video unavailable',
2119 },
2120 {
2121 # empty description results in an empty string
2122 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2123 'info_dict': {
2124 'id': 'x41yOUIvK2k',
2125 'ext': 'mp4',
2126 'title': 'IMG 3456',
2127 'description': '',
2128 'upload_date': '20170613',
2129 'view_count': int,
2130 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2131 'like_count': int,
2132 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2133 'tags': [],
2134 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2135 'availability': 'public',
2136 'age_limit': 0,
2137 'categories': ['Pets & Animals'],
2138 'duration': 7,
2139 'playable_in_embed': True,
2140 'live_status': 'not_live',
2141 'channel': 'l\'Or Vert asbl',
2142 'channel_follower_count': int,
2143 'uploader': 'l\'Or Vert asbl',
2144 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2145 'uploader_id': '@ElevageOrVert',
2146 },
2147 'params': {
2148 'skip_download': True,
2149 },
2150 },
2151 {
2152 # with '};' inside yt initial data (see [1])
2153 # see [2] for an example with '};' inside ytInitialPlayerResponse
2154 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2155 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2156 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2157 'info_dict': {
2158 'id': 'CHqg6qOn4no',
2159 'ext': 'mp4',
2160 'title': 'Part 77 Sort a list of simple types in c#',
2161 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2162 'upload_date': '20130831',
2163 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2164 'like_count': int,
2165 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2166 'live_status': 'not_live',
2167 'categories': ['Education'],
2168 'availability': 'public',
2169 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2170 'tags': 'count:12',
2171 'playable_in_embed': True,
2172 'age_limit': 0,
2173 'view_count': int,
2174 'duration': 522,
2175 'channel': 'kudvenkat',
2176 'comment_count': int,
2177 'channel_follower_count': int,
2178 'chapters': list,
2179 'uploader': 'kudvenkat',
2180 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2181 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2182 'channel_is_verified': True,
2183 'heatmap': 'count:100',
2184 },
2185 'params': {
2186 'skip_download': True,
2187 },
2188 },
2189 {
2190 # another example of '};' in ytInitialData
2191 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2192 'only_matching': True,
2193 },
2194 {
2195 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2196 'only_matching': True,
2197 },
2198 {
2199 # https://github.com/ytdl-org/youtube-dl/pull/28094
2200 'url': 'OtqTfy26tG0',
2201 'info_dict': {
2202 'id': 'OtqTfy26tG0',
2203 'ext': 'mp4',
2204 'title': 'Burn Out',
2205 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2206 'upload_date': '20141120',
2207 'artist': 'The Cinematic Orchestra',
2208 'track': 'Burn Out',
2209 'album': 'Every Day',
2210 'like_count': int,
2211 'live_status': 'not_live',
2212 'alt_title': 'Burn Out',
2213 'duration': 614,
2214 'age_limit': 0,
2215 'view_count': int,
2216 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2217 'creator': 'The Cinematic Orchestra',
2218 'channel': 'The Cinematic Orchestra',
2219 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2220 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2221 'availability': 'public',
2222 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2223 'categories': ['Music'],
2224 'playable_in_embed': True,
2225 'channel_follower_count': int,
2226 'uploader': 'The Cinematic Orchestra',
2227 'comment_count': int,
2228 },
2229 'params': {
2230 'skip_download': True,
2231 },
2232 },
2233 {
2234 # controversial video, only works with bpctr when authenticated with cookies
2235 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2236 'only_matching': True,
2237 },
2238 {
2239 # controversial video, requires bpctr/contentCheckOk
2240 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2241 'info_dict': {
2242 'id': 'SZJvDhaSDnc',
2243 'ext': 'mp4',
2244 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2245 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2246 'upload_date': '20140716',
2247 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2248 'duration': 170,
2249 'categories': ['News & Politics'],
2250 'view_count': int,
2251 'channel': 'CBS Mornings',
2252 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2253 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2254 'age_limit': 18,
2255 'availability': 'needs_auth',
2256 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2257 'like_count': int,
2258 'live_status': 'not_live',
2259 'playable_in_embed': True,
2260 'channel_follower_count': int,
2261 'uploader': 'CBS Mornings',
2262 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2263 'uploader_id': '@CBSMornings',
2264 'comment_count': int,
2265 'channel_is_verified': True,
2266 }
2267 },
2268 {
2269 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2270 'url': 'cBvYw8_A0vQ',
2271 'info_dict': {
2272 'id': 'cBvYw8_A0vQ',
2273 'ext': 'mp4',
2274 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2275 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2276 'upload_date': '20201120',
2277 'duration': 1456,
2278 'categories': ['Travel & Events'],
2279 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2280 'view_count': int,
2281 'channel': 'Walk around Japan',
2282 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2283 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2284 'age_limit': 0,
2285 'availability': 'public',
2286 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2287 'live_status': 'not_live',
2288 'playable_in_embed': True,
2289 'channel_follower_count': int,
2290 'uploader': 'Walk around Japan',
2291 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2292 'uploader_id': '@walkaroundjapan7124',
2293 },
2294 'params': {
2295 'skip_download': True,
2296 },
2297 }, {
2298 # Has multiple audio streams
2299 'url': 'WaOKSUlf4TM',
2300 'only_matching': True
2301 }, {
2302 # Requires Premium: has format 141 when requested using YTM url
2303 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2304 'only_matching': True
2305 }, {
2306 # multiple subtitles with same lang_code
2307 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2308 'only_matching': True,
2309 }, {
2310 # Force use android client fallback
2311 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2312 'info_dict': {
2313 'id': 'YOelRv7fMxY',
2314 'title': 'DIGGING A SECRET TUNNEL Part 1',
2315 'ext': '3gp',
2316 'upload_date': '20210624',
2317 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2318 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2319 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2320 'duration': 596,
2321 'categories': ['Entertainment'],
2322 'view_count': int,
2323 'channel': 'colinfurze',
2324 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2325 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2326 'age_limit': 0,
2327 'availability': 'public',
2328 'like_count': int,
2329 'live_status': 'not_live',
2330 'playable_in_embed': True,
2331 'channel_follower_count': int,
2332 'chapters': list,
2333 'uploader': 'colinfurze',
2334 'uploader_url': 'https://www.youtube.com/@colinfurze',
2335 'uploader_id': '@colinfurze',
2336 'comment_count': int,
2337 'channel_is_verified': True,
2338 'heatmap': 'count:100',
2339 },
2340 'params': {
2341 'format': '17', # 3gp format available on android
2342 'extractor_args': {'youtube': {'player_client': ['android']}},
2343 },
2344 },
2345 {
2346 # Skip download of additional client configs (remix client config in this case)
2347 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2348 'only_matching': True,
2349 'params': {
2350 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2351 },
2352 }, {
2353 # shorts
2354 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2355 'only_matching': True,
2356 }, {
2357 'note': 'Storyboards',
2358 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2359 'info_dict': {
2360 'id': '5KLPxDtMqe8',
2361 'ext': 'mhtml',
2362 'format_id': 'sb0',
2363 'title': 'Your Brain is Plastic',
2364 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2365 'upload_date': '20140324',
2366 'like_count': int,
2367 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2368 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2369 'view_count': int,
2370 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2371 'playable_in_embed': True,
2372 'tags': 'count:12',
2373 'availability': 'public',
2374 'channel': 'SciShow',
2375 'live_status': 'not_live',
2376 'duration': 248,
2377 'categories': ['Education'],
2378 'age_limit': 0,
2379 'channel_follower_count': int,
2380 'chapters': list,
2381 'uploader': 'SciShow',
2382 'uploader_url': 'https://www.youtube.com/@SciShow',
2383 'uploader_id': '@SciShow',
2384 'comment_count': int,
2385 'channel_is_verified': True,
2386 'heatmap': 'count:100',
2387 }, 'params': {'format': 'mhtml', 'skip_download': True}
2388 }, {
2389 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2390 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2391 'info_dict': {
2392 'id': '2NUZ8W2llS4',
2393 'ext': 'mp4',
2394 'title': 'The NP that test your phone performance 🙂',
2395 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2396 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2397 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2398 'duration': 21,
2399 'view_count': int,
2400 'age_limit': 0,
2401 'categories': ['Gaming'],
2402 'tags': 'count:23',
2403 'playable_in_embed': True,
2404 'live_status': 'not_live',
2405 'upload_date': '20220103',
2406 'like_count': int,
2407 'availability': 'public',
2408 'channel': 'Leon Nguyen',
2409 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2410 'comment_count': int,
2411 'channel_follower_count': int,
2412 'uploader': 'Leon Nguyen',
2413 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2414 'uploader_id': '@LeonNguyen',
2415 'heatmap': 'count:100',
2416 }
2417 }, {
2418 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2419 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2420 'info_dict': {
2421 'id': '2NUZ8W2llS4',
2422 'ext': 'mp4',
2423 'title': 'The NP that test your phone performance 🙂',
2424 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2425 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2426 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2427 'duration': 21,
2428 'view_count': int,
2429 'age_limit': 0,
2430 'categories': ['Gaming'],
2431 'tags': 'count:23',
2432 'playable_in_embed': True,
2433 'live_status': 'not_live',
2434 'upload_date': '20220102',
2435 'like_count': int,
2436 'availability': 'public',
2437 'channel': 'Leon Nguyen',
2438 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2439 'comment_count': int,
2440 'channel_follower_count': int,
2441 'uploader': 'Leon Nguyen',
2442 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2443 'uploader_id': '@LeonNguyen',
2444 'heatmap': 'count:100',
2445 },
2446 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2447 }, {
2448 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2449 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2450 'info_dict': {
2451 'id': 'mzZzzBU6lrM',
2452 'ext': 'mp4',
2453 'title': 'I Met GeorgeNotFound In Real Life...',
2454 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2455 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2456 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2457 'duration': 955,
2458 'view_count': int,
2459 'age_limit': 0,
2460 'categories': ['Entertainment'],
2461 'tags': 'count:26',
2462 'playable_in_embed': True,
2463 'live_status': 'not_live',
2464 'release_timestamp': 1641172509,
2465 'release_date': '20220103',
2466 'upload_date': '20220103',
2467 'like_count': int,
2468 'availability': 'public',
2469 'channel': 'Quackity',
2470 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2471 'channel_follower_count': int,
2472 'uploader': 'Quackity',
2473 'uploader_id': '@Quackity',
2474 'uploader_url': 'https://www.youtube.com/@Quackity',
2475 'comment_count': int,
2476 'channel_is_verified': True,
2477 'heatmap': 'count:100',
2478 }
2479 },
2480 { # continuous livestream. Microformat upload date should be preferred.
2481 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2482 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2483 'info_dict': {
2484 'id': 'kgx4WGK0oNU',
2485 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2486 'ext': 'mp4',
2487 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2488 'availability': 'public',
2489 'age_limit': 0,
2490 'release_timestamp': 1637975704,
2491 'upload_date': '20210619',
2492 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2493 'live_status': 'is_live',
2494 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2495 'channel': 'Abao in Tokyo',
2496 'channel_follower_count': int,
2497 'release_date': '20211127',
2498 'tags': 'count:39',
2499 'categories': ['People & Blogs'],
2500 'like_count': int,
2501 'view_count': int,
2502 'playable_in_embed': True,
2503 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2504 'concurrent_view_count': int,
2505 'uploader': 'Abao in Tokyo',
2506 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2507 'uploader_id': '@abaointokyo',
2508 },
2509 'params': {'skip_download': True}
2510 }, {
2511 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2512 'info_dict': {
2513 'id': 'tjjjtzRLHvA',
2514 'ext': 'mp4',
2515 'title': 'ハッシュタグ無し };if window.ytcsi',
2516 'upload_date': '20220323',
2517 'like_count': int,
2518 'availability': 'unlisted',
2519 'channel': 'Lesmiscore',
2520 'thumbnail': r're:^https?://.*\.jpg',
2521 'age_limit': 0,
2522 'categories': ['Music'],
2523 'view_count': int,
2524 'description': '',
2525 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2526 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2527 'live_status': 'not_live',
2528 'playable_in_embed': True,
2529 'channel_follower_count': int,
2530 'duration': 6,
2531 'tags': [],
2532 'uploader_id': '@lesmiscore',
2533 'uploader': 'Lesmiscore',
2534 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2535 }
2536 }, {
2537 # Prefer primary title+description language metadata by default
2538 # Do not prefer translated description if primary is empty
2539 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2540 'info_dict': {
2541 'id': 'el3E4MbxRqQ',
2542 'ext': 'mp4',
2543 'title': 'dlp test video 2 - primary sv no desc',
2544 'description': '',
2545 'channel': 'cole-dlp-test-acc',
2546 'tags': [],
2547 'view_count': int,
2548 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2549 'like_count': int,
2550 'playable_in_embed': True,
2551 'availability': 'unlisted',
2552 'thumbnail': r're:^https?://.*\.jpg',
2553 'age_limit': 0,
2554 'duration': 5,
2555 'live_status': 'not_live',
2556 'upload_date': '20220908',
2557 'categories': ['People & Blogs'],
2558 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2559 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2560 'uploader_id': '@coletdjnz',
2561 'uploader': 'cole-dlp-test-acc',
2562 },
2563 'params': {'skip_download': True}
2564 }, {
2565 # Extractor argument: prefer translated title+description
2566 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2567 'info_dict': {
2568 'id': 'gHKT4uU8Zng',
2569 'ext': 'mp4',
2570 'channel': 'cole-dlp-test-acc',
2571 'tags': [],
2572 'duration': 5,
2573 'live_status': 'not_live',
2574 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2575 'upload_date': '20220728',
2576 'view_count': int,
2577 'categories': ['People & Blogs'],
2578 'thumbnail': r're:^https?://.*\.jpg',
2579 'title': 'dlp test video title translated (fr)',
2580 'availability': 'public',
2581 'age_limit': 0,
2582 'description': 'dlp test video description translated (fr)',
2583 'playable_in_embed': True,
2584 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2585 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2586 'uploader_id': '@coletdjnz',
2587 'uploader': 'cole-dlp-test-acc',
2588 },
2589 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2590 'expected_warnings': [r'Preferring "fr" translated fields'],
2591 }, {
2592 'note': '6 channel audio',
2593 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2594 'only_matching': True,
2595 }, {
2596 'note': 'Multiple HLS formats with same itag',
2597 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2598 'info_dict': {
2599 'id': 'kX3nB4PpJko',
2600 'ext': 'mp4',
2601 'categories': ['Entertainment'],
2602 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2603 'live_status': 'not_live',
2604 'duration': 937,
2605 'channel_follower_count': int,
2606 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2607 'title': 'Last To Take Hand Off Jet, Keeps It!',
2608 'channel': 'MrBeast',
2609 'playable_in_embed': True,
2610 'view_count': int,
2611 'upload_date': '20221112',
2612 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2613 'age_limit': 0,
2614 'availability': 'public',
2615 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2616 'like_count': int,
2617 'tags': [],
2618 'uploader': 'MrBeast',
2619 'uploader_url': 'https://www.youtube.com/@MrBeast',
2620 'uploader_id': '@MrBeast',
2621 'comment_count': int,
2622 'channel_is_verified': True,
2623 'heatmap': 'count:100',
2624 },
2625 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2626 }, {
2627 'note': 'Audio formats with Dynamic Range Compression',
2628 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2629 'info_dict': {
2630 'id': 'Tq92D6wQ1mg',
2631 'ext': 'webm',
2632 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2633 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2634 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2635 'channel_follower_count': int,
2636 'description': 'md5:17eccca93a786d51bc67646756894066',
2637 'upload_date': '20191228',
2638 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2639 'playable_in_embed': True,
2640 'like_count': int,
2641 'categories': ['Entertainment'],
2642 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2643 'age_limit': 18,
2644 'channel': 'Projekt Melody',
2645 'view_count': int,
2646 'availability': 'needs_auth',
2647 'comment_count': int,
2648 'live_status': 'not_live',
2649 'duration': 106,
2650 'uploader': 'Projekt Melody',
2651 'uploader_id': '@ProjektMelody',
2652 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2653 },
2654 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2655 },
2656 {
2657 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2658 'info_dict': {
2659 'id': 'qVv6vCqciTM',
2660 'ext': 'mp4',
2661 'age_limit': 0,
2662 'comment_count': int,
2663 'chapters': 'count:13',
2664 'upload_date': '20221223',
2665 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2666 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2667 'like_count': int,
2668 'release_date': '20221223',
2669 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2670 'title': '【 #インターネット女クリスマス 】3Dで歌ってはしゃぐインターネットの女たち【月ノ美兎/名取さな】',
2671 'view_count': int,
2672 'playable_in_embed': True,
2673 'duration': 4438,
2674 'availability': 'public',
2675 'channel_follower_count': int,
2676 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2677 'categories': ['Entertainment'],
2678 'live_status': 'was_live',
2679 'release_timestamp': 1671793345,
2680 'channel': 'さなちゃんねる',
2681 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2682 'uploader': 'さなちゃんねる',
2683 'uploader_url': 'https://www.youtube.com/@sana_natori',
2684 'uploader_id': '@sana_natori',
2685 'channel_is_verified': True,
2686 'heatmap': 'count:100',
2687 },
2688 },
2689 {
2690 # Fallbacks when webpage and web client is unavailable
2691 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2692 'info_dict': {
2693 'id': 'wSSmNUl9Snw',
2694 'ext': 'mp4',
2695 # 'categories': ['Science & Technology'],
2696 'view_count': int,
2697 'chapters': 'count:2',
2698 'channel': 'Scott Manley',
2699 'like_count': int,
2700 'age_limit': 0,
2701 # 'availability': 'public',
2702 'channel_follower_count': int,
2703 'live_status': 'not_live',
2704 'upload_date': '20170831',
2705 'duration': 682,
2706 'tags': 'count:8',
2707 'uploader_url': 'https://www.youtube.com/@scottmanley',
2708 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2709 'uploader': 'Scott Manley',
2710 'uploader_id': '@scottmanley',
2711 'title': 'The Computer Hack That Saved Apollo 14',
2712 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2713 'thumbnail': r're:^https?://.*\.webp',
2714 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2715 'playable_in_embed': True,
2716 'comment_count': int,
2717 'channel_is_verified': True,
2718 'heatmap': 'count:100',
2719 },
2720 'params': {
2721 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2722 },
2723 },
2724 ]
2725
2726 _WEBPAGE_TESTS = [
2727 # YouTube <object> embed
2728 {
2729 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2730 'md5': '873c81d308b979f0e23ee7e620b312a3',
2731 'info_dict': {
2732 'id': 'msN87y-iEx0',
2733 'ext': 'mp4',
2734 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2735 'upload_date': '20080526',
2736 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2737 'age_limit': 0,
2738 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2739 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2740 'playable_in_embed': True,
2741 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2742 'like_count': int,
2743 'comment_count': int,
2744 'channel': 'Christopher Sykes',
2745 'live_status': 'not_live',
2746 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2747 'availability': 'public',
2748 'duration': 195,
2749 'view_count': int,
2750 'categories': ['Science & Technology'],
2751 'channel_follower_count': int,
2752 'uploader': 'Christopher Sykes',
2753 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2754 'uploader_id': '@ChristopherSykesDocumentaries',
2755 'heatmap': 'count:100',
2756 },
2757 'params': {
2758 'skip_download': True,
2759 }
2760 },
2761 ]
2762
2763 @classmethod
2764 def suitable(cls, url):
2765 from ..utils import parse_qs
2766
2767 qs = parse_qs(url)
2768 if qs.get('list', [None])[0]:
2769 return False
2770 return super().suitable(url)
2771
2772 def __init__(self, *args, **kwargs):
2773 super().__init__(*args, **kwargs)
2774 self._code_cache = {}
2775 self._player_cache = {}
2776
2777 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2778 lock = threading.Lock()
2779 start_time = time.time()
2780 formats = [f for f in formats if f.get('is_from_start')]
2781
2782 def refetch_manifest(format_id, delay):
2783 nonlocal formats, start_time, is_live
2784 if time.time() <= start_time + delay:
2785 return
2786
2787 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2788 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2789 microformats = traverse_obj(
2790 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2791 expected_type=dict)
2792 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2793 is_live = live_status == 'is_live'
2794 start_time = time.time()
2795
2796 def mpd_feed(format_id, delay):
2797 """
2798 @returns (manifest_url, manifest_stream_number, is_live) or None
2799 """
2800 for retry in self.RetryManager(fatal=False):
2801 with lock:
2802 refetch_manifest(format_id, delay)
2803
2804 f = next((f for f in formats if f['format_id'] == format_id), None)
2805 if not f:
2806 if not is_live:
2807 retry.error = f'{video_id}: Video is no longer live'
2808 else:
2809 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2810 continue
2811 return f['manifest_url'], f['manifest_stream_number'], is_live
2812 return None
2813
2814 for f in formats:
2815 f['is_live'] = is_live
2816 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2817 live_start_time, mpd_feed, not is_live and f.copy())
2818 if is_live:
2819 f['fragments'] = gen
2820 f['protocol'] = 'http_dash_segments_generator'
2821 else:
2822 f['fragments'] = LazyList(gen({}))
2823 del f['is_from_start']
2824
2825 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2826 FETCH_SPAN, MAX_DURATION = 5, 432000
2827
2828 mpd_url, stream_number, is_live = None, None, True
2829
2830 begin_index = 0
2831 download_start_time = ctx.get('start') or time.time()
2832
2833 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2834 if lack_early_segments:
2835 self.report_warning(bug_reports_message(
2836 'Starting download from the last 120 hours of the live stream since '
2837 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2838 lack_early_segments = True
2839
2840 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2841 fragments, fragment_base_url = None, None
2842
2843 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2844 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2845 # Obtain from MPD's maximum seq value
2846 old_mpd_url = mpd_url
2847 last_error = ctx.pop('last_error', None)
2848 expire_fast = immediate or last_error and isinstance(last_error, HTTPError) and last_error.status == 403
2849 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2850 or (mpd_url, stream_number, False))
2851 if not refresh_sequence:
2852 if expire_fast and not is_live:
2853 return False, last_seq
2854 elif old_mpd_url == mpd_url:
2855 return True, last_seq
2856 if manifestless_orig_fmt:
2857 fmt_info = manifestless_orig_fmt
2858 else:
2859 try:
2860 fmts, _ = self._extract_mpd_formats_and_subtitles(
2861 mpd_url, None, note=False, errnote=False, fatal=False)
2862 except ExtractorError:
2863 fmts = None
2864 if not fmts:
2865 no_fragment_score += 2
2866 return False, last_seq
2867 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2868 fragments = fmt_info['fragments']
2869 fragment_base_url = fmt_info['fragment_base_url']
2870 assert fragment_base_url
2871
2872 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2873 return True, _last_seq
2874
2875 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2876 while is_live:
2877 fetch_time = time.time()
2878 if no_fragment_score > 30:
2879 return
2880 if last_segment_url:
2881 # Obtain from "X-Head-Seqnum" header value from each segment
2882 try:
2883 urlh = self._request_webpage(
2884 last_segment_url, None, note=False, errnote=False, fatal=False)
2885 except ExtractorError:
2886 urlh = None
2887 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2888 if last_seq is None:
2889 no_fragment_score += 2
2890 last_segment_url = None
2891 continue
2892 else:
2893 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2894 no_fragment_score += 2
2895 if not should_continue:
2896 continue
2897
2898 if known_idx > last_seq:
2899 last_segment_url = None
2900 continue
2901
2902 last_seq += 1
2903
2904 if begin_index < 0 and known_idx < 0:
2905 # skip from the start when it's negative value
2906 known_idx = last_seq + begin_index
2907 if lack_early_segments:
2908 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2909 try:
2910 for idx in range(known_idx, last_seq):
2911 # do not update sequence here or you'll get skipped some part of it
2912 should_continue, _ = _extract_sequence_from_mpd(False, False)
2913 if not should_continue:
2914 known_idx = idx - 1
2915 raise ExtractorError('breaking out of outer loop')
2916 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2917 yield {
2918 'url': last_segment_url,
2919 'fragment_count': last_seq,
2920 }
2921 if known_idx == last_seq:
2922 no_fragment_score += 5
2923 else:
2924 no_fragment_score = 0
2925 known_idx = last_seq
2926 except ExtractorError:
2927 continue
2928
2929 if manifestless_orig_fmt:
2930 # Stop at the first iteration if running for post-live manifestless;
2931 # fragment count no longer increase since it starts
2932 break
2933
2934 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2935
2936 def _extract_player_url(self, *ytcfgs, webpage=None):
2937 player_url = traverse_obj(
2938 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2939 get_all=False, expected_type=str)
2940 if not player_url:
2941 return
2942 return urljoin('https://www.youtube.com', player_url)
2943
2944 def _download_player_url(self, video_id, fatal=False):
2945 res = self._download_webpage(
2946 'https://www.youtube.com/iframe_api',
2947 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2948 if res:
2949 player_version = self._search_regex(
2950 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2951 if player_version:
2952 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2953
2954 def _signature_cache_id(self, example_sig):
2955 """ Return a string representation of a signature """
2956 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2957
2958 @classmethod
2959 def _extract_player_info(cls, player_url):
2960 for player_re in cls._PLAYER_INFO_RE:
2961 id_m = re.search(player_re, player_url)
2962 if id_m:
2963 break
2964 else:
2965 raise ExtractorError('Cannot identify player %r' % player_url)
2966 return id_m.group('id')
2967
2968 def _load_player(self, video_id, player_url, fatal=True):
2969 player_id = self._extract_player_info(player_url)
2970 if player_id not in self._code_cache:
2971 code = self._download_webpage(
2972 player_url, video_id, fatal=fatal,
2973 note='Downloading player ' + player_id,
2974 errnote='Download of %s failed' % player_url)
2975 if code:
2976 self._code_cache[player_id] = code
2977 return self._code_cache.get(player_id)
2978
2979 def _extract_signature_function(self, video_id, player_url, example_sig):
2980 player_id = self._extract_player_info(player_url)
2981
2982 # Read from filesystem cache
2983 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2984 assert os.path.basename(func_id) == func_id
2985
2986 self.write_debug(f'Extracting signature function {func_id}')
2987 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2988
2989 if not cache_spec:
2990 code = self._load_player(video_id, player_url)
2991 if code:
2992 res = self._parse_sig_js(code)
2993 test_string = ''.join(map(chr, range(len(example_sig))))
2994 cache_spec = [ord(c) for c in res(test_string)]
2995 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2996
2997 return lambda s: ''.join(s[i] for i in cache_spec)
2998
2999 def _print_sig_code(self, func, example_sig):
3000 if not self.get_param('youtube_print_sig_code'):
3001 return
3002
3003 def gen_sig_code(idxs):
3004 def _genslice(start, end, step):
3005 starts = '' if start == 0 else str(start)
3006 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
3007 steps = '' if step == 1 else (':%d' % step)
3008 return f's[{starts}{ends}{steps}]'
3009
3010 step = None
3011 # Quelch pyflakes warnings - start will be set when step is set
3012 start = '(Never used)'
3013 for i, prev in zip(idxs[1:], idxs[:-1]):
3014 if step is not None:
3015 if i - prev == step:
3016 continue
3017 yield _genslice(start, prev, step)
3018 step = None
3019 continue
3020 if i - prev in [-1, 1]:
3021 step = i - prev
3022 start = prev
3023 continue
3024 else:
3025 yield 's[%d]' % prev
3026 if step is None:
3027 yield 's[%d]' % i
3028 else:
3029 yield _genslice(start, i, step)
3030
3031 test_string = ''.join(map(chr, range(len(example_sig))))
3032 cache_res = func(test_string)
3033 cache_spec = [ord(c) for c in cache_res]
3034 expr_code = ' + '.join(gen_sig_code(cache_spec))
3035 signature_id_tuple = '(%s)' % (
3036 ', '.join(str(len(p)) for p in example_sig.split('.')))
3037 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
3038 ' return %s\n') % (signature_id_tuple, expr_code)
3039 self.to_screen('Extracted signature function:\n' + code)
3040
3041 def _parse_sig_js(self, jscode):
3042 funcname = self._search_regex(
3043 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3044 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3045 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3046 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3047 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)(?:;[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\))?',
3048 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3049 # Obsolete patterns
3050 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3051 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3052 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3053 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3054 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3055 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3056 jscode, 'Initial JS player signature function name', group='sig')
3057
3058 jsi = JSInterpreter(jscode)
3059 initial_function = jsi.extract_function(funcname)
3060 return lambda s: initial_function([s])
3061
3062 def _cached(self, func, *cache_id):
3063 def inner(*args, **kwargs):
3064 if cache_id not in self._player_cache:
3065 try:
3066 self._player_cache[cache_id] = func(*args, **kwargs)
3067 except ExtractorError as e:
3068 self._player_cache[cache_id] = e
3069 except Exception as e:
3070 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3071
3072 ret = self._player_cache[cache_id]
3073 if isinstance(ret, Exception):
3074 raise ret
3075 return ret
3076 return inner
3077
3078 def _decrypt_signature(self, s, video_id, player_url):
3079 """Turn the encrypted s field into a working signature"""
3080 extract_sig = self._cached(
3081 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3082 func = extract_sig(video_id, player_url, s)
3083 self._print_sig_code(func, s)
3084 return func(s)
3085
3086 def _decrypt_nsig(self, s, video_id, player_url):
3087 """Turn the encrypted n field into a working signature"""
3088 if player_url is None:
3089 raise ExtractorError('Cannot decrypt nsig without player_url')
3090 player_url = urljoin('https://www.youtube.com', player_url)
3091
3092 try:
3093 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3094 except ExtractorError as e:
3095 raise ExtractorError('Unable to extract nsig function code', cause=e)
3096 if self.get_param('youtube_print_sig_code'):
3097 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3098
3099 try:
3100 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3101 ret = extract_nsig(jsi, func_code)(s)
3102 except JSInterpreter.Exception as e:
3103 try:
3104 jsi = PhantomJSwrapper(self, timeout=5000)
3105 except ExtractorError:
3106 raise e
3107 self.report_warning(
3108 f'Native nsig extraction failed: Trying with PhantomJS\n'
3109 f' n = {s} ; player = {player_url}', video_id)
3110 self.write_debug(e, only_once=True)
3111
3112 args, func_body = func_code
3113 ret = jsi.execute(
3114 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3115 video_id=video_id, note='Executing signature code').strip()
3116
3117 self.write_debug(f'Decrypted nsig {s} => {ret}')
3118 return ret
3119
3120 def _extract_n_function_name(self, jscode):
3121 funcname, idx = self._search_regex(
3122 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3123 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3124 if not idx:
3125 return funcname
3126
3127 return json.loads(js_to_json(self._search_regex(
3128 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
3129 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3130
3131 def _extract_n_function_code(self, video_id, player_url):
3132 player_id = self._extract_player_info(player_url)
3133 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3134 jscode = func_code or self._load_player(video_id, player_url)
3135 jsi = JSInterpreter(jscode)
3136
3137 if func_code:
3138 return jsi, player_id, func_code
3139
3140 func_name = self._extract_n_function_name(jscode)
3141
3142 # For redundancy
3143 func_code = self._search_regex(
3144 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3145 # NB: The end of the regex is intentionally kept strict
3146 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3147 jscode, 'nsig function', group=('var', 'code'), default=None)
3148 if func_code:
3149 func_code = ([func_code[0]], func_code[1])
3150 else:
3151 self.write_debug('Extracting nsig function with jsinterp')
3152 func_code = jsi.extract_function_code(func_name)
3153
3154 self.cache.store('youtube-nsig', player_id, func_code)
3155 return jsi, player_id, func_code
3156
3157 def _extract_n_function_from_code(self, jsi, func_code):
3158 func = jsi.extract_function_from_code(*func_code)
3159
3160 def extract_nsig(s):
3161 try:
3162 ret = func([s])
3163 except JSInterpreter.Exception:
3164 raise
3165 except Exception as e:
3166 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3167
3168 if ret.startswith('enhanced_except_'):
3169 raise JSInterpreter.Exception('Signature function returned an exception')
3170 return ret
3171
3172 return extract_nsig
3173
3174 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3175 """
3176 Extract signatureTimestamp (sts)
3177 Required to tell API what sig/player version is in use.
3178 """
3179 sts = None
3180 if isinstance(ytcfg, dict):
3181 sts = int_or_none(ytcfg.get('STS'))
3182
3183 if not sts:
3184 # Attempt to extract from player
3185 if player_url is None:
3186 error_msg = 'Cannot extract signature timestamp without player_url.'
3187 if fatal:
3188 raise ExtractorError(error_msg)
3189 self.report_warning(error_msg)
3190 return
3191 code = self._load_player(video_id, player_url, fatal=fatal)
3192 if code:
3193 sts = int_or_none(self._search_regex(
3194 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3195 'JS player signature timestamp', group='sts', fatal=fatal))
3196 return sts
3197
3198 def _mark_watched(self, video_id, player_responses):
3199 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3200 label = 'fully ' if is_full else ''
3201 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3202 expected_type=url_or_none)
3203 if not url:
3204 self.report_warning(f'Unable to mark {label}watched')
3205 return
3206 parsed_url = urllib.parse.urlparse(url)
3207 qs = urllib.parse.parse_qs(parsed_url.query)
3208
3209 # cpn generation algorithm is reverse engineered from base.js.
3210 # In fact it works even with dummy cpn.
3211 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3212 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3213
3214 # # more consistent results setting it to right before the end
3215 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3216
3217 qs.update({
3218 'ver': ['2'],
3219 'cpn': [cpn],
3220 'cmt': video_length,
3221 'el': 'detailpage', # otherwise defaults to "shorts"
3222 })
3223
3224 if is_full:
3225 # these seem to mark watchtime "history" in the real world
3226 # they're required, so send in a single value
3227 qs.update({
3228 'st': 0,
3229 'et': video_length,
3230 })
3231
3232 url = urllib.parse.urlunparse(
3233 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3234
3235 self._download_webpage(
3236 url, video_id, f'Marking {label}watched',
3237 'Unable to mark watched', fatal=False)
3238
3239 @classmethod
3240 def _extract_from_webpage(cls, url, webpage):
3241 # Invidious Instances
3242 # https://github.com/yt-dlp/yt-dlp/issues/195
3243 # https://github.com/iv-org/invidious/pull/1730
3244 mobj = re.search(
3245 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3246 webpage)
3247 if mobj:
3248 yield cls.url_result(mobj.group('url'), cls)
3249 raise cls.StopExtraction()
3250
3251 yield from super()._extract_from_webpage(url, webpage)
3252
3253 # lazyYT YouTube embed
3254 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3255 yield cls.url_result(unescapeHTML(id_), cls, id_)
3256
3257 # Wordpress "YouTube Video Importer" plugin
3258 for m in re.findall(r'''(?x)<div[^>]+
3259 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3260 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3261 yield cls.url_result(m[-1], cls, m[-1])
3262
3263 @classmethod
3264 def extract_id(cls, url):
3265 video_id = cls.get_temp_id(url)
3266 if not video_id:
3267 raise ExtractorError(f'Invalid URL: {url}')
3268 return video_id
3269
3270 def _extract_chapters_from_json(self, data, duration):
3271 chapter_list = traverse_obj(
3272 data, (
3273 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3274 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3275 ), expected_type=list)
3276
3277 return self._extract_chapters_helper(
3278 chapter_list,
3279 start_function=lambda chapter: float_or_none(
3280 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3281 title_function=lambda chapter: traverse_obj(
3282 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3283 duration=duration)
3284
3285 def _extract_chapters_from_engagement_panel(self, data, duration):
3286 content_list = traverse_obj(
3287 data,
3288 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3289 expected_type=list)
3290 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3291 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3292
3293 return next(filter(None, (
3294 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3295 chapter_time, chapter_title, duration)
3296 for contents in content_list)), [])
3297
3298 def _extract_heatmap(self, data):
3299 return traverse_obj(data, (
3300 'frameworkUpdates', 'entityBatchUpdate', 'mutations',
3301 lambda _, v: v['payload']['macroMarkersListEntity']['markersList']['markerType'] == 'MARKER_TYPE_HEATMAP',
3302 'payload', 'macroMarkersListEntity', 'markersList', 'markers', ..., {
3303 'start_time': ('startMillis', {functools.partial(float_or_none, scale=1000)}),
3304 'end_time': {lambda x: (int(x['startMillis']) + int(x['durationMillis'])) / 1000},
3305 'value': ('intensityScoreNormalized', {float_or_none}),
3306 })) or None
3307
3308 def _extract_comment(self, comment_renderer, parent=None):
3309 comment_id = comment_renderer.get('commentId')
3310 if not comment_id:
3311 return
3312
3313 info = {
3314 'id': comment_id,
3315 'text': self._get_text(comment_renderer, 'contentText'),
3316 'like_count': self._get_count(comment_renderer, 'voteCount'),
3317 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3318 'author': self._get_text(comment_renderer, 'authorText'),
3319 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3320 'parent': parent or 'root',
3321 }
3322
3323 # Timestamp is an estimate calculated from the current time and time_text
3324 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3325 timestamp = self._parse_time_text(time_text)
3326
3327 info.update({
3328 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3329 '_time_text': time_text,
3330 'timestamp': timestamp,
3331 })
3332
3333 info['author_url'] = urljoin(
3334 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3335 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3336 expected_type=str, get_all=False))
3337
3338 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3339 if author_is_uploader is not None:
3340 info['author_is_uploader'] = author_is_uploader
3341
3342 comment_abr = traverse_obj(
3343 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3344 if comment_abr is not None:
3345 info['is_favorited'] = 'creatorHeart' in comment_abr
3346
3347 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3348 if self._has_badge(badges, BadgeType.VERIFIED):
3349 info['author_is_verified'] = True
3350
3351 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3352 if is_pinned:
3353 info['is_pinned'] = True
3354
3355 return info
3356
3357 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3358
3359 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3360
3361 def extract_header(contents):
3362 _continuation = None
3363 for content in contents:
3364 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3365 expected_comment_count = self._get_count(
3366 comments_header_renderer, 'countText', 'commentsCount')
3367
3368 if expected_comment_count is not None:
3369 tracker['est_total'] = expected_comment_count
3370 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3371 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3372
3373 sort_menu_item = try_get(
3374 comments_header_renderer,
3375 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3376 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3377
3378 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3379 if not _continuation:
3380 continue
3381
3382 sort_text = str_or_none(sort_menu_item.get('title'))
3383 if not sort_text:
3384 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3385 self.to_screen('Sorting comments by %s' % sort_text.lower())
3386 break
3387 return _continuation
3388
3389 def extract_thread(contents):
3390 if not parent:
3391 tracker['current_page_thread'] = 0
3392 for content in contents:
3393 if not parent and tracker['total_parent_comments'] >= max_parents:
3394 yield
3395 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3396 comment_renderer = get_first(
3397 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3398 expected_type=dict, default={})
3399
3400 comment = self._extract_comment(comment_renderer, parent)
3401 if not comment:
3402 continue
3403 comment_id = comment['id']
3404 if comment.get('is_pinned'):
3405 tracker['pinned_comment_ids'].add(comment_id)
3406 # Sometimes YouTube may break and give us infinite looping comments.
3407 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3408 if comment_id in tracker['seen_comment_ids']:
3409 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3410 # Pinned comments may appear a second time in newest first sort
3411 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3412 continue
3413 self.report_warning(
3414 'Detected YouTube comments looping. Stopping comment extraction '
3415 f'{"for this thread" if parent else ""} as we probably cannot get any more.')
3416 yield
3417 else:
3418 tracker['seen_comment_ids'].add(comment['id'])
3419
3420 tracker['running_total'] += 1
3421 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3422 yield comment
3423
3424 # Attempt to get the replies
3425 comment_replies_renderer = try_get(
3426 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3427
3428 if comment_replies_renderer:
3429 tracker['current_page_thread'] += 1
3430 comment_entries_iter = self._comment_entries(
3431 comment_replies_renderer, ytcfg, video_id,
3432 parent=comment.get('id'), tracker=tracker)
3433 yield from itertools.islice(comment_entries_iter, min(
3434 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3435
3436 # Keeps track of counts across recursive calls
3437 if not tracker:
3438 tracker = dict(
3439 running_total=0,
3440 est_total=None,
3441 current_page_thread=0,
3442 total_parent_comments=0,
3443 total_reply_comments=0,
3444 seen_comment_ids=set(),
3445 pinned_comment_ids=set()
3446 )
3447
3448 # TODO: Deprecated
3449 # YouTube comments have a max depth of 2
3450 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3451 if max_depth:
3452 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3453 'Set max replies in the max-comments extractor argument instead')
3454 if max_depth == 1 and parent:
3455 return
3456
3457 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3458 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3459
3460 continuation = self._extract_continuation(root_continuation_data)
3461
3462 response = None
3463 is_forced_continuation = False
3464 is_first_continuation = parent is None
3465 if is_first_continuation and not continuation:
3466 # Sometimes you can get comments by generating the continuation yourself,
3467 # even if YouTube initially reports them being disabled - e.g. stories comments.
3468 # Note: if the comment section is actually disabled, YouTube may return a response with
3469 # required check_get_keys missing. So we will disable that check initially in this case.
3470 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3471 is_forced_continuation = True
3472
3473 continuation_items_path = (
3474 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3475 for page_num in itertools.count(0):
3476 if not continuation:
3477 break
3478 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3479 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3480 if page_num == 0:
3481 if is_first_continuation:
3482 note_prefix = 'Downloading comment section API JSON'
3483 else:
3484 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3485 tracker['current_page_thread'], comment_prog_str)
3486 else:
3487 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3488 ' ' if parent else '', ' replies' if parent else '',
3489 page_num, comment_prog_str)
3490
3491 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3492 # Ignore check if YouTube says the comment count is 0.
3493 check_get_keys = None
3494 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3495 check_get_keys = [[*continuation_items_path, ..., (
3496 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
3497 try:
3498 response = self._extract_response(
3499 item_id=None, query=continuation,
3500 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3501 check_get_keys=check_get_keys)
3502 except ExtractorError as e:
3503 # Ignore incomplete data error for replies if retries didn't work.
3504 # This is to allow any other parent comments and comment threads to be downloaded.
3505 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3506 if 'incomplete data' in str(e).lower() and parent:
3507 if self.get_param('ignoreerrors') in (True, 'only_download'):
3508 self.report_warning(
3509 'Received incomplete data for a comment reply thread and retrying did not help. '
3510 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
3511 return
3512 else:
3513 raise ExtractorError(
3514 'Incomplete data received for comment reply thread. '
3515 'Pass --ignore-errors to ignore and allow rest of comments to download.',
3516 expected=True)
3517 raise
3518 is_forced_continuation = False
3519 continuation = None
3520 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3521 if is_first_continuation:
3522 continuation = extract_header(continuation_items)
3523 is_first_continuation = False
3524 if continuation:
3525 break
3526 continue
3527
3528 for entry in extract_thread(continuation_items):
3529 if not entry:
3530 return
3531 yield entry
3532 continuation = self._extract_continuation({'contents': continuation_items})
3533 if continuation:
3534 break
3535
3536 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3537 if message and not parent and tracker['running_total'] == 0:
3538 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3539 raise self.CommentsDisabled
3540
3541 @staticmethod
3542 def _generate_comment_continuation(video_id):
3543 """
3544 Generates initial comment section continuation token from given video id
3545 """
3546 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3547 return base64.b64encode(token.encode()).decode()
3548
3549 def _get_comments(self, ytcfg, video_id, contents, webpage):
3550 """Entry for comment extraction"""
3551 def _real_comment_extract(contents):
3552 renderer = next((
3553 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3554 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3555 yield from self._comment_entries(renderer, ytcfg, video_id)
3556
3557 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3558 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3559
3560 @staticmethod
3561 def _get_checkok_params():
3562 return {'contentCheckOk': True, 'racyCheckOk': True}
3563
3564 @classmethod
3565 def _generate_player_context(cls, sts=None):
3566 context = {
3567 'html5Preference': 'HTML5_PREF_WANTS',
3568 }
3569 if sts is not None:
3570 context['signatureTimestamp'] = sts
3571 return {
3572 'playbackContext': {
3573 'contentPlaybackContext': context
3574 },
3575 **cls._get_checkok_params()
3576 }
3577
3578 @staticmethod
3579 def _is_agegated(player_response):
3580 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3581 return True
3582
3583 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3584 AGE_GATE_REASONS = (
3585 'confirm your age', 'age-restricted', 'inappropriate', # reason
3586 'age_verification_required', 'age_check_required', # status
3587 )
3588 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3589
3590 @staticmethod
3591 def _is_unplayable(player_response):
3592 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3593
3594 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3595
3596 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3597 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3598 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3599 headers = self.generate_api_headers(
3600 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3601
3602 yt_query = {
3603 'videoId': video_id,
3604 }
3605 if _split_innertube_client(client)[0] == 'android':
3606 yt_query['params'] = 'CgIQBg=='
3607
3608 pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0]
3609 if pp_arg:
3610 yt_query['params'] = pp_arg
3611
3612 yt_query.update(self._generate_player_context(sts))
3613 return self._extract_response(
3614 item_id=video_id, ep='player', query=yt_query,
3615 ytcfg=player_ytcfg, headers=headers, fatal=True,
3616 default_client=client,
3617 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3618 ) or None
3619
3620 def _get_requested_clients(self, url, smuggled_data):
3621 requested_clients = []
3622 default = ['ios', 'android', 'web']
3623 allowed_clients = sorted(
3624 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3625 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3626 for client in self._configuration_arg('player_client'):
3627 if client in allowed_clients:
3628 requested_clients.append(client)
3629 elif client == 'default':
3630 requested_clients.extend(default)
3631 elif client == 'all':
3632 requested_clients.extend(allowed_clients)
3633 else:
3634 self.report_warning(f'Skipping unsupported client {client}')
3635 if not requested_clients:
3636 requested_clients = default
3637
3638 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3639 requested_clients.extend(
3640 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3641
3642 return orderedSet(requested_clients)
3643
3644 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3645 initial_pr = None
3646 if webpage:
3647 initial_pr = self._search_json(
3648 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3649
3650 all_clients = set(clients)
3651 clients = clients[::-1]
3652 prs = []
3653
3654 def append_client(*client_names):
3655 """ Append the first client name that exists but not already used """
3656 for client_name in client_names:
3657 actual_client = _split_innertube_client(client_name)[0]
3658 if actual_client in INNERTUBE_CLIENTS:
3659 if actual_client not in all_clients:
3660 clients.append(client_name)
3661 all_clients.add(actual_client)
3662 return
3663
3664 # Android player_response does not have microFormats which are needed for
3665 # extraction of some data. So we return the initial_pr with formats
3666 # stripped out even if not requested by the user
3667 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3668 if initial_pr:
3669 pr = dict(initial_pr)
3670 pr['streamingData'] = None
3671 prs.append(pr)
3672
3673 last_error = None
3674 tried_iframe_fallback = False
3675 player_url = None
3676 while clients:
3677 client, base_client, variant = _split_innertube_client(clients.pop())
3678 player_ytcfg = master_ytcfg if client == 'web' else {}
3679 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3680 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3681
3682 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3683 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3684 if 'js' in self._configuration_arg('player_skip'):
3685 require_js_player = False
3686 player_url = None
3687
3688 if not player_url and not tried_iframe_fallback and require_js_player:
3689 player_url = self._download_player_url(video_id)
3690 tried_iframe_fallback = True
3691
3692 try:
3693 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3694 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3695 except ExtractorError as e:
3696 if last_error:
3697 self.report_warning(last_error)
3698 last_error = e
3699 continue
3700
3701 if pr:
3702 # YouTube may return a different video player response than expected.
3703 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3704 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3705 if pr_video_id and pr_video_id != video_id:
3706 self.report_warning(
3707 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3708 else:
3709 # Save client name for introspection later
3710 name = short_client_name(client)
3711 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3712 sd[STREAMING_DATA_CLIENT_NAME] = name
3713 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3714 f[STREAMING_DATA_CLIENT_NAME] = name
3715 prs.append(pr)
3716
3717 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3718 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3719 append_client(f'{base_client}_creator')
3720 elif self._is_agegated(pr):
3721 if variant == 'tv_embedded':
3722 append_client(f'{base_client}_embedded')
3723 elif not variant:
3724 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3725
3726 if last_error:
3727 if not len(prs):
3728 raise last_error
3729 self.report_warning(last_error)
3730 return prs, player_url
3731
3732 def _needs_live_processing(self, live_status, duration):
3733 if (live_status == 'is_live' and self.get_param('live_from_start')
3734 or live_status == 'post_live' and (duration or 0) > 2 * 3600):
3735 return live_status
3736
3737 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3738 CHUNK_SIZE = 10 << 20
3739 itags, stream_ids = collections.defaultdict(set), []
3740 itag_qualities, res_qualities = {}, {0: None}
3741 q = qualities([
3742 # Normally tiny is the smallest video-only formats. But
3743 # audio-only formats with unknown quality may get tagged as tiny
3744 'tiny',
3745 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3746 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3747 ])
3748 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3749 format_types = self._configuration_arg('formats')
3750 all_formats = 'duplicate' in format_types
3751 if self._configuration_arg('include_duplicate_formats'):
3752 all_formats = True
3753 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
3754 'Use formats=duplicate extractor argument instead')
3755
3756 def build_fragments(f):
3757 return LazyList({
3758 'url': update_url_query(f['url'], {
3759 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3760 })
3761 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3762
3763 for fmt in streaming_formats:
3764 if fmt.get('targetDurationSec'):
3765 continue
3766
3767 itag = str_or_none(fmt.get('itag'))
3768 audio_track = fmt.get('audioTrack') or {}
3769 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3770 if not all_formats:
3771 if stream_id in stream_ids:
3772 continue
3773
3774 quality = fmt.get('quality')
3775 height = int_or_none(fmt.get('height'))
3776 if quality == 'tiny' or not quality:
3777 quality = fmt.get('audioQuality', '').lower() or quality
3778 # The 3gp format (17) in android client has a quality of "small",
3779 # but is actually worse than other formats
3780 if itag == '17':
3781 quality = 'tiny'
3782 if quality:
3783 if itag:
3784 itag_qualities[itag] = quality
3785 if height:
3786 res_qualities[height] = quality
3787 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3788 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3789 # number of fragment that would subsequently requested with (`&sq=N`)
3790 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3791 continue
3792
3793 fmt_url = fmt.get('url')
3794 if not fmt_url:
3795 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3796 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3797 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3798 if not all((sc, fmt_url, player_url, encrypted_sig)):
3799 continue
3800 try:
3801 fmt_url += '&%s=%s' % (
3802 traverse_obj(sc, ('sp', -1)) or 'signature',
3803 self._decrypt_signature(encrypted_sig, video_id, player_url)
3804 )
3805 except ExtractorError as e:
3806 self.report_warning('Signature extraction failed: Some formats may be missing',
3807 video_id=video_id, only_once=True)
3808 self.write_debug(e, only_once=True)
3809 continue
3810
3811 query = parse_qs(fmt_url)
3812 throttled = False
3813 if query.get('n'):
3814 try:
3815 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3816 fmt_url = update_url_query(fmt_url, {
3817 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3818 })
3819 except ExtractorError as e:
3820 phantomjs_hint = ''
3821 if isinstance(e, JSInterpreter.Exception):
3822 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3823 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3824 if player_url:
3825 self.report_warning(
3826 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3827 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3828 self.write_debug(e, only_once=True)
3829 else:
3830 self.report_warning(
3831 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3832 video_id=video_id, only_once=True)
3833 throttled = True
3834
3835 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3836 language_preference = (
3837 10 if audio_track.get('audioIsDefault') and 10
3838 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3839 else -1)
3840 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3841 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3842 # Make sure to avoid false positives with small duration differences.
3843 # E.g. __2ABJjxzNo, ySuUZEjARPY
3844 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3845 if is_damaged:
3846 self.report_warning(
3847 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3848
3849 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3850 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
3851 fps = int_or_none(fmt.get('fps')) or 0
3852 dct = {
3853 'asr': int_or_none(fmt.get('audioSampleRate')),
3854 'filesize': int_or_none(fmt.get('contentLength')),
3855 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3856 'format_note': join_nonempty(
3857 join_nonempty(audio_track.get('displayName'),
3858 language_preference > 0 and ' (default)', delim=''),
3859 name, fmt.get('isDrc') and 'DRC',
3860 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3861 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3862 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3863 (self.get_param('verbose') or all_formats) and client_name,
3864 delim=', '),
3865 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3866 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
3867 + (100 if 'Premium' in name else 0)),
3868 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
3869 'audio_channels': fmt.get('audioChannels'),
3870 'height': height,
3871 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3872 'has_drm': bool(fmt.get('drmFamilies')),
3873 'tbr': tbr,
3874 'url': fmt_url,
3875 'width': int_or_none(fmt.get('width')),
3876 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3877 'desc' if language_preference < -1 else '') or None,
3878 'language_preference': language_preference,
3879 # Strictly de-prioritize damaged and 3gp formats
3880 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3881 }
3882 mime_mobj = re.match(
3883 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3884 if mime_mobj:
3885 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3886 dct.update(parse_codecs(mime_mobj.group(2)))
3887 if itag:
3888 itags[itag].add(('https', dct.get('language')))
3889 stream_ids.append(stream_id)
3890 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3891 if single_stream and dct.get('ext'):
3892 dct['container'] = dct['ext'] + '_dash'
3893
3894 if (all_formats or 'dashy' in format_types) and dct['filesize']:
3895 yield {
3896 **dct,
3897 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3898 'protocol': 'http_dash_segments',
3899 'fragments': build_fragments(dct),
3900 }
3901 if all_formats or 'dashy' not in format_types:
3902 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3903 yield dct
3904
3905 needs_live_processing = self._needs_live_processing(live_status, duration)
3906 skip_bad_formats = 'incomplete' not in format_types
3907 if self._configuration_arg('include_incomplete_formats'):
3908 skip_bad_formats = False
3909 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
3910 'Use formats=incomplete extractor argument instead')
3911
3912 skip_manifests = set(self._configuration_arg('skip'))
3913 if (not self.get_param('youtube_include_hls_manifest', True)
3914 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3915 or needs_live_processing and skip_bad_formats):
3916 skip_manifests.add('hls')
3917
3918 if not self.get_param('youtube_include_dash_manifest', True):
3919 skip_manifests.add('dash')
3920 if self._configuration_arg('include_live_dash'):
3921 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3922 'Use formats=incomplete extractor argument instead')
3923 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3924 skip_manifests.add('dash')
3925
3926 def process_manifest_format(f, proto, client_name, itag):
3927 key = (proto, f.get('language'))
3928 if not all_formats and key in itags[itag]:
3929 return False
3930 itags[itag].add(key)
3931
3932 if itag and all_formats:
3933 f['format_id'] = f'{itag}-{proto}'
3934 elif any(p != proto for p, _ in itags[itag]):
3935 f['format_id'] = f'{itag}-{proto}'
3936 elif itag:
3937 f['format_id'] = itag
3938
3939 if f.get('source_preference') is None:
3940 f['source_preference'] = -1
3941
3942 if itag in ('616', '235'):
3943 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
3944 f['source_preference'] += 100
3945
3946 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3947 if f['quality'] == -1 and f.get('height'):
3948 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3949 if self.get_param('verbose') or all_formats:
3950 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3951 if f.get('fps') and f['fps'] <= 1:
3952 del f['fps']
3953
3954 if proto == 'hls' and f.get('has_drm'):
3955 f['has_drm'] = 'maybe'
3956 f['source_preference'] -= 5
3957 return True
3958
3959 subtitles = {}
3960 for sd in streaming_data:
3961 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3962
3963 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3964 if hls_manifest_url:
3965 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3966 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3967 subtitles = self._merge_subtitles(subs, subtitles)
3968 for f in fmts:
3969 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3970 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3971 yield f
3972
3973 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3974 if dash_manifest_url:
3975 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3976 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3977 for f in formats:
3978 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3979 f['filesize'] = int_or_none(self._search_regex(
3980 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3981 if needs_live_processing:
3982 f['is_from_start'] = True
3983
3984 yield f
3985 yield subtitles
3986
3987 def _extract_storyboard(self, player_responses, duration):
3988 spec = get_first(
3989 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3990 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3991 if not base_url:
3992 return
3993 L = len(spec) - 1
3994 for i, args in enumerate(spec):
3995 args = args.split('#')
3996 counts = list(map(int_or_none, args[:5]))
3997 if len(args) != 8 or not all(counts):
3998 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3999 continue
4000 width, height, frame_count, cols, rows = counts
4001 N, sigh = args[6:]
4002
4003 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
4004 fragment_count = frame_count / (cols * rows)
4005 fragment_duration = duration / fragment_count
4006 yield {
4007 'format_id': f'sb{i}',
4008 'format_note': 'storyboard',
4009 'ext': 'mhtml',
4010 'protocol': 'mhtml',
4011 'acodec': 'none',
4012 'vcodec': 'none',
4013 'url': url,
4014 'width': width,
4015 'height': height,
4016 'fps': frame_count / duration,
4017 'rows': rows,
4018 'columns': cols,
4019 'fragments': [{
4020 'url': url.replace('$M', str(j)),
4021 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4022 } for j in range(math.ceil(fragment_count))],
4023 }
4024
4025 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4026 webpage = None
4027 if 'webpage' not in self._configuration_arg('player_skip'):
4028 query = {'bpctr': '9999999999', 'has_verified': '1'}
4029 pp = self._configuration_arg('player_params', [None], casesense=True)[0]
4030 if pp:
4031 query['pp'] = pp
4032 webpage = self._download_webpage(
4033 webpage_url, video_id, fatal=False, query=query)
4034
4035 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4036
4037 player_responses, player_url = self._extract_player_responses(
4038 self._get_requested_clients(url, smuggled_data),
4039 video_id, webpage, master_ytcfg, smuggled_data)
4040
4041 return webpage, master_ytcfg, player_responses, player_url
4042
4043 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4044 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4045 is_live = get_first(video_details, 'isLive')
4046 if is_live is None:
4047 is_live = get_first(live_broadcast_details, 'isLiveNow')
4048 live_content = get_first(video_details, 'isLiveContent')
4049 is_upcoming = get_first(video_details, 'isUpcoming')
4050 post_live = get_first(video_details, 'isPostLiveDvr')
4051 live_status = ('post_live' if post_live
4052 else 'is_live' if is_live
4053 else 'is_upcoming' if is_upcoming
4054 else 'was_live' if live_content
4055 else 'not_live' if False in (is_live, live_content)
4056 else None)
4057 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4058 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4059 if all(f.get('has_drm') for f in formats):
4060 # If there are no formats that definitely don't have DRM, all have DRM
4061 for f in formats:
4062 f['has_drm'] = True
4063
4064 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4065
4066 def _real_extract(self, url):
4067 url, smuggled_data = unsmuggle_url(url, {})
4068 video_id = self._match_id(url)
4069
4070 base_url = self.http_scheme() + '//www.youtube.com/'
4071 webpage_url = base_url + 'watch?v=' + video_id
4072
4073 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4074
4075 playability_statuses = traverse_obj(
4076 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4077
4078 trailer_video_id = get_first(
4079 playability_statuses,
4080 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4081 expected_type=str)
4082 if trailer_video_id:
4083 return self.url_result(
4084 trailer_video_id, self.ie_key(), trailer_video_id)
4085
4086 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4087 if webpage else (lambda x: None))
4088
4089 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4090 microformats = traverse_obj(
4091 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4092 expected_type=dict)
4093
4094 translated_title = self._get_text(microformats, (..., 'title'))
4095 video_title = (self._preferred_lang and translated_title
4096 or get_first(video_details, 'title') # primary
4097 or translated_title
4098 or search_meta(['og:title', 'twitter:title', 'title']))
4099 translated_description = self._get_text(microformats, (..., 'description'))
4100 original_description = get_first(video_details, 'shortDescription')
4101 video_description = (
4102 self._preferred_lang and translated_description
4103 # If original description is blank, it will be an empty string.
4104 # Do not prefer translated description in this case.
4105 or original_description if original_description is not None else translated_description)
4106
4107 multifeed_metadata_list = get_first(
4108 player_responses,
4109 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4110 expected_type=str)
4111 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4112 if self.get_param('noplaylist'):
4113 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4114 else:
4115 entries = []
4116 feed_ids = []
4117 for feed in multifeed_metadata_list.split(','):
4118 # Unquote should take place before split on comma (,) since textual
4119 # fields may contain comma as well (see
4120 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4121 feed_data = urllib.parse.parse_qs(
4122 urllib.parse.unquote_plus(feed))
4123
4124 def feed_entry(name):
4125 return try_get(
4126 feed_data, lambda x: x[name][0], str)
4127
4128 feed_id = feed_entry('id')
4129 if not feed_id:
4130 continue
4131 feed_title = feed_entry('title')
4132 title = video_title
4133 if feed_title:
4134 title += ' (%s)' % feed_title
4135 entries.append({
4136 '_type': 'url_transparent',
4137 'ie_key': 'Youtube',
4138 'url': smuggle_url(
4139 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4140 {'force_singlefeed': True}),
4141 'title': title,
4142 })
4143 feed_ids.append(feed_id)
4144 self.to_screen(
4145 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4146 % (', '.join(feed_ids), video_id))
4147 return self.playlist_result(
4148 entries, video_id, video_title, video_description)
4149
4150 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4151 or int_or_none(get_first(microformats, 'lengthSeconds'))
4152 or parse_duration(search_meta('duration')) or None)
4153
4154 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4155 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4156 if live_status == 'post_live':
4157 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4158
4159 if not formats:
4160 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4161 self.report_drm(video_id)
4162 pemr = get_first(
4163 playability_statuses,
4164 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4165 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4166 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4167 if subreason:
4168 if subreason == 'The uploader has not made this video available in your country.':
4169 countries = get_first(microformats, 'availableCountries')
4170 if not countries:
4171 regions_allowed = search_meta('regionsAllowed')
4172 countries = regions_allowed.split(',') if regions_allowed else None
4173 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4174 reason += f'. {subreason}'
4175 if reason:
4176 self.raise_no_formats(reason, expected=True)
4177
4178 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4179 if not keywords and webpage:
4180 keywords = [
4181 unescapeHTML(m.group('content'))
4182 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4183 for keyword in keywords:
4184 if keyword.startswith('yt:stretch='):
4185 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4186 if mobj:
4187 # NB: float is intentional for forcing float division
4188 w, h = (float(v) for v in mobj.groups())
4189 if w > 0 and h > 0:
4190 ratio = w / h
4191 for f in formats:
4192 if f.get('vcodec') != 'none':
4193 f['stretched_ratio'] = ratio
4194 break
4195 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4196 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4197 if thumbnail_url:
4198 thumbnails.append({
4199 'url': thumbnail_url,
4200 })
4201 original_thumbnails = thumbnails.copy()
4202
4203 # The best resolution thumbnails sometimes does not appear in the webpage
4204 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4205 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4206 thumbnail_names = [
4207 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4208 # in resolution, these are not the custom thumbnail. So de-prioritize them
4209 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4210 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4211 ]
4212 n_thumbnail_names = len(thumbnail_names)
4213 thumbnails.extend({
4214 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4215 video_id=video_id, name=name, ext=ext,
4216 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4217 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4218 for thumb in thumbnails:
4219 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4220 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4221 self._remove_duplicate_formats(thumbnails)
4222 self._downloader._sort_thumbnails(original_thumbnails)
4223
4224 category = get_first(microformats, 'category') or search_meta('genre')
4225 channel_id = self.ucid_or_none(str_or_none(
4226 get_first(video_details, 'channelId')
4227 or get_first(microformats, 'externalChannelId')
4228 or search_meta('channelId')))
4229 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4230
4231 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4232 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4233 if not duration and live_end_time and live_start_time:
4234 duration = live_end_time - live_start_time
4235
4236 needs_live_processing = self._needs_live_processing(live_status, duration)
4237
4238 def is_bad_format(fmt):
4239 if needs_live_processing and not fmt.get('is_from_start'):
4240 return True
4241 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4242 and fmt.get('protocol') == 'http_dash_segments'):
4243 return True
4244
4245 for fmt in filter(is_bad_format, formats):
4246 fmt['preference'] = (fmt.get('preference') or -1) - 10
4247 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4248
4249 if needs_live_processing:
4250 self._prepare_live_from_start_formats(
4251 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4252
4253 formats.extend(self._extract_storyboard(player_responses, duration))
4254
4255 channel_handle = self.handle_from_url(owner_profile_url)
4256
4257 info = {
4258 'id': video_id,
4259 'title': video_title,
4260 'formats': formats,
4261 'thumbnails': thumbnails,
4262 # The best thumbnail that we are sure exists. Prevents unnecessary
4263 # URL checking if user don't care about getting the best possible thumbnail
4264 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4265 'description': video_description,
4266 'channel_id': channel_id,
4267 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4268 'duration': duration,
4269 'view_count': int_or_none(
4270 get_first((video_details, microformats), (..., 'viewCount'))
4271 or search_meta('interactionCount')),
4272 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4273 'age_limit': 18 if (
4274 get_first(microformats, 'isFamilySafe') is False
4275 or search_meta('isFamilyFriendly') == 'false'
4276 or search_meta('og:restrictions:age') == '18+') else 0,
4277 'webpage_url': webpage_url,
4278 'categories': [category] if category else None,
4279 'tags': keywords,
4280 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4281 'live_status': live_status,
4282 'release_timestamp': live_start_time,
4283 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4284 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4285 }
4286
4287 subtitles = {}
4288 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4289 if pctr:
4290 def get_lang_code(track):
4291 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4292 or track.get('languageCode'))
4293
4294 # Converted into dicts to remove duplicates
4295 captions = {
4296 get_lang_code(sub): sub
4297 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4298 translation_languages = {
4299 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4300 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4301
4302 def process_language(container, base_url, lang_code, sub_name, query):
4303 lang_subs = container.setdefault(lang_code, [])
4304 for fmt in self._SUBTITLE_FORMATS:
4305 query.update({
4306 'fmt': fmt,
4307 })
4308 lang_subs.append({
4309 'ext': fmt,
4310 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4311 'name': sub_name,
4312 })
4313
4314 # NB: Constructing the full subtitle dictionary is slow
4315 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4316 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4317 for lang_code, caption_track in captions.items():
4318 base_url = caption_track.get('baseUrl')
4319 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4320 if not base_url:
4321 continue
4322 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4323 if caption_track.get('kind') != 'asr':
4324 if not lang_code:
4325 continue
4326 process_language(
4327 subtitles, base_url, lang_code, lang_name, {})
4328 if not caption_track.get('isTranslatable'):
4329 continue
4330 for trans_code, trans_name in translation_languages.items():
4331 if not trans_code:
4332 continue
4333 orig_trans_code = trans_code
4334 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4335 if not get_translated_subs:
4336 continue
4337 trans_code += f'-{lang_code}'
4338 trans_name += format_field(lang_name, None, ' from %s')
4339 if lang_code == f'a-{orig_trans_code}':
4340 # Set audio language based on original subtitles
4341 for f in formats:
4342 if f.get('acodec') != 'none' and not f.get('language'):
4343 f['language'] = orig_trans_code
4344 # Add an "-orig" label to the original language so that it can be distinguished.
4345 # The subs are returned without "-orig" as well for compatibility
4346 process_language(
4347 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4348 # Setting tlang=lang returns damaged subtitles.
4349 process_language(automatic_captions, base_url, trans_code, trans_name,
4350 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4351
4352 info['automatic_captions'] = automatic_captions
4353 info['subtitles'] = subtitles
4354
4355 parsed_url = urllib.parse.urlparse(url)
4356 for component in [parsed_url.fragment, parsed_url.query]:
4357 query = urllib.parse.parse_qs(component)
4358 for k, v in query.items():
4359 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4360 d_k += '_time'
4361 if d_k not in info and k in s_ks:
4362 info[d_k] = parse_duration(query[k][0])
4363
4364 # Youtube Music Auto-generated description
4365 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4366 # XXX: Causes catastrophic backtracking if description has "·"
4367 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4368 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4369 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4370 mobj = re.search(
4371 r'''(?xs)
4372 (?=(?P<track>[^\n·]+))(?P=track)·
4373 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4374 (?=(?P<album>[^\n]+))(?P=album)\n
4375 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4376 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4377 (.+?\nArtist\s*:\s*
4378 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4379 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4380 ''', video_description)
4381 if mobj:
4382 release_year = mobj.group('release_year')
4383 release_date = mobj.group('release_date')
4384 if release_date:
4385 release_date = release_date.replace('-', '')
4386 if not release_year:
4387 release_year = release_date[:4]
4388 info.update({
4389 'album': mobj.group('album'.strip()),
4390 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4391 'track': mobj.group('track').strip(),
4392 'release_date': release_date,
4393 'release_year': int_or_none(release_year),
4394 })
4395
4396 initial_data = None
4397 if webpage:
4398 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4399 if not traverse_obj(initial_data, 'contents'):
4400 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4401 initial_data = None
4402 if not initial_data:
4403 query = {'videoId': video_id}
4404 query.update(self._get_checkok_params())
4405 initial_data = self._extract_response(
4406 item_id=video_id, ep='next', fatal=False,
4407 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4408 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4409 note='Downloading initial data API JSON')
4410
4411 info['comment_count'] = traverse_obj(initial_data, (
4412 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4413 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4414 ), (
4415 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4416 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4417 ), expected_type=self._get_count, get_all=False)
4418
4419 try: # This will error if there is no livechat
4420 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4421 except (KeyError, IndexError, TypeError):
4422 pass
4423 else:
4424 info.setdefault('subtitles', {})['live_chat'] = [{
4425 # url is needed to set cookies
4426 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4427 'video_id': video_id,
4428 'ext': 'json',
4429 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4430 else 'youtube_live_chat_replay'),
4431 }]
4432
4433 if initial_data:
4434 info['chapters'] = (
4435 self._extract_chapters_from_json(initial_data, duration)
4436 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4437 or self._extract_chapters_from_description(video_description, duration)
4438 or None)
4439
4440 info['heatmap'] = self._extract_heatmap(initial_data)
4441
4442 contents = traverse_obj(
4443 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4444 expected_type=list, default=[])
4445
4446 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4447 if vpir:
4448 stl = vpir.get('superTitleLink')
4449 if stl:
4450 stl = self._get_text(stl)
4451 if try_get(
4452 vpir,
4453 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4454 info['location'] = stl
4455 else:
4456 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4457 if mobj:
4458 info.update({
4459 'series': mobj.group(1),
4460 'season_number': int(mobj.group(2)),
4461 'episode_number': int(mobj.group(3)),
4462 })
4463 for tlb in (try_get(
4464 vpir,
4465 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4466 list) or []):
4467 tbrs = variadic(
4468 traverse_obj(
4469 tlb, ('toggleButtonRenderer', ...),
4470 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4471 for tbr in tbrs:
4472 for getter, regex in [(
4473 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4474 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4475 lambda x: x['accessibility'],
4476 lambda x: x['accessibilityData']['accessibilityData'],
4477 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4478 label = (try_get(tbr, getter, dict) or {}).get('label')
4479 if label:
4480 mobj = re.match(regex, label)
4481 if mobj:
4482 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4483 break
4484 sbr_tooltip = try_get(
4485 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4486 if sbr_tooltip:
4487 like_count, dislike_count = sbr_tooltip.split(' / ')
4488 info.update({
4489 'like_count': str_to_int(like_count),
4490 'dislike_count': str_to_int(dislike_count),
4491 })
4492 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4493 if vcr:
4494 vc = self._get_count(vcr, 'viewCount')
4495 # Upcoming premieres with waiting count are treated as live here
4496 if vcr.get('isLive'):
4497 info['concurrent_view_count'] = vc
4498 elif info.get('view_count') is None:
4499 info['view_count'] = vc
4500
4501 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4502 if vsir:
4503 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4504 info.update({
4505 'channel': self._get_text(vor, 'title'),
4506 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4507
4508 if not channel_handle:
4509 channel_handle = self.handle_from_url(
4510 traverse_obj(vor, (
4511 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4512 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4513 {str}), get_all=False))
4514
4515 rows = try_get(
4516 vsir,
4517 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4518 list) or []
4519 multiple_songs = False
4520 for row in rows:
4521 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4522 multiple_songs = True
4523 break
4524 for row in rows:
4525 mrr = row.get('metadataRowRenderer') or {}
4526 mrr_title = mrr.get('title')
4527 if not mrr_title:
4528 continue
4529 mrr_title = self._get_text(mrr, 'title')
4530 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4531 if mrr_title == 'License':
4532 info['license'] = mrr_contents_text
4533 elif not multiple_songs:
4534 if mrr_title == 'Album':
4535 info['album'] = mrr_contents_text
4536 elif mrr_title == 'Artist':
4537 info['artist'] = mrr_contents_text
4538 elif mrr_title == 'Song':
4539 info['track'] = mrr_contents_text
4540 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4541 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4542 info['channel_is_verified'] = True
4543
4544 info.update({
4545 'uploader': info.get('channel'),
4546 'uploader_id': channel_handle,
4547 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4548 })
4549 # The upload date for scheduled, live and past live streams / premieres in microformats
4550 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4551 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4552 upload_date = (
4553 unified_strdate(get_first(microformats, 'uploadDate'))
4554 or unified_strdate(search_meta('uploadDate')))
4555 if not upload_date or (
4556 live_status in ('not_live', None)
4557 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4558 ):
4559 upload_date = strftime_or_none(
4560 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4561 info['upload_date'] = upload_date
4562
4563 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4564 v = info.get(s_k)
4565 if v:
4566 info[d_k] = v
4567
4568 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4569
4570 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4571 or get_first(video_details, 'isPrivate', expected_type=bool))
4572
4573 info['availability'] = (
4574 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4575 else self._availability(
4576 is_private=is_private,
4577 needs_premium=(
4578 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4579 or False if initial_data and is_private is not None else None),
4580 needs_subscription=(
4581 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4582 or False if initial_data and is_private is not None else None),
4583 needs_auth=info['age_limit'] >= 18,
4584 is_unlisted=None if is_private is None else (
4585 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4586 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4587
4588 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4589
4590 self.mark_watched(video_id, player_responses)
4591
4592 return info
4593
4594
4595 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4596 @staticmethod
4597 def passthrough_smuggled_data(func):
4598 def _smuggle(info, smuggled_data):
4599 if info.get('_type') not in ('url', 'url_transparent'):
4600 return info
4601 if smuggled_data.get('is_music_url'):
4602 parsed_url = urllib.parse.urlparse(info['url'])
4603 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4604 smuggled_data.pop('is_music_url')
4605 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4606 if smuggled_data:
4607 info['url'] = smuggle_url(info['url'], smuggled_data)
4608 return info
4609
4610 @functools.wraps(func)
4611 def wrapper(self, url):
4612 url, smuggled_data = unsmuggle_url(url, {})
4613 if self.is_music_url(url):
4614 smuggled_data['is_music_url'] = True
4615 info_dict = func(self, url, smuggled_data)
4616 if smuggled_data:
4617 _smuggle(info_dict, smuggled_data)
4618 if info_dict.get('entries'):
4619 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4620 return info_dict
4621 return wrapper
4622
4623 @staticmethod
4624 def _extract_basic_item_renderer(item):
4625 # Modified from _extract_grid_item_renderer
4626 known_basic_renderers = (
4627 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4628 )
4629 for key, renderer in item.items():
4630 if not isinstance(renderer, dict):
4631 continue
4632 elif key in known_basic_renderers:
4633 return renderer
4634 elif key.startswith('grid') and key.endswith('Renderer'):
4635 return renderer
4636
4637 def _extract_channel_renderer(self, renderer):
4638 channel_id = self.ucid_or_none(renderer['channelId'])
4639 title = self._get_text(renderer, 'title')
4640 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4641 channel_handle = self.handle_from_url(
4642 traverse_obj(renderer, (
4643 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4644 ('browseEndpoint', 'canonicalBaseUrl')),
4645 {str}), get_all=False))
4646 if not channel_handle:
4647 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
4648 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
4649 return {
4650 '_type': 'url',
4651 'url': channel_url,
4652 'id': channel_id,
4653 'ie_key': YoutubeTabIE.ie_key(),
4654 'channel': title,
4655 'uploader': title,
4656 'channel_id': channel_id,
4657 'channel_url': channel_url,
4658 'title': title,
4659 'uploader_id': channel_handle,
4660 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4661 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
4662 # However, in feed/channels this is set correctly to the subscriber count
4663 'channel_follower_count': traverse_obj(
4664 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
4665 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4666 'playlist_count': (
4667 # videoCountText may be the subscriber count
4668 self._get_count(renderer, 'videoCountText')
4669 if self._get_count(renderer, 'subscriberCountText') is not None else None),
4670 'description': self._get_text(renderer, 'descriptionSnippet'),
4671 'channel_is_verified': True if self._has_badge(
4672 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
4673 }
4674
4675 def _grid_entries(self, grid_renderer):
4676 for item in grid_renderer['items']:
4677 if not isinstance(item, dict):
4678 continue
4679 renderer = self._extract_basic_item_renderer(item)
4680 if not isinstance(renderer, dict):
4681 continue
4682 title = self._get_text(renderer, 'title')
4683
4684 # playlist
4685 playlist_id = renderer.get('playlistId')
4686 if playlist_id:
4687 yield self.url_result(
4688 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4689 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4690 video_title=title)
4691 continue
4692 # video
4693 video_id = renderer.get('videoId')
4694 if video_id:
4695 yield self._extract_video(renderer)
4696 continue
4697 # channel
4698 channel_id = renderer.get('channelId')
4699 if channel_id:
4700 yield self._extract_channel_renderer(renderer)
4701 continue
4702 # generic endpoint URL support
4703 ep_url = urljoin('https://www.youtube.com/', try_get(
4704 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4705 str))
4706 if ep_url:
4707 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4708 if ie.suitable(ep_url):
4709 yield self.url_result(
4710 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4711 break
4712
4713 def _music_reponsive_list_entry(self, renderer):
4714 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4715 if video_id:
4716 title = traverse_obj(renderer, (
4717 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4718 'text', 'runs', 0, 'text'))
4719 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4720 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4721 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4722 if playlist_id:
4723 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4724 if video_id:
4725 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4726 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4727 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4728 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4729 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4730 if browse_id:
4731 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4732 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4733
4734 def _shelf_entries_from_content(self, shelf_renderer):
4735 content = shelf_renderer.get('content')
4736 if not isinstance(content, dict):
4737 return
4738 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4739 if renderer:
4740 # TODO: add support for nested playlists so each shelf is processed
4741 # as separate playlist
4742 # TODO: this includes only first N items
4743 yield from self._grid_entries(renderer)
4744 renderer = content.get('horizontalListRenderer')
4745 if renderer:
4746 # TODO
4747 pass
4748
4749 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4750 ep = try_get(
4751 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4752 str)
4753 shelf_url = urljoin('https://www.youtube.com', ep)
4754 if shelf_url:
4755 # Skipping links to another channels, note that checking for
4756 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4757 # will not work
4758 if skip_channels and '/channels?' in shelf_url:
4759 return
4760 title = self._get_text(shelf_renderer, 'title')
4761 yield self.url_result(shelf_url, video_title=title)
4762 # Shelf may not contain shelf URL, fallback to extraction from content
4763 yield from self._shelf_entries_from_content(shelf_renderer)
4764
4765 def _playlist_entries(self, video_list_renderer):
4766 for content in video_list_renderer['contents']:
4767 if not isinstance(content, dict):
4768 continue
4769 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4770 if not isinstance(renderer, dict):
4771 continue
4772 video_id = renderer.get('videoId')
4773 if not video_id:
4774 continue
4775 yield self._extract_video(renderer)
4776
4777 def _rich_entries(self, rich_grid_renderer):
4778 renderer = traverse_obj(
4779 rich_grid_renderer,
4780 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4781 video_id = renderer.get('videoId')
4782 if video_id:
4783 yield self._extract_video(renderer)
4784 return
4785 playlist_id = renderer.get('playlistId')
4786 if playlist_id:
4787 yield self.url_result(
4788 f'https://www.youtube.com/playlist?list={playlist_id}',
4789 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4790 video_title=self._get_text(renderer, 'title'))
4791 return
4792
4793 def _video_entry(self, video_renderer):
4794 video_id = video_renderer.get('videoId')
4795 if video_id:
4796 return self._extract_video(video_renderer)
4797
4798 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4799 url = urljoin('https://youtube.com', traverse_obj(
4800 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4801 if url:
4802 return self.url_result(
4803 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4804
4805 def _post_thread_entries(self, post_thread_renderer):
4806 post_renderer = try_get(
4807 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4808 if not post_renderer:
4809 return
4810 # video attachment
4811 video_renderer = try_get(
4812 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4813 video_id = video_renderer.get('videoId')
4814 if video_id:
4815 entry = self._extract_video(video_renderer)
4816 if entry:
4817 yield entry
4818 # playlist attachment
4819 playlist_id = try_get(
4820 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4821 if playlist_id:
4822 yield self.url_result(
4823 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4824 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4825 # inline video links
4826 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4827 for run in runs:
4828 if not isinstance(run, dict):
4829 continue
4830 ep_url = try_get(
4831 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4832 if not ep_url:
4833 continue
4834 if not YoutubeIE.suitable(ep_url):
4835 continue
4836 ep_video_id = YoutubeIE._match_id(ep_url)
4837 if video_id == ep_video_id:
4838 continue
4839 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4840
4841 def _post_thread_continuation_entries(self, post_thread_continuation):
4842 contents = post_thread_continuation.get('contents')
4843 if not isinstance(contents, list):
4844 return
4845 for content in contents:
4846 renderer = content.get('backstagePostThreadRenderer')
4847 if isinstance(renderer, dict):
4848 yield from self._post_thread_entries(renderer)
4849 continue
4850 renderer = content.get('videoRenderer')
4851 if isinstance(renderer, dict):
4852 yield self._video_entry(renderer)
4853
4854 r''' # unused
4855 def _rich_grid_entries(self, contents):
4856 for content in contents:
4857 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4858 if video_renderer:
4859 entry = self._video_entry(video_renderer)
4860 if entry:
4861 yield entry
4862 '''
4863
4864 def _report_history_entries(self, renderer):
4865 for url in traverse_obj(renderer, (
4866 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4867 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4868 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4869 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4870
4871 def _extract_entries(self, parent_renderer, continuation_list):
4872 # continuation_list is modified in-place with continuation_list = [continuation_token]
4873 continuation_list[:] = [None]
4874 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4875 for content in contents:
4876 if not isinstance(content, dict):
4877 continue
4878 is_renderer = traverse_obj(
4879 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4880 expected_type=dict)
4881 if not is_renderer:
4882 if content.get('richItemRenderer'):
4883 for entry in self._rich_entries(content['richItemRenderer']):
4884 yield entry
4885 continuation_list[0] = self._extract_continuation(parent_renderer)
4886 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4887 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4888 yield from self._report_history_entries(table)
4889 continuation_list[0] = self._extract_continuation(table)
4890 continue
4891
4892 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4893 for isr_content in isr_contents:
4894 if not isinstance(isr_content, dict):
4895 continue
4896
4897 known_renderers = {
4898 'playlistVideoListRenderer': self._playlist_entries,
4899 'gridRenderer': self._grid_entries,
4900 'reelShelfRenderer': self._grid_entries,
4901 'shelfRenderer': self._shelf_entries,
4902 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4903 'backstagePostThreadRenderer': self._post_thread_entries,
4904 'videoRenderer': lambda x: [self._video_entry(x)],
4905 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4906 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4907 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
4908 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
4909 }
4910 for key, renderer in isr_content.items():
4911 if key not in known_renderers:
4912 continue
4913 for entry in known_renderers[key](renderer):
4914 if entry:
4915 yield entry
4916 continuation_list[0] = self._extract_continuation(renderer)
4917 break
4918
4919 if not continuation_list[0]:
4920 continuation_list[0] = self._extract_continuation(is_renderer)
4921
4922 if not continuation_list[0]:
4923 continuation_list[0] = self._extract_continuation(parent_renderer)
4924
4925 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4926 continuation_list = [None]
4927 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4928 tab_content = try_get(tab, lambda x: x['content'], dict)
4929 if not tab_content:
4930 return
4931 parent_renderer = (
4932 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4933 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4934 yield from extract_entries(parent_renderer)
4935 continuation = continuation_list[0]
4936 seen_continuations = set()
4937 for page_num in itertools.count(1):
4938 if not continuation:
4939 break
4940 continuation_token = continuation.get('continuation')
4941 if continuation_token is not None and continuation_token in seen_continuations:
4942 self.write_debug('Detected YouTube feed looping - assuming end of feed.')
4943 break
4944 seen_continuations.add(continuation_token)
4945 headers = self.generate_api_headers(
4946 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4947 response = self._extract_response(
4948 item_id=f'{item_id} page {page_num}',
4949 query=continuation, headers=headers, ytcfg=ytcfg,
4950 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4951
4952 if not response:
4953 break
4954 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4955 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4956 visitor_data = self._extract_visitor_data(response) or visitor_data
4957
4958 known_renderers = {
4959 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4960 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4961 'gridVideoRenderer': (self._grid_entries, 'items'),
4962 'gridChannelRenderer': (self._grid_entries, 'items'),
4963 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4964 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4965 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4966 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4967 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4968 'playlistVideoListContinuation': (self._playlist_entries, None),
4969 'gridContinuation': (self._grid_entries, None),
4970 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4971 'sectionListContinuation': (extract_entries, None), # for feeds
4972 }
4973
4974 continuation_items = traverse_obj(response, (
4975 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4976 'appendContinuationItemsAction', 'continuationItems'
4977 ), 'continuationContents', get_all=False)
4978 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4979
4980 video_items_renderer = None
4981 for key in continuation_item.keys():
4982 if key not in known_renderers:
4983 continue
4984 func, parent_key = known_renderers[key]
4985 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4986 continuation_list = [None]
4987 yield from func(video_items_renderer)
4988 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4989
4990 if not video_items_renderer:
4991 break
4992
4993 @staticmethod
4994 def _extract_selected_tab(tabs, fatal=True):
4995 for tab_renderer in tabs:
4996 if tab_renderer.get('selected'):
4997 return tab_renderer
4998 if fatal:
4999 raise ExtractorError('Unable to find selected tab')
5000
5001 @staticmethod
5002 def _extract_tab_renderers(response):
5003 return traverse_obj(
5004 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
5005
5006 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
5007 metadata = self._extract_metadata_from_tabs(item_id, data)
5008
5009 selected_tab = self._extract_selected_tab(tabs)
5010 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
5011 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
5012
5013 return self.playlist_result(
5014 self._entries(
5015 selected_tab, metadata['id'], ytcfg,
5016 self._extract_account_syncid(ytcfg, data),
5017 self._extract_visitor_data(data, ytcfg)),
5018 **metadata)
5019
5020 def _extract_metadata_from_tabs(self, item_id, data):
5021 info = {'id': item_id}
5022
5023 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
5024 if metadata_renderer:
5025 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
5026 ('channelUrl', {self.ucid_from_url}))
5027 info.update({
5028 'channel': metadata_renderer.get('title'),
5029 'channel_id': channel_id,
5030 })
5031 if info['channel_id']:
5032 info['id'] = info['channel_id']
5033 else:
5034 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5035
5036 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5037 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5038 def _get_uncropped(url):
5039 return url_or_none((url or '').split('=')[0] + '=s0')
5040
5041 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5042 if avatar_thumbnails:
5043 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5044 if uncropped_avatar:
5045 avatar_thumbnails.append({
5046 'url': uncropped_avatar,
5047 'id': 'avatar_uncropped',
5048 'preference': 1
5049 })
5050
5051 channel_banners = self._extract_thumbnails(
5052 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5053 for banner in channel_banners:
5054 banner['preference'] = -10
5055
5056 if channel_banners:
5057 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5058 if uncropped_banner:
5059 channel_banners.append({
5060 'url': uncropped_banner,
5061 'id': 'banner_uncropped',
5062 'preference': -5
5063 })
5064
5065 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5066 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5067 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5068
5069 primary_thumbnails = self._extract_thumbnails(
5070 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5071 playlist_thumbnails = self._extract_thumbnails(
5072 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5073
5074 info.update({
5075 'title': (traverse_obj(metadata_renderer, 'title')
5076 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5077 or info['id']),
5078 'availability': self._extract_availability(data),
5079 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
5080 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5081 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
5082 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5083 })
5084
5085 channel_handle = (
5086 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5087 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5088
5089 if channel_handle:
5090 info.update({
5091 'uploader_id': channel_handle,
5092 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5093 })
5094
5095 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5096 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5097 info['channel_is_verified'] = True
5098 # Playlist stats is a text runs array containing [video count, view count, last updated].
5099 # last updated or (view count and last updated) may be missing.
5100 playlist_stats = get_first(
5101 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5102
5103 last_updated_unix = self._parse_time_text(
5104 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5105 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5106 info['modified_date'] = strftime_or_none(last_updated_unix)
5107
5108 info['view_count'] = self._get_count(playlist_stats, 1)
5109 if info['view_count'] is None: # 0 is allowed
5110 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5111 if info['view_count'] is None:
5112 info['view_count'] = self._get_count(data, (
5113 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5114 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5115
5116 info['playlist_count'] = self._get_count(playlist_stats, 0)
5117 if info['playlist_count'] is None: # 0 is allowed
5118 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5119
5120 if not info.get('channel_id'):
5121 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5122 if not owner: # Deprecated
5123 owner = traverse_obj(
5124 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5125 ('videoOwner', 'videoOwnerRenderer', 'title'))
5126 owner_text = self._get_text(owner)
5127 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5128 info.update({
5129 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5130 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5131 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
5132 })
5133
5134 info.update({
5135 'uploader': info['channel'],
5136 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5137 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5138 })
5139
5140 return info
5141
5142 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5143 first_id = last_id = response = None
5144 for page_num in itertools.count(1):
5145 videos = list(self._playlist_entries(playlist))
5146 if not videos:
5147 return
5148 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5149 if start >= len(videos):
5150 return
5151 yield from videos[start:]
5152 first_id = first_id or videos[0]['id']
5153 last_id = videos[-1]['id']
5154 watch_endpoint = try_get(
5155 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5156 headers = self.generate_api_headers(
5157 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5158 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5159 query = {
5160 'playlistId': playlist_id,
5161 'videoId': watch_endpoint.get('videoId') or last_id,
5162 'index': watch_endpoint.get('index') or len(videos),
5163 'params': watch_endpoint.get('params') or 'OAE%3D'
5164 }
5165 response = self._extract_response(
5166 item_id='%s page %d' % (playlist_id, page_num),
5167 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5168 check_get_keys='contents'
5169 )
5170 playlist = try_get(
5171 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5172
5173 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5174 title = playlist.get('title') or try_get(
5175 data, lambda x: x['titleText']['simpleText'], str)
5176 playlist_id = playlist.get('playlistId') or item_id
5177
5178 # Delegating everything except mix playlists to regular tab-based playlist URL
5179 playlist_url = urljoin(url, try_get(
5180 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5181 str))
5182
5183 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5184 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5185 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5186
5187 if playlist_url and playlist_url != url and not is_known_unviewable:
5188 return self.url_result(
5189 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5190 video_title=title)
5191
5192 return self.playlist_result(
5193 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5194 playlist_id=playlist_id, playlist_title=title)
5195
5196 def _extract_availability(self, data):
5197 """
5198 Gets the availability of a given playlist/tab.
5199 Note: Unless YouTube tells us explicitly, we do not assume it is public
5200 @param data: response
5201 """
5202 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5203 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5204 player_header_privacy = playlist_header_renderer.get('privacy')
5205
5206 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5207
5208 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5209 privacy_setting_icon = get_first(
5210 (playlist_header_renderer, sidebar_renderer),
5211 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5212 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5213 expected_type=str)
5214
5215 microformats_is_unlisted = traverse_obj(
5216 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5217
5218 return (
5219 'public' if (
5220 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5221 or player_header_privacy == 'PUBLIC'
5222 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5223 else self._availability(
5224 is_private=(
5225 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5226 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5227 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5228 is_unlisted=(
5229 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5230 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5231 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5232 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5233 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5234 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5235 needs_auth=False))
5236
5237 @staticmethod
5238 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5239 sidebar_renderer = try_get(
5240 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5241 for item in sidebar_renderer:
5242 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5243 if renderer:
5244 return renderer
5245
5246 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5247 """
5248 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5249 """
5250 is_playlist = bool(traverse_obj(
5251 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5252 if not is_playlist:
5253 return
5254 headers = self.generate_api_headers(
5255 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5256 visitor_data=self._extract_visitor_data(data, ytcfg))
5257 query = {
5258 'params': 'wgYCCAA=',
5259 'browseId': f'VL{item_id}'
5260 }
5261 return self._extract_response(
5262 item_id=item_id, headers=headers, query=query,
5263 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5264 note='Redownloading playlist API JSON with unavailable videos')
5265
5266 @functools.cached_property
5267 def skip_webpage(self):
5268 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5269
5270 def _extract_webpage(self, url, item_id, fatal=True):
5271 webpage, data = None, None
5272 for retry in self.RetryManager(fatal=fatal):
5273 try:
5274 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5275 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5276 except ExtractorError as e:
5277 if isinstance(e.cause, network_exceptions):
5278 if not isinstance(e.cause, HTTPError) or e.cause.status not in (403, 429):
5279 retry.error = e
5280 continue
5281 self._error_or_warning(e, fatal=fatal)
5282 break
5283
5284 try:
5285 self._extract_and_report_alerts(data)
5286 except ExtractorError as e:
5287 self._error_or_warning(e, fatal=fatal)
5288 break
5289
5290 # Sometimes youtube returns a webpage with incomplete ytInitialData
5291 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5292 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5293 retry.error = ExtractorError('Incomplete yt initial data received')
5294 continue
5295
5296 return webpage, data
5297
5298 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5299 """Use if failed to extract ytcfg (and data) from initial webpage"""
5300 if not ytcfg and self.is_authenticated:
5301 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5302 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5303 raise ExtractorError(
5304 f'{msg}. If you are not downloading private content, or '
5305 'your cookies are only for the first account and channel,'
5306 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5307 expected=True)
5308 self.report_warning(msg, only_once=True)
5309
5310 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5311 data = None
5312 if not self.skip_webpage:
5313 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5314 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5315 # Reject webpage data if redirected to home page without explicitly requesting
5316 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5317 if (url != 'https://www.youtube.com/feed/recommended'
5318 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5319 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5320 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5321 if fatal:
5322 raise ExtractorError(msg, expected=True)
5323 self.report_warning(msg, only_once=True)
5324 if not data:
5325 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5326 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5327 return data, ytcfg
5328
5329 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5330 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5331 resolve_response = self._extract_response(
5332 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5333 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5334 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5335 for ep_key, ep in endpoints.items():
5336 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5337 if params:
5338 return self._extract_response(
5339 item_id=item_id, query=params, ep=ep, headers=headers,
5340 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5341 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5342 err_note = 'Failed to resolve url (does the playlist exist?)'
5343 if fatal:
5344 raise ExtractorError(err_note, expected=True)
5345 self.report_warning(err_note, item_id)
5346
5347 _SEARCH_PARAMS = None
5348
5349 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5350 data = {'query': query}
5351 if params is NO_DEFAULT:
5352 params = self._SEARCH_PARAMS
5353 if params:
5354 data['params'] = params
5355
5356 content_keys = (
5357 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5358 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5359 # ytmusic search
5360 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5361 ('continuationContents', ),
5362 )
5363 display_id = f'query "{query}"'
5364 check_get_keys = tuple({keys[0] for keys in content_keys})
5365 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5366 self._report_playlist_authcheck(ytcfg, fatal=False)
5367
5368 continuation_list = [None]
5369 search = None
5370 for page_num in itertools.count(1):
5371 data.update(continuation_list[0] or {})
5372 headers = self.generate_api_headers(
5373 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5374 search = self._extract_response(
5375 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5376 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5377 slr_contents = traverse_obj(search, *content_keys)
5378 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5379 if not continuation_list[0]:
5380 break
5381
5382
5383 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5384 IE_DESC = 'YouTube Tabs'
5385 _VALID_URL = r'''(?x:
5386 https?://
5387 (?!consent\.)(?:\w+\.)?
5388 (?:
5389 youtube(?:kids)?\.com|
5390 %(invidious)s
5391 )/
5392 (?:
5393 (?P<channel_type>channel|c|user|browse)/|
5394 (?P<not_channel>
5395 feed/|hashtag/|
5396 (?:playlist|watch)\?.*?\blist=
5397 )|
5398 (?!(?:%(reserved_names)s)\b) # Direct URLs
5399 )
5400 (?P<id>[^/?\#&]+)
5401 )''' % {
5402 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5403 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5404 }
5405 IE_NAME = 'youtube:tab'
5406
5407 _TESTS = [{
5408 'note': 'playlists, multipage',
5409 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5410 'playlist_mincount': 94,
5411 'info_dict': {
5412 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5413 'title': 'Igor Kleiner - Playlists',
5414 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5415 'uploader': 'Igor Kleiner',
5416 'uploader_id': '@IgorDataScience',
5417 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5418 'channel': 'Igor Kleiner',
5419 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5420 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5421 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5422 'channel_follower_count': int
5423 },
5424 }, {
5425 'note': 'playlists, multipage, different order',
5426 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5427 'playlist_mincount': 94,
5428 'info_dict': {
5429 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5430 'title': 'Igor Kleiner - Playlists',
5431 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5432 'uploader': 'Igor Kleiner',
5433 'uploader_id': '@IgorDataScience',
5434 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5435 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5436 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5437 'channel': 'Igor Kleiner',
5438 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5439 'channel_follower_count': int
5440 },
5441 }, {
5442 'note': 'playlists, series',
5443 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5444 'playlist_mincount': 5,
5445 'info_dict': {
5446 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5447 'title': '3Blue1Brown - Playlists',
5448 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5449 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5450 'channel': '3Blue1Brown',
5451 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5452 'uploader_id': '@3blue1brown',
5453 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5454 'uploader': '3Blue1Brown',
5455 'tags': ['Mathematics'],
5456 'channel_follower_count': int,
5457 'channel_is_verified': True,
5458 },
5459 }, {
5460 'note': 'playlists, singlepage',
5461 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5462 'playlist_mincount': 4,
5463 'info_dict': {
5464 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5465 'title': 'ThirstForScience - Playlists',
5466 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5467 'uploader': 'ThirstForScience',
5468 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5469 'uploader_id': '@ThirstForScience',
5470 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5471 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5472 'tags': 'count:13',
5473 'channel': 'ThirstForScience',
5474 'channel_follower_count': int
5475 }
5476 }, {
5477 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5478 'only_matching': True,
5479 }, {
5480 'note': 'basic, single video playlist',
5481 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5482 'info_dict': {
5483 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5484 'title': 'youtube-dl public playlist',
5485 'description': '',
5486 'tags': [],
5487 'view_count': int,
5488 'modified_date': '20201130',
5489 'channel': 'Sergey M.',
5490 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5491 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5492 'availability': 'public',
5493 'uploader': 'Sergey M.',
5494 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5495 'uploader_id': '@sergeym.6173',
5496 },
5497 'playlist_count': 1,
5498 }, {
5499 'note': 'empty playlist',
5500 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5501 'info_dict': {
5502 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5503 'title': 'youtube-dl empty playlist',
5504 'tags': [],
5505 'channel': 'Sergey M.',
5506 'description': '',
5507 'modified_date': '20160902',
5508 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5509 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5510 'availability': 'public',
5511 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5512 'uploader_id': '@sergeym.6173',
5513 'uploader': 'Sergey M.',
5514 },
5515 'playlist_count': 0,
5516 }, {
5517 'note': 'Home tab',
5518 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5519 'info_dict': {
5520 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5521 'title': 'lex will - Home',
5522 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5523 'uploader': 'lex will',
5524 'uploader_id': '@lexwill718',
5525 'channel': 'lex will',
5526 'tags': ['bible', 'history', 'prophesy'],
5527 'uploader_url': 'https://www.youtube.com/@lexwill718',
5528 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5529 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5530 'channel_follower_count': int
5531 },
5532 'playlist_mincount': 2,
5533 }, {
5534 'note': 'Videos tab',
5535 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5536 'info_dict': {
5537 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5538 'title': 'lex will - Videos',
5539 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5540 'uploader': 'lex will',
5541 'uploader_id': '@lexwill718',
5542 'tags': ['bible', 'history', 'prophesy'],
5543 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5544 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5545 'uploader_url': 'https://www.youtube.com/@lexwill718',
5546 'channel': 'lex will',
5547 'channel_follower_count': int
5548 },
5549 'playlist_mincount': 975,
5550 }, {
5551 'note': 'Videos tab, sorted by popular',
5552 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5553 'info_dict': {
5554 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5555 'title': 'lex will - Videos',
5556 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5557 'uploader': 'lex will',
5558 'uploader_id': '@lexwill718',
5559 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5560 'uploader_url': 'https://www.youtube.com/@lexwill718',
5561 'channel': 'lex will',
5562 'tags': ['bible', 'history', 'prophesy'],
5563 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5564 'channel_follower_count': int
5565 },
5566 'playlist_mincount': 199,
5567 }, {
5568 'note': 'Playlists tab',
5569 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5570 'info_dict': {
5571 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5572 'title': 'lex will - Playlists',
5573 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5574 'uploader': 'lex will',
5575 'uploader_id': '@lexwill718',
5576 'uploader_url': 'https://www.youtube.com/@lexwill718',
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 },
5583 'playlist_mincount': 17,
5584 }, {
5585 'note': 'Community tab',
5586 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5587 'info_dict': {
5588 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5589 'title': 'lex will - Community',
5590 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5591 'channel': 'lex will',
5592 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5593 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5594 'tags': ['bible', 'history', 'prophesy'],
5595 'channel_follower_count': int,
5596 'uploader_url': 'https://www.youtube.com/@lexwill718',
5597 'uploader_id': '@lexwill718',
5598 'uploader': 'lex will',
5599 },
5600 'playlist_mincount': 18,
5601 }, {
5602 'note': 'Channels tab',
5603 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5604 'info_dict': {
5605 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5606 'title': 'lex will - Channels',
5607 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5608 'channel': 'lex will',
5609 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5610 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5611 'tags': ['bible', 'history', 'prophesy'],
5612 'channel_follower_count': int,
5613 'uploader_url': 'https://www.youtube.com/@lexwill718',
5614 'uploader_id': '@lexwill718',
5615 'uploader': 'lex will',
5616 },
5617 'playlist_mincount': 12,
5618 }, {
5619 'note': 'Search tab',
5620 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5621 'playlist_mincount': 40,
5622 'info_dict': {
5623 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5624 'title': '3Blue1Brown - Search - linear algebra',
5625 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5626 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5627 'tags': ['Mathematics'],
5628 'channel': '3Blue1Brown',
5629 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5630 'channel_follower_count': int,
5631 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5632 'uploader_id': '@3blue1brown',
5633 'uploader': '3Blue1Brown',
5634 'channel_is_verified': True,
5635 },
5636 }, {
5637 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5638 'only_matching': True,
5639 }, {
5640 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5641 'only_matching': True,
5642 }, {
5643 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5644 'only_matching': True,
5645 }, {
5646 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5647 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5648 'info_dict': {
5649 'title': '29C3: Not my department',
5650 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5651 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5652 'tags': [],
5653 'view_count': int,
5654 'modified_date': '20150605',
5655 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5656 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5657 'channel': 'Christiaan008',
5658 'availability': 'public',
5659 'uploader_id': '@ChRiStIaAn008',
5660 'uploader': 'Christiaan008',
5661 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5662 },
5663 'playlist_count': 96,
5664 }, {
5665 'note': 'Large playlist',
5666 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5667 'info_dict': {
5668 'title': 'Uploads from Cauchemar',
5669 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5670 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5671 'tags': [],
5672 'modified_date': r're:\d{8}',
5673 'channel': 'Cauchemar',
5674 'view_count': int,
5675 'description': '',
5676 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5677 'availability': 'public',
5678 'uploader_id': '@Cauchemar89',
5679 'uploader': 'Cauchemar',
5680 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5681 },
5682 'playlist_mincount': 1123,
5683 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5684 }, {
5685 'note': 'even larger playlist, 8832 videos',
5686 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5687 'only_matching': True,
5688 }, {
5689 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5690 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5691 'info_dict': {
5692 'title': 'Uploads from Interstellar Movie',
5693 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5694 'tags': [],
5695 'view_count': int,
5696 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5697 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5698 'channel': 'Interstellar Movie',
5699 'description': '',
5700 'modified_date': r're:\d{8}',
5701 'availability': 'public',
5702 'uploader_id': '@InterstellarMovie',
5703 'uploader': 'Interstellar Movie',
5704 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5705 },
5706 'playlist_mincount': 21,
5707 }, {
5708 'note': 'Playlist with "show unavailable videos" button',
5709 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5710 'info_dict': {
5711 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5712 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5713 'view_count': int,
5714 'channel': 'Phim Siêu Nhân Nhật Bản',
5715 'tags': [],
5716 'description': '',
5717 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5718 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5719 'modified_date': r're:\d{8}',
5720 'availability': 'public',
5721 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5722 'uploader_id': '@phimsieunhannhatban',
5723 'uploader': 'Phim Siêu Nhân Nhật Bản',
5724 },
5725 'playlist_mincount': 200,
5726 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5727 }, {
5728 'note': 'Playlist with unavailable videos in page 7',
5729 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5730 'info_dict': {
5731 'title': 'Uploads from BlankTV',
5732 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5733 'channel': 'BlankTV',
5734 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5735 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5736 'view_count': int,
5737 'tags': [],
5738 'modified_date': r're:\d{8}',
5739 'description': '',
5740 'availability': 'public',
5741 'uploader_id': '@blanktv',
5742 'uploader': 'BlankTV',
5743 'uploader_url': 'https://www.youtube.com/@blanktv',
5744 },
5745 'playlist_mincount': 1000,
5746 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5747 }, {
5748 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5749 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5750 'info_dict': {
5751 'title': 'Data Analysis with Dr Mike Pound',
5752 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5753 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5754 'tags': [],
5755 'view_count': int,
5756 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5757 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5758 'channel': 'Computerphile',
5759 'availability': 'public',
5760 'modified_date': '20190712',
5761 'uploader_id': '@Computerphile',
5762 'uploader': 'Computerphile',
5763 'uploader_url': 'https://www.youtube.com/@Computerphile',
5764 },
5765 'playlist_mincount': 11,
5766 }, {
5767 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5768 'only_matching': True,
5769 }, {
5770 'note': 'Playlist URL that does not actually serve a playlist',
5771 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5772 'info_dict': {
5773 'id': 'FqZTN594JQw',
5774 'ext': 'webm',
5775 'title': "Smiley's People 01 detective, Adventure Series, Action",
5776 'upload_date': '20150526',
5777 'license': 'Standard YouTube License',
5778 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5779 'categories': ['People & Blogs'],
5780 'tags': list,
5781 'view_count': int,
5782 'like_count': int,
5783 },
5784 'params': {
5785 'skip_download': True,
5786 },
5787 'skip': 'This video is not available.',
5788 'add_ie': [YoutubeIE.ie_key()],
5789 }, {
5790 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5791 'only_matching': True,
5792 }, {
5793 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5794 'only_matching': True,
5795 }, {
5796 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5797 'info_dict': {
5798 'id': 'hGkQjiJLjWQ', # This will keep changing
5799 'ext': 'mp4',
5800 'title': str,
5801 'upload_date': r're:\d{8}',
5802 'description': str,
5803 'categories': ['News & Politics'],
5804 'tags': list,
5805 'like_count': int,
5806 'release_timestamp': int,
5807 'channel': 'Sky News',
5808 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5809 'age_limit': 0,
5810 'view_count': int,
5811 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5812 'playable_in_embed': True,
5813 'release_date': r're:\d+',
5814 'availability': 'public',
5815 'live_status': 'is_live',
5816 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5817 'channel_follower_count': int,
5818 'concurrent_view_count': int,
5819 'uploader_url': 'https://www.youtube.com/@SkyNews',
5820 'uploader_id': '@SkyNews',
5821 'uploader': 'Sky News',
5822 'channel_is_verified': True,
5823 },
5824 'params': {
5825 'skip_download': True,
5826 },
5827 'expected_warnings': ['Ignoring subtitle tracks found in '],
5828 }, {
5829 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5830 'info_dict': {
5831 'id': 'a48o2S1cPoo',
5832 'ext': 'mp4',
5833 'title': 'The Young Turks - Live Main Show',
5834 'upload_date': '20150715',
5835 'license': 'Standard YouTube License',
5836 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5837 'categories': ['News & Politics'],
5838 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5839 'like_count': int,
5840 },
5841 'params': {
5842 'skip_download': True,
5843 },
5844 'only_matching': True,
5845 }, {
5846 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5847 'only_matching': True,
5848 }, {
5849 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5850 'only_matching': True,
5851 }, {
5852 'note': 'A channel that is not live. Should raise error',
5853 'url': 'https://www.youtube.com/user/numberphile/live',
5854 'only_matching': True,
5855 }, {
5856 'url': 'https://www.youtube.com/feed/trending',
5857 'only_matching': True,
5858 }, {
5859 'url': 'https://www.youtube.com/feed/library',
5860 'only_matching': True,
5861 }, {
5862 'url': 'https://www.youtube.com/feed/history',
5863 'only_matching': True,
5864 }, {
5865 'url': 'https://www.youtube.com/feed/subscriptions',
5866 'only_matching': True,
5867 }, {
5868 'url': 'https://www.youtube.com/feed/watch_later',
5869 'only_matching': True,
5870 }, {
5871 'note': 'Recommended - redirects to home page.',
5872 'url': 'https://www.youtube.com/feed/recommended',
5873 'only_matching': True,
5874 }, {
5875 'note': 'inline playlist with not always working continuations',
5876 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5877 'only_matching': True,
5878 }, {
5879 'url': 'https://www.youtube.com/course',
5880 'only_matching': True,
5881 }, {
5882 'url': 'https://www.youtube.com/zsecurity',
5883 'only_matching': True,
5884 }, {
5885 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5886 'only_matching': True,
5887 }, {
5888 'url': 'https://www.youtube.com/TheYoungTurks/live',
5889 'only_matching': True,
5890 }, {
5891 'url': 'https://www.youtube.com/hashtag/cctv9',
5892 'info_dict': {
5893 'id': 'cctv9',
5894 'title': '#cctv9',
5895 'tags': [],
5896 },
5897 'playlist_mincount': 300, # not consistent but should be over 300
5898 }, {
5899 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5900 'only_matching': True,
5901 }, {
5902 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5903 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5904 'only_matching': True
5905 }, {
5906 'note': '/browse/ should redirect to /channel/',
5907 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5908 'only_matching': True
5909 }, {
5910 'note': 'VLPL, should redirect to playlist?list=PL...',
5911 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5912 'info_dict': {
5913 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5914 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5915 'title': 'NCS : All Releases 💿',
5916 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5917 'modified_date': r're:\d{8}',
5918 'view_count': int,
5919 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5920 'tags': [],
5921 'channel': 'NoCopyrightSounds',
5922 'availability': 'public',
5923 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5924 'uploader': 'NoCopyrightSounds',
5925 'uploader_id': '@NoCopyrightSounds',
5926 },
5927 'playlist_mincount': 166,
5928 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5929 }, {
5930 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5931 'note': 'Topic, should redirect to playlist?list=UU...',
5932 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5933 'info_dict': {
5934 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5935 'title': 'Uploads from Royalty Free Music - Topic',
5936 'tags': [],
5937 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5938 'channel': 'Royalty Free Music - Topic',
5939 'view_count': int,
5940 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5941 'modified_date': r're:\d{8}',
5942 'description': '',
5943 'availability': 'public',
5944 'uploader': 'Royalty Free Music - Topic',
5945 },
5946 'playlist_mincount': 101,
5947 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5948 }, {
5949 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5950 # Treat as a general feed
5951 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5952 'info_dict': {
5953 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5954 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5955 'tags': [],
5956 },
5957 'playlist_mincount': 9,
5958 }, {
5959 'note': 'Youtube music Album',
5960 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5961 'info_dict': {
5962 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5963 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5964 'tags': [],
5965 'view_count': int,
5966 'description': '',
5967 'availability': 'unlisted',
5968 'modified_date': r're:\d{8}',
5969 },
5970 'playlist_count': 50,
5971 'expected_warnings': ['YouTube Music is not directly supported'],
5972 }, {
5973 'note': 'unlisted single video playlist',
5974 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5975 'info_dict': {
5976 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5977 'title': 'yt-dlp unlisted playlist test',
5978 'availability': 'unlisted',
5979 'tags': [],
5980 'modified_date': '20220418',
5981 'channel': 'colethedj',
5982 'view_count': int,
5983 'description': '',
5984 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5985 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5986 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5987 'uploader_id': '@colethedj1894',
5988 'uploader': 'colethedj',
5989 },
5990 'playlist': [{
5991 'info_dict': {
5992 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
5993 'id': 'BaW_jenozKc',
5994 '_type': 'url',
5995 'ie_key': 'Youtube',
5996 'duration': 10,
5997 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
5998 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
5999 'view_count': int,
6000 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
6001 'channel': 'Philipp Hagemeister',
6002 'uploader_id': '@PhilippHagemeister',
6003 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
6004 'uploader': 'Philipp Hagemeister',
6005 }
6006 }],
6007 'playlist_count': 1,
6008 'params': {'extract_flat': True},
6009 }, {
6010 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
6011 'url': 'https://www.youtube.com/feed/recommended',
6012 'info_dict': {
6013 'id': 'recommended',
6014 'title': 'recommended',
6015 'tags': [],
6016 },
6017 'playlist_mincount': 50,
6018 'params': {
6019 'skip_download': True,
6020 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6021 },
6022 }, {
6023 'note': 'API Fallback: /videos tab, sorted by oldest first',
6024 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
6025 'info_dict': {
6026 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6027 'title': 'Cody\'sLab - Videos',
6028 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6029 'channel': 'Cody\'sLab',
6030 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6031 'tags': [],
6032 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6033 'channel_follower_count': int
6034 },
6035 'playlist_mincount': 650,
6036 'params': {
6037 'skip_download': True,
6038 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6039 },
6040 'skip': 'Query for sorting no longer works',
6041 }, {
6042 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6043 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6044 'info_dict': {
6045 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6046 'title': 'Uploads from Royalty Free Music - Topic',
6047 'modified_date': r're:\d{8}',
6048 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6049 'description': '',
6050 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6051 'tags': [],
6052 'channel': 'Royalty Free Music - Topic',
6053 'view_count': int,
6054 'availability': 'public',
6055 'uploader': 'Royalty Free Music - Topic',
6056 },
6057 'playlist_mincount': 101,
6058 'params': {
6059 'skip_download': True,
6060 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
6061 },
6062 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6063 }, {
6064 'note': 'non-standard redirect to regional channel',
6065 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6066 'only_matching': True
6067 }, {
6068 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6069 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6070 'info_dict': {
6071 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6072 'modified_date': '20220407',
6073 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6074 'tags': [],
6075 'availability': 'unlisted',
6076 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6077 'channel': 'pukkandan',
6078 'description': 'Test for collaborative playlist',
6079 'title': 'yt-dlp test - collaborative playlist',
6080 'view_count': int,
6081 'uploader_url': 'https://www.youtube.com/@pukkandan',
6082 'uploader_id': '@pukkandan',
6083 'uploader': 'pukkandan',
6084 },
6085 'playlist_mincount': 2
6086 }, {
6087 'note': 'translated tab name',
6088 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6089 'info_dict': {
6090 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6091 'tags': [],
6092 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6093 'description': 'test description',
6094 'title': 'cole-dlp-test-acc - 再生リスト',
6095 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6096 'channel': 'cole-dlp-test-acc',
6097 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6098 'uploader_id': '@coletdjnz',
6099 'uploader': 'cole-dlp-test-acc',
6100 },
6101 'playlist_mincount': 1,
6102 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6103 'expected_warnings': ['Preferring "ja"'],
6104 }, {
6105 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6106 'note': 'preferred lang set with playlist with translated video titles',
6107 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6108 'info_dict': {
6109 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6110 'tags': [],
6111 'view_count': int,
6112 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6113 'channel': 'cole-dlp-test-acc',
6114 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6115 'description': 'test',
6116 'title': 'dlp test playlist',
6117 'availability': 'public',
6118 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6119 'uploader_id': '@coletdjnz',
6120 'uploader': 'cole-dlp-test-acc',
6121 },
6122 'playlist_mincount': 1,
6123 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6124 'expected_warnings': ['Preferring "ja"'],
6125 }, {
6126 # shorts audio pivot for 2GtVksBMYFM.
6127 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6128 'info_dict': {
6129 'id': 'sfv_audio_pivot',
6130 'title': 'sfv_audio_pivot',
6131 'tags': [],
6132 },
6133 'playlist_mincount': 50,
6134
6135 }, {
6136 # Channel with a real live tab (not to be mistaken with streams tab)
6137 # Do not treat like it should redirect to live stream
6138 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6139 'info_dict': {
6140 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6141 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6142 'tags': [],
6143 },
6144 'playlist_mincount': 20,
6145 }, {
6146 # Tab name is not the same as tab id
6147 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6148 'info_dict': {
6149 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6150 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6151 'tags': [],
6152 },
6153 'playlist_mincount': 8,
6154 }, {
6155 # Home tab id is literally home. Not to get mistaken with featured
6156 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6157 'info_dict': {
6158 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6159 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6160 'tags': [],
6161 },
6162 'playlist_mincount': 8,
6163 }, {
6164 # Should get three playlists for videos, shorts and streams tabs
6165 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6166 'info_dict': {
6167 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6168 'title': 'Polka Ch. 尾丸ポルカ',
6169 'channel_follower_count': int,
6170 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6171 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6172 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6173 'channel': 'Polka Ch. 尾丸ポルカ',
6174 'tags': 'count:35',
6175 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6176 'uploader': 'Polka Ch. 尾丸ポルカ',
6177 'uploader_id': '@OmaruPolka',
6178 },
6179 'playlist_count': 3,
6180 }, {
6181 # Shorts tab with channel with handle
6182 # TODO: fix channel description
6183 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6184 'info_dict': {
6185 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6186 'title': 'Not Just Bikes - Shorts',
6187 'tags': 'count:12',
6188 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6189 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6190 'channel_follower_count': int,
6191 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6192 'channel': 'Not Just Bikes',
6193 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6194 'uploader': 'Not Just Bikes',
6195 'uploader_id': '@NotJustBikes',
6196 },
6197 'playlist_mincount': 10,
6198 }, {
6199 # Streams tab
6200 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6201 'info_dict': {
6202 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6203 'title': '中村悠一 - Live',
6204 'tags': 'count:7',
6205 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6206 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6207 'channel': '中村悠一',
6208 'channel_follower_count': int,
6209 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6210 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6211 'uploader_id': '@Yuichi-Nakamura',
6212 'uploader': '中村悠一',
6213 },
6214 'playlist_mincount': 60,
6215 }, {
6216 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6217 # See test_youtube_lists
6218 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6219 'only_matching': True,
6220 }, {
6221 # No uploads and no UCID given. Should fail with no uploads error
6222 # See test_youtube_lists
6223 'url': 'https://www.youtube.com/news',
6224 'only_matching': True
6225 }, {
6226 # No videos tab but has a shorts tab
6227 'url': 'https://www.youtube.com/c/TKFShorts',
6228 'info_dict': {
6229 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6230 'title': 'Shorts Break - Shorts',
6231 'tags': 'count:48',
6232 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6233 'channel': 'Shorts Break',
6234 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6235 'channel_follower_count': int,
6236 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6237 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6238 'uploader': 'Shorts Break',
6239 'uploader_id': '@ShortsBreak_Official',
6240 },
6241 'playlist_mincount': 30,
6242 }, {
6243 # Trending Now Tab. tab id is empty
6244 'url': 'https://www.youtube.com/feed/trending',
6245 'info_dict': {
6246 'id': 'trending',
6247 'title': 'trending - Now',
6248 'tags': [],
6249 },
6250 'playlist_mincount': 30,
6251 }, {
6252 # Trending Gaming Tab. tab id is empty
6253 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6254 'info_dict': {
6255 'id': 'trending',
6256 'title': 'trending - Gaming',
6257 'tags': [],
6258 },
6259 'playlist_mincount': 30,
6260 }, {
6261 # Shorts url result in shorts tab
6262 # TODO: Fix channel id extraction
6263 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6264 'info_dict': {
6265 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6266 'title': 'cole-dlp-test-acc - Shorts',
6267 'channel': 'cole-dlp-test-acc',
6268 'description': 'test description',
6269 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6270 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6271 'tags': [],
6272 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6273 'uploader_id': '@coletdjnz',
6274 'uploader': 'cole-dlp-test-acc',
6275 },
6276 'playlist': [{
6277 'info_dict': {
6278 # Channel data is not currently available for short renderers (as of 2023-03-01)
6279 '_type': 'url',
6280 'ie_key': 'Youtube',
6281 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6282 'id': 'sSM9J5YH_60',
6283 'title': 'SHORT short',
6284 'view_count': int,
6285 'thumbnails': list,
6286 }
6287 }],
6288 'params': {'extract_flat': True},
6289 }, {
6290 # Live video status should be extracted
6291 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6292 'info_dict': {
6293 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6294 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6295 'tags': []
6296 },
6297 'playlist': [{
6298 'info_dict': {
6299 '_type': 'url',
6300 'ie_key': 'Youtube',
6301 'url': 'startswith:https://www.youtube.com/watch?v=',
6302 'id': str,
6303 'title': str,
6304 'live_status': 'is_live',
6305 'channel_id': str,
6306 'channel_url': str,
6307 'concurrent_view_count': int,
6308 'channel': str,
6309 'uploader': str,
6310 'uploader_url': str,
6311 'uploader_id': str,
6312 'channel_is_verified': bool, # this will keep changing
6313 }
6314 }],
6315 'params': {'extract_flat': True, 'playlist_items': '1'},
6316 'playlist_mincount': 1
6317 }, {
6318 # Channel renderer metadata. Contains number of videos on the channel
6319 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6320 'info_dict': {
6321 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6322 'title': 'cole-dlp-test-acc - Channels',
6323 'channel': 'cole-dlp-test-acc',
6324 'description': 'test description',
6325 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6326 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6327 'tags': [],
6328 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6329 'uploader_id': '@coletdjnz',
6330 'uploader': 'cole-dlp-test-acc',
6331 },
6332 'playlist': [{
6333 'info_dict': {
6334 '_type': 'url',
6335 'ie_key': 'YoutubeTab',
6336 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6337 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6338 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6339 'title': 'PewDiePie',
6340 'channel': 'PewDiePie',
6341 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6342 'thumbnails': list,
6343 'channel_follower_count': int,
6344 'playlist_count': int,
6345 'uploader': 'PewDiePie',
6346 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6347 'uploader_id': '@PewDiePie',
6348 'channel_is_verified': True,
6349 }
6350 }],
6351 'params': {'extract_flat': True},
6352 }, {
6353 'url': 'https://www.youtube.com/@3blue1brown/about',
6354 'info_dict': {
6355 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6356 'tags': ['Mathematics'],
6357 'title': '3Blue1Brown - About',
6358 'channel_follower_count': int,
6359 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6360 'channel': '3Blue1Brown',
6361 'view_count': int,
6362 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6363 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6364 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6365 'uploader_id': '@3blue1brown',
6366 'uploader': '3Blue1Brown',
6367 'channel_is_verified': True,
6368 },
6369 'playlist_count': 0,
6370 }, {
6371 # Podcasts tab, with rich entry playlistRenderers
6372 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6373 'info_dict': {
6374 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6375 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6376 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6377 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6378 'title': '99 Percent Invisible - Podcasts',
6379 'uploader': '99 Percent Invisible',
6380 'channel_follower_count': int,
6381 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6382 'tags': [],
6383 'channel': '99 Percent Invisible',
6384 'uploader_id': '@99percentinvisiblepodcast',
6385 },
6386 'playlist_count': 1,
6387 }, {
6388 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6389 'url': 'https://www.youtube.com/@AHimitsu/releases',
6390 'info_dict': {
6391 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6392 'channel': 'A Himitsu',
6393 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6394 'title': 'A Himitsu - Releases',
6395 'uploader_id': '@AHimitsu',
6396 'uploader': 'A Himitsu',
6397 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6398 'tags': 'count:16',
6399 'description': 'I make music',
6400 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6401 'channel_follower_count': int,
6402 'channel_is_verified': True,
6403 },
6404 'playlist_mincount': 10,
6405 }, {
6406 # Playlist with only shorts, shown as reel renderers
6407 # FIXME: future: YouTube currently doesn't give continuation for this,
6408 # may do in future.
6409 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
6410 'info_dict': {
6411 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
6412 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
6413 'view_count': int,
6414 'uploader_id': '@BangyShorts',
6415 'description': '',
6416 'uploader_url': 'https://www.youtube.com/@BangyShorts',
6417 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
6418 'channel': 'Bangy Shorts',
6419 'uploader': 'Bangy Shorts',
6420 'tags': [],
6421 'availability': 'public',
6422 'modified_date': '20230626',
6423 'title': 'Uploads from Bangy Shorts',
6424 },
6425 'playlist_mincount': 100,
6426 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6427 }]
6428
6429 @classmethod
6430 def suitable(cls, url):
6431 return False if YoutubeIE.suitable(url) else super().suitable(url)
6432
6433 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6434
6435 def _get_url_mobj(self, url):
6436 mobj = self._URL_RE.match(url).groupdict()
6437 mobj.update((k, '') for k, v in mobj.items() if v is None)
6438 return mobj
6439
6440 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6441 tab_name = (tab.get('title') or '').lower()
6442 tab_url = urljoin(base_url, traverse_obj(
6443 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6444
6445 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6446 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6447 if tab_id:
6448 return {
6449 'TAB_ID_SPONSORSHIPS': 'membership',
6450 }.get(tab_id, tab_id), tab_name
6451
6452 # Fallback to tab name if we cannot get the tab id.
6453 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6454 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6455 if tab_name:
6456 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6457 return {
6458 'home': 'featured',
6459 'live': 'streams',
6460 }.get(tab_name, tab_name), tab_name
6461
6462 def _has_tab(self, tabs, tab_id):
6463 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6464
6465 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6466 def _real_extract(self, url, smuggled_data):
6467 item_id = self._match_id(url)
6468 url = urllib.parse.urlunparse(
6469 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6470 compat_opts = self.get_param('compat_opts', [])
6471
6472 mobj = self._get_url_mobj(url)
6473 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6474 if is_channel and smuggled_data.get('is_music_url'):
6475 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6476 return self.url_result(
6477 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6478 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6479 mdata = self._extract_tab_endpoint(
6480 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6481 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6482 get_all=False, expected_type=str)
6483 if not murl:
6484 raise ExtractorError('Failed to resolve album to playlist')
6485 return self.url_result(murl, YoutubeTabIE)
6486 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6487 return self.url_result(
6488 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6489
6490 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6491 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6492 url = f'{pre}/videos{post}'
6493 if smuggled_data.get('is_music_url'):
6494 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6495
6496 # Handle both video/playlist URLs
6497 qs = parse_qs(url)
6498 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6499 if not video_id and mobj['not_channel'].startswith('watch'):
6500 if not playlist_id:
6501 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6502 raise ExtractorError('A video URL was given without video ID', expected=True)
6503 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6504 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6505 return self.url_result(
6506 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6507
6508 if not self._yes_playlist(playlist_id, video_id):
6509 return self.url_result(
6510 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6511
6512 data, ytcfg = self._extract_data(url, display_id)
6513
6514 # YouTube may provide a non-standard redirect to the regional channel
6515 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6516 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6517 redirect_url = traverse_obj(
6518 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6519 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6520 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6521 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6522 return self.url_result(redirect_url, YoutubeTabIE)
6523
6524 tabs, extra_tabs = self._extract_tab_renderers(data), []
6525 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6526 selected_tab = self._extract_selected_tab(tabs)
6527 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6528 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6529
6530 if not original_tab_id and selected_tab_name:
6531 self.to_screen('Downloading all uploads of the channel. '
6532 'To download only the videos in a specific tab, pass the tab\'s URL')
6533 if self._has_tab(tabs, 'streams'):
6534 extra_tabs.append(''.join((pre, '/streams', post)))
6535 if self._has_tab(tabs, 'shorts'):
6536 extra_tabs.append(''.join((pre, '/shorts', post)))
6537 # XXX: Members-only tab should also be extracted
6538
6539 if not extra_tabs and selected_tab_id != 'videos':
6540 # Channel does not have streams, shorts or videos tabs
6541 if item_id[:2] != 'UC':
6542 raise ExtractorError('This channel has no uploads', expected=True)
6543
6544 # Topic channels don't have /videos. Use the equivalent playlist instead
6545 pl_id = f'UU{item_id[2:]}'
6546 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6547 try:
6548 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6549 except ExtractorError:
6550 raise ExtractorError('This channel has no uploads', expected=True)
6551 else:
6552 item_id, url = pl_id, pl_url
6553 self.to_screen(
6554 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6555
6556 elif extra_tabs and selected_tab_id != 'videos':
6557 # When there are shorts/live tabs but not videos tab
6558 url, data = f'{pre}{post}', None
6559
6560 elif (original_tab_id or 'videos') != selected_tab_id:
6561 if original_tab_id == 'live':
6562 # Live tab should have redirected to the video
6563 # Except in the case the channel has an actual live tab
6564 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6565 raise UserNotLive(video_id=item_id)
6566 elif selected_tab_name:
6567 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6568
6569 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6570 url = f'{pre}{post}'
6571
6572 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6573 if 'no-youtube-unavailable-videos' not in compat_opts:
6574 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6575 self._extract_and_report_alerts(data, only_once=True)
6576
6577 tabs, entries = self._extract_tab_renderers(data), []
6578 if tabs:
6579 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6580 entries[0].update({
6581 'extractor_key': YoutubeTabIE.ie_key(),
6582 'extractor': YoutubeTabIE.IE_NAME,
6583 'webpage_url': url,
6584 })
6585 if self.get_param('playlist_items') == '0':
6586 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6587 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6588 entries.extend(map(self._real_extract, extra_tabs))
6589
6590 if len(entries) == 1:
6591 return entries[0]
6592 elif entries:
6593 metadata = self._extract_metadata_from_tabs(item_id, data)
6594 uploads_url = 'the Uploads (UU) playlist URL'
6595 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6596 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6597 self.to_screen(
6598 'Downloading as multiple playlists, separated by tabs. '
6599 f'To download as a single playlist instead, pass {uploads_url}')
6600 return self.playlist_result(entries, item_id, **metadata)
6601
6602 # Inline playlist
6603 playlist = traverse_obj(
6604 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6605 if playlist:
6606 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6607
6608 video_id = traverse_obj(
6609 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6610 if video_id:
6611 if tab != '/live': # live tab is expected to redirect to video
6612 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6613 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6614
6615 raise ExtractorError('Unable to recognize tab page')
6616
6617
6618 class YoutubePlaylistIE(InfoExtractor):
6619 IE_DESC = 'YouTube playlists'
6620 _VALID_URL = r'''(?x)(?:
6621 (?:https?://)?
6622 (?:\w+\.)?
6623 (?:
6624 (?:
6625 youtube(?:kids)?\.com|
6626 %(invidious)s
6627 )
6628 /.*?\?.*?\blist=
6629 )?
6630 (?P<id>%(playlist_id)s)
6631 )''' % {
6632 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6633 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6634 }
6635 IE_NAME = 'youtube:playlist'
6636 _TESTS = [{
6637 'note': 'issue #673',
6638 'url': 'PLBB231211A4F62143',
6639 'info_dict': {
6640 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6641 'id': 'PLBB231211A4F62143',
6642 'uploader': 'Wickman',
6643 'uploader_id': '@WickmanVT',
6644 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6645 'view_count': int,
6646 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6647 'modified_date': r're:\d{8}',
6648 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6649 'channel': 'Wickman',
6650 'tags': [],
6651 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6652 'availability': 'public',
6653 },
6654 'playlist_mincount': 29,
6655 }, {
6656 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6657 'info_dict': {
6658 'title': 'YDL_safe_search',
6659 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6660 },
6661 'playlist_count': 2,
6662 'skip': 'This playlist is private',
6663 }, {
6664 'note': 'embedded',
6665 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6666 'playlist_count': 4,
6667 'info_dict': {
6668 'title': 'JODA15',
6669 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6670 'uploader': 'milan',
6671 'uploader_id': '@milan5503',
6672 'description': '',
6673 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6674 'tags': [],
6675 'modified_date': '20140919',
6676 'view_count': int,
6677 'channel': 'milan',
6678 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6679 'uploader_url': 'https://www.youtube.com/@milan5503',
6680 'availability': 'public',
6681 },
6682 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6683 }, {
6684 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6685 'playlist_mincount': 455,
6686 'info_dict': {
6687 'title': '2018 Chinese New Singles (11/6 updated)',
6688 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6689 'uploader': 'LBK',
6690 'uploader_id': '@music_king',
6691 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6692 'channel': 'LBK',
6693 'view_count': int,
6694 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6695 'tags': [],
6696 'uploader_url': 'https://www.youtube.com/@music_king',
6697 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6698 'modified_date': r're:\d{8}',
6699 'availability': 'public',
6700 },
6701 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6702 }, {
6703 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6704 'only_matching': True,
6705 }, {
6706 # music album playlist
6707 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6708 'only_matching': True,
6709 }]
6710
6711 @classmethod
6712 def suitable(cls, url):
6713 if YoutubeTabIE.suitable(url):
6714 return False
6715 from ..utils import parse_qs
6716 qs = parse_qs(url)
6717 if qs.get('v', [None])[0]:
6718 return False
6719 return super().suitable(url)
6720
6721 def _real_extract(self, url):
6722 playlist_id = self._match_id(url)
6723 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6724 url = update_url_query(
6725 'https://www.youtube.com/playlist',
6726 parse_qs(url) or {'list': playlist_id})
6727 if is_music_url:
6728 url = smuggle_url(url, {'is_music_url': True})
6729 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6730
6731
6732 class YoutubeYtBeIE(InfoExtractor):
6733 IE_DESC = 'youtu.be'
6734 _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}
6735 _TESTS = [{
6736 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6737 'info_dict': {
6738 'id': 'yeWKywCrFtk',
6739 'ext': 'mp4',
6740 'title': 'Small Scale Baler and Braiding Rugs',
6741 'uploader': 'Backus-Page House Museum',
6742 'uploader_id': '@backuspagemuseum',
6743 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6744 'upload_date': '20161008',
6745 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6746 'categories': ['Nonprofits & Activism'],
6747 'tags': list,
6748 'like_count': int,
6749 'age_limit': 0,
6750 'playable_in_embed': True,
6751 'thumbnail': r're:^https?://.*\.webp',
6752 'channel': 'Backus-Page House Museum',
6753 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6754 'live_status': 'not_live',
6755 'view_count': int,
6756 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6757 'availability': 'public',
6758 'duration': 59,
6759 'comment_count': int,
6760 'channel_follower_count': int
6761 },
6762 'params': {
6763 'noplaylist': True,
6764 'skip_download': True,
6765 },
6766 }, {
6767 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6768 'only_matching': True,
6769 }]
6770
6771 def _real_extract(self, url):
6772 mobj = self._match_valid_url(url)
6773 video_id = mobj.group('id')
6774 playlist_id = mobj.group('playlist_id')
6775 return self.url_result(
6776 update_url_query('https://www.youtube.com/watch', {
6777 'v': video_id,
6778 'list': playlist_id,
6779 'feature': 'youtu.be',
6780 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6781
6782
6783 class YoutubeLivestreamEmbedIE(InfoExtractor):
6784 IE_DESC = 'YouTube livestream embeds'
6785 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6786 _TESTS = [{
6787 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6788 'only_matching': True,
6789 }]
6790
6791 def _real_extract(self, url):
6792 channel_id = self._match_id(url)
6793 return self.url_result(
6794 f'https://www.youtube.com/channel/{channel_id}/live',
6795 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6796
6797
6798 class YoutubeYtUserIE(InfoExtractor):
6799 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6800 IE_NAME = 'youtube:user'
6801 _VALID_URL = r'ytuser:(?P<id>.+)'
6802 _TESTS = [{
6803 'url': 'ytuser:phihag',
6804 'only_matching': True,
6805 }]
6806
6807 def _real_extract(self, url):
6808 user_id = self._match_id(url)
6809 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6810
6811
6812 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6813 IE_NAME = 'youtube:favorites'
6814 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6815 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6816 _LOGIN_REQUIRED = True
6817 _TESTS = [{
6818 'url': ':ytfav',
6819 'only_matching': True,
6820 }, {
6821 'url': ':ytfavorites',
6822 'only_matching': True,
6823 }]
6824
6825 def _real_extract(self, url):
6826 return self.url_result(
6827 'https://www.youtube.com/playlist?list=LL',
6828 ie=YoutubeTabIE.ie_key())
6829
6830
6831 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6832 IE_NAME = 'youtube:notif'
6833 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6834 _VALID_URL = r':ytnotif(?:ication)?s?'
6835 _LOGIN_REQUIRED = True
6836 _TESTS = [{
6837 'url': ':ytnotif',
6838 'only_matching': True,
6839 }, {
6840 'url': ':ytnotifications',
6841 'only_matching': True,
6842 }]
6843
6844 def _extract_notification_menu(self, response, continuation_list):
6845 notification_list = traverse_obj(
6846 response,
6847 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6848 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6849 expected_type=list) or []
6850 continuation_list[0] = None
6851 for item in notification_list:
6852 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6853 if entry:
6854 yield entry
6855 continuation = item.get('continuationItemRenderer')
6856 if continuation:
6857 continuation_list[0] = continuation
6858
6859 def _extract_notification_renderer(self, notification):
6860 video_id = traverse_obj(
6861 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6862 url = f'https://www.youtube.com/watch?v={video_id}'
6863 channel_id = None
6864 if not video_id:
6865 browse_ep = traverse_obj(
6866 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6867 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6868 post_id = self._search_regex(
6869 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6870 'post id', default=None)
6871 if not channel_id or not post_id:
6872 return
6873 # The direct /post url redirects to this in the browser
6874 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6875
6876 channel = traverse_obj(
6877 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6878 expected_type=str)
6879 notification_title = self._get_text(notification, 'shortMessage')
6880 if notification_title:
6881 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6882 # TODO: handle recommended videos
6883 title = self._search_regex(
6884 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6885 'video title', default=None)
6886 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6887 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6888 else None)
6889 return {
6890 '_type': 'url',
6891 'url': url,
6892 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6893 'video_id': video_id,
6894 'title': title,
6895 'channel_id': channel_id,
6896 'channel': channel,
6897 'uploader': channel,
6898 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6899 'timestamp': timestamp,
6900 }
6901
6902 def _notification_menu_entries(self, ytcfg):
6903 continuation_list = [None]
6904 response = None
6905 for page in itertools.count(1):
6906 ctoken = traverse_obj(
6907 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6908 response = self._extract_response(
6909 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6910 ep='notification/get_notification_menu', check_get_keys='actions',
6911 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6912 yield from self._extract_notification_menu(response, continuation_list)
6913 if not continuation_list[0]:
6914 break
6915
6916 def _real_extract(self, url):
6917 display_id = 'notifications'
6918 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6919 self._report_playlist_authcheck(ytcfg)
6920 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6921
6922
6923 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6924 IE_DESC = 'YouTube search'
6925 IE_NAME = 'youtube:search'
6926 _SEARCH_KEY = 'ytsearch'
6927 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6928 _TESTS = [{
6929 'url': 'ytsearch5:youtube-dl test video',
6930 'playlist_count': 5,
6931 'info_dict': {
6932 'id': 'youtube-dl test video',
6933 'title': 'youtube-dl test video',
6934 }
6935 }]
6936
6937
6938 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6939 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6940 _SEARCH_KEY = 'ytsearchdate'
6941 IE_DESC = 'YouTube search, newest videos first'
6942 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6943 _TESTS = [{
6944 'url': 'ytsearchdate5:youtube-dl test video',
6945 'playlist_count': 5,
6946 'info_dict': {
6947 'id': 'youtube-dl test video',
6948 'title': 'youtube-dl test video',
6949 }
6950 }]
6951
6952
6953 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6954 IE_DESC = 'YouTube search URLs with sorting and filter support'
6955 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6956 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6957 _TESTS = [{
6958 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6959 'playlist_mincount': 5,
6960 'info_dict': {
6961 'id': 'youtube-dl test video',
6962 'title': 'youtube-dl test video',
6963 }
6964 }, {
6965 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6966 'playlist_mincount': 5,
6967 'info_dict': {
6968 'id': 'python',
6969 'title': 'python',
6970 }
6971 }, {
6972 'url': 'https://www.youtube.com/results?search_query=%23cats',
6973 'playlist_mincount': 1,
6974 'info_dict': {
6975 'id': '#cats',
6976 'title': '#cats',
6977 # The test suite does not have support for nested playlists
6978 # 'entries': [{
6979 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6980 # 'title': '#cats',
6981 # }],
6982 },
6983 }, {
6984 # Channel results
6985 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
6986 'info_dict': {
6987 'id': 'kurzgesagt',
6988 'title': 'kurzgesagt',
6989 },
6990 'playlist': [{
6991 'info_dict': {
6992 '_type': 'url',
6993 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6994 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6995 'ie_key': 'YoutubeTab',
6996 'channel': 'Kurzgesagt – In a Nutshell',
6997 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
6998 'title': 'Kurzgesagt – In a Nutshell',
6999 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7000 # No longer available for search as it is set to the handle.
7001 # 'playlist_count': int,
7002 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7003 'thumbnails': list,
7004 'uploader_id': '@kurzgesagt',
7005 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
7006 'uploader': 'Kurzgesagt – In a Nutshell',
7007 'channel_is_verified': True,
7008 'channel_follower_count': int,
7009 }
7010 }],
7011 'params': {'extract_flat': True, 'playlist_items': '1'},
7012 'playlist_mincount': 1,
7013 }, {
7014 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
7015 'only_matching': True,
7016 }]
7017
7018 def _real_extract(self, url):
7019 qs = parse_qs(url)
7020 query = (qs.get('search_query') or qs.get('q'))[0]
7021 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
7022
7023
7024 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
7025 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
7026 IE_NAME = 'youtube:music:search_url'
7027 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7028 _TESTS = [{
7029 'url': 'https://music.youtube.com/search?q=royalty+free+music',
7030 'playlist_count': 16,
7031 'info_dict': {
7032 'id': 'royalty free music',
7033 'title': 'royalty free music',
7034 }
7035 }, {
7036 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7037 'playlist_mincount': 30,
7038 'info_dict': {
7039 'id': 'royalty free music - songs',
7040 'title': 'royalty free music - songs',
7041 },
7042 'params': {'extract_flat': 'in_playlist'}
7043 }, {
7044 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7045 'playlist_mincount': 30,
7046 'info_dict': {
7047 'id': 'royalty free music - community playlists',
7048 'title': 'royalty free music - community playlists',
7049 },
7050 'params': {'extract_flat': 'in_playlist'}
7051 }]
7052
7053 _SECTIONS = {
7054 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7055 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7056 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7057 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7058 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7059 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7060 }
7061
7062 def _real_extract(self, url):
7063 qs = parse_qs(url)
7064 query = (qs.get('search_query') or qs.get('q'))[0]
7065 params = qs.get('sp', (None,))[0]
7066 if params:
7067 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7068 else:
7069 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
7070 params = self._SECTIONS.get(section)
7071 if not params:
7072 section = None
7073 title = join_nonempty(query, section, delim=' - ')
7074 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7075
7076
7077 class YoutubeFeedsInfoExtractor(InfoExtractor):
7078 """
7079 Base class for feed extractors
7080 Subclasses must re-define the _FEED_NAME property.
7081 """
7082 _LOGIN_REQUIRED = True
7083 _FEED_NAME = 'feeds'
7084
7085 def _real_initialize(self):
7086 YoutubeBaseInfoExtractor._check_login_required(self)
7087
7088 @classproperty
7089 def IE_NAME(self):
7090 return f'youtube:{self._FEED_NAME}'
7091
7092 def _real_extract(self, url):
7093 return self.url_result(
7094 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7095
7096
7097 class YoutubeWatchLaterIE(InfoExtractor):
7098 IE_NAME = 'youtube:watchlater'
7099 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7100 _VALID_URL = r':ytwatchlater'
7101 _TESTS = [{
7102 'url': ':ytwatchlater',
7103 'only_matching': True,
7104 }]
7105
7106 def _real_extract(self, url):
7107 return self.url_result(
7108 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7109
7110
7111 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7112 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7113 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7114 _FEED_NAME = 'recommended'
7115 _LOGIN_REQUIRED = False
7116 _TESTS = [{
7117 'url': ':ytrec',
7118 'only_matching': True,
7119 }, {
7120 'url': ':ytrecommended',
7121 'only_matching': True,
7122 }, {
7123 'url': 'https://youtube.com',
7124 'only_matching': True,
7125 }]
7126
7127
7128 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7129 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7130 _VALID_URL = r':ytsub(?:scription)?s?'
7131 _FEED_NAME = 'subscriptions'
7132 _TESTS = [{
7133 'url': ':ytsubs',
7134 'only_matching': True,
7135 }, {
7136 'url': ':ytsubscriptions',
7137 'only_matching': True,
7138 }]
7139
7140
7141 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7142 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7143 _VALID_URL = r':ythis(?:tory)?'
7144 _FEED_NAME = 'history'
7145 _TESTS = [{
7146 'url': ':ythistory',
7147 'only_matching': True,
7148 }]
7149
7150
7151 class YoutubeShortsAudioPivotIE(InfoExtractor):
7152 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7153 IE_NAME = 'youtube:shorts:pivot:audio'
7154 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7155 _TESTS = [{
7156 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7157 'only_matching': True,
7158 }]
7159
7160 @staticmethod
7161 def _generate_audio_pivot_params(video_id):
7162 """
7163 Generates sfv_audio_pivot browse params for this video id
7164 """
7165 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7166 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7167
7168 def _real_extract(self, url):
7169 video_id = self._match_id(url)
7170 return self.url_result(
7171 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7172 ie=YoutubeTabIE)
7173
7174
7175 class YoutubeTruncatedURLIE(InfoExtractor):
7176 IE_NAME = 'youtube:truncated_url'
7177 IE_DESC = False # Do not list
7178 _VALID_URL = r'''(?x)
7179 (?:https?://)?
7180 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7181 (?:watch\?(?:
7182 feature=[a-z_]+|
7183 annotation_id=annotation_[^&]+|
7184 x-yt-cl=[0-9]+|
7185 hl=[^&]*|
7186 t=[0-9]+
7187 )?
7188 |
7189 attribution_link\?a=[^&]+
7190 )
7191 $
7192 '''
7193
7194 _TESTS = [{
7195 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7196 'only_matching': True,
7197 }, {
7198 'url': 'https://www.youtube.com/watch?',
7199 'only_matching': True,
7200 }, {
7201 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7202 'only_matching': True,
7203 }, {
7204 'url': 'https://www.youtube.com/watch?feature=foo',
7205 'only_matching': True,
7206 }, {
7207 'url': 'https://www.youtube.com/watch?hl=en-GB',
7208 'only_matching': True,
7209 }, {
7210 'url': 'https://www.youtube.com/watch?t=2372',
7211 'only_matching': True,
7212 }]
7213
7214 def _real_extract(self, url):
7215 raise ExtractorError(
7216 'Did you forget to quote the URL? Remember that & is a meta '
7217 'character in most shells, so you want to put the URL in quotes, '
7218 'like youtube-dl '
7219 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7220 ' or simply youtube-dl BaW_jenozKc .',
7221 expected=True)
7222
7223
7224 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7225 IE_NAME = 'youtube:clip'
7226 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7227 _TESTS = [{
7228 # FIXME: Other metadata should be extracted from the clip, not from the base video
7229 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7230 'info_dict': {
7231 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7232 'ext': 'mp4',
7233 'section_start': 29.0,
7234 'section_end': 39.7,
7235 'duration': 10.7,
7236 'age_limit': 0,
7237 'availability': 'public',
7238 'categories': ['Gaming'],
7239 'channel': 'Scott The Woz',
7240 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7241 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7242 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7243 'like_count': int,
7244 'playable_in_embed': True,
7245 'tags': 'count:17',
7246 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7247 'title': 'Mobile Games on Console - Scott The Woz',
7248 'upload_date': '20210920',
7249 'uploader': 'Scott The Woz',
7250 'uploader_id': '@ScottTheWoz',
7251 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7252 'view_count': int,
7253 'live_status': 'not_live',
7254 'channel_follower_count': int,
7255 'chapters': 'count:20',
7256 'comment_count': int,
7257 'heatmap': 'count:100',
7258 }
7259 }]
7260
7261 def _real_extract(self, url):
7262 clip_id = self._match_id(url)
7263 _, data = self._extract_webpage(url, clip_id)
7264
7265 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7266 if not video_id:
7267 raise ExtractorError('Unable to find video ID')
7268
7269 clip_data = traverse_obj(data, (
7270 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7271 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7272 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7273 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7274
7275 return {
7276 '_type': 'url_transparent',
7277 'url': f'https://www.youtube.com/watch?v={video_id}',
7278 'ie_key': YoutubeIE.ie_key(),
7279 'id': clip_id,
7280 'section_start': int(clip_data['startTimeMs']) / 1000,
7281 'section_end': int(clip_data['endTimeMs']) / 1000,
7282 }
7283
7284
7285 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7286 IE_NAME = 'youtube:consent'
7287 IE_DESC = False # Do not list
7288 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7289 _TESTS = [{
7290 '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',
7291 'info_dict': {
7292 'id': 'qVv6vCqciTM',
7293 'ext': 'mp4',
7294 'age_limit': 0,
7295 'uploader_id': '@sana_natori',
7296 'comment_count': int,
7297 'chapters': 'count:13',
7298 'upload_date': '20221223',
7299 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7300 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7301 'uploader_url': 'https://www.youtube.com/@sana_natori',
7302 'like_count': int,
7303 'release_date': '20221223',
7304 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7305 '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】',
7306 'view_count': int,
7307 'playable_in_embed': True,
7308 'duration': 4438,
7309 'availability': 'public',
7310 'channel_follower_count': int,
7311 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7312 'categories': ['Entertainment'],
7313 'live_status': 'was_live',
7314 'release_timestamp': 1671793345,
7315 'channel': 'さなちゃんねる',
7316 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7317 'uploader': 'さなちゃんねる',
7318 'channel_is_verified': True,
7319 'heatmap': 'count:100',
7320 },
7321 'add_ie': ['Youtube'],
7322 'params': {'skip_download': 'Youtube'},
7323 }]
7324
7325 def _real_extract(self, url):
7326 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7327 if not redirect_url:
7328 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7329 return self.url_result(redirect_url)
7330
7331
7332 class YoutubeTruncatedIDIE(InfoExtractor):
7333 IE_NAME = 'youtube:truncated_id'
7334 IE_DESC = False # Do not list
7335 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7336
7337 _TESTS = [{
7338 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7339 'only_matching': True,
7340 }]
7341
7342 def _real_extract(self, url):
7343 video_id = self._match_id(url)
7344 raise ExtractorError(
7345 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7346 expected=True)