]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[ie/youtube] Extract all formats from multi-language m3u8s (#9875)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import collections
4 import copy
5 import datetime as dt
6 import enum
7 import hashlib
8 import itertools
9 import json
10 import math
11 import os.path
12 import random
13 import re
14 import shlex
15 import sys
16 import threading
17 import time
18 import traceback
19 import urllib.parse
20
21 from .common import InfoExtractor, SearchInfoExtractor
22 from .openload import PhantomJSwrapper
23 from ..compat import functools
24 from ..jsinterp import JSInterpreter
25 from ..networking.exceptions import HTTPError, network_exceptions
26 from ..utils import (
27 NO_DEFAULT,
28 ExtractorError,
29 LazyList,
30 UserNotLive,
31 bug_reports_message,
32 classproperty,
33 clean_html,
34 datetime_from_str,
35 dict_get,
36 filesize_from_tbr,
37 filter_dict,
38 float_or_none,
39 format_field,
40 get_first,
41 int_or_none,
42 is_html,
43 join_nonempty,
44 js_to_json,
45 mimetype2ext,
46 orderedSet,
47 parse_codecs,
48 parse_count,
49 parse_duration,
50 parse_iso8601,
51 parse_qs,
52 qualities,
53 remove_start,
54 smuggle_url,
55 str_or_none,
56 str_to_int,
57 strftime_or_none,
58 traverse_obj,
59 try_call,
60 try_get,
61 unescapeHTML,
62 unified_strdate,
63 unified_timestamp,
64 unsmuggle_url,
65 update_url_query,
66 url_or_none,
67 urljoin,
68 variadic,
69 )
70
71 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
72 # any clients starting with _ cannot be explicitly requested by the user
73 INNERTUBE_CLIENTS = {
74 'web': {
75 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
76 'INNERTUBE_CONTEXT': {
77 'client': {
78 'clientName': 'WEB',
79 'clientVersion': '2.20220801.00.00',
80 },
81 },
82 'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
83 },
84 'web_embedded': {
85 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
86 'INNERTUBE_CONTEXT': {
87 'client': {
88 'clientName': 'WEB_EMBEDDED_PLAYER',
89 'clientVersion': '1.20220731.00.00',
90 },
91 },
92 'INNERTUBE_CONTEXT_CLIENT_NAME': 56,
93 },
94 'web_music': {
95 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
96 'INNERTUBE_HOST': 'music.youtube.com',
97 'INNERTUBE_CONTEXT': {
98 'client': {
99 'clientName': 'WEB_REMIX',
100 'clientVersion': '1.20220727.01.00',
101 },
102 },
103 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
104 },
105 'web_creator': {
106 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
107 'INNERTUBE_CONTEXT': {
108 'client': {
109 'clientName': 'WEB_CREATOR',
110 'clientVersion': '1.20220726.00.00',
111 },
112 },
113 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
114 },
115 'android': {
116 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
117 'INNERTUBE_CONTEXT': {
118 'client': {
119 'clientName': 'ANDROID',
120 'clientVersion': '19.09.37',
121 'androidSdkVersion': 30,
122 'userAgent': 'com.google.android.youtube/19.09.37 (Linux; U; Android 11) gzip',
123 },
124 },
125 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
126 'REQUIRE_JS_PLAYER': False,
127 },
128 'android_embedded': {
129 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
130 'INNERTUBE_CONTEXT': {
131 'client': {
132 'clientName': 'ANDROID_EMBEDDED_PLAYER',
133 'clientVersion': '19.09.37',
134 'androidSdkVersion': 30,
135 'userAgent': 'com.google.android.youtube/19.09.37 (Linux; U; Android 11) gzip',
136 },
137 },
138 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
139 'REQUIRE_JS_PLAYER': False,
140 },
141 'android_music': {
142 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
143 'INNERTUBE_CONTEXT': {
144 'client': {
145 'clientName': 'ANDROID_MUSIC',
146 'clientVersion': '6.42.52',
147 'androidSdkVersion': 30,
148 'userAgent': 'com.google.android.apps.youtube.music/6.42.52 (Linux; U; Android 11) gzip',
149 },
150 },
151 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
152 'REQUIRE_JS_PLAYER': False,
153 },
154 'android_creator': {
155 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
156 'INNERTUBE_CONTEXT': {
157 'client': {
158 'clientName': 'ANDROID_CREATOR',
159 'clientVersion': '22.30.100',
160 'androidSdkVersion': 30,
161 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip',
162 },
163 },
164 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
165 'REQUIRE_JS_PLAYER': False,
166 },
167 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
168 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
169 'ios': {
170 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
171 'INNERTUBE_CONTEXT': {
172 'client': {
173 'clientName': 'IOS',
174 'clientVersion': '19.09.3',
175 'deviceModel': 'iPhone14,3',
176 'userAgent': 'com.google.ios.youtube/19.09.3 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)',
177 },
178 },
179 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
180 'REQUIRE_JS_PLAYER': False,
181 },
182 'ios_embedded': {
183 'INNERTUBE_CONTEXT': {
184 'client': {
185 'clientName': 'IOS_MESSAGES_EXTENSION',
186 'clientVersion': '19.09.3',
187 'deviceModel': 'iPhone14,3',
188 'userAgent': 'com.google.ios.youtube/19.09.3 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)',
189 },
190 },
191 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
192 'REQUIRE_JS_PLAYER': False,
193 },
194 'ios_music': {
195 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
196 'INNERTUBE_CONTEXT': {
197 'client': {
198 'clientName': 'IOS_MUSIC',
199 'clientVersion': '6.33.3',
200 'deviceModel': 'iPhone14,3',
201 'userAgent': 'com.google.ios.youtubemusic/6.33.3 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)',
202 },
203 },
204 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
205 'REQUIRE_JS_PLAYER': False,
206 },
207 'ios_creator': {
208 'INNERTUBE_CONTEXT': {
209 'client': {
210 'clientName': 'IOS_CREATOR',
211 'clientVersion': '22.33.101',
212 'deviceModel': 'iPhone14,3',
213 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)',
214 },
215 },
216 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
217 'REQUIRE_JS_PLAYER': False,
218 },
219 # mweb has 'ultralow' formats
220 # See: https://github.com/yt-dlp/yt-dlp/pull/557
221 'mweb': {
222 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
223 'INNERTUBE_CONTEXT': {
224 'client': {
225 'clientName': 'MWEB',
226 'clientVersion': '2.20220801.00.00',
227 },
228 },
229 'INNERTUBE_CONTEXT_CLIENT_NAME': 2,
230 },
231 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
232 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
233 'tv_embedded': {
234 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
235 'INNERTUBE_CONTEXT': {
236 'client': {
237 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
238 'clientVersion': '2.0',
239 },
240 },
241 'INNERTUBE_CONTEXT_CLIENT_NAME': 85,
242 },
243 # This client has pre-merged video+audio 720p/1080p streams
244 'mediaconnect': {
245 'INNERTUBE_CONTEXT': {
246 'client': {
247 'clientName': 'MEDIA_CONNECT_FRONTEND',
248 'clientVersion': '0.1',
249 },
250 },
251 'INNERTUBE_CONTEXT_CLIENT_NAME': 95,
252 },
253 }
254
255
256 def _split_innertube_client(client_name):
257 variant, *base = client_name.rsplit('.', 1)
258 if base:
259 return variant, base[0], variant
260 base, *variant = client_name.split('_', 1)
261 return client_name, base, variant[0] if variant else None
262
263
264 def short_client_name(client_name):
265 main, *parts = _split_innertube_client(client_name)[0].replace('embedscreen', 'e_s').split('_')
266 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
267
268
269 def build_innertube_clients():
270 THIRD_PARTY = {
271 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
272 }
273 BASE_CLIENTS = ('ios', 'android', 'web', 'tv', 'mweb')
274 priority = qualities(BASE_CLIENTS[::-1])
275
276 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
277 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
278 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
279 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
280 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
281
282 _, base_client, variant = _split_innertube_client(client)
283 ytcfg['priority'] = 10 * priority(base_client)
284
285 if not variant:
286 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
287 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
288 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
289 embedscreen['priority'] -= 3
290 elif variant == 'embedded':
291 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
292 ytcfg['priority'] -= 2
293 else:
294 ytcfg['priority'] -= 3
295
296
297 build_innertube_clients()
298
299
300 class BadgeType(enum.Enum):
301 AVAILABILITY_UNLISTED = enum.auto()
302 AVAILABILITY_PRIVATE = enum.auto()
303 AVAILABILITY_PUBLIC = enum.auto()
304 AVAILABILITY_PREMIUM = enum.auto()
305 AVAILABILITY_SUBSCRIPTION = enum.auto()
306 LIVE_NOW = enum.auto()
307 VERIFIED = enum.auto()
308
309
310 class YoutubeBaseInfoExtractor(InfoExtractor):
311 """Provide base functions for Youtube extractors"""
312
313 _RESERVED_NAMES = (
314 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
315 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
316 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
317 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
318
319 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
320
321 # _NETRC_MACHINE = 'youtube'
322
323 # If True it will raise an error if no login info is provided
324 _LOGIN_REQUIRED = False
325
326 _INVIDIOUS_SITES = (
327 # invidious-redirect websites
328 r'(?:www\.)?redirect\.invidious\.io',
329 r'(?:(?:www|dev)\.)?invidio\.us',
330 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
331 r'(?:www\.)?invidious\.pussthecat\.org',
332 r'(?:www\.)?invidious\.zee\.li',
333 r'(?:www\.)?invidious\.ethibox\.fr',
334 r'(?:www\.)?iv\.ggtyler\.dev',
335 r'(?:www\.)?inv\.vern\.i2p',
336 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
337 r'(?:www\.)?inv\.riverside\.rocks',
338 r'(?:www\.)?invidious\.silur\.me',
339 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
340 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
341 r'(?:www\.)?invidious\.slipfox\.xyz',
342 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
343 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
344 r'(?:www\.)?invidious\.tiekoetter\.com',
345 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
346 r'(?:www\.)?invidious\.nerdvpn\.de',
347 r'(?:www\.)?invidious\.weblibre\.org',
348 r'(?:www\.)?inv\.odyssey346\.dev',
349 r'(?:www\.)?invidious\.dhusch\.de',
350 r'(?:www\.)?iv\.melmac\.space',
351 r'(?:www\.)?watch\.thekitty\.zone',
352 r'(?:www\.)?invidious\.privacydev\.net',
353 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
354 r'(?:www\.)?invidious\.drivet\.xyz',
355 r'(?:www\.)?vid\.priv\.au',
356 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
357 r'(?:www\.)?inv\.vern\.cc',
358 r'(?:www\.)?invidious\.esmailelbob\.xyz',
359 r'(?:www\.)?invidious\.sethforprivacy\.com',
360 r'(?:www\.)?yt\.oelrichsgarcia\.de',
361 r'(?:www\.)?yt\.artemislena\.eu',
362 r'(?:www\.)?invidious\.flokinet\.to',
363 r'(?:www\.)?invidious\.baczek\.me',
364 r'(?:www\.)?y\.com\.sb',
365 r'(?:www\.)?invidious\.epicsite\.xyz',
366 r'(?:www\.)?invidious\.lidarshield\.cloud',
367 r'(?:www\.)?yt\.funami\.tech',
368 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
369 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
370 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
371 # youtube-dl invidious instances list
372 r'(?:(?:www|no)\.)?invidiou\.sh',
373 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
374 r'(?:www\.)?invidious\.kabi\.tk',
375 r'(?:www\.)?invidious\.mastodon\.host',
376 r'(?:www\.)?invidious\.zapashcanon\.fr',
377 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
378 r'(?:www\.)?invidious\.tinfoil-hat\.net',
379 r'(?:www\.)?invidious\.himiko\.cloud',
380 r'(?:www\.)?invidious\.reallyancient\.tech',
381 r'(?:www\.)?invidious\.tube',
382 r'(?:www\.)?invidiou\.site',
383 r'(?:www\.)?invidious\.site',
384 r'(?:www\.)?invidious\.xyz',
385 r'(?:www\.)?invidious\.nixnet\.xyz',
386 r'(?:www\.)?invidious\.048596\.xyz',
387 r'(?:www\.)?invidious\.drycat\.fr',
388 r'(?:www\.)?inv\.skyn3t\.in',
389 r'(?:www\.)?tube\.poal\.co',
390 r'(?:www\.)?tube\.connect\.cafe',
391 r'(?:www\.)?vid\.wxzm\.sx',
392 r'(?:www\.)?vid\.mint\.lgbt',
393 r'(?:www\.)?vid\.puffyan\.us',
394 r'(?:www\.)?yewtu\.be',
395 r'(?:www\.)?yt\.elukerio\.org',
396 r'(?:www\.)?yt\.lelux\.fi',
397 r'(?:www\.)?invidious\.ggc-project\.de',
398 r'(?:www\.)?yt\.maisputain\.ovh',
399 r'(?:www\.)?ytprivate\.com',
400 r'(?:www\.)?invidious\.13ad\.de',
401 r'(?:www\.)?invidious\.toot\.koeln',
402 r'(?:www\.)?invidious\.fdn\.fr',
403 r'(?:www\.)?watch\.nettohikari\.com',
404 r'(?:www\.)?invidious\.namazso\.eu',
405 r'(?:www\.)?invidious\.silkky\.cloud',
406 r'(?:www\.)?invidious\.exonip\.de',
407 r'(?:www\.)?invidious\.riverside\.rocks',
408 r'(?:www\.)?invidious\.blamefran\.net',
409 r'(?:www\.)?invidious\.moomoo\.de',
410 r'(?:www\.)?ytb\.trom\.tf',
411 r'(?:www\.)?yt\.cyberhost\.uk',
412 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
413 r'(?:www\.)?qklhadlycap4cnod\.onion',
414 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
415 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
416 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
417 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
418 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
419 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
420 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
421 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
422 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
423 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
424 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
425 r'(?:www\.)?piped\.kavin\.rocks',
426 r'(?:www\.)?piped\.tokhmi\.xyz',
427 r'(?:www\.)?piped\.syncpundit\.io',
428 r'(?:www\.)?piped\.mha\.fi',
429 r'(?:www\.)?watch\.whatever\.social',
430 r'(?:www\.)?piped\.garudalinux\.org',
431 r'(?:www\.)?piped\.rivo\.lol',
432 r'(?:www\.)?piped-libre\.kavin\.rocks',
433 r'(?:www\.)?yt\.jae\.fi',
434 r'(?:www\.)?piped\.mint\.lgbt',
435 r'(?:www\.)?il\.ax',
436 r'(?:www\.)?piped\.esmailelbob\.xyz',
437 r'(?:www\.)?piped\.projectsegfau\.lt',
438 r'(?:www\.)?piped\.privacydev\.net',
439 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
440 r'(?:www\.)?piped\.smnz\.de',
441 r'(?:www\.)?piped\.adminforge\.de',
442 r'(?:www\.)?watch\.whatevertinfoil\.de',
443 r'(?:www\.)?piped\.qdi\.fi',
444 r'(?:(?:www|cf)\.)?piped\.video',
445 r'(?:www\.)?piped\.aeong\.one',
446 r'(?:www\.)?piped\.moomoo\.me',
447 r'(?:www\.)?piped\.chauvet\.pro',
448 r'(?:www\.)?watch\.leptons\.xyz',
449 r'(?:www\.)?pd\.vern\.cc',
450 r'(?:www\.)?piped\.hostux\.net',
451 r'(?:www\.)?piped\.lunar\.icu',
452 # Hyperpipe instances from https://hyperpipe.codeberg.page/
453 r'(?:www\.)?hyperpipe\.surge\.sh',
454 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
455 r'(?:www\.)?listen\.whatever\.social',
456 r'(?:www\.)?music\.adminforge\.de',
457 )
458
459 # extracted from account/account_menu ep
460 # XXX: These are the supported YouTube UI and API languages,
461 # which is slightly different from languages supported for translation in YouTube studio
462 _SUPPORTED_LANG_CODES = [
463 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
464 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
465 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
466 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
467 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
468 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko',
469 ]
470
471 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
472
473 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
474 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
475
476 def ucid_or_none(self, ucid):
477 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
478
479 def handle_or_none(self, handle):
480 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
481
482 def handle_from_url(self, url):
483 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
484 url, 'channel handle', default=None)
485
486 def ucid_from_url(self, url):
487 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
488 url, 'channel id', default=None)
489
490 @functools.cached_property
491 def _preferred_lang(self):
492 """
493 Returns a language code supported by YouTube for the user preferred language.
494 Returns None if no preferred language set.
495 """
496 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
497 if not preferred_lang:
498 return
499 if preferred_lang not in self._SUPPORTED_LANG_CODES:
500 raise ExtractorError(
501 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
502 expected=True)
503 elif preferred_lang != 'en':
504 self.report_warning(
505 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
506 return preferred_lang
507
508 def _initialize_consent(self):
509 cookies = self._get_cookies('https://www.youtube.com/')
510 if cookies.get('__Secure-3PSID'):
511 return
512 socs = cookies.get('SOCS')
513 if socs and not socs.value.startswith('CAA'): # not consented
514 return
515 self._set_cookie('.youtube.com', 'SOCS', 'CAI', secure=True) # accept all (required for mixes)
516
517 def _initialize_pref(self):
518 cookies = self._get_cookies('https://www.youtube.com/')
519 pref_cookie = cookies.get('PREF')
520 pref = {}
521 if pref_cookie:
522 try:
523 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
524 except ValueError:
525 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
526 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
527 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
528
529 def _real_initialize(self):
530 self._initialize_pref()
531 self._initialize_consent()
532 self._check_login_required()
533
534 def _check_login_required(self):
535 if self._LOGIN_REQUIRED and not self._cookies_passed:
536 self.raise_login_required('Login details are needed to download this content', method='cookies')
537
538 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
539 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
540
541 def _get_default_ytcfg(self, client='web'):
542 return copy.deepcopy(INNERTUBE_CLIENTS[client])
543
544 def _get_innertube_host(self, client='web'):
545 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
546
547 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
548 # try_get but with fallback to default ytcfg client values when present
549 _func = lambda y: try_get(y, getter, expected_type)
550 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
551
552 def _extract_client_name(self, ytcfg, default_client='web'):
553 return self._ytcfg_get_safe(
554 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
555 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
556
557 def _extract_client_version(self, ytcfg, default_client='web'):
558 return self._ytcfg_get_safe(
559 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
560 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
561
562 def _select_api_hostname(self, req_api_hostname, default_client=None):
563 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
564 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
565
566 def _extract_api_key(self, ytcfg=None, default_client='web'):
567 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
568
569 def _extract_context(self, ytcfg=None, default_client='web'):
570 context = get_first(
571 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
572 # Enforce language and tz for extraction
573 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
574 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
575 return context
576
577 _SAPISID = None
578
579 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
580 time_now = round(time.time())
581 if self._SAPISID is None:
582 yt_cookies = self._get_cookies('https://www.youtube.com')
583 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
584 # See: https://github.com/yt-dlp/yt-dlp/issues/393
585 sapisid_cookie = dict_get(
586 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
587 if sapisid_cookie and sapisid_cookie.value:
588 self._SAPISID = sapisid_cookie.value
589 self.write_debug('Extracted SAPISID cookie')
590 # SAPISID cookie is required if not already present
591 if not yt_cookies.get('SAPISID'):
592 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
593 self._set_cookie(
594 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
595 else:
596 self._SAPISID = False
597 if not self._SAPISID:
598 return None
599 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
600 sapisidhash = hashlib.sha1(
601 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
602 return f'SAPISIDHASH {time_now}_{sapisidhash}'
603
604 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
605 note='Downloading API JSON', errnote='Unable to download API page',
606 context=None, api_key=None, api_hostname=None, default_client='web'):
607
608 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
609 data.update(query)
610 real_headers = self.generate_api_headers(default_client=default_client)
611 real_headers.update({'content-type': 'application/json'})
612 if headers:
613 real_headers.update(headers)
614 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
615 or api_key or self._extract_api_key(default_client=default_client))
616 return self._download_json(
617 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
618 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
619 data=json.dumps(data).encode('utf8'), headers=real_headers,
620 query={'key': api_key, 'prettyPrint': 'false'})
621
622 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
623 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
624
625 @staticmethod
626 def _extract_session_index(*data):
627 """
628 Index of current account in account list.
629 See: https://github.com/yt-dlp/yt-dlp/pull/519
630 """
631 for ytcfg in data:
632 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
633 if session_index is not None:
634 return session_index
635
636 # Deprecated?
637 def _extract_identity_token(self, ytcfg=None, webpage=None):
638 if ytcfg:
639 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
640 if token:
641 return token
642 if webpage:
643 return self._search_regex(
644 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
645 'identity token', default=None, fatal=False)
646
647 @staticmethod
648 def _extract_account_syncid(*args):
649 """
650 Extract syncId required to download private playlists of secondary channels
651 @params response and/or ytcfg
652 """
653 for data in args:
654 # ytcfg includes channel_syncid if on secondary channel
655 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
656 if delegated_sid:
657 return delegated_sid
658 sync_ids = (try_get(
659 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
660 lambda x: x['DATASYNC_ID']), str) or '').split('||')
661 if len(sync_ids) >= 2 and sync_ids[1]:
662 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
663 # and just "user_syncid||" for primary channel. We only want the channel_syncid
664 return sync_ids[0]
665
666 @staticmethod
667 def _extract_visitor_data(*args):
668 """
669 Extracts visitorData from an API response or ytcfg
670 Appears to be used to track session state
671 """
672 return get_first(
673 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
674 expected_type=str)
675
676 @functools.cached_property
677 def is_authenticated(self):
678 return bool(self._generate_sapisidhash_header())
679
680 def extract_ytcfg(self, video_id, webpage):
681 if not webpage:
682 return {}
683 return self._parse_json(
684 self._search_regex(
685 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
686 default='{}'), video_id, fatal=False) or {}
687
688 def generate_api_headers(
689 self, *, ytcfg=None, account_syncid=None, session_index=None,
690 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
691
692 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
693 headers = {
694 'X-YouTube-Client-Name': str(
695 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
696 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
697 'Origin': origin,
698 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
699 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
700 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
701 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client),
702 }
703 if session_index is None:
704 session_index = self._extract_session_index(ytcfg)
705 if account_syncid or session_index is not None:
706 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
707
708 auth = self._generate_sapisidhash_header(origin)
709 if auth is not None:
710 headers['Authorization'] = auth
711 headers['X-Origin'] = origin
712 return filter_dict(headers)
713
714 def _download_ytcfg(self, client, video_id):
715 url = {
716 'web': 'https://www.youtube.com',
717 'web_music': 'https://music.youtube.com',
718 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1',
719 }.get(client)
720 if not url:
721 return {}
722 webpage = self._download_webpage(
723 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
724 return self.extract_ytcfg(video_id, webpage) or {}
725
726 @staticmethod
727 def _build_api_continuation_query(continuation, ctp=None):
728 query = {
729 'continuation': continuation,
730 }
731 # TODO: Inconsistency with clickTrackingParams.
732 # Currently we have a fixed ctp contained within context (from ytcfg)
733 # and a ctp in root query for continuation.
734 if ctp:
735 query['clickTracking'] = {'clickTrackingParams': ctp}
736 return query
737
738 @classmethod
739 def _extract_next_continuation_data(cls, renderer):
740 next_continuation = try_get(
741 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
742 lambda x: x['continuation']['reloadContinuationData']), dict)
743 if not next_continuation:
744 return
745 continuation = next_continuation.get('continuation')
746 if not continuation:
747 return
748 ctp = next_continuation.get('clickTrackingParams')
749 return cls._build_api_continuation_query(continuation, ctp)
750
751 @classmethod
752 def _extract_continuation_ep_data(cls, continuation_ep: dict):
753 if isinstance(continuation_ep, dict):
754 continuation = try_get(
755 continuation_ep, lambda x: x['continuationCommand']['token'], str)
756 if not continuation:
757 return
758 ctp = continuation_ep.get('clickTrackingParams')
759 return cls._build_api_continuation_query(continuation, ctp)
760
761 @classmethod
762 def _extract_continuation(cls, renderer):
763 next_continuation = cls._extract_next_continuation_data(renderer)
764 if next_continuation:
765 return next_continuation
766
767 return traverse_obj(renderer, (
768 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
769 ('continuationEndpoint', ('button', 'buttonRenderer', 'command')),
770 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
771
772 @classmethod
773 def _extract_alerts(cls, data):
774 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
775 if not isinstance(alert_dict, dict):
776 continue
777 for alert in alert_dict.values():
778 alert_type = alert.get('type')
779 if not alert_type:
780 continue
781 message = cls._get_text(alert, 'text')
782 if message:
783 yield alert_type, message
784
785 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
786 errors, warnings = [], []
787 for alert_type, alert_message in alerts:
788 if alert_type.lower() == 'error' and fatal:
789 errors.append([alert_type, alert_message])
790 elif alert_message not in self._IGNORED_WARNINGS:
791 warnings.append([alert_type, alert_message])
792
793 for alert_type, alert_message in (warnings + errors[:-1]):
794 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
795 if errors:
796 raise ExtractorError(f'YouTube said: {errors[-1][1]}', expected=expected)
797
798 def _extract_and_report_alerts(self, data, *args, **kwargs):
799 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
800
801 def _extract_badges(self, badge_list: list):
802 """
803 Extract known BadgeType's from a list of badge renderers.
804 @returns [{'type': BadgeType}]
805 """
806 icon_type_map = {
807 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
808 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
809 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC,
810 'CHECK_CIRCLE_THICK': BadgeType.VERIFIED,
811 'OFFICIAL_ARTIST_BADGE': BadgeType.VERIFIED,
812 'CHECK': BadgeType.VERIFIED,
813 }
814
815 badge_style_map = {
816 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
817 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
818 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
819 'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
820 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED,
821 }
822
823 label_map = {
824 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
825 'private': BadgeType.AVAILABILITY_PRIVATE,
826 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
827 'live': BadgeType.LIVE_NOW,
828 'premium': BadgeType.AVAILABILITY_PREMIUM,
829 'verified': BadgeType.VERIFIED,
830 'official artist channel': BadgeType.VERIFIED,
831 }
832
833 badges = []
834 for badge in traverse_obj(badge_list, (..., lambda key, _: re.search(r'[bB]adgeRenderer$', key))):
835 badge_type = (
836 icon_type_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
837 or badge_style_map.get(traverse_obj(badge, 'style'))
838 )
839 if badge_type:
840 badges.append({'type': badge_type})
841 continue
842
843 # fallback, won't work in some languages
844 label = traverse_obj(
845 badge, 'label', ('accessibilityData', 'label'), 'tooltip', 'iconTooltip', get_all=False, expected_type=str, default='')
846 for match, label_badge_type in label_map.items():
847 if match in label.lower():
848 badges.append({'type': label_badge_type})
849 break
850
851 return badges
852
853 @staticmethod
854 def _has_badge(badges, badge_type):
855 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
856
857 @staticmethod
858 def _get_text(data, *path_list, max_runs=None):
859 for path in path_list or [None]:
860 if path is None:
861 obj = [data]
862 else:
863 obj = traverse_obj(data, path, default=[])
864 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
865 obj = [obj]
866 for item in obj:
867 text = try_get(item, lambda x: x['simpleText'], str)
868 if text:
869 return text
870 runs = try_get(item, lambda x: x['runs'], list) or []
871 if not runs and isinstance(item, list):
872 runs = item
873
874 runs = runs[:min(len(runs), max_runs or len(runs))]
875 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
876 if text:
877 return text
878
879 def _get_count(self, data, *path_list):
880 count_text = self._get_text(data, *path_list) or ''
881 count = parse_count(count_text)
882 if count is None:
883 count = str_to_int(
884 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
885 return count
886
887 @staticmethod
888 def _extract_thumbnails(data, *path_list, final_key='thumbnails'):
889 """
890 Extract thumbnails from thumbnails dict
891 @param path_list: path list to level that contains 'thumbnails' key
892 """
893 thumbnails = []
894 for path in path_list or [()]:
895 for thumbnail in traverse_obj(data, (*variadic(path), final_key, ...)):
896 thumbnail_url = url_or_none(thumbnail.get('url'))
897 if not thumbnail_url:
898 continue
899 # Sometimes youtube gives a wrong thumbnail URL. See:
900 # https://github.com/yt-dlp/yt-dlp/issues/233
901 # https://github.com/ytdl-org/youtube-dl/issues/28023
902 if 'maxresdefault' in thumbnail_url:
903 thumbnail_url = thumbnail_url.split('?')[0]
904 thumbnails.append({
905 'url': thumbnail_url,
906 'height': int_or_none(thumbnail.get('height')),
907 'width': int_or_none(thumbnail.get('width')),
908 })
909 return thumbnails
910
911 @staticmethod
912 def extract_relative_time(relative_time_text):
913 """
914 Extracts a relative time from string and converts to dt object
915 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today', '8 yr ago'
916 """
917
918 # XXX: this could be moved to a general function in utils/_utils.py
919 # The relative time text strings are roughly the same as what
920 # Javascript's Intl.RelativeTimeFormat function generates.
921 # See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat
922 mobj = re.search(
923 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',
924 relative_time_text)
925 if mobj:
926 start = mobj.group('start')
927 if start:
928 return datetime_from_str(start)
929 try:
930 return datetime_from_str('now-{}{}'.format(mobj.group('time'), mobj.group('unit')))
931 except ValueError:
932 return None
933
934 def _parse_time_text(self, text):
935 if not text:
936 return
937 dt_ = self.extract_relative_time(text)
938 timestamp = None
939 if isinstance(dt_, dt.datetime):
940 timestamp = calendar.timegm(dt_.timetuple())
941
942 if timestamp is None:
943 timestamp = (
944 unified_timestamp(text) or unified_timestamp(
945 self._search_regex(
946 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
947 text.lower(), 'time text', default=None)))
948
949 if text and timestamp is None and self._preferred_lang in (None, 'en'):
950 self.report_warning(
951 f'Cannot parse localized time text "{text}"', only_once=True)
952 return timestamp
953
954 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
955 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
956 default_client='web'):
957 raise_for_incomplete = bool(self._configuration_arg('raise_incomplete_data', ie_key=YoutubeIE))
958 # Incomplete Data should be a warning by default when retries are exhausted, while other errors should be fatal.
959 icd_retries = iter(self.RetryManager(fatal=raise_for_incomplete))
960 icd_rm = next(icd_retries)
961 main_retries = iter(self.RetryManager())
962 main_rm = next(main_retries)
963 # Manual retry loop for multiple RetryManagers
964 # The proper RetryManager MUST be advanced after an error
965 # and its result MUST be checked if the manager is non fatal
966 while True:
967 try:
968 response = self._call_api(
969 ep=ep, fatal=True, headers=headers,
970 video_id=item_id, query=query, note=note,
971 context=self._extract_context(ytcfg, default_client),
972 api_key=self._extract_api_key(ytcfg, default_client),
973 api_hostname=api_hostname, default_client=default_client)
974 except ExtractorError as e:
975 if not isinstance(e.cause, network_exceptions):
976 return self._error_or_warning(e, fatal=fatal)
977 elif not isinstance(e.cause, HTTPError):
978 main_rm.error = e
979 next(main_retries)
980 continue
981
982 first_bytes = e.cause.response.read(512)
983 if not is_html(first_bytes):
984 yt_error = try_get(
985 self._parse_json(
986 self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
987 lambda x: x['error']['message'], str)
988 if yt_error:
989 self._report_alerts([('ERROR', yt_error)], fatal=False)
990 # Downloading page may result in intermittent 5xx HTTP error
991 # Sometimes a 404 is also received. See: https://github.com/ytdl-org/youtube-dl/issues/28289
992 # We also want to catch all other network exceptions since errors in later pages can be troublesome
993 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
994 if e.cause.status not in (403, 429):
995 main_rm.error = e
996 next(main_retries)
997 continue
998 return self._error_or_warning(e, fatal=fatal)
999
1000 try:
1001 self._extract_and_report_alerts(response, only_once=True)
1002 except ExtractorError as e:
1003 # YouTube's servers may return errors we want to retry on in a 200 OK response
1004 # See: https://github.com/yt-dlp/yt-dlp/issues/839
1005 if 'unknown error' in e.msg.lower():
1006 main_rm.error = e
1007 next(main_retries)
1008 continue
1009 return self._error_or_warning(e, fatal=fatal)
1010 # Youtube sometimes sends incomplete data
1011 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1012 if not traverse_obj(response, *variadic(check_get_keys)):
1013 icd_rm.error = ExtractorError('Incomplete data received', expected=True)
1014 should_retry = next(icd_retries, None)
1015 if not should_retry:
1016 return None
1017 continue
1018
1019 return response
1020
1021 @staticmethod
1022 def is_music_url(url):
1023 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
1024
1025 def _extract_video(self, renderer):
1026 video_id = renderer.get('videoId')
1027
1028 reel_header_renderer = traverse_obj(renderer, (
1029 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
1030 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
1031
1032 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
1033 description = self._get_text(renderer, 'descriptionSnippet')
1034
1035 duration = int_or_none(renderer.get('lengthSeconds'))
1036 if duration is None:
1037 duration = parse_duration(self._get_text(
1038 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1039 if duration is None:
1040 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1041 duration = parse_duration(self._search_regex(
1042 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1043 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1044 video_id, default=None, group='duration'))
1045
1046 channel_id = traverse_obj(
1047 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1048 expected_type=str, get_all=False)
1049 if not channel_id:
1050 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1051
1052 channel_id = self.ucid_or_none(channel_id)
1053
1054 overlay_style = traverse_obj(
1055 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1056 get_all=False, expected_type=str)
1057 badges = self._extract_badges(traverse_obj(renderer, 'badges'))
1058 owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
1059 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1060 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1061 expected_type=str)) or ''
1062 url = f'https://www.youtube.com/watch?v={video_id}'
1063 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1064 url = f'https://www.youtube.com/shorts/{video_id}'
1065
1066 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1067 or self._get_text(reel_header_renderer, 'timestampText') or '')
1068 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1069
1070 live_status = (
1071 'is_upcoming' if scheduled_timestamp is not None
1072 else 'was_live' if 'streamed' in time_text.lower()
1073 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1074 else None)
1075
1076 # videoInfo is a string like '50K views • 10 years ago'.
1077 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1078 view_count = (0 if 'no views' in view_count_text.lower()
1079 else self._get_count({'simpleText': view_count_text}))
1080 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1081
1082 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1083 or self._get_text(reel_header_renderer, 'channelTitleText'))
1084
1085 channel_handle = traverse_obj(renderer, (
1086 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1087 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1088 expected_type=self.handle_from_url, get_all=False)
1089 return {
1090 '_type': 'url',
1091 'ie_key': YoutubeIE.ie_key(),
1092 'id': video_id,
1093 'url': url,
1094 'title': title,
1095 'description': description,
1096 'duration': duration,
1097 'channel_id': channel_id,
1098 'channel': channel,
1099 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1100 'uploader': channel,
1101 'uploader_id': channel_handle,
1102 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1103 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1104 'timestamp': (self._parse_time_text(time_text)
1105 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1106 else None),
1107 'release_timestamp': scheduled_timestamp,
1108 'availability':
1109 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1110 else self._availability(
1111 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1112 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1113 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1114 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1115 view_count_field: view_count,
1116 'live_status': live_status,
1117 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None,
1118 }
1119
1120
1121 class YoutubeIE(YoutubeBaseInfoExtractor):
1122 IE_DESC = 'YouTube'
1123 _VALID_URL = r'''(?x)^
1124 (
1125 (?:https?://|//) # http(s):// or protocol-independent URL
1126 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1127 (?:www\.)?deturl\.com/www\.youtube\.com|
1128 (?:www\.)?pwnyoutube\.com|
1129 (?:www\.)?hooktube\.com|
1130 (?:www\.)?yourepeat\.com|
1131 tube\.majestyc\.net|
1132 {invidious}|
1133 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1134 (?:.*?\#/)? # handle anchor (#/) redirect urls
1135 (?: # the various things that can precede the ID:
1136 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1137 |(?: # or the v= param in all its forms
1138 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1139 (?:\?|\#!?) # the params delimiter ? or # or #!
1140 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1141 v=
1142 )
1143 ))
1144 |(?:
1145 youtu\.be| # just youtu.be/xxxx
1146 vid\.plus| # or vid.plus/xxxx
1147 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1148 {invidious}
1149 )/
1150 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1151 )
1152 )? # all until now is optional -> you can pass the naked ID
1153 (?P<id>[0-9A-Za-z_-]{{11}}) # here is it! the YouTube video ID
1154 (?(1).+)? # if we found the ID, everything can follow
1155 (?:\#|$)'''.format(
1156 invidious='|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1157 )
1158 _EMBED_REGEX = [
1159 r'''(?x)
1160 (?:
1161 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1162 data-video-url=|
1163 <embed[^>]+?src=|
1164 embedSWF\(?:\s*|
1165 <object[^>]+data=|
1166 new\s+SWFObject\(
1167 )
1168 (["\'])
1169 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1170 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1171 \1''',
1172 # https://wordpress.org/plugins/lazy-load-for-videos/
1173 r'''(?xs)
1174 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1175 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1176 ]
1177 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1178
1179 _PLAYER_INFO_RE = (
1180 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1181 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$',
1182 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1183 )
1184 _formats = { # NB: Used in YoutubeWebArchiveIE and GoogleDriveIE
1185 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1186 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1187 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1188 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1189 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1190 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1191 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1192 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1193 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1194 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1195 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1196 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1197 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1198 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1199 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1200 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1201 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1202 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1203
1204
1205 # 3D videos
1206 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1207 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1208 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1209 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1210 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1211 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1212 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1213
1214 # Apple HTTP Live Streaming
1215 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1216 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1217 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1218 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1219 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1220 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1221 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1222 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1223
1224 # DASH mp4 video
1225 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1226 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1227 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1228 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1229 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1230 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1231 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1232 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1233 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1234 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1235 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1236 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1237
1238 # Dash mp4 audio
1239 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1240 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1241 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1242 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1243 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1244 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1245 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1246
1247 # Dash webm
1248 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1249 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1250 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1251 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1252 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1253 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1254 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1255 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1256 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1257 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1258 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1259 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1260 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1261 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1262 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1263 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1264 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1265 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1266 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1267 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1268 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1269 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1270
1271 # Dash webm audio
1272 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1273 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1274
1275 # Dash webm audio with opus inside
1276 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1277 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1278 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1279
1280 # RTMP (unnamed)
1281 '_rtmp': {'protocol': 'rtmp'},
1282
1283 # av01 video only formats sometimes served with "unknown" codecs
1284 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1285 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1286 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1287 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1288 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1289 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1290 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1291 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1292 }
1293 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1294
1295 _GEO_BYPASS = False
1296
1297 IE_NAME = 'youtube'
1298 _TESTS = [
1299 {
1300 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1301 'info_dict': {
1302 'id': 'BaW_jenozKc',
1303 'ext': 'mp4',
1304 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1305 'channel': 'Philipp Hagemeister',
1306 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1307 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1308 'upload_date': '20121002',
1309 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1310 'categories': ['Science & Technology'],
1311 'tags': ['youtube-dl'],
1312 'duration': 10,
1313 'view_count': int,
1314 'like_count': int,
1315 'availability': 'public',
1316 'playable_in_embed': True,
1317 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1318 'live_status': 'not_live',
1319 'age_limit': 0,
1320 'start_time': 1,
1321 'end_time': 9,
1322 'comment_count': int,
1323 'channel_follower_count': int,
1324 'uploader': 'Philipp Hagemeister',
1325 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1326 'uploader_id': '@PhilippHagemeister',
1327 'heatmap': 'count:100',
1328 'timestamp': 1349198244,
1329 },
1330 },
1331 {
1332 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1333 'note': 'Embed-only video (#1746)',
1334 'info_dict': {
1335 'id': 'yZIXLfi8CZQ',
1336 'ext': 'mp4',
1337 'upload_date': '20120608',
1338 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1339 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1340 'age_limit': 18,
1341 },
1342 'skip': 'Private video',
1343 },
1344 {
1345 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1346 'note': 'Use the first video ID in the URL',
1347 'info_dict': {
1348 'id': 'BaW_jenozKc',
1349 'ext': 'mp4',
1350 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1351 'channel': 'Philipp Hagemeister',
1352 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1353 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1354 'upload_date': '20121002',
1355 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1356 'categories': ['Science & Technology'],
1357 'tags': ['youtube-dl'],
1358 'duration': 10,
1359 'view_count': int,
1360 'like_count': int,
1361 'availability': 'public',
1362 'playable_in_embed': True,
1363 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1364 'live_status': 'not_live',
1365 'age_limit': 0,
1366 'comment_count': int,
1367 'channel_follower_count': int,
1368 'uploader': 'Philipp Hagemeister',
1369 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1370 'uploader_id': '@PhilippHagemeister',
1371 'heatmap': 'count:100',
1372 'timestamp': 1349198244,
1373 },
1374 'params': {
1375 'skip_download': True,
1376 },
1377 },
1378 {
1379 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1380 'note': '256k DASH audio (format 141) via DASH manifest',
1381 'info_dict': {
1382 'id': 'a9LDPn-MO4I',
1383 'ext': 'm4a',
1384 'upload_date': '20121002',
1385 'description': '',
1386 'title': 'UHDTV TEST 8K VIDEO.mp4',
1387 },
1388 'params': {
1389 'youtube_include_dash_manifest': True,
1390 'format': '141',
1391 },
1392 'skip': 'format 141 not served anymore',
1393 },
1394 # DASH manifest with encrypted signature
1395 {
1396 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1397 'info_dict': {
1398 'id': 'IB3lcPjvWLA',
1399 'ext': 'm4a',
1400 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1401 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1402 'duration': 244,
1403 'upload_date': '20131011',
1404 'abr': 129.495,
1405 'like_count': int,
1406 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1407 'playable_in_embed': True,
1408 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1409 'view_count': int,
1410 'track': 'The Spark',
1411 'live_status': 'not_live',
1412 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1413 'channel': 'Afrojack',
1414 'tags': 'count:19',
1415 'availability': 'public',
1416 'categories': ['Music'],
1417 'age_limit': 0,
1418 'alt_title': 'The Spark',
1419 'channel_follower_count': int,
1420 'uploader': 'Afrojack',
1421 'uploader_url': 'https://www.youtube.com/@Afrojack',
1422 'uploader_id': '@Afrojack',
1423 },
1424 'params': {
1425 'youtube_include_dash_manifest': True,
1426 'format': '141/bestaudio[ext=m4a]',
1427 },
1428 },
1429 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1430 {
1431 'note': 'Embed allowed age-gate video',
1432 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1433 'info_dict': {
1434 'id': 'HtVdAasjOgU',
1435 'ext': 'mp4',
1436 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1437 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1438 'duration': 142,
1439 'upload_date': '20140605',
1440 'age_limit': 18,
1441 'categories': ['Gaming'],
1442 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1443 'availability': 'needs_auth',
1444 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1445 'like_count': int,
1446 'channel': 'The Witcher',
1447 'live_status': 'not_live',
1448 'tags': 'count:17',
1449 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1450 'playable_in_embed': True,
1451 'view_count': int,
1452 'channel_follower_count': int,
1453 'uploader': 'The Witcher',
1454 'uploader_url': 'https://www.youtube.com/@thewitcher',
1455 'uploader_id': '@thewitcher',
1456 'comment_count': int,
1457 'channel_is_verified': True,
1458 'heatmap': 'count:100',
1459 'timestamp': 1401991663,
1460 },
1461 },
1462 {
1463 'note': 'Age-gate video with embed allowed in public site',
1464 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1465 'info_dict': {
1466 'id': 'HsUATh_Nc2U',
1467 'ext': 'mp4',
1468 'title': 'Godzilla 2 (Official Video)',
1469 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1470 'upload_date': '20200408',
1471 'age_limit': 18,
1472 'availability': 'needs_auth',
1473 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1474 'channel': 'FlyingKitty',
1475 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1476 'view_count': int,
1477 'categories': ['Entertainment'],
1478 'live_status': 'not_live',
1479 'tags': ['Flyingkitty', 'godzilla 2'],
1480 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1481 'like_count': int,
1482 'duration': 177,
1483 'playable_in_embed': True,
1484 'channel_follower_count': int,
1485 'uploader': 'FlyingKitty',
1486 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1487 'uploader_id': '@FlyingKitty900',
1488 'comment_count': int,
1489 'channel_is_verified': True,
1490 },
1491 },
1492 {
1493 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1494 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1495 'info_dict': {
1496 'id': 'Tq92D6wQ1mg',
1497 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1498 'ext': 'mp4',
1499 'upload_date': '20191228',
1500 'description': 'md5:17eccca93a786d51bc67646756894066',
1501 'age_limit': 18,
1502 'like_count': int,
1503 'availability': 'needs_auth',
1504 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1505 'view_count': int,
1506 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1507 'channel': 'Projekt Melody',
1508 'live_status': 'not_live',
1509 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1510 'playable_in_embed': True,
1511 'categories': ['Entertainment'],
1512 'duration': 106,
1513 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1514 'comment_count': int,
1515 'channel_follower_count': int,
1516 'uploader': 'Projekt Melody',
1517 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1518 'uploader_id': '@ProjektMelody',
1519 'timestamp': 1577508724,
1520 },
1521 },
1522 {
1523 'note': 'Non-Agegated non-embeddable video',
1524 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1525 'info_dict': {
1526 'id': 'MeJVWBSsPAY',
1527 'ext': 'mp4',
1528 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1529 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1530 'upload_date': '20130730',
1531 'track': 'Such mich find mich',
1532 'age_limit': 0,
1533 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1534 'like_count': int,
1535 'playable_in_embed': False,
1536 'creator': 'OOMPH!',
1537 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1538 'view_count': int,
1539 'alt_title': 'Such mich find mich',
1540 'duration': 210,
1541 'channel': 'Herr Lurik',
1542 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1543 'categories': ['Music'],
1544 'availability': 'public',
1545 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1546 'live_status': 'not_live',
1547 'artist': 'OOMPH!',
1548 'channel_follower_count': int,
1549 'uploader': 'Herr Lurik',
1550 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1551 'uploader_id': '@HerrLurik',
1552 },
1553 },
1554 {
1555 'note': 'Non-bypassable age-gated video',
1556 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1557 'only_matching': True,
1558 },
1559 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1560 # YouTube Red ad is not captured for creator
1561 {
1562 'url': '__2ABJjxzNo',
1563 'info_dict': {
1564 'id': '__2ABJjxzNo',
1565 'ext': 'mp4',
1566 'duration': 266,
1567 'upload_date': '20100430',
1568 'creator': 'deadmau5',
1569 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1570 'title': 'Deadmau5 - Some Chords (HD)',
1571 'alt_title': 'Some Chords',
1572 'availability': 'public',
1573 'tags': 'count:14',
1574 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1575 'view_count': int,
1576 'live_status': 'not_live',
1577 'channel': 'deadmau5',
1578 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1579 'like_count': int,
1580 'track': 'Some Chords',
1581 'artist': 'deadmau5',
1582 'playable_in_embed': True,
1583 'age_limit': 0,
1584 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1585 'categories': ['Music'],
1586 'album': 'Some Chords',
1587 'channel_follower_count': int,
1588 'uploader': 'deadmau5',
1589 'uploader_url': 'https://www.youtube.com/@deadmau5',
1590 'uploader_id': '@deadmau5',
1591 },
1592 'expected_warnings': [
1593 'DASH manifest missing',
1594 ],
1595 },
1596 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1597 {
1598 'url': 'lqQg6PlCWgI',
1599 'info_dict': {
1600 'id': 'lqQg6PlCWgI',
1601 'ext': 'mp4',
1602 'duration': 6085,
1603 'upload_date': '20150827',
1604 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1605 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1606 'like_count': int,
1607 'release_timestamp': 1343767800,
1608 'playable_in_embed': True,
1609 'categories': ['Sports'],
1610 'release_date': '20120731',
1611 'channel': 'Olympics',
1612 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1613 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1614 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1615 'age_limit': 0,
1616 'availability': 'public',
1617 'live_status': 'was_live',
1618 'view_count': int,
1619 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1620 'channel_follower_count': int,
1621 'uploader': 'Olympics',
1622 'uploader_url': 'https://www.youtube.com/@Olympics',
1623 'uploader_id': '@Olympics',
1624 'channel_is_verified': True,
1625 'timestamp': 1440707674,
1626 },
1627 'params': {
1628 'skip_download': 'requires avconv',
1629 },
1630 },
1631 # Non-square pixels
1632 {
1633 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1634 'info_dict': {
1635 'id': '_b-2C3KPAM0',
1636 'ext': 'mp4',
1637 'stretched_ratio': 16 / 9.,
1638 'duration': 85,
1639 'upload_date': '20110310',
1640 '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',
1641 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1642 'playable_in_embed': True,
1643 'channel': '孫ᄋᄅ',
1644 'age_limit': 0,
1645 'tags': 'count:11',
1646 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1647 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1648 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1649 'view_count': int,
1650 'categories': ['People & Blogs'],
1651 'like_count': int,
1652 'live_status': 'not_live',
1653 'availability': 'unlisted',
1654 'comment_count': int,
1655 'channel_follower_count': int,
1656 'uploader': '孫ᄋᄅ',
1657 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1658 'uploader_id': '@AllenMeow',
1659 'timestamp': 1299776999,
1660 },
1661 },
1662 # url_encoded_fmt_stream_map is empty string
1663 {
1664 'url': 'qEJwOuvDf7I',
1665 'info_dict': {
1666 'id': 'qEJwOuvDf7I',
1667 'ext': 'webm',
1668 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1669 'description': '',
1670 'upload_date': '20150404',
1671 },
1672 'params': {
1673 'skip_download': 'requires avconv',
1674 },
1675 'skip': 'This live event has ended.',
1676 },
1677 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1678 {
1679 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1680 'info_dict': {
1681 'id': 'FIl7x6_3R5Y',
1682 'ext': 'webm',
1683 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1684 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1685 'duration': 220,
1686 'upload_date': '20150625',
1687 'formats': 'mincount:31',
1688 },
1689 'skip': 'not actual anymore',
1690 },
1691 # DASH manifest with segment_list
1692 {
1693 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1694 'md5': '8ce563a1d667b599d21064e982ab9e31',
1695 'info_dict': {
1696 'id': 'CsmdDsKjzN8',
1697 'ext': 'mp4',
1698 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1699 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1700 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1701 },
1702 'params': {
1703 'youtube_include_dash_manifest': True,
1704 'format': '135', # bestvideo
1705 },
1706 'skip': 'This live event has ended.',
1707 },
1708 {
1709 # Multifeed videos (multiple cameras), URL can be of any Camera
1710 # TODO: fix multifeed titles
1711 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1712 'info_dict': {
1713 'id': 'zaPI8MvL8pg',
1714 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1715 'description': 'md5:563ccbc698b39298481ca3c571169519',
1716 },
1717 'playlist': [{
1718 'info_dict': {
1719 'id': 'j5yGuxZ8lLU',
1720 'ext': 'mp4',
1721 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1722 'description': 'md5:563ccbc698b39298481ca3c571169519',
1723 'duration': 10120,
1724 'channel_follower_count': int,
1725 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1726 'availability': 'public',
1727 'playable_in_embed': True,
1728 'upload_date': '20131105',
1729 'categories': ['Gaming'],
1730 'live_status': 'was_live',
1731 'tags': 'count:24',
1732 'release_timestamp': 1383701910,
1733 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1734 'comment_count': int,
1735 'age_limit': 0,
1736 'like_count': int,
1737 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1738 'channel': 'WiiLikeToPlay',
1739 'view_count': int,
1740 'release_date': '20131106',
1741 'uploader': 'WiiLikeToPlay',
1742 'uploader_id': '@WLTP',
1743 'uploader_url': 'https://www.youtube.com/@WLTP',
1744 },
1745 }, {
1746 'info_dict': {
1747 'id': 'zaPI8MvL8pg',
1748 'ext': 'mp4',
1749 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1750 'availability': 'public',
1751 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1752 'channel': 'WiiLikeToPlay',
1753 'channel_follower_count': int,
1754 'description': 'md5:563ccbc698b39298481ca3c571169519',
1755 'duration': 10108,
1756 'age_limit': 0,
1757 'like_count': int,
1758 'tags': 'count:24',
1759 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1760 'release_timestamp': 1383701915,
1761 'comment_count': int,
1762 'upload_date': '20131105',
1763 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1764 'release_date': '20131106',
1765 'playable_in_embed': True,
1766 'live_status': 'was_live',
1767 'categories': ['Gaming'],
1768 'view_count': int,
1769 'uploader': 'WiiLikeToPlay',
1770 'uploader_id': '@WLTP',
1771 'uploader_url': 'https://www.youtube.com/@WLTP',
1772 },
1773 }, {
1774 'info_dict': {
1775 'id': 'R7r3vfO7Hao',
1776 'ext': 'mp4',
1777 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1778 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1779 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1780 'like_count': int,
1781 'availability': 'public',
1782 'playable_in_embed': True,
1783 'upload_date': '20131105',
1784 'description': 'md5:563ccbc698b39298481ca3c571169519',
1785 'channel_follower_count': int,
1786 'tags': 'count:24',
1787 'release_date': '20131106',
1788 'comment_count': int,
1789 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1790 'channel': 'WiiLikeToPlay',
1791 'categories': ['Gaming'],
1792 'release_timestamp': 1383701914,
1793 'live_status': 'was_live',
1794 'age_limit': 0,
1795 'duration': 10128,
1796 'view_count': int,
1797 'uploader': 'WiiLikeToPlay',
1798 'uploader_id': '@WLTP',
1799 'uploader_url': 'https://www.youtube.com/@WLTP',
1800 },
1801 }],
1802 'params': {'skip_download': True},
1803 'skip': 'Not multifeed anymore',
1804 },
1805 {
1806 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1807 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1808 'info_dict': {
1809 'id': 'gVfLd0zydlo',
1810 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1811 },
1812 'playlist_count': 2,
1813 'skip': 'Not multifeed anymore',
1814 },
1815 {
1816 'url': 'https://vid.plus/FlRa-iH7PGw',
1817 'only_matching': True,
1818 },
1819 {
1820 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1821 'only_matching': True,
1822 },
1823 {
1824 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1825 # Also tests cut-off URL expansion in video description (see
1826 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1827 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1828 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1829 'info_dict': {
1830 'id': 'lsguqyKfVQg',
1831 'ext': 'mp4',
1832 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1833 'alt_title': 'Dark Walk',
1834 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1835 'duration': 133,
1836 'upload_date': '20151119',
1837 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1838 'track': 'Dark Walk',
1839 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1840 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1841 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1842 'categories': ['Film & Animation'],
1843 'view_count': int,
1844 'live_status': 'not_live',
1845 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1846 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1847 'tags': 'count:13',
1848 'availability': 'public',
1849 'channel': 'IronSoulElf',
1850 'playable_in_embed': True,
1851 'like_count': int,
1852 'age_limit': 0,
1853 'channel_follower_count': int,
1854 },
1855 'params': {
1856 'skip_download': True,
1857 },
1858 },
1859 {
1860 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1861 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1862 'only_matching': True,
1863 },
1864 {
1865 # Video with yt:stretch=17:0
1866 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1867 'info_dict': {
1868 'id': 'Q39EVAstoRM',
1869 'ext': 'mp4',
1870 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1871 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1872 'upload_date': '20151107',
1873 },
1874 'params': {
1875 'skip_download': True,
1876 },
1877 'skip': 'This video does not exist.',
1878 },
1879 {
1880 # Video with incomplete 'yt:stretch=16:'
1881 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1882 'only_matching': True,
1883 },
1884 {
1885 # Video licensed under Creative Commons
1886 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1887 'info_dict': {
1888 'id': 'M4gD1WSo5mA',
1889 'ext': 'mp4',
1890 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1891 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1892 'duration': 721,
1893 'upload_date': '20150128',
1894 'license': 'Creative Commons Attribution license (reuse allowed)',
1895 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1896 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1897 'like_count': int,
1898 'age_limit': 0,
1899 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1900 'channel': 'The Berkman Klein Center for Internet & Society',
1901 'availability': 'public',
1902 'view_count': int,
1903 'categories': ['Education'],
1904 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1905 'live_status': 'not_live',
1906 'playable_in_embed': True,
1907 'channel_follower_count': int,
1908 'chapters': list,
1909 'uploader': 'The Berkman Klein Center for Internet & Society',
1910 'uploader_id': '@BKCHarvard',
1911 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1912 'timestamp': 1422422076,
1913 },
1914 'params': {
1915 'skip_download': True,
1916 },
1917 },
1918 {
1919 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1920 'info_dict': {
1921 'id': 'eQcmzGIKrzg',
1922 'ext': 'mp4',
1923 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1924 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1925 'duration': 4060,
1926 'upload_date': '20151120',
1927 'license': 'Creative Commons Attribution license (reuse allowed)',
1928 'playable_in_embed': True,
1929 'tags': 'count:12',
1930 'like_count': int,
1931 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1932 'age_limit': 0,
1933 'availability': 'public',
1934 'categories': ['News & Politics'],
1935 'channel': 'Bernie Sanders',
1936 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1937 'view_count': int,
1938 'live_status': 'not_live',
1939 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1940 'comment_count': int,
1941 'channel_follower_count': int,
1942 'chapters': list,
1943 'uploader': 'Bernie Sanders',
1944 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1945 'uploader_id': '@BernieSanders',
1946 'channel_is_verified': True,
1947 'heatmap': 'count:100',
1948 'timestamp': 1447987198,
1949 },
1950 'params': {
1951 'skip_download': True,
1952 },
1953 },
1954 {
1955 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1956 'only_matching': True,
1957 },
1958 {
1959 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1960 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1961 'only_matching': True,
1962 },
1963 {
1964 # Rental video preview
1965 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1966 'info_dict': {
1967 'id': 'uGpuVWrhIzE',
1968 'ext': 'mp4',
1969 'title': 'Piku - Trailer',
1970 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1971 'upload_date': '20150811',
1972 'license': 'Standard YouTube License',
1973 },
1974 'params': {
1975 'skip_download': True,
1976 },
1977 'skip': 'This video is not available.',
1978 },
1979 {
1980 # YouTube Red video with episode data
1981 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1982 'info_dict': {
1983 'id': 'iqKdEhx-dD4',
1984 'ext': 'mp4',
1985 'title': 'Isolation - Mind Field (Ep 1)',
1986 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1987 'duration': 2085,
1988 'upload_date': '20170118',
1989 'series': 'Mind Field',
1990 'season_number': 1,
1991 'episode_number': 1,
1992 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1993 'tags': 'count:12',
1994 'view_count': int,
1995 'availability': 'public',
1996 'age_limit': 0,
1997 'channel': 'Vsauce',
1998 'episode': 'Episode 1',
1999 'categories': ['Entertainment'],
2000 'season': 'Season 1',
2001 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
2002 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
2003 'like_count': int,
2004 'playable_in_embed': True,
2005 'live_status': 'not_live',
2006 'channel_follower_count': int,
2007 'uploader': 'Vsauce',
2008 'uploader_url': 'https://www.youtube.com/@Vsauce',
2009 'uploader_id': '@Vsauce',
2010 'comment_count': int,
2011 'channel_is_verified': True,
2012 'timestamp': 1484761047,
2013 },
2014 'params': {
2015 'skip_download': True,
2016 },
2017 'expected_warnings': [
2018 'Skipping DASH manifest',
2019 ],
2020 },
2021 {
2022 # The following content has been identified by the YouTube community
2023 # as inappropriate or offensive to some audiences.
2024 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
2025 'info_dict': {
2026 'id': '6SJNVb0GnPI',
2027 'ext': 'mp4',
2028 'title': 'Race Differences in Intelligence',
2029 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
2030 'duration': 965,
2031 'upload_date': '20140124',
2032 },
2033 'params': {
2034 'skip_download': True,
2035 },
2036 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
2037 },
2038 {
2039 # itag 212
2040 'url': '1t24XAntNCY',
2041 'only_matching': True,
2042 },
2043 {
2044 # geo restricted to JP
2045 'url': 'sJL6WA-aGkQ',
2046 'only_matching': True,
2047 },
2048 {
2049 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
2050 'only_matching': True,
2051 },
2052 {
2053 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
2054 'only_matching': True,
2055 },
2056 {
2057 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2058 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2059 'only_matching': True,
2060 },
2061 {
2062 # DRM protected
2063 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2064 'only_matching': True,
2065 },
2066 {
2067 # Video with unsupported adaptive stream type formats
2068 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2069 'info_dict': {
2070 'id': 'Z4Vy8R84T1U',
2071 'ext': 'mp4',
2072 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2073 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2074 'duration': 433,
2075 'upload_date': '20130923',
2076 'formats': 'maxcount:10',
2077 },
2078 'params': {
2079 'skip_download': True,
2080 'youtube_include_dash_manifest': False,
2081 },
2082 'skip': 'not actual anymore',
2083 },
2084 {
2085 # Youtube Music Auto-generated description
2086 # TODO: fix metadata extraction
2087 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2088 'info_dict': {
2089 'id': 'MgNrAu2pzNs',
2090 'ext': 'mp4',
2091 'title': 'Voyeur Girl',
2092 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2093 'upload_date': '20190312',
2094 'artists': ['Stephen'],
2095 'creators': ['Stephen'],
2096 'track': 'Voyeur Girl',
2097 'album': 'it\'s too much love to know my dear',
2098 'release_date': '20190313',
2099 'alt_title': 'Voyeur Girl',
2100 'view_count': int,
2101 'playable_in_embed': True,
2102 'like_count': int,
2103 'categories': ['Music'],
2104 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2105 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2106 'uploader': 'Stephen',
2107 'availability': 'public',
2108 'duration': 169,
2109 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2110 'age_limit': 0,
2111 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2112 'tags': 'count:11',
2113 'live_status': 'not_live',
2114 'channel_follower_count': int,
2115 },
2116 'params': {
2117 'skip_download': True,
2118 },
2119 },
2120 {
2121 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2122 'only_matching': True,
2123 },
2124 {
2125 # invalid -> valid video id redirection
2126 'url': 'DJztXj2GPfl',
2127 'info_dict': {
2128 'id': 'DJztXj2GPfk',
2129 'ext': 'mp4',
2130 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2131 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2132 'upload_date': '20090125',
2133 'artist': 'Panjabi MC',
2134 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2135 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2136 },
2137 'params': {
2138 'skip_download': True,
2139 },
2140 'skip': 'Video unavailable',
2141 },
2142 {
2143 # empty description results in an empty string
2144 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2145 'info_dict': {
2146 'id': 'x41yOUIvK2k',
2147 'ext': 'mp4',
2148 'title': 'IMG 3456',
2149 'description': '',
2150 'upload_date': '20170613',
2151 'view_count': int,
2152 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2153 'like_count': int,
2154 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2155 'tags': [],
2156 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2157 'availability': 'public',
2158 'age_limit': 0,
2159 'categories': ['Pets & Animals'],
2160 'duration': 7,
2161 'playable_in_embed': True,
2162 'live_status': 'not_live',
2163 'channel': 'l\'Or Vert asbl',
2164 'channel_follower_count': int,
2165 'uploader': 'l\'Or Vert asbl',
2166 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2167 'uploader_id': '@ElevageOrVert',
2168 'timestamp': 1497343210,
2169 },
2170 'params': {
2171 'skip_download': True,
2172 },
2173 },
2174 {
2175 # with '};' inside yt initial data (see [1])
2176 # see [2] for an example with '};' inside ytInitialPlayerResponse
2177 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2178 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2179 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2180 'info_dict': {
2181 'id': 'CHqg6qOn4no',
2182 'ext': 'mp4',
2183 'title': 'Part 77 Sort a list of simple types in c#',
2184 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2185 'upload_date': '20130831',
2186 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2187 'like_count': int,
2188 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2189 'live_status': 'not_live',
2190 'categories': ['Education'],
2191 'availability': 'public',
2192 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2193 'tags': 'count:12',
2194 'playable_in_embed': True,
2195 'age_limit': 0,
2196 'view_count': int,
2197 'duration': 522,
2198 'channel': 'kudvenkat',
2199 'comment_count': int,
2200 'channel_follower_count': int,
2201 'chapters': list,
2202 'uploader': 'kudvenkat',
2203 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2204 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2205 'channel_is_verified': True,
2206 'heatmap': 'count:100',
2207 'timestamp': 1377976349,
2208 },
2209 'params': {
2210 'skip_download': True,
2211 },
2212 },
2213 {
2214 # another example of '};' in ytInitialData
2215 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2216 'only_matching': True,
2217 },
2218 {
2219 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2220 'only_matching': True,
2221 },
2222 {
2223 # https://github.com/ytdl-org/youtube-dl/pull/28094
2224 'url': 'OtqTfy26tG0',
2225 'info_dict': {
2226 'id': 'OtqTfy26tG0',
2227 'ext': 'mp4',
2228 'title': 'Burn Out',
2229 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2230 'upload_date': '20141120',
2231 'artist': 'The Cinematic Orchestra',
2232 'track': 'Burn Out',
2233 'album': 'Every Day',
2234 'like_count': int,
2235 'live_status': 'not_live',
2236 'alt_title': 'Burn Out',
2237 'duration': 614,
2238 'age_limit': 0,
2239 'view_count': int,
2240 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2241 'creator': 'The Cinematic Orchestra',
2242 'channel': 'The Cinematic Orchestra',
2243 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2244 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2245 'availability': 'public',
2246 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2247 'categories': ['Music'],
2248 'playable_in_embed': True,
2249 'channel_follower_count': int,
2250 'uploader': 'The Cinematic Orchestra',
2251 'comment_count': int,
2252 },
2253 'params': {
2254 'skip_download': True,
2255 },
2256 },
2257 {
2258 # controversial video, only works with bpctr when authenticated with cookies
2259 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2260 'only_matching': True,
2261 },
2262 {
2263 # controversial video, requires bpctr/contentCheckOk
2264 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2265 'info_dict': {
2266 'id': 'SZJvDhaSDnc',
2267 'ext': 'mp4',
2268 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2269 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2270 'upload_date': '20140716',
2271 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2272 'duration': 170,
2273 'categories': ['News & Politics'],
2274 'view_count': int,
2275 'channel': 'CBS Mornings',
2276 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2277 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2278 'age_limit': 18,
2279 'availability': 'needs_auth',
2280 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2281 'like_count': int,
2282 'live_status': 'not_live',
2283 'playable_in_embed': True,
2284 'channel_follower_count': int,
2285 'uploader': 'CBS Mornings',
2286 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2287 'uploader_id': '@CBSMornings',
2288 'comment_count': int,
2289 'channel_is_verified': True,
2290 'timestamp': 1405513526,
2291 },
2292 },
2293 {
2294 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2295 'url': 'cBvYw8_A0vQ',
2296 'info_dict': {
2297 'id': 'cBvYw8_A0vQ',
2298 'ext': 'mp4',
2299 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2300 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2301 'upload_date': '20201120',
2302 'duration': 1456,
2303 'categories': ['Travel & Events'],
2304 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2305 'view_count': int,
2306 'channel': 'Walk around Japan',
2307 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2308 'thumbnail': 'https://i.ytimg.com/vi/cBvYw8_A0vQ/hqdefault.jpg',
2309 'age_limit': 0,
2310 'availability': 'public',
2311 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2312 'live_status': 'not_live',
2313 'playable_in_embed': True,
2314 'channel_follower_count': int,
2315 'uploader': 'Walk around Japan',
2316 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2317 'uploader_id': '@walkaroundjapan7124',
2318 'timestamp': 1605884416,
2319 },
2320 'params': {
2321 'skip_download': True,
2322 },
2323 }, {
2324 # Has multiple audio streams
2325 'url': 'WaOKSUlf4TM',
2326 'only_matching': True,
2327 }, {
2328 # Requires Premium: has format 141 when requested using YTM url
2329 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2330 'only_matching': True,
2331 }, {
2332 # multiple subtitles with same lang_code
2333 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2334 'only_matching': True,
2335 }, {
2336 # Force use android client fallback
2337 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2338 'info_dict': {
2339 'id': 'YOelRv7fMxY',
2340 'title': 'DIGGING A SECRET TUNNEL Part 1',
2341 'ext': '3gp',
2342 'upload_date': '20210624',
2343 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2344 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2345 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2346 'duration': 596,
2347 'categories': ['Entertainment'],
2348 'view_count': int,
2349 'channel': 'colinfurze',
2350 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2351 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2352 'age_limit': 0,
2353 'availability': 'public',
2354 'like_count': int,
2355 'live_status': 'not_live',
2356 'playable_in_embed': True,
2357 'channel_follower_count': int,
2358 'chapters': list,
2359 'uploader': 'colinfurze',
2360 'uploader_url': 'https://www.youtube.com/@colinfurze',
2361 'uploader_id': '@colinfurze',
2362 'comment_count': int,
2363 'channel_is_verified': True,
2364 'heatmap': 'count:100',
2365 },
2366 'params': {
2367 'format': '17', # 3gp format available on android
2368 'extractor_args': {'youtube': {'player_client': ['android']}},
2369 },
2370 'skip': 'android client broken',
2371 },
2372 {
2373 # Skip download of additional client configs (remix client config in this case)
2374 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2375 'only_matching': True,
2376 'params': {
2377 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2378 },
2379 }, {
2380 # shorts
2381 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2382 'only_matching': True,
2383 }, {
2384 'note': 'Storyboards',
2385 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2386 'info_dict': {
2387 'id': '5KLPxDtMqe8',
2388 'ext': 'mhtml',
2389 'format_id': 'sb0',
2390 'title': 'Your Brain is Plastic',
2391 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2392 'upload_date': '20140324',
2393 'like_count': int,
2394 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2395 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2396 'view_count': int,
2397 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2398 'playable_in_embed': True,
2399 'tags': 'count:12',
2400 'availability': 'public',
2401 'channel': 'SciShow',
2402 'live_status': 'not_live',
2403 'duration': 248,
2404 'categories': ['Education'],
2405 'age_limit': 0,
2406 'channel_follower_count': int,
2407 'chapters': list,
2408 'uploader': 'SciShow',
2409 'uploader_url': 'https://www.youtube.com/@SciShow',
2410 'uploader_id': '@SciShow',
2411 'comment_count': int,
2412 'channel_is_verified': True,
2413 'heatmap': 'count:100',
2414 'timestamp': 1395685455,
2415 }, 'params': {'format': 'mhtml', 'skip_download': True},
2416 }, {
2417 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2418 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2419 'info_dict': {
2420 'id': '2NUZ8W2llS4',
2421 'ext': 'mp4',
2422 'title': 'The NP that test your phone performance 🙂',
2423 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2424 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2425 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2426 'duration': 21,
2427 'view_count': int,
2428 'age_limit': 0,
2429 'categories': ['Gaming'],
2430 'tags': 'count:23',
2431 'playable_in_embed': True,
2432 'live_status': 'not_live',
2433 'upload_date': '20220103',
2434 'like_count': int,
2435 'availability': 'public',
2436 'channel': 'Leon Nguyen',
2437 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2438 'comment_count': int,
2439 'channel_follower_count': int,
2440 'uploader': 'Leon Nguyen',
2441 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2442 'uploader_id': '@LeonNguyen',
2443 'heatmap': 'count:100',
2444 'timestamp': 1641170939,
2445 },
2446 }, {
2447 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2448 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2449 'info_dict': {
2450 'id': 'mzZzzBU6lrM',
2451 'ext': 'mp4',
2452 'title': 'I Met GeorgeNotFound In Real Life...',
2453 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2454 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2455 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2456 'duration': 955,
2457 'view_count': int,
2458 'age_limit': 0,
2459 'categories': ['Entertainment'],
2460 'tags': 'count:26',
2461 'playable_in_embed': True,
2462 'live_status': 'not_live',
2463 'release_timestamp': 1641172509,
2464 'release_date': '20220103',
2465 'upload_date': '20220103',
2466 'like_count': int,
2467 'availability': 'public',
2468 'channel': 'Quackity',
2469 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2470 'channel_follower_count': int,
2471 'uploader': 'Quackity',
2472 'uploader_id': '@Quackity',
2473 'uploader_url': 'https://www.youtube.com/@Quackity',
2474 'comment_count': int,
2475 'channel_is_verified': True,
2476 'heatmap': 'count:100',
2477 'timestamp': 1641172509,
2478 },
2479 },
2480 { # continuous livestream.
2481 # Upload date was 2022-07-12T05:12:29-07:00, while stream start is 2022-07-12T15:59:30+00:00
2482 'url': 'https://www.youtube.com/watch?v=jfKfPfyJRdk',
2483 'info_dict': {
2484 'id': 'jfKfPfyJRdk',
2485 'ext': 'mp4',
2486 'channel_id': 'UCSJ4gkVC6NrvII8umztf0Ow',
2487 'like_count': int,
2488 'uploader': 'Lofi Girl',
2489 'categories': ['Music'],
2490 'concurrent_view_count': int,
2491 'playable_in_embed': True,
2492 'timestamp': 1657627949,
2493 'release_date': '20220712',
2494 'channel_url': 'https://www.youtube.com/channel/UCSJ4gkVC6NrvII8umztf0Ow',
2495 'description': 'md5:13a6f76df898f5674f9127139f3df6f7',
2496 'age_limit': 0,
2497 'thumbnail': 'https://i.ytimg.com/vi/jfKfPfyJRdk/maxresdefault.jpg',
2498 'release_timestamp': 1657641570,
2499 'uploader_url': 'https://www.youtube.com/@LofiGirl',
2500 'channel_follower_count': int,
2501 'channel_is_verified': True,
2502 'title': r're:^lofi hip hop radio 📚 - beats to relax/study to',
2503 'view_count': int,
2504 'live_status': 'is_live',
2505 'tags': 'count:32',
2506 'channel': 'Lofi Girl',
2507 'availability': 'public',
2508 'upload_date': '20220712',
2509 'uploader_id': '@LofiGirl',
2510 },
2511 'params': {'skip_download': True},
2512 }, {
2513 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2514 'info_dict': {
2515 'id': 'tjjjtzRLHvA',
2516 'ext': 'mp4',
2517 'title': 'ハッシュタグ無し };if window.ytcsi',
2518 'upload_date': '20220323',
2519 'like_count': int,
2520 'availability': 'unlisted',
2521 'channel': 'Lesmiscore',
2522 'thumbnail': r're:^https?://.*\.jpg',
2523 'age_limit': 0,
2524 'categories': ['Music'],
2525 'view_count': int,
2526 'description': '',
2527 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2528 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2529 'live_status': 'not_live',
2530 'playable_in_embed': True,
2531 'channel_follower_count': int,
2532 'duration': 6,
2533 'tags': [],
2534 'uploader_id': '@lesmiscore',
2535 'uploader': 'Lesmiscore',
2536 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2537 'timestamp': 1648005313,
2538 },
2539 }, {
2540 # Prefer primary title+description language metadata by default
2541 # Do not prefer translated description if primary is empty
2542 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2543 'info_dict': {
2544 'id': 'el3E4MbxRqQ',
2545 'ext': 'mp4',
2546 'title': 'dlp test video 2 - primary sv no desc',
2547 'description': '',
2548 'channel': 'cole-dlp-test-acc',
2549 'tags': [],
2550 'view_count': int,
2551 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2552 'like_count': int,
2553 'playable_in_embed': True,
2554 'availability': 'unlisted',
2555 'thumbnail': r're:^https?://.*\.jpg',
2556 'age_limit': 0,
2557 'duration': 5,
2558 'live_status': 'not_live',
2559 'upload_date': '20220908',
2560 'categories': ['People & Blogs'],
2561 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2562 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2563 'uploader_id': '@coletdjnz',
2564 'uploader': 'cole-dlp-test-acc',
2565 'timestamp': 1662677394,
2566 },
2567 'params': {'skip_download': True},
2568 }, {
2569 # Extractor argument: prefer translated title+description
2570 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2571 'info_dict': {
2572 'id': 'gHKT4uU8Zng',
2573 'ext': 'mp4',
2574 'channel': 'cole-dlp-test-acc',
2575 'tags': [],
2576 'duration': 5,
2577 'live_status': 'not_live',
2578 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2579 'upload_date': '20220729',
2580 'view_count': int,
2581 'categories': ['People & Blogs'],
2582 'thumbnail': r're:^https?://.*\.jpg',
2583 'title': 'dlp test video title translated (fr)',
2584 'availability': 'public',
2585 'age_limit': 0,
2586 'description': 'dlp test video description translated (fr)',
2587 'playable_in_embed': True,
2588 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2589 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2590 'uploader_id': '@coletdjnz',
2591 'uploader': 'cole-dlp-test-acc',
2592 'timestamp': 1659073275,
2593 'like_count': int,
2594 },
2595 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2596 'expected_warnings': [r'Preferring "fr" translated fields'],
2597 }, {
2598 'note': '6 channel audio',
2599 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2600 'only_matching': True,
2601 }, {
2602 'note': 'Multiple HLS formats with same itag',
2603 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2604 'info_dict': {
2605 'id': 'kX3nB4PpJko',
2606 'ext': 'mp4',
2607 'categories': ['Entertainment'],
2608 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2609 'live_status': 'not_live',
2610 'duration': 937,
2611 'channel_follower_count': int,
2612 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2613 'title': 'Last To Take Hand Off Jet, Keeps It!',
2614 'channel': 'MrBeast',
2615 'playable_in_embed': True,
2616 'view_count': int,
2617 'upload_date': '20221112',
2618 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2619 'age_limit': 0,
2620 'availability': 'public',
2621 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2622 'like_count': int,
2623 'tags': [],
2624 'uploader': 'MrBeast',
2625 'uploader_url': 'https://www.youtube.com/@MrBeast',
2626 'uploader_id': '@MrBeast',
2627 'comment_count': int,
2628 'channel_is_verified': True,
2629 'heatmap': 'count:100',
2630 },
2631 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2632 }, {
2633 'note': 'Audio formats with Dynamic Range Compression',
2634 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2635 'info_dict': {
2636 'id': 'Tq92D6wQ1mg',
2637 'ext': 'webm',
2638 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2639 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2640 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2641 'channel_follower_count': int,
2642 'description': 'md5:17eccca93a786d51bc67646756894066',
2643 'upload_date': '20191228',
2644 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2645 'playable_in_embed': True,
2646 'like_count': int,
2647 'categories': ['Entertainment'],
2648 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2649 'age_limit': 18,
2650 'channel': 'Projekt Melody',
2651 'view_count': int,
2652 'availability': 'needs_auth',
2653 'comment_count': int,
2654 'live_status': 'not_live',
2655 'duration': 106,
2656 'uploader': 'Projekt Melody',
2657 'uploader_id': '@ProjektMelody',
2658 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2659 'timestamp': 1577508724,
2660 },
2661 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2662 },
2663 {
2664 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2665 'info_dict': {
2666 'id': 'qVv6vCqciTM',
2667 'ext': 'mp4',
2668 'age_limit': 0,
2669 'comment_count': int,
2670 'chapters': 'count:13',
2671 'upload_date': '20221223',
2672 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2673 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2674 'like_count': int,
2675 'release_date': '20221223',
2676 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2677 '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】',
2678 'view_count': int,
2679 'playable_in_embed': True,
2680 'duration': 4438,
2681 'availability': 'public',
2682 'channel_follower_count': int,
2683 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2684 'categories': ['Entertainment'],
2685 'live_status': 'was_live',
2686 'release_timestamp': 1671793345,
2687 'channel': 'さなちゃんねる',
2688 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2689 'uploader': 'さなちゃんねる',
2690 'uploader_url': 'https://www.youtube.com/@sana_natori',
2691 'uploader_id': '@sana_natori',
2692 'channel_is_verified': True,
2693 'heatmap': 'count:100',
2694 'timestamp': 1671798112,
2695 },
2696 },
2697 {
2698 # Fallbacks when webpage and web client is unavailable
2699 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2700 'info_dict': {
2701 'id': 'wSSmNUl9Snw',
2702 'ext': 'mp4',
2703 # 'categories': ['Science & Technology'],
2704 'view_count': int,
2705 'chapters': 'count:2',
2706 'channel': 'Scott Manley',
2707 'like_count': int,
2708 'age_limit': 0,
2709 # 'availability': 'public',
2710 'channel_follower_count': int,
2711 'live_status': 'not_live',
2712 'upload_date': '20170831',
2713 'duration': 682,
2714 'tags': 'count:8',
2715 'uploader_url': 'https://www.youtube.com/@scottmanley',
2716 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2717 'uploader': 'Scott Manley',
2718 'uploader_id': '@scottmanley',
2719 'title': 'The Computer Hack That Saved Apollo 14',
2720 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2721 'thumbnail': r're:^https?://.*\.webp',
2722 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2723 'playable_in_embed': True,
2724 'comment_count': int,
2725 'channel_is_verified': True,
2726 'heatmap': 'count:100',
2727 },
2728 'params': {
2729 'extractor_args': {'youtube': {'player_client': ['ios'], 'player_skip': ['webpage']}},
2730 },
2731 },
2732 ]
2733
2734 _WEBPAGE_TESTS = [
2735 # YouTube <object> embed
2736 {
2737 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2738 'md5': '873c81d308b979f0e23ee7e620b312a3',
2739 'info_dict': {
2740 'id': 'msN87y-iEx0',
2741 'ext': 'mp4',
2742 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2743 'upload_date': '20080526',
2744 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2745 'age_limit': 0,
2746 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2747 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2748 'playable_in_embed': True,
2749 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2750 'like_count': int,
2751 'comment_count': int,
2752 'channel': 'Christopher Sykes',
2753 'live_status': 'not_live',
2754 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2755 'availability': 'public',
2756 'duration': 195,
2757 'view_count': int,
2758 'categories': ['Science & Technology'],
2759 'channel_follower_count': int,
2760 'uploader': 'Christopher Sykes',
2761 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2762 'uploader_id': '@ChristopherSykesDocumentaries',
2763 'heatmap': 'count:100',
2764 'timestamp': 1211825920,
2765 },
2766 'params': {
2767 'skip_download': True,
2768 },
2769 },
2770 ]
2771
2772 @classmethod
2773 def suitable(cls, url):
2774 from ..utils import parse_qs
2775
2776 qs = parse_qs(url)
2777 if qs.get('list', [None])[0]:
2778 return False
2779 return super().suitable(url)
2780
2781 def __init__(self, *args, **kwargs):
2782 super().__init__(*args, **kwargs)
2783 self._code_cache = {}
2784 self._player_cache = {}
2785
2786 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2787 lock = threading.Lock()
2788 start_time = time.time()
2789 formats = [f for f in formats if f.get('is_from_start')]
2790
2791 def refetch_manifest(format_id, delay):
2792 nonlocal formats, start_time, is_live
2793 if time.time() <= start_time + delay:
2794 return
2795
2796 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2797 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2798 microformats = traverse_obj(
2799 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2800 expected_type=dict)
2801 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2802 is_live = live_status == 'is_live'
2803 start_time = time.time()
2804
2805 def mpd_feed(format_id, delay):
2806 """
2807 @returns (manifest_url, manifest_stream_number, is_live) or None
2808 """
2809 for retry in self.RetryManager(fatal=False):
2810 with lock:
2811 refetch_manifest(format_id, delay)
2812
2813 f = next((f for f in formats if f['format_id'] == format_id), None)
2814 if not f:
2815 if not is_live:
2816 retry.error = f'{video_id}: Video is no longer live'
2817 else:
2818 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2819 continue
2820 return f['manifest_url'], f['manifest_stream_number'], is_live
2821 return None
2822
2823 for f in formats:
2824 f['is_live'] = is_live
2825 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2826 live_start_time, mpd_feed, not is_live and f.copy())
2827 if is_live:
2828 f['fragments'] = gen
2829 f['protocol'] = 'http_dash_segments_generator'
2830 else:
2831 f['fragments'] = LazyList(gen({}))
2832 del f['is_from_start']
2833
2834 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2835 FETCH_SPAN, MAX_DURATION = 5, 432000
2836
2837 mpd_url, stream_number, is_live = None, None, True
2838
2839 begin_index = 0
2840 download_start_time = ctx.get('start') or time.time()
2841
2842 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2843 if lack_early_segments:
2844 self.report_warning(bug_reports_message(
2845 'Starting download from the last 120 hours of the live stream since '
2846 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2847 lack_early_segments = True
2848
2849 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2850 fragments, fragment_base_url = None, None
2851
2852 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2853 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2854 # Obtain from MPD's maximum seq value
2855 old_mpd_url = mpd_url
2856 last_error = ctx.pop('last_error', None)
2857 expire_fast = immediate or last_error and isinstance(last_error, HTTPError) and last_error.status == 403
2858 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2859 or (mpd_url, stream_number, False))
2860 if not refresh_sequence:
2861 if expire_fast and not is_live:
2862 return False, last_seq
2863 elif old_mpd_url == mpd_url:
2864 return True, last_seq
2865 if manifestless_orig_fmt:
2866 fmt_info = manifestless_orig_fmt
2867 else:
2868 try:
2869 fmts, _ = self._extract_mpd_formats_and_subtitles(
2870 mpd_url, None, note=False, errnote=False, fatal=False)
2871 except ExtractorError:
2872 fmts = None
2873 if not fmts:
2874 no_fragment_score += 2
2875 return False, last_seq
2876 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2877 fragments = fmt_info['fragments']
2878 fragment_base_url = fmt_info['fragment_base_url']
2879 assert fragment_base_url
2880
2881 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2882 return True, _last_seq
2883
2884 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2885 while is_live:
2886 fetch_time = time.time()
2887 if no_fragment_score > 30:
2888 return
2889 if last_segment_url:
2890 # Obtain from "X-Head-Seqnum" header value from each segment
2891 try:
2892 urlh = self._request_webpage(
2893 last_segment_url, None, note=False, errnote=False, fatal=False)
2894 except ExtractorError:
2895 urlh = None
2896 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2897 if last_seq is None:
2898 no_fragment_score += 2
2899 last_segment_url = None
2900 continue
2901 else:
2902 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2903 no_fragment_score += 2
2904 if not should_continue:
2905 continue
2906
2907 if known_idx > last_seq:
2908 last_segment_url = None
2909 continue
2910
2911 last_seq += 1
2912
2913 if begin_index < 0 and known_idx < 0:
2914 # skip from the start when it's negative value
2915 known_idx = last_seq + begin_index
2916 if lack_early_segments:
2917 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2918 try:
2919 for idx in range(known_idx, last_seq):
2920 # do not update sequence here or you'll get skipped some part of it
2921 should_continue, _ = _extract_sequence_from_mpd(False, False)
2922 if not should_continue:
2923 known_idx = idx - 1
2924 raise ExtractorError('breaking out of outer loop')
2925 last_segment_url = urljoin(fragment_base_url, f'sq/{idx}')
2926 yield {
2927 'url': last_segment_url,
2928 'fragment_count': last_seq,
2929 }
2930 if known_idx == last_seq:
2931 no_fragment_score += 5
2932 else:
2933 no_fragment_score = 0
2934 known_idx = last_seq
2935 except ExtractorError:
2936 continue
2937
2938 if manifestless_orig_fmt:
2939 # Stop at the first iteration if running for post-live manifestless;
2940 # fragment count no longer increase since it starts
2941 break
2942
2943 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2944
2945 def _extract_player_url(self, *ytcfgs, webpage=None):
2946 player_url = traverse_obj(
2947 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2948 get_all=False, expected_type=str)
2949 if not player_url:
2950 return
2951 return urljoin('https://www.youtube.com', player_url)
2952
2953 def _download_player_url(self, video_id, fatal=False):
2954 res = self._download_webpage(
2955 'https://www.youtube.com/iframe_api',
2956 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2957 if res:
2958 player_version = self._search_regex(
2959 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2960 if player_version:
2961 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2962
2963 def _signature_cache_id(self, example_sig):
2964 """ Return a string representation of a signature """
2965 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2966
2967 @classmethod
2968 def _extract_player_info(cls, player_url):
2969 for player_re in cls._PLAYER_INFO_RE:
2970 id_m = re.search(player_re, player_url)
2971 if id_m:
2972 break
2973 else:
2974 raise ExtractorError(f'Cannot identify player {player_url!r}')
2975 return id_m.group('id')
2976
2977 def _load_player(self, video_id, player_url, fatal=True):
2978 player_id = self._extract_player_info(player_url)
2979 if player_id not in self._code_cache:
2980 code = self._download_webpage(
2981 player_url, video_id, fatal=fatal,
2982 note='Downloading player ' + player_id,
2983 errnote=f'Download of {player_url} failed')
2984 if code:
2985 self._code_cache[player_id] = code
2986 return self._code_cache.get(player_id)
2987
2988 def _extract_signature_function(self, video_id, player_url, example_sig):
2989 player_id = self._extract_player_info(player_url)
2990
2991 # Read from filesystem cache
2992 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2993 assert os.path.basename(func_id) == func_id
2994
2995 self.write_debug(f'Extracting signature function {func_id}')
2996 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2997
2998 if not cache_spec:
2999 code = self._load_player(video_id, player_url)
3000 if code:
3001 res = self._parse_sig_js(code)
3002 test_string = ''.join(map(chr, range(len(example_sig))))
3003 cache_spec = [ord(c) for c in res(test_string)]
3004 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
3005
3006 return lambda s: ''.join(s[i] for i in cache_spec)
3007
3008 def _print_sig_code(self, func, example_sig):
3009 if not self.get_param('youtube_print_sig_code'):
3010 return
3011
3012 def gen_sig_code(idxs):
3013 def _genslice(start, end, step):
3014 starts = '' if start == 0 else str(start)
3015 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
3016 steps = '' if step == 1 else (':%d' % step)
3017 return f's[{starts}{ends}{steps}]'
3018
3019 step = None
3020 # Quelch pyflakes warnings - start will be set when step is set
3021 start = '(Never used)'
3022 for i, prev in zip(idxs[1:], idxs[:-1]):
3023 if step is not None:
3024 if i - prev == step:
3025 continue
3026 yield _genslice(start, prev, step)
3027 step = None
3028 continue
3029 if i - prev in [-1, 1]:
3030 step = i - prev
3031 start = prev
3032 continue
3033 else:
3034 yield 's[%d]' % prev
3035 if step is None:
3036 yield 's[%d]' % i
3037 else:
3038 yield _genslice(start, i, step)
3039
3040 test_string = ''.join(map(chr, range(len(example_sig))))
3041 cache_res = func(test_string)
3042 cache_spec = [ord(c) for c in cache_res]
3043 expr_code = ' + '.join(gen_sig_code(cache_spec))
3044 signature_id_tuple = '({})'.format(', '.join(str(len(p)) for p in example_sig.split('.')))
3045 code = (f'if tuple(len(p) for p in s.split(\'.\')) == {signature_id_tuple}:\n'
3046 f' return {expr_code}\n')
3047 self.to_screen('Extracted signature function:\n' + code)
3048
3049 def _parse_sig_js(self, jscode):
3050 funcname = self._search_regex(
3051 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3052 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3053 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3054 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3055 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}\(a,\d+\))?',
3056 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3057 # Obsolete patterns
3058 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3059 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3060 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3061 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3062 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3063 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3064 jscode, 'Initial JS player signature function name', group='sig')
3065
3066 jsi = JSInterpreter(jscode)
3067 initial_function = jsi.extract_function(funcname)
3068 return lambda s: initial_function([s])
3069
3070 def _cached(self, func, *cache_id):
3071 def inner(*args, **kwargs):
3072 if cache_id not in self._player_cache:
3073 try:
3074 self._player_cache[cache_id] = func(*args, **kwargs)
3075 except ExtractorError as e:
3076 self._player_cache[cache_id] = e
3077 except Exception as e:
3078 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3079
3080 ret = self._player_cache[cache_id]
3081 if isinstance(ret, Exception):
3082 raise ret
3083 return ret
3084 return inner
3085
3086 def _decrypt_signature(self, s, video_id, player_url):
3087 """Turn the encrypted s field into a working signature"""
3088 extract_sig = self._cached(
3089 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3090 func = extract_sig(video_id, player_url, s)
3091 self._print_sig_code(func, s)
3092 return func(s)
3093
3094 def _decrypt_nsig(self, s, video_id, player_url):
3095 """Turn the encrypted n field into a working signature"""
3096 if player_url is None:
3097 raise ExtractorError('Cannot decrypt nsig without player_url')
3098 player_url = urljoin('https://www.youtube.com', player_url)
3099
3100 try:
3101 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3102 except ExtractorError as e:
3103 raise ExtractorError('Unable to extract nsig function code', cause=e)
3104 if self.get_param('youtube_print_sig_code'):
3105 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3106
3107 try:
3108 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3109 ret = extract_nsig(jsi, func_code)(s)
3110 except JSInterpreter.Exception as e:
3111 try:
3112 jsi = PhantomJSwrapper(self, timeout=5000)
3113 except ExtractorError:
3114 raise e
3115 self.report_warning(
3116 f'Native nsig extraction failed: Trying with PhantomJS\n'
3117 f' n = {s} ; player = {player_url}', video_id)
3118 self.write_debug(e, only_once=True)
3119
3120 args, func_body = func_code
3121 ret = jsi.execute(
3122 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3123 video_id=video_id, note='Executing signature code').strip()
3124
3125 self.write_debug(f'Decrypted nsig {s} => {ret}')
3126 return ret
3127
3128 def _extract_n_function_name(self, jscode):
3129 funcname, idx = self._search_regex(
3130 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3131 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3132 if not idx:
3133 return funcname
3134
3135 return json.loads(js_to_json(self._search_regex(
3136 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
3137 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3138
3139 def _extract_n_function_code(self, video_id, player_url):
3140 player_id = self._extract_player_info(player_url)
3141 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3142 jscode = func_code or self._load_player(video_id, player_url)
3143 jsi = JSInterpreter(jscode)
3144
3145 if func_code:
3146 return jsi, player_id, func_code
3147
3148 func_name = self._extract_n_function_name(jscode)
3149
3150 # For redundancy
3151 func_code = self._search_regex(
3152 rf'''(?xs){func_name}\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3153 # NB: The end of the regex is intentionally kept strict
3154 {{(?P<code>.+?}}\s*return\ [\w$]+.join\(""\))}};''',
3155 jscode, 'nsig function', group=('var', 'code'), default=None)
3156 if func_code:
3157 func_code = ([func_code[0]], func_code[1])
3158 else:
3159 self.write_debug('Extracting nsig function with jsinterp')
3160 func_code = jsi.extract_function_code(func_name)
3161
3162 self.cache.store('youtube-nsig', player_id, func_code)
3163 return jsi, player_id, func_code
3164
3165 def _extract_n_function_from_code(self, jsi, func_code):
3166 func = jsi.extract_function_from_code(*func_code)
3167
3168 def extract_nsig(s):
3169 try:
3170 ret = func([s])
3171 except JSInterpreter.Exception:
3172 raise
3173 except Exception as e:
3174 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3175
3176 if ret.startswith('enhanced_except_'):
3177 raise JSInterpreter.Exception('Signature function returned an exception')
3178 return ret
3179
3180 return extract_nsig
3181
3182 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3183 """
3184 Extract signatureTimestamp (sts)
3185 Required to tell API what sig/player version is in use.
3186 """
3187 sts = None
3188 if isinstance(ytcfg, dict):
3189 sts = int_or_none(ytcfg.get('STS'))
3190
3191 if not sts:
3192 # Attempt to extract from player
3193 if player_url is None:
3194 error_msg = 'Cannot extract signature timestamp without player_url.'
3195 if fatal:
3196 raise ExtractorError(error_msg)
3197 self.report_warning(error_msg)
3198 return
3199 code = self._load_player(video_id, player_url, fatal=fatal)
3200 if code:
3201 sts = int_or_none(self._search_regex(
3202 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3203 'JS player signature timestamp', group='sts', fatal=fatal))
3204 return sts
3205
3206 def _mark_watched(self, video_id, player_responses):
3207 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3208 label = 'fully ' if is_full else ''
3209 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3210 expected_type=url_or_none)
3211 if not url:
3212 self.report_warning(f'Unable to mark {label}watched')
3213 return
3214 parsed_url = urllib.parse.urlparse(url)
3215 qs = urllib.parse.parse_qs(parsed_url.query)
3216
3217 # cpn generation algorithm is reverse engineered from base.js.
3218 # In fact it works even with dummy cpn.
3219 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3220 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(16))
3221
3222 # # more consistent results setting it to right before the end
3223 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3224
3225 qs.update({
3226 'ver': ['2'],
3227 'cpn': [cpn],
3228 'cmt': video_length,
3229 'el': 'detailpage', # otherwise defaults to "shorts"
3230 })
3231
3232 if is_full:
3233 # these seem to mark watchtime "history" in the real world
3234 # they're required, so send in a single value
3235 qs.update({
3236 'st': 0,
3237 'et': video_length,
3238 })
3239
3240 url = urllib.parse.urlunparse(
3241 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3242
3243 self._download_webpage(
3244 url, video_id, f'Marking {label}watched',
3245 'Unable to mark watched', fatal=False)
3246
3247 @classmethod
3248 def _extract_from_webpage(cls, url, webpage):
3249 # Invidious Instances
3250 # https://github.com/yt-dlp/yt-dlp/issues/195
3251 # https://github.com/iv-org/invidious/pull/1730
3252 mobj = re.search(
3253 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3254 webpage)
3255 if mobj:
3256 yield cls.url_result(mobj.group('url'), cls)
3257 raise cls.StopExtraction
3258
3259 yield from super()._extract_from_webpage(url, webpage)
3260
3261 # lazyYT YouTube embed
3262 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3263 yield cls.url_result(unescapeHTML(id_), cls, id_)
3264
3265 # Wordpress "YouTube Video Importer" plugin
3266 for m in re.findall(r'''(?x)<div[^>]+
3267 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3268 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3269 yield cls.url_result(m[-1], cls, m[-1])
3270
3271 @classmethod
3272 def extract_id(cls, url):
3273 video_id = cls.get_temp_id(url)
3274 if not video_id:
3275 raise ExtractorError(f'Invalid URL: {url}')
3276 return video_id
3277
3278 def _extract_chapters_from_json(self, data, duration):
3279 chapter_list = traverse_obj(
3280 data, (
3281 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3282 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters',
3283 ), expected_type=list)
3284
3285 return self._extract_chapters_helper(
3286 chapter_list,
3287 start_function=lambda chapter: float_or_none(
3288 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3289 title_function=lambda chapter: traverse_obj(
3290 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3291 duration=duration)
3292
3293 def _extract_chapters_from_engagement_panel(self, data, duration):
3294 content_list = traverse_obj(
3295 data,
3296 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3297 expected_type=list)
3298 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3299 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3300
3301 return next(filter(None, (
3302 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3303 chapter_time, chapter_title, duration)
3304 for contents in content_list)), [])
3305
3306 def _extract_heatmap(self, data):
3307 return traverse_obj(data, (
3308 'frameworkUpdates', 'entityBatchUpdate', 'mutations',
3309 lambda _, v: v['payload']['macroMarkersListEntity']['markersList']['markerType'] == 'MARKER_TYPE_HEATMAP',
3310 'payload', 'macroMarkersListEntity', 'markersList', 'markers', ..., {
3311 'start_time': ('startMillis', {functools.partial(float_or_none, scale=1000)}),
3312 'end_time': {lambda x: (int(x['startMillis']) + int(x['durationMillis'])) / 1000},
3313 'value': ('intensityScoreNormalized', {float_or_none}),
3314 })) or None
3315
3316 def _extract_comment(self, entities, parent=None):
3317 comment_entity_payload = get_first(entities, ('payload', 'commentEntityPayload', {dict}))
3318 if not (comment_id := traverse_obj(comment_entity_payload, ('properties', 'commentId', {str}))):
3319 return
3320
3321 toolbar_entity_payload = get_first(entities, ('payload', 'engagementToolbarStateEntityPayload', {dict}))
3322 time_text = traverse_obj(comment_entity_payload, ('properties', 'publishedTime', {str})) or ''
3323
3324 return {
3325 'id': comment_id,
3326 'parent': parent or 'root',
3327 **traverse_obj(comment_entity_payload, {
3328 'text': ('properties', 'content', 'content', {str}),
3329 'like_count': ('toolbar', 'likeCountA11y', {parse_count}),
3330 'author_id': ('author', 'channelId', {self.ucid_or_none}),
3331 'author': ('author', 'displayName', {str}),
3332 'author_thumbnail': ('author', 'avatarThumbnailUrl', {url_or_none}),
3333 'author_is_uploader': ('author', 'isCreator', {bool}),
3334 'author_is_verified': ('author', 'isVerified', {bool}),
3335 'author_url': ('author', 'channelCommand', 'innertubeCommand', (
3336 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'),
3337 ), {lambda x: urljoin('https://www.youtube.com', x)}),
3338 }, get_all=False),
3339 'is_favorited': (None if toolbar_entity_payload is None else
3340 toolbar_entity_payload.get('heartState') == 'TOOLBAR_HEART_STATE_HEARTED'),
3341 '_time_text': time_text, # FIXME: non-standard, but we need a way of showing that it is an estimate.
3342 'timestamp': self._parse_time_text(time_text),
3343 }
3344
3345 def _extract_comment_old(self, comment_renderer, parent=None):
3346 comment_id = comment_renderer.get('commentId')
3347 if not comment_id:
3348 return
3349
3350 info = {
3351 'id': comment_id,
3352 'text': self._get_text(comment_renderer, 'contentText'),
3353 'like_count': self._get_count(comment_renderer, 'voteCount'),
3354 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3355 'author': self._get_text(comment_renderer, 'authorText'),
3356 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3357 'parent': parent or 'root',
3358 }
3359
3360 # Timestamp is an estimate calculated from the current time and time_text
3361 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3362 timestamp = self._parse_time_text(time_text)
3363
3364 info.update({
3365 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3366 '_time_text': time_text,
3367 'timestamp': timestamp,
3368 })
3369
3370 info['author_url'] = urljoin(
3371 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3372 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3373 expected_type=str, get_all=False))
3374
3375 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3376 if author_is_uploader is not None:
3377 info['author_is_uploader'] = author_is_uploader
3378
3379 comment_abr = traverse_obj(
3380 comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3381 if comment_abr is not None:
3382 info['is_favorited'] = 'creatorHeart' in comment_abr
3383
3384 badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
3385 if self._has_badge(badges, BadgeType.VERIFIED):
3386 info['author_is_verified'] = True
3387
3388 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3389 if is_pinned:
3390 info['is_pinned'] = True
3391
3392 return info
3393
3394 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3395
3396 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3397
3398 def extract_header(contents):
3399 _continuation = None
3400 for content in contents:
3401 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3402 expected_comment_count = self._get_count(
3403 comments_header_renderer, 'countText', 'commentsCount')
3404
3405 if expected_comment_count is not None:
3406 tracker['est_total'] = expected_comment_count
3407 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3408 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3409
3410 sort_menu_item = try_get(
3411 comments_header_renderer,
3412 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3413 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3414
3415 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3416 if not _continuation:
3417 continue
3418
3419 sort_text = str_or_none(sort_menu_item.get('title'))
3420 if not sort_text:
3421 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3422 self.to_screen(f'Sorting comments by {sort_text.lower()}')
3423 break
3424 return _continuation
3425
3426 def extract_thread(contents, entity_payloads):
3427 if not parent:
3428 tracker['current_page_thread'] = 0
3429 for content in contents:
3430 if not parent and tracker['total_parent_comments'] >= max_parents:
3431 yield
3432 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3433
3434 # old comment format
3435 if not entity_payloads:
3436 comment_renderer = get_first(
3437 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3438 expected_type=dict, default={})
3439
3440 comment = self._extract_comment_old(comment_renderer, parent)
3441
3442 # new comment format
3443 else:
3444 view_model = (
3445 traverse_obj(comment_thread_renderer, ('commentViewModel', 'commentViewModel', {dict}))
3446 or traverse_obj(content, ('commentViewModel', {dict})))
3447 comment_keys = traverse_obj(view_model, (('commentKey', 'toolbarStateKey'), {str}))
3448 if not comment_keys:
3449 continue
3450 entities = traverse_obj(entity_payloads, lambda _, v: v['entityKey'] in comment_keys)
3451 comment = self._extract_comment(entities, parent)
3452 if comment:
3453 comment['is_pinned'] = traverse_obj(view_model, ('pinnedText', {str})) is not None
3454
3455 if not comment:
3456 continue
3457 comment_id = comment['id']
3458
3459 if comment.get('is_pinned'):
3460 tracker['pinned_comment_ids'].add(comment_id)
3461 # Sometimes YouTube may break and give us infinite looping comments.
3462 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3463 if comment_id in tracker['seen_comment_ids']:
3464 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3465 # Pinned comments may appear a second time in newest first sort
3466 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3467 continue
3468 self.report_warning(
3469 'Detected YouTube comments looping. Stopping comment extraction '
3470 f'{"for this thread" if parent else ""} as we probably cannot get any more.')
3471 yield
3472 else:
3473 tracker['seen_comment_ids'].add(comment['id'])
3474
3475 tracker['running_total'] += 1
3476 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3477 yield comment
3478
3479 # Attempt to get the replies
3480 comment_replies_renderer = try_get(
3481 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3482
3483 if comment_replies_renderer:
3484 tracker['current_page_thread'] += 1
3485 comment_entries_iter = self._comment_entries(
3486 comment_replies_renderer, ytcfg, video_id,
3487 parent=comment.get('id'), tracker=tracker)
3488 yield from itertools.islice(comment_entries_iter, min(
3489 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3490
3491 # Keeps track of counts across recursive calls
3492 if not tracker:
3493 tracker = {
3494 'running_total': 0,
3495 'est_total': None,
3496 'current_page_thread': 0,
3497 'total_parent_comments': 0,
3498 'total_reply_comments': 0,
3499 'seen_comment_ids': set(),
3500 'pinned_comment_ids': set(),
3501 }
3502
3503 # TODO: Deprecated
3504 # YouTube comments have a max depth of 2
3505 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3506 if max_depth:
3507 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3508 'Set max replies in the max-comments extractor argument instead')
3509 if max_depth == 1 and parent:
3510 return
3511
3512 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = (
3513 int_or_none(p, default=sys.maxsize) for p in self._configuration_arg('max_comments') + [''] * 4)
3514
3515 continuation = self._extract_continuation(root_continuation_data)
3516
3517 response = None
3518 is_forced_continuation = False
3519 is_first_continuation = parent is None
3520 if is_first_continuation and not continuation:
3521 # Sometimes you can get comments by generating the continuation yourself,
3522 # even if YouTube initially reports them being disabled - e.g. stories comments.
3523 # Note: if the comment section is actually disabled, YouTube may return a response with
3524 # required check_get_keys missing. So we will disable that check initially in this case.
3525 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3526 is_forced_continuation = True
3527
3528 continuation_items_path = (
3529 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3530 for page_num in itertools.count(0):
3531 if not continuation:
3532 break
3533 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3534 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3535 if page_num == 0:
3536 if is_first_continuation:
3537 note_prefix = 'Downloading comment section API JSON'
3538 else:
3539 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3540 tracker['current_page_thread'], comment_prog_str)
3541 else:
3542 note_prefix = '{}Downloading comment{} API JSON page {} {}'.format(
3543 ' ' if parent else '', ' replies' if parent else '',
3544 page_num, comment_prog_str)
3545
3546 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3547 # Ignore check if YouTube says the comment count is 0.
3548 check_get_keys = None
3549 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3550 check_get_keys = [[*continuation_items_path, ..., (
3551 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentViewModel', 'commentRenderer'))]]
3552 try:
3553 response = self._extract_response(
3554 item_id=None, query=continuation,
3555 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3556 check_get_keys=check_get_keys)
3557 except ExtractorError as e:
3558 # Ignore incomplete data error for replies if retries didn't work.
3559 # This is to allow any other parent comments and comment threads to be downloaded.
3560 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3561 if 'incomplete data' in str(e).lower() and parent:
3562 if self.get_param('ignoreerrors') in (True, 'only_download'):
3563 self.report_warning(
3564 'Received incomplete data for a comment reply thread and retrying did not help. '
3565 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
3566 return
3567 else:
3568 raise ExtractorError(
3569 'Incomplete data received for comment reply thread. '
3570 'Pass --ignore-errors to ignore and allow rest of comments to download.',
3571 expected=True)
3572 raise
3573 is_forced_continuation = False
3574 continuation = None
3575 mutations = traverse_obj(response, ('frameworkUpdates', 'entityBatchUpdate', 'mutations', ..., {dict}))
3576 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3577 if is_first_continuation:
3578 continuation = extract_header(continuation_items)
3579 is_first_continuation = False
3580 if continuation:
3581 break
3582 continue
3583
3584 for entry in extract_thread(continuation_items, mutations):
3585 if not entry:
3586 return
3587 yield entry
3588 continuation = self._extract_continuation({'contents': continuation_items})
3589 if continuation:
3590 break
3591
3592 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3593 if message and not parent and tracker['running_total'] == 0:
3594 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3595 raise self.CommentsDisabled
3596
3597 @staticmethod
3598 def _generate_comment_continuation(video_id):
3599 """
3600 Generates initial comment section continuation token from given video id
3601 """
3602 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3603 return base64.b64encode(token.encode()).decode()
3604
3605 def _get_comments(self, ytcfg, video_id, contents, webpage):
3606 """Entry for comment extraction"""
3607 def _real_comment_extract(contents):
3608 renderer = next((
3609 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3610 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3611 yield from self._comment_entries(renderer, ytcfg, video_id)
3612
3613 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3614 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3615
3616 @staticmethod
3617 def _get_checkok_params():
3618 return {'contentCheckOk': True, 'racyCheckOk': True}
3619
3620 @classmethod
3621 def _generate_player_context(cls, sts=None):
3622 context = {
3623 'html5Preference': 'HTML5_PREF_WANTS',
3624 }
3625 if sts is not None:
3626 context['signatureTimestamp'] = sts
3627 return {
3628 'playbackContext': {
3629 'contentPlaybackContext': context,
3630 },
3631 **cls._get_checkok_params(),
3632 }
3633
3634 @staticmethod
3635 def _is_agegated(player_response):
3636 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3637 return True
3638
3639 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3640 AGE_GATE_REASONS = (
3641 'confirm your age', 'age-restricted', 'inappropriate', # reason
3642 'age_verification_required', 'age_check_required', # status
3643 )
3644 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3645
3646 @staticmethod
3647 def _is_unplayable(player_response):
3648 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3649
3650 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3651
3652 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3653 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3654 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3655 headers = self.generate_api_headers(
3656 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3657
3658 yt_query = {
3659 'videoId': video_id,
3660 }
3661
3662 pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0]
3663 if pp_arg:
3664 yt_query['params'] = pp_arg
3665
3666 yt_query.update(self._generate_player_context(sts))
3667 return self._extract_response(
3668 item_id=video_id, ep='player', query=yt_query,
3669 ytcfg=player_ytcfg, headers=headers, fatal=True,
3670 default_client=client,
3671 note='Downloading {} player API JSON'.format(client.replace('_', ' ').strip()),
3672 ) or None
3673
3674 def _get_requested_clients(self, url, smuggled_data):
3675 requested_clients = []
3676 android_clients = []
3677 default = ['ios', 'web']
3678 allowed_clients = sorted(
3679 (client for client in INNERTUBE_CLIENTS if client[:1] != '_'),
3680 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3681 for client in self._configuration_arg('player_client'):
3682 if client == 'default':
3683 requested_clients.extend(default)
3684 elif client == 'all':
3685 requested_clients.extend(allowed_clients)
3686 elif client not in allowed_clients:
3687 self.report_warning(f'Skipping unsupported client {client}')
3688 elif client.startswith('android'):
3689 android_clients.append(client)
3690 else:
3691 requested_clients.append(client)
3692 # Force deprioritization of broken Android clients for format de-duplication
3693 requested_clients.extend(android_clients)
3694 if not requested_clients:
3695 requested_clients = default
3696
3697 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3698 requested_clients.extend(
3699 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3700
3701 return orderedSet(requested_clients)
3702
3703 def _invalid_player_response(self, pr, video_id):
3704 # YouTube may return a different video player response than expected.
3705 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3706 if (pr_id := traverse_obj(pr, ('videoDetails', 'videoId'))) != video_id:
3707 return pr_id
3708
3709 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3710 initial_pr = None
3711 if webpage:
3712 initial_pr = self._search_json(
3713 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3714
3715 prs = []
3716 if initial_pr and not self._invalid_player_response(initial_pr, video_id):
3717 # Android player_response does not have microFormats which are needed for
3718 # extraction of some data. So we return the initial_pr with formats
3719 # stripped out even if not requested by the user
3720 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3721 prs.append({**initial_pr, 'streamingData': None})
3722
3723 all_clients = set(clients)
3724 clients = clients[::-1]
3725
3726 def append_client(*client_names):
3727 """ Append the first client name that exists but not already used """
3728 for client_name in client_names:
3729 actual_client = _split_innertube_client(client_name)[0]
3730 if actual_client in INNERTUBE_CLIENTS:
3731 if actual_client not in all_clients:
3732 clients.append(client_name)
3733 all_clients.add(actual_client)
3734 return
3735
3736 tried_iframe_fallback = False
3737 player_url = None
3738 skipped_clients = {}
3739 while clients:
3740 client, base_client, variant = _split_innertube_client(clients.pop())
3741 player_ytcfg = master_ytcfg if client == 'web' else {}
3742 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3743 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3744
3745 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3746 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3747 if 'js' in self._configuration_arg('player_skip'):
3748 require_js_player = False
3749 player_url = None
3750
3751 if not player_url and not tried_iframe_fallback and require_js_player:
3752 player_url = self._download_player_url(video_id)
3753 tried_iframe_fallback = True
3754
3755 try:
3756 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3757 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3758 except ExtractorError as e:
3759 self.report_warning(e)
3760 continue
3761
3762 if pr_id := self._invalid_player_response(pr, video_id):
3763 skipped_clients[client] = pr_id
3764 elif pr:
3765 # Save client name for introspection later
3766 name = short_client_name(client)
3767 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3768 sd[STREAMING_DATA_CLIENT_NAME] = name
3769 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3770 f[STREAMING_DATA_CLIENT_NAME] = name
3771 prs.append(pr)
3772
3773 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3774 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3775 append_client(f'{base_client}_creator')
3776 elif self._is_agegated(pr):
3777 if variant == 'tv_embedded':
3778 append_client(f'{base_client}_embedded')
3779 elif not variant:
3780 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3781
3782 if skipped_clients:
3783 self.report_warning(
3784 f'Skipping player responses from {"/".join(skipped_clients)} clients '
3785 f'(got player responses for video "{"/".join(set(skipped_clients.values()))}" instead of "{video_id}")')
3786 if not prs:
3787 raise ExtractorError(
3788 'All player responses are invalid. Your IP is likely being blocked by Youtube', expected=True)
3789 elif not prs:
3790 raise ExtractorError('Failed to extract any player response')
3791 return prs, player_url
3792
3793 def _needs_live_processing(self, live_status, duration):
3794 if (live_status == 'is_live' and self.get_param('live_from_start')
3795 or live_status == 'post_live' and (duration or 0) > 2 * 3600):
3796 return live_status
3797
3798 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3799 CHUNK_SIZE = 10 << 20
3800 PREFERRED_LANG_VALUE = 10
3801 original_language = None
3802 itags, stream_ids = collections.defaultdict(set), []
3803 itag_qualities, res_qualities = {}, {0: None}
3804 q = qualities([
3805 # Normally tiny is the smallest video-only formats. But
3806 # audio-only formats with unknown quality may get tagged as tiny
3807 'tiny',
3808 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3809 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres',
3810 ])
3811 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3812 format_types = self._configuration_arg('formats')
3813 all_formats = 'duplicate' in format_types
3814 if self._configuration_arg('include_duplicate_formats'):
3815 all_formats = True
3816 self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
3817 'Use formats=duplicate extractor argument instead')
3818
3819 def build_fragments(f):
3820 return LazyList({
3821 'url': update_url_query(f['url'], {
3822 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}',
3823 }),
3824 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3825
3826 for fmt in streaming_formats:
3827 if fmt.get('targetDurationSec'):
3828 continue
3829
3830 itag = str_or_none(fmt.get('itag'))
3831 audio_track = fmt.get('audioTrack') or {}
3832 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3833 if not all_formats:
3834 if stream_id in stream_ids:
3835 continue
3836
3837 quality = fmt.get('quality')
3838 height = int_or_none(fmt.get('height'))
3839 if quality == 'tiny' or not quality:
3840 quality = fmt.get('audioQuality', '').lower() or quality
3841 # The 3gp format (17) in android client has a quality of "small",
3842 # but is actually worse than other formats
3843 if itag == '17':
3844 quality = 'tiny'
3845 if quality:
3846 if itag:
3847 itag_qualities[itag] = quality
3848 if height:
3849 res_qualities[height] = quality
3850 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3851 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3852 # number of fragment that would subsequently requested with (`&sq=N`)
3853 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3854 continue
3855
3856 fmt_url = fmt.get('url')
3857 if not fmt_url:
3858 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3859 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3860 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3861 if not all((sc, fmt_url, player_url, encrypted_sig)):
3862 continue
3863 try:
3864 fmt_url += '&{}={}'.format(
3865 traverse_obj(sc, ('sp', -1)) or 'signature',
3866 self._decrypt_signature(encrypted_sig, video_id, player_url),
3867 )
3868 except ExtractorError as e:
3869 self.report_warning('Signature extraction failed: Some formats may be missing',
3870 video_id=video_id, only_once=True)
3871 self.write_debug(e, only_once=True)
3872 continue
3873
3874 query = parse_qs(fmt_url)
3875 throttled = False
3876 if query.get('n'):
3877 try:
3878 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3879 fmt_url = update_url_query(fmt_url, {
3880 'n': decrypt_nsig(query['n'][0], video_id, player_url),
3881 })
3882 except ExtractorError as e:
3883 phantomjs_hint = ''
3884 if isinstance(e, JSInterpreter.Exception):
3885 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3886 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3887 if player_url:
3888 self.report_warning(
3889 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3890 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3891 self.write_debug(e, only_once=True)
3892 else:
3893 self.report_warning(
3894 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3895 video_id=video_id, only_once=True)
3896 throttled = True
3897
3898 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3899 is_default = audio_track.get('audioIsDefault')
3900 is_descriptive = 'descriptive' in (audio_track.get('displayName') or '').lower()
3901 language_code = audio_track.get('id', '').split('.')[0]
3902 if language_code and is_default:
3903 original_language = language_code
3904
3905 format_duration = traverse_obj(fmt, ('approxDurationMs', {lambda x: float_or_none(x, 1000)}))
3906 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3907 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3908 # Make sure to avoid false positives with small duration differences.
3909 # E.g. __2ABJjxzNo, ySuUZEjARPY
3910 is_damaged = try_call(lambda: format_duration < duration // 2)
3911 if is_damaged:
3912 self.report_warning(
3913 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3914
3915 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3916 # Android client formats are broken due to integrity check enforcement
3917 # Ref: https://github.com/yt-dlp/yt-dlp/issues/9554
3918 is_broken = client_name and client_name.startswith(short_client_name('android'))
3919 if is_broken:
3920 self.report_warning(
3921 f'{video_id}: Android client formats are broken and may yield HTTP Error 403. '
3922 'They will be deprioritized', only_once=True)
3923
3924 name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
3925 fps = int_or_none(fmt.get('fps')) or 0
3926 dct = {
3927 'asr': int_or_none(fmt.get('audioSampleRate')),
3928 'filesize': int_or_none(fmt.get('contentLength')),
3929 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3930 'format_note': join_nonempty(
3931 join_nonempty(audio_track.get('displayName'), is_default and ' (default)', delim=''),
3932 name, fmt.get('isDrc') and 'DRC',
3933 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3934 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3935 throttled and 'THROTTLED', is_damaged and 'DAMAGED', is_broken and 'BROKEN',
3936 (self.get_param('verbose') or all_formats) and client_name,
3937 delim=', '),
3938 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3939 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
3940 + (100 if 'Premium' in name else 0)),
3941 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
3942 'audio_channels': fmt.get('audioChannels'),
3943 'height': height,
3944 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3945 'has_drm': bool(fmt.get('drmFamilies')),
3946 'tbr': tbr,
3947 'filesize_approx': filesize_from_tbr(tbr, format_duration),
3948 'url': fmt_url,
3949 'width': int_or_none(fmt.get('width')),
3950 'language': join_nonempty(language_code, 'desc' if is_descriptive else '') or None,
3951 'language_preference': PREFERRED_LANG_VALUE if is_default else -10 if is_descriptive else -1,
3952 # Strictly de-prioritize broken, damaged and 3gp formats
3953 'preference': -20 if is_broken else -10 if is_damaged else -2 if itag == '17' else None,
3954 }
3955 mime_mobj = re.match(
3956 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3957 if mime_mobj:
3958 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3959 dct.update(parse_codecs(mime_mobj.group(2)))
3960 if itag:
3961 itags[itag].add(('https', dct.get('language')))
3962 stream_ids.append(stream_id)
3963 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3964 if single_stream and dct.get('ext'):
3965 dct['container'] = dct['ext'] + '_dash'
3966
3967 if (all_formats or 'dashy' in format_types) and dct['filesize']:
3968 yield {
3969 **dct,
3970 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3971 'protocol': 'http_dash_segments',
3972 'fragments': build_fragments(dct),
3973 }
3974 if all_formats or 'dashy' not in format_types:
3975 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3976 yield dct
3977
3978 needs_live_processing = self._needs_live_processing(live_status, duration)
3979 skip_bad_formats = 'incomplete' not in format_types
3980 if self._configuration_arg('include_incomplete_formats'):
3981 skip_bad_formats = False
3982 self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
3983 'Use formats=incomplete extractor argument instead')
3984
3985 skip_manifests = set(self._configuration_arg('skip'))
3986 if (not self.get_param('youtube_include_hls_manifest', True)
3987 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3988 or needs_live_processing and skip_bad_formats):
3989 skip_manifests.add('hls')
3990
3991 if not self.get_param('youtube_include_dash_manifest', True):
3992 skip_manifests.add('dash')
3993 if self._configuration_arg('include_live_dash'):
3994 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3995 'Use formats=incomplete extractor argument instead')
3996 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3997 skip_manifests.add('dash')
3998
3999 def process_manifest_format(f, proto, client_name, itag):
4000 key = (proto, f.get('language'))
4001 if not all_formats and key in itags[itag]:
4002 return False
4003 itags[itag].add(key)
4004
4005 if itag and all_formats:
4006 f['format_id'] = f'{itag}-{proto}'
4007 elif any(p != proto for p, _ in itags[itag]):
4008 f['format_id'] = f'{itag}-{proto}'
4009 elif itag:
4010 f['format_id'] = itag
4011
4012 if original_language and f.get('language') == original_language:
4013 f['format_note'] = join_nonempty(f.get('format_note'), '(default)', delim=' ')
4014 f['language_preference'] = PREFERRED_LANG_VALUE
4015
4016 if f.get('source_preference') is None:
4017 f['source_preference'] = -1
4018
4019 if itag in ('616', '235'):
4020 f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
4021 f['source_preference'] += 100
4022
4023 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
4024 if f['quality'] == -1 and f.get('height'):
4025 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
4026 if self.get_param('verbose') or all_formats:
4027 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
4028 if f.get('fps') and f['fps'] <= 1:
4029 del f['fps']
4030
4031 if proto == 'hls' and f.get('has_drm'):
4032 f['has_drm'] = 'maybe'
4033 f['source_preference'] -= 5
4034 return True
4035
4036 subtitles = {}
4037 for sd in streaming_data:
4038 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
4039
4040 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
4041 if hls_manifest_url:
4042 fmts, subs = self._extract_m3u8_formats_and_subtitles(
4043 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
4044 subtitles = self._merge_subtitles(subs, subtitles)
4045 for f in fmts:
4046 if process_manifest_format(f, 'hls', client_name, self._search_regex(
4047 r'/itag/(\d+)', f['url'], 'itag', default=None)):
4048 yield f
4049
4050 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
4051 if dash_manifest_url:
4052 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
4053 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
4054 for f in formats:
4055 if process_manifest_format(f, 'dash', client_name, f['format_id']):
4056 f['filesize'] = int_or_none(self._search_regex(
4057 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
4058 if needs_live_processing:
4059 f['is_from_start'] = True
4060
4061 yield f
4062 yield subtitles
4063
4064 def _extract_storyboard(self, player_responses, duration):
4065 spec = get_first(
4066 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
4067 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
4068 if not base_url:
4069 return
4070 L = len(spec) - 1
4071 for i, args in enumerate(spec):
4072 args = args.split('#')
4073 counts = list(map(int_or_none, args[:5]))
4074 if len(args) != 8 or not all(counts):
4075 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
4076 continue
4077 width, height, frame_count, cols, rows = counts
4078 N, sigh = args[6:]
4079
4080 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
4081 fragment_count = frame_count / (cols * rows)
4082 fragment_duration = duration / fragment_count
4083 yield {
4084 'format_id': f'sb{i}',
4085 'format_note': 'storyboard',
4086 'ext': 'mhtml',
4087 'protocol': 'mhtml',
4088 'acodec': 'none',
4089 'vcodec': 'none',
4090 'url': url,
4091 'width': width,
4092 'height': height,
4093 'fps': frame_count / duration,
4094 'rows': rows,
4095 'columns': cols,
4096 'fragments': [{
4097 'url': url.replace('$M', str(j)),
4098 'duration': min(fragment_duration, duration - (j * fragment_duration)),
4099 } for j in range(math.ceil(fragment_count))],
4100 }
4101
4102 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
4103 webpage = None
4104 if 'webpage' not in self._configuration_arg('player_skip'):
4105 query = {'bpctr': '9999999999', 'has_verified': '1'}
4106 pp = self._configuration_arg('player_params', [None], casesense=True)[0]
4107 if pp:
4108 query['pp'] = pp
4109 webpage = self._download_webpage(
4110 webpage_url, video_id, fatal=False, query=query)
4111
4112 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
4113
4114 player_responses, player_url = self._extract_player_responses(
4115 self._get_requested_clients(url, smuggled_data),
4116 video_id, webpage, master_ytcfg, smuggled_data)
4117
4118 return webpage, master_ytcfg, player_responses, player_url
4119
4120 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
4121 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
4122 is_live = get_first(video_details, 'isLive')
4123 if is_live is None:
4124 is_live = get_first(live_broadcast_details, 'isLiveNow')
4125 live_content = get_first(video_details, 'isLiveContent')
4126 is_upcoming = get_first(video_details, 'isUpcoming')
4127 post_live = get_first(video_details, 'isPostLiveDvr')
4128 live_status = ('post_live' if post_live
4129 else 'is_live' if is_live
4130 else 'is_upcoming' if is_upcoming
4131 else 'was_live' if live_content
4132 else 'not_live' if False in (is_live, live_content)
4133 else None)
4134 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
4135 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
4136 if all(f.get('has_drm') for f in formats):
4137 # If there are no formats that definitely don't have DRM, all have DRM
4138 for f in formats:
4139 f['has_drm'] = True
4140
4141 return live_broadcast_details, live_status, streaming_data, formats, subtitles
4142
4143 def _real_extract(self, url):
4144 url, smuggled_data = unsmuggle_url(url, {})
4145 video_id = self._match_id(url)
4146
4147 base_url = self.http_scheme() + '//www.youtube.com/'
4148 webpage_url = base_url + 'watch?v=' + video_id
4149
4150 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4151
4152 playability_statuses = traverse_obj(
4153 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4154
4155 trailer_video_id = get_first(
4156 playability_statuses,
4157 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4158 expected_type=str)
4159 if trailer_video_id:
4160 return self.url_result(
4161 trailer_video_id, self.ie_key(), trailer_video_id)
4162
4163 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4164 if webpage else (lambda x: None))
4165
4166 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4167 microformats = traverse_obj(
4168 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4169 expected_type=dict)
4170
4171 translated_title = self._get_text(microformats, (..., 'title'))
4172 video_title = (self._preferred_lang and translated_title
4173 or get_first(video_details, 'title') # primary
4174 or translated_title
4175 or search_meta(['og:title', 'twitter:title', 'title']))
4176 translated_description = self._get_text(microformats, (..., 'description'))
4177 original_description = get_first(video_details, 'shortDescription')
4178 video_description = (
4179 self._preferred_lang and translated_description
4180 # If original description is blank, it will be an empty string.
4181 # Do not prefer translated description in this case.
4182 or original_description if original_description is not None else translated_description)
4183
4184 multifeed_metadata_list = get_first(
4185 player_responses,
4186 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4187 expected_type=str)
4188 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4189 if self.get_param('noplaylist'):
4190 self.to_screen(f'Downloading just video {video_id} because of --no-playlist')
4191 else:
4192 entries = []
4193 feed_ids = []
4194 for feed in multifeed_metadata_list.split(','):
4195 # Unquote should take place before split on comma (,) since textual
4196 # fields may contain comma as well (see
4197 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4198 feed_data = urllib.parse.parse_qs(
4199 urllib.parse.unquote_plus(feed))
4200
4201 def feed_entry(name):
4202 return try_get(
4203 feed_data, lambda x: x[name][0], str)
4204
4205 feed_id = feed_entry('id')
4206 if not feed_id:
4207 continue
4208 feed_title = feed_entry('title')
4209 title = video_title
4210 if feed_title:
4211 title += f' ({feed_title})'
4212 entries.append({
4213 '_type': 'url_transparent',
4214 'ie_key': 'Youtube',
4215 'url': smuggle_url(
4216 '{}watch?v={}'.format(base_url, feed_data['id'][0]),
4217 {'force_singlefeed': True}),
4218 'title': title,
4219 })
4220 feed_ids.append(feed_id)
4221 self.to_screen(
4222 'Downloading multifeed video ({}) - add --no-playlist to just download video {}'.format(
4223 ', '.join(feed_ids), video_id))
4224 return self.playlist_result(
4225 entries, video_id, video_title, video_description)
4226
4227 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4228 or int_or_none(get_first(microformats, 'lengthSeconds'))
4229 or parse_duration(search_meta('duration')) or None)
4230
4231 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4232 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4233 if live_status == 'post_live':
4234 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4235
4236 if not formats:
4237 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4238 self.report_drm(video_id)
4239 pemr = get_first(
4240 playability_statuses,
4241 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4242 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4243 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4244 if subreason:
4245 if subreason == 'The uploader has not made this video available in your country.':
4246 countries = get_first(microformats, 'availableCountries')
4247 if not countries:
4248 regions_allowed = search_meta('regionsAllowed')
4249 countries = regions_allowed.split(',') if regions_allowed else None
4250 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4251 reason += f'. {subreason}'
4252 if reason:
4253 self.raise_no_formats(reason, expected=True)
4254
4255 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4256 if not keywords and webpage:
4257 keywords = [
4258 unescapeHTML(m.group('content'))
4259 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4260 for keyword in keywords:
4261 if keyword.startswith('yt:stretch='):
4262 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4263 if mobj:
4264 # NB: float is intentional for forcing float division
4265 w, h = (float(v) for v in mobj.groups())
4266 if w > 0 and h > 0:
4267 ratio = w / h
4268 for f in formats:
4269 if f.get('vcodec') != 'none':
4270 f['stretched_ratio'] = ratio
4271 break
4272 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4273 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4274 if thumbnail_url:
4275 thumbnails.append({
4276 'url': thumbnail_url,
4277 })
4278 original_thumbnails = thumbnails.copy()
4279
4280 # The best resolution thumbnails sometimes does not appear in the webpage
4281 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4282 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4283 thumbnail_names = [
4284 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4285 # in resolution, these are not the custom thumbnail. So de-prioritize them
4286 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4287 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3',
4288 ]
4289 n_thumbnail_names = len(thumbnail_names)
4290 thumbnails.extend({
4291 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4292 video_id=video_id, name=name, ext=ext,
4293 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4294 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4295 for thumb in thumbnails:
4296 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4297 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4298 self._remove_duplicate_formats(thumbnails)
4299 self._downloader._sort_thumbnails(original_thumbnails)
4300
4301 category = get_first(microformats, 'category') or search_meta('genre')
4302 channel_id = self.ucid_or_none(str_or_none(
4303 get_first(video_details, 'channelId')
4304 or get_first(microformats, 'externalChannelId')
4305 or search_meta('channelId')))
4306 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4307
4308 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4309 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4310 if not duration and live_end_time and live_start_time:
4311 duration = live_end_time - live_start_time
4312
4313 needs_live_processing = self._needs_live_processing(live_status, duration)
4314
4315 def is_bad_format(fmt):
4316 if needs_live_processing and not fmt.get('is_from_start'):
4317 return True
4318 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4319 and fmt.get('protocol') == 'http_dash_segments'):
4320 return True
4321
4322 for fmt in filter(is_bad_format, formats):
4323 fmt['preference'] = (fmt.get('preference') or -1) - 10
4324 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
4325
4326 if needs_live_processing:
4327 self._prepare_live_from_start_formats(
4328 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4329
4330 formats.extend(self._extract_storyboard(player_responses, duration))
4331
4332 channel_handle = self.handle_from_url(owner_profile_url)
4333
4334 info = {
4335 'id': video_id,
4336 'title': video_title,
4337 'formats': formats,
4338 'thumbnails': thumbnails,
4339 # The best thumbnail that we are sure exists. Prevents unnecessary
4340 # URL checking if user don't care about getting the best possible thumbnail
4341 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4342 'description': video_description,
4343 'channel_id': channel_id,
4344 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4345 'duration': duration,
4346 'view_count': int_or_none(
4347 get_first((video_details, microformats), (..., 'viewCount'))
4348 or search_meta('interactionCount')),
4349 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4350 'age_limit': 18 if (
4351 get_first(microformats, 'isFamilySafe') is False
4352 or search_meta('isFamilyFriendly') == 'false'
4353 or search_meta('og:restrictions:age') == '18+') else 0,
4354 'webpage_url': webpage_url,
4355 'categories': [category] if category else None,
4356 'tags': keywords,
4357 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4358 'live_status': live_status,
4359 'release_timestamp': live_start_time,
4360 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4361 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto'),
4362 }
4363
4364 subtitles = {}
4365 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4366 if pctr:
4367 def get_lang_code(track):
4368 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4369 or track.get('languageCode'))
4370
4371 # Converted into dicts to remove duplicates
4372 captions = {
4373 get_lang_code(sub): sub
4374 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4375 translation_languages = {
4376 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4377 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4378
4379 def process_language(container, base_url, lang_code, sub_name, query):
4380 lang_subs = container.setdefault(lang_code, [])
4381 for fmt in self._SUBTITLE_FORMATS:
4382 query.update({
4383 'fmt': fmt,
4384 })
4385 lang_subs.append({
4386 'ext': fmt,
4387 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4388 'name': sub_name,
4389 })
4390
4391 # NB: Constructing the full subtitle dictionary is slow
4392 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4393 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4394 for lang_code, caption_track in captions.items():
4395 base_url = caption_track.get('baseUrl')
4396 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4397 if not base_url:
4398 continue
4399 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4400 if caption_track.get('kind') != 'asr':
4401 if not lang_code:
4402 continue
4403 process_language(
4404 subtitles, base_url, lang_code, lang_name, {})
4405 if not caption_track.get('isTranslatable'):
4406 continue
4407 for trans_code, trans_name in translation_languages.items():
4408 if not trans_code:
4409 continue
4410 orig_trans_code = trans_code
4411 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4412 if not get_translated_subs:
4413 continue
4414 trans_code += f'-{lang_code}'
4415 trans_name += format_field(lang_name, None, ' from %s')
4416 if lang_code == f'a-{orig_trans_code}':
4417 # Set audio language based on original subtitles
4418 for f in formats:
4419 if f.get('acodec') != 'none' and not f.get('language'):
4420 f['language'] = orig_trans_code
4421 # Add an "-orig" label to the original language so that it can be distinguished.
4422 # The subs are returned without "-orig" as well for compatibility
4423 process_language(
4424 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4425 # Setting tlang=lang returns damaged subtitles.
4426 process_language(automatic_captions, base_url, trans_code, trans_name,
4427 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4428
4429 info['automatic_captions'] = automatic_captions
4430 info['subtitles'] = subtitles
4431
4432 parsed_url = urllib.parse.urlparse(url)
4433 for component in [parsed_url.fragment, parsed_url.query]:
4434 query = urllib.parse.parse_qs(component)
4435 for k, v in query.items():
4436 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4437 d_k += '_time'
4438 if d_k not in info and k in s_ks:
4439 info[d_k] = parse_duration(v[0])
4440
4441 # Youtube Music Auto-generated description
4442 if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
4443 # XXX: Causes catastrophic backtracking if description has "·"
4444 # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
4445 # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
4446 # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
4447 mobj = re.search(
4448 r'''(?xs)
4449 (?=(?P<track>[^\n·]+))(?P=track)·
4450 (?=(?P<artist>[^\n]+))(?P=artist)\n+
4451 (?=(?P<album>[^\n]+))(?P=album)\n
4452 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4453 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4454 (.+?\nArtist\s*:\s*
4455 (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
4456 )?.+\nAuto-generated\ by\ YouTube\.\s*$
4457 ''', video_description)
4458 if mobj:
4459 release_year = mobj.group('release_year')
4460 release_date = mobj.group('release_date')
4461 if release_date:
4462 release_date = release_date.replace('-', '')
4463 if not release_year:
4464 release_year = release_date[:4]
4465 info.update({
4466 'album': mobj.group('album'.strip()),
4467 'artists': ([a] if (a := mobj.group('clean_artist'))
4468 else [a.strip() for a in mobj.group('artist').split('·')]),
4469 'track': mobj.group('track').strip(),
4470 'release_date': release_date,
4471 'release_year': int_or_none(release_year),
4472 })
4473
4474 initial_data = None
4475 if webpage:
4476 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4477 if not traverse_obj(initial_data, 'contents'):
4478 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4479 initial_data = None
4480 if not initial_data:
4481 query = {'videoId': video_id}
4482 query.update(self._get_checkok_params())
4483 initial_data = self._extract_response(
4484 item_id=video_id, ep='next', fatal=False,
4485 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4486 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4487 note='Downloading initial data API JSON')
4488
4489 info['comment_count'] = traverse_obj(initial_data, (
4490 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4491 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount',
4492 ), (
4493 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4494 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo',
4495 ), expected_type=self._get_count, get_all=False)
4496
4497 try: # This will error if there is no livechat
4498 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4499 except (KeyError, IndexError, TypeError):
4500 pass
4501 else:
4502 info.setdefault('subtitles', {})['live_chat'] = [{
4503 # url is needed to set cookies
4504 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4505 'video_id': video_id,
4506 'ext': 'json',
4507 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4508 else 'youtube_live_chat_replay'),
4509 }]
4510
4511 if initial_data:
4512 info['chapters'] = (
4513 self._extract_chapters_from_json(initial_data, duration)
4514 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4515 or self._extract_chapters_from_description(video_description, duration)
4516 or None)
4517
4518 info['heatmap'] = self._extract_heatmap(initial_data)
4519
4520 contents = traverse_obj(
4521 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4522 expected_type=list, default=[])
4523
4524 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4525 if vpir:
4526 stl = vpir.get('superTitleLink')
4527 if stl:
4528 stl = self._get_text(stl)
4529 if try_get(
4530 vpir,
4531 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4532 info['location'] = stl
4533 else:
4534 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4535 if mobj:
4536 info.update({
4537 'series': mobj.group(1),
4538 'season_number': int(mobj.group(2)),
4539 'episode_number': int(mobj.group(3)),
4540 })
4541 for tlb in (try_get(
4542 vpir,
4543 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4544 list) or []):
4545 tbrs = variadic(
4546 traverse_obj(
4547 tlb, ('toggleButtonRenderer', ...),
4548 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4549 for tbr in tbrs:
4550 for getter, regex in [(
4551 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4552 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4553 lambda x: x['accessibility'],
4554 lambda x: x['accessibilityData']['accessibilityData'],
4555 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4556 label = (try_get(tbr, getter, dict) or {}).get('label')
4557 if label:
4558 mobj = re.match(regex, label)
4559 if mobj:
4560 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4561 break
4562
4563 info['like_count'] = traverse_obj(vpir, (
4564 'videoActions', 'menuRenderer', 'topLevelButtons', ...,
4565 'segmentedLikeDislikeButtonViewModel', 'likeButtonViewModel', 'likeButtonViewModel',
4566 'toggleButtonViewModel', 'toggleButtonViewModel', 'defaultButtonViewModel',
4567 'buttonViewModel', 'accessibilityText', {parse_count}), get_all=False)
4568
4569 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4570 if vcr:
4571 vc = self._get_count(vcr, 'viewCount')
4572 # Upcoming premieres with waiting count are treated as live here
4573 if vcr.get('isLive'):
4574 info['concurrent_view_count'] = vc
4575 elif info.get('view_count') is None:
4576 info['view_count'] = vc
4577
4578 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4579 if vsir:
4580 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4581 info.update({
4582 'channel': self._get_text(vor, 'title'),
4583 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4584
4585 if not channel_handle:
4586 channel_handle = self.handle_from_url(
4587 traverse_obj(vor, (
4588 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4589 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4590 {str}), get_all=False))
4591
4592 rows = try_get(
4593 vsir,
4594 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4595 list) or []
4596 multiple_songs = False
4597 for row in rows:
4598 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4599 multiple_songs = True
4600 break
4601 for row in rows:
4602 mrr = row.get('metadataRowRenderer') or {}
4603 mrr_title = mrr.get('title')
4604 if not mrr_title:
4605 continue
4606 mrr_title = self._get_text(mrr, 'title')
4607 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4608 if mrr_title == 'License':
4609 info['license'] = mrr_contents_text
4610 elif not multiple_songs:
4611 if mrr_title == 'Album':
4612 info['album'] = mrr_contents_text
4613 elif mrr_title == 'Artist':
4614 info['artists'] = [mrr_contents_text] if mrr_contents_text else None
4615 elif mrr_title == 'Song':
4616 info['track'] = mrr_contents_text
4617 owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
4618 if self._has_badge(owner_badges, BadgeType.VERIFIED):
4619 info['channel_is_verified'] = True
4620
4621 info.update({
4622 'uploader': info.get('channel'),
4623 'uploader_id': channel_handle,
4624 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4625 })
4626
4627 # We only want timestamp IF it has time precision AND a timezone
4628 # Currently the uploadDate in microformats appears to be in US/Pacific timezone.
4629 timestamp = (
4630 parse_iso8601(get_first(microformats, 'uploadDate'), timezone=NO_DEFAULT)
4631 or parse_iso8601(search_meta('uploadDate'), timezone=NO_DEFAULT)
4632 )
4633 upload_date = (
4634 dt.datetime.fromtimestamp(timestamp, dt.timezone.utc).strftime('%Y%m%d') if timestamp else
4635 (
4636 unified_strdate(get_first(microformats, 'uploadDate'))
4637 or unified_strdate(search_meta('uploadDate'))
4638 ))
4639
4640 # In the case we cannot get the timestamp:
4641 # The upload date for scheduled, live and past live streams / premieres in microformats
4642 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4643 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4644 if not upload_date or (not timestamp and live_status in ('not_live', None)):
4645 # this should be in UTC, as configured in the cookie/client context
4646 upload_date = strftime_or_none(
4647 self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
4648
4649 info['upload_date'] = upload_date
4650 info['timestamp'] = timestamp
4651
4652 if upload_date and live_status not in ('is_live', 'post_live', 'is_upcoming'):
4653 # Newly uploaded videos' HLS formats are potentially problematic and need to be checked
4654 upload_datetime = datetime_from_str(upload_date).replace(tzinfo=dt.timezone.utc)
4655 if upload_datetime >= datetime_from_str('today-2days'):
4656 for fmt in info['formats']:
4657 if fmt.get('protocol') == 'm3u8_native':
4658 fmt['__needs_testing'] = True
4659
4660 for s_k, d_k in [('artists', 'creators'), ('track', 'alt_title')]:
4661 v = info.get(s_k)
4662 if v:
4663 info[d_k] = v
4664
4665 badges = self._extract_badges(traverse_obj(vpir, 'badges'))
4666
4667 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4668 or get_first(video_details, 'isPrivate', expected_type=bool))
4669
4670 info['availability'] = (
4671 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4672 else self._availability(
4673 is_private=is_private,
4674 needs_premium=(
4675 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4676 or False if initial_data and is_private is not None else None),
4677 needs_subscription=(
4678 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4679 or False if initial_data and is_private is not None else None),
4680 needs_auth=info['age_limit'] >= 18,
4681 is_unlisted=None if is_private is None else (
4682 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4683 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4684
4685 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4686
4687 self.mark_watched(video_id, player_responses)
4688
4689 return info
4690
4691
4692 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4693 @staticmethod
4694 def passthrough_smuggled_data(func):
4695 def _smuggle(info, smuggled_data):
4696 if info.get('_type') not in ('url', 'url_transparent'):
4697 return info
4698 if smuggled_data.get('is_music_url'):
4699 parsed_url = urllib.parse.urlparse(info['url'])
4700 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4701 smuggled_data.pop('is_music_url')
4702 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4703 if smuggled_data:
4704 info['url'] = smuggle_url(info['url'], smuggled_data)
4705 return info
4706
4707 @functools.wraps(func)
4708 def wrapper(self, url):
4709 url, smuggled_data = unsmuggle_url(url, {})
4710 if self.is_music_url(url):
4711 smuggled_data['is_music_url'] = True
4712 info_dict = func(self, url, smuggled_data)
4713 if smuggled_data:
4714 _smuggle(info_dict, smuggled_data)
4715 if info_dict.get('entries'):
4716 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4717 return info_dict
4718 return wrapper
4719
4720 @staticmethod
4721 def _extract_basic_item_renderer(item):
4722 # Modified from _extract_grid_item_renderer
4723 known_basic_renderers = (
4724 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer',
4725 )
4726 for key, renderer in item.items():
4727 if not isinstance(renderer, dict):
4728 continue
4729 elif key in known_basic_renderers:
4730 return renderer
4731 elif key.startswith('grid') and key.endswith('Renderer'):
4732 return renderer
4733
4734 def _extract_channel_renderer(self, renderer):
4735 channel_id = self.ucid_or_none(renderer['channelId'])
4736 title = self._get_text(renderer, 'title')
4737 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4738 channel_handle = self.handle_from_url(
4739 traverse_obj(renderer, (
4740 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4741 ('browseEndpoint', 'canonicalBaseUrl')),
4742 {str}), get_all=False))
4743 if not channel_handle:
4744 # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
4745 channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
4746 return {
4747 '_type': 'url',
4748 'url': channel_url,
4749 'id': channel_id,
4750 'ie_key': YoutubeTabIE.ie_key(),
4751 'channel': title,
4752 'uploader': title,
4753 'channel_id': channel_id,
4754 'channel_url': channel_url,
4755 'title': title,
4756 'uploader_id': channel_handle,
4757 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4758 # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
4759 # However, in feed/channels this is set correctly to the subscriber count
4760 'channel_follower_count': traverse_obj(
4761 renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
4762 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4763 'playlist_count': (
4764 # videoCountText may be the subscriber count
4765 self._get_count(renderer, 'videoCountText')
4766 if self._get_count(renderer, 'subscriberCountText') is not None else None),
4767 'description': self._get_text(renderer, 'descriptionSnippet'),
4768 'channel_is_verified': True if self._has_badge(
4769 self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
4770 }
4771
4772 def _grid_entries(self, grid_renderer):
4773 for item in grid_renderer['items']:
4774 if not isinstance(item, dict):
4775 continue
4776 renderer = self._extract_basic_item_renderer(item)
4777 if not isinstance(renderer, dict):
4778 continue
4779 title = self._get_text(renderer, 'title')
4780
4781 # playlist
4782 playlist_id = renderer.get('playlistId')
4783 if playlist_id:
4784 yield self.url_result(
4785 f'https://www.youtube.com/playlist?list={playlist_id}',
4786 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4787 video_title=title)
4788 continue
4789 # video
4790 video_id = renderer.get('videoId')
4791 if video_id:
4792 yield self._extract_video(renderer)
4793 continue
4794 # channel
4795 channel_id = renderer.get('channelId')
4796 if channel_id:
4797 yield self._extract_channel_renderer(renderer)
4798 continue
4799 # generic endpoint URL support
4800 ep_url = urljoin('https://www.youtube.com/', try_get(
4801 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4802 str))
4803 if ep_url:
4804 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4805 if ie.suitable(ep_url):
4806 yield self.url_result(
4807 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4808 break
4809
4810 def _music_reponsive_list_entry(self, renderer):
4811 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4812 if video_id:
4813 title = traverse_obj(renderer, (
4814 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4815 'text', 'runs', 0, 'text'))
4816 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4817 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4818 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4819 if playlist_id:
4820 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4821 if video_id:
4822 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4823 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4824 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4825 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4826 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4827 if browse_id:
4828 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4829 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4830
4831 def _shelf_entries_from_content(self, shelf_renderer):
4832 content = shelf_renderer.get('content')
4833 if not isinstance(content, dict):
4834 return
4835 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4836 if renderer:
4837 # TODO: add support for nested playlists so each shelf is processed
4838 # as separate playlist
4839 # TODO: this includes only first N items
4840 yield from self._grid_entries(renderer)
4841 renderer = content.get('horizontalListRenderer')
4842 if renderer:
4843 # TODO: handle case
4844 pass
4845
4846 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4847 ep = try_get(
4848 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4849 str)
4850 shelf_url = urljoin('https://www.youtube.com', ep)
4851 if shelf_url:
4852 # Skipping links to another channels, note that checking for
4853 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4854 # will not work
4855 if skip_channels and '/channels?' in shelf_url:
4856 return
4857 title = self._get_text(shelf_renderer, 'title')
4858 yield self.url_result(shelf_url, video_title=title)
4859 # Shelf may not contain shelf URL, fallback to extraction from content
4860 yield from self._shelf_entries_from_content(shelf_renderer)
4861
4862 def _playlist_entries(self, video_list_renderer):
4863 for content in video_list_renderer['contents']:
4864 if not isinstance(content, dict):
4865 continue
4866 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4867 if not isinstance(renderer, dict):
4868 continue
4869 video_id = renderer.get('videoId')
4870 if not video_id:
4871 continue
4872 yield self._extract_video(renderer)
4873
4874 def _rich_entries(self, rich_grid_renderer):
4875 renderer = traverse_obj(
4876 rich_grid_renderer,
4877 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4878 video_id = renderer.get('videoId')
4879 if video_id:
4880 yield self._extract_video(renderer)
4881 return
4882 playlist_id = renderer.get('playlistId')
4883 if playlist_id:
4884 yield self.url_result(
4885 f'https://www.youtube.com/playlist?list={playlist_id}',
4886 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4887 video_title=self._get_text(renderer, 'title'))
4888 return
4889
4890 def _video_entry(self, video_renderer):
4891 video_id = video_renderer.get('videoId')
4892 if video_id:
4893 return self._extract_video(video_renderer)
4894
4895 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4896 url = urljoin('https://youtube.com', traverse_obj(
4897 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4898 if url:
4899 return self.url_result(
4900 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4901
4902 def _post_thread_entries(self, post_thread_renderer):
4903 post_renderer = try_get(
4904 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4905 if not post_renderer:
4906 return
4907 # video attachment
4908 video_renderer = try_get(
4909 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4910 video_id = video_renderer.get('videoId')
4911 if video_id:
4912 entry = self._extract_video(video_renderer)
4913 if entry:
4914 yield entry
4915 # playlist attachment
4916 playlist_id = try_get(
4917 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4918 if playlist_id:
4919 yield self.url_result(
4920 f'https://www.youtube.com/playlist?list={playlist_id}',
4921 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4922 # inline video links
4923 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4924 for run in runs:
4925 if not isinstance(run, dict):
4926 continue
4927 ep_url = try_get(
4928 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4929 if not ep_url:
4930 continue
4931 if not YoutubeIE.suitable(ep_url):
4932 continue
4933 ep_video_id = YoutubeIE._match_id(ep_url)
4934 if video_id == ep_video_id:
4935 continue
4936 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4937
4938 def _post_thread_continuation_entries(self, post_thread_continuation):
4939 contents = post_thread_continuation.get('contents')
4940 if not isinstance(contents, list):
4941 return
4942 for content in contents:
4943 renderer = content.get('backstagePostThreadRenderer')
4944 if isinstance(renderer, dict):
4945 yield from self._post_thread_entries(renderer)
4946 continue
4947 renderer = content.get('videoRenderer')
4948 if isinstance(renderer, dict):
4949 yield self._video_entry(renderer)
4950
4951 r''' # unused
4952 def _rich_grid_entries(self, contents):
4953 for content in contents:
4954 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4955 if video_renderer:
4956 entry = self._video_entry(video_renderer)
4957 if entry:
4958 yield entry
4959 '''
4960
4961 def _report_history_entries(self, renderer):
4962 for url in traverse_obj(renderer, (
4963 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4964 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4965 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4966 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4967
4968 def _extract_entries(self, parent_renderer, continuation_list):
4969 # continuation_list is modified in-place with continuation_list = [continuation_token]
4970 continuation_list[:] = [None]
4971 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4972 for content in contents:
4973 if not isinstance(content, dict):
4974 continue
4975 is_renderer = traverse_obj(
4976 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4977 expected_type=dict)
4978 if not is_renderer:
4979 if content.get('richItemRenderer'):
4980 for entry in self._rich_entries(content['richItemRenderer']):
4981 yield entry
4982 continuation_list[0] = self._extract_continuation(parent_renderer)
4983 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4984 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4985 yield from self._report_history_entries(table)
4986 continuation_list[0] = self._extract_continuation(table)
4987 continue
4988
4989 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4990 for isr_content in isr_contents:
4991 if not isinstance(isr_content, dict):
4992 continue
4993
4994 known_renderers = {
4995 'playlistVideoListRenderer': self._playlist_entries,
4996 'gridRenderer': self._grid_entries,
4997 'reelShelfRenderer': self._grid_entries,
4998 'shelfRenderer': self._shelf_entries,
4999 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
5000 'backstagePostThreadRenderer': self._post_thread_entries,
5001 'videoRenderer': lambda x: [self._video_entry(x)],
5002 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
5003 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
5004 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
5005 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
5006 }
5007 for key, renderer in isr_content.items():
5008 if key not in known_renderers:
5009 continue
5010 for entry in known_renderers[key](renderer):
5011 if entry:
5012 yield entry
5013 continuation_list[0] = self._extract_continuation(renderer)
5014 break
5015
5016 if not continuation_list[0]:
5017 continuation_list[0] = self._extract_continuation(is_renderer)
5018
5019 if not continuation_list[0]:
5020 continuation_list[0] = self._extract_continuation(parent_renderer)
5021
5022 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
5023 continuation_list = [None]
5024 extract_entries = lambda x: self._extract_entries(x, continuation_list)
5025 tab_content = try_get(tab, lambda x: x['content'], dict)
5026 if not tab_content:
5027 return
5028 parent_renderer = (
5029 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
5030 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
5031 yield from extract_entries(parent_renderer)
5032 continuation = continuation_list[0]
5033 seen_continuations = set()
5034 for page_num in itertools.count(1):
5035 if not continuation:
5036 break
5037 continuation_token = continuation.get('continuation')
5038 if continuation_token is not None and continuation_token in seen_continuations:
5039 self.write_debug('Detected YouTube feed looping - assuming end of feed.')
5040 break
5041 seen_continuations.add(continuation_token)
5042 headers = self.generate_api_headers(
5043 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
5044 response = self._extract_response(
5045 item_id=f'{item_id} page {page_num}',
5046 query=continuation, headers=headers, ytcfg=ytcfg,
5047 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
5048
5049 if not response:
5050 break
5051 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
5052 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
5053 visitor_data = self._extract_visitor_data(response) or visitor_data
5054
5055 known_renderers = {
5056 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
5057 'gridPlaylistRenderer': (self._grid_entries, 'items'),
5058 'gridVideoRenderer': (self._grid_entries, 'items'),
5059 'gridChannelRenderer': (self._grid_entries, 'items'),
5060 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
5061 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
5062 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
5063 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
5064 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
5065 'playlistVideoListContinuation': (self._playlist_entries, None),
5066 'gridContinuation': (self._grid_entries, None),
5067 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
5068 'sectionListContinuation': (extract_entries, None), # for feeds
5069 }
5070
5071 continuation_items = traverse_obj(response, (
5072 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
5073 'appendContinuationItemsAction', 'continuationItems',
5074 ), 'continuationContents', get_all=False)
5075 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
5076
5077 video_items_renderer = None
5078 for key in continuation_item:
5079 if key not in known_renderers:
5080 continue
5081 func, parent_key = known_renderers[key]
5082 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
5083 continuation_list = [None]
5084 yield from func(video_items_renderer)
5085 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
5086
5087 if not video_items_renderer:
5088 break
5089
5090 @staticmethod
5091 def _extract_selected_tab(tabs, fatal=True):
5092 for tab_renderer in tabs:
5093 if tab_renderer.get('selected'):
5094 return tab_renderer
5095 if fatal:
5096 raise ExtractorError('Unable to find selected tab')
5097
5098 @staticmethod
5099 def _extract_tab_renderers(response):
5100 return traverse_obj(
5101 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
5102
5103 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
5104 metadata = self._extract_metadata_from_tabs(item_id, data)
5105
5106 selected_tab = self._extract_selected_tab(tabs)
5107 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
5108 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
5109
5110 return self.playlist_result(
5111 self._entries(
5112 selected_tab, metadata['id'], ytcfg,
5113 self._extract_account_syncid(ytcfg, data),
5114 self._extract_visitor_data(data, ytcfg)),
5115 **metadata)
5116
5117 def _extract_metadata_from_tabs(self, item_id, data):
5118 info = {'id': item_id}
5119
5120 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
5121 if metadata_renderer:
5122 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
5123 ('channelUrl', {self.ucid_from_url}))
5124 info.update({
5125 'channel': metadata_renderer.get('title'),
5126 'channel_id': channel_id,
5127 })
5128 if info['channel_id']:
5129 info['id'] = info['channel_id']
5130 else:
5131 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
5132
5133 # pageHeaderViewModel slow rollout began April 2024
5134 page_header_view_model = traverse_obj(data, (
5135 'header', 'pageHeaderRenderer', 'content', 'pageHeaderViewModel', {dict}))
5136
5137 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
5138 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
5139 def _get_uncropped(url):
5140 return url_or_none((url or '').split('=')[0] + '=s0')
5141
5142 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
5143 if avatar_thumbnails:
5144 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
5145 if uncropped_avatar:
5146 avatar_thumbnails.append({
5147 'url': uncropped_avatar,
5148 'id': 'avatar_uncropped',
5149 'preference': 1,
5150 })
5151
5152 channel_banners = (
5153 self._extract_thumbnails(data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
5154 or self._extract_thumbnails(
5155 page_header_view_model, ('banner', 'imageBannerViewModel', 'image'), final_key='sources'))
5156 for banner in channel_banners:
5157 banner['preference'] = -10
5158
5159 if channel_banners:
5160 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
5161 if uncropped_banner:
5162 channel_banners.append({
5163 'url': uncropped_banner,
5164 'id': 'banner_uncropped',
5165 'preference': -5,
5166 })
5167
5168 # Deprecated - remove primary_sidebar_renderer when layout discontinued
5169 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
5170 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
5171
5172 primary_thumbnails = self._extract_thumbnails(
5173 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
5174 playlist_thumbnails = self._extract_thumbnails(
5175 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
5176
5177 info.update({
5178 'title': (traverse_obj(metadata_renderer, 'title')
5179 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
5180 or info['id']),
5181 'availability': self._extract_availability(data),
5182 'channel_follower_count': (
5183 self._get_count(data, ('header', ..., 'subscriberCountText'))
5184 or traverse_obj(page_header_view_model, (
5185 'metadata', 'contentMetadataViewModel', 'metadataRows', ..., 'metadataParts',
5186 lambda _, v: 'subscribers' in v['text']['content'], 'text', 'content', {parse_count}, any))),
5187 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
5188 'tags': (traverse_obj(data, ('microformat', 'microformatDataRenderer', 'tags', ..., {str}))
5189 or traverse_obj(metadata_renderer, ('keywords', {lambda x: x and shlex.split(x)}, ...))),
5190 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
5191 })
5192
5193 channel_handle = (
5194 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
5195 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
5196
5197 if channel_handle:
5198 info.update({
5199 'uploader_id': channel_handle,
5200 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
5201 })
5202
5203 channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
5204 if self._has_badge(channel_badges, BadgeType.VERIFIED):
5205 info['channel_is_verified'] = True
5206 # Playlist stats is a text runs array containing [video count, view count, last updated].
5207 # last updated or (view count and last updated) may be missing.
5208 playlist_stats = get_first(
5209 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
5210
5211 last_updated_unix = self._parse_time_text(
5212 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5213 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5214 info['modified_date'] = strftime_or_none(last_updated_unix)
5215
5216 info['view_count'] = self._get_count(playlist_stats, 1)
5217 if info['view_count'] is None: # 0 is allowed
5218 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5219 if info['view_count'] is None:
5220 info['view_count'] = self._get_count(data, (
5221 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5222 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5223
5224 info['playlist_count'] = self._get_count(playlist_stats, 0)
5225 if info['playlist_count'] is None: # 0 is allowed
5226 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5227
5228 if not info.get('channel_id'):
5229 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5230 if not owner: # Deprecated
5231 owner = traverse_obj(
5232 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5233 ('videoOwner', 'videoOwnerRenderer', 'title'))
5234 owner_text = self._get_text(owner)
5235 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5236 info.update({
5237 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5238 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5239 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl'))),
5240 })
5241
5242 info.update({
5243 'uploader': info['channel'],
5244 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5245 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5246 })
5247
5248 return info
5249
5250 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5251 first_id = last_id = response = None
5252 for page_num in itertools.count(1):
5253 videos = list(self._playlist_entries(playlist))
5254 if not videos:
5255 return
5256 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5257 if start >= len(videos):
5258 return
5259 yield from videos[start:]
5260 first_id = first_id or videos[0]['id']
5261 last_id = videos[-1]['id']
5262 watch_endpoint = try_get(
5263 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5264 headers = self.generate_api_headers(
5265 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5266 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5267 query = {
5268 'playlistId': playlist_id,
5269 'videoId': watch_endpoint.get('videoId') or last_id,
5270 'index': watch_endpoint.get('index') or len(videos),
5271 'params': watch_endpoint.get('params') or 'OAE%3D',
5272 }
5273 response = self._extract_response(
5274 item_id=f'{playlist_id} page {page_num}',
5275 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5276 check_get_keys='contents',
5277 )
5278 playlist = try_get(
5279 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5280
5281 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5282 title = playlist.get('title') or try_get(
5283 data, lambda x: x['titleText']['simpleText'], str)
5284 playlist_id = playlist.get('playlistId') or item_id
5285
5286 # Delegating everything except mix playlists to regular tab-based playlist URL
5287 playlist_url = urljoin(url, try_get(
5288 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5289 str))
5290
5291 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5292 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5293 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5294
5295 if playlist_url and playlist_url != url and not is_known_unviewable:
5296 return self.url_result(
5297 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5298 video_title=title)
5299
5300 return self.playlist_result(
5301 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5302 playlist_id=playlist_id, playlist_title=title)
5303
5304 def _extract_availability(self, data):
5305 """
5306 Gets the availability of a given playlist/tab.
5307 Note: Unless YouTube tells us explicitly, we do not assume it is public
5308 @param data: response
5309 """
5310 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5311 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5312 player_header_privacy = playlist_header_renderer.get('privacy')
5313
5314 badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
5315
5316 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5317 privacy_setting_icon = get_first(
5318 (playlist_header_renderer, sidebar_renderer),
5319 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5320 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5321 expected_type=str)
5322
5323 microformats_is_unlisted = traverse_obj(
5324 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5325
5326 return (
5327 'public' if (
5328 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5329 or player_header_privacy == 'PUBLIC'
5330 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5331 else self._availability(
5332 is_private=(
5333 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5334 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5335 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5336 is_unlisted=(
5337 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5338 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5339 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5340 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5341 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5342 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5343 needs_auth=False))
5344
5345 @staticmethod
5346 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5347 sidebar_renderer = try_get(
5348 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5349 for item in sidebar_renderer:
5350 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5351 if renderer:
5352 return renderer
5353
5354 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5355 """
5356 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5357 """
5358 is_playlist = bool(traverse_obj(
5359 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5360 if not is_playlist:
5361 return
5362 headers = self.generate_api_headers(
5363 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5364 visitor_data=self._extract_visitor_data(data, ytcfg))
5365 query = {
5366 'params': 'wgYCCAA=',
5367 'browseId': f'VL{item_id}',
5368 }
5369 return self._extract_response(
5370 item_id=item_id, headers=headers, query=query,
5371 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5372 note='Redownloading playlist API JSON with unavailable videos')
5373
5374 @functools.cached_property
5375 def skip_webpage(self):
5376 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5377
5378 def _extract_webpage(self, url, item_id, fatal=True):
5379 webpage, data = None, None
5380 for retry in self.RetryManager(fatal=fatal):
5381 try:
5382 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5383 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5384 except ExtractorError as e:
5385 if isinstance(e.cause, network_exceptions):
5386 if not isinstance(e.cause, HTTPError) or e.cause.status not in (403, 429):
5387 retry.error = e
5388 continue
5389 self._error_or_warning(e, fatal=fatal)
5390 break
5391
5392 try:
5393 self._extract_and_report_alerts(data)
5394 except ExtractorError as e:
5395 self._error_or_warning(e, fatal=fatal)
5396 break
5397
5398 # Sometimes youtube returns a webpage with incomplete ytInitialData
5399 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5400 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5401 retry.error = ExtractorError('Incomplete yt initial data received')
5402 data = None
5403 continue
5404
5405 return webpage, data
5406
5407 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5408 """Use if failed to extract ytcfg (and data) from initial webpage"""
5409 if not ytcfg and self.is_authenticated:
5410 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5411 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5412 raise ExtractorError(
5413 f'{msg}. If you are not downloading private content, or '
5414 'your cookies are only for the first account and channel,'
5415 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5416 expected=True)
5417 self.report_warning(msg, only_once=True)
5418
5419 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5420 data = None
5421 if not self.skip_webpage:
5422 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5423 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5424 # Reject webpage data if redirected to home page without explicitly requesting
5425 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5426 if (url != 'https://www.youtube.com/feed/recommended'
5427 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5428 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5429 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5430 if fatal:
5431 raise ExtractorError(msg, expected=True)
5432 self.report_warning(msg, only_once=True)
5433 if not data:
5434 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5435 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5436 return data, ytcfg
5437
5438 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5439 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5440 resolve_response = self._extract_response(
5441 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5442 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5443 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5444 for ep_key, ep in endpoints.items():
5445 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5446 if params:
5447 return self._extract_response(
5448 item_id=item_id, query=params, ep=ep, headers=headers,
5449 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5450 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5451 err_note = 'Failed to resolve url (does the playlist exist?)'
5452 if fatal:
5453 raise ExtractorError(err_note, expected=True)
5454 self.report_warning(err_note, item_id)
5455
5456 _SEARCH_PARAMS = None
5457
5458 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5459 data = {'query': query}
5460 if params is NO_DEFAULT:
5461 params = self._SEARCH_PARAMS
5462 if params:
5463 data['params'] = params
5464
5465 content_keys = (
5466 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5467 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5468 # ytmusic search
5469 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5470 ('continuationContents', ),
5471 )
5472 display_id = f'query "{query}"'
5473 check_get_keys = tuple({keys[0] for keys in content_keys})
5474 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5475 self._report_playlist_authcheck(ytcfg, fatal=False)
5476
5477 continuation_list = [None]
5478 search = None
5479 for page_num in itertools.count(1):
5480 data.update(continuation_list[0] or {})
5481 headers = self.generate_api_headers(
5482 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5483 search = self._extract_response(
5484 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5485 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5486 slr_contents = traverse_obj(search, *content_keys)
5487 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5488 if not continuation_list[0]:
5489 break
5490
5491
5492 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5493 IE_DESC = 'YouTube Tabs'
5494 _VALID_URL = r'''(?x:
5495 https?://
5496 (?!consent\.)(?:\w+\.)?
5497 (?:
5498 youtube(?:kids)?\.com|
5499 {invidious}
5500 )/
5501 (?:
5502 (?P<channel_type>channel|c|user|browse)/|
5503 (?P<not_channel>
5504 feed/|hashtag/|
5505 (?:playlist|watch)\?.*?\blist=
5506 )|
5507 (?!(?:{reserved_names})\b) # Direct URLs
5508 )
5509 (?P<id>[^/?\#&]+)
5510 )'''.format(
5511 reserved_names=YoutubeBaseInfoExtractor._RESERVED_NAMES,
5512 invidious='|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5513 )
5514 IE_NAME = 'youtube:tab'
5515
5516 _TESTS = [{
5517 'note': 'playlists, multipage',
5518 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5519 'playlist_mincount': 94,
5520 'info_dict': {
5521 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5522 'title': 'Igor Kleiner Ph.D. - Playlists',
5523 'description': 'md5:15d7dd9e333cb987907fcb0d604b233a',
5524 'uploader': 'Igor Kleiner Ph.D.',
5525 'uploader_id': '@IgorDataScience',
5526 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5527 'channel': 'Igor Kleiner Ph.D.',
5528 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5529 'tags': ['критическое мышление', 'наука просто', 'математика', 'анализ данных'],
5530 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5531 'channel_follower_count': int,
5532 },
5533 }, {
5534 'note': 'playlists, multipage, different order',
5535 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5536 'playlist_mincount': 94,
5537 'info_dict': {
5538 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5539 'title': 'Igor Kleiner Ph.D. - Playlists',
5540 'description': 'md5:15d7dd9e333cb987907fcb0d604b233a',
5541 'uploader': 'Igor Kleiner Ph.D.',
5542 'uploader_id': '@IgorDataScience',
5543 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5544 'tags': ['критическое мышление', 'наука просто', 'математика', 'анализ данных'],
5545 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5546 'channel': 'Igor Kleiner Ph.D.',
5547 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5548 'channel_follower_count': int,
5549 },
5550 }, {
5551 'note': 'playlists, series',
5552 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5553 'playlist_mincount': 5,
5554 'info_dict': {
5555 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5556 'title': '3Blue1Brown - Playlists',
5557 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
5558 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5559 'channel': '3Blue1Brown',
5560 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5561 'uploader_id': '@3blue1brown',
5562 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5563 'uploader': '3Blue1Brown',
5564 'tags': ['Mathematics'],
5565 'channel_follower_count': int,
5566 'channel_is_verified': True,
5567 },
5568 }, {
5569 'note': 'playlists, singlepage',
5570 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5571 'playlist_mincount': 4,
5572 'info_dict': {
5573 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5574 'title': 'ThirstForScience - Playlists',
5575 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5576 'uploader': 'ThirstForScience',
5577 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5578 'uploader_id': '@ThirstForScience',
5579 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5580 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5581 'tags': 'count:12',
5582 'channel': 'ThirstForScience',
5583 'channel_follower_count': int,
5584 },
5585 }, {
5586 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5587 'only_matching': True,
5588 }, {
5589 'note': 'basic, single video playlist',
5590 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5591 'info_dict': {
5592 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5593 'title': 'youtube-dl public playlist',
5594 'description': '',
5595 'tags': [],
5596 'view_count': int,
5597 'modified_date': '20201130',
5598 'channel': 'Sergey M.',
5599 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5600 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5601 'availability': 'public',
5602 'uploader': 'Sergey M.',
5603 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5604 'uploader_id': '@sergeym.6173',
5605 },
5606 'playlist_count': 1,
5607 }, {
5608 'note': 'empty playlist',
5609 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5610 'info_dict': {
5611 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5612 'title': 'youtube-dl empty playlist',
5613 'tags': [],
5614 'channel': 'Sergey M.',
5615 'description': '',
5616 'modified_date': '20230921',
5617 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5618 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5619 'availability': 'unlisted',
5620 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5621 'uploader_id': '@sergeym.6173',
5622 'uploader': 'Sergey M.',
5623 },
5624 'playlist_count': 0,
5625 }, {
5626 'note': 'Home tab',
5627 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5628 'info_dict': {
5629 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5630 'title': 'lex will - Home',
5631 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5632 'uploader': 'lex will',
5633 'uploader_id': '@lexwill718',
5634 'channel': 'lex will',
5635 'tags': ['bible', 'history', 'prophesy'],
5636 'uploader_url': 'https://www.youtube.com/@lexwill718',
5637 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5638 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5639 'channel_follower_count': int,
5640 },
5641 'playlist_mincount': 2,
5642 }, {
5643 'note': 'Videos tab',
5644 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5645 'info_dict': {
5646 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5647 'title': 'lex will - Videos',
5648 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5649 'uploader': 'lex will',
5650 'uploader_id': '@lexwill718',
5651 'tags': ['bible', 'history', 'prophesy'],
5652 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5653 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5654 'uploader_url': 'https://www.youtube.com/@lexwill718',
5655 'channel': 'lex will',
5656 'channel_follower_count': int,
5657 },
5658 'playlist_mincount': 975,
5659 }, {
5660 'note': 'Videos tab, sorted by popular',
5661 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5662 'info_dict': {
5663 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5664 'title': 'lex will - Videos',
5665 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5666 'uploader': 'lex will',
5667 'uploader_id': '@lexwill718',
5668 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5669 'uploader_url': 'https://www.youtube.com/@lexwill718',
5670 'channel': 'lex will',
5671 'tags': ['bible', 'history', 'prophesy'],
5672 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5673 'channel_follower_count': int,
5674 },
5675 'playlist_mincount': 199,
5676 }, {
5677 'note': 'Playlists tab',
5678 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5679 'info_dict': {
5680 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5681 'title': 'lex will - Playlists',
5682 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5683 'uploader': 'lex will',
5684 'uploader_id': '@lexwill718',
5685 'uploader_url': 'https://www.youtube.com/@lexwill718',
5686 'channel': 'lex will',
5687 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5688 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5689 'tags': ['bible', 'history', 'prophesy'],
5690 'channel_follower_count': int,
5691 },
5692 'playlist_mincount': 17,
5693 }, {
5694 'note': 'Community tab',
5695 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5696 'info_dict': {
5697 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5698 'title': 'lex will - Community',
5699 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5700 'channel': 'lex will',
5701 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5702 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5703 'tags': ['bible', 'history', 'prophesy'],
5704 'channel_follower_count': int,
5705 'uploader_url': 'https://www.youtube.com/@lexwill718',
5706 'uploader_id': '@lexwill718',
5707 'uploader': 'lex will',
5708 },
5709 'playlist_mincount': 18,
5710 }, {
5711 'note': 'Channels tab',
5712 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5713 'info_dict': {
5714 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5715 'title': 'lex will - Channels',
5716 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5717 'channel': 'lex will',
5718 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5719 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5720 'tags': ['bible', 'history', 'prophesy'],
5721 'channel_follower_count': int,
5722 'uploader_url': 'https://www.youtube.com/@lexwill718',
5723 'uploader_id': '@lexwill718',
5724 'uploader': 'lex will',
5725 },
5726 'playlist_mincount': 12,
5727 }, {
5728 'note': 'Search tab',
5729 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5730 'playlist_mincount': 40,
5731 'info_dict': {
5732 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5733 'title': '3Blue1Brown - Search - linear algebra',
5734 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
5735 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5736 'tags': ['Mathematics'],
5737 'channel': '3Blue1Brown',
5738 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5739 'channel_follower_count': int,
5740 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5741 'uploader_id': '@3blue1brown',
5742 'uploader': '3Blue1Brown',
5743 'channel_is_verified': True,
5744 },
5745 }, {
5746 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5747 'only_matching': True,
5748 }, {
5749 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5750 'only_matching': True,
5751 }, {
5752 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5753 'only_matching': True,
5754 }, {
5755 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5756 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5757 'info_dict': {
5758 'title': '29C3: Not my department',
5759 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5760 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5761 'tags': [],
5762 'view_count': int,
5763 'modified_date': '20150605',
5764 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5765 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5766 'channel': 'Christiaan008',
5767 'availability': 'public',
5768 'uploader_id': '@ChRiStIaAn008',
5769 'uploader': 'Christiaan008',
5770 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5771 },
5772 'playlist_count': 96,
5773 }, {
5774 'note': 'Large playlist',
5775 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5776 'info_dict': {
5777 'title': 'Uploads from Cauchemar',
5778 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5779 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5780 'tags': [],
5781 'modified_date': r're:\d{8}',
5782 'channel': 'Cauchemar',
5783 'view_count': int,
5784 'description': '',
5785 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5786 'availability': 'public',
5787 'uploader_id': '@Cauchemar89',
5788 'uploader': 'Cauchemar',
5789 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5790 },
5791 'playlist_mincount': 1123,
5792 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5793 }, {
5794 'note': 'even larger playlist, 8832 videos',
5795 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5796 'only_matching': True,
5797 }, {
5798 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5799 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5800 'info_dict': {
5801 'title': 'Uploads from Interstellar Movie',
5802 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5803 'tags': [],
5804 'view_count': int,
5805 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5806 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5807 'channel': 'Interstellar Movie',
5808 'description': '',
5809 'modified_date': r're:\d{8}',
5810 'availability': 'public',
5811 'uploader_id': '@InterstellarMovie',
5812 'uploader': 'Interstellar Movie',
5813 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5814 },
5815 'playlist_mincount': 21,
5816 }, {
5817 'note': 'Playlist with "show unavailable videos" button',
5818 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5819 'info_dict': {
5820 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5821 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5822 'view_count': int,
5823 'channel': 'Phim Siêu Nhân Nhật Bản',
5824 'tags': [],
5825 'description': '',
5826 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5827 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5828 'modified_date': r're:\d{8}',
5829 'availability': 'public',
5830 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5831 'uploader_id': '@phimsieunhannhatban',
5832 'uploader': 'Phim Siêu Nhân Nhật Bản',
5833 },
5834 'playlist_mincount': 200,
5835 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5836 }, {
5837 'note': 'Playlist with unavailable videos in page 7',
5838 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5839 'info_dict': {
5840 'title': 'Uploads from BlankTV',
5841 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5842 'channel': 'BlankTV',
5843 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5844 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5845 'view_count': int,
5846 'tags': [],
5847 'modified_date': r're:\d{8}',
5848 'description': '',
5849 'availability': 'public',
5850 'uploader_id': '@blanktv',
5851 'uploader': 'BlankTV',
5852 'uploader_url': 'https://www.youtube.com/@blanktv',
5853 },
5854 'playlist_mincount': 1000,
5855 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5856 }, {
5857 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5858 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5859 'info_dict': {
5860 'title': 'Data Analysis with Dr Mike Pound',
5861 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5862 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5863 'tags': [],
5864 'view_count': int,
5865 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5866 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5867 'channel': 'Computerphile',
5868 'availability': 'public',
5869 'modified_date': '20190712',
5870 'uploader_id': '@Computerphile',
5871 'uploader': 'Computerphile',
5872 'uploader_url': 'https://www.youtube.com/@Computerphile',
5873 },
5874 'playlist_mincount': 11,
5875 }, {
5876 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5877 'only_matching': True,
5878 }, {
5879 'note': 'Playlist URL that does not actually serve a playlist',
5880 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5881 'info_dict': {
5882 'id': 'FqZTN594JQw',
5883 'ext': 'webm',
5884 'title': "Smiley's People 01 detective, Adventure Series, Action",
5885 'upload_date': '20150526',
5886 'license': 'Standard YouTube License',
5887 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5888 'categories': ['People & Blogs'],
5889 'tags': list,
5890 'view_count': int,
5891 'like_count': int,
5892 },
5893 'params': {
5894 'skip_download': True,
5895 },
5896 'skip': 'This video is not available.',
5897 'add_ie': [YoutubeIE.ie_key()],
5898 }, {
5899 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5900 'only_matching': True,
5901 }, {
5902 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5903 'only_matching': True,
5904 }, {
5905 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5906 'info_dict': {
5907 'id': 'hGkQjiJLjWQ', # This will keep changing
5908 'ext': 'mp4',
5909 'title': str,
5910 'upload_date': r're:\d{8}',
5911 'description': str,
5912 'categories': ['News & Politics'],
5913 'tags': list,
5914 'like_count': int,
5915 'release_timestamp': int,
5916 'channel': 'Sky News',
5917 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5918 'age_limit': 0,
5919 'view_count': int,
5920 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5921 'playable_in_embed': True,
5922 'release_date': r're:\d+',
5923 'availability': 'public',
5924 'live_status': 'is_live',
5925 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5926 'channel_follower_count': int,
5927 'concurrent_view_count': int,
5928 'uploader_url': 'https://www.youtube.com/@SkyNews',
5929 'uploader_id': '@SkyNews',
5930 'uploader': 'Sky News',
5931 'channel_is_verified': True,
5932 },
5933 'params': {
5934 'skip_download': True,
5935 },
5936 'expected_warnings': ['Ignoring subtitle tracks found in '],
5937 }, {
5938 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5939 'info_dict': {
5940 'id': 'a48o2S1cPoo',
5941 'ext': 'mp4',
5942 'title': 'The Young Turks - Live Main Show',
5943 'upload_date': '20150715',
5944 'license': 'Standard YouTube License',
5945 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5946 'categories': ['News & Politics'],
5947 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5948 'like_count': int,
5949 },
5950 'params': {
5951 'skip_download': True,
5952 },
5953 'only_matching': True,
5954 }, {
5955 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5956 'only_matching': True,
5957 }, {
5958 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5959 'only_matching': True,
5960 }, {
5961 'note': 'A channel that is not live. Should raise error',
5962 'url': 'https://www.youtube.com/user/numberphile/live',
5963 'only_matching': True,
5964 }, {
5965 'url': 'https://www.youtube.com/feed/trending',
5966 'only_matching': True,
5967 }, {
5968 'url': 'https://www.youtube.com/feed/library',
5969 'only_matching': True,
5970 }, {
5971 'url': 'https://www.youtube.com/feed/history',
5972 'only_matching': True,
5973 }, {
5974 'url': 'https://www.youtube.com/feed/subscriptions',
5975 'only_matching': True,
5976 }, {
5977 'url': 'https://www.youtube.com/feed/watch_later',
5978 'only_matching': True,
5979 }, {
5980 'note': 'Recommended - redirects to home page.',
5981 'url': 'https://www.youtube.com/feed/recommended',
5982 'only_matching': True,
5983 }, {
5984 'note': 'inline playlist with not always working continuations',
5985 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5986 'only_matching': True,
5987 }, {
5988 'url': 'https://www.youtube.com/course',
5989 'only_matching': True,
5990 }, {
5991 'url': 'https://www.youtube.com/zsecurity',
5992 'only_matching': True,
5993 }, {
5994 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5995 'only_matching': True,
5996 }, {
5997 'url': 'https://www.youtube.com/TheYoungTurks/live',
5998 'only_matching': True,
5999 }, {
6000 'url': 'https://www.youtube.com/hashtag/cctv9',
6001 'info_dict': {
6002 'id': 'cctv9',
6003 'title': 'cctv9 - All',
6004 'tags': [],
6005 },
6006 'playlist_mincount': 300, # not consistent but should be over 300
6007 }, {
6008 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
6009 'only_matching': True,
6010 }, {
6011 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
6012 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
6013 'only_matching': True,
6014 }, {
6015 'note': '/browse/ should redirect to /channel/',
6016 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
6017 'only_matching': True,
6018 }, {
6019 'note': 'VLPL, should redirect to playlist?list=PL...',
6020 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
6021 'info_dict': {
6022 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
6023 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
6024 'title': 'NCS : All Releases 💿',
6025 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
6026 'modified_date': r're:\d{8}',
6027 'view_count': int,
6028 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
6029 'tags': [],
6030 'channel': 'NoCopyrightSounds',
6031 'availability': 'public',
6032 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
6033 'uploader': 'NoCopyrightSounds',
6034 'uploader_id': '@NoCopyrightSounds',
6035 },
6036 'playlist_mincount': 166,
6037 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
6038 }, {
6039 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
6040 'note': 'Topic, should redirect to playlist?list=UU...',
6041 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6042 'info_dict': {
6043 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6044 'title': 'Uploads from Royalty Free Music - Topic',
6045 'tags': [],
6046 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6047 'channel': 'Royalty Free Music - Topic',
6048 'view_count': int,
6049 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6050 'modified_date': r're:\d{8}',
6051 'description': '',
6052 'availability': 'public',
6053 'uploader': 'Royalty Free Music - Topic',
6054 },
6055 'playlist_mincount': 101,
6056 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6057 }, {
6058 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
6059 # Treat as a general feed
6060 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
6061 'info_dict': {
6062 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
6063 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
6064 'tags': [],
6065 },
6066 'playlist_mincount': 9,
6067 }, {
6068 'note': 'Youtube music Album',
6069 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
6070 'info_dict': {
6071 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
6072 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
6073 'tags': [],
6074 'view_count': int,
6075 'description': '',
6076 'availability': 'unlisted',
6077 'modified_date': r're:\d{8}',
6078 },
6079 'playlist_count': 50,
6080 'expected_warnings': ['YouTube Music is not directly supported'],
6081 }, {
6082 'note': 'unlisted single video playlist',
6083 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
6084 'info_dict': {
6085 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
6086 'title': 'yt-dlp unlisted playlist test',
6087 'availability': 'unlisted',
6088 'tags': [],
6089 'modified_date': '20220418',
6090 'channel': 'colethedj',
6091 'view_count': int,
6092 'description': '',
6093 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
6094 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
6095 'uploader_url': 'https://www.youtube.com/@colethedj1894',
6096 'uploader_id': '@colethedj1894',
6097 'uploader': 'colethedj',
6098 },
6099 'playlist': [{
6100 'info_dict': {
6101 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
6102 'id': 'BaW_jenozKc',
6103 '_type': 'url',
6104 'ie_key': 'Youtube',
6105 'duration': 10,
6106 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
6107 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
6108 'view_count': int,
6109 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
6110 'channel': 'Philipp Hagemeister',
6111 'uploader_id': '@PhilippHagemeister',
6112 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
6113 'uploader': 'Philipp Hagemeister',
6114 },
6115 }],
6116 'playlist_count': 1,
6117 'params': {'extract_flat': True},
6118 }, {
6119 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
6120 'url': 'https://www.youtube.com/feed/recommended',
6121 'info_dict': {
6122 'id': 'recommended',
6123 'title': 'recommended',
6124 'tags': [],
6125 },
6126 'playlist_mincount': 50,
6127 'params': {
6128 'skip_download': True,
6129 'extractor_args': {'youtubetab': {'skip': ['webpage']}},
6130 },
6131 }, {
6132 'note': 'API Fallback: /videos tab, sorted by oldest first',
6133 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
6134 'info_dict': {
6135 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6136 'title': 'Cody\'sLab - Videos',
6137 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
6138 'channel': 'Cody\'sLab',
6139 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
6140 'tags': [],
6141 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
6142 'channel_follower_count': int,
6143 },
6144 'playlist_mincount': 650,
6145 'params': {
6146 'skip_download': True,
6147 'extractor_args': {'youtubetab': {'skip': ['webpage']}},
6148 },
6149 'skip': 'Query for sorting no longer works',
6150 }, {
6151 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
6152 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
6153 'info_dict': {
6154 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
6155 'title': 'Uploads from Royalty Free Music - Topic',
6156 'modified_date': r're:\d{8}',
6157 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
6158 'description': '',
6159 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
6160 'tags': [],
6161 'channel': 'Royalty Free Music - Topic',
6162 'view_count': int,
6163 'availability': 'public',
6164 'uploader': 'Royalty Free Music - Topic',
6165 },
6166 'playlist_mincount': 101,
6167 'params': {
6168 'skip_download': True,
6169 'extractor_args': {'youtubetab': {'skip': ['webpage']}},
6170 },
6171 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
6172 }, {
6173 'note': 'non-standard redirect to regional channel',
6174 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
6175 'only_matching': True,
6176 }, {
6177 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
6178 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6179 'info_dict': {
6180 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
6181 'modified_date': '20220407',
6182 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
6183 'tags': [],
6184 'availability': 'unlisted',
6185 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
6186 'channel': 'pukkandan',
6187 'description': 'Test for collaborative playlist',
6188 'title': 'yt-dlp test - collaborative playlist',
6189 'view_count': int,
6190 'uploader_url': 'https://www.youtube.com/@pukkandan',
6191 'uploader_id': '@pukkandan',
6192 'uploader': 'pukkandan',
6193 },
6194 'playlist_mincount': 2,
6195 }, {
6196 'note': 'translated tab name',
6197 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
6198 'info_dict': {
6199 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6200 'tags': [],
6201 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6202 'description': 'test description',
6203 'title': 'cole-dlp-test-acc - 再生リスト',
6204 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6205 'channel': 'cole-dlp-test-acc',
6206 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6207 'uploader_id': '@coletdjnz',
6208 'uploader': 'cole-dlp-test-acc',
6209 },
6210 'playlist_mincount': 1,
6211 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6212 'expected_warnings': ['Preferring "ja"'],
6213 }, {
6214 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6215 'note': 'preferred lang set with playlist with translated video titles',
6216 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6217 'info_dict': {
6218 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6219 'tags': [],
6220 'view_count': int,
6221 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6222 'channel': 'cole-dlp-test-acc',
6223 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6224 'description': 'test',
6225 'title': 'dlp test playlist',
6226 'availability': 'public',
6227 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6228 'uploader_id': '@coletdjnz',
6229 'uploader': 'cole-dlp-test-acc',
6230 },
6231 'playlist_mincount': 1,
6232 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6233 'expected_warnings': ['Preferring "ja"'],
6234 }, {
6235 # shorts audio pivot for 2GtVksBMYFM.
6236 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6237 'info_dict': {
6238 'id': 'sfv_audio_pivot',
6239 'title': 'sfv_audio_pivot',
6240 'tags': [],
6241 },
6242 'playlist_mincount': 50,
6243
6244 }, {
6245 # Channel with a real live tab (not to be mistaken with streams tab)
6246 # Do not treat like it should redirect to live stream
6247 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6248 'info_dict': {
6249 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6250 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6251 'tags': [],
6252 },
6253 'playlist_mincount': 20,
6254 }, {
6255 # Tab name is not the same as tab id
6256 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6257 'info_dict': {
6258 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6259 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6260 'tags': [],
6261 },
6262 'playlist_mincount': 8,
6263 }, {
6264 # Home tab id is literally home. Not to get mistaken with featured
6265 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6266 'info_dict': {
6267 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6268 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6269 'tags': [],
6270 },
6271 'playlist_mincount': 8,
6272 }, {
6273 # Should get three playlists for videos, shorts and streams tabs
6274 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6275 'info_dict': {
6276 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6277 'title': 'Polka Ch. 尾丸ポルカ',
6278 'channel_follower_count': int,
6279 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6280 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6281 'description': 'md5:49809d8bf9da539bc48ed5d1f83c33f2',
6282 'channel': 'Polka Ch. 尾丸ポルカ',
6283 'tags': 'count:35',
6284 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6285 'uploader': 'Polka Ch. 尾丸ポルカ',
6286 'uploader_id': '@OmaruPolka',
6287 'channel_is_verified': True,
6288 },
6289 'playlist_count': 3,
6290 }, {
6291 # Shorts tab with channel with handle
6292 # TODO: fix channel description
6293 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6294 'info_dict': {
6295 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6296 'title': 'Not Just Bikes - Shorts',
6297 'tags': 'count:10',
6298 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6299 'description': 'md5:5e82545b3a041345927a92d0585df247',
6300 'channel_follower_count': int,
6301 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6302 'channel': 'Not Just Bikes',
6303 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6304 'uploader': 'Not Just Bikes',
6305 'uploader_id': '@NotJustBikes',
6306 'channel_is_verified': True,
6307 },
6308 'playlist_mincount': 10,
6309 }, {
6310 # Streams tab
6311 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6312 'info_dict': {
6313 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6314 'title': '中村悠一 - Live',
6315 'tags': 'count:7',
6316 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6317 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6318 'channel': '中村悠一',
6319 'channel_follower_count': int,
6320 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6321 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6322 'uploader_id': '@Yuichi-Nakamura',
6323 'uploader': '中村悠一',
6324 },
6325 'playlist_mincount': 60,
6326 }, {
6327 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6328 # See test_youtube_lists
6329 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6330 'only_matching': True,
6331 }, {
6332 # No uploads and no UCID given. Should fail with no uploads error
6333 # See test_youtube_lists
6334 'url': 'https://www.youtube.com/news',
6335 'only_matching': True,
6336 }, {
6337 # No videos tab but has a shorts tab
6338 'url': 'https://www.youtube.com/c/TKFShorts',
6339 'info_dict': {
6340 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6341 'title': 'Shorts Break - Shorts',
6342 'tags': 'count:48',
6343 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6344 'channel': 'Shorts Break',
6345 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6346 'channel_follower_count': int,
6347 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6348 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6349 'uploader': 'Shorts Break',
6350 'uploader_id': '@ShortsBreak_Official',
6351 },
6352 'playlist_mincount': 30,
6353 }, {
6354 # Trending Now Tab. tab id is empty
6355 'url': 'https://www.youtube.com/feed/trending',
6356 'info_dict': {
6357 'id': 'trending',
6358 'title': 'trending - Now',
6359 'tags': [],
6360 },
6361 'playlist_mincount': 30,
6362 }, {
6363 # Trending Gaming Tab. tab id is empty
6364 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6365 'info_dict': {
6366 'id': 'trending',
6367 'title': 'trending - Gaming',
6368 'tags': [],
6369 },
6370 'playlist_mincount': 30,
6371 }, {
6372 # Shorts url result in shorts tab
6373 # TODO: Fix channel id extraction
6374 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6375 'info_dict': {
6376 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6377 'title': 'cole-dlp-test-acc - Shorts',
6378 'channel': 'cole-dlp-test-acc',
6379 'description': 'test description',
6380 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6381 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6382 'tags': [],
6383 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6384 'uploader_id': '@coletdjnz',
6385 'uploader': 'cole-dlp-test-acc',
6386 },
6387 'playlist': [{
6388 'info_dict': {
6389 # Channel data is not currently available for short renderers (as of 2023-03-01)
6390 '_type': 'url',
6391 'ie_key': 'Youtube',
6392 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6393 'id': 'sSM9J5YH_60',
6394 'title': 'SHORT short',
6395 'view_count': int,
6396 'thumbnails': list,
6397 },
6398 }],
6399 'params': {'extract_flat': True},
6400 }, {
6401 # Live video status should be extracted
6402 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6403 'info_dict': {
6404 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6405 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO: should be Minecraft - Live or Minecraft - Topic - Live
6406 'tags': [],
6407 },
6408 'playlist': [{
6409 'info_dict': {
6410 '_type': 'url',
6411 'ie_key': 'Youtube',
6412 'url': 'startswith:https://www.youtube.com/watch?v=',
6413 'id': str,
6414 'title': str,
6415 'live_status': 'is_live',
6416 'channel_id': str,
6417 'channel_url': str,
6418 'concurrent_view_count': int,
6419 'channel': str,
6420 'uploader': str,
6421 'uploader_url': str,
6422 'uploader_id': str,
6423 'channel_is_verified': bool, # this will keep changing
6424 },
6425 }],
6426 'params': {'extract_flat': True, 'playlist_items': '1'},
6427 'playlist_mincount': 1,
6428 }, {
6429 # Channel renderer metadata. Contains number of videos on the channel
6430 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6431 'info_dict': {
6432 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6433 'title': 'cole-dlp-test-acc - Channels',
6434 'channel': 'cole-dlp-test-acc',
6435 'description': 'test description',
6436 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6437 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6438 'tags': [],
6439 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6440 'uploader_id': '@coletdjnz',
6441 'uploader': 'cole-dlp-test-acc',
6442 },
6443 'playlist': [{
6444 'info_dict': {
6445 '_type': 'url',
6446 'ie_key': 'YoutubeTab',
6447 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6448 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6449 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6450 'title': 'PewDiePie',
6451 'channel': 'PewDiePie',
6452 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6453 'thumbnails': list,
6454 'channel_follower_count': int,
6455 'playlist_count': int,
6456 'uploader': 'PewDiePie',
6457 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6458 'uploader_id': '@PewDiePie',
6459 'channel_is_verified': True,
6460 },
6461 }],
6462 'params': {'extract_flat': True},
6463 }, {
6464 'url': 'https://www.youtube.com/@3blue1brown/about',
6465 'info_dict': {
6466 'id': '@3blue1brown',
6467 'tags': ['Mathematics'],
6468 'title': '3Blue1Brown',
6469 'channel_follower_count': int,
6470 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6471 'channel': '3Blue1Brown',
6472 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6473 'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
6474 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6475 'uploader_id': '@3blue1brown',
6476 'uploader': '3Blue1Brown',
6477 'channel_is_verified': True,
6478 },
6479 'playlist_count': 0,
6480 }, {
6481 # Podcasts tab, with rich entry playlistRenderers
6482 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6483 'info_dict': {
6484 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6485 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6486 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6487 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6488 'title': '99 Percent Invisible - Podcasts',
6489 'uploader': '99 Percent Invisible',
6490 'channel_follower_count': int,
6491 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6492 'tags': [],
6493 'channel': '99 Percent Invisible',
6494 'uploader_id': '@99percentinvisiblepodcast',
6495 },
6496 'playlist_count': 0,
6497 }, {
6498 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6499 'url': 'https://www.youtube.com/@AHimitsu/releases',
6500 'info_dict': {
6501 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6502 'channel': 'A Himitsu',
6503 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6504 'title': 'A Himitsu - Releases',
6505 'uploader_id': '@AHimitsu',
6506 'uploader': 'A Himitsu',
6507 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6508 'tags': 'count:12',
6509 'description': 'I make music',
6510 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6511 'channel_follower_count': int,
6512 'channel_is_verified': True,
6513 },
6514 'playlist_mincount': 10,
6515 }, {
6516 # Playlist with only shorts, shown as reel renderers
6517 # FIXME: future: YouTube currently doesn't give continuation for this,
6518 # may do in future.
6519 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
6520 'info_dict': {
6521 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
6522 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
6523 'view_count': int,
6524 'uploader_id': '@BangyShorts',
6525 'description': '',
6526 'uploader_url': 'https://www.youtube.com/@BangyShorts',
6527 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
6528 'channel': 'Bangy Shorts',
6529 'uploader': 'Bangy Shorts',
6530 'tags': [],
6531 'availability': 'public',
6532 'modified_date': r're:\d{8}',
6533 'title': 'Uploads from Bangy Shorts',
6534 },
6535 'playlist_mincount': 100,
6536 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6537 }, {
6538 'note': 'Tags containing spaces',
6539 'url': 'https://www.youtube.com/channel/UC7_YxT-KID8kRbqZo7MyscQ',
6540 'playlist_count': 3,
6541 'info_dict': {
6542 'id': 'UC7_YxT-KID8kRbqZo7MyscQ',
6543 'channel': 'Markiplier',
6544 'channel_id': 'UC7_YxT-KID8kRbqZo7MyscQ',
6545 'title': 'Markiplier',
6546 'channel_follower_count': int,
6547 'description': 'md5:0c010910558658824402809750dc5d97',
6548 'uploader_id': '@markiplier',
6549 'uploader_url': 'https://www.youtube.com/@markiplier',
6550 'uploader': 'Markiplier',
6551 'channel_url': 'https://www.youtube.com/channel/UC7_YxT-KID8kRbqZo7MyscQ',
6552 'channel_is_verified': True,
6553 'tags': ['markiplier', 'comedy', 'gaming', 'funny videos', 'funny moments',
6554 'sketch comedy', 'laughing', 'lets play', 'challenge videos', 'hilarious',
6555 'challenges', 'sketches', 'scary games', 'funny games', 'rage games',
6556 'mark fischbach'],
6557 },
6558 }]
6559
6560 @classmethod
6561 def suitable(cls, url):
6562 return False if YoutubeIE.suitable(url) else super().suitable(url)
6563
6564 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6565
6566 def _get_url_mobj(self, url):
6567 mobj = self._URL_RE.match(url).groupdict()
6568 mobj.update((k, '') for k, v in mobj.items() if v is None)
6569 return mobj
6570
6571 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6572 tab_name = (tab.get('title') or '').lower()
6573 tab_url = urljoin(base_url, traverse_obj(
6574 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6575
6576 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6577 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6578 if tab_id:
6579 return {
6580 'TAB_ID_SPONSORSHIPS': 'membership',
6581 }.get(tab_id, tab_id), tab_name
6582
6583 # Fallback to tab name if we cannot get the tab id.
6584 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6585 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6586 if tab_name:
6587 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6588 return {
6589 'home': 'featured',
6590 'live': 'streams',
6591 }.get(tab_name, tab_name), tab_name
6592
6593 def _has_tab(self, tabs, tab_id):
6594 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6595
6596 def _empty_playlist(self, item_id, data):
6597 return self.playlist_result([], item_id, **self._extract_metadata_from_tabs(item_id, data))
6598
6599 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6600 def _real_extract(self, url, smuggled_data):
6601 item_id = self._match_id(url)
6602 url = urllib.parse.urlunparse(
6603 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6604 compat_opts = self.get_param('compat_opts', [])
6605
6606 mobj = self._get_url_mobj(url)
6607 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6608 if is_channel and smuggled_data.get('is_music_url'):
6609 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6610 return self.url_result(
6611 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6612 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6613 mdata = self._extract_tab_endpoint(
6614 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6615 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6616 get_all=False, expected_type=str)
6617 if not murl:
6618 raise ExtractorError('Failed to resolve album to playlist')
6619 return self.url_result(murl, YoutubeTabIE)
6620 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6621 return self.url_result(
6622 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6623
6624 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6625 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6626 url = f'{pre}/videos{post}'
6627 if smuggled_data.get('is_music_url'):
6628 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6629
6630 # Handle both video/playlist URLs
6631 qs = parse_qs(url)
6632 video_id, playlist_id = (traverse_obj(qs, (key, 0)) for key in ('v', 'list'))
6633 if not video_id and mobj['not_channel'].startswith('watch'):
6634 if not playlist_id:
6635 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6636 raise ExtractorError('A video URL was given without video ID', expected=True)
6637 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6638 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6639 return self.url_result(
6640 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6641
6642 if not self._yes_playlist(playlist_id, video_id):
6643 return self.url_result(
6644 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6645
6646 data, ytcfg = self._extract_data(url, display_id)
6647
6648 # YouTube may provide a non-standard redirect to the regional channel
6649 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6650 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6651 redirect_url = traverse_obj(
6652 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6653 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6654 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6655 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6656 return self.url_result(redirect_url, YoutubeTabIE)
6657
6658 tabs, extra_tabs = self._extract_tab_renderers(data), []
6659 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6660 selected_tab = self._extract_selected_tab(tabs)
6661 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6662 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6663
6664 # /about is no longer a tab
6665 if original_tab_id == 'about':
6666 return self._empty_playlist(item_id, data)
6667
6668 if not original_tab_id and selected_tab_name:
6669 self.to_screen('Downloading all uploads of the channel. '
6670 'To download only the videos in a specific tab, pass the tab\'s URL')
6671 if self._has_tab(tabs, 'streams'):
6672 extra_tabs.append(''.join((pre, '/streams', post)))
6673 if self._has_tab(tabs, 'shorts'):
6674 extra_tabs.append(''.join((pre, '/shorts', post)))
6675 # XXX: Members-only tab should also be extracted
6676
6677 if not extra_tabs and selected_tab_id != 'videos':
6678 # Channel does not have streams, shorts or videos tabs
6679 if item_id[:2] != 'UC':
6680 return self._empty_playlist(item_id, data)
6681
6682 # Topic channels don't have /videos. Use the equivalent playlist instead
6683 pl_id = f'UU{item_id[2:]}'
6684 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6685 try:
6686 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6687 except ExtractorError:
6688 return self._empty_playlist(item_id, data)
6689 else:
6690 item_id, url = pl_id, pl_url
6691 self.to_screen(
6692 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6693
6694 elif extra_tabs and selected_tab_id != 'videos':
6695 # When there are shorts/live tabs but not videos tab
6696 url, data = f'{pre}{post}', None
6697
6698 elif (original_tab_id or 'videos') != selected_tab_id:
6699 if original_tab_id == 'live':
6700 # Live tab should have redirected to the video
6701 # Except in the case the channel has an actual live tab
6702 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6703 raise UserNotLive(video_id=item_id)
6704 elif selected_tab_name:
6705 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6706
6707 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6708 url = f'{pre}{post}'
6709
6710 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6711 if 'no-youtube-unavailable-videos' not in compat_opts:
6712 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6713 self._extract_and_report_alerts(data, only_once=True)
6714
6715 tabs, entries = self._extract_tab_renderers(data), []
6716 if tabs:
6717 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6718 entries[0].update({
6719 'extractor_key': YoutubeTabIE.ie_key(),
6720 'extractor': YoutubeTabIE.IE_NAME,
6721 'webpage_url': url,
6722 })
6723 if self.get_param('playlist_items') == '0':
6724 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6725 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6726 entries.extend(map(self._real_extract, extra_tabs))
6727
6728 if len(entries) == 1:
6729 return entries[0]
6730 elif entries:
6731 metadata = self._extract_metadata_from_tabs(item_id, data)
6732 uploads_url = 'the Uploads (UU) playlist URL'
6733 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6734 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6735 self.to_screen(
6736 'Downloading as multiple playlists, separated by tabs. '
6737 f'To download as a single playlist instead, pass {uploads_url}')
6738 return self.playlist_result(entries, item_id, **metadata)
6739
6740 # Inline playlist
6741 playlist = traverse_obj(
6742 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6743 if playlist:
6744 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6745
6746 video_id = traverse_obj(
6747 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6748 if video_id:
6749 if tab != '/live': # live tab is expected to redirect to video
6750 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6751 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6752
6753 raise ExtractorError('Unable to recognize tab page')
6754
6755
6756 class YoutubePlaylistIE(InfoExtractor):
6757 IE_DESC = 'YouTube playlists'
6758 _VALID_URL = r'''(?x)(?:
6759 (?:https?://)?
6760 (?:\w+\.)?
6761 (?:
6762 (?:
6763 youtube(?:kids)?\.com|
6764 {invidious}
6765 )
6766 /.*?\?.*?\blist=
6767 )?
6768 (?P<id>{playlist_id})
6769 )'''.format(
6770 playlist_id=YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6771 invidious='|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6772 )
6773 IE_NAME = 'youtube:playlist'
6774 _TESTS = [{
6775 'note': 'issue #673',
6776 'url': 'PLBB231211A4F62143',
6777 'info_dict': {
6778 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6779 'id': 'PLBB231211A4F62143',
6780 'uploader': 'Wickman',
6781 'uploader_id': '@WickmanVT',
6782 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6783 'view_count': int,
6784 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6785 'modified_date': r're:\d{8}',
6786 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6787 'channel': 'Wickman',
6788 'tags': [],
6789 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6790 'availability': 'public',
6791 },
6792 'playlist_mincount': 29,
6793 }, {
6794 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6795 'info_dict': {
6796 'title': 'YDL_safe_search',
6797 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6798 },
6799 'playlist_count': 2,
6800 'skip': 'This playlist is private',
6801 }, {
6802 'note': 'embedded',
6803 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6804 'playlist_count': 4,
6805 'info_dict': {
6806 'title': 'JODA15',
6807 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6808 'uploader': 'milan',
6809 'uploader_id': '@milan5503',
6810 'description': '',
6811 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6812 'tags': [],
6813 'modified_date': '20140919',
6814 'view_count': int,
6815 'channel': 'milan',
6816 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6817 'uploader_url': 'https://www.youtube.com/@milan5503',
6818 'availability': 'public',
6819 },
6820 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden', 'Retrying', 'Giving up'],
6821 }, {
6822 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6823 'playlist_mincount': 455,
6824 'info_dict': {
6825 'title': '2018 Chinese New Singles (11/6 updated)',
6826 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6827 'uploader': 'LBK',
6828 'uploader_id': '@music_king',
6829 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6830 'channel': 'LBK',
6831 'view_count': int,
6832 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6833 'tags': [],
6834 'uploader_url': 'https://www.youtube.com/@music_king',
6835 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6836 'modified_date': r're:\d{8}',
6837 'availability': 'public',
6838 },
6839 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6840 }, {
6841 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6842 'only_matching': True,
6843 }, {
6844 # music album playlist
6845 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6846 'only_matching': True,
6847 }]
6848
6849 @classmethod
6850 def suitable(cls, url):
6851 if YoutubeTabIE.suitable(url):
6852 return False
6853 from ..utils import parse_qs
6854 qs = parse_qs(url)
6855 if qs.get('v', [None])[0]:
6856 return False
6857 return super().suitable(url)
6858
6859 def _real_extract(self, url):
6860 playlist_id = self._match_id(url)
6861 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6862 url = update_url_query(
6863 'https://www.youtube.com/playlist',
6864 parse_qs(url) or {'list': playlist_id})
6865 if is_music_url:
6866 url = smuggle_url(url, {'is_music_url': True})
6867 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6868
6869
6870 class YoutubeYtBeIE(InfoExtractor):
6871 IE_DESC = 'youtu.be'
6872 _VALID_URL = rf'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{{11}})/*?.*?\blist=(?P<playlist_id>{YoutubeBaseInfoExtractor._PLAYLIST_ID_RE})'
6873 _TESTS = [{
6874 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6875 'info_dict': {
6876 'id': 'yeWKywCrFtk',
6877 'ext': 'mp4',
6878 'title': 'Small Scale Baler and Braiding Rugs',
6879 'uploader': 'Backus-Page House Museum',
6880 'uploader_id': '@backuspagemuseum',
6881 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6882 'upload_date': '20161008',
6883 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6884 'categories': ['Nonprofits & Activism'],
6885 'tags': list,
6886 'like_count': int,
6887 'age_limit': 0,
6888 'playable_in_embed': True,
6889 'thumbnail': r're:^https?://.*\.webp',
6890 'channel': 'Backus-Page House Museum',
6891 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6892 'live_status': 'not_live',
6893 'view_count': int,
6894 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6895 'availability': 'public',
6896 'duration': 59,
6897 'comment_count': int,
6898 'channel_follower_count': int,
6899 },
6900 'params': {
6901 'noplaylist': True,
6902 'skip_download': True,
6903 },
6904 }, {
6905 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6906 'only_matching': True,
6907 }]
6908
6909 def _real_extract(self, url):
6910 mobj = self._match_valid_url(url)
6911 video_id = mobj.group('id')
6912 playlist_id = mobj.group('playlist_id')
6913 return self.url_result(
6914 update_url_query('https://www.youtube.com/watch', {
6915 'v': video_id,
6916 'list': playlist_id,
6917 'feature': 'youtu.be',
6918 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6919
6920
6921 class YoutubeLivestreamEmbedIE(InfoExtractor):
6922 IE_DESC = 'YouTube livestream embeds'
6923 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6924 _TESTS = [{
6925 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6926 'only_matching': True,
6927 }]
6928
6929 def _real_extract(self, url):
6930 channel_id = self._match_id(url)
6931 return self.url_result(
6932 f'https://www.youtube.com/channel/{channel_id}/live',
6933 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6934
6935
6936 class YoutubeYtUserIE(InfoExtractor):
6937 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6938 IE_NAME = 'youtube:user'
6939 _VALID_URL = r'ytuser:(?P<id>.+)'
6940 _TESTS = [{
6941 'url': 'ytuser:phihag',
6942 'only_matching': True,
6943 }]
6944
6945 def _real_extract(self, url):
6946 user_id = self._match_id(url)
6947 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6948
6949
6950 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6951 IE_NAME = 'youtube:favorites'
6952 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6953 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6954 _LOGIN_REQUIRED = True
6955 _TESTS = [{
6956 'url': ':ytfav',
6957 'only_matching': True,
6958 }, {
6959 'url': ':ytfavorites',
6960 'only_matching': True,
6961 }]
6962
6963 def _real_extract(self, url):
6964 return self.url_result(
6965 'https://www.youtube.com/playlist?list=LL',
6966 ie=YoutubeTabIE.ie_key())
6967
6968
6969 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6970 IE_NAME = 'youtube:notif'
6971 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6972 _VALID_URL = r':ytnotif(?:ication)?s?'
6973 _LOGIN_REQUIRED = True
6974 _TESTS = [{
6975 'url': ':ytnotif',
6976 'only_matching': True,
6977 }, {
6978 'url': ':ytnotifications',
6979 'only_matching': True,
6980 }]
6981
6982 def _extract_notification_menu(self, response, continuation_list):
6983 notification_list = traverse_obj(
6984 response,
6985 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6986 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6987 expected_type=list) or []
6988 continuation_list[0] = None
6989 for item in notification_list:
6990 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6991 if entry:
6992 yield entry
6993 continuation = item.get('continuationItemRenderer')
6994 if continuation:
6995 continuation_list[0] = continuation
6996
6997 def _extract_notification_renderer(self, notification):
6998 video_id = traverse_obj(
6999 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
7000 url = f'https://www.youtube.com/watch?v={video_id}'
7001 channel_id = None
7002 if not video_id:
7003 browse_ep = traverse_obj(
7004 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
7005 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
7006 post_id = self._search_regex(
7007 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
7008 'post id', default=None)
7009 if not channel_id or not post_id:
7010 return
7011 # The direct /post url redirects to this in the browser
7012 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
7013
7014 channel = traverse_obj(
7015 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
7016 expected_type=str)
7017 notification_title = self._get_text(notification, 'shortMessage')
7018 if notification_title:
7019 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
7020 # TODO: handle recommended videos
7021 title = self._search_regex(
7022 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
7023 'video title', default=None)
7024 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
7025 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
7026 else None)
7027 return {
7028 '_type': 'url',
7029 'url': url,
7030 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
7031 'video_id': video_id,
7032 'title': title,
7033 'channel_id': channel_id,
7034 'channel': channel,
7035 'uploader': channel,
7036 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
7037 'timestamp': timestamp,
7038 }
7039
7040 def _notification_menu_entries(self, ytcfg):
7041 continuation_list = [None]
7042 response = None
7043 for page in itertools.count(1):
7044 ctoken = traverse_obj(
7045 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
7046 response = self._extract_response(
7047 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
7048 ep='notification/get_notification_menu', check_get_keys='actions',
7049 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
7050 yield from self._extract_notification_menu(response, continuation_list)
7051 if not continuation_list[0]:
7052 break
7053
7054 def _real_extract(self, url):
7055 display_id = 'notifications'
7056 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
7057 self._report_playlist_authcheck(ytcfg)
7058 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
7059
7060
7061 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
7062 IE_DESC = 'YouTube search'
7063 IE_NAME = 'youtube:search'
7064 _SEARCH_KEY = 'ytsearch'
7065 _SEARCH_PARAMS = 'EgIQAfABAQ==' # Videos only
7066 _TESTS = [{
7067 'url': 'ytsearch5:youtube-dl test video',
7068 'playlist_count': 5,
7069 'info_dict': {
7070 'id': 'youtube-dl test video',
7071 'title': 'youtube-dl test video',
7072 },
7073 }, {
7074 'note': 'Suicide/self-harm search warning',
7075 'url': 'ytsearch1:i hate myself and i wanna die',
7076 'playlist_count': 1,
7077 'info_dict': {
7078 'id': 'i hate myself and i wanna die',
7079 'title': 'i hate myself and i wanna die',
7080 },
7081 }]
7082
7083
7084 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
7085 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
7086 _SEARCH_KEY = 'ytsearchdate'
7087 IE_DESC = 'YouTube search, newest videos first'
7088 _SEARCH_PARAMS = 'CAISAhAB8AEB' # Videos only, sorted by date
7089 _TESTS = [{
7090 'url': 'ytsearchdate5:youtube-dl test video',
7091 'playlist_count': 5,
7092 'info_dict': {
7093 'id': 'youtube-dl test video',
7094 'title': 'youtube-dl test video',
7095 },
7096 }]
7097
7098
7099 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
7100 IE_DESC = 'YouTube search URLs with sorting and filter support'
7101 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
7102 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7103 _TESTS = [{
7104 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
7105 'playlist_mincount': 5,
7106 'info_dict': {
7107 'id': 'youtube-dl test video',
7108 'title': 'youtube-dl test video',
7109 },
7110 }, {
7111 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
7112 'playlist_mincount': 5,
7113 'info_dict': {
7114 'id': 'python',
7115 'title': 'python',
7116 },
7117 }, {
7118 'url': 'https://www.youtube.com/results?search_query=%23cats',
7119 'playlist_mincount': 1,
7120 'info_dict': {
7121 'id': '#cats',
7122 'title': '#cats',
7123 # The test suite does not have support for nested playlists
7124 # 'entries': [{
7125 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
7126 # 'title': '#cats',
7127 # }],
7128 },
7129 }, {
7130 # Channel results
7131 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
7132 'info_dict': {
7133 'id': 'kurzgesagt',
7134 'title': 'kurzgesagt',
7135 },
7136 'playlist': [{
7137 'info_dict': {
7138 '_type': 'url',
7139 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7140 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7141 'ie_key': 'YoutubeTab',
7142 'channel': 'Kurzgesagt – In a Nutshell',
7143 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
7144 'title': 'Kurzgesagt – In a Nutshell',
7145 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
7146 # No longer available for search as it is set to the handle.
7147 # 'playlist_count': int,
7148 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
7149 'thumbnails': list,
7150 'uploader_id': '@kurzgesagt',
7151 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
7152 'uploader': 'Kurzgesagt – In a Nutshell',
7153 'channel_is_verified': True,
7154 'channel_follower_count': int,
7155 },
7156 }],
7157 'params': {'extract_flat': True, 'playlist_items': '1'},
7158 'playlist_mincount': 1,
7159 }, {
7160 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
7161 'only_matching': True,
7162 }]
7163
7164 def _real_extract(self, url):
7165 qs = parse_qs(url)
7166 query = (qs.get('search_query') or qs.get('q'))[0]
7167 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
7168
7169
7170 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
7171 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
7172 IE_NAME = 'youtube:music:search_url'
7173 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
7174 _TESTS = [{
7175 'url': 'https://music.youtube.com/search?q=royalty+free+music',
7176 'playlist_count': 16,
7177 'info_dict': {
7178 'id': 'royalty free music',
7179 'title': 'royalty free music',
7180 },
7181 }, {
7182 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
7183 'playlist_mincount': 30,
7184 'info_dict': {
7185 'id': 'royalty free music - songs',
7186 'title': 'royalty free music - songs',
7187 },
7188 'params': {'extract_flat': 'in_playlist'},
7189 }, {
7190 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
7191 'playlist_mincount': 30,
7192 'info_dict': {
7193 'id': 'royalty free music - community playlists',
7194 'title': 'royalty free music - community playlists',
7195 },
7196 'params': {'extract_flat': 'in_playlist'},
7197 }]
7198
7199 _SECTIONS = {
7200 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
7201 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
7202 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
7203 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
7204 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
7205 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
7206 }
7207
7208 def _real_extract(self, url):
7209 qs = parse_qs(url)
7210 query = (qs.get('search_query') or qs.get('q'))[0]
7211 params = qs.get('sp', (None,))[0]
7212 if params:
7213 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
7214 else:
7215 section = urllib.parse.unquote_plus(([*url.split('#'), ''])[1]).lower()
7216 params = self._SECTIONS.get(section)
7217 if not params:
7218 section = None
7219 title = join_nonempty(query, section, delim=' - ')
7220 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
7221
7222
7223 class YoutubeFeedsInfoExtractor(InfoExtractor):
7224 """
7225 Base class for feed extractors
7226 Subclasses must re-define the _FEED_NAME property.
7227 """
7228 _LOGIN_REQUIRED = True
7229 _FEED_NAME = 'feeds'
7230
7231 def _real_initialize(self):
7232 YoutubeBaseInfoExtractor._check_login_required(self)
7233
7234 @classproperty
7235 def IE_NAME(cls):
7236 return f'youtube:{cls._FEED_NAME}'
7237
7238 def _real_extract(self, url):
7239 return self.url_result(
7240 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
7241
7242
7243 class YoutubeWatchLaterIE(InfoExtractor):
7244 IE_NAME = 'youtube:watchlater'
7245 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
7246 _VALID_URL = r':ytwatchlater'
7247 _TESTS = [{
7248 'url': ':ytwatchlater',
7249 'only_matching': True,
7250 }]
7251
7252 def _real_extract(self, url):
7253 return self.url_result(
7254 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
7255
7256
7257 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7258 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
7259 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
7260 _FEED_NAME = 'recommended'
7261 _LOGIN_REQUIRED = False
7262 _TESTS = [{
7263 'url': ':ytrec',
7264 'only_matching': True,
7265 }, {
7266 'url': ':ytrecommended',
7267 'only_matching': True,
7268 }, {
7269 'url': 'https://youtube.com',
7270 'only_matching': True,
7271 }]
7272
7273
7274 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
7275 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
7276 _VALID_URL = r':ytsub(?:scription)?s?'
7277 _FEED_NAME = 'subscriptions'
7278 _TESTS = [{
7279 'url': ':ytsubs',
7280 'only_matching': True,
7281 }, {
7282 'url': ':ytsubscriptions',
7283 'only_matching': True,
7284 }]
7285
7286
7287 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7288 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7289 _VALID_URL = r':ythis(?:tory)?'
7290 _FEED_NAME = 'history'
7291 _TESTS = [{
7292 'url': ':ythistory',
7293 'only_matching': True,
7294 }]
7295
7296
7297 class YoutubeShortsAudioPivotIE(InfoExtractor):
7298 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7299 IE_NAME = 'youtube:shorts:pivot:audio'
7300 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7301 _TESTS = [{
7302 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7303 'only_matching': True,
7304 }]
7305
7306 @staticmethod
7307 def _generate_audio_pivot_params(video_id):
7308 """
7309 Generates sfv_audio_pivot browse params for this video id
7310 """
7311 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7312 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7313
7314 def _real_extract(self, url):
7315 video_id = self._match_id(url)
7316 return self.url_result(
7317 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7318 ie=YoutubeTabIE)
7319
7320
7321 class YoutubeTruncatedURLIE(InfoExtractor):
7322 IE_NAME = 'youtube:truncated_url'
7323 IE_DESC = False # Do not list
7324 _VALID_URL = r'''(?x)
7325 (?:https?://)?
7326 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7327 (?:watch\?(?:
7328 feature=[a-z_]+|
7329 annotation_id=annotation_[^&]+|
7330 x-yt-cl=[0-9]+|
7331 hl=[^&]*|
7332 t=[0-9]+
7333 )?
7334 |
7335 attribution_link\?a=[^&]+
7336 )
7337 $
7338 '''
7339
7340 _TESTS = [{
7341 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7342 'only_matching': True,
7343 }, {
7344 'url': 'https://www.youtube.com/watch?',
7345 'only_matching': True,
7346 }, {
7347 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7348 'only_matching': True,
7349 }, {
7350 'url': 'https://www.youtube.com/watch?feature=foo',
7351 'only_matching': True,
7352 }, {
7353 'url': 'https://www.youtube.com/watch?hl=en-GB',
7354 'only_matching': True,
7355 }, {
7356 'url': 'https://www.youtube.com/watch?t=2372',
7357 'only_matching': True,
7358 }]
7359
7360 def _real_extract(self, url):
7361 raise ExtractorError(
7362 'Did you forget to quote the URL? Remember that & is a meta '
7363 'character in most shells, so you want to put the URL in quotes, '
7364 'like youtube-dl '
7365 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7366 ' or simply youtube-dl BaW_jenozKc .',
7367 expected=True)
7368
7369
7370 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7371 IE_NAME = 'youtube:clip'
7372 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7373 _TESTS = [{
7374 # FIXME: Other metadata should be extracted from the clip, not from the base video
7375 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7376 'info_dict': {
7377 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7378 'ext': 'mp4',
7379 'section_start': 29.0,
7380 'section_end': 39.7,
7381 'duration': 10.7,
7382 'age_limit': 0,
7383 'availability': 'public',
7384 'categories': ['Gaming'],
7385 'channel': 'Scott The Woz',
7386 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7387 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7388 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7389 'like_count': int,
7390 'playable_in_embed': True,
7391 'tags': 'count:17',
7392 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7393 'title': 'Mobile Games on Console - Scott The Woz',
7394 'upload_date': '20210920',
7395 'uploader': 'Scott The Woz',
7396 'uploader_id': '@ScottTheWoz',
7397 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7398 'view_count': int,
7399 'live_status': 'not_live',
7400 'channel_follower_count': int,
7401 'chapters': 'count:20',
7402 'comment_count': int,
7403 'heatmap': 'count:100',
7404 },
7405 }]
7406
7407 def _real_extract(self, url):
7408 clip_id = self._match_id(url)
7409 _, data = self._extract_webpage(url, clip_id)
7410
7411 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7412 if not video_id:
7413 raise ExtractorError('Unable to find video ID')
7414
7415 clip_data = traverse_obj(data, (
7416 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7417 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7418 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7419 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7420
7421 return {
7422 '_type': 'url_transparent',
7423 'url': f'https://www.youtube.com/watch?v={video_id}',
7424 'ie_key': YoutubeIE.ie_key(),
7425 'id': clip_id,
7426 'section_start': int(clip_data['startTimeMs']) / 1000,
7427 'section_end': int(clip_data['endTimeMs']) / 1000,
7428 }
7429
7430
7431 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7432 IE_NAME = 'youtube:consent'
7433 IE_DESC = False # Do not list
7434 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7435 _TESTS = [{
7436 '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',
7437 'info_dict': {
7438 'id': 'qVv6vCqciTM',
7439 'ext': 'mp4',
7440 'age_limit': 0,
7441 'uploader_id': '@sana_natori',
7442 'comment_count': int,
7443 'chapters': 'count:13',
7444 'upload_date': '20221223',
7445 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7446 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7447 'uploader_url': 'https://www.youtube.com/@sana_natori',
7448 'like_count': int,
7449 'release_date': '20221223',
7450 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7451 '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】',
7452 'view_count': int,
7453 'playable_in_embed': True,
7454 'duration': 4438,
7455 'availability': 'public',
7456 'channel_follower_count': int,
7457 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7458 'categories': ['Entertainment'],
7459 'live_status': 'was_live',
7460 'release_timestamp': 1671793345,
7461 'channel': 'さなちゃんねる',
7462 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7463 'uploader': 'さなちゃんねる',
7464 'channel_is_verified': True,
7465 'heatmap': 'count:100',
7466 },
7467 'add_ie': ['Youtube'],
7468 'params': {'skip_download': 'Youtube'},
7469 }]
7470
7471 def _real_extract(self, url):
7472 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7473 if not redirect_url:
7474 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7475 return self.url_result(redirect_url)
7476
7477
7478 class YoutubeTruncatedIDIE(InfoExtractor):
7479 IE_NAME = 'youtube:truncated_id'
7480 IE_DESC = False # Do not list
7481 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7482
7483 _TESTS = [{
7484 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7485 'only_matching': True,
7486 }]
7487
7488 def _real_extract(self, url):
7489 video_id = self._match_id(url)
7490 raise ExtractorError(
7491 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7492 expected=True)