]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[extractor/youtube:music:search_url] Extract title (#7102)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 import base64
2 import calendar
3 import collections
4 import copy
5 import datetime
6 import enum
7 import hashlib
8 import itertools
9 import json
10 import math
11 import os.path
12 import random
13 import re
14 import sys
15 import threading
16 import time
17 import traceback
18 import urllib.error
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 ..utils import (
26 NO_DEFAULT,
27 ExtractorError,
28 LazyList,
29 UserNotLive,
30 bug_reports_message,
31 classproperty,
32 clean_html,
33 datetime_from_str,
34 dict_get,
35 filter_dict,
36 float_or_none,
37 format_field,
38 get_first,
39 int_or_none,
40 is_html,
41 join_nonempty,
42 js_to_json,
43 mimetype2ext,
44 network_exceptions,
45 orderedSet,
46 parse_codecs,
47 parse_count,
48 parse_duration,
49 parse_iso8601,
50 parse_qs,
51 qualities,
52 remove_start,
53 smuggle_url,
54 str_or_none,
55 str_to_int,
56 strftime_or_none,
57 traverse_obj,
58 try_get,
59 unescapeHTML,
60 unified_strdate,
61 unified_timestamp,
62 unsmuggle_url,
63 update_url_query,
64 url_or_none,
65 urljoin,
66 variadic,
67 )
68
69
70 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
71 # any clients starting with _ cannot be explicitly requested by the user
72 INNERTUBE_CLIENTS = {
73 'web': {
74 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
75 'INNERTUBE_CONTEXT': {
76 'client': {
77 'clientName': 'WEB',
78 'clientVersion': '2.20220801.00.00',
79 }
80 },
81 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
82 },
83 'web_embedded': {
84 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
85 'INNERTUBE_CONTEXT': {
86 'client': {
87 'clientName': 'WEB_EMBEDDED_PLAYER',
88 'clientVersion': '1.20220731.00.00',
89 },
90 },
91 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
92 },
93 'web_music': {
94 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
95 'INNERTUBE_HOST': 'music.youtube.com',
96 'INNERTUBE_CONTEXT': {
97 'client': {
98 'clientName': 'WEB_REMIX',
99 'clientVersion': '1.20220727.01.00',
100 }
101 },
102 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
103 },
104 'web_creator': {
105 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
106 'INNERTUBE_CONTEXT': {
107 'client': {
108 'clientName': 'WEB_CREATOR',
109 'clientVersion': '1.20220726.00.00',
110 }
111 },
112 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
113 },
114 'android': {
115 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
116 'INNERTUBE_CONTEXT': {
117 'client': {
118 'clientName': 'ANDROID',
119 'clientVersion': '17.31.35',
120 'androidSdkVersion': 30,
121 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
122 }
123 },
124 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
125 'REQUIRE_JS_PLAYER': False
126 },
127 'android_embedded': {
128 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
129 'INNERTUBE_CONTEXT': {
130 'client': {
131 'clientName': 'ANDROID_EMBEDDED_PLAYER',
132 'clientVersion': '17.31.35',
133 'androidSdkVersion': 30,
134 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
135 },
136 },
137 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
138 'REQUIRE_JS_PLAYER': False
139 },
140 'android_music': {
141 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
142 'INNERTUBE_CONTEXT': {
143 'client': {
144 'clientName': 'ANDROID_MUSIC',
145 'clientVersion': '5.16.51',
146 'androidSdkVersion': 30,
147 'userAgent': 'com.google.android.apps.youtube.music/5.16.51 (Linux; U; Android 11) gzip'
148 }
149 },
150 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
151 'REQUIRE_JS_PLAYER': False
152 },
153 'android_creator': {
154 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
155 'INNERTUBE_CONTEXT': {
156 'client': {
157 'clientName': 'ANDROID_CREATOR',
158 'clientVersion': '22.30.100',
159 'androidSdkVersion': 30,
160 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip'
161 },
162 },
163 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
164 'REQUIRE_JS_PLAYER': False
165 },
166 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
167 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
168 'ios': {
169 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
170 'INNERTUBE_CONTEXT': {
171 'client': {
172 'clientName': 'IOS',
173 'clientVersion': '17.33.2',
174 'deviceModel': 'iPhone14,3',
175 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
176 }
177 },
178 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
179 'REQUIRE_JS_PLAYER': False
180 },
181 'ios_embedded': {
182 'INNERTUBE_CONTEXT': {
183 'client': {
184 'clientName': 'IOS_MESSAGES_EXTENSION',
185 'clientVersion': '17.33.2',
186 'deviceModel': 'iPhone14,3',
187 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
188 },
189 },
190 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
191 'REQUIRE_JS_PLAYER': False
192 },
193 'ios_music': {
194 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
195 'INNERTUBE_CONTEXT': {
196 'client': {
197 'clientName': 'IOS_MUSIC',
198 'clientVersion': '5.21',
199 'deviceModel': 'iPhone14,3',
200 'userAgent': 'com.google.ios.youtubemusic/5.21 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
201 },
202 },
203 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
204 'REQUIRE_JS_PLAYER': False
205 },
206 'ios_creator': {
207 'INNERTUBE_CONTEXT': {
208 'client': {
209 'clientName': 'IOS_CREATOR',
210 'clientVersion': '22.33.101',
211 'deviceModel': 'iPhone14,3',
212 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
213 },
214 },
215 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
216 'REQUIRE_JS_PLAYER': False
217 },
218 # mweb has 'ultralow' formats
219 # See: https://github.com/yt-dlp/yt-dlp/pull/557
220 'mweb': {
221 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
222 'INNERTUBE_CONTEXT': {
223 'client': {
224 'clientName': 'MWEB',
225 'clientVersion': '2.20220801.00.00',
226 }
227 },
228 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
229 },
230 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
231 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
232 'tv_embedded': {
233 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
234 'INNERTUBE_CONTEXT': {
235 'client': {
236 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
237 'clientVersion': '2.0',
238 },
239 },
240 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
241 },
242 }
243
244
245 def _split_innertube_client(client_name):
246 variant, *base = client_name.rsplit('.', 1)
247 if base:
248 return variant, base[0], variant
249 base, *variant = client_name.split('_', 1)
250 return client_name, base, variant[0] if variant else None
251
252
253 def short_client_name(client_name):
254 main, *parts = _split_innertube_client(client_name)[0].replace('embedscreen', 'e_s').split('_')
255 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
256
257
258 def build_innertube_clients():
259 THIRD_PARTY = {
260 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
261 }
262 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
263 priority = qualities(BASE_CLIENTS[::-1])
264
265 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
266 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
267 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
268 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
269 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
270
271 _, base_client, variant = _split_innertube_client(client)
272 ytcfg['priority'] = 10 * priority(base_client)
273
274 if not variant:
275 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
276 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
277 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
278 embedscreen['priority'] -= 3
279 elif variant == 'embedded':
280 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
281 ytcfg['priority'] -= 2
282 else:
283 ytcfg['priority'] -= 3
284
285
286 build_innertube_clients()
287
288
289 class BadgeType(enum.Enum):
290 AVAILABILITY_UNLISTED = enum.auto()
291 AVAILABILITY_PRIVATE = enum.auto()
292 AVAILABILITY_PUBLIC = enum.auto()
293 AVAILABILITY_PREMIUM = enum.auto()
294 AVAILABILITY_SUBSCRIPTION = enum.auto()
295 LIVE_NOW = enum.auto()
296
297
298 class YoutubeBaseInfoExtractor(InfoExtractor):
299 """Provide base functions for Youtube extractors"""
300
301 _RESERVED_NAMES = (
302 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
303 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
304 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
305 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
306
307 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
308
309 # _NETRC_MACHINE = 'youtube'
310
311 # If True it will raise an error if no login info is provided
312 _LOGIN_REQUIRED = False
313
314 _INVIDIOUS_SITES = (
315 # invidious-redirect websites
316 r'(?:www\.)?redirect\.invidious\.io',
317 r'(?:(?:www|dev)\.)?invidio\.us',
318 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
319 r'(?:www\.)?invidious\.pussthecat\.org',
320 r'(?:www\.)?invidious\.zee\.li',
321 r'(?:www\.)?invidious\.ethibox\.fr',
322 r'(?:www\.)?iv\.ggtyler\.dev',
323 r'(?:www\.)?inv\.vern\.i2p',
324 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
325 r'(?:www\.)?inv\.riverside\.rocks',
326 r'(?:www\.)?invidious\.silur\.me',
327 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
328 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
329 r'(?:www\.)?invidious\.slipfox\.xyz',
330 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
331 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
332 r'(?:www\.)?invidious\.tiekoetter\.com',
333 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
334 r'(?:www\.)?invidious\.nerdvpn\.de',
335 r'(?:www\.)?invidious\.weblibre\.org',
336 r'(?:www\.)?inv\.odyssey346\.dev',
337 r'(?:www\.)?invidious\.dhusch\.de',
338 r'(?:www\.)?iv\.melmac\.space',
339 r'(?:www\.)?watch\.thekitty\.zone',
340 r'(?:www\.)?invidious\.privacydev\.net',
341 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
342 r'(?:www\.)?invidious\.drivet\.xyz',
343 r'(?:www\.)?vid\.priv\.au',
344 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
345 r'(?:www\.)?inv\.vern\.cc',
346 r'(?:www\.)?invidious\.esmailelbob\.xyz',
347 r'(?:www\.)?invidious\.sethforprivacy\.com',
348 r'(?:www\.)?yt\.oelrichsgarcia\.de',
349 r'(?:www\.)?yt\.artemislena\.eu',
350 r'(?:www\.)?invidious\.flokinet\.to',
351 r'(?:www\.)?invidious\.baczek\.me',
352 r'(?:www\.)?y\.com\.sb',
353 r'(?:www\.)?invidious\.epicsite\.xyz',
354 r'(?:www\.)?invidious\.lidarshield\.cloud',
355 r'(?:www\.)?yt\.funami\.tech',
356 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
357 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
358 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
359 # youtube-dl invidious instances list
360 r'(?:(?:www|no)\.)?invidiou\.sh',
361 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
362 r'(?:www\.)?invidious\.kabi\.tk',
363 r'(?:www\.)?invidious\.mastodon\.host',
364 r'(?:www\.)?invidious\.zapashcanon\.fr',
365 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
366 r'(?:www\.)?invidious\.tinfoil-hat\.net',
367 r'(?:www\.)?invidious\.himiko\.cloud',
368 r'(?:www\.)?invidious\.reallyancient\.tech',
369 r'(?:www\.)?invidious\.tube',
370 r'(?:www\.)?invidiou\.site',
371 r'(?:www\.)?invidious\.site',
372 r'(?:www\.)?invidious\.xyz',
373 r'(?:www\.)?invidious\.nixnet\.xyz',
374 r'(?:www\.)?invidious\.048596\.xyz',
375 r'(?:www\.)?invidious\.drycat\.fr',
376 r'(?:www\.)?inv\.skyn3t\.in',
377 r'(?:www\.)?tube\.poal\.co',
378 r'(?:www\.)?tube\.connect\.cafe',
379 r'(?:www\.)?vid\.wxzm\.sx',
380 r'(?:www\.)?vid\.mint\.lgbt',
381 r'(?:www\.)?vid\.puffyan\.us',
382 r'(?:www\.)?yewtu\.be',
383 r'(?:www\.)?yt\.elukerio\.org',
384 r'(?:www\.)?yt\.lelux\.fi',
385 r'(?:www\.)?invidious\.ggc-project\.de',
386 r'(?:www\.)?yt\.maisputain\.ovh',
387 r'(?:www\.)?ytprivate\.com',
388 r'(?:www\.)?invidious\.13ad\.de',
389 r'(?:www\.)?invidious\.toot\.koeln',
390 r'(?:www\.)?invidious\.fdn\.fr',
391 r'(?:www\.)?watch\.nettohikari\.com',
392 r'(?:www\.)?invidious\.namazso\.eu',
393 r'(?:www\.)?invidious\.silkky\.cloud',
394 r'(?:www\.)?invidious\.exonip\.de',
395 r'(?:www\.)?invidious\.riverside\.rocks',
396 r'(?:www\.)?invidious\.blamefran\.net',
397 r'(?:www\.)?invidious\.moomoo\.de',
398 r'(?:www\.)?ytb\.trom\.tf',
399 r'(?:www\.)?yt\.cyberhost\.uk',
400 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
401 r'(?:www\.)?qklhadlycap4cnod\.onion',
402 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
403 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
404 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
405 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
406 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
407 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
408 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
409 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
410 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
411 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
412 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
413 r'(?:www\.)?piped\.kavin\.rocks',
414 r'(?:www\.)?piped\.tokhmi\.xyz',
415 r'(?:www\.)?piped\.syncpundit\.io',
416 r'(?:www\.)?piped\.mha\.fi',
417 r'(?:www\.)?watch\.whatever\.social',
418 r'(?:www\.)?piped\.garudalinux\.org',
419 r'(?:www\.)?piped\.rivo\.lol',
420 r'(?:www\.)?piped-libre\.kavin\.rocks',
421 r'(?:www\.)?yt\.jae\.fi',
422 r'(?:www\.)?piped\.mint\.lgbt',
423 r'(?:www\.)?il\.ax',
424 r'(?:www\.)?piped\.esmailelbob\.xyz',
425 r'(?:www\.)?piped\.projectsegfau\.lt',
426 r'(?:www\.)?piped\.privacydev\.net',
427 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
428 r'(?:www\.)?piped\.smnz\.de',
429 r'(?:www\.)?piped\.adminforge\.de',
430 r'(?:www\.)?watch\.whatevertinfoil\.de',
431 r'(?:www\.)?piped\.qdi\.fi',
432 r'(?:www\.)?piped\.video',
433 r'(?:www\.)?piped\.aeong\.one',
434 r'(?:www\.)?piped\.moomoo\.me',
435 r'(?:www\.)?piped\.chauvet\.pro',
436 r'(?:www\.)?watch\.leptons\.xyz',
437 r'(?:www\.)?pd\.vern\.cc',
438 r'(?:www\.)?piped\.hostux\.net',
439 r'(?:www\.)?piped\.lunar\.icu',
440 # Hyperpipe instances from https://hyperpipe.codeberg.page/
441 r'(?:www\.)?hyperpipe\.surge\.sh',
442 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
443 r'(?:www\.)?listen\.whatever\.social',
444 r'(?:www\.)?music\.adminforge\.de',
445 )
446
447 # extracted from account/account_menu ep
448 # XXX: These are the supported YouTube UI and API languages,
449 # which is slightly different from languages supported for translation in YouTube studio
450 _SUPPORTED_LANG_CODES = [
451 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
452 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
453 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
454 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
455 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
456 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
457 ]
458
459 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
460
461 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
462 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
463
464 def ucid_or_none(self, ucid):
465 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
466
467 def handle_or_none(self, handle):
468 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
469
470 def handle_from_url(self, url):
471 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
472 url, 'channel handle', default=None)
473
474 def ucid_from_url(self, url):
475 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
476 url, 'channel id', default=None)
477
478 @functools.cached_property
479 def _preferred_lang(self):
480 """
481 Returns a language code supported by YouTube for the user preferred language.
482 Returns None if no preferred language set.
483 """
484 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
485 if not preferred_lang:
486 return
487 if preferred_lang not in self._SUPPORTED_LANG_CODES:
488 raise ExtractorError(
489 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
490 expected=True)
491 elif preferred_lang != 'en':
492 self.report_warning(
493 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
494 return preferred_lang
495
496 def _initialize_consent(self):
497 cookies = self._get_cookies('https://www.youtube.com/')
498 if cookies.get('__Secure-3PSID'):
499 return
500 consent_id = None
501 consent = cookies.get('CONSENT')
502 if consent:
503 if 'YES' in consent.value:
504 return
505 consent_id = self._search_regex(
506 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
507 if not consent_id:
508 consent_id = random.randint(100, 999)
509 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
510
511 def _initialize_pref(self):
512 cookies = self._get_cookies('https://www.youtube.com/')
513 pref_cookie = cookies.get('PREF')
514 pref = {}
515 if pref_cookie:
516 try:
517 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
518 except ValueError:
519 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
520 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
521 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
522
523 def _real_initialize(self):
524 self._initialize_pref()
525 self._initialize_consent()
526 self._check_login_required()
527
528 def _check_login_required(self):
529 if self._LOGIN_REQUIRED and not self._cookies_passed:
530 self.raise_login_required('Login details are needed to download this content', method='cookies')
531
532 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
533 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
534
535 def _get_default_ytcfg(self, client='web'):
536 return copy.deepcopy(INNERTUBE_CLIENTS[client])
537
538 def _get_innertube_host(self, client='web'):
539 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
540
541 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
542 # try_get but with fallback to default ytcfg client values when present
543 _func = lambda y: try_get(y, getter, expected_type)
544 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
545
546 def _extract_client_name(self, ytcfg, default_client='web'):
547 return self._ytcfg_get_safe(
548 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
549 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
550
551 def _extract_client_version(self, ytcfg, default_client='web'):
552 return self._ytcfg_get_safe(
553 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
554 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
555
556 def _select_api_hostname(self, req_api_hostname, default_client=None):
557 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
558 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
559
560 def _extract_api_key(self, ytcfg=None, default_client='web'):
561 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
562
563 def _extract_context(self, ytcfg=None, default_client='web'):
564 context = get_first(
565 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
566 # Enforce language and tz for extraction
567 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
568 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
569 return context
570
571 _SAPISID = None
572
573 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
574 time_now = round(time.time())
575 if self._SAPISID is None:
576 yt_cookies = self._get_cookies('https://www.youtube.com')
577 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
578 # See: https://github.com/yt-dlp/yt-dlp/issues/393
579 sapisid_cookie = dict_get(
580 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
581 if sapisid_cookie and sapisid_cookie.value:
582 self._SAPISID = sapisid_cookie.value
583 self.write_debug('Extracted SAPISID cookie')
584 # SAPISID cookie is required if not already present
585 if not yt_cookies.get('SAPISID'):
586 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
587 self._set_cookie(
588 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
589 else:
590 self._SAPISID = False
591 if not self._SAPISID:
592 return None
593 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
594 sapisidhash = hashlib.sha1(
595 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
596 return f'SAPISIDHASH {time_now}_{sapisidhash}'
597
598 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
599 note='Downloading API JSON', errnote='Unable to download API page',
600 context=None, api_key=None, api_hostname=None, default_client='web'):
601
602 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
603 data.update(query)
604 real_headers = self.generate_api_headers(default_client=default_client)
605 real_headers.update({'content-type': 'application/json'})
606 if headers:
607 real_headers.update(headers)
608 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
609 or api_key or self._extract_api_key(default_client=default_client))
610 return self._download_json(
611 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
612 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
613 data=json.dumps(data).encode('utf8'), headers=real_headers,
614 query={'key': api_key, 'prettyPrint': 'false'})
615
616 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
617 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
618
619 @staticmethod
620 def _extract_session_index(*data):
621 """
622 Index of current account in account list.
623 See: https://github.com/yt-dlp/yt-dlp/pull/519
624 """
625 for ytcfg in data:
626 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
627 if session_index is not None:
628 return session_index
629
630 # Deprecated?
631 def _extract_identity_token(self, ytcfg=None, webpage=None):
632 if ytcfg:
633 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
634 if token:
635 return token
636 if webpage:
637 return self._search_regex(
638 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
639 'identity token', default=None, fatal=False)
640
641 @staticmethod
642 def _extract_account_syncid(*args):
643 """
644 Extract syncId required to download private playlists of secondary channels
645 @params response and/or ytcfg
646 """
647 for data in args:
648 # ytcfg includes channel_syncid if on secondary channel
649 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
650 if delegated_sid:
651 return delegated_sid
652 sync_ids = (try_get(
653 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
654 lambda x: x['DATASYNC_ID']), str) or '').split('||')
655 if len(sync_ids) >= 2 and sync_ids[1]:
656 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
657 # and just "user_syncid||" for primary channel. We only want the channel_syncid
658 return sync_ids[0]
659
660 @staticmethod
661 def _extract_visitor_data(*args):
662 """
663 Extracts visitorData from an API response or ytcfg
664 Appears to be used to track session state
665 """
666 return get_first(
667 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
668 expected_type=str)
669
670 @functools.cached_property
671 def is_authenticated(self):
672 return bool(self._generate_sapisidhash_header())
673
674 def extract_ytcfg(self, video_id, webpage):
675 if not webpage:
676 return {}
677 return self._parse_json(
678 self._search_regex(
679 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
680 default='{}'), video_id, fatal=False) or {}
681
682 def generate_api_headers(
683 self, *, ytcfg=None, account_syncid=None, session_index=None,
684 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
685
686 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
687 headers = {
688 'X-YouTube-Client-Name': str(
689 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
690 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
691 'Origin': origin,
692 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
693 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
694 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
695 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
696 }
697 if session_index is None:
698 session_index = self._extract_session_index(ytcfg)
699 if account_syncid or session_index is not None:
700 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
701
702 auth = self._generate_sapisidhash_header(origin)
703 if auth is not None:
704 headers['Authorization'] = auth
705 headers['X-Origin'] = origin
706 return filter_dict(headers)
707
708 def _download_ytcfg(self, client, video_id):
709 url = {
710 'web': 'https://www.youtube.com',
711 'web_music': 'https://music.youtube.com',
712 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
713 }.get(client)
714 if not url:
715 return {}
716 webpage = self._download_webpage(
717 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
718 return self.extract_ytcfg(video_id, webpage) or {}
719
720 @staticmethod
721 def _build_api_continuation_query(continuation, ctp=None):
722 query = {
723 'continuation': continuation
724 }
725 # TODO: Inconsistency with clickTrackingParams.
726 # Currently we have a fixed ctp contained within context (from ytcfg)
727 # and a ctp in root query for continuation.
728 if ctp:
729 query['clickTracking'] = {'clickTrackingParams': ctp}
730 return query
731
732 @classmethod
733 def _extract_next_continuation_data(cls, renderer):
734 next_continuation = try_get(
735 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
736 lambda x: x['continuation']['reloadContinuationData']), dict)
737 if not next_continuation:
738 return
739 continuation = next_continuation.get('continuation')
740 if not continuation:
741 return
742 ctp = next_continuation.get('clickTrackingParams')
743 return cls._build_api_continuation_query(continuation, ctp)
744
745 @classmethod
746 def _extract_continuation_ep_data(cls, continuation_ep: dict):
747 if isinstance(continuation_ep, dict):
748 continuation = try_get(
749 continuation_ep, lambda x: x['continuationCommand']['token'], str)
750 if not continuation:
751 return
752 ctp = continuation_ep.get('clickTrackingParams')
753 return cls._build_api_continuation_query(continuation, ctp)
754
755 @classmethod
756 def _extract_continuation(cls, renderer):
757 next_continuation = cls._extract_next_continuation_data(renderer)
758 if next_continuation:
759 return next_continuation
760
761 return traverse_obj(renderer, (
762 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
763 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
764 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
765
766 @classmethod
767 def _extract_alerts(cls, data):
768 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
769 if not isinstance(alert_dict, dict):
770 continue
771 for alert in alert_dict.values():
772 alert_type = alert.get('type')
773 if not alert_type:
774 continue
775 message = cls._get_text(alert, 'text')
776 if message:
777 yield alert_type, message
778
779 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
780 errors, warnings = [], []
781 for alert_type, alert_message in alerts:
782 if alert_type.lower() == 'error' and fatal:
783 errors.append([alert_type, alert_message])
784 elif alert_message not in self._IGNORED_WARNINGS:
785 warnings.append([alert_type, alert_message])
786
787 for alert_type, alert_message in (warnings + errors[:-1]):
788 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
789 if errors:
790 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
791
792 def _extract_and_report_alerts(self, data, *args, **kwargs):
793 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
794
795 def _extract_badges(self, renderer: dict):
796 privacy_icon_map = {
797 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
798 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
799 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC
800 }
801
802 badge_style_map = {
803 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
804 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
805 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW
806 }
807
808 label_map = {
809 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
810 'private': BadgeType.AVAILABILITY_PRIVATE,
811 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
812 'live': BadgeType.LIVE_NOW,
813 'premium': BadgeType.AVAILABILITY_PREMIUM
814 }
815
816 badges = []
817 for badge in traverse_obj(renderer, ('badges', ..., 'metadataBadgeRenderer')):
818 badge_type = (
819 privacy_icon_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
820 or badge_style_map.get(traverse_obj(badge, 'style'))
821 )
822 if badge_type:
823 badges.append({'type': badge_type})
824 continue
825
826 # fallback, won't work in some languages
827 label = traverse_obj(badge, 'label', expected_type=str, default='')
828 for match, label_badge_type in label_map.items():
829 if match in label.lower():
830 badges.append({'type': badge_type})
831 continue
832
833 return badges
834
835 @staticmethod
836 def _has_badge(badges, badge_type):
837 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
838
839 @staticmethod
840 def _get_text(data, *path_list, max_runs=None):
841 for path in path_list or [None]:
842 if path is None:
843 obj = [data]
844 else:
845 obj = traverse_obj(data, path, default=[])
846 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
847 obj = [obj]
848 for item in obj:
849 text = try_get(item, lambda x: x['simpleText'], str)
850 if text:
851 return text
852 runs = try_get(item, lambda x: x['runs'], list) or []
853 if not runs and isinstance(item, list):
854 runs = item
855
856 runs = runs[:min(len(runs), max_runs or len(runs))]
857 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
858 if text:
859 return text
860
861 def _get_count(self, data, *path_list):
862 count_text = self._get_text(data, *path_list) or ''
863 count = parse_count(count_text)
864 if count is None:
865 count = str_to_int(
866 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
867 return count
868
869 @staticmethod
870 def _extract_thumbnails(data, *path_list):
871 """
872 Extract thumbnails from thumbnails dict
873 @param path_list: path list to level that contains 'thumbnails' key
874 """
875 thumbnails = []
876 for path in path_list or [()]:
877 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...)):
878 thumbnail_url = url_or_none(thumbnail.get('url'))
879 if not thumbnail_url:
880 continue
881 # Sometimes youtube gives a wrong thumbnail URL. See:
882 # https://github.com/yt-dlp/yt-dlp/issues/233
883 # https://github.com/ytdl-org/youtube-dl/issues/28023
884 if 'maxresdefault' in thumbnail_url:
885 thumbnail_url = thumbnail_url.split('?')[0]
886 thumbnails.append({
887 'url': thumbnail_url,
888 'height': int_or_none(thumbnail.get('height')),
889 'width': int_or_none(thumbnail.get('width')),
890 })
891 return thumbnails
892
893 @staticmethod
894 def extract_relative_time(relative_time_text):
895 """
896 Extracts a relative time from string and converts to dt object
897 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today'
898 """
899 mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text)
900 if mobj:
901 start = mobj.group('start')
902 if start:
903 return datetime_from_str(start)
904 try:
905 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
906 except ValueError:
907 return None
908
909 def _parse_time_text(self, text):
910 if not text:
911 return
912 dt = self.extract_relative_time(text)
913 timestamp = None
914 if isinstance(dt, datetime.datetime):
915 timestamp = calendar.timegm(dt.timetuple())
916
917 if timestamp is None:
918 timestamp = (
919 unified_timestamp(text) or unified_timestamp(
920 self._search_regex(
921 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
922 text.lower(), 'time text', default=None)))
923
924 if text and timestamp is None and self._preferred_lang in (None, 'en'):
925 self.report_warning(
926 f'Cannot parse localized time text "{text}"', only_once=True)
927 return timestamp
928
929 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
930 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
931 default_client='web'):
932 for retry in self.RetryManager():
933 try:
934 response = self._call_api(
935 ep=ep, fatal=True, headers=headers,
936 video_id=item_id, query=query, note=note,
937 context=self._extract_context(ytcfg, default_client),
938 api_key=self._extract_api_key(ytcfg, default_client),
939 api_hostname=api_hostname, default_client=default_client)
940 except ExtractorError as e:
941 if not isinstance(e.cause, network_exceptions):
942 return self._error_or_warning(e, fatal=fatal)
943 elif not isinstance(e.cause, urllib.error.HTTPError):
944 retry.error = e
945 continue
946
947 first_bytes = e.cause.read(512)
948 if not is_html(first_bytes):
949 yt_error = try_get(
950 self._parse_json(
951 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
952 lambda x: x['error']['message'], str)
953 if yt_error:
954 self._report_alerts([('ERROR', yt_error)], fatal=False)
955 # Downloading page may result in intermittent 5xx HTTP error
956 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
957 # We also want to catch all other network exceptions since errors in later pages can be troublesome
958 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
959 if e.cause.code not in (403, 429):
960 retry.error = e
961 continue
962 return self._error_or_warning(e, fatal=fatal)
963
964 try:
965 self._extract_and_report_alerts(response, only_once=True)
966 except ExtractorError as e:
967 # YouTube servers may return errors we want to retry on in a 200 OK response
968 # See: https://github.com/yt-dlp/yt-dlp/issues/839
969 if 'unknown error' in e.msg.lower():
970 retry.error = e
971 continue
972 return self._error_or_warning(e, fatal=fatal)
973 # Youtube sometimes sends incomplete data
974 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
975 if not traverse_obj(response, *variadic(check_get_keys)):
976 retry.error = ExtractorError('Incomplete data received', expected=True)
977 continue
978
979 return response
980
981 @staticmethod
982 def is_music_url(url):
983 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
984
985 def _extract_video(self, renderer):
986 video_id = renderer.get('videoId')
987
988 reel_header_renderer = traverse_obj(renderer, (
989 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
990 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
991
992 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
993 description = self._get_text(renderer, 'descriptionSnippet')
994
995 duration = int_or_none(renderer.get('lengthSeconds'))
996 if duration is None:
997 duration = parse_duration(self._get_text(
998 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
999 if duration is None:
1000 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1001 duration = parse_duration(self._search_regex(
1002 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1003 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1004 video_id, default=None, group='duration'))
1005
1006 channel_id = traverse_obj(
1007 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1008 expected_type=str, get_all=False)
1009 if not channel_id:
1010 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1011
1012 channel_id = self.ucid_or_none(channel_id)
1013
1014 overlay_style = traverse_obj(
1015 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1016 get_all=False, expected_type=str)
1017 badges = self._extract_badges(renderer)
1018
1019 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1020 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1021 expected_type=str)) or ''
1022 url = f'https://www.youtube.com/watch?v={video_id}'
1023 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1024 url = f'https://www.youtube.com/shorts/{video_id}'
1025
1026 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1027 or self._get_text(reel_header_renderer, 'timestampText') or '')
1028 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1029
1030 live_status = (
1031 'is_upcoming' if scheduled_timestamp is not None
1032 else 'was_live' if 'streamed' in time_text.lower()
1033 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1034 else None)
1035
1036 # videoInfo is a string like '50K views • 10 years ago'.
1037 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1038 view_count = (0 if 'no views' in view_count_text.lower()
1039 else self._get_count({'simpleText': view_count_text}))
1040 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1041
1042 return {
1043 '_type': 'url',
1044 'ie_key': YoutubeIE.ie_key(),
1045 'id': video_id,
1046 'url': url,
1047 'title': title,
1048 'description': description,
1049 'duration': duration,
1050 'channel_id': channel_id,
1051 'channel': (self._get_text(renderer, 'ownerText', 'shortBylineText')
1052 or self._get_text(reel_header_renderer, 'channelTitleText')),
1053 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1054 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1055 'timestamp': (self._parse_time_text(time_text)
1056 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1057 else None),
1058 'release_timestamp': scheduled_timestamp,
1059 'availability':
1060 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1061 else self._availability(
1062 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1063 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1064 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1065 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1066 view_count_field: view_count,
1067 'live_status': live_status
1068 }
1069
1070
1071 class YoutubeIE(YoutubeBaseInfoExtractor):
1072 IE_DESC = 'YouTube'
1073 _VALID_URL = r"""(?x)^
1074 (
1075 (?:https?://|//) # http(s):// or protocol-independent URL
1076 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1077 (?:www\.)?deturl\.com/www\.youtube\.com|
1078 (?:www\.)?pwnyoutube\.com|
1079 (?:www\.)?hooktube\.com|
1080 (?:www\.)?yourepeat\.com|
1081 tube\.majestyc\.net|
1082 %(invidious)s|
1083 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1084 (?:.*?\#/)? # handle anchor (#/) redirect urls
1085 (?: # the various things that can precede the ID:
1086 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1087 |(?: # or the v= param in all its forms
1088 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1089 (?:\?|\#!?) # the params delimiter ? or # or #!
1090 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1091 v=
1092 )
1093 ))
1094 |(?:
1095 youtu\.be| # just youtu.be/xxxx
1096 vid\.plus| # or vid.plus/xxxx
1097 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1098 %(invidious)s
1099 )/
1100 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1101 )
1102 )? # all until now is optional -> you can pass the naked ID
1103 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1104 (?(1).+)? # if we found the ID, everything can follow
1105 (?:\#|$)""" % {
1106 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1107 }
1108 _EMBED_REGEX = [
1109 r'''(?x)
1110 (?:
1111 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1112 data-video-url=|
1113 <embed[^>]+?src=|
1114 embedSWF\(?:\s*|
1115 <object[^>]+data=|
1116 new\s+SWFObject\(
1117 )
1118 (["\'])
1119 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1120 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1121 \1''',
1122 # https://wordpress.org/plugins/lazy-load-for-videos/
1123 r'''(?xs)
1124 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1125 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1126 ]
1127 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1128
1129 _PLAYER_INFO_RE = (
1130 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1131 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$',
1132 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1133 )
1134 _formats = {
1135 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1136 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1137 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1138 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1139 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1140 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1141 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1142 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1143 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1144 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1145 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1146 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1147 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1148 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1149 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1150 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1151 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1152 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1153
1154
1155 # 3D videos
1156 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1157 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1158 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1159 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1160 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1161 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1162 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1163
1164 # Apple HTTP Live Streaming
1165 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1166 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1167 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1168 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1169 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1170 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1171 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1172 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1173
1174 # DASH mp4 video
1175 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1176 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1177 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1178 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1179 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1180 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1181 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1182 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1183 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1184 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1185 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1186 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1187
1188 # Dash mp4 audio
1189 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1190 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1191 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1192 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1193 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1194 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1195 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1196
1197 # Dash webm
1198 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1199 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1200 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1201 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1202 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1203 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1204 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1205 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1206 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1207 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1208 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1209 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1210 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1211 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1212 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1213 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1214 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1215 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1216 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1217 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1218 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1219 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1220
1221 # Dash webm audio
1222 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1223 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1224
1225 # Dash webm audio with opus inside
1226 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1227 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1228 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1229
1230 # RTMP (unnamed)
1231 '_rtmp': {'protocol': 'rtmp'},
1232
1233 # av01 video only formats sometimes served with "unknown" codecs
1234 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1235 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1236 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1237 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1238 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1239 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1240 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1241 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1242 }
1243 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1244
1245 _GEO_BYPASS = False
1246
1247 IE_NAME = 'youtube'
1248 _TESTS = [
1249 {
1250 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1251 'info_dict': {
1252 'id': 'BaW_jenozKc',
1253 'ext': 'mp4',
1254 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1255 'channel': 'Philipp Hagemeister',
1256 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1257 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1258 'upload_date': '20121002',
1259 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1260 'categories': ['Science & Technology'],
1261 'tags': ['youtube-dl'],
1262 'duration': 10,
1263 'view_count': int,
1264 'like_count': int,
1265 'availability': 'public',
1266 'playable_in_embed': True,
1267 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1268 'live_status': 'not_live',
1269 'age_limit': 0,
1270 'start_time': 1,
1271 'end_time': 9,
1272 'comment_count': int,
1273 'channel_follower_count': int,
1274 'uploader': 'Philipp Hagemeister',
1275 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1276 'uploader_id': '@PhilippHagemeister',
1277 }
1278 },
1279 {
1280 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1281 'note': 'Embed-only video (#1746)',
1282 'info_dict': {
1283 'id': 'yZIXLfi8CZQ',
1284 'ext': 'mp4',
1285 'upload_date': '20120608',
1286 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1287 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1288 'age_limit': 18,
1289 },
1290 'skip': 'Private video',
1291 },
1292 {
1293 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1294 'note': 'Use the first video ID in the URL',
1295 'info_dict': {
1296 'id': 'BaW_jenozKc',
1297 'ext': 'mp4',
1298 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1299 'channel': 'Philipp Hagemeister',
1300 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1301 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1302 'upload_date': '20121002',
1303 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1304 'categories': ['Science & Technology'],
1305 'tags': ['youtube-dl'],
1306 'duration': 10,
1307 'view_count': int,
1308 'like_count': int,
1309 'availability': 'public',
1310 'playable_in_embed': True,
1311 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1312 'live_status': 'not_live',
1313 'age_limit': 0,
1314 'comment_count': int,
1315 'channel_follower_count': int,
1316 'uploader': 'Philipp Hagemeister',
1317 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1318 'uploader_id': '@PhilippHagemeister',
1319 },
1320 'params': {
1321 'skip_download': True,
1322 },
1323 },
1324 {
1325 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1326 'note': '256k DASH audio (format 141) via DASH manifest',
1327 'info_dict': {
1328 'id': 'a9LDPn-MO4I',
1329 'ext': 'm4a',
1330 'upload_date': '20121002',
1331 'description': '',
1332 'title': 'UHDTV TEST 8K VIDEO.mp4'
1333 },
1334 'params': {
1335 'youtube_include_dash_manifest': True,
1336 'format': '141',
1337 },
1338 'skip': 'format 141 not served anymore',
1339 },
1340 # DASH manifest with encrypted signature
1341 {
1342 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1343 'info_dict': {
1344 'id': 'IB3lcPjvWLA',
1345 'ext': 'm4a',
1346 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1347 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1348 'duration': 244,
1349 'upload_date': '20131011',
1350 'abr': 129.495,
1351 'like_count': int,
1352 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1353 'playable_in_embed': True,
1354 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1355 'view_count': int,
1356 'track': 'The Spark',
1357 'live_status': 'not_live',
1358 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1359 'channel': 'Afrojack',
1360 'tags': 'count:19',
1361 'availability': 'public',
1362 'categories': ['Music'],
1363 'age_limit': 0,
1364 'alt_title': 'The Spark',
1365 'channel_follower_count': int,
1366 'uploader': 'Afrojack',
1367 'uploader_url': 'https://www.youtube.com/@Afrojack',
1368 'uploader_id': '@Afrojack',
1369 },
1370 'params': {
1371 'youtube_include_dash_manifest': True,
1372 'format': '141/bestaudio[ext=m4a]',
1373 },
1374 },
1375 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1376 {
1377 'note': 'Embed allowed age-gate video',
1378 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1379 'info_dict': {
1380 'id': 'HtVdAasjOgU',
1381 'ext': 'mp4',
1382 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1383 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1384 'duration': 142,
1385 'upload_date': '20140605',
1386 'age_limit': 18,
1387 'categories': ['Gaming'],
1388 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1389 'availability': 'needs_auth',
1390 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1391 'like_count': int,
1392 'channel': 'The Witcher',
1393 'live_status': 'not_live',
1394 'tags': 'count:17',
1395 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1396 'playable_in_embed': True,
1397 'view_count': int,
1398 'channel_follower_count': int,
1399 'uploader': 'The Witcher',
1400 'uploader_url': 'https://www.youtube.com/@thewitcher',
1401 'uploader_id': '@thewitcher',
1402 },
1403 },
1404 {
1405 'note': 'Age-gate video with embed allowed in public site',
1406 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1407 'info_dict': {
1408 'id': 'HsUATh_Nc2U',
1409 'ext': 'mp4',
1410 'title': 'Godzilla 2 (Official Video)',
1411 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1412 'upload_date': '20200408',
1413 'age_limit': 18,
1414 'availability': 'needs_auth',
1415 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1416 'channel': 'FlyingKitty',
1417 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1418 'view_count': int,
1419 'categories': ['Entertainment'],
1420 'live_status': 'not_live',
1421 'tags': ['Flyingkitty', 'godzilla 2'],
1422 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1423 'like_count': int,
1424 'duration': 177,
1425 'playable_in_embed': True,
1426 'channel_follower_count': int,
1427 'uploader': 'FlyingKitty',
1428 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1429 'uploader_id': '@FlyingKitty900',
1430 },
1431 },
1432 {
1433 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1434 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1435 'info_dict': {
1436 'id': 'Tq92D6wQ1mg',
1437 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1438 'ext': 'mp4',
1439 'upload_date': '20191228',
1440 'description': 'md5:17eccca93a786d51bc67646756894066',
1441 'age_limit': 18,
1442 'like_count': int,
1443 'availability': 'needs_auth',
1444 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1445 'view_count': int,
1446 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1447 'channel': 'Projekt Melody',
1448 'live_status': 'not_live',
1449 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1450 'playable_in_embed': True,
1451 'categories': ['Entertainment'],
1452 'duration': 106,
1453 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1454 'comment_count': int,
1455 'channel_follower_count': int,
1456 'uploader': 'Projekt Melody',
1457 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1458 'uploader_id': '@ProjektMelody',
1459 },
1460 },
1461 {
1462 'note': 'Non-Agegated non-embeddable video',
1463 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1464 'info_dict': {
1465 'id': 'MeJVWBSsPAY',
1466 'ext': 'mp4',
1467 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1468 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1469 'upload_date': '20130730',
1470 'track': 'Such mich find mich',
1471 'age_limit': 0,
1472 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1473 'like_count': int,
1474 'playable_in_embed': False,
1475 'creator': 'OOMPH!',
1476 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1477 'view_count': int,
1478 'alt_title': 'Such mich find mich',
1479 'duration': 210,
1480 'channel': 'Herr Lurik',
1481 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1482 'categories': ['Music'],
1483 'availability': 'public',
1484 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1485 'live_status': 'not_live',
1486 'artist': 'OOMPH!',
1487 'channel_follower_count': int,
1488 'uploader': 'Herr Lurik',
1489 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1490 'uploader_id': '@HerrLurik',
1491 },
1492 },
1493 {
1494 'note': 'Non-bypassable age-gated video',
1495 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1496 'only_matching': True,
1497 },
1498 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1499 # YouTube Red ad is not captured for creator
1500 {
1501 'url': '__2ABJjxzNo',
1502 'info_dict': {
1503 'id': '__2ABJjxzNo',
1504 'ext': 'mp4',
1505 'duration': 266,
1506 'upload_date': '20100430',
1507 'creator': 'deadmau5',
1508 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1509 'title': 'Deadmau5 - Some Chords (HD)',
1510 'alt_title': 'Some Chords',
1511 'availability': 'public',
1512 'tags': 'count:14',
1513 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1514 'view_count': int,
1515 'live_status': 'not_live',
1516 'channel': 'deadmau5',
1517 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1518 'like_count': int,
1519 'track': 'Some Chords',
1520 'artist': 'deadmau5',
1521 'playable_in_embed': True,
1522 'age_limit': 0,
1523 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1524 'categories': ['Music'],
1525 'album': 'Some Chords',
1526 'channel_follower_count': int,
1527 'uploader': 'deadmau5',
1528 'uploader_url': 'https://www.youtube.com/@deadmau5',
1529 'uploader_id': '@deadmau5',
1530 },
1531 'expected_warnings': [
1532 'DASH manifest missing',
1533 ]
1534 },
1535 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1536 {
1537 'url': 'lqQg6PlCWgI',
1538 'info_dict': {
1539 'id': 'lqQg6PlCWgI',
1540 'ext': 'mp4',
1541 'duration': 6085,
1542 'upload_date': '20150827',
1543 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1544 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1545 'like_count': int,
1546 'release_timestamp': 1343767800,
1547 'playable_in_embed': True,
1548 'categories': ['Sports'],
1549 'release_date': '20120731',
1550 'channel': 'Olympics',
1551 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1552 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1553 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1554 'age_limit': 0,
1555 'availability': 'public',
1556 'live_status': 'was_live',
1557 'view_count': int,
1558 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1559 'channel_follower_count': int,
1560 'uploader': 'Olympics',
1561 'uploader_url': 'https://www.youtube.com/@Olympics',
1562 'uploader_id': '@Olympics',
1563 },
1564 'params': {
1565 'skip_download': 'requires avconv',
1566 }
1567 },
1568 # Non-square pixels
1569 {
1570 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1571 'info_dict': {
1572 'id': '_b-2C3KPAM0',
1573 'ext': 'mp4',
1574 'stretched_ratio': 16 / 9.,
1575 'duration': 85,
1576 'upload_date': '20110310',
1577 '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',
1578 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1579 'playable_in_embed': True,
1580 'channel': '孫ᄋᄅ',
1581 'age_limit': 0,
1582 'tags': 'count:11',
1583 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1584 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1585 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1586 'view_count': int,
1587 'categories': ['People & Blogs'],
1588 'like_count': int,
1589 'live_status': 'not_live',
1590 'availability': 'unlisted',
1591 'comment_count': int,
1592 'channel_follower_count': int,
1593 'uploader': '孫ᄋᄅ',
1594 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1595 'uploader_id': '@AllenMeow',
1596 },
1597 },
1598 # url_encoded_fmt_stream_map is empty string
1599 {
1600 'url': 'qEJwOuvDf7I',
1601 'info_dict': {
1602 'id': 'qEJwOuvDf7I',
1603 'ext': 'webm',
1604 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1605 'description': '',
1606 'upload_date': '20150404',
1607 },
1608 'params': {
1609 'skip_download': 'requires avconv',
1610 },
1611 'skip': 'This live event has ended.',
1612 },
1613 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1614 {
1615 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1616 'info_dict': {
1617 'id': 'FIl7x6_3R5Y',
1618 'ext': 'webm',
1619 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1620 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1621 'duration': 220,
1622 'upload_date': '20150625',
1623 'formats': 'mincount:31',
1624 },
1625 'skip': 'not actual anymore',
1626 },
1627 # DASH manifest with segment_list
1628 {
1629 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1630 'md5': '8ce563a1d667b599d21064e982ab9e31',
1631 'info_dict': {
1632 'id': 'CsmdDsKjzN8',
1633 'ext': 'mp4',
1634 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1635 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1636 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1637 },
1638 'params': {
1639 'youtube_include_dash_manifest': True,
1640 'format': '135', # bestvideo
1641 },
1642 'skip': 'This live event has ended.',
1643 },
1644 {
1645 # Multifeed videos (multiple cameras), URL can be of any Camera
1646 # TODO: fix multifeed titles
1647 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1648 'info_dict': {
1649 'id': 'zaPI8MvL8pg',
1650 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1651 'description': 'md5:563ccbc698b39298481ca3c571169519',
1652 },
1653 'playlist': [{
1654 'info_dict': {
1655 'id': 'j5yGuxZ8lLU',
1656 'ext': 'mp4',
1657 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1658 'description': 'md5:563ccbc698b39298481ca3c571169519',
1659 'duration': 10120,
1660 'channel_follower_count': int,
1661 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1662 'availability': 'public',
1663 'playable_in_embed': True,
1664 'upload_date': '20131105',
1665 'categories': ['Gaming'],
1666 'live_status': 'was_live',
1667 'tags': 'count:24',
1668 'release_timestamp': 1383701910,
1669 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1670 'comment_count': int,
1671 'age_limit': 0,
1672 'like_count': int,
1673 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1674 'channel': 'WiiLikeToPlay',
1675 'view_count': int,
1676 'release_date': '20131106',
1677 'uploader': 'WiiLikeToPlay',
1678 'uploader_id': '@WLTP',
1679 'uploader_url': 'https://www.youtube.com/@WLTP',
1680 },
1681 }, {
1682 'info_dict': {
1683 'id': 'zaPI8MvL8pg',
1684 'ext': 'mp4',
1685 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1686 'availability': 'public',
1687 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1688 'channel': 'WiiLikeToPlay',
1689 'channel_follower_count': int,
1690 'description': 'md5:563ccbc698b39298481ca3c571169519',
1691 'duration': 10108,
1692 'age_limit': 0,
1693 'like_count': int,
1694 'tags': 'count:24',
1695 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1696 'release_timestamp': 1383701915,
1697 'comment_count': int,
1698 'upload_date': '20131105',
1699 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1700 'release_date': '20131106',
1701 'playable_in_embed': True,
1702 'live_status': 'was_live',
1703 'categories': ['Gaming'],
1704 'view_count': int,
1705 'uploader': 'WiiLikeToPlay',
1706 'uploader_id': '@WLTP',
1707 'uploader_url': 'https://www.youtube.com/@WLTP',
1708 },
1709 }, {
1710 'info_dict': {
1711 'id': 'R7r3vfO7Hao',
1712 'ext': 'mp4',
1713 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1714 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1715 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1716 'like_count': int,
1717 'availability': 'public',
1718 'playable_in_embed': True,
1719 'upload_date': '20131105',
1720 'description': 'md5:563ccbc698b39298481ca3c571169519',
1721 'channel_follower_count': int,
1722 'tags': 'count:24',
1723 'release_date': '20131106',
1724 'comment_count': int,
1725 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1726 'channel': 'WiiLikeToPlay',
1727 'categories': ['Gaming'],
1728 'release_timestamp': 1383701914,
1729 'live_status': 'was_live',
1730 'age_limit': 0,
1731 'duration': 10128,
1732 'view_count': int,
1733 'uploader': 'WiiLikeToPlay',
1734 'uploader_id': '@WLTP',
1735 'uploader_url': 'https://www.youtube.com/@WLTP',
1736 },
1737 }],
1738 'params': {'skip_download': True},
1739 },
1740 {
1741 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1742 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1743 'info_dict': {
1744 'id': 'gVfLd0zydlo',
1745 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1746 },
1747 'playlist_count': 2,
1748 'skip': 'Not multifeed anymore',
1749 },
1750 {
1751 'url': 'https://vid.plus/FlRa-iH7PGw',
1752 'only_matching': True,
1753 },
1754 {
1755 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1756 'only_matching': True,
1757 },
1758 {
1759 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1760 # Also tests cut-off URL expansion in video description (see
1761 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1762 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1763 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1764 'info_dict': {
1765 'id': 'lsguqyKfVQg',
1766 'ext': 'mp4',
1767 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1768 'alt_title': 'Dark Walk',
1769 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1770 'duration': 133,
1771 'upload_date': '20151119',
1772 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1773 'track': 'Dark Walk',
1774 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1775 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1776 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1777 'categories': ['Film & Animation'],
1778 'view_count': int,
1779 'live_status': 'not_live',
1780 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1781 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1782 'tags': 'count:13',
1783 'availability': 'public',
1784 'channel': 'IronSoulElf',
1785 'playable_in_embed': True,
1786 'like_count': int,
1787 'age_limit': 0,
1788 'channel_follower_count': int
1789 },
1790 'params': {
1791 'skip_download': True,
1792 },
1793 },
1794 {
1795 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1796 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1797 'only_matching': True,
1798 },
1799 {
1800 # Video with yt:stretch=17:0
1801 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1802 'info_dict': {
1803 'id': 'Q39EVAstoRM',
1804 'ext': 'mp4',
1805 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1806 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1807 'upload_date': '20151107',
1808 },
1809 'params': {
1810 'skip_download': True,
1811 },
1812 'skip': 'This video does not exist.',
1813 },
1814 {
1815 # Video with incomplete 'yt:stretch=16:'
1816 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1817 'only_matching': True,
1818 },
1819 {
1820 # Video licensed under Creative Commons
1821 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1822 'info_dict': {
1823 'id': 'M4gD1WSo5mA',
1824 'ext': 'mp4',
1825 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1826 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1827 'duration': 721,
1828 'upload_date': '20150128',
1829 'license': 'Creative Commons Attribution license (reuse allowed)',
1830 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1831 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1832 'like_count': int,
1833 'age_limit': 0,
1834 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1835 'channel': 'The Berkman Klein Center for Internet & Society',
1836 'availability': 'public',
1837 'view_count': int,
1838 'categories': ['Education'],
1839 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1840 'live_status': 'not_live',
1841 'playable_in_embed': True,
1842 'channel_follower_count': int,
1843 'chapters': list,
1844 'uploader': 'The Berkman Klein Center for Internet & Society',
1845 'uploader_id': '@BKCHarvard',
1846 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1847 },
1848 'params': {
1849 'skip_download': True,
1850 },
1851 },
1852 {
1853 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1854 'info_dict': {
1855 'id': 'eQcmzGIKrzg',
1856 'ext': 'mp4',
1857 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1858 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1859 'duration': 4060,
1860 'upload_date': '20151120',
1861 'license': 'Creative Commons Attribution license (reuse allowed)',
1862 'playable_in_embed': True,
1863 'tags': 'count:12',
1864 'like_count': int,
1865 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1866 'age_limit': 0,
1867 'availability': 'public',
1868 'categories': ['News & Politics'],
1869 'channel': 'Bernie Sanders',
1870 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1871 'view_count': int,
1872 'live_status': 'not_live',
1873 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1874 'comment_count': int,
1875 'channel_follower_count': int,
1876 'chapters': list,
1877 'uploader': 'Bernie Sanders',
1878 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1879 'uploader_id': '@BernieSanders',
1880 },
1881 'params': {
1882 'skip_download': True,
1883 },
1884 },
1885 {
1886 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1887 'only_matching': True,
1888 },
1889 {
1890 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1891 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1892 'only_matching': True,
1893 },
1894 {
1895 # Rental video preview
1896 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1897 'info_dict': {
1898 'id': 'uGpuVWrhIzE',
1899 'ext': 'mp4',
1900 'title': 'Piku - Trailer',
1901 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1902 'upload_date': '20150811',
1903 'license': 'Standard YouTube License',
1904 },
1905 'params': {
1906 'skip_download': True,
1907 },
1908 'skip': 'This video is not available.',
1909 },
1910 {
1911 # YouTube Red video with episode data
1912 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1913 'info_dict': {
1914 'id': 'iqKdEhx-dD4',
1915 'ext': 'mp4',
1916 'title': 'Isolation - Mind Field (Ep 1)',
1917 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1918 'duration': 2085,
1919 'upload_date': '20170118',
1920 'series': 'Mind Field',
1921 'season_number': 1,
1922 'episode_number': 1,
1923 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1924 'tags': 'count:12',
1925 'view_count': int,
1926 'availability': 'public',
1927 'age_limit': 0,
1928 'channel': 'Vsauce',
1929 'episode': 'Episode 1',
1930 'categories': ['Entertainment'],
1931 'season': 'Season 1',
1932 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1933 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1934 'like_count': int,
1935 'playable_in_embed': True,
1936 'live_status': 'not_live',
1937 'channel_follower_count': int,
1938 'uploader': 'Vsauce',
1939 'uploader_url': 'https://www.youtube.com/@Vsauce',
1940 'uploader_id': '@Vsauce',
1941 },
1942 'params': {
1943 'skip_download': True,
1944 },
1945 'expected_warnings': [
1946 'Skipping DASH manifest',
1947 ],
1948 },
1949 {
1950 # The following content has been identified by the YouTube community
1951 # as inappropriate or offensive to some audiences.
1952 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1953 'info_dict': {
1954 'id': '6SJNVb0GnPI',
1955 'ext': 'mp4',
1956 'title': 'Race Differences in Intelligence',
1957 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1958 'duration': 965,
1959 'upload_date': '20140124',
1960 },
1961 'params': {
1962 'skip_download': True,
1963 },
1964 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1965 },
1966 {
1967 # itag 212
1968 'url': '1t24XAntNCY',
1969 'only_matching': True,
1970 },
1971 {
1972 # geo restricted to JP
1973 'url': 'sJL6WA-aGkQ',
1974 'only_matching': True,
1975 },
1976 {
1977 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1978 'only_matching': True,
1979 },
1980 {
1981 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1982 'only_matching': True,
1983 },
1984 {
1985 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1986 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1987 'only_matching': True,
1988 },
1989 {
1990 # DRM protected
1991 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1992 'only_matching': True,
1993 },
1994 {
1995 # Video with unsupported adaptive stream type formats
1996 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1997 'info_dict': {
1998 'id': 'Z4Vy8R84T1U',
1999 'ext': 'mp4',
2000 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2001 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2002 'duration': 433,
2003 'upload_date': '20130923',
2004 'formats': 'maxcount:10',
2005 },
2006 'params': {
2007 'skip_download': True,
2008 'youtube_include_dash_manifest': False,
2009 },
2010 'skip': 'not actual anymore',
2011 },
2012 {
2013 # Youtube Music Auto-generated description
2014 # TODO: fix metadata extraction
2015 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2016 'info_dict': {
2017 'id': 'MgNrAu2pzNs',
2018 'ext': 'mp4',
2019 'title': 'Voyeur Girl',
2020 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2021 'upload_date': '20190312',
2022 'artist': 'Stephen',
2023 'track': 'Voyeur Girl',
2024 'album': 'it\'s too much love to know my dear',
2025 'release_date': '20190313',
2026 'release_year': 2019,
2027 'alt_title': 'Voyeur Girl',
2028 'view_count': int,
2029 'playable_in_embed': True,
2030 'like_count': int,
2031 'categories': ['Music'],
2032 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2033 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2034 'uploader': 'Stephen',
2035 'availability': 'public',
2036 'creator': 'Stephen',
2037 'duration': 169,
2038 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2039 'age_limit': 0,
2040 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2041 'tags': 'count:11',
2042 'live_status': 'not_live',
2043 'channel_follower_count': int
2044 },
2045 'params': {
2046 'skip_download': True,
2047 },
2048 },
2049 {
2050 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2051 'only_matching': True,
2052 },
2053 {
2054 # invalid -> valid video id redirection
2055 'url': 'DJztXj2GPfl',
2056 'info_dict': {
2057 'id': 'DJztXj2GPfk',
2058 'ext': 'mp4',
2059 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2060 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2061 'upload_date': '20090125',
2062 'artist': 'Panjabi MC',
2063 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2064 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2065 },
2066 'params': {
2067 'skip_download': True,
2068 },
2069 'skip': 'Video unavailable',
2070 },
2071 {
2072 # empty description results in an empty string
2073 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2074 'info_dict': {
2075 'id': 'x41yOUIvK2k',
2076 'ext': 'mp4',
2077 'title': 'IMG 3456',
2078 'description': '',
2079 'upload_date': '20170613',
2080 'view_count': int,
2081 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2082 'like_count': int,
2083 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2084 'tags': [],
2085 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2086 'availability': 'public',
2087 'age_limit': 0,
2088 'categories': ['Pets & Animals'],
2089 'duration': 7,
2090 'playable_in_embed': True,
2091 'live_status': 'not_live',
2092 'channel': 'l\'Or Vert asbl',
2093 'channel_follower_count': int,
2094 'uploader': 'l\'Or Vert asbl',
2095 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2096 'uploader_id': '@ElevageOrVert',
2097 },
2098 'params': {
2099 'skip_download': True,
2100 },
2101 },
2102 {
2103 # with '};' inside yt initial data (see [1])
2104 # see [2] for an example with '};' inside ytInitialPlayerResponse
2105 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2106 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2107 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2108 'info_dict': {
2109 'id': 'CHqg6qOn4no',
2110 'ext': 'mp4',
2111 'title': 'Part 77 Sort a list of simple types in c#',
2112 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2113 'upload_date': '20130831',
2114 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2115 'like_count': int,
2116 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2117 'live_status': 'not_live',
2118 'categories': ['Education'],
2119 'availability': 'public',
2120 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2121 'tags': 'count:12',
2122 'playable_in_embed': True,
2123 'age_limit': 0,
2124 'view_count': int,
2125 'duration': 522,
2126 'channel': 'kudvenkat',
2127 'comment_count': int,
2128 'channel_follower_count': int,
2129 'chapters': list,
2130 'uploader': 'kudvenkat',
2131 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2132 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2133 },
2134 'params': {
2135 'skip_download': True,
2136 },
2137 },
2138 {
2139 # another example of '};' in ytInitialData
2140 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2141 'only_matching': True,
2142 },
2143 {
2144 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2145 'only_matching': True,
2146 },
2147 {
2148 # https://github.com/ytdl-org/youtube-dl/pull/28094
2149 'url': 'OtqTfy26tG0',
2150 'info_dict': {
2151 'id': 'OtqTfy26tG0',
2152 'ext': 'mp4',
2153 'title': 'Burn Out',
2154 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2155 'upload_date': '20141120',
2156 'artist': 'The Cinematic Orchestra',
2157 'track': 'Burn Out',
2158 'album': 'Every Day',
2159 'like_count': int,
2160 'live_status': 'not_live',
2161 'alt_title': 'Burn Out',
2162 'duration': 614,
2163 'age_limit': 0,
2164 'view_count': int,
2165 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2166 'creator': 'The Cinematic Orchestra',
2167 'channel': 'The Cinematic Orchestra',
2168 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2169 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2170 'availability': 'public',
2171 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2172 'categories': ['Music'],
2173 'playable_in_embed': True,
2174 'channel_follower_count': int,
2175 'uploader': 'The Cinematic Orchestra',
2176 'comment_count': int,
2177 },
2178 'params': {
2179 'skip_download': True,
2180 },
2181 },
2182 {
2183 # controversial video, only works with bpctr when authenticated with cookies
2184 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2185 'only_matching': True,
2186 },
2187 {
2188 # controversial video, requires bpctr/contentCheckOk
2189 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2190 'info_dict': {
2191 'id': 'SZJvDhaSDnc',
2192 'ext': 'mp4',
2193 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2194 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2195 'upload_date': '20140716',
2196 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2197 'duration': 170,
2198 'categories': ['News & Politics'],
2199 'view_count': int,
2200 'channel': 'CBS Mornings',
2201 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2202 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2203 'age_limit': 18,
2204 'availability': 'needs_auth',
2205 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2206 'like_count': int,
2207 'live_status': 'not_live',
2208 'playable_in_embed': True,
2209 'channel_follower_count': int,
2210 'uploader': 'CBS Mornings',
2211 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2212 'uploader_id': '@CBSMornings',
2213 }
2214 },
2215 {
2216 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2217 'url': 'cBvYw8_A0vQ',
2218 'info_dict': {
2219 'id': 'cBvYw8_A0vQ',
2220 'ext': 'mp4',
2221 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2222 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2223 'upload_date': '20201120',
2224 'duration': 1456,
2225 'categories': ['Travel & Events'],
2226 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2227 'view_count': int,
2228 'channel': 'Walk around Japan',
2229 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2230 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2231 'age_limit': 0,
2232 'availability': 'public',
2233 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2234 'live_status': 'not_live',
2235 'playable_in_embed': True,
2236 'channel_follower_count': int,
2237 'uploader': 'Walk around Japan',
2238 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2239 'uploader_id': '@walkaroundjapan7124',
2240 },
2241 'params': {
2242 'skip_download': True,
2243 },
2244 }, {
2245 # Has multiple audio streams
2246 'url': 'WaOKSUlf4TM',
2247 'only_matching': True
2248 }, {
2249 # Requires Premium: has format 141 when requested using YTM url
2250 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2251 'only_matching': True
2252 }, {
2253 # multiple subtitles with same lang_code
2254 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2255 'only_matching': True,
2256 }, {
2257 # Force use android client fallback
2258 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2259 'info_dict': {
2260 'id': 'YOelRv7fMxY',
2261 'title': 'DIGGING A SECRET TUNNEL Part 1',
2262 'ext': '3gp',
2263 'upload_date': '20210624',
2264 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2265 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2266 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2267 'duration': 596,
2268 'categories': ['Entertainment'],
2269 'view_count': int,
2270 'channel': 'colinfurze',
2271 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2272 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2273 'age_limit': 0,
2274 'availability': 'public',
2275 'like_count': int,
2276 'live_status': 'not_live',
2277 'playable_in_embed': True,
2278 'channel_follower_count': int,
2279 'chapters': list,
2280 'uploader': 'colinfurze',
2281 'uploader_url': 'https://www.youtube.com/@colinfurze',
2282 'uploader_id': '@colinfurze',
2283 },
2284 'params': {
2285 'format': '17', # 3gp format available on android
2286 'extractor_args': {'youtube': {'player_client': ['android']}},
2287 },
2288 },
2289 {
2290 # Skip download of additional client configs (remix client config in this case)
2291 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2292 'only_matching': True,
2293 'params': {
2294 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2295 },
2296 }, {
2297 # shorts
2298 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2299 'only_matching': True,
2300 }, {
2301 'note': 'Storyboards',
2302 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2303 'info_dict': {
2304 'id': '5KLPxDtMqe8',
2305 'ext': 'mhtml',
2306 'format_id': 'sb0',
2307 'title': 'Your Brain is Plastic',
2308 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2309 'upload_date': '20140324',
2310 'like_count': int,
2311 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2312 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2313 'view_count': int,
2314 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2315 'playable_in_embed': True,
2316 'tags': 'count:12',
2317 'availability': 'public',
2318 'channel': 'SciShow',
2319 'live_status': 'not_live',
2320 'duration': 248,
2321 'categories': ['Education'],
2322 'age_limit': 0,
2323 'channel_follower_count': int,
2324 'chapters': list,
2325 'uploader': 'SciShow',
2326 'uploader_url': 'https://www.youtube.com/@SciShow',
2327 'uploader_id': '@SciShow',
2328 }, 'params': {'format': 'mhtml', 'skip_download': True}
2329 }, {
2330 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2331 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2332 'info_dict': {
2333 'id': '2NUZ8W2llS4',
2334 'ext': 'mp4',
2335 'title': 'The NP that test your phone performance 🙂',
2336 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2337 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2338 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2339 'duration': 21,
2340 'view_count': int,
2341 'age_limit': 0,
2342 'categories': ['Gaming'],
2343 'tags': 'count:23',
2344 'playable_in_embed': True,
2345 'live_status': 'not_live',
2346 'upload_date': '20220103',
2347 'like_count': int,
2348 'availability': 'public',
2349 'channel': 'Leon Nguyen',
2350 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2351 'comment_count': int,
2352 'channel_follower_count': int,
2353 'uploader': 'Leon Nguyen',
2354 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2355 'uploader_id': '@LeonNguyen',
2356 }
2357 }, {
2358 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2359 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2360 'info_dict': {
2361 'id': '2NUZ8W2llS4',
2362 'ext': 'mp4',
2363 'title': 'The NP that test your phone performance 🙂',
2364 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2365 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2366 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2367 'duration': 21,
2368 'view_count': int,
2369 'age_limit': 0,
2370 'categories': ['Gaming'],
2371 'tags': 'count:23',
2372 'playable_in_embed': True,
2373 'live_status': 'not_live',
2374 'upload_date': '20220102',
2375 'like_count': int,
2376 'availability': 'public',
2377 'channel': 'Leon Nguyen',
2378 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2379 'comment_count': int,
2380 'channel_follower_count': int,
2381 'uploader': 'Leon Nguyen',
2382 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2383 'uploader_id': '@LeonNguyen',
2384 },
2385 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2386 }, {
2387 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2388 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2389 'info_dict': {
2390 'id': 'mzZzzBU6lrM',
2391 'ext': 'mp4',
2392 'title': 'I Met GeorgeNotFound In Real Life...',
2393 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2394 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2395 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2396 'duration': 955,
2397 'view_count': int,
2398 'age_limit': 0,
2399 'categories': ['Entertainment'],
2400 'tags': 'count:26',
2401 'playable_in_embed': True,
2402 'live_status': 'not_live',
2403 'release_timestamp': 1641172509,
2404 'release_date': '20220103',
2405 'upload_date': '20220103',
2406 'like_count': int,
2407 'availability': 'public',
2408 'channel': 'Quackity',
2409 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2410 'channel_follower_count': int,
2411 'uploader': 'Quackity',
2412 'uploader_id': '@Quackity',
2413 'uploader_url': 'https://www.youtube.com/@Quackity',
2414 }
2415 },
2416 { # continuous livestream. Microformat upload date should be preferred.
2417 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2418 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2419 'info_dict': {
2420 'id': 'kgx4WGK0oNU',
2421 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}',
2422 'ext': 'mp4',
2423 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2424 'availability': 'public',
2425 'age_limit': 0,
2426 'release_timestamp': 1637975704,
2427 'upload_date': '20210619',
2428 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2429 'live_status': 'is_live',
2430 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2431 'channel': 'Abao in Tokyo',
2432 'channel_follower_count': int,
2433 'release_date': '20211127',
2434 'tags': 'count:39',
2435 'categories': ['People & Blogs'],
2436 'like_count': int,
2437 'view_count': int,
2438 'playable_in_embed': True,
2439 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2440 'concurrent_view_count': int,
2441 'uploader': 'Abao in Tokyo',
2442 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2443 'uploader_id': '@abaointokyo',
2444 },
2445 'params': {'skip_download': True}
2446 }, {
2447 # Story. Requires specific player params to work.
2448 'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2449 'info_dict': {
2450 'id': 'vv8qTUWmulI',
2451 'ext': 'mp4',
2452 'availability': 'unlisted',
2453 'view_count': int,
2454 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2455 'upload_date': '20220526',
2456 'categories': ['Education'],
2457 'title': 'Story',
2458 'channel': 'IT\'S HISTORY',
2459 'description': '',
2460 'duration': 12,
2461 'playable_in_embed': True,
2462 'age_limit': 0,
2463 'live_status': 'not_live',
2464 'tags': [],
2465 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2466 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2467 },
2468 'skip': 'stories get removed after some period of time',
2469 }, {
2470 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2471 'info_dict': {
2472 'id': 'tjjjtzRLHvA',
2473 'ext': 'mp4',
2474 'title': 'ハッシュタグ無し };if window.ytcsi',
2475 'upload_date': '20220323',
2476 'like_count': int,
2477 'availability': 'unlisted',
2478 'channel': 'Lesmiscore',
2479 'thumbnail': r're:^https?://.*\.jpg',
2480 'age_limit': 0,
2481 'categories': ['Music'],
2482 'view_count': int,
2483 'description': '',
2484 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2485 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2486 'live_status': 'not_live',
2487 'playable_in_embed': True,
2488 'channel_follower_count': int,
2489 'duration': 6,
2490 'tags': [],
2491 'uploader_id': '@lesmiscore',
2492 'uploader': 'Lesmiscore',
2493 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2494 }
2495 }, {
2496 # Prefer primary title+description language metadata by default
2497 # Do not prefer translated description if primary is empty
2498 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2499 'info_dict': {
2500 'id': 'el3E4MbxRqQ',
2501 'ext': 'mp4',
2502 'title': 'dlp test video 2 - primary sv no desc',
2503 'description': '',
2504 'channel': 'cole-dlp-test-acc',
2505 'tags': [],
2506 'view_count': int,
2507 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2508 'like_count': int,
2509 'playable_in_embed': True,
2510 'availability': 'unlisted',
2511 'thumbnail': r're:^https?://.*\.jpg',
2512 'age_limit': 0,
2513 'duration': 5,
2514 'live_status': 'not_live',
2515 'upload_date': '20220908',
2516 'categories': ['People & Blogs'],
2517 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2518 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2519 'uploader_id': '@coletdjnz',
2520 'uploader': 'cole-dlp-test-acc',
2521 },
2522 'params': {'skip_download': True}
2523 }, {
2524 # Extractor argument: prefer translated title+description
2525 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2526 'info_dict': {
2527 'id': 'gHKT4uU8Zng',
2528 'ext': 'mp4',
2529 'channel': 'cole-dlp-test-acc',
2530 'tags': [],
2531 'duration': 5,
2532 'live_status': 'not_live',
2533 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2534 'upload_date': '20220728',
2535 'view_count': int,
2536 'categories': ['People & Blogs'],
2537 'thumbnail': r're:^https?://.*\.jpg',
2538 'title': 'dlp test video title translated (fr)',
2539 'availability': 'public',
2540 'age_limit': 0,
2541 'description': 'dlp test video description translated (fr)',
2542 'playable_in_embed': True,
2543 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2544 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2545 'uploader_id': '@coletdjnz',
2546 'uploader': 'cole-dlp-test-acc',
2547 },
2548 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2549 'expected_warnings': [r'Preferring "fr" translated fields'],
2550 }, {
2551 'note': '6 channel audio',
2552 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2553 'only_matching': True,
2554 }, {
2555 'note': 'Multiple HLS formats with same itag',
2556 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2557 'info_dict': {
2558 'id': 'kX3nB4PpJko',
2559 'ext': 'mp4',
2560 'categories': ['Entertainment'],
2561 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2562 'live_status': 'not_live',
2563 'duration': 937,
2564 'channel_follower_count': int,
2565 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2566 'title': 'Last To Take Hand Off Jet, Keeps It!',
2567 'channel': 'MrBeast',
2568 'playable_in_embed': True,
2569 'view_count': int,
2570 'upload_date': '20221112',
2571 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2572 'age_limit': 0,
2573 'availability': 'public',
2574 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2575 'like_count': int,
2576 'tags': [],
2577 'uploader': 'MrBeast',
2578 'uploader_url': 'https://www.youtube.com/@MrBeast',
2579 'uploader_id': '@MrBeast',
2580 },
2581 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2582 }, {
2583 'note': 'Audio formats with Dynamic Range Compression',
2584 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2585 'info_dict': {
2586 'id': 'Tq92D6wQ1mg',
2587 'ext': 'webm',
2588 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2589 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2590 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2591 'channel_follower_count': int,
2592 'description': 'md5:17eccca93a786d51bc67646756894066',
2593 'upload_date': '20191228',
2594 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2595 'playable_in_embed': True,
2596 'like_count': int,
2597 'categories': ['Entertainment'],
2598 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2599 'age_limit': 18,
2600 'channel': 'Projekt Melody',
2601 'view_count': int,
2602 'availability': 'needs_auth',
2603 'comment_count': int,
2604 'live_status': 'not_live',
2605 'duration': 106,
2606 'uploader': 'Projekt Melody',
2607 'uploader_id': '@ProjektMelody',
2608 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2609 },
2610 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2611 },
2612 {
2613 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2614 'info_dict': {
2615 'id': 'qVv6vCqciTM',
2616 'ext': 'mp4',
2617 'age_limit': 0,
2618 'comment_count': int,
2619 'chapters': 'count:13',
2620 'upload_date': '20221223',
2621 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2622 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2623 'like_count': int,
2624 'release_date': '20221223',
2625 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2626 '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】',
2627 'view_count': int,
2628 'playable_in_embed': True,
2629 'duration': 4438,
2630 'availability': 'public',
2631 'channel_follower_count': int,
2632 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2633 'categories': ['Entertainment'],
2634 'live_status': 'was_live',
2635 'release_timestamp': 1671793345,
2636 'channel': 'さなちゃんねる',
2637 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2638 'uploader': 'さなちゃんねる',
2639 'uploader_url': 'https://www.youtube.com/@sana_natori',
2640 'uploader_id': '@sana_natori',
2641 },
2642 },
2643 {
2644 # Fallbacks when webpage and web client is unavailable
2645 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2646 'info_dict': {
2647 'id': 'wSSmNUl9Snw',
2648 'ext': 'mp4',
2649 # 'categories': ['Science & Technology'],
2650 'view_count': int,
2651 'chapters': 'count:2',
2652 'channel': 'Scott Manley',
2653 'like_count': int,
2654 'age_limit': 0,
2655 # 'availability': 'public',
2656 'channel_follower_count': int,
2657 'live_status': 'not_live',
2658 'upload_date': '20170831',
2659 'duration': 682,
2660 'tags': 'count:8',
2661 'uploader_url': 'https://www.youtube.com/@scottmanley',
2662 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2663 'uploader': 'Scott Manley',
2664 'uploader_id': '@scottmanley',
2665 'title': 'The Computer Hack That Saved Apollo 14',
2666 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2667 'thumbnail': r're:^https?://.*\.webp',
2668 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2669 'playable_in_embed': True,
2670 },
2671 'params': {
2672 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2673 },
2674 },
2675 ]
2676
2677 _WEBPAGE_TESTS = [
2678 # YouTube <object> embed
2679 {
2680 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2681 'md5': '873c81d308b979f0e23ee7e620b312a3',
2682 'info_dict': {
2683 'id': 'msN87y-iEx0',
2684 'ext': 'mp4',
2685 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2686 'upload_date': '20080526',
2687 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2688 'age_limit': 0,
2689 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2690 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2691 'playable_in_embed': True,
2692 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2693 'like_count': int,
2694 'comment_count': int,
2695 'channel': 'Christopher Sykes',
2696 'live_status': 'not_live',
2697 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2698 'availability': 'public',
2699 'duration': 195,
2700 'view_count': int,
2701 'categories': ['Science & Technology'],
2702 'channel_follower_count': int,
2703 'uploader': 'Christopher Sykes',
2704 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2705 'uploader_id': '@ChristopherSykesDocumentaries',
2706 },
2707 'params': {
2708 'skip_download': True,
2709 }
2710 },
2711 ]
2712
2713 @classmethod
2714 def suitable(cls, url):
2715 from ..utils import parse_qs
2716
2717 qs = parse_qs(url)
2718 if qs.get('list', [None])[0]:
2719 return False
2720 return super().suitable(url)
2721
2722 def __init__(self, *args, **kwargs):
2723 super().__init__(*args, **kwargs)
2724 self._code_cache = {}
2725 self._player_cache = {}
2726
2727 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2728 lock = threading.Lock()
2729 start_time = time.time()
2730 formats = [f for f in formats if f.get('is_from_start')]
2731
2732 def refetch_manifest(format_id, delay):
2733 nonlocal formats, start_time, is_live
2734 if time.time() <= start_time + delay:
2735 return
2736
2737 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2738 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2739 microformats = traverse_obj(
2740 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2741 expected_type=dict)
2742 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2743 is_live = live_status == 'is_live'
2744 start_time = time.time()
2745
2746 def mpd_feed(format_id, delay):
2747 """
2748 @returns (manifest_url, manifest_stream_number, is_live) or None
2749 """
2750 for retry in self.RetryManager(fatal=False):
2751 with lock:
2752 refetch_manifest(format_id, delay)
2753
2754 f = next((f for f in formats if f['format_id'] == format_id), None)
2755 if not f:
2756 if not is_live:
2757 retry.error = f'{video_id}: Video is no longer live'
2758 else:
2759 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2760 continue
2761 return f['manifest_url'], f['manifest_stream_number'], is_live
2762 return None
2763
2764 for f in formats:
2765 f['is_live'] = is_live
2766 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2767 live_start_time, mpd_feed, not is_live and f.copy())
2768 if is_live:
2769 f['fragments'] = gen
2770 f['protocol'] = 'http_dash_segments_generator'
2771 else:
2772 f['fragments'] = LazyList(gen({}))
2773 del f['is_from_start']
2774
2775 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2776 FETCH_SPAN, MAX_DURATION = 5, 432000
2777
2778 mpd_url, stream_number, is_live = None, None, True
2779
2780 begin_index = 0
2781 download_start_time = ctx.get('start') or time.time()
2782
2783 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2784 if lack_early_segments:
2785 self.report_warning(bug_reports_message(
2786 'Starting download from the last 120 hours of the live stream since '
2787 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2788 lack_early_segments = True
2789
2790 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2791 fragments, fragment_base_url = None, None
2792
2793 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2794 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2795 # Obtain from MPD's maximum seq value
2796 old_mpd_url = mpd_url
2797 last_error = ctx.pop('last_error', None)
2798 expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
2799 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2800 or (mpd_url, stream_number, False))
2801 if not refresh_sequence:
2802 if expire_fast and not is_live:
2803 return False, last_seq
2804 elif old_mpd_url == mpd_url:
2805 return True, last_seq
2806 if manifestless_orig_fmt:
2807 fmt_info = manifestless_orig_fmt
2808 else:
2809 try:
2810 fmts, _ = self._extract_mpd_formats_and_subtitles(
2811 mpd_url, None, note=False, errnote=False, fatal=False)
2812 except ExtractorError:
2813 fmts = None
2814 if not fmts:
2815 no_fragment_score += 2
2816 return False, last_seq
2817 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2818 fragments = fmt_info['fragments']
2819 fragment_base_url = fmt_info['fragment_base_url']
2820 assert fragment_base_url
2821
2822 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2823 return True, _last_seq
2824
2825 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2826 while is_live:
2827 fetch_time = time.time()
2828 if no_fragment_score > 30:
2829 return
2830 if last_segment_url:
2831 # Obtain from "X-Head-Seqnum" header value from each segment
2832 try:
2833 urlh = self._request_webpage(
2834 last_segment_url, None, note=False, errnote=False, fatal=False)
2835 except ExtractorError:
2836 urlh = None
2837 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2838 if last_seq is None:
2839 no_fragment_score += 2
2840 last_segment_url = None
2841 continue
2842 else:
2843 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2844 no_fragment_score += 2
2845 if not should_continue:
2846 continue
2847
2848 if known_idx > last_seq:
2849 last_segment_url = None
2850 continue
2851
2852 last_seq += 1
2853
2854 if begin_index < 0 and known_idx < 0:
2855 # skip from the start when it's negative value
2856 known_idx = last_seq + begin_index
2857 if lack_early_segments:
2858 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2859 try:
2860 for idx in range(known_idx, last_seq):
2861 # do not update sequence here or you'll get skipped some part of it
2862 should_continue, _ = _extract_sequence_from_mpd(False, False)
2863 if not should_continue:
2864 known_idx = idx - 1
2865 raise ExtractorError('breaking out of outer loop')
2866 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2867 yield {
2868 'url': last_segment_url,
2869 'fragment_count': last_seq,
2870 }
2871 if known_idx == last_seq:
2872 no_fragment_score += 5
2873 else:
2874 no_fragment_score = 0
2875 known_idx = last_seq
2876 except ExtractorError:
2877 continue
2878
2879 if manifestless_orig_fmt:
2880 # Stop at the first iteration if running for post-live manifestless;
2881 # fragment count no longer increase since it starts
2882 break
2883
2884 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2885
2886 def _extract_player_url(self, *ytcfgs, webpage=None):
2887 player_url = traverse_obj(
2888 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2889 get_all=False, expected_type=str)
2890 if not player_url:
2891 return
2892 return urljoin('https://www.youtube.com', player_url)
2893
2894 def _download_player_url(self, video_id, fatal=False):
2895 res = self._download_webpage(
2896 'https://www.youtube.com/iframe_api',
2897 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2898 if res:
2899 player_version = self._search_regex(
2900 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2901 if player_version:
2902 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2903
2904 def _signature_cache_id(self, example_sig):
2905 """ Return a string representation of a signature """
2906 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2907
2908 @classmethod
2909 def _extract_player_info(cls, player_url):
2910 for player_re in cls._PLAYER_INFO_RE:
2911 id_m = re.search(player_re, player_url)
2912 if id_m:
2913 break
2914 else:
2915 raise ExtractorError('Cannot identify player %r' % player_url)
2916 return id_m.group('id')
2917
2918 def _load_player(self, video_id, player_url, fatal=True):
2919 player_id = self._extract_player_info(player_url)
2920 if player_id not in self._code_cache:
2921 code = self._download_webpage(
2922 player_url, video_id, fatal=fatal,
2923 note='Downloading player ' + player_id,
2924 errnote='Download of %s failed' % player_url)
2925 if code:
2926 self._code_cache[player_id] = code
2927 return self._code_cache.get(player_id)
2928
2929 def _extract_signature_function(self, video_id, player_url, example_sig):
2930 player_id = self._extract_player_info(player_url)
2931
2932 # Read from filesystem cache
2933 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2934 assert os.path.basename(func_id) == func_id
2935
2936 self.write_debug(f'Extracting signature function {func_id}')
2937 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2938
2939 if not cache_spec:
2940 code = self._load_player(video_id, player_url)
2941 if code:
2942 res = self._parse_sig_js(code)
2943 test_string = ''.join(map(chr, range(len(example_sig))))
2944 cache_spec = [ord(c) for c in res(test_string)]
2945 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2946
2947 return lambda s: ''.join(s[i] for i in cache_spec)
2948
2949 def _print_sig_code(self, func, example_sig):
2950 if not self.get_param('youtube_print_sig_code'):
2951 return
2952
2953 def gen_sig_code(idxs):
2954 def _genslice(start, end, step):
2955 starts = '' if start == 0 else str(start)
2956 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2957 steps = '' if step == 1 else (':%d' % step)
2958 return f's[{starts}{ends}{steps}]'
2959
2960 step = None
2961 # Quelch pyflakes warnings - start will be set when step is set
2962 start = '(Never used)'
2963 for i, prev in zip(idxs[1:], idxs[:-1]):
2964 if step is not None:
2965 if i - prev == step:
2966 continue
2967 yield _genslice(start, prev, step)
2968 step = None
2969 continue
2970 if i - prev in [-1, 1]:
2971 step = i - prev
2972 start = prev
2973 continue
2974 else:
2975 yield 's[%d]' % prev
2976 if step is None:
2977 yield 's[%d]' % i
2978 else:
2979 yield _genslice(start, i, step)
2980
2981 test_string = ''.join(map(chr, range(len(example_sig))))
2982 cache_res = func(test_string)
2983 cache_spec = [ord(c) for c in cache_res]
2984 expr_code = ' + '.join(gen_sig_code(cache_spec))
2985 signature_id_tuple = '(%s)' % (
2986 ', '.join(str(len(p)) for p in example_sig.split('.')))
2987 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
2988 ' return %s\n') % (signature_id_tuple, expr_code)
2989 self.to_screen('Extracted signature function:\n' + code)
2990
2991 def _parse_sig_js(self, jscode):
2992 funcname = self._search_regex(
2993 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2994 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
2995 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
2996 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
2997 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}\(a,\d+\)',
2998 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\(\s*""\s*\)',
2999 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3000 # Obsolete patterns
3001 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3002 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3003 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3004 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3005 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3006 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3007 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3008 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3009 jscode, 'Initial JS player signature function name', group='sig')
3010
3011 jsi = JSInterpreter(jscode)
3012 initial_function = jsi.extract_function(funcname)
3013 return lambda s: initial_function([s])
3014
3015 def _cached(self, func, *cache_id):
3016 def inner(*args, **kwargs):
3017 if cache_id not in self._player_cache:
3018 try:
3019 self._player_cache[cache_id] = func(*args, **kwargs)
3020 except ExtractorError as e:
3021 self._player_cache[cache_id] = e
3022 except Exception as e:
3023 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3024
3025 ret = self._player_cache[cache_id]
3026 if isinstance(ret, Exception):
3027 raise ret
3028 return ret
3029 return inner
3030
3031 def _decrypt_signature(self, s, video_id, player_url):
3032 """Turn the encrypted s field into a working signature"""
3033 extract_sig = self._cached(
3034 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3035 func = extract_sig(video_id, player_url, s)
3036 self._print_sig_code(func, s)
3037 return func(s)
3038
3039 def _decrypt_nsig(self, s, video_id, player_url):
3040 """Turn the encrypted n field into a working signature"""
3041 if player_url is None:
3042 raise ExtractorError('Cannot decrypt nsig without player_url')
3043 player_url = urljoin('https://www.youtube.com', player_url)
3044
3045 try:
3046 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3047 except ExtractorError as e:
3048 raise ExtractorError('Unable to extract nsig function code', cause=e)
3049 if self.get_param('youtube_print_sig_code'):
3050 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3051
3052 try:
3053 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3054 ret = extract_nsig(jsi, func_code)(s)
3055 except JSInterpreter.Exception as e:
3056 try:
3057 jsi = PhantomJSwrapper(self, timeout=5000)
3058 except ExtractorError:
3059 raise e
3060 self.report_warning(
3061 f'Native nsig extraction failed: Trying with PhantomJS\n'
3062 f' n = {s} ; player = {player_url}', video_id)
3063 self.write_debug(e, only_once=True)
3064
3065 args, func_body = func_code
3066 ret = jsi.execute(
3067 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3068 video_id=video_id, note='Executing signature code').strip()
3069
3070 self.write_debug(f'Decrypted nsig {s} => {ret}')
3071 return ret
3072
3073 def _extract_n_function_name(self, jscode):
3074 funcname, idx = self._search_regex(
3075 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3076 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3077 if not idx:
3078 return funcname
3079
3080 return json.loads(js_to_json(self._search_regex(
3081 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\]);', jscode,
3082 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3083
3084 def _extract_n_function_code(self, video_id, player_url):
3085 player_id = self._extract_player_info(player_url)
3086 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3087 jscode = func_code or self._load_player(video_id, player_url)
3088 jsi = JSInterpreter(jscode)
3089
3090 if func_code:
3091 return jsi, player_id, func_code
3092
3093 func_name = self._extract_n_function_name(jscode)
3094
3095 # For redundancy
3096 func_code = self._search_regex(
3097 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3098 # NB: The end of the regex is intentionally kept strict
3099 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3100 jscode, 'nsig function', group=('var', 'code'), default=None)
3101 if func_code:
3102 func_code = ([func_code[0]], func_code[1])
3103 else:
3104 self.write_debug('Extracting nsig function with jsinterp')
3105 func_code = jsi.extract_function_code(func_name)
3106
3107 self.cache.store('youtube-nsig', player_id, func_code)
3108 return jsi, player_id, func_code
3109
3110 def _extract_n_function_from_code(self, jsi, func_code):
3111 func = jsi.extract_function_from_code(*func_code)
3112
3113 def extract_nsig(s):
3114 try:
3115 ret = func([s])
3116 except JSInterpreter.Exception:
3117 raise
3118 except Exception as e:
3119 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3120
3121 if ret.startswith('enhanced_except_'):
3122 raise JSInterpreter.Exception('Signature function returned an exception')
3123 return ret
3124
3125 return extract_nsig
3126
3127 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3128 """
3129 Extract signatureTimestamp (sts)
3130 Required to tell API what sig/player version is in use.
3131 """
3132 sts = None
3133 if isinstance(ytcfg, dict):
3134 sts = int_or_none(ytcfg.get('STS'))
3135
3136 if not sts:
3137 # Attempt to extract from player
3138 if player_url is None:
3139 error_msg = 'Cannot extract signature timestamp without player_url.'
3140 if fatal:
3141 raise ExtractorError(error_msg)
3142 self.report_warning(error_msg)
3143 return
3144 code = self._load_player(video_id, player_url, fatal=fatal)
3145 if code:
3146 sts = int_or_none(self._search_regex(
3147 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3148 'JS player signature timestamp', group='sts', fatal=fatal))
3149 return sts
3150
3151 def _mark_watched(self, video_id, player_responses):
3152 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3153 label = 'fully ' if is_full else ''
3154 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3155 expected_type=url_or_none)
3156 if not url:
3157 self.report_warning(f'Unable to mark {label}watched')
3158 return
3159 parsed_url = urllib.parse.urlparse(url)
3160 qs = urllib.parse.parse_qs(parsed_url.query)
3161
3162 # cpn generation algorithm is reverse engineered from base.js.
3163 # In fact it works even with dummy cpn.
3164 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3165 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3166
3167 # # more consistent results setting it to right before the end
3168 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3169
3170 qs.update({
3171 'ver': ['2'],
3172 'cpn': [cpn],
3173 'cmt': video_length,
3174 'el': 'detailpage', # otherwise defaults to "shorts"
3175 })
3176
3177 if is_full:
3178 # these seem to mark watchtime "history" in the real world
3179 # they're required, so send in a single value
3180 qs.update({
3181 'st': 0,
3182 'et': video_length,
3183 })
3184
3185 url = urllib.parse.urlunparse(
3186 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3187
3188 self._download_webpage(
3189 url, video_id, f'Marking {label}watched',
3190 'Unable to mark watched', fatal=False)
3191
3192 @classmethod
3193 def _extract_from_webpage(cls, url, webpage):
3194 # Invidious Instances
3195 # https://github.com/yt-dlp/yt-dlp/issues/195
3196 # https://github.com/iv-org/invidious/pull/1730
3197 mobj = re.search(
3198 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3199 webpage)
3200 if mobj:
3201 yield cls.url_result(mobj.group('url'), cls)
3202 raise cls.StopExtraction()
3203
3204 yield from super()._extract_from_webpage(url, webpage)
3205
3206 # lazyYT YouTube embed
3207 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3208 yield cls.url_result(unescapeHTML(id_), cls, id_)
3209
3210 # Wordpress "YouTube Video Importer" plugin
3211 for m in re.findall(r'''(?x)<div[^>]+
3212 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3213 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3214 yield cls.url_result(m[-1], cls, m[-1])
3215
3216 @classmethod
3217 def extract_id(cls, url):
3218 video_id = cls.get_temp_id(url)
3219 if not video_id:
3220 raise ExtractorError(f'Invalid URL: {url}')
3221 return video_id
3222
3223 def _extract_chapters_from_json(self, data, duration):
3224 chapter_list = traverse_obj(
3225 data, (
3226 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3227 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3228 ), expected_type=list)
3229
3230 return self._extract_chapters_helper(
3231 chapter_list,
3232 start_function=lambda chapter: float_or_none(
3233 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3234 title_function=lambda chapter: traverse_obj(
3235 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3236 duration=duration)
3237
3238 def _extract_chapters_from_engagement_panel(self, data, duration):
3239 content_list = traverse_obj(
3240 data,
3241 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3242 expected_type=list)
3243 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3244 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3245
3246 return next(filter(None, (
3247 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3248 chapter_time, chapter_title, duration)
3249 for contents in content_list)), [])
3250
3251 def _extract_comment(self, comment_renderer, parent=None):
3252 comment_id = comment_renderer.get('commentId')
3253 if not comment_id:
3254 return
3255
3256 text = self._get_text(comment_renderer, 'contentText')
3257
3258 # Timestamp is an estimate calculated from the current time and time_text
3259 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3260 timestamp = self._parse_time_text(time_text)
3261
3262 author = self._get_text(comment_renderer, 'authorText')
3263 author_id = try_get(comment_renderer,
3264 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], str)
3265
3266 votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
3267 lambda x: x['likeCount']), str)) or 0
3268 author_thumbnail = try_get(comment_renderer,
3269 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], str)
3270
3271 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
3272 is_favorited = 'creatorHeart' in (try_get(
3273 comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {})
3274 return {
3275 'id': comment_id,
3276 'text': text,
3277 'timestamp': timestamp,
3278 'time_text': time_text,
3279 'like_count': votes,
3280 'is_favorited': is_favorited,
3281 'author': author,
3282 'author_id': author_id,
3283 'author_thumbnail': author_thumbnail,
3284 'author_is_uploader': author_is_uploader,
3285 'parent': parent or 'root'
3286 }
3287
3288 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3289
3290 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3291
3292 def extract_header(contents):
3293 _continuation = None
3294 for content in contents:
3295 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3296 expected_comment_count = self._get_count(
3297 comments_header_renderer, 'countText', 'commentsCount')
3298
3299 if expected_comment_count:
3300 tracker['est_total'] = expected_comment_count
3301 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3302 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3303
3304 sort_menu_item = try_get(
3305 comments_header_renderer,
3306 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3307 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3308
3309 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3310 if not _continuation:
3311 continue
3312
3313 sort_text = str_or_none(sort_menu_item.get('title'))
3314 if not sort_text:
3315 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3316 self.to_screen('Sorting comments by %s' % sort_text.lower())
3317 break
3318 return _continuation
3319
3320 def extract_thread(contents):
3321 if not parent:
3322 tracker['current_page_thread'] = 0
3323 for content in contents:
3324 if not parent and tracker['total_parent_comments'] >= max_parents:
3325 yield
3326 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3327 comment_renderer = get_first(
3328 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3329 expected_type=dict, default={})
3330
3331 comment = self._extract_comment(comment_renderer, parent)
3332 if not comment:
3333 continue
3334 is_pinned = bool(traverse_obj(comment_renderer, 'pinnedCommentBadge'))
3335 comment_id = comment['id']
3336 if is_pinned:
3337 tracker['pinned_comment_ids'].add(comment_id)
3338 # Sometimes YouTube may break and give us infinite looping comments.
3339 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3340 if comment_id in tracker['seen_comment_ids']:
3341 if comment_id in tracker['pinned_comment_ids'] and not is_pinned:
3342 # Pinned comments may appear a second time in newest first sort
3343 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3344 continue
3345 self.report_warning('Detected YouTube comments looping. Stopping comment extraction as we probably cannot get any more.')
3346 yield
3347 else:
3348 tracker['seen_comment_ids'].add(comment['id'])
3349
3350 tracker['running_total'] += 1
3351 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3352 yield comment
3353
3354 # Attempt to get the replies
3355 comment_replies_renderer = try_get(
3356 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3357
3358 if comment_replies_renderer:
3359 tracker['current_page_thread'] += 1
3360 comment_entries_iter = self._comment_entries(
3361 comment_replies_renderer, ytcfg, video_id,
3362 parent=comment.get('id'), tracker=tracker)
3363 yield from itertools.islice(comment_entries_iter, min(
3364 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3365
3366 # Keeps track of counts across recursive calls
3367 if not tracker:
3368 tracker = dict(
3369 running_total=0,
3370 est_total=0,
3371 current_page_thread=0,
3372 total_parent_comments=0,
3373 total_reply_comments=0,
3374 seen_comment_ids=set(),
3375 pinned_comment_ids=set()
3376 )
3377
3378 # TODO: Deprecated
3379 # YouTube comments have a max depth of 2
3380 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3381 if max_depth:
3382 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3383 'Set max replies in the max-comments extractor argument instead')
3384 if max_depth == 1 and parent:
3385 return
3386
3387 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3388 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3389
3390 continuation = self._extract_continuation(root_continuation_data)
3391
3392 response = None
3393 is_forced_continuation = False
3394 is_first_continuation = parent is None
3395 if is_first_continuation and not continuation:
3396 # Sometimes you can get comments by generating the continuation yourself,
3397 # even if YouTube initially reports them being disabled - e.g. stories comments.
3398 # Note: if the comment section is actually disabled, YouTube may return a response with
3399 # required check_get_keys missing. So we will disable that check initially in this case.
3400 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3401 is_forced_continuation = True
3402
3403 for page_num in itertools.count(0):
3404 if not continuation:
3405 break
3406 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3407 comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})"
3408 if page_num == 0:
3409 if is_first_continuation:
3410 note_prefix = 'Downloading comment section API JSON'
3411 else:
3412 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3413 tracker['current_page_thread'], comment_prog_str)
3414 else:
3415 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3416 ' ' if parent else '', ' replies' if parent else '',
3417 page_num, comment_prog_str)
3418 try:
3419 response = self._extract_response(
3420 item_id=None, query=continuation,
3421 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3422 check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None)
3423 except ExtractorError as e:
3424 # Ignore incomplete data error for replies if retries didn't work.
3425 # This is to allow any other parent comments and comment threads to be downloaded.
3426 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3427 if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True:
3428 self.report_warning(
3429 'Received incomplete data for a comment reply thread and retrying did not help. '
3430 'Ignoring to let other comments be downloaded.')
3431 else:
3432 raise
3433 is_forced_continuation = False
3434 continuation_contents = traverse_obj(
3435 response, 'onResponseReceivedEndpoints', expected_type=list, default=[])
3436
3437 continuation = None
3438 for continuation_section in continuation_contents:
3439 continuation_items = traverse_obj(
3440 continuation_section,
3441 (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'),
3442 get_all=False, expected_type=list) or []
3443 if is_first_continuation:
3444 continuation = extract_header(continuation_items)
3445 is_first_continuation = False
3446 if continuation:
3447 break
3448 continue
3449
3450 for entry in extract_thread(continuation_items):
3451 if not entry:
3452 return
3453 yield entry
3454 continuation = self._extract_continuation({'contents': continuation_items})
3455 if continuation:
3456 break
3457
3458 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3459 if message and not parent and tracker['running_total'] == 0:
3460 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3461 raise self.CommentsDisabled
3462
3463 @staticmethod
3464 def _generate_comment_continuation(video_id):
3465 """
3466 Generates initial comment section continuation token from given video id
3467 """
3468 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3469 return base64.b64encode(token.encode()).decode()
3470
3471 def _get_comments(self, ytcfg, video_id, contents, webpage):
3472 """Entry for comment extraction"""
3473 def _real_comment_extract(contents):
3474 renderer = next((
3475 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3476 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3477 yield from self._comment_entries(renderer, ytcfg, video_id)
3478
3479 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3480 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3481
3482 @staticmethod
3483 def _get_checkok_params():
3484 return {'contentCheckOk': True, 'racyCheckOk': True}
3485
3486 @classmethod
3487 def _generate_player_context(cls, sts=None):
3488 context = {
3489 'html5Preference': 'HTML5_PREF_WANTS',
3490 }
3491 if sts is not None:
3492 context['signatureTimestamp'] = sts
3493 return {
3494 'playbackContext': {
3495 'contentPlaybackContext': context
3496 },
3497 **cls._get_checkok_params()
3498 }
3499
3500 @staticmethod
3501 def _is_agegated(player_response):
3502 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3503 return True
3504
3505 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3506 AGE_GATE_REASONS = (
3507 'confirm your age', 'age-restricted', 'inappropriate', # reason
3508 'age_verification_required', 'age_check_required', # status
3509 )
3510 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3511
3512 @staticmethod
3513 def _is_unplayable(player_response):
3514 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3515
3516 _STORY_PLAYER_PARAMS = '8AEB'
3517
3518 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3519
3520 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3521 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3522 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3523 headers = self.generate_api_headers(
3524 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3525
3526 yt_query = {
3527 'videoId': video_id,
3528 }
3529 if smuggled_data.get('is_story') or _split_innertube_client(client)[0] == 'android':
3530 yt_query['params'] = self._STORY_PLAYER_PARAMS
3531
3532 yt_query.update(self._generate_player_context(sts))
3533 return self._extract_response(
3534 item_id=video_id, ep='player', query=yt_query,
3535 ytcfg=player_ytcfg, headers=headers, fatal=True,
3536 default_client=client,
3537 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3538 ) or None
3539
3540 def _get_requested_clients(self, url, smuggled_data):
3541 requested_clients = []
3542 default = ['android', 'web']
3543 allowed_clients = sorted(
3544 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3545 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3546 for client in self._configuration_arg('player_client'):
3547 if client in allowed_clients:
3548 requested_clients.append(client)
3549 elif client == 'default':
3550 requested_clients.extend(default)
3551 elif client == 'all':
3552 requested_clients.extend(allowed_clients)
3553 else:
3554 self.report_warning(f'Skipping unsupported client {client}')
3555 if not requested_clients:
3556 requested_clients = default
3557
3558 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3559 requested_clients.extend(
3560 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3561
3562 return orderedSet(requested_clients)
3563
3564 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3565 initial_pr = None
3566 if webpage:
3567 initial_pr = self._search_json(
3568 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3569
3570 all_clients = set(clients)
3571 clients = clients[::-1]
3572 prs = []
3573
3574 def append_client(*client_names):
3575 """ Append the first client name that exists but not already used """
3576 for client_name in client_names:
3577 actual_client = _split_innertube_client(client_name)[0]
3578 if actual_client in INNERTUBE_CLIENTS:
3579 if actual_client not in all_clients:
3580 clients.append(client_name)
3581 all_clients.add(actual_client)
3582 return
3583
3584 # Android player_response does not have microFormats which are needed for
3585 # extraction of some data. So we return the initial_pr with formats
3586 # stripped out even if not requested by the user
3587 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3588 if initial_pr:
3589 pr = dict(initial_pr)
3590 pr['streamingData'] = None
3591 prs.append(pr)
3592
3593 last_error = None
3594 tried_iframe_fallback = False
3595 player_url = None
3596 while clients:
3597 client, base_client, variant = _split_innertube_client(clients.pop())
3598 player_ytcfg = master_ytcfg if client == 'web' else {}
3599 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3600 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3601
3602 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3603 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3604 if 'js' in self._configuration_arg('player_skip'):
3605 require_js_player = False
3606 player_url = None
3607
3608 if not player_url and not tried_iframe_fallback and require_js_player:
3609 player_url = self._download_player_url(video_id)
3610 tried_iframe_fallback = True
3611
3612 try:
3613 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3614 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3615 except ExtractorError as e:
3616 if last_error:
3617 self.report_warning(last_error)
3618 last_error = e
3619 continue
3620
3621 if pr:
3622 # YouTube may return a different video player response than expected.
3623 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3624 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3625 if pr_video_id and pr_video_id != video_id:
3626 self.report_warning(
3627 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3628 else:
3629 # Save client name for introspection later
3630 name = short_client_name(client)
3631 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3632 sd[STREAMING_DATA_CLIENT_NAME] = name
3633 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3634 f[STREAMING_DATA_CLIENT_NAME] = name
3635 prs.append(pr)
3636
3637 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3638 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3639 append_client(f'{base_client}_creator')
3640 elif self._is_agegated(pr):
3641 if variant == 'tv_embedded':
3642 append_client(f'{base_client}_embedded')
3643 elif not variant:
3644 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3645
3646 if last_error:
3647 if not len(prs):
3648 raise last_error
3649 self.report_warning(last_error)
3650 return prs, player_url
3651
3652 def _needs_live_processing(self, live_status, duration):
3653 if (live_status == 'is_live' and self.get_param('live_from_start')
3654 or live_status == 'post_live' and (duration or 0) > 4 * 3600):
3655 return live_status
3656
3657 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3658 CHUNK_SIZE = 10 << 20
3659 itags, stream_ids = collections.defaultdict(set), []
3660 itag_qualities, res_qualities = {}, {0: None}
3661 q = qualities([
3662 # Normally tiny is the smallest video-only formats. But
3663 # audio-only formats with unknown quality may get tagged as tiny
3664 'tiny',
3665 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3666 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3667 ])
3668 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3669 all_formats = self._configuration_arg('include_duplicate_formats')
3670
3671 def build_fragments(f):
3672 return LazyList({
3673 'url': update_url_query(f['url'], {
3674 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3675 })
3676 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3677
3678 for fmt in streaming_formats:
3679 if fmt.get('targetDurationSec'):
3680 continue
3681
3682 itag = str_or_none(fmt.get('itag'))
3683 audio_track = fmt.get('audioTrack') or {}
3684 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3685 if not all_formats:
3686 if stream_id in stream_ids:
3687 continue
3688
3689 quality = fmt.get('quality')
3690 height = int_or_none(fmt.get('height'))
3691 if quality == 'tiny' or not quality:
3692 quality = fmt.get('audioQuality', '').lower() or quality
3693 # The 3gp format (17) in android client has a quality of "small",
3694 # but is actually worse than other formats
3695 if itag == '17':
3696 quality = 'tiny'
3697 if quality:
3698 if itag:
3699 itag_qualities[itag] = quality
3700 if height:
3701 res_qualities[height] = quality
3702 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3703 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3704 # number of fragment that would subsequently requested with (`&sq=N`)
3705 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3706 continue
3707
3708 fmt_url = fmt.get('url')
3709 if not fmt_url:
3710 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3711 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3712 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3713 if not all((sc, fmt_url, player_url, encrypted_sig)):
3714 continue
3715 try:
3716 fmt_url += '&%s=%s' % (
3717 traverse_obj(sc, ('sp', -1)) or 'signature',
3718 self._decrypt_signature(encrypted_sig, video_id, player_url)
3719 )
3720 except ExtractorError as e:
3721 self.report_warning('Signature extraction failed: Some formats may be missing',
3722 video_id=video_id, only_once=True)
3723 self.write_debug(e, only_once=True)
3724 continue
3725
3726 query = parse_qs(fmt_url)
3727 throttled = False
3728 if query.get('n'):
3729 try:
3730 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3731 fmt_url = update_url_query(fmt_url, {
3732 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3733 })
3734 except ExtractorError as e:
3735 phantomjs_hint = ''
3736 if isinstance(e, JSInterpreter.Exception):
3737 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3738 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3739 if player_url:
3740 self.report_warning(
3741 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3742 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3743 self.write_debug(e, only_once=True)
3744 else:
3745 self.report_warning(
3746 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3747 video_id=video_id, only_once=True)
3748 throttled = True
3749
3750 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3751 language_preference = (
3752 10 if audio_track.get('audioIsDefault') and 10
3753 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3754 else -1)
3755 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3756 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3757 # Make sure to avoid false positives with small duration differences.
3758 # E.g. __2ABJjxzNo, ySuUZEjARPY
3759 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3760 if is_damaged:
3761 self.report_warning(
3762 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3763
3764 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3765 dct = {
3766 'asr': int_or_none(fmt.get('audioSampleRate')),
3767 'filesize': int_or_none(fmt.get('contentLength')),
3768 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3769 'format_note': join_nonempty(
3770 join_nonempty(audio_track.get('displayName'),
3771 language_preference > 0 and ' (default)', delim=''),
3772 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3773 fmt.get('isDrc') and 'DRC',
3774 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3775 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3776 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3777 (self.get_param('verbose') or all_formats) and client_name,
3778 delim=', '),
3779 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3780 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3781 'fps': int_or_none(fmt.get('fps')) or None,
3782 'audio_channels': fmt.get('audioChannels'),
3783 'height': height,
3784 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3785 'has_drm': bool(fmt.get('drmFamilies')),
3786 'tbr': tbr,
3787 'url': fmt_url,
3788 'width': int_or_none(fmt.get('width')),
3789 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3790 'desc' if language_preference < -1 else '') or None,
3791 'language_preference': language_preference,
3792 # Strictly de-prioritize damaged and 3gp formats
3793 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3794 }
3795 mime_mobj = re.match(
3796 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3797 if mime_mobj:
3798 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3799 dct.update(parse_codecs(mime_mobj.group(2)))
3800 if itag:
3801 itags[itag].add(('https', dct.get('language')))
3802 stream_ids.append(stream_id)
3803 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3804 if single_stream and dct.get('ext'):
3805 dct['container'] = dct['ext'] + '_dash'
3806
3807 if all_formats and dct['filesize']:
3808 yield {
3809 **dct,
3810 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3811 'protocol': 'http_dash_segments',
3812 'fragments': build_fragments(dct),
3813 }
3814 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3815 yield dct
3816
3817 needs_live_processing = self._needs_live_processing(live_status, duration)
3818 skip_bad_formats = not self._configuration_arg('include_incomplete_formats')
3819
3820 skip_manifests = set(self._configuration_arg('skip'))
3821 if (not self.get_param('youtube_include_hls_manifest', True)
3822 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3823 or needs_live_processing and skip_bad_formats):
3824 skip_manifests.add('hls')
3825
3826 if not self.get_param('youtube_include_dash_manifest', True):
3827 skip_manifests.add('dash')
3828 if self._configuration_arg('include_live_dash'):
3829 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3830 'Use include_incomplete_formats extractor argument instead')
3831 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3832 skip_manifests.add('dash')
3833
3834 def process_manifest_format(f, proto, client_name, itag):
3835 key = (proto, f.get('language'))
3836 if not all_formats and key in itags[itag]:
3837 return False
3838 itags[itag].add(key)
3839
3840 if itag and all_formats:
3841 f['format_id'] = f'{itag}-{proto}'
3842 elif any(p != proto for p, _ in itags[itag]):
3843 f['format_id'] = f'{itag}-{proto}'
3844 elif itag:
3845 f['format_id'] = itag
3846
3847 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3848 if f['quality'] == -1 and f.get('height'):
3849 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3850 if self.get_param('verbose'):
3851 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3852 return True
3853
3854 subtitles = {}
3855 for sd in streaming_data:
3856 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3857
3858 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3859 if hls_manifest_url:
3860 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3861 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3862 subtitles = self._merge_subtitles(subs, subtitles)
3863 for f in fmts:
3864 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3865 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3866 yield f
3867
3868 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3869 if dash_manifest_url:
3870 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3871 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3872 for f in formats:
3873 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3874 f['filesize'] = int_or_none(self._search_regex(
3875 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3876 if needs_live_processing:
3877 f['is_from_start'] = True
3878
3879 yield f
3880 yield subtitles
3881
3882 def _extract_storyboard(self, player_responses, duration):
3883 spec = get_first(
3884 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3885 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3886 if not base_url:
3887 return
3888 L = len(spec) - 1
3889 for i, args in enumerate(spec):
3890 args = args.split('#')
3891 counts = list(map(int_or_none, args[:5]))
3892 if len(args) != 8 or not all(counts):
3893 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3894 continue
3895 width, height, frame_count, cols, rows = counts
3896 N, sigh = args[6:]
3897
3898 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3899 fragment_count = frame_count / (cols * rows)
3900 fragment_duration = duration / fragment_count
3901 yield {
3902 'format_id': f'sb{i}',
3903 'format_note': 'storyboard',
3904 'ext': 'mhtml',
3905 'protocol': 'mhtml',
3906 'acodec': 'none',
3907 'vcodec': 'none',
3908 'url': url,
3909 'width': width,
3910 'height': height,
3911 'fps': frame_count / duration,
3912 'rows': rows,
3913 'columns': cols,
3914 'fragments': [{
3915 'url': url.replace('$M', str(j)),
3916 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3917 } for j in range(math.ceil(fragment_count))],
3918 }
3919
3920 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3921 webpage = None
3922 if 'webpage' not in self._configuration_arg('player_skip'):
3923 query = {'bpctr': '9999999999', 'has_verified': '1'}
3924 if smuggled_data.get('is_story'):
3925 query['pp'] = self._STORY_PLAYER_PARAMS
3926 webpage = self._download_webpage(
3927 webpage_url, video_id, fatal=False, query=query)
3928
3929 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3930
3931 player_responses, player_url = self._extract_player_responses(
3932 self._get_requested_clients(url, smuggled_data),
3933 video_id, webpage, master_ytcfg, smuggled_data)
3934
3935 return webpage, master_ytcfg, player_responses, player_url
3936
3937 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3938 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3939 is_live = get_first(video_details, 'isLive')
3940 if is_live is None:
3941 is_live = get_first(live_broadcast_details, 'isLiveNow')
3942 live_content = get_first(video_details, 'isLiveContent')
3943 is_upcoming = get_first(video_details, 'isUpcoming')
3944 post_live = get_first(video_details, 'isPostLiveDvr')
3945 live_status = ('post_live' if post_live
3946 else 'is_live' if is_live
3947 else 'is_upcoming' if is_upcoming
3948 else 'was_live' if live_content
3949 else 'not_live' if False in (is_live, live_content)
3950 else None)
3951 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
3952 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
3953
3954 return live_broadcast_details, live_status, streaming_data, formats, subtitles
3955
3956 def _real_extract(self, url):
3957 url, smuggled_data = unsmuggle_url(url, {})
3958 video_id = self._match_id(url)
3959
3960 base_url = self.http_scheme() + '//www.youtube.com/'
3961 webpage_url = base_url + 'watch?v=' + video_id
3962
3963 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
3964
3965 playability_statuses = traverse_obj(
3966 player_responses, (..., 'playabilityStatus'), expected_type=dict)
3967
3968 trailer_video_id = get_first(
3969 playability_statuses,
3970 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
3971 expected_type=str)
3972 if trailer_video_id:
3973 return self.url_result(
3974 trailer_video_id, self.ie_key(), trailer_video_id)
3975
3976 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
3977 if webpage else (lambda x: None))
3978
3979 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
3980 microformats = traverse_obj(
3981 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
3982 expected_type=dict)
3983
3984 translated_title = self._get_text(microformats, (..., 'title'))
3985 video_title = (self._preferred_lang and translated_title
3986 or get_first(video_details, 'title') # primary
3987 or translated_title
3988 or search_meta(['og:title', 'twitter:title', 'title']))
3989 translated_description = self._get_text(microformats, (..., 'description'))
3990 original_description = get_first(video_details, 'shortDescription')
3991 video_description = (
3992 self._preferred_lang and translated_description
3993 # If original description is blank, it will be an empty string.
3994 # Do not prefer translated description in this case.
3995 or original_description if original_description is not None else translated_description)
3996
3997 multifeed_metadata_list = get_first(
3998 player_responses,
3999 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4000 expected_type=str)
4001 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4002 if self.get_param('noplaylist'):
4003 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4004 else:
4005 entries = []
4006 feed_ids = []
4007 for feed in multifeed_metadata_list.split(','):
4008 # Unquote should take place before split on comma (,) since textual
4009 # fields may contain comma as well (see
4010 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4011 feed_data = urllib.parse.parse_qs(
4012 urllib.parse.unquote_plus(feed))
4013
4014 def feed_entry(name):
4015 return try_get(
4016 feed_data, lambda x: x[name][0], str)
4017
4018 feed_id = feed_entry('id')
4019 if not feed_id:
4020 continue
4021 feed_title = feed_entry('title')
4022 title = video_title
4023 if feed_title:
4024 title += ' (%s)' % feed_title
4025 entries.append({
4026 '_type': 'url_transparent',
4027 'ie_key': 'Youtube',
4028 'url': smuggle_url(
4029 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4030 {'force_singlefeed': True}),
4031 'title': title,
4032 })
4033 feed_ids.append(feed_id)
4034 self.to_screen(
4035 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4036 % (', '.join(feed_ids), video_id))
4037 return self.playlist_result(
4038 entries, video_id, video_title, video_description)
4039
4040 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4041 or int_or_none(get_first(microformats, 'lengthSeconds'))
4042 or parse_duration(search_meta('duration')) or None)
4043
4044 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4045 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4046 if live_status == 'post_live':
4047 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4048
4049 if not formats:
4050 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4051 self.report_drm(video_id)
4052 pemr = get_first(
4053 playability_statuses,
4054 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4055 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4056 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4057 if subreason:
4058 if subreason == 'The uploader has not made this video available in your country.':
4059 countries = get_first(microformats, 'availableCountries')
4060 if not countries:
4061 regions_allowed = search_meta('regionsAllowed')
4062 countries = regions_allowed.split(',') if regions_allowed else None
4063 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4064 reason += f'. {subreason}'
4065 if reason:
4066 self.raise_no_formats(reason, expected=True)
4067
4068 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4069 if not keywords and webpage:
4070 keywords = [
4071 unescapeHTML(m.group('content'))
4072 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4073 for keyword in keywords:
4074 if keyword.startswith('yt:stretch='):
4075 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4076 if mobj:
4077 # NB: float is intentional for forcing float division
4078 w, h = (float(v) for v in mobj.groups())
4079 if w > 0 and h > 0:
4080 ratio = w / h
4081 for f in formats:
4082 if f.get('vcodec') != 'none':
4083 f['stretched_ratio'] = ratio
4084 break
4085 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4086 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4087 if thumbnail_url:
4088 thumbnails.append({
4089 'url': thumbnail_url,
4090 })
4091 original_thumbnails = thumbnails.copy()
4092
4093 # The best resolution thumbnails sometimes does not appear in the webpage
4094 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4095 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4096 thumbnail_names = [
4097 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4098 # in resolution, these are not the custom thumbnail. So de-prioritize them
4099 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4100 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4101 ]
4102 n_thumbnail_names = len(thumbnail_names)
4103 thumbnails.extend({
4104 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4105 video_id=video_id, name=name, ext=ext,
4106 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4107 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4108 for thumb in thumbnails:
4109 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4110 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4111 self._remove_duplicate_formats(thumbnails)
4112 self._downloader._sort_thumbnails(original_thumbnails)
4113
4114 category = get_first(microformats, 'category') or search_meta('genre')
4115 channel_id = self.ucid_or_none(str_or_none(
4116 get_first(video_details, 'channelId')
4117 or get_first(microformats, 'externalChannelId')
4118 or search_meta('channelId')))
4119 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4120
4121 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4122 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4123 if not duration and live_end_time and live_start_time:
4124 duration = live_end_time - live_start_time
4125
4126 needs_live_processing = self._needs_live_processing(live_status, duration)
4127
4128 def is_bad_format(fmt):
4129 if needs_live_processing and not fmt.get('is_from_start'):
4130 return True
4131 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4132 and fmt.get('protocol') == 'http_dash_segments'):
4133 return True
4134
4135 for fmt in filter(is_bad_format, formats):
4136 fmt['preference'] = (fmt.get('preference') or -1) - 10
4137 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 4 hours)', delim=' ')
4138
4139 if needs_live_processing:
4140 self._prepare_live_from_start_formats(
4141 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4142
4143 formats.extend(self._extract_storyboard(player_responses, duration))
4144
4145 channel_handle = self.handle_from_url(owner_profile_url)
4146
4147 info = {
4148 'id': video_id,
4149 'title': video_title,
4150 'formats': formats,
4151 'thumbnails': thumbnails,
4152 # The best thumbnail that we are sure exists. Prevents unnecessary
4153 # URL checking if user don't care about getting the best possible thumbnail
4154 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4155 'description': video_description,
4156 'channel_id': channel_id,
4157 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4158 'duration': duration,
4159 'view_count': int_or_none(
4160 get_first((video_details, microformats), (..., 'viewCount'))
4161 or search_meta('interactionCount')),
4162 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4163 'age_limit': 18 if (
4164 get_first(microformats, 'isFamilySafe') is False
4165 or search_meta('isFamilyFriendly') == 'false'
4166 or search_meta('og:restrictions:age') == '18+') else 0,
4167 'webpage_url': webpage_url,
4168 'categories': [category] if category else None,
4169 'tags': keywords,
4170 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4171 'live_status': live_status,
4172 'release_timestamp': live_start_time,
4173 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4174 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4175 }
4176
4177 subtitles = {}
4178 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4179 if pctr:
4180 def get_lang_code(track):
4181 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4182 or track.get('languageCode'))
4183
4184 # Converted into dicts to remove duplicates
4185 captions = {
4186 get_lang_code(sub): sub
4187 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4188 translation_languages = {
4189 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4190 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4191
4192 def process_language(container, base_url, lang_code, sub_name, query):
4193 lang_subs = container.setdefault(lang_code, [])
4194 for fmt in self._SUBTITLE_FORMATS:
4195 query.update({
4196 'fmt': fmt,
4197 })
4198 lang_subs.append({
4199 'ext': fmt,
4200 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4201 'name': sub_name,
4202 })
4203
4204 # NB: Constructing the full subtitle dictionary is slow
4205 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4206 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4207 for lang_code, caption_track in captions.items():
4208 base_url = caption_track.get('baseUrl')
4209 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4210 if not base_url:
4211 continue
4212 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4213 if caption_track.get('kind') != 'asr':
4214 if not lang_code:
4215 continue
4216 process_language(
4217 subtitles, base_url, lang_code, lang_name, {})
4218 if not caption_track.get('isTranslatable'):
4219 continue
4220 for trans_code, trans_name in translation_languages.items():
4221 if not trans_code:
4222 continue
4223 orig_trans_code = trans_code
4224 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4225 if not get_translated_subs:
4226 continue
4227 trans_code += f'-{lang_code}'
4228 trans_name += format_field(lang_name, None, ' from %s')
4229 # Add an "-orig" label to the original language so that it can be distinguished.
4230 # The subs are returned without "-orig" as well for compatibility
4231 if lang_code == f'a-{orig_trans_code}':
4232 process_language(
4233 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4234 # Setting tlang=lang returns damaged subtitles.
4235 process_language(automatic_captions, base_url, trans_code, trans_name,
4236 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4237
4238 info['automatic_captions'] = automatic_captions
4239 info['subtitles'] = subtitles
4240
4241 parsed_url = urllib.parse.urlparse(url)
4242 for component in [parsed_url.fragment, parsed_url.query]:
4243 query = urllib.parse.parse_qs(component)
4244 for k, v in query.items():
4245 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4246 d_k += '_time'
4247 if d_k not in info and k in s_ks:
4248 info[d_k] = parse_duration(query[k][0])
4249
4250 # Youtube Music Auto-generated description
4251 if video_description:
4252 mobj = re.search(
4253 r'''(?xs)
4254 (?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+
4255 (?P<album>[^\n]+)
4256 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4257 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4258 (.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?
4259 .+\nAuto-generated\ by\ YouTube\.\s*$
4260 ''', video_description)
4261 if mobj:
4262 release_year = mobj.group('release_year')
4263 release_date = mobj.group('release_date')
4264 if release_date:
4265 release_date = release_date.replace('-', '')
4266 if not release_year:
4267 release_year = release_date[:4]
4268 info.update({
4269 'album': mobj.group('album'.strip()),
4270 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4271 'track': mobj.group('track').strip(),
4272 'release_date': release_date,
4273 'release_year': int_or_none(release_year),
4274 })
4275
4276 initial_data = None
4277 if webpage:
4278 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4279 if not traverse_obj(initial_data, 'contents'):
4280 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4281 initial_data = None
4282 if not initial_data:
4283 query = {'videoId': video_id}
4284 query.update(self._get_checkok_params())
4285 initial_data = self._extract_response(
4286 item_id=video_id, ep='next', fatal=False,
4287 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4288 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4289 note='Downloading initial data API JSON')
4290
4291 info['comment_count'] = traverse_obj(initial_data, (
4292 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4293 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4294 ), (
4295 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4296 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4297 ), expected_type=self._get_count, get_all=False)
4298
4299 try: # This will error if there is no livechat
4300 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4301 except (KeyError, IndexError, TypeError):
4302 pass
4303 else:
4304 info.setdefault('subtitles', {})['live_chat'] = [{
4305 # url is needed to set cookies
4306 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4307 'video_id': video_id,
4308 'ext': 'json',
4309 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4310 else 'youtube_live_chat_replay'),
4311 }]
4312
4313 if initial_data:
4314 info['chapters'] = (
4315 self._extract_chapters_from_json(initial_data, duration)
4316 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4317 or self._extract_chapters_from_description(video_description, duration)
4318 or None)
4319
4320 contents = traverse_obj(
4321 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4322 expected_type=list, default=[])
4323
4324 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4325 if vpir:
4326 stl = vpir.get('superTitleLink')
4327 if stl:
4328 stl = self._get_text(stl)
4329 if try_get(
4330 vpir,
4331 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4332 info['location'] = stl
4333 else:
4334 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4335 if mobj:
4336 info.update({
4337 'series': mobj.group(1),
4338 'season_number': int(mobj.group(2)),
4339 'episode_number': int(mobj.group(3)),
4340 })
4341 for tlb in (try_get(
4342 vpir,
4343 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4344 list) or []):
4345 tbrs = variadic(
4346 traverse_obj(
4347 tlb, ('toggleButtonRenderer', ...),
4348 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4349 for tbr in tbrs:
4350 for getter, regex in [(
4351 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4352 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4353 lambda x: x['accessibility'],
4354 lambda x: x['accessibilityData']['accessibilityData'],
4355 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4356 label = (try_get(tbr, getter, dict) or {}).get('label')
4357 if label:
4358 mobj = re.match(regex, label)
4359 if mobj:
4360 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4361 break
4362 sbr_tooltip = try_get(
4363 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4364 if sbr_tooltip:
4365 like_count, dislike_count = sbr_tooltip.split(' / ')
4366 info.update({
4367 'like_count': str_to_int(like_count),
4368 'dislike_count': str_to_int(dislike_count),
4369 })
4370 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4371 if vcr:
4372 vc = self._get_count(vcr, 'viewCount')
4373 # Upcoming premieres with waiting count are treated as live here
4374 if vcr.get('isLive'):
4375 info['concurrent_view_count'] = vc
4376 elif info.get('view_count') is None:
4377 info['view_count'] = vc
4378
4379 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4380 if vsir:
4381 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4382 info.update({
4383 'channel': self._get_text(vor, 'title'),
4384 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4385
4386 if not channel_handle:
4387 channel_handle = self.handle_from_url(
4388 traverse_obj(vor, (
4389 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4390 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4391 {str}), get_all=False))
4392
4393 rows = try_get(
4394 vsir,
4395 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4396 list) or []
4397 multiple_songs = False
4398 for row in rows:
4399 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4400 multiple_songs = True
4401 break
4402 for row in rows:
4403 mrr = row.get('metadataRowRenderer') or {}
4404 mrr_title = mrr.get('title')
4405 if not mrr_title:
4406 continue
4407 mrr_title = self._get_text(mrr, 'title')
4408 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4409 if mrr_title == 'License':
4410 info['license'] = mrr_contents_text
4411 elif not multiple_songs:
4412 if mrr_title == 'Album':
4413 info['album'] = mrr_contents_text
4414 elif mrr_title == 'Artist':
4415 info['artist'] = mrr_contents_text
4416 elif mrr_title == 'Song':
4417 info['track'] = mrr_contents_text
4418
4419 info.update({
4420 'uploader': info.get('channel'),
4421 'uploader_id': channel_handle,
4422 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4423 })
4424 # The upload date for scheduled, live and past live streams / premieres in microformats
4425 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4426 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4427 upload_date = (
4428 unified_strdate(get_first(microformats, 'uploadDate'))
4429 or unified_strdate(search_meta('uploadDate')))
4430 if not upload_date or (
4431 live_status in ('not_live', None)
4432 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4433 ):
4434 upload_date = strftime_or_none(
4435 self._parse_time_text(self._get_text(vpir, 'dateText')), '%Y%m%d') or upload_date
4436 info['upload_date'] = upload_date
4437
4438 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4439 v = info.get(s_k)
4440 if v:
4441 info[d_k] = v
4442
4443 badges = self._extract_badges(traverse_obj(contents, (..., 'videoPrimaryInfoRenderer'), get_all=False))
4444
4445 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4446 or get_first(video_details, 'isPrivate', expected_type=bool))
4447
4448 info['availability'] = (
4449 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4450 else self._availability(
4451 is_private=is_private,
4452 needs_premium=(
4453 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4454 or False if initial_data and is_private is not None else None),
4455 needs_subscription=(
4456 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4457 or False if initial_data and is_private is not None else None),
4458 needs_auth=info['age_limit'] >= 18,
4459 is_unlisted=None if is_private is None else (
4460 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4461 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4462
4463 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4464
4465 self.mark_watched(video_id, player_responses)
4466
4467 return info
4468
4469
4470 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4471 @staticmethod
4472 def passthrough_smuggled_data(func):
4473 def _smuggle(info, smuggled_data):
4474 if info.get('_type') not in ('url', 'url_transparent'):
4475 return info
4476 if smuggled_data.get('is_music_url'):
4477 parsed_url = urllib.parse.urlparse(info['url'])
4478 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4479 smuggled_data.pop('is_music_url')
4480 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4481 if smuggled_data:
4482 info['url'] = smuggle_url(info['url'], smuggled_data)
4483 return info
4484
4485 @functools.wraps(func)
4486 def wrapper(self, url):
4487 url, smuggled_data = unsmuggle_url(url, {})
4488 if self.is_music_url(url):
4489 smuggled_data['is_music_url'] = True
4490 info_dict = func(self, url, smuggled_data)
4491 if smuggled_data:
4492 _smuggle(info_dict, smuggled_data)
4493 if info_dict.get('entries'):
4494 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4495 return info_dict
4496 return wrapper
4497
4498 @staticmethod
4499 def _extract_basic_item_renderer(item):
4500 # Modified from _extract_grid_item_renderer
4501 known_basic_renderers = (
4502 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4503 )
4504 for key, renderer in item.items():
4505 if not isinstance(renderer, dict):
4506 continue
4507 elif key in known_basic_renderers:
4508 return renderer
4509 elif key.startswith('grid') and key.endswith('Renderer'):
4510 return renderer
4511
4512 def _extract_channel_renderer(self, renderer):
4513 channel_id = self.ucid_or_none(renderer['channelId'])
4514 title = self._get_text(renderer, 'title')
4515 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4516 # As of 2023-03-01 YouTube doesn't use the channel handles on these renderers yet.
4517 # However we can expect them to change that in the future.
4518 channel_handle = self.handle_from_url(
4519 traverse_obj(renderer, (
4520 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4521 ('browseEndpoint', 'canonicalBaseUrl')),
4522 {str}), get_all=False))
4523 return {
4524 '_type': 'url',
4525 'url': channel_url,
4526 'id': channel_id,
4527 'ie_key': YoutubeTabIE.ie_key(),
4528 'channel': title,
4529 'uploader': title,
4530 'channel_id': channel_id,
4531 'channel_url': channel_url,
4532 'title': title,
4533 'uploader_id': channel_handle,
4534 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4535 'channel_follower_count': self._get_count(renderer, 'subscriberCountText'),
4536 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4537 'playlist_count': self._get_count(renderer, 'videoCountText'),
4538 'description': self._get_text(renderer, 'descriptionSnippet'),
4539 }
4540
4541 def _grid_entries(self, grid_renderer):
4542 for item in grid_renderer['items']:
4543 if not isinstance(item, dict):
4544 continue
4545 renderer = self._extract_basic_item_renderer(item)
4546 if not isinstance(renderer, dict):
4547 continue
4548 title = self._get_text(renderer, 'title')
4549
4550 # playlist
4551 playlist_id = renderer.get('playlistId')
4552 if playlist_id:
4553 yield self.url_result(
4554 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4555 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4556 video_title=title)
4557 continue
4558 # video
4559 video_id = renderer.get('videoId')
4560 if video_id:
4561 yield self._extract_video(renderer)
4562 continue
4563 # channel
4564 channel_id = renderer.get('channelId')
4565 if channel_id:
4566 yield self._extract_channel_renderer(renderer)
4567 continue
4568 # generic endpoint URL support
4569 ep_url = urljoin('https://www.youtube.com/', try_get(
4570 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4571 str))
4572 if ep_url:
4573 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4574 if ie.suitable(ep_url):
4575 yield self.url_result(
4576 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4577 break
4578
4579 def _music_reponsive_list_entry(self, renderer):
4580 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4581 if video_id:
4582 title = traverse_obj(renderer, (
4583 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4584 'text', 'runs', 0, 'text'))
4585 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4586 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4587 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4588 if playlist_id:
4589 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4590 if video_id:
4591 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4592 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4593 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4594 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4595 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4596 if browse_id:
4597 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4598 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4599
4600 def _shelf_entries_from_content(self, shelf_renderer):
4601 content = shelf_renderer.get('content')
4602 if not isinstance(content, dict):
4603 return
4604 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4605 if renderer:
4606 # TODO: add support for nested playlists so each shelf is processed
4607 # as separate playlist
4608 # TODO: this includes only first N items
4609 yield from self._grid_entries(renderer)
4610 renderer = content.get('horizontalListRenderer')
4611 if renderer:
4612 # TODO
4613 pass
4614
4615 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4616 ep = try_get(
4617 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4618 str)
4619 shelf_url = urljoin('https://www.youtube.com', ep)
4620 if shelf_url:
4621 # Skipping links to another channels, note that checking for
4622 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4623 # will not work
4624 if skip_channels and '/channels?' in shelf_url:
4625 return
4626 title = self._get_text(shelf_renderer, 'title')
4627 yield self.url_result(shelf_url, video_title=title)
4628 # Shelf may not contain shelf URL, fallback to extraction from content
4629 yield from self._shelf_entries_from_content(shelf_renderer)
4630
4631 def _playlist_entries(self, video_list_renderer):
4632 for content in video_list_renderer['contents']:
4633 if not isinstance(content, dict):
4634 continue
4635 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4636 if not isinstance(renderer, dict):
4637 continue
4638 video_id = renderer.get('videoId')
4639 if not video_id:
4640 continue
4641 yield self._extract_video(renderer)
4642
4643 def _rich_entries(self, rich_grid_renderer):
4644 renderer = traverse_obj(
4645 rich_grid_renderer,
4646 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4647 video_id = renderer.get('videoId')
4648 if video_id:
4649 yield self._extract_video(renderer)
4650 return
4651 playlist_id = renderer.get('playlistId')
4652 if playlist_id:
4653 yield self.url_result(
4654 f'https://www.youtube.com/playlist?list={playlist_id}',
4655 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4656 video_title=self._get_text(renderer, 'title'))
4657 return
4658
4659 def _video_entry(self, video_renderer):
4660 video_id = video_renderer.get('videoId')
4661 if video_id:
4662 return self._extract_video(video_renderer)
4663
4664 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4665 url = urljoin('https://youtube.com', traverse_obj(
4666 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4667 if url:
4668 return self.url_result(
4669 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4670
4671 def _post_thread_entries(self, post_thread_renderer):
4672 post_renderer = try_get(
4673 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4674 if not post_renderer:
4675 return
4676 # video attachment
4677 video_renderer = try_get(
4678 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4679 video_id = video_renderer.get('videoId')
4680 if video_id:
4681 entry = self._extract_video(video_renderer)
4682 if entry:
4683 yield entry
4684 # playlist attachment
4685 playlist_id = try_get(
4686 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4687 if playlist_id:
4688 yield self.url_result(
4689 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4690 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4691 # inline video links
4692 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4693 for run in runs:
4694 if not isinstance(run, dict):
4695 continue
4696 ep_url = try_get(
4697 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4698 if not ep_url:
4699 continue
4700 if not YoutubeIE.suitable(ep_url):
4701 continue
4702 ep_video_id = YoutubeIE._match_id(ep_url)
4703 if video_id == ep_video_id:
4704 continue
4705 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4706
4707 def _post_thread_continuation_entries(self, post_thread_continuation):
4708 contents = post_thread_continuation.get('contents')
4709 if not isinstance(contents, list):
4710 return
4711 for content in contents:
4712 renderer = content.get('backstagePostThreadRenderer')
4713 if isinstance(renderer, dict):
4714 yield from self._post_thread_entries(renderer)
4715 continue
4716 renderer = content.get('videoRenderer')
4717 if isinstance(renderer, dict):
4718 yield self._video_entry(renderer)
4719
4720 r''' # unused
4721 def _rich_grid_entries(self, contents):
4722 for content in contents:
4723 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4724 if video_renderer:
4725 entry = self._video_entry(video_renderer)
4726 if entry:
4727 yield entry
4728 '''
4729
4730 def _report_history_entries(self, renderer):
4731 for url in traverse_obj(renderer, (
4732 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4733 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4734 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4735 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4736
4737 def _extract_entries(self, parent_renderer, continuation_list):
4738 # continuation_list is modified in-place with continuation_list = [continuation_token]
4739 continuation_list[:] = [None]
4740 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4741 for content in contents:
4742 if not isinstance(content, dict):
4743 continue
4744 is_renderer = traverse_obj(
4745 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4746 expected_type=dict)
4747 if not is_renderer:
4748 if content.get('richItemRenderer'):
4749 for entry in self._rich_entries(content['richItemRenderer']):
4750 yield entry
4751 continuation_list[0] = self._extract_continuation(parent_renderer)
4752 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4753 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4754 yield from self._report_history_entries(table)
4755 continuation_list[0] = self._extract_continuation(table)
4756 continue
4757
4758 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4759 for isr_content in isr_contents:
4760 if not isinstance(isr_content, dict):
4761 continue
4762
4763 known_renderers = {
4764 'playlistVideoListRenderer': self._playlist_entries,
4765 'gridRenderer': self._grid_entries,
4766 'reelShelfRenderer': self._grid_entries,
4767 'shelfRenderer': self._shelf_entries,
4768 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4769 'backstagePostThreadRenderer': self._post_thread_entries,
4770 'videoRenderer': lambda x: [self._video_entry(x)],
4771 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4772 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4773 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4774 }
4775 for key, renderer in isr_content.items():
4776 if key not in known_renderers:
4777 continue
4778 for entry in known_renderers[key](renderer):
4779 if entry:
4780 yield entry
4781 continuation_list[0] = self._extract_continuation(renderer)
4782 break
4783
4784 if not continuation_list[0]:
4785 continuation_list[0] = self._extract_continuation(is_renderer)
4786
4787 if not continuation_list[0]:
4788 continuation_list[0] = self._extract_continuation(parent_renderer)
4789
4790 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4791 continuation_list = [None]
4792 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4793 tab_content = try_get(tab, lambda x: x['content'], dict)
4794 if not tab_content:
4795 return
4796 parent_renderer = (
4797 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4798 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4799 yield from extract_entries(parent_renderer)
4800 continuation = continuation_list[0]
4801
4802 for page_num in itertools.count(1):
4803 if not continuation:
4804 break
4805 headers = self.generate_api_headers(
4806 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4807 response = self._extract_response(
4808 item_id=f'{item_id} page {page_num}',
4809 query=continuation, headers=headers, ytcfg=ytcfg,
4810 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4811
4812 if not response:
4813 break
4814 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4815 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4816 visitor_data = self._extract_visitor_data(response) or visitor_data
4817
4818 known_renderers = {
4819 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4820 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4821 'gridVideoRenderer': (self._grid_entries, 'items'),
4822 'gridChannelRenderer': (self._grid_entries, 'items'),
4823 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4824 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4825 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4826 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4827 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4828 'playlistVideoListContinuation': (self._playlist_entries, None),
4829 'gridContinuation': (self._grid_entries, None),
4830 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4831 'sectionListContinuation': (extract_entries, None), # for feeds
4832 }
4833
4834 continuation_items = traverse_obj(response, (
4835 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4836 'appendContinuationItemsAction', 'continuationItems'
4837 ), 'continuationContents', get_all=False)
4838 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4839
4840 video_items_renderer = None
4841 for key in continuation_item.keys():
4842 if key not in known_renderers:
4843 continue
4844 func, parent_key = known_renderers[key]
4845 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4846 continuation_list = [None]
4847 yield from func(video_items_renderer)
4848 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4849
4850 if not video_items_renderer:
4851 break
4852
4853 @staticmethod
4854 def _extract_selected_tab(tabs, fatal=True):
4855 for tab_renderer in tabs:
4856 if tab_renderer.get('selected'):
4857 return tab_renderer
4858 if fatal:
4859 raise ExtractorError('Unable to find selected tab')
4860
4861 @staticmethod
4862 def _extract_tab_renderers(response):
4863 return traverse_obj(
4864 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
4865
4866 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4867 metadata = self._extract_metadata_from_tabs(item_id, data)
4868
4869 selected_tab = self._extract_selected_tab(tabs)
4870 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
4871 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
4872
4873 return self.playlist_result(
4874 self._entries(
4875 selected_tab, metadata['id'], ytcfg,
4876 self._extract_account_syncid(ytcfg, data),
4877 self._extract_visitor_data(data, ytcfg)),
4878 **metadata)
4879
4880 def _extract_metadata_from_tabs(self, item_id, data):
4881 info = {'id': item_id}
4882
4883 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
4884 if metadata_renderer:
4885 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
4886 ('channelUrl', {self.ucid_from_url}))
4887 info.update({
4888 'channel': metadata_renderer.get('title'),
4889 'channel_id': channel_id,
4890 })
4891 if info['channel_id']:
4892 info['id'] = info['channel_id']
4893 else:
4894 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
4895
4896 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4897 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4898 def _get_uncropped(url):
4899 return url_or_none((url or '').split('=')[0] + '=s0')
4900
4901 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
4902 if avatar_thumbnails:
4903 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4904 if uncropped_avatar:
4905 avatar_thumbnails.append({
4906 'url': uncropped_avatar,
4907 'id': 'avatar_uncropped',
4908 'preference': 1
4909 })
4910
4911 channel_banners = self._extract_thumbnails(
4912 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
4913 for banner in channel_banners:
4914 banner['preference'] = -10
4915
4916 if channel_banners:
4917 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4918 if uncropped_banner:
4919 channel_banners.append({
4920 'url': uncropped_banner,
4921 'id': 'banner_uncropped',
4922 'preference': -5
4923 })
4924
4925 # Deprecated - remove primary_sidebar_renderer when layout discontinued
4926 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4927 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
4928
4929 primary_thumbnails = self._extract_thumbnails(
4930 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4931 playlist_thumbnails = self._extract_thumbnails(
4932 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
4933
4934 info.update({
4935 'title': (traverse_obj(metadata_renderer, 'title')
4936 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
4937 or info['id']),
4938 'availability': self._extract_availability(data),
4939 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4940 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
4941 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
4942 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
4943 })
4944
4945 channel_handle = (
4946 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
4947 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
4948
4949 if channel_handle:
4950 info.update({
4951 'uploader_id': channel_handle,
4952 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4953 })
4954 # Playlist stats is a text runs array containing [video count, view count, last updated].
4955 # last updated or (view count and last updated) may be missing.
4956 playlist_stats = get_first(
4957 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
4958
4959 last_updated_unix = self._parse_time_text(
4960 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
4961 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
4962 info['modified_date'] = strftime_or_none(last_updated_unix, '%Y%m%d')
4963
4964 info['view_count'] = self._get_count(playlist_stats, 1)
4965 if info['view_count'] is None: # 0 is allowed
4966 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
4967 if info['view_count'] is None:
4968 info['view_count'] = self._get_count(data, (
4969 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
4970 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
4971
4972 info['playlist_count'] = self._get_count(playlist_stats, 0)
4973 if info['playlist_count'] is None: # 0 is allowed
4974 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
4975
4976 if not info.get('channel_id'):
4977 owner = traverse_obj(playlist_header_renderer, 'ownerText')
4978 if not owner: # Deprecated
4979 owner = traverse_obj(
4980 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
4981 ('videoOwner', 'videoOwnerRenderer', 'title'))
4982 owner_text = self._get_text(owner)
4983 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
4984 info.update({
4985 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
4986 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
4987 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
4988 })
4989
4990 info.update({
4991 'uploader': info['channel'],
4992 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
4993 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
4994 })
4995
4996 return info
4997
4998 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
4999 first_id = last_id = response = None
5000 for page_num in itertools.count(1):
5001 videos = list(self._playlist_entries(playlist))
5002 if not videos:
5003 return
5004 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5005 if start >= len(videos):
5006 return
5007 yield from videos[start:]
5008 first_id = first_id or videos[0]['id']
5009 last_id = videos[-1]['id']
5010 watch_endpoint = try_get(
5011 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5012 headers = self.generate_api_headers(
5013 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5014 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5015 query = {
5016 'playlistId': playlist_id,
5017 'videoId': watch_endpoint.get('videoId') or last_id,
5018 'index': watch_endpoint.get('index') or len(videos),
5019 'params': watch_endpoint.get('params') or 'OAE%3D'
5020 }
5021 response = self._extract_response(
5022 item_id='%s page %d' % (playlist_id, page_num),
5023 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5024 check_get_keys='contents'
5025 )
5026 playlist = try_get(
5027 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5028
5029 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5030 title = playlist.get('title') or try_get(
5031 data, lambda x: x['titleText']['simpleText'], str)
5032 playlist_id = playlist.get('playlistId') or item_id
5033
5034 # Delegating everything except mix playlists to regular tab-based playlist URL
5035 playlist_url = urljoin(url, try_get(
5036 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5037 str))
5038
5039 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5040 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5041 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5042
5043 if playlist_url and playlist_url != url and not is_known_unviewable:
5044 return self.url_result(
5045 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5046 video_title=title)
5047
5048 return self.playlist_result(
5049 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5050 playlist_id=playlist_id, playlist_title=title)
5051
5052 def _extract_availability(self, data):
5053 """
5054 Gets the availability of a given playlist/tab.
5055 Note: Unless YouTube tells us explicitly, we do not assume it is public
5056 @param data: response
5057 """
5058 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5059 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5060 player_header_privacy = playlist_header_renderer.get('privacy')
5061
5062 badges = self._extract_badges(sidebar_renderer)
5063
5064 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5065 privacy_setting_icon = get_first(
5066 (playlist_header_renderer, sidebar_renderer),
5067 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5068 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5069 expected_type=str)
5070
5071 microformats_is_unlisted = traverse_obj(
5072 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5073
5074 return (
5075 'public' if (
5076 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5077 or player_header_privacy == 'PUBLIC'
5078 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5079 else self._availability(
5080 is_private=(
5081 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5082 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5083 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5084 is_unlisted=(
5085 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5086 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5087 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5088 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5089 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5090 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5091 needs_auth=False))
5092
5093 @staticmethod
5094 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5095 sidebar_renderer = try_get(
5096 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5097 for item in sidebar_renderer:
5098 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5099 if renderer:
5100 return renderer
5101
5102 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5103 """
5104 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5105 """
5106 is_playlist = bool(traverse_obj(
5107 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5108 if not is_playlist:
5109 return
5110 headers = self.generate_api_headers(
5111 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5112 visitor_data=self._extract_visitor_data(data, ytcfg))
5113 query = {
5114 'params': 'wgYCCAA=',
5115 'browseId': f'VL{item_id}'
5116 }
5117 return self._extract_response(
5118 item_id=item_id, headers=headers, query=query,
5119 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5120 note='Redownloading playlist API JSON with unavailable videos')
5121
5122 @functools.cached_property
5123 def skip_webpage(self):
5124 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5125
5126 def _extract_webpage(self, url, item_id, fatal=True):
5127 webpage, data = None, None
5128 for retry in self.RetryManager(fatal=fatal):
5129 try:
5130 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5131 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5132 except ExtractorError as e:
5133 if isinstance(e.cause, network_exceptions):
5134 if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
5135 retry.error = e
5136 continue
5137 self._error_or_warning(e, fatal=fatal)
5138 break
5139
5140 try:
5141 self._extract_and_report_alerts(data)
5142 except ExtractorError as e:
5143 self._error_or_warning(e, fatal=fatal)
5144 break
5145
5146 # Sometimes youtube returns a webpage with incomplete ytInitialData
5147 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5148 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5149 retry.error = ExtractorError('Incomplete yt initial data received')
5150 continue
5151
5152 return webpage, data
5153
5154 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5155 """Use if failed to extract ytcfg (and data) from initial webpage"""
5156 if not ytcfg and self.is_authenticated:
5157 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5158 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5159 raise ExtractorError(
5160 f'{msg}. If you are not downloading private content, or '
5161 'your cookies are only for the first account and channel,'
5162 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5163 expected=True)
5164 self.report_warning(msg, only_once=True)
5165
5166 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5167 data = None
5168 if not self.skip_webpage:
5169 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5170 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5171 # Reject webpage data if redirected to home page without explicitly requesting
5172 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5173 if (url != 'https://www.youtube.com/feed/recommended'
5174 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5175 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5176 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5177 if fatal:
5178 raise ExtractorError(msg, expected=True)
5179 self.report_warning(msg, only_once=True)
5180 if not data:
5181 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5182 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5183 return data, ytcfg
5184
5185 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5186 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5187 resolve_response = self._extract_response(
5188 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5189 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5190 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5191 for ep_key, ep in endpoints.items():
5192 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5193 if params:
5194 return self._extract_response(
5195 item_id=item_id, query=params, ep=ep, headers=headers,
5196 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5197 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5198 err_note = 'Failed to resolve url (does the playlist exist?)'
5199 if fatal:
5200 raise ExtractorError(err_note, expected=True)
5201 self.report_warning(err_note, item_id)
5202
5203 _SEARCH_PARAMS = None
5204
5205 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5206 data = {'query': query}
5207 if params is NO_DEFAULT:
5208 params = self._SEARCH_PARAMS
5209 if params:
5210 data['params'] = params
5211
5212 content_keys = (
5213 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5214 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5215 # ytmusic search
5216 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5217 ('continuationContents', ),
5218 )
5219 display_id = f'query "{query}"'
5220 check_get_keys = tuple({keys[0] for keys in content_keys})
5221 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5222 self._report_playlist_authcheck(ytcfg, fatal=False)
5223
5224 continuation_list = [None]
5225 search = None
5226 for page_num in itertools.count(1):
5227 data.update(continuation_list[0] or {})
5228 headers = self.generate_api_headers(
5229 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5230 search = self._extract_response(
5231 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5232 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5233 slr_contents = traverse_obj(search, *content_keys)
5234 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5235 if not continuation_list[0]:
5236 break
5237
5238
5239 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5240 IE_DESC = 'YouTube Tabs'
5241 _VALID_URL = r'''(?x:
5242 https?://
5243 (?!consent\.)(?:\w+\.)?
5244 (?:
5245 youtube(?:kids)?\.com|
5246 %(invidious)s
5247 )/
5248 (?:
5249 (?P<channel_type>channel|c|user|browse)/|
5250 (?P<not_channel>
5251 feed/|hashtag/|
5252 (?:playlist|watch)\?.*?\blist=
5253 )|
5254 (?!(?:%(reserved_names)s)\b) # Direct URLs
5255 )
5256 (?P<id>[^/?\#&]+)
5257 )''' % {
5258 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5259 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5260 }
5261 IE_NAME = 'youtube:tab'
5262
5263 _TESTS = [{
5264 'note': 'playlists, multipage',
5265 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5266 'playlist_mincount': 94,
5267 'info_dict': {
5268 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5269 'title': 'Igor Kleiner - Playlists',
5270 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5271 'uploader': 'Igor Kleiner',
5272 'uploader_id': '@IgorDataScience',
5273 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5274 'channel': 'Igor Kleiner',
5275 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5276 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5277 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5278 'channel_follower_count': int
5279 },
5280 }, {
5281 'note': 'playlists, multipage, different order',
5282 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5283 'playlist_mincount': 94,
5284 'info_dict': {
5285 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5286 'title': 'Igor Kleiner - Playlists',
5287 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5288 'uploader': 'Igor Kleiner',
5289 'uploader_id': '@IgorDataScience',
5290 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5291 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5292 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5293 'channel': 'Igor Kleiner',
5294 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5295 'channel_follower_count': int
5296 },
5297 }, {
5298 'note': 'playlists, series',
5299 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5300 'playlist_mincount': 5,
5301 'info_dict': {
5302 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5303 'title': '3Blue1Brown - Playlists',
5304 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5305 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5306 'channel': '3Blue1Brown',
5307 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5308 'uploader_id': '@3blue1brown',
5309 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5310 'uploader': '3Blue1Brown',
5311 'tags': ['Mathematics'],
5312 'channel_follower_count': int
5313 },
5314 }, {
5315 'note': 'playlists, singlepage',
5316 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5317 'playlist_mincount': 4,
5318 'info_dict': {
5319 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5320 'title': 'ThirstForScience - Playlists',
5321 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5322 'uploader': 'ThirstForScience',
5323 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5324 'uploader_id': '@ThirstForScience',
5325 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5326 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5327 'tags': 'count:13',
5328 'channel': 'ThirstForScience',
5329 'channel_follower_count': int
5330 }
5331 }, {
5332 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5333 'only_matching': True,
5334 }, {
5335 'note': 'basic, single video playlist',
5336 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5337 'info_dict': {
5338 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5339 'title': 'youtube-dl public playlist',
5340 'description': '',
5341 'tags': [],
5342 'view_count': int,
5343 'modified_date': '20201130',
5344 'channel': 'Sergey M.',
5345 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5346 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5347 'availability': 'public',
5348 'uploader': 'Sergey M.',
5349 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5350 'uploader_id': '@sergeym.6173',
5351 },
5352 'playlist_count': 1,
5353 }, {
5354 'note': 'empty playlist',
5355 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5356 'info_dict': {
5357 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5358 'title': 'youtube-dl empty playlist',
5359 'tags': [],
5360 'channel': 'Sergey M.',
5361 'description': '',
5362 'modified_date': '20160902',
5363 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5364 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5365 'availability': 'public',
5366 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5367 'uploader_id': '@sergeym.6173',
5368 'uploader': 'Sergey M.',
5369 },
5370 'playlist_count': 0,
5371 }, {
5372 'note': 'Home tab',
5373 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5374 'info_dict': {
5375 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5376 'title': 'lex will - Home',
5377 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5378 'uploader': 'lex will',
5379 'uploader_id': '@lexwill718',
5380 'channel': 'lex will',
5381 'tags': ['bible', 'history', 'prophesy'],
5382 'uploader_url': 'https://www.youtube.com/@lexwill718',
5383 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5384 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5385 'channel_follower_count': int
5386 },
5387 'playlist_mincount': 2,
5388 }, {
5389 'note': 'Videos tab',
5390 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5391 'info_dict': {
5392 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5393 'title': 'lex will - Videos',
5394 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5395 'uploader': 'lex will',
5396 'uploader_id': '@lexwill718',
5397 'tags': ['bible', 'history', 'prophesy'],
5398 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5399 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5400 'uploader_url': 'https://www.youtube.com/@lexwill718',
5401 'channel': 'lex will',
5402 'channel_follower_count': int
5403 },
5404 'playlist_mincount': 975,
5405 }, {
5406 'note': 'Videos tab, sorted by popular',
5407 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5408 'info_dict': {
5409 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5410 'title': 'lex will - Videos',
5411 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5412 'uploader': 'lex will',
5413 'uploader_id': '@lexwill718',
5414 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5415 'uploader_url': 'https://www.youtube.com/@lexwill718',
5416 'channel': 'lex will',
5417 'tags': ['bible', 'history', 'prophesy'],
5418 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5419 'channel_follower_count': int
5420 },
5421 'playlist_mincount': 199,
5422 }, {
5423 'note': 'Playlists tab',
5424 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5425 'info_dict': {
5426 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5427 'title': 'lex will - Playlists',
5428 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5429 'uploader': 'lex will',
5430 'uploader_id': '@lexwill718',
5431 'uploader_url': 'https://www.youtube.com/@lexwill718',
5432 'channel': 'lex will',
5433 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5434 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5435 'tags': ['bible', 'history', 'prophesy'],
5436 'channel_follower_count': int
5437 },
5438 'playlist_mincount': 17,
5439 }, {
5440 'note': 'Community tab',
5441 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5442 'info_dict': {
5443 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5444 'title': 'lex will - Community',
5445 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5446 'channel': 'lex will',
5447 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5448 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5449 'tags': ['bible', 'history', 'prophesy'],
5450 'channel_follower_count': int,
5451 'uploader_url': 'https://www.youtube.com/@lexwill718',
5452 'uploader_id': '@lexwill718',
5453 'uploader': 'lex will',
5454 },
5455 'playlist_mincount': 18,
5456 }, {
5457 'note': 'Channels tab',
5458 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5459 'info_dict': {
5460 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5461 'title': 'lex will - Channels',
5462 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5463 'channel': 'lex will',
5464 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5465 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5466 'tags': ['bible', 'history', 'prophesy'],
5467 'channel_follower_count': int,
5468 'uploader_url': 'https://www.youtube.com/@lexwill718',
5469 'uploader_id': '@lexwill718',
5470 'uploader': 'lex will',
5471 },
5472 'playlist_mincount': 12,
5473 }, {
5474 'note': 'Search tab',
5475 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5476 'playlist_mincount': 40,
5477 'info_dict': {
5478 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5479 'title': '3Blue1Brown - Search - linear algebra',
5480 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5481 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5482 'tags': ['Mathematics'],
5483 'channel': '3Blue1Brown',
5484 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5485 'channel_follower_count': int,
5486 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5487 'uploader_id': '@3blue1brown',
5488 'uploader': '3Blue1Brown',
5489 },
5490 }, {
5491 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5492 'only_matching': True,
5493 }, {
5494 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5495 'only_matching': True,
5496 }, {
5497 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5498 'only_matching': True,
5499 }, {
5500 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5501 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5502 'info_dict': {
5503 'title': '29C3: Not my department',
5504 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5505 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5506 'tags': [],
5507 'view_count': int,
5508 'modified_date': '20150605',
5509 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5510 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5511 'channel': 'Christiaan008',
5512 'availability': 'public',
5513 'uploader_id': '@ChRiStIaAn008',
5514 'uploader': 'Christiaan008',
5515 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5516 },
5517 'playlist_count': 96,
5518 }, {
5519 'note': 'Large playlist',
5520 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5521 'info_dict': {
5522 'title': 'Uploads from Cauchemar',
5523 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5524 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5525 'tags': [],
5526 'modified_date': r're:\d{8}',
5527 'channel': 'Cauchemar',
5528 'view_count': int,
5529 'description': '',
5530 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5531 'availability': 'public',
5532 'uploader_id': '@Cauchemar89',
5533 'uploader': 'Cauchemar',
5534 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5535 },
5536 'playlist_mincount': 1123,
5537 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5538 }, {
5539 'note': 'even larger playlist, 8832 videos',
5540 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5541 'only_matching': True,
5542 }, {
5543 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5544 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5545 'info_dict': {
5546 'title': 'Uploads from Interstellar Movie',
5547 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5548 'tags': [],
5549 'view_count': int,
5550 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5551 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5552 'channel': 'Interstellar Movie',
5553 'description': '',
5554 'modified_date': r're:\d{8}',
5555 'availability': 'public',
5556 'uploader_id': '@InterstellarMovie',
5557 'uploader': 'Interstellar Movie',
5558 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5559 },
5560 'playlist_mincount': 21,
5561 }, {
5562 'note': 'Playlist with "show unavailable videos" button',
5563 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5564 'info_dict': {
5565 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5566 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5567 'view_count': int,
5568 'channel': 'Phim Siêu Nhân Nhật Bản',
5569 'tags': [],
5570 'description': '',
5571 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5572 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5573 'modified_date': r're:\d{8}',
5574 'availability': 'public',
5575 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5576 'uploader_id': '@phimsieunhannhatban',
5577 'uploader': 'Phim Siêu Nhân Nhật Bản',
5578 },
5579 'playlist_mincount': 200,
5580 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5581 }, {
5582 'note': 'Playlist with unavailable videos in page 7',
5583 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5584 'info_dict': {
5585 'title': 'Uploads from BlankTV',
5586 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5587 'channel': 'BlankTV',
5588 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5589 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5590 'view_count': int,
5591 'tags': [],
5592 'modified_date': r're:\d{8}',
5593 'description': '',
5594 'availability': 'public',
5595 'uploader_id': '@blanktv',
5596 'uploader': 'BlankTV',
5597 'uploader_url': 'https://www.youtube.com/@blanktv',
5598 },
5599 'playlist_mincount': 1000,
5600 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5601 }, {
5602 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5603 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5604 'info_dict': {
5605 'title': 'Data Analysis with Dr Mike Pound',
5606 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5607 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5608 'tags': [],
5609 'view_count': int,
5610 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5611 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5612 'channel': 'Computerphile',
5613 'availability': 'public',
5614 'modified_date': '20190712',
5615 'uploader_id': '@Computerphile',
5616 'uploader': 'Computerphile',
5617 'uploader_url': 'https://www.youtube.com/@Computerphile',
5618 },
5619 'playlist_mincount': 11,
5620 }, {
5621 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5622 'only_matching': True,
5623 }, {
5624 'note': 'Playlist URL that does not actually serve a playlist',
5625 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5626 'info_dict': {
5627 'id': 'FqZTN594JQw',
5628 'ext': 'webm',
5629 'title': "Smiley's People 01 detective, Adventure Series, Action",
5630 'upload_date': '20150526',
5631 'license': 'Standard YouTube License',
5632 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5633 'categories': ['People & Blogs'],
5634 'tags': list,
5635 'view_count': int,
5636 'like_count': int,
5637 },
5638 'params': {
5639 'skip_download': True,
5640 },
5641 'skip': 'This video is not available.',
5642 'add_ie': [YoutubeIE.ie_key()],
5643 }, {
5644 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5645 'only_matching': True,
5646 }, {
5647 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5648 'only_matching': True,
5649 }, {
5650 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5651 'info_dict': {
5652 'id': 'AlTsmyW4auo', # This will keep changing
5653 'ext': 'mp4',
5654 'title': str,
5655 'upload_date': r're:\d{8}',
5656 'description': str,
5657 'categories': ['News & Politics'],
5658 'tags': list,
5659 'like_count': int,
5660 'release_timestamp': int,
5661 'channel': 'Sky News',
5662 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5663 'age_limit': 0,
5664 'view_count': int,
5665 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5666 'playable_in_embed': True,
5667 'release_date': r're:\d+',
5668 'availability': 'public',
5669 'live_status': 'is_live',
5670 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5671 'channel_follower_count': int,
5672 'concurrent_view_count': int,
5673 'uploader_url': 'https://www.youtube.com/@SkyNews',
5674 'uploader_id': '@SkyNews',
5675 'uploader': 'Sky News',
5676 },
5677 'params': {
5678 'skip_download': True,
5679 },
5680 'expected_warnings': ['Ignoring subtitle tracks found in '],
5681 }, {
5682 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5683 'info_dict': {
5684 'id': 'a48o2S1cPoo',
5685 'ext': 'mp4',
5686 'title': 'The Young Turks - Live Main Show',
5687 'upload_date': '20150715',
5688 'license': 'Standard YouTube License',
5689 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5690 'categories': ['News & Politics'],
5691 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5692 'like_count': int,
5693 },
5694 'params': {
5695 'skip_download': True,
5696 },
5697 'only_matching': True,
5698 }, {
5699 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5700 'only_matching': True,
5701 }, {
5702 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5703 'only_matching': True,
5704 }, {
5705 'note': 'A channel that is not live. Should raise error',
5706 'url': 'https://www.youtube.com/user/numberphile/live',
5707 'only_matching': True,
5708 }, {
5709 'url': 'https://www.youtube.com/feed/trending',
5710 'only_matching': True,
5711 }, {
5712 'url': 'https://www.youtube.com/feed/library',
5713 'only_matching': True,
5714 }, {
5715 'url': 'https://www.youtube.com/feed/history',
5716 'only_matching': True,
5717 }, {
5718 'url': 'https://www.youtube.com/feed/subscriptions',
5719 'only_matching': True,
5720 }, {
5721 'url': 'https://www.youtube.com/feed/watch_later',
5722 'only_matching': True,
5723 }, {
5724 'note': 'Recommended - redirects to home page.',
5725 'url': 'https://www.youtube.com/feed/recommended',
5726 'only_matching': True,
5727 }, {
5728 'note': 'inline playlist with not always working continuations',
5729 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5730 'only_matching': True,
5731 }, {
5732 'url': 'https://www.youtube.com/course',
5733 'only_matching': True,
5734 }, {
5735 'url': 'https://www.youtube.com/zsecurity',
5736 'only_matching': True,
5737 }, {
5738 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5739 'only_matching': True,
5740 }, {
5741 'url': 'https://www.youtube.com/TheYoungTurks/live',
5742 'only_matching': True,
5743 }, {
5744 'url': 'https://www.youtube.com/hashtag/cctv9',
5745 'info_dict': {
5746 'id': 'cctv9',
5747 'title': '#cctv9',
5748 'tags': [],
5749 },
5750 'playlist_mincount': 300, # not consistent but should be over 300
5751 }, {
5752 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5753 'only_matching': True,
5754 }, {
5755 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5756 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5757 'only_matching': True
5758 }, {
5759 'note': '/browse/ should redirect to /channel/',
5760 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5761 'only_matching': True
5762 }, {
5763 'note': 'VLPL, should redirect to playlist?list=PL...',
5764 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5765 'info_dict': {
5766 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5767 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5768 'title': 'NCS : All Releases 💿',
5769 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5770 'modified_date': r're:\d{8}',
5771 'view_count': int,
5772 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5773 'tags': [],
5774 'channel': 'NoCopyrightSounds',
5775 'availability': 'public',
5776 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5777 'uploader': 'NoCopyrightSounds',
5778 'uploader_id': '@NoCopyrightSounds',
5779 },
5780 'playlist_mincount': 166,
5781 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5782 }, {
5783 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5784 'note': 'Topic, should redirect to playlist?list=UU...',
5785 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5786 'info_dict': {
5787 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5788 'title': 'Uploads from Royalty Free Music - Topic',
5789 'tags': [],
5790 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5791 'channel': 'Royalty Free Music - Topic',
5792 'view_count': int,
5793 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5794 'modified_date': r're:\d{8}',
5795 'description': '',
5796 'availability': 'public',
5797 'uploader': 'Royalty Free Music - Topic',
5798 },
5799 'playlist_mincount': 101,
5800 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5801 }, {
5802 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5803 # Treat as a general feed
5804 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5805 'info_dict': {
5806 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5807 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5808 'tags': [],
5809 },
5810 'playlist_mincount': 9,
5811 }, {
5812 'note': 'Youtube music Album',
5813 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5814 'info_dict': {
5815 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5816 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5817 'tags': [],
5818 'view_count': int,
5819 'description': '',
5820 'availability': 'unlisted',
5821 'modified_date': r're:\d{8}',
5822 },
5823 'playlist_count': 50,
5824 'expected_warnings': ['YouTube Music is not directly supported'],
5825 }, {
5826 'note': 'unlisted single video playlist',
5827 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5828 'info_dict': {
5829 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5830 'title': 'yt-dlp unlisted playlist test',
5831 'availability': 'unlisted',
5832 'tags': [],
5833 'modified_date': '20220418',
5834 'channel': 'colethedj',
5835 'view_count': int,
5836 'description': '',
5837 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5838 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5839 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5840 'uploader_id': '@colethedj1894',
5841 'uploader': 'colethedj',
5842 },
5843 'playlist_count': 1,
5844 }, {
5845 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5846 'url': 'https://www.youtube.com/feed/recommended',
5847 'info_dict': {
5848 'id': 'recommended',
5849 'title': 'recommended',
5850 'tags': [],
5851 },
5852 'playlist_mincount': 50,
5853 'params': {
5854 'skip_download': True,
5855 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5856 },
5857 }, {
5858 'note': 'API Fallback: /videos tab, sorted by oldest first',
5859 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5860 'info_dict': {
5861 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5862 'title': 'Cody\'sLab - Videos',
5863 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5864 'channel': 'Cody\'sLab',
5865 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5866 'tags': [],
5867 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5868 'channel_follower_count': int
5869 },
5870 'playlist_mincount': 650,
5871 'params': {
5872 'skip_download': True,
5873 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5874 },
5875 'skip': 'Query for sorting no longer works',
5876 }, {
5877 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5878 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5879 'info_dict': {
5880 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5881 'title': 'Uploads from Royalty Free Music - Topic',
5882 'modified_date': r're:\d{8}',
5883 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5884 'description': '',
5885 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5886 'tags': [],
5887 'channel': 'Royalty Free Music - Topic',
5888 'view_count': int,
5889 'availability': 'public',
5890 'uploader': 'Royalty Free Music - Topic',
5891 },
5892 'playlist_mincount': 101,
5893 'params': {
5894 'skip_download': True,
5895 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5896 },
5897 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5898 }, {
5899 'note': 'non-standard redirect to regional channel',
5900 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5901 'only_matching': True
5902 }, {
5903 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5904 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5905 'info_dict': {
5906 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5907 'modified_date': '20220407',
5908 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5909 'tags': [],
5910 'availability': 'unlisted',
5911 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5912 'channel': 'pukkandan',
5913 'description': 'Test for collaborative playlist',
5914 'title': 'yt-dlp test - collaborative playlist',
5915 'view_count': int,
5916 'uploader_url': 'https://www.youtube.com/@pukkandan',
5917 'uploader_id': '@pukkandan',
5918 'uploader': 'pukkandan',
5919 },
5920 'playlist_mincount': 2
5921 }, {
5922 'note': 'translated tab name',
5923 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
5924 'info_dict': {
5925 'id': 'UCiu-3thuViMebBjw_5nWYrA',
5926 'tags': [],
5927 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5928 'description': 'test description',
5929 'title': 'cole-dlp-test-acc - 再生リスト',
5930 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5931 'channel': 'cole-dlp-test-acc',
5932 'uploader_url': 'https://www.youtube.com/@coletdjnz',
5933 'uploader_id': '@coletdjnz',
5934 'uploader': 'cole-dlp-test-acc',
5935 },
5936 'playlist_mincount': 1,
5937 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
5938 'expected_warnings': ['Preferring "ja"'],
5939 }, {
5940 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
5941 'note': 'preferred lang set with playlist with translated video titles',
5942 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
5943 'info_dict': {
5944 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
5945 'tags': [],
5946 'view_count': int,
5947 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5948 'channel': 'cole-dlp-test-acc',
5949 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5950 'description': 'test',
5951 'title': 'dlp test playlist',
5952 'availability': 'public',
5953 'uploader_url': 'https://www.youtube.com/@coletdjnz',
5954 'uploader_id': '@coletdjnz',
5955 'uploader': 'cole-dlp-test-acc',
5956 },
5957 'playlist_mincount': 1,
5958 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
5959 'expected_warnings': ['Preferring "ja"'],
5960 }, {
5961 # shorts audio pivot for 2GtVksBMYFM.
5962 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
5963 'info_dict': {
5964 'id': 'sfv_audio_pivot',
5965 'title': 'sfv_audio_pivot',
5966 'tags': [],
5967 },
5968 'playlist_mincount': 50,
5969
5970 }, {
5971 # Channel with a real live tab (not to be mistaken with streams tab)
5972 # Do not treat like it should redirect to live stream
5973 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
5974 'info_dict': {
5975 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
5976 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
5977 'tags': [],
5978 },
5979 'playlist_mincount': 20,
5980 }, {
5981 # Tab name is not the same as tab id
5982 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
5983 'info_dict': {
5984 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
5985 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
5986 'tags': [],
5987 },
5988 'playlist_mincount': 8,
5989 }, {
5990 # Home tab id is literally home. Not to get mistaken with featured
5991 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
5992 'info_dict': {
5993 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
5994 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
5995 'tags': [],
5996 },
5997 'playlist_mincount': 8,
5998 }, {
5999 # Should get three playlists for videos, shorts and streams tabs
6000 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6001 'info_dict': {
6002 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6003 'title': 'Polka Ch. 尾丸ポルカ',
6004 'channel_follower_count': int,
6005 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6006 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6007 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6008 'channel': 'Polka Ch. 尾丸ポルカ',
6009 'tags': 'count:35',
6010 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6011 'uploader': 'Polka Ch. 尾丸ポルカ',
6012 'uploader_id': '@OmaruPolka',
6013 },
6014 'playlist_count': 3,
6015 }, {
6016 # Shorts tab with channel with handle
6017 # TODO: fix channel description
6018 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6019 'info_dict': {
6020 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6021 'title': 'Not Just Bikes - Shorts',
6022 'tags': 'count:12',
6023 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6024 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6025 'channel_follower_count': int,
6026 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6027 'channel': 'Not Just Bikes',
6028 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6029 'uploader': 'Not Just Bikes',
6030 'uploader_id': '@NotJustBikes',
6031 },
6032 'playlist_mincount': 10,
6033 }, {
6034 # Streams tab
6035 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6036 'info_dict': {
6037 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6038 'title': '中村悠一 - Live',
6039 'tags': 'count:7',
6040 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6041 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6042 'channel': '中村悠一',
6043 'channel_follower_count': int,
6044 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6045 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6046 'uploader_id': '@Yuichi-Nakamura',
6047 'uploader': '中村悠一',
6048 },
6049 'playlist_mincount': 60,
6050 }, {
6051 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6052 # See test_youtube_lists
6053 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6054 'only_matching': True,
6055 }, {
6056 # No uploads and no UCID given. Should fail with no uploads error
6057 # See test_youtube_lists
6058 'url': 'https://www.youtube.com/news',
6059 'only_matching': True
6060 }, {
6061 # No videos tab but has a shorts tab
6062 'url': 'https://www.youtube.com/c/TKFShorts',
6063 'info_dict': {
6064 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6065 'title': 'Shorts Break - Shorts',
6066 'tags': 'count:48',
6067 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6068 'channel': 'Shorts Break',
6069 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6070 'channel_follower_count': int,
6071 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6072 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6073 'uploader': 'Shorts Break',
6074 'uploader_id': '@ShortsBreak_Official',
6075 },
6076 'playlist_mincount': 30,
6077 }, {
6078 # Trending Now Tab. tab id is empty
6079 'url': 'https://www.youtube.com/feed/trending',
6080 'info_dict': {
6081 'id': 'trending',
6082 'title': 'trending - Now',
6083 'tags': [],
6084 },
6085 'playlist_mincount': 30,
6086 }, {
6087 # Trending Gaming Tab. tab id is empty
6088 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6089 'info_dict': {
6090 'id': 'trending',
6091 'title': 'trending - Gaming',
6092 'tags': [],
6093 },
6094 'playlist_mincount': 30,
6095 }, {
6096 # Shorts url result in shorts tab
6097 # TODO: Fix channel id extraction
6098 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6099 'info_dict': {
6100 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6101 'title': 'cole-dlp-test-acc - Shorts',
6102 'channel': 'cole-dlp-test-acc',
6103 'description': 'test description',
6104 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6105 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6106 'tags': [],
6107 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6108 'uploader_id': '@coletdjnz',
6109 'uploader': 'cole-dlp-test-acc',
6110 },
6111 'playlist': [{
6112 'info_dict': {
6113 # Channel data is not currently available for short renderers (as of 2023-03-01)
6114 '_type': 'url',
6115 'ie_key': 'Youtube',
6116 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6117 'id': 'sSM9J5YH_60',
6118 'title': 'SHORT short',
6119 'view_count': int,
6120 'thumbnails': list,
6121 }
6122 }],
6123 'params': {'extract_flat': True},
6124 }, {
6125 # Live video status should be extracted
6126 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6127 'info_dict': {
6128 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6129 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6130 'tags': []
6131 },
6132 'playlist': [{
6133 'info_dict': {
6134 '_type': 'url',
6135 'ie_key': 'Youtube',
6136 'url': 'startswith:https://www.youtube.com/watch?v=',
6137 'id': str,
6138 'title': str,
6139 'live_status': 'is_live',
6140 'channel_id': str,
6141 'channel_url': str,
6142 'concurrent_view_count': int,
6143 'channel': str,
6144 }
6145 }],
6146 'params': {'extract_flat': True, 'playlist_items': '1'},
6147 'playlist_mincount': 1
6148 }, {
6149 # Channel renderer metadata. Contains number of videos on the channel
6150 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6151 'info_dict': {
6152 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6153 'title': 'cole-dlp-test-acc - Channels',
6154 'channel': 'cole-dlp-test-acc',
6155 'description': 'test description',
6156 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6157 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6158 'tags': [],
6159 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6160 'uploader_id': '@coletdjnz',
6161 'uploader': 'cole-dlp-test-acc',
6162 },
6163 'playlist': [{
6164 'info_dict': {
6165 '_type': 'url',
6166 'ie_key': 'YoutubeTab',
6167 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6168 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6169 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6170 'title': 'PewDiePie',
6171 'channel': 'PewDiePie',
6172 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6173 'thumbnails': list,
6174 'channel_follower_count': int,
6175 'playlist_count': int,
6176 'uploader': 'PewDiePie',
6177 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6178 'uploader_id': '@PewDiePie',
6179 }
6180 }],
6181 'params': {'extract_flat': True},
6182 }, {
6183 'url': 'https://www.youtube.com/@3blue1brown/about',
6184 'info_dict': {
6185 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6186 'tags': ['Mathematics'],
6187 'title': '3Blue1Brown - About',
6188 'channel_follower_count': int,
6189 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6190 'channel': '3Blue1Brown',
6191 'view_count': int,
6192 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6193 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6194 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6195 'uploader_id': '@3blue1brown',
6196 'uploader': '3Blue1Brown',
6197 },
6198 'playlist_count': 0,
6199 }, {
6200 # Podcasts tab, with rich entry playlistRenderers
6201 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6202 'info_dict': {
6203 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6204 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6205 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6206 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6207 'title': '99 Percent Invisible - Podcasts',
6208 'uploader': '99 Percent Invisible',
6209 'channel_follower_count': int,
6210 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6211 'tags': [],
6212 'channel': '99 Percent Invisible',
6213 'uploader_id': '@99percentinvisiblepodcast',
6214 },
6215 'playlist_count': 1,
6216 }, {
6217 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6218 'url': 'https://www.youtube.com/@AHimitsu/releases',
6219 'info_dict': {
6220 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6221 'channel': 'A Himitsu',
6222 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6223 'title': 'A Himitsu - Releases',
6224 'uploader_id': '@AHimitsu',
6225 'uploader': 'A Himitsu',
6226 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6227 'tags': 'count:16',
6228 'description': 'I make music',
6229 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6230 'channel_follower_count': int,
6231 },
6232 'playlist_mincount': 10,
6233 }]
6234
6235 @classmethod
6236 def suitable(cls, url):
6237 return False if YoutubeIE.suitable(url) else super().suitable(url)
6238
6239 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6240
6241 def _get_url_mobj(self, url):
6242 mobj = self._URL_RE.match(url).groupdict()
6243 mobj.update((k, '') for k, v in mobj.items() if v is None)
6244 return mobj
6245
6246 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6247 tab_name = (tab.get('title') or '').lower()
6248 tab_url = urljoin(base_url, traverse_obj(
6249 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6250
6251 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6252 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6253 if tab_id:
6254 return {
6255 'TAB_ID_SPONSORSHIPS': 'membership',
6256 }.get(tab_id, tab_id), tab_name
6257
6258 # Fallback to tab name if we cannot get the tab id.
6259 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6260 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6261 if tab_name:
6262 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6263 return {
6264 'home': 'featured',
6265 'live': 'streams',
6266 }.get(tab_name, tab_name), tab_name
6267
6268 def _has_tab(self, tabs, tab_id):
6269 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6270
6271 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6272 def _real_extract(self, url, smuggled_data):
6273 item_id = self._match_id(url)
6274 url = urllib.parse.urlunparse(
6275 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6276 compat_opts = self.get_param('compat_opts', [])
6277
6278 mobj = self._get_url_mobj(url)
6279 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6280 if is_channel and smuggled_data.get('is_music_url'):
6281 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6282 return self.url_result(
6283 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6284 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6285 mdata = self._extract_tab_endpoint(
6286 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6287 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6288 get_all=False, expected_type=str)
6289 if not murl:
6290 raise ExtractorError('Failed to resolve album to playlist')
6291 return self.url_result(murl, YoutubeTabIE)
6292 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6293 return self.url_result(
6294 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6295
6296 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6297 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6298 url = f'{pre}/videos{post}'
6299 if smuggled_data.get('is_music_url'):
6300 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6301
6302 # Handle both video/playlist URLs
6303 qs = parse_qs(url)
6304 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6305 if not video_id and mobj['not_channel'].startswith('watch'):
6306 if not playlist_id:
6307 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6308 raise ExtractorError('A video URL was given without video ID', expected=True)
6309 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6310 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6311 return self.url_result(
6312 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6313
6314 if not self._yes_playlist(playlist_id, video_id):
6315 return self.url_result(
6316 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6317
6318 data, ytcfg = self._extract_data(url, display_id)
6319
6320 # YouTube may provide a non-standard redirect to the regional channel
6321 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6322 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6323 redirect_url = traverse_obj(
6324 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6325 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6326 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6327 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6328 return self.url_result(redirect_url, YoutubeTabIE)
6329
6330 tabs, extra_tabs = self._extract_tab_renderers(data), []
6331 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6332 selected_tab = self._extract_selected_tab(tabs)
6333 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6334 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6335
6336 if not original_tab_id and selected_tab_name:
6337 self.to_screen('Downloading all uploads of the channel. '
6338 'To download only the videos in a specific tab, pass the tab\'s URL')
6339 if self._has_tab(tabs, 'streams'):
6340 extra_tabs.append(''.join((pre, '/streams', post)))
6341 if self._has_tab(tabs, 'shorts'):
6342 extra_tabs.append(''.join((pre, '/shorts', post)))
6343 # XXX: Members-only tab should also be extracted
6344
6345 if not extra_tabs and selected_tab_id != 'videos':
6346 # Channel does not have streams, shorts or videos tabs
6347 if item_id[:2] != 'UC':
6348 raise ExtractorError('This channel has no uploads', expected=True)
6349
6350 # Topic channels don't have /videos. Use the equivalent playlist instead
6351 pl_id = f'UU{item_id[2:]}'
6352 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6353 try:
6354 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6355 except ExtractorError:
6356 raise ExtractorError('This channel has no uploads', expected=True)
6357 else:
6358 item_id, url = pl_id, pl_url
6359 self.to_screen(
6360 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6361
6362 elif extra_tabs and selected_tab_id != 'videos':
6363 # When there are shorts/live tabs but not videos tab
6364 url, data = f'{pre}{post}', None
6365
6366 elif (original_tab_id or 'videos') != selected_tab_id:
6367 if original_tab_id == 'live':
6368 # Live tab should have redirected to the video
6369 # Except in the case the channel has an actual live tab
6370 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6371 raise UserNotLive(video_id=item_id)
6372 elif selected_tab_name:
6373 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6374
6375 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6376 url = f'{pre}{post}'
6377
6378 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6379 if 'no-youtube-unavailable-videos' not in compat_opts:
6380 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6381 self._extract_and_report_alerts(data, only_once=True)
6382
6383 tabs, entries = self._extract_tab_renderers(data), []
6384 if tabs:
6385 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6386 entries[0].update({
6387 'extractor_key': YoutubeTabIE.ie_key(),
6388 'extractor': YoutubeTabIE.IE_NAME,
6389 'webpage_url': url,
6390 })
6391 if self.get_param('playlist_items') == '0':
6392 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6393 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6394 entries.extend(map(self._real_extract, extra_tabs))
6395
6396 if len(entries) == 1:
6397 return entries[0]
6398 elif entries:
6399 metadata = self._extract_metadata_from_tabs(item_id, data)
6400 uploads_url = 'the Uploads (UU) playlist URL'
6401 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6402 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6403 self.to_screen(
6404 'Downloading as multiple playlists, separated by tabs. '
6405 f'To download as a single playlist instead, pass {uploads_url}')
6406 return self.playlist_result(entries, item_id, **metadata)
6407
6408 # Inline playlist
6409 playlist = traverse_obj(
6410 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6411 if playlist:
6412 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6413
6414 video_id = traverse_obj(
6415 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6416 if video_id:
6417 if tab != '/live': # live tab is expected to redirect to video
6418 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6419 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6420
6421 raise ExtractorError('Unable to recognize tab page')
6422
6423
6424 class YoutubePlaylistIE(InfoExtractor):
6425 IE_DESC = 'YouTube playlists'
6426 _VALID_URL = r'''(?x)(?:
6427 (?:https?://)?
6428 (?:\w+\.)?
6429 (?:
6430 (?:
6431 youtube(?:kids)?\.com|
6432 %(invidious)s
6433 )
6434 /.*?\?.*?\blist=
6435 )?
6436 (?P<id>%(playlist_id)s)
6437 )''' % {
6438 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6439 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6440 }
6441 IE_NAME = 'youtube:playlist'
6442 _TESTS = [{
6443 'note': 'issue #673',
6444 'url': 'PLBB231211A4F62143',
6445 'info_dict': {
6446 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6447 'id': 'PLBB231211A4F62143',
6448 'uploader': 'Wickman',
6449 'uploader_id': '@WickmanVT',
6450 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6451 'view_count': int,
6452 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6453 'modified_date': r're:\d{8}',
6454 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6455 'channel': 'Wickman',
6456 'tags': [],
6457 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6458 'availability': 'public',
6459 },
6460 'playlist_mincount': 29,
6461 }, {
6462 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6463 'info_dict': {
6464 'title': 'YDL_safe_search',
6465 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6466 },
6467 'playlist_count': 2,
6468 'skip': 'This playlist is private',
6469 }, {
6470 'note': 'embedded',
6471 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6472 'playlist_count': 4,
6473 'info_dict': {
6474 'title': 'JODA15',
6475 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6476 'uploader': 'milan',
6477 'uploader_id': '@milan5503',
6478 'description': '',
6479 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6480 'tags': [],
6481 'modified_date': '20140919',
6482 'view_count': int,
6483 'channel': 'milan',
6484 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6485 'uploader_url': 'https://www.youtube.com/@milan5503',
6486 'availability': 'public',
6487 },
6488 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6489 }, {
6490 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6491 'playlist_mincount': 455,
6492 'info_dict': {
6493 'title': '2018 Chinese New Singles (11/6 updated)',
6494 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6495 'uploader': 'LBK',
6496 'uploader_id': '@music_king',
6497 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6498 'channel': 'LBK',
6499 'view_count': int,
6500 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6501 'tags': [],
6502 'uploader_url': 'https://www.youtube.com/@music_king',
6503 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6504 'modified_date': r're:\d{8}',
6505 'availability': 'public',
6506 },
6507 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6508 }, {
6509 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6510 'only_matching': True,
6511 }, {
6512 # music album playlist
6513 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6514 'only_matching': True,
6515 }]
6516
6517 @classmethod
6518 def suitable(cls, url):
6519 if YoutubeTabIE.suitable(url):
6520 return False
6521 from ..utils import parse_qs
6522 qs = parse_qs(url)
6523 if qs.get('v', [None])[0]:
6524 return False
6525 return super().suitable(url)
6526
6527 def _real_extract(self, url):
6528 playlist_id = self._match_id(url)
6529 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6530 url = update_url_query(
6531 'https://www.youtube.com/playlist',
6532 parse_qs(url) or {'list': playlist_id})
6533 if is_music_url:
6534 url = smuggle_url(url, {'is_music_url': True})
6535 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6536
6537
6538 class YoutubeYtBeIE(InfoExtractor):
6539 IE_DESC = 'youtu.be'
6540 _VALID_URL = r'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{11})/*?.*?\blist=(?P<playlist_id>%(playlist_id)s)' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
6541 _TESTS = [{
6542 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6543 'info_dict': {
6544 'id': 'yeWKywCrFtk',
6545 'ext': 'mp4',
6546 'title': 'Small Scale Baler and Braiding Rugs',
6547 'uploader': 'Backus-Page House Museum',
6548 'uploader_id': '@backuspagemuseum',
6549 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6550 'upload_date': '20161008',
6551 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6552 'categories': ['Nonprofits & Activism'],
6553 'tags': list,
6554 'like_count': int,
6555 'age_limit': 0,
6556 'playable_in_embed': True,
6557 'thumbnail': r're:^https?://.*\.webp',
6558 'channel': 'Backus-Page House Museum',
6559 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6560 'live_status': 'not_live',
6561 'view_count': int,
6562 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6563 'availability': 'public',
6564 'duration': 59,
6565 'comment_count': int,
6566 'channel_follower_count': int
6567 },
6568 'params': {
6569 'noplaylist': True,
6570 'skip_download': True,
6571 },
6572 }, {
6573 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6574 'only_matching': True,
6575 }]
6576
6577 def _real_extract(self, url):
6578 mobj = self._match_valid_url(url)
6579 video_id = mobj.group('id')
6580 playlist_id = mobj.group('playlist_id')
6581 return self.url_result(
6582 update_url_query('https://www.youtube.com/watch', {
6583 'v': video_id,
6584 'list': playlist_id,
6585 'feature': 'youtu.be',
6586 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6587
6588
6589 class YoutubeLivestreamEmbedIE(InfoExtractor):
6590 IE_DESC = 'YouTube livestream embeds'
6591 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6592 _TESTS = [{
6593 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6594 'only_matching': True,
6595 }]
6596
6597 def _real_extract(self, url):
6598 channel_id = self._match_id(url)
6599 return self.url_result(
6600 f'https://www.youtube.com/channel/{channel_id}/live',
6601 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6602
6603
6604 class YoutubeYtUserIE(InfoExtractor):
6605 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6606 IE_NAME = 'youtube:user'
6607 _VALID_URL = r'ytuser:(?P<id>.+)'
6608 _TESTS = [{
6609 'url': 'ytuser:phihag',
6610 'only_matching': True,
6611 }]
6612
6613 def _real_extract(self, url):
6614 user_id = self._match_id(url)
6615 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6616
6617
6618 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6619 IE_NAME = 'youtube:favorites'
6620 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6621 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6622 _LOGIN_REQUIRED = True
6623 _TESTS = [{
6624 'url': ':ytfav',
6625 'only_matching': True,
6626 }, {
6627 'url': ':ytfavorites',
6628 'only_matching': True,
6629 }]
6630
6631 def _real_extract(self, url):
6632 return self.url_result(
6633 'https://www.youtube.com/playlist?list=LL',
6634 ie=YoutubeTabIE.ie_key())
6635
6636
6637 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6638 IE_NAME = 'youtube:notif'
6639 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6640 _VALID_URL = r':ytnotif(?:ication)?s?'
6641 _LOGIN_REQUIRED = True
6642 _TESTS = [{
6643 'url': ':ytnotif',
6644 'only_matching': True,
6645 }, {
6646 'url': ':ytnotifications',
6647 'only_matching': True,
6648 }]
6649
6650 def _extract_notification_menu(self, response, continuation_list):
6651 notification_list = traverse_obj(
6652 response,
6653 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6654 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6655 expected_type=list) or []
6656 continuation_list[0] = None
6657 for item in notification_list:
6658 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6659 if entry:
6660 yield entry
6661 continuation = item.get('continuationItemRenderer')
6662 if continuation:
6663 continuation_list[0] = continuation
6664
6665 def _extract_notification_renderer(self, notification):
6666 video_id = traverse_obj(
6667 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6668 url = f'https://www.youtube.com/watch?v={video_id}'
6669 channel_id = None
6670 if not video_id:
6671 browse_ep = traverse_obj(
6672 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6673 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6674 post_id = self._search_regex(
6675 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6676 'post id', default=None)
6677 if not channel_id or not post_id:
6678 return
6679 # The direct /post url redirects to this in the browser
6680 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6681
6682 channel = traverse_obj(
6683 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6684 expected_type=str)
6685 notification_title = self._get_text(notification, 'shortMessage')
6686 if notification_title:
6687 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6688 # TODO: handle recommended videos
6689 title = self._search_regex(
6690 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6691 'video title', default=None)
6692 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6693 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6694 else None)
6695 return {
6696 '_type': 'url',
6697 'url': url,
6698 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6699 'video_id': video_id,
6700 'title': title,
6701 'channel_id': channel_id,
6702 'channel': channel,
6703 'uploader': channel,
6704 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6705 'timestamp': timestamp,
6706 }
6707
6708 def _notification_menu_entries(self, ytcfg):
6709 continuation_list = [None]
6710 response = None
6711 for page in itertools.count(1):
6712 ctoken = traverse_obj(
6713 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6714 response = self._extract_response(
6715 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6716 ep='notification/get_notification_menu', check_get_keys='actions',
6717 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6718 yield from self._extract_notification_menu(response, continuation_list)
6719 if not continuation_list[0]:
6720 break
6721
6722 def _real_extract(self, url):
6723 display_id = 'notifications'
6724 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6725 self._report_playlist_authcheck(ytcfg)
6726 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6727
6728
6729 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6730 IE_DESC = 'YouTube search'
6731 IE_NAME = 'youtube:search'
6732 _SEARCH_KEY = 'ytsearch'
6733 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6734 _TESTS = [{
6735 'url': 'ytsearch5:youtube-dl test video',
6736 'playlist_count': 5,
6737 'info_dict': {
6738 'id': 'youtube-dl test video',
6739 'title': 'youtube-dl test video',
6740 }
6741 }]
6742
6743
6744 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6745 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6746 _SEARCH_KEY = 'ytsearchdate'
6747 IE_DESC = 'YouTube search, newest videos first'
6748 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6749 _TESTS = [{
6750 'url': 'ytsearchdate5:youtube-dl test video',
6751 'playlist_count': 5,
6752 'info_dict': {
6753 'id': 'youtube-dl test video',
6754 'title': 'youtube-dl test video',
6755 }
6756 }]
6757
6758
6759 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6760 IE_DESC = 'YouTube search URLs with sorting and filter support'
6761 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6762 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6763 _TESTS = [{
6764 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6765 'playlist_mincount': 5,
6766 'info_dict': {
6767 'id': 'youtube-dl test video',
6768 'title': 'youtube-dl test video',
6769 }
6770 }, {
6771 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6772 'playlist_mincount': 5,
6773 'info_dict': {
6774 'id': 'python',
6775 'title': 'python',
6776 }
6777 }, {
6778 'url': 'https://www.youtube.com/results?search_query=%23cats',
6779 'playlist_mincount': 1,
6780 'info_dict': {
6781 'id': '#cats',
6782 'title': '#cats',
6783 # The test suite does not have support for nested playlists
6784 # 'entries': [{
6785 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6786 # 'title': '#cats',
6787 # }],
6788 },
6789 }, {
6790 # Channel results
6791 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
6792 'info_dict': {
6793 'id': 'kurzgesagt',
6794 'title': 'kurzgesagt',
6795 },
6796 'playlist': [{
6797 'info_dict': {
6798 '_type': 'url',
6799 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6800 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6801 'ie_key': 'YoutubeTab',
6802 'channel': 'Kurzgesagt – In a Nutshell',
6803 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
6804 'title': 'Kurzgesagt – In a Nutshell',
6805 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6806 'playlist_count': int, # XXX: should have a way of saying > 1
6807 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6808 'thumbnails': list,
6809 'uploader_id': '@kurzgesagt',
6810 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
6811 'uploader': 'Kurzgesagt – In a Nutshell',
6812 }
6813 }],
6814 'params': {'extract_flat': True, 'playlist_items': '1'},
6815 'playlist_mincount': 1,
6816 }, {
6817 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
6818 'only_matching': True,
6819 }]
6820
6821 def _real_extract(self, url):
6822 qs = parse_qs(url)
6823 query = (qs.get('search_query') or qs.get('q'))[0]
6824 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
6825
6826
6827 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
6828 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
6829 IE_NAME = 'youtube:music:search_url'
6830 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6831 _TESTS = [{
6832 'url': 'https://music.youtube.com/search?q=royalty+free+music',
6833 'playlist_count': 16,
6834 'info_dict': {
6835 'id': 'royalty free music',
6836 'title': 'royalty free music',
6837 }
6838 }, {
6839 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
6840 'playlist_mincount': 30,
6841 'info_dict': {
6842 'id': 'royalty free music - songs',
6843 'title': 'royalty free music - songs',
6844 },
6845 'params': {'extract_flat': 'in_playlist'}
6846 }, {
6847 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
6848 'playlist_mincount': 30,
6849 'info_dict': {
6850 'id': 'royalty free music - community playlists',
6851 'title': 'royalty free music - community playlists',
6852 },
6853 'params': {'extract_flat': 'in_playlist'}
6854 }]
6855
6856 _SECTIONS = {
6857 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
6858 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
6859 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
6860 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
6861 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
6862 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
6863 }
6864
6865 def _real_extract(self, url):
6866 qs = parse_qs(url)
6867 query = (qs.get('search_query') or qs.get('q'))[0]
6868 params = qs.get('sp', (None,))[0]
6869 if params:
6870 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
6871 else:
6872 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
6873 params = self._SECTIONS.get(section)
6874 if not params:
6875 section = None
6876 title = join_nonempty(query, section, delim=' - ')
6877 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
6878
6879
6880 class YoutubeFeedsInfoExtractor(InfoExtractor):
6881 """
6882 Base class for feed extractors
6883 Subclasses must re-define the _FEED_NAME property.
6884 """
6885 _LOGIN_REQUIRED = True
6886 _FEED_NAME = 'feeds'
6887
6888 def _real_initialize(self):
6889 YoutubeBaseInfoExtractor._check_login_required(self)
6890
6891 @classproperty
6892 def IE_NAME(self):
6893 return f'youtube:{self._FEED_NAME}'
6894
6895 def _real_extract(self, url):
6896 return self.url_result(
6897 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
6898
6899
6900 class YoutubeWatchLaterIE(InfoExtractor):
6901 IE_NAME = 'youtube:watchlater'
6902 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
6903 _VALID_URL = r':ytwatchlater'
6904 _TESTS = [{
6905 'url': ':ytwatchlater',
6906 'only_matching': True,
6907 }]
6908
6909 def _real_extract(self, url):
6910 return self.url_result(
6911 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
6912
6913
6914 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
6915 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
6916 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
6917 _FEED_NAME = 'recommended'
6918 _LOGIN_REQUIRED = False
6919 _TESTS = [{
6920 'url': ':ytrec',
6921 'only_matching': True,
6922 }, {
6923 'url': ':ytrecommended',
6924 'only_matching': True,
6925 }, {
6926 'url': 'https://youtube.com',
6927 'only_matching': True,
6928 }]
6929
6930
6931 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
6932 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
6933 _VALID_URL = r':ytsub(?:scription)?s?'
6934 _FEED_NAME = 'subscriptions'
6935 _TESTS = [{
6936 'url': ':ytsubs',
6937 'only_matching': True,
6938 }, {
6939 'url': ':ytsubscriptions',
6940 'only_matching': True,
6941 }]
6942
6943
6944 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
6945 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
6946 _VALID_URL = r':ythis(?:tory)?'
6947 _FEED_NAME = 'history'
6948 _TESTS = [{
6949 'url': ':ythistory',
6950 'only_matching': True,
6951 }]
6952
6953
6954 class YoutubeStoriesIE(InfoExtractor):
6955 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
6956 IE_NAME = 'youtube:stories'
6957 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
6958 _TESTS = [{
6959 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
6960 'only_matching': True,
6961 }]
6962
6963 def _real_extract(self, url):
6964 playlist_id = f'RLTD{self._match_id(url)}'
6965 return self.url_result(
6966 smuggle_url(f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1', {'is_story': True}),
6967 ie=YoutubeTabIE, video_id=playlist_id)
6968
6969
6970 class YoutubeShortsAudioPivotIE(InfoExtractor):
6971 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
6972 IE_NAME = 'youtube:shorts:pivot:audio'
6973 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
6974 _TESTS = [{
6975 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
6976 'only_matching': True,
6977 }]
6978
6979 @staticmethod
6980 def _generate_audio_pivot_params(video_id):
6981 """
6982 Generates sfv_audio_pivot browse params for this video id
6983 """
6984 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
6985 return urllib.parse.quote(base64.b64encode(pb_params).decode())
6986
6987 def _real_extract(self, url):
6988 video_id = self._match_id(url)
6989 return self.url_result(
6990 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
6991 ie=YoutubeTabIE)
6992
6993
6994 class YoutubeTruncatedURLIE(InfoExtractor):
6995 IE_NAME = 'youtube:truncated_url'
6996 IE_DESC = False # Do not list
6997 _VALID_URL = r'''(?x)
6998 (?:https?://)?
6999 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7000 (?:watch\?(?:
7001 feature=[a-z_]+|
7002 annotation_id=annotation_[^&]+|
7003 x-yt-cl=[0-9]+|
7004 hl=[^&]*|
7005 t=[0-9]+
7006 )?
7007 |
7008 attribution_link\?a=[^&]+
7009 )
7010 $
7011 '''
7012
7013 _TESTS = [{
7014 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7015 'only_matching': True,
7016 }, {
7017 'url': 'https://www.youtube.com/watch?',
7018 'only_matching': True,
7019 }, {
7020 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7021 'only_matching': True,
7022 }, {
7023 'url': 'https://www.youtube.com/watch?feature=foo',
7024 'only_matching': True,
7025 }, {
7026 'url': 'https://www.youtube.com/watch?hl=en-GB',
7027 'only_matching': True,
7028 }, {
7029 'url': 'https://www.youtube.com/watch?t=2372',
7030 'only_matching': True,
7031 }]
7032
7033 def _real_extract(self, url):
7034 raise ExtractorError(
7035 'Did you forget to quote the URL? Remember that & is a meta '
7036 'character in most shells, so you want to put the URL in quotes, '
7037 'like youtube-dl '
7038 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7039 ' or simply youtube-dl BaW_jenozKc .',
7040 expected=True)
7041
7042
7043 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7044 IE_NAME = 'youtube:clip'
7045 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7046 _TESTS = [{
7047 # FIXME: Other metadata should be extracted from the clip, not from the base video
7048 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7049 'info_dict': {
7050 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7051 'ext': 'mp4',
7052 'section_start': 29.0,
7053 'section_end': 39.7,
7054 'duration': 10.7,
7055 'age_limit': 0,
7056 'availability': 'public',
7057 'categories': ['Gaming'],
7058 'channel': 'Scott The Woz',
7059 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7060 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7061 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7062 'like_count': int,
7063 'playable_in_embed': True,
7064 'tags': 'count:17',
7065 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7066 'title': 'Mobile Games on Console - Scott The Woz',
7067 'upload_date': '20210920',
7068 'uploader': 'Scott The Woz',
7069 'uploader_id': '@ScottTheWoz',
7070 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7071 'view_count': int,
7072 'live_status': 'not_live',
7073 'channel_follower_count': int,
7074 'chapters': 'count:20',
7075 }
7076 }]
7077
7078 def _real_extract(self, url):
7079 clip_id = self._match_id(url)
7080 _, data = self._extract_webpage(url, clip_id)
7081
7082 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7083 if not video_id:
7084 raise ExtractorError('Unable to find video ID')
7085
7086 clip_data = traverse_obj(data, (
7087 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7088 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7089 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7090 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7091
7092 return {
7093 '_type': 'url_transparent',
7094 'url': f'https://www.youtube.com/watch?v={video_id}',
7095 'ie_key': YoutubeIE.ie_key(),
7096 'id': clip_id,
7097 'section_start': int(clip_data['startTimeMs']) / 1000,
7098 'section_end': int(clip_data['endTimeMs']) / 1000,
7099 }
7100
7101
7102 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7103 IE_NAME = 'youtube:consent'
7104 IE_DESC = False # Do not list
7105 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7106 _TESTS = [{
7107 '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',
7108 'info_dict': {
7109 'id': 'qVv6vCqciTM',
7110 'ext': 'mp4',
7111 'age_limit': 0,
7112 'uploader_id': '@sana_natori',
7113 'comment_count': int,
7114 'chapters': 'count:13',
7115 'upload_date': '20221223',
7116 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7117 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7118 'uploader_url': 'https://www.youtube.com/@sana_natori',
7119 'like_count': int,
7120 'release_date': '20221223',
7121 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7122 '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】',
7123 'view_count': int,
7124 'playable_in_embed': True,
7125 'duration': 4438,
7126 'availability': 'public',
7127 'channel_follower_count': int,
7128 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7129 'categories': ['Entertainment'],
7130 'live_status': 'was_live',
7131 'release_timestamp': 1671793345,
7132 'channel': 'さなちゃんねる',
7133 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7134 'uploader': 'さなちゃんねる',
7135 },
7136 'add_ie': ['Youtube'],
7137 'params': {'skip_download': 'Youtube'},
7138 }]
7139
7140 def _real_extract(self, url):
7141 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7142 if not redirect_url:
7143 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7144 return self.url_result(redirect_url)
7145
7146
7147 class YoutubeTruncatedIDIE(InfoExtractor):
7148 IE_NAME = 'youtube:truncated_id'
7149 IE_DESC = False # Do not list
7150 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7151
7152 _TESTS = [{
7153 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7154 'only_matching': True,
7155 }]
7156
7157 def _real_extract(self, url):
7158 video_id = self._match_id(url)
7159 raise ExtractorError(
7160 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7161 expected=True)