]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
[extractor/youtube] Support shorter relative time format (#7191)
[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 STREAMING_DATA_CLIENT_NAME = '__yt_dlp_client'
70 # any clients starting with _ cannot be explicitly requested by the user
71 INNERTUBE_CLIENTS = {
72 'web': {
73 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
74 'INNERTUBE_CONTEXT': {
75 'client': {
76 'clientName': 'WEB',
77 'clientVersion': '2.20220801.00.00',
78 }
79 },
80 'INNERTUBE_CONTEXT_CLIENT_NAME': 1
81 },
82 'web_embedded': {
83 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
84 'INNERTUBE_CONTEXT': {
85 'client': {
86 'clientName': 'WEB_EMBEDDED_PLAYER',
87 'clientVersion': '1.20220731.00.00',
88 },
89 },
90 'INNERTUBE_CONTEXT_CLIENT_NAME': 56
91 },
92 'web_music': {
93 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30',
94 'INNERTUBE_HOST': 'music.youtube.com',
95 'INNERTUBE_CONTEXT': {
96 'client': {
97 'clientName': 'WEB_REMIX',
98 'clientVersion': '1.20220727.01.00',
99 }
100 },
101 'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
102 },
103 'web_creator': {
104 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo',
105 'INNERTUBE_CONTEXT': {
106 'client': {
107 'clientName': 'WEB_CREATOR',
108 'clientVersion': '1.20220726.00.00',
109 }
110 },
111 'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
112 },
113 'android': {
114 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w',
115 'INNERTUBE_CONTEXT': {
116 'client': {
117 'clientName': 'ANDROID',
118 'clientVersion': '17.31.35',
119 'androidSdkVersion': 30,
120 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
121 }
122 },
123 'INNERTUBE_CONTEXT_CLIENT_NAME': 3,
124 'REQUIRE_JS_PLAYER': False
125 },
126 'android_embedded': {
127 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw',
128 'INNERTUBE_CONTEXT': {
129 'client': {
130 'clientName': 'ANDROID_EMBEDDED_PLAYER',
131 'clientVersion': '17.31.35',
132 'androidSdkVersion': 30,
133 'userAgent': 'com.google.android.youtube/17.31.35 (Linux; U; Android 11) gzip'
134 },
135 },
136 'INNERTUBE_CONTEXT_CLIENT_NAME': 55,
137 'REQUIRE_JS_PLAYER': False
138 },
139 'android_music': {
140 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI',
141 'INNERTUBE_CONTEXT': {
142 'client': {
143 'clientName': 'ANDROID_MUSIC',
144 'clientVersion': '5.16.51',
145 'androidSdkVersion': 30,
146 'userAgent': 'com.google.android.apps.youtube.music/5.16.51 (Linux; U; Android 11) gzip'
147 }
148 },
149 'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
150 'REQUIRE_JS_PLAYER': False
151 },
152 'android_creator': {
153 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8',
154 'INNERTUBE_CONTEXT': {
155 'client': {
156 'clientName': 'ANDROID_CREATOR',
157 'clientVersion': '22.30.100',
158 'androidSdkVersion': 30,
159 'userAgent': 'com.google.android.apps.youtube.creator/22.30.100 (Linux; U; Android 11) gzip'
160 },
161 },
162 'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
163 'REQUIRE_JS_PLAYER': False
164 },
165 # iOS clients have HLS live streams. Setting device model to get 60fps formats.
166 # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
167 'ios': {
168 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc',
169 'INNERTUBE_CONTEXT': {
170 'client': {
171 'clientName': 'IOS',
172 'clientVersion': '17.33.2',
173 'deviceModel': 'iPhone14,3',
174 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
175 }
176 },
177 'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
178 'REQUIRE_JS_PLAYER': False
179 },
180 'ios_embedded': {
181 'INNERTUBE_CONTEXT': {
182 'client': {
183 'clientName': 'IOS_MESSAGES_EXTENSION',
184 'clientVersion': '17.33.2',
185 'deviceModel': 'iPhone14,3',
186 'userAgent': 'com.google.ios.youtube/17.33.2 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
187 },
188 },
189 'INNERTUBE_CONTEXT_CLIENT_NAME': 66,
190 'REQUIRE_JS_PLAYER': False
191 },
192 'ios_music': {
193 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s',
194 'INNERTUBE_CONTEXT': {
195 'client': {
196 'clientName': 'IOS_MUSIC',
197 'clientVersion': '5.21',
198 'deviceModel': 'iPhone14,3',
199 'userAgent': 'com.google.ios.youtubemusic/5.21 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
200 },
201 },
202 'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
203 'REQUIRE_JS_PLAYER': False
204 },
205 'ios_creator': {
206 'INNERTUBE_CONTEXT': {
207 'client': {
208 'clientName': 'IOS_CREATOR',
209 'clientVersion': '22.33.101',
210 'deviceModel': 'iPhone14,3',
211 'userAgent': 'com.google.ios.ytcreator/22.33.101 (iPhone14,3; U; CPU iOS 15_6 like Mac OS X)'
212 },
213 },
214 'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
215 'REQUIRE_JS_PLAYER': False
216 },
217 # mweb has 'ultralow' formats
218 # See: https://github.com/yt-dlp/yt-dlp/pull/557
219 'mweb': {
220 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
221 'INNERTUBE_CONTEXT': {
222 'client': {
223 'clientName': 'MWEB',
224 'clientVersion': '2.20220801.00.00',
225 }
226 },
227 'INNERTUBE_CONTEXT_CLIENT_NAME': 2
228 },
229 # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
230 # See: https://github.com/zerodytrash/YouTube-Internal-Clients
231 'tv_embedded': {
232 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8',
233 'INNERTUBE_CONTEXT': {
234 'client': {
235 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER',
236 'clientVersion': '2.0',
237 },
238 },
239 'INNERTUBE_CONTEXT_CLIENT_NAME': 85
240 },
241 }
242
243
244 def _split_innertube_client(client_name):
245 variant, *base = client_name.rsplit('.', 1)
246 if base:
247 return variant, base[0], variant
248 base, *variant = client_name.split('_', 1)
249 return client_name, base, variant[0] if variant else None
250
251
252 def short_client_name(client_name):
253 main, *parts = _split_innertube_client(client_name)[0].replace('embedscreen', 'e_s').split('_')
254 return join_nonempty(main[:4], ''.join(x[0] for x in parts)).upper()
255
256
257 def build_innertube_clients():
258 THIRD_PARTY = {
259 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
260 }
261 BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
262 priority = qualities(BASE_CLIENTS[::-1])
263
264 for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
265 ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8')
266 ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
267 ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
268 ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
269
270 _, base_client, variant = _split_innertube_client(client)
271 ytcfg['priority'] = 10 * priority(base_client)
272
273 if not variant:
274 INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg)
275 embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED'
276 embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
277 embedscreen['priority'] -= 3
278 elif variant == 'embedded':
279 ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY
280 ytcfg['priority'] -= 2
281 else:
282 ytcfg['priority'] -= 3
283
284
285 build_innertube_clients()
286
287
288 class BadgeType(enum.Enum):
289 AVAILABILITY_UNLISTED = enum.auto()
290 AVAILABILITY_PRIVATE = enum.auto()
291 AVAILABILITY_PUBLIC = enum.auto()
292 AVAILABILITY_PREMIUM = enum.auto()
293 AVAILABILITY_SUBSCRIPTION = enum.auto()
294 LIVE_NOW = enum.auto()
295
296
297 class YoutubeBaseInfoExtractor(InfoExtractor):
298 """Provide base functions for Youtube extractors"""
299
300 _RESERVED_NAMES = (
301 r'channel|c|user|playlist|watch|w|v|embed|e|live|watch_popup|clip|'
302 r'shorts|movies|results|search|shared|hashtag|trending|explore|feed|feeds|'
303 r'browse|oembed|get_video_info|iframe_api|s/player|source|'
304 r'storefront|oops|index|account|t/terms|about|upload|signin|logout')
305
306 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
307
308 # _NETRC_MACHINE = 'youtube'
309
310 # If True it will raise an error if no login info is provided
311 _LOGIN_REQUIRED = False
312
313 _INVIDIOUS_SITES = (
314 # invidious-redirect websites
315 r'(?:www\.)?redirect\.invidious\.io',
316 r'(?:(?:www|dev)\.)?invidio\.us',
317 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/docs/instances.md
318 r'(?:www\.)?invidious\.pussthecat\.org',
319 r'(?:www\.)?invidious\.zee\.li',
320 r'(?:www\.)?invidious\.ethibox\.fr',
321 r'(?:www\.)?iv\.ggtyler\.dev',
322 r'(?:www\.)?inv\.vern\.i2p',
323 r'(?:www\.)?am74vkcrjp2d5v36lcdqgsj2m6x36tbrkhsruoegwfcizzabnfgf5zyd\.onion',
324 r'(?:www\.)?inv\.riverside\.rocks',
325 r'(?:www\.)?invidious\.silur\.me',
326 r'(?:www\.)?inv\.bp\.projectsegfau\.lt',
327 r'(?:www\.)?invidious\.g4c3eya4clenolymqbpgwz3q3tawoxw56yhzk4vugqrl6dtu3ejvhjid\.onion',
328 r'(?:www\.)?invidious\.slipfox\.xyz',
329 r'(?:www\.)?invidious\.esmail5pdn24shtvieloeedh7ehz3nrwcdivnfhfcedl7gf4kwddhkqd\.onion',
330 r'(?:www\.)?inv\.vernccvbvyi5qhfzyqengccj7lkove6bjot2xhh5kajhwvidqafczrad\.onion',
331 r'(?:www\.)?invidious\.tiekoetter\.com',
332 r'(?:www\.)?iv\.odysfvr23q5wgt7i456o5t3trw2cw5dgn56vbjfbq2m7xsc5vqbqpcyd\.onion',
333 r'(?:www\.)?invidious\.nerdvpn\.de',
334 r'(?:www\.)?invidious\.weblibre\.org',
335 r'(?:www\.)?inv\.odyssey346\.dev',
336 r'(?:www\.)?invidious\.dhusch\.de',
337 r'(?:www\.)?iv\.melmac\.space',
338 r'(?:www\.)?watch\.thekitty\.zone',
339 r'(?:www\.)?invidious\.privacydev\.net',
340 r'(?:www\.)?ng27owmagn5amdm7l5s3rsqxwscl5ynppnis5dqcasogkyxcfqn7psid\.onion',
341 r'(?:www\.)?invidious\.drivet\.xyz',
342 r'(?:www\.)?vid\.priv\.au',
343 r'(?:www\.)?euxxcnhsynwmfidvhjf6uzptsmh4dipkmgdmcmxxuo7tunp3ad2jrwyd\.onion',
344 r'(?:www\.)?inv\.vern\.cc',
345 r'(?:www\.)?invidious\.esmailelbob\.xyz',
346 r'(?:www\.)?invidious\.sethforprivacy\.com',
347 r'(?:www\.)?yt\.oelrichsgarcia\.de',
348 r'(?:www\.)?yt\.artemislena\.eu',
349 r'(?:www\.)?invidious\.flokinet\.to',
350 r'(?:www\.)?invidious\.baczek\.me',
351 r'(?:www\.)?y\.com\.sb',
352 r'(?:www\.)?invidious\.epicsite\.xyz',
353 r'(?:www\.)?invidious\.lidarshield\.cloud',
354 r'(?:www\.)?yt\.funami\.tech',
355 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
356 r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion',
357 r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.onion',
358 # youtube-dl invidious instances list
359 r'(?:(?:www|no)\.)?invidiou\.sh',
360 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
361 r'(?:www\.)?invidious\.kabi\.tk',
362 r'(?:www\.)?invidious\.mastodon\.host',
363 r'(?:www\.)?invidious\.zapashcanon\.fr',
364 r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks',
365 r'(?:www\.)?invidious\.tinfoil-hat\.net',
366 r'(?:www\.)?invidious\.himiko\.cloud',
367 r'(?:www\.)?invidious\.reallyancient\.tech',
368 r'(?:www\.)?invidious\.tube',
369 r'(?:www\.)?invidiou\.site',
370 r'(?:www\.)?invidious\.site',
371 r'(?:www\.)?invidious\.xyz',
372 r'(?:www\.)?invidious\.nixnet\.xyz',
373 r'(?:www\.)?invidious\.048596\.xyz',
374 r'(?:www\.)?invidious\.drycat\.fr',
375 r'(?:www\.)?inv\.skyn3t\.in',
376 r'(?:www\.)?tube\.poal\.co',
377 r'(?:www\.)?tube\.connect\.cafe',
378 r'(?:www\.)?vid\.wxzm\.sx',
379 r'(?:www\.)?vid\.mint\.lgbt',
380 r'(?:www\.)?vid\.puffyan\.us',
381 r'(?:www\.)?yewtu\.be',
382 r'(?:www\.)?yt\.elukerio\.org',
383 r'(?:www\.)?yt\.lelux\.fi',
384 r'(?:www\.)?invidious\.ggc-project\.de',
385 r'(?:www\.)?yt\.maisputain\.ovh',
386 r'(?:www\.)?ytprivate\.com',
387 r'(?:www\.)?invidious\.13ad\.de',
388 r'(?:www\.)?invidious\.toot\.koeln',
389 r'(?:www\.)?invidious\.fdn\.fr',
390 r'(?:www\.)?watch\.nettohikari\.com',
391 r'(?:www\.)?invidious\.namazso\.eu',
392 r'(?:www\.)?invidious\.silkky\.cloud',
393 r'(?:www\.)?invidious\.exonip\.de',
394 r'(?:www\.)?invidious\.riverside\.rocks',
395 r'(?:www\.)?invidious\.blamefran\.net',
396 r'(?:www\.)?invidious\.moomoo\.de',
397 r'(?:www\.)?ytb\.trom\.tf',
398 r'(?:www\.)?yt\.cyberhost\.uk',
399 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
400 r'(?:www\.)?qklhadlycap4cnod\.onion',
401 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
402 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
403 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
404 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
405 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
406 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
407 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
408 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
409 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
410 r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
411 # piped instances from https://github.com/TeamPiped/Piped/wiki/Instances
412 r'(?:www\.)?piped\.kavin\.rocks',
413 r'(?:www\.)?piped\.tokhmi\.xyz',
414 r'(?:www\.)?piped\.syncpundit\.io',
415 r'(?:www\.)?piped\.mha\.fi',
416 r'(?:www\.)?watch\.whatever\.social',
417 r'(?:www\.)?piped\.garudalinux\.org',
418 r'(?:www\.)?piped\.rivo\.lol',
419 r'(?:www\.)?piped-libre\.kavin\.rocks',
420 r'(?:www\.)?yt\.jae\.fi',
421 r'(?:www\.)?piped\.mint\.lgbt',
422 r'(?:www\.)?il\.ax',
423 r'(?:www\.)?piped\.esmailelbob\.xyz',
424 r'(?:www\.)?piped\.projectsegfau\.lt',
425 r'(?:www\.)?piped\.privacydev\.net',
426 r'(?:www\.)?piped\.palveluntarjoaja\.eu',
427 r'(?:www\.)?piped\.smnz\.de',
428 r'(?:www\.)?piped\.adminforge\.de',
429 r'(?:www\.)?watch\.whatevertinfoil\.de',
430 r'(?:www\.)?piped\.qdi\.fi',
431 r'(?:www\.)?piped\.video',
432 r'(?:www\.)?piped\.aeong\.one',
433 r'(?:www\.)?piped\.moomoo\.me',
434 r'(?:www\.)?piped\.chauvet\.pro',
435 r'(?:www\.)?watch\.leptons\.xyz',
436 r'(?:www\.)?pd\.vern\.cc',
437 r'(?:www\.)?piped\.hostux\.net',
438 r'(?:www\.)?piped\.lunar\.icu',
439 # Hyperpipe instances from https://hyperpipe.codeberg.page/
440 r'(?:www\.)?hyperpipe\.surge\.sh',
441 r'(?:www\.)?hyperpipe\.esmailelbob\.xyz',
442 r'(?:www\.)?listen\.whatever\.social',
443 r'(?:www\.)?music\.adminforge\.de',
444 )
445
446 # extracted from account/account_menu ep
447 # XXX: These are the supported YouTube UI and API languages,
448 # which is slightly different from languages supported for translation in YouTube studio
449 _SUPPORTED_LANG_CODES = [
450 'af', 'az', 'id', 'ms', 'bs', 'ca', 'cs', 'da', 'de', 'et', 'en-IN', 'en-GB', 'en', 'es',
451 'es-419', 'es-US', 'eu', 'fil', 'fr', 'fr-CA', 'gl', 'hr', 'zu', 'is', 'it', 'sw', 'lv',
452 'lt', 'hu', 'nl', 'no', 'uz', 'pl', 'pt-PT', 'pt', 'ro', 'sq', 'sk', 'sl', 'sr-Latn', 'fi',
453 'sv', 'vi', 'tr', 'be', 'bg', 'ky', 'kk', 'mk', 'mn', 'ru', 'sr', 'uk', 'el', 'hy', 'iw',
454 'ur', 'ar', 'fa', 'ne', 'mr', 'hi', 'as', 'bn', 'pa', 'gu', 'or', 'ta', 'te', 'kn', 'ml',
455 'si', 'th', 'lo', 'my', 'ka', 'am', 'km', 'zh-CN', 'zh-TW', 'zh-HK', 'ja', 'ko'
456 ]
457
458 _IGNORED_WARNINGS = {'Unavailable videos will be hidden during playback'}
459
460 _YT_HANDLE_RE = r'@[\w.-]{3,30}' # https://support.google.com/youtube/answer/11585688?hl=en
461 _YT_CHANNEL_UCID_RE = r'UC[\w-]{22}'
462
463 def ucid_or_none(self, ucid):
464 return self._search_regex(rf'^({self._YT_CHANNEL_UCID_RE})$', ucid, 'UC-id', default=None)
465
466 def handle_or_none(self, handle):
467 return self._search_regex(rf'^({self._YT_HANDLE_RE})$', handle, '@-handle', default=None)
468
469 def handle_from_url(self, url):
470 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_HANDLE_RE})',
471 url, 'channel handle', default=None)
472
473 def ucid_from_url(self, url):
474 return self._search_regex(rf'^(?:https?://(?:www\.)?youtube\.com)?/({self._YT_CHANNEL_UCID_RE})',
475 url, 'channel id', default=None)
476
477 @functools.cached_property
478 def _preferred_lang(self):
479 """
480 Returns a language code supported by YouTube for the user preferred language.
481 Returns None if no preferred language set.
482 """
483 preferred_lang = self._configuration_arg('lang', ie_key='Youtube', casesense=True, default=[''])[0]
484 if not preferred_lang:
485 return
486 if preferred_lang not in self._SUPPORTED_LANG_CODES:
487 raise ExtractorError(
488 f'Unsupported language code: {preferred_lang}. Supported language codes (case-sensitive): {join_nonempty(*self._SUPPORTED_LANG_CODES, delim=", ")}.',
489 expected=True)
490 elif preferred_lang != 'en':
491 self.report_warning(
492 f'Preferring "{preferred_lang}" translated fields. Note that some metadata extraction may fail or be incorrect.')
493 return preferred_lang
494
495 def _initialize_consent(self):
496 cookies = self._get_cookies('https://www.youtube.com/')
497 if cookies.get('__Secure-3PSID'):
498 return
499 consent_id = None
500 consent = cookies.get('CONSENT')
501 if consent:
502 if 'YES' in consent.value:
503 return
504 consent_id = self._search_regex(
505 r'PENDING\+(\d+)', consent.value, 'consent', default=None)
506 if not consent_id:
507 consent_id = random.randint(100, 999)
508 self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id)
509
510 def _initialize_pref(self):
511 cookies = self._get_cookies('https://www.youtube.com/')
512 pref_cookie = cookies.get('PREF')
513 pref = {}
514 if pref_cookie:
515 try:
516 pref = dict(urllib.parse.parse_qsl(pref_cookie.value))
517 except ValueError:
518 self.report_warning('Failed to parse user PREF cookie' + bug_reports_message())
519 pref.update({'hl': self._preferred_lang or 'en', 'tz': 'UTC'})
520 self._set_cookie('.youtube.com', name='PREF', value=urllib.parse.urlencode(pref))
521
522 def _real_initialize(self):
523 self._initialize_pref()
524 self._initialize_consent()
525 self._check_login_required()
526
527 def _check_login_required(self):
528 if self._LOGIN_REQUIRED and not self._cookies_passed:
529 self.raise_login_required('Login details are needed to download this content', method='cookies')
530
531 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*='
532 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*='
533
534 def _get_default_ytcfg(self, client='web'):
535 return copy.deepcopy(INNERTUBE_CLIENTS[client])
536
537 def _get_innertube_host(self, client='web'):
538 return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST']
539
540 def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'):
541 # try_get but with fallback to default ytcfg client values when present
542 _func = lambda y: try_get(y, getter, expected_type)
543 return _func(ytcfg) or _func(self._get_default_ytcfg(default_client))
544
545 def _extract_client_name(self, ytcfg, default_client='web'):
546 return self._ytcfg_get_safe(
547 ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'],
548 lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), str, default_client)
549
550 def _extract_client_version(self, ytcfg, default_client='web'):
551 return self._ytcfg_get_safe(
552 ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'],
553 lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), str, default_client)
554
555 def _select_api_hostname(self, req_api_hostname, default_client=None):
556 return (self._configuration_arg('innertube_host', [''], ie_key=YoutubeIE.ie_key())[0]
557 or req_api_hostname or self._get_innertube_host(default_client or 'web'))
558
559 def _extract_api_key(self, ytcfg=None, default_client='web'):
560 return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], str, default_client)
561
562 def _extract_context(self, ytcfg=None, default_client='web'):
563 context = get_first(
564 (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict)
565 # Enforce language and tz for extraction
566 client_context = traverse_obj(context, 'client', expected_type=dict, default={})
567 client_context.update({'hl': self._preferred_lang or 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0})
568 return context
569
570 _SAPISID = None
571
572 def _generate_sapisidhash_header(self, origin='https://www.youtube.com'):
573 time_now = round(time.time())
574 if self._SAPISID is None:
575 yt_cookies = self._get_cookies('https://www.youtube.com')
576 # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is.
577 # See: https://github.com/yt-dlp/yt-dlp/issues/393
578 sapisid_cookie = dict_get(
579 yt_cookies, ('__Secure-3PAPISID', 'SAPISID'))
580 if sapisid_cookie and sapisid_cookie.value:
581 self._SAPISID = sapisid_cookie.value
582 self.write_debug('Extracted SAPISID cookie')
583 # SAPISID cookie is required if not already present
584 if not yt_cookies.get('SAPISID'):
585 self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie')
586 self._set_cookie(
587 '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600)
588 else:
589 self._SAPISID = False
590 if not self._SAPISID:
591 return None
592 # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323
593 sapisidhash = hashlib.sha1(
594 f'{time_now} {self._SAPISID} {origin}'.encode()).hexdigest()
595 return f'SAPISIDHASH {time_now}_{sapisidhash}'
596
597 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
598 note='Downloading API JSON', errnote='Unable to download API page',
599 context=None, api_key=None, api_hostname=None, default_client='web'):
600
601 data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)}
602 data.update(query)
603 real_headers = self.generate_api_headers(default_client=default_client)
604 real_headers.update({'content-type': 'application/json'})
605 if headers:
606 real_headers.update(headers)
607 api_key = (self._configuration_arg('innertube_key', [''], ie_key=YoutubeIE.ie_key(), casesense=True)[0]
608 or api_key or self._extract_api_key(default_client=default_client))
609 return self._download_json(
610 f'https://{self._select_api_hostname(api_hostname, default_client)}/youtubei/v1/{ep}',
611 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
612 data=json.dumps(data).encode('utf8'), headers=real_headers,
613 query={'key': api_key, 'prettyPrint': 'false'})
614
615 def extract_yt_initial_data(self, item_id, webpage, fatal=True):
616 return self._search_json(self._YT_INITIAL_DATA_RE, webpage, 'yt initial data', item_id, fatal=fatal)
617
618 @staticmethod
619 def _extract_session_index(*data):
620 """
621 Index of current account in account list.
622 See: https://github.com/yt-dlp/yt-dlp/pull/519
623 """
624 for ytcfg in data:
625 session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX']))
626 if session_index is not None:
627 return session_index
628
629 # Deprecated?
630 def _extract_identity_token(self, ytcfg=None, webpage=None):
631 if ytcfg:
632 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], str)
633 if token:
634 return token
635 if webpage:
636 return self._search_regex(
637 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
638 'identity token', default=None, fatal=False)
639
640 @staticmethod
641 def _extract_account_syncid(*args):
642 """
643 Extract syncId required to download private playlists of secondary channels
644 @params response and/or ytcfg
645 """
646 for data in args:
647 # ytcfg includes channel_syncid if on secondary channel
648 delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], str)
649 if delegated_sid:
650 return delegated_sid
651 sync_ids = (try_get(
652 data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'],
653 lambda x: x['DATASYNC_ID']), str) or '').split('||')
654 if len(sync_ids) >= 2 and sync_ids[1]:
655 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
656 # and just "user_syncid||" for primary channel. We only want the channel_syncid
657 return sync_ids[0]
658
659 @staticmethod
660 def _extract_visitor_data(*args):
661 """
662 Extracts visitorData from an API response or ytcfg
663 Appears to be used to track session state
664 """
665 return get_first(
666 args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))],
667 expected_type=str)
668
669 @functools.cached_property
670 def is_authenticated(self):
671 return bool(self._generate_sapisidhash_header())
672
673 def extract_ytcfg(self, video_id, webpage):
674 if not webpage:
675 return {}
676 return self._parse_json(
677 self._search_regex(
678 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
679 default='{}'), video_id, fatal=False) or {}
680
681 def generate_api_headers(
682 self, *, ytcfg=None, account_syncid=None, session_index=None,
683 visitor_data=None, identity_token=None, api_hostname=None, default_client='web'):
684
685 origin = 'https://' + (self._select_api_hostname(api_hostname, default_client))
686 headers = {
687 'X-YouTube-Client-Name': str(
688 self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)),
689 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client),
690 'Origin': origin,
691 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg),
692 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg),
693 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
694 'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client)
695 }
696 if session_index is None:
697 session_index = self._extract_session_index(ytcfg)
698 if account_syncid or session_index is not None:
699 headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0
700
701 auth = self._generate_sapisidhash_header(origin)
702 if auth is not None:
703 headers['Authorization'] = auth
704 headers['X-Origin'] = origin
705 return filter_dict(headers)
706
707 def _download_ytcfg(self, client, video_id):
708 url = {
709 'web': 'https://www.youtube.com',
710 'web_music': 'https://music.youtube.com',
711 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1'
712 }.get(client)
713 if not url:
714 return {}
715 webpage = self._download_webpage(
716 url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
717 return self.extract_ytcfg(video_id, webpage) or {}
718
719 @staticmethod
720 def _build_api_continuation_query(continuation, ctp=None):
721 query = {
722 'continuation': continuation
723 }
724 # TODO: Inconsistency with clickTrackingParams.
725 # Currently we have a fixed ctp contained within context (from ytcfg)
726 # and a ctp in root query for continuation.
727 if ctp:
728 query['clickTracking'] = {'clickTrackingParams': ctp}
729 return query
730
731 @classmethod
732 def _extract_next_continuation_data(cls, renderer):
733 next_continuation = try_get(
734 renderer, (lambda x: x['continuations'][0]['nextContinuationData'],
735 lambda x: x['continuation']['reloadContinuationData']), dict)
736 if not next_continuation:
737 return
738 continuation = next_continuation.get('continuation')
739 if not continuation:
740 return
741 ctp = next_continuation.get('clickTrackingParams')
742 return cls._build_api_continuation_query(continuation, ctp)
743
744 @classmethod
745 def _extract_continuation_ep_data(cls, continuation_ep: dict):
746 if isinstance(continuation_ep, dict):
747 continuation = try_get(
748 continuation_ep, lambda x: x['continuationCommand']['token'], str)
749 if not continuation:
750 return
751 ctp = continuation_ep.get('clickTrackingParams')
752 return cls._build_api_continuation_query(continuation, ctp)
753
754 @classmethod
755 def _extract_continuation(cls, renderer):
756 next_continuation = cls._extract_next_continuation_data(renderer)
757 if next_continuation:
758 return next_continuation
759
760 return traverse_obj(renderer, (
761 ('contents', 'items', 'rows'), ..., 'continuationItemRenderer',
762 ('continuationEndpoint', ('button', 'buttonRenderer', 'command'))
763 ), get_all=False, expected_type=cls._extract_continuation_ep_data)
764
765 @classmethod
766 def _extract_alerts(cls, data):
767 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
768 if not isinstance(alert_dict, dict):
769 continue
770 for alert in alert_dict.values():
771 alert_type = alert.get('type')
772 if not alert_type:
773 continue
774 message = cls._get_text(alert, 'text')
775 if message:
776 yield alert_type, message
777
778 def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False):
779 errors, warnings = [], []
780 for alert_type, alert_message in alerts:
781 if alert_type.lower() == 'error' and fatal:
782 errors.append([alert_type, alert_message])
783 elif alert_message not in self._IGNORED_WARNINGS:
784 warnings.append([alert_type, alert_message])
785
786 for alert_type, alert_message in (warnings + errors[:-1]):
787 self.report_warning(f'YouTube said: {alert_type} - {alert_message}', only_once=only_once)
788 if errors:
789 raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected)
790
791 def _extract_and_report_alerts(self, data, *args, **kwargs):
792 return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
793
794 def _extract_badges(self, renderer: dict):
795 privacy_icon_map = {
796 'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
797 'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
798 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC
799 }
800
801 badge_style_map = {
802 'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
803 'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
804 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW
805 }
806
807 label_map = {
808 'unlisted': BadgeType.AVAILABILITY_UNLISTED,
809 'private': BadgeType.AVAILABILITY_PRIVATE,
810 'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
811 'live': BadgeType.LIVE_NOW,
812 'premium': BadgeType.AVAILABILITY_PREMIUM
813 }
814
815 badges = []
816 for badge in traverse_obj(renderer, ('badges', ..., 'metadataBadgeRenderer')):
817 badge_type = (
818 privacy_icon_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
819 or badge_style_map.get(traverse_obj(badge, 'style'))
820 )
821 if badge_type:
822 badges.append({'type': badge_type})
823 continue
824
825 # fallback, won't work in some languages
826 label = traverse_obj(badge, 'label', expected_type=str, default='')
827 for match, label_badge_type in label_map.items():
828 if match in label.lower():
829 badges.append({'type': badge_type})
830 continue
831
832 return badges
833
834 @staticmethod
835 def _has_badge(badges, badge_type):
836 return bool(traverse_obj(badges, lambda _, v: v['type'] == badge_type))
837
838 @staticmethod
839 def _get_text(data, *path_list, max_runs=None):
840 for path in path_list or [None]:
841 if path is None:
842 obj = [data]
843 else:
844 obj = traverse_obj(data, path, default=[])
845 if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)):
846 obj = [obj]
847 for item in obj:
848 text = try_get(item, lambda x: x['simpleText'], str)
849 if text:
850 return text
851 runs = try_get(item, lambda x: x['runs'], list) or []
852 if not runs and isinstance(item, list):
853 runs = item
854
855 runs = runs[:min(len(runs), max_runs or len(runs))]
856 text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str))
857 if text:
858 return text
859
860 def _get_count(self, data, *path_list):
861 count_text = self._get_text(data, *path_list) or ''
862 count = parse_count(count_text)
863 if count is None:
864 count = str_to_int(
865 self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None))
866 return count
867
868 @staticmethod
869 def _extract_thumbnails(data, *path_list):
870 """
871 Extract thumbnails from thumbnails dict
872 @param path_list: path list to level that contains 'thumbnails' key
873 """
874 thumbnails = []
875 for path in path_list or [()]:
876 for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...)):
877 thumbnail_url = url_or_none(thumbnail.get('url'))
878 if not thumbnail_url:
879 continue
880 # Sometimes youtube gives a wrong thumbnail URL. See:
881 # https://github.com/yt-dlp/yt-dlp/issues/233
882 # https://github.com/ytdl-org/youtube-dl/issues/28023
883 if 'maxresdefault' in thumbnail_url:
884 thumbnail_url = thumbnail_url.split('?')[0]
885 thumbnails.append({
886 'url': thumbnail_url,
887 'height': int_or_none(thumbnail.get('height')),
888 'width': int_or_none(thumbnail.get('width')),
889 })
890 return thumbnails
891
892 @staticmethod
893 def extract_relative_time(relative_time_text):
894 """
895 Extracts a relative time from string and converts to dt object
896 e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today', '8 yr ago'
897 """
898
899 # XXX: this could be moved to a general function in utils.py
900 # The relative time text strings are roughly the same as what
901 # Javascript's Intl.RelativeTimeFormat function generates.
902 # See: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/RelativeTimeFormat
903 mobj = re.search(
904 r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>sec(?:ond)?|s|min(?:ute)?|h(?:our|r)?|d(?:ay)?|w(?:eek|k)?|mo(?:nth)?|y(?:ear|r)?)s?\s*ago',
905 relative_time_text)
906 if mobj:
907 start = mobj.group('start')
908 if start:
909 return datetime_from_str(start)
910 try:
911 return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit')))
912 except ValueError:
913 return None
914
915 def _parse_time_text(self, text):
916 if not text:
917 return
918 dt = self.extract_relative_time(text)
919 timestamp = None
920 if isinstance(dt, datetime.datetime):
921 timestamp = calendar.timegm(dt.timetuple())
922
923 if timestamp is None:
924 timestamp = (
925 unified_timestamp(text) or unified_timestamp(
926 self._search_regex(
927 (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'),
928 text.lower(), 'time text', default=None)))
929
930 if text and timestamp is None and self._preferred_lang in (None, 'en'):
931 self.report_warning(
932 f'Cannot parse localized time text "{text}"', only_once=True)
933 return timestamp
934
935 def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
936 ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
937 default_client='web'):
938 for retry in self.RetryManager():
939 try:
940 response = self._call_api(
941 ep=ep, fatal=True, headers=headers,
942 video_id=item_id, query=query, note=note,
943 context=self._extract_context(ytcfg, default_client),
944 api_key=self._extract_api_key(ytcfg, default_client),
945 api_hostname=api_hostname, default_client=default_client)
946 except ExtractorError as e:
947 if not isinstance(e.cause, network_exceptions):
948 return self._error_or_warning(e, fatal=fatal)
949 elif not isinstance(e.cause, urllib.error.HTTPError):
950 retry.error = e
951 continue
952
953 first_bytes = e.cause.read(512)
954 if not is_html(first_bytes):
955 yt_error = try_get(
956 self._parse_json(
957 self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
958 lambda x: x['error']['message'], str)
959 if yt_error:
960 self._report_alerts([('ERROR', yt_error)], fatal=False)
961 # Downloading page may result in intermittent 5xx HTTP error
962 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
963 # We also want to catch all other network exceptions since errors in later pages can be troublesome
964 # See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
965 if e.cause.code not in (403, 429):
966 retry.error = e
967 continue
968 return self._error_or_warning(e, fatal=fatal)
969
970 try:
971 self._extract_and_report_alerts(response, only_once=True)
972 except ExtractorError as e:
973 # YouTube servers may return errors we want to retry on in a 200 OK response
974 # See: https://github.com/yt-dlp/yt-dlp/issues/839
975 if 'unknown error' in e.msg.lower():
976 retry.error = e
977 continue
978 return self._error_or_warning(e, fatal=fatal)
979 # Youtube sometimes sends incomplete data
980 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
981 if not traverse_obj(response, *variadic(check_get_keys)):
982 retry.error = ExtractorError('Incomplete data received', expected=True)
983 continue
984
985 return response
986
987 @staticmethod
988 def is_music_url(url):
989 return re.match(r'(https?://)?music\.youtube\.com/', url) is not None
990
991 def _extract_video(self, renderer):
992 video_id = renderer.get('videoId')
993
994 reel_header_renderer = traverse_obj(renderer, (
995 'navigationEndpoint', 'reelWatchEndpoint', 'overlay', 'reelPlayerOverlayRenderer',
996 'reelPlayerHeaderSupportedRenderers', 'reelPlayerHeaderRenderer'))
997
998 title = self._get_text(renderer, 'title', 'headline') or self._get_text(reel_header_renderer, 'reelTitleText')
999 description = self._get_text(renderer, 'descriptionSnippet')
1000
1001 duration = int_or_none(renderer.get('lengthSeconds'))
1002 if duration is None:
1003 duration = parse_duration(self._get_text(
1004 renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text')))
1005 if duration is None:
1006 # XXX: should write a parser to be more general to support more cases (e.g. shorts in shorts tab)
1007 duration = parse_duration(self._search_regex(
1008 r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$',
1009 traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str),
1010 video_id, default=None, group='duration'))
1011
1012 channel_id = traverse_obj(
1013 renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'),
1014 expected_type=str, get_all=False)
1015 if not channel_id:
1016 channel_id = traverse_obj(reel_header_renderer, ('channelNavigationEndpoint', 'browseEndpoint', 'browseId'))
1017
1018 channel_id = self.ucid_or_none(channel_id)
1019
1020 overlay_style = traverse_obj(
1021 renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
1022 get_all=False, expected_type=str)
1023 badges = self._extract_badges(renderer)
1024
1025 navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
1026 renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
1027 expected_type=str)) or ''
1028 url = f'https://www.youtube.com/watch?v={video_id}'
1029 if overlay_style == 'SHORTS' or '/shorts/' in navigation_url:
1030 url = f'https://www.youtube.com/shorts/{video_id}'
1031
1032 time_text = (self._get_text(renderer, 'publishedTimeText', 'videoInfo')
1033 or self._get_text(reel_header_renderer, 'timestampText') or '')
1034 scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False))
1035
1036 live_status = (
1037 'is_upcoming' if scheduled_timestamp is not None
1038 else 'was_live' if 'streamed' in time_text.lower()
1039 else 'is_live' if overlay_style == 'LIVE' or self._has_badge(badges, BadgeType.LIVE_NOW)
1040 else None)
1041
1042 # videoInfo is a string like '50K views • 10 years ago'.
1043 view_count_text = self._get_text(renderer, 'viewCountText', 'shortViewCountText', 'videoInfo') or ''
1044 view_count = (0 if 'no views' in view_count_text.lower()
1045 else self._get_count({'simpleText': view_count_text}))
1046 view_count_field = 'concurrent_view_count' if live_status in ('is_live', 'is_upcoming') else 'view_count'
1047
1048 channel = (self._get_text(renderer, 'ownerText', 'shortBylineText')
1049 or self._get_text(reel_header_renderer, 'channelTitleText'))
1050
1051 channel_handle = traverse_obj(renderer, (
1052 'shortBylineText', 'runs', ..., 'navigationEndpoint',
1053 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl'))),
1054 expected_type=self.handle_from_url, get_all=False)
1055 return {
1056 '_type': 'url',
1057 'ie_key': YoutubeIE.ie_key(),
1058 'id': video_id,
1059 'url': url,
1060 'title': title,
1061 'description': description,
1062 'duration': duration,
1063 'channel_id': channel_id,
1064 'channel': channel,
1065 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None,
1066 'uploader': channel,
1067 'uploader_id': channel_handle,
1068 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
1069 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
1070 'timestamp': (self._parse_time_text(time_text)
1071 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
1072 else None),
1073 'release_timestamp': scheduled_timestamp,
1074 'availability':
1075 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
1076 else self._availability(
1077 is_private=self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE) or None,
1078 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
1079 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
1080 is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
1081 view_count_field: view_count,
1082 'live_status': live_status
1083 }
1084
1085
1086 class YoutubeIE(YoutubeBaseInfoExtractor):
1087 IE_DESC = 'YouTube'
1088 _VALID_URL = r"""(?x)^
1089 (
1090 (?:https?://|//) # http(s):// or protocol-independent URL
1091 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
1092 (?:www\.)?deturl\.com/www\.youtube\.com|
1093 (?:www\.)?pwnyoutube\.com|
1094 (?:www\.)?hooktube\.com|
1095 (?:www\.)?yourepeat\.com|
1096 tube\.majestyc\.net|
1097 %(invidious)s|
1098 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
1099 (?:.*?\#/)? # handle anchor (#/) redirect urls
1100 (?: # the various things that can precede the ID:
1101 (?:(?:v|embed|e|shorts|live)/(?!videoseries|live_stream)) # v/ or embed/ or e/ or shorts/
1102 |(?: # or the v= param in all its forms
1103 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
1104 (?:\?|\#!?) # the params delimiter ? or # or #!
1105 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
1106 v=
1107 )
1108 ))
1109 |(?:
1110 youtu\.be| # just youtu.be/xxxx
1111 vid\.plus| # or vid.plus/xxxx
1112 zwearz\.com/watch| # or zwearz.com/watch/xxxx
1113 %(invidious)s
1114 )/
1115 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
1116 )
1117 )? # all until now is optional -> you can pass the naked ID
1118 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
1119 (?(1).+)? # if we found the ID, everything can follow
1120 (?:\#|$)""" % {
1121 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
1122 }
1123 _EMBED_REGEX = [
1124 r'''(?x)
1125 (?:
1126 <(?:[0-9A-Za-z-]+?)?iframe[^>]+?src=|
1127 data-video-url=|
1128 <embed[^>]+?src=|
1129 embedSWF\(?:\s*|
1130 <object[^>]+data=|
1131 new\s+SWFObject\(
1132 )
1133 (["\'])
1134 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1135 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1136 \1''',
1137 # https://wordpress.org/plugins/lazy-load-for-videos/
1138 r'''(?xs)
1139 <a\s[^>]*\bhref="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"
1140 \s[^>]*\bclass="[^"]*\blazy-load-youtube''',
1141 ]
1142 _RETURN_TYPE = 'video' # XXX: How to handle multifeed?
1143
1144 _PLAYER_INFO_RE = (
1145 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
1146 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$',
1147 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
1148 )
1149 _formats = {
1150 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1151 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
1152 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
1153 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
1154 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
1155 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1156 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1157 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1158 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
1159 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
1160 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1161 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
1162 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1163 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
1164 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1165 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
1166 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1167 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
1168
1169
1170 # 3D videos
1171 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1172 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
1173 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1174 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
1175 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
1176 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1177 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
1178
1179 # Apple HTTP Live Streaming
1180 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1181 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1182 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1183 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
1184 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1185 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
1186 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
1187 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
1188
1189 # DASH mp4 video
1190 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
1191 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
1192 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1193 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
1194 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
1195 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
1196 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
1197 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
1198 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
1199 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1200 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
1201 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
1202
1203 # Dash mp4 audio
1204 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
1205 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
1206 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
1207 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1208 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
1209 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
1210 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
1211
1212 # Dash webm
1213 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1214 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1215 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1216 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1217 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1218 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
1219 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
1220 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1221 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1222 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1223 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1224 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1225 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1226 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1227 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1228 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
1229 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1230 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1231 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1232 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1233 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
1234 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
1235
1236 # Dash webm audio
1237 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
1238 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
1239
1240 # Dash webm audio with opus inside
1241 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
1242 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
1243 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
1244
1245 # RTMP (unnamed)
1246 '_rtmp': {'protocol': 'rtmp'},
1247
1248 # av01 video only formats sometimes served with "unknown" codecs
1249 '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1250 '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'},
1251 '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'},
1252 '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'},
1253 '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'},
1254 '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'},
1255 '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1256 '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'},
1257 }
1258 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
1259
1260 _GEO_BYPASS = False
1261
1262 IE_NAME = 'youtube'
1263 _TESTS = [
1264 {
1265 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
1266 'info_dict': {
1267 'id': 'BaW_jenozKc',
1268 'ext': 'mp4',
1269 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1270 'channel': 'Philipp Hagemeister',
1271 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1272 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1273 'upload_date': '20121002',
1274 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1275 'categories': ['Science & Technology'],
1276 'tags': ['youtube-dl'],
1277 'duration': 10,
1278 'view_count': int,
1279 'like_count': int,
1280 'availability': 'public',
1281 'playable_in_embed': True,
1282 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1283 'live_status': 'not_live',
1284 'age_limit': 0,
1285 'start_time': 1,
1286 'end_time': 9,
1287 'comment_count': int,
1288 'channel_follower_count': int,
1289 'uploader': 'Philipp Hagemeister',
1290 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1291 'uploader_id': '@PhilippHagemeister',
1292 'heatmap': 'count:100',
1293 }
1294 },
1295 {
1296 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
1297 'note': 'Embed-only video (#1746)',
1298 'info_dict': {
1299 'id': 'yZIXLfi8CZQ',
1300 'ext': 'mp4',
1301 'upload_date': '20120608',
1302 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
1303 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
1304 'age_limit': 18,
1305 },
1306 'skip': 'Private video',
1307 },
1308 {
1309 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
1310 'note': 'Use the first video ID in the URL',
1311 'info_dict': {
1312 'id': 'BaW_jenozKc',
1313 'ext': 'mp4',
1314 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
1315 'channel': 'Philipp Hagemeister',
1316 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
1317 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
1318 'upload_date': '20121002',
1319 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22',
1320 'categories': ['Science & Technology'],
1321 'tags': ['youtube-dl'],
1322 'duration': 10,
1323 'view_count': int,
1324 'like_count': int,
1325 'availability': 'public',
1326 'playable_in_embed': True,
1327 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg',
1328 'live_status': 'not_live',
1329 'age_limit': 0,
1330 'comment_count': int,
1331 'channel_follower_count': int,
1332 'uploader': 'Philipp Hagemeister',
1333 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
1334 'uploader_id': '@PhilippHagemeister',
1335 },
1336 'params': {
1337 'skip_download': True,
1338 },
1339 },
1340 {
1341 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
1342 'note': '256k DASH audio (format 141) via DASH manifest',
1343 'info_dict': {
1344 'id': 'a9LDPn-MO4I',
1345 'ext': 'm4a',
1346 'upload_date': '20121002',
1347 'description': '',
1348 'title': 'UHDTV TEST 8K VIDEO.mp4'
1349 },
1350 'params': {
1351 'youtube_include_dash_manifest': True,
1352 'format': '141',
1353 },
1354 'skip': 'format 141 not served anymore',
1355 },
1356 # DASH manifest with encrypted signature
1357 {
1358 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
1359 'info_dict': {
1360 'id': 'IB3lcPjvWLA',
1361 'ext': 'm4a',
1362 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
1363 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
1364 'duration': 244,
1365 'upload_date': '20131011',
1366 'abr': 129.495,
1367 'like_count': int,
1368 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg',
1369 'playable_in_embed': True,
1370 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg',
1371 'view_count': int,
1372 'track': 'The Spark',
1373 'live_status': 'not_live',
1374 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp',
1375 'channel': 'Afrojack',
1376 'tags': 'count:19',
1377 'availability': 'public',
1378 'categories': ['Music'],
1379 'age_limit': 0,
1380 'alt_title': 'The Spark',
1381 'channel_follower_count': int,
1382 'uploader': 'Afrojack',
1383 'uploader_url': 'https://www.youtube.com/@Afrojack',
1384 'uploader_id': '@Afrojack',
1385 },
1386 'params': {
1387 'youtube_include_dash_manifest': True,
1388 'format': '141/bestaudio[ext=m4a]',
1389 },
1390 },
1391 # Age-gate videos. See https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-888837000
1392 {
1393 'note': 'Embed allowed age-gate video',
1394 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
1395 'info_dict': {
1396 'id': 'HtVdAasjOgU',
1397 'ext': 'mp4',
1398 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
1399 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
1400 'duration': 142,
1401 'upload_date': '20140605',
1402 'age_limit': 18,
1403 'categories': ['Gaming'],
1404 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp',
1405 'availability': 'needs_auth',
1406 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg',
1407 'like_count': int,
1408 'channel': 'The Witcher',
1409 'live_status': 'not_live',
1410 'tags': 'count:17',
1411 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg',
1412 'playable_in_embed': True,
1413 'view_count': int,
1414 'channel_follower_count': int,
1415 'uploader': 'The Witcher',
1416 'uploader_url': 'https://www.youtube.com/@thewitcher',
1417 'uploader_id': '@thewitcher',
1418 },
1419 },
1420 {
1421 'note': 'Age-gate video with embed allowed in public site',
1422 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U',
1423 'info_dict': {
1424 'id': 'HsUATh_Nc2U',
1425 'ext': 'mp4',
1426 'title': 'Godzilla 2 (Official Video)',
1427 'description': 'md5:bf77e03fcae5529475e500129b05668a',
1428 'upload_date': '20200408',
1429 'age_limit': 18,
1430 'availability': 'needs_auth',
1431 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg',
1432 'channel': 'FlyingKitty',
1433 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg',
1434 'view_count': int,
1435 'categories': ['Entertainment'],
1436 'live_status': 'not_live',
1437 'tags': ['Flyingkitty', 'godzilla 2'],
1438 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg',
1439 'like_count': int,
1440 'duration': 177,
1441 'playable_in_embed': True,
1442 'channel_follower_count': int,
1443 'uploader': 'FlyingKitty',
1444 'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
1445 'uploader_id': '@FlyingKitty900',
1446 'comment_count': int,
1447 },
1448 },
1449 {
1450 'note': 'Age-gate video embedable only with clientScreen=EMBED',
1451 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg',
1452 'info_dict': {
1453 'id': 'Tq92D6wQ1mg',
1454 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
1455 'ext': 'mp4',
1456 'upload_date': '20191228',
1457 'description': 'md5:17eccca93a786d51bc67646756894066',
1458 'age_limit': 18,
1459 'like_count': int,
1460 'availability': 'needs_auth',
1461 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
1462 'view_count': int,
1463 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp',
1464 'channel': 'Projekt Melody',
1465 'live_status': 'not_live',
1466 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
1467 'playable_in_embed': True,
1468 'categories': ['Entertainment'],
1469 'duration': 106,
1470 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
1471 'comment_count': int,
1472 'channel_follower_count': int,
1473 'uploader': 'Projekt Melody',
1474 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
1475 'uploader_id': '@ProjektMelody',
1476 },
1477 },
1478 {
1479 'note': 'Non-Agegated non-embeddable video',
1480 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY',
1481 'info_dict': {
1482 'id': 'MeJVWBSsPAY',
1483 'ext': 'mp4',
1484 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)',
1485 'description': 'Fan Video. Music & Lyrics by OOMPH!.',
1486 'upload_date': '20130730',
1487 'track': 'Such mich find mich',
1488 'age_limit': 0,
1489 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'],
1490 'like_count': int,
1491 'playable_in_embed': False,
1492 'creator': 'OOMPH!',
1493 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg',
1494 'view_count': int,
1495 'alt_title': 'Such mich find mich',
1496 'duration': 210,
1497 'channel': 'Herr Lurik',
1498 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA',
1499 'categories': ['Music'],
1500 'availability': 'public',
1501 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA',
1502 'live_status': 'not_live',
1503 'artist': 'OOMPH!',
1504 'channel_follower_count': int,
1505 'uploader': 'Herr Lurik',
1506 'uploader_url': 'https://www.youtube.com/@HerrLurik',
1507 'uploader_id': '@HerrLurik',
1508 },
1509 },
1510 {
1511 'note': 'Non-bypassable age-gated video',
1512 'url': 'https://youtube.com/watch?v=Cr381pDsSsA',
1513 'only_matching': True,
1514 },
1515 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
1516 # YouTube Red ad is not captured for creator
1517 {
1518 'url': '__2ABJjxzNo',
1519 'info_dict': {
1520 'id': '__2ABJjxzNo',
1521 'ext': 'mp4',
1522 'duration': 266,
1523 'upload_date': '20100430',
1524 'creator': 'deadmau5',
1525 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
1526 'title': 'Deadmau5 - Some Chords (HD)',
1527 'alt_title': 'Some Chords',
1528 'availability': 'public',
1529 'tags': 'count:14',
1530 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ',
1531 'view_count': int,
1532 'live_status': 'not_live',
1533 'channel': 'deadmau5',
1534 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp',
1535 'like_count': int,
1536 'track': 'Some Chords',
1537 'artist': 'deadmau5',
1538 'playable_in_embed': True,
1539 'age_limit': 0,
1540 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ',
1541 'categories': ['Music'],
1542 'album': 'Some Chords',
1543 'channel_follower_count': int,
1544 'uploader': 'deadmau5',
1545 'uploader_url': 'https://www.youtube.com/@deadmau5',
1546 'uploader_id': '@deadmau5',
1547 },
1548 'expected_warnings': [
1549 'DASH manifest missing',
1550 ]
1551 },
1552 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
1553 {
1554 'url': 'lqQg6PlCWgI',
1555 'info_dict': {
1556 'id': 'lqQg6PlCWgI',
1557 'ext': 'mp4',
1558 'duration': 6085,
1559 'upload_date': '20150827',
1560 'description': 'md5:04bbbf3ccceb6795947572ca36f45904',
1561 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
1562 'like_count': int,
1563 'release_timestamp': 1343767800,
1564 'playable_in_embed': True,
1565 'categories': ['Sports'],
1566 'release_date': '20120731',
1567 'channel': 'Olympics',
1568 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'],
1569 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q',
1570 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg',
1571 'age_limit': 0,
1572 'availability': 'public',
1573 'live_status': 'was_live',
1574 'view_count': int,
1575 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q',
1576 'channel_follower_count': int,
1577 'uploader': 'Olympics',
1578 'uploader_url': 'https://www.youtube.com/@Olympics',
1579 'uploader_id': '@Olympics',
1580 },
1581 'params': {
1582 'skip_download': 'requires avconv',
1583 }
1584 },
1585 # Non-square pixels
1586 {
1587 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
1588 'info_dict': {
1589 'id': '_b-2C3KPAM0',
1590 'ext': 'mp4',
1591 'stretched_ratio': 16 / 9.,
1592 'duration': 85,
1593 'upload_date': '20110310',
1594 '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',
1595 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
1596 'playable_in_embed': True,
1597 'channel': '孫ᄋᄅ',
1598 'age_limit': 0,
1599 'tags': 'count:11',
1600 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw',
1601 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw',
1602 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg',
1603 'view_count': int,
1604 'categories': ['People & Blogs'],
1605 'like_count': int,
1606 'live_status': 'not_live',
1607 'availability': 'unlisted',
1608 'comment_count': int,
1609 'channel_follower_count': int,
1610 'uploader': '孫ᄋᄅ',
1611 'uploader_url': 'https://www.youtube.com/@AllenMeow',
1612 'uploader_id': '@AllenMeow',
1613 },
1614 },
1615 # url_encoded_fmt_stream_map is empty string
1616 {
1617 'url': 'qEJwOuvDf7I',
1618 'info_dict': {
1619 'id': 'qEJwOuvDf7I',
1620 'ext': 'webm',
1621 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
1622 'description': '',
1623 'upload_date': '20150404',
1624 },
1625 'params': {
1626 'skip_download': 'requires avconv',
1627 },
1628 'skip': 'This live event has ended.',
1629 },
1630 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
1631 {
1632 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
1633 'info_dict': {
1634 'id': 'FIl7x6_3R5Y',
1635 'ext': 'webm',
1636 'title': 'md5:7b81415841e02ecd4313668cde88737a',
1637 'description': 'md5:116377fd2963b81ec4ce64b542173306',
1638 'duration': 220,
1639 'upload_date': '20150625',
1640 'formats': 'mincount:31',
1641 },
1642 'skip': 'not actual anymore',
1643 },
1644 # DASH manifest with segment_list
1645 {
1646 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
1647 'md5': '8ce563a1d667b599d21064e982ab9e31',
1648 'info_dict': {
1649 'id': 'CsmdDsKjzN8',
1650 'ext': 'mp4',
1651 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
1652 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
1653 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
1654 },
1655 'params': {
1656 'youtube_include_dash_manifest': True,
1657 'format': '135', # bestvideo
1658 },
1659 'skip': 'This live event has ended.',
1660 },
1661 {
1662 # Multifeed videos (multiple cameras), URL can be of any Camera
1663 # TODO: fix multifeed titles
1664 'url': 'https://www.youtube.com/watch?v=zaPI8MvL8pg',
1665 'info_dict': {
1666 'id': 'zaPI8MvL8pg',
1667 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04',
1668 'description': 'md5:563ccbc698b39298481ca3c571169519',
1669 },
1670 'playlist': [{
1671 'info_dict': {
1672 'id': 'j5yGuxZ8lLU',
1673 'ext': 'mp4',
1674 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Chris)',
1675 'description': 'md5:563ccbc698b39298481ca3c571169519',
1676 'duration': 10120,
1677 'channel_follower_count': int,
1678 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1679 'availability': 'public',
1680 'playable_in_embed': True,
1681 'upload_date': '20131105',
1682 'categories': ['Gaming'],
1683 'live_status': 'was_live',
1684 'tags': 'count:24',
1685 'release_timestamp': 1383701910,
1686 'thumbnail': 'https://i.ytimg.com/vi/j5yGuxZ8lLU/maxresdefault.jpg',
1687 'comment_count': int,
1688 'age_limit': 0,
1689 'like_count': int,
1690 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1691 'channel': 'WiiLikeToPlay',
1692 'view_count': int,
1693 'release_date': '20131106',
1694 'uploader': 'WiiLikeToPlay',
1695 'uploader_id': '@WLTP',
1696 'uploader_url': 'https://www.youtube.com/@WLTP',
1697 },
1698 }, {
1699 'info_dict': {
1700 'id': 'zaPI8MvL8pg',
1701 'ext': 'mp4',
1702 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Tyson)',
1703 'availability': 'public',
1704 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1705 'channel': 'WiiLikeToPlay',
1706 'channel_follower_count': int,
1707 'description': 'md5:563ccbc698b39298481ca3c571169519',
1708 'duration': 10108,
1709 'age_limit': 0,
1710 'like_count': int,
1711 'tags': 'count:24',
1712 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1713 'release_timestamp': 1383701915,
1714 'comment_count': int,
1715 'upload_date': '20131105',
1716 'thumbnail': 'https://i.ytimg.com/vi/zaPI8MvL8pg/maxresdefault.jpg',
1717 'release_date': '20131106',
1718 'playable_in_embed': True,
1719 'live_status': 'was_live',
1720 'categories': ['Gaming'],
1721 'view_count': int,
1722 'uploader': 'WiiLikeToPlay',
1723 'uploader_id': '@WLTP',
1724 'uploader_url': 'https://www.youtube.com/@WLTP',
1725 },
1726 }, {
1727 'info_dict': {
1728 'id': 'R7r3vfO7Hao',
1729 'ext': 'mp4',
1730 'title': 'Terraria 1.2 Live Stream | Let\'s Play - Part 04 (Spencer)',
1731 'thumbnail': 'https://i.ytimg.com/vi/R7r3vfO7Hao/maxresdefault.jpg',
1732 'channel_id': 'UCN2XePorRokPB9TEgRZpddg',
1733 'like_count': int,
1734 'availability': 'public',
1735 'playable_in_embed': True,
1736 'upload_date': '20131105',
1737 'description': 'md5:563ccbc698b39298481ca3c571169519',
1738 'channel_follower_count': int,
1739 'tags': 'count:24',
1740 'release_date': '20131106',
1741 'comment_count': int,
1742 'channel_url': 'https://www.youtube.com/channel/UCN2XePorRokPB9TEgRZpddg',
1743 'channel': 'WiiLikeToPlay',
1744 'categories': ['Gaming'],
1745 'release_timestamp': 1383701914,
1746 'live_status': 'was_live',
1747 'age_limit': 0,
1748 'duration': 10128,
1749 'view_count': int,
1750 'uploader': 'WiiLikeToPlay',
1751 'uploader_id': '@WLTP',
1752 'uploader_url': 'https://www.youtube.com/@WLTP',
1753 },
1754 }],
1755 'params': {'skip_download': True},
1756 },
1757 {
1758 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
1759 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
1760 'info_dict': {
1761 'id': 'gVfLd0zydlo',
1762 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
1763 },
1764 'playlist_count': 2,
1765 'skip': 'Not multifeed anymore',
1766 },
1767 {
1768 'url': 'https://vid.plus/FlRa-iH7PGw',
1769 'only_matching': True,
1770 },
1771 {
1772 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
1773 'only_matching': True,
1774 },
1775 {
1776 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1777 # Also tests cut-off URL expansion in video description (see
1778 # https://github.com/ytdl-org/youtube-dl/issues/1892,
1779 # https://github.com/ytdl-org/youtube-dl/issues/8164)
1780 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
1781 'info_dict': {
1782 'id': 'lsguqyKfVQg',
1783 'ext': 'mp4',
1784 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
1785 'alt_title': 'Dark Walk',
1786 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
1787 'duration': 133,
1788 'upload_date': '20151119',
1789 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1790 'track': 'Dark Walk',
1791 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan',
1792 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
1793 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp',
1794 'categories': ['Film & Animation'],
1795 'view_count': int,
1796 'live_status': 'not_live',
1797 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ',
1798 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ',
1799 'tags': 'count:13',
1800 'availability': 'public',
1801 'channel': 'IronSoulElf',
1802 'playable_in_embed': True,
1803 'like_count': int,
1804 'age_limit': 0,
1805 'channel_follower_count': int
1806 },
1807 'params': {
1808 'skip_download': True,
1809 },
1810 },
1811 {
1812 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
1813 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
1814 'only_matching': True,
1815 },
1816 {
1817 # Video with yt:stretch=17:0
1818 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
1819 'info_dict': {
1820 'id': 'Q39EVAstoRM',
1821 'ext': 'mp4',
1822 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
1823 'description': 'md5:ee18a25c350637c8faff806845bddee9',
1824 'upload_date': '20151107',
1825 },
1826 'params': {
1827 'skip_download': True,
1828 },
1829 'skip': 'This video does not exist.',
1830 },
1831 {
1832 # Video with incomplete 'yt:stretch=16:'
1833 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI',
1834 'only_matching': True,
1835 },
1836 {
1837 # Video licensed under Creative Commons
1838 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1839 'info_dict': {
1840 'id': 'M4gD1WSo5mA',
1841 'ext': 'mp4',
1842 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1843 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1844 'duration': 721,
1845 'upload_date': '20150128',
1846 'license': 'Creative Commons Attribution license (reuse allowed)',
1847 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA',
1848 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA',
1849 'like_count': int,
1850 'age_limit': 0,
1851 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'],
1852 'channel': 'The Berkman Klein Center for Internet & Society',
1853 'availability': 'public',
1854 'view_count': int,
1855 'categories': ['Education'],
1856 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp',
1857 'live_status': 'not_live',
1858 'playable_in_embed': True,
1859 'channel_follower_count': int,
1860 'chapters': list,
1861 'uploader': 'The Berkman Klein Center for Internet & Society',
1862 'uploader_id': '@BKCHarvard',
1863 'uploader_url': 'https://www.youtube.com/@BKCHarvard',
1864 },
1865 'params': {
1866 'skip_download': True,
1867 },
1868 },
1869 {
1870 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1871 'info_dict': {
1872 'id': 'eQcmzGIKrzg',
1873 'ext': 'mp4',
1874 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1875 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
1876 'duration': 4060,
1877 'upload_date': '20151120',
1878 'license': 'Creative Commons Attribution license (reuse allowed)',
1879 'playable_in_embed': True,
1880 'tags': 'count:12',
1881 'like_count': int,
1882 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1883 'age_limit': 0,
1884 'availability': 'public',
1885 'categories': ['News & Politics'],
1886 'channel': 'Bernie Sanders',
1887 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp',
1888 'view_count': int,
1889 'live_status': 'not_live',
1890 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1891 'comment_count': int,
1892 'channel_follower_count': int,
1893 'chapters': list,
1894 'uploader': 'Bernie Sanders',
1895 'uploader_url': 'https://www.youtube.com/@BernieSanders',
1896 'uploader_id': '@BernieSanders',
1897 },
1898 'params': {
1899 'skip_download': True,
1900 },
1901 },
1902 {
1903 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1904 'only_matching': True,
1905 },
1906 {
1907 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1908 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1909 'only_matching': True,
1910 },
1911 {
1912 # Rental video preview
1913 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1914 'info_dict': {
1915 'id': 'uGpuVWrhIzE',
1916 'ext': 'mp4',
1917 'title': 'Piku - Trailer',
1918 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1919 'upload_date': '20150811',
1920 'license': 'Standard YouTube License',
1921 },
1922 'params': {
1923 'skip_download': True,
1924 },
1925 'skip': 'This video is not available.',
1926 },
1927 {
1928 # YouTube Red video with episode data
1929 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1930 'info_dict': {
1931 'id': 'iqKdEhx-dD4',
1932 'ext': 'mp4',
1933 'title': 'Isolation - Mind Field (Ep 1)',
1934 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1935 'duration': 2085,
1936 'upload_date': '20170118',
1937 'series': 'Mind Field',
1938 'season_number': 1,
1939 'episode_number': 1,
1940 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp',
1941 'tags': 'count:12',
1942 'view_count': int,
1943 'availability': 'public',
1944 'age_limit': 0,
1945 'channel': 'Vsauce',
1946 'episode': 'Episode 1',
1947 'categories': ['Entertainment'],
1948 'season': 'Season 1',
1949 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA',
1950 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA',
1951 'like_count': int,
1952 'playable_in_embed': True,
1953 'live_status': 'not_live',
1954 'channel_follower_count': int,
1955 'uploader': 'Vsauce',
1956 'uploader_url': 'https://www.youtube.com/@Vsauce',
1957 'uploader_id': '@Vsauce',
1958 },
1959 'params': {
1960 'skip_download': True,
1961 },
1962 'expected_warnings': [
1963 'Skipping DASH manifest',
1964 ],
1965 },
1966 {
1967 # The following content has been identified by the YouTube community
1968 # as inappropriate or offensive to some audiences.
1969 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1970 'info_dict': {
1971 'id': '6SJNVb0GnPI',
1972 'ext': 'mp4',
1973 'title': 'Race Differences in Intelligence',
1974 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1975 'duration': 965,
1976 'upload_date': '20140124',
1977 },
1978 'params': {
1979 'skip_download': True,
1980 },
1981 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1982 },
1983 {
1984 # itag 212
1985 'url': '1t24XAntNCY',
1986 'only_matching': True,
1987 },
1988 {
1989 # geo restricted to JP
1990 'url': 'sJL6WA-aGkQ',
1991 'only_matching': True,
1992 },
1993 {
1994 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1995 'only_matching': True,
1996 },
1997 {
1998 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1999 'only_matching': True,
2000 },
2001 {
2002 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
2003 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
2004 'only_matching': True,
2005 },
2006 {
2007 # DRM protected
2008 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
2009 'only_matching': True,
2010 },
2011 {
2012 # Video with unsupported adaptive stream type formats
2013 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
2014 'info_dict': {
2015 'id': 'Z4Vy8R84T1U',
2016 'ext': 'mp4',
2017 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
2018 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
2019 'duration': 433,
2020 'upload_date': '20130923',
2021 'formats': 'maxcount:10',
2022 },
2023 'params': {
2024 'skip_download': True,
2025 'youtube_include_dash_manifest': False,
2026 },
2027 'skip': 'not actual anymore',
2028 },
2029 {
2030 # Youtube Music Auto-generated description
2031 # TODO: fix metadata extraction
2032 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2033 'info_dict': {
2034 'id': 'MgNrAu2pzNs',
2035 'ext': 'mp4',
2036 'title': 'Voyeur Girl',
2037 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
2038 'upload_date': '20190312',
2039 'artist': 'Stephen',
2040 'track': 'Voyeur Girl',
2041 'album': 'it\'s too much love to know my dear',
2042 'release_date': '20190313',
2043 'release_year': 2019,
2044 'alt_title': 'Voyeur Girl',
2045 'view_count': int,
2046 'playable_in_embed': True,
2047 'like_count': int,
2048 'categories': ['Music'],
2049 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA',
2050 'channel': 'Stephen', # TODO: should be "Stephen - Topic"
2051 'uploader': 'Stephen',
2052 'availability': 'public',
2053 'creator': 'Stephen',
2054 'duration': 169,
2055 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp',
2056 'age_limit': 0,
2057 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA',
2058 'tags': 'count:11',
2059 'live_status': 'not_live',
2060 'channel_follower_count': int
2061 },
2062 'params': {
2063 'skip_download': True,
2064 },
2065 },
2066 {
2067 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
2068 'only_matching': True,
2069 },
2070 {
2071 # invalid -> valid video id redirection
2072 'url': 'DJztXj2GPfl',
2073 'info_dict': {
2074 'id': 'DJztXj2GPfk',
2075 'ext': 'mp4',
2076 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
2077 'description': 'md5:bf577a41da97918e94fa9798d9228825',
2078 'upload_date': '20090125',
2079 'artist': 'Panjabi MC',
2080 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
2081 'album': 'Beware of the Boys (Mundian To Bach Ke)',
2082 },
2083 'params': {
2084 'skip_download': True,
2085 },
2086 'skip': 'Video unavailable',
2087 },
2088 {
2089 # empty description results in an empty string
2090 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
2091 'info_dict': {
2092 'id': 'x41yOUIvK2k',
2093 'ext': 'mp4',
2094 'title': 'IMG 3456',
2095 'description': '',
2096 'upload_date': '20170613',
2097 'view_count': int,
2098 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp',
2099 'like_count': int,
2100 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw',
2101 'tags': [],
2102 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw',
2103 'availability': 'public',
2104 'age_limit': 0,
2105 'categories': ['Pets & Animals'],
2106 'duration': 7,
2107 'playable_in_embed': True,
2108 'live_status': 'not_live',
2109 'channel': 'l\'Or Vert asbl',
2110 'channel_follower_count': int,
2111 'uploader': 'l\'Or Vert asbl',
2112 'uploader_url': 'https://www.youtube.com/@ElevageOrVert',
2113 'uploader_id': '@ElevageOrVert',
2114 },
2115 'params': {
2116 'skip_download': True,
2117 },
2118 },
2119 {
2120 # with '};' inside yt initial data (see [1])
2121 # see [2] for an example with '};' inside ytInitialPlayerResponse
2122 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
2123 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
2124 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
2125 'info_dict': {
2126 'id': 'CHqg6qOn4no',
2127 'ext': 'mp4',
2128 'title': 'Part 77 Sort a list of simple types in c#',
2129 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
2130 'upload_date': '20130831',
2131 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q',
2132 'like_count': int,
2133 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q',
2134 'live_status': 'not_live',
2135 'categories': ['Education'],
2136 'availability': 'public',
2137 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg',
2138 'tags': 'count:12',
2139 'playable_in_embed': True,
2140 'age_limit': 0,
2141 'view_count': int,
2142 'duration': 522,
2143 'channel': 'kudvenkat',
2144 'comment_count': int,
2145 'channel_follower_count': int,
2146 'chapters': list,
2147 'uploader': 'kudvenkat',
2148 'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
2149 'uploader_id': '@Csharp-video-tutorialsBlogspot',
2150 },
2151 'params': {
2152 'skip_download': True,
2153 },
2154 },
2155 {
2156 # another example of '};' in ytInitialData
2157 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
2158 'only_matching': True,
2159 },
2160 {
2161 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
2162 'only_matching': True,
2163 },
2164 {
2165 # https://github.com/ytdl-org/youtube-dl/pull/28094
2166 'url': 'OtqTfy26tG0',
2167 'info_dict': {
2168 'id': 'OtqTfy26tG0',
2169 'ext': 'mp4',
2170 'title': 'Burn Out',
2171 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
2172 'upload_date': '20141120',
2173 'artist': 'The Cinematic Orchestra',
2174 'track': 'Burn Out',
2175 'album': 'Every Day',
2176 'like_count': int,
2177 'live_status': 'not_live',
2178 'alt_title': 'Burn Out',
2179 'duration': 614,
2180 'age_limit': 0,
2181 'view_count': int,
2182 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
2183 'creator': 'The Cinematic Orchestra',
2184 'channel': 'The Cinematic Orchestra',
2185 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'],
2186 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
2187 'availability': 'public',
2188 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg',
2189 'categories': ['Music'],
2190 'playable_in_embed': True,
2191 'channel_follower_count': int,
2192 'uploader': 'The Cinematic Orchestra',
2193 'comment_count': int,
2194 },
2195 'params': {
2196 'skip_download': True,
2197 },
2198 },
2199 {
2200 # controversial video, only works with bpctr when authenticated with cookies
2201 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
2202 'only_matching': True,
2203 },
2204 {
2205 # controversial video, requires bpctr/contentCheckOk
2206 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc',
2207 'info_dict': {
2208 'id': 'SZJvDhaSDnc',
2209 'ext': 'mp4',
2210 'title': 'San Diego teen commits suicide after bullying over embarrassing video',
2211 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ',
2212 'upload_date': '20140716',
2213 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7',
2214 'duration': 170,
2215 'categories': ['News & Politics'],
2216 'view_count': int,
2217 'channel': 'CBS Mornings',
2218 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'],
2219 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg',
2220 'age_limit': 18,
2221 'availability': 'needs_auth',
2222 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ',
2223 'like_count': int,
2224 'live_status': 'not_live',
2225 'playable_in_embed': True,
2226 'channel_follower_count': int,
2227 'uploader': 'CBS Mornings',
2228 'uploader_url': 'https://www.youtube.com/@CBSMornings',
2229 'uploader_id': '@CBSMornings',
2230 }
2231 },
2232 {
2233 # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
2234 'url': 'cBvYw8_A0vQ',
2235 'info_dict': {
2236 'id': 'cBvYw8_A0vQ',
2237 'ext': 'mp4',
2238 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き',
2239 'description': 'md5:ea770e474b7cd6722b4c95b833c03630',
2240 'upload_date': '20201120',
2241 'duration': 1456,
2242 'categories': ['Travel & Events'],
2243 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw',
2244 'view_count': int,
2245 'channel': 'Walk around Japan',
2246 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'],
2247 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp',
2248 'age_limit': 0,
2249 'availability': 'public',
2250 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw',
2251 'live_status': 'not_live',
2252 'playable_in_embed': True,
2253 'channel_follower_count': int,
2254 'uploader': 'Walk around Japan',
2255 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124',
2256 'uploader_id': '@walkaroundjapan7124',
2257 },
2258 'params': {
2259 'skip_download': True,
2260 },
2261 }, {
2262 # Has multiple audio streams
2263 'url': 'WaOKSUlf4TM',
2264 'only_matching': True
2265 }, {
2266 # Requires Premium: has format 141 when requested using YTM url
2267 'url': 'https://music.youtube.com/watch?v=XclachpHxis',
2268 'only_matching': True
2269 }, {
2270 # multiple subtitles with same lang_code
2271 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug',
2272 'only_matching': True,
2273 }, {
2274 # Force use android client fallback
2275 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY',
2276 'info_dict': {
2277 'id': 'YOelRv7fMxY',
2278 'title': 'DIGGING A SECRET TUNNEL Part 1',
2279 'ext': '3gp',
2280 'upload_date': '20210624',
2281 'channel_id': 'UCp68_FLety0O-n9QU6phsgw',
2282 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw',
2283 'description': 'md5:5d5991195d599b56cd0c4148907eec50',
2284 'duration': 596,
2285 'categories': ['Entertainment'],
2286 'view_count': int,
2287 'channel': 'colinfurze',
2288 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'],
2289 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg',
2290 'age_limit': 0,
2291 'availability': 'public',
2292 'like_count': int,
2293 'live_status': 'not_live',
2294 'playable_in_embed': True,
2295 'channel_follower_count': int,
2296 'chapters': list,
2297 'uploader': 'colinfurze',
2298 'uploader_url': 'https://www.youtube.com/@colinfurze',
2299 'uploader_id': '@colinfurze',
2300 },
2301 'params': {
2302 'format': '17', # 3gp format available on android
2303 'extractor_args': {'youtube': {'player_client': ['android']}},
2304 },
2305 },
2306 {
2307 # Skip download of additional client configs (remix client config in this case)
2308 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
2309 'only_matching': True,
2310 'params': {
2311 'extractor_args': {'youtube': {'player_skip': ['configs']}},
2312 },
2313 }, {
2314 # shorts
2315 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY',
2316 'only_matching': True,
2317 }, {
2318 'note': 'Storyboards',
2319 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8',
2320 'info_dict': {
2321 'id': '5KLPxDtMqe8',
2322 'ext': 'mhtml',
2323 'format_id': 'sb0',
2324 'title': 'Your Brain is Plastic',
2325 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc',
2326 'upload_date': '20140324',
2327 'like_count': int,
2328 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow',
2329 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow',
2330 'view_count': int,
2331 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg',
2332 'playable_in_embed': True,
2333 'tags': 'count:12',
2334 'availability': 'public',
2335 'channel': 'SciShow',
2336 'live_status': 'not_live',
2337 'duration': 248,
2338 'categories': ['Education'],
2339 'age_limit': 0,
2340 'channel_follower_count': int,
2341 'chapters': list,
2342 'uploader': 'SciShow',
2343 'uploader_url': 'https://www.youtube.com/@SciShow',
2344 'uploader_id': '@SciShow',
2345 }, 'params': {'format': 'mhtml', 'skip_download': True}
2346 }, {
2347 # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
2348 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2349 'info_dict': {
2350 'id': '2NUZ8W2llS4',
2351 'ext': 'mp4',
2352 'title': 'The NP that test your phone performance 🙂',
2353 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2354 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2355 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2356 'duration': 21,
2357 'view_count': int,
2358 'age_limit': 0,
2359 'categories': ['Gaming'],
2360 'tags': 'count:23',
2361 'playable_in_embed': True,
2362 'live_status': 'not_live',
2363 'upload_date': '20220103',
2364 'like_count': int,
2365 'availability': 'public',
2366 'channel': 'Leon Nguyen',
2367 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2368 'comment_count': int,
2369 'channel_follower_count': int,
2370 'uploader': 'Leon Nguyen',
2371 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2372 'uploader_id': '@LeonNguyen',
2373 }
2374 }, {
2375 # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
2376 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4',
2377 'info_dict': {
2378 'id': '2NUZ8W2llS4',
2379 'ext': 'mp4',
2380 'title': 'The NP that test your phone performance 🙂',
2381 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d',
2382 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA',
2383 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA',
2384 'duration': 21,
2385 'view_count': int,
2386 'age_limit': 0,
2387 'categories': ['Gaming'],
2388 'tags': 'count:23',
2389 'playable_in_embed': True,
2390 'live_status': 'not_live',
2391 'upload_date': '20220102',
2392 'like_count': int,
2393 'availability': 'public',
2394 'channel': 'Leon Nguyen',
2395 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp',
2396 'comment_count': int,
2397 'channel_follower_count': int,
2398 'uploader': 'Leon Nguyen',
2399 'uploader_url': 'https://www.youtube.com/@LeonNguyen',
2400 'uploader_id': '@LeonNguyen',
2401 },
2402 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']}
2403 }, {
2404 # date text is premiered video, ensure upload date in UTC (published 1641172509)
2405 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM',
2406 'info_dict': {
2407 'id': 'mzZzzBU6lrM',
2408 'ext': 'mp4',
2409 'title': 'I Met GeorgeNotFound In Real Life...',
2410 'description': 'md5:978296ec9783a031738b684d4ebf302d',
2411 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q',
2412 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q',
2413 'duration': 955,
2414 'view_count': int,
2415 'age_limit': 0,
2416 'categories': ['Entertainment'],
2417 'tags': 'count:26',
2418 'playable_in_embed': True,
2419 'live_status': 'not_live',
2420 'release_timestamp': 1641172509,
2421 'release_date': '20220103',
2422 'upload_date': '20220103',
2423 'like_count': int,
2424 'availability': 'public',
2425 'channel': 'Quackity',
2426 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg',
2427 'channel_follower_count': int,
2428 'uploader': 'Quackity',
2429 'uploader_id': '@Quackity',
2430 'uploader_url': 'https://www.youtube.com/@Quackity',
2431 }
2432 },
2433 { # continuous livestream. Microformat upload date should be preferred.
2434 # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27
2435 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU',
2436 'info_dict': {
2437 'id': 'kgx4WGK0oNU',
2438 '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}',
2439 'ext': 'mp4',
2440 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA',
2441 'availability': 'public',
2442 'age_limit': 0,
2443 'release_timestamp': 1637975704,
2444 'upload_date': '20210619',
2445 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA',
2446 'live_status': 'is_live',
2447 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg',
2448 'channel': 'Abao in Tokyo',
2449 'channel_follower_count': int,
2450 'release_date': '20211127',
2451 'tags': 'count:39',
2452 'categories': ['People & Blogs'],
2453 'like_count': int,
2454 'view_count': int,
2455 'playable_in_embed': True,
2456 'description': 'md5:2ef1d002cad520f65825346e2084e49d',
2457 'concurrent_view_count': int,
2458 'uploader': 'Abao in Tokyo',
2459 'uploader_url': 'https://www.youtube.com/@abaointokyo',
2460 'uploader_id': '@abaointokyo',
2461 },
2462 'params': {'skip_download': True}
2463 }, {
2464 # Story. Requires specific player params to work.
2465 'url': 'https://www.youtube.com/watch?v=vv8qTUWmulI',
2466 'info_dict': {
2467 'id': 'vv8qTUWmulI',
2468 'ext': 'mp4',
2469 'availability': 'unlisted',
2470 'view_count': int,
2471 'channel_id': 'UCzIZ8HrzDgc-pNQDUG6avBA',
2472 'upload_date': '20220526',
2473 'categories': ['Education'],
2474 'title': 'Story',
2475 'channel': 'IT\'S HISTORY',
2476 'description': '',
2477 'duration': 12,
2478 'playable_in_embed': True,
2479 'age_limit': 0,
2480 'live_status': 'not_live',
2481 'tags': [],
2482 'thumbnail': 'https://i.ytimg.com/vi_webp/vv8qTUWmulI/maxresdefault.webp',
2483 'channel_url': 'https://www.youtube.com/channel/UCzIZ8HrzDgc-pNQDUG6avBA',
2484 },
2485 'skip': 'stories get removed after some period of time',
2486 }, {
2487 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA',
2488 'info_dict': {
2489 'id': 'tjjjtzRLHvA',
2490 'ext': 'mp4',
2491 'title': 'ハッシュタグ無し };if window.ytcsi',
2492 'upload_date': '20220323',
2493 'like_count': int,
2494 'availability': 'unlisted',
2495 'channel': 'Lesmiscore',
2496 'thumbnail': r're:^https?://.*\.jpg',
2497 'age_limit': 0,
2498 'categories': ['Music'],
2499 'view_count': int,
2500 'description': '',
2501 'channel_url': 'https://www.youtube.com/channel/UCdqltm_7iv1Vs6kp6Syke5A',
2502 'channel_id': 'UCdqltm_7iv1Vs6kp6Syke5A',
2503 'live_status': 'not_live',
2504 'playable_in_embed': True,
2505 'channel_follower_count': int,
2506 'duration': 6,
2507 'tags': [],
2508 'uploader_id': '@lesmiscore',
2509 'uploader': 'Lesmiscore',
2510 'uploader_url': 'https://www.youtube.com/@lesmiscore',
2511 }
2512 }, {
2513 # Prefer primary title+description language metadata by default
2514 # Do not prefer translated description if primary is empty
2515 'url': 'https://www.youtube.com/watch?v=el3E4MbxRqQ',
2516 'info_dict': {
2517 'id': 'el3E4MbxRqQ',
2518 'ext': 'mp4',
2519 'title': 'dlp test video 2 - primary sv no desc',
2520 'description': '',
2521 'channel': 'cole-dlp-test-acc',
2522 'tags': [],
2523 'view_count': int,
2524 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2525 'like_count': int,
2526 'playable_in_embed': True,
2527 'availability': 'unlisted',
2528 'thumbnail': r're:^https?://.*\.jpg',
2529 'age_limit': 0,
2530 'duration': 5,
2531 'live_status': 'not_live',
2532 'upload_date': '20220908',
2533 'categories': ['People & Blogs'],
2534 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2535 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2536 'uploader_id': '@coletdjnz',
2537 'uploader': 'cole-dlp-test-acc',
2538 },
2539 'params': {'skip_download': True}
2540 }, {
2541 # Extractor argument: prefer translated title+description
2542 'url': 'https://www.youtube.com/watch?v=gHKT4uU8Zng',
2543 'info_dict': {
2544 'id': 'gHKT4uU8Zng',
2545 'ext': 'mp4',
2546 'channel': 'cole-dlp-test-acc',
2547 'tags': [],
2548 'duration': 5,
2549 'live_status': 'not_live',
2550 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
2551 'upload_date': '20220728',
2552 'view_count': int,
2553 'categories': ['People & Blogs'],
2554 'thumbnail': r're:^https?://.*\.jpg',
2555 'title': 'dlp test video title translated (fr)',
2556 'availability': 'public',
2557 'age_limit': 0,
2558 'description': 'dlp test video description translated (fr)',
2559 'playable_in_embed': True,
2560 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
2561 'uploader_url': 'https://www.youtube.com/@coletdjnz',
2562 'uploader_id': '@coletdjnz',
2563 'uploader': 'cole-dlp-test-acc',
2564 },
2565 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}},
2566 'expected_warnings': [r'Preferring "fr" translated fields'],
2567 }, {
2568 'note': '6 channel audio',
2569 'url': 'https://www.youtube.com/watch?v=zgdo7-RRjgo',
2570 'only_matching': True,
2571 }, {
2572 'note': 'Multiple HLS formats with same itag',
2573 'url': 'https://www.youtube.com/watch?v=kX3nB4PpJko',
2574 'info_dict': {
2575 'id': 'kX3nB4PpJko',
2576 'ext': 'mp4',
2577 'categories': ['Entertainment'],
2578 'description': 'md5:e8031ff6e426cdb6a77670c9b81f6fa6',
2579 'live_status': 'not_live',
2580 'duration': 937,
2581 'channel_follower_count': int,
2582 'thumbnail': 'https://i.ytimg.com/vi_webp/kX3nB4PpJko/maxresdefault.webp',
2583 'title': 'Last To Take Hand Off Jet, Keeps It!',
2584 'channel': 'MrBeast',
2585 'playable_in_embed': True,
2586 'view_count': int,
2587 'upload_date': '20221112',
2588 'channel_url': 'https://www.youtube.com/channel/UCX6OQ3DkcsbYNE6H8uQQuVA',
2589 'age_limit': 0,
2590 'availability': 'public',
2591 'channel_id': 'UCX6OQ3DkcsbYNE6H8uQQuVA',
2592 'like_count': int,
2593 'tags': [],
2594 'uploader': 'MrBeast',
2595 'uploader_url': 'https://www.youtube.com/@MrBeast',
2596 'uploader_id': '@MrBeast',
2597 },
2598 'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
2599 }, {
2600 'note': 'Audio formats with Dynamic Range Compression',
2601 'url': 'https://www.youtube.com/watch?v=Tq92D6wQ1mg',
2602 'info_dict': {
2603 'id': 'Tq92D6wQ1mg',
2604 'ext': 'webm',
2605 'title': '[MMD] Adios - EVERGLOW [+Motion DL]',
2606 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ',
2607 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ',
2608 'channel_follower_count': int,
2609 'description': 'md5:17eccca93a786d51bc67646756894066',
2610 'upload_date': '20191228',
2611 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'],
2612 'playable_in_embed': True,
2613 'like_count': int,
2614 'categories': ['Entertainment'],
2615 'thumbnail': 'https://i.ytimg.com/vi/Tq92D6wQ1mg/sddefault.jpg',
2616 'age_limit': 18,
2617 'channel': 'Projekt Melody',
2618 'view_count': int,
2619 'availability': 'needs_auth',
2620 'comment_count': int,
2621 'live_status': 'not_live',
2622 'duration': 106,
2623 'uploader': 'Projekt Melody',
2624 'uploader_id': '@ProjektMelody',
2625 'uploader_url': 'https://www.youtube.com/@ProjektMelody',
2626 },
2627 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
2628 },
2629 {
2630 'url': 'https://www.youtube.com/live/qVv6vCqciTM',
2631 'info_dict': {
2632 'id': 'qVv6vCqciTM',
2633 'ext': 'mp4',
2634 'age_limit': 0,
2635 'comment_count': int,
2636 'chapters': 'count:13',
2637 'upload_date': '20221223',
2638 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
2639 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
2640 'like_count': int,
2641 'release_date': '20221223',
2642 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
2643 '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】',
2644 'view_count': int,
2645 'playable_in_embed': True,
2646 'duration': 4438,
2647 'availability': 'public',
2648 'channel_follower_count': int,
2649 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
2650 'categories': ['Entertainment'],
2651 'live_status': 'was_live',
2652 'release_timestamp': 1671793345,
2653 'channel': 'さなちゃんねる',
2654 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
2655 'uploader': 'さなちゃんねる',
2656 'uploader_url': 'https://www.youtube.com/@sana_natori',
2657 'uploader_id': '@sana_natori',
2658 },
2659 },
2660 {
2661 # Fallbacks when webpage and web client is unavailable
2662 'url': 'https://www.youtube.com/watch?v=wSSmNUl9Snw',
2663 'info_dict': {
2664 'id': 'wSSmNUl9Snw',
2665 'ext': 'mp4',
2666 # 'categories': ['Science & Technology'],
2667 'view_count': int,
2668 'chapters': 'count:2',
2669 'channel': 'Scott Manley',
2670 'like_count': int,
2671 'age_limit': 0,
2672 # 'availability': 'public',
2673 'channel_follower_count': int,
2674 'live_status': 'not_live',
2675 'upload_date': '20170831',
2676 'duration': 682,
2677 'tags': 'count:8',
2678 'uploader_url': 'https://www.youtube.com/@scottmanley',
2679 'description': 'md5:f4bed7b200404b72a394c2f97b782c02',
2680 'uploader': 'Scott Manley',
2681 'uploader_id': '@scottmanley',
2682 'title': 'The Computer Hack That Saved Apollo 14',
2683 'channel_id': 'UCxzC4EngIsMrPmbm6Nxvb-A',
2684 'thumbnail': r're:^https?://.*\.webp',
2685 'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
2686 'playable_in_embed': True,
2687 },
2688 'params': {
2689 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
2690 },
2691 },
2692 ]
2693
2694 _WEBPAGE_TESTS = [
2695 # YouTube <object> embed
2696 {
2697 'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
2698 'md5': '873c81d308b979f0e23ee7e620b312a3',
2699 'info_dict': {
2700 'id': 'msN87y-iEx0',
2701 'ext': 'mp4',
2702 'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
2703 'upload_date': '20080526',
2704 'description': 'md5:873c81d308b979f0e23ee7e620b312a3',
2705 'age_limit': 0,
2706 'tags': ['feynman', 'mirror', 'science', 'physics', 'imagination', 'fun', 'cool', 'puzzle'],
2707 'channel_id': 'UCCeo--lls1vna5YJABWAcVA',
2708 'playable_in_embed': True,
2709 'thumbnail': 'https://i.ytimg.com/vi/msN87y-iEx0/hqdefault.jpg',
2710 'like_count': int,
2711 'comment_count': int,
2712 'channel': 'Christopher Sykes',
2713 'live_status': 'not_live',
2714 'channel_url': 'https://www.youtube.com/channel/UCCeo--lls1vna5YJABWAcVA',
2715 'availability': 'public',
2716 'duration': 195,
2717 'view_count': int,
2718 'categories': ['Science & Technology'],
2719 'channel_follower_count': int,
2720 'uploader': 'Christopher Sykes',
2721 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
2722 'uploader_id': '@ChristopherSykesDocumentaries',
2723 },
2724 'params': {
2725 'skip_download': True,
2726 }
2727 },
2728 ]
2729
2730 @classmethod
2731 def suitable(cls, url):
2732 from ..utils import parse_qs
2733
2734 qs = parse_qs(url)
2735 if qs.get('list', [None])[0]:
2736 return False
2737 return super().suitable(url)
2738
2739 def __init__(self, *args, **kwargs):
2740 super().__init__(*args, **kwargs)
2741 self._code_cache = {}
2742 self._player_cache = {}
2743
2744 def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data, is_live):
2745 lock = threading.Lock()
2746 start_time = time.time()
2747 formats = [f for f in formats if f.get('is_from_start')]
2748
2749 def refetch_manifest(format_id, delay):
2750 nonlocal formats, start_time, is_live
2751 if time.time() <= start_time + delay:
2752 return
2753
2754 _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
2755 video_details = traverse_obj(prs, (..., 'videoDetails'), expected_type=dict)
2756 microformats = traverse_obj(
2757 prs, (..., 'microformat', 'playerMicroformatRenderer'),
2758 expected_type=dict)
2759 _, live_status, _, formats, _ = self._list_formats(video_id, microformats, video_details, prs, player_url)
2760 is_live = live_status == 'is_live'
2761 start_time = time.time()
2762
2763 def mpd_feed(format_id, delay):
2764 """
2765 @returns (manifest_url, manifest_stream_number, is_live) or None
2766 """
2767 for retry in self.RetryManager(fatal=False):
2768 with lock:
2769 refetch_manifest(format_id, delay)
2770
2771 f = next((f for f in formats if f['format_id'] == format_id), None)
2772 if not f:
2773 if not is_live:
2774 retry.error = f'{video_id}: Video is no longer live'
2775 else:
2776 retry.error = f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}'
2777 continue
2778 return f['manifest_url'], f['manifest_stream_number'], is_live
2779 return None
2780
2781 for f in formats:
2782 f['is_live'] = is_live
2783 gen = functools.partial(self._live_dash_fragments, video_id, f['format_id'],
2784 live_start_time, mpd_feed, not is_live and f.copy())
2785 if is_live:
2786 f['fragments'] = gen
2787 f['protocol'] = 'http_dash_segments_generator'
2788 else:
2789 f['fragments'] = LazyList(gen({}))
2790 del f['is_from_start']
2791
2792 def _live_dash_fragments(self, video_id, format_id, live_start_time, mpd_feed, manifestless_orig_fmt, ctx):
2793 FETCH_SPAN, MAX_DURATION = 5, 432000
2794
2795 mpd_url, stream_number, is_live = None, None, True
2796
2797 begin_index = 0
2798 download_start_time = ctx.get('start') or time.time()
2799
2800 lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION
2801 if lack_early_segments:
2802 self.report_warning(bug_reports_message(
2803 'Starting download from the last 120 hours of the live stream since '
2804 'YouTube does not have data before that. If you think this is wrong,'), only_once=True)
2805 lack_early_segments = True
2806
2807 known_idx, no_fragment_score, last_segment_url = begin_index, 0, None
2808 fragments, fragment_base_url = None, None
2809
2810 def _extract_sequence_from_mpd(refresh_sequence, immediate):
2811 nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url
2812 # Obtain from MPD's maximum seq value
2813 old_mpd_url = mpd_url
2814 last_error = ctx.pop('last_error', None)
2815 expire_fast = immediate or last_error and isinstance(last_error, urllib.error.HTTPError) and last_error.code == 403
2816 mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
2817 or (mpd_url, stream_number, False))
2818 if not refresh_sequence:
2819 if expire_fast and not is_live:
2820 return False, last_seq
2821 elif old_mpd_url == mpd_url:
2822 return True, last_seq
2823 if manifestless_orig_fmt:
2824 fmt_info = manifestless_orig_fmt
2825 else:
2826 try:
2827 fmts, _ = self._extract_mpd_formats_and_subtitles(
2828 mpd_url, None, note=False, errnote=False, fatal=False)
2829 except ExtractorError:
2830 fmts = None
2831 if not fmts:
2832 no_fragment_score += 2
2833 return False, last_seq
2834 fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number)
2835 fragments = fmt_info['fragments']
2836 fragment_base_url = fmt_info['fragment_base_url']
2837 assert fragment_base_url
2838
2839 _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1))
2840 return True, _last_seq
2841
2842 self.write_debug(f'[{video_id}] Generating fragments for format {format_id}')
2843 while is_live:
2844 fetch_time = time.time()
2845 if no_fragment_score > 30:
2846 return
2847 if last_segment_url:
2848 # Obtain from "X-Head-Seqnum" header value from each segment
2849 try:
2850 urlh = self._request_webpage(
2851 last_segment_url, None, note=False, errnote=False, fatal=False)
2852 except ExtractorError:
2853 urlh = None
2854 last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum']))
2855 if last_seq is None:
2856 no_fragment_score += 2
2857 last_segment_url = None
2858 continue
2859 else:
2860 should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15)
2861 no_fragment_score += 2
2862 if not should_continue:
2863 continue
2864
2865 if known_idx > last_seq:
2866 last_segment_url = None
2867 continue
2868
2869 last_seq += 1
2870
2871 if begin_index < 0 and known_idx < 0:
2872 # skip from the start when it's negative value
2873 known_idx = last_seq + begin_index
2874 if lack_early_segments:
2875 known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration']))
2876 try:
2877 for idx in range(known_idx, last_seq):
2878 # do not update sequence here or you'll get skipped some part of it
2879 should_continue, _ = _extract_sequence_from_mpd(False, False)
2880 if not should_continue:
2881 known_idx = idx - 1
2882 raise ExtractorError('breaking out of outer loop')
2883 last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx)
2884 yield {
2885 'url': last_segment_url,
2886 'fragment_count': last_seq,
2887 }
2888 if known_idx == last_seq:
2889 no_fragment_score += 5
2890 else:
2891 no_fragment_score = 0
2892 known_idx = last_seq
2893 except ExtractorError:
2894 continue
2895
2896 if manifestless_orig_fmt:
2897 # Stop at the first iteration if running for post-live manifestless;
2898 # fragment count no longer increase since it starts
2899 break
2900
2901 time.sleep(max(0, FETCH_SPAN + fetch_time - time.time()))
2902
2903 def _extract_player_url(self, *ytcfgs, webpage=None):
2904 player_url = traverse_obj(
2905 ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'),
2906 get_all=False, expected_type=str)
2907 if not player_url:
2908 return
2909 return urljoin('https://www.youtube.com', player_url)
2910
2911 def _download_player_url(self, video_id, fatal=False):
2912 res = self._download_webpage(
2913 'https://www.youtube.com/iframe_api',
2914 note='Downloading iframe API JS', video_id=video_id, fatal=fatal)
2915 if res:
2916 player_version = self._search_regex(
2917 r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal)
2918 if player_version:
2919 return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js'
2920
2921 def _signature_cache_id(self, example_sig):
2922 """ Return a string representation of a signature """
2923 return '.'.join(str(len(part)) for part in example_sig.split('.'))
2924
2925 @classmethod
2926 def _extract_player_info(cls, player_url):
2927 for player_re in cls._PLAYER_INFO_RE:
2928 id_m = re.search(player_re, player_url)
2929 if id_m:
2930 break
2931 else:
2932 raise ExtractorError('Cannot identify player %r' % player_url)
2933 return id_m.group('id')
2934
2935 def _load_player(self, video_id, player_url, fatal=True):
2936 player_id = self._extract_player_info(player_url)
2937 if player_id not in self._code_cache:
2938 code = self._download_webpage(
2939 player_url, video_id, fatal=fatal,
2940 note='Downloading player ' + player_id,
2941 errnote='Download of %s failed' % player_url)
2942 if code:
2943 self._code_cache[player_id] = code
2944 return self._code_cache.get(player_id)
2945
2946 def _extract_signature_function(self, video_id, player_url, example_sig):
2947 player_id = self._extract_player_info(player_url)
2948
2949 # Read from filesystem cache
2950 func_id = f'js_{player_id}_{self._signature_cache_id(example_sig)}'
2951 assert os.path.basename(func_id) == func_id
2952
2953 self.write_debug(f'Extracting signature function {func_id}')
2954 cache_spec, code = self.cache.load('youtube-sigfuncs', func_id), None
2955
2956 if not cache_spec:
2957 code = self._load_player(video_id, player_url)
2958 if code:
2959 res = self._parse_sig_js(code)
2960 test_string = ''.join(map(chr, range(len(example_sig))))
2961 cache_spec = [ord(c) for c in res(test_string)]
2962 self.cache.store('youtube-sigfuncs', func_id, cache_spec)
2963
2964 return lambda s: ''.join(s[i] for i in cache_spec)
2965
2966 def _print_sig_code(self, func, example_sig):
2967 if not self.get_param('youtube_print_sig_code'):
2968 return
2969
2970 def gen_sig_code(idxs):
2971 def _genslice(start, end, step):
2972 starts = '' if start == 0 else str(start)
2973 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
2974 steps = '' if step == 1 else (':%d' % step)
2975 return f's[{starts}{ends}{steps}]'
2976
2977 step = None
2978 # Quelch pyflakes warnings - start will be set when step is set
2979 start = '(Never used)'
2980 for i, prev in zip(idxs[1:], idxs[:-1]):
2981 if step is not None:
2982 if i - prev == step:
2983 continue
2984 yield _genslice(start, prev, step)
2985 step = None
2986 continue
2987 if i - prev in [-1, 1]:
2988 step = i - prev
2989 start = prev
2990 continue
2991 else:
2992 yield 's[%d]' % prev
2993 if step is None:
2994 yield 's[%d]' % i
2995 else:
2996 yield _genslice(start, i, step)
2997
2998 test_string = ''.join(map(chr, range(len(example_sig))))
2999 cache_res = func(test_string)
3000 cache_spec = [ord(c) for c in cache_res]
3001 expr_code = ' + '.join(gen_sig_code(cache_spec))
3002 signature_id_tuple = '(%s)' % (
3003 ', '.join(str(len(p)) for p in example_sig.split('.')))
3004 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
3005 ' return %s\n') % (signature_id_tuple, expr_code)
3006 self.to_screen('Extracted signature function:\n' + code)
3007
3008 def _parse_sig_js(self, jscode):
3009 funcname = self._search_regex(
3010 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3011 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3012 r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
3013 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
3014 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}\(a,\d+\))?',
3015 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
3016 # Obsolete patterns
3017 r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3018 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
3019 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3020 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3021 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
3022 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
3023 jscode, 'Initial JS player signature function name', group='sig')
3024
3025 jsi = JSInterpreter(jscode)
3026 initial_function = jsi.extract_function(funcname)
3027 return lambda s: initial_function([s])
3028
3029 def _cached(self, func, *cache_id):
3030 def inner(*args, **kwargs):
3031 if cache_id not in self._player_cache:
3032 try:
3033 self._player_cache[cache_id] = func(*args, **kwargs)
3034 except ExtractorError as e:
3035 self._player_cache[cache_id] = e
3036 except Exception as e:
3037 self._player_cache[cache_id] = ExtractorError(traceback.format_exc(), cause=e)
3038
3039 ret = self._player_cache[cache_id]
3040 if isinstance(ret, Exception):
3041 raise ret
3042 return ret
3043 return inner
3044
3045 def _decrypt_signature(self, s, video_id, player_url):
3046 """Turn the encrypted s field into a working signature"""
3047 extract_sig = self._cached(
3048 self._extract_signature_function, 'sig', player_url, self._signature_cache_id(s))
3049 func = extract_sig(video_id, player_url, s)
3050 self._print_sig_code(func, s)
3051 return func(s)
3052
3053 def _decrypt_nsig(self, s, video_id, player_url):
3054 """Turn the encrypted n field into a working signature"""
3055 if player_url is None:
3056 raise ExtractorError('Cannot decrypt nsig without player_url')
3057 player_url = urljoin('https://www.youtube.com', player_url)
3058
3059 try:
3060 jsi, player_id, func_code = self._extract_n_function_code(video_id, player_url)
3061 except ExtractorError as e:
3062 raise ExtractorError('Unable to extract nsig function code', cause=e)
3063 if self.get_param('youtube_print_sig_code'):
3064 self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n')
3065
3066 try:
3067 extract_nsig = self._cached(self._extract_n_function_from_code, 'nsig func', player_url)
3068 ret = extract_nsig(jsi, func_code)(s)
3069 except JSInterpreter.Exception as e:
3070 try:
3071 jsi = PhantomJSwrapper(self, timeout=5000)
3072 except ExtractorError:
3073 raise e
3074 self.report_warning(
3075 f'Native nsig extraction failed: Trying with PhantomJS\n'
3076 f' n = {s} ; player = {player_url}', video_id)
3077 self.write_debug(e, only_once=True)
3078
3079 args, func_body = func_code
3080 ret = jsi.execute(
3081 f'console.log(function({", ".join(args)}) {{ {func_body} }}({s!r}));',
3082 video_id=video_id, note='Executing signature code').strip()
3083
3084 self.write_debug(f'Decrypted nsig {s} => {ret}')
3085 return ret
3086
3087 def _extract_n_function_name(self, jscode):
3088 funcname, idx = self._search_regex(
3089 r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)',
3090 jscode, 'Initial JS player n function name', group=('nfunc', 'idx'))
3091 if not idx:
3092 return funcname
3093
3094 return json.loads(js_to_json(self._search_regex(
3095 rf'var {re.escape(funcname)}\s*=\s*(\[.+?\]);', jscode,
3096 f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
3097
3098 def _extract_n_function_code(self, video_id, player_url):
3099 player_id = self._extract_player_info(player_url)
3100 func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
3101 jscode = func_code or self._load_player(video_id, player_url)
3102 jsi = JSInterpreter(jscode)
3103
3104 if func_code:
3105 return jsi, player_id, func_code
3106
3107 func_name = self._extract_n_function_name(jscode)
3108
3109 # For redundancy
3110 func_code = self._search_regex(
3111 r'''(?xs)%s\s*=\s*function\s*\((?P<var>[\w$]+)\)\s*
3112 # NB: The end of the regex is intentionally kept strict
3113 {(?P<code>.+?}\s*return\ [\w$]+.join\(""\))};''' % func_name,
3114 jscode, 'nsig function', group=('var', 'code'), default=None)
3115 if func_code:
3116 func_code = ([func_code[0]], func_code[1])
3117 else:
3118 self.write_debug('Extracting nsig function with jsinterp')
3119 func_code = jsi.extract_function_code(func_name)
3120
3121 self.cache.store('youtube-nsig', player_id, func_code)
3122 return jsi, player_id, func_code
3123
3124 def _extract_n_function_from_code(self, jsi, func_code):
3125 func = jsi.extract_function_from_code(*func_code)
3126
3127 def extract_nsig(s):
3128 try:
3129 ret = func([s])
3130 except JSInterpreter.Exception:
3131 raise
3132 except Exception as e:
3133 raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
3134
3135 if ret.startswith('enhanced_except_'):
3136 raise JSInterpreter.Exception('Signature function returned an exception')
3137 return ret
3138
3139 return extract_nsig
3140
3141 def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False):
3142 """
3143 Extract signatureTimestamp (sts)
3144 Required to tell API what sig/player version is in use.
3145 """
3146 sts = None
3147 if isinstance(ytcfg, dict):
3148 sts = int_or_none(ytcfg.get('STS'))
3149
3150 if not sts:
3151 # Attempt to extract from player
3152 if player_url is None:
3153 error_msg = 'Cannot extract signature timestamp without player_url.'
3154 if fatal:
3155 raise ExtractorError(error_msg)
3156 self.report_warning(error_msg)
3157 return
3158 code = self._load_player(video_id, player_url, fatal=fatal)
3159 if code:
3160 sts = int_or_none(self._search_regex(
3161 r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code,
3162 'JS player signature timestamp', group='sts', fatal=fatal))
3163 return sts
3164
3165 def _mark_watched(self, video_id, player_responses):
3166 for is_full, key in enumerate(('videostatsPlaybackUrl', 'videostatsWatchtimeUrl')):
3167 label = 'fully ' if is_full else ''
3168 url = get_first(player_responses, ('playbackTracking', key, 'baseUrl'),
3169 expected_type=url_or_none)
3170 if not url:
3171 self.report_warning(f'Unable to mark {label}watched')
3172 return
3173 parsed_url = urllib.parse.urlparse(url)
3174 qs = urllib.parse.parse_qs(parsed_url.query)
3175
3176 # cpn generation algorithm is reverse engineered from base.js.
3177 # In fact it works even with dummy cpn.
3178 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
3179 cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
3180
3181 # # more consistent results setting it to right before the end
3182 video_length = [str(float((qs.get('len') or ['1.5'])[0]) - 1)]
3183
3184 qs.update({
3185 'ver': ['2'],
3186 'cpn': [cpn],
3187 'cmt': video_length,
3188 'el': 'detailpage', # otherwise defaults to "shorts"
3189 })
3190
3191 if is_full:
3192 # these seem to mark watchtime "history" in the real world
3193 # they're required, so send in a single value
3194 qs.update({
3195 'st': 0,
3196 'et': video_length,
3197 })
3198
3199 url = urllib.parse.urlunparse(
3200 parsed_url._replace(query=urllib.parse.urlencode(qs, True)))
3201
3202 self._download_webpage(
3203 url, video_id, f'Marking {label}watched',
3204 'Unable to mark watched', fatal=False)
3205
3206 @classmethod
3207 def _extract_from_webpage(cls, url, webpage):
3208 # Invidious Instances
3209 # https://github.com/yt-dlp/yt-dlp/issues/195
3210 # https://github.com/iv-org/invidious/pull/1730
3211 mobj = re.search(
3212 r'<link rel="alternate" href="(?P<url>https://www\.youtube\.com/watch\?v=[0-9A-Za-z_-]{11})"',
3213 webpage)
3214 if mobj:
3215 yield cls.url_result(mobj.group('url'), cls)
3216 raise cls.StopExtraction()
3217
3218 yield from super()._extract_from_webpage(url, webpage)
3219
3220 # lazyYT YouTube embed
3221 for id_ in re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage):
3222 yield cls.url_result(unescapeHTML(id_), cls, id_)
3223
3224 # Wordpress "YouTube Video Importer" plugin
3225 for m in re.findall(r'''(?x)<div[^>]+
3226 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
3227 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage):
3228 yield cls.url_result(m[-1], cls, m[-1])
3229
3230 @classmethod
3231 def extract_id(cls, url):
3232 video_id = cls.get_temp_id(url)
3233 if not video_id:
3234 raise ExtractorError(f'Invalid URL: {url}')
3235 return video_id
3236
3237 def _extract_chapters_from_json(self, data, duration):
3238 chapter_list = traverse_obj(
3239 data, (
3240 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer',
3241 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters'
3242 ), expected_type=list)
3243
3244 return self._extract_chapters_helper(
3245 chapter_list,
3246 start_function=lambda chapter: float_or_none(
3247 traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000),
3248 title_function=lambda chapter: traverse_obj(
3249 chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str),
3250 duration=duration)
3251
3252 def _extract_chapters_from_engagement_panel(self, data, duration):
3253 content_list = traverse_obj(
3254 data,
3255 ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'),
3256 expected_type=list)
3257 chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription'))
3258 chapter_title = lambda chapter: self._get_text(chapter, 'title')
3259
3260 return next(filter(None, (
3261 self._extract_chapters_helper(traverse_obj(contents, (..., 'macroMarkersListItemRenderer')),
3262 chapter_time, chapter_title, duration)
3263 for contents in content_list)), [])
3264
3265 def _extract_heatmap_from_player_overlay(self, data):
3266 content_list = traverse_obj(data, (
3267 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer', 'decoratedPlayerBarRenderer', 'playerBar',
3268 'multiMarkersPlayerBarRenderer', 'markersMap', ..., 'value', 'heatmap', 'heatmapRenderer', 'heatMarkers', {list}))
3269 return next(filter(None, (
3270 traverse_obj(contents, (..., 'heatMarkerRenderer', {
3271 'start_time': ('timeRangeStartMillis', {functools.partial(float_or_none, scale=1000)}),
3272 'end_time': {lambda x: (x['timeRangeStartMillis'] + x['markerDurationMillis']) / 1000},
3273 'value': ('heatMarkerIntensityScoreNormalized', {float_or_none}),
3274 })) for contents in content_list)), None)
3275
3276 def _extract_comment(self, comment_renderer, parent=None):
3277 comment_id = comment_renderer.get('commentId')
3278 if not comment_id:
3279 return
3280
3281 info = {
3282 'id': comment_id,
3283 'text': self._get_text(comment_renderer, 'contentText'),
3284 'like_count': self._get_count(comment_renderer, 'voteCount'),
3285 'author_id': traverse_obj(comment_renderer, ('authorEndpoint', 'browseEndpoint', 'browseId', {self.ucid_or_none})),
3286 'author': self._get_text(comment_renderer, 'authorText'),
3287 'author_thumbnail': traverse_obj(comment_renderer, ('authorThumbnail', 'thumbnails', -1, 'url', {url_or_none})),
3288 'parent': parent or 'root',
3289 }
3290
3291 # Timestamp is an estimate calculated from the current time and time_text
3292 time_text = self._get_text(comment_renderer, 'publishedTimeText') or ''
3293 timestamp = self._parse_time_text(time_text)
3294
3295 info.update({
3296 # FIXME: non-standard, but we need a way of showing that it is an estimate.
3297 '_time_text': time_text,
3298 'timestamp': timestamp,
3299 })
3300
3301 info['author_url'] = urljoin(
3302 'https://www.youtube.com', traverse_obj(comment_renderer, ('authorEndpoint', (
3303 ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url'))),
3304 expected_type=str, get_all=False))
3305
3306 author_is_uploader = traverse_obj(comment_renderer, 'authorIsChannelOwner')
3307 if author_is_uploader is not None:
3308 info['author_is_uploader'] = author_is_uploader
3309
3310 comment_abr = traverse_obj(
3311 comment_renderer, ('actionsButtons', 'commentActionButtonsRenderer'), expected_type=dict)
3312 if comment_abr is not None:
3313 info['is_favorited'] = 'creatorHeart' in comment_abr
3314
3315 comment_ab_icontype = traverse_obj(
3316 comment_renderer, ('authorCommentBadge', 'authorCommentBadgeRenderer', 'icon', 'iconType'))
3317 if comment_ab_icontype is not None:
3318 info['author_is_verified'] = comment_ab_icontype in ('CHECK_CIRCLE_THICK', 'OFFICIAL_ARTIST_BADGE')
3319
3320 is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
3321 if is_pinned:
3322 info['is_pinned'] = True
3323
3324 return info
3325
3326 def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None):
3327
3328 get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0]
3329
3330 def extract_header(contents):
3331 _continuation = None
3332 for content in contents:
3333 comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer')
3334 expected_comment_count = self._get_count(
3335 comments_header_renderer, 'countText', 'commentsCount')
3336
3337 if expected_comment_count is not None:
3338 tracker['est_total'] = expected_comment_count
3339 self.to_screen(f'Downloading ~{expected_comment_count} comments')
3340 comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top
3341
3342 sort_menu_item = try_get(
3343 comments_header_renderer,
3344 lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {}
3345 sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {}
3346
3347 _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item)
3348 if not _continuation:
3349 continue
3350
3351 sort_text = str_or_none(sort_menu_item.get('title'))
3352 if not sort_text:
3353 sort_text = 'top comments' if comment_sort_index == 0 else 'newest first'
3354 self.to_screen('Sorting comments by %s' % sort_text.lower())
3355 break
3356 return _continuation
3357
3358 def extract_thread(contents):
3359 if not parent:
3360 tracker['current_page_thread'] = 0
3361 for content in contents:
3362 if not parent and tracker['total_parent_comments'] >= max_parents:
3363 yield
3364 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
3365 comment_renderer = get_first(
3366 (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]],
3367 expected_type=dict, default={})
3368
3369 comment = self._extract_comment(comment_renderer, parent)
3370 if not comment:
3371 continue
3372 comment_id = comment['id']
3373 if comment.get('is_pinned'):
3374 tracker['pinned_comment_ids'].add(comment_id)
3375 # Sometimes YouTube may break and give us infinite looping comments.
3376 # See: https://github.com/yt-dlp/yt-dlp/issues/6290
3377 if comment_id in tracker['seen_comment_ids']:
3378 if comment_id in tracker['pinned_comment_ids'] and not comment.get('is_pinned'):
3379 # Pinned comments may appear a second time in newest first sort
3380 # See: https://github.com/yt-dlp/yt-dlp/issues/6712
3381 continue
3382 self.report_warning('Detected YouTube comments looping. Stopping comment extraction as we probably cannot get any more.')
3383 yield
3384 else:
3385 tracker['seen_comment_ids'].add(comment['id'])
3386
3387 tracker['running_total'] += 1
3388 tracker['total_reply_comments' if parent else 'total_parent_comments'] += 1
3389 yield comment
3390
3391 # Attempt to get the replies
3392 comment_replies_renderer = try_get(
3393 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
3394
3395 if comment_replies_renderer:
3396 tracker['current_page_thread'] += 1
3397 comment_entries_iter = self._comment_entries(
3398 comment_replies_renderer, ytcfg, video_id,
3399 parent=comment.get('id'), tracker=tracker)
3400 yield from itertools.islice(comment_entries_iter, min(
3401 max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments'])))
3402
3403 # Keeps track of counts across recursive calls
3404 if not tracker:
3405 tracker = dict(
3406 running_total=0,
3407 est_total=None,
3408 current_page_thread=0,
3409 total_parent_comments=0,
3410 total_reply_comments=0,
3411 seen_comment_ids=set(),
3412 pinned_comment_ids=set()
3413 )
3414
3415 # TODO: Deprecated
3416 # YouTube comments have a max depth of 2
3417 max_depth = int_or_none(get_single_config_arg('max_comment_depth'))
3418 if max_depth:
3419 self._downloader.deprecated_feature('[youtube] max_comment_depth extractor argument is deprecated. '
3420 'Set max replies in the max-comments extractor argument instead')
3421 if max_depth == 1 and parent:
3422 return
3423
3424 max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map(
3425 lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4)
3426
3427 continuation = self._extract_continuation(root_continuation_data)
3428
3429 response = None
3430 is_forced_continuation = False
3431 is_first_continuation = parent is None
3432 if is_first_continuation and not continuation:
3433 # Sometimes you can get comments by generating the continuation yourself,
3434 # even if YouTube initially reports them being disabled - e.g. stories comments.
3435 # Note: if the comment section is actually disabled, YouTube may return a response with
3436 # required check_get_keys missing. So we will disable that check initially in this case.
3437 continuation = self._build_api_continuation_query(self._generate_comment_continuation(video_id))
3438 is_forced_continuation = True
3439
3440 continuation_items_path = (
3441 'onResponseReceivedEndpoints', ..., ('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems')
3442 for page_num in itertools.count(0):
3443 if not continuation:
3444 break
3445 headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response))
3446 comment_prog_str = f"({tracker['running_total']}/~{tracker['est_total']})"
3447 if page_num == 0:
3448 if is_first_continuation:
3449 note_prefix = 'Downloading comment section API JSON'
3450 else:
3451 note_prefix = ' Downloading comment API JSON reply thread %d %s' % (
3452 tracker['current_page_thread'], comment_prog_str)
3453 else:
3454 note_prefix = '%sDownloading comment%s API JSON page %d %s' % (
3455 ' ' if parent else '', ' replies' if parent else '',
3456 page_num, comment_prog_str)
3457
3458 # Do a deep check for incomplete data as sometimes YouTube may return no comments for a continuation
3459 # Ignore check if YouTube says the comment count is 0.
3460 check_get_keys = None
3461 if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0):
3462 check_get_keys = [[*continuation_items_path, ..., (
3463 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]]
3464 try:
3465 response = self._extract_response(
3466 item_id=None, query=continuation,
3467 ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix,
3468 check_get_keys=check_get_keys)
3469 except ExtractorError as e:
3470 # Ignore incomplete data error for replies if retries didn't work.
3471 # This is to allow any other parent comments and comment threads to be downloaded.
3472 # See: https://github.com/yt-dlp/yt-dlp/issues/4669
3473 if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True:
3474 self.report_warning(
3475 'Received incomplete data for a comment reply thread and retrying did not help. '
3476 'Ignoring to let other comments be downloaded.')
3477 else:
3478 raise
3479 is_forced_continuation = False
3480 continuation = None
3481 for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
3482 if is_first_continuation:
3483 continuation = extract_header(continuation_items)
3484 is_first_continuation = False
3485 if continuation:
3486 break
3487 continue
3488
3489 for entry in extract_thread(continuation_items):
3490 if not entry:
3491 return
3492 yield entry
3493 continuation = self._extract_continuation({'contents': continuation_items})
3494 if continuation:
3495 break
3496
3497 message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1)
3498 if message and not parent and tracker['running_total'] == 0:
3499 self.report_warning(f'Youtube said: {message}', video_id=video_id, only_once=True)
3500 raise self.CommentsDisabled
3501
3502 @staticmethod
3503 def _generate_comment_continuation(video_id):
3504 """
3505 Generates initial comment section continuation token from given video id
3506 """
3507 token = f'\x12\r\x12\x0b{video_id}\x18\x062\'"\x11"\x0b{video_id}0\x00x\x020\x00B\x10comments-section'
3508 return base64.b64encode(token.encode()).decode()
3509
3510 def _get_comments(self, ytcfg, video_id, contents, webpage):
3511 """Entry for comment extraction"""
3512 def _real_comment_extract(contents):
3513 renderer = next((
3514 item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={})
3515 if item.get('sectionIdentifier') == 'comment-item-section'), None)
3516 yield from self._comment_entries(renderer, ytcfg, video_id)
3517
3518 max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0])
3519 return itertools.islice(_real_comment_extract(contents), 0, max_comments)
3520
3521 @staticmethod
3522 def _get_checkok_params():
3523 return {'contentCheckOk': True, 'racyCheckOk': True}
3524
3525 @classmethod
3526 def _generate_player_context(cls, sts=None):
3527 context = {
3528 'html5Preference': 'HTML5_PREF_WANTS',
3529 }
3530 if sts is not None:
3531 context['signatureTimestamp'] = sts
3532 return {
3533 'playbackContext': {
3534 'contentPlaybackContext': context
3535 },
3536 **cls._get_checkok_params()
3537 }
3538
3539 @staticmethod
3540 def _is_agegated(player_response):
3541 if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')):
3542 return True
3543
3544 reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')))
3545 AGE_GATE_REASONS = (
3546 'confirm your age', 'age-restricted', 'inappropriate', # reason
3547 'age_verification_required', 'age_check_required', # status
3548 )
3549 return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons)
3550
3551 @staticmethod
3552 def _is_unplayable(player_response):
3553 return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
3554
3555 _STORY_PLAYER_PARAMS = '8AEB'
3556
3557 def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
3558
3559 session_index = self._extract_session_index(player_ytcfg, master_ytcfg)
3560 syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr)
3561 sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None
3562 headers = self.generate_api_headers(
3563 ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client)
3564
3565 yt_query = {
3566 'videoId': video_id,
3567 }
3568 if smuggled_data.get('is_story') or _split_innertube_client(client)[0] == 'android':
3569 yt_query['params'] = self._STORY_PLAYER_PARAMS
3570
3571 yt_query.update(self._generate_player_context(sts))
3572 return self._extract_response(
3573 item_id=video_id, ep='player', query=yt_query,
3574 ytcfg=player_ytcfg, headers=headers, fatal=True,
3575 default_client=client,
3576 note='Downloading %s player API JSON' % client.replace('_', ' ').strip()
3577 ) or None
3578
3579 def _get_requested_clients(self, url, smuggled_data):
3580 requested_clients = []
3581 default = ['android', 'web']
3582 allowed_clients = sorted(
3583 (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
3584 key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
3585 for client in self._configuration_arg('player_client'):
3586 if client in allowed_clients:
3587 requested_clients.append(client)
3588 elif client == 'default':
3589 requested_clients.extend(default)
3590 elif client == 'all':
3591 requested_clients.extend(allowed_clients)
3592 else:
3593 self.report_warning(f'Skipping unsupported client {client}')
3594 if not requested_clients:
3595 requested_clients = default
3596
3597 if smuggled_data.get('is_music_url') or self.is_music_url(url):
3598 requested_clients.extend(
3599 f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS)
3600
3601 return orderedSet(requested_clients)
3602
3603 def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg, smuggled_data):
3604 initial_pr = None
3605 if webpage:
3606 initial_pr = self._search_json(
3607 self._YT_INITIAL_PLAYER_RESPONSE_RE, webpage, 'initial player response', video_id, fatal=False)
3608
3609 all_clients = set(clients)
3610 clients = clients[::-1]
3611 prs = []
3612
3613 def append_client(*client_names):
3614 """ Append the first client name that exists but not already used """
3615 for client_name in client_names:
3616 actual_client = _split_innertube_client(client_name)[0]
3617 if actual_client in INNERTUBE_CLIENTS:
3618 if actual_client not in all_clients:
3619 clients.append(client_name)
3620 all_clients.add(actual_client)
3621 return
3622
3623 # Android player_response does not have microFormats which are needed for
3624 # extraction of some data. So we return the initial_pr with formats
3625 # stripped out even if not requested by the user
3626 # See: https://github.com/yt-dlp/yt-dlp/issues/501
3627 if initial_pr:
3628 pr = dict(initial_pr)
3629 pr['streamingData'] = None
3630 prs.append(pr)
3631
3632 last_error = None
3633 tried_iframe_fallback = False
3634 player_url = None
3635 while clients:
3636 client, base_client, variant = _split_innertube_client(clients.pop())
3637 player_ytcfg = master_ytcfg if client == 'web' else {}
3638 if 'configs' not in self._configuration_arg('player_skip') and client != 'web':
3639 player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg
3640
3641 player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage)
3642 require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER')
3643 if 'js' in self._configuration_arg('player_skip'):
3644 require_js_player = False
3645 player_url = None
3646
3647 if not player_url and not tried_iframe_fallback and require_js_player:
3648 player_url = self._download_player_url(video_id)
3649 tried_iframe_fallback = True
3650
3651 try:
3652 pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response(
3653 client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr, smuggled_data)
3654 except ExtractorError as e:
3655 if last_error:
3656 self.report_warning(last_error)
3657 last_error = e
3658 continue
3659
3660 if pr:
3661 # YouTube may return a different video player response than expected.
3662 # See: https://github.com/TeamNewPipe/NewPipe/issues/8713
3663 pr_video_id = traverse_obj(pr, ('videoDetails', 'videoId'))
3664 if pr_video_id and pr_video_id != video_id:
3665 self.report_warning(
3666 f'Skipping player response from {client} client (got player response for video "{pr_video_id}" instead of "{video_id}")' + bug_reports_message())
3667 else:
3668 # Save client name for introspection later
3669 name = short_client_name(client)
3670 sd = traverse_obj(pr, ('streamingData', {dict})) or {}
3671 sd[STREAMING_DATA_CLIENT_NAME] = name
3672 for f in traverse_obj(sd, (('formats', 'adaptiveFormats'), ..., {dict})):
3673 f[STREAMING_DATA_CLIENT_NAME] = name
3674 prs.append(pr)
3675
3676 # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in
3677 if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated:
3678 append_client(f'{base_client}_creator')
3679 elif self._is_agegated(pr):
3680 if variant == 'tv_embedded':
3681 append_client(f'{base_client}_embedded')
3682 elif not variant:
3683 append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded')
3684
3685 if last_error:
3686 if not len(prs):
3687 raise last_error
3688 self.report_warning(last_error)
3689 return prs, player_url
3690
3691 def _needs_live_processing(self, live_status, duration):
3692 if (live_status == 'is_live' and self.get_param('live_from_start')
3693 or live_status == 'post_live' and (duration or 0) > 4 * 3600):
3694 return live_status
3695
3696 def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
3697 CHUNK_SIZE = 10 << 20
3698 itags, stream_ids = collections.defaultdict(set), []
3699 itag_qualities, res_qualities = {}, {0: None}
3700 q = qualities([
3701 # Normally tiny is the smallest video-only formats. But
3702 # audio-only formats with unknown quality may get tagged as tiny
3703 'tiny',
3704 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats
3705 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
3706 ])
3707 streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
3708 all_formats = self._configuration_arg('include_duplicate_formats')
3709
3710 def build_fragments(f):
3711 return LazyList({
3712 'url': update_url_query(f['url'], {
3713 'range': f'{range_start}-{min(range_start + CHUNK_SIZE - 1, f["filesize"])}'
3714 })
3715 } for range_start in range(0, f['filesize'], CHUNK_SIZE))
3716
3717 for fmt in streaming_formats:
3718 if fmt.get('targetDurationSec'):
3719 continue
3720
3721 itag = str_or_none(fmt.get('itag'))
3722 audio_track = fmt.get('audioTrack') or {}
3723 stream_id = (itag, audio_track.get('id'), fmt.get('isDrc'))
3724 if not all_formats:
3725 if stream_id in stream_ids:
3726 continue
3727
3728 quality = fmt.get('quality')
3729 height = int_or_none(fmt.get('height'))
3730 if quality == 'tiny' or not quality:
3731 quality = fmt.get('audioQuality', '').lower() or quality
3732 # The 3gp format (17) in android client has a quality of "small",
3733 # but is actually worse than other formats
3734 if itag == '17':
3735 quality = 'tiny'
3736 if quality:
3737 if itag:
3738 itag_qualities[itag] = quality
3739 if height:
3740 res_qualities[height] = quality
3741 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
3742 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
3743 # number of fragment that would subsequently requested with (`&sq=N`)
3744 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
3745 continue
3746
3747 fmt_url = fmt.get('url')
3748 if not fmt_url:
3749 sc = urllib.parse.parse_qs(fmt.get('signatureCipher'))
3750 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
3751 encrypted_sig = try_get(sc, lambda x: x['s'][0])
3752 if not all((sc, fmt_url, player_url, encrypted_sig)):
3753 continue
3754 try:
3755 fmt_url += '&%s=%s' % (
3756 traverse_obj(sc, ('sp', -1)) or 'signature',
3757 self._decrypt_signature(encrypted_sig, video_id, player_url)
3758 )
3759 except ExtractorError as e:
3760 self.report_warning('Signature extraction failed: Some formats may be missing',
3761 video_id=video_id, only_once=True)
3762 self.write_debug(e, only_once=True)
3763 continue
3764
3765 query = parse_qs(fmt_url)
3766 throttled = False
3767 if query.get('n'):
3768 try:
3769 decrypt_nsig = self._cached(self._decrypt_nsig, 'nsig', query['n'][0])
3770 fmt_url = update_url_query(fmt_url, {
3771 'n': decrypt_nsig(query['n'][0], video_id, player_url)
3772 })
3773 except ExtractorError as e:
3774 phantomjs_hint = ''
3775 if isinstance(e, JSInterpreter.Exception):
3776 phantomjs_hint = (f' Install {self._downloader._format_err("PhantomJS", self._downloader.Styles.EMPHASIS)} '
3777 f'to workaround the issue. {PhantomJSwrapper.INSTALL_HINT}\n')
3778 if player_url:
3779 self.report_warning(
3780 f'nsig extraction failed: You may experience throttling for some formats\n{phantomjs_hint}'
3781 f' n = {query["n"][0]} ; player = {player_url}', video_id=video_id, only_once=True)
3782 self.write_debug(e, only_once=True)
3783 else:
3784 self.report_warning(
3785 'Cannot decrypt nsig without player_url: You may experience throttling for some formats',
3786 video_id=video_id, only_once=True)
3787 throttled = True
3788
3789 tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
3790 language_preference = (
3791 10 if audio_track.get('audioIsDefault') and 10
3792 else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10
3793 else -1)
3794 # Some formats may have much smaller duration than others (possibly damaged during encoding)
3795 # E.g. 2-nOtRESiUc Ref: https://github.com/yt-dlp/yt-dlp/issues/2823
3796 # Make sure to avoid false positives with small duration differences.
3797 # E.g. __2ABJjxzNo, ySuUZEjARPY
3798 is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500)
3799 if is_damaged:
3800 self.report_warning(
3801 f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
3802
3803 client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
3804 dct = {
3805 'asr': int_or_none(fmt.get('audioSampleRate')),
3806 'filesize': int_or_none(fmt.get('contentLength')),
3807 'format_id': f'{itag}{"-drc" if fmt.get("isDrc") else ""}',
3808 'format_note': join_nonempty(
3809 join_nonempty(audio_track.get('displayName'),
3810 language_preference > 0 and ' (default)', delim=''),
3811 fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
3812 fmt.get('isDrc') and 'DRC',
3813 try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()),
3814 try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()),
3815 throttled and 'THROTTLED', is_damaged and 'DAMAGED',
3816 (self.get_param('verbose') or all_formats) and client_name,
3817 delim=', '),
3818 # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372
3819 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
3820 'fps': int_or_none(fmt.get('fps')) or None,
3821 'audio_channels': fmt.get('audioChannels'),
3822 'height': height,
3823 'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
3824 'has_drm': bool(fmt.get('drmFamilies')),
3825 'tbr': tbr,
3826 'url': fmt_url,
3827 'width': int_or_none(fmt.get('width')),
3828 'language': join_nonempty(audio_track.get('id', '').split('.')[0],
3829 'desc' if language_preference < -1 else '') or None,
3830 'language_preference': language_preference,
3831 # Strictly de-prioritize damaged and 3gp formats
3832 'preference': -10 if is_damaged else -2 if itag == '17' else None,
3833 }
3834 mime_mobj = re.match(
3835 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '')
3836 if mime_mobj:
3837 dct['ext'] = mimetype2ext(mime_mobj.group(1))
3838 dct.update(parse_codecs(mime_mobj.group(2)))
3839 if itag:
3840 itags[itag].add(('https', dct.get('language')))
3841 stream_ids.append(stream_id)
3842 single_stream = 'none' in (dct.get('acodec'), dct.get('vcodec'))
3843 if single_stream and dct.get('ext'):
3844 dct['container'] = dct['ext'] + '_dash'
3845
3846 if all_formats and dct['filesize']:
3847 yield {
3848 **dct,
3849 'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
3850 'protocol': 'http_dash_segments',
3851 'fragments': build_fragments(dct),
3852 }
3853 dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
3854 yield dct
3855
3856 needs_live_processing = self._needs_live_processing(live_status, duration)
3857 skip_bad_formats = not self._configuration_arg('include_incomplete_formats')
3858
3859 skip_manifests = set(self._configuration_arg('skip'))
3860 if (not self.get_param('youtube_include_hls_manifest', True)
3861 or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
3862 or needs_live_processing and skip_bad_formats):
3863 skip_manifests.add('hls')
3864
3865 if not self.get_param('youtube_include_dash_manifest', True):
3866 skip_manifests.add('dash')
3867 if self._configuration_arg('include_live_dash'):
3868 self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
3869 'Use include_incomplete_formats extractor argument instead')
3870 elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
3871 skip_manifests.add('dash')
3872
3873 def process_manifest_format(f, proto, client_name, itag):
3874 key = (proto, f.get('language'))
3875 if not all_formats and key in itags[itag]:
3876 return False
3877 itags[itag].add(key)
3878
3879 if itag and all_formats:
3880 f['format_id'] = f'{itag}-{proto}'
3881 elif any(p != proto for p, _ in itags[itag]):
3882 f['format_id'] = f'{itag}-{proto}'
3883 elif itag:
3884 f['format_id'] = itag
3885
3886 f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
3887 if f['quality'] == -1 and f.get('height'):
3888 f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
3889 if self.get_param('verbose'):
3890 f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
3891 return True
3892
3893 subtitles = {}
3894 for sd in streaming_data:
3895 client_name = sd.get(STREAMING_DATA_CLIENT_NAME)
3896
3897 hls_manifest_url = 'hls' not in skip_manifests and sd.get('hlsManifestUrl')
3898 if hls_manifest_url:
3899 fmts, subs = self._extract_m3u8_formats_and_subtitles(
3900 hls_manifest_url, video_id, 'mp4', fatal=False, live=live_status == 'is_live')
3901 subtitles = self._merge_subtitles(subs, subtitles)
3902 for f in fmts:
3903 if process_manifest_format(f, 'hls', client_name, self._search_regex(
3904 r'/itag/(\d+)', f['url'], 'itag', default=None)):
3905 yield f
3906
3907 dash_manifest_url = 'dash' not in skip_manifests and sd.get('dashManifestUrl')
3908 if dash_manifest_url:
3909 formats, subs = self._extract_mpd_formats_and_subtitles(dash_manifest_url, video_id, fatal=False)
3910 subtitles = self._merge_subtitles(subs, subtitles) # Prioritize HLS subs over DASH
3911 for f in formats:
3912 if process_manifest_format(f, 'dash', client_name, f['format_id']):
3913 f['filesize'] = int_or_none(self._search_regex(
3914 r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None))
3915 if needs_live_processing:
3916 f['is_from_start'] = True
3917
3918 yield f
3919 yield subtitles
3920
3921 def _extract_storyboard(self, player_responses, duration):
3922 spec = get_first(
3923 player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1]
3924 base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None))
3925 if not base_url:
3926 return
3927 L = len(spec) - 1
3928 for i, args in enumerate(spec):
3929 args = args.split('#')
3930 counts = list(map(int_or_none, args[:5]))
3931 if len(args) != 8 or not all(counts):
3932 self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}')
3933 continue
3934 width, height, frame_count, cols, rows = counts
3935 N, sigh = args[6:]
3936
3937 url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}'
3938 fragment_count = frame_count / (cols * rows)
3939 fragment_duration = duration / fragment_count
3940 yield {
3941 'format_id': f'sb{i}',
3942 'format_note': 'storyboard',
3943 'ext': 'mhtml',
3944 'protocol': 'mhtml',
3945 'acodec': 'none',
3946 'vcodec': 'none',
3947 'url': url,
3948 'width': width,
3949 'height': height,
3950 'fps': frame_count / duration,
3951 'rows': rows,
3952 'columns': cols,
3953 'fragments': [{
3954 'url': url.replace('$M', str(j)),
3955 'duration': min(fragment_duration, duration - (j * fragment_duration)),
3956 } for j in range(math.ceil(fragment_count))],
3957 }
3958
3959 def _download_player_responses(self, url, smuggled_data, video_id, webpage_url):
3960 webpage = None
3961 if 'webpage' not in self._configuration_arg('player_skip'):
3962 query = {'bpctr': '9999999999', 'has_verified': '1'}
3963 if smuggled_data.get('is_story'):
3964 query['pp'] = self._STORY_PLAYER_PARAMS
3965 webpage = self._download_webpage(
3966 webpage_url, video_id, fatal=False, query=query)
3967
3968 master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
3969
3970 player_responses, player_url = self._extract_player_responses(
3971 self._get_requested_clients(url, smuggled_data),
3972 video_id, webpage, master_ytcfg, smuggled_data)
3973
3974 return webpage, master_ytcfg, player_responses, player_url
3975
3976 def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None):
3977 live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails'))
3978 is_live = get_first(video_details, 'isLive')
3979 if is_live is None:
3980 is_live = get_first(live_broadcast_details, 'isLiveNow')
3981 live_content = get_first(video_details, 'isLiveContent')
3982 is_upcoming = get_first(video_details, 'isUpcoming')
3983 post_live = get_first(video_details, 'isPostLiveDvr')
3984 live_status = ('post_live' if post_live
3985 else 'is_live' if is_live
3986 else 'is_upcoming' if is_upcoming
3987 else 'was_live' if live_content
3988 else 'not_live' if False in (is_live, live_content)
3989 else None)
3990 streaming_data = traverse_obj(player_responses, (..., 'streamingData'))
3991 *formats, subtitles = self._extract_formats_and_subtitles(streaming_data, video_id, player_url, live_status, duration)
3992
3993 return live_broadcast_details, live_status, streaming_data, formats, subtitles
3994
3995 def _real_extract(self, url):
3996 url, smuggled_data = unsmuggle_url(url, {})
3997 video_id = self._match_id(url)
3998
3999 base_url = self.http_scheme() + '//www.youtube.com/'
4000 webpage_url = base_url + 'watch?v=' + video_id
4001
4002 webpage, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url)
4003
4004 playability_statuses = traverse_obj(
4005 player_responses, (..., 'playabilityStatus'), expected_type=dict)
4006
4007 trailer_video_id = get_first(
4008 playability_statuses,
4009 ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'),
4010 expected_type=str)
4011 if trailer_video_id:
4012 return self.url_result(
4013 trailer_video_id, self.ie_key(), trailer_video_id)
4014
4015 search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None))
4016 if webpage else (lambda x: None))
4017
4018 video_details = traverse_obj(player_responses, (..., 'videoDetails'), expected_type=dict)
4019 microformats = traverse_obj(
4020 player_responses, (..., 'microformat', 'playerMicroformatRenderer'),
4021 expected_type=dict)
4022
4023 translated_title = self._get_text(microformats, (..., 'title'))
4024 video_title = (self._preferred_lang and translated_title
4025 or get_first(video_details, 'title') # primary
4026 or translated_title
4027 or search_meta(['og:title', 'twitter:title', 'title']))
4028 translated_description = self._get_text(microformats, (..., 'description'))
4029 original_description = get_first(video_details, 'shortDescription')
4030 video_description = (
4031 self._preferred_lang and translated_description
4032 # If original description is blank, it will be an empty string.
4033 # Do not prefer translated description in this case.
4034 or original_description if original_description is not None else translated_description)
4035
4036 multifeed_metadata_list = get_first(
4037 player_responses,
4038 ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'),
4039 expected_type=str)
4040 if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'):
4041 if self.get_param('noplaylist'):
4042 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
4043 else:
4044 entries = []
4045 feed_ids = []
4046 for feed in multifeed_metadata_list.split(','):
4047 # Unquote should take place before split on comma (,) since textual
4048 # fields may contain comma as well (see
4049 # https://github.com/ytdl-org/youtube-dl/issues/8536)
4050 feed_data = urllib.parse.parse_qs(
4051 urllib.parse.unquote_plus(feed))
4052
4053 def feed_entry(name):
4054 return try_get(
4055 feed_data, lambda x: x[name][0], str)
4056
4057 feed_id = feed_entry('id')
4058 if not feed_id:
4059 continue
4060 feed_title = feed_entry('title')
4061 title = video_title
4062 if feed_title:
4063 title += ' (%s)' % feed_title
4064 entries.append({
4065 '_type': 'url_transparent',
4066 'ie_key': 'Youtube',
4067 'url': smuggle_url(
4068 '%swatch?v=%s' % (base_url, feed_data['id'][0]),
4069 {'force_singlefeed': True}),
4070 'title': title,
4071 })
4072 feed_ids.append(feed_id)
4073 self.to_screen(
4074 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
4075 % (', '.join(feed_ids), video_id))
4076 return self.playlist_result(
4077 entries, video_id, video_title, video_description)
4078
4079 duration = (int_or_none(get_first(video_details, 'lengthSeconds'))
4080 or int_or_none(get_first(microformats, 'lengthSeconds'))
4081 or parse_duration(search_meta('duration')) or None)
4082
4083 live_broadcast_details, live_status, streaming_data, formats, automatic_captions = \
4084 self._list_formats(video_id, microformats, video_details, player_responses, player_url, duration)
4085 if live_status == 'post_live':
4086 self.write_debug(f'{video_id}: Video is in Post-Live Manifestless mode')
4087
4088 if not formats:
4089 if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')):
4090 self.report_drm(video_id)
4091 pemr = get_first(
4092 playability_statuses,
4093 ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {}
4094 reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason')
4095 subreason = clean_html(self._get_text(pemr, 'subreason') or '')
4096 if subreason:
4097 if subreason == 'The uploader has not made this video available in your country.':
4098 countries = get_first(microformats, 'availableCountries')
4099 if not countries:
4100 regions_allowed = search_meta('regionsAllowed')
4101 countries = regions_allowed.split(',') if regions_allowed else None
4102 self.raise_geo_restricted(subreason, countries, metadata_available=True)
4103 reason += f'. {subreason}'
4104 if reason:
4105 self.raise_no_formats(reason, expected=True)
4106
4107 keywords = get_first(video_details, 'keywords', expected_type=list) or []
4108 if not keywords and webpage:
4109 keywords = [
4110 unescapeHTML(m.group('content'))
4111 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
4112 for keyword in keywords:
4113 if keyword.startswith('yt:stretch='):
4114 mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword)
4115 if mobj:
4116 # NB: float is intentional for forcing float division
4117 w, h = (float(v) for v in mobj.groups())
4118 if w > 0 and h > 0:
4119 ratio = w / h
4120 for f in formats:
4121 if f.get('vcodec') != 'none':
4122 f['stretched_ratio'] = ratio
4123 break
4124 thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail'))
4125 thumbnail_url = search_meta(['og:image', 'twitter:image'])
4126 if thumbnail_url:
4127 thumbnails.append({
4128 'url': thumbnail_url,
4129 })
4130 original_thumbnails = thumbnails.copy()
4131
4132 # The best resolution thumbnails sometimes does not appear in the webpage
4133 # See: https://github.com/yt-dlp/yt-dlp/issues/340
4134 # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029>
4135 thumbnail_names = [
4136 # While the *1,*2,*3 thumbnails are just below their corresponding "*default" variants
4137 # in resolution, these are not the custom thumbnail. So de-prioritize them
4138 'maxresdefault', 'hq720', 'sddefault', 'hqdefault', '0', 'mqdefault', 'default',
4139 'sd1', 'sd2', 'sd3', 'hq1', 'hq2', 'hq3', 'mq1', 'mq2', 'mq3', '1', '2', '3'
4140 ]
4141 n_thumbnail_names = len(thumbnail_names)
4142 thumbnails.extend({
4143 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format(
4144 video_id=video_id, name=name, ext=ext,
4145 webp='_webp' if ext == 'webp' else '', live='_live' if live_status == 'is_live' else ''),
4146 } for name in thumbnail_names for ext in ('webp', 'jpg'))
4147 for thumb in thumbnails:
4148 i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names)
4149 thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i)
4150 self._remove_duplicate_formats(thumbnails)
4151 self._downloader._sort_thumbnails(original_thumbnails)
4152
4153 category = get_first(microformats, 'category') or search_meta('genre')
4154 channel_id = self.ucid_or_none(str_or_none(
4155 get_first(video_details, 'channelId')
4156 or get_first(microformats, 'externalChannelId')
4157 or search_meta('channelId')))
4158 owner_profile_url = get_first(microformats, 'ownerProfileUrl')
4159
4160 live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp'))
4161 live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp'))
4162 if not duration and live_end_time and live_start_time:
4163 duration = live_end_time - live_start_time
4164
4165 needs_live_processing = self._needs_live_processing(live_status, duration)
4166
4167 def is_bad_format(fmt):
4168 if needs_live_processing and not fmt.get('is_from_start'):
4169 return True
4170 elif (live_status == 'is_live' and needs_live_processing != 'is_live'
4171 and fmt.get('protocol') == 'http_dash_segments'):
4172 return True
4173
4174 for fmt in filter(is_bad_format, formats):
4175 fmt['preference'] = (fmt.get('preference') or -1) - 10
4176 fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 4 hours)', delim=' ')
4177
4178 if needs_live_processing:
4179 self._prepare_live_from_start_formats(
4180 formats, video_id, live_start_time, url, webpage_url, smuggled_data, live_status == 'is_live')
4181
4182 formats.extend(self._extract_storyboard(player_responses, duration))
4183
4184 channel_handle = self.handle_from_url(owner_profile_url)
4185
4186 info = {
4187 'id': video_id,
4188 'title': video_title,
4189 'formats': formats,
4190 'thumbnails': thumbnails,
4191 # The best thumbnail that we are sure exists. Prevents unnecessary
4192 # URL checking if user don't care about getting the best possible thumbnail
4193 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')),
4194 'description': video_description,
4195 'channel_id': channel_id,
4196 'channel_url': format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None),
4197 'duration': duration,
4198 'view_count': int_or_none(
4199 get_first((video_details, microformats), (..., 'viewCount'))
4200 or search_meta('interactionCount')),
4201 'average_rating': float_or_none(get_first(video_details, 'averageRating')),
4202 'age_limit': 18 if (
4203 get_first(microformats, 'isFamilySafe') is False
4204 or search_meta('isFamilyFriendly') == 'false'
4205 or search_meta('og:restrictions:age') == '18+') else 0,
4206 'webpage_url': webpage_url,
4207 'categories': [category] if category else None,
4208 'tags': keywords,
4209 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'),
4210 'live_status': live_status,
4211 'release_timestamp': live_start_time,
4212 '_format_sort_fields': ( # source_preference is lower for throttled/potentially damaged formats
4213 'quality', 'res', 'fps', 'hdr:12', 'source', 'vcodec:vp9.2', 'channels', 'acodec', 'lang', 'proto')
4214 }
4215
4216 subtitles = {}
4217 pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict)
4218 if pctr:
4219 def get_lang_code(track):
4220 return (remove_start(track.get('vssId') or '', '.').replace('.', '-')
4221 or track.get('languageCode'))
4222
4223 # Converted into dicts to remove duplicates
4224 captions = {
4225 get_lang_code(sub): sub
4226 for sub in traverse_obj(pctr, (..., 'captionTracks', ...))}
4227 translation_languages = {
4228 lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1)
4229 for lang in traverse_obj(pctr, (..., 'translationLanguages', ...))}
4230
4231 def process_language(container, base_url, lang_code, sub_name, query):
4232 lang_subs = container.setdefault(lang_code, [])
4233 for fmt in self._SUBTITLE_FORMATS:
4234 query.update({
4235 'fmt': fmt,
4236 })
4237 lang_subs.append({
4238 'ext': fmt,
4239 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)),
4240 'name': sub_name,
4241 })
4242
4243 # NB: Constructing the full subtitle dictionary is slow
4244 get_translated_subs = 'translated_subs' not in self._configuration_arg('skip') and (
4245 self.get_param('writeautomaticsub', False) or self.get_param('listsubtitles'))
4246 for lang_code, caption_track in captions.items():
4247 base_url = caption_track.get('baseUrl')
4248 orig_lang = parse_qs(base_url).get('lang', [None])[-1]
4249 if not base_url:
4250 continue
4251 lang_name = self._get_text(caption_track, 'name', max_runs=1)
4252 if caption_track.get('kind') != 'asr':
4253 if not lang_code:
4254 continue
4255 process_language(
4256 subtitles, base_url, lang_code, lang_name, {})
4257 if not caption_track.get('isTranslatable'):
4258 continue
4259 for trans_code, trans_name in translation_languages.items():
4260 if not trans_code:
4261 continue
4262 orig_trans_code = trans_code
4263 if caption_track.get('kind') != 'asr' and trans_code != 'und':
4264 if not get_translated_subs:
4265 continue
4266 trans_code += f'-{lang_code}'
4267 trans_name += format_field(lang_name, None, ' from %s')
4268 # Add an "-orig" label to the original language so that it can be distinguished.
4269 # The subs are returned without "-orig" as well for compatibility
4270 if lang_code == f'a-{orig_trans_code}':
4271 process_language(
4272 automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
4273 # Setting tlang=lang returns damaged subtitles.
4274 process_language(automatic_captions, base_url, trans_code, trans_name,
4275 {} if orig_lang == orig_trans_code else {'tlang': trans_code})
4276
4277 info['automatic_captions'] = automatic_captions
4278 info['subtitles'] = subtitles
4279
4280 parsed_url = urllib.parse.urlparse(url)
4281 for component in [parsed_url.fragment, parsed_url.query]:
4282 query = urllib.parse.parse_qs(component)
4283 for k, v in query.items():
4284 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
4285 d_k += '_time'
4286 if d_k not in info and k in s_ks:
4287 info[d_k] = parse_duration(query[k][0])
4288
4289 # Youtube Music Auto-generated description
4290 if video_description:
4291 mobj = re.search(
4292 r'''(?xs)
4293 (?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+
4294 (?P<album>[^\n]+)
4295 (?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
4296 (?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
4297 (.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?
4298 .+\nAuto-generated\ by\ YouTube\.\s*$
4299 ''', video_description)
4300 if mobj:
4301 release_year = mobj.group('release_year')
4302 release_date = mobj.group('release_date')
4303 if release_date:
4304 release_date = release_date.replace('-', '')
4305 if not release_year:
4306 release_year = release_date[:4]
4307 info.update({
4308 'album': mobj.group('album'.strip()),
4309 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
4310 'track': mobj.group('track').strip(),
4311 'release_date': release_date,
4312 'release_year': int_or_none(release_year),
4313 })
4314
4315 initial_data = None
4316 if webpage:
4317 initial_data = self.extract_yt_initial_data(video_id, webpage, fatal=False)
4318 if not traverse_obj(initial_data, 'contents'):
4319 self.report_warning('Incomplete data received in embedded initial data; re-fetching using API.')
4320 initial_data = None
4321 if not initial_data:
4322 query = {'videoId': video_id}
4323 query.update(self._get_checkok_params())
4324 initial_data = self._extract_response(
4325 item_id=video_id, ep='next', fatal=False,
4326 ytcfg=master_ytcfg, query=query, check_get_keys='contents',
4327 headers=self.generate_api_headers(ytcfg=master_ytcfg),
4328 note='Downloading initial data API JSON')
4329
4330 info['comment_count'] = traverse_obj(initial_data, (
4331 'contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents', ..., 'itemSectionRenderer',
4332 'contents', ..., 'commentsEntryPointHeaderRenderer', 'commentCount'
4333 ), (
4334 'engagementPanels', lambda _, v: v['engagementPanelSectionListRenderer']['panelIdentifier'] == 'comment-item-section',
4335 'engagementPanelSectionListRenderer', 'header', 'engagementPanelTitleHeaderRenderer', 'contextualInfo'
4336 ), expected_type=self._get_count, get_all=False)
4337
4338 try: # This will error if there is no livechat
4339 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
4340 except (KeyError, IndexError, TypeError):
4341 pass
4342 else:
4343 info.setdefault('subtitles', {})['live_chat'] = [{
4344 # url is needed to set cookies
4345 'url': f'https://www.youtube.com/watch?v={video_id}&bpctr=9999999999&has_verified=1',
4346 'video_id': video_id,
4347 'ext': 'json',
4348 'protocol': ('youtube_live_chat' if live_status in ('is_live', 'is_upcoming')
4349 else 'youtube_live_chat_replay'),
4350 }]
4351
4352 if initial_data:
4353 info['chapters'] = (
4354 self._extract_chapters_from_json(initial_data, duration)
4355 or self._extract_chapters_from_engagement_panel(initial_data, duration)
4356 or self._extract_chapters_from_description(video_description, duration)
4357 or None)
4358
4359 info['heatmap'] = self._extract_heatmap_from_player_overlay(initial_data)
4360
4361 contents = traverse_obj(
4362 initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'),
4363 expected_type=list, default=[])
4364
4365 vpir = get_first(contents, 'videoPrimaryInfoRenderer')
4366 if vpir:
4367 stl = vpir.get('superTitleLink')
4368 if stl:
4369 stl = self._get_text(stl)
4370 if try_get(
4371 vpir,
4372 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
4373 info['location'] = stl
4374 else:
4375 mobj = re.search(r'(.+?)\s*S(\d+)\s*•?\s*E(\d+)', stl)
4376 if mobj:
4377 info.update({
4378 'series': mobj.group(1),
4379 'season_number': int(mobj.group(2)),
4380 'episode_number': int(mobj.group(3)),
4381 })
4382 for tlb in (try_get(
4383 vpir,
4384 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
4385 list) or []):
4386 tbrs = variadic(
4387 traverse_obj(
4388 tlb, ('toggleButtonRenderer', ...),
4389 ('segmentedLikeDislikeButtonRenderer', ..., 'toggleButtonRenderer')))
4390 for tbr in tbrs:
4391 for getter, regex in [(
4392 lambda x: x['defaultText']['accessibility']['accessibilityData'],
4393 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
4394 lambda x: x['accessibility'],
4395 lambda x: x['accessibilityData']['accessibilityData'],
4396 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
4397 label = (try_get(tbr, getter, dict) or {}).get('label')
4398 if label:
4399 mobj = re.match(regex, label)
4400 if mobj:
4401 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
4402 break
4403 sbr_tooltip = try_get(
4404 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
4405 if sbr_tooltip:
4406 like_count, dislike_count = sbr_tooltip.split(' / ')
4407 info.update({
4408 'like_count': str_to_int(like_count),
4409 'dislike_count': str_to_int(dislike_count),
4410 })
4411 vcr = traverse_obj(vpir, ('viewCount', 'videoViewCountRenderer'))
4412 if vcr:
4413 vc = self._get_count(vcr, 'viewCount')
4414 # Upcoming premieres with waiting count are treated as live here
4415 if vcr.get('isLive'):
4416 info['concurrent_view_count'] = vc
4417 elif info.get('view_count') is None:
4418 info['view_count'] = vc
4419
4420 vsir = get_first(contents, 'videoSecondaryInfoRenderer')
4421 if vsir:
4422 vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer'))
4423 info.update({
4424 'channel': self._get_text(vor, 'title'),
4425 'channel_follower_count': self._get_count(vor, 'subscriberCountText')})
4426
4427 if not channel_handle:
4428 channel_handle = self.handle_from_url(
4429 traverse_obj(vor, (
4430 ('navigationEndpoint', ('title', 'runs', ..., 'navigationEndpoint')),
4431 (('commandMetadata', 'webCommandMetadata', 'url'), ('browseEndpoint', 'canonicalBaseUrl')),
4432 {str}), get_all=False))
4433
4434 rows = try_get(
4435 vsir,
4436 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
4437 list) or []
4438 multiple_songs = False
4439 for row in rows:
4440 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
4441 multiple_songs = True
4442 break
4443 for row in rows:
4444 mrr = row.get('metadataRowRenderer') or {}
4445 mrr_title = mrr.get('title')
4446 if not mrr_title:
4447 continue
4448 mrr_title = self._get_text(mrr, 'title')
4449 mrr_contents_text = self._get_text(mrr, ('contents', 0))
4450 if mrr_title == 'License':
4451 info['license'] = mrr_contents_text
4452 elif not multiple_songs:
4453 if mrr_title == 'Album':
4454 info['album'] = mrr_contents_text
4455 elif mrr_title == 'Artist':
4456 info['artist'] = mrr_contents_text
4457 elif mrr_title == 'Song':
4458 info['track'] = mrr_contents_text
4459
4460 info.update({
4461 'uploader': info.get('channel'),
4462 'uploader_id': channel_handle,
4463 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4464 })
4465 # The upload date for scheduled, live and past live streams / premieres in microformats
4466 # may be different from the stream date. Although not in UTC, we will prefer it in this case.
4467 # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139
4468 upload_date = (
4469 unified_strdate(get_first(microformats, 'uploadDate'))
4470 or unified_strdate(search_meta('uploadDate')))
4471 if not upload_date or (
4472 live_status in ('not_live', None)
4473 and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
4474 ):
4475 upload_date = strftime_or_none(
4476 self._parse_time_text(self._get_text(vpir, 'dateText')), '%Y%m%d') or upload_date
4477 info['upload_date'] = upload_date
4478
4479 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
4480 v = info.get(s_k)
4481 if v:
4482 info[d_k] = v
4483
4484 badges = self._extract_badges(traverse_obj(contents, (..., 'videoPrimaryInfoRenderer'), get_all=False))
4485
4486 is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
4487 or get_first(video_details, 'isPrivate', expected_type=bool))
4488
4489 info['availability'] = (
4490 'public' if self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
4491 else self._availability(
4492 is_private=is_private,
4493 needs_premium=(
4494 self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM)
4495 or False if initial_data and is_private is not None else None),
4496 needs_subscription=(
4497 self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION)
4498 or False if initial_data and is_private is not None else None),
4499 needs_auth=info['age_limit'] >= 18,
4500 is_unlisted=None if is_private is None else (
4501 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
4502 or get_first(microformats, 'isUnlisted', expected_type=bool))))
4503
4504 info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage)
4505
4506 self.mark_watched(video_id, player_responses)
4507
4508 return info
4509
4510
4511 class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
4512 @staticmethod
4513 def passthrough_smuggled_data(func):
4514 def _smuggle(info, smuggled_data):
4515 if info.get('_type') not in ('url', 'url_transparent'):
4516 return info
4517 if smuggled_data.get('is_music_url'):
4518 parsed_url = urllib.parse.urlparse(info['url'])
4519 if parsed_url.netloc in ('www.youtube.com', 'music.youtube.com'):
4520 smuggled_data.pop('is_music_url')
4521 info['url'] = urllib.parse.urlunparse(parsed_url._replace(netloc='music.youtube.com'))
4522 if smuggled_data:
4523 info['url'] = smuggle_url(info['url'], smuggled_data)
4524 return info
4525
4526 @functools.wraps(func)
4527 def wrapper(self, url):
4528 url, smuggled_data = unsmuggle_url(url, {})
4529 if self.is_music_url(url):
4530 smuggled_data['is_music_url'] = True
4531 info_dict = func(self, url, smuggled_data)
4532 if smuggled_data:
4533 _smuggle(info_dict, smuggled_data)
4534 if info_dict.get('entries'):
4535 info_dict['entries'] = (_smuggle(i, smuggled_data.copy()) for i in info_dict['entries'])
4536 return info_dict
4537 return wrapper
4538
4539 @staticmethod
4540 def _extract_basic_item_renderer(item):
4541 # Modified from _extract_grid_item_renderer
4542 known_basic_renderers = (
4543 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer'
4544 )
4545 for key, renderer in item.items():
4546 if not isinstance(renderer, dict):
4547 continue
4548 elif key in known_basic_renderers:
4549 return renderer
4550 elif key.startswith('grid') and key.endswith('Renderer'):
4551 return renderer
4552
4553 def _extract_channel_renderer(self, renderer):
4554 channel_id = self.ucid_or_none(renderer['channelId'])
4555 title = self._get_text(renderer, 'title')
4556 channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
4557 # As of 2023-03-01 YouTube doesn't use the channel handles on these renderers yet.
4558 # However we can expect them to change that in the future.
4559 channel_handle = self.handle_from_url(
4560 traverse_obj(renderer, (
4561 'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
4562 ('browseEndpoint', 'canonicalBaseUrl')),
4563 {str}), get_all=False))
4564 return {
4565 '_type': 'url',
4566 'url': channel_url,
4567 'id': channel_id,
4568 'ie_key': YoutubeTabIE.ie_key(),
4569 'channel': title,
4570 'uploader': title,
4571 'channel_id': channel_id,
4572 'channel_url': channel_url,
4573 'title': title,
4574 'uploader_id': channel_handle,
4575 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4576 'channel_follower_count': self._get_count(renderer, 'subscriberCountText'),
4577 'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
4578 'playlist_count': self._get_count(renderer, 'videoCountText'),
4579 'description': self._get_text(renderer, 'descriptionSnippet'),
4580 }
4581
4582 def _grid_entries(self, grid_renderer):
4583 for item in grid_renderer['items']:
4584 if not isinstance(item, dict):
4585 continue
4586 renderer = self._extract_basic_item_renderer(item)
4587 if not isinstance(renderer, dict):
4588 continue
4589 title = self._get_text(renderer, 'title')
4590
4591 # playlist
4592 playlist_id = renderer.get('playlistId')
4593 if playlist_id:
4594 yield self.url_result(
4595 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4596 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4597 video_title=title)
4598 continue
4599 # video
4600 video_id = renderer.get('videoId')
4601 if video_id:
4602 yield self._extract_video(renderer)
4603 continue
4604 # channel
4605 channel_id = renderer.get('channelId')
4606 if channel_id:
4607 yield self._extract_channel_renderer(renderer)
4608 continue
4609 # generic endpoint URL support
4610 ep_url = urljoin('https://www.youtube.com/', try_get(
4611 renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
4612 str))
4613 if ep_url:
4614 for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE):
4615 if ie.suitable(ep_url):
4616 yield self.url_result(
4617 ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title)
4618 break
4619
4620 def _music_reponsive_list_entry(self, renderer):
4621 video_id = traverse_obj(renderer, ('playlistItemData', 'videoId'))
4622 if video_id:
4623 title = traverse_obj(renderer, (
4624 'flexColumns', 0, 'musicResponsiveListItemFlexColumnRenderer',
4625 'text', 'runs', 0, 'text'))
4626 return self.url_result(f'https://music.youtube.com/watch?v={video_id}',
4627 ie=YoutubeIE.ie_key(), video_id=video_id, title=title)
4628 playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId'))
4629 if playlist_id:
4630 video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId'))
4631 if video_id:
4632 return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}',
4633 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4634 return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}',
4635 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4636 browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId'))
4637 if browse_id:
4638 return self.url_result(f'https://music.youtube.com/browse/{browse_id}',
4639 ie=YoutubeTabIE.ie_key(), video_id=browse_id)
4640
4641 def _shelf_entries_from_content(self, shelf_renderer):
4642 content = shelf_renderer.get('content')
4643 if not isinstance(content, dict):
4644 return
4645 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
4646 if renderer:
4647 # TODO: add support for nested playlists so each shelf is processed
4648 # as separate playlist
4649 # TODO: this includes only first N items
4650 yield from self._grid_entries(renderer)
4651 renderer = content.get('horizontalListRenderer')
4652 if renderer:
4653 # TODO
4654 pass
4655
4656 def _shelf_entries(self, shelf_renderer, skip_channels=False):
4657 ep = try_get(
4658 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
4659 str)
4660 shelf_url = urljoin('https://www.youtube.com', ep)
4661 if shelf_url:
4662 # Skipping links to another channels, note that checking for
4663 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
4664 # will not work
4665 if skip_channels and '/channels?' in shelf_url:
4666 return
4667 title = self._get_text(shelf_renderer, 'title')
4668 yield self.url_result(shelf_url, video_title=title)
4669 # Shelf may not contain shelf URL, fallback to extraction from content
4670 yield from self._shelf_entries_from_content(shelf_renderer)
4671
4672 def _playlist_entries(self, video_list_renderer):
4673 for content in video_list_renderer['contents']:
4674 if not isinstance(content, dict):
4675 continue
4676 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
4677 if not isinstance(renderer, dict):
4678 continue
4679 video_id = renderer.get('videoId')
4680 if not video_id:
4681 continue
4682 yield self._extract_video(renderer)
4683
4684 def _rich_entries(self, rich_grid_renderer):
4685 renderer = traverse_obj(
4686 rich_grid_renderer,
4687 ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer')), get_all=False) or {}
4688 video_id = renderer.get('videoId')
4689 if video_id:
4690 yield self._extract_video(renderer)
4691 return
4692 playlist_id = renderer.get('playlistId')
4693 if playlist_id:
4694 yield self.url_result(
4695 f'https://www.youtube.com/playlist?list={playlist_id}',
4696 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
4697 video_title=self._get_text(renderer, 'title'))
4698 return
4699
4700 def _video_entry(self, video_renderer):
4701 video_id = video_renderer.get('videoId')
4702 if video_id:
4703 return self._extract_video(video_renderer)
4704
4705 def _hashtag_tile_entry(self, hashtag_tile_renderer):
4706 url = urljoin('https://youtube.com', traverse_obj(
4707 hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url')))
4708 if url:
4709 return self.url_result(
4710 url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag'))
4711
4712 def _post_thread_entries(self, post_thread_renderer):
4713 post_renderer = try_get(
4714 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
4715 if not post_renderer:
4716 return
4717 # video attachment
4718 video_renderer = try_get(
4719 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {}
4720 video_id = video_renderer.get('videoId')
4721 if video_id:
4722 entry = self._extract_video(video_renderer)
4723 if entry:
4724 yield entry
4725 # playlist attachment
4726 playlist_id = try_get(
4727 post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], str)
4728 if playlist_id:
4729 yield self.url_result(
4730 'https://www.youtube.com/playlist?list=%s' % playlist_id,
4731 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
4732 # inline video links
4733 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
4734 for run in runs:
4735 if not isinstance(run, dict):
4736 continue
4737 ep_url = try_get(
4738 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], str)
4739 if not ep_url:
4740 continue
4741 if not YoutubeIE.suitable(ep_url):
4742 continue
4743 ep_video_id = YoutubeIE._match_id(ep_url)
4744 if video_id == ep_video_id:
4745 continue
4746 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id)
4747
4748 def _post_thread_continuation_entries(self, post_thread_continuation):
4749 contents = post_thread_continuation.get('contents')
4750 if not isinstance(contents, list):
4751 return
4752 for content in contents:
4753 renderer = content.get('backstagePostThreadRenderer')
4754 if isinstance(renderer, dict):
4755 yield from self._post_thread_entries(renderer)
4756 continue
4757 renderer = content.get('videoRenderer')
4758 if isinstance(renderer, dict):
4759 yield self._video_entry(renderer)
4760
4761 r''' # unused
4762 def _rich_grid_entries(self, contents):
4763 for content in contents:
4764 video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict)
4765 if video_renderer:
4766 entry = self._video_entry(video_renderer)
4767 if entry:
4768 yield entry
4769 '''
4770
4771 def _report_history_entries(self, renderer):
4772 for url in traverse_obj(renderer, (
4773 'rows', ..., 'reportHistoryTableRowRenderer', 'cells', ...,
4774 'reportHistoryTableCellRenderer', 'cell', 'reportHistoryTableTextCellRenderer', 'text', 'runs', ...,
4775 'navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')):
4776 yield self.url_result(urljoin('https://www.youtube.com', url), YoutubeIE)
4777
4778 def _extract_entries(self, parent_renderer, continuation_list):
4779 # continuation_list is modified in-place with continuation_list = [continuation_token]
4780 continuation_list[:] = [None]
4781 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
4782 for content in contents:
4783 if not isinstance(content, dict):
4784 continue
4785 is_renderer = traverse_obj(
4786 content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation',
4787 expected_type=dict)
4788 if not is_renderer:
4789 if content.get('richItemRenderer'):
4790 for entry in self._rich_entries(content['richItemRenderer']):
4791 yield entry
4792 continuation_list[0] = self._extract_continuation(parent_renderer)
4793 elif content.get('reportHistorySectionRenderer'): # https://www.youtube.com/reporthistory
4794 table = traverse_obj(content, ('reportHistorySectionRenderer', 'table', 'tableRenderer'))
4795 yield from self._report_history_entries(table)
4796 continuation_list[0] = self._extract_continuation(table)
4797 continue
4798
4799 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
4800 for isr_content in isr_contents:
4801 if not isinstance(isr_content, dict):
4802 continue
4803
4804 known_renderers = {
4805 'playlistVideoListRenderer': self._playlist_entries,
4806 'gridRenderer': self._grid_entries,
4807 'reelShelfRenderer': self._grid_entries,
4808 'shelfRenderer': self._shelf_entries,
4809 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)],
4810 'backstagePostThreadRenderer': self._post_thread_entries,
4811 'videoRenderer': lambda x: [self._video_entry(x)],
4812 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
4813 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
4814 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
4815 }
4816 for key, renderer in isr_content.items():
4817 if key not in known_renderers:
4818 continue
4819 for entry in known_renderers[key](renderer):
4820 if entry:
4821 yield entry
4822 continuation_list[0] = self._extract_continuation(renderer)
4823 break
4824
4825 if not continuation_list[0]:
4826 continuation_list[0] = self._extract_continuation(is_renderer)
4827
4828 if not continuation_list[0]:
4829 continuation_list[0] = self._extract_continuation(parent_renderer)
4830
4831 def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data):
4832 continuation_list = [None]
4833 extract_entries = lambda x: self._extract_entries(x, continuation_list)
4834 tab_content = try_get(tab, lambda x: x['content'], dict)
4835 if not tab_content:
4836 return
4837 parent_renderer = (
4838 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
4839 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
4840 yield from extract_entries(parent_renderer)
4841 continuation = continuation_list[0]
4842
4843 for page_num in itertools.count(1):
4844 if not continuation:
4845 break
4846 headers = self.generate_api_headers(
4847 ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data)
4848 response = self._extract_response(
4849 item_id=f'{item_id} page {page_num}',
4850 query=continuation, headers=headers, ytcfg=ytcfg,
4851 check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints'))
4852
4853 if not response:
4854 break
4855 # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases
4856 # See: https://github.com/ytdl-org/youtube-dl/issues/28702
4857 visitor_data = self._extract_visitor_data(response) or visitor_data
4858
4859 known_renderers = {
4860 'videoRenderer': (self._grid_entries, 'items'), # for membership tab
4861 'gridPlaylistRenderer': (self._grid_entries, 'items'),
4862 'gridVideoRenderer': (self._grid_entries, 'items'),
4863 'gridChannelRenderer': (self._grid_entries, 'items'),
4864 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
4865 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
4866 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
4867 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents'),
4868 'reportHistoryTableRowRenderer': (self._report_history_entries, 'rows'),
4869 'playlistVideoListContinuation': (self._playlist_entries, None),
4870 'gridContinuation': (self._grid_entries, None),
4871 'itemSectionContinuation': (self._post_thread_continuation_entries, None),
4872 'sectionListContinuation': (extract_entries, None), # for feeds
4873 }
4874
4875 continuation_items = traverse_obj(response, (
4876 ('onResponseReceivedActions', 'onResponseReceivedEndpoints'), ...,
4877 'appendContinuationItemsAction', 'continuationItems'
4878 ), 'continuationContents', get_all=False)
4879 continuation_item = traverse_obj(continuation_items, 0, None, expected_type=dict, default={})
4880
4881 video_items_renderer = None
4882 for key in continuation_item.keys():
4883 if key not in known_renderers:
4884 continue
4885 func, parent_key = known_renderers[key]
4886 video_items_renderer = {parent_key: continuation_items} if parent_key else continuation_items
4887 continuation_list = [None]
4888 yield from func(video_items_renderer)
4889 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
4890
4891 if not video_items_renderer:
4892 break
4893
4894 @staticmethod
4895 def _extract_selected_tab(tabs, fatal=True):
4896 for tab_renderer in tabs:
4897 if tab_renderer.get('selected'):
4898 return tab_renderer
4899 if fatal:
4900 raise ExtractorError('Unable to find selected tab')
4901
4902 @staticmethod
4903 def _extract_tab_renderers(response):
4904 return traverse_obj(
4905 response, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., ('tabRenderer', 'expandableTabRenderer')), expected_type=dict)
4906
4907 def _extract_from_tabs(self, item_id, ytcfg, data, tabs):
4908 metadata = self._extract_metadata_from_tabs(item_id, data)
4909
4910 selected_tab = self._extract_selected_tab(tabs)
4911 metadata['title'] += format_field(selected_tab, 'title', ' - %s')
4912 metadata['title'] += format_field(selected_tab, 'expandedText', ' - %s')
4913
4914 return self.playlist_result(
4915 self._entries(
4916 selected_tab, metadata['id'], ytcfg,
4917 self._extract_account_syncid(ytcfg, data),
4918 self._extract_visitor_data(data, ytcfg)),
4919 **metadata)
4920
4921 def _extract_metadata_from_tabs(self, item_id, data):
4922 info = {'id': item_id}
4923
4924 metadata_renderer = traverse_obj(data, ('metadata', 'channelMetadataRenderer'), expected_type=dict)
4925 if metadata_renderer:
4926 channel_id = traverse_obj(metadata_renderer, ('externalId', {self.ucid_or_none}),
4927 ('channelUrl', {self.ucid_from_url}))
4928 info.update({
4929 'channel': metadata_renderer.get('title'),
4930 'channel_id': channel_id,
4931 })
4932 if info['channel_id']:
4933 info['id'] = info['channel_id']
4934 else:
4935 metadata_renderer = traverse_obj(data, ('metadata', 'playlistMetadataRenderer'), expected_type=dict)
4936
4937 # We can get the uncropped banner/avatar by replacing the crop params with '=s0'
4938 # See: https://github.com/yt-dlp/yt-dlp/issues/2237#issuecomment-1013694714
4939 def _get_uncropped(url):
4940 return url_or_none((url or '').split('=')[0] + '=s0')
4941
4942 avatar_thumbnails = self._extract_thumbnails(metadata_renderer, 'avatar')
4943 if avatar_thumbnails:
4944 uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url'])
4945 if uncropped_avatar:
4946 avatar_thumbnails.append({
4947 'url': uncropped_avatar,
4948 'id': 'avatar_uncropped',
4949 'preference': 1
4950 })
4951
4952 channel_banners = self._extract_thumbnails(
4953 data, ('header', ..., ('banner', 'mobileBanner', 'tvBanner')))
4954 for banner in channel_banners:
4955 banner['preference'] = -10
4956
4957 if channel_banners:
4958 uncropped_banner = _get_uncropped(channel_banners[0]['url'])
4959 if uncropped_banner:
4960 channel_banners.append({
4961 'url': uncropped_banner,
4962 'id': 'banner_uncropped',
4963 'preference': -5
4964 })
4965
4966 # Deprecated - remove primary_sidebar_renderer when layout discontinued
4967 primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer')
4968 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer'), expected_type=dict)
4969
4970 primary_thumbnails = self._extract_thumbnails(
4971 primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail'))
4972 playlist_thumbnails = self._extract_thumbnails(
4973 playlist_header_renderer, ('playlistHeaderBanner', 'heroPlaylistThumbnailRenderer', 'thumbnail'))
4974
4975 info.update({
4976 'title': (traverse_obj(metadata_renderer, 'title')
4977 or self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag'))
4978 or info['id']),
4979 'availability': self._extract_availability(data),
4980 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')),
4981 'description': try_get(metadata_renderer, lambda x: x.get('description', '')),
4982 'tags': try_get(metadata_renderer or {}, lambda x: x.get('keywords', '').split()),
4983 'thumbnails': (primary_thumbnails or playlist_thumbnails) + avatar_thumbnails + channel_banners,
4984 })
4985
4986 channel_handle = (
4987 traverse_obj(metadata_renderer, (('vanityChannelUrl', ('ownerUrls', ...)), {self.handle_from_url}), get_all=False)
4988 or traverse_obj(data, ('header', ..., 'channelHandleText', {self.handle_or_none}), get_all=False))
4989
4990 if channel_handle:
4991 info.update({
4992 'uploader_id': channel_handle,
4993 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
4994 })
4995 # Playlist stats is a text runs array containing [video count, view count, last updated].
4996 # last updated or (view count and last updated) may be missing.
4997 playlist_stats = get_first(
4998 (primary_sidebar_renderer, playlist_header_renderer), (('stats', 'briefStats', 'numVideosText'), ))
4999
5000 last_updated_unix = self._parse_time_text(
5001 self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
5002 or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
5003 info['modified_date'] = strftime_or_none(last_updated_unix, '%Y%m%d')
5004
5005 info['view_count'] = self._get_count(playlist_stats, 1)
5006 if info['view_count'] is None: # 0 is allowed
5007 info['view_count'] = self._get_count(playlist_header_renderer, 'viewCountText')
5008 if info['view_count'] is None:
5009 info['view_count'] = self._get_count(data, (
5010 'contents', 'twoColumnBrowseResultsRenderer', 'tabs', ..., 'tabRenderer', 'content', 'sectionListRenderer',
5011 'contents', ..., 'itemSectionRenderer', 'contents', ..., 'channelAboutFullMetadataRenderer', 'viewCountText'))
5012
5013 info['playlist_count'] = self._get_count(playlist_stats, 0)
5014 if info['playlist_count'] is None: # 0 is allowed
5015 info['playlist_count'] = self._get_count(playlist_header_renderer, ('byline', 0, 'playlistBylineRenderer', 'text'))
5016
5017 if not info.get('channel_id'):
5018 owner = traverse_obj(playlist_header_renderer, 'ownerText')
5019 if not owner: # Deprecated
5020 owner = traverse_obj(
5021 self._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer'),
5022 ('videoOwner', 'videoOwnerRenderer', 'title'))
5023 owner_text = self._get_text(owner)
5024 browse_ep = traverse_obj(owner, ('runs', 0, 'navigationEndpoint', 'browseEndpoint')) or {}
5025 info.update({
5026 'channel': self._search_regex(r'^by (.+) and \d+ others?$', owner_text, 'uploader', default=owner_text),
5027 'channel_id': self.ucid_or_none(browse_ep.get('browseId')),
5028 'uploader_id': self.handle_from_url(urljoin('https://www.youtube.com', browse_ep.get('canonicalBaseUrl')))
5029 })
5030
5031 info.update({
5032 'uploader': info['channel'],
5033 'channel_url': format_field(info.get('channel_id'), None, 'https://www.youtube.com/channel/%s', default=None),
5034 'uploader_url': format_field(info.get('uploader_id'), None, 'https://www.youtube.com/%s', default=None),
5035 })
5036
5037 return info
5038
5039 def _extract_inline_playlist(self, playlist, playlist_id, data, ytcfg):
5040 first_id = last_id = response = None
5041 for page_num in itertools.count(1):
5042 videos = list(self._playlist_entries(playlist))
5043 if not videos:
5044 return
5045 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
5046 if start >= len(videos):
5047 return
5048 yield from videos[start:]
5049 first_id = first_id or videos[0]['id']
5050 last_id = videos[-1]['id']
5051 watch_endpoint = try_get(
5052 playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint'])
5053 headers = self.generate_api_headers(
5054 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5055 visitor_data=self._extract_visitor_data(response, data, ytcfg))
5056 query = {
5057 'playlistId': playlist_id,
5058 'videoId': watch_endpoint.get('videoId') or last_id,
5059 'index': watch_endpoint.get('index') or len(videos),
5060 'params': watch_endpoint.get('params') or 'OAE%3D'
5061 }
5062 response = self._extract_response(
5063 item_id='%s page %d' % (playlist_id, page_num),
5064 query=query, ep='next', headers=headers, ytcfg=ytcfg,
5065 check_get_keys='contents'
5066 )
5067 playlist = try_get(
5068 response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
5069
5070 def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg):
5071 title = playlist.get('title') or try_get(
5072 data, lambda x: x['titleText']['simpleText'], str)
5073 playlist_id = playlist.get('playlistId') or item_id
5074
5075 # Delegating everything except mix playlists to regular tab-based playlist URL
5076 playlist_url = urljoin(url, try_get(
5077 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
5078 str))
5079
5080 # Some playlists are unviewable but YouTube still provides a link to the (broken) playlist page [1]
5081 # [1] MLCT, RLTDwFCb4jeqaKWnciAYM-ZVHg
5082 is_known_unviewable = re.fullmatch(r'MLCT|RLTD[\w-]{22}', playlist_id)
5083
5084 if playlist_url and playlist_url != url and not is_known_unviewable:
5085 return self.url_result(
5086 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
5087 video_title=title)
5088
5089 return self.playlist_result(
5090 self._extract_inline_playlist(playlist, playlist_id, data, ytcfg),
5091 playlist_id=playlist_id, playlist_title=title)
5092
5093 def _extract_availability(self, data):
5094 """
5095 Gets the availability of a given playlist/tab.
5096 Note: Unless YouTube tells us explicitly, we do not assume it is public
5097 @param data: response
5098 """
5099 sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {}
5100 playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
5101 player_header_privacy = playlist_header_renderer.get('privacy')
5102
5103 badges = self._extract_badges(sidebar_renderer)
5104
5105 # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
5106 privacy_setting_icon = get_first(
5107 (playlist_header_renderer, sidebar_renderer),
5108 ('privacyForm', 'dropdownFormFieldRenderer', 'dropdown', 'dropdownRenderer', 'entries',
5109 lambda _, v: v['privacyDropdownItemRenderer']['isSelected'], 'privacyDropdownItemRenderer', 'icon', 'iconType'),
5110 expected_type=str)
5111
5112 microformats_is_unlisted = traverse_obj(
5113 data, ('microformat', 'microformatDataRenderer', 'unlisted'), expected_type=bool)
5114
5115 return (
5116 'public' if (
5117 self._has_badge(badges, BadgeType.AVAILABILITY_PUBLIC)
5118 or player_header_privacy == 'PUBLIC'
5119 or privacy_setting_icon == 'PRIVACY_PUBLIC')
5120 else self._availability(
5121 is_private=(
5122 self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
5123 or player_header_privacy == 'PRIVATE' if player_header_privacy is not None
5124 else privacy_setting_icon == 'PRIVACY_PRIVATE' if privacy_setting_icon is not None else None),
5125 is_unlisted=(
5126 self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED)
5127 or player_header_privacy == 'UNLISTED' if player_header_privacy is not None
5128 else privacy_setting_icon == 'PRIVACY_UNLISTED' if privacy_setting_icon is not None
5129 else microformats_is_unlisted if microformats_is_unlisted is not None else None),
5130 needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
5131 needs_premium=self._has_badge(badges, BadgeType.AVAILABILITY_PREMIUM) or None,
5132 needs_auth=False))
5133
5134 @staticmethod
5135 def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict):
5136 sidebar_renderer = try_get(
5137 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or []
5138 for item in sidebar_renderer:
5139 renderer = try_get(item, lambda x: x[info_renderer], expected_type)
5140 if renderer:
5141 return renderer
5142
5143 def _reload_with_unavailable_videos(self, item_id, data, ytcfg):
5144 """
5145 Reload playlists with unavailable videos (e.g. private videos, region blocked, etc.)
5146 """
5147 is_playlist = bool(traverse_obj(
5148 data, ('metadata', 'playlistMetadataRenderer'), ('header', 'playlistHeaderRenderer')))
5149 if not is_playlist:
5150 return
5151 headers = self.generate_api_headers(
5152 ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data),
5153 visitor_data=self._extract_visitor_data(data, ytcfg))
5154 query = {
5155 'params': 'wgYCCAA=',
5156 'browseId': f'VL{item_id}'
5157 }
5158 return self._extract_response(
5159 item_id=item_id, headers=headers, query=query,
5160 check_get_keys='contents', fatal=False, ytcfg=ytcfg,
5161 note='Redownloading playlist API JSON with unavailable videos')
5162
5163 @functools.cached_property
5164 def skip_webpage(self):
5165 return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key())
5166
5167 def _extract_webpage(self, url, item_id, fatal=True):
5168 webpage, data = None, None
5169 for retry in self.RetryManager(fatal=fatal):
5170 try:
5171 webpage = self._download_webpage(url, item_id, note='Downloading webpage')
5172 data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
5173 except ExtractorError as e:
5174 if isinstance(e.cause, network_exceptions):
5175 if not isinstance(e.cause, urllib.error.HTTPError) or e.cause.code not in (403, 429):
5176 retry.error = e
5177 continue
5178 self._error_or_warning(e, fatal=fatal)
5179 break
5180
5181 try:
5182 self._extract_and_report_alerts(data)
5183 except ExtractorError as e:
5184 self._error_or_warning(e, fatal=fatal)
5185 break
5186
5187 # Sometimes youtube returns a webpage with incomplete ytInitialData
5188 # See: https://github.com/yt-dlp/yt-dlp/issues/116
5189 if not traverse_obj(data, 'contents', 'currentVideoEndpoint', 'onResponseReceivedActions'):
5190 retry.error = ExtractorError('Incomplete yt initial data received')
5191 continue
5192
5193 return webpage, data
5194
5195 def _report_playlist_authcheck(self, ytcfg, fatal=True):
5196 """Use if failed to extract ytcfg (and data) from initial webpage"""
5197 if not ytcfg and self.is_authenticated:
5198 msg = 'Playlists that require authentication may not extract correctly without a successful webpage download'
5199 if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal:
5200 raise ExtractorError(
5201 f'{msg}. If you are not downloading private content, or '
5202 'your cookies are only for the first account and channel,'
5203 ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check',
5204 expected=True)
5205 self.report_warning(msg, only_once=True)
5206
5207 def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'):
5208 data = None
5209 if not self.skip_webpage:
5210 webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal)
5211 ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage)
5212 # Reject webpage data if redirected to home page without explicitly requesting
5213 selected_tab = self._extract_selected_tab(self._extract_tab_renderers(data), fatal=False) or {}
5214 if (url != 'https://www.youtube.com/feed/recommended'
5215 and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page
5216 and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])):
5217 msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page'
5218 if fatal:
5219 raise ExtractorError(msg, expected=True)
5220 self.report_warning(msg, only_once=True)
5221 if not data:
5222 self._report_playlist_authcheck(ytcfg, fatal=fatal)
5223 data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client)
5224 return data, ytcfg
5225
5226 def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'):
5227 headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client)
5228 resolve_response = self._extract_response(
5229 item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal,
5230 ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client)
5231 endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'}
5232 for ep_key, ep in endpoints.items():
5233 params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict)
5234 if params:
5235 return self._extract_response(
5236 item_id=item_id, query=params, ep=ep, headers=headers,
5237 ytcfg=ytcfg, fatal=fatal, default_client=default_client,
5238 check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions'))
5239 err_note = 'Failed to resolve url (does the playlist exist?)'
5240 if fatal:
5241 raise ExtractorError(err_note, expected=True)
5242 self.report_warning(err_note, item_id)
5243
5244 _SEARCH_PARAMS = None
5245
5246 def _search_results(self, query, params=NO_DEFAULT, default_client='web'):
5247 data = {'query': query}
5248 if params is NO_DEFAULT:
5249 params = self._SEARCH_PARAMS
5250 if params:
5251 data['params'] = params
5252
5253 content_keys = (
5254 ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'),
5255 ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'),
5256 # ytmusic search
5257 ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'),
5258 ('continuationContents', ),
5259 )
5260 display_id = f'query "{query}"'
5261 check_get_keys = tuple({keys[0] for keys in content_keys})
5262 ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {}
5263 self._report_playlist_authcheck(ytcfg, fatal=False)
5264
5265 continuation_list = [None]
5266 search = None
5267 for page_num in itertools.count(1):
5268 data.update(continuation_list[0] or {})
5269 headers = self.generate_api_headers(
5270 ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client)
5271 search = self._extract_response(
5272 item_id=f'{display_id} page {page_num}', ep='search', query=data,
5273 default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers)
5274 slr_contents = traverse_obj(search, *content_keys)
5275 yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list)
5276 if not continuation_list[0]:
5277 break
5278
5279
5280 class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
5281 IE_DESC = 'YouTube Tabs'
5282 _VALID_URL = r'''(?x:
5283 https?://
5284 (?!consent\.)(?:\w+\.)?
5285 (?:
5286 youtube(?:kids)?\.com|
5287 %(invidious)s
5288 )/
5289 (?:
5290 (?P<channel_type>channel|c|user|browse)/|
5291 (?P<not_channel>
5292 feed/|hashtag/|
5293 (?:playlist|watch)\?.*?\blist=
5294 )|
5295 (?!(?:%(reserved_names)s)\b) # Direct URLs
5296 )
5297 (?P<id>[^/?\#&]+)
5298 )''' % {
5299 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
5300 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
5301 }
5302 IE_NAME = 'youtube:tab'
5303
5304 _TESTS = [{
5305 'note': 'playlists, multipage',
5306 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
5307 'playlist_mincount': 94,
5308 'info_dict': {
5309 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5310 'title': 'Igor Kleiner - Playlists',
5311 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5312 'uploader': 'Igor Kleiner',
5313 'uploader_id': '@IgorDataScience',
5314 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5315 'channel': 'Igor Kleiner',
5316 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5317 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5318 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5319 'channel_follower_count': int
5320 },
5321 }, {
5322 'note': 'playlists, multipage, different order',
5323 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
5324 'playlist_mincount': 94,
5325 'info_dict': {
5326 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
5327 'title': 'Igor Kleiner - Playlists',
5328 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
5329 'uploader': 'Igor Kleiner',
5330 'uploader_id': '@IgorDataScience',
5331 'uploader_url': 'https://www.youtube.com/@IgorDataScience',
5332 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'],
5333 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg',
5334 'channel': 'Igor Kleiner',
5335 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg',
5336 'channel_follower_count': int
5337 },
5338 }, {
5339 'note': 'playlists, series',
5340 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3',
5341 'playlist_mincount': 5,
5342 'info_dict': {
5343 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5344 'title': '3Blue1Brown - Playlists',
5345 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5346 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5347 'channel': '3Blue1Brown',
5348 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5349 'uploader_id': '@3blue1brown',
5350 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5351 'uploader': '3Blue1Brown',
5352 'tags': ['Mathematics'],
5353 'channel_follower_count': int
5354 },
5355 }, {
5356 'note': 'playlists, singlepage',
5357 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
5358 'playlist_mincount': 4,
5359 'info_dict': {
5360 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5361 'title': 'ThirstForScience - Playlists',
5362 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
5363 'uploader': 'ThirstForScience',
5364 'uploader_url': 'https://www.youtube.com/@ThirstForScience',
5365 'uploader_id': '@ThirstForScience',
5366 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
5367 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ',
5368 'tags': 'count:13',
5369 'channel': 'ThirstForScience',
5370 'channel_follower_count': int
5371 }
5372 }, {
5373 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
5374 'only_matching': True,
5375 }, {
5376 'note': 'basic, single video playlist',
5377 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5378 'info_dict': {
5379 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5380 'title': 'youtube-dl public playlist',
5381 'description': '',
5382 'tags': [],
5383 'view_count': int,
5384 'modified_date': '20201130',
5385 'channel': 'Sergey M.',
5386 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5387 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5388 'availability': 'public',
5389 'uploader': 'Sergey M.',
5390 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5391 'uploader_id': '@sergeym.6173',
5392 },
5393 'playlist_count': 1,
5394 }, {
5395 'note': 'empty playlist',
5396 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5397 'info_dict': {
5398 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
5399 'title': 'youtube-dl empty playlist',
5400 'tags': [],
5401 'channel': 'Sergey M.',
5402 'description': '',
5403 'modified_date': '20160902',
5404 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
5405 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5406 'availability': 'public',
5407 'uploader_url': 'https://www.youtube.com/@sergeym.6173',
5408 'uploader_id': '@sergeym.6173',
5409 'uploader': 'Sergey M.',
5410 },
5411 'playlist_count': 0,
5412 }, {
5413 'note': 'Home tab',
5414 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
5415 'info_dict': {
5416 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5417 'title': 'lex will - Home',
5418 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5419 'uploader': 'lex will',
5420 'uploader_id': '@lexwill718',
5421 'channel': 'lex will',
5422 'tags': ['bible', 'history', 'prophesy'],
5423 'uploader_url': 'https://www.youtube.com/@lexwill718',
5424 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5425 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5426 'channel_follower_count': int
5427 },
5428 'playlist_mincount': 2,
5429 }, {
5430 'note': 'Videos tab',
5431 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
5432 'info_dict': {
5433 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5434 'title': 'lex will - Videos',
5435 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5436 'uploader': 'lex will',
5437 'uploader_id': '@lexwill718',
5438 'tags': ['bible', 'history', 'prophesy'],
5439 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5440 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5441 'uploader_url': 'https://www.youtube.com/@lexwill718',
5442 'channel': 'lex will',
5443 'channel_follower_count': int
5444 },
5445 'playlist_mincount': 975,
5446 }, {
5447 'note': 'Videos tab, sorted by popular',
5448 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
5449 'info_dict': {
5450 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5451 'title': 'lex will - Videos',
5452 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5453 'uploader': 'lex will',
5454 'uploader_id': '@lexwill718',
5455 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5456 'uploader_url': 'https://www.youtube.com/@lexwill718',
5457 'channel': 'lex will',
5458 'tags': ['bible', 'history', 'prophesy'],
5459 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5460 'channel_follower_count': int
5461 },
5462 'playlist_mincount': 199,
5463 }, {
5464 'note': 'Playlists tab',
5465 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
5466 'info_dict': {
5467 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5468 'title': 'lex will - Playlists',
5469 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5470 'uploader': 'lex will',
5471 'uploader_id': '@lexwill718',
5472 'uploader_url': 'https://www.youtube.com/@lexwill718',
5473 'channel': 'lex will',
5474 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5475 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5476 'tags': ['bible', 'history', 'prophesy'],
5477 'channel_follower_count': int
5478 },
5479 'playlist_mincount': 17,
5480 }, {
5481 'note': 'Community tab',
5482 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
5483 'info_dict': {
5484 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5485 'title': 'lex will - Community',
5486 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5487 'channel': 'lex will',
5488 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5489 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5490 'tags': ['bible', 'history', 'prophesy'],
5491 'channel_follower_count': int,
5492 'uploader_url': 'https://www.youtube.com/@lexwill718',
5493 'uploader_id': '@lexwill718',
5494 'uploader': 'lex will',
5495 },
5496 'playlist_mincount': 18,
5497 }, {
5498 'note': 'Channels tab',
5499 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
5500 'info_dict': {
5501 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5502 'title': 'lex will - Channels',
5503 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
5504 'channel': 'lex will',
5505 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
5506 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
5507 'tags': ['bible', 'history', 'prophesy'],
5508 'channel_follower_count': int,
5509 'uploader_url': 'https://www.youtube.com/@lexwill718',
5510 'uploader_id': '@lexwill718',
5511 'uploader': 'lex will',
5512 },
5513 'playlist_mincount': 12,
5514 }, {
5515 'note': 'Search tab',
5516 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra',
5517 'playlist_mincount': 40,
5518 'info_dict': {
5519 'id': 'UCYO_jab_esuFRV4b17AJtAw',
5520 'title': '3Blue1Brown - Search - linear algebra',
5521 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
5522 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
5523 'tags': ['Mathematics'],
5524 'channel': '3Blue1Brown',
5525 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
5526 'channel_follower_count': int,
5527 'uploader_url': 'https://www.youtube.com/@3blue1brown',
5528 'uploader_id': '@3blue1brown',
5529 'uploader': '3Blue1Brown',
5530 },
5531 }, {
5532 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5533 'only_matching': True,
5534 }, {
5535 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5536 'only_matching': True,
5537 }, {
5538 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
5539 'only_matching': True,
5540 }, {
5541 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
5542 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5543 'info_dict': {
5544 'title': '29C3: Not my department',
5545 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
5546 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
5547 'tags': [],
5548 'view_count': int,
5549 'modified_date': '20150605',
5550 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
5551 'channel_url': 'https://www.youtube.com/channel/UCEPzS1rYsrkqzSLNp76nrcg',
5552 'channel': 'Christiaan008',
5553 'availability': 'public',
5554 'uploader_id': '@ChRiStIaAn008',
5555 'uploader': 'Christiaan008',
5556 'uploader_url': 'https://www.youtube.com/@ChRiStIaAn008',
5557 },
5558 'playlist_count': 96,
5559 }, {
5560 'note': 'Large playlist',
5561 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
5562 'info_dict': {
5563 'title': 'Uploads from Cauchemar',
5564 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
5565 'channel_url': 'https://www.youtube.com/channel/UCBABnxM4Ar9ten8Mdjj1j0Q',
5566 'tags': [],
5567 'modified_date': r're:\d{8}',
5568 'channel': 'Cauchemar',
5569 'view_count': int,
5570 'description': '',
5571 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
5572 'availability': 'public',
5573 'uploader_id': '@Cauchemar89',
5574 'uploader': 'Cauchemar',
5575 'uploader_url': 'https://www.youtube.com/@Cauchemar89',
5576 },
5577 'playlist_mincount': 1123,
5578 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5579 }, {
5580 'note': 'even larger playlist, 8832 videos',
5581 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
5582 'only_matching': True,
5583 }, {
5584 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
5585 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
5586 'info_dict': {
5587 'title': 'Uploads from Interstellar Movie',
5588 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
5589 'tags': [],
5590 'view_count': int,
5591 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
5592 'channel_url': 'https://www.youtube.com/channel/UCXw-G3eDE9trcvY2sBMM_aA',
5593 'channel': 'Interstellar Movie',
5594 'description': '',
5595 'modified_date': r're:\d{8}',
5596 'availability': 'public',
5597 'uploader_id': '@InterstellarMovie',
5598 'uploader': 'Interstellar Movie',
5599 'uploader_url': 'https://www.youtube.com/@InterstellarMovie',
5600 },
5601 'playlist_mincount': 21,
5602 }, {
5603 'note': 'Playlist with "show unavailable videos" button',
5604 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q',
5605 'info_dict': {
5606 'title': 'Uploads from Phim Siêu Nhân Nhật Bản',
5607 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q',
5608 'view_count': int,
5609 'channel': 'Phim Siêu Nhân Nhật Bản',
5610 'tags': [],
5611 'description': '',
5612 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q',
5613 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q',
5614 'modified_date': r're:\d{8}',
5615 'availability': 'public',
5616 'uploader_url': 'https://www.youtube.com/@phimsieunhannhatban',
5617 'uploader_id': '@phimsieunhannhatban',
5618 'uploader': 'Phim Siêu Nhân Nhật Bản',
5619 },
5620 'playlist_mincount': 200,
5621 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5622 }, {
5623 'note': 'Playlist with unavailable videos in page 7',
5624 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w',
5625 'info_dict': {
5626 'title': 'Uploads from BlankTV',
5627 'id': 'UU8l9frL61Yl5KFOl87nIm2w',
5628 'channel': 'BlankTV',
5629 'channel_url': 'https://www.youtube.com/channel/UC8l9frL61Yl5KFOl87nIm2w',
5630 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w',
5631 'view_count': int,
5632 'tags': [],
5633 'modified_date': r're:\d{8}',
5634 'description': '',
5635 'availability': 'public',
5636 'uploader_id': '@blanktv',
5637 'uploader': 'BlankTV',
5638 'uploader_url': 'https://www.youtube.com/@blanktv',
5639 },
5640 'playlist_mincount': 1000,
5641 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
5642 }, {
5643 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844',
5644 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5645 'info_dict': {
5646 'title': 'Data Analysis with Dr Mike Pound',
5647 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
5648 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
5649 'tags': [],
5650 'view_count': int,
5651 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA',
5652 'channel_url': 'https://www.youtube.com/channel/UC9-y-6csu5WGm29I7JiwpnA',
5653 'channel': 'Computerphile',
5654 'availability': 'public',
5655 'modified_date': '20190712',
5656 'uploader_id': '@Computerphile',
5657 'uploader': 'Computerphile',
5658 'uploader_url': 'https://www.youtube.com/@Computerphile',
5659 },
5660 'playlist_mincount': 11,
5661 }, {
5662 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
5663 'only_matching': True,
5664 }, {
5665 'note': 'Playlist URL that does not actually serve a playlist',
5666 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
5667 'info_dict': {
5668 'id': 'FqZTN594JQw',
5669 'ext': 'webm',
5670 'title': "Smiley's People 01 detective, Adventure Series, Action",
5671 'upload_date': '20150526',
5672 'license': 'Standard YouTube License',
5673 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
5674 'categories': ['People & Blogs'],
5675 'tags': list,
5676 'view_count': int,
5677 'like_count': int,
5678 },
5679 'params': {
5680 'skip_download': True,
5681 },
5682 'skip': 'This video is not available.',
5683 'add_ie': [YoutubeIE.ie_key()],
5684 }, {
5685 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
5686 'only_matching': True,
5687 }, {
5688 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
5689 'only_matching': True,
5690 }, {
5691 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
5692 'info_dict': {
5693 'id': 'AlTsmyW4auo', # This will keep changing
5694 'ext': 'mp4',
5695 'title': str,
5696 'upload_date': r're:\d{8}',
5697 'description': str,
5698 'categories': ['News & Politics'],
5699 'tags': list,
5700 'like_count': int,
5701 'release_timestamp': int,
5702 'channel': 'Sky News',
5703 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ',
5704 'age_limit': 0,
5705 'view_count': int,
5706 'thumbnail': r're:https?://i\.ytimg\.com/vi/[^/]+/maxresdefault(?:_live)?\.jpg',
5707 'playable_in_embed': True,
5708 'release_date': r're:\d+',
5709 'availability': 'public',
5710 'live_status': 'is_live',
5711 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ',
5712 'channel_follower_count': int,
5713 'concurrent_view_count': int,
5714 'uploader_url': 'https://www.youtube.com/@SkyNews',
5715 'uploader_id': '@SkyNews',
5716 'uploader': 'Sky News',
5717 },
5718 'params': {
5719 'skip_download': True,
5720 },
5721 'expected_warnings': ['Ignoring subtitle tracks found in '],
5722 }, {
5723 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
5724 'info_dict': {
5725 'id': 'a48o2S1cPoo',
5726 'ext': 'mp4',
5727 'title': 'The Young Turks - Live Main Show',
5728 'upload_date': '20150715',
5729 'license': 'Standard YouTube License',
5730 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
5731 'categories': ['News & Politics'],
5732 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
5733 'like_count': int,
5734 },
5735 'params': {
5736 'skip_download': True,
5737 },
5738 'only_matching': True,
5739 }, {
5740 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
5741 'only_matching': True,
5742 }, {
5743 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
5744 'only_matching': True,
5745 }, {
5746 'note': 'A channel that is not live. Should raise error',
5747 'url': 'https://www.youtube.com/user/numberphile/live',
5748 'only_matching': True,
5749 }, {
5750 'url': 'https://www.youtube.com/feed/trending',
5751 'only_matching': True,
5752 }, {
5753 'url': 'https://www.youtube.com/feed/library',
5754 'only_matching': True,
5755 }, {
5756 'url': 'https://www.youtube.com/feed/history',
5757 'only_matching': True,
5758 }, {
5759 'url': 'https://www.youtube.com/feed/subscriptions',
5760 'only_matching': True,
5761 }, {
5762 'url': 'https://www.youtube.com/feed/watch_later',
5763 'only_matching': True,
5764 }, {
5765 'note': 'Recommended - redirects to home page.',
5766 'url': 'https://www.youtube.com/feed/recommended',
5767 'only_matching': True,
5768 }, {
5769 'note': 'inline playlist with not always working continuations',
5770 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
5771 'only_matching': True,
5772 }, {
5773 'url': 'https://www.youtube.com/course',
5774 'only_matching': True,
5775 }, {
5776 'url': 'https://www.youtube.com/zsecurity',
5777 'only_matching': True,
5778 }, {
5779 'url': 'http://www.youtube.com/NASAgovVideo/videos',
5780 'only_matching': True,
5781 }, {
5782 'url': 'https://www.youtube.com/TheYoungTurks/live',
5783 'only_matching': True,
5784 }, {
5785 'url': 'https://www.youtube.com/hashtag/cctv9',
5786 'info_dict': {
5787 'id': 'cctv9',
5788 'title': '#cctv9',
5789 'tags': [],
5790 },
5791 'playlist_mincount': 300, # not consistent but should be over 300
5792 }, {
5793 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU',
5794 'only_matching': True,
5795 }, {
5796 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist',
5797 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5798 'only_matching': True
5799 }, {
5800 'note': '/browse/ should redirect to /channel/',
5801 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng',
5802 'only_matching': True
5803 }, {
5804 'note': 'VLPL, should redirect to playlist?list=PL...',
5805 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5806 'info_dict': {
5807 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq',
5808 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!',
5809 'title': 'NCS : All Releases 💿',
5810 'channel_url': 'https://www.youtube.com/channel/UC_aEa8K-EOJ3D6gOs7HcyNg',
5811 'modified_date': r're:\d{8}',
5812 'view_count': int,
5813 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg',
5814 'tags': [],
5815 'channel': 'NoCopyrightSounds',
5816 'availability': 'public',
5817 'uploader_url': 'https://www.youtube.com/@NoCopyrightSounds',
5818 'uploader': 'NoCopyrightSounds',
5819 'uploader_id': '@NoCopyrightSounds',
5820 },
5821 'playlist_mincount': 166,
5822 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden', 'YouTube Music is not directly supported'],
5823 }, {
5824 # TODO: fix 'unviewable' issue with this playlist when reloading with unavailable videos
5825 'note': 'Topic, should redirect to playlist?list=UU...',
5826 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5827 'info_dict': {
5828 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5829 'title': 'Uploads from Royalty Free Music - Topic',
5830 'tags': [],
5831 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5832 'channel': 'Royalty Free Music - Topic',
5833 'view_count': int,
5834 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5835 'modified_date': r're:\d{8}',
5836 'description': '',
5837 'availability': 'public',
5838 'uploader': 'Royalty Free Music - Topic',
5839 },
5840 'playlist_mincount': 101,
5841 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5842 }, {
5843 # Destination channel with only a hidden self tab (tab id is UCtFRv9O2AHqOZjjynzrv-xg)
5844 # Treat as a general feed
5845 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg',
5846 'info_dict': {
5847 'id': 'UCtFRv9O2AHqOZjjynzrv-xg',
5848 'title': 'UCtFRv9O2AHqOZjjynzrv-xg',
5849 'tags': [],
5850 },
5851 'playlist_mincount': 9,
5852 }, {
5853 'note': 'Youtube music Album',
5854 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE',
5855 'info_dict': {
5856 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0',
5857 'title': 'Album - Royalty Free Music Library V2 (50 Songs)',
5858 'tags': [],
5859 'view_count': int,
5860 'description': '',
5861 'availability': 'unlisted',
5862 'modified_date': r're:\d{8}',
5863 },
5864 'playlist_count': 50,
5865 'expected_warnings': ['YouTube Music is not directly supported'],
5866 }, {
5867 'note': 'unlisted single video playlist',
5868 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5869 'info_dict': {
5870 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf',
5871 'title': 'yt-dlp unlisted playlist test',
5872 'availability': 'unlisted',
5873 'tags': [],
5874 'modified_date': '20220418',
5875 'channel': 'colethedj',
5876 'view_count': int,
5877 'description': '',
5878 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q',
5879 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q',
5880 'uploader_url': 'https://www.youtube.com/@colethedj1894',
5881 'uploader_id': '@colethedj1894',
5882 'uploader': 'colethedj',
5883 },
5884 'playlist': [{
5885 'info_dict': {
5886 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
5887 'id': 'BaW_jenozKc',
5888 '_type': 'url',
5889 'ie_key': 'Youtube',
5890 'duration': 10,
5891 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
5892 'channel_url': 'https://www.youtube.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
5893 'view_count': int,
5894 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc',
5895 'channel': 'Philipp Hagemeister',
5896 'uploader_id': '@PhilippHagemeister',
5897 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
5898 'uploader': 'Philipp Hagemeister',
5899 }
5900 }],
5901 'playlist_count': 1,
5902 'params': {'extract_flat': True},
5903 }, {
5904 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData',
5905 'url': 'https://www.youtube.com/feed/recommended',
5906 'info_dict': {
5907 'id': 'recommended',
5908 'title': 'recommended',
5909 'tags': [],
5910 },
5911 'playlist_mincount': 50,
5912 'params': {
5913 'skip_download': True,
5914 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5915 },
5916 }, {
5917 'note': 'API Fallback: /videos tab, sorted by oldest first',
5918 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid',
5919 'info_dict': {
5920 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5921 'title': 'Cody\'sLab - Videos',
5922 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa',
5923 'channel': 'Cody\'sLab',
5924 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw',
5925 'tags': [],
5926 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw',
5927 'channel_follower_count': int
5928 },
5929 'playlist_mincount': 650,
5930 'params': {
5931 'skip_download': True,
5932 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5933 },
5934 'skip': 'Query for sorting no longer works',
5935 }, {
5936 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...',
5937 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw',
5938 'info_dict': {
5939 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw',
5940 'title': 'Uploads from Royalty Free Music - Topic',
5941 'modified_date': r're:\d{8}',
5942 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw',
5943 'description': '',
5944 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw',
5945 'tags': [],
5946 'channel': 'Royalty Free Music - Topic',
5947 'view_count': int,
5948 'availability': 'public',
5949 'uploader': 'Royalty Free Music - Topic',
5950 },
5951 'playlist_mincount': 101,
5952 'params': {
5953 'skip_download': True,
5954 'extractor_args': {'youtubetab': {'skip': ['webpage']}}
5955 },
5956 'expected_warnings': ['YouTube Music is not directly supported', r'[Uu]navailable videos (are|will be) hidden'],
5957 }, {
5958 'note': 'non-standard redirect to regional channel',
5959 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ',
5960 'only_matching': True
5961 }, {
5962 'note': 'collaborative playlist (uploader name in the form "by <uploader> and x other(s)")',
5963 'url': 'https://www.youtube.com/playlist?list=PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5964 'info_dict': {
5965 'id': 'PLx-_-Kk4c89oOHEDQAojOXzEzemXxoqx6',
5966 'modified_date': '20220407',
5967 'channel_url': 'https://www.youtube.com/channel/UCKcqXmCcyqnhgpA5P0oHH_Q',
5968 'tags': [],
5969 'availability': 'unlisted',
5970 'channel_id': 'UCKcqXmCcyqnhgpA5P0oHH_Q',
5971 'channel': 'pukkandan',
5972 'description': 'Test for collaborative playlist',
5973 'title': 'yt-dlp test - collaborative playlist',
5974 'view_count': int,
5975 'uploader_url': 'https://www.youtube.com/@pukkandan',
5976 'uploader_id': '@pukkandan',
5977 'uploader': 'pukkandan',
5978 },
5979 'playlist_mincount': 2
5980 }, {
5981 'note': 'translated tab name',
5982 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/playlists',
5983 'info_dict': {
5984 'id': 'UCiu-3thuViMebBjw_5nWYrA',
5985 'tags': [],
5986 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
5987 'description': 'test description',
5988 'title': 'cole-dlp-test-acc - 再生リスト',
5989 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
5990 'channel': 'cole-dlp-test-acc',
5991 'uploader_url': 'https://www.youtube.com/@coletdjnz',
5992 'uploader_id': '@coletdjnz',
5993 'uploader': 'cole-dlp-test-acc',
5994 },
5995 'playlist_mincount': 1,
5996 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
5997 'expected_warnings': ['Preferring "ja"'],
5998 }, {
5999 # XXX: this should really check flat playlist entries, but the test suite doesn't support that
6000 'note': 'preferred lang set with playlist with translated video titles',
6001 'url': 'https://www.youtube.com/playlist?list=PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6002 'info_dict': {
6003 'id': 'PLt5yu3-wZAlQAaPZ5Z-rJoTdbT-45Q7c0',
6004 'tags': [],
6005 'view_count': int,
6006 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6007 'channel': 'cole-dlp-test-acc',
6008 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6009 'description': 'test',
6010 'title': 'dlp test playlist',
6011 'availability': 'public',
6012 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6013 'uploader_id': '@coletdjnz',
6014 'uploader': 'cole-dlp-test-acc',
6015 },
6016 'playlist_mincount': 1,
6017 'params': {'extractor_args': {'youtube': {'lang': ['ja']}}},
6018 'expected_warnings': ['Preferring "ja"'],
6019 }, {
6020 # shorts audio pivot for 2GtVksBMYFM.
6021 'url': 'https://www.youtube.com/feed/sfv_audio_pivot?bp=8gUrCikSJwoLMkd0VmtzQk1ZRk0SCzJHdFZrc0JNWUZNGgsyR3RWa3NCTVlGTQ==',
6022 'info_dict': {
6023 'id': 'sfv_audio_pivot',
6024 'title': 'sfv_audio_pivot',
6025 'tags': [],
6026 },
6027 'playlist_mincount': 50,
6028
6029 }, {
6030 # Channel with a real live tab (not to be mistaken with streams tab)
6031 # Do not treat like it should redirect to live stream
6032 'url': 'https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live',
6033 'info_dict': {
6034 'id': 'UCEH7P7kyJIkS_gJf93VYbmg',
6035 'title': 'UCEH7P7kyJIkS_gJf93VYbmg - Live',
6036 'tags': [],
6037 },
6038 'playlist_mincount': 20,
6039 }, {
6040 # Tab name is not the same as tab id
6041 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/letsplay',
6042 'info_dict': {
6043 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6044 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Let\'s play',
6045 'tags': [],
6046 },
6047 'playlist_mincount': 8,
6048 }, {
6049 # Home tab id is literally home. Not to get mistaken with featured
6050 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/home',
6051 'info_dict': {
6052 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6053 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Home',
6054 'tags': [],
6055 },
6056 'playlist_mincount': 8,
6057 }, {
6058 # Should get three playlists for videos, shorts and streams tabs
6059 'url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6060 'info_dict': {
6061 'id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6062 'title': 'Polka Ch. 尾丸ポルカ',
6063 'channel_follower_count': int,
6064 'channel_id': 'UCK9V2B22uJYu3N7eR_BT9QA',
6065 'channel_url': 'https://www.youtube.com/channel/UCK9V2B22uJYu3N7eR_BT9QA',
6066 'description': 'md5:e56b74b5bb7e9c701522162e9abfb822',
6067 'channel': 'Polka Ch. 尾丸ポルカ',
6068 'tags': 'count:35',
6069 'uploader_url': 'https://www.youtube.com/@OmaruPolka',
6070 'uploader': 'Polka Ch. 尾丸ポルカ',
6071 'uploader_id': '@OmaruPolka',
6072 },
6073 'playlist_count': 3,
6074 }, {
6075 # Shorts tab with channel with handle
6076 # TODO: fix channel description
6077 'url': 'https://www.youtube.com/@NotJustBikes/shorts',
6078 'info_dict': {
6079 'id': 'UC0intLFzLaudFG-xAvUEO-A',
6080 'title': 'Not Just Bikes - Shorts',
6081 'tags': 'count:12',
6082 'channel_url': 'https://www.youtube.com/channel/UC0intLFzLaudFG-xAvUEO-A',
6083 'description': 'md5:26bc55af26855a608a5cf89dfa595c8d',
6084 'channel_follower_count': int,
6085 'channel_id': 'UC0intLFzLaudFG-xAvUEO-A',
6086 'channel': 'Not Just Bikes',
6087 'uploader_url': 'https://www.youtube.com/@NotJustBikes',
6088 'uploader': 'Not Just Bikes',
6089 'uploader_id': '@NotJustBikes',
6090 },
6091 'playlist_mincount': 10,
6092 }, {
6093 # Streams tab
6094 'url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig/streams',
6095 'info_dict': {
6096 'id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6097 'title': '中村悠一 - Live',
6098 'tags': 'count:7',
6099 'channel_id': 'UC3eYAvjCVwNHgkaGbXX3sig',
6100 'channel_url': 'https://www.youtube.com/channel/UC3eYAvjCVwNHgkaGbXX3sig',
6101 'channel': '中村悠一',
6102 'channel_follower_count': int,
6103 'description': 'md5:e744f6c93dafa7a03c0c6deecb157300',
6104 'uploader_url': 'https://www.youtube.com/@Yuichi-Nakamura',
6105 'uploader_id': '@Yuichi-Nakamura',
6106 'uploader': '中村悠一',
6107 },
6108 'playlist_mincount': 60,
6109 }, {
6110 # Channel with no uploads and hence no videos, streams, shorts tabs or uploads playlist. This should fail.
6111 # See test_youtube_lists
6112 'url': 'https://www.youtube.com/channel/UC2yXPzFejc422buOIzn_0CA',
6113 'only_matching': True,
6114 }, {
6115 # No uploads and no UCID given. Should fail with no uploads error
6116 # See test_youtube_lists
6117 'url': 'https://www.youtube.com/news',
6118 'only_matching': True
6119 }, {
6120 # No videos tab but has a shorts tab
6121 'url': 'https://www.youtube.com/c/TKFShorts',
6122 'info_dict': {
6123 'id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6124 'title': 'Shorts Break - Shorts',
6125 'tags': 'count:48',
6126 'channel_id': 'UCgJ5_1F6yJhYLnyMszUdmUg',
6127 'channel': 'Shorts Break',
6128 'description': 'md5:6de33c5e7ba686e5f3efd4e19c7ef499',
6129 'channel_follower_count': int,
6130 'channel_url': 'https://www.youtube.com/channel/UCgJ5_1F6yJhYLnyMszUdmUg',
6131 'uploader_url': 'https://www.youtube.com/@ShortsBreak_Official',
6132 'uploader': 'Shorts Break',
6133 'uploader_id': '@ShortsBreak_Official',
6134 },
6135 'playlist_mincount': 30,
6136 }, {
6137 # Trending Now Tab. tab id is empty
6138 'url': 'https://www.youtube.com/feed/trending',
6139 'info_dict': {
6140 'id': 'trending',
6141 'title': 'trending - Now',
6142 'tags': [],
6143 },
6144 'playlist_mincount': 30,
6145 }, {
6146 # Trending Gaming Tab. tab id is empty
6147 'url': 'https://www.youtube.com/feed/trending?bp=4gIcGhpnYW1pbmdfY29ycHVzX21vc3RfcG9wdWxhcg%3D%3D',
6148 'info_dict': {
6149 'id': 'trending',
6150 'title': 'trending - Gaming',
6151 'tags': [],
6152 },
6153 'playlist_mincount': 30,
6154 }, {
6155 # Shorts url result in shorts tab
6156 # TODO: Fix channel id extraction
6157 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/shorts',
6158 'info_dict': {
6159 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6160 'title': 'cole-dlp-test-acc - Shorts',
6161 'channel': 'cole-dlp-test-acc',
6162 'description': 'test description',
6163 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6164 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6165 'tags': [],
6166 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6167 'uploader_id': '@coletdjnz',
6168 'uploader': 'cole-dlp-test-acc',
6169 },
6170 'playlist': [{
6171 'info_dict': {
6172 # Channel data is not currently available for short renderers (as of 2023-03-01)
6173 '_type': 'url',
6174 'ie_key': 'Youtube',
6175 'url': 'https://www.youtube.com/shorts/sSM9J5YH_60',
6176 'id': 'sSM9J5YH_60',
6177 'title': 'SHORT short',
6178 'view_count': int,
6179 'thumbnails': list,
6180 }
6181 }],
6182 'params': {'extract_flat': True},
6183 }, {
6184 # Live video status should be extracted
6185 'url': 'https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/live',
6186 'info_dict': {
6187 'id': 'UCQvWX73GQygcwXOTSf_VDVg',
6188 'title': 'UCQvWX73GQygcwXOTSf_VDVg - Live', # TODO, should be Minecraft - Live or Minecraft - Topic - Live
6189 'tags': []
6190 },
6191 'playlist': [{
6192 'info_dict': {
6193 '_type': 'url',
6194 'ie_key': 'Youtube',
6195 'url': 'startswith:https://www.youtube.com/watch?v=',
6196 'id': str,
6197 'title': str,
6198 'live_status': 'is_live',
6199 'channel_id': str,
6200 'channel_url': str,
6201 'concurrent_view_count': int,
6202 'channel': str,
6203 'uploader': str,
6204 'uploader_url': str,
6205 'uploader_id': str
6206 }
6207 }],
6208 'params': {'extract_flat': True, 'playlist_items': '1'},
6209 'playlist_mincount': 1
6210 }, {
6211 # Channel renderer metadata. Contains number of videos on the channel
6212 'url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA/channels',
6213 'info_dict': {
6214 'id': 'UCiu-3thuViMebBjw_5nWYrA',
6215 'title': 'cole-dlp-test-acc - Channels',
6216 'channel': 'cole-dlp-test-acc',
6217 'description': 'test description',
6218 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA',
6219 'channel_url': 'https://www.youtube.com/channel/UCiu-3thuViMebBjw_5nWYrA',
6220 'tags': [],
6221 'uploader_url': 'https://www.youtube.com/@coletdjnz',
6222 'uploader_id': '@coletdjnz',
6223 'uploader': 'cole-dlp-test-acc',
6224 },
6225 'playlist': [{
6226 'info_dict': {
6227 '_type': 'url',
6228 'ie_key': 'YoutubeTab',
6229 'url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6230 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6231 'channel_id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6232 'title': 'PewDiePie',
6233 'channel': 'PewDiePie',
6234 'channel_url': 'https://www.youtube.com/channel/UC-lHJZR3Gqxm24_Vd_AJ5Yw',
6235 'thumbnails': list,
6236 'channel_follower_count': int,
6237 'playlist_count': int,
6238 'uploader': 'PewDiePie',
6239 'uploader_url': 'https://www.youtube.com/@PewDiePie',
6240 'uploader_id': '@PewDiePie',
6241 }
6242 }],
6243 'params': {'extract_flat': True},
6244 }, {
6245 'url': 'https://www.youtube.com/@3blue1brown/about',
6246 'info_dict': {
6247 'id': 'UCYO_jab_esuFRV4b17AJtAw',
6248 'tags': ['Mathematics'],
6249 'title': '3Blue1Brown - About',
6250 'channel_follower_count': int,
6251 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
6252 'channel': '3Blue1Brown',
6253 'view_count': int,
6254 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
6255 'description': 'md5:e1384e8a133307dd10edee76e875d62f',
6256 'uploader_url': 'https://www.youtube.com/@3blue1brown',
6257 'uploader_id': '@3blue1brown',
6258 'uploader': '3Blue1Brown',
6259 },
6260 'playlist_count': 0,
6261 }, {
6262 # Podcasts tab, with rich entry playlistRenderers
6263 'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
6264 'info_dict': {
6265 'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6266 'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6267 'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
6268 'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
6269 'title': '99 Percent Invisible - Podcasts',
6270 'uploader': '99 Percent Invisible',
6271 'channel_follower_count': int,
6272 'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
6273 'tags': [],
6274 'channel': '99 Percent Invisible',
6275 'uploader_id': '@99percentinvisiblepodcast',
6276 },
6277 'playlist_count': 1,
6278 }, {
6279 # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
6280 'url': 'https://www.youtube.com/@AHimitsu/releases',
6281 'info_dict': {
6282 'id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6283 'channel': 'A Himitsu',
6284 'uploader_url': 'https://www.youtube.com/@AHimitsu',
6285 'title': 'A Himitsu - Releases',
6286 'uploader_id': '@AHimitsu',
6287 'uploader': 'A Himitsu',
6288 'channel_id': 'UCgFwu-j5-xNJml2FtTrrB3A',
6289 'tags': 'count:16',
6290 'description': 'I make music',
6291 'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
6292 'channel_follower_count': int,
6293 },
6294 'playlist_mincount': 10,
6295 }]
6296
6297 @classmethod
6298 def suitable(cls, url):
6299 return False if YoutubeIE.suitable(url) else super().suitable(url)
6300
6301 _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/[^?#/]+))?(?P<post>.*)$')
6302
6303 def _get_url_mobj(self, url):
6304 mobj = self._URL_RE.match(url).groupdict()
6305 mobj.update((k, '') for k, v in mobj.items() if v is None)
6306 return mobj
6307
6308 def _extract_tab_id_and_name(self, tab, base_url='https://www.youtube.com'):
6309 tab_name = (tab.get('title') or '').lower()
6310 tab_url = urljoin(base_url, traverse_obj(
6311 tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
6312
6313 tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
6314 or traverse_obj(tab, 'tabIdentifier', expected_type=str))
6315 if tab_id:
6316 return {
6317 'TAB_ID_SPONSORSHIPS': 'membership',
6318 }.get(tab_id, tab_id), tab_name
6319
6320 # Fallback to tab name if we cannot get the tab id.
6321 # XXX: should we strip non-ascii letters? e.g. in case of 'let's play' tab example on special gaming channel
6322 # Note that in the case of translated tab name this may result in an empty string, which we don't want.
6323 if tab_name:
6324 self.write_debug(f'Falling back to selected tab name: {tab_name}')
6325 return {
6326 'home': 'featured',
6327 'live': 'streams',
6328 }.get(tab_name, tab_name), tab_name
6329
6330 def _has_tab(self, tabs, tab_id):
6331 return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
6332
6333 @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
6334 def _real_extract(self, url, smuggled_data):
6335 item_id = self._match_id(url)
6336 url = urllib.parse.urlunparse(
6337 urllib.parse.urlparse(url)._replace(netloc='www.youtube.com'))
6338 compat_opts = self.get_param('compat_opts', [])
6339
6340 mobj = self._get_url_mobj(url)
6341 pre, tab, post, is_channel = mobj['pre'], mobj['tab'], mobj['post'], not mobj['not_channel']
6342 if is_channel and smuggled_data.get('is_music_url'):
6343 if item_id[:2] == 'VL': # Youtube music VL channels have an equivalent playlist
6344 return self.url_result(
6345 f'https://music.youtube.com/playlist?list={item_id[2:]}', YoutubeTabIE, item_id[2:])
6346 elif item_id[:2] == 'MP': # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist
6347 mdata = self._extract_tab_endpoint(
6348 f'https://music.youtube.com/channel/{item_id}', item_id, default_client='web_music')
6349 murl = traverse_obj(mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'),
6350 get_all=False, expected_type=str)
6351 if not murl:
6352 raise ExtractorError('Failed to resolve album to playlist')
6353 return self.url_result(murl, YoutubeTabIE)
6354 elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/
6355 return self.url_result(
6356 f'https://music.youtube.com/channel/{item_id}{tab}{post}', YoutubeTabIE, item_id)
6357
6358 original_tab_id, display_id = tab[1:], f'{item_id}{tab}'
6359 if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts:
6360 url = f'{pre}/videos{post}'
6361 if smuggled_data.get('is_music_url'):
6362 self.report_warning(f'YouTube Music is not directly supported. Redirecting to {url}')
6363
6364 # Handle both video/playlist URLs
6365 qs = parse_qs(url)
6366 video_id, playlist_id = [traverse_obj(qs, (key, 0)) for key in ('v', 'list')]
6367 if not video_id and mobj['not_channel'].startswith('watch'):
6368 if not playlist_id:
6369 # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable
6370 raise ExtractorError('A video URL was given without video ID', expected=True)
6371 # Common mistake: https://www.youtube.com/watch?list=playlist_id
6372 self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}')
6373 return self.url_result(
6374 f'https://www.youtube.com/playlist?list={playlist_id}', YoutubeTabIE, playlist_id)
6375
6376 if not self._yes_playlist(playlist_id, video_id):
6377 return self.url_result(
6378 f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6379
6380 data, ytcfg = self._extract_data(url, display_id)
6381
6382 # YouTube may provide a non-standard redirect to the regional channel
6383 # See: https://github.com/yt-dlp/yt-dlp/issues/2694
6384 # https://support.google.com/youtube/answer/2976814#zippy=,conditional-redirects
6385 redirect_url = traverse_obj(
6386 data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False)
6387 if redirect_url and 'no-youtube-channel-redirect' not in compat_opts:
6388 redirect_url = ''.join((urljoin('https://www.youtube.com', redirect_url), tab, post))
6389 self.to_screen(f'This playlist is likely not available in your region. Following conditional redirect to {redirect_url}')
6390 return self.url_result(redirect_url, YoutubeTabIE)
6391
6392 tabs, extra_tabs = self._extract_tab_renderers(data), []
6393 if is_channel and tabs and 'no-youtube-channel-redirect' not in compat_opts:
6394 selected_tab = self._extract_selected_tab(tabs)
6395 selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
6396 self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
6397
6398 if not original_tab_id and selected_tab_name:
6399 self.to_screen('Downloading all uploads of the channel. '
6400 'To download only the videos in a specific tab, pass the tab\'s URL')
6401 if self._has_tab(tabs, 'streams'):
6402 extra_tabs.append(''.join((pre, '/streams', post)))
6403 if self._has_tab(tabs, 'shorts'):
6404 extra_tabs.append(''.join((pre, '/shorts', post)))
6405 # XXX: Members-only tab should also be extracted
6406
6407 if not extra_tabs and selected_tab_id != 'videos':
6408 # Channel does not have streams, shorts or videos tabs
6409 if item_id[:2] != 'UC':
6410 raise ExtractorError('This channel has no uploads', expected=True)
6411
6412 # Topic channels don't have /videos. Use the equivalent playlist instead
6413 pl_id = f'UU{item_id[2:]}'
6414 pl_url = f'https://www.youtube.com/playlist?list={pl_id}'
6415 try:
6416 data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
6417 except ExtractorError:
6418 raise ExtractorError('This channel has no uploads', expected=True)
6419 else:
6420 item_id, url = pl_id, pl_url
6421 self.to_screen(
6422 f'The channel does not have a videos, shorts, or live tab. Redirecting to playlist {pl_id} instead')
6423
6424 elif extra_tabs and selected_tab_id != 'videos':
6425 # When there are shorts/live tabs but not videos tab
6426 url, data = f'{pre}{post}', None
6427
6428 elif (original_tab_id or 'videos') != selected_tab_id:
6429 if original_tab_id == 'live':
6430 # Live tab should have redirected to the video
6431 # Except in the case the channel has an actual live tab
6432 # Example: https://www.youtube.com/channel/UCEH7P7kyJIkS_gJf93VYbmg/live
6433 raise UserNotLive(video_id=item_id)
6434 elif selected_tab_name:
6435 raise ExtractorError(f'This channel does not have a {original_tab_id} tab', expected=True)
6436
6437 # For channels such as https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg
6438 url = f'{pre}{post}'
6439
6440 # YouTube sometimes provides a button to reload playlist with unavailable videos.
6441 if 'no-youtube-unavailable-videos' not in compat_opts:
6442 data = self._reload_with_unavailable_videos(display_id, data, ytcfg) or data
6443 self._extract_and_report_alerts(data, only_once=True)
6444
6445 tabs, entries = self._extract_tab_renderers(data), []
6446 if tabs:
6447 entries = [self._extract_from_tabs(item_id, ytcfg, data, tabs)]
6448 entries[0].update({
6449 'extractor_key': YoutubeTabIE.ie_key(),
6450 'extractor': YoutubeTabIE.IE_NAME,
6451 'webpage_url': url,
6452 })
6453 if self.get_param('playlist_items') == '0':
6454 entries.extend(self.url_result(u, YoutubeTabIE) for u in extra_tabs)
6455 else: # Users expect to get all `video_id`s even with `--flat-playlist`. So don't return `url_result`
6456 entries.extend(map(self._real_extract, extra_tabs))
6457
6458 if len(entries) == 1:
6459 return entries[0]
6460 elif entries:
6461 metadata = self._extract_metadata_from_tabs(item_id, data)
6462 uploads_url = 'the Uploads (UU) playlist URL'
6463 if try_get(metadata, lambda x: x['channel_id'].startswith('UC')):
6464 uploads_url = f'https://www.youtube.com/playlist?list=UU{metadata["channel_id"][2:]}'
6465 self.to_screen(
6466 'Downloading as multiple playlists, separated by tabs. '
6467 f'To download as a single playlist instead, pass {uploads_url}')
6468 return self.playlist_result(entries, item_id, **metadata)
6469
6470 # Inline playlist
6471 playlist = traverse_obj(
6472 data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict)
6473 if playlist:
6474 return self._extract_from_playlist(item_id, url, data, playlist, ytcfg)
6475
6476 video_id = traverse_obj(
6477 data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id
6478 if video_id:
6479 if tab != '/live': # live tab is expected to redirect to video
6480 self.report_warning(f'Unable to recognize playlist. Downloading just video {video_id}')
6481 return self.url_result(f'https://www.youtube.com/watch?v={video_id}', YoutubeIE, video_id)
6482
6483 raise ExtractorError('Unable to recognize tab page')
6484
6485
6486 class YoutubePlaylistIE(InfoExtractor):
6487 IE_DESC = 'YouTube playlists'
6488 _VALID_URL = r'''(?x)(?:
6489 (?:https?://)?
6490 (?:\w+\.)?
6491 (?:
6492 (?:
6493 youtube(?:kids)?\.com|
6494 %(invidious)s
6495 )
6496 /.*?\?.*?\blist=
6497 )?
6498 (?P<id>%(playlist_id)s)
6499 )''' % {
6500 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
6501 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
6502 }
6503 IE_NAME = 'youtube:playlist'
6504 _TESTS = [{
6505 'note': 'issue #673',
6506 'url': 'PLBB231211A4F62143',
6507 'info_dict': {
6508 'title': '[OLD]Team Fortress 2 (Class-based LP)',
6509 'id': 'PLBB231211A4F62143',
6510 'uploader': 'Wickman',
6511 'uploader_id': '@WickmanVT',
6512 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2',
6513 'view_count': int,
6514 'uploader_url': 'https://www.youtube.com/@WickmanVT',
6515 'modified_date': r're:\d{8}',
6516 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
6517 'channel': 'Wickman',
6518 'tags': [],
6519 'channel_url': 'https://www.youtube.com/channel/UCKSpbfbl5kRQpTdL7kMc-1Q',
6520 'availability': 'public',
6521 },
6522 'playlist_mincount': 29,
6523 }, {
6524 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6525 'info_dict': {
6526 'title': 'YDL_safe_search',
6527 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
6528 },
6529 'playlist_count': 2,
6530 'skip': 'This playlist is private',
6531 }, {
6532 'note': 'embedded',
6533 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6534 'playlist_count': 4,
6535 'info_dict': {
6536 'title': 'JODA15',
6537 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
6538 'uploader': 'milan',
6539 'uploader_id': '@milan5503',
6540 'description': '',
6541 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw',
6542 'tags': [],
6543 'modified_date': '20140919',
6544 'view_count': int,
6545 'channel': 'milan',
6546 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
6547 'uploader_url': 'https://www.youtube.com/@milan5503',
6548 'availability': 'public',
6549 },
6550 'expected_warnings': [r'[Uu]navailable videos? (is|are|will be) hidden'],
6551 }, {
6552 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6553 'playlist_mincount': 455,
6554 'info_dict': {
6555 'title': '2018 Chinese New Singles (11/6 updated)',
6556 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
6557 'uploader': 'LBK',
6558 'uploader_id': '@music_king',
6559 'description': 'md5:da521864744d60a198e3a88af4db0d9d',
6560 'channel': 'LBK',
6561 'view_count': int,
6562 'channel_url': 'https://www.youtube.com/channel/UC21nz3_MesPLqtDqwdvnoxA',
6563 'tags': [],
6564 'uploader_url': 'https://www.youtube.com/@music_king',
6565 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
6566 'modified_date': r're:\d{8}',
6567 'availability': 'public',
6568 },
6569 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
6570 }, {
6571 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
6572 'only_matching': True,
6573 }, {
6574 # music album playlist
6575 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
6576 'only_matching': True,
6577 }]
6578
6579 @classmethod
6580 def suitable(cls, url):
6581 if YoutubeTabIE.suitable(url):
6582 return False
6583 from ..utils import parse_qs
6584 qs = parse_qs(url)
6585 if qs.get('v', [None])[0]:
6586 return False
6587 return super().suitable(url)
6588
6589 def _real_extract(self, url):
6590 playlist_id = self._match_id(url)
6591 is_music_url = YoutubeBaseInfoExtractor.is_music_url(url)
6592 url = update_url_query(
6593 'https://www.youtube.com/playlist',
6594 parse_qs(url) or {'list': playlist_id})
6595 if is_music_url:
6596 url = smuggle_url(url, {'is_music_url': True})
6597 return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6598
6599
6600 class YoutubeYtBeIE(InfoExtractor):
6601 IE_DESC = 'youtu.be'
6602 _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}
6603 _TESTS = [{
6604 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
6605 'info_dict': {
6606 'id': 'yeWKywCrFtk',
6607 'ext': 'mp4',
6608 'title': 'Small Scale Baler and Braiding Rugs',
6609 'uploader': 'Backus-Page House Museum',
6610 'uploader_id': '@backuspagemuseum',
6611 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@backuspagemuseum',
6612 'upload_date': '20161008',
6613 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
6614 'categories': ['Nonprofits & Activism'],
6615 'tags': list,
6616 'like_count': int,
6617 'age_limit': 0,
6618 'playable_in_embed': True,
6619 'thumbnail': r're:^https?://.*\.webp',
6620 'channel': 'Backus-Page House Museum',
6621 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw',
6622 'live_status': 'not_live',
6623 'view_count': int,
6624 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw',
6625 'availability': 'public',
6626 'duration': 59,
6627 'comment_count': int,
6628 'channel_follower_count': int
6629 },
6630 'params': {
6631 'noplaylist': True,
6632 'skip_download': True,
6633 },
6634 }, {
6635 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
6636 'only_matching': True,
6637 }]
6638
6639 def _real_extract(self, url):
6640 mobj = self._match_valid_url(url)
6641 video_id = mobj.group('id')
6642 playlist_id = mobj.group('playlist_id')
6643 return self.url_result(
6644 update_url_query('https://www.youtube.com/watch', {
6645 'v': video_id,
6646 'list': playlist_id,
6647 'feature': 'youtu.be',
6648 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
6649
6650
6651 class YoutubeLivestreamEmbedIE(InfoExtractor):
6652 IE_DESC = 'YouTube livestream embeds'
6653 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)'
6654 _TESTS = [{
6655 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA',
6656 'only_matching': True,
6657 }]
6658
6659 def _real_extract(self, url):
6660 channel_id = self._match_id(url)
6661 return self.url_result(
6662 f'https://www.youtube.com/channel/{channel_id}/live',
6663 ie=YoutubeTabIE.ie_key(), video_id=channel_id)
6664
6665
6666 class YoutubeYtUserIE(InfoExtractor):
6667 IE_DESC = 'YouTube user videos; "ytuser:" prefix'
6668 IE_NAME = 'youtube:user'
6669 _VALID_URL = r'ytuser:(?P<id>.+)'
6670 _TESTS = [{
6671 'url': 'ytuser:phihag',
6672 'only_matching': True,
6673 }]
6674
6675 def _real_extract(self, url):
6676 user_id = self._match_id(url)
6677 return self.url_result(f'https://www.youtube.com/user/{user_id}', YoutubeTabIE, user_id)
6678
6679
6680 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
6681 IE_NAME = 'youtube:favorites'
6682 IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)'
6683 _VALID_URL = r':ytfav(?:ou?rite)?s?'
6684 _LOGIN_REQUIRED = True
6685 _TESTS = [{
6686 'url': ':ytfav',
6687 'only_matching': True,
6688 }, {
6689 'url': ':ytfavorites',
6690 'only_matching': True,
6691 }]
6692
6693 def _real_extract(self, url):
6694 return self.url_result(
6695 'https://www.youtube.com/playlist?list=LL',
6696 ie=YoutubeTabIE.ie_key())
6697
6698
6699 class YoutubeNotificationsIE(YoutubeTabBaseInfoExtractor):
6700 IE_NAME = 'youtube:notif'
6701 IE_DESC = 'YouTube notifications; ":ytnotif" keyword (requires cookies)'
6702 _VALID_URL = r':ytnotif(?:ication)?s?'
6703 _LOGIN_REQUIRED = True
6704 _TESTS = [{
6705 'url': ':ytnotif',
6706 'only_matching': True,
6707 }, {
6708 'url': ':ytnotifications',
6709 'only_matching': True,
6710 }]
6711
6712 def _extract_notification_menu(self, response, continuation_list):
6713 notification_list = traverse_obj(
6714 response,
6715 ('actions', 0, 'openPopupAction', 'popup', 'multiPageMenuRenderer', 'sections', 0, 'multiPageMenuNotificationSectionRenderer', 'items'),
6716 ('actions', 0, 'appendContinuationItemsAction', 'continuationItems'),
6717 expected_type=list) or []
6718 continuation_list[0] = None
6719 for item in notification_list:
6720 entry = self._extract_notification_renderer(item.get('notificationRenderer'))
6721 if entry:
6722 yield entry
6723 continuation = item.get('continuationItemRenderer')
6724 if continuation:
6725 continuation_list[0] = continuation
6726
6727 def _extract_notification_renderer(self, notification):
6728 video_id = traverse_obj(
6729 notification, ('navigationEndpoint', 'watchEndpoint', 'videoId'), expected_type=str)
6730 url = f'https://www.youtube.com/watch?v={video_id}'
6731 channel_id = None
6732 if not video_id:
6733 browse_ep = traverse_obj(
6734 notification, ('navigationEndpoint', 'browseEndpoint'), expected_type=dict)
6735 channel_id = self.ucid_or_none(traverse_obj(browse_ep, 'browseId', expected_type=str))
6736 post_id = self._search_regex(
6737 r'/post/(.+)', traverse_obj(browse_ep, 'canonicalBaseUrl', expected_type=str),
6738 'post id', default=None)
6739 if not channel_id or not post_id:
6740 return
6741 # The direct /post url redirects to this in the browser
6742 url = f'https://www.youtube.com/channel/{channel_id}/community?lb={post_id}'
6743
6744 channel = traverse_obj(
6745 notification, ('contextualMenu', 'menuRenderer', 'items', 1, 'menuServiceItemRenderer', 'text', 'runs', 1, 'text'),
6746 expected_type=str)
6747 notification_title = self._get_text(notification, 'shortMessage')
6748 if notification_title:
6749 notification_title = notification_title.replace('\xad', '') # remove soft hyphens
6750 # TODO: handle recommended videos
6751 title = self._search_regex(
6752 rf'{re.escape(channel or "")}[^:]+: (.+)', notification_title,
6753 'video title', default=None)
6754 timestamp = (self._parse_time_text(self._get_text(notification, 'sentTimeText'))
6755 if self._configuration_arg('approximate_date', ie_key=YoutubeTabIE)
6756 else None)
6757 return {
6758 '_type': 'url',
6759 'url': url,
6760 'ie_key': (YoutubeIE if video_id else YoutubeTabIE).ie_key(),
6761 'video_id': video_id,
6762 'title': title,
6763 'channel_id': channel_id,
6764 'channel': channel,
6765 'uploader': channel,
6766 'thumbnails': self._extract_thumbnails(notification, 'videoThumbnail'),
6767 'timestamp': timestamp,
6768 }
6769
6770 def _notification_menu_entries(self, ytcfg):
6771 continuation_list = [None]
6772 response = None
6773 for page in itertools.count(1):
6774 ctoken = traverse_obj(
6775 continuation_list, (0, 'continuationEndpoint', 'getNotificationMenuEndpoint', 'ctoken'), expected_type=str)
6776 response = self._extract_response(
6777 item_id=f'page {page}', query={'ctoken': ctoken} if ctoken else {}, ytcfg=ytcfg,
6778 ep='notification/get_notification_menu', check_get_keys='actions',
6779 headers=self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)))
6780 yield from self._extract_notification_menu(response, continuation_list)
6781 if not continuation_list[0]:
6782 break
6783
6784 def _real_extract(self, url):
6785 display_id = 'notifications'
6786 ytcfg = self._download_ytcfg('web', display_id) if not self.skip_webpage else {}
6787 self._report_playlist_authcheck(ytcfg)
6788 return self.playlist_result(self._notification_menu_entries(ytcfg), display_id, display_id)
6789
6790
6791 class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6792 IE_DESC = 'YouTube search'
6793 IE_NAME = 'youtube:search'
6794 _SEARCH_KEY = 'ytsearch'
6795 _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only
6796 _TESTS = [{
6797 'url': 'ytsearch5:youtube-dl test video',
6798 'playlist_count': 5,
6799 'info_dict': {
6800 'id': 'youtube-dl test video',
6801 'title': 'youtube-dl test video',
6802 }
6803 }]
6804
6805
6806 class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor):
6807 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
6808 _SEARCH_KEY = 'ytsearchdate'
6809 IE_DESC = 'YouTube search, newest videos first'
6810 _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date
6811 _TESTS = [{
6812 'url': 'ytsearchdate5:youtube-dl test video',
6813 'playlist_count': 5,
6814 'info_dict': {
6815 'id': 'youtube-dl test video',
6816 'title': 'youtube-dl test video',
6817 }
6818 }]
6819
6820
6821 class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor):
6822 IE_DESC = 'YouTube search URLs with sorting and filter support'
6823 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
6824 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6825 _TESTS = [{
6826 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
6827 'playlist_mincount': 5,
6828 'info_dict': {
6829 'id': 'youtube-dl test video',
6830 'title': 'youtube-dl test video',
6831 }
6832 }, {
6833 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D',
6834 'playlist_mincount': 5,
6835 'info_dict': {
6836 'id': 'python',
6837 'title': 'python',
6838 }
6839 }, {
6840 'url': 'https://www.youtube.com/results?search_query=%23cats',
6841 'playlist_mincount': 1,
6842 'info_dict': {
6843 'id': '#cats',
6844 'title': '#cats',
6845 # The test suite does not have support for nested playlists
6846 # 'entries': [{
6847 # 'url': r're:https://(www\.)?youtube\.com/hashtag/cats',
6848 # 'title': '#cats',
6849 # }],
6850 },
6851 }, {
6852 # Channel results
6853 'url': 'https://www.youtube.com/results?search_query=kurzgesagt&sp=EgIQAg%253D%253D',
6854 'info_dict': {
6855 'id': 'kurzgesagt',
6856 'title': 'kurzgesagt',
6857 },
6858 'playlist': [{
6859 'info_dict': {
6860 '_type': 'url',
6861 'id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6862 'url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6863 'ie_key': 'YoutubeTab',
6864 'channel': 'Kurzgesagt – In a Nutshell',
6865 'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
6866 'title': 'Kurzgesagt – In a Nutshell',
6867 'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
6868 'playlist_count': int, # XXX: should have a way of saying > 1
6869 'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
6870 'thumbnails': list,
6871 'uploader_id': '@kurzgesagt',
6872 'uploader_url': 'https://www.youtube.com/@kurzgesagt',
6873 'uploader': 'Kurzgesagt – In a Nutshell',
6874 }
6875 }],
6876 'params': {'extract_flat': True, 'playlist_items': '1'},
6877 'playlist_mincount': 1,
6878 }, {
6879 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
6880 'only_matching': True,
6881 }]
6882
6883 def _real_extract(self, url):
6884 qs = parse_qs(url)
6885 query = (qs.get('search_query') or qs.get('q'))[0]
6886 return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query)
6887
6888
6889 class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor):
6890 IE_DESC = 'YouTube music search URLs with selectable sections, e.g. #songs'
6891 IE_NAME = 'youtube:music:search_url'
6892 _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)'
6893 _TESTS = [{
6894 'url': 'https://music.youtube.com/search?q=royalty+free+music',
6895 'playlist_count': 16,
6896 'info_dict': {
6897 'id': 'royalty free music',
6898 'title': 'royalty free music',
6899 }
6900 }, {
6901 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D',
6902 'playlist_mincount': 30,
6903 'info_dict': {
6904 'id': 'royalty free music - songs',
6905 'title': 'royalty free music - songs',
6906 },
6907 'params': {'extract_flat': 'in_playlist'}
6908 }, {
6909 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists',
6910 'playlist_mincount': 30,
6911 'info_dict': {
6912 'id': 'royalty free music - community playlists',
6913 'title': 'royalty free music - community playlists',
6914 },
6915 'params': {'extract_flat': 'in_playlist'}
6916 }]
6917
6918 _SECTIONS = {
6919 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==',
6920 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==',
6921 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF',
6922 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==',
6923 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==',
6924 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==',
6925 }
6926
6927 def _real_extract(self, url):
6928 qs = parse_qs(url)
6929 query = (qs.get('search_query') or qs.get('q'))[0]
6930 params = qs.get('sp', (None,))[0]
6931 if params:
6932 section = next((k for k, v in self._SECTIONS.items() if v == params), params)
6933 else:
6934 section = urllib.parse.unquote_plus((url.split('#') + [''])[1]).lower()
6935 params = self._SECTIONS.get(section)
6936 if not params:
6937 section = None
6938 title = join_nonempty(query, section, delim=' - ')
6939 return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title)
6940
6941
6942 class YoutubeFeedsInfoExtractor(InfoExtractor):
6943 """
6944 Base class for feed extractors
6945 Subclasses must re-define the _FEED_NAME property.
6946 """
6947 _LOGIN_REQUIRED = True
6948 _FEED_NAME = 'feeds'
6949
6950 def _real_initialize(self):
6951 YoutubeBaseInfoExtractor._check_login_required(self)
6952
6953 @classproperty
6954 def IE_NAME(self):
6955 return f'youtube:{self._FEED_NAME}'
6956
6957 def _real_extract(self, url):
6958 return self.url_result(
6959 f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key())
6960
6961
6962 class YoutubeWatchLaterIE(InfoExtractor):
6963 IE_NAME = 'youtube:watchlater'
6964 IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)'
6965 _VALID_URL = r':ytwatchlater'
6966 _TESTS = [{
6967 'url': ':ytwatchlater',
6968 'only_matching': True,
6969 }]
6970
6971 def _real_extract(self, url):
6972 return self.url_result(
6973 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
6974
6975
6976 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
6977 IE_DESC = 'YouTube recommended videos; ":ytrec" keyword'
6978 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
6979 _FEED_NAME = 'recommended'
6980 _LOGIN_REQUIRED = False
6981 _TESTS = [{
6982 'url': ':ytrec',
6983 'only_matching': True,
6984 }, {
6985 'url': ':ytrecommended',
6986 'only_matching': True,
6987 }, {
6988 'url': 'https://youtube.com',
6989 'only_matching': True,
6990 }]
6991
6992
6993 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
6994 IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)'
6995 _VALID_URL = r':ytsub(?:scription)?s?'
6996 _FEED_NAME = 'subscriptions'
6997 _TESTS = [{
6998 'url': ':ytsubs',
6999 'only_matching': True,
7000 }, {
7001 'url': ':ytsubscriptions',
7002 'only_matching': True,
7003 }]
7004
7005
7006 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7007 IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)'
7008 _VALID_URL = r':ythis(?:tory)?'
7009 _FEED_NAME = 'history'
7010 _TESTS = [{
7011 'url': ':ythistory',
7012 'only_matching': True,
7013 }]
7014
7015
7016 class YoutubeStoriesIE(InfoExtractor):
7017 IE_DESC = 'YouTube channel stories; "ytstories:" prefix'
7018 IE_NAME = 'youtube:stories'
7019 _VALID_URL = r'ytstories:UC(?P<id>[A-Za-z0-9_-]{21}[AQgw])$'
7020 _TESTS = [{
7021 'url': 'ytstories:UCwFCb4jeqaKWnciAYM-ZVHg',
7022 'only_matching': True,
7023 }]
7024
7025 def _real_extract(self, url):
7026 playlist_id = f'RLTD{self._match_id(url)}'
7027 return self.url_result(
7028 smuggle_url(f'https://www.youtube.com/playlist?list={playlist_id}&playnext=1', {'is_story': True}),
7029 ie=YoutubeTabIE, video_id=playlist_id)
7030
7031
7032 class YoutubeShortsAudioPivotIE(InfoExtractor):
7033 IE_DESC = 'YouTube Shorts audio pivot (Shorts using audio of a given video)'
7034 IE_NAME = 'youtube:shorts:pivot:audio'
7035 _VALID_URL = r'https?://(?:www\.)?youtube\.com/source/(?P<id>[\w-]{11})/shorts'
7036 _TESTS = [{
7037 'url': 'https://www.youtube.com/source/Lyj-MZSAA9o/shorts',
7038 'only_matching': True,
7039 }]
7040
7041 @staticmethod
7042 def _generate_audio_pivot_params(video_id):
7043 """
7044 Generates sfv_audio_pivot browse params for this video id
7045 """
7046 pb_params = b'\xf2\x05+\n)\x12\'\n\x0b%b\x12\x0b%b\x1a\x0b%b' % ((video_id.encode(),) * 3)
7047 return urllib.parse.quote(base64.b64encode(pb_params).decode())
7048
7049 def _real_extract(self, url):
7050 video_id = self._match_id(url)
7051 return self.url_result(
7052 f'https://www.youtube.com/feed/sfv_audio_pivot?bp={self._generate_audio_pivot_params(video_id)}',
7053 ie=YoutubeTabIE)
7054
7055
7056 class YoutubeTruncatedURLIE(InfoExtractor):
7057 IE_NAME = 'youtube:truncated_url'
7058 IE_DESC = False # Do not list
7059 _VALID_URL = r'''(?x)
7060 (?:https?://)?
7061 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
7062 (?:watch\?(?:
7063 feature=[a-z_]+|
7064 annotation_id=annotation_[^&]+|
7065 x-yt-cl=[0-9]+|
7066 hl=[^&]*|
7067 t=[0-9]+
7068 )?
7069 |
7070 attribution_link\?a=[^&]+
7071 )
7072 $
7073 '''
7074
7075 _TESTS = [{
7076 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
7077 'only_matching': True,
7078 }, {
7079 'url': 'https://www.youtube.com/watch?',
7080 'only_matching': True,
7081 }, {
7082 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
7083 'only_matching': True,
7084 }, {
7085 'url': 'https://www.youtube.com/watch?feature=foo',
7086 'only_matching': True,
7087 }, {
7088 'url': 'https://www.youtube.com/watch?hl=en-GB',
7089 'only_matching': True,
7090 }, {
7091 'url': 'https://www.youtube.com/watch?t=2372',
7092 'only_matching': True,
7093 }]
7094
7095 def _real_extract(self, url):
7096 raise ExtractorError(
7097 'Did you forget to quote the URL? Remember that & is a meta '
7098 'character in most shells, so you want to put the URL in quotes, '
7099 'like youtube-dl '
7100 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
7101 ' or simply youtube-dl BaW_jenozKc .',
7102 expected=True)
7103
7104
7105 class YoutubeClipIE(YoutubeTabBaseInfoExtractor):
7106 IE_NAME = 'youtube:clip'
7107 _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/(?P<id>[^/?#]+)'
7108 _TESTS = [{
7109 # FIXME: Other metadata should be extracted from the clip, not from the base video
7110 'url': 'https://www.youtube.com/clip/UgytZKpehg-hEMBSn3F4AaABCQ',
7111 'info_dict': {
7112 'id': 'UgytZKpehg-hEMBSn3F4AaABCQ',
7113 'ext': 'mp4',
7114 'section_start': 29.0,
7115 'section_end': 39.7,
7116 'duration': 10.7,
7117 'age_limit': 0,
7118 'availability': 'public',
7119 'categories': ['Gaming'],
7120 'channel': 'Scott The Woz',
7121 'channel_id': 'UC4rqhyiTs7XyuODcECvuiiQ',
7122 'channel_url': 'https://www.youtube.com/channel/UC4rqhyiTs7XyuODcECvuiiQ',
7123 'description': 'md5:7a4517a17ea9b4bd98996399d8bb36e7',
7124 'like_count': int,
7125 'playable_in_embed': True,
7126 'tags': 'count:17',
7127 'thumbnail': 'https://i.ytimg.com/vi_webp/ScPX26pdQik/maxresdefault.webp',
7128 'title': 'Mobile Games on Console - Scott The Woz',
7129 'upload_date': '20210920',
7130 'uploader': 'Scott The Woz',
7131 'uploader_id': '@ScottTheWoz',
7132 'uploader_url': 'https://www.youtube.com/@ScottTheWoz',
7133 'view_count': int,
7134 'live_status': 'not_live',
7135 'channel_follower_count': int,
7136 'chapters': 'count:20',
7137 }
7138 }]
7139
7140 def _real_extract(self, url):
7141 clip_id = self._match_id(url)
7142 _, data = self._extract_webpage(url, clip_id)
7143
7144 video_id = traverse_obj(data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'))
7145 if not video_id:
7146 raise ExtractorError('Unable to find video ID')
7147
7148 clip_data = traverse_obj(data, (
7149 'engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'clipSectionRenderer',
7150 'contents', ..., 'clipAttributionRenderer', 'onScrubExit', 'commandExecutorCommand', 'commands', ...,
7151 'openPopupAction', 'popup', 'notificationActionRenderer', 'actionButton', 'buttonRenderer', 'command',
7152 'commandExecutorCommand', 'commands', ..., 'loopCommand'), get_all=False)
7153
7154 return {
7155 '_type': 'url_transparent',
7156 'url': f'https://www.youtube.com/watch?v={video_id}',
7157 'ie_key': YoutubeIE.ie_key(),
7158 'id': clip_id,
7159 'section_start': int(clip_data['startTimeMs']) / 1000,
7160 'section_end': int(clip_data['endTimeMs']) / 1000,
7161 }
7162
7163
7164 class YoutubeConsentRedirectIE(YoutubeBaseInfoExtractor):
7165 IE_NAME = 'youtube:consent'
7166 IE_DESC = False # Do not list
7167 _VALID_URL = r'https?://consent\.youtube\.com/m\?'
7168 _TESTS = [{
7169 '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',
7170 'info_dict': {
7171 'id': 'qVv6vCqciTM',
7172 'ext': 'mp4',
7173 'age_limit': 0,
7174 'uploader_id': '@sana_natori',
7175 'comment_count': int,
7176 'chapters': 'count:13',
7177 'upload_date': '20221223',
7178 'thumbnail': 'https://i.ytimg.com/vi/qVv6vCqciTM/maxresdefault.jpg',
7179 'channel_url': 'https://www.youtube.com/channel/UCIdEIHpS0TdkqRkHL5OkLtA',
7180 'uploader_url': 'https://www.youtube.com/@sana_natori',
7181 'like_count': int,
7182 'release_date': '20221223',
7183 'tags': ['Vtuber', '月ノ美兎', '名取さな', 'にじさんじ', 'クリスマス', '3D配信'],
7184 '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】',
7185 'view_count': int,
7186 'playable_in_embed': True,
7187 'duration': 4438,
7188 'availability': 'public',
7189 'channel_follower_count': int,
7190 'channel_id': 'UCIdEIHpS0TdkqRkHL5OkLtA',
7191 'categories': ['Entertainment'],
7192 'live_status': 'was_live',
7193 'release_timestamp': 1671793345,
7194 'channel': 'さなちゃんねる',
7195 'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
7196 'uploader': 'さなちゃんねる',
7197 },
7198 'add_ie': ['Youtube'],
7199 'params': {'skip_download': 'Youtube'},
7200 }]
7201
7202 def _real_extract(self, url):
7203 redirect_url = url_or_none(parse_qs(url).get('continue', [None])[-1])
7204 if not redirect_url:
7205 raise ExtractorError('Invalid cookie consent redirect URL', expected=True)
7206 return self.url_result(redirect_url)
7207
7208
7209 class YoutubeTruncatedIDIE(InfoExtractor):
7210 IE_NAME = 'youtube:truncated_id'
7211 IE_DESC = False # Do not list
7212 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
7213
7214 _TESTS = [{
7215 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
7216 'only_matching': True,
7217 }]
7218
7219 def _real_extract(self, url):
7220 video_id = self._match_id(url)
7221 raise ExtractorError(
7222 f'Incomplete YouTube ID {video_id}. URL {url} looks truncated.',
7223 expected=True)